Slashdot Mirror


Michael Robertson Says Root is Safe

Kez writes "HEXUS.net caught up with Michael Robertson, CEO of Linspire, at the UK launch of Linspire 5. Their interview with Mr. Robertson covers everything from hardware support to software patents, but a comment from Mr. Robertson on using root is perhaps the most interesting: "I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer. They say 'oh, yeah, it is!', but it really isn't." I would imagine a few Slashdotters would dispute that."

1,174 comments

  1. Okay now... by DarkHelmet · · Score: 5, Insightful
    Let's see
    • Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful. Running something like apache as root, and any vulnerability in programs such as phpMyAdmin will make your whole server go poof.
    • rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.
    • ActiveX and a lot of spyware is contained in windows when running as non-administrator. It's running as admin (like most people do), that cause the majority of problems with things.
    This kind of talk is pandering to the lowest common denominator of user. Honestly, I feel users SHOULD learn a little bit about privileges before being handed the machine, and clicking on that file attachment.

    I know Slashdot attempts to soundbite things just like any other modern news media, so I'll quote:

    Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

    MySQL, for instance, runs as a separate user. If I so desired, I could limit the login / password for my MySQL account to only allow row INSERTs and SELECTs, but no DELETEs or DROPs. If someone were to break into my account, they could see my data, but at least they couldn't delete from the table. As root, they could stop and start the actual service, and wipe out the whole directory for that matter.

    I generally see what he's saying about data being king. But if your data is that important, you'll have other safeguards for protecting it, typically via (dun dun dun), user management! For instance, keep your accounting files under a different user, home directory chmodded to 700. Stuff like that.

    Then you could say "Well, it's not really about your data, it's that people could accidentally mess things up!". Well, you could accidentally drive into a wall as well, it doesn't mean we should make all cars drive at 10 miles an hour. So, I don't see the added benefit.

    Cars happen to have seat belts. Roads also have speed limits, so this analogy is flawed.

    The best way for Linux to break into the market isn't to emulate windows entirely. The best way is to take the best of what windows has to offer, and augment it with the best of what Linux has to offer. After all, look at Firefox. Firefox didn't choose to adopt ActiveX, or adopt Microsoft's proprietary style transitions, or render CSS in the same broken way, right? Neither should Linux, or in this case, Linspire.

    --
    /^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$/i
    1. Re:Okay now... by malfunct · · Score: 5, Insightful

      Even if user data is the most important thing, if you run as root on a multi user box you put every users data at risk instead of only your own.

      The other thing, and this isn't easy to do in many OS's, that would be nice is granular escalation of privledge. As you point out in your SQL example, if you need someone to do inserts you shouldn't have to allow them to delete.

      --

      "You can now flame me, I am full of love,"

    2. Re:Okay now... by HiThere · · Score: 1

      This is why I don't consider Linspire to actually be a Linux variant, though admittedly it's quite close, and can be turned into a true Linux with minimal effort (but why bother?).

      He really doesn't understand why certain choices are bad. It's not malice, it's ... well, a focus on other areas, coupled with a conviction that since he knows he's a smart businessman, his technical choices must also be right.

      --

      I think we've pushed this "anyone can grow up to be president" thing too far.
    3. Re:Okay now... by Phleg · · Score: 5, Informative

      rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.

      I dare you to try this. Dare.

      Note: you may wish to back your home directory up first. Preferably somewhere not under /, or using with someone else's permissions.

      --
      No comment.
    4. Re:Okay now... by bfields · · Score: 4, Insightful
      Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful. Running something like apache as root, and any vulnerability in programs such as phpMyAdmin will make your whole server go poof.

      He's not talking about daemons--presumably apache, mysql, etc. are still run as a separate user under Linspire, as they are in Debian. There's no reason to change that, since those users don't have usernames that people need to enter.

      He's talking about the user account that's used by the real physical user of a desktop system.

      In that case, no local exploit is needed--the attacker either uses sudo, or just sniffs the password the next time the user uses su (or whatever graphical equivalent pops up next time they try to upgrade some software).

      rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.

      For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root. Although I have heard of somewhat similar catastrophes. I doubt the typical gui/finder-like interface makes this so easy, but perhaps I'm wrong. In any case, as he points out, in the case of a single-user desktop, the most important data is in /home/joeuser. Once "joeuser" has deleted that, they're almost back to square one anyway.

      --Bruce Fields

    5. Re:Okay now... by crowemojo · · Score: 1

      For all the above reasons as well as several others, I can think of no more effective way that Michael Robertson could have made me refuse to take him and any products that he is responsible for seriously. We are talking about fundamental, basic security concepts that are obviously flying over Mr. Robertson's head.

    6. Re:Okay now... by As+Seen+On+TV · · Score: 1, Insightful

      Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful.

      That's fine, but he has a point. How much actual real-world good does that do? It does plenty of theoretical good, but so does making the speed limit 10 MPH. By far the better solution is to make sure that the system is safe from remote attacks.

      rm -Rf / as nonroot will make you give a sigh of relief.

      That sounds like a workaround to make up for a design flaw in the command-line interface to me.

      ActiveX and a lot of spyware is contained in windows when running as non-administrator.

      I don't know the first thing about spyware or Active X or Windows, so I certainly don't care. But since this isn't Windows we're talking about here, I fail to see how this is applicable.

      Honestly, I feel users SHOULD learn a little bit about privileges before being handed the machine

      Well, fortunately you're not making the decisions. The "users should have to learn" mentality is what keeps computers complicated and difficult to use.

      The best way is to take the best of what windows has to offer, and augment it with the best of what Linux has to offer.

      I think you'd have a hard time convincing anybody that things like "rm -Rf /" and "users SHOULD learn" and "limit the login / password for my MySQL account to only allow row INSERTs and SELECTs" and "home directory chmodded to 700" is the best of anything.

    7. Re:Okay now... by iamwahoo2 · · Score: 2, Funny

      Better yet, try it on someone else's computer! They'll thank you.

    8. Re:Okay now... by sketerpot · · Score: 2, Insightful
      That's fine, but he has a point. How much actual real-world good does that do? It does plenty of theoretical good, but so does making the speed limit 10 MPH. By far the better solution is to make sure that the system is safe from remote attacks.

      Which is easier: running a program as non-root or ensuring that it has no remote vulnerabilities? And can you be as sure about the second one as you can be about the first?

    9. Re:Okay now... by PDAllen · · Score: 1

      You, the person on the chair in front of the machine, can run as root without compromising security.

      You lose all protection from your own fuckups, of course, and that includes being social-engineered into clicking on the interesting file someone emailed you. But no security compromise, assuming you know enough to recognise a crack attempt when you see one, so won't actively cooperate.

      Why? Suppose I'm logged in as peter, and you try to hack my box via my mySQL installation. You need to hack mySQL, then you need to run an escalation exploit to get root access, then you have my box.

      Now suppose I'm logged in as root, and you try to hack my box via my mySQL installation. You need to hack mySQL, then you need to run an escalation exploit to get root access, then you have my box.

      Hey... I just wrote the same thing twice. Why did I do that? Maybe because mySQL does not run as root, whether or not I am logged in as root.

      However 99% of UNIX users will not recognise a hack attempt when they see it, and will run the pretty file they got by email, or install the dodgy WaReZ. So 99% of UNIX users should not ever run as root because they can't be trusted. Then 9 out of 10 of the people who can be trusted prefer to be protected from their typoes in unfortunate places and only run as root when they need to. The 1 in 1000 people left over can be trusted to run as root and are confident that they won't accidentally rm-rf whatever. They can run as root, their box is no more likely to be owned than anyone else's.

    10. Re:Okay now... by Anonymous Coward · · Score: 1, Interesting

      Actually, slightly off topic, but you have just highlighted/reminded me what I believe to be one of the problems with permissions on *nix generally.

      What we lack is that fine tuning - I should be able to specify that a particular UID can listen on ifname:80, not kick off a process as root, then setuid it...

      A heirarchical permissions set on the process tree could also be very handy... (think ACLs for the proc tree), although this could get pretty damned difficult to drive very fast if implemented badly.

    11. Re:Okay now... by Hatta · · Score: 1

      I know Slashdot attempts to soundbite things just like any other modern news media,

      Are you seriously comparing Slashdot to USA Today or Fox News? Wait.. nevermind.

      --
      Give me Classic Slashdot or give me death!
    12. Re:Okay now... by Anonymous Coward · · Score: 0
      Roads also have speed limits, so this analogy is flawed.


      The autobahn has no speed limit and half the fatality rate of the US highway system. Whose analogy is flawed now?
    13. Re:Okay now... by anthony_dipierro · · Score: 1

      Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful.

      What harmful thing can be done with root access which can't be done with access to the primary user's account?

      Running something like apache as root, and any vulnerability in programs such as phpMyAdmin will make your whole server go poof.

      As was said, this doesn't really apply to daemons. I ssh to my machines as root, but I still run apache as www.

      ActiveX and a lot of spyware is contained in windows when running as non-administrator. It's running as admin (like most people do), that cause the majority of problems with things.

      Spyware doesn't need root privileges to spy on you.

    14. Re:Okay now... by Gary+Destruction · · Score: 5, Insightful

      I totally agree with you about privilege levels. I was all about running as a non-privileged user. That was until realism and idealism clashed. Some programs literally won't work right without for example administrator rights on Windows. In the corporate environment, at least the Windows corporate environment, there are too many programs that need administrator privileges. Without Administrator rights, Citrix Client will open, try to initiate a session, fail and then close without error.

      While this is a Windows problem, it can result in a misconception that could end up being applied to other platforms. If people are used to using administrator privileges because of programs requiring them, they might think that they'll have to do the same on Linux and other systems. Avoiding Microsoft's mistakes is one thing. Undoing its influence is another.

    15. Re:Okay now... by bosewicht · · Score: 0

      Actually 'rm -Rf /' won't do what you think it will, I've tried it.... :)

      --
      There are 10 kinds of people in the world - those who understand binary and those who don't
    16. Re:Okay now... by Mr.+Slippery · · Score: 5, Insightful
      The "users should have to learn" mentality is what keeps computers complicated and difficult to use.
      Computers are complicated and difficult to use properly, and until we all admit that the picture for safety and security remains bleak.

      Automobiles are much less complicated, but we don't try to hide that complexity; we assume that people must be trained in their proper use. Why not computers?

      --
      Tom Swiss | the infamous tms | my blog
      You cannot wash away blood with blood
    17. Re:Okay now... by Mr2cents · · Score: 2, Funny

      Don't you get it? He sure does! You see, Windows has 95% of the users, and 99% of the virusses. By making it easier to hijack the system, he hopes to attract some of those great Windows hackers to Linux. Inevitably, users will follow when they see their favorite virusses are now also available on Linux!

      In the future he'll be making statements like "Passwords are for pussies!" and "Bah, firewalls, a lot of hot air I tell you!". It's part of the plan..

      --
      "It's too bad that stupidity isn't painful." - Anton LaVey
    18. Re:Okay now... by maraist · · Score: 4, Informative

      I should be able to specify that a particular UID can listen on ifname:80

      Have you looked into selinux? I don't know if it allows port 80 access from an initially non root user, but it allows you to run a locked-down root process. Problem is that it's apparently very complicated so only supports a scant few products out of the box. But web serving is one of them.

      --
      -Michael
    19. Re:Okay now... by bmw · · Score: 4, Insightful

      "rm -Rf / as nonroot will make you give a sigh of relief."

      That sounds like a workaround to make up for a design flaw in the command-line interface to me.


      How is this a design flaw? If you ask me, it is the command-line's greatest strength. You tell it to do something and it does it. If you wanted to be safe and have it confirm your request before it does each and every action you shouldn't specify the 'force' option. This is a GOOD THING!

    20. Re:Okay now... by maraist · · Score: 4, Insightful

      Don't forget, as a smart businessman, he knows how to sell his product.. Logging in is REALLY hard to sell.. Even for XP users (notice the pretty typing-free login icons in XP).. If XP required people to memorize passwords to do anything, then people would be use to it, and wouldn't bitch about it in Linux. Thus to have people adopt his product, he needs to soften the hard-core UNIX advocates's argument. Plus XP has one thing over Lin-whatever-the-hell-they-call-themselves, XP has a super-root account which nobody but MS has access to. It just isn't needed for any software/hardware installation.. I'm speaking out of my hat; I don't even know much about win-Administrator.

      --
      -Michael
    21. Re:Okay now... by As+Seen+On+TV · · Score: 2, Funny

      Computers are complicated and difficult to use properly

      They used to say the same thing about elevators. When's the last time you met an elevator operator?

    22. Re:Okay now... by EvilSporkMan · · Score: 4, Funny

      I had a friend who typed rm -rf *>tar instead of rm -rf *.tar. He was left with an empty file named tar and a whole lot of space to keep it...

      --
      -insert a witty something-
    23. Re:Okay now... by composer777 · · Score: 2, Insightful

      I ghost my machine every week or two, it seems to work fairly well as far as data protection goes. I think that if you properly back up, then the amount of time saved by running as root is actually higher than the time spent when disaster strikes from running as root.

      I really think the usage model is important. If you use linux like a windows user, and are constantly installing desktop applications (i.e. games, office apps, etc.), then the convenience of running as root is difficult to beat. If, on the other hand, you want to run a multi-user, commandline environment, then the separation of priveledges makes quite a bit of sense. For me, the convenience of root outweighs the drawbacks of the occasional rm -rf disaster. Proper backups mitigate any real risk.

    24. Re:Okay now... by ScuzzMonkey · · Score: 5, Funny

      When's the last time your elevator blue-screened and sent you and the other occupants hurtling screaming to the bottom of the shaft?

      --
      No relation to Happy Monkey
    25. Re:Okay now... by maraist · · Score: 1

      try
      cd some-dir
      ls
      rm -r some-other-dir
      cd some-other-other-dir
      ls
      rm -r some-other-other-other-dir
      cd /usr/local
      ls
      rm -r some-o-o-o-dir
      cd /var
      ls
      rm -r /etc

      oops, dislexic brain fart!

      I've seen this done on a server.. It's really really fun to see what still runs; you'd be amazed.

      --
      -Michael
    26. Re:Okay now... by NatasRevol · · Score: 0, Troll

      Ohhh, oohh, is that an "analogy"?

      Cause it's a really, really, really dumb one.

      --
      There are two types of people in the world: Those who crave closure
    27. Re:Okay now... by bcrowell · · Score: 2, Interesting
      Your arguments all make sense, but notice how some of them really apply more to a server. For instance,
      • MySQL, for instance, runs as a separate user. [...] For instance, keep your accounting files under a different user
      Well, sure, but most Linspire users probably don't run MySQL or keep accounting files for a business on their Linspire box. I mean, from the article, it's clearly aimed and Grandma who want to web surf and send e-mail.
      • Running something like apache as root, and any vulnerability in programs such as phpMyAdmin will make your whole server go poof.
      Same comment. Grandma isn't running a server, or using phpMyAdmin.

      • Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful.
      Well, the point he's making in the article is that on a personal desktop machine, it's the data in your own user account that's valuable. The exploitable program running as user gramma can still delete all of Gramma's files, without escalating to root.
      • rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.
      Well, Gramma's not likely to type that obscure command anyway. But even if she's not root, what if she types rm -Rf ~? From her point of view, on a single-user machine, that's just as bad -- she's back to a fresh install.

      And remember, when Gramma fires up her Linspire box for the first time, she doesn't have any services turned on, so actually there's not much that anyone from the outside can do without convincing her to execute an e-mail attachment or something (which Linux mail readers typically don't make it easy to do casually). Give her a hardware router between the machine and the wall, and bang, she's got a pretty decent hardware firewall as well (and it's a firewall that she doesn't need to configure or maintain).

      And suppose Gramma creates a root account, but the password she chooses is her dog's name, because she figures nobody can guess that? If I was helping her set things up, I'd be more concerned with explaining to her about how to choose a good password than with convincing her to set up a separate root account.

      Actually I think MacOS X has done a really nice job on this kind of stuff, and their strategy should probably be emulated, especially by distros aimed at home users. Everything is done using sudo. Any time you want to install a printer driver or whatever, it makes sure you're a user who's got administrator privs, and it makes you type in your password. For example, on my wife's MacOS X box, she and I both have admin privs, but our kids don't. I can't even remember the last time I had to do an su root on her box.

    28. Re:Okay now... by Anonymous Coward · · Score: 2, Informative

      You can access the SYSTEM account in WinXP if you are already administrator, by scheduling an interactive command prompt. (Basically, let the equivalent of cron start a shell with privledges set to super-root)

      Normally, however, you would not be using it for anything - there's no point, not much that can do an admin can't.

    29. Re:Okay now... by anagama · · Score: 5, Funny

      • I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.

      I did.

      I had two hard drives with RH on them, one slightly newer. I didn't want to upgrade my main system and risk it going all screwy, so I just took out the HD, put in a blank one, and installed the new system. Then I put my old HD in usb enclosure, copied over everything I needed, and then decided I'd erase the old drive. I had it in /mnt/usbhd, and of course, the directory structure looked practically identical to my new system from there on down. I mucked about making sure I had copied over everything I wanted, figured I was all set, and then, with the intent of going to / of the usb drive, I typed "cd /" and them "rm -rf *". I walked away for a while ....

      So I get back -- anyway, you can imagine the sick sort of dizy feeling that mistake can generate as one slowly begins to comprehend the magnitude of one's error. It only takes a second, one stray thought - "do I want a coke or a coffee" - while typing and out comes a "cd /" instead of a "cd /mnt/usbhd".
      --
      What changed under Obama? Nothing Good
    30. Re:Okay now... by Anonymous Coward · · Score: 5, Interesting

      That's why you set the /home directory to non-executable. No program, including rm, will walk into it unless you are root. Note that this doesn't affect the ability of non-root users to access any correctly permissioned sub-directory of /home.

    31. Re:Okay now... by Anonymous Coward · · Score: 1, Insightful

      I don't know the first thing about spyware or Active X or Windows, so I certainly don't care. But since this isn't Windows we're talking about here, I fail to see how this is applicable.

      It is remarkable, with comments like these, how the parent post got modded up.

    32. Re:Okay now... by Anonymous Coward · · Score: 0

      The elevators in my building run Windows and they regularly blue-screen. The plummeting to the bottom of the shaft bit hasn't happened yet though...

    33. Re:Okay now... by Anonymous Coward · · Score: 0

      ah, that's why i never use the shift key. the shift key is for sissies 1111. oh wait.

    34. Re:Okay now... by ilikejam · · Score: 2, Informative

      He should ask himself why he used the -r arg in combination with *.tar
      Unless he had directories which ended in '.tar', of course...

      --
      C-x C-s C-x k
    35. Re:Okay now... by andreyw · · Score: 1

      If you run as root, you or a malicious program can hose an entire system down - whether it's malware, a badly written shellscript/makefile, PEBCAK CLI problems, or a clueless luser making more space for all that horse pr0n he/she just torrented. Is this really so hard to understand?

      As a non-root user, you do *not* have the priviledges to fuck around with system files. How the hell is that a *bad* thing?

      Don't want to deal with logging-in, in order to pander to Sally Sockermom and Lou Serr? Then make the system log in, by default, into a NON-ROOT account. Really, there is nothing that a person can't live without being a USER and not a SYSTEM ADMINISTRATOR. Sheesh.

    36. Re:Okay now... by InadequateCamel · · Score: 2, Interesting

      Elevators go up and down. The only thing that straightforward on a computer is the CD drive (and even that sometimes causes my system to freeze :-) )

      I'm not suggesting that the usability of computers cannot be improved; far from it. But just as some people are simply very bad drivers, some people will not be able to use some programs because they don't have the training, they aren't willing to practice, or they just don't "get it". Trying to cater to these people by writing programs that a 5-year-old could use probably results in programs that only a 5-year-old would want to use.

    37. Re:Okay now... by alienw · · Score: 1

      rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.

      Yeah, let's see, what's better. As root, the command will delete a few files I can easily reinstall from CD (I tend to notice when an rm command takes more than half a second). As a user, this command will delete all of my documents, my code, my email archive, my MP3 collection, and so on. So, tell me, what's worse?

    38. Re:Okay now... by Rimbo · · Score: 4, Funny

      That's because elevators have emergency brakes that automatically latch on at the first sign of anything going wrong.

      Now that I think of it, it's like running as a user instead of root, isn't it? Ehrm, wait a minute...

    39. Re:Okay now... by germanStefan · · Score: 1
      "I don't know the first thing about spyware or Active X or Windows, so I certainly don't care. But since this isn't Windows we're talking about here, I fail to see how this is applicable."
      He means that any scumware that would eventually be written for Linux/MacOSx will only be able to run as the user who unknowingly (knowingly) installed it. Thus if you can't remove the offending software bc it hides, all you have to do is copy documents and data, them remove that users home directory then its gone because it can't copy itself to the systems program directories

      and how is "rm -Rf / as nonroot will make you give a sigh of relief." "a design flaw in the command-line interface to me."
      If your root then you should be able to do whatever you want. It has happend to before that I did chmod 700 /* and left out the . infront of the /. However since I was running as a restricted user I got a bunch of errors about not having permission. If your running as root by default, then no errors about permissions, it just does it.

      And finally "I think you'd have a hard time convincing anybody that things like "rm -Rf /" and "users SHOULD learn" and "limit the login / password for my MySQL account to only allow row INSERTs and SELECTs" and "home directory chmodded to 700" is the best of anything." While it is a sad fact that people still need to know a little bit about computer to use them, thats how I think it will be for a while. I don't know how many computers I have fixed and set up so that they (hopefully) can't destroy it again. I do however try to teach them a little about computers and tell them how spyware/viruses can get on their system. Whether they follow my advice or not is up to them. Telling someone to chmod their home directory to 700 wont accomplish much I agree, but telling them that running as administrator is is not fine. While I don't use Windows, that is one of my major gripes...I do think that longhorn will finally have true priveleges for users so that you don't need to be admin to do everything.

    40. Re:Okay now... by rapidweather · · Score: 1
      Knoppix does not run as root. Many of the things that have to be done require a su to root, and that is done on a per case basis.

      Perhaps Linspire runs as root to make things easier. That is not an entirely Live CD distro, however, like Knoppix. It can be installed on the hard drive.

      I wonder what kind of damage or intrusions Linspire users could expect, as compared to a Knoppix Live CD distro.

    41. Re:Okay now... by mcg1969 · · Score: 1

      I once typed "rm -rf .*" in a directory just below root on an Ultrix system. Turns out it matched "..", descended into root, and blew it away.

    42. Re:Okay now... by dilvish_the_damned · · Score: 2, Funny

      It was a dark and stormy night. Had a few beers, I was tired. I was young and did not fully comprehend the power I was weilding. I needed to remove /etc/ppp/dilvish before giving up the machine to the new admin. I typed 'rm -r /etc' and my right pinky was getting lazy from my too long of session at the keyboard and the return key was hair trigger. I didnt mean to shoot the machine, but what is done, is done. Or rather I spent the next 48 hours recovering files one inode at a time.
      Only then did I comprehend the awsome evil power of root.

      --
      I think you underestimate just how much I just dont care.
    43. Re:Okay now... by ticktockticktock · · Score: 2, Informative
      What we lack is that fine tuning - I should be able to specify that a particular UID can listen on ifname:80, not kick off a process as root, then setuid it...

      Or you could run the process non-root and setup iptables rules to redirect port 80 requests to a port a non-root user can open. I think one can also set rules so that iptables only allows certain incoming ports to certain user accounts, so that no one else can run their own apache and take over the port, although I am not 100% sure on this.

    44. Re:Okay now... by pherthyl · · Score: 1

      That's fine, but he has a point. How much actual real-world good does that do? It does plenty of theoretical good, but so does making the speed limit 10 MPH. By far the better solution is to make sure that the system is safe from remote attacks.

      Let's not kid ourselves and think we can do this.

      That sounds like a workaround to make up for a design flaw in the command-line interface to me.

      How exactly is that a flaw in the CLI? Arguably a flaw in the permission model that it will let you destroy your system with no complaint, but it's not the CLI's fault. Doing the same from a graphical file manager will destroy your system too.

      I don't know the first thing about spyware or Active X or Windows, so I certainly don't care.

      If you don't know anything about it, why are you dismissing the idea? The GP is saying that if windows users ran as non-administrator, it would be far easier to rescue a spyware ridden windows installation since the core system files would not be corruptable. This is exactly the same issue that this story is about.

      Well, fortunately you're not making the decisions. The "users should have to learn" mentality is what keeps computers complicated and difficult to use.

      Ever used OS X? People don't seem to think that's too hard to use. And anyway, the argument that "users shouldn't have to learn" is not entirely correct either. Any reasonably complex device takes some learning. Cars require extensive training to operate, and yet they aren't capable of nearly as many different tasks as computers are.

      I think you'd have a hard time convincing anybody that things like "rm -Rf /" and "users SHOULD learn" and "limit the login / password for my MySQL account to only allow row INSERTs and SELECTs" and "home directory chmodded to 700" is the best of anything.

      What are you talking about? No-one is suggesting that users learn that. I'd advise you to re-read the GP post.

      As has been mentioned in the past, even Windows is going towards a security model where normal user will not run as administrator. Going the opposite way has been proven to be a bad idea by Microsoft.

    45. Re:Okay now... by eno2001 · · Score: 1
      Cars happen to have seat belts. Roads also have speed limits, so this analogy is flawed.

      Yes. But there are a lot of idiots who don't want to wear seat belts and see them as an inconvenience. There are also people who don't want to follow the speed limits as my little flamewar yesterday indicates. I would guess that the same people who think seat belts and speed limits are a bad idea are the same kind of people who think it's OK to run as 'root' or Administrator. Hmmm... I wonder if Robertson was involved in my flamewar yesterday? ;P

      I agree that running as root is a bad idea. But the problem is that most people can't even begin to understand the reasons why no matter how clearly you try to explain it to them. I think Robertson is just saying this in the hopes that he can get some mindshare and the accompanying market share. I also think Robertson is being really thick.

      --
      -"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
    46. Re:Okay now... by Lord+Ender · · Score: 1

      Linspire is for single-user systems, right? "rm -rf" would be horrible as a single user because it would delete all the data you care about.

      --
      A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
    47. Re:Okay now... by clem · · Score: 2, Interesting

      What I'd be interested to know is if there's a means to switch between user sessions on a Linux system without logging off. This is something I actually miss from XP.

      I suppose that I could rig something that required multiple X sessions that you go between by hitting
      the CTRL-ALT-F# keys. However, it'd be nice to have something that simple folk can use.

      --
      Your courageous and selfless spelling corrections have made me a better person.
    48. Re:Okay now... by anomalous+cohort · · Score: 1
      Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

      If you wish to run a program whose trustworthiness is uncertain, then log in as a different user than what you normally log in when accessing your precious data. Neither of these user accounts should be root.

    49. Re:Okay now... by CoderJoe · · Score: 1

      Okay, Now suppose you're logged in as peter, and you are running program Y. Y just happens to have an exploitable bug in it, and access the internet. If someone were to use that bug, they would only have the access of the user named peter, and would still need an escalation exploit to gain root.

      But, if you are logged in as root and running program Y, any attacker that exploits that bug in Y now has root access directly.

      Rememer, services are not the only programs running on a system that can be exploited. User programs can as well.

    50. Re:Okay now... by cortana · · Score: 1, Insightful

      More to the point: why not do both?

      Security is about layers!

    51. Re:Okay now... by drsmithy · · Score: 2, Insightful
      MySQL, for instance, runs as a separate user.

      You are no longer talking about scenarios within the realm of the typical end-user desktop and, thus, are talking about a completely different target market to Linspires.

      You also talk about not being able to do any "damage" as a non-root user. That's right - except to your data, the most important data on 99% of machines.

      You have completely missed the point. Thanks for playing.

    52. Re:Okay now... by coma_bug · · Score: 1

      If I so desired, I could limit the login / password for my MySQL account to only allow row INSERTs and SELECTs, but no DELETEs or DROPs

      This is a bit of a kludge, no?

      Have a look at capability systems like EROS. Capabilities are like file descriptors in Unix: only processes holding the file descriptors may access the corresponding files, and there is a specific function (the open system call) to obtain file descriptors.

      Now suppose that only the initial shell may open files and that all other processes receive the content of files by redirection (established by the shell). A compromised process cannot access just any file but only those files given to it.

      Improving the redirection mechanism along the lines of Plan 9 or, similarly, MSH produces a very powerful and secure system.

    53. Re:Okay now... by cortana · · Score: 1

      Recent versions of xscreensaver and gdm (and the kde screensaver and kdm) do this.

    54. Re:Okay now... by Anonymous Coward · · Score: 0

      Run as?
      SUDO?

    55. Re:Okay now... by Dimensio · · Score: 4, Insightful

      Without Administrator rights, Citrix Client will open, try to initiate a session, fail and then close without error.

      1) A lot of programs where this happens can be fixed by adjusting configuration, or copying registry keys rather than giving the user full Admin rights.

      2) Developers who write software that absolutely requires Administrative rights for common use, and the program is not designed to alter fundamental hardware or OS configuration (such as a registry editor or a graphics driver tweak utility) are incompetent and should be killed.

    56. Re:Okay now... by John_Sauter · · Score: 3, Insightful
      ...Some programs literally won't work right without for example administrator rights on Windows. In the corporate environment, at least the Windows corporate environment, there are too many programs that need administrator privileges. Without Administrator rights, Citrix Client will open, try to initiate a session, fail and then close without error.

      In my shop, administrative rights are strictly limited, and so I see this effect also. There is some Kodak camera-handling software that complains if you run it without administrative rights (though it seems to work just fine) and a weather display application that fails like Citrix Client unless it is run as an administrator. I am sure there are other examples.

      My answer to this class of problems is to declare the software not working, and suggest that the user ask the vendor for a version that will run without administrator privileges. I have yet to see a software vendor respond positively to this request, but in the long run I think it is the only solution. I am not willing to give my users administrator privileges so they can run some poorly-written application!
      John Sauter (J_Sauter@Empire.Net)

    57. Re:Okay now... by andreyw · · Score: 1

      Actually.... as a root you risk wiping out your system without realizing what even happened - your files are gone too, bub. As a user blowing out your home directory, you then have, as root, the possibililty of recovering your files, modulo the type of filesystem you have (ext2fs good, reiser bad). Considering I have a CVS repo set up for ALL my files, whether schoolwork or source code, I would be pissed off if I wiped out my homedir - but suicidal if I wiped out the entire contents of my disk.

    58. Re:Okay now... by lakeland · · Score: 1

      KDE added support for this some time ago (try hitting 'lock screen'). Alternatively, some distros start KDM(/GDM/XDM) automatically on F7, F8, F9 -- see /etc/X11/xdm/settings or somesuch...

      But you're right, it isn't as easy as it should be.

      I guess they're happy with non-geeks hitting logout and relying on KDE to correctly restore your session when you get back (which it does a moderately good job of)

    59. Re:Okay now... by andreyw · · Score: 1

      That, and please forget about Windows/DOS single-user mentality. Linux as a multi-user system. Have fun explaining to your significant other or parents on why their tax filing information, personal documents, letters, porn, photos and other stuff is missing simply because it was "too inconventient" for you to be like everyone else and run non-root?

    60. Re:Okay now... by Anubis350 · · Score: 2, Interesting

      even better, firemen and other individuals with authority can gain "root access" by using a key and thus gain full control of and override ability on the elevator.

      --
      "goodbye and hello, as always" ~Prince Corwin, from Zelazny's Amber series
    61. Re:Okay now... by Anonymous Coward · · Score: 0

      So you need to run as root in order to "do anything harmful"? rm -rf / as nonroot will "give you a sigh of relief"?

      I don't think so.
      --
      perl -e '$??s:;s:s;;$?::s;;=]=>%-{<-|}<&|`{;;y; -/:-@[-`{-};`-{~" -;;s;;$_;see'

    62. Re:Okay now... by pHDNgell · · Score: 4, Insightful

      Logging in is REALLY hard to sell

      See NeXTSTEP and MacOS X. Users were not root. Users seem to be getting along just fine. Login optional.

      --
      -- The world is watching America, and America is watching TV.
    63. Re:Okay now... by Anonymous Coward · · Score: 0

      That's fine, but he has a point. How much actual real-world good does that do?

      Consider this: WHEN your data goes poof, you need to perform a restore of some sort. The issues of backing up aside (which is one place where Linux could easily take the market, as no other "user" operating system bothers to make backups mandatory and simple enough for everyone). Do you want to be able to restore your user data on a secure machine, or one where you ran as a root account all the time? Can you even trust your backups if you have been running as root and something may have corrupted the backup process itself? There is something to be said for privilege separation even on home user machines. In no case should anything the user does allow writing to system files, the boot partition, or accessing hardware improperly. Those silly BIOS corruption viruses wouldn't run as non-root, etc.

      There are plenty of reasons to keep system resources protected from every day usage.

    64. Re:Okay now... by ebyrob · · Score: 2, Interesting

      The "users should have to learn" mentality is what keeps computers complicated and difficult to use.

      Actually, my opinion is and always has been that assuming users are stupid and incapable of learning the most basic idioms is the real problem with computing. I mean, if we can't even expect to teach people what a "directory tree" is and means, how do we expect them to learn to organize information? Sure, google can claim you should "search instead of organize," but the fact remains there are times when searching is useful and times when indexing and organizing are useful. Knowing both is computing 101.

      The trick for developers is creating minimal yet powerful knowledge-space for users to occupy and NOT CHANGING IT! (Note: this doesn't mean the back-end doesn't change, just that the controls remain familiar... and every change is designed specifically to make usage easier, and with an eye toward disruption costs.)

      I mean really. The basic distribution model:
      1) Download application to known location.
      2) Execute application at known location.

      Hasn't changed since the very first personal computers, so why is it we even need things like ActiveX? (ie: if it's worth running, it's probably worth the trouble to purposely install...)

      Note: For moving around alot or organizations, replace "application" with "appliciation suite".

      And food for thought: Why can't I just grab the contents of my "programs" directory and move it to a new machine?

    65. Re:Okay now... by chill · · Score: 1

      The autobahn may not have a speed limit (in areas), but it has dozens of other rules, the likes of which aren't really in place or enforced in the U.S.

      --
      Learning HOW to think is more important than learning WHAT to think.
    66. Re:Okay now... by nihilogos · · Score: 1

      rm -Rf / as nonroot will make you give a sigh of relief

      I once demonstrated this to someone. 'Relief' is not quite what I felt on realizing I'd just deleted my home directory.

      --
      :wq
    67. Re:Okay now... by Anonymous Coward · · Score: 0

      I got a bluescreen in mine the other day. Admittedly, I shouldn't have been jumping up and down.

    68. Re:Okay now... by Anonymous Coward · · Score: 2, Insightful

      Uh, computers are hard to use?

      The User Interface on my Microwave Oven has never crashed. Nor has my alram clock, or stereo. My digital answering machine works every time. The ATM at the bank has a really ugly UI, but it seems to me that it has always just plain worked.

      Oh wait, you didn't know that those were all computers? Sorry, I guess I have been working in embedded systems for too long....

      Oh yeah, Windows does some ugly things at odd times that I just cannot understand. Linux does some really ugly things sometimes to, but I have the source code, so I can fix it. Ummm, no thanks, I spend all day writing code. For example, why do the printer drivers for my Laserjet IID stink so bad under Linux? Windows drives the thing beautifully. I solved that by upgrading to a 4M+ with lots-o-memory and a PS cartridge, but it still doesn't excuse the stinky printer support in Linux.

      Both are a pain in the butt to install and use. And hey, shut up, I have spent the last 20 years in Unix lala land. System 7, System V.2, V.4, HPUX, SunOS 4.x, Solaris, Linux, FreeBSD, NetBSD, and QNX.

      This isn't intended to be a Flame, just my general observation. The products that I have developed were certainly smaller than either Linux or Windows (only 20KNCSL to 80KNCSL for my part of each 'thing') but the final products were stable, and pretty easy to use.

      I think the closest you get in the general computer world to having a complex system with an easy to use, easy to learn UI would be games. It gaming systems are so easy to use, why is the 'desktop' so stinking hard?

      Grow up. Get your head out of your pants. You your system once while trying to pretend not to know anything. You'll soon see what a mess it is.....

    69. Re:Okay now... by JanneM · · Score: 1

      When's the last time you met an elevator operator?

      Last week, at the Takashimaya department store north of Namba in Osaka.

      Why?

      --
      Trust the Computer. The Computer is your friend.
    70. Re:Okay now... by alienw · · Score: 1

      If I type in rm -rf / instead of something else, I would notice when it takes a few seconds. It will try to delete /bin, /boot, /dev, and /etc before it gets to /home. I will most likely notice it before it gets there. As a user, it will very quickly figure out it doesn't have write permissions to anything in / and will start deleting the home directory.

    71. Re:Okay now... by GerbilSoft · · Score: 1
      I've done worse.

      cat /dev/mem > /dev/hda

      Don't ask how or why. Luckily, I had backups of my Linux partitions. (My Windows partitions weren't as lucky.)
    72. Re:Okay now... by MegaManXcalibur · · Score: 1

      Well, fortunately you're not making the decisions. The "users should have to learn" mentality is what keeps computers complicated and difficult to use.

      Computers, like any complex device, are going to require some the user to learn something. Lets use the car analogy here. A car is a complex piece of machinery and people are required to learn how to safely operate one before they are given a license to drive. The same thing goes for many complex piece of equipment.

      You can't just let somebody use something without any knowlege on how to use it. Do you know how dangerous roads would be if everybody wasn't required to take some basic training? It would be far worse then it already is (just because you complete the training doesn't mean you'll be a good driver but at least you can't claim you didn't know anything when you get into an accident).

      In fact it would be similar to how the computing environment is now. Needless to say Microsoft labeled Windows as so easy anybody could run it without much in the way of training. Where did that get us? Well the majority of the computing environment (i.e. desktops running Windows) is now one riddles with zombie servers, viruses, and identify theft because the majority of people don't know the first thing about basic computer security.

      The point I'm trying to make is if people were required to take some basic computer training before being allows to operate a computer (or at least go online with one) we wouldn't have such a dire situation. Sure there would be plenty of zombie servers out there from people being lazy but many people would also take some basic security practices so we wouldn't have as many problems as we do today.

      Like any complex piece of equipment computers should require at least some basic training to use (maybe mandatory classes in school). If people aren't willing to learn the basics of computing then they shouldn't be running their own systems in the first place.

    73. Re:Okay now... by Anonymous Coward · · Score: 0

      sorry to hear about your data loss :(

      but it sure is amusing.

      I've trained myself to never use relative paths and wildcards when running rm -rf. It's just not worth the timesavings to say:

      cd /tmp/trashdir
      rm -rf ./*

      instead of

      rm -rf /tmp/trashdir

      a single typo in the cd command above will leave you in the dir you were in before. better hope you notice before you rm -rf !!

    74. Re:Okay now... by maraist · · Score: 1
      Well, fortunately you're not making the decisions. The "users should have to learn" mentality is what keeps computers complicated and difficult to use.

      Ok, so people SHOULD NOT have to learn how to drive a car.. It should be as intuitive, and easy to use as possible.. We shouldn't confuse them with keys or remote-entry button-dethingies; you're discriminating against older people that don't have good tactile agility. You shouldn't require that people take tests to drive cars.. They're consumers, let them do what they want; so long as they give us good business.. Don't FORCE them to have air-bags, because we're worried about how air-bags kill babys or worse, cost a lot of money to repair the dashboard if you get into a minor fender-bender (I was one of these paranoid types for a while, mind you).

      Ok, so a computer and a car are different.. I get your point. I'll grant you that.. With computers, you can't steal people's social security numbers, bank-account numbers, secret-access-codes-to-terrorist-interested-facili ties, you can't cause a MS-windows oriented network connected medical device to crash, you can't cause infect a hard drive or a cell phone and cause it to overload and self desctruct (yes, you read correctly), you can't be held responsible because a hacker has 0wn3d your computer and used it to perform a highly illegal activity (and in this day and age, the government isn't paranoid about suspending civil liberties in it's cursade against Muslims (I mean terrorists)).

      So no, I see you're point.. Test people for planes, trains and automobiles, because they have immediate visible consequences to the consumer for misuse, but leave our virtual-lives to the playground of incompetents.

      That rant being made...

      rm -Rf / as nonroot will make you give a sigh of relief.

      That sounds like a workaround to make up for a design flaw in the command-line interface to me.

      Except that there are good reasons why you'd want to rm -r the root of a partition. Or even of the "chroot"ed directory. I won't go further into detail. Don't claim ignorance and then profess a flaw.

      ActiveX and a lot of spyware is contained in windows when running as non-administrator.

      I don't know the first thing about spyware or Active X or Windows, so I certainly don't care. But since this isn't Windows we're talking about here, I fail to see how this is applicable.

      Because Linspire is aspiring to have universal interpolibility of applications LIKE activeX.. Look at open-office, or gnome, or KDE. Here applications all talk with each other... They're all part of one big global community of trusted software.. They're trusted because they're of the same user-id. BUT, each user has independent application-spaces. (for gnome/kde, it's by user-name-named UNIX sockets). I think open-office literally is just one running application. Whenever a central widget in KDE dies, all of my KDE apps become unresponsive.. So I use a mixture of gnome for desktop and KDE for applications.. Thus at least I can click on my stupid task-bar. The EXACT same phenomena occurs in windows.. One shared component stalls for any reason and like a traffic-jam, every other intersection clogs. Ok, fine, great.. But if you're running these applications as ROOT! I'll elaborate no more.

      Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful.

      That's fine, but he has a point. How much actual real-world good does that do?

      Well, UNIX wasn't built in a Windows-wannabe world. 90% of the Linux machines I use are not sitting in front of me. They're servers or peer-work-stations. And likewise, others at my company or part of my social demographic or whatever are likewise on those same machines. It isn't a one machine one

      --
      -Michael
    75. Re:Okay now... by DShard · · Score: 2

      Of course it is a good thing. If you go the other way you make it harder to do the right thing whenever you figure out what that may be. Coddling the users makes systems insecure, overbearing and uncooperative. The GP is a certified moron for posting in a linux thread that an administrator shouldn't be able to do whatever the hell he pleases.

      Computers are the most complicated tools that _everyone_ can use but few are willing to learn. Hey maybe longhorn will meet this need, but I bet you Microsoft doesn't consider running as Administrator as a viable option anymore.

    76. Re:Okay now... by Anonymous Coward · · Score: 0

      not in that particular case, nom since you ARE using a the forced flag.

      however, unix has a long list of examples of horrible error handling and plain silly UI design.

      Take tar. The syntax is tar -cf archive.tar foo bar.

      What is archive.tar exists? It simply overwrites it. What if you don't provide foor or bar? It makes an empty file.

      Simple shit that didnt need to happen.

    77. Re:Okay now... by Anonymous Coward · · Score: 0

      Suse has a "Switch User" button in the panel menu. Not much simpler than that...

    78. Re:Okay now... by The+Wookie · · Score: 1

      For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.

      Many years ago, I was trying to free up space on a Sparcstation 1. I was trying to clear stuff out of /usr/demo or something similar just off /usr. The directory had several subdirectories starting with . that weren't picked up by the rm -rf *. So, I did rm -rf .* which unfortunately included .. and it started cleaning out /usr. I freed up a bit more space than I intended.

    79. Re:Okay now... by Anonymous Coward · · Score: 1, Insightful

      This seems like a silly debate. Why must it be one size fits all? Cars aren't built that way, to use the example du jour. You can shift your gears yourself, or you can have 'em shifted for you. Consumer choice.

      In one way I agree with the otherwise brain-dead parent post: the proper way to focus computer interface design these days is maximizing its ability to adapt to a spectrum of user sophistication and experience, native intelligence, and taste. I'm old enough to remember being charged $1000/hour for CPU time (1980s dollars, so multiply accordingly for 2005 dollars). Under those circumstances obviously you minimize the fussing you do with the computer, and you expect people to get trained before they use it. The cost of human time spent training is nothing compared to computer time that might be wasted if humans are untrained.

      But those days are gone. In any modern firm, human time is fantastically more expensive than computer time, which has become dirt cheap. It's even generally -- sorry slashdotters -- lots cheaper than programmer time, when you are talking upper management and the programming can probably be outsourced to Bangladesh anyway.

      It just makes very little sense to be sending $150/hour people to take yearly training seminars when you can spend the equivalent of an extra $2/hour in salary for them making sure their computing technology adapts itself to their every quirk and whim.

      Think of it this way: the CEO of a big multinational needs to communicate with important folks in many other countries. But he doesn't invest 3-4 years becoming proficient in each language he might have occasion to use. That's a crazy waste of his time. Instead the company hires a few top-notch translators, permanently or on a contract basis, and they plug them in as needed. It should be the same way with ubiquitous computing technology. The power of the hardware and the cleverness in the software should be making sure the computing resources adapt themselves to the peculiarities of their human operators. That's the route to maximizing efficiency in the system.

    80. Re:Okay now... by dknj · · Score: 0, Troll

      and this is why you set your shell prompt to your hostname and current path. you get lazy, you pay the price... sorry

      -dk

    81. Re:Okay now... by identity0 · · Score: 2, Insightful

      Right, while I have never done an "rm -Rf /", I have hosed my /lib directory when I thought I was just deleting some game's stupid lib directory.

      I didn't know just how important /lib was until I did that and found a lot of basic CLI utils didn't work anymore.

      The thing is, unless you have your shell prompt display the full path you're in or do a "pwd" before every delete, UNIX's dir naming traditions make it easy to mistake top-level dirs for some subdirectories. For example, a lot of programs will have their own 'bin' or 'lib' directories, some of the top-level dirs are repeated in /usr, so on.

    82. Re:Okay now... by man_of_mr_e · · Score: 1

      No. There is almost *NO* software that *REQUIRES* Admin rights to run, other than tools that are, by their nature, admin tools.

      Most people are just too lazy to figure out the correct permissions and/or privileges needed to NOT run as an admin. There are lots of tools to use to troubleshoot this when a vendor doesn't bother to document them.

    83. Re:Okay now... by fbjon · · Score: 1
      A long time ago I almost did that, but not recursively. Had to recompile my kernel.

      I've also done the same in windows, a long time ago. Had to figure out what my intricate config.sys tunings and autoexec configs were. Notice that "del" doesn't automatically delete hidden, ro, or system files, which saved me. Both times would have been totally irrelevant if I had been unable to do such a thing.

      --
      True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
    84. Re:Okay now... by maraist · · Score: 1

      When's the last time you met an elevator operator?

      It's called your computer.. HA! I crack myself up.

      --
      -Michael
    85. Re:Okay now... by Ramadog · · Score: 1
      I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root

      Very early in my linux experience I that. It was an install I wanted to get rid and was wondering how linux would handle the filesystem being deleted from a running system.

      The drive containing windows was mounted. Bit of a hint something was wrong when windows would not boot afterwards.

      These days I prefer dd if=/dev/zero of=/dev/hda for killing a running system.

    86. Re:Okay now... by hawk · · Score: 1

      Elevators go up and down. The only thing that straightforward on a computer is the CD drive (and even that sometimes causes my system to freeze :-) )

      Quite obviously, you should be putting hotter coffee in your cupholder . . .

    87. Re:Okay now... by kyliaar · · Score: 1

      I don't disagree with you on either point. However, you have to look at this within context.

      Lindows is attempting to provide a better alternative to Windows. They have to give up certain things that we hold as truisms in the Unix world.

      I personally feel, especially after reading the interview, that the approach of running the user as root for lindows is a good idea. It does offer the choice of setting up users but doesn't force you too.

      Lindows is meant for a workstation platform, not a windows platform. A proper default confirugation for a workstation will still be incredibly more secure and stable than a Windows workstation will be, even if all local user apps are running as root.

      If this is a home user, he is going to be just as responsible for his security as a windows user. Here education and experience is needed in either case. At least, he won't wind up with an infected machine in the 20 minutes it takes him to connect to the internet and download any needed security patches.

      In a corporate environment, the IT department will be responsible for deciding which configuration gets deployed and can easily take root privileges from the user since that is an option during install.

    88. Re:Okay now... by lamber45 · · Score: 1
      A while ago in a local LUG we talked about Linux-side systems that do things like this. Really, the best place for this is the display-manager, although it might need tighter integration with the screensaver to work well.

      On the other hand, Linux can be set up so that login/logout is really fast, especially if you use good hardware and a lightweight windowmanager like, oh, say, FVWM, along with a session manager. You won't necessarily get your documents back at the same page and line after a logout/login sequence, but no information will be lost. I've seen Windows take five minutes to "apply the system security policy" or whatever; if Linux or any Unix takes less than 30 seconds to display a usable desktop, something is probably broken, for instance a bogus NFS mount.

    89. Re:Okay now... by Anonymous Coward · · Score: 0
      2) Developers who write software that absolutely requires Administrative rights for common use, and the program is not designed to alter fundamental hardware or OS configuration (such as a registry editor or a graphics driver tweak utility) are incompetent and should be killed.

      Hey, could I ask you for a favor? I've got a couple Russian VB programmers who are in need of your services.
    90. Re:Okay now... by brianosaurus · · Score: 3, Interesting

      I'd like to add the fact elevators didn't always have light-up buttons labelled for each floor. There used to be a lever to make it go up or down. Stopping at a floor was a skill. It was more convenient to have an operator than have people miss the floor by 3 feet and break their ankles climbing out, or maybe cutting each other in half by accidentally bumping the lever when exiting.

      Now there is a much simpler and intuitive interface that anyone can use, so a dedicated operator is not needed (though I hear Congress still has elevator operators so those busy politicians don't have to worry about breaking their nails, or something).

      If you had a computer with a set of buttons for each of a few trivial operations available to the user, and those are the only operations, it probably doesn't matter if you run as root or not.
      Such a system would also suck as a general purpose home computer.

      If you're going to do anything beyond trivial actions, and perhaps getting into complex stuff that you don't necessarily understand, its probably best NOT to be running as root.

      Think of it as 2 sets of operations:

      - the ones that can mess up your stuff
      - the ones that can mess up the whole system

      Both sets have the ability to wipe out your data, but the latter can wipe out other people's data, critical system files, raw hard drives... pretty much screw your data, and your machine.

      Both your user account and root have the ability to mess up your stuff. A regular user account typically cannot mess up other accounts' data or the operating system, without using "su" or "sudo" or some other method to escalate privliges.

      MacOSX has root separate from the user account. A user can be an "Administrator", which gives the user sudo capability. GIU operations (software installs, editing user accounts, and other system configuration) do a graphical equivalent to sudo, prompting the user for their admin password. Its not that complicated. Its an extra layer of protection, and lets the user know that they're doing something out of the ordinary. Its not that complicated.

      Even my parents understand it.

      --
      blog
    91. Re:Okay now... by jasperc · · Score: 1

      Good lord... The data is the most important part of your comptuer? I call bullshit. The computer plus its broadband connectiom are the most imprtant part, since that gives you shells and the ability to run a proxy- or spamming mail server.

      --
      I'm not an actor, but I play one on TV.
    92. Re:Okay now... by TheKarateMaster · · Score: 1

      I have a theory: The longer the post, the more likely the sheep are to mod it +1 informative / interesting.

      see? I drew a nice graph.
      .......0
      s.....0.
      c....0..
      o...0...
      r..0....
      e.0..... .length.

    93. Re:Okay now... by Anonymous Coward · · Score: 0

      I did too, rm -Rf / as root that is.

      I was a newbie and wrote a bash script that executed it. I was watching it happen and managed to stop it after seeing a whole bunch of errors flow past saying that /dev/whatever couldn't be deleted. But it was too late because /bin was already gone and I had no commands left. The funny thing was that it was still serving web pages and probably would have continued until the next reboot killed it.

    94. Re:Okay now... by Artega+VH · · Score: 3, Informative

      urr doesn't that make the directory non browseable?

      --
      groklaw, wired and slashdot. The holy trinity of work based time wasting.
    95. Re:Okay now... by killjoe · · Score: 4, Insightful

      It's not logging in that he is concerned about, it's click N run. Apt won't install under non root user.

      I would think that it would take some serious hacking to allow apt to install software into your own home directory but it would probably be a worthwhile effort.

      --
      evil is as evil does
    96. Re:Okay now... by pg110404 · · Score: 1

      I dare you to try this

      The difference here between that and "deltree c:\ /y" for example is that in windows, a lot of system files get nuked in the process and you can ONLY reinstall the entire OS because enough files got nuked to make it unusable. The worst that happens when you do that as a regular user in unix/linux is that you manage to wipe out your home directory and possibly all the files in /tmp. The system is still intact, but all your personal data files are gone.

      I think the original point was that unlike windows, as a non-privileged user, you can only screw yourself and not take down the whole system.

    97. Re:Okay now... by Mr.+Slippery · · Score: 1
      Oh wait, you didn't know that those were all computers? Sorry, I guess I have been working in embedded systems for too long....

      Then I'm sure that you understand the difference between a general-purpose personal computer and an embedded systems, neh? One of them being that end users tend not to install other software on an embedded system. Or demand a theme-able skinnable GUI. Or run your software on incompatible hardware platforms. Speaking of which...

      For example, why do the printer drivers for my Laserjet IID stink so bad under Linux?

      Ask HP. Perhaps if you want to run Linux you should buy a printer that works more than "Mostly" with Linux? (I recommed Samsung's ML-1430, it's worked great for me.)

      It gaming systems are so easy to use, why is the 'desktop' so stinking hard?

      Because the things you do in games are fewer and less complex than things you do on the desktop, as are the things you do them to and with.

      Game designed get to contrain actions, subjects, and objects, but computer users want to create their own objects (files), and install their own software (creating new actions and subjects).

      --
      Tom Swiss | the infamous tms | my blog
      You cannot wash away blood with blood
    98. Re:Okay now... by NatteringNabob · · Score: 1

      >For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root. I've done it. Once.

    99. Re:Okay now... by Frumious+Wombat · · Score: 1

      I was called in to fix coworkers blowing away /dev on a new Sparcstation once. I told them they could restore it from /dev/cdrom, /dev/tape, or type it in manually from /dev/keyboard.

      --
      the more accurate the calculations became, the more the concepts tended to vanish into thin air. R. S. Mulliken
    100. Re:Okay now... by krunk4ever · · Score: 1

      rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.

      i really think linux should implement a trash/recycle bin like system for rm. right now, my friend has aliased rm to move to /trash/ instead of the default delete. so incases like the above mentioned, what you get isn't a whole bunch of data gone, but just moved to a temporary location which you can empty or really delete later.

    101. Re:Okay now... by Aeiri · · Score: 1

      iptables -A INPUT -p tcp --dport 80 -m state --state NEW -m owner --uid-owner NON_ROOT_ACCOUNT --cmd-owner httpd -j ACCEPT

      Given all other "new" packets are dropped afterwards.

    102. Re:Okay now... by MrZaius · · Score: 4, Informative

      >urr doesn't that make the directory non browseable?

      Yes. That's a good thing, for the reasons described in the parent post. It bears repeating that he did NOT say to set /home/* non-executable, but only the /home/ directory itself. This allows users access to subdirectories of /home/, but only the ones they know about independently.

      An "ls -l --recursive /home/" will fail to find any world-readable directories, because it won't be able to get a listing of /home/

      An "ls -l /home/bob/public_stuffs" will work just fine, however, with the permissions set properly.

    103. Re:Okay now... by Philodoxx · · Score: 1

      While this is a Windows problem, it can result in a misconception that could end up being applied to other platforms. If people are used to using administrator privileges because of programs requiring them, they might think that they'll have to do the same on Linux and other systems. Avoiding Microsoft's mistakes is one thing. Undoing its influence is another.

      Chalk this up to whatever you want, but when I set up my first linux system I was irritated at the notion that I should run as anything less than a super-user. This was coming from 15ish years of running windows exclusively, so the idea that I should not have all the power at by beck and call seemed absurd (oh the indescretion of my youth).

      Omnipotent godlike powers make the system user lazy, and it's very easy to get entrenched in the kind of thinking that being an admin at all times brings with with it. That being said, I was glad to read that windows is adopting a more UNIX-like permisisons model for longhorn. Now it will just be a matter of convincing software developers and average joes that this is the right way of doing things.

      --
      Oh, a lesson in history from Mr. I'm my own grandpa.
    104. Re:Okay now... by roju · · Score: 1
      Spyware doesn't need root privileges to spy on you.

      But it sure helps. For instance, a regular user has no way to run a process without it appearing in the process list. A regular user can't load kernel modules. On the other hand, root can do both those things.

      Spyware installed as root can become invisible, by taking advantage of root's powers. It makes the problem that much harder to detect and diagnose. If you run as root all the time, there's just no way to know if you've been owned. On the other hand, running under a regular user account, you can be a lot more confident that there isn't an invisible process/module watching you.

      Not to mention that a lot of viruses like to guarantee their execution by piggybacking into system files. Remember when instead of worms, we had viruses that would infect .exe files? If all of your apps are root:root r-xr-xr-x and you aren't running as root, then it's a lot harder for a virus to add itself to a system binary.
    105. Re:Okay now... by Anonymous Coward · · Score: 0
      The only thing that straightforward on a computer is the CD drive

      Is that next to the cup holder?

      I think the cooling fans are pretty straight forward.

    106. Re:Okay now... by brianosaurus · · Score: 1

      If everyone was using root instead of user accounts, all those people you've known who have typed "rm -rf *" or "rm -rf /" as nonroot would have been doing it as root.

      You're not seriously advocating that single users should just run as root, are you?

      --
      blog
    107. Re:Okay now... by labratuk · · Score: 1

      Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

      I'll tell you all something. All these windows desktop users getting their machines cracked & infected aren't because of elite hackers wanting to gain access to the highly important secretive letters to grandma on peoples hard drives. They want to use your computer as a platform to do their nasty things for them.

      --
      Malike Bamiyi wanted my assistance.
    108. Re:Okay now... by PostItNote · · Score: 1
      For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.


      No, but I have typed rm -Rf * when in the root directory, by mistyping a previous cd. That sucked a lot, even though I caught it in 15 seconds.
    109. Re:Okay now... by gone_bush · · Score: 1
      rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare

      When I was learning *nix I managed to perform rm -Rf /usr/bin. At that moment I graduated from the University of Adversity, School of Hard Knocks. You learn a lot at that place but, boy, are the fees high!

      --
      Two roads diverged in a wood, and I - I took the one less travelled by. (Robert Frost, 1916)
    110. Re:Okay now... by muzza · · Score: 1

      I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.

      A user on a system I administered ran rm -Rf /. Fortunately he wasn't root. Deleted a bunch of data in a shared directory but otherwise we were OK. I recoverd the data from the previous nights backup and celebrated; if it had happened a few months earlier it would have been a disaster- before I took over the administration every-one logged in as root. Just as well it was a Red Hat box and not a Linspire one :-)

    111. Re:Okay now... by Anonymous Coward · · Score: 0

      If by Longhorn you mean Tiger.

    112. Re:Okay now... by Dimensio · · Score: 1

      Hey, could I ask you for a favor? I've got a couple Russian VB programmers who are in need of your services.

      They'll have to get in line behind the Microsoft programmers who wrote the Age of Empires games.

    113. Re:Okay now... by XMyth · · Score: 1

      Re: MySQL

      How about UPDATE ? Certainly you'll need to change your data from time to time, eh? :)

    114. Re:Okay now... by OzRoy · · Score: 1

      I had a guy do it to one of our servers. Although he accidently did rm -rf ..

      Luckily he realised what was happening and stopped it before it took out the entire disk and we were able to restore a lot of it by copying all the system files from another machine. As long as the machine was never rebooted it was fine.

    115. Re:Okay now... by Profane+MuthaFucka · · Score: 2, Funny

      you can imagine the sick sort of dizy feeling

      Once I was making root and boot rescue disks, had the images all set, and typed dd if=/tmp/root.img of=/dev/hda1

      I meant to type dd if=/tmp/root.img of=/dev/fd0h1440

      Oops.

      --
      Fascism trolls keeping me up every night. When I starts a preachin', he HITS ME WITH HIS REICH!
    116. Re:Okay now... by maraist · · Score: 1

      While what you say is true.. I think you're missing the argument. It isn't that having multiple user-accounts for different tasks isn't good or useful.. The article (and it's supporters) are saying the data is king.. And the OS itself holds no interesting data to people that would purchase Linspire or even win XP Home. The data is the only thing of value.. And since the data is owned by the same person that would have to log in. Thus there is no greater security than if there were only one user for the whole system.

      I have argued several times in this thread that there are other reasons just as great as the data (namely the integrity of the hardware that you've purchased or spent time setting up, or register-per-install-software). But the issue of your personal data (your spread-sheet files, word-processor documents, pictures, music, bookmarks, etc).. All this is available to any hacked application run by the user. So an open-office macro that runs file-system operations, or an exploited firefox instance, or interesting perl-hooked-into-GNORBA tainted input, can all be just as dangerous as on windows everyone gets a free Admin-car! You get an Admin, and you get an Admin,..

      Really, there are still ways you can sell pre-configured machines that are easy to use but hard to exploit. Have a separate user account for each major application-suite. Your gnucash owned by gnuuser. Your open-office by oouser. Your desktop (KDE/Gnome) by another user. Especially your browser and email reader as their own users. Even if you set their passwords all the same or establish a single-sign on, each program won't have the ability to directly touch the files of the other applications.. So long as only GUI applications can make use of such inter-user file-access (via behind-the-scenes sudoing or what-have-you), you're about as safe as you can be.. Course this doesn't work so well with the gnome mentality of taskbar+email+firefox integration.. But that's the price you pay.

      --
      -Michael
    117. Re:Okay now... by anthony_dipierro · · Score: 1

      For instance, a regular user has no way to run a process without it appearing in the process list.

      They could replace the program that a user uses for listing processes though.

      A regular user can't load kernel modules.

      You don't need to run kernel modules to spy on people.

      On the other hand, root can do both those things.

      Maybe, maybe not. Depends how secure your OS is.

      If all of your apps are root:root r-xr-xr-x and you aren't running as root, then it's a lot harder for a virus to add itself to a system binary.

      But how is that necessary? You only need to infect one executable. Or even easier, you can just infect the .bashrc.

    118. Re:Okay now... by Anonymous Coward · · Score: 0

      Just about as bad... I was installing Gentoo, and had all my data on hard drive /dev/hda, wanting to put Gentoo on /dev/hdb. It was just that time to make the file system on my drive I didn't care about--no worries about backing up since I was working on a different drive, right? Right.

      mke2fs -j /dev/hda3 (instead of /dev/hdb3)
      Ten seconds later I realized what I did. Ouch.

    119. Re:Okay now... by anthony_dipierro · · Score: 1

      rm -Rf / as nonroot will make you give a sigh of relief. As root will be your nightmare.

      -bash-2.05b# whoami
      root
      -bash-2.05b# rm -rf /
      rm: cannot remove `//lost+found': Permission denied
      rm: cannot remove `//package': Permission denied
      rm: cannot remove `//bin': Permission denied
      rm: cannot remove `//boot': Permission denied
      rm: cannot remove `//dev': Permission denied
      rm: cannot remove `//etc': Permission denied
      rm: cannot remove `//home': Permission denied
      rm: cannot remove `//lib': Permission denied
      rm: cannot remove `//mnt': Permission denied
      rm: cannot remove `//proc': Permission denied
      rm: cannot remove `//root': Permission denied
      rm: cannot remove `//sbin': Permission denied
      rm: cannot remove `//tmp': Permission denied
      rm: cannot remove `//usr': Permission denied
      rm: cannot remove `//var': Permission denied
      rm: cannot remove `//opt': Permission denied
      rm: cannot remove `//.journal': Permission denied
      rm: cannot remove `//initrd': Permission denied
      rm: cannot remove `//www': Permission denied
      rm: cannot remove `//command': Permission denied
      rm: cannot remove `//service': Permission denied
      rm: cannot remove `//tools': Permission denied
      -bash-2.05b# lsattr -d /
      ----i-------- /

    120. Re:Okay now... by Anonymous Coward · · Score: 0

      I use both Windows and Linux. With Linux, I always run as a non-root account, have nicely configured directories separating things by needed permission, &c. With Windows, it's Administrator all the way... just because it's such a hassle to try otherwise. My experience separating permissions in Windows has definitely been negative.

    121. Re:Okay now... by SamAMac · · Score: 1

      You know what drives me insane? The Windows OpenAFS client requires Admin rights. Turns out that to thwart user stupidity, they have the daemon put open ports through the Windows Firewall when it starts. This of course requires Admin rights. (Actually, considering that most people run with Admin rights all of the time, being able to modify the firewall config programmatically defeats the whole purpose of having a firewall.)

    122. Re:Okay now... by TCM · · Score: 1

      Wrong. Any element in a path needs to have execute permission or traversing stops there. "ls -l /home/bob/public_stuffs" will not work if /home has no execute bit set.

      --
      Of course it runs NetBSD. BTC: 1NT7QvbetmANwaMzhpVL6
    123. Re:Okay now... by TCM · · Score: 1

      Wrong. See my post below.

      --
      Of course it runs NetBSD. BTC: 1NT7QvbetmANwaMzhpVL6
    124. Re:Okay now... by Anonymous Coward · · Score: 0

      As another poster mentioned, SuSE (9.2 at least) has this automated under "Switch User".

    125. Re:Okay now... by Lumpy · · Score: 1

      Administrative rights for common use, and the program is not designed to alter fundamental hardware or OS configuration (such as a registry editor or a graphics driver tweak utility) are incompetent and should be killed.

      I am printing that out in 300point on the hp plotter in engineering and I am going to hang that in the IS department where the programmers work tonight.

      I am sure that I will get talked to by management but it is so worth it.

      THANKS!

      --
      Do not look at laser with remaining good eye.
    126. Re:Okay now... by maraist · · Score: 4, Insightful

      There are some good replies here, but nobody's talked about "su" and friends.. I know su's not a user-friendly application, but damnit, I use it all the time. After several OS upgrades, whenever something fishy is going on with an application, I open a terminal window, login as a dummy userand run the application from there with a fresh configuration. Viola, proper settings, it must be my dot-files being mangled in the upgrade.. Time to hunt-save, and rm -r that dot-directory. Harder to do in gnome since they're all in a common tree. And yes, this is more of a power-user thing.

      But if I want to visit some illicit web site, and I don't trust that my cookie files won't be sought out by some clever Ajax tricks (hey, it's new, we can fear it), I at least launch a different one of the dozens of install browsers, or if I'm really paranoid, I log in as the dummy user. (again takes half a second from a terminal window). With the exception of X-atom-based consolidation of browsers, so long as I run a different base application (epiphony, mozilla, firefox, galean, etc), I can have two different users displaying graphics on the X-session.

      Again, I know.. power-user stuff.. But you could have (as I've pushed for in other posts) applications on the task bar launching applications of different users.. Especially if you're the distribution writer.. And ESPECIALLY if you're a single-user-signon distribution.

      --
      -Michael
    127. Re:Okay now... by Anonymous Coward · · Score: 0
      He should ask himself why he used the -r arg in combination with *.tar
      Unless he had directories which ended in '.tar', of course...


      Because he wanted to remove all the tar files in the current dir and all sub-dirs?
    128. Re:Okay now... by noidentity · · Score: 1

      Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

      Right, so I back up my data regularly. What's the next thing? My system's stability. I don't want executables compromised since they can then cause further problems. If my data gets corrupt, it's not going to spread.

      Er well I guess this is all theoretical at the moment, since I still run Classic Mac OS 9 :)

    129. Re:Okay now... by roju · · Score: 1
      But you ignored my point. I wasn't arguing that it was necessary to be root in order to spy on someone. It's just that it's possible to do a better job as root. Security isn't about absolutes, it's about taking precautions. Just because I have a lock on my door, doesn't mean no-one can break into my house. However, it makes it less-likely.
      For instance, a regular user has no way to run a process without it appearing in the process list.
      They could replace the program that a user uses for listing processes though.

      No, they can't. Well, that's not quite what I mean. They could alias ps to /home/foo/.trojan/ps, for the low-priv user. Thing is, when Joe User calls his nerdy cousin Dwight up because his computer is slow as shit, and Dwight comes over and logs in to the console as root, typing /bin/ps aux is going to show the trojan running.
    130. Re:Okay now... by fossa · · Score: 1

      How is this a design flaw? If you ask me, it is the command-line's greatest strength. You tell it to do something and it does it.

      In a nutshell, humans mistakes are inevitable. The command line is like a mischievious genie, doing exactly what you asked for. A proper design takes human imperfection into account and never assumes the human always means exactly what she says.

      Note that an annoying "are you sure?" command line is also improper design. Fairly obvious that, as it obnoxiously gets in the way.

      Undo is a good thing. It is my opinion that a system designed with universal undo in mind would not find much that couldn't be undone, even rm -r /. Of course, Unix is not like this and so we must choose between the mischievious genie and the annoying questions that are soon ignored and thus less than useless.

    131. Re:Okay now... by nolife · · Score: 1

      MySQL, for instance, runs as a separate user.

      I agree with your entire post and I do not agree with running as root either but...
      The whole purpose of and market for Linspire/Lindows/Lyroris and whatever, is not going to be the AVERAGE person and the average person is not using a SQL database to store and reference their data. It will placed as regular files under /home/user/my\ documents just like in Windows. That is what he is refering too.
      The possible vector to discount his theory of the data being owned by the user so who cares would be a rogue program that would need root privleges to run could then wipe out the users data. That would be the likely situation, just as Windows suffers from now. I actually have shortcuts to start IE as a reglaur user on my Win systems. It works well.

      --
      Bad boys rape our young girls but Violet gives willingly.
    132. Re:Okay now... by Anonymous Coward · · Score: 0

      I used to work on an old Unix (forget which) which had an 'unlink' command. It was much quicker than 'rm'. Amazed onlookers that the box stayed running.
      # unlink /.* /*

    133. Re:Okay now... by mcslappy · · Score: 1

      *raises hand*

      i've done it, and killed a redhat box. wasn't much of a loss, i threw debian on right after.

    134. Re:Okay now... by CylanR77 · · Score: 1

      I've started to feel as though the ausomobile:computer analogy is a very bad, misleading one. The car is designed for one purpose: to go fowards. That's it.

      We've put comfortable baskets for us to ride in on top of them, made these baskets climate controlled, added devices to play music, and so forth. But you can remove all of these things, and the car is still functionally a car: it will go fowards.

      But cars are complicated. They are built with intricate machines at their heart, much like computers with their chips and cards [made with hundreds of millions of transistors]. This is the similarity between the two, and this is the basis for most people using the cars:computers analogy. Unfortunately, it ignores the fact that the uses of computers are far, far more broad than a car's uses could ever hope to be.

      A computer is an information-enabling device. They're used to play games. They're used to solve complicated math problems. They're used to write essays. They're used to create special effects for movies. They're used to instantly communicate with someone across the continent. They're used to access an incredibly vast amount of mostly unrelated information. They're used as jukeboxes. They're used as photography studios. They're used to access an incredibly vast number of mindless amusements. They're used as shopping malls. They're used as repositories of personal information.

      Each of these functions is as primary a function of the computer as the next. Or in other words, the computer's specific purpose is to be a general purpose device. It has an untold number of exclusive intended uses, while a car only has one. This is why I feel that the car:computer analogy is a poor one.

      *Note: I know that some could argue that the computer's primary purpose is to perform the execution of code, and that things such as racing, driving to the store, etc. are roughly analogous to the computer's uses I mentioned, but these activities in a car are all variations of the same thing to most everybody; in all cases, you're going fowards. With a computer though, ask your average person how similar playing UT2004 is to managing your finances on a computer. The computer's executing code in both cases, but I imagine that almost everyone would agree that these two activities practically have nothing in common with each other.

      --
      http://cylan.deviantart.com/gallery/
    135. Re:Okay now... by yoshi_mon · · Score: 1

      Automobiles are much less complicated, but we don't try to hide that complexity; we assume that people must be trained in their proper use. Why not computers?

      Because unless you decide to pick up a computer and brain someone with it they are not a lethal weapon.

      Apples and oranges my friend.

      --

      Really, I know what I'm doing...Ohhhh, look at the shiny buttons!
    136. Re:Okay now... by prockcore · · Score: 1


      They used to say the same thing about elevators. When's the last time you met an elevator operator?


      I'm sure a certain delivery man would have appreciated an elevator operator a few weeks ago in new york.

    137. Re:Okay now... by Eric+Damron · · Score: 1

      In this world of software viruses and trogans I can't think of any reason to run as root. A virus will run with the permissions of the user who starts it. That is reason enough not to run as the root user unless you really need to.

      --
      The race isn't always to the swift... but that's the way to bet!
    138. Re:Okay now... by Anonymous Coward · · Score: 0

      boot -r ?

    139. Re:Okay now... by jon855 · · Score: 0

      Last night.

      --
      May /. rule the /.ing realm
    140. Re:Okay now... by Anonymous Coward · · Score: 0

      Years ago my friend had a NSF presentation to get funding to continue his project. Day before the presentation, he typed

      cd ; \rm -rf *

      of course he had the wrong shell; he acidentally typed it in on the shell he was root in. The only reason he noticed was the command was taking too long to return. The beautiful thing is that there were NFS mounted disks on there... He lost a lot, much of it not even his.

      Anyway, he pulled off the funding with a lot of bandaid and creative presentation. But people do accidentally do this kind of thing. In this case, proper permissions could not overcome user incompetence.

    141. Re:Okay now... by Anonymous Coward · · Score: 0

      Will that stop rm from descending?
      /home and other directories are probably not immutable.

    142. Re:Okay now... by quetzalc0atl · · Score: 1

      i also had a friend who typed "rm * .gz" instead of "rm *.gz" and accidentally deleted about 5,000 lines of source code that he had written and not backed up.

    143. Re:Okay now... by Red+Alastor · · Score: 1

      He meant : Logging is really hard to sell to Windows Users.

      --
      Slashdot anagrams to "Sad Sloth"
    144. Re:Okay now... by swimin · · Score: 1

      What about doing what everyone else does: logging in as user, then using kdesu?

    145. Re:Okay now... by rscrawford · · Score: 1

      I did it myself. Admittedly it required that the previous administrator had:

      a. Aliased rm to follow symbolic links; and
      b. Put a symbolic link to / in his home directory.

      It was still amazing that I got to keep my job, though.

      --
      -- The reason it's called the right wing? Irony.
    146. Re:Okay now... by Pete · · Score: 1

      Systrace is the kind of thing you're looking for. As far as I can tell, it's available for NetBSD, OpenBSD and Linux. I'm not surely if it's actively maintained.

    147. Re:Okay now... by anthony_dipierro · · Score: 1

      I wasn't arguing that it was necessary to be root in order to spy on someone. It's just that it's possible to do a better job as root.

      You haven't presented anything better that you can do as root. There might be more ways to do the same thing, but that isn't any better.

      Security isn't about absolutes, it's about taking precautions.

      Security is about making the cost to an attacker breaking the security more than the worth derived by the attacker (although at low costs you have to remember the worth to some attackers of simply enjoying screwing up someone's system). I just don't see how running as a non-root user puts you over the top. Sure, if you run as non-root, and everyone else runs as root, then you're probably safer just because the most widespread scripts are going to be targetting others, but this just seems to me like security through obscurity.

      No, they can't. Well, that's not quite what I mean. They could alias ps to /home/foo/.trojan/ps, for the low-priv user. Thing is, when Joe User calls his nerdy cousin Dwight up because his computer is slow as shit, and Dwight comes over and logs in to the console as root, typing /bin/ps aux is going to show the trojan running.

      What if they changed the LD_LIBRARY_PATH?

      Really, the best solution is for no one, including root, to be able to change /bin/ps. I believe BSD supports this through the immutable flag. In order to change a file marked immutable, you have to boot into single user mode.

    148. Re:Okay now... by anthony_dipierro · · Score: 1

      Yes. My /home directory is not immutable, but it wasn't touched when I ran "rm -rf /".

    149. Re:Okay now... by Red+Alastor · · Score: 1

      I think it's better to script a new command (you could name it del) that will have safeguards (like moving file to trash instead). This way, apps that wants to use the real rm won't have problems.

      --
      Slashdot anagrams to "Sad Sloth"
    150. Re:Okay now... by _Sprocket_ · · Score: 1
      ...the CEO of a big multinational needs to communicate with important folks in many other countries. But he doesn't invest 3-4 years becoming proficient in each language he might have occasion to use. That's a crazy waste of his time. Instead the company hires a few top-notch translators, permanently or on a contract basis, and they plug them in as needed.

      That's just crazy. You're telling me you expect a CEO to have to deal with the intricacies of culture? And then have to hire expensive translators? Just to talk to another person? Bah. This language thing just isn't going to get anywhere until linguists get together and create a uniform language that can be understood by all speakers while being fluid enough to be modified by each individual speaker.

      Some things are just a bit more complex than you seem to imply. I appreciate the idea - and there's some nice goals in there. But some things are simply complex. Everything isn't a toaster.

      Once again we have the automobile analogy. Here's another angle. Early automobiles were very difficult to operate. They broke down constantly. They were difficult to operate (imagine life without an electric starter). They were expensive. Yet they've become imbiquious in society and a common analogy of ease-of-use. No wonder. The modern automobile has little resemblance to the early technology. But is it really simplified?

      Modern automobiles are mechanically complex; so much that the term "shade tree mechanic" has become an anachronism. Few people even know how to change their oil - some don't even understand the importance of doing so. Automobile controls are a wide selection of buttons and leavers of varying importance in any given situation - all of them potential distractions. And distractions are dangerous. The infrastructure to support automobiles is increadably more advanced but involve a complex set of rules. Those rules only further adds to the increased demands of today's driving conditions that involve considerably more traffic at much, much faster speeds.

      In short, today's automobile is a vast improvement over early technology. The new technology allows for considerably more capability. But is this common icon of simple technology really all that simple?

      And this is a technology that has, for the most part, a single purpose.
    151. Re:Okay now... by Anonymous Coward · · Score: 0

      ere are lots of tools to use totroubleshoot this when a vendor doesn't bother to document them.

      such as?

    152. Re:Okay now... by cbowland · · Score: 1
      Solaris 10 does this with "Process Rights Management." Essentially, this allows a process to have a defined subset of root priviledges without full root access. Running a web server on port 80 is a perfect example. Using process rights management, the apache user can be allow to open port 80 and that's it. No buffer overflow to gain root priviledge.

      See Sun for more info.

      --

      Give a man a fish and he will eat for a day.
      Teach him to eat and he will fish forever.

    153. Re:Okay now... by jerw134 · · Score: 1

      First of all, the deltree command is gone. Second of all, running an equivalent command would absolutely not hose the system. You would lose all of your personal stuff, but no system files could be touched.

      Please, quit trolling. You know nothing about how Windows works.

    154. Re:Okay now... by roju · · Score: 1
      Just so we get the tone of this conversation right, I'm enjoying this exchange. It's causing me to evaluate my assumptions and to question my beliefs (well, my beliefs wrt linux security).
      You haven't presented anything better that you can do as root. There might be more ways to do the same thing, but that isn't any better.
      It is possible, as root, to load code at runtime into the kernel that is totally invisible to anyone poking at the computer. It is NOT possible (within some epsilon of NOT, anyway) for an arbitrary user to hide running code from root, without first gaining root.

      My SOP for dealing with weirdness is to log into a console and run top as root. If I always ran as root, the spykit could be completely invisible. If I typically ran as user, then I'll see something weird.
      What if they changed the LD_LIBRARY_PATH?
      They can't change root's LD_LIBRARY_PATH. That's the whole point of having multiple users.
    155. Re:Okay now... by Anonymous Coward · · Score: 0
      I have a friend who went back from ASIC design to Elevator Repair. He apparently paid his way through college doing elevator repair; and now he's practically the only guy in the Union with a EE degree and a union membership - so he gets paid about 2X what he did doing asic design.

      Because of his EE/CS background, he gets assigned all the modern computerized elevators in the company's buildings in SF.

      He said the most common repair is to reboot when they blue-screen.

    156. Re:Okay now... by Mr.+Slippery · · Score: 1
      Or in other words, the computer's specific purpose is to be a general purpose device. It has an untold number of exclusive intended uses, while a car only has one. This is why I feel that the car:computer analogy is a poor one.

      Depends on the context of the analogy. To fuse your argument with mine, if a car which has only one purpose requires training for it to be used effectively, why would be beleive for a moment that a general purpose device could be used effectively by the ignorant?

      --
      Tom Swiss | the infamous tms | my blog
      You cannot wash away blood with blood
    157. Re:Okay now... by nokilli · · Score: 1

      What I don't understand is why the *nixes don't implement something like the Mac's trash can.

      OK, strike that, I'm sure it's been implemented, so maybe the question is, why isn't something like that installed as a default?

      Create an invisible directory under each and every mount that is called .trash, and when *any* user does *any* rm command, instead of deleting the files outright, simply move them into the .trash directory.

      It would let you recover from some of the more catastrophic rm's, and it wouldn't/shouldn't impair the time it takes to execute the command.

      When finally you are shopping around for disk space, only then do you consider emptying the trash. Ideally you do this on semi-regular schedule that is decoupled from the act of rm'ing, but even if the situation required the rm followed immediately by the empty, you'd at least have one more chance to not be an idiot.

    158. Re:Okay now... by ilikejam · · Score: 1
      #!/bin/bash

      mkdir test
      cd test
      touch tartest.tar
      cd ..
      rm -rf *.tar
      cd test
      ls

      Nope. Try again AC.

      --
      C-x C-s C-x k
    159. Re:Okay now... by giantsfan89 · · Score: 1

      Without Administrator rights, Citrix Client will open, try to initiate a session, fail and then close without error.

      Are you talking about the protocol driver error? Yes, you have to install the Citrix ICA client (8.1.*) as an admin, but it can be run as any user.

      And on Linux, you can install it either as non-root or root, and run as non-root or root.

      --
      Don't ping my cheese with your bandwidth!
    160. Re:Okay now... by AvantLegion · · Score: 1
      >> 'm old enough to remember being charged $1000/hour for CPU time (1980s dollars, so multiply accordingly for 2005 dollars)

      *calculates using the current "strength" of the US dollar...*

      "35 cents".

    161. Re:Okay now... by Tony+Hoyle · · Score: 1

      The deltree command is *not* gone. It's called rmdir /s now.

      Running it from c:\ will very rapidly hose your system. Try it if you're so certain it won't.

    162. Re:Okay now... by Anonymous Coward · · Score: 0

      Not correct. There's many games who require admin privs for copyprotection or anti-cheating purposes. I suppose you could grant the user account "Act as part of the OS", "Impersonate another user", and so on, but you might as well run as Administrator at that point.

    163. Re:Okay now... by man_of_mr_e · · Score: 1

      Regmon, Filemon, etc...

    164. Re:Okay now... by As+Seen+On+TV · · Score: 1

      Of course, Unix is not like this and so we must choose

      Not exactly.

    165. Re:Okay now... by As+Seen+On+TV · · Score: 1

      What you should have comprehended was the awesomely bad design of a user interface that assumes you're infallible.

    166. Re:Okay now... by Anonymous Coward · · Score: 0

      How the hell did the first reply manage to get so long?

    167. Re:Okay now... by As+Seen+On+TV · · Score: 1

      Doing the same from a graphical file manager will destroy your system too.

      It certainly will not. Because when you drag files to the trash, you can put them back where they came from. Dragging a file to the trash destroys nothing. Dragging a file to the trash, emptying the trash and clicking "okay" destroys it.

      The argument was that we need a user-hostile privileges system because it prevents people from entering easily mis-typed commands that do drastic things. That's dumb. What we need to fix is the interface. Throwing a whole layer of crap on top to try to work around that fundamental flaw is not a good solution.

      Ever used OS X?

      Heh. And here I thought my reputation had preceded me.

    168. Re:Okay now... by Tony+Hoyle · · Score: 1

      I always keep a static busybox around for those eventualities.

      I have in fact quite recently done rm -rf * in the root. I was trying an app and it immediately dumped itself in the root directory.. and refused to go anywhere else (lots of hardcoded paths). I decided it sucked, and decided to delete it. Forgot I was in the root for a second...

      It got as far as /etc before I killed it... lost all the configuration files, most of /dev, /bin, /boot...

      Much fun was had by all.

    169. Re:Okay now... by jerw134 · · Score: 1

      The deltree command is gone. I know that they have replaced it, but it's still gone. Typing "deltree" and pressing Enter will spit back an error.

      Remember, we're talking about a non-Administrator. Of course running it as an admin will screw things up, but Users cannot screw up anything outside of their profile.

    170. Re:Okay now... by Mr.+Slippery · · Score: 1
      Because unless you decide to pick up a computer and brain someone with it they are not a lethal weapon.

      The issue is not the consequences of mis-use, but the complexity of learning proper use.

      Anyway - bad software kills. Probably the best known incident is the Therac 25 software failure.

      Peter G. Neumann notes over 700 deaths in 24 incidents - and that was just by 1986. (All software developers should read PGN's RISKS Forum.) Some more fatal incidents can be found here.

      --
      Tom Swiss | the infamous tms | my blog
      You cannot wash away blood with blood
    171. Re:Okay now... by As+Seen+On+TV · · Score: 1

      I mean, if we can't even expect to teach people what a "directory tree" is and means, how do we expect them to learn to organize information?

      Expecting people to change the way they work to accommodate design flaws in the tool you've built for them is the wrong idea entirely. There's a better way.

    172. Re:Okay now... by Tony+Hoyle · · Score: 2, Informative

      Users are administrators by default on Windows boxes.

      That's why they are so prone to viruses, becoming spam zombies, etc.

      A properly admined box wouldn't have that issue, but then half the coporate machines I've used haven't been properly admined let alone the home ones.

      The only OS I know of besides Unix that enforces proper user/admin separation by default is OSX (it does it really nicely in fact).

    173. Re:Okay now... by Lockz · · Score: 1

      And you fried it without going -f? Couldn't just reboot when it asked "are you sure you want to be stupid?"

      --
      Life is the sport of champions. Those who lose, die.
    174. Re:Okay now... by As+Seen+On+TV · · Score: 1

      Why not do both? Didn't you read the interview? The system has fundamental design flaws, okay? One of those design flaws is that it's not at all robust or forgiving. Delete the wrong file, the computer stops working. I mean it just stops working. That's obviously a massive design flaw.

      The solution to that massive design flaw, according to some, is to add a whole layer to prevent the user from doing certain operations. This makes perfect sense on a multiuser system, particularly a timesharing multiuser system, but it makes no sense at all on a single-user system. It results in users being told that they don't have sufficient privileges to install software and things like that. It's not a good solution.

      The problem arises when people use this added layer of multiuser security as an excuse for not fixing the fundamental design flaws. "We don't need to make the system survivable," goes the theory, "because we'll just tell people to accept a high degree of inconvenience in the name of 'security.'"

      The problem is that so many people choose to sit on their hands denying that the flaw even exists rather than putting their big ol' brains toward solving it.

    175. Re:Okay now... by kenaaker · · Score: 1
      This happened on a AIX system longer ago than I like to believe. I just wanted to clean out a bunch of . files from a user's home directory. So I entered

      rm -rf .*

      in the directory in question. By the time I realized that .. matched .* most of the system was gone.

    176. Re:Okay now... by fossa · · Score: 1

      Can one undo a commandline "rm" in OSX? Can one undo a "move to trash, empty trash" in OSX? An honest question; I don't know much about OSX. Or can search for "deleted but not yet overwritten files", and OSX makes an effort to overwrite oldest files first and only when necessary? (that would actually be pretty nice...)

    177. Re:Okay now... by jerw134 · · Score: 1

      I understand that they're admins by default, and I would not argue with that. But again, I am specifically talking about non-admins, and only non-admins.

      Admin by default is definitely a weak point for Windows, and it's something Microsoft is focusing on for Longhorn. It will finally be a thing of the past.

    178. Re:Okay now... by anthony_dipierro · · Score: 1

      It is possible, as root, to load code at runtime into the kernel that is totally invisible to anyone poking at the computer.

      Under linux, with kernel modules, this is possible. Even without kernel modules, it's possible to do it upon the next reboot (in Linux, under any distro I know of that doesn't run off a CD or other non-writeable filesystem).

      I guess my lack of attention to this detail comes from the fact that I learned unix on FreeBSD, and in FreeBSD you can set the immutable flag on the kernel so that no one, including root, has permission to modify it. In order to remove the immutable flag, you have to be running in single user mode. It was actually just today that I came to the full realization that most Linux distros don't do this, and as far as I can tell it isn't even possible to do it on Linux.

      My SOP for dealing with weirdness is to log into a console and run top as root. If I always ran as root, the spykit could be completely invisible. If I typically ran as user, then I'll see something weird.

      OK, but you know what you're doing. The average Joe isn't going to. And if you've already noticed the weirdness, I would assume you're going to continue exploring further and further until you eventually find the culprit, even if that means booting from a CD and reinstalling everything.

      They can't change root's LD_LIBRARY_PATH. That's the whole point of having multiple users.

      It just seems to me that as you increase the number of things that you need root access to do, you also increase the ease with which an attacker can get a user to run a trojan. After all, if one needs root to install most programs, then a user isn't going to think twice when running an installer for a trojan which tells them to use sudo or su or whatever. And then, when it comes down to you and me, well we always check the MD5sum of the program before we compile and install it, whether we're root or not, right?

      In my opinion, in order to properly protect an end-user from himself you need a whole paradigm shift. Firefox shouldn't run as root, but it shouldn't run as "anthony" either. Firefox should run as "firefox". And it should be installed as "firefox". Now there's something which can actually protect your data when you accidently click on a link to a website which exploits a hole in firefox. On my server system I've set up these "package users", although perhaps ironically, my only login to the system is root (it makes sense to me, because I only log in to the system to perform maintenance on it).

    179. Re:Okay now... by bfields · · Score: 1
      A virus will run with the permissions of the user who starts it.

      Right. But if that user su's to root on a regular basis (as will be the case if it's the only user on a single-user desktop--they need to upgrade every now and then...), then escalating to root privileges will be trivial.

      --Bruce Fields

    180. Re:Okay now... by ogl_codemonkey · · Score: 1
      The best way is to take the best of what windows has to offer, and augment it with the best of what Linux has to offer.


      Where exactly do you `take' a 90+% user and software market share from?
    181. Re:Okay now... by man_of_mr_e · · Score: 1

      Games don't need administrator either. They just need physical write access to the CD device, a privilege you can grant explicitly.

    182. Re:Okay now... by papaskunk · · Score: 1

      Okay, here we go, trying it right now. rm -Rf / , Okay, nothing happening, nosiree, everything here seems to be just fi

    183. Re:Okay now... by Anonymous Coward · · Score: 0

      Not unless they can change all the applications that currently rely on being admin to function correctly.

    184. Re:Okay now... by maglor_83 · · Score: 1

      I did that but with * .c Subversion was my friend that day.

    185. Re:Okay now... by As+Seen+On+TV · · Score: 0

      Can one undo a commandline "rm" in OSX?

      One would not do such a thing in Mac OS X.

      Can one undo a "move to trash, empty trash" in OSX?

      One can't undo an "empty trash." But in order to do an "empty trash," you have to first put the file in the trash, then when asked you have to confirm that you really mean to empty the trash. You can't undo a nuclear missile launch, but you can require that two people be there to turn the keys. Same basic idea.

    186. Re:Okay now... by Anonymous Coward · · Score: 0

      Well, I'm not going to try "rm -rf /", but:

      # ls testdir/subdir
      file
      # lsattr -d testdir testdir/subdir
      ----i-------- testdir
      ------------- testdir/subdir
      # rm -rf testdir
      rm: cannot remove directory `testdir/subdir': Permission denied
      # ls testdir/subdir
      #


      That's about what I'd expect. I feel like I'm missing some detail.

    187. Re:Okay now... by shadowbearer · · Score: 1

      we assume that people must be trained in their proper use

      It's an assumption, all right :-D

      SB

      --
      It's old. The more humans I meet, the more I like my cats. At least they are honest.
    188. Re:Okay now... by strangemonkey · · Score: 1

      In this case, the Citrix client requires write access to HKLM\Software\Microsoft\MSLicensing\Store (see this Citrix KB article) So you don't have to give the users full admin rights.

      It's a software development problem, not strictly a "Windows problem". With most applications, developers assume users have admin rights, and don't test their products any other way. Even big guys like Citrix.

      I suppose it's a Windows problem for reasons of history: in Win9x everyone was a root user, so Windows programmers developed bad habits. By and large, OSS developers (usually) assume you have minimal rights.

      This stuff is still a nightmare to administer. If you want to remove admin rights from users, you're forced to spend hours finding permissions tweaks like this for every new application. (Use regmon & filemon from sysinternals.com - that's how I found the Citrix tweak.)

    189. Re:Okay now... by mcc · · Score: 1

      Logging in is REALLY hard to sell

      OS X seems to do okay with it. Of course, they still have a default user when the computer switches on.

      OS X of course doesn't even have a root user. As in, there's no root account. There are only accounts which happen to have sudo access.

    190. Re:Okay now... by Justin205 · · Score: 1

      While that was an attempt to be funny, may I remind you why that will not happen by simply removing files: Any currently used programs are in RAM or in swap. The only case where it would cut off the program is if it tried to access something later that had been deleted, and was needed, and the program crashed on itself because it couldn't access the expected file. But that wouldn't happen in most programs. At most they'd spit an error or two out.

      For example, if you remove the rm binary while running rm, it won't magically stop. It'll continue until that command is finished, and then, once done, it will not be able to use it again, until you restore your binary or install a new one.

      --
      "Your effort to remain what you are is what limits you."
    191. Re:Okay now... by anthony_dipierro · · Score: 1

      I dunno, maybe you're using a different version of rm?

      -bash-2.05b# mkdir testdir
      -bash-2.05b# mkdir testdir/subdir
      -bash-2.05b# touch testdir/subdir/file
      -bash-2.05b# ls testdir/subdir
      file
      -bash-2.05b# chattr +i testdir
      -bash-2.05b# rm -rf testdir
      rm: cannot remove `testdir/subdir': Permission denied
      -bash-2.05b# ls testdir/subdir
      file
      -bash-2.05b# lsattr -d testdir testdir/subdir
      ----i-------- testdir
      ------------- testdir/subdir

    192. Re:Okay now... by dimer0 · · Score: 1

      I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.

      I didn't do that, but I do remember one night at about 3am, remotely ssh'd into a box across town, I su'd up to root, was working on some backups, one was running way too long, I did a 'jobs', saw my first job in the background was never gunna finish, so I quickly did a:

      kill 1 (instead of kill %1)

      [enter].. wait.. [enter].. wait.. (sinking feeling)

      Damn, time to jump in the car and drive 20 miles to the data center.

      There's a lot of things besides rm that are dangerous when you're root.

    193. Re:Okay now... by Dwonis · · Score: 1

      heh. Just make sure you don't hit Enter too soon...

    194. Re:Okay now... by Anonymous Coward · · Score: 0

      I did a chflags -R schg * once on OpenBSD in the root directory. Now, someone (my freind) was owning me as I typed it, and the results were . . . interesting. (note: its been a long time since I used OpenBSD, and that might not be the right syntax, you might want to try it out . . . )

    195. Re:Okay now... by Anonymous Coward · · Score: 0

      yeah.. whats sudo?

    196. Re:Okay now... by Anonymous Coward · · Score: 0

      Yeah, I guess it looks that way.

    197. Re:Okay now... by squiggleslash · · Score: 1
      non-readable. Not non-executable.

      Non readable allows subdirectory access (as long as executable is set) but means no process can get a directory listing (and therefore recursive commands like the above do not work.) Doing it your way doesn't really give you any security.

      --
      You are not alone. This is not normal. None of this is normal.
    198. Re:Okay now... by Anonymous Coward · · Score: 0

      Yep, or in my case relying on shell filename completion to fill most but not all characters:
      rm partialprefix*
      expecting:
      rm prefix_*
      but got:
      rm prefix_*
      because I have a file with the exact "prefix_" name. Changed my filename completion settings ever since.

    199. Re:Okay now... by TClevenger · · Score: 2, Informative

      Actually, there is a root account. It's just disabled by default.

    200. Re:Okay now... by Anonymous Coward · · Score: 0

      It's not up to Microsoft to make those changes. It's up to the lazy ass people who developed the software incorrectly in the first place.

    201. Re:Okay now... by Minna+Kirai · · Score: 2, Interesting

      What I don't understand is why the *nixes don't implement something like the Mac's trash can.

      First, notice that if you run "rm" on Mac OS X, even it won't use the trash can.

      The behavior of Linux and Mac is actually quite similar in this instance. On either platform, removing a file with the GUI tool brings it to a trash holder, but the command line deletes immediately.

      Create an invisible directory under each and every mount that is called .trash, and when *any* user does *any* rm command, instead of deleting the files outright, simply move them into the .trash directory.

      Simple, practical obstacles: ~/.trash won't work for files which are on other disks, network shares, removable media, etc. It would have to move the file to the same hard drive as your ~ directory first, which will at best take time, and at worst will overfill your own disk.

      More fundamental, and historical explanation: Unix was designed as a operating system, a framework for applications. To keep the job managable, they added in things that were necessary for the OS (like files, copying, and deleting), but not things that could be better handled at the application level. ~/trash is GUI sugar: just a minor way to make it more difficult for users to input commands that they likely didn't intend.

      So, then the question becomes, why did application-level implementations of a two-stage file deletion become popular? And here, the answer is the old canard "Good is the enemy of great". Because the native "rm" command was adequate for more than 98% of all usages, there was little demand to shift to something more complex, even if it would be occasionally safer.

      When finally you are shopping around for disk space, only then do you consider emptying the trash.

      Unix is a server-oriented OS, both historically and still today. Servers are expected to go weeks and months without a user sitting at them. Needing a person on-hand to Empty Trash just because the webserver has been creating and deleting a bunch of cache files is a bad thing.

    202. Re:Okay now... by Anonymous Coward · · Score: 0

      He meant "read" rather than "execute." But in your infinite wisdom you must already have know that and just decided not to share it.

    203. Re:Okay now... by Anonymous Coward · · Score: 0

      You have the linux source code, you can turn off the below 1024 restriction.

      Enjoy.

    204. Re:Okay now... by pg110404 · · Score: 1

      running an equivalent command would absolutely not hose the system

      I stand corrected. While it's true I'm more comfortable in unix than I am in windows, the fact of the matter is virtually all system files in unix are owned by root and you have to be root to the root user group to delete most of those files. Belonging to the root user group is not enough, it's root or not at all for most system files.

      Unix/linux defaults to non-administrator type user and thus can't easily cause system wide damage. You (read system administrator) have to want that added user to be 'root like' which is something you have to go out of your way to accomplish (add 'root' as additional groups that user belongs to). Also you can't be same-as-root, you can only be root-like and even then many system files are only modifiable changeable by root.

      Windows on the other hand defaults to administrator and any one of those additional administrator accounts are just as capable of destroying those critical system files. You have to go out of your way to be a neutered non administrator user and to NOT be an administrator capable of destroying the system simply because you forgot to change the radio button from 'administrator' to 'non-privileged' user

      My argument of 'non-privileged' user was incorrect and I should have said 'any account with default access privileges other than the standard administrator account which can obviously do that', but for the most part, my argument that any account other than the one and only 'administrator' account still being cabable of hosing the entire system still holds. There is a difference between non-administrator and non-administrator, and there are many people out there who would not understand the difference.

      As to what I do or don't know about windows, please don't assume I don't know anything.

    205. Re:Okay now... by Minna+Kirai · · Score: 1

      I personally feel, especially after reading the interview, that the approach of running the user as root for lindows is a good idea. It does offer the choice of setting up users but doesn't force you too.

      The Apple Macintosh system has a good reputation for stability, security, and user-friendliness. And, do they give the normal user root permissions by default? Absolutely not. Why do we think this is?

      Anyway, "Copy what Apple does" is rarely a bad starting point for configuring a desktop user interface.

      Here education and experience is needed in either case.

      Yes, and that's why non-root is a good default. When the user is poking around and seeing what things he can change, he won't be able to lose the libc.so file just because he dragged some mysterious icon a few pixels. The request for a root password to complete that operation is a great warning sign that he probably shouldn't be screwing around with whatever he just clicked on, and is a fine signal that more education might be needed before proceeding.

      Consider how police dogs are trained- they learn to attack and bite people only when wearing a tight collar, which makes them safer to be around the 99% of the time when there are no burglars to chase. The root login is like a spiked collar that reminds the dog-handler to be extra careful, because now major damage can be done.

    206. Re:Okay now... by Minna+Kirai · · Score: 1

      Games don't need administrator either. They just need physical write access to the CD device, a privilege you can grant explicitly.

      Nope. Some games need access to 100% of ram, so they can scan for unapproved device drivers or other things that may provide an unfair advantage to online competitors.

    207. Re:Okay now... by Miamicanes · · Score: 1

      For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root

      OK, maybe it wasn't QUITE that bad... but way back in '97 when I first started playing with Linux, I did actually delete some very important data files as root. At first, I didn't panic... undeleting files on a FAT32 drive was a pain, but no big deal. EXT2 couldn't be any worse, right? Needless to say, I wasn't amused when I found the following (paraphrased from memory) quote after desperately searching for undelete info on Yahoo:

      On a Unix system, "root" is God.

      God is perfect and incapable of making mistakes.

      THEREFORE, when root-god deletes a file, He obviously intended for it to stay deleted.

      THUS, there is no way to undelete files deleted as root.

      It's funny to laugh about now, but I was pretty traumatized over it at the time...

    208. Re:Okay now... by rookworm · · Score: 1
      I think he says it best himself:

      "I think we've largely sucked" ;)

      --
      The toad can't burp - and for some reason can't fart either, so it swells up and eventually explodes. --Anonymous Coward
    209. Re:Okay now... by fvwmfan · · Score: 1

      Yes, it does.

      That isn't a problem for /home, because the directory you want to browse is /home/you .

      There is no need for anyone to be browsing /home, and this fix makes the whole system more secure.

      I hadn't thought of this as a solution to the problem of running "rm -rf /" by mistake, but it is a really good solution. Dispite being pretty careful with such things, I did recently run "rm -rf ." whilst in /, and I DID breathe a sigh of relief when I realised I was a normal user. I hit Ctrl-D pretty darn fast, believe me. The only thing that saved me was the fact that I was running as a normal user.

    210. Re:Okay now... by anthony_dipierro · · Score: 1

      I checked. My version of rm is from coreutils-5.0, and the applicable portion of code is in src/remove.c:

      if ((errno != EISDIR && errno != EPERM) || ! x->recursive)
      { /* [comment omitted] */
      error (0, errno, _("cannot remove %s"),
      quote (full_filename (filename)));
      return RM_ERROR;
      }

      Linux is returning EACCES. I'm not sure if that's the proper behavior or not (man unlink doesn't talk about the immutable flag).

    211. Re:Okay now... by leshert · · Score: 1

      Developers who write software that absolutely requires Administrative rights for common use... are incompetent and should be killed.

      Indeed, they should.

      Although the KB article only lists up to VC++ 4.2, this bit me with VC++ 6.0 on Win2k the one and only time I tried decreasing to non-Admin privileges.

      I haven't seen it myself, but I've heard that the same is true with EVC 4.2, which is still the only (released) game in town for developing on CE.

    212. Re:Okay now... by Anonymous Coward · · Score: 0

      Ok, then don't do the same thing. Just drag your windows directory anywhere else on your hard drive. Go on. It's easy.

      Really. Have you NEVER mistakenly moved a file (I mean not even MEANING to move a file in the first place) due to your mouse glitching, or a finger spasm (or brain spasm.. whatever)? I thought so. This is why it's bad to run a gui with root permissions, whether it's in Windows or Linux. Shit happens.

    213. Re:Okay now... by Anonymous Coward · · Score: 0

      Did that... mounted my root partition under some other directory, thinking that I've mounted root partion of an other Linux installation. Then I intentionally wrote 'rm -Rf *' to remove all files from the other linux installation. Did I felt like a jackass afterwards!?! :)

      Under an user-account I've only managed to delete all my mail, again intentionally writing 'rm -Rf Mail'. Within a nanosecond I've realized I should be removing an empty directory named 'mail', not the one with capital letter M... My attempts, as a user, to do something like 'rm -Rf .*' were allways stopped... :)

      And I hold a masters degree in distributed computer systems, meaning I have some experience and knowledge about computer systems.

      shit happens... I know...

      Sorry about my English... It's 6 am and I'm on my first coffee...

    214. Re:Okay now... by burns210 · · Score: 2, Interesting
      "Can one undo a commandline "rm" in OSX?
      One would not do such a thing in Mac OS X."

      Granted, I use finder to delete files 95% of the time, but on occasion I use the rm command to delete.. Not only can I not undo this, rm does not act the same way finder's delte does.. rm does not put files into the trash.

      This seems like a design flaw. The Mac is a great platform(my Tiger dvd is in the mail, I am hooked) and the Tiger features that make mv and cp more mac-native are great. Having said that, the GUI operations that have a CLI counterpart (delete in finder vs. the rm command) should operate the the same way and be interchangeable wherever possible.

    215. Re:Okay now... by ejtttje · · Score: 1

      I had a shell script that, on a certain command, was supposed to erase the contents of a directory specified by the user. E.g.:
      rm -rf ${dir}/* Looks reasonable, right? Except one very important user happens to type a space at the end of the directory name, guess what that expands to?
      rm -rf /path/to/dir /*
      Not only deletes the directory itself (least of concerns at this point), but /* as well!
      Sigh. Moral of the story? Be very very careful with user input:
      rm -rf "${dir}/"*
      Also, we now prepend the command with a "are you sure" dialog, listing the command about to be executed, just in case.

      It could happen to you!

    216. Re:Okay now... by Anonymous Coward · · Score: 0

      undelete.exe. DOS is greater than UNIX.

    217. Re:Okay now... by nokilli · · Score: 1

      Of course, I was speaking of the classic Mac's trash can.

      And I did say that the .trash directories would be created on each mount, or volume. So the time spent moving the files is really no greater than that deleting them, since they would be moved onto the same volume.

      A cron job could do the job of emptying the trash very well.

      I still fail to see why this is a bad idea.

    218. Re:Okay now... by Baricom · · Score: 1

      I have read elsewhere (can't remember where, though it may have been UNIX for Dummies) that the main reason is that even if it is installed on some platforms by default, it wouldn't be on others, and people may become complacent typing "rm", knowing they can always get their files back - until they do it on a machine that doesn't have trash set up.

    219. Re:Okay now... by istartedi · · Score: 2, Informative

      This isn't a problem with CLIs. The GUI analogy is the Windows pop-up that asks you if you're sure you want to delete a file. Raise your hand if you use Windows and you've gotten into the habit of smaking your enter-key, sometimes before that dialog even displays.

      The problem is that people want to do things quickly, so you've got people training themselves to use -f because they're in the habit of recursivily deleteing files on a regular basis and they don't feel like interupting the flow responding to a prompt. This works really well until they don't mean to do it. The Windows recycling bin is not a bad solution to this problem; there is no widely adopted *NIX equivalent.

      --
      For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
    220. Re:Okay now... by cp.tar · · Score: 1
      Don't forget, as a smart businessman, he knows how to sell his product.. Logging in is REALLY hard to sell.. Even for XP users (notice the pretty typing-free login icons in XP).. If XP required people to memorize passwords to do anything, then people would be use to it, and wouldn't bitch about it in Linux.

      I just recently bought my father his first computer.

      Dual-boot, WinXP and S.u.S.E Linux (it came recommended by a "Linux user", as he calls himself).

      My father was completely computer illiterate, and his girlfriend even more so. However, they both have passwords on their Linux accounts - I'd set them up so as to have the same passwords their mobile phones have. It might not be great security, although they don't as yet even have Internet, so the whole question is quite moot. However, it is a good way to introduce passwords - my father said he understood that his password should be more complex, but that he's too old to remember that along with a million other things he's learning (he's even having problems with click, double click, click-and-hold and click-and-drag): therefore, when he finally does learn all the basic stuff, a new password will not be as complex a thing - especially when I finally get that .rpm that gtypist depends upon, so that he can learn how to type.

      I didn't bother setting up their WinXP accounts to have passwords simply because they only ever use Windows to watch DVDs (and play Solitaire, sometimes) - I have yet to install libdvdcss (and gtypist, but that's another issue) on S.u.S.E. Besides, when they finally do get an Internet connection, they will only surf from their Linux accounts, especially because I'm quite certain they will both go looking for pr0n.

      I find it quite amazing that a complete computer illiterate actually finds Linux/KDE combination more comfortable than WinXP. But that's maybe because he doesn't have any prejudices... except from what I'd told him. ;)

      Thus to have people adopt his product, he needs to soften the hard-core UNIX advocates's argument.

      I told my father that surfing the Internet on Windows or from a root account is equal to having sex without protection... and on the Internet, you meet a whole lot of people. It's amazing what you can do with a good analogy.

      --
      Ignore this signature. By order.
    221. Re:Okay now... by As+Seen+On+TV · · Score: 1

      Um. No. I can honestly say that I've never mistakenly moved a file that I didn't intend to move. That sounds incredibly unlikely.

      But even if I did, the solution is a quick command-Z away. Unlike that silly command-line example that everybody's still straw-manning.

    222. Re:Okay now... by Lorkki · · Score: 4, Interesting
      See NeXTSTEP and MacOS X. Users were not root. Users seem to be getting along just fine. Login optional.

      Ubuntu does this too. The default installation has the root account disabled for login purposes. What few administration tasks require root access is done through sudo using the user's password for authentication. Login could just as well be automatic.

      I fail to see entirely what Linspire needs continuous root-level access for.

    223. Re:Okay now... by As+Seen+On+TV · · Score: 1

      This seems like a design flaw.

      It is a HUGE design flaw. We've been over this and over this internally. But like so many things, it's a compromise. On the one hand, everybody recognizes that the UNIX command line is a massively stupid thing to keep around. On the other, lots of our users want it, and WE want POSIX compliance. So we strike a compromise. We keep it, but we ensure that nobody will ever have to use it for anything, ever.

      Here's the important thing, though. We go back and forth on this with every release. We're aware of the problem, and we're constantly looking for solutions to it. Compare and contrast with the mouth-breathing idiots who continue to insist that it's just FINE that the command-line interface is the equivalent of hammering in a nail with the butt of a loaded pistol.

    224. Re:Okay now... by pegacat · · Score: 0

      I see your analogy, and I raise it:

      * automatic transmission
      * power assisted steering
      * synchromesh clutches
      * automatic choke
      * anti-lock brakes
      * traction control
      etc. etc. etc.

      Cars *are* complicated, and a lot of effort has gone into hiding that complexity in a modern car, so that all the user has to do these days is set speed and direction. Try driving an old, big car (or truck!) where you have to double de-clutch, crank the engine to start, manually set the mixture with the choke and be built like Arnie to steer the thing. Sure users still crash cars, but it's usually 'pilot error' now rather than 'mechanical failure', or 'workload induced failure'.

      So (with the obvious exception of Mac OS X :-) ), I claim most computers are still at the usability stage of automobiles in the 30s (they've had similar years of development) - sure you don't have to prime each cylinder with fuel before you start, but they're still very complex to run.

      But the fact they're complex doesn't mean that you can't hide that from a casual user attempting simple tasks.

      --
      Wer mit Ungeheuern kämpft, mag zusehn, dass er nicht dabei zum Ungeheuer wird.
    225. Re:Okay now... by Anonymous Coward · · Score: 0

      Um... wouldn't that mean that the economy was GOOD?

    226. Re:Okay now... by s-orbital · · Score: 1

      Actually, when I was a kid, I got in the elevator in the Pantages Theatre in Tacoma, and pressed the up button, and... the doors closed, and the elevator did nothing. A crowd gathered around the outside during the half hour I was trapped in there, and people tried in vain to open the doors. People were blaming it on "some kid in there who was messing with the elevator"... well, it turns out that a computer controlling the elevator crashed. Fucking computers....`

      --
      Patent: from Latin patere, to be open
    227. Re:Okay now... by ebyrob · · Score: 1

      Was the point that searching isn't always the best method of finding data completely lost on you? Do you leave your room a mess thinking "oh, I have a great search tool"?

      Is 30+ years of computer science a design flaw? Is the fact cars don't stop instantly when you hit the brakes (inertia) a design flaw? What about physics and information theory in general? Heck engineers used to talk about "essence" and "accident" in terms of design costs. Am I expected to believe everything is an accident, essence doesn't exist and creating such a seperation is a waste of time? (Yes, they blur and we're never sure, but that hardly makes the distinctinction unimportant)

      My entire point was that different != better when it comes to user interfaces. Replacing file-trees with search and removing the ability to keep file-trees is worse than doing nothing in most environments.

      For the record: The first time I sat down at an OS-X workstation I couldn't even find the web-browser let alone documents. That was having already seen screen-shots with the big-ugly toolbar at the bottom. (Silly me thinking there was an easy alternative to the darned thing...) In theory I think OS-X is really cool, but in practice I don't have much use for anything except the command line.

    228. Re:Okay now... by adamruck · · Score: 3, Informative

      Some more information for you.. this is an blurb from the iptables man page

      ----------------

      owner
      This module attempts to match various characteristics of the packet creator, for locally-generated packets. It is only valid in the OUTPUT chain, and even this some packets (such as ICMP ping responses) may have no owner, and hence never match.

      --uid-owner userid
      Matches if the packet was created by a process with the given effective user id.
      --gid-owner groupid
      Matches if the packet was created by a process with the given effective group id.
      --pid-owner processid
      Matches if the packet was created by a process with the given process id.
      --sid-owner sessionid
      Matches if the packet was created by a process in the given session group.

      ------

      You can filter network traffic based off of the same system that you can use to filter access to files. Even more fun is the ability to filter network traffic based off of a process id.

      --
      Selling software wont make you money, selling a service will.
    229. Re:Okay now... by R@Bastard · · Score: 1

      Or, you could use authbind to do this without iptables:

      http://packages.debian.org/stable/utils/authbind

      --
      Mucous membranes are the part of your brain that, like, make you think about mucous. --Beavis
    230. Re:Okay now... by lahvak · · Score: 1

      Once I have accidentally done rm -Rf /etc.

      I did some extensive changes in /etc (some experimenting), and I was affraid I would mess up, so I made a copy of all the files I wanted to change to ~/etc, did the hanges there, went through them again, checked everything, su'ed to root, copied all of them to /etc. Then I wanted to clean up. It was about 3am, I was sleepy, and I accidentaly did rm -rf /etc instead of rm -rf ~/etc. It was fun (not). I didn't get much sleep that night.

      Once on my work machine, windows NT, I needed to find information about some driver, so I unziped this file I downloaded from HP. It unzipped to C:/, and there was about 200 files with all sorts of weird names. I wanted to clean it up, but was too lazy to figure out which files were they. I tried to edit AUTOEXEC.BAT, and I couldn't, so I figured I wasn't Administrator, and I would not be able to erase anything important. So I opened c:/ in explorer, selected everything and erased it. I could hear disk spin for a sec, and then the screen went blank. After that, I wasn't even able to boot the system. I turned the box over to our NT guys, and they were not able to do anything with it, they had to re-install the system. Aparently, even though I wasn't Administrator, they gave me some sort of elevated priviledges so I could install software on the box, which allowed me to erase some sort of vital file.

      I think there is a huge difference between runnning as a root and running as a regular user. Root can very easily destroy the entire system, to the point where there is no way to access any data from the disk whatsoever. Regular user can delete all user data. The system will still be running, and in many cases at least some of the files can still be recovered. As a matter of fact, in the above example, explorer just moved the files to the trashcan. It was enough, though, to make the disk totally unaccessible.

      --
      AccountKiller
    231. Re:Okay now... by ABaumann · · Score: 1

      Who needs the command line? I actually deleted about 6000 MP3's about a month ago for doing a "Select All" -> "Delete" when I didn't know that I was in my iTunes music directory. ...Yeah, I know... that's what I get for downloading all of that music without paying for it.

    232. Re:Okay now... by Anonymous Coward · · Score: 0

      "Without Administrator rights, Citrix Client will open, try to initiate a session, fail and then close without error."

      I fixed that... Install Citrix client and then install MS Terminal Server Client...

    233. Re:Okay now... by yupa · · Score: 1

      Well people often use shift + delete, so they don't use the recycling bin...

    234. Re:Okay now... by trawg · · Score: 1
      If you ask me, it is the command-line's greatest strength. You tell it to do something and it does it. If you wanted to be safe and have it confirm your request before it does each and every action you shouldn't specify the 'force' option.
      Heh, I totally agree. Its like putting a gun to your head and pulling the trigger and being annoyed when your brains are plastered all over the walls. Pull the trigger with -i, people!
    235. Re:Okay now... by sebster · · Score: 1

      One time I when I was really really tired late at night I wanted to rm something in /var/log. So I start typing:

      rm -rf /

      with var/log/... to be next.

      Then I figured I didn't need the -rf, that was just a sleepy mistake, so I decided to get rid of it (don't know why), and pressed backspace...

      Or so I thought....

      It took me about 30 seconds to realize what happened (I was sleepy remember), by that time my machine was thoroughly screwed. Fortunately I still had my user data when I pressed ^C.

      Before this, I'd have never believed it to be possible to accidently type rm -rf / as root!

      Greetz,
      Seb*

    236. Re:Okay now... by yupa · · Score: 1

      Well it's not quite trivial to implment an recycling bin.

      First problem : which layer use :
      - libc : in user space, portable to near all *NIX, only intercept the call using libc.
      - fs : OS specific, fs specific
      - kernel syscall interception : os specific.

      There are solution which were implemented for that (http://pwp.netcabo.pt/0154115101/software/libtras h/,
      http://e2undel.sourceforge.net/,
      http://www. shirka.org/recycled4linux/), but I think it don't fit well with *NIX.
      Admin prefer using backup.

    237. Re:Okay now... by Anonymous Coward · · Score: 0

      I was working for my High School. I ran their RH server and we had just got the faculty used to using the mail from it. Im putty'd in to the server. I need to clear out a users directory. I typed 'rm -rf *'. It chugs, and throws up an error in /dev. FUCK! Ctrl-C Ctrl-C Ctrl-C Ctrl-C Ctrl-C. I was in /. Faculty tech guy says 'Im going to go have a smoke, when i get back tell me what happened.', i was rolling on the floor. 5 mins later I built up the balls to check out the damage. System was gone, luckily i stopped it before it hit the home dirs. The server was down for three days because of a "hard drive crash".

    238. Re:Okay now... by nokilli · · Score: 1

      That makes sense, but wouldn't that also argue against having *any* safeguards, such as the confirm-before-deleting prompt normal users receive?

    239. Re:Okay now... by HomeworkJunkie · · Score: 1

      At our company part of our interview process involves a UNIX test to ensure they are as competent as they say they are (we've had problems with agencies sending us people who don't have a clue).

      One question was something like: what could go wrong with the following 'rm -fr $(FRED)/'?

      Also, one of our more "experienced" engineers decided to show a newbie this example at a console...whoops...he pressed enter :) In the time he managed to frantically press CTRL-C a number of root folders had gone, most notably /dev . He's never lived that one down :)

      A few years earlier a non-techy (support) was using one of the UNIX machines for some testing and was worried about diskspace. He didn't like or understand the contents of /dev/ so he deleted it!! I think his thought process was something like: "What are these for? I don't know and they make the file structure look messy, I'll delete them"

      --
      "Why take life seriously, you're not coming out of it alive anyway."
    240. Re:Okay now... by Nevyn · · Score: 1

      So now you have a somewhat opaque dep. on iptables ... and it allows all local users to also bind to the web server port. Not exactly an improvement

      SELinux doesn't even really help, you just have to start the httpd as root and "know" it'll just bind to the port/chroot/whatever and then drop privs.

      I know my web server does that, and I apache-httpd seems to (although I'm not sure how easy it is to chroot it).

      --
      ustr: Managed string API with ave. 44% overhead over strdup(), for 0-20B
    241. Re:Okay now... by Rimbo · · Score: 1

      Can you imagine anyone who rides on an elevator having that kind of access?

      Including 13-year-old punkass kids?*

      *Note: I do not refer to the vast majority of non punkass 13-year-olds; rather to that particular beast for whom pushing every button on the elevator, hitting the "emergency stop" and trying to climb out when it's halfway down, etc. are fun things to do

    242. Re:Okay now... by As+Seen+On+TV · · Score: 2, Insightful

      Was the point that searching isn't always the best method of finding data completely lost on you?

      Basically, yeah. Because you see, searching is the best way to find things. It's the best way we have. We don't know of a better way. If you think of one, great, I hope you become fabulously rich. But right now, searching is the acme of human accomplishment when it comes to finding things.

      See, the computer is there to keep track of relationships for you. That's its job. It's not just a storage device. It's a device that's capable of storing and retrieving things. If you want information related to scuba diving, you shouldn't have to go looking with lots of clicky-clicky-clicky. You should be able to just ask for it. That's why we have computers instead of, say, shoeboxes full of paper.

      Is 30+ years of computer science a design flaw?

      It is when that 30+ years of work leads to something that's since been supplanted. It took us thousands of years to come up with Aristotle's theory of nested crystal spheres, but it was still wrong.

      My entire point was that different != better when it comes to user interfaces.

      And my point is that better is better. You're talking in meaningless and nonsensical abstractions. I'm talking about an actual computer program.

      Replacing file-trees with search and removing the ability to keep file-trees

      Who said anything about removing anything? Of course, if somebody did do that, it would not be worse. Have you looked at a database lately? Databases preserve relationships between entities in a way that's totally opaque to the user. You get to the data by going through a semantic interface. This is superior for managing large volumes of data. In fact, it's superior for managing small volumes of data, too. It's just that the activation energy used to be so high that it wasn't worth creating a relational database to store, say, your e-mail, chat transcripts, contact information and calendar entries. Spotlight reduces this activation energy to zero, meaning you can create that relational database with no investment at all. Just plug in the data as you normally would, and Spotlight handles building the relationship models for you.

      Example: Just a few hours ago, I got an iChat from a person whose name I recognized, but I drew a complete blank on who he was. I spotlighted his name and instantly found an e-mail that he'd sent me two years ago. As soon as I saw it I knew exactly who he was; I just needed to be reminded. At the same time, I saw that he was on the attendees list for a interdepartmental meeting that's scheduled for April 28. It was in my calendar, you see.

      See what I mean? The computer goes from being a big shoebox full of paper to a machine that actually knows things and that can answer questions. This is good. This is important.

      Again: You're talking in meaningless abstractions. I'm telling you about an actual, working, shipping product. Not maybe-someday, but today, tonight, right now. (Well, right now for us. Ten days from now for everybody else.)

      first time I sat down at an OS-X workstation I couldn't even find the web-browser let alone documents.

      Hm. I'm having a hard time coming up with an explanation for this that doesn't involve massive, almost comical stupidity on your part. Maybe I should refrain from jumping to conclusions ... but you're not making it easy on me.

      I found in particular your question from two comments ago to be freaking hilarious in light of what you said about not having much use for Mac OS X. You asked, "Why can't I just grab the contents of my 'programs' directory and move it to a new machine?" If you'd spent ten minutes using a Mac instead of complaining that the Dock is, to use your word, "ugly," you'd know that that's precisely how things work on the Mac. Applications are self-contained little packages that can be run from anywhere and simply dragged from one computer to another.

      That cracked me up.

    243. Re:Okay now... by Anonymous Coward · · Score: 0

      Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

      As far as data loss goes: I log in as a user and I have a job which periodically backs up my user data to another directory but all with read-only privs. I can go back in time to the last hour, or any day in the last 7, or any month in the last 12. I also rsync my important data across multiple machines spanning multiple architectures and multiple operating systems, along with removable media like USB drives and CDRW and DVDRW.

      As far as data privacy goes: Encrypt it.

      How hard is this Mr. Robertson?

      BTW, one mistake as root and you might be fucked. Staying logged in as root allows one to become complacent after a while. By using su, every time you invoke it you are reminded of the care that is required. Simply by the fact that you either know you need to use su because you are doing something potentially dangerous or are reminded of it when something does not work with your user privs.

    244. Re:Okay now... by kasperd · · Score: 1

      The beautiful thing is that there were NFS mounted disks on there... He lost a lot, much of it not even his.

      There are a few thingss in this story that sounds strange. In a normal configuration doing an rm -rf /home as root on an NFS client will not do much harm, because the squash root feature will mean the server really see you as nobody. Are you saying the server had squash root disabled, or he did this on the server? In either case I'd say that demonstrates somebody has been pretty incompetent. The other thing I wonder about is if there really still are systems where root have / as his homedir? On the systems I use root will use /root as homedir, in which case the command you gave would be almost harmless.

      --

      Do you care about the security of your wireless mouse?
    245. Re:Okay now... by Anonymous Coward · · Score: 0

      Even if this were to work on a given configuration --for instance it certainly wouldn't do shit if I were using a LiveCD image loaded into RAM which I use almost exclusively for all my computing needs-- how would it be any scarrier than a DOS user typing format c:? In the days of DOS, how many times did you ever actually do such a thing on accident? I may have done it a few times, but I almost always had backups.

    246. Re:Okay now... by jnf · · Score: 1

      there are these things called 'linux capabilities', its broken by default because the developers don't think its secure, but it will allow you to do things like that, and things like ntpd as a non-root user, etc.

      Then you have posix fs acl's which add a lot of granularity. Then you have third party things like grsec which allow you to do all of the above (and specifically adds /proc restrictions)

      Simply put, just because you don't know how to do it doesn't mean it can't be done. Also, if you code your program correctly you don't even need to get funky. Open your socket but don't accept, drop priv's and chug along. Or, open a named pipe and have the priv'd process open the socket and hand the fd pair off to the child (priv seperation).

      My point is that there is nothing wrong with a program initially needing root provided it drops its priv's in a correct manner.

      And just to be an ass, but a serious ass, dig into your kernel and modify the system calls to do a

      if ( current->uid == MAGIC ) { [ ... ] }

      That's the beauty of open source, you can change the things you don't like about it.

    247. Re:Okay now... by Anonymous Coward · · Score: 0

      do you have any more info on this?

    248. Re:Okay now... by kasperd · · Score: 1

      That are all good arguments why running as root is a bad choice. But I'd like to add, that I actually backup /home from my cron.daily. So even if somebody could take over my user account and damage my data, they still wouldn't have permission to damage the backups. My next step will be to upload the backups to a friends computer, where a sgid program will ensure I cannot damage a backup once it has been uploaded. When I get that working a backup will be safe even from root.

      --

      Do you care about the security of your wireless mouse?
    249. Re:Okay now... by ZephyrXero · · Score: 1

      I haven't had any trouble with my users using Citrix on their restricted accounts. I didn't have to jump thru any hoops or anything, it just worked.

      --
      "A truly wise man realizes he knows nothing."
    250. Re:Okay now... by Anonymous Coward · · Score: 0


      I generally see what he's saying about data being king. But if your data is that important, you'll have other safeguards for protecting it, typically via (dun dun dun), user management! For instance, keep your accounting files under a different user, home directory chmodded to 700. Stuff like that.

      You're right, of course. But let's not forget that the product this man sells is aimed at individual home users who don't need multiple user accounts (after all, it's only them using the system) and they're never going to want to run the likes of Apache or MySQL.

    251. Re:Okay now... by achim · · Score: 1

      Windows on the other hand defaults to administrator and any one of those additional administrator accounts are just as capable of destroying those critical system files. You have to go out of your way to be a neutered non administrator user and to NOT be an administrator capable of destroying the system simply because you forgot to change the radio button from 'administrator' to 'non-privileged' user

      Even worse, one usually cannot run quite a number of programs if he is not Administrator. So even a guy like me that knows how it should be like is forced to add Administrator privileges to his account to run ordinary apps like CorelDraw.

      It was reported recently that MS wants application vendors to write their applications correctly, so even non-Administrator users can run them, but the intended additional certification

      • will be for Longhorn, and thus in the future,
      • seems to be unpopular with application vendors
      • and is only an additional certification that nobody is forced to support.

      So, all in all, local security on Windows will reamin a problem in the foreseeable future though Windows in theory brings all the requirements for more secure use.

    252. Re:Okay now... by DrXym · · Score: 1
      Better yet...

      Linspire users are deemed to be dumb (by their CEOs implicit admission) to handle the concept of an admin password. In other words, they're easy marks for phishers.

      Therefore a simple malicious attack might be easy to pull off - just scrape for email addresses and Linspire in the same page and mass email them all a trojan. Using the appropriate social engineering techniques, direct them all to a hacked ClickNRun site or promise them an extra year's free sub if they download and run the attached survey programme.

      You can see where I'm going with this. Let's see what happens to the "no one explaine why root is insecure" argument after the hilarity ensues.

    253. Re:Okay now... by Anonymous Coward · · Score: 0

      not if you mount /home partition with noexec !

    254. Re:Okay now... by Anonymous Coward · · Score: 0

      i threw a party and a non geek friend asked a room full of drunk geeks what a cool command to throw into my linux box would be - almost in unison we rattled off rm -rf * - my friend also being drunk assumed we wouldnt give him the command, whilst being logged in as root, to say bye bye to the box.... anyway a long story shory, he hit enter a roomful of geeks dived at a keyboard but no - my linuxbox/server was running in memory

      dont run as root unless you want bad things to happen.....

    255. Re:Okay now... by Narchie+Troll · · Score: 2, Informative

      With GNU rm, you only get a prompt if
      a. you attempt to delete a write-protected file
      b. you use the -i switch, which some distros automatically stick into the global bashrc

    256. Re:Okay now... by Leto2 · · Score: 1
      2) Developers who write software that absolutely requires Administrative rights for common use, and the program is not designed to alter fundamental hardware or OS configuration (such as a registry editor or a graphics driver tweak utility) are incompetent and should be killed.

      You mean like Firefox...

      --
      <grub> Reading /. at -1 is like driving through Cracktown in a convertible that is stuck in 1st
    257. Re:Okay now... by Mornelithe · · Score: 1

      How much time to you spend installing applications versus using them?

      How much time do you actually waste by typing in a password once in a while?

      I can type my password in 1 second. Do you install enough software that password typing actually uses up a significant portion of your week?

      --

      I've come for the woman, and your head.

    258. Re:Okay now... by Anonymous Coward · · Score: 0

      I would rather run rm -rf "/usr" or "/var" than my home directory...

      I can replace /usr. the files in my home directory are irreplacable, so I don't see how this is an argument. Bad use of rm -rf is bad whatever user you are. At least with /usr, it takes so long, you get enough time to hit ctrl-c :-D

    259. Re:Okay now... by Anonymous Coward · · Score: 0

      You are arguing on his side.

      This was about why it's a bad idea to run as root/Administrator always. The reasons are the same no matter what system you are on - Linux, Windows or OS/X.

      Windows can be pretty secure if you don't run as administrator, unfortunately many apps (especially games and especially Microsoft products for some reason) won't run in this situation.

    260. Re:Okay now... by Ed+Avis · · Score: 1
      Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful.
      What is 'harmful'? If something running as your ordinary user account is exploited - for example, a bug in your web browser lets it be taken over by some hostile web page - then your user account is cracked. Why is it necessary to escalate that to root to do any damage? All the files I care about are in my home directory; all the sensitive operations (like typing credit card numbers into a web page) happen when running as my user account. Root access is more or less irrelevant, unless there are other user accounts on the machine. But most PCs these days have only one real user.
      rm -Rf / as nonroot will make you give a sigh of relief.
      I don't know how often the typical Linspire user runs such a command, or its equivalent. But as nonroot 'rm -rf /' will still trash my home directory, which is all that matters. Honestly, I don't much care about the contents of /usr/ or /etc/ - those can easily be reinstalled.

      The example you give for MySQL makes sense, however Linspire users aren't running MySQL. Having a separate user for accounting files might be a good idea if you are very security-conscious, but do you honestly believe grandma will or should set up her machine like this?

      I think it would be a good idea for anyone responding to Robertson's arguments to give a practical *example* of where running as non-root makes a home user's system more secure. Remember this is a machine with only one user, it doesn't run MySQL or Apache or other server things, and the data that really matters is the user's personal files. I think Robertson is probably right: people tend to respond with a lot of bluster and 'of course non-root is more secure', seat belt analogies and moralizing, but find it hard to give a real scenario where Linspire's always-root policy worsens security.
      --
      -- Ed Avis ed@membled.com
    261. Re:Okay now... by esbjerg · · Score: 2, Insightful

      Quote:
      "In that case, no local exploit is needed--the attacker either uses sudo, or just sniffs the password the next time the user uses su..."

      Unless the user has . in his/her PATH the attacker will not be able to "replace" any programs like the sh. The attacker cannot log keyboards strokes or snif the tty without root privileges. Hence if your system is set up in a good manner the attacker cannot just get the root password.
      Unfortunately most users can be tricked into writing their own password and most systems like ubuntu and MacOS has sudo ALL as default.

    262. Re:Okay now... by jbolden · · Score: 1

      I did a variant of it

      rm -Rf /var/(space)log/httpd

    263. Re:Okay now... by m50d · · Score: 1

      There are exploits for that too. Recently there was a buffer overflow in xine that let you execute arbitrary code by giving it the right stream - something you can do over the internet. It's not in the wild but iirc there was a proof-of-concept exploit available. I'd be willing to bet there's at least one arbitrary code vulnerability we don't know about in firefox, gaim, evolution or some other internet-facing program. If you're root, an exploit like that gets you 0wned. If you're a user, not so much. They can delete your latest data, but not your nightly backups. They can't install a rootkit because they have no access to the kernel image or init scripts. They can't change anything in /bin to trojan it.

      In that case, no local exploit is needed--the attacker either uses sudo, or just sniffs the password the next time the user uses su (or whatever graphical equivalent pops up next time they try to upgrade some software).

      That's why you're always told to use /bin/su rather than just typing su. They can change your .bash_profile to have your PATH pointing to trojan versions, but they can't get /bin/bash to execute something other than /bin/su when that's what you type, and they can't change /bin/su. Normal users have no ability to sudo on my system and probably many others, they don't need to. A determined attacker can probably get the root password somehow, but it makes it far harder, which is all you can hope for with security - make it so it's not worth an attacker's while to crack your system.

      --
      I am trolling
    264. Re:Okay now... by petermgreen · · Score: 1

      what i don't understand is if you have made the normal user account basically god what have you gained over having the user run as root from a security pov (protection from accidents is another thing entirely but this is supposed to be a security disscussion) btw when you use su on the shell. If theyve comprimised your user account then how do you know you aren't being keylogged as you give your password to su? unless you keep your use of root TOTALLY seperated from your normal account and you control that seperation in ways your normal account can't mess with then someone with access to your normal account is going to be able to get root fairly easilly.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    265. Re:Okay now... by m50d · · Score: 1

      Some of it's part of braindead windows design though. Win2k required you to have admin rights to burn cds! So of course users had to run as admin (I know there are now workarounds)

      --
      I am trolling
    266. Re:Okay now... by Anonymous Coward · · Score: 0

      So even a guy like me that knows how it should be like is forced to add Administrator privileges to his account to run ordinary apps like CorelDraw.

      Why don't you just take steps to fix it instead of accepting the slashdot mantra?

      Start here. Download regmon and filemon. Find out what CorelDraw is accessing that is being denied, and give your user (or better: a user group) appropriate permissions to make it work. I have had to do this for a number of apps, and I've not yet found one that REALLY DOES need admin privs - it's just that by default only admin has the rights needed to complete the task (usually it's just write-access to the application's own directory).

    267. Re:Okay now... by yruf · · Score: 1

      i can imagine, anagama doesn't really want this modded 'Funny', you bastards...

    268. Re:Okay now... by Anonymous Coward · · Score: 0
      That sounds like a workaround to make up for a design flaw in the command-line interface to me.

      What!?
      The f in -Rf is the force option, you give it that option because you don't want to confirm every action.
      If you want confirmation for each file/directory before removing it you simply omit the f.
      There's no design flaw, it's perfect!
    269. Re:Okay now... by Anonymous Coward · · Score: 0

      I didn't want to upgrade my main system and risk it going all screwy, so I just took out the HD, put in a blank one, and installed the new system. Then I put my old HD in usb enclosure, ...

      So, you were in a middle of a task, which required root privileges to succeed. How would have using non-root privileges prevented you to do the mistake, without preventing you to do the task at all?

    270. Re:Okay now... by Anonymous Coward · · Score: 0

      I insist that it's just fine that commands exist to do exactly what I say, and thank you very much for the abuse. I will try not to let your unprofessional attitude affect my decision to purchase a couple of Macs.

      As for a solution, could you not provide two alternate command line environments, one for picky people like me who don't like to be second guessed, and one for people who would like rm to move files to the trash?

      I am sure the picky people would have no problem adding an "export PICKY_BITCH_USER=1" to their shell init script, so that the user-friendly version could be the default (I assume you would actually find a better way to achieve the same goal, since Apple are so good at that sort of thing).

    271. Re:Okay now... by Anonymous Coward · · Score: 0

      "people" my ass. If you take a quick survey of 100 random windows users, I can assure you that less than 10% of them will even be aware of shift+delete.

    272. Re:Okay now... by baadger · · Score: 1

      If i'm not mistaken,

      Unlike Linux, Windows will not delete any dll or exe currently in memory, or for that matter any file that has an open handle. It will however allow you to rename (ren) them.

      Here's one comparison on how easy it is to hose Windows and Linux

    273. Re:Okay now... by baadger · · Score: 1

      How does that work?

      Windows 2000 doesn't come with native CD burning software, so isn't it down to the vendor CD burning software you chose to use?

      Admitedly Windows doesn't exactly offer brick walls between user accounts out of the box but with some configuration it isn't at all that bad.

    274. Re:Okay now... by meringuoid · · Score: 1
      How would have using non-root privileges prevented you to do the mistake, without preventing you to do the task at all?

      Not the point. We're all competent geeks who would never run as root unless we had good reason to do so and yet we've all still got horror stories to tell about the rm that went wrong.

      If we ran as root the whole time there'd be a whole lot more stories like that. People would wipe out their systems when they meant to back up their mp3 collection. Worse, if Joe Schmo ran as root the whole time... well, look at the Windows world for just some of the horrors we can expect.

      --
      Real Daleks don't climb stairs - they level the building.
    275. Re:Okay now... by KiloByte · · Score: 1

      weather display application

      Is that BonziWhatever, perhaps? You see, it has a lot of value-added stuff bundled with it, and it needs root access to install them and then re-install them after every boot if something bad happened to them...

      --
      The creatures outside looked from Alt-Right to Antifa; but already it was impossible to say which was which.
    276. Re:Okay now... by maccallr · · Score: 1

      I put the -rf in at the last minute and never use dot or star on its own. This is what I'd normally do:

      cd some/old/junkdir (ENTER)
      ls (ENTER)
      # see all the files I don't want
      cd .. (ENTER)
      rm junkdir/*
      # left-arrow or ctrl-b several times then add the -rf)
      rm -rf junkdir/* (ENTER)

      Even if you had some weird symbolic link which left you in a strange place after the cd .., it's unlikely that the junkdir/* will match unless you're in the right place.

    277. Re:Okay now... by Anonymous Coward · · Score: 0

      I did 'cp file /dev/hdxy', thereby overwriting the bootsector of another partition and losing everything on it.

    278. Re:Okay now... by Anonymous Coward · · Score: 0

      Note: If you happen to be a noob, please do not actually enter the command rm -Rf / as root. I'm serious. I know it's written out all pretty there, but don't do it. You'll be sorry.

    279. Re:Okay now... by petermgreen · · Score: 1

      so you wan't rm to move files to the trash?

      thats just fine until you run a script that expects rm to MEAN remove and your trash ends up full of useless temp files created by the script

      you could patch it so that rm did different things in scripts than when run interactively but thats going to cause confusion when people try something interactively before putting it in a script.

      maybe you could add a new command like rmmac or mtt (move to trash) but most of the guys who use the command line wouldn't realise it existed anyway.

      btw windows also only uses its trash eqivilent (recycle bin) for deletes from the windows gui not deletes by apps or from the

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    280. Re:Okay now... by bankman · · Score: 1
      It's not logging in that he is concerned about, it's click N run. Apt won't install under non root user.

      I would think that it would take some serious hacking to allow apt to install software into your own home directory but it would probably be a worthwhile effort.

      Fercryinoutloud, why? On a single user desktop machine this doesn't make any sense whatsoever, it would just break compatibility with other distros (ie. Debian) and LSB. On a multi-user machine it would become an instant maintenance and, quite possibly, security nightmare.

      Click'n'run installation is about the most idiotic thing ever devised for IT systems. As a system's administrator, you don't need this at all. Modern distros give you a centralised point of software installation (eg. YaST, yum or apt and various frontends). These systems will check for package integrity and authenticity and will keep track of dependencies and your system in a healthy state. Click'n'run installation can't do that for you (yet). Of course you can break these mechanisms in Debian, RedHat and SuSE by installing arbitrary software off the net, but it's not exactly encouraged.

      --
      I feel so sig.
    281. Re:Okay now... by Nite_Hawk · · Score: 1

      It was about 10 years ago now that I accidently did an rm -rf. I was trying to delete a file in some subdirectory, probably in my home directory or something. What should have read as:

      rm -rf /home/nh/file.foo

      was actually typed as:

      rm -rf / home/nh/file.foo

      I managed to catch it just as it had finished destroying bin, dev, and most of etc. I actually tried rebuilding those directories by hand at the time, but it was more work than it was worth. This was on redhat 5 and the event that sparked me to install Debian for the first time.

    282. Re:Okay now... by Anonymous Coward · · Score: 0

      It's neat that you're looking into this.

      I'm using coreutils-5.2.1. It looks like that bit of code is around line 776 in my remove.c.

    283. Re:Okay now... by IchBinEinPenguin · · Score: 1

      I am not willing to give my users administrator privileges so they can run some poorly-written application!

      Meanwhile, in the RealWorld(tm), management mandates that Application X is required to make $$$ and _will_ therefore be supported. (OK.. maybe not a weather widget)
      Oh how I wish I could ban things because they are a Bad Thing, if only it weren't for the fact that the PC's are just means to an end.
      X will make $$$ and cause problem Y which will take $ to fix. $$$ > $, so X is approved. (Note that typically Management, not IT, come up with the numbers).

    284. Re:Okay now... by Lagged2Death · · Score: 1
      Users are administrators by default on Windows boxes.

      No, that's not the case, and it has not been the case (IIRC) in any version of Windows that had user privilege settings.

      It's true that it can be difficult or impossible to get many Windows programs to work correctly in non-administrator accounts. And it's true that in many offices, the IT department has "solved" the problem by making every account an administrator.

      It's not a good situation, but I think everyone's better served by accurate information about why it is so. In this case, it's mostly the third-party software developers (i.e., not Microsoft) who have been slow to adopt a user/administrator model, or who have not made their software easy to use in such an environment.

      Microsoft is an evil, evil corporation, but you can't really dump the blame on them for this one.

    285. Re:Okay now... by anthony_dipierro · · Score: 1

      I just upgraded to 5.2.1, and it still works the same way. Maybe your kernel is returning a different error code. Not sure which error code is correct.

    286. Re:Okay now... by G-Spot · · Score: 1

      MS seems to be fixing this problem.

    287. Re:Okay now... by Anonymous Coward · · Score: 0

      (though I hear Congress still has elevator operators so those busy politicians don't have to worry about breaking their nails, or something).

      That's not the reason.
      The operator is there because otherwise the politicians would never go anywhere, being
      fundamentally unable to decide which button to push - seeing as how there's no one around to pay them for pushing the 'correct' button.

    288. Re:Okay now... by QuietLagoon · · Score: 1
      Here's why: What's the most important thing on your desktop? It's the data.

      In actuality, it is the data and my ability to access that data. Running as root increases the chances that I will not be able to access my data for all the reasons noted elsewhere.

      The car analogy is specious at best.

    289. Re:Okay now... by Uzik2 · · Score: 1

      >> Well, you could accidentally drive into a wall as well, it doesn't mean we should make all cars drive at 10 miles an hour. So, I don't see the added benefit.

      >Cars happen to have seat belts. Roads also have speed limits, so this analogy is flawed.

      This counter argument is flawed as well.
      Your car will still drive without the
      belt buckled and will go faster than the speed
      limit.

      --
      -- Programming with boost is like building a house with lego. It's a cool but I wouldn't want to live in it
    290. Re:Okay now... by Uzik2 · · Score: 1

      >Even if user data is the most important thing, if you run as root on a multi user box you put every users data at risk instead of only your own.

      There's only one users data on almost every
      computer. In 99% of the cases other user's
      data isn't secured against access/deletion
      by non root users either.

      --
      -- Programming with boost is like building a house with lego. It's a cool but I wouldn't want to live in it
    291. Re:Okay now... by larien · · Score: 1

      I came close to doing rm -rf / as root. I was doing some housekeeping on my desktop box and had some files I wanted to delete. They weren't owned by my user so I figured I'd su to root and remove them; su by default leaves the shell in the same directory, but I did 'su -' out of habit. One rm -r * later caused some errors, so I Ctrl-C'd it, but by that time I'd rendered my system unusable. Luckily it hadn't hit /export/home (Solaris system), so my data wasn't putzed, just the OS on the box. I just reinstalled the OS and the system was fine after that.

    292. Re:Okay now... by junkcannibal · · Score: 1

      Any program can be run as another user in windows as well. I have a special shortcut that opens up windows explorer as an admin after asking you for an administrator password. I can then run anything as admin or fiddle with other admin stuff while logged in as a limited user. Fast-user switching is for assholes. (troll) I try to emulate the fundamental security attitudes of unix whenever I have to use windows, too often these days. Granted, programs shouldn't require admin privileges, but there are ways of limiting access when programmers don't play nice with regular users.

    293. Re:Okay now... by Anonymous Coward · · Score: 0

      It's sort of a pain to post a real strace excerpt (junk characters), but here is what I'm seeing. I guess it could be a behavioral change in the linux kernel. I'm using 2.6.11-ck4 which includes patch-2.6.11.7.

      unlink("testdir")
      EPERM (Operation not permitted)

      unlink("file")=0

      rmdir("subdir")
      EACCES (Permission denied)

    294. Re:Okay now... by higginbo · · Score: 1

      What you are saying is true, but doesn't really give the whole story. For example, If you chmod 700 $HOME, and some one goes to http://yoururl/~yourhome, apache will say access is forbidden. but if you chmod 711 $HOME and do the same thing, it will not.

    295. Re:Okay now... by Anonymous Coward · · Score: 0

      > Raise your hand if you use Windows and you've gotten into the habit of smaking your enter-key, sometimes before that dialog even displays.

      Yes.

      I always do: "shift-delete + enter" to delete files.

      Of course, if the directory contains too many files, windows pop up a dialog that says 'please wait', with a cancel button, and it is the one that receives the "enter" keypress. Of course, the window don't stay long enough, so I often ends-up thinking I had delete files, while I had just cancelled the deletion...

      Windows brain-damage.

    296. Re:Okay now... by Anonymous Coward · · Score: 0

      Yes, chflags schg sets the system immutable flag on the BSDs. I protect the /bin, /sbin, /boot etc directories this way, so if I ever execute rm -rf / as root, I don't destroy my system completely.

    297. Re:Okay now... by bfields · · Score: 1
      Unless the user has . in his/her PATH the attacker will not be able to "replace" any programs like the sh. The attacker cannot log keyboards strokes or snif the tty without root privileges. Hence if your system is set up in a good manner the attacker cannot just get the root password.

      Oh, come on. The "." in PATH trick is just one of a billion tricks they could use. Off the top of my head:

      • They could *add* "." to your PATH, even if you don't have it there already.
      • They could use shell aliases to remap common commands.
      • They could intercept library calls with LD_PRELOAD tricks.
      • I know nothing about X, but I can't believe it would be that hard to get your own X session to intercept keystrokes for you.

      --Bruce Fields

    298. Re:Okay now... by bfields · · Score: 1
      That's why you're always told to use /bin/su rather than just typing su. They can change your .bash_profile to have your PATH pointing to trojan versions, but they can't get /bin/bash to execute something other than /bin/su when that's what you type, and they can't change /bin/su.

      Once they have control of a process with your uid, none of this does any good. If all else fails, I think it should be possible to read and write bytes to arbitrary offsets in any process with the same uid with ptrace(2).

      A determined attacker can probably get the root password somehow, but it makes it far harder, which is all you can hope for with security - make it so it's not worth an attacker's while to crack your system.

      I'd expect all of this to be simpler than the original stack smasher they used as the local exploit....

      --Bruce Fields

    299. Re:Okay now... by zakharin · · Score: 0

      I have this problem with the Ambient/Wunderground weather application. I must use it, though, because in all other apps relying on "official" readings the official data point is 12 miles (and currently 4 degrees) away.

    300. Re:Okay now... by marcosdumay · · Score: 1

      I even add that command line is hard to learn. Because of this, it intuitively is designed for advanced users. If you don't know what you are doing, you will not be able to use it!

      Anyway, modern versions rm often refuse to errase '/' so easily.

    301. Re:Okay now... by randomencounter · · Score: 1
      Run MSOffice 95-2000 as a non administrator user under Windows XP and tell me again that Microsoft isn't to blame.

      Yes, you can do it, but it takes quite a bit of administrative acumen that most people simply don't have. Frankly, the whole system is set up to require admin access at a frequency that means either having a full time admin, or having the user run as admin.
      Failure to provide the needed level of administration either gives you a malware-ridden useless installation, or results in a slow degradation into uselessness from accumulated errors.

      --
      Forget diamonds, copyright is forever.
    302. Re:Okay now... by tokabola · · Score: 1

      There is on my Linux Box. You can do a "new login", and get the GDM login screen. When the new user logs out, you're right back to the first user.

      Even more usefull is the option for "new login in a nested window". This causes the login screen to appear in a window on the original users desktop. The second user can log in, even using a different desktop (you can run KDE in a window on Gnome, for instance), and actually have both users signed in simultaneously. Although it's not the best idea, you can log in as root in that nested login, and run all your graphic config utilities as root.

      I suspect some distros don't enable this feature since it could be (and likely would be) abused and result in security problems. However, nested logins are a feature of the X-Window system (XFree86, don't know if Xorg has it) so any distro with a recent XFree could be configured to allow this.

      Tommy
      --
      Open Source for Open Minds
    303. Re:Okay now... by Dimensio · · Score: 1

      Win2k required you to have admin rights to burn cds!

      What software were you using? When I used Win2K to burn CDs (usually with DiscJuggler or Nero), I could do it as a Power User.

      True, Nero required additional software for non-Admins, but at least Ahead stated it up front and made obtaining and configuring the software very easy.

    304. Re:Okay now... by As+Seen+On+TV · · Score: 1

      I am sure the picky people would have no problem adding an "export PICKY_BITCH_USER=1" to their shell init script

      You're really not clear on this whole "Mac" thing, are you?

    305. Re:Okay now... by m50d · · Score: 1

      They need to have access to the device, windows 2000 doesn't give normal users enough. Unlike *nix you can't add them to a "burning" group, you need to give them root privs. More recent versions of cd burning apps come with a "run as" program that works like sudo, but for a while there was nothing like that, and if you can't afford the upgrade you're forced to have everyone admin.

      --
      I am trolling
    306. Re:Okay now... by m50d · · Score: 1
      Once they have control of a process with your uid, none of this does any good. If all else fails, I think it should be possible to read and write bytes to arbitrary offsets in any process with the same uid with ptrace(2).

      Very likely true. But I've yet to see an "off the shelf" program to let you do that. It's not impossible, but it's tricky. And what about people who don't use su, logging in on a separate console when they need to be root?

      I'd expect all of this to be simpler than the original stack smasher they used as the local exploit....

      Yes it is. But script kiddies outnumber exploit writers at least a hundred to one. Most exploits I've seen in use there has been a program/image/movie/etc. available to give you a shell from it. The guy attacking you is probably just pushing a button or running a script he got from a friend. And I haven't seen a simple script or similar to get local root, except where a bug exists (like the recent kernel local root exploit).

      --
      I am trolling
    307. Re:Okay now... by m50d · · Score: 1

      Nero, iirc 5, at a small educational institution that couldn't afford to upgrade. I don't know if that version works with extra software from ahead and didn't have time to find out about it and getting it set up - policy was just to have an admin account with the login details known by everyone. That's the only time I've seen win2k, everyone else seems to have stuck with 98 or moved on to xp.

      --
      I am trolling
    308. Re:Okay now... by Anonymous Coward · · Score: 0
      On Windows, devices have ACLs (access control lists), just like files and other NT objects. The correct way to give a user write access to a device is to add an ACE (access control entry) for the user, with the appropriate privileges, to the device object's ACL. Alternatively, if you want to use a 'burning' group to manage access, you could create such a group, add an ACE for it to the device, then add the user(s) to the group.

      The real problem with Windows is it's much harder to figure out how to do things like this than it is to use simple groups and mode bits on Unix-like systems, where devices are presented as files. The SysInternals WinObj tool helps a lot, but it's still more confusing.

    309. Re:Okay now... by Mr.+Slippery · · Score: 1
      and a lot of effort has gone into hiding that complexity in a modern car

      Insulating is not quite the same as hiding.

      The difference is that with cars a user knows there is underlying complexity, and is prepared to recieve dozens of hours of training and take it to experts (and pay real money!) for maintenance and when problems develop - even undergo government mandated licencing and safety and pollution inspections. Yet many expect a computer, orders of magnitude more complex, to "just work" for them without any effort on their part.

      I'm not saying that using a computer effectively has to be hard, just that it can never be trivially easy.

      --
      Tom Swiss | the infamous tms | my blog
      You cannot wash away blood with blood
    310. Re:Okay now... by dmolavi · · Score: 1

      The best way for Linux to break into the market isn't to emulate windows entirely. Oh, but it is: http://www.pen.k12.va.us/Div/Winchester/jhhs/math/ humor/comics/computer/linux.html hehe..

    311. Re:Okay now... by RogerWilco · · Score: 1

      Well, not rm -rf /
      but yesterday I did rm * in /etc
      (to much windows open, and focus follow mouse)
      I had to ask YaST to fix it, and still need to reconfigure some stuff.

      --
      RogerWilco the Adventurous Janitor
    312. Re:Okay now... by vidnet · · Score: 1
      Let's summarize (reply with additions/modifications):

      Running as root:

      • Trojans/Exploits (hereafter referred to as Things) can delete everything, not just your own files.
      • Lets Things load kernel modules to override /proc and various system calls, to hide itself.
      • Things can open raw sockets/low ports and sniff connections for various networking scams
      Running as a user:
      • Data storage facilities running as other users, like databases and maybe some backup services, won't be erasable.
      • If you have none of them, since you're the only user, Things can erase all user files. The rest are stock files available on the CD or from Click&Run.
      • Things can add LD_PRELOAD hacks to hide itself. While not as resilient as kernel modules, it would throw most people off.
      • Things can't get overly messy with the network.
    313. Re:Okay now... by Syberghost · · Score: 1

      I've never done that, but just last week I typed "grep ###### *" (where ###### was an employee number, and I was in a crucial configuration directory for our single-sign-on solution) and hit enter, only to realize I'd typed "rm" instead of "grep"...

      Thank goodness for backups.

    314. Re:Okay now... by Anonymous Coward · · Score: 0

      That's because autoconf sucks ass by default. Actually, it does that even if you change default behavior.

    315. Re:Okay now... by Anonymous Coward · · Score: 0

      Why did Apple ever hire a prick like you? Did you even read the text I put in brackets?

      Here's hoping you lose your job soon, asshole.

    316. Re:Okay now... by ebyrob · · Score: 1
      Was the point that searching isn't always the best method of finding data completely lost on you?
      Basically, yeah
      Replacing file-trees with search and removing the ability to keep file-trees
      Who said anything about removing anything?

      Searching is "the" method and at the same time you're not advocating removing existing functionality. Ri-ight.

      You're talking in meaningless and nonsensical abstractions. I'm talking about an actual computer program.

      I'm trying to have a conversation about useability, you seem to be looking for an "ooh that's cool" and a pat on your back for finding some useless(to me) search tool.
      first time I sat down at an OS-X workstation I couldn't even find the web-browser let alone documents.
      Hm. I'm having a hard time coming up with an explanation for this that doesn't involve massive, almost comma-cal, stupidity on your part.

      Your opinions about my intelligence might carry more weight if properly punctuated.

      in light of what you said about not having much use for Mac OS X. You asked, "Why can't I just grab the contents of my 'programs' directory and move it to a new machine?"

      Well... I don't use MAC's, and until I see this in action I doubt it really is that simple despite all claims to the contrary. I've heard a lot of similar claims and they've all been bunk. In every environment I've ever worked in, nothing ever is "that simple". In fact I have a co-worker who regularly tells me to try some new program because it works great and does all these things and doesn't break. So I download it and run it and without fail and within 10 minutes I break it in a way he has no idea how to fix, so I throw it away and move on to something else. (I reserve my time for programs that take me at least 30 minutes to break, or have some really nifty feature I can't get elsewhere...)

      If you want information related to scuba diving, you shouldn't have to go looking with lots of clicky-clicky-clicky. ...
      See what I mean? The computer goes from being a big shoebox full of paper to a machine that actually knows things and that can answer questions. This is good. This is important.


      No I don't. This is a pipedream. Computers are nothing more than stupid automatons doing exactly as they are told no matter what the user intended. Maybe someday in the distant future this might change, but we're a long ways from there. If you know the words "scuba diving" and how to quote it as a string, searching is just fine. But what if you're interested in methods of breathing underwater and don't know the name, or can't remember it? Trying to find this kind of information via a generically-indexed database or search engine can be painful at best. Also, as the dataset grows and changes the results for any given set of keywords changes over time. This might be great for entertainment, but it hardly fills the bill when you need to keep track of important information, none of which can fall through the cracks.

      To reiterate (and no matter how cool you think some app or class of apps is): Searching is a part, but not all, of the equation. The drawback of searching is that I have to find keywords, remember them, worry about varied results and go through complete and sometimes slow iterations to refine what I'm after. The drawback of file-trees is I have to take the time to organize, go clickety-clickety and remember something about how I organize. In either case I'd hate to have to go without tools or force someone off of a methodology that works for them...
    317. Re:Okay now... by HiThere · · Score: 1

      What you gain is that root level access doesn't happen by accident. (Unless you are running in a root shell...bad idea! Option should be hidden from novices.)

      It's true that you can sudo anything with your password, but that doesn't happen by accident. You need to explicitly type it in. On a single-user system that's as close to good protection as is feasible.

      --

      I think we've pushed this "anyone can grow up to be president" thing too far.
    318. Re:Okay now... by As+Seen+On+TV · · Score: 1

      Well... I don't use MAC's

      It's "Mac," short for "Macintosh" and not an acronym of anything. And I can't help but notice that the fact that you don't use them didn't stop you from telling me all about how Spotlight is supposedly useless. What strikes you as immediately and obviously wrong about this?

      Computers are nothing more than stupid automatons doing exactly as they are told no matter what the user intended.

      There are none so blind as those who will not see.

      Let's just go ahead and cut to the chase scene here, okay? You believe, with all the conviction of a precocious seventeen year old, that you understand everything, and that you know exactly how things oughta be. You speak in sweeping generalizations from a point of ignorance that you don't merely admit to but actively brag about. "I have no idea what you're talking about and in fact can't even spell the three-letter name of the product your company sells, but sit back and let me tell you exactly how things are."

      When I come to you with specific information about a shipping product, you close your eyes tightly and stick your fingers in your ears and insist, all facts to the contrary, that it can't possibly work.

      You have contributed nothing, and served only to make yourself look like a fool.

      This is not going well for you.

      I have some advice: Take a huge step back from that promontory of perfect certainty onto which you've heaved your considerable girth. Do something to diminish your ignorance before spouting off to the whole wide world from The Gospel According to Whatever Your Name Is.

      Just a suggestion. Take it or leave it.

    319. Re:Okay now... by Anonymous Coward · · Score: 0
      For anyone who is interested in how to get other X apps running under a different user context as described above, check out the xsu script:

      http://yolinux.com/MINI-HOWTO/Remote-X-Apps.html

      Great stuff.

    320. Re:Okay now... by Anonymous Coward · · Score: 0

      Always look at where you are before you do that command. This is why people who claim CLIs are so much better than GUIs just misunderstand having the screen store your state information in front of your face. It's a computer after all -- why not use it to do your work for you?

    321. Re:Okay now... by Eric+Damron · · Score: 1

      That's not exactly true. Unless the user starts the virus as root.

      I run Linux all the time and I rarly su to root. My system starts automatic updates that run as root but even if I get a virus it would have to execute some kind of exploit to gain root access.

      --
      The race isn't always to the swift... but that's the way to bet!
    322. Re:Okay now... by papaskunk · · Score: 1

      While that may have been an attempt to sound smart and sophisticated, may I remind you that any attempt to haughtily correct a 'humorous' post will usually result in a backlash, such as this. No, you may not remind me of what I already know! Take a joke!

    323. Re:Okay now... by bfields · · Score: 1
      That's not exactly true. Unless the user starts the virus as root.

      I'm not sure I've made it clear the sort of attack I'm talking about:

      1. The "virus" first gets access as user joe.
      2. Using joe's privileges, it installs some sort of wrapper for su, or keystroke logger, or whatever. It can't, of course, replace /bin/su, or log every keystroke from the keyboard, because it doesn't have root yet; but it should be able to do the equivalent for the purposes of capturing joe's activity, by modifying dot-files, etc.
      3. Wait for joe to su to root.

      At this point it shouldn't be hard to capture the root password that joe uses. No local root exploit required.

      Whether there are actually automated scripts that make this easy right now, I don't know. I don't imagine it would be that difficult to write one. So it'll probably happen some day.

      I rarly su to root.

      Well, that should slow it down at least.

      --Bruce Fields

    324. Re:Okay now... by John_Sauter · · Score: 1

      I am in the perhaps unusual position of having the confidence of my management. If something new comes in the door, I will try to get it to work within our data processing environment. If I cannot, I say that it is defective, and, so far at least, nobody has second-guessed me.
      John Sauter (J_Sauter@Empire.Net)

    325. Re:Okay now... by fuzzybunny · · Score: 1

      Use whatever's convenient for you, as in whatever works. I have an XP laptop for work with customers, a FreeBSD file & print server at home, a M0n0wall firewall, a second playing-around drive for my laptop with Debian Woody, a couple of live filesystem CDs with Auditor and other similar security-relevant distros as well as a Knoppix CD for recovery, and I'm buying a Powerbook soon to get real work done (network security analysis type stuff, PITA under Windows.)

      OS evangelism is stupid, and you have some good points about usability.

      As for your printing woes, please do have a look at CUPS--it's the mutt's nutts for UNIX printing as far as I'm concerned.

      --
      Cole's Law: Thinly sliced cabbage
    326. Re:Okay now... by composer777 · · Score: 1

      Good question, right now, I haven't been spending much time installing software, but the problem is, when I'm installing the system and setting things up, I tend to run as root, and by the time I'm done, I have gotten used to running as root.

      When I'm at work, I try to get work done, and log in as a regular user. But when I'm at home I tend to hack around a lot on linux. Back when I was running LFS (linux from scratch), set up could take weeks, and most of the time I was running LFS I was installing a piece of software or setting something up. Now, I'm sure your reaction is going to be that my usage of linux wasn't serious, but I don't think that we should underestimate the hobby user when you talk about linux. I also tend to experiment quite a bit with different distros, right now I'm running 64 bit fedora when I boot into linux, and colinux from my 32 bit fedora partition when I'm booted into Windows. Again, not a serious use of linux, but when I run at home I'm a hobbyist, not a sysadmin.

    327. Re:Okay now... by Anonymous Coward · · Score: 0

      Those 'lazy ass people who developed the software incorrectly' such that it requires running as an administrator *include* Microsoft!

    328. Re:Okay now... by Mornelithe · · Score: 1

      Well, I've never done LFS, myself, but...

      What does it take to install a piece of software? Compiling it? You can do that as a user, and then do 'sudo make install'? Editing configuration files? Even if you have to do many in a row, you can resort to 'sudo -s' and then edit in that root shell. It doesn't require being root for absolutely everything you do.

      I have to be root to install software on my machine (Gentoo), but with a proper sudoers file (letting me run 'emerge' as root without a password), and an alias (emerge='sudo emerge'), installing software as a user is as simple as installing it as root, and yet, it doesn't require me to run my web browser and everything else as root. I realize you can't alias things that easily for LFS, but then, that's a rather special case.

      In any case, a hobbyist argument still doesn't justify the opinion in this article, considering that Linspire is definitely not a hobbyist distro. Having an average user run as root just makes it that much easier to turn the machine into a zombie, and having to type in your password once a week isn't enough of a pain to justify that situation.

      I don't know what constitutes a "serious" use of Linux. Hell, what do I use it for? Browsing the web and homework? Is that significantly more serious than tooling around with it as a hobbyist?

      The only way your use would be less serious is if you can sincerely say: "I don't care if my machine becomes a spam zombie, or all my activities/keystrokes are logged by some unknown party." If you agree with that, then fine, run as root. If not, then running as a normal user and only switching to root when needed is just one additional line of defense that makes it less likely for your box to get totally owned.

      --

      I've come for the woman, and your head.

    329. Re:Okay now... by ebyrob · · Score: 1

      You have contributed nothing, and served only to make yourself look like a fool.

      Ah good. I'll consider us equals in this endeavor.

      Cheers.

    330. Re:Okay now... by Anonymous Coward · · Score: 0

      ooh burn

      and the 'cheers' is what really sold it. like you're to the manor born or some shit. 'i'm a twit, cheerio, ta, pip pip, 1066 and all that.'

      cockmunch

    331. Re:Okay now... by CRCampbell · · Score: 1
      * ActiveX and a lot of spyware is contained in windows when running as non-administrator. It's running as admin (like most people do), that cause the majority of problems with things. *

      NOT ENTIRELY TRUE! Users can still destroy machines with Active X whilst logged in as a non Admin. They do it all the time in my organization, because Active X isn't filtered out as a matter of security policy.

      So, Windows (2000, NT) is even worse than you thought.

    332. Re:Okay now... by theLOUDroom · · Score: 1

      I totally agree with you about privilege levels. I was all about running as a non-privileged user. That was until realism and idealism clashed. Some programs literally won't work right without for example administrator rights on Windows.

      That's not realism vs idealism, that's broken software.

      What you're doing is saying that it's not practical to lock the doors on a house because your particular door locks are jammed. You're not illustrating the idea as unworkable so much as your unwillingness to fix things that are broken.

      --
      Life is too short to proofread.
    333. Re:Okay now... by Anonymous Coward · · Score: 0
      so you wan't rm to move files to the trash?

      What English word is being contracted here with the apostrophe? "wan't"="wa not"? I think you mean to use the word WANT, not a self-constructed contraction of a non-existent word, "wa".

  2. Mr. Lindows is just stirring shit as usual... by garcia · · Score: 4, Insightful

    Michael: I think, like everything, it's a question of balance. Ease of use, versus security. I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer. They say "oh, yeah, it is!", but it really isn't. Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

    Techincally it's gaining control over your system without you knowing it and running exploitable programs as root makes that easier. If the hackers get access to your libraries, programs, etc, they can do far more damage to you by sniffing your data w/o your knowledge. Hackers aren't going to just steal your data and run. If they can gain easy access to the system they are going to modify it and snoop everything and keep getting what they came for.

    Michael: Then you could say "Well, it's not really about your data, it's that people could accidentally mess things up!". Well, you could accidentally drive into a wall as well, it doesn't mean we should make all cars drive at 10 miles an hour. So, I don't see the added benefit. I DO see it's an added pain in the ass when grandma tries to change her wallpaper, and it tells her "you don't have root privileges". What are you talking about, man? I'm just trying to use my computer, or change the clock, or any one of a hundred other things. So, people always say "it's less secure", but I defy anyone to point out a single instance, and people all go "Well, I, erm, it's theoretical!". There's no one area I think you can point out - In this instance, a machine that's run with the root user could be compromised, in this instance one couldn't be compromised.

    I am in no way a master of Linux/UNIX and I never claimed to be but even I know that if you are exploited while running something as root more damage can be done to a lot more services, files, etc, than if you were just running it as a user. It's not theoretical. It's fucking very real and it's idiots like this guy that make it easier and easier for more zombie boxes to get out there. Look at Windows... Yeah, no, we don't need Linux to end up like that too.

    I want to know who the hell this guy is talking to that don't give him a valid argument. I have a feeling they are and he isn't listening.

    Michael: I know the hardcore geeks feel differently, that's fine. When somebody installs Linspire, we say "do you want to set up users, yes or no", we give them the choice, right there when they start up for the first time. If they want to set up multiple users, they're welcome to do that, but we don't force them to. That's the difference we have.

    It shouldn't even be a choice. Prompt for a password (like OS X) when something that needs root privledges runs. If it has succeeded with the Mac then it can with Linspire users too. If you are so concerned about making the users have a positive Linux experience rewrite the dialog boxes when they ask for "root priveledges" so that they are human readable. Don't just eliminate it and say that there's no valid reason not to. Taking the easy way out doesn't solve the problem.

    Since when is Michael Roberson a trusted source? He's an asshole that's just into pushing the envelope and making waves (remember Lindows and MP3.com?) Right now he's doing exactly the same thing. "See, those Linux users are trying to make it hard for the layperson to use "their" OS and I'm trying to make it easy. Listen to me! I'm trustworthy!"

    1. Re:Mr. Lindows is just stirring shit as usual... by Owndapan · · Score: 2, Insightful
      I DO see it's an added pain in the ass when grandma tries to change her wallpaper, and it tells her "you don't have root privileges"

      Correct me if I'm wrong, but I don't remember ever requiring root access to change your desktop wallpaper.

    2. Re:Mr. Lindows is just stirring shit as usual... by 0racle · · Score: 4, Funny

      Its the root window, it must require root privileges to change.

      --
      "I use a Mac because I'm just better than you are."
    3. Re:Mr. Lindows is just stirring shit as usual... by Mysticalfruit · · Score: 1

      What? If I'm in Gnome and I right click on the background and choose "Change Wallpaper" I can pick any JPG/PNG/etc I want and it'll set that as my background.

      I don't remember root privlages getting involved at all.

      --
      Yes Francis, the world has gone crazy.
    4. Re:Mr. Lindows is just stirring shit as usual... by dgatwood · · Score: 1, Funny

      As someone else once put it....

      .* <--joke

      .o
      -|- <-- you
      / \

      --

      Check out my sci-fi/humor trilogy at PatriotsBooks.

    5. Re:Mr. Lindows is just stirring shit as usual... by berenddeboer · · Score: 1

      > Prompt for a password

      How is that more secure? The typical grandma will be baffled, and type in the password whatever requested it.

      Michael has a point here and it's not easily argued away.

      --
      If I had a sig, I would put it here.
    6. Re:Mr. Lindows is just stirring shit as usual... by NevermindPhreak · · Score: 1
      When somebody installs Linspire, we say "do you want to set up users, yes or no", we give them the choice, right there when they start up for the first time.

      It shouldn't even be a choice.

      why shouldnt it be a chioce? its my computer, why cant i set it up the way i want to? if i want the added security of user priviledges then ill choose to use them. if i want to be lazy and not use them, so be it.

      these are the things a lot of people think about when not deciding to use linux. instead of forcing people to use your way to do things, give them options and tell them in what ways which is better.

    7. Re:Mr. Lindows is just stirring shit as usual... by drgonzo59 · · Score: 1
      I think Michael is talking about home desktop system where users just want to change wallpaper, burn CDs, connect their usb toys to the computer, use a joystick, install and un-install programs and other such things. You are thinking about servers connected to the network that hold lots of important data, have to administered by an expert and perhaps serve many users. Those are the two extremes, typically Linux is on the server extreme, while most poeple at home still run Windows 98 that came with their computers in the late 90's. Then there the more realistic grey area in the middle, where people as myself are, that want the machine to be both a desktop but sometimes run some services and be secure too.


      Now Linspire is trying to move Linux into that home desktop market, which I have to give it to them, is a lot harder than have Linux replace traditional UNIXes in the server market (which already is happening at a fast rate). So while I agree that it is a lot more secure to run services not under root, I don't think the average home user (think moms, grandmas, and sociology teachers ;) should be expected to know about root, suid bits, and quotas. Most will unaliterally choose the more usable system over a more secure one if they couldn't get both.


      So the best that can happen is to find some compromise, like accept the fact that some deskto p home systems will be less secure but still expect the user to type in an "administrative" password once in a while.

    8. Re:Mr. Lindows is just stirring shit as usual... by Anonymous Coward · · Score: 0

      I really don't think it would be difficult for people to understand if, during the install process or when the computer first boots up after being sold, it asks you to set the root password. All you really need is a brief explanation saying something like

      "The root password is required to make important changes to your system. This is neccessary both to prevent unauthorized use and to ensure that users can't make permanent changes to the system without intending to."

      "To keep your system secure, never send this password over the internet."

      I'm sure someone else could write up something better, but I think I covered the basic points. Sure, a lot of people will skip over it, but those who really need to see this information, such as your archetypical grandmother, will be certain to read everything very carefully.

    9. Re:Mr. Lindows is just stirring shit as usual... by Anonymous Coward · · Score: 0

      The whole "It's just a desktop, why secure it?" approach is the reason there are so many wonderful botnets turning out spam and scam emails all the world around.

      We need people who use computers to get a clue when it comes to security, but that clue has to come from the OSs they use. Running by default as non-root and asking good, simple questions when asking for a gksudo password would be a great start. Things like:

      "POTENTIAL SECURITY PROBLEM: You are trying to add a new program onto your computer. Do you trust the authors of this program with your personal emails, bank details and anything else you may type into this computer? If not, do not enter your password. If you are unsure, please contact [whoever] for technical assistance".

      Grandmothers aren't generally stupid, they just have a steep learning curve late in life for something extremely complex. Explain it in good, plain English and they'll be fine, for the most part.

      Getting people thinking about who they want to trust would be a great start. The [whoever] bit could be an option at install time - "Please enter the name of somebody who can assist with the management of this computer". That would give the OS the ability to prompt Mrs Grandma to call Young Grandson The Geek whenever she's about to do something potentially problematic.

      Obviously you'd ask different questions if she wanted to change the clock rather than install an app. But that's pretty trivial desktop-environment type stuff to sort out.

    10. Re:Mr. Lindows is just stirring shit as usual... by hackstraw · · Score: 3, Informative


      If this Michael guy has ever seen a rooted Linux system with one of those groovy kernel modules loaded to hide the doings of the people that rooted the box, then he would guess a 2nd time about his assertion that its OK to run Linux as root all the time.

      You think that WIndows zombie boxes are a problem? However, those systems are able to be fixed (to my knowledge, don't use windows). A rooted box with a kernel module installed to hide itself, has to be completely restored.

      I'm glad you mentioned OS X. I believe that it is a beautiful compromise between running as a user and asking for permission to escalate the privileges when needed. The best part of it is that it _rarely_ asks for administrator privilege, and when it does it makes sense. If someone opened an email attachment and it asked for administrator privileges, that would be a bit fishy (although some people would fall for it).

    11. Re:Mr. Lindows is just stirring shit as usual... by lachlan76 · · Score: 1

      However, those systems are able to be fixed (to my knowledge, don't use windows). A rooted box with a kernel module installed to hide itself, has to be completely restored.

      IIRC, there was an article on slashdot a while back about a Windows kernel-level rootkit. And this is a nice piece of writing about how to make one for Solaris. As for Linux, I'm not so sure how easy it is, but I could probably work it out if I had the time.

    12. Re:Mr. Lindows is just stirring shit as usual... by rpdillon · · Score: 1

      Sure it is. This is about spyware. This is about spam-sending zombie boxes. This is about programs like BackOrifice.

      If you're not installing software, and a box pops up and says "Please type in the admin password to install Mhfyshc456.sh", even a novice user will be suspicious.

      Now, you can argue "Anytime a computer asks for a password, someone will enter it!", but you're making what is a technical issue into a social engineering issue. I would assume that most slashdotters know that social engineering will win out every time...it is simply the easiest way to break into a system 90% of the time. So sure, the system is vulnerable, we knew that.

      But to say that since someone can be convinced to supply a password is a reason to scrap passwords is obviously incorrect. Michael *doesn't* have a point. Under an unprivileged user scenario, 90% of power users and 30% of normal users will be wise to the ruse, and it won't work. The remaining folks will have at least been notified that something occurred, and if they start to notice odd behavior, they will have an idea where it came from. With all users having root, 100% of users will never be with wiser that their system was just comprimised.

      You can't argue one system is just as secure as the other.

    13. Re:Mr. Lindows is just stirring shit as usual... by ImaLamer · · Score: 1

      From your topic I hope you can agree the whole topic is flamebait and the submitter has just trolled us into going after Linspire...

      Aren't there better things to talk about?

    14. Re:Mr. Lindows is just stirring shit as usual... by Anonymous Coward · · Score: 0

      "If they want to set up multiple users, they're welcome to do that"

      wait a minute. WHy is it that not setting up multiple users means that the default user is root?

      Shouldn't there be some (non-root) default user account that the user is assigned to?

      If they are too much of a dumb ass to deal with an admin password that may be needed occaisionally --- they really shouldn't have a computer. Not these days.

      I know. People (especially unkowledgable ones) can be tricked into using a root password. The lock and your door can probably be broken quite easily; but its still there.

      Knowingly leaving the front door unlocked is at least a bit irresponsible. All we need are more spam zombies and other machines with spys in their wires.

    15. Re:Mr. Lindows is just stirring shit as usual... by More+Trouble · · Score: 1

      A rooted box with a kernel module installed to hide itself, has to be completely restored.

      You can repair such a root kitted box by 1) booting from clean media, and 2) running a tool like radmind that knows how the system is supposed to be and can fix it.

      :w

    16. Re:Mr. Lindows is just stirring shit as usual... by lahvak · · Score: 2, Insightful

      Now Linspire is trying to move Linux into that home desktop market, which I have to give it to them, is a lot harder than have Linux replace traditional UNIXes in the server market (which already is happening at a fast rate). So while I agree that it is a lot more secure to run services not under root, I don't think the average home user (think moms, grandmas, and sociology teachers ;) should be expected to know about root, suid bits, and quotas. Most will unaliterally choose the more usable system over a more secure one if they couldn't get both.

      But the point is, if the user interface to the system is well designed, they don't have to know anything about root, suid bits and quotas. They can have a machine that is both more usable and more secure, and it is job of companies as Linspire to give it to them. Instead, Robertson is saying: we can't be bothered about this, our user interface is a piece of crap, so we will simply let everybody run as root and forget about it. That's exactly what Microsoft did, and that's exactly the reason there is now so much spyware and other crap.

      --
      AccountKiller
    17. Re:Mr. Lindows is just stirring shit as usual... by MrResistor · · Score: 1

      its my computer, why cant i set it up the way i want to?

      Because your spam-spewing zombie box is a problem for the rest of society. As they say: your right to swing your fist ends at my face.

      --
      Under capitalism man exploits man. Under communism it's the other way around.
    18. Re:Mr. Lindows is just stirring shit as usual... by MrResistor · · Score: 1

      I don't know. Linspire is the primary OS on those $199 Walmart PCs.

      And before you say only Linux people will buy those, you should consider that Clark Howard has one, and mentions how satisfied he is with it on a regular basis (he also talked up OpenOffice recently).

      There are plenty of people in the world that never install anything on their computers beyond what it came with, and those are exactly the kind of people those cheap Linspire systems will appeal to. Should they really be left practically naked to the world like that?

      --
      Under capitalism man exploits man. Under communism it's the other way around.
  3. Full article link and observations on root by ZiZ · · Score: 5, Insightful
    An easier-to-read 'formatted-for-print' version is here. (Not here, as I tried after decoding the base64-encoded GET, but that's beside the point.)

    Not running as root works like this. Your data is no more inherently safe than it is when you /are/ running as root, but nobody ELSE'S data will fall prey to your screwup, nor will the central integrity of the system. (For granny, this means that grandson Billy can ssh in, recover this morning's backups from the write-once partition, and she can keep going, having lost minimal data.)

    Running as root is like pointing a loaded gun at everyone just in case they're a criminal.

    Not running as root is like fastening your seat belt. Sure, you're not intending to get in an accident...

    Running as root is like driving down the highway with your hood open and your oil cap off.

    Not running as root is like locking your door when you leave.

    Running as root is like posting to slashdot without reading TFA. :)

    --
    This flies in the face of science.
    1. Re:Full article link and observations on root by nxtr · · Score: 5, Funny

      Running as root is like posting to slashdot without reading TFA. :) Exactly; you get what you want done faster, like get the much coveted +5 Funny.

    2. Re:Full article link and observations on root by IthnkImParanoid · · Score: 1, Funny

      Metaphors should be like driving: a privilege, not a right. Reading those was like stuffing strips of colored paper into my bleeding eye sockets and calling it a ticker-tape parade.

      --
      It's nothing but crumpled porno and Ayn Rand.
    3. Re:Full article link and observations on root by mabinogi · · Score: 1

      Nice simile.

      --
      Advanced users are users too!
    4. Re:Full article link and observations on root by Anonymous Coward · · Score: 0

      You just made Robertson's point for him. Go back and read his quote.

    5. Re:Full article link and observations on root by Anonymous Coward · · Score: 0

      *Not running as root works like this. Your data is no more inherently safe than it is when you /are/ running as root, but nobody ELSE'S data will fall prey to your screwup, nor will the central integrity of the system. (For granny, this means that grandson Billy can ssh in, recover this morning's backups from the write-once partition, and she can keep going, having lost minimal data.)*

      linspire is not meant a) to be remote used, b) to be multi-user - so really not running as not root wouldnt protect anything, the default use is supposed to be able to do all the root stuff on linspire anyways, so just naming it something else wouldn't really change anything.

    6. Re:Full article link and observations on root by Anonymous Coward · · Score: 0

      His point, which everyone seems to miss, is that running as root is no less safe if you want your operating system to emulate Windows.

  4. What a great question! by ScentCone · · Score: 1

    Coming from the Windows side, I hear this warning constantly, but rarely hear about the practical fallout. OK, splain, Lucy.

    --
    Don't disappoint your bird dog. Go to the range.
    1. Re:What a great question! by spectre_240sx · · Score: 4, Insightful

      One word: Spyware. You run as Administrator, it hoses your machine. If developers would actually write software so that users didn't have to run as Admin just to open up notepad, then spyware wouldn't be anywhere near as big a problem as it is right now.

      Now take that one step further and consider a malicious virus being accidentally executed by the same user that thought Bonzi Buddy was cute. Spyware is bad, but that virus might, oh... kill all your .jpg .doc and .mp3 files.

      Now how do you feel about running as Administrator?

    2. Re:What a great question! by Homology · · Score: 1
      Coming from the Windows side, I hear this warning constantly, but rarely hear about the practical fallout. OK, splain, Lucy.

      When you run as root, you can easily hose your entire system. Small mistakes/typos can ruin your whole system. For instance "rm -fr / somedir/*" (not the space after first /) will happily remove just about any file. There are many commands available to root that should be used with care. Applications have bugs that when run as root can do much more damage.Better not run as root for daily use.

      From a security point of view : If you run as root it's easier to get a compromised system. If many people runs as root, you can be sure that viruses will be a plague for Linux as well (or any other *BSD/Unix)

    3. Re:What a great question! by Anonymous Coward · · Score: 0

      Spyware is bad, but that virus might, oh... kill all your .jpg .doc and .mp3 files.

      Oh no, my .jpg, .doc, and .mp3 pr0n!

    4. Re:What a great question! by aslate · · Score: 1

      Fine, i've never had a real problem with spyware, and as i've grown older i've eliminated it. Spyware isn't an issue for me because i'm not stupid enough to go around installing and clicking on whatever i see. Just did a random scan on Ad-Aware and got 0 results.

      Viruses? I think i've had a couple of false positives if that's what you mean...

      These aren't really problems that affect you if you know what you're doing, or even if you simply don't click on everything you see. It may also help that i use Firefox and Thunderbird and only IE whenever i need to (Windows update, my bank now likes Firefox).

    5. Re:What a great question! by spectre_240sx · · Score: 1

      No, these aren't problems that should affect a skilled user, but we're not talking about skilled users. Linspire is aimed towards many people, including the type that doesn't know enough to be careful about what they click on. And windows? Well, I think we all know what the majority of windows users are like. Also, don't forget, humans aren't perfect. Even the smartest sysadmin does stupid things sometimes, I'm sure of it. Running as a non-privledged user prevents those mistakes from being disasters.

    6. Re:What a great question! by drsmithy · · Score: 1
      One word: Spyware. You run as Administrator, it hoses your machine. If developers would actually write software so that users didn't have to run as Admin just to open up notepad, then spyware wouldn't be anywhere near as big a problem as it is right now.

      This about the only really compelling argument - and the only reason it's compelling is because *today's* malware assumes the user is running as admin. However, there's very little that malware actually does that requires admin access, so once non-admin users become commonplace, malware authors will simply rewrite their software to assume the user doesn't have admin access.

      Now take that one step further and consider a malicious virus being accidentally executed by the same user that thought Bonzi Buddy was cute. Spyware is bad, but that virus might, oh... kill all your .jpg .doc and .mp3 files.

      Destructive viruses are old and busted. Teh n3w h0tn3ss is mass-mailing and DDoS zombies.

      Now how do you feel about running as Administrator?

      Personally I don't, but I'm not under any illusions that it offers a great deal more inherent security (from a long term perspective) to a single-user desktop. Once malware authors adjust their software to do away with the "running as admin" assumption, the problems will start all over again.

    7. Re:What a great question! by lachlan76 · · Score: 1

      Once malware authors adjust their software to do away with the "running as admin" assumption, the problems will start all over again.

      Well a user isn't gonna be running an open mail relay are they? Unless they decide to chmod 777 the kernel :-|

    8. Re:What a great question! by spectre_240sx · · Score: 1
      This about the only really compelling argument - and the only reason it's compelling is because *today's* malware assumes the user is running as admin. However, there's very little that malware actually does that requires admin access, so once non-admin users become commonplace, malware authors will simply rewrite their software to assume the user doesn't have admin access.
      Ok, so here's where OS security comes in. Say there's a provision that keeps a non-privledged user from accessing any startup routines. In this case, the malware won't be able to start itself every boot and becomes MUCH easier to remove. Actually, I think the operating system should warn ANY user of whether an installation or process is modifying startup config, but that's beside the point.
      Destructive viruses are old and busted. Teh n3w h0tn3ss is mass-mailing and DDoS zombies.
      Destructive viruses haven't been a problem for a while. I'll grant you that. However, there's nothing that says they won't make a return. Even one destructive virus can do a lot of harm if it's well written. Why chance it?
    9. Re:What a great question! by Anonymous Coward · · Score: 0

      The "Running as admin" problem is the reason that anti-virus software has to play such a constant game of catch-up with virus authors. They're fighting on even ground - admin vs admin.

      The problem with AV software at the moment is that it's an addiction - you can't block the problematic behavior (after all, most worms do pretty much the same stuff) because another worm just finds a new way to cut the throat of the AV package. Then the upstream provider has to patch and update all the users to prevent this happening in the wild, and the next round starts.

      If that fight is a user-level virus versus a root-level AV package it's a lot easier to win. Sure, it just means that viruses have to include a rootkit. But that then leaves you with a bigger challenge than just, say, getting a user to click on a zip file. You've got to have another vulnerability after that which will give you root permissions before you can even start working on killing any running AV software. Assuming viruses on Linux become an issue then it's still easier for a root-level AV package to keep an eye out for things trying to root your box before they have an opportunity to attack the AV package itself.

    10. Re:What a great question! by dr_d_19 · · Score: 1

      Now take that one step further and consider a malicious virus being accidentally executed by the same user that thought Bonzi Buddy was cute. Spyware is bad, but that virus might, oh... kill all your .jpg .doc and .mp3 files.

      Those files would have been deleted for this user even if he/she does not run them as root. Of course, the files of other users would not have been affected, but there's still a lot of damage right there.

    11. Re:What a great question! by drsmithy · · Score: 1
      Well a user isn't gonna be running an open mail relay are they?

      However, they're quite capable of running some other daemon that listens on a non-privileged port for a message and a list of addresses (or even just a web page to grab the details from) to mass-mail out spam.

    12. Re:What a great question! by drsmithy · · Score: 1
      Ok, so here's where OS security comes in. Say there's a provision that keeps a non-privledged user from accessing any startup routines. In this case, the malware won't be able to start itself every boot and becomes MUCH easier to remove.

      Most OSes/environments I'm aware of have a per-user method of starting applications on login ("Startup Folder" in Windows, "Login Items" in OS X, etc).

      The vast bulk of computers are *not* centrally-managed multiuser machines, a simple fact a rather large proportion of /. has trouble grasping - even the ones who would never have been in a situation to see such an environment. When 99% of the time the computer is in use it's in use by the same person, 0wning that person's "startup items" is the same thing as having access to the system's "startup items".

      Actually, I think the operating system should warn ANY user of whether an installation or process is modifying startup config, but that's beside the point.

      To give them another meaningless dialog to click "Yes" to ?

      However, there's nothing that says they won't make a return.

      Ah, but there is. In the past creating viruses and other "malware" were the domain of bored and/or malicious geeks. Today, they're mostly the domain of organised crime and professional spammers - neither of who have anything to gain by letting you know their software is present by damaging stuff.

    13. Re:What a great question! by spectre_240sx · · Score: 1

      Most OSes/environments I'm aware of have a per-user method of starting applications on login ("Startup Folder" in Windows, "Login Items" in OS X, etc).

      This isn't news to me. Reread my statement again and you might find that I'm talking about limiting that ability.


      To give them another meaningless dialog to click "Yes" to ?


      Who said that? Apple has already come up with a partial solution to this. Use VERBS instead of a simple yes or no when it comes to buttons. Instead of Yes, it would be "Make this run on startup" or something similar. Most people aren't stupid, they're lazy and don't want to read dialog boxes. Giving them a short cliffs note will inform them better as well as possibly interesting them enough to read the entire dialog.


      Ah, but there is. In the past creating viruses and other "malware" were the domain of bored and/or malicious geeks. Today, they're mostly the domain of organised crime and professional spammers - neither of who have anything to gain by letting you know their software is present by damaging stuff.


      And how exactly does that prove to me that kids won't get bored and start writing viruses again? The nature of malware changed once, and it will certainly change again in the future. There's nothing to say it won't go back to what it was before. Beside that, who knows what else it could turn into next. All the more reason to not run as root.
  5. Uhhh by elid · · Score: 1

    So that IE users don't end up downloading + installing every ActiveX spyware app in existence and messing up the entire computer?

    1. Re:Uhhh by Anonymous Coward · · Score: 0

      The difference being that instead of having it automatically done, they get trained to type in their password every time it asks, rather than clicking "yes"?

      Yeah. That's nice. I'm sure that has no chance of being abused.

    2. Re:Uhhh by ink · · Score: 2, Informative

      Unfortunately, a normal user can install any browser plugin that they want to. Running as root would simply allow the user to install plugins for other users as well. For the curious, you can install them in $HOME/.mozilla/plugins (among other locations). Running as a normal user will not prevent your box from becoming a zombie, unless you have some kick-ass SELinux rules in place.

      --
      The wheel is turning, but the hamster is dead.
    3. Re:Uhhh by Greyfox · · Score: 1
      Security is an ongoing process, but having a user level app compromised on a Linux system is a far cry from having the entire system compromised from an administrative account. If you keep your system updated (Now what did we say about that...?) then the chances of your attacker finding a privlidge escallation from userspace is pretty small.

      Isolation procedures from user space are a lot easier too. rm -rf the user's home directory. Problem solved. You can lock his user ID and password if you want to pick through his stuff before you do that. Note that in MOST cases, you'll be the user you're doing this for so it's not a BOFH sadistic sort of process. You're just cleaning up after a compromise. Contrast this with what would happen if your app is compromised while you're root -- at that point the only way to really be sure is the ol' Fdisk, Format, Reinstall procedure.

      It's also a lot harder to hide your tracks as a regular user. You can't edit logs or install rootkits so that netstat doesn't show your high socket servers that you set up. You can't just open the ethernet card in promiscuous mode and watch user IDs and passwords go by and you can't create a new user to hide your activities in.

      So yeah, you can still have your system taken over by crapware if you run as a normal user, but the impact that crapware can have on your system will be greatly reduced. Not running as root is simply one of many security measures that you HAVE to take if you don't want your system taken over on the Internet. It's not the first or last one but it certainly is an important one. Not taking every step you can to insure your safety is a sure sign of foolishness (How many people drive without seatbelts anymore? It's the same damn thing.)

      --

      I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

  6. Wow by bmw · · Score: 3, Insightful

    You've got to be kidding me. Is this just a big troll or is this guy actually that ignorant? Who the hell has he been talking to anyway? The reasons for doing day-to-day things as a non super user is one of the most basic security concepts ever. Even my parents understand this. The reason you don't run everything as root is to avoid COMPROMISING THE ENTIRE MACHINE if some random application has a vulnerability. You don't want each and every little program you run to potentially allow someone to gain full access to everything on your computer. Not to mention protecting the computer from the application itself. I don't want some poorly written piece of software accidentally deleting important system files or some other user's data. And how about protecting the system from the user themselves? How many people here have accidentally rm'd a bunch of important system files (or all of / for that matter) on accident? I know I have and I consider myself a very careful person when it comes to such things.

    C'mon... How fucking retarded can you be?

    He does _almost_ make a good argument for his case though...

    Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

    That statement does have some merit but it definitely isn't always true and even then, I would much rather compromise only my data than have someone gain access to the entire system. If they only get my data, that's all they get. If they gain access to the entire system there is no limit to what they can do... What if they want to setup a very well hidden rootkit and snoop around on my box (watching traffic, capture credit cards, etc. etc.) for as long as possible? Not to mention multi-user systems... A compromised super user gives them full access to EVERYONE's stuff.

    And of course, after he says something nearly sensible he goes on to completely shoot himself in the foot by making another completely ridiculous challenge...

    So, people always say "it's less secure", but I defy anyone to point out a single instance, and people all go "Well, I, erm, it's theoretical!". There's no one area I think you can point out - In this instance, a machine that's run with the root user could be compromised, in this instance one couldn't be compromised.

    What world does this guy live in? Is he completely surrounded by idiots? Remind me never to go anywhere near Linspire.

    1. Re:Wow by dougmc · · Score: 1
      That statement does have some merit but it definitely isn't always true and even then, I would much rather compromise only my data than have someone gain access to the entire system.
      The statement has a lot of merit. But even so, this merit does not make `running as root' as safe as `not running as root', because `running as root' has additional dangers.

      Of course, as a general rule of thumb, once you have access to a box, there's almost always a way to `hack root'. This assumes that it's a knowledgable person who's gotten in though -- in most cases, it's either a clueless script kiddie, or even more likely some sort of worm or automated scanner.

      Still, only an idiot would claim that `running as root' is as safe as `not running as root' -- at least without adding a long list of qualifiers to that statement.

      What world does this guy live in? Is he completely surrounded by idiots? Remind me never to go anywhere near Linspire.
      Oh bmw -- don't ever go anywhere near Linspire.

      As for what he said, remember who his target audience is -- the target audience for Linspire is idiots. Or at least those not really familiar with computers. Microsoft and AOL go after the same target audience as well.

    2. Re:Wow by Anonymous Coward · · Score: 0

      I dont think the slashdot crowd really gets it, This guy is talking about SINGLE USER SYSTEMS for single users if thier local account is comprimised the entire system is basically comprimised. I really dont think he is advocating giving all users of a multi-user system root access.....

      Start thinking a little differently fellas, not everything is black and white.

    3. Re:Wow by bmw · · Score: 1

      Microsoft and AOL go after the same target audience as well.

      Indeed. And look what a wonderful world they have given us.

    4. Re:Wow by bmw · · Score: 1

      I dont think the slashdot crowd really gets it, This guy is talking about SINGLE USER SYSTEMS for single users if thier local account is comprimised the entire system is basically comprimised.

      This is where you're wrong. Even with only a single user at least the attacker only has access to your personal files and is unable to trash the entire system.

      I really dont think he is advocating giving all users of a multi-user system root access.....

      No one said he was. It still isn't a good idea for a variety of reasons. Some of which I listed above, including protecting your system from a poorly written piece of software itself as well as protecting the system from the user making mistakes and damaging important system files.

      There's a reason that the majority of systems have had some sort of priveleged account system. It IS more secure.

    5. Re:Wow by javaxman · · Score: 1
      What world does this guy live in? Is he completely surrounded by idiots? Remind me never to go anywhere near Linspire.

      He lives in a world where he's this rich guy boss-man who scoffs at the silly common people beneath him and fires idiots who can't understand that he's always right?

      Seriously, he may be a great guy, I'm just goofing on him above, but... his position in the article is pretty pointy-haired-boss, don't you think? Ignore the idea of multiple users, ignore the idea of protecting you from you mistakes, in short, design a craptastic product, and maybe his ideas don't sound totally stupid... if you're willing to ignore real life for a moment, and design a completely different system from the one that will work...

    6. Re:Wow by anthony_dipierro · · Score: 1

      Even with only a single user at least the attacker only has access to your personal files and is unable to trash the entire system.

      If an attacker gets access to my personal files you can bet I'm going to reinstall my entire system anyway. So what's the difference?

      Some of which I listed above, including protecting your system from a poorly written piece of software itself as well as protecting the system from the user making mistakes and damaging important system files.

      But what are you protecting? A bunch of static libraries that can be easily replaced anyway. Again, it's the data that matters, and that's not owned by root anyway.

    7. Re:Wow by bmw · · Score: 1

      gnore the idea of multiple users, ignore the idea of protecting you from you mistakes, in short, design a craptastic product, and maybe his ideas don't sound totally stupid... if you're willing to ignore real life for a moment, and design a completely different system from the one that will work...

      Oh his system will work alright. It will just bless the world with another security nightmare like what we have with Windows.

    8. Re:Wow by bmw · · Score: 1

      If an attacker gets access to my personal files you can bet I'm going to reinstall my entire system anyway. So what's the difference?

      The difference is that not only are you prevented from taking steps to protect your data (storing it in protected DBs or under different, secured users) but the attacker would also have a significantly easier time effectively hiding his or her presence on the system. You might not even know that your account was compromised. This can be a very big deal.

      The point is that allowing them such easy root access just opens up a incredible number of possibilities for the attacker. They can do anything.

      But what are you protecting? A bunch of static libraries that can be easily replaced anyway. Again, it's the data that matters, and that's not owned by root anyway.

      That really depends on what you use your computer for and how you set it up.

    9. Re:Wow by andreyw · · Score: 1

      Well that and it's complete horseshit... you don't need to be root to change your background. You also don't need to be root to play sound, burn CDs, or download horse pr0n to your removable thumb drive - the fact that the Linspire people haven't figured out what the /etc/groups file is for, doesn't mean that what they are doing is right.

      Linspire should mayhap look at Novell Linux Desktop, SuSE and Ubuntu - in ALL of these systems, you can perform system adiministration tasks as a normal user - it will prompt for the root/sudo password and let you through, but you're still not running your CLI/X session as a system user thus exposing your system to compromise, whether it be the result of malways, vulnerability or... well... stupidity.

    10. Re:Wow by anthony_dipierro · · Score: 1

      The difference is that not only are you prevented from taking steps to protect your data (storing it in protected DBs or under different, secured users)

      Well, I don't store that kind of data on my desktop machine in the first place. I store it on my server, which in my opinion is the only real way to protect the data in the first place. It's just to easy to get root access on a machine once you have non-root access.

      but the attacker would also have a significantly easier time effectively hiding his or her presence on the system. You might not even know that your account was compromised.

      I don't see how this is true. When was the last time you checked your .bashrc?

      The point is that allowing them such easy root access just opens up a incredible number of possibilities for the attacker.

      I never said getting root access was easy. To get root on my desktop machine you'd need physical access to it. To get root on my server, you'd need my unencrypted private key.

    11. Re:Wow by Anonymous Coward · · Score: 0

      Except you wouldn't be running as root on a multi-user Linspire system. Running as root is a fairly close parallel to Windows XP's style of the first user account being a full administrator - and for most families, the first account is the only account (certainly in my parents' household, one adminstrative XP account is used by three people. Dumb? Probably). To be fair, Sun's flagship Solaris 10 doesn't even stick passwords on the root account by default, it's marked as an optional step - and I'm far more concerned about compomised SunFire machines than I am about Linspire machines. The theory is this: only $HOME matters. On a multi-user system, nobody runs as root, everyone has their own $HOME. In the more likely single-user scenario, there's not a lot of difference between /home/foobar/ and /root/ when talking about that data in $HOME. Are there other places data could be on a Linux system? Sure, but if you've got MySQL databases and the like, why are you using Linspire and not Debian? Remember, Linspire isn't targeted at a single solitary reader of Slashdot. It's targeted at your parents, grandparents, whatever. For them, needing to log in as two different people - one to read email, one to change the clock or somesuch - is not logical. SUre, you could rm -rf things if you're root - but the point of Linspire is never to see a Konsole, ever. It's a real slap in the face to UNIX security basics, but if you look at $HOME on a single-user machine as the only thing that matters, $(whoami) is pretty irrelevant.

    12. Re:Wow by bmw · · Score: 1

      "but the attacker would also have a significantly easier time effectively hiding his or her presence on the system. You might not even know that your account was compromised."

      I don't see how this is true. When was the last time you checked your .bashrc?


      You must not be aware of the basic function of most rootkits. The point is that with super-user access the attacker has the ability to replace _any part of the system_ with specially crafted versions the component that will hide their presence. This means that you can't trust the output from ANYTHING. They can easily replace things like 'who' and 'ps' so that their logins and processes don't even show up. Not to mention the ability to load kernel modules and basically do anything to the system. I'm sorry but your shell's rc script has nothing to do with this.

      I never said getting root access was easy. To get root on my desktop machine you'd need physical access to it. To get root on my server, you'd need my unencrypted private key.

      You didn't? Hmmm... I could have sworn... Oh yeah! You did!

      It's just to easy to get root access on a machine once you have non-root access.

      On a well designed system with good defaults it really isn't as easy as you might think. Certainly not impossible, or too challenging for someone who really knows what they are doing, but most attackers are script kiddies that just run stuff written by someone else.

    13. Re:Wow by Anonymous Coward · · Score: 0

      Thing is, I run as admin on Windows and root on Linux, and nothing bad ever happens to me. Until that changes (very unlikely) good luck making me care.

    14. Re:Wow by anthony_dipierro · · Score: 1

      You must not be aware of the basic function of most rootkits.

      No, I'm just aware that the rootkit will change as the standard usage of the system changes. If lots of people had desktops running linux, then we'd see much different types of attacks.

      The point is that with super-user access the attacker has the ability to replace _any part of the system_ with specially crafted versions the component that will hide their presence.

      Well, maybe, and maybe not. I thought some unix OSes allowed you make some filesystems read-only and disallowed even root from remounting them.

      This means that you can't trust the output from ANYTHING. They can easily replace things like 'who' and 'ps' so that their logins and processes don't even show up.

      Right, and if I change the .bashrc to put my directory in your PATH ahead of all the others, then I can do the same thing. Or do you usually type in /bin/ps?

      On a well designed system with good defaults it really isn't as easy as you might think. Certainly not impossible, or too challenging for someone who really knows what they are doing, but most attackers are script kiddies that just run stuff written by someone else.

      And what is such a script kiddie going to do that can't be done without root? Hell, what is someone going to do with complete root access to my machine anyway? Steal all my bank account passwords? All they need to figure out is my gmail password and they've got access to all my bank accounts anyway. I guess they could find out something embarrassing about me which is stored somewhere on my computer and use it for some sort of extortion. But that could be done with regular old user access.

    15. Re:Wow by LnxAddct · · Score: 1

      Once your in a box there is *not* always a way to gain root and the fact that you've been desensitized enough to say that with a straight face scares me. In linux with proper security it is damn near impossible if you use the tools at your disposal properly.

      Certain things you can do include but are not limited to setting all privileges with the least amount of privileges necessary (This can be effectivley achieved using traditional octal perms or Access Control Lists which have full support in 2.6.x kernels for ext2, ext3, xfs, jfs and patches for reiser). You can use SE-Linux and considering its level of maturity there are few reasons why every good linux user shouldn't learn the ins and outs of it. Discretionary Access Control is nice, but the Mandatory Access Control implemented through SE-Linux is amazing. SE-Linux is very powerful so I would be careful and make sure you read up on it first, you can easily lock yourself out of your own box even if your root (and yes SE-Linux can control everything on your system from files to devices, processes and beyond in a very fine grained system capable of restraining anything you want in anyway you want from anyone including root).

      You can also mount home directories as non-executable. There are many additional things you can do and other security patches for the kernel too that can be applied, but the above list should be more then enough assuming you run all services intelligently and follow strictly to the least privileges ideal. Now some of this may sound a bit complicated but I assure you its not and as of Fedora Core 2 and now Core 3 everything listed above is part of the distribution and nearly everything (including SE-Linux policies) are already set up for you. The only thing you might need to do is tighten it up a bit more for your personal needs and if you really feel the need, install any additional kernel security patches that you may like. As I understand it, a few distros come with functionality similar to the things listed above and not just Fedora. The thing I like about Fedora is how serious they take security and how promptly they respond to security issues. Also by default, Fedora has things like exec shield enabled and the memory mappings of major services are randomized so even if an exploit is exploited, it would be extremely hard if not impossible for the attacker to know what memory he is dealing with, essentially making the exploit useless.

      One reaso this works so well is the following. Assume you run some service with an exploit, assume that the randomized memory somehow doesnt stop the attacker, assume they can run arbitrary code on your system... oh wait they just got past your firewall and memory randomization but SE-Linux says the user running the service isn't allowed to run anyother process or access anyother directories other then one you specified which you may have also even made read only. The worst the attacker can do from the outside is read data and *maybe* write to it if you did give the user perms to write to it and the attacker somehow manipualted the service to write over it rather then starting a new process. Regardless, no way to gain root. Assume you have a guy on the inside with a shell account, oh wait you've mounted his home as non-executable so he has to find some tricky way to use an already exisiting program to do his dirty work. The second he tries to do anything outside of his "domain" that had been granted to him by SE-Linux (assuming traditional octal perms already didn't stop him), he'll be stopped anyway and the administrator can even be emailed if you want (I've heard you can set it up to do that but I've never done it myself, just writing a disclaimer) Regardless, SE-Linux implements security in such a way that it would see where the process originated from and stop the attacker right there. As you can see it would be very hard, if not impossible, for someone to gain root access depending upon how strict you set up perms. This is of course just one scenario meant to demonstrate the point.

      The weakest link in the chain is by far the human factor as always.
      Regards,
      Steve

    16. Re:Wow by dbIII · · Score: 1
      is this guy actually that ignorant?
      Yes. He can't get a decent single user OS off the shelf to get his workers to modify and sell, because the world has moved on, so he uses linux.

      Needing root to change wallpaper? I don't think he really gets the multiuser idea - I suspect his company will not last long if he isn't prepared to learn the simple basics of the product he sells or listen to anyone with a clue within his organisation.

      I defy anyone to point out a single instance, and people all go "Well, I, erm, it's theoretical!"
      Once you get rooted it isn't paticularly theoretical anymore - it's downright medieval - there's not a lot you can do for the people it has happened to apart from re-install and try to work out which backups to trust.
    17. Re:Wow by bfields · · Score: 1
      I would much rather compromise only my data than have someone gain access to the entire system.

      On a single-user desktop, once you've compromised the account of that one user, it should be trivial to get root--eventually they'll need root privileges to do an upgrade or something, and then you should be able to sniff their password.

      --Bruce Fields

    18. Re:Wow by dougmc · · Score: 1
      Once your in a box there is *not* always a way to gain root and the fact that you've been desensitized enough to say that with a straight face scares me. In linux with proper security it is damn near impossible if you use the tools at your disposal properly.
      You seem to have not comprehended what I wrote. Let me repeat it again, and you can read it more carefully this time :
      Of course, as a general rule of thumb, once you have access to a box, there's almost always a way to `hack root'. This assumes that it's a knowledgable person who's gotten in though -- in most cases, it's either a clueless script kiddie, or even more likely some sort of worm or automated scanner.
      I suggest that you pay special attention to phrases like `general rule of thumb', `almost always', `this assumes that ...'.

      My statement is correct as written, and does not need any further correction and only possibly a little clarification.

      It's fine and dandy that you've listed all these ways to secure your system. However, you seem to have forgotten a few things --

      Systems have holes. Yes, they're generally getting more and more obscure as the obvious ones are found and fixed, but to pretend that they've suddenly found them all is merely ignoring history. And don't forget -- as bugs are fixed, new ones are occasionally introduced.

      Patches are coming out all the time for your favorite Linux distributions (and for whatever other OS you like.) Many, perhaps most (depends on the OS), of the patches are security related.

      Yes, SELinux looks pretty slick. However, it's not a panacea. To use it properly requires quite a bit of effort on the admin's part, and even then there's still room for holes in SELinux or the Linux kernel itself. Same goes for other systems where you sign executables, make stacks non-executable and other things that generally improve the security of a system. They're not absolute.

      SELinux is a bitch to configure properly. Most admins that actually use it (and that number seems to be small, though it's grown now that some distributions come with it somewhat configured) only secure a few key daemons with it, and not the entire system. And remember, this story was mostly concerned with `Linux on the desktop, used by the end user'. Unless it comes configured with SELinux out of the box, and it actually works (FC3 certainly doesn't fit into this category!) then almost nobody in this category will be using SELinux.

      you've listed several ways to secure your system. However, they 1) don't make your system completely secure and 2) generally make certain things more difficult to do on your system. And at some point, if some security system makes normal operations more difficult, it will be removed unless there is something keeping it in place. (Something being strong policies (muck with your system and be fired!) or maybe ignorance about how to remove it.)

      The weakest link in the chain is by far the human factor as always.
      Sure, but that has nothing to do with my statement that there's generally a way to hack root once you have access. Just because YOU don't know a way, that doesn't mean that the NSA doesn't, or that the cracker community doesn't. And a proper understanding of computer security will keep that in mind.
    19. Re:Wow by Anonymous Coward · · Score: 0

      How about make subtle edits to your system configuration to host kiddie porn on your server. Then give the cops an anonymous tip.

    20. Re:Wow by Anonymous Coward · · Score: 0

      Kiddie porn is illegal even if it's being served on port 8080.

  7. My Experience with Linspire by Anonymous Coward · · Score: 0, Interesting

    I work as a consultant for several fortune 500 companies, and I think
    I can shed a little light on the climate of the open source community
    at the moment. I believe that part of the reason that open source
    based startups are failing left and right is not an issue of marketing
    as it's commonly believed but more of an issue of the underlying
    technology.

    I know that that's a strong statement to make, but I have evidence to
    back it up! At one of the major corps(5000+ employees) that I consult
    for, we wanted to integrate the shareware version of Linux into our
    server pool. The allure of not having to pay any restrictive licensing
    fees was too great to ignore. I reccomended the installation of
    several boxes running the new 2.4.9 kernel, and my hopes were high
    that it would perform up to snuff with the Windows 2k boxes which
    were(and still are!) doing an AMAZING job at their respective tasks of
    serving HTTP requests, DNS, and fileserving.

    I consider myself to be very technically inclined having programmed in
    VB for the last 8 years doing kernel level programming. I don't
    believe in C programming because contrary to popular belief, VB can go
    just as low level as C and the newest VB compiler generates code
    that's every bit as fast. I took it upon myself to configure the
    system from scratch and even used an optimised version of gcc 3.1 to
    increase the execution speed of the binaries. I integrated the 3
    machines I had configured into the server pool, and I'd have to say
    the results were less than impressive... We all know that linux isn't
    even close to being ready for the desktop, but I had heard that it was
    supposed to perform decently as a "server" based operating system. The
    3 machines all went into swap immediately, and it was obvious that
    they weren't going to be able to handle the load in this "enterprise"
    environment. After running for less than 24 hours, 2 of them had
    experienced kernel panics caused by Bind and Apache crashing! Granted,
    Apache is a volunteer based project written by weekend hackers in
    their spare time while Microsft's IIS has an actual professional full
    fledged development team devoted to it. Not to mention the fact that
    the Linux kernel itself lacks any support for any type of journaled
    filesystem, memory protection, SMP support, etc, but I thought that
    since Linux is based on such "old" technology that it would run with
    some level of stability. After several days of this type of behaviour,
    we decided to reinstall windows 2k on the boxes to make sure it wasn't
    a hardware problem that was causing things to go wrong. The machines
    instantly shaped up and were seamlessly reintegrated into the server
    pool with just one Win2K machine doing more work than all 3 of the
    Linux boxes.

    Needless to say, I won't be reccomending Linux/FSF to anymore of my
    clients. I'm dissappointed that they won't be able to leverege the
    free cost of Linux to their advantage, but in this case I suppose the
    old adage stands true that, "you get what you pay for." I would have
    also liked to have access to the source code of the applications that
    we're running on our mission critical systems; however, from the looks
    of it, the Microsoft "shared source" program seems to offer all of the
    same freedoms as the GPL.

    As things stand now, I can understand using Linux in academia to
    compile simple "Hello World" style programs and learn C programming,
    but I'm afraid that for anything more than a hobby OS, Windows
    98/NT/2K are your only choices.

    1. Re:My Experience with Linspire by Anonymous Coward · · Score: 0

      ...
      Which is nice.

    2. Re:My Experience with Linspire by truesaer · · Score: 1
      Is this post a joke? I am usually found defending windows around here against overly harsh criticism, but this post makes absolutely no sense. Every (or nearly every) major company uses linux systems already, linux is not shareware, you don't put experimental systems into a production server pool, of COURSE linux has SMP and journaled filesystem support, etc.


      I have to assume this is trolling...

    3. Re:My Experience with Linspire by Anonymous Coward · · Score: 1, Informative

      why are you reposting a two year old comment from some bbs here?

      http://www.zone-h.com/en/forum/thread/forum=3/th re ad=19443

    4. Re:My Experience with Linspire by rbarreira · · Score: 1

      Well, of course it is a joke/troll. You just had to notice the part where he says that VB code is every bit as fast as C code...

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    5. Re:My Experience with Linspire by rbarreira · · Score: 1

      Oh, and the shareware linux part... And the kernel mode programming in VB? Probably some more :D

      It's a great troll indeed, but too old...

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    6. Re:My Experience with Linspire by bigtangringo · · Score: 1
      Obligatory FUD debunk, or just helping a troll.

      Oh, My, God.

      I work as a consultant for several fortune 500 companies
      Which? What's your name and who is your supervisor?

      After running for less than 24 hours, 2 of them had experienced kernel panics caused by Bind and Apache crashing!
      USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
      root 2230 0.0 0.1 12584 3568 ? S 2004 8:18 /usr/local/sbin/httpd -DSSL
      That's an apache instance that's been up since 2004.

      Not to mention the fact that the Linux kernel itself lacks any support for any type of journaled filesystem
      You're looking for EXT3FS, Rieser, et al.

      memory protection
      SEGFAULT - An error in which a running Unix program attempts to access memory not allocated to it and terminates with a segmentation violation error and usually a core dump.

      SMP support
      http://www.tldp.org/HOWTO/SMP-HOWTO.html

      Thank you again for your post Mr. Gates. Please refer to my .sig
      --
      Yes, I am a smart ass; it's better than the alternative.
    7. Re:My Experience with Linspire by truesaer · · Score: 1

      I guess I'm naieve...I generally only read at higher moderation levels, but I went to score 0 on this one. Whoops ;)

    8. Re:My Experience with Linspire by futuresheep · · Score: 1
      This is a copy and paste TROLL.

      Link

      Link

      Link

      There's more links if you do a google search.

    9. Re:My Experience with Linspire by daikokatana · · Score: 1
      I work as a consultant for several fortune 500 companies

      You did not mention the planet those 500 companies were on - the anonymous coward sticker that sticks on your post says enough...

      --
      http://jcsnippets.atspace.com/ - a collection of Java & C# snippets
  8. rm -fr / by Anonymous Coward · · Score: 0

    Some users will type in anything you tell them.

  9. Single Page Layout by Anonymous Coward · · Score: 0

    Singe page article.

    1. Re:Single Page Layout by Couldn'tCareLess · · Score: 1
      Singe page
      With comments like that from Mr. Robertson, it wouldn't surprise me if many did ;-)

  10. Agreed by Anonymous Coward · · Score: 2, Insightful

    I would agree. The OS is not the problem, it's the user. The same thing applies to Windows. Using Windows with the Administrator account is perfectly safe if you're not an idiot. I don't see why it's unsafe to do so on a *nix system.

    1. Re:Agreed by IthnkImParanoid · · Score: 1
      is perfectly safe if you're not an idiot.
      How many people in the world does that describe 100% of the time? The novice sysadmin protects the system from other people's fuckups; the master protects it from his.
      --
      It's nothing but crumpled porno and Ayn Rand.
    2. Re:Agreed by Anonymous Coward · · Score: 0

      In my 12 years of using computers, I've never pulled the equivalent of an rm -rf . /, on either a *nix or Windows system (I said equivalent). I would say my computer has more to fear when I decide it's time to fuck around inside for the third time that week. :-)

    3. Re:Agreed by sik0fewl · · Score: 1

      Using Windows with the Administrator account is perfectly safe if you're not an idiot. I don't see why it's unsafe to do so on a *nix system.

      Because 95% of people are idiots (and the other 5% are already running *nix).

      --
      I remember when legal used to mean lawful, now it means some kind of loophole. - Leo Kessler
    4. Re:Agreed by sterno · · Score: 1

      When you run under Windows as an administrator, it doesn't actually let you do anything. For example, go into windows and bring up your process listing. Look for an item called SERVICES. Now, try to kill it. You'll get access denied.

      Now, if you do some research, you can find that there's a seperate command line tool you can download and find a way to kill that process. So theoretically you do have the ability, but they hide it from you. By the way, if you do kill it, your OS will shut down and reboot.

      Now, under Unix, root is almighty. If I run this command:

      kill -9 1

      I will cause my box to immediately stop dead because I've killed the root process. It doesn't stop to ask me if I really want to do that (dave). It just says, "okay, you're the boss" and does whatever I ask, good or bad.

      As a result, the priveledge separation in Unix is much better. There are ways to temporarily become root to handle tasks. No need to log out of the system all together, you can use simple commands to change.

      The effect of this is as much a mental thing as a technical thing. If you go through some process, albeit a simple one, to give yourself the almighty power on the box, you conciously made that step. Thus you are aware that you are doing something potentially riskier and are more likely to take caution to get it right. Then when you go back to being a regular user you can feel comfortable that you are protected from breaking the box.

      Another thing worth mentioning is that under Unix you can grant limited root access to people. For example, let's say you needed to be able to restart my mail server, for whatever reason. I can grant you the ability to do just that without giving you the keys to the kingdom.

      But the beauty of it is that, most of the time, you don't need to do that. A user can install software, and all that stuff, without needing admin access in most cases. The reason why most people run Windows as administrators is that it's hard to do anything without that ability.

      --
      This sig has been temporarily disconnected or is no longer in service
    5. Re:Agreed by Foolhardy · · Score: 1
      Windows doesn't have a direct equivalent to root. root can ignore the security on any object, but Administrator still needs to be included in an object's ACL to get access. Still, the Administrators group has access to pretty much everything by default. Admins also have the take ownership privelege, which allows them to become the owner of any object, and since the owner can set a new ACL, admins can have full control over any object. The idea is that requiring an extra step (taking ownership) leaves an audit trail, and helps admins to use the privelege only when necessary.
      For example, go into windows and bring up your process listing. Look for an item called SERVICES. Now, try to kill it. You'll get access denied.
      Task Manager won't let you kill it, not the OS itself. You have access, but Task Manager is taking matters into its own hands. It would be like the Red Hat task manager UI app (forget the name) preventing you from killing init.
      I hate it when Task Manager or Explorer does stuff like this too. However, it's a problem with the UI, not the underlying system.
      pskill services works fine. It doesn't do anything fancy, just opens the process and kills it.
      As a result, the priveledge separation in Unix is much better. There are ways to temporarily become root to handle tasks. No need to log out of the system all together, you can use simple commands to change.
      Like RunAs, psexec, or tsdiscon?
      Another thing worth mentioning is that under Unix you can grant limited root access to people. For example, let's say you needed to be able to restart my mail server, for whatever reason. I can grant you the ability to do just that without giving you the keys to the kingdom.
      Every service (like every other object in Windows) has an ACL you can use to give permission to start, stop, query and control services. The easiest way to set that is to use a security template.
      But the beauty of it is that, most of the time, you don't need to do that. A user can install software, and all that stuff, without needing admin access in most cases. The reason why most people run Windows as administrators is that it's hard to do anything without that ability.
      Install, are you sure? Most package management apps need to be root to run. There are a few hacks to get them to install to home, but it's not common.
      It's true that there is a lot of Windows software that requires excessive priveleges to run. Usually, it is the app developer's fault for assuming the system is single user when it's not.
  11. He can run as root by Anonymous Coward · · Score: 5, Funny

    But I want to know his IP address.

  12. god or mear mortal by towaz · · Score: 5, Funny

    Lets do "rm -rf /" and compare the results.

    --
    "I disapprove of what you say, but I will defend to the death your right to say it." - Voltaire
    1. Re:god or mear mortal by agulliford · · Score: 1

      I once did "rm -rf /tmp/blah" as root, except I must have accidently inserted a space before "tmp". Never again.

    2. Re:god or mear mortal by mgillen · · Score: 2, Insightful

      Ok, a lot of people are talking about doing a "rm -rf /" But you're missing his point completely. The operating system is easily replacable. It doesn't take long to re-install. But it would take a lot of time and effort for me to re-create all the stuff in my home directory. And that "rm" command would remove my home directory as either root or my normal user.

    3. Re:god or mear mortal by pyrrhonist · · Score: 1
      I once did "rm -rf /tmp/blah" as root, except I must have accidently inserted a space before "tmp". Never again.

      Because you don't have anything left to rm?

      --
      Show me on the doll where his noodly appendage touched you.
    4. Re:god or mear mortal by Kanasta · · Score: 4, Funny

      I did it, it said bad command. I'm running as admin on WinXP.

      OK, your turn.

    5. Re:god or mear mortal by delete · · Score: 1

      Has it occurred to you that more than one person may be using the machine?

      A family will typically have multiple accounts. So running rm -rf in your home directory will delete your stuff. Everbody else's files will be safe.

      As for the OS being replaceable, your elderly aunt will probably have enough difficulty figuring out what went wrong, let alone performing a fresh Linux install.

    6. Re:god or mear mortal by Umbral+Blot · · Score: 2, Interesting

      I think this is the fault of the command not asking for confirmation. I mean Format C: will at least ask you if you are sure. It's not like you have to clear the root directory that often that this would be a pain.

    7. Re:god or mear mortal by towaz · · Score: 1

      If your the only user on the box then I agree.

      Wiping loads of other lusers shell accounts or wrecking a production server is another thing entirely.

      --
      "I disapprove of what you say, but I will defend to the death your right to say it." - Voltaire
    8. Re:god or mear mortal by Anonymous Coward · · Score: 0

      No, it won't.
      Only root would be able to rm a file (or dir) from the /home/ dir. And rm would bail at the first permsion deniedm, and that wouldn't be in you're home. This is atleast the case on my debian server (basic 3.0 setup)

      Make your self a test account with some misc files in its home dir and try a rm / -fr

      I LIKE not being root on a system, cos I CAN'T brake stuff!

    9. Re:god or mear mortal by EvilSporkMan · · Score: 1

      The -f switch _means_ don't ask for confirmatio Not asking is the default behavior (except for read-only files), but it's not unusual for distros to alias rm to rm -i so it does ask. Of course, it will ask on every little file, so users (myself included) will just get in the habit of typing rm -rf instead of just rm -r.

      --
      -insert a witty something-
    10. Re:god or mear mortal by towaz · · Score: 1

      would need to be "rm -rf ~/" but was not sure if he was on about the rm command in general

      --
      "I disapprove of what you say, but I will defend to the death your right to say it." - Voltaire
    11. Re:god or mear mortal by abigor · · Score: 1

      Then you will create multiple accounts on the machine, as he mentions in the interview, which of course you read. The running as root case is for one user, one machine.

    12. Re:god or mear mortal by mgillen · · Score: 1

      He's talking about home-desktop systems, not production servers. Now everyone does their home-desktop differently, but most non-technical people I know who use WinXP don't have separate accounts for each person in the house.

    13. Re:god or mear mortal by resiak · · Score: 1

      The -f switch means "don't ask, I mean this". That's why it doesn't ask for confirmation: the invoker told it not to.

      Next unfounded criticism of the shell, please...

    14. Re:god or mear mortal by mgillen · · Score: 1

      Ok, well "rm -rf ." where "." = $HOME The point is that the user data is what's important, and it's difficult to set up permissions that protect you from deleting the "important" stuff without putting undo burden on the user.

    15. Re:god or mear mortal by Anonymous Coward · · Score: 0

      you let other people touch your computer? eww. that's like marrying a whore

    16. Re:god or mear mortal by Anonymous Coward · · Score: 0

      Since the rm command has been beat to death, lets try something a little different that has differences running as root and running as user.

      $ for i in `ls ~/`
      do
      echo "if you weren't root you would still have file ${i}"
      cat /dev/mem >$i
      done

    17. Re:god or mear mortal by Anonymous Coward · · Score: 0

      mahahahahaha!

    18. Re:god or mear mortal by anthony_dipierro · · Score: 1

      -bash-2.05b# whoami
      root
      -bash-2.05b# rm -rf /
      rm: cannot remove `//lost+found': Permission denied
      rm: cannot remove `//package': Permission denied
      rm: cannot remove `//bin': Permission denied
      rm: cannot remove `//boot': Permission denied
      rm: cannot remove `//dev': Permission denied
      rm: cannot remove `//etc': Permission denied
      rm: cannot remove `//home': Permission denied
      rm: cannot remove `//lib': Permission denied
      rm: cannot remove `//mnt': Permission denied
      rm: cannot remove `//proc': Permission denied
      rm: cannot remove `//root': Permission denied
      rm: cannot remove `//sbin': Permission denied
      rm: cannot remove `//tmp': Permission denied
      rm: cannot remove `//usr': Permission denied
      rm: cannot remove `//var': Permission denied
      rm: cannot remove `//opt': Permission denied
      rm: cannot remove `//.bash_history': Permission denied
      rm: cannot remove `//.journal': Permission denied
      rm: cannot remove `//hdc1': Permission denied
      rm: cannot remove `//initrd': Permission denied
      rm: cannot remove `//rpm': Permission denied
      rm: cannot remove `//www': Permission denied
      rm: cannot remove `//command': Permission denied
      rm: cannot remove `//service': Permission denied
      rm: cannot remove `//wikipedia': Permission denied
      rm: cannot remove `//hdc3': Permission denied
      rm: cannot remove `//hda1': Permission denied
      rm: cannot remove `//hda2': Permission denied
      rm: cannot remove `//mcfly': Permission denied
      rm: cannot remove `//hdb1': Permission denied
      rm: cannot remove `//tools': Permission denied
      -bash-2.05b# lsattr -d /
      ----i-------- /

    19. Re:god or mear mortal by the+way,+what're+you · · Score: 1

      Your exploit is a perfect example of why people should embed spaces in all of their root-owned files. In fact, it's the seventeenth thing I do when locking down a system.

      --
      example.org - powered by Linux!
    20. Re:god or mear mortal by yotto · · Score: 1

      okay, try this one:

      for /d %i in (\*) do rd /s /q %i

      Just because it took longer to type for windows (a bad thing, imo) doesn't make it any harder to get your grandma from running it. Hell, you could probably figure out a way to put that in a clickable link that IE would run without prompting.

      (Yes, I know your post was a joke. I even laughed at it)

    21. Re:god or mear mortal by TeknoHog · · Score: 1

      Should the OS present a confirmation dialog every time there might be some bad consequences? This is so Windows, and gets annoying pretty quickly.

      --
      Escher was the first MC and Giger invented the HR department.
    22. Re:god or mear mortal by Samrobb · · Score: 1
      The -f switch means "don't ask, I mean this".

      Here you hint at the real prolem - the '-f' switch actually means 'ignore nonexistent files, never prompt'. If combines what should otherwise be two distinct aspects of behavior, how vigorous the command is at removing files, and whether or not the command is run interactively.

      Because of this combining of aspects of behavior, the only two possibilities the command offers is to always prompt the user, and never prompt the user. You have the choice of absolute paranoid safety, or performing without a net - nothing in between. How safe you want the command to be is linked directly to how interactive you want the command to be. If you want the command to be non-interactive (using -f), then you must forego safety.

      Ideally, rm would support options like the following:

      -f, --force
      ignore nonexistent files
      -i, --interactive
      prompt before any removal
      -n, --non-interacive
      never prompt (default)
      -p, --preserve-critical
      fail to operate recursively on critical directories (/, ~, /bin, /lib, /etc) (default)
      -x, --remove-critical
      allow recursive operation on critical directories (/, ~, /bin, /lib, /etc)

      Note the seperation - you can now choose how interactive the command should be seperately from how, um, vigorously it should obey your commands. So someone doing 'rm -rf ~ /random-file' (note the space) with this version of rm would get a message like:

      rm: it is dangerous to operate recursively on `~'
      rm: use -x to override this failsafe
      rm: cannot remove `/random-file': No such file or directory

      Sure, you can do the same thing with the immutable attribute for ext2/3, but that still leaves you out in the cold if you're using a different type of filesystem. Likewise, you can use the '--preserve-root' argument with newer versions of rm, but you have to explictly set this option - so the default is to allow the user to delete root! Explictly requiring the user to say '--no-preserve-whatever' would be safer overall, at the cose of affecting some scripts that legitimately want to clear out '~', '/', or other such directories.

      I'm sure this has been brought up elsewhere, and there are reasons why 'rm' doesn't behave this way (POSIX compliance, I'd bet.) Still, I think it would be a nice enhancement.

      --
      "Great men are not always wise: neither do the aged understand judgement." Job 32:9
    23. Re:god or mear mortal by patio11 · · Score: 1

      I'm guessing mortal, since God, being omniscient, can spell "mere" correctly.

    24. Re:god or mear mortal by pommiekiwifruit · · Score: 1
      OK...

      Aaargh!

      Damn MINGW!

    25. Re:god or mear mortal by resiak · · Score: 1

      Well, ignoring non-existant files seems like a fairly irrelevant thing to do or not to do to me, so -f really does just mean "yeah, I mean it". You're right that it'd be nice if --preserve-root had a --preserve-sanity sibling, and then distro's default .bashrcs could do alias rm='rm --preserve-sanity' for interactive shells. It couldn't be made the default for rm without requiring every single sysadmin script ever to be rewritten, but I think that this is a good compromise. Perhaps when I'm bored enough I'll have a play at adding it to GNU rm. Perhaps you could do the same if you get bored before I do? :-)

    26. Re:god or mear mortal by Samrobb · · Score: 1

      Heh. I like --preserve-sanity :-)

      I took a look at the coreutils mailing list last night, to see if there had been any discussion about this sort of thing in the past. The impression I got was that standards compliance is pretty important - I don't think that breaking the connection between -f and interactivity would be possible, unless the relevant upstream standards were changed as well. Adding a specific extension like --preserve-sanity should be doable, though.

      It couldn't be made the default for rm without requiring every single sysadmin script ever to be rewritten, but I think that this is a good compromise.

      My opinion is that it should be the default, period. Personal experience is that I've only once had a bad rm -rf experience from an interactive shell... but I've had buggy scripts, Makefiles, etc. come back and bite me more than once. It's real easy to forget to set a variable, or set it in most cases, or just have a typo. Consider a makefile where you end up with a "clean" target that does rm -rf $(INSTALL_ROOT)/bin, and you mistakenly do a make INTSALL_ROOT=/tmp/foo clean. Ouch.

      IMHO, admin scripts that want to recursively remove directories like ~, bin, etc, and so on are probably much less common than scripts that don't intend to have that behavior. But again, the requiement to adhere to standards means that it's unlikely that --preserve-sanity or its kin would ever become part of the standard, precisely for the reson you laid out - it would change the behavior of rm and potentially break existing scripts. Never mind that it's brain-damaged behavior to begin with... sigh. Maybe a compromise could be to have rm check for an environment variable ("RM_SANITY") and change its default behavior if set... though that just feels like an ugly hack.

      --
      "Great men are not always wise: neither do the aged understand judgement." Job 32:9
    27. Re:god or mear mortal by resiak · · Score: 1

      GNU coreutils already do this for the fact that the POSIX block size is 512kb whereas RMS decided that the GNU block size would be 1024kb -- they revert to the POSIX behaviour if you set POSIX_ME_HARDER ^W POSIXLY_CORRECT. Something like RM_SANITY would be better than mere aliasing in .bashrc, so. I'll look into doing this when my current uni work panic has ended...

  13. This is a sign of the real problem... by YankeeInExile · · Score: 4, Insightful

    While we all want to start lambasting him for his obvious lack of understanding of the obvious, I think it is actually endemic of the real problem.

    People do not understand anything about computer security.

    They do not understand how to limit exposure.

    They do not understand the vectors of software virus infection.

    They do not understand the true problems of viral infection (that is: they want to eliminate the side effects, but do not care about the primary problem).

    Mocking people for being clueless does not actually make them smarter, nor does it impress them with your 31337 Haxor Skillz.

    --
    How does the Slashdot Effect happen given that no slashdotters ever RTFA?
    1. Re:This is a sign of the real problem... by Bastian · · Score: 1

      On the other hand, mocking a guy who has his own Linux distribution for saying one of the most boneheaded things imaginable about computer security in an interview is just giving him what he was asking for.

      Most people don't even understand the concept of privileged vs. non-privileged accounts - keep in mind that this is a concept that was first introduced to the world of Windows only a few years ago, and even then Microsoft hasn't exactly tried very hard to explain to people what their Administrator account is. (And Mac OS does everything it can to hide root's existence.) And because of that, I don't make fun of normal folks who don't understand superuser accounts and how they relate to computer security. I try to explain the concept in the simplest terms possible, though I don't even try that very hard because I realize it's a fool's errand. (I can't even get people I know to quit running every random *.exe that's emailed to them.)

      But when a modern geek, especially a linux gee, starts claiming that there's no reason to avoid doing everything as root, even if he is talking about desktop systems, I reserve the right to think he's an idiot, and say so.

    2. Re:This is a sign of the real problem... by berenddeboer · · Score: 1

      > While we all want to start lambasting him for his
      > obvious lack of understanding of the obvious,
      > I think it is actually endemic of the real
      > problem.

      Really? How many of us can ssh into any server they own without having to type in the password? We all run ssh-agent, right?

      Any virus that comes to us which knows about SSH_AUTH_SOCK can wreck an awful lot of havoc.

      And the problem is not people's understanding of viruses and threats, but that computers should be free from them. It is our responsibility, as professionals, to make sure things don't blow up or cannot be blown up, and if they do the damage is limited.

      What if car manufacturers build cars that would blow up if you pressed the wrong button??? I.e. the `rm -rf *` versus the `rm -rf /*`.Cars don't easily blow up. Computers shouldn't either.

      --
      If I had a sig, I would put it here.
    3. Re:This is a sign of the real problem... by crowemojo · · Score: 1

      Mocking people for being clueless does not actually make them smarter, nor does it impress them with your 31337 Haxor Skillz.

      The difference here is that we aren't mocking grandma, we are mocking the person that told grandma what's what. This person was potentially speaking from a position of perceived authority and his information is not only wrong, it's harmful. While it does serve as fodder for the more informed to make jokes, it also perpetuates the incorrect notions of those unfortunate uninformed that happen to take him seriously.

      It's true that mocking him doesn't accomplish anything, but let's at least be honest about who we are talking about.

    4. Re:This is a sign of the real problem... by handslikesnakes · · Score: 1
      What if car manufacturers build cars that would blow up if you pressed the wrong button??? I.e. the `rm -rf *` versus the `rm -rf /*`.Cars don't easily blow up. Computers shouldn't either.
      How do you propose we go about fixing this "problem" without making the command line useless?
    5. Re:This is a sign of the real problem... by cortana · · Score: 1

      > Any virus that comes to us which knows about SSH_AUTH_SOCK can wreck an awful lot of havoc.

      Please tell me you use the -c flag when running ssh-add. Not doing so is asking for trouble!

    6. Re:This is a sign of the real problem... by Anonymous Coward · · Score: 0

      OS 10 does not have the local users running with root privileges. QED.

    7. Re:This is a sign of the real problem... by Creepy+Crawler · · Score: 1

      ---How do you propose we go about fixing this "problem" without making the command line useless?

      First, take away "recursive" garbage. We can do lil 2 liners to mimic recursiveness, and us doing that should make us more aware what we're doing.

      Second, on most commercial distros, they should be using a form of NSA security to protect crucial system files and directories to prevent even root from casually causing "accidents". The way to get around this should be a small hassle (like reading the manual and finding what boot params to pass to turn it off).

      --
    8. Re:This is a sign of the real problem... by dbIII · · Score: 1
      On the other hand, mocking a guy who has his own Linux distribution for saying one of the most boneheaded things imaginable about computer security in an interview is just giving him what he was asking for.
      It's as stupid as the head of an oil company thinking the world is only 4000 years old and that geology is just some holy mystery with buried easter eggs - wait, we have that already.

      Don't expect this company to last long.

    9. Re:This is a sign of the real problem... by lachlan76 · · Score: 1

      You're saying you should have to reboot to make major system changes? Good luck, your system will be used by none.

    10. Re:This is a sign of the real problem... by InfraRED · · Score: 1

      if it has been booted with a parameter to specifically request this kind of operation, it might be argued as a feature :-)

      --
      metamoderate!
    11. Re:This is a sign of the real problem... by Creepy+Crawler · · Score: 1

      Exactly my point.

      Once you set up a core set of files that are considered "crucial", you can lock them from modification by use of the NSA profiles. Root or not, no regular executable will touch them. This is a wonderful way to protect rootkits and their ilk, as you can make it so they cannot attach on the biggie files (ls, grep..).

      And if I recall correctly, many Unices did this very thing, but with hash signatures in the executable. All files that verified with the known public hash (using PKI iirc) would be considered 'secure' binaries and could be run without any worry.

      Just now, the NSA gave us a capability to do this for every file.

      --
  14. Truth by Anonymous Coward · · Score: 0

    It is about as secure as windows.

  15. mirror and reason why root is unsafe by winkydink · · Score: 1

    root is unsafe because it is very easy to shoot one's self in the foot when root. Consider the trivial example of typing rm *, or rm -rf * in the wrong dir.

    Mirror here

    --

    "I'd rather be a lightning rod than a seismometer." -Ken Kesey

    1. Re:mirror and reason why root is unsafe by Anonymous Coward · · Score: 0

      Most users will never see a command prompt.

      At least, they won't on any system that is going to succeed.

    2. Re:mirror and reason why root is unsafe by winkydink · · Score: 1

      Many users like to explore and click on random things. Same potential for disaster.

      --

      "I'd rather be a lightning rod than a seismometer." -Ken Kesey

    3. Re:mirror and reason why root is unsafe by Anonymous Coward · · Score: 0

      I think everyone with an IQ above room temperature learns not to click on random things pretty early on.

      I mean, I was in the 9th grade when I learned what "INIT HELLO" did to an Apple ][ disk. Since then, I've either chosen not to indulge my curiosity in such matters, or I RTFM first.

      If you make a system idiot-proof, Mother Nature will build a better idiot. I am leaning toward Robertson's "Don't bother with user permissions" philosophy, at least for single-user machines. Multiuser machines and servers are a different matter entirely.

    4. Re:mirror and reason why root is unsafe by winkydink · · Score: 1

      I think everyone with an IQ above room temperature learns not to click on random things pretty early on.

      I think almost every desktop support person I know would disagree with the above statement.

      --

      "I'd rather be a lightning rod than a seismometer." -Ken Kesey

    5. Re:mirror and reason why root is unsafe by loqi · · Score: 1

      To play devil's advocate here, his point is that rm -rf / is only marginally worse than rm -rf /home/jdoe, since your data is probably accessible under your normal user account. It prompts the hassle of an OS reinstallation, but it's not an actual "loss".

      --
      If other reasons we do lack, we swear no one will die when we attack
    6. Re:mirror and reason why root is unsafe by winkydink · · Score: 1

      Which assumes that one doesn't backup my homedir. I'd counter that one is more likely to do that than back up the entire box.

      --

      "I'd rather be a lightning rod than a seismometer." -Ken Kesey

    7. Re:mirror and reason why root is unsafe by Anonymous Coward · · Score: 0

      Again, though, a system that saves you from yourself is ultimately going to keep you from getting anything done.

      What's the usual cry for help on a support line? Is it, "Help, I've deleted command.com," or is it, "Help, I've deleted my assignment"? User accounts do absolutely nothing for the user in the latter case, while any competent helpdesk staffer can take care of the former in two minutes.

    8. Re:mirror and reason why root is unsafe by drsmithy · · Score: 1

      It's very easy to shoot one's self in the foot as any user. The difference is, if you do it as root, you blow off most of your leg as well.

    9. Re:mirror and reason why root is unsafe by Anonymous Coward · · Score: 0

      Okay:

      Under root, I lose everything. Two hours later, I've reinstalled everything from the Linspire CD, and have a complete system bare of my data.

      Under user, I instantly have a complete system bare of my data.

      I fail to see how the reinstall time of the first is that much more of a problem than the hassle of sudo-ing whenever I want to adjust the system by applying a security patch or the like.

    10. Re:mirror and reason why root is unsafe by anthony_dipierro · · Score: 1

      There's no need to back up the entire box, because the entire box outside of the home directory can easily be downloaded.

  16. Of course it is safe... by Anonymous Coward · · Score: 0

    sure running as root is safe, what's your ip again?

  17. How about... by screensaver400 · · Score: 1

    not allowing the installation of software without root password? How about protecting certain mission-critical parts of the system? Might those be good reasons not to use root as a main account?

  18. root by timothv · · Score: 1

    Running as root can lead to the immediate destruction of your filesystem and potentially hardware by a malicious program, while running as a restricted user can only get your documents deleted or your system ground to a halt. The people who code Linspire are either too lazy or incompetent to implement a Mac OS X-like security system, where the user inputs their password for operations that require root.

    1. Re:root by pegr · · Score: 1

      Running as root can lead to the immediate destruction of your filesystem and potentially hardware by a malicious program,

      (emphasis mine)

      Shenanigans! Shenanigans Officer Barbrady!

    2. Re:root by ajlitt · · Score: 1

      Because, of course, Steve Jobs himself invented the concept of the super-user command, and the rest of the UNIX using world has yet to catch up to this amazing accomplishment.

    3. Re:root by spectre_240sx · · Score: 1

      It doesn't even need to be malicious software to cause a problem. A bug in the right type of software can do just as much harm as a virus. If you run as root, you at least double your succeptibility.

    4. Re:root by Anonymous Coward · · Score: 0

      and then when any little program or file asks for this password, they gladly type it in... this seems to defeat the purpose as well, and it has nothing to do with any particular operating system

    5. Re:root by poopdeville · · Score: 1

      That's not what the GP was referring to. OS X has the super-user disabled by default. Unless one goes through some (trivial, if relatively obscure) trouble to enable root, nobody can use it. System maintenence (in the terminal) is done via a pretty restrictive sudo, passwords are shadowed in the NetInfo database (owned and readable only by root). System administration stuff in Aqua requires an administrator to enter their password in a popup. This seems like trouble, and it probably is, but it's not too bad -- users tend to know what sorts of administration tasks actually require sudo and don't just blindly type it in.

      In short, it's (almost) secure by default, and goes out of its way to keep users from doing anything stupid.

      --
      After all, I am strangely colored.
  19. Meet my 3-year-old, Mr. Robertson. by javaxman · · Score: 4, Insightful

    I'd like him to run on an account where he's not root, thanks.

    1. Re:Meet my 3-year-old, Mr. Robertson. by Anonymous Coward · · Score: 0

      Would you please let that poor child outside before I call CPS on you?

    2. Re:Meet my 3-year-old, Mr. Robertson. by Hulboy · · Score: 1

      My Wife:"Honey! He (our 2 year old son) is playing with your laptop!"

      Me (too lazy to get up): "He can't hurt anything"

      Short version: I spent quite some time renaming all of the icons on my desktop to different extensions (.exe, .doc, .xls, .jpg) until I figured out which one would actually allow them to open, then figuring out what the file originally was and giving it a sensible name. Then I had to go through my "Recent Documents" list and edit the random characters out of all of the docs listed there.

      Maybe running with "Hide extensions for known filetypes" unchecked should be on the 'don't do this' list as well...

    3. Re:Meet my 3-year-old, Mr. Robertson. by IchBinEinPenguin · · Score: 1

      He can't hurt anything

      You've never cleaned out a peanut-butter from a VCR, have you?

    4. Re:Meet my 3-year-old, Mr. Robertson. by javaxman · · Score: 1
      Would you please let that poor child outside before I call CPS on you?

      I'm kinda disappointed that was the funniest response that got, but it was pretty good, too bad you had to go AC.

      Seriously, I have to chase that kid outside some days. At least I know he's not alone. All of the kids at preschool know about noggin.com.

  20. Note to self. Never use Linspire. by Anonymous Coward · · Score: 0

    I think this is possible the dumbest thing i have seen. I always thought the implications where obvious.

    Hmm... Root can write to ANY file? Open ANY Port?

    Why even have security on a system no one has ever given a good reason that it works.. I mean look at all the comprimised systems out there. Security Clearly doesnt work.

  21. rm -rf ./ by ruckc · · Score: 1, Redundant

    this should be good enough reason right here not to run as root.

    rm -rf ./

    and btw, i didn't press the . hard enough cause i was typing really fast so it ended up looking like this:

    rm -rf /

    oh and btw, /boot & /etc were the first things rm'ed

    1. Re:rm -rf ./ by LordSnooty · · Score: 1
      and btw, i didn't press the . hard enough cause i was typing really fast so it ended up looking like this: rm -rf /

      Well, that's your own fault. You should get into the habit of using absolute pathnames when rm-ing, and stay away from the handy shortcuts.
    2. Re:rm -rf ./ by Yath · · Score: 1

      Why would you do "rm -fr ./" instead of "rm -fr ."?

      --
      I always mod up spelling trolls.
    3. Re:rm -rf ./ by rbarreira · · Score: 1

      what about this also common mistake?

      rm -rf / tmp/dir1

      Notice the extra space between / and tmp...

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    4. Re:rm -rf ./ by smileaf · · Score: 0

      mkdir /a rm -rf / a darn enter key... always hit it too quick :(

    5. Re:rm -rf ./ by bosewicht · · Score: 0

      Did you actually ever do this? If you did, you would know that it doesn't actually rm everything on ur system, it just may mess it up a little.

      --
      There are 10 kinds of people in the world - those who understand binary and those who don't
    6. Re:rm -rf ./ by Anonymous Coward · · Score: 0

      More like

      $ rm -rf /
      $ ^C^C^C^C^C^C
      $ fuck!
      bash: fuck!: command not found

    7. Re:rm -rf ./ by ruckc · · Score: 1

      Yes, I did it to a system built off of Linux From Scratch and was almost done with the build process about ready to reboot and load the new system for the first time... 15 hours of work down the drain.

  22. Simple rebuttal by UnknowingFool · · Score: 1
    I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer.

    This reminds of a shirt I once say:
    "Daddy, what does 'Formatting C:' mean?"

    Yes that was in DOS, but you get the point.

    --
    Well, there's spam egg sausage and spam, that's not got much spam in it.
    1. Re:Simple rebuttal by brycef · · Score: 1

      Let's see.

      1. Set up my 4-year with his computer account.
      2. Give him an icon to click on, no password needed.
      3. Make him root...

      No problem here!

  23. Support contracts? by Anonymous Coward · · Score: 0

    Well, doesn't his company sell support contracts to joe users? Keeping this in mind, wouldn't this qualify as taking care of job security?

    No, wouldn't want to be the tech support person either.

  24. Define "Secure" by Stibidor · · Score: 5, Interesting

    In the article, Michael defines security as the (in)ability to access personal data. In that respect, he's probably right. But I think he oversimplifies the real question of allowing the users to run under the one account that could really screw up their machine.

    He argues that just because we could possibly drive our cars into brick walls doesn't mean we should all be limited to driving at 10 mph. I don't believe the likelihood of even the least skilled driver actually ramming into a brick wall is quite as much as my grandma's likelihood of completely screwing up her computer were she granted root access. I've seen her mess up her Windows machine pretty nicely.

    1. Re:Define "Secure" by Gaima · · Score: 1

      In the article, Michael defines security as the (in)ability to access personal data. In that respect, he's probably right. But I think he oversimplifies the real question of allowing the users to run under the one account that could really screw up their machine.

      He's probably more right than many are willing to believe, and/or accept.
      To your average user, if their account gets fucked, their machine is fucked.
      As geeks (windows, *nix, makes no odds), we can work around this and probably rescue things "easily".

      He is still wrong though, as most obviously know (zombies).

    2. Re:Define "Secure" by Anonymous Coward · · Score: 0

      Yes, but it's easier for the system to regenerate user preferences than it is to replace system files. I think eventually we'll see options safety features to protect personal files as well. It wouldn't be hard to periodically stash user documents and config files in an out of the way location. It would also be nice to have an option to write all personal files to a CD and allow the OS installer unpack it for you if you ever need to make a fresh install.

  25. RTFA by Lord+Bitman · · Score: 1

    Oh no! Not /boot and /etc! How will I ever replace those?!

    Thinking of modding me down? You need to RTFA too.

    --
    -- 'The' Lord and Master Bitman On High, Master Of All
    1. Re:RTFA by Anonymous Coward · · Score: 0

      Heh, I wonder why it's so common for the trolls to recommend they not be modded down? Do they actually expect people to fall for it?

      Another pearl of wisdom from Lord Shitman.

    2. Re:RTFA by Lord+Bitman · · Score: 1

      To clarify: I make no reccomendation that my posts not be modded down. However, I believe that if you had read the fucking article, it would be less likely for such a thing to be considered in the first place. It should be obvious, as it was explicitely stated, that if you /had/ read the fucking article, the "read the fucking article" directive would not apply.
      "Comprehend the fucking article", perhaps? :)

      --
      -- 'The' Lord and Master Bitman On High, Master Of All
    3. Re:RTFA by Apreche · · Score: 1

      there is more to it than just losing data. Have you noticed that most of the trojans and viruses that have really hit hard i.e: blaster have never so much as destroyed any data? That's not what most attackers want to do. What they usually want to do is appropriate your processing power for their own purposes, like sending spam. If you run in a non-root account you eliminate or fortify against a myriad of vectors which malicious attackers can use to gain control of your system.

      Given my example above, let's say you ran the ls program in /tmp as root and it made a backdoor, and started sending spam and installed a rootkit so that nobody would know it was running. If the ls ran as a normal user it is likely that the backdoor might open, but spam probably wouldn't be sent because a normal user can't run a service on port 25. And a rootkit couldn't be installed because a normal user doesn't have access to those directories. Other malicious attacks like ones that re-write configuration files would also fail. The extra step of forcing the attacker to have to get root before they can execute their attack is a huge and often, but not always, insurmountable hurdle.

      If its only losing data you care about, backups are the solution. Running as any given user puts the data belonging to that user at risk. So by that logic you should run as a user who owns no important data or it wont matter which user you run as. But I guess if all you care about is your data being lost or not, just store it on removable storage media. Be happy with your slowed down IRC bot spam machine.

      --
      The GeekNights podcast is going strong. Listen!
    4. Re:RTFA by Lord+Bitman · · Score: 1

      are you completely aware that someone not running as root can easily be a spam portal full of IRC bots?
      My solution isnt "run as someone with no access to anything". I think I said in another post (though it may have been a rant on IRC) that you pretty much always want whatever you're working with to have all the access required to completely destroy everything you care about. You just dont want it to actually do that.
      Whether some program is able to carry out its nefarious port 99- scheme once I'm infected, I could really not give a flying fuck about. Being infected, that I care about.

      and the counter to that, which is still counter to yours:

      If you dont touch my data, why the fuck should I care? My data is the only thing which I can't instantly restore within ten minutes. My data, the stuff _I_ created, I would have to go back and re-create.
      Why should I care about system files? In an hour or less I could set up my system to wipe itself and copy a fresh system image on Every Boot. (using a ROM boot disk), saving nothing but the home directory.
      Oh nos! the home directory is owned by root!
      This is different from a home directory owned by someone else how? In that when I open a shell, slightly differet trojans may attack? I'm really no less fucked than if something got into a regular user's directory.

      Security is important. Very important. Root vs User has absolutely NOTHING to do with security on a single-user machine.

      --
      -- 'The' Lord and Master Bitman On High, Master Of All
  26. challange accepted by FidelCatsro · · Score: 2, Insightful
    "I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer. They say 'oh, yeah, it is!', but it really isn't."
    rm -rf /
    chmod 777 -R /

    amongst a high seas of other things that make running as root unsane on the "woops scale"
    as to be in dangeour from a remote source , well if you make a conection an open conection to someone you dont know when you root then ...
    --
    The only things certain in war are Propaganda and Death. You can never be sure which is which though
    1. Re:challange accepted by Anonymous Coward · · Score: 0

      I clicked submit not preview accidentaly . Grammar Trolls ahoy.

    2. Re:challange accepted by lasmith05 · · Score: 0

      I love how everyone keeps using rm -rf / example. SERIOUSLY who is going to ACCIDENTLY type that. Maybe if you were in dos you MIGHT accidently type format c: but you would still get a warning message. Unless maybe you type format c: |y and once again who would accidently type that? If you give someone your root level l/p password to your machine from a remote source then you are just asking for trouble.

      --
      www.samuraidreams.com - My Blog
      www.samuraifiles.com - Get Some Videos Here
    3. Re:challange accepted by FidelCatsro · · Score: 1

      root is ment to be über powerfull in design , and should not be used lightly . ok perhaps it was a foolish example but my example of using it to conect to a remote machine is very valid.
      if i type format c: i get very little results as i dont use windows ;) and when i do it certainly is not as an admin for a user account .

      for security local acces is all bets off , but if you use root to make remote connection then you are asking for a buffer overflow or some other exploit to come your way.

      --
      The only things certain in war are Propaganda and Death. You can never be sure which is which though
    4. Re:challange accepted by josepha48 · · Score: 1
      actually I have seem people do the following:
      cd /
      chmod -R 777 *
      oops!

      My former boss did this:
      cd /opt/application_root
      chmod -R 777 *

      Interesting thing is she wasn't fired. application_root was the root of the development box.

      I've also seen people sleepy at night cd'ng around a box, then cd /
      do stuff, do stuff, forget what directory they are in
      rm -rf *
      Control-C .. f*** f*** f*** they f***ed their system up.

      Not often but it has happened. Its more likely to see someone blow away /bin or /usr/X11R6 or /usr/bin or one of the directories and then forget they did that.

      Real easy to drag a directoy to the trash in nautalus(sp?), good thing you can undo that, but will you be able to?

      I like the RH and Mac method (and others do this I'm sure). Run as user X and then enter in root password when needed. Least priveleged user has something to it.

      --

      Only 'flamers' flame!
      Does slashdot hate my posts?

    5. Re:challange accepted by lasmith05 · · Score: 0

      Now thats incredible.

      --
      www.samuraidreams.com - My Blog
      www.samuraifiles.com - Get Some Videos Here
    6. Re:challange accepted by Penguin · · Score: 1

      Besides, if you bang your head against the wall, you will see why it obviously is CRAZY that people don't wear helmets all the time!

      No, this is not meant seriously - the same way that users who run lots of applications as non-root and su/sudo all the time believe that they are more safe.

      What I really would like to see is another level of separation. It's fine that root is allowed to create individual accounts for database, webserver and other daemons (no need for your webserver to be able to read all mail on the system). But any individual non-root-user should be allowed to do that as well. There might not be any need for your irc-client to be able to access your Maildir or access other of the processes run by the same user.

      Any user/application should be able to dynamically create a sub-user that can only access its own processes and files. It could be performed crudely using groups, where a sub-user is created with a new UID/GID and the original user is added to the group of the new UID (being able to control that user) - umask for that new user should allow group read/write.

      It could be refined a lot. I really don't see any reason why a process de facto should be able to inherit all the power and privileges of the user running it. Of course there are exceptions (one might want one process to actually be able to have all the same privileges as one self), but these should be handled as such exceptions.

      I just don't see why running a simple application should be able to risk your entire account (and maybe even the root account if you su from the compromised account).

      --
      - Peter Brodersen; professional nerd
    7. Re:challange accepted by SEE · · Score: 1

      rm -rf /

      So, um, why is this so much more devastating if it kills more than just all my data? Software is the easiest thing to replace; my data is what's important, and I lose it no matter whether it's stored in a user account I nuke as a user or in the root account and I nuke it as root.

    8. Re:challange accepted by Anonymous Coward · · Score: 0
      amongst a high seas of other things that make running as root unsane on the "woops scale"
      What part of that didn't seem like a joke to you

    9. Re:challange accepted by FidelCatsro · · Score: 1

      that was but one example and a rather poor one if i do say so myself(doh)
      A compromised user account is a dangeour only to your data , if the system is secure.
      A compromised root account is a dangeour to other peoples data and your own(Local and remote if the computer is set used by someone for illicet)

      --
      The only things certain in war are Propaganda and Death. You can never be sure which is which though
    10. Re:challange accepted by FidelCatsro · · Score: 1

      No that proves that the act of banging your head off a table is insecure by default without an added level of security such as a helmet .

      --
      The only things certain in war are Propaganda and Death. You can never be sure which is which though
    11. Re:challange accepted by marcosdumay · · Score: 1

      chmod 777 -R /

      That is useless if you run as root.

      I think that the bigest danger is some worm/trojan/... making itself invisible and spying your system forever (at least, with Windows people have to format their HDs some times).

  27. Ok, show of hands by Anonymous Coward · · Score: 0

    Who didn't see this coming?

    What, so the name "Lindows" didn't clue you in from the start?

  28. MOD DOWN by XanC · · Score: 0, Offtopic

    Random anti-Linux gibberish; nothing to do with this (or any, really) topic.

  29. I use root every time I log in... by Anonymous Coward · · Score: 0

    But then again, unlike everyone else, I'm not an idiot!!!!!!1

    On a serious note, he really shouldn't have said that. I know he's just defending a practice that's encouraged natively by his product, but saying something like that kills a lot of credibility instantly.

    Running as root is bad because it unnecessarily allows room for user (or program) error which can often have devistating results.

    It also allows normal security glitches in programs the same access to cause such undesirable effects. The stupid buffer overflow in Mozilla just went from erasing your home directory (full of scanned comic books and editor preferences) to erasing 5 years of corporate data. Woops!

    Not to mention that at one point or another, everyone screws up a command. Be it typing rm -Rf /usr/ instead of /usr/samba/ (mind fart), or accidently hitting enter before you can complete "kill 14914", and ending up with "kill 1"... nasty results, stupidly unnecessary.

    These things can't be prevented entirely, but running as a non-priviledged user goes a long way, not to mention it helps encourage proper access and operational policies in the system.

  30. Ignorance by El · · Score: 4, Insightful

    This is exactly the kind of attitude that I'd expect from someone that learned everything they know about computers from working with MS-DOS... he can't seem to conceive of the notion that there might be more than one person's data on a single machine!

    --

    "Freedom means freedom for everybody" -- Dick Cheney

    1. Re:Ignorance by Short+Circuit · · Score: 1

      On a Linspire machine, he might be right.

    2. Re:Ignorance by Anonymous Coward · · Score: 0

      RTFA

      He is talking about logging in as the root user, when there are no user accounts. If you have multiple users on the machine, then you have obviously set up user accounts and would use your own.

    3. Re:Ignorance by Val314 · · Score: 1

      > he can't seem to conceive of the notion that there might be more than one person's data on a single machine!

      Most Home Users have a single user and auto Login turned on.
      In fact i havent seen a single Home PC(*) that has more than one User.

      (*) That means PC from "normal Guys" not the ones that read /.

  31. Chuckle by The+Bungi · · Score: 1

    Let's just wait until he has 100 million customers (98% of which will be totally clueless if the Windows user base is any indication) and we'll see if it's such a good idea.

    1. Re:Chuckle by BigTunaCan · · Score: 0

      Those 98% already can hose their Windows boxes now so it won't be any different. If you read the article he makes a good point anyway. On top of that, how many "clueless" users are going to pull up an X-term anyway and start running things from the command line? Clueless users run the mouse, which makes it much less likely they will destroy everything.

  32. 500,000 windows zombies by Linux_ho · · Score: 5, Insightful

    Why is it more secure not to run as root?

    500,000 Windows zombies should be the only answer you need.

    --
    include $sig;
    1;
    1. Re:500,000 windows zombies by IthnkImParanoid · · Score: 1
      500,000 Windows zombies
      ppppaaaaaaaaaaannnneeeessssss.
      --
      It's nothing but crumpled porno and Ayn Rand.
    2. Re:500,000 windows zombies by Anonymous Coward · · Score: 0

      And how exactly would running as a non root user prevent this?

    3. Re:500,000 windows zombies by Jeff+DeMaagd · · Score: 2, Interesting

      In that case, I think running in administrator mode just makes it harder to remove the infection. I think it's trivial to trojan people into running bots that run in user space rather than system space. It's just not necessary to make such a program because it's easier to assume they are running as admin.

    4. Re:500,000 windows zombies by Anonymous Coward · · Score: 0

      If nothing else it *might* make you think "hmmm, why does this screensaver weather thingie need complete and unrestriced access to my computer to install?"

    5. Re:500,000 windows zombies by gnuman99 · · Score: 1

      Running bots under normal users will show them up as processes. Bots running under root can hide themselves from being detected.

    6. Re:500,000 windows zombies by Anonymous Coward · · Score: 0

      The type of user who opens BritnaysTitz.rpm in from their email isn't exactly checking their process list.

    7. Re:500,000 windows zombies by MikeBabcock · · Score: 1

      But their virus scanner is.

      Think for a moment of Windows -- if that virus / spyware / etc. didn't have administrator access, and the virus scanner running in the background does, then the scanner can 0WN the virus in no time.

      If however the virus also has admin power, it can hide itself, shut off the virus scanner (I love that one), or even uninstall the definitions files.

      --
      - Michael T. Babcock (Yes, I blog)
    8. Re:500,000 windows zombies by dangitman · · Score: 1
      500,000 Windows zombies should be the only answer you need.

      Zombie eat brains, but zombie cannot swallow injustice of using Windows!

      --
      ... and then they built the supercollider.
    9. Re:500,000 windows zombies by Anonymous Coward · · Score: 0

      You've not used Mac OS X have you? Most installation programs require you enter your password to allow for root access. I doubt many OS X users even give it a second thought.

  33. Is this the Anti-Theo by 0xdeaddead · · Score: 1

    Really someone ought to put him in a room with Theo to see if they cancel eachother out.

    1. Re:Is this the Anti-Theo by Anonymous Coward · · Score: 0

      Hehe. I laughed. Your mission is fulfilled! :-)

  34. Perfect Example by Apreche · · Score: 1, Informative

    So every user on a system usually can make files in /tmp. Let's say that a malicious user of the system goes into /tmp and makes an executable file named ls. That executable file contains the code which opens up a backdoor onto the system via netcat. If you were running as a normal user and ran ls in /tmp then you would not open up any backdoor. In fact, you might realize what's going on and be able to fix it. If you were root however, the backdoor would open wide and let the whole world have a root shell on your machine. This particular problem can be averted by removing . from $PATH of all users including root. But does Linspire do this? I don't know for sure, but I doubt it.

    Linspire, Linux dumbed down for dummies by dummies.

    --
    The GeekNights podcast is going strong. Listen!
    1. Re:Perfect Example by plj · · Score: 1
      This is ultimately the most insightful comment attached to this story that I've read so far. It clearly underlines the real reasons why no one should run as root. For all others who did not RTFA and neither thought about it by themselves:
      1. Normal boxes only have one user. Most people have no clue whatsoever about multiple user accounts. This means that your own account is all that matters.
      2. Due to the fact mentioned above, rm -rf in ~/Documents/ is as bad for most people as rm -rf in /. Most people are also very lazy in taking backups.
      3. Normal users do not run and should have no business running any daemons that are listening to non-local ports either.
      But parent was the first one who clearly described an excellent reason to stay as normal user even under conditions I stated above: it just greatly helps in avoiding troijans and other malicious programs.
      --
      “Wait for Hurd if you want something real” –Linus
    2. Re:Perfect Example by ink_13 · · Score: 2, Insightful

      Except that in a properly configured system, . should never be in your $PATH.

    3. Re:Perfect Example by prockcore · · Score: 2, Insightful

      This particular problem can be averted by removing . from $PATH of all users including root

      I've never seen a linux distro that even *had* CWD in $PATH.

      But hey, maybe you like going around trying to run "./ls"

      Finally, PATH is searched in order. So /bin:/usr/bin:. would find ls in /bin and run it.. even if ls was also in /usr/bin or cwd

  35. Excellent commentary... by ta+bu+shi+da+yu · · Score: 1, Interesting

    ... however, your comment about FireFox not adopting ActiveX, I would put to you, is actually not a good thing. Many, many Microsoft software developers are exploiting this, and without ActiveX compatibility they aren't going to migrate to FireFox very quickly (if at all).

    On a side note: this is sort of like Word and Excel macros and OpenOffice.org. Without them, Oo.org is missing quite a few companies.

    --
    XML is like violence. If it doesn't solve the problem, use more.
    1. Re:Excellent commentary... by NanoGator · · Score: 3, Insightful

      "... however, your comment about FireFox not adopting ActiveX, I would put to you, is actually not a good thing."

      Lack of ActiveX support actually prevented my previous company from switching to OpenOffice or Mozilla. The attitude that it's better that these two apps don't support it seriously pisses me off. If Microsoft can't get away with being arrogant, than the OSS Community can't either. /rant

      --
      "Derp de derp."
    2. Re:Excellent commentary... by someonewhois · · Score: 1

      Very true. The only thing is a LOT of people use macros in Office, but relatively speaking, not too many people use ActiveX for a legitimate reason, and therefore it doesn't cause as big of a dilemma.

    3. Re:Excellent commentary... by Monkelectric · · Score: 5, Insightful

      How could open source applications support ActiveX? The WHOLE point of ActiveX was to add a proprietary MS extension to the web to keep companies like yours locked in.

      --

      Religion is a gateway psychosis. -- Dave Foley

    4. Re:Excellent commentary... by NanoGator · · Score: 5, Informative

      "How could open source applications support ActiveX?"

      Ask these guys.

      BTW, you REALLY don't understand what ActiveX is. Heh. Non-MS products can open ActiveX plugins.

      --
      "Derp de derp."
    5. Re:Excellent commentary... by NanoGator · · Score: 1, Insightful

      "BTW, you REALLY don't understand what ActiveX is. Heh. Non-MS products can open ActiveX plugins."

      Hmm.. I misread your comment. Sorry. I'd like to replace the above statement with this one:

      How can OSS use ActiveX? A better question is: How can they expect change to happen if stuff works in MS products and doesn't work in OSS software? All of your ideals go flying out the window if one can't do their basic job. You're basically handing Microsoft a good reason continue being their customer, not the other way around.

      Make all the excuses you want, at the end of the day what matters is if the product does what it needs to or not. That's why my previous company uses Office and not OpenOffice. You can tell them they're wrong all day and all night, but they'll reply with "I don't give a shit, I can't control my need for AX."

      --
      "Derp de derp."
    6. Re:Excellent commentary... by Ost99 · · Score: 2, Insightful


      BTW, you REALLY don't understand what ActiveX is. Heh. Non-MS products can open ActiveX plugins.

      No, you don't understand.
      It's still a proprietary MS extension even if you can add it to non-MS products.

      Oh and you can't add it to Linux products.

      - Stian

      --
      ---- Sig. gone.
    7. Re:Excellent commentary... by One+Childish+N00b · · Score: 4, Insightful

      Refusal to support one of the biggest vectors of spyware infiltration is not arrogant, it is common sense, at least for the average desktop user.

      For the business world, admittedly, with the entrenched position of ActiveX-based systems on corporate intranets, it's perhaps a little silly and a bit of a barrier to business adoption, but for home users one of the biggest complaints about Windows is the fact their machine can be 0wned by Virtual Bouncer, CoolWebSearch, ABetterInternet and God knows how many other drive-by-installed apps and toolbars just by visiting a slightly wrong-side-of-the-tracks website.

      --
      Dealing with lawyers would be a lot less tedious if they all looked like Casey Novak.
    8. Re:Excellent commentary... by Monkelectric · · Score: 1
      BTW, you REALLY don't understand what ActiveX is. Heh. Non-MS products can open ActiveX plugins.

      Eh, no. On WINDOWS platform, applications can open ActiveX. On every other platform and architecture -- its hopeless. Us open source guys don't live in a monoculture.

      --

      Religion is a gateway psychosis. -- Dave Foley

    9. Re:Excellent commentary... by NanoGator · · Score: 1, Troll

      "No, you don't understand."

      I think you missed it, but I added a correction to my previous post. I misread what that person had said.

      "Oh and you can't add it to Linux products."

      Um yeah. A Windows user trying to switch to FireFox or OO doesn't give a flying fuck if AX will work on Linux or not. Pardon my bluntness, but you're rationalizing NOT putting a feature in that some people need. That's bullshit.

      --
      "Derp de derp."
    10. Re:Excellent commentary... by Anonymous Coward · · Score: 1
      The attitude that it's better that these two apps don't support it seriously pisses me off. If Microsoft can't get away with being arrogant, than the OSS Community can't either.

      There is nothing arrogant in the stance of avoiding implementing support for ActiveX (the way MS implemented it); quite on contrary. It WAS arrogant that Microsoft implemented it the way they did, even after pointed out by everyone how dangerous it was. Not repeating the same mistake is if not wisdom, at least common sense survival.

      The fundamental problem, of course, is that of distinction between "safe" intra-nets (that can allow wider access; and do benefit from extra functionality of such extensions) and general web, which makes it hard to create a one-size-fits-all solution. That particular problem has been mostly solved by Java (applets, java web start); too bad it was done in a too-little-too-late way. :-/

      At any rate, "just support ActiveX" is a wrong answer, and repeating it doesn't make it any better.

    11. Re:Excellent commentary... by DShard · · Score: 1

      Great... how does it work in linux?

      AxtiveX is the Microsoft API in answer to Java. It lacks sandbox execution so any ActiveX embedded code can trivially access anything on Windows. The reason even microsoft doesn't think you should blindly execute any activeX component is that it is equivalent to a local binary in terms of system access.

      If you fail to see why it is insecure and bad for an untrusted network then please change your settings for activeX in the "internet" zone to enable and browse until you get the point.

    12. Re:Excellent commentary... by Anonymous Coward · · Score: 0
      '0wned'?


      Get out of your parents basement much?

    13. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      Lack of ActiveX support actually prevented my previous company from switching to OpenOffice or Mozilla.

      So you wanted to switch to Mozilla and OpenOffice, but STICK with ActiveX? That's a very odd decision to make. That's like modding your (old style) VW bug, replacing the body with one from a prius, then saying you want to keep the old engine. Now you're mad that the prius can't waste as much gas or produce as much CO. Why not get rid of the crap while you're at it and move on to something better?

    14. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      [i]Lack of ActiveX support actually prevented my previous company from switching to OpenOffice or Mozilla. The attitude that it's better that these two apps don't support it seriously pisses me off. If Microsoft can't get away with being arrogant, than the OSS Community can't either.[/i]

      I'm sorry that a choice based not on arrogance, but on examining and weighing the difficulty of developing and emulating a proprietary interface, the benefits to users like yourself that this would bring, and the drawbacks of limited developer resources (may volunteer), and the security problems that the ActiveX model has at a fundamental level. Making a decision like this is not based in any way on arrogance. Now there may be Open Source pundits who may be arrogant in their derision of ActiveX, Microsoft, or MSIE. That's an entirely different issue than the very logical and reasoned decisions the developers and contributors to Firefox have made.

      The wonderful thing about open source is if there is an itch, you can scratch it. So if there are enough people like you who could really, really use ActiveX support, there is ample opportunity for you to pool your resources and attempt to develop just such support. You might have to fork the source tree and call your modified ActiveX-enabled Firefox-based browser something else, or perhaps once you've developed it and debugged it, you can make a convincing argument to the Firefox source tree maintainers to integrate your support.

      I hope that in spite of your understandable frustration that you were unable to use Firefox because of lack of ActiveX at the previous company you worked at, and in spite of your also-understandable irritation at those people DO respond arrogantly about such things when calm reason should do, that you recognize that venting your frustration by calling Firefox developers arrogant is really just a form of the same thing that irritates you when you're on the receiving end.

      Now we interrupt this post to bring you a trite cliche: "Can't we all just get along?" *smirk*

    15. Re:Excellent commentary... by NanoGator · · Score: 1

      "At any rate, "just support ActiveX" is a wrong answer, and repeating it doesn't make it any better."

      Ignoring it and hoping it goes away is not the right answer, either.

      --
      "Derp de derp."
    16. Re:Excellent commentary... by Ost99 · · Score: 2

      Um yeah. A Windows user trying to switch to FireFox or OO doesn't give a flying fuck if AX will work on Linux or not. Pardon my bluntness, but you're rationalizing NOT putting a feature in that some people need. That's bullshit.


      I'd rather have a true cross-platform product. MS can keep their shit to them selves.

      Besides, ActiveX is a security nightmare. It is simply not worth it; whatever the problem is, ActiveX is never the solution :-)

      - Ost
      --
      ---- Sig. gone.
    17. Re:Excellent commentary... by dvdeug · · Score: 1

      Lack of ActiveX support actually prevented my previous company from switching to OpenOffice or Mozilla. The attitude that it's better that these two apps don't support it seriously pisses me off. If Microsoft can't get away with being arrogant, than the OSS Community can't either.

      Why does it piss you off? Agree with it or not, but there's very valid reasons behind the thought that it is better that OpenOffice and Mozilla can't run arbitrary code. Not everyone who doesn't see things your way is being arrogant.

    18. Re:Excellent commentary... by NanoGator · · Score: 2, Interesting

      "It is simply not worth it; whatever the problem is, ActiveX is never the solution :-)"

      *Sigh* This is what I'm talking about! I know AX ain't great. I'm no fan of it, either. But when it's needed, it's NEEDED. Since OO and FireFox wouldn't support it, we had to use a MORE INSECURE office and browsing app! You cannot honestly tell me that the OSS Community couldn't develop something to support AX and maintain security. Heck, all it would really need is to be off by default and the user has to either turn it on or install a special module. I don't care. It certainly would have been infinitely better than what we had.

      Whatever. I seriously doubt this has been given serious consideration. Flipping off MS is fun, but you're also flipping off some people who can't switch.

      --
      "Derp de derp."
    19. Re:Excellent commentary... by fbjon · · Score: 2, Interesting
      Is there som obstacle to adding support for activeX in only the windows version? Like this:

      Default turned off. If a page has some activex thingys, block, display small text that a thingy was blocked. If user wants to run it, click here and blabla, the url gets added to "Allow" list. Done. Other platforms need not even bother.

      --
      True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
    20. Re:Excellent commentary... by SQLz · · Score: 1

      LOL...that was the plan man! How soon people forget. The whole point of ActiveX is vendor lock. Their plan is obviously workin but now that 'vendor lock in' is a hot topic when people develop a new system from scratch they are not opting to use ActiveX. Those of you stuck with ActiveX, thats your own fault.

    21. Re:Excellent commentary... by Ost99 · · Score: 2, Interesting

      The new netscape based on firefox is supposed to support AX on windows.

      Transgaming is working on a mozilla plugin for AX, for linux running winex / cedega.

      For openoffice, I think macros (with import from ms formats) would be more useful than AX (who uses AX in a document?)

      - Ost

      --
      ---- Sig. gone.
    22. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      "If you fail to see why it is insecure and bad for an untrusted network then please change your settings for activeX in the "internet" zone to enable and browse until you get the point. "

      He didn't say that AX was secure he said that he needed it and the only way to use it was with Internet Exploder.

      These stupid comments like yours are really quite amusing given his comments about arrogance.

    23. Re:Excellent commentary... by saforrest · · Score: 1

      Whatever. I seriously doubt this has been given serious consideration. Flipping off MS is fun, but you're also flipping off some people who can't switch.

      I think there are enough people out there who have to use ActiveX that support will eventually be added as a special module or something.

      It's just that its status as Microsoft's attempt at vendor lock-in has given it enough of a bad taste that it's pretty far down on the priority list of most volunteer developers.

      It's surprising to what degree open-source development is still driven by the private whims of its contributors.

    24. Re:Excellent commentary... by Monkelectric · · Score: 2, Interesting
      Well, what the grandparent is pissed at -- and he has half a point -- is that firefox COULD support activex -- on windows only, by using the activex api.

      However, activeX is a security nightmare. And regardless it *IS* a proprietary MS extension -- and nobody wants to A: support MS and their bullcrap, B: Firefox has a reputation as a secure alternative to IE. If FireFox supports the hopelessly insecure ActiveX -- they really have nothing to offer anyone anymore as their reputation is *done*.

      --

      Religion is a gateway psychosis. -- Dave Foley

    25. Re:Excellent commentary... by DShard · · Score: 1

      You said this need is filled by a plugin. OO supports scripting. What is it that wouldn't fill the need?

      It isn't about flipping microsoft any animal. It is about best practices. You shouldn't consider activeX as a viable alternative _ever_. It sends the wrong messages to users and developers. So that doesn't suit you, the guy who depends on this junk for your job, well somebody has already stepped forward and provided a solution. Security be damned, there it is.

      OO isn't a web browser so why would it ever? Maybe you mean VBA... assuming yes, the answer is they can't and no one besides microsoft or a licensee can. They don't license it for free so it isn't possible.

    26. Re:Excellent commentary... by Anonymous Coward · · Score: 2, Insightful

      "B: Firefox has a reputation as a secure alternative to IE. "

      Except that it cannot completely be an alternative to IE because IE supports something that FireFox doesn't.

      I believe that is the point that other dude was trying to make, but everybody's busy telling him he's wrong.

    27. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      He already mentioned a plugin that could handle it brainiac. He was asking why they didn't ship it.

    28. Re:Excellent commentary... by fbjon · · Score: 1

      Good point. So how about an obscure plugin that the really persistent winwebdesigners can point at? With the same ultra-restrictive behavior, including extra blocks for the really insecure stuff.

      --
      True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
    29. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      ... and everybody arrogantly shook their heads and told him no.

      Hypocritical assholes.

    30. Re:Excellent commentary... by Jeremi · · Score: 1
      You cannot honestly tell me that the OSS Community couldn't develop something to support AX and maintain security.


      They certainly could, but why should they when the general consensus is that ActiveX was a bad idea? To make you happy? What's in it for them? Perhaps if you (or your locked-in company) paid them $$$ to do it.


      --


      I don't care if it's 90,000 hectares. That lake was not my doing.
    31. Re:Excellent commentary... by sbrown123 · · Score: 5, Insightful

      Okay, I will make it easy for you. Why does Firefox and OpenOffice not use ActiveX? Heres why:

      (1) It does not work cross-platform. Both Firefox and OpenOffice work on platforms other than Windows. Both platforms keep this compatibility by not introducin technology that could possibly limit this capability.

      (2) It is proprietary. You may be confused on what this means. Basically, the technology is owned by Microsoft. This very same reason is why PNG exists despite the existance of GIFs. GIF technology was proprietary and, thus, could not be placed into a product that had a open source license (Linux).

      (3) Firefox has no need for ActiveX since it has, in my opinion, a better technology with XPCOM. OpenOffice, if I remember, can be extended with Java plugins. Java has built-in security unlike ActiveX. Both XPCOM and Java are cross-platform which goes back to my point #1.

      (4) Active X is not very secure. You will hear this time and time again. Microsoft even knows this and turned them off by default in SP2!

      Make all the excuses you want, at the end of the day what matters is if the product does what it needs to or not.

      As stated in point #3 above both Firefox and OpenOffice support technologies that give them quite a bit of power to get any job done.

      I have a pretty good memory and I remember correcting you on these issues before:

      http://slashdot.org/comments.pl?sid=144131&thres ho ld=0&commentsort=0&tid=109&mode=thread&cid=1207754 3

      Before you go spouting about some guy who would have no job if it werent for ActiveX or some other non-sense drivel, answer this:

      What does ActiveX do that XPCOM and Java are incapable of performing?

      If you are not able to answer that question, you have no base to stand on.

    32. Re:Excellent commentary... by Anonymous Coward · · Score: 0
      Besides, ActiveX is a security nightmare. It is simply not worth it; whatever the problem is, ActiveX is never the solution :-)

      Why is ActiveX worse than, for example, plugins? Both allow native code to execute, call native APIs, etc. What's the difference, apart from the fact that ActiveX is much more widely used, especially for bespoke applications?

    33. Re:Excellent commentary... by Anonymous Coward · · Score: 0
      Yeah, it's really surprising that people who develope software for free tend to not work on stuff that they're opposed to. Man, I wish I had been sharp enough to figure that out on my own.

      Seriously though, if you or your company were stupid enough to get locked into microsoft software, then stick with it. You deserve each other.

    34. Re:Excellent commentary... by fyngyrz · · Score: 4, Insightful
      Except that it cannot completely be an alternative to IE because IE supports something that FireFox doesn't.

      Fine. But FireFox (and others, such as Mac's Safari) support something highly worthwhile that IE most definitely does not. Namely, a reasonably safe and secure browsing experience.

      Some markets will opt for security and safety, using technologies that are (compared to active x) much (duh) safer and more secure.

      Others will continue to endure spyware, viri, adware and various trojans and other invasive garbage. Those are "IE features" FireFox doesn't want to offer. Or let me put it this way -- they are "features" that this FF user doesn't want to be "given", because they are inevitably prefaced with the command "bend over."

      I truly think that to impress ActiveX upon FireFox would be just about the worst thing the FF developers could do. FireFox provides a better experience. That's why it's doing so amazingly well. Put ActiveX in there, and that experience is going to begin to degrade. It may go as far as to be as risky to surf with FF as it is to surf with with IE.

      Does anyone really want that, other than the companies who have embraced and extended Microsoft's Active-X? Is there anything truly significant you can do with Active-X that you cannot also do with Java?

      Sure... you pick a technology that is proprietary to one browser, that browser starts to lose favor with the user community, and definitely, you will have work to do. Time to start studying Java. It's not time for the junk technology to be imported into FF to extend the EOL of some Active-X product.

      Java was designed to be secure. It's been remarkably successful at it, too.

      --
      I've fallen off your lawn, and I can't get up.
    35. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      "What does ActiveX do that XPCOM and Java are incapable of performing?"

      It runs currently available ActiveX controls that haven't been ported to XPCOM and Java.

      Next question?

    36. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      > Fine. But FireFox (and others, such as Mac's Safari) support something highly worthwhile that IE most definitely does not. Namely, a reasonably safe and secure browsing experience.

      Fine. Turn it off by default. Simple.

      >Does anyone really want that, other than the companies who have embraced and extended Microsoft's Active-X? Is there anything truly significant you can do with Active-X that you cannot also do with Java?

      Yeah. You can run Active X controls that haven't been portd to Java.

      So, I have to ask, did you actually put any serious thought into what the guy said? Or, did all your thought go into telling him why it cannot be done? If it's the latter, which I strongly suspect, then blammo, there's the arrogance he was talking about.

    37. Re:Excellent commentary... by Anonymous Coward · · Score: 2, Informative
      AxtiveX is the Microsoft API in answer to Java.

      That's not at all true. ActiveX is just COM/OLE, which is older than Java. The origins of COM/OLE go back to the 1980s, and OLE 1 was publicly distributed with MS Office in 1991. OLE 1 wasn't based on COM, however, so is to some extent irrelevant. The first release of COM-based OLE (called OLE 2) came in 1993, at a time when Microsoft were still ignoring the Internet, with OLE controls (now called ActiveX controls) added to Visual Basic in 1994.

      The first release of Java only came in 1996, and whilst it almost certainly did inspire Microsoft's rebranding of COM as ActiveX, the ActiveX technology itself was not in any way an answer to Java (and obviously couldn't have been, since it's older).

    38. Re:Excellent commentary... by Zphbeeblbrox · · Score: 1

      The real question is why are people still writing stuff in ActiveX? Sure lots of people are still using IE but the number is dwindling. Wouldn't it be better to start shifting so you don't get left behind? For that matter why are people still programming in XPCOM and Java? What's wrong with some good old XML-RPC X-HTML? Honestly I could do without all those silly downloadable embedded applications. All they ever did was make my life difficult when I was a windows user. If it needs ActiveX then it doesn't belong as part of a website. Make a downloadable application that runs seperately. I even dislike Java's WebStart technology.

      --
      If you see spelling or grammatical errors don't blame me. I tried to preview but IE here at work borked the CSS
    39. Re:Excellent commentary... by amliebsch · · Score: 2, Funny
      What does ActiveX do that XPCOM and Java are incapable of performing?

      It can run ActiveX applications that have already been designed and implemented.

      --
      If you don't know where you are going, you will wind up somewhere else.
    40. Re:Excellent commentary... by lgw · · Score: 1

      OO scripting doesn't fill the need to run the VBA already in my spreadsheet. Therefore, OO is dead to me.

      What's stopping anyone from clean room reverse engineering VBA? It's just a programming language, shouldn't be that hard to write from scratch (especially the pre-.NET VBA, which is most of the VBA code base).

      --
      Socialism: a lie told by totalitarians and believed by fools.
    41. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      What does ActiveX do that XPCOM and Java are incapable of performing?

      ActiveX makes a MS manager happy. This makes my life easier and gives me a better performance review. So I guess it makes it look like I perform better. Java and XPCOM don't accomplish this.

    42. Re:Excellent commentary... by Eric+Damron · · Score: 1

      Non-MS products running on a MS OS that is. Lockin does occur.

      BTW I do understand active X and have written ActiveX components for my Department.

      --
      The race isn't always to the swift... but that's the way to bet!
    43. Re:Excellent commentary... by fyngyrz · · Score: 4, Insightful
      The argument being made here is that Active-X is dangerous. You seem to give that a nod by saying it should be off.

      Fine.

      However, the next implication is that it can be turned on. This is not fine. Why? Because it is dangerous. The average user does not comprehend that it is dangerous. Like the argument here that one should not run as root (which I agree with for most people in most situations) the idea is that if you're not smart enough to handle a tool, you should not be handed that tool.

      It's not arrogance to say that it is not a happy worldview to see people's computers being trashed by junkware let in by badly designed software -- Active-X -- it simply isn't a good thing. You can't make it a good thing.

      Now, if a company has invested time in developing for this proprietary (but very dangerous) technology, and the marketplace leaves them behind, as it is showing definite signs of doing, then if that company wants to survive, it needs to lose the dangerous technology, get with the program, and use the safe technology. That's called evolutionary pressure. I'm part of that pressure. I don't use IE. If you use IE-specific technologies on your site, you've lost me (and at least 10% of the rest of the world, and more every day.) Now, you can only ignore this for so long before you (a) solve the problem by losing the junkware, or (b) are driven from the business space by competitors who are able to recognize and resolve the problem.

      From a user perspective, I'm just one guy. I won't use IE.

      From an applications standpoint, I own several companies and we don't use Active-X (or Java, for that matter) as a matter of course. We do server-side apps, because (a) we have total control over them and (b) because all users, that's 100% of them, can use our apps. We give up some glitz, certainly, but we've never, ever had to give up anything important.

      So my outlook does have some effect. If Active-X were to go away, it wouldn't touch me at all, other than to make the web more accessible to me and perhaps give my competitors a more stable place to stand. Do I worry about the people who invested in Active-X? No. And, really -- why should I?

      Arrogant? No. I'm entitled to my opinion, just as you are entitled to yours. As for putting any thought into it, apparently you didn't notice my sig. This isn't an issue I just picked up on this afternoon. I have indeed thought about it, and this is where I ended up.

      --
      I've fallen off your lawn, and I can't get up.
    44. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      Sure lots of people are still using IE but the number is dwindling.

      By "dwindling" you mean "increasing at a slower rate."

    45. Re:Excellent commentary... by saforrest · · Score: 1

      Yeah, it's really surprising that people who develope software for free tend to not work on stuff that they're opposed to. Man, I wish I had been sharp enough to figure that out on my own.

      My point wasn't as trite as that, but I didn't think I needed to make it because it's been made so often before.

      The reason most open-source developers contribute their time, aside from the sheer love of making software, is to build a better computing world.

      Part of building this world is attracting interest and money to support open-source projects. Doing this kind of work -- interfacing with proprietary formats, etc. -- isn't as much fun, but helps to spread the word.

      Clearly we are motivated to spread the world; what surprises me is the degree to which our goals are still driven by our (possibly obscure) personal interests rather than some idea of what the world needs.

      I'm not saying it should necessarily be otherwise, or that I do anything other than create software motivated by my personal interest. I just would have thought that by now there would be much larger communities of people developing general-interest 'software for the masses' to lure them off proprietary stuff.

    46. Re:Excellent commentary... by cgenman · · Score: 4, Insightful

      Good post overall. However, you're simplifying the relationship between GIF and PNG, and you imply that by including proprietary renderers in Linux that they too must be open source. This is completely incorrect, as many binary drivers, patent-encumbered applications, and even closed applications are distributed with Linux. Debian has an open-only policy, but that reflects their outlook, not a requirement of the Open Source license.

      PNG was developed not because it was impossible to put GIF support in Linux, but because it was feared that Compuserve (which discovered it held a patent on one of the processes used in GIF compression / decompression) would abuse it's power on all platforms. In the early days, they talked about levying a fee on all clients, users... anything that interacted with GIFs. At which point development of PNG began. I believe CompuServe finally settled on the less unreasonable 5c per paid application that can encode GIF's, with no fee for decoders. That fee is no longer with us, as the patent has expired.

      On the other hand, PNG has surpassed GIF's by adding alpha layer transparency... in other words, you can have certain pixels that are 100% opaque, or 10%, or 55% solid, or whatever. This would make working with images on the WWW so much easier, if MS would just bloody well implement proper PNG support like they promised as a feature for I.E. 4.

      What does ActiveX do that XPCOM and Java are incapable of performing?

      Install very convienient password management apps, automatically, like Gator.

    47. Re:Excellent commentary... by Waffle+Iron · · Score: 3, Insightful
      Heck, all it would really need is to be off by default and the user has to either turn it on or install a special module.

      With ActiveX, you're using IE as a custom client UI for your apps, not as a web browser. Why should other web browsers turn themselves into a general-purpose Win32 UI platform? That's not their focus.

      What would be wrong with just staying with IE for your Win32 application? You can still keep it around just as a container for your custom-coded UI clients. If you want to actually *browse* the wold-wide-web instead of running little Win32 applications, nothing's stopping you from using other more modern browsers.

    48. Re:Excellent commentary... by sbrown123 · · Score: 0, Redundant

      It can run ActiveX applications that have already been designed and implemented.

      ActiveX is nothing more than simple COM. It is not very difficult to use Java or XPCOM to communicate to ActiveX controls, and vice versa.

    49. Re:Excellent commentary... by sbrown123 · · Score: 2, Interesting

      Nice try. ActiveX is nothing more than simple COM. It is not very difficult to use Java or XPCOM to communicate to ActiveX controls, and vice versa. Try again.

    50. Re:Excellent commentary... by Electroly · · Score: 1

      GIF technology was proprietary and, thus, could not be placed into a product that had a open source license (Linux).

      Last I checked, open source Mozilla supports GIFs.

    51. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      I agree with NanoGator - you have no idea what ActiveX is. if XPCOM == GOOD then ActiveX == GOOD too - ActiveX == a handful (dozen) (XP)COM interfaces --- Hmmm ActiveX has hmmm - maybe 5 actual functions to 'implement' for an API --- perhaps you should reference to Mozilla Help on XPCOM -- notice how it references msdn.microsoft.com for help on COM? The only real difference between a MS's ActiveX controls is the 'seamless' download. Netscape plugins require a seperate naviagtion/download/install. Both require user interaction - just some more than others. Microsoft's model is more prone to social engineering problems.

    52. Re:Excellent commentary... by msuarezalvarez · · Score: 1
      It's surprising to what degree open-source development is still driven by the private whims of its contributors.

      What else would it be mostly driven by?

    53. Re:Excellent commentary... by metamatic · · Score: 1
      You cannot honestly tell me that the OSS Community couldn't develop something to support AX and maintain security.

      Sure I can. ActiveX controls are native code, and have access to any API in Windows. They run with the full permissions of the user running the browser, which is generally an administrative user because so many applications break if you run them as a non-admin user. Hence ActiveX is inherently insecure--sure, you can have code signing, but once that ActiveX control runs, it can do anything from formatting the hard drive to rewriting the BIOS. The only way to make it "safe" would be to run the entire thing in a virtual machine--basically, have a VMware-like virtual Windows to run your ActiveX controls in. Which would probably ruin the point of most of them, because typically people use ActiveX when they want access to the local hard drive etc.

      --
      GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
    54. Re:Excellent commentary... by NanoGator · · Score: 1

      "Sure I can. ActiveX controls are native code, and have access to any API in Windows."

      Right, but FireFox doesn't have to allow download or execution of an AX control. Set it up with a white-list, and then it's entirely up to the user. (as opposed to arbitrarily deciding I cannot have it at all, which Microsoft gets flambe'd about all the time.)

      --
      "Derp de derp."
    55. Re:Excellent commentary... by sbrown123 · · Score: 1

      However, you're simplifying the relationship between GIF and PNG

      Yes, I did simplify the relationship. There is quite a history behind both formats. I will not get into preferences though since I use both and think they are both great.

      This is completely incorrect, as many binary drivers, patent-encumbered applications, and even closed applications are distributed with Linux.

      There is no patent-encumbered technology in Linux since that would contradict the purpose of the GPL license. Are you talking about Linux distros? I am aware of patented technology in these. NVidia and ATI drivers are good examples of binary installs. I have yet to see these incorporated into any distro though. Do you know of one?

      PNG was developed not because it was impossible to put GIF support in Linux, but because it was feared that Compuserve (which discovered it held a patent on one of the processes used in GIF compression / decompression) would abuse it's power on all platforms.

      Actually it was the LZW algorithm in GIF that was the problem. Unisys owned the patent on this technology and in 1994 started demanding royalities. And I don't believe its an abuse of one's power to require payment of royalities. I'm sure others will see differently though and software patents are sometimes rather stupid.

    56. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      So, smart ass, tell us how easy it would be to take an existing ActiveX control and port it to either Java or XPCOM. Then tell me that Internet Explorer supports it. THEN tell me that IE will install it, too.

      If you can give me a serious answer, I'll apologize for jumping your butt.

    57. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      "Seriously though, if you or your company were stupid enough to get locked into microsoft software, then stick with it. You deserve each other."

      It could have been worse. They could have been stuck with non-MS software and not gotten their jobs done.

    58. Re:Excellent commentary... by metamatic · · Score: 1

      IE gives users exactly that choice of whether to run ActiveX controls or not. It doesn't solve the problem, as you may have noticed.

      --
      GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
    59. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      "So you wanted to switch to Mozilla and OpenOffice, but STICK with ActiveX? That's a very odd decision to make. "

      That really depends on if the company he worked for developed the Active X control or not. You OSS zealots love pointing out the big problem with proprietary software. I'm amazed at how quickly that is forgotten when one is trying to discredit another.

    60. Re:Excellent commentary... by jonadab · · Score: 2, Interesting

      > Lack of ActiveX support actually prevented my previous company from switching
      > to OpenOffice or Mozilla. The attitude that it's better that these two apps
      > don't support it seriously pisses me off. If Microsoft can't get away with
      > being arrogant, than the OSS Community can't either.

      Arrogance has nothing to do with it; this decision is about (and can only be about) security. Applications that care about security *cannot* support ActiveX, full stop.

      It's not just better; it's *VITAL* that they not support ActiveX. If Mozilla for instance did support ActiveX, anyone even the slightest bit conscious of basic security issues would migrate away to another browser immediately (Opera, most likely). If you think ActiveX is a good thing, you have no idea what ActiveX is, or no understanding of security at all. Fundamentally, by design, ActiveX allows any website you visit to do, quite literally, whatever it wants on your computer[1]. A well-behaved site is *supposed* to be nice and just draw stuff in the browser window, but fundamentally it can do whatever it likes, because that's how ActiveX was designed. Microsoft created ActiveX during the era when they considered security to be 100% Somebody Else's Problem, so they didn't give this a second thought; now that they are making some attempt to take security seriously, they regret ever having developed ActiveX in the first place; sooner or later they will have to discontinue support for it in a service pack or upgrade, because there is no secure way to support it.

      It was a mistake for Microsoft to develop ActiveX and start supporting it; it would be a mistake for *any* application to support it that doesn't already, and the ones that do already (mainly, MSIE) will eventually have to bite the backward-compatibility bullet and stop supporting it. Mozilla.org absolutely cannot afford to make that kind of mistake; security has been and is one of the major factors driving Firefox adoption; if Firefox supported ActiveX, it would actively lose most of its market share virtually overnight. That kind of wide-open security hole is never EVER worth the risk. OpenOffice *might* be able to get away with it better, because it is used mostly with internal documents, not content off the internet, but it would still be a major security headache, and not supporting ActiveX is still substantially the right decision.

      Lack of ActiveX support is not about lack of developer time; it is not about needing to reverse-engineer protocols; it is not about platform parity; it is not about open standards, and it is certainly not about arrogance; it is about security, and it is so essential to security that no other issue can matter.

      It is Windows users who would suffer if these applications supported ActiveX on Windows. Yes, Windows has other security problems, but ActiveX dwarfs relatively little things like Shatter attacks (a form of privilege escalation attack that exploits a design flaw in the Win32 API), because it is so much easier to exploit; it is not so much a security vulnerability as a complete abdication of all pretenses of security. Right now, Windows users have a choice; they can use MSIE, and pray nobody ever sends them a link to a site with a less-than-scrupulous webmaster, or they can download a browser with basic security. Don't take that choice away from them.

      ---
      [1] The design has now had user approval retrofitted onto it, so that a site
      now can only do whatever it wants after the user frobs the "Ok" button.
      But the user (and the computer, for that matter) has no way to tell
      before doing so whether the site intends to draw pictures in the browser
      window, scroll text across the status bar, or scour the user's Documents
      directory for credit card details and other personal information and send
      it back to the site. In fact, it's not easy to tell what a site's ActiveX
      programs (called "controls" in ActiveX parlance) have done even afterwards.

      --
      Cut that out, or I will ship you to Norilsk in a box.
    61. Re:Excellent commentary... by corvair2k1 · · Score: 1

      I believe the operative words that he used were "could not be placed into a product that had a [sic] open source license". You know, past tense. =)

    62. Re:Excellent commentary... by sbrown123 · · Score: 1

      I never said to port anything. Porting has never been mentioned and is not required. Read the posts before responding blindly.

    63. Re:Excellent commentary... by Curtman · · Score: 4, Insightful

      What does ActiveX do that XPCOM and Java are incapable of performing?

      I think the correct answer is marketing. The gecko browsers are packed full of some really cool toys for developers. But it's very very hard to sort through it all. Every so often I start playing with various features common to Mozillaish browsers like XPI, XSLT, and Javascript. It always strikes me how much potential there is to make some very cool applications using these. One pet project of mine is to see if I could create a set of XSLT documents that would transform glade projects into XUL applications, which could be themed via css.

      It's coming along pretty well, but I find it very difficult to wade through the developer documentation. XULplanet is a great resource, and there's a few others like the DOM ref on moz.org, but it seems pretty scattered, and sometimes out dated, and sometimes it just completely disappears like DevEdge (which there was some talk about being resurrected). In some cases, the only reliable documentation is the moz source itself, which is very hard to navigate without a fair bit of research.

      I've never done anything with ActiveX at all, or dealt with Microsoft API's very often at all, but I've seen their documentation, and it seems like its quite a bit more focused, and easy to find things.

      Having had to go looking for documentation myself, I think I can see why companies would be reluctant to use Mozilla technologies in house. Is there anybody at the Mozilla foundation that deals strictly with promoting moz as a developer platform, rather than focusing on the browser itself?

    64. Re:Excellent commentary... by Plaid+Phantom · · Score: 1

      Notice he said "was" proprietary. It seems to me that the status of GIF has changed since then.

      --
      All comments are properties and trademarks of the voices in my head. Not like I'm gonna claim them.
    65. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      Nice dodge.

    66. Re:Excellent commentary... by Ibag · · Score: 2, Insightful

      "What does ActiveX do that XPCOM and Java are incapable of performing?"

      ActiveX runs things that were previously designed to run via ActiveX. The question is akin to "Why use windows over linux?" There are many ActiveX controls already written that cannot or will not be rerwritten that people depend on, and unless someone takes the time and money to make clones for the controls, people will continue using the pltforms thtat the controls tie them to.

      There are good reasons for people to replace ActiveX with something else, but there are also good reasons for users to continue using the programs that support their controls. Inertia is a powerful thing.

    67. Re:Excellent commentary... by ultranova · · Score: 1

      ... however, your comment about FireFox not adopting ActiveX, I would put to you, is actually not a good thing. Many, many Microsoft software developers are exploiting this, and without ActiveX compatibility they aren't going to migrate to FireFox very quickly (if at all).

      AFAIK there's an ActiveX plugin for Firefox. Can't remember where I found it, nor do I know how well it works, but it was there.

      I also have to point out that a web browser really should just allow the user to browse the Web, not act as a software development platform - especially if doing so makes browsing the Web less secure.

      On a side note: this is sort of like Word and Excel macros and OpenOffice.org. Without them, Oo.org is missing quite a few companies.

      It's missing quite a lot of viruses, too :). Embedding executable code in data documents is a bad idea, now matter what kind of desperate database-lookalikes it might let one turn text or spreadsheet documents into.

      --

      Forget magic. Any technology distinguishable from divine power is insufficiently advanced.

    68. Re:Excellent commentary... by Durandal64 · · Score: 1

      Except if they're running on a non-Windows, non-x86 architecture, like Mac OS X. Then the system is simply incapable of executing the binary code in the plug-in. I've run into this little problem at work with the oh-so-wonderful Magic Helpdesk software. This kludge requires Internet Explorer 6 on Windows because it uses proprietary ActiveX controls which are not portable. And they actually advertise it as being able to support multiple platforms! Really? So anyone doing Mac support has to fire up a PC just to log tickets? That's brilliant!

      In my case, we got around it by setting up a remote Windows 2000 terminal server that the other Mac analyst and I could access from our G5s with RDC for OS X. Still a patently asinine solution.

    69. Re:Excellent commentary... by reub2000 · · Score: 1

      Yeah, but Richard M. Stallman hates propriary software, and thinks it shouldn't even exit.

    70. Re:Excellent commentary... by sumdumass · · Score: 1

      umm.. yea, and microsoft already said it is "part of the operating system" so it isn't like you can uninstall it or anything.

      Well you might be able to uninstall it with alot of prying and a third party app or somethign.

    71. Re:Excellent commentary... by Lorkki · · Score: 1
      Non-MS products can open ActiveX plugins.

      Yes, all they need is the Win32 API. I'm sure it'll be available for Linux, MacOS and Solaris any day now.

    72. Re:Excellent commentary... by s2kdave · · Score: 1

      GIFs aren't proprietary and they never were. It was the LZW compression algorithm used in GIFs that was patented. So anyone could freely read GIFs, but you needed a license to create a GIF. But now that patent has since expired so it's free all around.

    73. Re:Excellent commentary... by sumdumass · · Score: 1

      I don't think it is a decision to not let you have it at all. It is more of a decision of the need for it compared to the amount of development time. If it is so easy to implement safley then people would already have done it. As a matter of fact some have.

      When you are developing a product with the goal of the majority of code being recycled for other platforms, the focuse is on keeping that possible. I doubt someone is sitting in the mozilla development hut flipping us off saying "you'll get nothign and like it." AX is just insecure and takes time away from other important stuff. IF AX was so important, people would just add it and use it. Here is a link from another post with someon who though it was neccesary.

    74. Re:Excellent commentary... by ColMustard · · Score: 1

      You think the Internet Explorer user base is dwindling? Not even close.

      --
      Moof.
    75. Re:Excellent commentary... by Luke-Jr · · Score: 1

      ActiveX has never been a standard... no sane web developer would ever use ActiveX on their website...

      --
      Luke-Jr
    76. Re:Excellent commentary... by koreaman · · Score: 1

      Your entire argument revolves around the supposed evils of things that can do what you want, if you let them.

      Oooooh, wow, I'm scared. rm -rf / can do scary shit to my computer too, if I let it. This is not much different.

    77. Re:Excellent commentary... by ddig83 · · Score: 0

      "What does ActiveX do that XPCOM and Java are incapable of performing? If you are not able to answer that question, you have no base to stand on."

      How about millions of dollars invested capital? Whether or not XPCOM and Java do it is irrelevant. The question is, 'Are companies going to spend millions to redevelop ActiveX dependent applications?' I can tell you right now, (unless you show a significant cost benefit of switching) the answer is no.

    78. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      ActiveX has the same problem as root; The lack of separation of concerns.

      A policy of 'only whats needed' applies to both programs' and users' rights to commands and portions of the file system, the same way a firewall clamps down on ports for applications.

      ActiveX gives programmic access to applications under one hood, which is more insecure, because every program shares in every other programs' holes in this trusted environment.

      An Attacker might use one program to send bad data to the other program, to gain access to a home PC. He only needs one point of access to attack the hole. Now he has many chances for access.

      Right now, I'm on windows and this user has 0 permissions. WHy? I'm using IE, and IE has access to Excel, word, the command prompt, the registry, to everything, if it has the rights. Don't believe me? By a VB Book for ActiveX. ActiveX runs under the rights of the user running the program. My user can't change the time, let alone anything else.

      Further, to understand ActiveX's standing, you have to look at the other environments.

      Look at MAC's. They have the cocoa language, and whole bunch more.

      Linux has scripting for the Gnome Desktop and the KDE Desktop, and many other scripting languages.

      Sun's Java Desktop has the java enviro, ECMA, and a few others that I don't know yet.

      Open Office allows scripting to ECMA scripting, and plugins for Gnome, KDE, and MAC (cocoa).

      You probably think the market is fragmented, and you are right. However, it is a very tough thing to develop interoperability and provide only what's needed for each function to run. ActiveX opens the door very wide, and it's more convenient to work with, but that's insecure.

      Microsoft owns 9x% of the desktop PC's.
      It was nice for MS to make a scripting language for their office suite, Internet browser, email client, and file system interface, which interfaces with the rest of the operating system, but this is very insecure.

      The one diff between ActiveX and root is that if it's you running root, atleast at the time you sit at the console, you trust yourself not to destroy things, whereas ActiveX runs on compiled code and you can't see everything running.

      I take that back. An old mozilla unpatched with an unpatched RH 5.2 will probably get you a root kit, albeit probably not within the 30-45 seconds the same unpatched XP on either an infected windows domain or unfire-walled internet connection.

      Notice, that neither mentions running as root or as admin, because the attacker doesn't need you to? Well. In both windows and unices there are ways to limit the shell access to a point where accessing commands to get to these levels would be impossible. If you go through all that, and run as root/admin, you shoot yourself in the foot.

      So, if atal possible, I never run anything as root/admin, and never run activeX, unless I have an app that requires it, because both allow evil changes to the system.

    79. Re:Excellent commentary... by masklinn · · Score: 1
      it was feared that Compuserve would abuse it's power on all platforms
      Actually it wasn't Compuserve who owned the patent but Unisys. They got friendly with Compuserve because Compuserve was the introducer of the technology (in 1987)
      On the other hand, PNG has surpassed GIF's by adding alpha layer transparency... in other words, you can have certain pixels that are 100% opaque, or 10%, or 55% solid, or whatever.
      That's far from the only thing PNG has:
      • Better compression than GIF on 95% of the images you can find (try recompressing your GIFs with the impressive PNG Optimizer if you don't believe me), usually around 20%
      • Ability to handle many color formats (8bits and 16bits grayscale, 8bits palletized or 24bits truecolors)
      • 8bits alpha channel (progressive alpha is also supposed to be avaible on palettized PNGs)
      • gamma correction
      • Interlacing (much better than GIF's)
      • File integrity check
      The only thing the PNG is currently missing is the equivalent to animated GIFs, which happens to exist as the MNG file format but barely has implementations avaible
      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    80. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      Microsoft's platforms are, typically speaking, much more robustly documented than free software projects.

      Even if you look at the "undocumented" kernel books for NT, or those for writing for the WDM, and then look at the handful of Linux kernel books you'll see just how much better and stable the documentation is, and it's hard to get much more proprietary than NT.

      There's lots of in-depth technical documentation about COM, DirectX, the ATL, and even MFC, whereas the documentation for XPCOM is a bit spotty, SDL doesn't even contain a lot of the functionality of DirectX (least of all its documentation), and Gtk+ has about as poor a collection of documentation as can be found. (To stave off comments: Qt is of course better in that regard, and also the primary product of a software development company)

      Mozilla used to be pushed as a platform. That's where tools like Komodo come from. Somewhere that goal kind of died. Probably because Mozilla stopped being funded by AOL, or maybe because pretty much no one was interested in using it. The platform is pretty-much as memory-hungry as the browser itself.

    81. Re:Excellent commentary... by masklinn · · Score: 1
      But it's very very hard to sort through it all. Every so often I start playing with various features common to Mozillaish browsers like XPI, XSLT, and Javascript. It always strikes me how much potential there is to make some very cool applications using these. One pet project of mine is to see if I could create a set of XSLT documents that would transform glade projects into XUL applications, which could be themed via css.
      XPI is the only feature of Mozilla browsers.
      XSLT is a W3C standard, and supposed to be handled by the MSXML parser (which in fact often fails at that on the most complicated rules, better use a standalone parser such as Saxon or Xalan)
      Javascript is an ECMA standard (ECMA-262) and supposedly uses the DOM, first created by Netscape (DOM-0) and standardized by W3C as DOM-1 (and DOM-2 soon) and DOM Events. Here again, MSIE is supposed to handle them but fails 9 times out of 10 (while Moz browsers "only" fail half the time)

      BTW if you're looking for informations on Javascript/ECMAScript (and you don't already know about that website), you should visit Quirksmode, it's one of the most extensive, complete, precise and fully true Javascript resources I've ever seen. And it's got some CSS infos on top.
      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    82. Re:Excellent commentary... by masklinn · · Score: 1
      Um yeah. A Windows user trying to switch to FireFox or OO doesn't give a flying fuck if AX will work on Linux or not. Pardon my bluntness, but you're rationalizing NOT putting a feature in that some people need. That's bullshit.
      The worst issue of ActiveX (the one that you missed, and that prevents AX to be implemented in his project by any sane dev) is that it has no security AT ALL.

      MS didn't disable ActiveX in WXP SP2 for the fun, they did it because half MSIE's security issues come from ActiveX...

      Implementing ActiveX is not only dumb, it's harmful for god's sake, it's like opening your door and putting giant neons saying "hey guys, my door is unlocked, i'm gone for two months and i have no alarm or security device, but please don't come break in my house thanks".


      Oh and BTW i'm pretty sure there is an ActiveX extension for Firefox (remember, The Firefox Rule: if it isn't avaible out of the box and can be interresting, someone has done an extension for it... Doesn't fail often, that rule of thumb)
      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    83. Re:Excellent commentary... by Curtman · · Score: 1
      I was with you right up to:
      • "
      • Gtk+ has about as poor a collection of documentation as can be found"


      The GTK+ documentation is awesome. It's exactly like what XUL needs.
    84. Re:Excellent commentary... by masklinn · · Score: 4, Informative
      I think there are enough people out there who have to use ActiveX that support will eventually be added as a special module or something.
      Too bad you don't think like the Mozilla.org foundation does.

      It's been stated repeatedly that Mozilla.org products will never implement ActiveX out of the box... ever...

      There are extensions, if there weren't you could develop them, it's up to you to implement ActiveX in moz/fox and degrade your security, but THAT won't come from the foundation.

      Try again.
      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    85. Re:Excellent commentary... by masklinn · · Score: 1

      The API between ActiveX and the browser, or plugins and the browser, or ActiveX and the OS itself (hint: Mozilla plugins cannot toy around with your OS, they're restricted to the OS itself and even in the OS they can't do whatever they want. And clearly can't install themselves silently as the default setting.)

      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    86. Re:Excellent commentary... by timmarhy · · Score: 1

      mate, any site or application thats uses AX or FLASH for that matter, i refuse to use or visit. the only way to get through to developers and companies that these technologies are NOT good, is to vote with your wallet and feet.

      --
      If you mod me down, I will become more powerful than you can imagine....
    87. Re:Excellent commentary... by Curtman · · Score: 1

      XSLT is a W3C standard

      Which comes with every mozilla browser via the TransforMiiX module.

      Javascript is an ECMA standard

      Which is what Mozilla's javascript implementation is based on. To quote the home page, "Netscape's JavaScript is a superset of the ECMA-262 Edition 3".

      standardized by W3C as DOM-1 (and DOM-2 soon)

      Which the Gecko DOM resembles.

      That's my point. What's yours again?

    88. Re:Excellent commentary... by masklinn · · Score: 1
      Ignoring it and hoping it goes away is not the right answer, either.
      And is still the one chosen by the mozilla foundation.
      Thank you, drive thru
      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    89. Re:Excellent commentary... by masklinn · · Score: 1

      My point is that (tadaa) XSLT and Javascript are not "Mozilla features" as in "something they have others don't", they merely are standards that have actually been implemented in Moz

      But maybe HTML/XHTML and CSS are "mozilla features" to you, in which case i'll agree that XSLT and JS are, too.

      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    90. Re:Excellent commentary... by urbaneassault · · Score: 1

      Why did this get modded down?
      The parent is 100% correct on this. When IT dev costs are at an all-time high and most projects go over in time and money, existing apps aren't going to be rewritten unless they really have to. If a company has already dumped $1M into their java, VB, COBOL, ActiveX, or whatever codebase, they aren't going to rewrite the same apps unless support is dropped or costs are too high. End of story.

    91. Re:Excellent commentary... by LittleBigLui · · Score: 1
      Make a downloadable application that runs seperately. I even dislike Java's WebStart technology.


      Java WebStart is a mechanism for packaging applications that run separately.
      --
      Free as in mason.
    92. Re:Excellent commentary... by Curtman · · Score: 1

      they merely are standards that have actually been implemented in Moz

      I see.. So you're saying that support for something isn't a feature. That's interesting.

      But maybe HTML/XHTML and CSS are "mozilla features"

      Yes.. XHTML is a feature of Mozilla. It's not exactly one of IE's. CSS to a certain extent as well.

    93. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      CompuServe finally settled on the less unreasonable 5c per paid application that can encode GIF's, with no fee for decoders. That fee is no longer with us, as the patent has expired.

      the known unisys patents (and releated 'gif tax') may have expired in 2004 or earlier, however, there is a duplicate patent (all praise the uspto!) owned by ibm that does not expire until 11 august 2006 (4,814,746).

      ref:
      http://www.gnu.org/philosophy/gif.html#venuenote
      http://www.kuro5hin.org/story/2003/6/19/35919/4079

      and unisys' web site states they have patents pending on 'improvements' to the now-expired patent.

      ref:
      http://www.unisys.com/about__unisys/lzw

      so who knows if a simiilar fiasco will pop up again in the future.......

      one industry does that all the time. when a profitable and popular prescription drug is about to expire and get in the hands of (legal) generic manufacturers, a "new" version of the drug is "invented" (typically the same exact forumla in an 'extended-release' form, which changes the inactive ingredients, not the drug itself) and that is enough to keep the "original" forumla from going generic.

    94. Re:Excellent commentary... by MPolo · · Score: 1

      Crossover Office at least claims to have a working Active-X plugin. I have personally never had the need to turn it on (and it is turned off by default), so I can't comment as to how well it works. That might be a possibility for you. Of course, you have to pay something to use Crossover.

    95. Re:Excellent commentary... by jnf · · Score: 1

      that some people need.
      mod me a troll, but seriously if you _need_ activex, you need a new programmer.

    96. Re:Excellent commentary... by marsu_k · · Score: 1
      NVidia and ATI drivers are good examples of binary installs. I have yet to see these incorporated into any distro though. Do you know of one?
      Actually, they can be found in quite many distributions. I recon most non-free (as in beer) desktop-oriented distributions include these, at least Mandrake (or Mandriva) includes them in the sold CDs, not in the download version though. My current distro (Arch Linux) has Nvidia drivers in the extra repo, upon install it shows where the licence is located and asks you to remove the driver if you don't like the licence.
    97. Re:Excellent commentary... by Taladar · · Score: 1

      Sadly, sane web developers are a minority.

    98. Re:Excellent commentary... by Taladar · · Score: 1

      It is a feature but since it is a standard it doesn't have to be documented completely by Mozilla.

    99. Re:Excellent commentary... by Taladar · · Score: 1

      See it as a way to force people to abandon ActiveX. The easier we make it to stay with ActiveX the more users don't even consider using something else.

    100. Re:Excellent commentary... by PJBonoVox · · Score: 1

      Oh and you can't add it to Linux products. Why does that matter? What products do you know of that currently use ActiveX and run under Linux? None, probably. So what's the difference?

    101. Re:Excellent commentary... by Taladar · · Score: 1

      While I agree with you I think we must think of a new analogy for "voting with..." for the web. Most web sites don't cost money ("wallet" doesn't fit) and the web doesn't require you to walk anywhere ("feet" doesn't fit either).

    102. Re:Excellent commentary... by Narchie+Troll · · Score: 1

      Why is it NEEDED? If you want ActiveX, you can use IE on Windows and risk the security complications, many of which are directly due to ActiveX.

    103. Re:Excellent commentary... by Taladar · · Score: 1

      ActiveX is the same as having a system where websites can execute shell commands, you don't know which one they will execute before giving them permission and one of them runs "rm -Rf /" with you running as root because half of the other shell commands legimitate websites use don't work as normal user.

    104. Re:Excellent commentary... by DrXym · · Score: 1
      Great... how does it work in linux?


      I think the crossover plugin has support from some ActiveX controls via a plugin. Or whoever wrote the control could port it via winelib. Though of course someone could turn the question around and ask - does a natively compiled Linux application run on Windows?


      AxtiveX is the Microsoft API in answer to Java. It lacks sandbox execution so any ActiveX embedded code can trivially access anything on Windows. The reason even microsoft doesn't think you should blindly execute any activeX component is that it is equivalent to a local binary in terms of system access.


      Making it absolutely no different from any other binary format - a.out, ELF etc. ActiveX is unsafe for untrusted HTML content, but it is a perfectly legitimate and safe (as safe as any other binary format) for producing apps that run locally.


      The problem is not ActiveX controls but the manner in which IE fetches and installs them from the Internet. Once that is switched off or restricted to the intranet, there is no safety issue.

    105. Re:Excellent commentary... by Curtman · · Score: 1

      It is a feature but since it is a standard it doesn't have to be documented completely by Mozilla.

      Yes, that's very true, but Mozilla the platform would probably be much more popular if it was. That's all I'm saying. It would be very nice if there was a central location to go to when you wanted to learn about any aspect of Mozilla like there is in the Microsoft world. A revived and improved DevEdge would do great things for Mozilla I think.

    106. Re:Excellent commentary... by Ost99 · · Score: 1
      Oh and you can't add it to Linux products.


      Why does that matter? What products do you know of that currently use ActiveX and run under Linux? None, probably. So what's the difference?


      You missed the context, he argued that ActiveX was not a proprietary extension since you could add ActiveX plugins to non-MS producs. That is only true for windows programs, and it doesn't change the fact that it *is* a proprietary extension.

      His point was that Firefox should have ActiveX support, and my point was that his approach would only work with the windows version (using a MS plugin). The new Netscape does this.

      - Ost
      --
      ---- Sig. gone.
    107. Re:Excellent commentary... by petermgreen · · Score: 1

      hmm your list of color depths for png was a little incomplete ;). 1 2 4 8 and 16 bit greyscale 1 2 4 and 8 bit palletted 24 and 48 bit truecolor. note: theese values are in bits per pixel because thats what most editing software uses. The PNH specification actually talks in terms of bits per channel. greyscale and truecolor images can have an alpha channel of the same depth as one of the main channels (ie a 24 bit rgb image can have an 8 bit alpha channel) greyscale and truecolor images can have a single color declared as transparent (usefull if you don't want the overhead of an alpha channel and only need binary transparancy) pallette entrys are 24 bit rgb. They may also have an alpha value associated.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    108. Re:Excellent commentary... by PJBonoVox · · Score: 1

      Ah, I see, my apologies. I'm a victim of posting without reading the thread properly.

      ...and also a victim of posting HTML formatted comments without any tags since I didn't realise Slashdot works this way.

      I ahve to say I'm glad ActiveX is possible in Mozilla for those with a little knowhow and not 'generally' available. Some corporations really need it to 'make the switch'.

    109. Re:Excellent commentary... by petermgreen · · Score: 1

      GRR i left it set to html formatted so it crunched out the line breaks in that post sorry

      i also missspelt png in png specification

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    110. Re:Excellent commentary... by DrXym · · Score: 1
      (1) It does not work cross-platform. Both Firefox and OpenOffice work on platforms other than Windows. Both platforms keep this compatibility by not introducin technology that could possibly limit this capability.

      Plugins are not cross-platform either. The API might be more or less platform neutral (although there are differences), but everything behind that API is absolutely platform-specific. A flash plugin won't play on Linux if its compiled to Windows because the small part dealing with the API (which could be ActiveX or NPAPI) attaches to a huge other part that is intimately tied to various GDI calls, multimedia etc.

      Neither is XPCOM cross-platform. Most of the interfaces and IDL are designed to be cross-platform, but the implementations of those interfaces are natively compiled and at the lower levels make platform specific calls. For example nsILocalFile is implemented multiple times, once for each platform. Even Gecko's use of XPCOM only covers so much ground - once you start looking at the graphics / rendering aspects of Gecko you realise XPCOM has given way to a mess of pseudo-XPCOM (with ref counting abuse and handwritten C++ interfaces) and native C++ classes. This is definitely not for public consumption and definitely not portable either.

      Secondly, both OpenOffice and Firefox do support platform specific features. For example, the native theme engine, drag and drop, clipboard. On Windows, Firefox supports shortcuts, mime types, DDE launching & printing, MSAA accessibility.

      There would be nothing to stop Firefox supporting ActiveX controls. It wouldn't have to impact on its cross-platform abilities in other areas. In fact the Netscape 7.2 browser actually does support certain ActiveX controls so its quite possible to achieve.

    111. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      "What does ActiveX do that XPCOM and Java are incapable of performing?"

      Citrix Deployment is much easier with the ActiveX plugin. ActiveX is able to install itself and run with almost no user interaction, which looks better to business types. Also, drive mapping is more difficult to set from the server side using the Java Client, as the user must set it up themselves. Printing becomes infinately more complicated as you must print "through" the Java software.

      Though I don't have any real disputes with your post, your link to the article where you corrected the gentleman before is broken.

    112. Re:Excellent commentary... by masklinn · · Score: 1

      Ok, I see your point now, thanks for the clarification.

      Well, the issue I have there is that you can get the documentation from their "real" sources (W3C for XSLT, ...), while the Microsoft source for "external" standards they used sometimes (often?) have the proprietary MS extensions embedded in them as "legid" (supposedly) features of the language, which means that you can hardly use them as cross platforms documentations.

      I'd much prefer links to the real source to a full central documentation that may not be up to date or may not stick to the standards.

      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    113. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      Oh well in that case ActiveX is no good because it can't run currently available XPCOM and Java objects that havn't been ported to ActiveX. Check.

    114. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      FireFox supports many things that IE doesn't, therefore IE cannot completely be an alternative to FireFox.

    115. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      "What does ActiveX do that XPCOM and Java are incapable of performing?"

      HOTMAIL and GMAIL (now, it just changed to use Active Scripting apparently unless my webbrowser's lying to me, Opera8 final) for starters!

      They're BOTH "killer web apps" too, mind you... they're email! Email is, after all, THE 'killer' web app even today.

      What about ActiveDocuments?

      They're what MOST/MANY VB5-VB6 apps can be turned into so they run via a webbrowser as a corporate INTRANET style app!

      (and VB apps? Made with those two versions of MIcrosoft compilers?? Is OUT THERE HUGELY running Info. Systems worldwide! Doubtless millions of them that are CUSTOM apps, not out-of-the-box turnkey solutions. Ones built specifically around custom data & IT/IS/MIS systems and customized backend office setups on SQLServer, Oracle, or DB/2)

      * There's a few for ya!

      APK

      P.S.=> If the person here can show you various apps that have .ocx dependencies? He has your answers WITH EXAMPLES like these...

      (Sure, you can/could take the time AND MONEY to rebuild those apps into analogs that do some of the same via the other programming tools you mention, but would you? Could you absolutely guarantee 100% functional analogs?? Would the money minders/numbers crunchers allow it?? Doubt it... if a watch runs, why fix it would be their rationale. Unless you could show some HUGE security hole, I doubt it. Intranet bound apps are safe from Active X-ploits, provided the network security is 100% safe imo. It all starts with the smarts & saavy of your network admin & his staff of subordinate techs/engineers & the DBA there @ any company really, as well as the programmers who wrote said mentioned custom MIS/IS/IT apps!)... apk

    116. Re:Excellent commentary... by RidiculousPie · · Score: 1

      does a natively compiled Linux application run on Windows?

      Yes, it can, using coLinux, provided it was built for x86 linux.

      Or it could be ported using MSYS, or Cygwin.

      --
      ah, mod points ... now where is my crack?
    117. Re:Excellent commentary... by DrXym · · Score: 1

      So in other words you're saying host an entire Linux session to get an app to work or recompile it. How is that any different from running an ActiveX control under VMWare, or recompiling it against winelib?

    118. Re:Excellent commentary... by PHP+Addict · · Score: 1

      Does anyone really want that, other than the companies who have embraced and extended Microsoft's Active-X? Is there anything truly significant you can do with Active-X that you cannot also do with Java?

      Unfortunately, it's not what can or can't be done with ActiveX or Java that's usually the issue. It's what has already been done. After IE flattened Netscape in the browser wars a few years ago, large numbers of companies began mass-producing ActiveX components, thinking that the vast majority would be using IE for a long time to come. Thankfully, that is not the case; however, it now leaves oh so many people stuck using Microsoft's inferior, non-standard technology, regardless of their opinions on either browser.

      As a student studying for a BS in Software Engineering, I hear stories from countless professors and lecturers about systems still using long-replaced languages and technologies. The Legacy Code isn't going anywhere, much to my dismay.

      Finally, as an employee of a web development firm, I would love to see IE wiped off the map (or at least adhere to a f!@#ing standard, but I'm not sure which one is more likely). Unfortunately, this isn't going away either. Cross-browser and cross-platform testing incorporates about 25% of the development time of a project. It's also the least favorite part of my job, but it's absolutely necessary.

      <flamebait>I, for one, would be all for some sort of ActiveX sandbox extension (extention, as in NOT initially included), much like Java. And, if it were written in Java, it could even be portable to other platforms *gasp*.</flamebait>

      In an attempt to mitigate the flames inevitably coming my way regarding that previous paragraph, I'd like to point out that I absolutely HATE non-standard technologies and would send them all to an eternity in the cleansing fires of the sun if it were in my power. I'm simply trying to propose a solution to those others out there, (like me :{ ) who are constantly haunted by incompatibilities.

      Have a nice day.

      --
      Laziness, check. Impatience, check. Hubris, double check!
    119. Re:Excellent commentary... by JoloK · · Score: 0

      Boy, you're really confused, huh? Why should ANY other browser mire themselves down in the muck that Microsoft created called Active X ? Please, no!! It _is_ better that they don't support it -- Active X; are you serious?

      --
      JoloK
    120. Re:Excellent commentary... by SolusSD · · Score: 1

      ActiveX is a problem that is best fixed by not using it. It is the leasing exploit in Internet Explorer. I work for a company that *still* develops all of its web apps using activex. Not only are they tying our customers into using Internet Explorer, they are promoting a broken, buggy platform, activex. The best thing for any company to do is stop. stop using active X NOW. Trust me, I'm sure firefox, as well as mozilla developers thought about it long and hard. they could have implemented active x if they wanted to (specification is open). Also, most people/companies greatly exaggerate the number of websites that depend on active x.. there really isnt that many. If a company isn't able to fully adopt Firefox, why not a partial implementation? At least that puts them on the right path.

    121. Re:Excellent commentary... by CastrTroy · · Score: 2, Interesting

      You can get firefox to use active-x. It just doesn't do it by default. There's some stuff you can change in your profile to make active-x stuff work. It's not a good idea, but it can be done. As for openoffice, well, I'm not sure there. But if running compiled code in your office suite is something you can't live without, maybe you need to review the reasons behind doing stuff like this in the first place.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    122. Re:Excellent commentary... by HangingChad · · Score: 1
      Lack of ActiveX support actually prevented my previous company from switching to OpenOffice or Mozilla.

      Lack of ActiveX support isn't what prevented the transition, it was lack of planning and foresight in application development. You built or bought applications without considering compatibility with other browser platforms, then blame those other platforms for not supporting you. Seems like the tail wagging the dog to me.

      If your application developers or IT management never bothered to ask themselves what would happen if another browser became popular or if IE wasn't around, the fault rightly rests with the decision makers, not the Mozilla developers.

      It reminds me of people who voted for Bush blaming the Democrats for not picking a better candidate.

      --
      That's our life, the big wheel of shit. - The Fat Man, Blue Tango Salvage
    123. Re:Excellent commentary... by orasio · · Score: 1

      OK.
      If you need activeX, you can run IE.
      There's no need that everybody switches to IE.
      The enterprise that relies on IE+ActiveX does already have many security issues to care about, and many of the improvements of Firefox over IE would be redundant.

      For me, Firefox is free (as in "freedom") and IE is non-free (as in "proprietary"). It's more secure by default. It's more streamlined, easy to configure. Runs in Linux.

      (Of course, it has lots of extensions that I need for development, but I'm trying to think as a user)
      Those are many of the reasons I like Firefox.

      Maybe if I relied on ActiveX, I would like that FF could handle ActiveX!

      But wait!
      Firefox + ActiveX is non-free. It's non-secure.
      It would be another feature to configure (security zones?). It is Windows-only.

      I already have a product with those characteristics, and it's called IE!!

      So, there's no need for Firefox + ActiveX.
      Maybe as an abscure extension that the mozilla developers don't have to support, but I don't want ActiveX in my browser!!

    124. Re:Excellent commentary... by 4of12 · · Score: 1
      If Microsoft can't get away with being arrogant, than the OSS Community can't either.

      Worry not! OpenArrogance 1.2.1 has been under active development for years with a strong developer community. It has a broader feature set than VisualArrogance 2005 Pro and costs less, too!

      --
      "Provided by the management for your protection."
    125. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      > It is a feature but since it is a standard it doesn't have to be documented completely by Mozilla.

      That's exactly the problem. Imagine if msdn did not include documentation for stdio because it is standard, or stdc++, or opengl, or anything standard...

      Beeing standard is cool. But is not a replacement at beeing documented. Even if the doc is just a pointer of the standard and explanation on how to use it within the mozilla framework...

    126. Re:Excellent commentary... by PIBM · · Score: 1

      Notice that (as was said before) the patent expired a few years ago in most country and anyway, they were only going after "compressor", meaning apps that allow to save a file as a gif, starting from another file type. from gnu.org: The Unisys patent expired on 20 June 2003 in the USA, in Europe it expired on 18 June 2004, in Japan patent expired on 20 June 2004 and in Canada it expired on 7 July 2004. The U.S. IBM patent expires 11 August 2006,

    127. Re:Excellent commentary... by saforrest · · Score: 1

      Too bad you don't think like the Mozilla.org foundation does.

      When I said 'special module', I was thinking 'third-party add-on', not necessarily anything from the Mozilla Foundation.

      If the Mozilla Foundation doesn't want to support ActiveX, more power to 'em.

    128. Re:Excellent commentary... by saforrest · · Score: 1

      What else would it be mostly driven by?

      Some notion of what is needed to supplant proprietary software.

      Probably, somewhere, there's some proprietary software package that is 1) necessary, but 2) incredibly boring to write and support.

      So no open-source equivalent exists because no one's bothered to write it, because it isn't the private hobbyhorse of any volunteer developer.

      This isn't criticism of open-source contributors per se. It's their damn free time, so why should they spend it any other way?

      But clearly they donate their time for other reasons than the joy of coding: to popularize open-source software, and there is likely some low-hanging fruit in the 'boring software' category which could be written.

      However, I think that before this happens, we'll need more institutions to organize the writing of such software.

    129. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      It can run ActiveX applications that have already been designed and implemented.

      Like viruses?

    130. Re:Excellent commentary... by EXrider · · Score: 1
      What does ActiveX do that XPCOM and Java are incapable of performing?


      Pfft! ActiveX can format drives, all from the convenience of visiting a web page! http://www.uic.edu/depts/accc/newsletter/adn20/act ive.html Let's see Java or XPCOM do that!
      --
      grep -iw skynet /etc/services
    131. Re:Excellent commentary... by hesiod · · Score: 1

      Vote with your mouse.

    132. Re:Excellent commentary... by Foolomon · · Score: 1
      If you are not able to answer that question, you have no base to stand on.

      Your argument would have had more weight if you had said: If you are not able to answer that question, all your base are belong to us!!!1!!1!

    133. Re:Excellent commentary... by masklinn · · Score: 1
      That's exactly the problem. Imagine if msdn did not include documentation for stdio because it is standard, or stdc++, or opengl, or anything standard...
      Oh my god, you'd have to ...
      look for documentation !!!
      and the real one, that is, like real documentation... or man pages... or something !!!

      Come on, do you think that the MSDN is the only fucking C/C++ documentation out there?

      (oh, and OpenGL documentation in the MSDN, please, if you want to troll do it well, hell would freeze over before you found OpenGL documentation in a Microsoft turf)
      --
      "The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
    134. Re:Excellent commentary... by Zaiff+Urgulbunger · · Score: 1
      I can see how this is an issue to you on a practical level, however, I would have to ask _why_ you want to migrate (presumably) from MSIE/MS-Office to Moz(or Firefox)/OOo?

      If it is for the security benefits, then I would hope you could see that these benefits exist in no small part _because_ the FOSS community ignores technologies that are inferior to existing FOSS implementations, or that have legal issues that prevents their use.

      The other reason you might wish to switch from MS or FOSS could be to save money on licence fees. In which case, I feel you pain. But by the same token, I'm sorry to say that that's life! You made your own bed and now you've got to lie in it (or possibly your predecessor did... but same difference!).

      So back to the practical, if you have that much code that requires ActiveX you can either:
      • Create wrappers for the ActiveX components. I'd imagine this would be easy in Moz/Firefox but might be harder in OOo.
      • Use the MS and FOSS products side-by-side. Again, using IE for intranet and Moz for extranet might be easier than using MS-Office and OOo side-by-side.
      But basically you need to look at where you are now and where you want to go. If the MS software is working for you then maybe you stick with it until you such time as a major revamp of your existing code. Otherwise, look at how you manage a migration. But either way, take care with your technology choices.

      Best of luck!
    135. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      Hey, slow down, cow boy.

      > Come on, do you think that the MSDN is the only fucking C/C++ documentation out there?

      First point, I seldom use visual studio. Second point, I hate msdn, as it is a loosy documentation (okay, java doc is worse, but beeing better than java doc is too easy). Inside Mac ruled.

      But, when I use visual to do something, I am happy that *all* the doc is avalaible, from the Win32 SDK to libc++, DirectX or OpenGL. Yes. Opengl.

      > (oh, and OpenGL documentation in the MSDN, please, if you want to troll do it well, hell would freeze over before you found OpenGL documentation in a Microsoft turf)

      I have a bad news for you.

      From MSDN, October 2001:

      Platform SDK: OpenGL
      GL Functions
      The OpenGL commands, in alphabetical order are presented here. Each reference page describes one or more function. See also GLU Functions.

      Or, if you prefer, type 'msdn opengl' in goolge and go to the first link.

      MS doc sucks by many aspect, but it is much better than having to look at 10 different documentations.

      So, who is trolling who ?

    136. Re:Excellent commentary... by fyngyrz · · Score: 1
      Many times, market forces leave older technologies high and dry. That's just the way it is. You may have invested thousands of dollars in Betamax tapes or in 8-tracks, yet you cannot buy a new player today. You may have bought a DeLorean, or a Next, or a video disc player. On the company front, investments have been made over and over again in technologies that didn't stay around. Motorola's comm satellites, C-band television broadcasting, etc. Just because a technology exists and people are using it is not sufficient reason to keep it around. This is established fact.

      But that's not exactly what is going on here. In this particular case, the technology is dangerous and it is causing trouble all the time. Major trouble; time lost, data lost, people's lives are being disrupted.

      Now, you propose to sandbox it so that for FF, the technology becomes non-risky. In the meantime, the IE userbase continues to suffer the slings and arrows of Microsoft's incompetence (and mind you, there is no assurance that a "sandboxed" version of an Active-X control would work properly -- because they can access anything in the system, they are likely to have accessed someting in the system, as the Windows API is like a candy store -- why would you rewrite that functionality if you didn't have to?)

      If I were to compare Active-X to a product, I think I'd choose thalidomide. Everything looks fine when you first take it. It seems to be doing what it is supposed to be doing. Until your PC pups out some malformed spyware, that is. Or simply dies without warning. They took Thalidomide off the market. They should do the same with Active-X. It is toxic.

      My position is that giving Active-X a longer life span is a bad choice for developers to make. They should work to kill it sooner, rather than later. In the meantime, any company with half a wit will be working to get out of the technological jail they so blindly walked into. Maybe next time, they'll Think, Marketplan and Develop instead of Marketplan, Develop, and finally Think.

      --
      I've fallen off your lawn, and I can't get up.
    137. Re:Excellent commentary... by PastaLover · · Score: 1

      There is no patent-encumbered technology in Linux since that would contradict the purpose of the GPL license. Are you talking about Linux distros? I am aware of patented technology in these. NVidia and ATI drivers are good examples of binary installs. I have yet to see these incorporated into any distro though. Do you know of one?

      I would bet you that there is plenty of patent-encumbered technology in "linux". (I suspect you are talking about the kernel?) Only a couple of months ago there was an article which claimed there were quite a number of techniques used throughout the linux kernel that were patented. (I think they had a list of about 90 patent violations, not entirely sure) Lots of other GPL'ed software incorporates patent-encumbered techniques (for instance, about every video encoding library)

      As for nvidia and ATI I was under the impression Suse (now part of Novell) distributed them on their installer cds. Not quite sure though. Most distros only seem to provide installers. But in general, any propietary app can be redistributed with a distro, as long as it license doesn't prohibit redistribution, and some other things. I believe the reason Debian doesn't carry the java binaries from Sun is because the license requires you to indemnify for instance. But the propietary nature of the software is not the reason not to redistribute it.

    138. Re:Excellent commentary... by Anonymous Coward · · Score: 0
      The API between ActiveX and the browser, or plugins and the browser, or ActiveX and the OS itself (hint: Mozilla plugins cannot toy around with your OS, they're restricted to the OS itself and even in the OS they can't do whatever they want.

      What do you mean they're 'restricted to the OS itself'? I can't make any sense of that comment. (Sorry!)

      Please correct me if I'm wrong, but as far as I know, a Mozilla plug-in is just a shared library which is mapped into the browser process, no different in that respect to an ActiveX control. In both cases, the library provides entry points which have to be called by the browser, but once one of those entry points has been called, the library code can do whatever it wants: it's native code, executing with the same privileges as any other code in that process.

      The above problem is why the promise of Java (or .NET) is so compelling. If Mozilla plugins running native code are really restricted, that's quite impressive, but I'm sceptical. If it were so easy to restrict native code executing with a given process context, there would be no need for Java.

      And clearly can't install themselves silently as the default setting.)

      MSIE doesn't allow ActiveX controls to silently install themselves by default either. Maybe it did in the past; I don't know, but I just created a new user in XP (SP2), and checked the Internet settings: the user is prompted to install signed ActiveX controls (ie the user has to explicitly install them), and unsigned controls are disabled completely (ie they can't be installed at all).

    139. Re:Excellent commentary... by xilet · · Score: 1

      Well that is exactly it. Most open source devs are not paid to do this, they do it for the love of what they are doing [or occasionally egotism, but hey at least its done in a positive way]. If they are not enjoying what they are working on why would they do it. Get a corporation/government that wants to get off of M$, support folks to put it together.

    140. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      It wasn't Compuserve that held the patent, it was Unisys.

    141. Re:Excellent commentary... by jonadab · · Score: 1

      > Your entire argument revolves around the supposed evils of things that can
      > do what you want, if you let them.

      In theory.

      So, then, do you want your web browser to provide a DOM method that runs rm -rf ~/ if the user clicks Ok on a dialog box that asks if you want to view enhanced content? What if it runs an arbitrary *nix command, specified by the site, if the user clicks Ok, so that it wouldn't *necessarily* be destructive, but easily could be? That, in a nutshell, is ActiveX.

      Of course, if the user wants to download an executable and run it, he always can. But the way ActiveX is presented is rather different from that; it is presented as a rendering component that is just part of the page, and the retrofitted approval dialog is fundamentally no scarrier than the one you get when you use a search engine ("Oh, horrors, you're sending unencrypted information to this website!") Even that dialog has been retrofitted; the original design was for ActiveX controls to be executed automatically. Even with the approval, all the user has to do is tap the Enter key or bump the spacebar, and the system is compromised. This is fundamentally different from choosing to download and deliberately execute something.

      Yes, I suppose that in theory, if ActiveX "support" were done in a way that required the user to expressly choose an "Execute ActiveX Programs" item on a menu somewhere in order to start the site's ActiveX controls, and then presented a warning dialog that ActiveX Programs can do whatever they want, and you should only run them if you fully trust the site, and if the buttons were "Trust this site" and "Keep me safe", with the latter being the default, then that could be construed as secure. However, that level of support would not make ActiveX advocates happy (I can hear them now: "Why should the user have to click a menu item?!? Some users wont click it!! I need my sitez ActiveX to run no matter what!!1 Tell me how to do that on my site, or Foxfire sucks!!"), and few other people care (except the people who are actively against it), so it would be kinda pointless.

      Let me just close by saying that ActiveX is a *mistake* Microsoft made, *not* a feature. They've attempted to cover over the worst of the problem by retrofitting user approval, but eventually they will discontinue ActiveX support entirely. It is somewhat deprecated already in SP2, will be fully deprecated in Longhorn (which will include a replacement technology) and eventually discontinued, probably in a security-oriented service pack. Even a feature-for-feature clone of Microsoft software (and neither Mozilla nor OpenOffice is intended to be that) would not probably choose to implement ActiveX, unless it were intended to be an *exact*, bug-for-bug type of clone.

      --
      Cut that out, or I will ship you to Norilsk in a box.
    142. Re:Excellent commentary... by Anonymous Coward · · Score: 0
      ActiveX is actually useful in some cases, eg for developing bespoke applications for use within secure corporate networks, where the client machines are locked down, and managed by policy.

      Using ActiveX on the Internet is highly questionable, but the same applies to all native-code plugin architectures. The basic problem is that security granularity for native code is at the process level, so any native-code plugin (whether it uses ActiveX or NSAPI) automatically gains all the privileges of the host browser process.

      Microsoft mitigated this to some extent by adding a signing mechanism to ActiveX, which allows users to install ActiveX controls from trusted sources, but not others. Although it helped, there are a lot of signed ActiveX controls out there, and all it takes is one buggy control to compromise the browser process. (The same applies to plugins using NSAPI of course, and as far as I know, there isn't any signing mechanism for them, so they're theoretically even worse.)

      Java and .NET implement sub-process security mechanisms via the JVM and .NET runtime, which allow different code within the same process to run with different privileges. Because of this, they're much more suitable technologies for use in an Internet environment than native-code plugins (either NSAPI or ActiveX). To the extent that they satisfy corporate demands as well, they should be general purpose replacements for native-code plugins (both ActiveX and NSAPI).

    143. Re:Excellent commentary... by |<amikaze · · Score: 1

      Pardon my bluntness, but you're rationalizing NOT putting a feature in that some people need. That's bullshit.

      That's what you have to do with any project. If you can find a rational reason not to do something, and it outweighs the reasons to do it, then you make the decision to not. In this case it seems the developers have considered platform neutrality to be more important that implementing ActiveX.

    144. Re:Excellent commentary... by redheaded_stepchild · · Score: 1

      How about...
      ...allowing viruses & spyware onto end users machines without their knowledge or permission?
      ...clobbering all attempts at online security? (specifically, IE)
      ...useful only because the entire Windows Update service was made for it?
      ...allows die-hard MS junkies to gripe about how OSS is 'not standards compliant', even though ActiveX was never a standard?

      --
      Don't use the Troll mod just because you disagree with me.
    145. Re:Excellent commentary... by Senzei · · Score: 1
      What does ActiveX do that XPCOM and Java are incapable of performing?

      Work with an already developed in-house ActiveX plugin used for a critical business purpose.

      Not using the plugin is a non-option, and there is (according to management) not enough value in FF/Mozilla to warrant all the effort required to make the switch in the first place.

      Actually your answer here brings up another point. ActiveX works in IE and Office. Which means that, theoretically at least, the same plugin could be used in both applications. AFAIK XPCOM objects are written in C/C++. That would mean parallel development if you needed to use both applications. Hell even if you could write XPCOM in java the hooks are entirely different, which still means at least partial duplication of effort.

      You are asking people to give up a known working solution in favor of TWO ways to do the same thing that will have to be developed, debugged, and implemented. If either FF or OO are going to take off they have to start by providing what people already have, and extend on it with better features. If you don't have the first part the switch will not happen in the corporate world.

      Only on slashdot could the guy that avoids half another person's points be the one that does not get troll votes.

      --
      Slashdot: Where anecdotes and generalizations can be freely substituted for facts, logic, or intelligence
    146. Re:Excellent commentary... by mattyrobinson69 · · Score: 1

      hmm - how about an activex plugin that only allows activeX plugins from a configurable subnetmask that is hardcoded to allow everything accept 0.0.0.0 (the whole world - therefore useless to home users but not to corporate types).

      Or maybe disabled and only enableable through a config file (that would scare most people off but could be distrobuted corporation wide as part of an MSI)

    147. Re:Excellent commentary... by mattyrobinson69 · · Score: 1

      I think it would be more likely for it to do this

      cd /tmp
      wget http://spamsite.com/downloads/openrelayinstaller.s h
      sh /tmp/openrelayinstaller.sh
      openrelayd
      renice -20 openrelayd

    148. Re:Excellent commentary... by RedBear · · Score: 1

      Instead of ranting and raving about open source, CROSS-PLATFORM software not supporting some proprietary junk that only works on a single platform, maybe you should be ranting and raving at your organization for locking itself into ActiveX in the first place, and doing your best to CONSTRUCTIVELY move them toward a better solution for the future. We realize that it would be all cute and convenient in the SHORT TERM for you to be able to use ActiveX in Firefox, but in the LONG TERM we would be damaging open source and computing choices in general by maintaining AX as if it were a viable choice. The fact that it doesn't work in non-Microsoft environments is an incentive to not use it in the future. I'm sure I'm not alone in the opinion that this is a positive thing.

      I think the problem you're having is that many of us geeky open source people are "big picture" folks who put the ideology of freedom of choice and compatibility ahead of everything else because we've seen (or foreseen) the consequences of what happens when you lock yourself into proprietary crap. You're dealing with those consequences right now. Wouldn't you prefer it if future generations didn't have to? Use your energy to convince your organization to move away from ActiveX right now instead of trying to convince open source developers and advocates that there is some benefit to perpetuating proprietary technologies as bad as ActiveX.

      I'm sure you realize that it's already possible to set up IE to always prompt the user for permission to run Javascript and ActiveX controls. That would be practically the same level of security as putting a plugin in Firefox. Also the annoyance of constantly having to respond to the prompts can help promote awareness of the fact that ActiveX is a security problem. When the users ask to have the prompts turned off you just explain to them in no uncertain terms that they are there to protect your computing infrastructure. If your supervisors tell you to disable the prompts and stop annoying people in IE, the same thing would happen with Firefox.

      So really, no matter how you look at it there is very little point to what you're trying to do, and it's NOT solely because everyone here is a bunch of jerks who like flipping off Microsoft. Some of us actually do think this stuff out, and as far as we are concerned supporting ActiveX is a bad idea. Now, if it's really that important to your organization, I'm sure you'd have no problem with paying some developers to make an ActiveX plugin for Firefox and make it work in OpenOffice. But it would almost certainly be cheaper in the long run and probably even the short run, to pay some developers to help you move away from ActiveX. Strike at the problem, not the symptom.

    149. Re:Excellent commentary... by NanoGator · · Score: 1

      "Instead of ranting and raving about open source, CROSS-PLATFORM software not supporting some proprietary junk that only works on a single platform, maybe you should be ranting and raving at your organization for locking itself into ActiveX in the first place..."

      One of the reasons behind the establishment of Mozilla was to unseat Internet Explorer as the de-facto browser and replace it with something that is cross-platform. By ignoring this feature of IE, they are making their jobs harder. By not supporting a non-cross-platform feature, they're slowing the growth of cross-platform software.

      Whether or not my previous company was acting intelligently or not, it doesn't have the slightest bearing on the behaviour of the Mozilla development team. It is not my nor my previous company's fault they intentionally chose not to support a feature that has some healthy use in the IE world.

      You can sit here and tell me I'm wrong all you want, but at the end of the day there are a lot of people wanting to leave IE and can't. Since the Mozilla team has the power to solve this problem, and won't, the fault is theirs.

      --
      "Derp de derp."
    150. Re:Excellent commentary... by NanoGator · · Score: 1

      If that's the attitude that the community is going to take, then they're no less evil than Microsoft.

      --
      "Derp de derp."
    151. Re:Excellent commentary... by Anonymous Coward · · Score: 0

      I agree with your sig. AC comments are misguided opinions and never fact.

    152. Re:Excellent commentary... by RedBear · · Score: 1

      World domination, huh? Saying that Mozilla is all about killing off IE is probably about as true as saying the point of Linux is to replace Windows on the desktop. Although we all may wish these things were true, they have little bearing on the actual reasons for the existence of these open source projects. Geeks just wanted a better browser, and they didn't want to use IE. If anything can be said to be the whole point of Mozilla, it would be to make a platform-agnostic high quality web browser. So far they have done pretty well at that.

      By not supporting a non-cross-platform feature, they're slowing the growth of cross-platform software.

      What are you smoking? I can't think of any other way to react to that. Did you even read what you just wrote? And by perpetuating Windows lock-in, that supports the growth of cross-platform software how, exactly? That's right, it doesn't.

      Whether or not my previous company was acting intelligently or not, it doesn't have the slightest bearing on the behaviour of the Mozilla development team. It is not my nor my previous company's fault they intentionally chose not to support a feature that has some healthy use in the IE world.

      No, as I've been trying to explain, you're looking at this whole situation bass-ackwards. The Mozilla team owes you nothing. It is not the Mozilla development team's fault that your company chose to lock themselves into a proprietary technology, and I'm sure not many other people would ever consider using the word "healthy" in the same sentence with ActiveX. It's more like a cancer that is eating your company's bottom line and destroying the ability to adapt. If it did have good uses there might be a different attitude toward supporting it.

      Again, if it's so important to your company to continue using ActiveX while switching to open source, why aren't you employing some developers to code up ActiveX support for the open source projects you want to be using? Obviously the Mozilla developers feel like they have more important cross-platform features to work on. Dare I say, some of the very same features drew you to the software in the first place. Also quite obviously, the vast majority of the community of users does not want Mozilla to inherently support ActiveX. We don't see it as a healthy use of the Mozilla team's time.

      I'm sorry, but open source is not solely about replacing or destroying Microsoft. It's about providing alternatives based on open standards. Asking cross-platform open source projects to spend time developing support for lock-in technologies is like asking them to sabotage their own projects. There is already at least one third party group working on an ActiveX plugin, so why don't you go tell them to work faster, or give them some money. You decide which will produce better results. And then if you really think that support for ActiveX will help Mozilla, donate that ability back to the community. Prove us all wrong.

      You can sit here and tell me I'm wrong all you want, but at the end of the day there are a lot of people wanting to leave IE and can't.

      "Ladies and gentlemen, this is Chewbacca. Therefore, my client is not guilty." Nobody said there aren't a lot of people locked into various Microsoft software, including IE. The problem is how you're going about trying to get that fixed. You're not wrong. You are definitely locked into IE/ActiveX. What's "wrong" is your proposed method of escaping that lock-in, because it's really no escape at all. It serves very little purpose to escape IE and continue to be locked into ActiveX, and it certainly doesn't help the community at large. ActiveX is your main security issue, not IE.

      Since the Mozilla team has the power to solve this problem, and won't, the fault is theirs.

      I don't know what makes you think it's such an easy fix, and again, the fault is definitely not theirs. Might as well say it's Sun's fault that your screwed up Java applets don't work on Linux because they're using propriet

    153. Re:Excellent commentary... by NanoGator · · Score: 1

      " And by perpetuating Windows lock-in, that supports the growth of cross-platform software how, exactly?"

      I actually fought with myself a bit figuring out if I should answer this or not. Not only have I clarified my view here, but it's also actually quite simple. If a user cannot switch because of lock-in, they're still locked in. I could not switch to Mozilla because I was locked in with MS.
      Mozilla perpetuated the lock-in. Simple simple simple.

      "Ladies and gentlemen, this is Chewbacca. Therefore, my client is not guilty."

      I didn't bother reading past that. I don't mind when people disagree with me, but you're intentionally not listening just to argue. Well I'm sorry I'm not singing the praises of Mozilla.

      --
      "Derp de derp."
    154. Re:Excellent commentary... by RedBear · · Score: 1

      " And by perpetuating Windows lock-in, that supports the growth of cross-platform software how, exactly?"

      I could not switch to Mozilla because I was locked in with MS.
      Mozilla perpetuated the lock-in. Simple simple simple.


      Didn't you just basically say something that makes absolutely no sense, which is that Mozilla perpetuates your Windows/ActiveX lock-in by NOT perpetuating Windows/ActiveX lock-in? And thus fail to answer the question on how perpetuating lock-in is a good thing that will support cross-platform growth? I realize it would be convenient for you to have Mozilla support ActiveX on Windows, but how would that NOT perpetuate ActiveX lock-in, and how would it support cross-platform growth? You'd still be locked into the Windows version of Mozilla because of ActiveX, which wouldn't run on the other platforms.

      Sure, you couldn't switch to Mozilla because of MS lock-in, but you also couldn't switch to Opera, K-Meleon, Safari, Konqueror, OmniWeb, or any other non-IE browser for the same reason, a reason created by MS. How is all of that Mozilla's fault? Why isn't it Apple's fault? Or Opera's fault? According to your anti-logic they should all be blamed for perpetuating your lock-in. After all, they aren't doing anything to NOT perpetuate your lock-in, so it really must be their fault that Microsoft developed this lock-in technology and your company chose to use it.

      Wait, what did I just say? "Microsoft developed this lock-in technology and your company chose to use it." Hmmm... There's something there, some hidden meaning I can't quite grasp... Maybe... Maybe some fault lies with Microsoft themselves for perpetuating your vendor lock-in? But no, that's silly.

      I didn't bother reading past that.

      It was a comment meant to point out that one part of your statement had nothing to do with the other. Same as your bizarre example above where Microsoft does something and somehow it's Mozilla's fault. I'm hearing you quite clearly, but you seem to have some cognitive dissonance going on on the receiving end. You have managed to convince yourself that it's logical to blame third parties for the actions of Microsoft and the past decisions of your company to use some bad technology. And so, you continue to completely miss the point. And, you continue to perpetuate your vendor lock-in by not moving away from ActiveX. It's the ActiveX that is your problem, not the browser running it.

    155. Re:Excellent commentary... by NanoGator · · Score: 1

      "Didn't you just basically say something that makes absolutely no sense,"

      Only if you're not in a listenin mood.

      " Maybe some fault lies with Microsoft themselves for perpetuating your vendor lock-in? But no, that's silly."

      This might be interesting had I said "This is entirely Mozilla's fault and not Microsoft's" However, I didn't. It's Mozilla's fault they don't support a widely accepted standard, not Microsoft's. I don't see the point, though. I'm talking to a wall, here.

      --
      "Derp de derp."
    156. Re:Excellent commentary... by RedBear · · Score: 1

      My point was that you aren't saying the way things actually are, which is that this is entirely Microsoft's fault and has nothing to do with Mozilla or any other web browser developer that doesn't support ActiveX, of which there are many.

      Look in the mirror to see the wall. The things you say just keep getting more ridiculous. "Widely accepted standard" my ass. ActiveX is not a bloody standard! It is not Mozilla's "fault" for not supporting a proprietary non-standard. This is exactly why you are getting nowhere trying to convince anyone that Mozilla should support ActiveX. They're busy supporting actual standards that are beneficial to the entire community instead of crap that only supports Microsoft Windows. Thank goodness.

      I give up, buddy. Have fun perpetuating your ActiveX "standard". I mean, have fun blaming other software for perpetuating it by not supporting it. I admit it, I am not smart enough to make that logical leap.

    157. Re:Excellent commentary... by NanoGator · · Score: 1

      ", which is that this is entirely Microsoft's fault and has nothing to do with Mozilla"

      Mozilla not support ActiveX. Mozilla fault. *Grunt*

      ""Widely accepted standard" my ass. ActiveX is not a bloody standard!"

      ActiveX is a standard. It is widely used. Yes, it is Windows only. However, Mozilla could easily support it.

      "I mean, have fun blaming other software for perpetuating it by not supporting it"

      Heh. "We're making sure that anybody dependent on MS's features can't leave MS, but we won't acknowledge that by doing so we're feeding Microsoft! Afterall, we ARE the good guys!"

      Yeah yeah, whatever.

      --
      "Derp de derp."
    158. Re:Excellent commentary... by MrResistor · · Score: 1

      I was under the impression Suse (now part of Novell) distributed them on their installer cds

      Nope, but their online update gives you the option of downloading the current Nvidia driver from Nvidia, and will then install it for you.

      --
      Under capitalism man exploits man. Under communism it's the other way around.
  36. IRC by laurent420 · · Score: 2, Informative

    default dcc save directory is ~ . many users of irc are accustomed to permitting auto accept of files. someone sends you a .profile or .bashrc . .profile is sourced on every login. hmm i wonder what happened to all my filesystems.

  37. It's dangerous because... by jmil · · Score: 1

    ...lots of people have a blank password. I've seen it countless times in a windows environment.

    Let's say user A has a blank password, and runs as root. Said user also turns on SSH. Say hello to remote access for anyone who knows your IP, and goodbye to your computer (unless he simply installs a trojan as root... what fun!).

    Interestingly, OS X allows you to enable and disable the ability to login as root as needed. Logging in as root by default is disabled (though commands can obviously be run as root with su).

    --
    I wish I were old enough to put "Computer" on my resume.
    1. Re:It's dangerous because... by tim_mcc · · Score: 1

      Interestingly, OS X allows you to enable and disable the ability to login as root as needed. Logging in as root by default is disabled (though commands can obviously be run as root with su). Good point, another good point is that anyone running the OpenSSH daemon can change their sshd_config file (if they haven't done already) to include the following: PermitRootLogin no This will make the same changes you described.

    2. Re:It's dangerous because... by jmil · · Score: 1

      anyone running the OpenSSH daemon can change their sshd_config file (if they haven't done already) to include the following: PermitRootLogin no This will make the same changes you described.

      This is true. Unfortunately, the users we are worrying about here do not know what a "daemon" is, much less where its configuration file resides or how to edit it properly.

      --
      I wish I were old enough to put "Computer" on my resume.
  38. A short list of reasons to NOT be root by davidwr · · Score: 2, Interesting

    1) It protects you from yourself. Nobody's perfect all the time.
    2) It limits damage from exploits. Go ahead and be root if you aren't networked and never insert media, or are running a perfectly-secure OS.
    3) it protects you from another user's malice. N/A for single-user machines.

    Examples of when it is OK to run as root:
    1) many non-networked embedded systems, e.g. your microwave oven
    2) the DOS box in the corner your kids play DOOM I on.
    3) Demo machines at trade shows, but only if they are not networked and have no removable media.

    Other examples where running as root isn't advisable but the damage is greatly mitigated include read-only systems like Knoppix.

    --
    Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
    1. Re:A short list of reasons to NOT be root by SPY_jmr1 · · Score: 1

      any links on NON rooted dos?

  39. This defeats the point by bcmm · · Score: 1

    Linux's (well Unix's) strenght is the ability to do everything. Thats why you need to be a user who isn't allowed to do some stuff :)

    This would be worse than running Windows as an admin, because in Windows they have made stuff impossible to stop it being done for the wrong reasons, rather than putting a password on it. Maybe MS was right... No wait I meant Linspire is wrong.

    --
    # cat /dev/mem | strings | grep -i llama
    Damn, my RAM is full of llamas.
  40. Removing the support from the door of the keep. by Polarism · · Score: 1

    Sure, you might have enough confidence in your setup to operate under root, but why would you intentionally do it when you could run things with fewer permissions? This is like saying "hey my router is l33t so i'm just gonna keep the DMZ on 24/7".

    Never remove support beams from the building if you don't have to, eh?

    --
    All your base are belong to Google.
  41. dear sir by Anonymous Coward · · Score: 0

    You suck. You've got nothing better to do than copy and paste stupid trolls to slashdot? This suggests one of two possibilities:

    1. You are at home and in a refractory period and are just killing time until you can get back to masturbating to internet porn.
    2. You work a job where your employer has stupidly given you internet access that you abuse to post idiotic shit to slashdot. If this is the case, you are only dragging down the economy by getting paid to be non-productive and should be shipped off to a third world country in some sort of "productive workers for our slacking assholes" exchange program.

    Before you counter with some sort of "well what about you, writing this stupid reply" type argument I should point out that I am only wasting my amazing intellect on a fuckhat like yourself because I'm currently waiting for several servers to come back up. It's either this or head off to the bathroom with my powerbook to masturbate to internet porn.

    1. Re:dear sir by Anonymous Coward · · Score: 0

      "You work a job where your employer has stupidly given you internet access that you abuse to post idiotic shit to slashdot."

      Thank you. That made me laugh.

  42. Absolutely brimming over with wrongability by emurphy42 · · Score: 3, Insightful
    From TFA:
    when grandma tries to change her wallpaper, and it tells her "you don't have root privileges".
    I don't know whether this is hyperbole or just a bad acid trip, but either way, it shoots holes in his credibility big enough to drive a truck through.
  43. I have to say I love the OSX solution by arete · · Score: 5, Informative

    I have to say I love the OSX solution. For those of you that aren't familiar:

    The method:
    By default you don't use root (although it does exist)

    By default a user may or may not be an "admin" user. An admin user may perform root-like operations by authenticating again, but they give their own same password to the OS to do things.

    It still knows you're you, you're just super-you. So default files are created with you as owner, for instance. This is safer because it reduces slightly the number of escalations necessary.

    The effects:
    The actual user password being compromised is not the reason you need a separate root account, so they removed your need for two passwords.

    Bad apps still need separate priv escalation to do any harm, even if you're running as admin.

    BUT you don't have to logout of your GUI session to have one app - or even ONE PART of one app - run with escalated privledges, if you authorize it to.

    This means you have NO REASON to ever run unnecessary apps as an admin. No downloading just that one file as root because you're in the middle of doing a rooty thing and forgot one.

    The similar linux hack:
    I know you can setup similar things with sudo and a little tweaking. But this is how every OSX box ships, and it ought to be how every GUI consumer linux box ships too.

    --
    Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
    1. Re:I have to say I love the OSX solution by anthony_dipierro · · Score: 1

      If I recall correctly, by default any admin user has access to change the root password. So yes, they're not root, but it's trivial for them to become root.

    2. Re:I have to say I love the OSX solution by Relyt · · Score: 4, Informative

      Well, Ubuntu Linux is set up with sudo all set up right off the bat, which is probably the way things will be setup in the future. The user can use his or her own password to get root privileges.

      I think that anyone who is considering buying a PC for Lindows would be much better served buying a Mac or Mac Mini and using OS X instead. They'll spend the same amount of money and have an OS that is better-designed and is backed by a corporation and a CEO who actually know what they are talking about.

    3. Re:I have to say I love the OSX solution by Empty+Yo · · Score: 1
      Actually ...

      Due to some ridiculous permissions issues, my copy of Virex will not scan other Admin accounts when I run a full system scan, stating that I don't have the priviledges necessary to scan the files. I can delete the entire account as an Admin user, but I can't scan it for viruses.

      The only solution, until the manufacturer gets its act together and fixes this ridiculous issue, is to log in as root and then run Virex. Once the scan is done, you log out and then back into the normal Admin account.

      So there is at least ONE instance where you have to log in as root ...

      --
      I'll tolerate anything except intolerance.
    4. Re:I have to say I love the OSX solution by Anonymous Coward · · Score: 2, Informative

      I believe Ubuntu/Kubuntu are going down the 'root is there if you really want it, but you don't actually need it because sudo is already set up for you' path.

    5. Re:I have to say I love the OSX solution by GileadGreene · · Score: 1

      In OS X the root account isn't even active (by default). Everything is done through sudo.

    6. Re:I have to say I love the OSX solution by aixou · · Score: 1

      There's no reason why someone couldn't write a malicious piece of software that simply asks the user to give it escalated privileges. Users have become so accustomed to entering in their username and password that it wouldn't be much of a stretch for malicious software to do the same. Bang! back to square one.

    7. Re:I have to say I love the OSX solution by arete · · Score: 1

      It's not "square one" "square one" is when everyone logs in as Administrator or root all the time so their installs work. Requiring OS controlled user intervention is at least square two.

      Having uneducated users administering (in any system) their own machines will never be truly safe.

      --
      Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
    8. Re:I have to say I love the OSX solution by Anonymous Coward · · Score: 0

      Ideally, Virex should prompt you for your admin password, then run a scan process as root over the other users' folders. It sounds like right now it just runs with your privileges, which (properly) do not allow you access to the files of any other user.

      Until Virex uses the correct method to gain superuser rights, try an app called Pseudo http://personalpages.tds.net/~brian_hill/pseudo.ht ml which lets you launch a single GUI app as root from your own login session.

    9. Re:I have to say I love the OSX solution by Scudsucker · · Score: 1

      but it's trivial for them to become root.

      Not really. You have to know where to look for the root account in Netinfo Manager, and enable it. It is HIGHLY unlikely that the average user will look through the dozens of settings there to find and enable root. From the command line you can use sudo, but if you know about sudo then you probably know not to screw around with it. Also, even once you enable the account, you can't (easily) log into the GUI as root.

    10. Re:I have to say I love the OSX solution by j14ast · · Score: 1

      or better yet knows he doesn't know, no slight on Jobs, but he has very little in the way of geek chops.

      --
      Damn the man!
    11. Re:I have to say I love the OSX solution by Anonymous Coward · · Score: 0

      Mandrake ships almost that way (though using the root password) (well you could open the control center with just one password, but it's much quicker just to run the gui tool you need directly and punch the pword in each time.) On the command line however you do have to super user or su -c (I wish this was the default behavior, rather than the other way around (also I wish it worked like xargs: everythig after the options is part of the command to be run)

    12. Re:I have to say I love the OSX solution by Anonymous Coward · · Score: 0

      This sounds just like Solaris RBAC.

    13. Re:I have to say I love the OSX solution by tengwar · · Score: 1
      Sorry, but OS X by default is not as safe as you think. Yes, most of the time you are not running as root, but many important actions are enabled by your group if rather than your user id - for instance installing/removing applications. If you are an admin user (i.e. one with access to sudo, if you use the OS X user manager) then you are a member of the wheel group, which gives you the relevant privileges. This removes most of the apparent protection that you get from not being root.

      For safety, I recommend that you make a new admin user, then use the user manager to set your own privileges to not be an admin. OS X will then ask you for the username and password of a admin user to perform certain actions such as installing apps, but other than that everything will continue to work as normal (unlike Windows). You may also find it convenient to add your own user id to the /etc/sudoers if you use the command line.

      BTW, another little-known weakness of OS X seems to be that the firewall only blocks TCP, not UDP.

    14. Re:I have to say I love the OSX solution by Anonymous Coward · · Score: 0

      And yet he has built an OS that is faster, more secure and far more usable than the much vaunted (and way overhyped) Linus Torvalds.

    15. Re:I have to say I love the OSX solution by geggo98 · · Score: 1

      I think that anyone who is considering buying a PC for Lindows would be much better served buying a Mac or Mac Mini and using OS X instead.
      Or he could buy one of these fancy HP notebooks that will ship with Ubuntu. Rumors say, HP develops a special Ubuntu version, that will support the complete hardware of these notebooks, including ACPI Suspend to RAM.
      Ubuntu has a security model that is similar to Mac OS X. Root cannot login; when root privileges are required, sudo will be used.

  44. Re:uh... by NichG · · Score: 1

    And how often have you actually done that by accident?

  45. He has a point by photon317 · · Score: 4, Insightful


    We all know the reasons not to run anything as root unneccesarily are many, but you have to think from his perspective as well. He's picturing clueless linux desktop users, using a shrinkwrapped distro at home for personal use. If they were to only log in as a user rather than root, what does it buy them? Whoever gets them to run malicious code by exploiting them or their software will still get access to all of their data, since it was all stored as that user. And they still get access to backdoor all of the software they use, since they can screw the user's environment (PATH, LD_LIBRARY_PATH, etc).

    About the only thing not running as root saves the poor nontechnical home end-user from is wiping out their hard drive, but all the data that's important to them contained therein is still destructable.

    His point is in fact arguable - why bother?

    --
    11*43+456^2
    1. Re:He has a point by anthony_dipierro · · Score: 2, Insightful

      Whoever gets them to run malicious code by exploiting them or their software will still get access to all of their data, since it was all stored as that user.

      It's generally worse than that, in practice, because you usually have to be root in order to install most software. So unless a user is reading every line of code and compiling it locally, you're just need to shift the exploit into the install program.

      Now look, it's theoretically possible to make it so that nearly all programs can be installed without root privileges, but this requires pretty much a complete rewrite of the unix hierarchy. Unix wasn't designed with home systems in mind.

    2. Re:He has a point by Anonymous Coward · · Score: 0

      Here's the answer:

      If you have an uncompromised super-user on the box, it can run services which monitor problem points of entry into the system and work to correct any changes that it suspects the user really shouldn't be messing with. Essentially, software you trust still has the upper hand on your box.

      If you're running as root, you're basically fucked, because software you don't trust has a free run at anything on your PC.

      Having root available and protected gives you more options in terms of dealing with problems AFTER the worst has happened and a userspace vuln has already been exploited. Obviously you want to do all you can to prevent that happening in the first place, but it seems stupid to throw away another line of defence against malicious code when it is there and provided by default.

    3. Re:He has a point by hackstraw · · Score: 1

      His point is in fact arguable - why bother?

      His point is that the user's data is the box, and it is always going to be available to that user, etc.

      The funny thing, is I haven't heard of an exploit that actually did anything with user's data in over ten years. About ten years ago there were viruses that completely wiped or formatted the hard disk. I personally don't know why viruses stopped doing this. If I wrote a virus, that is what I would do, but then again I wouldn't do it in the first place.

      I hate to repeat myself within a thread, but I'm going to. What is one of the biggest issue with compromised boxes today? Zombie windows boxes that send out spam for free for those clever guys selling rolexes, penis pills, and mortgages. Linux rootkits are fairly clever with their kernel modules that hide the evidence of the rootkit. I would imagine that Linux would have a very bad name very shortly with every grandma and pr0n surfer using Linux as root. I would see nothing of value from that.

    4. Re:He has a point by shadowbearer · · Score: 1



      Well, yes, but a properly set up installation would archive the user's data somewhere where a non-root user couldn't get at it (and allow the user to recover it easily :-)

      SB

      --
      It's old. The more humans I meet, the more I like my cats. At least they are honest.
    5. Re:He has a point by m50d · · Score: 1

      How about their bios? Root can reflash it and make their computer completely unbootable, unless they're lucky enough to have a socketed one, even then it's going to cost them a lot to get it reflashed. Normal user can't.

      --
      I am trolling
    6. Re:He has a point by SolusSD · · Score: 1

      Your forgetting the fact that future viruses and other malicious programs that will eventually be written for Linux will need root access to do any real damage (ie, the kind that requires you call over your tech buddy to reinstall the os and reconfigure everything). If the virus only gains control of the users account, ie. is running with user level privledges, it may only be able to nuke the users personal files, on his account. the system will still run. other reasons are numerous. keep yourself or a hacker from editing important system settings intentionally or accidentlky, for example.

  46. I don't agree, but... by Tim+C · · Score: 1, Insightful

    Consider this:

    a) an awful lot of home machines are either single user, or effectively single user (where everyone shares a single account)

    b) all the system files are backed up on the nice, shiny install media

    c) none of the user files are backed up

    If you're not talking about a server or other shared/critical environment, then the only things of any real value on the machine are the user's own files. Root or not, they can toast them. Lindows, in case you hadn't noticed, is *not* aimed at servers...

    1. Re:I don't agree, but... by Anonymous Coward · · Score: 0

      Spyware isn't aimed at servers either.

      It is the single biggest problem on single user systems. Everyone says the same thing, "I don't care if they get my data, I don't have anything worth taking." Sometimes you even get evolution rejects who say "I don't mind the ads. I just close them."

      Two weeks later the same people call back crying that their new computer is running slower than crap. They have also managed to put enough "important personal stuff" on to make the system restore Not An Option.

      But sure Root is safe.

    2. Re:I don't agree, but... by roju · · Score: 1
      If you're not talking about a server or other shared/critical environment, then the only things of any real value on the machine are the user's own files. Root or not, they can toast them. Lindows, in case you hadn't noticed, is *not* aimed at servers...

      Now, if I'm running as some random user, I'd fear rm -rf $HOME. However, I'd be able to download some software and say to it "find all deleted files on this partition and make them come back."

      If I'm running as root, the attacker could do something like dd if=/dev/urandom of=/dev/hda and then my chances of getting any data back are pretty much toast.
  47. Running rootless by MemoryDragon · · Score: 1

    means, that you dont have any standard user on the machine where you just have to guess the password. On the other hand sharing the same password for sudo and the normal user, like OSX and Ubuntu do it, is as much as a security risk as having a dedicated root because all you have to find out is how to get into the machine as a normal sudo user.

  48. Modded -1 Flamebait by HiredMan · · Score: 3, Interesting

    I knew Michael Robertson in college and he was a technological lamer and pretty much an A-hole. And he doesn't appear to have changed much. He's cobbling together whatever technologies he can get his hands on and then shamelessly pimping^H^H^H^H^H^H^H self promoting whatever his latest project is regardless of merit.

    He unfortunately seems to have learned that there is little fact checking in the business press - especially where technology is concerned - and that if he can create a stir he can probably create profit.

    It was several years before I realized that it was the same Michael but I visted the website and found his picture there - in multiple super high resolutions - seriously why would I want a 1435x1980 pixel image of him?
    Does he think he's desktop material? There's even information for booking him for speaking engagements... but it's not about ego. *SIGH*

    Look for the stock pump and dump scheme followed by an SEC investigation in 5 - 10 years...

    =tkk

    1. Re:Modded -1 Flamebait by tfoss · · Score: 1
      He's cobbling together whatever technologies he can get his hands on and then shamelessly pimping^H^H^H^H^H^H^H self promoting whatever his latest project is regardless of merit.

      What a novel idea for someone in business.

      -Ted

      --
      -=-=- Quantum physics - the dreams stuff are made of.
  49. root vs user by iguana · · Score: 1

    I challenge a non-root user to screw up a system as bad as this.

    dd if=bootimage.bin of=/dev/hda

    "Weird," I thought. "Why did it come back so fast? Usually floppy writes take a whole lot longer?"

    I had been doing

    dd if=bootimage.bin of=/dev/fd0

    and brainlocked.

  50. Robertson is the "Billy Mays" of the Linux world by scupper · · Score: 3, Informative
    I can't take this guy seriously. He's the Billy Mays of the Linux world.

    Just read his responses....[a few of my repiles]

    Jo: On the security front, I noticed during the presentation that you were running everything as root. Is that really a wise idea, to train users to run everything as the one user who can mess everything up whenever they feel like it? Should you not try to teach one basic UNIX security idea, that you really don't want to run things as root?

    Michael: I think, like everything, it's a question of balance. Ease of use, versus security. I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer. They say "oh, yeah, it is!", but it really isn't. Here's why: What's the most important thing on your desktop? It's the data.[Mikey, that's like saying the people in my car are important, but to hell with the rest of the motorist on the highway. Pretty reckless and selfish. Maybe Linspire should should start "LinNet-Home of the Bots and Trojans] If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.[Mikey, what is a bot? And how are they born?]

    Michael: Then you could say "Well, it's not really about your data, it's that people could accidentally mess things up!". Well, you could accidentally drive into a wall as well, it doesn't mean we should make all cars drive at 10 miles an hour. So, I don't see the added benefit. I DO see it's an added pain in the ass when grandma tries to change her wallpaper, and it tells her "you don't have root privileges". What are you talking about, man? I'm just trying to use my computer, or change the clock, or any one of a hundred other things. So, people always say "it's less secure", but I defy anyone to point out a single instance, and people all go "Well, I, erm, it's theoretical!". There's no one area I think you can point out - In this instance, a machine that's run with the root user could be compromised, in this instance one couldn't be compromised.

    Michael: I know the hardcore geeks feel differently, that's fine. When somebody installs Linspire, we say "do you want to set up users, yes or no", we give them the choice, right there when they start up for the first time. If they want to set up multiple users, they're welcome to do that, but we don't force them to. That's the difference we have.

  51. worse than windows? by greenrom · · Score: 4, Insightful

    Running as root is dangerous, but is more dangerous than the average home user is used to? Probably not. The average user probably runs windows from a single user account with admin rights. For most people, the recycle bin is the only protection from stupid mistakes.

    1. Re:worse than windows? by broKenfoLd · · Score: 1

      Power outages are the only thing that keeps my users from breaking their machines.

    2. Re:worse than windows? by Craster · · Score: 1

      The parent is not insightful, it is horrifying.

      To say that there is no security problem because it's "not as bad as windows" is the worst attitude to security you can possibly have.

    3. Re:worse than windows? by Cro+Magnon · · Score: 1

      No, it's business as usual for Joe Winpack, but is that a good thing? Linux is supposed to be better than Windows, not merely "no worse".

      --
      Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
  52. Accidents by iamacat · · Score: 3, Insightful

    Malicious software can always trick user into giving it administrator access. But if you always login and root, one bad mouse gesture in file explorer can make your system unusable. Just yesterday I saw someone with a master degree trying to store MP3 files in /Library on MacOSX.

    Besides, if you have a family PC why would you want everyone messing up each other's files if they can have nice separate home directories?

    1. Re:Accidents by Anonymous Coward · · Score: 0

      > Just yesterday I saw someone with a master degree trying to store MP3 files in /Library on MacOSX.

      OK, explain to me as a hypothetical computer newbie why I *shouldn't* store my MP3s in /Library?

    2. Re:Accidents by iamacat · · Score: 1

      Because later you might want to delete or rename stuff inside your music library and damage your system by mistake. For example, if you rename /Library/Audio to /Library/mp3, you will cause microphone on your webcam to stop working.

  53. Yeah... by Aldric · · Score: 1

    Just how often is there more than one person's data on a machine? At home my computer is mine, at work everyone have their own machine. Well, my coworker checks my email while I'm away from the office for a few days to make sure no disasters are missed and I do the same for him but that doesn't count. It's not 1980 anymore - most computers are single user machines.

    1. Re:Yeah... by El · · Score: 1

      I don't know about you, but on my Windows XP machine at home I have separate user accounts set up for myself, my wife, and my 4-year old so they don't screw up my settings. You must consider yourself fortunate that you don't have to share your equipment with anyone else.

      --

      "Freedom means freedom for everybody" -- Dick Cheney

    2. Re:Yeah... by Cro+Magnon · · Score: 1

      I don't have the money or the space for multiple desktops. If my GF moves in with me, we'd be sharing one computer. And her young niece would probably want to play games on it, and I DON'T want her to be able to mess up my financial records.

      --
      Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
  54. I want to see ... by Anonymous Coward · · Score: 0

    ... him having a chat with Theo de Raadt about this...

  55. security using system console by PenguinX · · Score: 1

    Aside from blaring user error (i.e. any command using -R) it is my opinion that the computer is already compromised if you are using have the system console. Gaining root or hardware access is fairly simple and things such as encrypted filesystems are obviously out of the Linspire's scope. While I believe it would have been "better" had they used a sudo-like implementation (e.g. ksudo / MacOSX) I can understand why a business would take the root approach.

  56. Before somebody picks on a point by arete · · Score: 1

    I thought I'd pick on myself before somebody else did.

    I know in linux you can, for instance, open a terminal, su, and execute a GUI app as root while in an X session not as root. However, there's no general linux way for doing this for a nonCLI user.

    I also know that in Windows you can "run as" by providing that alternate password, and you could set your Administrator and user passwords the same. But you get all sorts of problems doing this - for instance with an app that needs admin privs to install but not to run.

    Finally, I certainly know that not every single security thing Apple has done has been right. This thread is particularly about their admin-user design, which I think is an ideal DESIGN. If they left open a hole somewhere, they should fix it. But the design is brilliant.

    --
    Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
    1. Re:Before somebody picks on a point by Anubis350 · · Score: 1

      try man sudo and come back, you dont need to drop in root in order to execute root priveleged apps, assuming sudo is setup correctly.

      in a graphical enviroment there's gksu (su/sudo front end) and one for kde which I cant recall, not to mention others. you don't need the admin password or any command line experiance in order to use them if, again, they've been set-up properly (though I know of no distros that do that out of the box, it prolly wouldnt be hard to implement since this is basically what OSX does...)

      --
      "goodbye and hello, as always" ~Prince Corwin, from Zelazny's Amber series
    2. Re:Before somebody picks on a point by Anubis350 · · Score: 1

      woops, as someone pointed out in a later post, ubuntu sets this up out-of-box. I had forgotten that, so yeah, I'll add that in.

      --
      "goodbye and hello, as always" ~Prince Corwin, from Zelazny's Amber series
    3. Re:Before somebody picks on a point by Mishura · · Score: 2, Informative

      "I know in linux you can, for instance, open a terminal, su, and execute a GUI app as root while in an X session not as root. However, there's no general linux way for doing this for a nonCLI user."

      Observe, The KDE solution:
      K --> Run Command --> kdesu program_name

      The Gnome Solution (I Think):
      Gnome Foot --> Run --> gksu program_name

      Also, you can set program shortcuts in either the K/Gnome/XFCE/icewm/wtf wm you desire/ menus to start off with the gksu or kdesu to launch an app as root.

      Also, if you have a lax sudo set up, a "sudo app_name" works as well.

  57. Err? by binkzz · · Score: 1

    "we wanted to integrate the shareware version of Linux" Are you for real? Seriously? You're a consultant? You're being overpayed, or you're just blagging it.

    --
    'For we walk by faith, not by sight.' II Corinthians 5:7
    1. Re:Err? by binkzz · · Score: 1

      I found this quote which summed it up nicely:

      "To err is human... to really foul up requires the root password."

      --
      'For we walk by faith, not by sight.' II Corinthians 5:7
  58. His main argument FOR using computer as root... by Lord+Duran · · Score: 1

    Changing the wallpaper or the time? I don't believe you need root access for either. Even if you did, how difficult is it to change that? The car analogy is pretty weak. Does the average computer user - and the extensive computer user - take 10+ hours of computer-teaching lessons along with theory studies and months of needing your dad to sit by you as you type? Getting a driver's license requires you to know your car, more or less inside out. If every computer user would do that, MAYBE it'll be a valid analogy. Even so, not being root still saves you. People still drive into brick walls; Accidents still happen. And, using Robertson's analogy, people still type rm -rf / instead of whatever. You might say that all that is needed is a couple hours' work to reinstall everything, but isn't that the purpose of Linspire? To save time, for those to which that time is worth a lot of money?

  59. the real reasons by bcrowell · · Score: 1
    I think you're missing the most important reasons it would suck if everyone ran as root all the time:
    1. It would completely take the punch out of the bumper sticker that says Bow before me for I am root!
    2. What would we call it when you got your machine rooted? Getting usered?
    3. I like having a $ prompt sometimes and a # other times. It gives variety to my life. It's sort of like the little program you can download for Windows that makes your cursor into a little comet, but if you get tired of that you can change it into a picture of Spongebob.
    4. It makes me feel special when I have to type ./script instead of just script.
    5. A cool piece of geek-speak has to go through a certain life cycle. In the case of the term "slashdot," for instance -- first there was the time when I didn't know what it was; then there was the period when it seemed cool but I still didn't know why it was called that; then came the era when I pretended I was (un)cool enough to know why it was called that; and finally there came the day when it was explained on wikipedia, so I found out why it was called that, but so did everybody else who cared, so it wasn't any fun anymore. The term "root" has not yet had a chance to go through the full cycle.
  60. hardware damage IS possible by davidwr · · Score: 2, Informative

    With poorly designed hardware, it is possible to wear out the hardware. Cheap printers and disk drives are relatively easy to wear out in a worst-case scenario. Certain types of flash memory can be destroyed by flashing it a few thousand times. While your operating system may not require you to be root to overuse these components, in principle it COULD force you to be root to do this.

    If you can write to BIOS or other boot-control data, you can potentially leave the hardware unbootable. Technically it's not hurting the hardware but you've still got a boat-anchor until it gets serviced.

    Older monitors could be fried if set to a "bad state" and left there too long. Ironically, in X-Windows, you don't have to be root to change the video settings to such a "bad state."

    --
    Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
  61. Yes ... it's that bad by argoff · · Score: 2, Insightful

    Uh well, I think he is actually that ignorant (or lost is more like it). This is the guy who started mp3.com and thought that the music industry was going to give him a big pat on the back for it and let them into their billionaires club. Even worse is how he down-talks illegal copying like it's a back-alley dirty activity, when in truth nobody is doing any worse than he has been, is, and will likely continue to do for the rest of his life. IMHO, he is the epitomy of blind love for evil systems. No matter how poor it is, evil it is, he throws himself at it with pure optimisim and glossy eyes.

    1. Re:Yes ... it's that bad by Anonymous Coward · · Score: 1, Funny

      Thats no troll. Its a space station.

  62. Backseat Driver by Anonymous Coward · · Score: 0

    Then you could say "Well, it's not really about your data, it's that people could accidentally mess things up!". Well, you could accidentally drive into a wall as well, it doesn't mean we should make all cars drive at 10 miles an hour. So, I don't see the added benefit.

    The analogy is more than flawed. Adding a regulator to keep cars below 10 MPH would limit the utility of a car. What he's arguing for is more like replicating the controls and placing them at various points around the car. Sure, you can argue that's not inherently dangerous, but one does not have to think to hard to see scenerios where it could increase the odds of bad outcomes.

  63. Breaking News from DHS Privacy Board.... by scupper · · Score: 3, Funny

    Michael "Root" Robertson is appointed to the Department of Homeland Security's Privacy Board.

  64. Thank you! by Lord+Bitman · · Score: 1

    This is exactly what I am talking about. For a desktop system, a single-user environment, USER ACCOUNTS ARE STUPID.
    Not the dummy accounts with no access which various services run as (eg: mysql only has access to its DBs and nothing else), those are important.
    But the USER of the computer. In an environment where heshe is the only user (true in 99.999% of cases- wastefulness debates being an unrelated issue), there is no reason at all to not run as root. For exactly the reasons mentioned in the article: Nobody gives a flying crap about anything they dont have write access to.
    Why?
    Because the things they dont already have write access to, they did not write. If they did not write them, they were written by someone else, and can be EASILY REPLACED. This is debian based. Something fucked up? Oh nos! apt-get install and happy new year.

    We are not talking about a vast multi-user network on a secure system. We are talking about a desktop. A single person who wants to read e-mail. When shit happens, it can do just as much damage with a user account as it can without one.

    "But... Hardware!"

    If your hardware can be broken by saying not-nice things to it, that is a bug in the hardware. That is not something to be corrected by disallowing things from talking wrong. Why in the almighty fuck would you put up with hardware which dies if it is sent the "die now" command? This isnt the CIA, here.

    magical metaphor land:
    A man is known to kill people whenever someone says "hello!" instead of "knickerbockey boingydoo". Do you: lock him away and stare at him through a tiny hole forever, or: tell everyone in the universe that greetings will now be handled by a password-locked greeting machine whereby pushing the big green button will always produce the greeting "knickerbockeyboingydoo!" and attempting to give a greeting all by yourself will result in an error.

    back to reality:
    I dont use root because I prefer to see the word "shruggar" everywhere.

    --
    -- 'The' Lord and Master Bitman On High, Master Of All
    1. Re:Thank you! by Lord+Duran · · Score: 1

      Their clientele stick with what they have. They use the hardware that dies when it is told because THAT'S WHAT CAME WITH THE BUNDLE.

    2. Re:Thank you! by Lord+Bitman · · Score: 1

      and if anything goes wrong with their system, be it their user directory deleted or a failed peice of hardware, they take it to a service center which says "yeah, that's due to a defect in the hardware, here's a new one" and the old one is RMAd. You know, because it's defective.

      Slow down, cowboy! You seem to be capable of writing a short reply to a short reply in under an hour! What the hell are you on? Abusing methamphedamines can make you see CRAAAzzzzYY shit and will make you think these sort of limits kicking in after ONE post is a good idea!

      --
      -- 'The' Lord and Master Bitman On High, Master Of All
  65. Is this guy running a computer business by chance? by janoc · · Score: 1
    If yes, I have good advice for his customers:

    RUN!

    This guy is absolutely clueless about the basic security principles and even makes a fool of himself in public by showing how much wiser he is than the generations of researchers and engineers which established them :(

    I would expect a Linux company CEO to know better than this or at least have few smarter engineers to hit him with a cluestick when necessary.

    Even Microsoft learned that running as root is wrong and causes tons of problems. Now a Linux distributor comes preaching the opposite? Unbelievable ...

  66. He's wrong about the data, too. by khasim · · Score: 1

    If I have root access, I can change the data and you wouldn't even know that it had been changed unless you compared it with a known good backup.

    Root puts EVERYTHING at risk.

    Also, his car analogy is flawed because people DO accidentally drive into walls and trees and other cars. And many times, people DIE from that.

    But we accept the risk because of the massive benefits of being that mobile.

    What are the benefits of running as root instead of a regular account? I run Ubuntu and I never have a problem with my regular user account.

  67. This one time, in band camp ... by mios · · Score: 1

    I was running as root setting up our debian server at work to do some level 1 raid magic. I was real close to finished after working on it the better part of a day or two ... after jumping around editing this file, doing that, mirroring a partition of a hard drive, I wanted to undo something a copy I made by nuking a directory ...instead I typed someting dumb like: rm -rf /usr/* after hitting enter and realizing how dumb I really could be, I truly understood why you shouldn't be running around as Mr. Root. Though I needed root privs to edit some of the conf files and what not, you realize that sometimes running as root is like having sex without a condom. If you want to have kids, don't use a condom. All of the rest of the times, though ... it's a real good idea, because the rhythm method just does't hack it ...

  68. With unsecurity like that ... by houghi · · Score: 5, Funny

    ... he should rename his Linspire to something like Lindows.

    --
    Don't fight for your country, if your country does not fight for you.
  69. RTFA by Lord+Bitman · · Score: 1

    rm -rf ~

    Hey look, the system is screwed up the exact same amount, as far as absolutely anyone at all cares.

    --
    -- 'The' Lord and Master Bitman On High, Master Of All
  70. Devil's advocate by Concern · · Score: 4, Insightful

    OK, I'll bite.

    Keeping in mind Linspire is totally Desktop-centric, I can see why they might have a radically different view on the permissions system from most existing Linux users.

    I've already read lots of lengthy posts trashing this contrarian point of view. And they have a lot of good points, as yours does, but ultimately this reads like a single-user vs. multi-user culture clash.

    The fact is that on any operating system when you have a single, important user who runs malicious code, it doesn't matter much whether they're root or not, unless the machine has a security model more fine-grained and well-integrated than anything currently in wide use.

    If that user can access their own files, then their own files can be destroyed. If that user can access the internet, then the compromise can also send their files over it. Or it can simply make them a spam bot. Or a relay. If that user has an address book, then its contents can be targets for viral propagation. And so on, and so forth.

    Frankly, to do most things attackers want to do, "root" is unnecessary. Nothing within the unix "user management" repertoire really lets you deal effectively with this problem, and what few solutions you do have are, let's be honest, ugly, cumbersome, evil hacks.

    What stops all this? A real, heretofore unknown high-level security model, that actually says "The email program can access stored email data, preferences, and can talk to the network on this port, to these hosts" and "the word processor cannot talk IRC" and so forth. This requires a rich resource model, rethinking data storage metaphors, the whole nine yards. Unix does not have this. Windows hosts only have it in the crudest and most limited form with "personal firewalls" that to some extent at least police the network activities of applications.

    So for all the Unix folks, of course, this disdain for the security model is heresy, but for the desktop world (and really, servers benefit greatly from a fresh perspective as well), it's not such a bad point. Unix lacks a security model rich enough to be truly useful to everyday users, and by extension, companies like Linspire that cater to them.

    --
    Tired of Political Trolls? Opt Out!
    1. Re:Devil's advocate by ceswiedler · · Score: 1

      What stops all this? A real, heretofore unknown high-level security model, that actually says "The email program can access stored email data, preferences, and can talk to the network on this port, to these hosts" and "the word processor cannot talk IRC" and so forth.

      What you are referring to is SELinux. There's not a lot of work going on in the desktop portions of this, but Fedora is doing a lot to get fine-grained permissions set up for server-side daemons.

      Eventually SELinux will be applied to desktop apps too, and then Linux will be an extremely secure system. It will be a while, however, because it's very difficult to define exactly what a program is allowed to do; the goal is to restrict anything it doesn't need to do while letting it do anything it does need to do. If you err on one side it's a security hole (though admittedly small compared to today's apps) and if you err on the other side the app will fail when it tries to perform some action.

    2. Re:Devil's advocate by Anonymous Coward · · Score: 0

      I can see why some people might run a desktop system as root 'because its easier' but I have to say that its a bad idea. I've run Ubuntu and don't have any problems typing in a password to sudo to root when required, of course I've changed it a bit so I can login to a terminal as root if I need to do more that 1 thing.

      Back ontopic: DDoS's would be much easier to deal with if no desktop ran as the super user. The network layer can be tweaked (if it doesn't do this already) to allow only super users to change the "from IP" of outgoing packets, which leads to an easy way to block DDoS traffic at upstream providers. If they run everything as root then its possible to do much more damage with randomly generated "from IP"'s etc.

      I'm sure there are more examples like this where root has access to more flexible (more abuseable) utilities than a normal user.

    3. Re:Devil's advocate by Jeremi · · Score: 1
      The fact is that on any operating system when you have a single, important user who runs malicious code, it doesn't matter much whether they're root or not


      As far as malicious activity goes, you may have a point... but there is also the danger of the user to himself. If the user can easily delete/overwrite important parts of the operating system, he eventually will. If he can't, then he won't.

      --


      I don't care if it's 90,000 hectares. That lake was not my doing.
    4. Re:Devil's advocate by maraist · · Score: 1

      Unix lacks a security model rich enough to be truly useful to everyday users, and by extension, companies like Linspire that cater to them.

      Frankly, I don't see the security model as having anything to do with usability. It's not like you take the application CD and install it into the CD-ROM and say "give me Quicken" when you use Linux.. Pretty much the beauty of Linux these days is that everything is already installed.. Why? Because the only things you can even FIND for it are open-source, and thus there's no reason why you shouldn't try to fit it onto the 4 CD installation set.

      That says something negative about Linux since we haven't enticed 3'rd party commercial for-profit's our way.. A good video game here and there is worth paying for; you know?

      But my point is that with this current situation, what is there that you need to run as root for on the desktop? The only things I can think of are finite, and pretty well established.. So you set up sudo accounts to let any user run the "printer-setup-wizard" or even "system-config-network". There aren't that many exploits that I can think of, and you should be able to spend some think-time coming up with more locked-down script that are sudo-world worthy.

      So our user-wants-this-configured-this-way can run as root.. But in a highly isolated fashion. And everything else runs as a non-privledged user.

      Is there really a problem here?

      --
      -Michael
    5. Re:Devil's advocate by NutscrapeSucks · · Score: 1

      Your argument is basically "Linux distros are safe because all of your applicaitons come from a single vendor you can trust".

      Which is accurate, but doesn't mean that Unix permissions address the Windows/Mac reality where software comes from different vendors an not all those vendors have the same trust level. The canonical example is installing some random freeware like Kazaa and then finding that the installation program has installed a nasty browser plugin.

      An ideal "Personal Computer"-oriented security system wouldn't have this problem. Unfortunately all modern PC OSes (including Linux) come with a security model which was designed back in the Managed Multiuser Minicomputer days of the 1970s.

      --
      Whenever I hear the word 'Innovation', I reach for my pistol.
    6. Re:Devil's advocate by Rutulian · · Score: 1

      The fact is that on any operating system when you have a single, important user who runs malicious code, it doesn't matter much whether they're root or not

      If the purpose of the virus is to destroy your data, then you are right. There is no current mechanism available to prevent you (or a virus you run) from deleting your own data. But that is not the purpose of most viruses. Viruses are out there to propagate themselves. So they send themselves as email attachments, or they exploit networking vulnerabilities, or embed themselves in other programs, or they background themselves and chew away at your memory, or they log your keystrokes, or they intercept data passed between your computer and something else.... There are mechanisms in place to severely hamper all of these types of attacks IF you are not running as root. There is very little a virus cannot do if you are running as root. Running as a non-privileged user does not solve all security problems, but it is necessary for effective security measures to be taken. Running a firewall isn't a one stop security solution either, but it is an extra layer that makes your machine that much harder to exploit.

    7. Re:Devil's advocate by maraist · · Score: 1

      Your argument is basically "Linux distros are safe because all of your applicaitons come from a single vendor you can trust".

      Well, the distribution == security wasn't my argument, but that's not bad either I guess. It's more that Linux distro's can have everything you'd want to run installed at OS-installation time; a time where you don't need to log in. We're talking potential purchasers of something like lindows or Linspire; so keep that in mind. Lay people like my father aren't going to do anything with root; they're going to look at somtehing like knoppix and say "what's in my menu of fun today". They'll see a word processor, a browser, an IM client etc. By having stock equipment represent what a target audience needs, then after initial installation and configuration of certain hardware like printers / network connections, there is no further need for root use.

      I also acknowledged that this is a crippled way of general computing, as there isn't room for 3'rd parties (outside of the distribution's choice or free-licensing).

      --
      -Michael
    8. Re:Devil's advocate by Anonymous Coward · · Score: 0

      Bah, in OpenBSD, you can't destroy the files even if you ARE root. :)

    9. Re:Devil's advocate by Minna+Kirai · · Score: 1

      A good video game here and there is worth paying for; you know?

      Doom, Quake, Unreal, Neverwinter, Americas Army... oh, and anything in Cedega

    10. Re:Devil's advocate by real+gumby · · Score: 1
      What stops all this? A real, heretofore unknown high-level security model, that actually says "The email program can access stored email data, preferences, and can talk to the network on this port, to these hosts" ...
      Unknown? Perhaps to you. Multics had all this and more...back in 1965. In fact it's the direct ancestor of Unix (although they of course share no code).

      The mail system (your excerpted example) did not use this privilege model until 1970 though, so you could sorta claim it was unknown until 35 years ago (and the IM support took another four years!).

      Otherwise your points are legit.

    11. Re:Devil's advocate by Malor · · Score: 1

      The single biggest reason I can think of to not run as root is this: if your system has a root-level compromise, it can be hidden from you... so well, in fact, that the compromise is completely invisible to routine inspection. (there are some NASTY rootkits out there.) The only way to be absolutely certain one has disinfected such a system is to rebuild it completely, from scratch, wiping the disk entirely. If you are extremely knowledgeable, you can manually clean it, but all you have to miss is one bad file out of thousands of good ones to remain compromised. It would take a HUGE amount of work to be absolutely certain an installation was secure, far more than just rebuilding it, 99 times out of 100.

      User-level compromises, at least, can't be hidden. They can do just as much damage from the user's perspective (delete files, send spam, and whatnot), but they're obvious to anyone with the least clue, and removing them won't take a system rebuild. The system itself can still be trusted, and that's a big deal.

      A Linux box with a root-level compromise could be sitting on the net for YEARS before getting rebuilt. If the hacker is careful, the owner will never even notice that anything is wrong. I worked on a system, remotely, that had been hacked that way. It was impossible to 'see' the bad guys except indirectly. There was a bunch of space mysteriously missing on the drive, and there were ports open (to a remote scan) that shouldn't have been... but which didn't register as being open locally. Had they been smarter and used a port-knocking system, they might still own that box. It was on good bandwidth and, from traffic logs, it looked like they'd been using it at least a year before I got there.

      (This can also happen with Windows, of course... this is absolutely not a Linux-only problem.)

      Running as a normal user just makes it harder on the bad guys. They now must exploit at least two holes, instead of just one. Considering the number of local root compromises there have been in the 2.6 kernel, this may not be that much of a challenge, but anything is better than nothing. The more steps that are required, the more likely it is that one of them will fail.

      This isn't defense in depth, it's more like a double-thick eggshell, but it's still better than just a single layer.

    12. Re:Devil's advocate by Anonymous Coward · · Score: 0

      What stops all this? A real, heretofore unknown high-level security model, that actually says "The email program can access stored email data, preferences, and can talk to the network on this port, to these hosts" and "the word processor cannot talk IRC" and so forth. This requires a rich resource model, rethinking data storage metaphors, the whole nine yards. Unix does not have this.

      Unknown? Not really. OpenBSD has had this for years. Systrace will do exactly what you want. You define exactly what network access, what devices, what sockets, what directories an application can have access to.

    13. Re:Devil's advocate by JohnFluxx · · Score: 1

      Hi,
      I just want to double on that SELinux comment. It's not some fringe thing or anything, but a major effort funded by the NSA and due to be put into the linux kernel. It's going to be extremely interesting when distros support it better.

    14. Re:Devil's advocate by m50d · · Score: 1

      Spam relaying and virus mailing they can do, yes, but even your system wouldn't stop that - the email program needs to be able to access its address book and port 25 on the isp smtp server, and once it can do that it can forward itself to your adress book. But they can't do too much damage to files if you do what I do, and have a nightly backup that tars up important files and puts them in /root. Once you do that the worst non-root attackers can do is delete a day's work.

      --
      I am trolling
    15. Re:Devil's advocate by skubeedooo · · Score: 1
      But the OS itself is not really that important, it is usually neither sensitive nor valuable on desktop systems.

      At worst you have to put the install cd's back in and perhaps reinstall some of the custom apps, at best you just revert to a backup image. This pales into insignificance when compared to losing your entire music collection, photo archive, email archive etc, or having someone take your bank passwords.

      In the end it's a matter of priorities. The unix world is used to multi-user systems where the priority is to minimize the damage down to just single users. Since backing up data is quite cheap (in both time and money) on a large scale, the data can be kept safe easily. The desktop world however doesn't have to worry about multi-user security, and so their priority is to keep their own data safe. In most cases, Unix-like permissions do nothing to protect this. Whilst it is possible to do your accounts as one user, your games as another etc, the process of logging out and back in again can be very time consuming and is well beyond what the consumer is going to accept.

      As far as i can tell, SELinux is very much on the right track. I really hope that it will be enforced as standard in major distributions soon, because if it is not, then I would not be at all surprised to see Microsoft beat Linux at its own game in the coming years. After all, whilst Windows security is bad, it is improving at a much faster rate than Linux, which doesn't seem to have brought much to the table in the last decade. (IMHO)

    16. Re:Devil's advocate by Anonymous Coward · · Score: 0

      "Mechanisms"?

      What mechanisms?

      What on earth does a Virus need root for? It can do everything it needs to do without it - wipe your files, spread over the network, start whenver you log in, replace your shell, bug all your programs... anything.

    17. Re:Devil's advocate by Concern · · Score: 1

      I can fool you without root.

      I can goose your startup scripts and replace your shell. From there, I can substitute anything you see and do, using my own ps, df, ls, you name it. And that's just the crude way. Sooner or later someone would write something really wicked, say parsing the machine code of every executable image your "shell" runs for system calls and replacing them with calls to intercepts. Or is there a more elegant way? strace has some special magic for this, I forget how it works.

      If you always log in with one account, you'll never know. Even if you logged in as root and poked around, I could hide it in very clever ways using one of the many indirect ways to get things done in a shell script, and in the average user's homedir there are a million places to hide files.

      --
      Tired of Political Trolls? Opt Out!
    18. Re:Devil's advocate by Anonymous Coward · · Score: 0

      Not all that many. The point is that you don't need root to do most of the nasty stuff.

    19. Re:Devil's advocate by Patoski · · Score: 1
      Keeping in mind Linspire is totally Desktop-centric, I can see why they might have a radically different view on the permissions system from most existing Linux users.

      I've already read lots of lengthy posts trashing this contrarian point of view. And they have a lot of good points, as yours does, but ultimately this reads like a single-user vs. multi-user culture clash.

      When you say that it is a "single-user vs. multi-user" culture clash I think you are quite right. We're just beginning an age of interconnectivity and Microsoft is still living like it is still a single-user world. This is a computing culture change which will take time. Culture changes are frequently much more difficult than any technical hurdles which must be overcome.

      I think it is a great fallacy to believe that security and desktop usability are mutually exclusive concepts. OSX is a great example of a unix based OS with as good (or better) usability than what MS provides and still provides a secure system. In OSX the default user doesn't run as root. It works just fine for Apple and their users *expect* a high degree of usability. Apple's users are frequently non-technical (artists for example) who just want to use their computer (just like good ol' granny). Why would this not work elsewhere?
      I agree that it is indeed time to reexamine our ideas about security but we need to find ways to strengthen the model *AND* improve usability wherever possible instead of making life easier for malcontents.

      Disclaimer: I haven't ever owned a Mac (although the Minis are darn cool [me want!])...
      --
      G. Washington on Government "it is force. Like fire, it is a dangerous servant and a fearful master."
    20. Re:Devil's advocate by Anonymous Coward · · Score: 0

      No, it cannot bind to privileged ports, overwrite your system libraries, disable your firewall, trojan your ssh program, or overwrite the memory of other processes if it is not root. If you have other security mechanisms in place, you can prevent it from installing keyloggers, using sendmail, or sending out packets on unprivileged ports, if you are not root. Remember, security is a process, not a product.

    21. Re:Devil's advocate by Cro+Magnon · · Score: 1

      Desktop doesn't neccesarily mean single-user. I realize that this is Slashdot, but in the real world, people have wives! Some of them might even have one or more children! Having seperate user accounts is a clear advantage in this case, and it might be a good idea for the admin to run as a regular user too, so his wife doesn't make him sleep on the couch because he got r00ted and her favorite recipes got deleted.

      --
      Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
    22. Re:Devil's advocate by Anonymous Coward · · Score: 0

      bind to privileged ports

      Who cares.

      overwrite your system libraries

      But it can effectively replace them for the compromised user, by inserting itself in a startup script and replacing the user's shell; from the shell, you can control everything, substitute anything, even trap system calls...

      disable your firewall

      Can you access the internet? So can your virus.

      trojan your ssh program

      So, yes, it absolutely, most certainly can.

      or overwrite the memory of other processes if it is not root

      It can control every instruction of code you execute if it wants.

      If you have other security mechanisms in place

      What would these be?

      you can prevent it from installing keyloggers, using sendmail, or sending out packets on unprivileged ports, if you are not root

      How?

      Remember, security is a process, not a product.

      Of course. So?

    23. Re:Devil's advocate by McDutchie · · Score: 1

      SELinux is already in the linux kernel, at least I get it as an option when compiling 2.6.11. Of course, most distributions have not turned it on yet, but that's a different matter.

    24. Re:Devil's advocate by JohnFluxx · · Score: 1

      Yeah, it was promised to be included about 6 months ago, but I hadn't been following developments, so decided to err on the safe side.

  71. I agree... by anthony_dipierro · · Score: 1

    For a single user system, there's no good reason not to run as root. I regularly log in as root on all the systems I own.

    1. Re:I agree... by anthony_dipierro · · Score: 1

      not to mention the fact that having to use root password to do so many things is a pain in the ass and causes users to get too used to just typing in their password without knowing what they are doing

      Absolutely. I'd be fine with having a root and a non-root account on my desktop machine if the root account were basically never used. Essentially make root access needed as often as access to the bios, and make it just as hard to access. You can't do it by typing in a password, to access the root account you have to physically reboot the machine. Now that would be secure. Anything else is just a false sense of security.

    2. Re:I agree... by MikeBabcock · · Score: 1

      For the average user, using root is almost never necessary -- software upgrades are run in the background with appropriate privileges by cron, etc.

      Power users know how to set up sudo and or other systems to give themselves root access for specific reasons.

      PS, don't worry -- we don't hire people who log in as root here. I administer the boxes -- and I almost never need to use root.

      --
      - Michael T. Babcock (Yes, I blog)
    3. Re:I agree... by anthony_dipierro · · Score: 1

      software upgrades are run in the background with appropriate privileges by cron, etc

      And how does one get a software package into such a directory to be run by cron, etc?

      PS, don't worry -- we don't hire people who log in as root here. I administer the boxes -- and I almost never need to use root.

      If you can administer a box while almost never needing to use root, then I highly doubt your box is any more secure than your admin account. Administration requires admin access. Now maybe you've given such access to non-root accounts, but that seems more likely to bite you in the ass than just running as root in the first place.

    4. Re:I agree... by MikeBabcock · · Score: 1

      A) Fedora, RedHat, et. al. download and install such updates themselves. In the case of RedHat Enterprise Linux (not a desktop OS), you can schedule updates from their website.

      B) Root can read any directory on the computer, and since the scripts to do the updates should be installed by the vendor, your first question is moot.

      You don't administer many solid boxes do you?

      --
      - Michael T. Babcock (Yes, I blog)
    5. Re:I agree... by anthony_dipierro · · Score: 1

      Root can read any directory on the computer, and since the scripts to do the updates should be installed by the vendor, your first question is moot.

      If all you want to run is programs provided to you by the vendor, perhaps.

      You don't administer many solid boxes do you?

      Not using Fedora or RedHat. My distro is Linux From Scratch.

    6. Re:I agree... by MikeBabcock · · Score: 1

      No offence to yourself or anyone else who does "Linux from scratch" boxes, but without real personnel and research time, its often a complete waste of effort. It is very difficult to keep up with the patching that goes on across a few hundred disparate packages on your system. This is where a good vendor comes in.

      --
      - Michael T. Babcock (Yes, I blog)
    7. Re:I agree... by anthony_dipierro · · Score: 1

      Well, it's not a complete waste of effort, because I've learned a lot about Linux in the process. As for keeping up with patches, for the most part I don't. I don't have a hundred packages, and I'm the only user on the system. So getting the latest and greatest version of packages would be a waste of time.

      If you're maintaining a system for more than just yourself, this is of course a completely different story. Of course, I'm not sure I'd use Linux at that point. Not building a free system from scratch kind of defeats the whole purpose of Linux in my opinion.

    8. Re:I agree... by MikeBabcock · · Score: 1

      For context ...

      I maintain a home computer which is some amalgamation of RedHats 6.x through 9 + Fedora Core updates.

      For work, I maintain almost 100 RedHat based servers for clients, some with only dial-in access via modem for support.

      --
      - Michael T. Babcock (Yes, I blog)
  72. Yawn... by ubiquitin · · Score: 1


    Michael Robertson has such a long history of controversy for the sake of controversy that anything from this guy whatsover goes like this: | /dev/null

    A bigger deal is Apple withdrawing setuid/gid in one of their OS "updates." Ouch.

    --
    http://tinyurl.com/4ny52
  73. linspire strongly encourages you not to run as roo by acomj · · Score: 1

    linsprire during installs now strongly encourages you to set up a root and user account.

    So he's just posturing, but back in the day you had to assume anyone who got on your system could get root if they wanted it.

    He's right that your data is very important, most important than the OS files..

  74. how about toor... by grumpyman · · Score: 1

    We allowed one of our partner to log on to our system to do some setup/testings as 'toor' with a different password but with root privileges. Oh yeah, like the unimaginable, he did a 'rm -rf *' on the wrong window with / as current dir. At least he can't logon to our production server to do the same thing.

  75. Intoxicated root by VersatilePrimate · · Score: 1

    All I have to say is that running a Gentoo install while completely drunk is extremely entertaining, even if extremely hazardous for your health.

    Drunken root is not like being a drunken master of kung fu.

    1. Re:Intoxicated root by aventius · · Score: 1

      not as hazardous as drinking and dialing ex girlfriends... wait... this is slashdot... what are girlfriends?

      --
      [insert lame joke here]
    2. Re:Intoxicated root by Hank+Reardon · · Score: 1

      Bah! It's good practice!

      If, when you login after the box "turning off" in the middle of intoxistall, you can figure out where you screwed up in the install process and fix it, you'll be able to fix any Gentoo problem you run into!

      :)

      --
      There's so little difference between politics and jihad lately...
    3. Re:Intoxicated root by fishbowl · · Score: 1

      "All I have to say is that running a Gentoo install while completely drunk is extremely entertaining, even if extremely hazardous for your health."

      It's impressive that your metabolism allows you to remain "completely drunk" for that long!

      --
      -fb Everything not expressly forbidden is now mandatory.
  76. Re:uh... by mios · · Score: 1

    it only takes one time ... and yes, I've actually done something close enough to that to realize that one time is enough.

  77. No no, he means "safe for Microsoft". by Morgaine · · Score: 1

    He's dead right, Linux users running everything as root is the only strategy that would offer any safety for Microsoft.

    --
    "The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
  78. Another reason never to use Linspire by Husgaard · · Score: 1
    With the CEO of Linspire saying this I cannot trust Linspire to care about security.

    I never liked Linspire, but this statement has just become my #1 reason to recommend users against ever trying it.

    There are lots of fine Linux distributions around that care about security, so why choose one that doesn't care?

    1. Re:Another reason never to use Linspire by Jose · · Score: 1

      thats the difference between a CEO and CTO.

      --
      The basic sleazeware produced in a drunken fury by a bunch of UCBerkeley grad students was still the core of BIND. --PV
  79. Apartment Analogy by whib · · Score: 1

    I view security somewhat as an analogy to homes. If I live in my home, and I am the only person who lives in my home, I can have all of the doors keyed the same, and use the master key, and have the master key on me at all times. If my house is broken into, it is all my stuff. This is the same as running as root all the time. Now, I may have things that I'd prefer not to get stolen, then I'd have a safe. Now, I wouldn't have a house safe that is key operated that uses the same key as my front door. I just wouldn't. I might store a spare house key in my safe so that I know how many house keys are out there, etc, but I would not have my safe using the same key as my front door. This would be similar to having my own computer system, but not running as root. Instead I would use a non-privileged user account for most of my day to day operations and only use "root" or "administrator" in the case where I specifically need those privileges. Now, in a multi-person environment, things change. Let's use an apartment (or hotel, if you please) as the word picture this time. If I managed an apartment, I might have a master key for the entire building. My life would be a heck of a lot easier if all I ever carried was the master key. I could go into whatever apartment I wanted as I pleased. I would only ever have to worry about losing that one key. This is the same as running as root on a multi-user system. I have access to everything. Back to the apartment model. If the apartment manager is getting groceries one day, and someone steals his ONE key, the whole apartment is now wide open to the person who possesses the key. A smart apartment owner might instead have a safe where he stores that key. The only key he would carry would be the key to open his apartment, and that key would not open his safe. If his key were now stolen, there would be a lot less risk to the other people in the building. Similar in a multi-user computer situation. As a user (or a tenant), I'd rather have an admin (or manager) who is security conscious enough to take basic precautions. As an admin (or manager), I'd rather take some simple precautions as come cya. For a one person system, root or not root.... not a large difference.... that being said, not following "good practices" is a bad practice.

    --
    -- www.WhereHaveIBeen.com
  80. Re:uh... by evn · · Score: 1

    There's an old bit of Unix Folklore about recovering from an "rm -rf /" by Mario Wolczko that originally made the rounds on Usenet in 1986.

    It's one heck of a read.

  81. The most important thing on my computer... by harmic · · Score: 3, Insightful
    Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares?

    The most valuable thing on my computer is probably the user name and password to my internet banking facility.. Not that I store them on the machine but I do type them in. Maybe running as non-root does give you access to all the data in a users home dir but it sure makes it more difficult to overwrite those libraries he's talking about with keylogging trojans that will harvest my passwords.

    1. Re:The most important thing on my computer... by fishbowl · · Score: 1

      "The most valuable thing on my computer is probably the user name and password to my internet banking facility"

      If you stored that anywhere except in your head, your problems are not related to technology, but in your use of it.

      That said, I myself keep similar things on a gpg-encrypted file on a keychain usb thingy, and everything needed to recover it except the private key and the passphrase are stored on that keychain.

      I understand the risk of copying this encrypted file, e.g., putting it in public places, and that is the level of risk I'm comfortable with.

      On the other hand, I would be no less comfortable with this arrangement if there are "root" users or anyone else with access to it. Deleting it is another story, of course.

      When I hear that 2048-bit gpg is being casually cracked by people seeking access to nickle-and-dime bank accounts, I'll change my risk quotient.

      --
      -fb Everything not expressly forbidden is now mandatory.
  82. Re:Agreed - NO WAY! by Indes · · Score: 1

    It IS THE SYSTEM, NOT THE USER!

    Sorry for the caps but its true..

    The user can accidently run things that (s/)he may be ignorant of, and unfortunatly have unforseen side effects.

    For example surfing to a webpage that contains an virii/spyware/crap/etc that executes upon load because of bugs in specific browsers...

    Clicking an indiscriminant link by accident IS the users fault, but it shouldn't be up to the system to do everything the user says like a brainless zombie mornic mornon (yes thats moron x2 - because not thinking before doing anything is a moronic thing to do, System and user alike!)

    So as you can see, IF YOU FALL ASLEEP AT THE MOUSE, and you click something deadly because you're system is a moron - reply to this with your happy thoughts about your infected/lost data.

    Thanks! I love a good laugh :-)

  83. Stress Test by qbasicnewbie · · Score: 2, Funny

    This man obviously hasn't met my sister: She attempts to open music files by selecting them in groups of 100, resulting in 100 sessions of xmms...I can only imagine her with root access.

    1. Re:Stress Test by smart.id · · Score: 1

      In a better designed situation, such as the one I have on my Windows system with Winamp, selecting songs in Explorer queues them up to be played in the playlist. Wouldn't that make more sense? Why would you need two copies of xmms running at the same time anyway?

      --
      blog & fiction: jd87
  84. Single user machine by pyth · · Score: 1

    When your entire home directory is deleted - all your work, all your bookmarks, all your records, all your porn - does it really MATTER that the rest of your system is still intact?

    This message brought to you by the Society For Capability-Based Security (SFCBS).

    1. Re:Single user machine by Todd+Knarr · · Score: 1

      Yes, it does. If the rest of the system, particularly the system programs like your backup and restore utilities, is still intact you can quickly recover your deleted home directory from your backups. You did keep backups, right?

      Also, viruses and trojans will tend not to be able to infect your system, since the programs on the system the malware could infest to spread further will tend to be installed as part of the system and not be owned or writable by you.

      Just because a system only has one user doesn't make it a single-user system.

  85. Redundant Mod by Anonymous Coward · · Score: 0

    Is equally hilarious.

  86. I know he's talking about Linux by jim_v2000 · · Score: 1

    But I've been running my Windows box with the Admin account forever. Never have I had any trouble with haxors or viruses. I would say this is due to the firewall and AV software I have running. Now, could I run an account with less privileges and not worry about AV or firewall? Maybe, but I'd be thowing out a lot of convenience of being able to change settings and install software without putting in a password.

    Plus, I think some people are missing his point. OS's can be restalled. HD's can be reformatted. Files can be replaced. But once important personal data is in the wrong hands, there is not getting it back. He's saying that a system crash for most users is not as critical as having important data stolen. So as far as securing data, (for as single user) running as root is no more secure than running a limited account.

    --
    Don't take life so seriously. No one makes it out alive.
    1. Re:I know he's talking about Linux by PDAllen · · Score: 1

      Major difference between unix and windows.

      In unix, you personally can run as root, this allows you to screw up (more easily than a windows admin in some ways). Your background tasks, however, should be running in dummy users. mySQL runs as mysql, for example, it cannot e.g. see files outside its /dbs. If someone runs an exploit against it, then they only get access to a very small part of my system, not including most of my data.

      If on the other hand I try that with windows, then the mySQL exploit allows access to all my personal data straight away. I can't (easily) have mySQL running as a dummy user with minimal privileges.

      That's not really relevant to 'you should not run as root', though, that's a built in security flaw with windows.

  87. I'll run as root... by womanfiend · · Score: 1

    I'll run as root on my machine, but I think I'd better set up a seperate, non-privliged account, for Mr Robertson. I don't think he's quite ready for / power.

    --
    Jon Green Cheyenne
  88. Multiple keyboard confusion by Anonymous Coward · · Score: 0

    Ever been root on one keyboard, and email (or Instant Message) on another?

    Sometimes it's easy to start typing a message on the wrong keyboard, and into the root shell. Usually they won't be valid commands, but we all know Murphy's Law.

  89. STFU Michael Robertson by Greyfox · · Score: 1
    You're about as well informed as a slakware-running cluebie going into a linux IRC channel for the first time. Actually some of THOSE are better informed than you are. Not only are you ignoring the 35-some-odd years of collected UNIX sysadminning experience that says "don't do that," you're completely failing to take into account that running in a priviliged account full time is what makes things like virusses and trojans possible in the first place. Moreover, any possible user friendliness advantage to runing as root full time is miniscule, even (especially?) on a single user machine.

    Many of us are the de-facto support for Windows users, usually friends and relatives. Do you know how incredibly easy it is to have a Windows system infected? You pretty much have to turn the computer on and that'll do 'er. Do you know why this is? It isn't the crappy kernel anymore -- they went over to an NT style one in XP. The reason it's trivial to infect a windows system is that the home users run with the administrative account all the time. That's also why real IT companies are somewhat harder to infect than home users -- they won't typically even give their users the admin account/password and security policies usually make it a firable offense to run as the administrative user full time (Basically you can do it until your system gets pwned and fucks the company network, then you're out.)

    And while I typically expect upper management to be clueless, the CEO IS the spokeperson for their company and when a Linux Company tells me it's OK to run as root, I am typically discouraged from doing business with them because my impression of them is that they're a bunch of IDIOTS!

    --

    I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

    1. Re:STFU Michael Robertson by anthony_dipierro · · Score: 1

      you're completely failing to take into account that running in a priviliged account full time is what makes things like virusses and trojans possible in the first place

      So you are claiming that it's not possible to have a virus or a trojan which runs as a non-privileged user? Sounds like you're the one who has no clue.

    2. Re:STFU Michael Robertson by MikeBabcock · · Score: 1

      Its not possible on a normally configured *nix box for a virus running as a normal user to reconfigure the OS and harm anything but that user's files (arguably a big problem in and of itself).

      Given that executables are typically only writable by root (those under /usr/bin, /bin, etc.), these can't be infected the way Windows binaries are.

      Rootkits, the primary plague on *nix boxes depend on being able to get root on your computer to work properly (duh).

      --
      - Michael T. Babcock (Yes, I blog)
    3. Re:STFU Michael Robertson by anthony_dipierro · · Score: 1

      Its not possible on a normally configured *nix box for a virus running as a normal user to reconfigure the OS and harm anything but that user's files

      Even if this is true (it's arguable as most "normally configured" *nix boxes have root exploits), saying that a virus can't do certain things is not at all the same as saying that you can't have a virus.

      arguably a big problem in and of itself

      And that's the argument that I'm making.

      Rootkits, the primary plague on *nix boxes depend on being able to get root on your computer to work properly (duh).

      They also require the /bin directory not to be on a read-only volume, or for root to be able to remount the volume writable, which can't be said of a well-secured *nix installation.

    4. Re:STFU Michael Robertson by Greyfox · · Score: 1
      Son, I've been doing security when you were in diapers.

      Of course you can have a virus or trojan that runs in non-privileged space, but the scope of the attack is greatly limited. You can't compromise system files without first finding a local root exploit -- admittedly those are a lot easier to find than remote ones, but if you keep your system up to date it's well beyond the means of any skript kiddie and that's where the majority of the activity takes place. A while back there was a proof-of-concept linux virus that exploited the old su buffer overflow (IIRC) and compromised the system that way. And of course even back in the '80's it was pretty common for the universities to grind to a halt around christmas time when the christmas tree card trojan spread like wildfire through E-Mail.

      Running as a standard user account is simply another safety measure among many which form the ongoing process which is security. Every step that you cut out of the process raises the danger not only to your personal information but also every other person on the net. Microsoft's already done enough damage on that front, we don't need some fuckwit linux company to do any more damage to global network security.

      --

      I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

    5. Re:STFU Michael Robertson by anthony_dipierro · · Score: 1

      Son, I've been doing security when you were in diapers.

      Maybe your techniques made sense then.

      Of course you can have a virus or trojan that runs in non-privileged space, but the scope of the attack is greatly limited.

      Not on a single-user system it's not.

      You can't compromise system files without first finding a local root exploit

      So what? You don't have to compromise system files to make a serious exploit on a single-user system.

      Every step that you cut out of the process raises the danger not only to your personal information but also every other person on the net.

      If you're running a single user system, then your regular user account has access to your personal information, can bind to any high port, and can make a connection to anyone on the net. That covers just about every exploit already out there in the Windows world.

    6. Re:STFU Michael Robertson by Greyfox · · Score: 1
      Not on a single-user system it's not.

      At the very least it still comes down to rm -rf /home/user (Worst case) versus fdisk, format and reinstall. It still comes down to your logs remaining intact so that you have a chance in hell of tracking the offender. It still comes down to possibly noticing that files are popping up in your home directory or /tmp that shouldn't be there. Sure you might not notice the extra activity, but you're a lot more likely to notice stuff going on in your user space than elsewhere on the system, especially once steps have been taken to hide their tracks.

      --

      I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

    7. Re:STFU Michael Robertson by anthony_dipierro · · Score: 1

      At the very least it still comes down to rm -rf /home/user (Worst case) versus fdisk, format and reinstall.

      I don't know about you, but if I've been compromised at a user level I'm not going to trust that that's as far as things went. I'm going to fdisk, format, and reinstall.

      It still comes down to your logs remaining intact so that you have a chance in hell of tracking the offender.

      I'm not all that interested in tracking the offender, and I suspect the average computer user is even less interested than I am.

      It still comes down to possibly noticing that files are popping up in your home directory or /tmp that shouldn't be there. Sure you might not notice the extra activity, but you're a lot more likely to notice stuff going on in your user space than elsewhere on the system, especially once steps have been taken to hide their tracks.

      Seems quite unlikely, especially for the average computer user, but I suppose it's slightly more possible.

  90. I can reveal it by plj · · Score: 4, Funny

    127.0.0.1.

    Just don't tell anyone that it was me who told you.

    --
    “Wait for Hurd if you want something real” –Linus
    1. Re:I can reveal it by sharkey · · Score: 2, Funny

      Neato! He likes the same big-titty pr0n that I like.

      --

      --
      "Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
    2. Re:I can reveal it by johnashby · · Score: 1

      ping 127.0.0.1
      Destination host unreachable.

      Maybe I need to log in as root?

  91. rm -rf ~/ by Anonymous Coward · · Score: 0

    So they'd type that?

  92. Re:uh... by FuzzyBad-Mofo · · Score: 1

    It happens. I once accidentaly typed an extra space when deleting core files on a production system:

    rm -r * core
    Goodbye to a full day of work.. Of course, now that I am more experienced with *nix systems, I see that there were several issues with the system configuration:
    • The account we used for testing was equivalant to root (sound familiar?)
    • The system should have had core dumps disabled, no reason for that on a production system
    • The data files should have been placed in a separate directory, like /var, instead of just using the current directory.
  93. You're MISSING THE POINT! by jonesy16 · · Score: 3, Insightful

    Before you blow everything out of proportion, take a second to look at a few things from a different perspective:

    1) The end user of Linspire is most probably a windows user trying to switch to something cheaper. The odds of Linspire being heavily used in a multiuser environment are bleak at best.

    2) He makes a valid point, the most valuble information on your computer are things stored in your home directory. Credit card information, social security, emails, etc. Guess what . . . `rm -rf` will eliminate all of that even if you aren't root. Who cares if you accidentally wipe an X library, a reinstall will fix that, it won't get back your emails and resumes.

    3) Everyone's argument for the flaw of running as root seem to stem from services running as root, which is something the enduser of an operating system like Linspire shouldn't be expected to fix anyway, nor will most Linspire users be running apache servers and mysql servers, I'm just guessing at that.

    A windows user or a linux newbie doesn't want to remember several account passwords just to change the IP address of their computer, or to reboot, or mount an external hard drive, or start Samba, etc. They want to know that they have permission to do those things out of the box. That's how windows is set up, that's what they want. Security should be handled by turning chrooted service invocation, firewalling, etc.

    This isn't FreeBSD, tailor to your customers and make them happy, without them you don't have a business.

    1. Re:You're MISSING THE POINT! by fishbowl · · Score: 1


      "1) The end user of Linspire is most probably a windows user trying to switch to something cheaper. The odds of Linspire being heavily used in a multiuser environment are bleak at best."

      Bleak? Single computer system for a family with two or more members? Seems like a common enough scenario! Even the most basic voice mail plans arrange for several users to each have a separate mailbox with a different password. This belongs in the same category!

      Of course, I do appreciate that even with a slight clue, Linspire can be configured with a proper getty and/or xdm, users, even a file system with ACLs etc.

      --
      -fb Everything not expressly forbidden is now mandatory.
  94. Other insights from "Root" Robertson..... by scupper · · Score: 1
    Michael's Minute: Predictions for 2005
    http://www.linspire.com/lindows_michaelsminutes_ar chives.php?id=153

    Predictions for 2005

    It's a new year, so time for a predictions column. But first,we'll do a review of last year's predictions to see what I got right - and what I got wrong.

    1) Software and movie companies embrace P2P = Cheaper products for consumers
    Some entertainment companies are using P2P to deliver TV-quality video and games through companies like RedSwoosh. But sadly, the majority are seemingly following in the footsteps of the music labels by trying to sue to slow or block the technology rather than figure out how to use it to make more money. This is a miss.

    2) Microsoft moves from growth to profit
    This year, Microsoft announced the biggest ever one-time dividend, which is the classic move of a "value" company. Growth companies never pay dividends because they use their cash to continue to grow. In spite of a wide range of efforts spanning TV, PDA, game consoles, etc. Microsoft cannot find another profitable venture outside their core operating system and office suite business. This was a hit.

    3) $499 Linux laptops under the Christmas tree
    On December 17th, Walmart.com began selling a Linspire laptop for $498. I was $1 off, but this is a definite hit.

    4) Microsoft announces plans for Microsoft Office for Linux
    This was my President, Kevin Carmony's prediction. Microsoft has not yet announced any products for Linux. With OpenOffice continuing to make great strides and OpenOffice 2.0 on the horizon, there will be more pressure for Microsoft to respond. But for now, this is a miss. Remember that OpenOffice 2.0 will be on display at the Desktop Summit on February 9-11th, in San Diego. So, register now!

    50% accuracy from 2004, so let's look at 2005.

    1) After buying IBM, Lenovo leans toward Linux
    China-based Lenovo just received US government clearance to purchase IBM's PC business. IBM executives have assured the IT world that the quality and service will remain, and I hope it does because I'm a big Thinkpad fan and own several of the X series laptops. But something must change or Lenovo will have paid $1.75 billion for the right to lose money on every IBM PC they sell. Over the last 3.5 years, IBM has sold about 30 million computers and lost $965 million dollars - or approximately $33 per computer. To reverse their fortune, Lenovo needs to find a way to have $50 better economics on every PC so will they not only break even, but they will generate some profits. IBM already uses Chinese labor in their plant in Shenzen to manufacture their PCs - so there won't be much savings there. Lenovo may be able to buy hard disks, memory or other parts slightly cheaper than IBM because of greater economies of scales, but at best this will be less than $10 per machine. The only place where significant savings can be generated to turn their PC business around is the operating system and office suite. Instead of paying Microsoft $100-$300 per machine for Microsoft Windows XP and Microsoft Office, they will ship Linux with an office suite and pay just $5-$10 on some of their product line. This will give them distinction from the well-entrenched Dell and HP computers they must compete with.

    2) Windows Media Center suffers BSOD (Blue Screen Of Death)
    You don't have one mega kitchen appliance, you have dedicated appliances tuned to each task. The same is true in your living room. The Windows Media Center is an admirable goal, but misguided implementation. They are trying to cram every function into one box and it makes the device expensive and unreliable. That was illustrated with Bill Gates' recent high-profile demo at CES (Consumer Electronics Show) of Windows Media Center, which suffered an embarrassing BSOD. Remember this i

  95. What is scary by fermion · · Score: 1
    Is this guy is buildign consumer machines. If he was building geek machines or high end bussiness machines, it wouldn't matter. But he want to be in the low end consumer market. Where Lupe and Joe buy the computer thingy.

    And when building mass consumer devices, one has to add extra junk, even though it may not be strictly neccesary. You add signs not to open things that no one should ever open. You add grills so that no one sticks a finger in places where no one should stick a finger in. You put covers on parts that really don't need covers, but need to be left alone. Making a consumer device requires a bit of extra thinking. Things are done not becuase they are required, but because they are neccesary.

    So, consumers need a special mode to run safely. It is precisely the lack of such a mode that made windows crash and burn when it becam a commodity. No one needed it to start. But for some reason MS never developed the consumer OS. We have now have the chance to do so using Linux.

    --
    "She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
  96. The More The Merrier by Anonymous Coward · · Score: 0

    Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well.

    Actually, I'd think this would be an argument for more user seperation, rather than as a somewhat reasonable point. Under his paradigm, it would be better to seperate some tasks by (non-super) user, such as browsing for porn, online bill paying and educational games for your child. Clearly, the data, risks and reasonable problem vectors are different for these 3 activities and a quick out and login is not that much to ask for a little extra peice of mind.

  97. Computers vs. Cars by cyberia625 · · Score: 1
    "Well, you could accidentally drive into a wall as well, it doesn't mean we should make all cars drive at 10 miles an hour."

    While this is true, he neglects to mention one very important thing: you don't have to pass a test in order to mess around with your computer. Unless someone standardizes a competancy test for computers which loads on first login for each user, there ought to be some level of protection for all the users on that machine (or any other machine connected to the internet) to prevent the "unlicensed driver" from causing harm.
    1. Re:Computers vs. Cars by fishbowl · · Score: 1

      I don't need any permission from a state, or any kind of license to buy a car, have it delivered to my property, and drive it into a wall.

      However, I *do* need permission from the state to build the *wall*. How do you like that?

      --
      -fb Everything not expressly forbidden is now mandatory.
  98. Re:Agreed - NO WAY! by sterno · · Score: 1

    The user can accidently run things that (s/)he may be ignorant of, and unfortunatly have unforseen side effects.


    This brings up a very good point about people who are new to computers. Most of the people that I've seen who have difficulty learning how to use a computer fear them. Actually, not the computer per se, but rather they fear that they are going to break it. If they try doing something they don't fully understand, they are afraid they will cause irreperable harm to the computer.

    Strong priveledge seperation makes this a non-issue. As a regular user on linux, I would have a seriously difficult time causing serious harm to the box through anything I might do. I might screw up my data, or the appearance of things, but in the end, I cannot break the OS.

    Under Windows, I can do all kinds of bad things. I've heard countless stories of people who are smart about how they use their computer coming home to find that it's useless becuase their kid downloaded a worm. It's one thing when the kid breaks their account, it's another when the whole operating system has to be re-installed from scratch.

    --
    This sig has been temporarily disconnected or is no longer in service
  99. No reason to be root by DrXym · · Score: 1
    For day to day use, there is absolutely no reason at all to be root. In fact doing so is as people have said likely to see you (or some trojan) screw your machine in an irreparable way.

    OS X has it right in this regard. Even the administrator runs as a normal user for most of the time. If they want to install something such as an update, they must use sudo to do it. At the very least that means that trojans have an extra hoop to jump through.

    On Windows (or Linspire) if the vast majority of users are running with root it presents a much more attactive attack surface. Aside from that, it prevents users from themselves - e.g. stopping some dumb user from cleaning up their machine like deleting etc/ for example.

    Michael Robertson should know this. In fact I suspect he does know this and employs this kind of rhetoric because Linspire has traditionally thrown all the security switches to make it one of the least secure Unices ever. At least it offers you the choice of a logon and password which is a start. Really though Linspire should adopt the Mac model. The fact that OS X makes it virtually transparent means that it's not a massive burden on the user to figure out.

    Even from a business perspective, less destroyed machines and less exploits is obviously good sense. But who knows - perhaps Linspire loves support calls for the revenue they generate.

    1. Re:No reason to be root by Etcetera · · Score: 0

      On Windows (or Linspire) if the vast majority of users are running with root it presents a much more attactive attack surface. Aside from that, it prevents users from themselves - e.g. stopping some dumb user from cleaning up their machine like deleting etc/ for example.

      Well, here's one of those philosophical differences then... Why doesn't 'rm' put forth an extra warning when you delete a directory named "/etc"? Or any of the "system" directory paths known to be important?


      cd /
      rm -fR etc
      *** Deleting the '/etc' directory may cause your computer to be unable to reboot properly and may cause serious loss of data. Are you sure you want to do this? (yes/NO)


      It could have a list of directories that it checks again (/etc, /* /usr/local, etc...). Or perhaps a hashed listing of all directories owned by any given package (on RPM-based systems). Someone tries to delete it manually and they get a confirmation like this:


      *** The '/usr/local/foo' directory is owned by the foo-devel-3.25-1 package. To remove this directory, you should uninstall the package using rpm instead; manually deleting files may cause your system to fail. Proceed with delete? (yes/NO)


      Special warnings could be bypassed with rm -ffR /blah

      See this Slashdot thread for more discussion... but basically, why should I have to enter a separate and artificial "mode" in which to do things to my computer? Isn't it the system's fault if it allows me to do something it knows will be "bad" without warning me?

      Just some $.02

    2. Re:No reason to be root by fishbowl · · Score: 1

      For your example, "/etc/", sure you could configure rm to know about it. But how does it know what's critical and what's not? This is a wide open system, and to introduce constraints for the kind of safety you'd like, will also limit the flexibility of the system.

      It's easy to identify boundary cases, such as, "deleting / shouldn't be possible", but everything else is difficult.

      But a far more important element to "root" versus "user" is the fact that resource limits can be placed on users, that can't be placed on root. In my opinion this is being overlooked in the discussion, as is the whole notion of privacy among users of the system, or the ability of the user to hand over the keyboard to someone that he does not have to trust especially well.

      How can two mild adversaries use the same computer if they can't even nominally invoke some kind of privacy among each other? I don't mean "user versus hacker" here, I'm taking about maybe employees in the same role (who might compete), members of a family (who might have very good reasons to need privacy), or students (who might have ethical considerations).

      Privacy and security don't have to be "all or nothing" considerations, but often it seems only the worst-case scenarios are ever considered (e.g., breaking in, theft, etc.). There are tons of reasons for security to exist that don't require such extreme scenarios to be reasonable.

      --
      -fb Everything not expressly forbidden is now mandatory.
    3. Re:No reason to be root by smash · · Score: 1
      Your "solution" will break compatibility with a vast library of software, as well as script compatibility with virtually every other unix under the sun.

      The solution, as implemented by OS/X, Ubuntu, etc - is to deny normal user access to the filesystem, except via either a package manager, or the root account.

      Isn't it the system's fault if it allows me to do something it knows will be "bad" without warning me?
      Nope.

      This what package managers are for. If you can't be trusted to play with low level tools, with root access, then use the package manager provided to do things from a high level application, which WILL provide the relevant dependancy warnings, etc.

      Logging in as root and playing with rm is deliberately circumventing the "protection" offered by package management (which, if need be, can be made to run SUID, instead of requiring root access, if you're REALLY that against typing a password in for it).

      Put another way - is it the fault of Black and Decker if you remove the safety guard from your circular saw so you can go cutting things up "better", and you end up cutting your arm off with it?

      root is all powerful. By deliberately and consciously logging into a shell as root, you're taking things into your own hands. All safeguards are off. There's no point in extra command line switches - you've already bypassed the safety net.

      Your command line switch to rm could instead be implemented with sudo (funnily enough, as it already is with Ubuntu), and logging in without root access. To remove files that are likely to affect your system - you need to run sudo rm. :)

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    4. Re:No reason to be root by VGPowerlord · · Score: 1
      Any protection rm has, you're already circumventing with the -f option.

      -f
      Do not prompt for confirmation. Do not write diagnostic messages or modify the exit status in the case of nonexistent operands. Any previous occurrences of the -i option shall be ignored.
      -- The Single UNIX Specification, Version 3
      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    5. Re:No reason to be root by DrXym · · Score: 1
      Well, here's one of those philosophical differences then... Why doesn't 'rm' put forth an extra warning when you delete a directory named "/etc"? Or any of the "system" directory paths known to be important?


      Well I'm supposing you actually mean KDE since its most likely that the user would rearrange their machine from there... But rm does have a check to prevent directory deletion but its overridden with a switch. If you say "rm /etc" nothing happens. You must say "rm -rf" to forcibly and recursively delete what you tell it to.


      But on KDE, how should it know what is safe to delete and what isn't? After all, you're running as root so you can delete everything. I don't use KDE so perhaps it has some kind of crude checks already that warn you when you start dragging /etc, /usr, /sbin into the trashcan. But how does KDE know that some-random-file is safe for deletion when another one isn't? In a perfect world (which Linspire isn't since it wouldn't in this mess if it were), perhaps access controls are enabled and KDE might be able to figure it out what is a system file, or the group might be "system" or something. But that requires every package install using those controls which I rather doubt Linspire does either.

  100. I want my data kept safe! by Anonymous Coward · · Score: 0

    If we all login as root then everyone could see my hidden pr0n dir!!!

  101. i have root issues by Anonymous Coward · · Score: 0

    i have run os x and one of the ppc linux distros as root and and i find that doing so just messes stuff up. i can only say that it just never works right. i wish i had some specific instances, but i only know i don't do it regularly for a reason (or many reasons, as i've had the root gnomes bite me once too many times). thanks, but i'll be computing as admin, mr. linspire guy.

    clackerd (forgot my login at work)
    temporary signature

  102. 99% by blackbear · · Score: 2, Insightful

    The reason that Robertson didn't get the answer to why not to "run as root" is twofold.

    1.) He didn't want to hear the answer when it was told him.

    2.) probably 99% of people who know that you shouldn't "run as root" don't know absolutly why themselves. They have a pretty good idea, but someone they respect and trust (and who is correct) told them it was stupid.

    The other 1% who could have told him why, weren't consulted. Nor will they be.

    It's no accident that Linspire (Lindows) is modeled after Windows, and it contains Windows' greatest fundamental security flaw.

    1. Re:99% by alexmagni · · Score: 1

      ... and you too forgot to mention the reason why

  103. My question to "Root" Robertson by scupper · · Score: 2, Funny

    Mikey, what is a bot? And how are they born?

  104. Would this work? by Anonymous Coward · · Score: 0

    For most of my work, I log in as Fred.

    For anything on the net, I log in as Bill.

    Almost none of my data requires web access to process. Anything I download as Bill is readable by Fred.

    This seems pretty bullet proof to me. Am I kidding myself?

  105. Boiler-plate troll by AngryElmo · · Score: 2, Interesting

    almost Word for word, this guy has been posting this same text around different sites for 2 years. It has sort of reached goatse status (ie effing annoying). Just ignore it

    1. Re:Boiler-plate troll by Anonymous Coward · · Score: 0

      Googling around shows he goes by "Egg Troll" and apparently he's the admin of Kuro5hin.org (linux powered according to his site).

      http://www.kuro5hin.org/user/egg%20troll

      Gotta love whois:
      rusty@kuro5hin.org
      Rusty Foster
      8605 Cameron St. Suite 200
      Silver Spring, MD 20910
      USA

      (202) 276-6774

      Love,
      Jeff

  106. Unix Hater's Handbook by suitepotato · · Score: 1

    It says it all and you're a fool to ignore what Unix started out as and what dangerous idiocy is still contained within its frightening depths. Like a nuclear reactor, it has its uses but should always be treated with great respect. If you're relying on the reactor to produce needed power and not contaminate the countryside, then leaving the control room wide open to any child that walks in is like leaving the average person in root on any given *nix box that is being relied on. Either thing is stupid and asking for trouble.

    It's not for nothing I half-rejoice at the option of using Run As on XP Pro but still await the concept of the machine not running every freaking thing as root. Until then, Windows will still be massively vulnerable as will every *nix box administered by someone following this guy for advice. Thankfully, the OpenBSD adherent where I work would laugh themselves into an asthmatic attack if they read his statement.

    --
    If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
  107. If you want the technical answer..... by Anonymous Coward · · Score: 1, Informative

    As a professional hacker, I'll give you the 25 cent reasons.

    1) Only root can forge packets. I have to get root to steal IP addresses and adjust routing tables most of the time.

    2) Only root can install kernel modules. Kernel modules are a great way to hide from prying eyes.

    3) Root can debug any process. If I can debug another process, my program can spread to that process giving me complete control of it.

    4) Getting root is noisy. If actually wants that box and not just to use it as a relay of some sort, he generally will need root to take the credit card numbers, corporate bid info, etc. Local exploits are among the noisiest and are the most likely to get caught by a good syscall IDS.

    5) Only root has access to most of the log files. If I'm not perfect when I take over a box, I've got to adjust the logs. I have to be root in most cases.

    6) Only root has raw disk access. If I want to hide all my stuff, the best way is to directly modify the filesystem. You need raw disk access for that.

    etc, etc, etc

  108. "Local escalation" fallacies. by markv242 · · Score: 1
    I have a text document that is owned by me, sitting in my home directory. I run a program that has a backdoor in it. This backdoor triggers the command "rm -rf /".

    Yes, 99% of my machine will not be affected. But guess what? I'm still losing that text document, which, to me is a hell of a lot more important than losing /bin/ls (which I can just reinstall). I think this is where Michael was heading, before he went down that road of overgeneralization. There's no safe way for me to protect that document from a malicious program.

    Also, you speak about keeping important files listed under a different user-- here's the issue: you shouldn't need to do that. If I have to authenticate every time I want to save a gif from Photoshop, that doesn't make for a good user experience at all.

    Lastly, you say: If I so desired, I could limit the login / password for my MySQL account to only allow row INSERTs and SELECTs, but no DELETEs or DROPs.

    But you don't, right? Why not? Because you're willing to take the risk that your MySQL login won't get swiped. So where's the acceptable level of risk? Michael seems to think that Linux has good enough security to make running as root not that big of a deal.

    1. Re:"Local escalation" fallacies. by roju · · Score: 1
      Yes, 99% of my machine will not be affected. But guess what? I'm still losing that text document, which, to me is a hell of a lot more important than losing /bin/ls (which I can just reinstall).

      If you aren't running as root, when you see:
      rm: Permission denied: /dev/hda
      rm: Permission denied: /dev/hdb
      rm: Permission denied: /dev/hdc
      ...
      scrolling up the screen, you can pound ctrl-c and pray that you stop things before any damage is done. And then you can reboot, without worrying that you're missing essential system files.

      OTOH, if you were running as root, you'd HAVE to reinstall, or at least put a lot of effort into recovery.
    2. Re:"Local escalation" fallacies. by DarkHelmet · · Score: 1
      But you don't, right? Why not? Because you're willing to take the risk that your MySQL login won't get swiped. So where's the acceptable level of risk? Michael seems to think that Linux has good enough security to make running as root not that big of a deal.

      It depends entirely on what I'm doing. The general policy for anything I do is to grant it the least amount of privileges necessary to do what I want to do with it. Is something like this really so bad?

      In any case, what's so bad about "locking" files? One of these days I think I'm going to write a sudo program that changes user ownership to root, and grant read access to my groupname.

      --
      /^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$/i
    3. Re:"Local escalation" fallacies. by TheLink · · Score: 1

      If you have a lot of harddisk space then perhaps a filesystem that supports versioning/snapshots will be useful.

      So even if you delete something it's not totally gone.

      VMS did that long ago. Novell used to do that years ago. I believe Network Appliance has something similar.

      --
  109. Troll by newbie65536 · · Score: 1

    YHBT YHL HAND

    --
    Profanity is the language all programmers know best.
  110. Sorry - a couple of examples by AngryElmo · · Score: 1

    http://www.securityfocus.com/columnists/144/commen t/18387#MSG http://business.newsforge.com/comments.pl?sid=2415 9&op=&threshold=0&commentsort=0&mode=thread&tid=11 1&tid=2&tid=3&tid=31&pid=15234#15239

  111. rm -rf / by nurb432 · · Score: 0, Redundant

    oops i meant ./bladir

    If i wasnt root, i wouldnt have hurt a thing.. Instead, as being root i just killed my entire machine due to hitting enter by accident..

    There are of course MANY other examples of why root is bad, but this is an easy example.. And with how easy it is to typo...

    --
    ---- Booth was a patriot ----
  112. RTFA by Lord+Bitman · · Score: 1

    on any multi-user environment, you're already running multiple users. In a single-user environment, you might as well run as root. Anyone (you, or something pretending to be you) can make a file called "ls" in /tmp. If this you is a user-account, it can completely trash everything you care about. If this you is actually root, it can not only trash everything you care about, but also trash a bunch of things you care nothing about! And it can even open a back door in two fewer lines of C than if it didnt have root, using methods which people who attack desktop systems dont care about too much anymore because it's not as efficient to control!

    --
    -- 'The' Lord and Master Bitman On High, Master Of All
  113. What Happened In Vegas by srobert · · Score: 1

    "Honey, what's this journal entry about Las Vegas?" :-)

  114. Why use rm at all? by CommandoB · · Score: 1

    This seems to be drifting off topic, but while we're talking about ways to avoid the catastrophe of rm -rf / just consider replacing rm altogether.

    I just threw together a bash script to keep an indexed "trash" directory of everything I delete. Instead of typing

    $rm -rf ./this ./or ./that

    I just run the script

    $trash ./this ./or ./that

    and it gets moved into "${TRASH}/`date +%Y.%m.%d_%H.%M.%S`/". I keep an index of these files and run another script:

    $restore this

    to undelete it. A couple more scripts let me check out the size/contents of the trash directory, empty the trash, etc.

    It's been a few years since I instituted this command-line recycle bin, and I have never accidentally deleted anything since. I also have a really hard time believing I'm the only one who does this.

    --
    Not that I post on slashdot or anything.
    1. Re:Why use rm at all? by skiman1979 · · Score: 1

      can't you then alias rm to execute the trash script? That way if you forget to use trash and use rm instead (or if another user does) your trash gets used anyway. :)

      --
      Having a smoking section in a public restaurant is like having a peeing section in a public swimming pool.
    2. Re:Why use rm at all? by CommandoB · · Score: 1

      I thought about that, but

      - there are times when I really do want to use rm
      - I want to avoid recursive calls to trash :-)
      - trash isn't really ideal when your trashcan resides in a different filesystem...

      I did alias 't' to 'trash'. It sure is easier to type 't' instead of fretting and then typing 'rm' (and then maybe fretting again).

      --
      Not that I post on slashdot or anything.
  115. Workaround? by SuperKendall · · Score: 2, Insightful

    rm -Rf / as nonroot will make you give a sigh of relief.

    That sounds like a workaround to make up for a design flaw in the command-line interface to me.


    No, it illustrates that some portions of the computers storage space have need to be protected. And that sometimes users do need to alter them, but not generally.

    It's just as easy to take a big chunk of the Windows directory and start trashing stuff with the GUI as well. If you give users the means to manipulate persistant storage (which you do need to do because as the article says, data is king) then you also have to have some way to gently steer them away from utter disaster. How you not heard tales of users trying to free up space on an HD deleting some crucial part of Windows? I have!

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
  116. 65 thousand UIDs, and he's using only two? by Morgaine · · Score: 1

    The reason why he's able to even phrase some sort of odd point about root is that he hasn't *got* Unix yet. Given 65k possible UIDs on his private box, he chooses to use only one for his "non-root" usage, and then claim that all his personal eggs are in one basket.

    Doh! Of course they are, if you put them all in a single basket then they're all at risk of being broken together.

    But if that were how Unix were meant to be run then we'd only need two UIDs, one for "root" and one for "user". Well fortunately Unix presuposes that we're less dumb than that.

    If he got off his ass and put different datasets under different users and shared the lot through read-only permissions and separate write spaces, then he wouldn't be making such silly comments about root not being more risky.

    Jeez.

    --
    "The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
    1. Re:65 thousand UIDs, and he's using only two? by markv242 · · Score: 1
      Please enter user203's password to save "new.gif".

      Please enter user34023's password to read "Saved Email 2".

      Please enter user1180's password to send "Letter to Parents to buy me a Real OS".

      Yeah, sign me up for that.

    2. Re:65 thousand UIDs, and he's using only two? by Anonymous Coward · · Score: 0

      Separation of data into different user spaces saves you from the vast majority of possible mishaps even with empty passwords set, if you're the lazy sort.

    3. Re:65 thousand UIDs, and he's using only two? by Anonymous Coward · · Score: 0

      It does make sense to run things you don't quite trust as a junk user, at least.

      Also, I've just discovered chattr +i. It seems like we can also chattr +i (depending on the filesystem) a directory and subdirectories in which we store important stuff. Then "root" can't even blast the files away.

    4. Re:65 thousand UIDs, and he's using only two? by TheLink · · Score: 1

      sudo?

      --
  117. You still haven't given any good examples. by Anonymous Coward · · Score: 0

    I tend to agree with Robertson in this case.

    See, the problem is that if malicious hacker can convince a user to execute a program as a normal user, he/she might also be able to convince a person to su to root and execute/install the program. Root isn't the be all/end all of solutions that many unix users claim. Neither are antivirus apps. The engineering of the hack isn't usually a technical one. It's a social one.

    Believe me, I wish this wasn't the case. One day, I will experience bad judgement and be susceptible. For the MS bashers out there, Windows 2K and XP has admin privileges... has it helped?

    1. Re:You still haven't given any good examples. by Hank+Reardon · · Score: 1

      For the MS bashers out there, Windows 2K and XP has admin privileges... has it helped?

      Considering that under XP the first "user" created on the system is the admin user, and that most people use that account for their day-to-day activities, I think this makes a great case for exactly why day-to-day operations should not be run on an administrative account. Unless, of course, said activities are administrative in nature.

      The vast majority of Win2K machines out there are in business use and are normally comprimised in different ways; SQL-Slammer, unplugged NetBIOS holes, etc.

      On my kids laptop, I switched their access privileges from "Power User" to "User" and all spyware and virus infections disappeared. If a program they want to use requires Admin privileges to run, they don't get access to it or they get it hosted in an RDP connection to another box.

      --
      There's so little difference between politics and jihad lately...
  118. WTF do people even use Linspire?! by FluffyArmada · · Score: 0

    I don't know why people even use linspire.
    First of all, it offers NO choices for users.
    (eg. KDE and nothing else)
    Their dumb little Click `N Run or whatever "technology" isn't even fricking technology, its the results of some 14 year olds who think they know how to "m4x0r t3h l33t c0d3". And they make you pay like a dollar to download an OPEN SOURCE program. And the money they make off that? It doesn't go to the developers for the open source application. IT GOES RIGHT BACK TO LINSPIRE. There is NO credit where credit is due.
    ( @ 14 year olds : learn C; Its more fun )

    Another thing, their using a version of KDE thats a fricking year old! ( last time I checked )
    And they somehow manage to f*ck the Keramic theme up in a terrible terrible way that simple wasn't deserved.

    And now this moron is griping about other people griping about his idea to let users just be root?!
    For f*cks sake! Thats just assinine! Thats like sitting down and saying.. Hmm.. Wouldn't it be fun if I hit the wrong button and deleted all my most important files while playing with this pretty box with letters and numbers inside it that says BASH.

    I dunno... They may have improved since last time I used Linspire ( back when it was called lindows ( oh shit I'm gonna get sued )) but, it would need to improve A LOT.
    The one time when I used Linspire, I removed it and installed SuSE over it. ( I use Fedora Core now )..

    --
    If con is the opposite of pro. Then isn't congress the opposite of progress?
    1. Re:WTF do people even use Linspire?! by fishbowl · · Score: 1

      > I don't know why people even use linspire.

      I cannot explain the phenomenon that I have witnessed, but for many people, Linspire appears to be the first and last contact they have with Linux.

      I don't understand it, but I've seen it with my eyes.

      --
      -fb Everything not expressly forbidden is now mandatory.
    2. Re:WTF do people even use Linspire?! by pvdl · · Score: 1

      The Linspire distro is Linux's first and last great hope of actually making some inroads on the Window's desktop market. Each person has to make up their mind- do I want Linux to be a mainstream technology, or do I want it to be the exclusive preserve of a tiny geek-clique? If you want linux to be a mainstream technology, you need to start supporting Linspire.

    3. Re:WTF do people even use Linspire?! by fishbowl · · Score: 1

      "Each person has to make up their mind- do I want Linux to be a mainstream technology, or do I want it to be the exclusive preserve of a tiny geek-clique?"

      There's a mistaken reality assumption here. You seem to be holding on to some notion that there can be any control of Linux other than what the GPL ensures.

      The choice you suggest can no longer be made! The Genie is Out of the Bottle!

      I was in my hometown taking care of some business at my family's farm. One afternoon our fuel truck driver started telling me all about Linux. Obviously he understood that I was something of a computer geek (as was he, just a different part of the spectrum), and that I'd understand.

      I didn't let on to him just how much I understood, because I wanted to see how far he'd gotten. (In case it's not obvious, I've been involved with linux since 0.99, long enough that I've stopped counting the years.)

      Turns out, this total country bumpkin, completely uneducated guy (I'm not really being unkind here, he'd describe himself this way), has fully embraced linux as something of a salvation for the common man. It allows him to make use of the computers that get kicked down as city people throw out their old stuff, and he sets them up to browse the web and do email, and as a result, people in my hometown are *wired*. And *I* didn't have to do anything to make it happen!

      There's even what I'd call a grassroots initiative to have shared broadband. Thanks to our fuel truck driver's efforts, folks are connected and aware of Linux, and are even savvy enough to be forgiving of, and avoid, its limitations, like the tricky details of modem support and that you are lucky if a certain wireless card works.

      I doubt this experience is unique. In fact, I imagine that similar things are happening all over the place. The things that city people (like me) take for granted (like disposable income) aren't always taken for granted at all by country people.
      I have a good perspective on both ways of life, since I grew up on a farm, but I live in the city and have an engineering job (and firmly entrenched in the city life, believe me), and I'm also still involved in the business of operating my family's farm.

      But I never expected someone from that end to come to me, to try to tell ME all about how great Linux is. He even burned a linspire disc for me, which I must admit, was something I'd never seen before.

      And now I feel a little guilty, because I never even took that disc out of my bag, much less booted it to see for myself.

      --
      -fb Everything not expressly forbidden is now mandatory.
    4. Re:WTF do people even use Linspire?! by Bicol+Willem · · Score: 1

      Fire it up!

      You don't know what you're missing.

      Its great to see that this disk came from a "unexpected source". It proves that the Linspire policy works. Linux should be easy, powerful, enjoyable, affordable, like ... Linspire!

      --
      MY OS? ... Linspire!
    5. Re:WTF do people even use Linspire?! by fishbowl · · Score: 1


      >Its great to see that this disk came from a
      >"unexpected source".

      No kidding. A truck driver that works for my family, who literally goes by the name "Bubba."

      --
      -fb Everything not expressly forbidden is now mandatory.
  119. *scary* by MoobY · · Score: 1

    that's like running windows ...

    --
    --- Sigmentation Fault - Comments Dumped
  120. Um, what? by OwenMarshall · · Score: 1

    Any exploitable program you run as another user will still need a local escilation exploit in order to do anything harmful.

    That's fine, but he has a point. How much actual real-world good does that do? It does plenty of theoretical good, but so does making the speed limit 10 MPH. By far the better solution is to make sure that the system is safe from remote attacks.


    By far the better solution for safe sex is to get rid of all STD's.

    Seriously. Answer me this -- do you administer servers?

    I run *all* my daemons in chroot jails as non-root users. Why? If someone hacks in through an exploit in Apache, they have compromised a small subset of my system. I notice and react quickly, and they don't actually do any damage. But if I run as root, and someone compromises Apache, my system is not under my control anymore. At least, without a lot of hard work.

    Any program run as root/suid root can cause a hole, no matter how small or trivial the program is.

    So, As Seen On TV, you now have a new project.

    Verify that the code used in a Linux distro on the desktop is secure from all vulnerabilities. I would start with the kernel, then move to the X server and the window manager, and then the applications.

    See you in thirty years!

  121. "Dad!...... by srobert · · Score: 1

    "...Jimmy destroyed all my files for my book report! and it's due tomorrow!"

  122. Standards compliance by Lifewish · · Score: 4, Insightful

    Since a decent majority of open source developers actually give, to use your phrase, a flying fuck about standards, it's generally not in their best interests to promote use of something that isn't a standard, will never be a standard and would be completely undesirable as a standard. Additionally, if it can't be ported across architectures then including it would do a hell of a lot of damage to firefox's geek cred, and hence developer base.

    Having said that, I think a plugin that allowed you to use activeX is a cool idea. I just don't think that tying the browser down to any one platform is a great idea. If you're particularly keen to produce an ActiveX version, go fork the codebase.

    --
    For the love of God, please learn to spell "ridiculous"!!!
    1. Re:Standards compliance by dannannan · · Score: 1

      Standards and "cool features" are sometimes like chicken and egg. If open source developers respond with "fork off" to every non-standard idea, then open source projects will always be playing catch-up with a fragmented codebase.

    2. Re:Standards compliance by lgw · · Score: 2, Insightful

      There's nothing I hate more than a closed-minded standards bigot (and I say this as a member of a standards committee). The purpose of a standard is to capture, not dictate, what the majority of the marketplace is doing. Active-X is quite an important standard - it's the way most users expect things to work.

      You can argue that Active-X is not open, but you can't argue that it's not a standard - in fact, it's the prevaling standard. Adding Active-X support would not tie a browser down to one platform, it would just limit that functionality to one platform. That's what plugins are for.

      Clearly, some people value Geek cred over broader adoptability, which is a shame.

      --
      Socialism: a lie told by totalitarians and believed by fools.
    3. Re:Standards compliance by NutscrapeSucks · · Score: 1

      > something that isn't a standard, will never be a standard and would be completely undesirable as a standard

      I don't really want to get into this argument, but ActiveX is actually a documented standard.

      Furthermore, COM/ActiveX is a core Windows OS technology which makes it a lot more "standard" than Mozilla plugins or XPI (aka Mozilla's clone of ActiveX).

      --
      Whenever I hear the word 'Innovation', I reach for my pistol.
    4. Re:Standards compliance by Anonymous Coward · · Score: 0

      "Since a decent majority of open source developers actually give, to use your phrase, a flying fuck about standards, it's generally not in their best interests to promote use of something that isn't a standard..."

      ActiveX is a standard, and gee golly gosh, lots of AX controls have been developed by 'developers'.

    5. Re:Standards compliance by Anonymous Coward · · Score: 0

      "prevaling standard"

      Uh, yeah. Seems it's so important that millions of people have absolutly no use for it, and would have it thrown to the "gators". The ONLY reason it exists is that it's one of MS's attempts to take over the internet. If you are ok with that, then that is your problem, but don't try to force it on us (that will have no part of it) as a "standard".

    6. Re:Standards compliance by lgw · · Score: 1

      The standard is, sadly, the thing the most people use. It usually does suck. It's sometimes hard to tell what's worse: the design one company runs with and takes most of the market, or the design that came out of the committee, after too much compromise. :)

      Someone adding an Acitve-X module to a browser forces no one to use Active-X. It's the lame web designers who do that.

      --
      Socialism: a lie told by totalitarians and believed by fools.
    7. Re:Standards compliance by Anonymous Coward · · Score: 0

      ActiveX has jack-shit to do with web client plugins: it's an interface standard for writing components. Some of those will be embedded into instances of a web client, some of them will be used to develop a web client, and some still will embed the web client. COM is about 5,000,000 times more of a standard than XPCOM. It's also completely portable (although components aren't necessarily--surprise the same is true for XPCOM) because it's a well-documented binary interface that takes advantage of various aspects of how C++ objects are represented in memory. It's also more sophisticated than XPCOM since it's designed to support multiple threading models, proxies, and out-of-process components.

      Now DCOM is more of a problem, but since XPCOM doesn't even have an analogue to DCOM that's not even remotely important.

    8. Re:Standards compliance by Anonymous Coward · · Score: 0

      Damn that ActiveX with its annoying Flash, QuickTime, embedded support for WMP, Acrobat, and everything else it lets people use inside of their web client! Damn them to Hell! Let's have a half dozen plug-in interfaces instead!

    9. Re:Standards compliance by bankman · · Score: 1
      Active-X is quite an important standard - it's the way most users expect things to work.

      I see your point, but what good is this standard in particular, if every IT news outlet recommends turning it off for security reasons and the technology's provider (MS) turns it off in their browser to get somewhat secure defaults?

      As with regard to users expecting things to work the ActiveX way: How many users actually know that they are running ActiveX? Exactly. They only care about a technology when it's either not working or doing something that it shouldn't. And then they only want to know how to disable or get rid of it entirely.

      Clearly, some people value Geek cred over broader adoptability, which is a shame.

      And no, it's not about geek cred, but rather marketing. Many flaws in IE are ActiveX related, which is why it gets bad press and furthermore it's why people migrate towards Firefox. If developers were to implement this technology into Firefox, the browser would be associated with the bad press ActiveX is getting.

      --
      I feel so sig.
    10. Re:Standards compliance by Anonymous Coward · · Score: 0

      I agree, anyone who wants ActiveX in Mozilla: fork off!

    11. Re:Standards compliance by Anonymous Coward · · Score: 0

      > don't try to force it on us [...] as a "standard".

      Force it on you? Do you feel forced because COBOL77 is a standard language? Not a currently-used standard, but it is one nonetheless. Calling something a "standard" doesn't mean everyone else is forced to use it.

      That said, MS needs to burn in hell & take ActiveX with it, but that has nothing to do with standards.

    12. Re:Standards compliance by hesiod · · Score: 1

      > Damn that ActiveX with its annoying [list of things not requiring ActiveX]

      Damn that Mozilla, still being able to use all the things you listed without ActiveX. None of that stuff relies on AX. (Except maybe "embedded" WMP, I embed the files, not the player) I use Mozilla, but can still put all that crap on the Intranet site I've made.

      Funny how that (just) works...

    13. Re:Standards compliance by lgw · · Score: 1

      You bring up a good point about the security flaws. I think the average home user is still oblivious, and *wants* weatherbug to work, but for the business desktop this is becoming a big deal.

      Maybe we'll get lucky and Active-X will fade.

      --
      Socialism: a lie told by totalitarians and believed by fools.
  123. Small problem by Anonymous Coward · · Score: 0

    Microsoft can and does get away with being that arrogant. In the face of callous indifference to whether their software works and the number of heart attacks that problems with it cause, one might as well be arrogant back.

    This isn't the middle east peace process. Making a concession to Microsoft by including their vulnerable ActiveX tech does not mean they will make some concession by fixing X number of bugs in return. Once you have ActiveX in Firefox, Firefox isn't safe anymore and you might as well run IE.

    1. Re:Small problem by NanoGator · · Score: 1

      "Once you have ActiveX in Firefox, Firefox isn't safe anymore and you might as well run IE."

      The answer to that is simple: Don't enable it by default. Warn the hell out of the user when they turn it on. Only allow it to be on with certain domains, I don't care.

      I have an idea: Instead of shaking your head "no no no" in order to avoid supporting *gasp* a Microsoft feature, try working on how it COULD be implemented so that you can have a broader audience with FireFox or OO. Somehow I doubt serious thought (at least by the Slashdotters that have replied so far) has been put into making this feasible. Afterall, we're quite comfy where we are right now.

      --
      "Derp de derp."
    2. Re:Small problem by Anonymous Coward · · Score: 0

      Sorry, but you chose to develop for a proprietary extension that, at the time of development, you knew didn't work in any browser other than IE.

      That basically ties you into a firm belief that IE will remain a) secure and b) supported as a platform.

      Obviously the wisdom of that decision is now questionable, and any sensible person would be lobbying internally to say "Look, IE has known security issues - we'll work with it for now but any future developments really need to be done on open platforms with multiple implementations so that we can change vendors if things get too bad". Once you phase out the old systems your ActiveX worries go away. The bigger the worries, the sooner that should be.

      Sure, in the meantime you're stuck with IE. But since you were stuck with IE when you developed the stuff, I can't see that you're in any place to complain that the proprietary features you chose to use are not supported by other vendors. If you choose to develop IE-only code it's hardly fair to complain when other browsers don't support your site. Particularly open-source ones, which you're not even paying for. If you don't like having an insecure browser, stop lobbying other vendors to add the features that cause most of the problems in the first place!

    3. Re:Small problem by Anonymous Coward · · Score: 0

      It doesn't surprise me to hear somebody arguing FOR active x that has "Gator" as a part of their name :P.

    4. Re:Small problem by Anonymous Coward · · Score: 0

      Heh that was actually kinda funny. :)

    5. Re:Small problem by InfraRED · · Score: 1

      actually, secure activex support might be pssible through some kind of virtualization layer which then could properly confine the control

      --
      metamoderate!
  124. I like them by Eternally+optimistic · · Score: 1

    Linspire has a nice building, I used to walk past it every day. Bad traffic though in the afternoon.

    --
    What keeps me going is my inertia.
  125. It goes to more than multiuser by Anonymous Coward · · Score: 3, Insightful

    Even on a single-user system, there is a damn good reason to run non-root: otherwise, if an attack makes its way in, you'll have no way to know about it. That's because every utility you could use to verify the integrity of the binaries and libraries and kernel you use can be altered by root.

    Not everyone takes proper advantage of the root privelege separation. Popping up dialog boxes asking you to enter your root password, for example, was a terrible design decision on the part of most distros. And sudo is almost always misused. But properly done, root privsep is your only way you can reasonably know you're running a clean system.

    1. Re:It goes to more than multiuser by jandrese · · Score: 1

      Those root boxes are a terrible design, they're almost as bad as the alternatives. Basically, you're kinda stuck if you want to do administration with windowing programs like that, since the GTK libraries don't even like to start if you're running as root (so sudo admintool tends to break), and users won't stand for logging out and logging back in as root to change their IP address or some other trivial administrative task. Of course you can do everything from the command line (it's safer that way), but that goes counter to the whole neophyte friendly thing that most distros strive for (or in some cases, were designed for). GTK or Qt are too complex to be considered secure for ANY root programs anyway, but it's not like there are many other options.

      --

      I read the internet for the articles.
    2. Re:It goes to more than multiuser by jbolden · · Score: 1

      You can actually handle that pretty easily. You can add an entry to the sudoers file which allows you to run a particular program as root with no password needed.
      ray rushmore = NOPASSWD: /usr/X11R6/bin/GTKconfig

      So you can set up a configuration program which can make the trivial changes....

      I still personally like to be asked for a password but it rather easy to handle this in Unix.

  126. How I learned not to run as root by Anonymous Coward · · Score: 2, Interesting

    I hope I can remember the details of this correctly. Here goes. Some time ago (maybe 5 years ago) I was running linux on a ppc box. I wanted to play a .au file. The sound device was something like /dev/scd All I needed to do was
    cat soundfile.au > /dev/scd

    I typed
    cat soundfile.au > /dev/sda

    Whoops. Yes, there is a reason not to run as root. I admit the mistake was dumb but if I wasn't root I would have been protected from myself.

    1. Re:How I learned not to run as root by Al+Dimond · · Score: 1

      Would you have had permission to write to /dev/scd as non-root anyway? That probably varies from system to system, but you might have had to become root to do that anyway.

    2. Re:How I learned not to run as root by Anonymous Coward · · Score: 0

      No, that mistake wasn't dumb, that user interface is dumb, appalling even.

      (This is not a GUI vs CLI comment)

  127. No good rebuttals yet... by downwa · · Score: 2, Insightful

    I haven't seen a rebuttal that gives a good solution yet. Michael Robertson is right about data being the most important, but seems to think that having users run as non-root means that changing the date will require a password (which is a "no no" for ease of use).

    This problem can easily be solved by making all non-critical configuration tasks (e.g. setting the date) run using sudo behind the scenes. Changing the date as a non-root user shouldn't require a root password (unless you're real paranoid about bugs in your date setting code allowing crackers to exploit date-triggered viruses, or something...)

    As for requiring a password, the only real solution is biometric, and that will be a while in coming. But most grannies aren't going to be changing hardware settings unless they are comfortable entering passwords. And if they aren't, they shouldn't be changing hardware settings anyway...

    --
    Life's a lot like money-- you spend it, then it's gone. Spend wisely.
    1. Re:No good rebuttals yet... by nagora · · Score: 1
      I haven't seen a rebuttal that gives a good solution yet.

      Err, there's nothing to rebutt. There's no difficulty inherent in running as non-root and thousands of obvious scenarios where it's a bad idea to run as root, from accidental data and program loss to virus spreading. What is the point of running as root? There's no good solution to the problems it causes.

      Even in a house with a single machine, there are often multiple users who actually WANT to have their accounts separate.

      Changing the date as a non-root user shouldn't require a root password

      Why? What's wrong with preventing someone with a confused idea of what the date command does, or a kid playing with the clock on KDE from actually changing the system clock? What is the disadvantage of preventing someone screwing with the system?

      Just because someone's rich and uses Linux doesn't mean can't be a moron too, you know.

      TWW

      --
      "Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
    2. Re:No good rebuttals yet... by downwa · · Score: 1

      I'm assuming you actually want to attract dumb users (as Linspire is trying to do). In that case, requiring a password to change the date is going to turn them off. Especially if it is not the one they normally use. If any password must be given, at least sudo should be used, so the password can be the user's own password.

      As for the kid changing the clock, this points to the need for greater privilege granularity. I believe most users would want newly created accounts to, by default, allow dates to be changed without a password. But, as you point out, a parent might want accounts with even more restricted privileges. For that matter, a parent may want a kid's account to not be able to download and run software, even as their own user. Or not be able to listen on a port (not even a non-privileged port). Thus, the need for greater granularity.

      Another post mentioned that SELinux is working on this. Hopefully it comes to general Linux soon.

      --
      Life's a lot like money-- you spend it, then it's gone. Spend wisely.
    3. Re:No good rebuttals yet... by Todd+Knarr · · Score: 1

      Actually, changing the date should require an explicit password entry. Think about it. On a Unix system, most often the system date's maintained by the NTP daemon or something similar. If everything's working right, the date's always correct automatically. If you're trying to change it, either you're doing something very very strange or you have a problem with the system daemons. Either way, you should probably be thinking about why you're doing what you're doing before you do it. In all likelihood you should be fixing the underlying problem instead of trying to patch around a symptom. And if you need an incorrect date for some reason, you should be very very careful about why you want it since system-wide date changes can have lots of nasty consequences you weren't thinking about (viruses and trojans here are the least of your worries).

      When someone mentions "ease of use", I feel a need to point out that it's probably a bad idea to make it easier for a burglar to break into your home or an arsonist to burn it down. "Ease of use" should only be considered after "having a system to use" has been considered.

    4. Re:No good rebuttals yet... by downwa · · Score: 1

      You're still thinking in server or workstation terms, always on, always connected, etc. Linspire is a *Desktop* system.

      Yes, I know messing with the date can screw up compilation trees (timestamps dependencies), daemons, and many other things. But, desktop users are going to want to do it. Put them behind a firewall, then let them do what they want with their system.

      --
      Life's a lot like money-- you spend it, then it's gone. Spend wisely.
    5. Re:No good rebuttals yet... by Todd+Knarr · · Score: 1

      No, I'm thinking in desktop terms. Always on doesn't matter, the standard logic in Unix systems is to read the hardware clock early in the boot process, then force a sync to the time servers later just before (or as part of) starting the time-sync daemon. As long as there's a time-server accessible, the standard logic will keep your time correct without you having to set it manually whether you leave the machine on all the time or shut it down every day when you're done. Though frankly on a Unix system there's no reason to shut it down. Just enable power-saving features and let the hardware power down when you're not using it. My experience is that that nets you 90-95% of the power savings you'd get from turning the power completely off and reduces the wear and tear on the hardware by an order of magnitude.

      Just because a user wants to do something doesn't mean they should be allowed to do it. Look at the state of the Windows world, most of it comes from the insistence on letting users do whatever they want regardless of whether it's a good idea or not. We don't let people drive cars without knowing how to drive first, regardless of how much they might want to drive or how important it might be for them to be allowed to drive. I fail to see why the logic for computers should be different.

    6. Re:No good rebuttals yet... by downwa · · Score: 1

      I understand the usual time initialization process, having closely studied init scripts. Nonetheless, this depends on a time server being available (DSL, Cable, etc.) Many home users still don't have these even if they are available in their area.

      I concur on the wear and tear of shutdowns and restarts-- but that doesn't mean you're going to get granny to stop doing it. And in general, your comparison to driving can only be taken so far-- as far as it pertains to hurting others. Granny's computer won't be much of a threat as long as it is firewalled off and not running Outlook.

      And, even the worst Windows zombies don't do as much damage as a single intoxicated driver can do. Let's keep some perspective.

      Back to my main point. If we simply tear down Michael Robertson's method without examining why he's doing it, we won't convince him to use a better method. It would be better to make some concessions and get Linspire to run most things as a non-root user, than to have a bunch of grannies running as root, becoming zombies, and dragging the reputation of Linux in the mud.

      --
      Life's a lot like money-- you spend it, then it's gone. Spend wisely.
    7. Re:No good rebuttals yet... by Anonymous Coward · · Score: 0

      You don't *GET* 'greater privilage granularity' when users run as root!

    8. Re:No good rebuttals yet... by Anonymous Coward · · Score: 0

      Actually, I'd argue that the Windows zombies do *more* damage, it's just that the intoxicated driver's damage is more *concentrated* in one place.

  128. Root won't help here. by Anonymous Coward · · Score: 0

    I used to work on a p2p video conferencing system. These types of software are rarely, if at all, bundled with standard distributions. Further, most video conferencing software requires root access for installation, some even require that you're root in order to execute them. I'll be the first to admit that I've installed hardware drivers without knowing the source of all of the parties who developed it. At best, I'll have a certain degree of trust with the hardware vendor who distributes it, but who knows if even they know the true source of the code.

  129. Re:uh... by NichG · · Score: 1

    I'm curious though: it takes a fair amount of time to actually wipe out say, 10gb of data. So I wonder if you rm -rf / and then immediately hit ctrl-c just how much you lose (or if you even really lose anything since rm probably just removes the file entry, as opposed to overwriting the data with zeros or noise...)

  130. Ubuntu got this one right. by OwenMarshall · · Score: 1

    The root account is disabled automatically.

    You read that right. You want to do something that requires root access?

    sudo.

    Anything that requires root prompts you with a dialog box, explaining in mostly clear language.

    Running as root is like running with scissors.

    1. Re:Ubuntu got this one right. by fishbowl · · Score: 1


      >You read that right. You want to do something that
      >requires root access?
      >
      >sudo.

      Okay,

      $ sudo bash -login
      # su - owen
      $

      --
      -fb Everything not expressly forbidden is now mandatory.
  131. rm * from / by deian · · Score: 1

    I used to always work under root, until one night. I was extremely tired and no longer really knew what I was doing. I only remeber typing in "rm *" to remove all the content from the directory, however I was not in the folder I wanted to delete the files from, but in "/". Ofcource, except for a few files none of the important system data & folders were removed - i didnt type "rm -rf *".
    It was an extremely careless and stupid mistake, however I realized that it would be safer to just create a user account and "su" when I need to - so when I'm in another one of those funky moods I will be harmless because I'll never get my funky 19 char root password right :)

  132. not off topic, inspired by the topic. by itzdandy · · Score: 1

    here is MY solution to this.

    we need a stack file system. where the base file system is / and the stacked file systems are difference files for each user.

    for instance.

    /
    /home
    /home/username.diff.img

    notice no folders in home, just difference files.

    login
    mount -o diff_filesystem /home/username.diff.img /
    this mounts a filesystem right on top of the old file system. and gives full read/write access to the / BUT all changes are saved to the diff.img rather than the / filesystem. this would allow unpriviledged users to install programs(if allowed to by making rpm/apt-get/emerge/etc etc suid root or something) and alter many many settings BUT not be able to mess up the base system OR other users.
    i know that this is a generic and incomplete solution as no user should have access to another users files BUT this could be handled in the stack filesystem module forbidding access to other home directories or root definable files and folders.
    so root could modify /etc/difffilesystem.exemptions when logged in as root and adding a line
    /etc/difffilesystem.exemptions -wrx
    would make it so the average user could not wrx that file or
    /home/* -wrx
    /home/$USER.diff.img +wrx
    would deny access to /home and then allow access to that users home only.

    i do this because unix file permissions would not be valid because any changes to the files would be written in the difference file so no changes would be made. of course the exemptions file should have an option

    /path/to/file/or/folder/* =unixfilepermissions
    or something like that to use the base filesystems permissions for those files.

    so the AVERAGE user would have an exemptions file like

    #base config
    /* =unixfilepermissions
    /etc/difffilesystem.exemptions -wrx
    /home/$USER.diff.img +wrx
    #other options.

    while a more priviledged user would get

    ##commented out for priviledged user
    #/* =unixfilepermissions
    /home/* -wrx /etc/difffilesystem.exemptions -wrx
    /home/$USER.diff.img +wrx
    so, the exemptions file would allow some users to install software via traditional install programs like apt-get or rpm without the need to modify those programs OR run as root.

    it would allow you to limit the folders users could modify so they could not re-run daemons with different settings in their userspace after editing the configs.

    and it would allow a side effect of makeing the users home directory a single file with its own filesystem. the / could be ext3 but the users home folder and all modifications could be reiserfs.
    AND the root filesystem would always be intact.

    just some thoughts.

    thanks for your time

    1. Re:not off topic, inspired by the topic. by smash · · Score: 1
      The only *real* reason to require root access (other than dodgy code) is to allow direct access to devices - which is not addressed by your diff file system anyway.

      Everything else you can do with paths and/or symlinks.

      If you're only going to let users install programs for their own use (as your diff file system suggests) - this can already be done, but putting them under their home directory.

      Programs that don't work this way, and require specific installation/library file locations are broken and should be fixed - there's no need to design a new file system.

      Neat idea (diff filesystem), but I just don't see any real need for it. My 2c.

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    2. Re:not off topic, inspired by the topic. by itzdandy · · Score: 1
      Programs that don't work this way, and require specific installation/library file locations are broken and should be fixed - there's no need to design a new file system.
      that was my point actually. in an open system(i believe) such as open source, one must work within the system to integrate and not try to ignore other systems because they are not as efficient as they could be. we just dont have the resources. what we can do is modify some of the more important or at least highly used systems and try to steer development out way.

      also, many programs do not install from an rpm or apt-get or package manager in general into a home directory all that well. this would allow unmodified installs easing the process.

      dont get me wrong, i think that many systems need overhauled.
    3. Re:not off topic, inspired by the topic. by smash · · Score: 1
      Fair enough.

      I think you'll have more success in adding a feature to a package manager than implementing/debugging/gaining support for a new filesystem though :)

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    4. Re:not off topic, inspired by the topic. by fishbowl · · Score: 1

      Interesting idea but it sounds very expensive.

      I've dabbled with things like a read-only / and /usr
      with a tmpfs /tmp and mounted homes (also nfs home), and it's been interesting, but I haven't really stuck to it or standardized it.

      --
      -fb Everything not expressly forbidden is now mandatory.
  133. Oh dear, what's his background, anyway? by porky_pig_jr · · Score: 1

    Is he the guy who started with MP3.com? what does he know about Unix (and Linux) administration? He sounds like not too much.

  134. Whats his root password by bxbaser · · Score: 1

    cause if its safe he should post his login and pass here.

  135. For about 30 years ... by GNUALMAFUERTE · · Score: 1

    The _real_ operating systems out there (read UNIX) has teached us that permissions are one of the more important stuff in systems security. Unix takes this concept to a new point with the "everything is a file" aproach, so using a single set of permissions (file system permissions) you can limit a user to only see certain files, only use certain programs, or only use certain printers, only connect devices into a certain port, etc.
    Even windorze and apple understood that this was needed, and they implemented it (In the case of m$, implemented poorly, in the case of Apple, Stolen).
    And now this guy (Who has proven to be a bloodsuc ker trying to be the new bill gates using the efforts of the Free Software comunity, and not giving back a shit to it (since every piece of crap they make is propietary), they also have got a record on spyware and other shit) Trys to tell us that we were wrong, and that we should just drop permissions all along?, Come on ...

    --
    WTF am I doing replying to an AC at 5 A.M on a Friday night?
  136. then trap rm -Rf / by cheekyboy · · Score: 1

    So all these years later, and NONE of the shells do a trap for rm -Rf /, even the rm command should prevent that or at least ask an interactive question.

    I dont think the guy said, make every service run as root, I think he meant running your X/apps as root.

    But I like the OSX model of asking for the root password, though once I lost it and I actually used a normal user account to 'reset' the root password to what I wanted to - All I did was copy the crypted string of USERX into the string for ROOT in some System Admin Config util there and bingo I got root access easily....

    --
    Liberty freedom are no1, not dicks in suits.
    1. Re:then trap rm -Rf / by Al+Dimond · · Score: 1

      What kind of "System Admin Config util" were you able to get into as non-root?

      If that's really true, that could be a hell of a security leak.

  137. In one-syllable words by Anonymous Coward · · Score: 0

    "I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer. They say 'oh, yeah, it is!', but it really isn't."

    If you run bad stuff it can't fuck up your stack.

    (Hey! I did it!)

  138. If root = Administrator Then by psyburn · · Score: 1

    OS = Windows XP
    Administrator = All_Local_Users
    If History = Multiple_Viral_WebPages Then
    My_Day = Very_Bad
    End If
    End.

    I dare you to mod this redundant mod +0/-1

    --
    This was brought to you buy the Department of Redundancy Department
    1. Re:If root = Administrator Then by some_random_person · · Score: 1

      I dare you to mod this redundant...

      It's only redundant if people are actually listening the first time around.

  139. Good reason: Beta software. by Anonymous Coward · · Score: 0

    This is hearsay, but I believe there were some root exploits in some "older" versions of IRC. There was a time when remote users could easily transfer files to your system byway of IRC. Who knows what happened after the software was downloaded, but suffice it to say that software in pre-development usually don't come with polished security.

  140. Question your best practices! by jhantin · · Score: 2, Interesting

    There have been some very good research projects done on how to build a more secure system, and some of the most amazingly effective ones have been the ones that challenge the basic assumptions of "best practice".

    MIT Kerberos takes the view that no machine on the network can be implicitly trusted; access to network services is controlled by tickets, mediated by a ticket distribution service with which each user and service has a pre-shared key. This works even for systems in which the local operating systems have no internal access control mechanisms whatsoever.

    Capability-based systems essentially throw out the classic security model of users, roles and permissions, replacing them with a system of nonforgeable references by means of a combination of memory protection and cryptographically strong naming.

    Finally, people need to come to terms with the fundamental fact that content-based security schemes are a losing proposition (1, 2). Virus scanners, adware scanners, porn blockers, spam filters, and even national customs departments all face the same problem: they can only inspect what goes by and apply a list of tests to winnow bad items. There is strong economic pressure to find ways to bypass these types of checkpoints, so new tricks are constantly being invented, only to be compensated for by the guardians; thus the guardians are always a step behind.

    --
    ...when you're writing a game...tweak the difficulty of "Easy" to something [your mother] can cope with. -- onion2k
  141. Something I've been curious about... by JWtW · · Score: 1

    ...forgive my ignorance, but what about running Yast Online Update, or apt-get, where you have to be root,but running as a normal user?

    1. Re:Something I've been curious about... by Russellkhan · · Score: 1
      What exactly are you asking? What should one do? For Apt-get, I generally either
      su
      to root or
      sudo apt-get
      I haven't worked with SuSE, but I imagine the process is pretty similar for Yast. I don't think anyone is saying that you shouldn't ever run as root, if that's what you're asking. If it isn't, I can't make out the meaning of your question enough to answer it.

      Perhaps if you'd rephrase it, I or someone else here might be able to answer.
      --
      Information doesn't want to be anthropomorphized anymore.
    2. Re:Something I've been curious about... by JWtW · · Score: 1

      Sorry, what I was asking was--"isn't 'root' vulnerable at this point? I've done apt-get, and YOU(Yast Online Update), but I don't understand the mechanics. YOU is behind a GUI (sort of), and it doesn't let you(me) know who is vulnerable. If someone got lucky, and found an open port while I was updating--would I be comprimised? FWIW--since I've posted, I've gotten some off-the-wall files on my desktop... Maybe 'root' is just one of my problems...

    3. Re:Something I've been curious about... by Russellkhan · · Score: 1

      There is always a risk using a computer online, and the more power the logged in user has, the greater the risk, so yes, it is a risk to be root while doing things online.

      But, it is not quite as dire as you fear it might be. An open port (as in a port not blocked by a firewall) is not quite enough to do any real damage. In fact, the usual meaning of an open port is one where a service is listening and responding to connections. Even then, if a service is running, and responding to connections, a cracker trying to break into your box would still have to overcome the safeguards of the daemon running the service - and then, the cracker would only have whatever permissions that the daemon itself had, unless in overcoming the daemon's security he also overcame the OS's protections (as in certain buffer-overflow attacks), which would require there being some unpatched vulnerability on a service running on your system.

      All this really has pretty much nothing to do with the fact that *you* are using an app that requires root permissions to run. The biggest risk from that that I can think of at the moment is some sort of man in the middle attack, but I would imagine that YOU connects through some sort of secure connection, or at least uses some sort of authentication to verify that it's connecting to the correct server.

      I don't really have any idea what the deal would be with the odd files on your desktop, sorry.

      --
      Information doesn't want to be anthropomorphized anymore.
  142. trust by LordMyren · · Score: 1

    the main point is eventually we will have to trust ourselves as root; whether we're running SELinux or CAS (Code Access Security). security must be managed, not just locked away.

  143. Why computers are complicated by Mr.+Slippery · · Score: 1, Interesting
    They used to say the same thing about elevators.

    An elevator has only six possible states: going up, going down, or stopped, multiplied by doors open or doors closed. While getting into those states may have required skill in old elevators, the complexity was inherently limited.

    Your computer has a whole bunch more potential states of configuration and execution. Just assuming ten programs that may or may not be running at a given time, right there you've got 1,024 states. Then there's the state of each of those programs - say each program is not just running or not, but can be in one of five states (which is not unreasonable - not running, loading, reading, writing, and closing). Now you've got 5^10=9,765,625 possible states for your system to be in. Six orders of magnitude more complex than the elevator. Then assume a few variables of configuration - just ten binary values would take us up to ten billion states. (And that's assuming only ten programs - right now ps -ax | wc says I've got over 100 processes running.)

    It gets worse if you take a finer-grained view of what a state is - the RAM in your system can assume more states than the number of elementary particles in the Universe.

    Of coruse in theory, our operating system partitions that complexity, so you only have to deal with the states of one program at a time. And one way it does that it by separating user privileges.

    --
    Tom Swiss | the infamous tms | my blog
    You cannot wash away blood with blood
    1. Re:Why computers are complicated by stridebird · · Score: 1
      An elevator has only six possible states: going up, going down, or stopped, multiplied by doors open or doors closed.

      More, I would wager...several more. How about "going down too fast" for one?

      And where I live in Brixton: "freshly urinated" too.

  144. He may or may not have a point by NullProg · · Score: 1

    Windows 98 has no concept of root.

    Win98SE without IE and Netbios over tcp/ip is pretty secure. Neither the Wife's or kids machines have ever been compromised. Granted, we run ZoneAlarm on both machines and I don't allow the kids to download/install programs without permission (active network monitoring is in place).

    Still, I wouldn't recommend any user to run an internet connected computer as root. If I were a malicious person I could craft an XPI to own a linux box. It'd be just a matter of waiting for some clueless Luser to click on 'OK'.

    Personal gripe. The mozilla foundation needs to sign certified/sponsored XPI's. If the XPI is not signed, the installer dialog box should be RED and consume 75% of the screen. They could also turn this into a revenue stream if they certify plugins for a small license fee to third parties. My opinion and I digress.

    Enjoy,

    --
    It's just the normal noises in here.
  145. It completly depends on the use by grumbel · · Score: 1

    If you are running in a multiuser environment with possibly untrustworthy or inexperienced users then yep, not having them all have root privileges is a hell of a lot saver. It protects the users privacy, avoids other people deleting each others data be it by intendion or accident or whatever.

    In a home environment with just one user, who also happens to maintain the machine its however a whole differnt thing. All valuable data there is stored in $HOME anyway, so gaining root helps nothing to destroy valuable data, neither helps it with preventing spying of credit card info and such. Last not least switching from the user who maintains the machine to root is also rather trivial for some evil programm, either wait till 'sudo' is unlocked or install a trojan 'su' binary in the path or just listen to the X11 key events, sooner or later everybody will end up typing his password in on a self maintained machine. There are still a few things left like accidently 'dd'ing the harddisk with zeros or such, which are more or less prevented by not running at root, but then if you are dd'ing around you are probally running at root anyway. It might also help in making it impossible to accidently delete other partions like the window one. But there is really not much for which not running as root protects you on a single-user machine, the benefit is far more in that it clearly seperates the users data and the application data, so that programms don't end up storing user data in 'C:/Program Files/SomeApp/' like many do under windows, but in that they are forced to store in $HOME.

    So should one run as non-root? Yep, last not least because a bunch of programms simply refuse to run as root, which would be quite annoying. But one shouldn't really have any illusions that one gains any kind of real advance in security in a single-user environment.

    1. Re:It completly depends on the use by PenGun · · Score: 0

      I've been root for 12 years, on my box who really cares, it's more convenient and I can easily su to anyone with no questions asked. I've screwed up badly once, a long time ago. Keep a CD mounted, it makes rm -f / fail ;), if you mount it at /cdrom.

      When I appear on a networked machine it's to do work and I gotta be root. I don't hang about there so why do I need to be a user ... ever.

      PenGun
      Do What Now ??? ... Standards and Practices !

    2. Re:It completly depends on the use by Anonymous Coward · · Score: 0


      >I've been root for 12 years, on my box who really
      >cares

      Nobody cares, that's what makes it your box.

      But if you were to mention this in a job interview with me, it would pretty much get you blackballed.

  146. Right about the problem, wrong about solution by Prien715 · · Score: 1

    Basically, his point is that users care about their data and running as root vs not offers no such protection. Michael's solution is to throw up his hands and give up such security completely. Why not, instead, embrace the root model to protect user data just as your system critical libraries are protected?

    The ability to do so is already in the system: just create a different user. But I don't want to run Openoffice, etc al, as a different user.

    In large coding projects, CVS/SVN/etc are used to maintain version integrity. Here's a thought: integrate svn or cvs into the GUI. When a user is asked to save a file, it prompts them for a password, they enter and voila (modify the APIs for Xwindows et al so that individual apps don't even need to know). Security. Additionally, since you're using version management, it's now possible to revert to previous version of documents -- hard drive space is so cheap nowadays compared ot file sizes (50KB for a spreadsheet) why not combine both technologies? The key is merely integrating the technologies into the GUI. Not that this is easy, but I see no reason why it's not useful (I have dozens of different versions of most everything lying around, just in case).

    Essentially the entire comment thread has been Robertson bashing which is good -- since his conclusion is ultimately wrong -- but no one seems to want to offer up a solution to the one good point he makes.

    --
    -- Political fascism requires a Fuhrer.
    1. Re:Right about the problem, wrong about solution by fishbowl · · Score: 1


      "Basically, his point is that users care about their data and running as root vs not offers no such protection."

      He seems to only be considering "protection from himself" and "protection from unknown parties", but he's missing the more common modes. Especially, "protection of privacy from casual, or even accidental, exposure to a peer/co-worker/family memeber".

      I can turn people loose on my laptop with a high degree of confidence that they won't read or delete my mail spool, use my ssh keys, etc. I can hand it to them, say "login as wizard pasword not4u" and walk away, without the slightest worry that anything that would rather be private will not remain private.

      What's more, I can put quotas and ulimits on that user, and any number of restrictions (or special privilegss) also. And it's all very easy, and very secure. I'm not talking about the FBI being the casual user, mind you, or anyone who is highly motivated to screw me over. I know for a fact that my laptop is vulnerable to the attack of removing the drive and mounting it elsewhere, or even simply booting a CDRom. But I could fix those holes as well, if I were so motivated.

      What was my point? I don't know, but I don't give half a crap what people do with their computers, as long as they don't come crying to me when they ignored (or didn't seek) my advice, and it bit them.

      Sometimes I think some people WANT to "get hacked" or "get a virus." It makes them feel important and gives them something to complain about. More than one person I know has used it as an excuse to upgrade a perfectly good PC. On one hand, it's sad that spyware can be a sufficient aggravation for someone to simply walk away from a perfectly good, recent, high-spec PC. On the other hand, it's important to realize that sometimes they just needed the excuse to get a new one.

      --
      -fb Everything not expressly forbidden is now mandatory.
  147. To root, or not to root? by Anti-Trend · · Score: 1

    ...that is the question. As somebody who's actually deployed many Linux systems for Joe Sixpack, I can say with a great deal of honesty that when you tell them that they can't break the system as long as they're not in root, they damn well don't log into root unless it's for something important. Most users are afraid of breaking their systems anyway, so if they see a clear cut divide between what's safe and what isn't, most will willingly choose to err on the side of safety. As a result, I've never seen any of those boxes compromised in any way.

    Call me a troll, but I think that if Linspire wants to try and be a "better Windows", they're going to find that you can't beat the original, no matter how crappy it may be. On the other hand, if they strive to be a "more consumer palletable Linux", they should probably consider actually addressing Linux's weak points rather than ignoring its strong ones (like security).

    -AT

    --
    Working in a DevOps shop is like playing in a band made up entirely of keytarists.
  148. Root password to change clock?? by edxwelch · · Score: 1

    I know it's not secure to log in as root, but many users are driven to do this because of how restrictive Linux security is.
    I mean you need to type the root password just to connect your modem, or change the clock time. Is this really necessary on a single user desktop?

    1. Re:Root password to change clock?? by smash · · Score: 1
      No, and a well designed distribution will not require this. Set permissions for users on the modem device, and you're set.

      Most of the "but i need root access!" problems are caused by poor distribution implementation, not inherent problems with the restriction of user accounts.

      See ubuntu for a good demonstration of how to do things correctly.

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    2. Re:Root password to change clock?? by fishbowl · · Score: 1

      "I mean you need to type the root password just to connect your modem"

      You need admin privileges to change the mode of a device node, or to assign suexec on software that uses a device node. Good. Use escalated privileges to fix that. An argument for allowing routine root logins? I think not.

      > or change the clock time

      You need admin privileges to alter the clock time. Something that should be done periodically as an automated task anyway. Good.

      >Is this really necessary on a single user >desktop?

      Is a single user desktop really the norm? If so, do whatever you want. But please don't try to apply the same parameters to anything other than that particular degenerate case. Even home computing is more often than not multi-user.

      --
      -fb Everything not expressly forbidden is now mandatory.
    3. Re:Root password to change clock?? by edxwelch · · Score: 1

      In that case both Mandrake and Fedora Core are badly designed, because they both require root access to connect (bear in mind this is for a USB DSL modem)

  149. Two different security types by mnmn · · Score: 1

    You can either be secure by design and implementation.. (OpenBSD), or you can be secure by piling up difficulties in the way... (Windows XP SP2 + Norton Antivirus etc).

    Not using root is in the second category. Even OpenBSD has all services disabled by default and many Linux distros are enabling iptables to close most ports by default, both of which fall in the second security category. Ideally, the first category would be enough. In real life, you need both the good design and implementation, and the second security layers because nothing can be 100% secure. Even if it was 100% secure, the human error always exists, which is undenyable.

    So in theory root is secure, as long as you're perfect, your passwords are extremely difficult, and your OS something like OpenBSD, or like OpenVMS for the VAX, which very few hackers would even want to learn to hack. But in real life, stay away from root.

    --
    "Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky
  150. Least-Privileged User Accounts on Windows by VGPowerlord · · Score: 2, Informative
    Microsoft's Technet has an article named Using a Least-Privileged User Account (LUA), and they intend to force LUA on people in Longhorn.

    My reaction? It's about time! This will help far more than any "Trusted Computing" initiative will.

    Now before I continue, I'll comment that my workstation/gamestation is a Windows XP SP2 machine. My web services machine is a Debian Linux machine.

    I have two accounts on my XP machine: One Administrator and one Limited User. I use the Limited User Account on a day to day basis for my classwork, Applications, and Games. I use the Administrator account to install new programs and program updates.

    The biggest problem with a LUA policy on a Windows system is... Application manufacturers. Programs tend to be written with the impression that the program directory and HKEY_LOCAL_MACHINE part of the registry is always writable. Unfortunately, this is undoubtably because Windows 9x didn't have the concept of file or registry permissions.

    On XP, by default, Limited Users can only write to their Profile directory on C:, and can only write to the HKEY_CURRENT_USER part of the registry. These are where user specific files and settings belong! The %USERPROFILE% and %APPDATA% environment variables are even set up for them! There's even an %OS% environment variable that tells the installer that this is a Windows NT system (It's set to Windows_NT).

    The most recent offender for ignoring these restrictions, that I've installed, is World of Warcraft. Since it was written in 2004, its installer is aware of accounts and account types, and gave me an error that I needed to install it as an Administrator. That's all well and good, but it still tries to write files to %ProgramFiles%\World of Warcraft\WTF\Account\[USERNAME]\ heirarchy every time it runs. While the game seems to work even if it can't write its files, you also can't save any settings changes.

    --
    GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
  151. linspire dot bomb!!! by mr_z_beeblebrox · · Score: 1

    What kind of knucklehead would let his CEO have root?

  152. Make the logins disappear by Urusai · · Score: 0

    Honestly, the boot process should be an admin account anyway, and can hand it off to another user process. Most home computers can be considered physically secure, and so merely booting (or clicking on your icon) should be enough to get to your account. Windows actually does this right.

    As for how to handle admin type functions under a GUI without incessant PW prompting: give the GUI escalation privileges, but not programmatically. I.E., I click on a Disk Format icon, it runs as admin, but there be no exposed API to do be exploited automatically. The boot process could hand off limited admin privilege to the GUI, but this requires trusting your software. I guess this is too difficult for people to grok...

    1. Re:Make the logins disappear by fremar · · Score: 1

      Which of course implied that an exploitable bug (e.g. buffer overflow) in the GUI apps or libraries can mean instant root compromise. It is really hard to get good security and priviledge separation and at the same tim get userfriendlyness to the level of automatic login without passwords and simple administrative access as well, also without password. It's probably much better to make people get used to providing a password for certain actions. It makes them more aware of security as well.

  153. Any winders admins care to explain by Anonymous Coward · · Score: 0

    just how in the hell am I supposed to be able to add/remove/manage/defrag/update/etc if I'm not admin. I've tried it, on machines at school, the simplest things require admin rights. updating a freaking flash plugin you must be admin.

    I'd be logging off and on every day if I weren't a user in the admin group on my home windoze pc.

    I don't think it's realistic to be non-admin on windows, and I wonder how y'all do it on *nix.

    1. Re:Any winders admins care to explain by merdely · · Score: 1

      You familiar with sudo on *nix? Check out RunAs. Or, right click on an icon or command and choose Run As...

      Much safer. Just takes some getting used to.

      So did sudo.

    2. Re:Any winders admins care to explain by fishbowl · · Score: 1

      sudo is the solution to your problem;
      grant temporary privilege escalation to specific users in specific context, complete with logging and the ability to revoke the privileges, and never any need to share the root password.

      It works in my shop, which is a significant real-world multiplatform environment with billions of dollars of assets and high liability for data integrity, and it will work for you.

      --
      -fb Everything not expressly forbidden is now mandatory.
    3. Re:Any winders admins care to explain by lantastik · · Score: 1

      From an administrative side of things, my users are not allowed to do any of that. They are policied and locked down. All of the operations you describe are controlled through our systems management software. The server has a System level account that it logs in to the machine with to perform these operations. Otherwise, the user has just enough rights to do their job. As for my home PC, of course I run as an admin.

  154. Michael Robertson Needs His Head Re-examined by chadpnet · · Score: 1

    In addition to the 1000's of reasons above, imagine how boldy Michael would defend this insecurity ideology if he walked away from his box and his 8 year old child decided to play around on the computer. The computer with all his financial assets, personal e-mail, business contacts, music, etc. I don't trust anyone with my data, and I'll be damned if I am going to give my kid a leg up in kicking me down.

  155. running as root *is* inherently insecure by smash · · Score: 1
    Yes, yes - its your data that is important.

    However, your applications are also important.

    Run as a user - you can only lose your data.

    Run as root - you can lose your data *AND* your programs.

    Programs can be reinstalled, sure. So can data, by this amazing technology called *backups*.

    Plus, if you run as root, there's the potential that every other user on the machine has *their* data compromised as well, through your own incompetence/mistake/spyware, etc.

    There's also the possibility that, when run as root, changes may be made to your system that you will not notice. Random example - lets say I build a trojan to modify your C library to install a backdoor into any code you *or any user on your system* link against it?

    How about I modify your e-mail client's binary to silently attach copies of personal documents to any email it sends out?

    You may not notice, until its too late. There's FAR more damage that can be done with root access than not - and its not *just* about restoring from backup.

    smash.

    --
    I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  156. I call bull. by SanityInAnarchy · · Score: 1

    Making the root user the default is fine, because absolutely none of the things you mention matter unless there's also user backup by default, or user management by default. (Hint: for desktop apps, there isn't.)

    I don't run as root because I don't want to do anything to my system setup by accident. I like to try out new stuff which isn't likely to nuke my data, but would think nothing of polluting my /usr.

    But I don't think for ONE SECOND that I'm actually more secure that way.

    If we want Unix security to mean anything, we have to do it with software. If you were serious about your ActiveX thing, we'd be running Firefox's gecko and javascript support as user "nobody". We'd have the "secure by default" distros let us play games, but only give the games access to their own config data and savegames.

    This kind of thing is only feasable with massive distro-wide support, and becomes much more feasable with Namesys' ideas on filesystem "views". But it's still not foolproof.

    To summarize: Don't run as root, but don't think that makes you so much more secure/sane. It's a lot less like seat belts and a lot more like adding a bit of tissue paper on top of the seat belt, in case said belt should fail. If you crash and your belt fails, you WILL fly through the window, tissue paper or not.

    --
    Don't thank God, thank a doctor!
  157. He has the answer to his own question by sootman · · Score: 1

    "What's the most important thing on your desktop? It's the data."

    Right. THE data. Not just YOUR data, but EVERYONE ON THE MACHINE's data. If you do something stupid as a regular user, you're only risking your own data. If you do something dumb as root, you can wipe out your whole family's data. Won't somebody think of the children!

    --
    Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
  158. Militant Idiots Unite! by Anonymous Coward · · Score: 0

    That's right, the time has come for all those who are willfully ignorant and cognitively challenged to stand up and tell those know-it-alls, "I'm not going to take it anymore!" If you're a lawyer, CEO, advertising exec, accountant, fifty year old receptionist, psychologist, middle manager, celebrity, or an elementary school teacher, and you're sick of being pushed around by underlings just because they happen to have a clue, it's time for you to join an organization that understands your special status in society and your immunity from conscious, rational thought. Call 1-800-ASSHOLE and ask for a copy of the pamphlet, "The Power of Claiming You Know What the Hell You're Talking About When you Clearly Aren't" and ask about membership in Militant Idiots United. Thank you.

  159. Think about your audience by slapout · · Score: 1

    I think that the target audience for Linspire is the average Windows user who sees no problem running everything as root.

    --
    Coder's Stone: The programming language quick ref for iPad
  160. He's got it backwards by Bootard · · Score: 2, Insightful

    From time to time I help my grandmother and my girlfriend's aunt out with their windows boxes. Both of them are pretty clueless as far as computers go, and I suspect that they could really get into running as a user and not as root. Especially when they were first getting started, and even now, they both have expressed a fear of fucking up their computers. If I/this guy could tell them why it's in their benefit, ie that as long as they don't enter the root password, they can do whatever they want and they never ever will have to worry about their computer breaking, I would imagine both of them would sign up really quickly.

    The solution here isn't in dumbing things down. It's in giving a 1 minute presentation about the *nix security philosophy during the first login. I would think that most people, and ESPECIALLY newbies, would get behind that type of security if this guy would take 30 seconds to explain what it is and why it HELPS them instead of just portraying it as a nusience. In many ways, logging in as a user really does give you a freedom to explore and learn pretty risk free. People know that there are ways to break their computers. Telling them to go ahead and press any button you want to; if it's a button that will do something serious, it will make you enter a password would probably be attractive to most people learning an unfamiliar OS for the first time.

    --
    exceptio probat regulam in casibus non exceptis
  161. Why? by ta+bu+shi+da+yu · · Score: 1

    Corporations have often built functionality around ActiveX. They need to migrate existing functions to the new products they are trialing or integrating into their systems. Not everyone can do a straight cut over you know.

    --
    XML is like violence. If it doesn't solve the problem, use more.
  162. I'll take "non-root" for 500, Alex. by Java+Ape · · Score: 1
    Several previous posters have pointed out that not running as root protects OTHER peoples' data, and ensures that an attack that grants control of an application's environments still requires an escalation of privileges to inflict serious damage.

    I would like to point out that with a bit of effort, unix protections can protect a user's data as well. For example, in web-directories, grant the HTTP user only read access to the files. Database files (at least for Oracle and PostgreSQL) don't need to be accesible to the users at all. Chroot offers further levels of protection.

    For example, I run a web-server with an "apache" user. The user is chrooted, and there is no rm command available to apache, the logs are just pipes, and they point out to a mount point apache can't directly access. The only files available to poor apache are the program files and the files in the web-root. An attacker could work for days, and end up with access to exactly what the web-server was exposing in the first place. Compare that to running apache as root with a few badly-written CGI scripts (third-party, naturally!) to be exploited!

  163. Easier configuration by downwa · · Score: 1
    Eventually SELinux will be applied to desktop apps too, and then Linux will be an extremely secure system. It will be a while, however, because it's very difficult to define exactly what a program is allowed to do; the goal is to restrict anything it doesn't need to do while letting it do anything it does need to do. If you err on one side it's a security hole (though admittedly small compared to today's apps) and if you err on the other side the app will fail when it tries to perform some action.

    Ideally, the system would suspend the application while the user received a descriptive message of what the program was trying to do, at a high level. The administrator could then configure the app by running it with no privileges initially, but clicking "Allow Always" for each allowed operation, exercising each feature of the program that the users need, then capture that configuration and apply it to all users.

    Users could (if allowed by their own account) also grant privileges (up to what their account had) for functions the system admin may have neglected to exercise. However, they would probably soon get annoyed by clicking dialogs and ignore what they said, so admins would in that case do well to make sure the user had few privileges and the app was already preconfigured properly.

    --
    Life's a lot like money-- you spend it, then it's gone. Spend wisely.
  164. AMEN by SanityInAnarchy · · Score: 1

    The "force" option is the worst thing to happen to Windows security, EVER. 99% of the time, you're asking the user to say "Yes, I'm sure I want to do what I just fucking told you to do." The other 1% of the time, you're asking the user to say "Yes, I want to be rooted."

    For instance, what user actually reads and understands the SSH warnings about host keys? Hardly anyone.

    Probably 60% of users just type "Yes" right away. These are the people who would just click "OK" to dialogs that say "Do you want to allow hardcorepr0nspyware.com to install software on your machine?"

    Probably another 20% actually read it, decide that they don't understand what it's saying, and type "No", because they don't like to agree to things they don't know about. These are the kinds of people who say "No" to the browser warnings about sending information over the internet. You know the kind -- they come up whenever you type a search query into Google.

    The other 10% actually go read the docs to understand the message. Actually, 10% is a bit over-optimistic, but we're talking about ssh, after all.

    The same thing has happened to legal licenses. Since we aren't all lawyers and software engineers, we generally click through warnings from software and click/sign licences without a second thought.

    Creative Commons has a good start on the legal front, and the shell is a good start on the software front. When we design any system, we should only force interactivity where it's absolutely needed, and explain it SIMPLY to the user.

    --
    Don't thank God, thank a doctor!
  165. Ubuntu by Anonymous Coward · · Score: 0

    Does this by default.
    *ducks*

  166. Another easy one by crmartin · · Score: 1

    rm -rf / usr/local/src/myjunk

    Then there's the one where you put a file with the same name as a system command in a user directory....

    I'm afraid that some of us who are old enough to rmember why we learned not to run as root are old enough that we're starting to forget things.

  167. Here you go, Robert by melted · · Score: 1

    cd /
    rm -r *

  168. bias by dannannan · · Score: 1

    Those who have experienced such a thing can't post on Slashdot. Let's face it: Slashdot is biased against the dead.

  169. It's so easy... by Fungus+King · · Score: 1

    ...for a younger sibling/child to wander up to your box and type: dd if=/dev/null of=/boot/vmlinuz In seriousness though, he almost makes a good point by saying that anyone compromising the account has access to whatever's in your $HOME anyway... but surely anyone would realise that it doesn't mean you can just give away access to the rest of the system as well?! It's almost like inviting people to start installing rootkits and suchlike...

    1. Re:It's so easy... by Sigma+7 · · Score: 1
      ...for a younger sibling/child to wander up to your box and type: dd if=/dev/null of=/boot/vmlinuz
      Not really - younger childs don't know how to type in that stuff unless they themselves are Unix experts. (And in that case, you probably wouldn't be running root since you wouldn't trust them.)

      I'm not downplaying this, since I know first hand on why root can be dangerous. In debugging one application, I was redirecting standard error to /dev/tty6 - but accidently written to /dev/hda6 instead. This overwrote the first sectors of a 180 GB NTFS partition, which rendered the partition unusable.

      I had to get a sector editor to fix this - I could copy the backup information from another sector, but if the program ran any longer, it would have hosed that backup sector and cause me to lose all information on the partition. (I'm suprised that regular disk-checking utilities can't fix this problem.)

  170. beans by Anonymous Coward · · Score: 0

    The guy's not a techie. QED.

  171. This thread is scary! Here are the reasons... by John+Sokol · · Score: 1

    Root has full authority over the system. Root programs that are exploited also do.

    The idea was/is to make it impossible for a user to become root without going through the proper mechanisms. (su, setuid, group wheel, etc...)

    In the OLD SCHOOL days it was an open challenge to see if a user account (student account) could even crash the system or get root access. Few would succeed. This was then there would only be 1 or maybe 2 computers on the whole school campus. Pre-networking days.

    Something us SVR4 and BSD folks understood, but the many Linux guys totaly miss the point of why we do things a certain way.

    So if a user account or program is comprimized is like on a ship with water tight bulk heads and steel hatches that lock shut, preventing the whole ship from flooding and sinking.

    If a user account become compromized, as root you can still get in and fix it, completely!
    If root is compromized, you probably can even get in, or tell if you've been compromized. This is really bad when you don't have physical access to the BOX. Like when it's 10,000 miles away. ( I have actualy been on boxes, as a hacker as well as root, when my adversary was on at the same time )

    In a partly compromized box, as the hacker that had gained a user account, one can not fix up logs (*to cover ones tracks*), is restricted to IP port services above 1000. (this is why we use 8080). Can not add NEW IP address, send Raw packets (spoofing), or sniff the network.

    They can not alter other users account. So if a web CGI get compromized they can't tag the SITE!!.

    I can go on and on, but I hope you get the point. Root and USER accounts exist for some very good reasons. Layered security.

    But it only works when you make your OS correctly. ( see FreeBSD, or OpenBSD for an example )

    --
    I am always doing that which I can not do, in order that I may learn how to do it. - Pablo Picasso
  172. None of you /.ers listen/read... by ThreeE · · Score: 1

    Root vs. non-root doesn't matter! Data is king (and it's in a non-root account) and everything else can be re-installed. Additionally, if you back things up, your data is safe -- at least from being lost. In the end, your privacy is all that is at risk -- and that resides in a non-root account.

    Silly Linux boys...

    1. Re:None of you /.ers listen/read... by Kaelem · · Score: 4, Interesting

      There is an issue you've not addressed. How about when your data is not the target? (Honestly, most people's data is not worth stealing).

      What if an attacker just wishes to compromise your machine and use it to attack other machines, relay spam, etc? This is a huge problem with Windows.

      --
      "That's some catch, that Catch 22." "It's the best there is."
    2. Re:None of you /.ers listen/read... by ThreeE · · Score: 1

      A valid point, but arguably, none of these things hurt me, and a reinstall fixes the impact on others -- assuming I care.

    3. Re:None of you /.ers listen/read... by smash · · Score: 1
      So you don't mind paying for quicker cpus, more ram, and a bigger internet connection for other people's free use?

      Care to give me some hosting space? :D

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    4. Re:None of you /.ers listen/read... by ThreeE · · Score: 1

      Like I use all of what I have. If it becomes an issue, again, just reinstall.

      Kaboom.

    5. Re:None of you /.ers listen/read... by smash · · Score: 1
      You are the reason we have a spyware/virus epidemic, it seems :)

      What about the hundred/thousand machines your box was used to compromise before you realised what was going on?

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    6. Re:None of you /.ers listen/read... by KillerBob · · Score: 2, Informative

      That's Windows logic.

      In Linux, run as a user. A malicious script destroys your files and "toasts" your system, the only thing you've lost is your user account. As root, you can then destroy the user and user's files, and recreate the user. You've lost maybe 5 minutes of your time, and don't have to reinstall/recompile/reupdate your system.

      If you're running as root, however, the script can access the *entire* system. If it runs amok, you're completely lost, and are out several hours of reformatting, reinstalling, recompiling, and reupdating the system.

      This is especially important if you're running a multi-user system. When there's 3 people using the computer, if one of them gets a malicious script and runs it as root, then the entire system is pooped, and all 3 users are out of luck. When they're running as users, they can't touch each others' files, and as such, they can't screw each other over.

      --
      If you believe everything you read, you'd better not read. - Japanese proverb
    7. Re:None of you /.ers listen/read... by ThreeE · · Score: 1

      Tell me why I care?

    8. Re:None of you /.ers listen/read... by VGPowerlord · · Score: 1
      Tell me why I care?

      Your ISP/upstream provider can terminate your Internet connection because your machine is spewing out viruses, worms, and/or spam.

      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    9. Re:None of you /.ers listen/read... by ThreeE · · Score: 1

      A) They never really do.
      B) ISP providers are a dime a dozen.
      C) Again, fixed by a reinstall.

    10. Re:None of you /.ers listen/read... by fishbowl · · Score: 1


      > Tell me why I care?

      Does it need to be explained to you why you shouldn't litter in the park, or fart on a crowded elevator too? Similar principle.

      --
      -fb Everything not expressly forbidden is now mandatory.
    11. Re:None of you /.ers listen/read... by smash · · Score: 1
      I *have* terminated user accounts for breach of service agreement :)

      smash.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    12. Re:None of you /.ers listen/read... by Anonymous Coward · · Score: 0
      In Linux, run as a user. A malicious script destroys your files and "toasts" your system, the only thing you've lost is your user account
      Yup, and that user account is the only thing that matters. Everything else is on the CD that came with the computer.

      Lindows/Linspire/whatever is designed as a personal computer OS, it's not there to serve multiple users like a mainframe. Everything the user cares about - everything important is in the user account.

      Of course to the slashdot crowd it's your 13 months continuous uptime that's important, not any work you do.
    13. Re:None of you /.ers listen/read... by Anonymous Coward · · Score: 0

      You can connect to port 25 when running as user, right?

      You can accept connections above port 1000 when running as user, right?

      So how exactly does running as user prevent a Linspire box from being a zombie?

    14. Re:None of you /.ers listen/read... by Anonymous Coward · · Score: 0

      And a problem not solved by running as user

    15. Re:None of you /.ers listen/read... by Anonymous Coward · · Score: 0

      Someone sets up a service on your compromised box which serves email kiddie-porn. *Your* box is the one it's coming from. *You* are in posession of said kp, and are distributing it. At best, you're looking at accomplice/aiding and abetting charges.

      Now, tell us again why you don't care...

    16. Re:None of you /.ers listen/read... by Anonymous Coward · · Score: 0

      The average home has *multiple* people living in it. The average household has *one* computer. That means, by definition, the average computer is a *multi-user* system. If you hose your files no problem, after all it's your fault. If you hose your roomate's/spouse's/child's/parent's files, you've just cause someone *else* a world of pain. Your spouse/parent loses an important piece of paperwork, and loses a major client as a result, their job may not be far behind.

  173. I got a great idea by x_codingmonkey_x · · Score: 0
    IMHO I think this is a pretty good idea.

    In order to make it easier on the users (ie no passwords and such when installing new software) why not do the following:

    Instead of running the user as root and allowing them to install any god forsaken program there is why not give them a regular account. Then when the user wants to install something they click on it and the Linspire package managing software connects to a Linspire database where the program is checked against a white list. If the program isn't on the white list tell the user that the program can't be installed because it is most likely a virus or spyware. Then give the user a help number to call if they still really want to install it and let people help them out.

    Now, if the program is OK to install Linspire queries it's user database for your products root password and then fills it in and viola, the program installs.

    A shortcoming is that you might not be connected to the Internet, but, IMO, it's pretty safe to assume that someone buying a new computer is going to have Internet access as well. Just a thought.

  174. YHBT by BasharTeg · · Score: 1

    Obligatory FUD debunk, or just helping a troll.

    You have been trolled.

    And way to blame the fact that YHBT on Bill Gates. That makes a lot of sense.

  175. I learned the hard way by Anonymous Coward · · Score: 0

    Once on a Solaris 2.6 box rented at a high price by my employer for me to port it's product to this OS. I wanted to remove the file sI had used to install something so I just typed "rm -rf *", after a few seconds, I realized that there were line sof text scrolling on the screen saying that it wasn't able to delete /dev/*something* and by the time I had killed the process, the system was no longer usable. I had to spend four or five hours rebuilding the deleted parts after booting for a CD.

    Using root is not a problem in itself, it just makes the problems related to your mistakes far far worse.

  176. Doesn't work for me? by Dlugar · · Score: 1

    I just gave it a shot (FC3) and it won't let me browse /home/bob/ if /home/ is chmod a-x, even if /home/bob/ is chmod a+x.

    Even typing "cd ~" as a normal user gives me "-bash: cd: home: Permission denied".

    Dlugar

    --
    Computer Go: Writing Software to Play the Ancient Game of Go
    1. Re:Doesn't work for me? by Anonymous Coward · · Score: 0

      Doesn't work here either (FreeBSD). I smell a troll.

  177. Let's continue the car analogy... by FooAtWFU · · Score: 1
    Running as root is like riding in your car without a seat belt. But driving without a seatbelt is safe! After all, if you get in a crash, you car's going to be more or less wrecked anyway, so why should we even bother?

    Yes, it's a bit less convenient, and it may be a hassle to put on, and maybe your passanger needs to reach something so they unbuckle themselves for a minute... but it can be the difference between Bad Stuff and Really, Really Bad Stuff.

    --
    The World Wide Web is dying. Soon, we shall have only the Internet.
  178. Am I missing something? by hsoft · · Score: 1

    Or the only way to be sure that my firewall blocks all nasty stuff that I might run locally (Ohh, annakounikova.exe!) is to run as non-root? If I run as root, a malicious script can change the firewall rules, and compromise my data (I mean, let someone else than me have access to it remotely.)

    Yeah, sure, whether I run as root or not, a malicious script can destroy my data, but I have backup, and if my firewall is set correctly, I won't have it stolen.

    Now, if the firewall is on another machine, well yeah, running as root or non root makes no difference (for a single user machine).

    --
    perception is reality
  179. His Point by Anonymous Coward · · Score: 0

    Look, here's the bottom line as far as running as root. If my grandma is using her computer, and she can't run the programs that she wants to, she gets aggrivated, and if it doesn't work she will most likely give it up. And there is one less user. Sure, it is definitely much better to run as a normal user, but if the software DOESN'T WORK, it really is pointless. If the O/S can be intelligent enough to consistently ask for a password in an intelligent manner, then perhaps it would be acceptable to the normal user. However my experience in both linux and in windows that, for many programs it just doesn't ask, and then proceeds to not work. And to top it off, the root/Administrator accounts should have a secure password, and as such, are usually long, and difficult for the average user to remember, and if nothing else they are a complete pain. Sure, I don't run my Linux box as root, but I do run my windows machine as admin, because simply put MANY, MANY programs REQUIRE that in order to work properly. Is it right? No. But that is how it is for the moment. I definitely agree that it needs to change, but you have to realize that the average ex-windows user has NO clue what root is, and doesn't want to be bothered with a password prompt every time he wants to change his background.

  180. BTW, you REALLY don't understand what ActiveX is by Anonymous Coward · · Score: 0

    I dont' think ANYONE understands what ActiveX is, least of all, the programmers at Microsoft.

  181. Re:Okay now... (Mod parent down?) by MrZaius · · Score: 3, Informative

    Correction:
    replace "executable" with "readable"

    chmod a-r /home/, and user bob executing rm -rf /home/ fails to eliminate a bob-owned /home/bob/, as it fails to get a listing of /home/

  182. Not Insightful by Anonymous Coward · · Score: 0

    From the article:"I DO see it's an added pain in the ass when grandma tries to change her wallpaper, and it tells her "you don't have root privileges"
    No, I don't think Grandma should learn about priviliges to change the desktop wallpaper, you elitist grandma-hater!

  183. Can you say "rootkit"? by zik · · Score: 1

    Michael Robertson isn't very "linspiring" as a CEO of a software company. He hasn't really thought this through, has he?

    Let's think about the kinds of data that might exist on a computer:

    The data you already have

    The OS, application programs and configuration

    Other peoples data

    Data you may create in the future

    Running a vulnerable program as a user exposes the data you already have to damage, so only one of the four categories is at risk. But running it as root means that all your system setup and any other programs also cop it. And even better trojans and rootkits can be installed on your system so all your future data is also vulnerable.

    If I'm going to have a disaster I'd prefer it to just screw me over once rather than make my entire life a misery.

  184. rm -Rf by noda132 · · Score: 1

    I dare you to try this. Dare.

    Not that I've tried it recently, but I think I did this a year or so ago (by accident) as non-root. I didn't lose any data whatsoever, since so many error messages began to get printed I hit Ctrl-C before it ever got to /home. So yeah, I heaved a huge sigh of relief.

    Or maybe that was just a weird dream.

  185. Re:only multiuser! by Anonymous Coward · · Score: 0

    the article mentions Grandma wanting to change desktop wallpaper. She is not going to know anymore about the attack if she's root or non-root. I think that is true for the vast majority of desktop users, who are lucky if they even know someone who can use that information. They also never run MySQL, do rm -Rf /, or any of the other things mentioned in the original post. duh!

  186. Four Words... by Sam+Nitzberg · · Score: 1

    "I defy anybody to tell me why is it more secure to not run as root. Nobody really has a good answer.

    PRINCIPLE OF LEAST PRIVILEGE

  187. Root separation is way overrated by alphorn · · Score: 1

    One word: Spyware

    Where is the problem in creating spyware that runs as user? Simply write the binary as dot-file to the home directory and append a call to .bashrc. Same goes for software that turns the computer in a spam-spewing zombie.

    He does have a good point. The separation of root and user on single user machines is greatly overrated. It would only make a minor difference if a root-only software firewall were to block some outgoing connections, e.g. SMTP. But as is, the damage root can do is only slightly more than the damage a user can do - not to mention that root exploits are not uncommon.

  188. I don't get it... by Anonymous Coward · · Score: 0

    The only good reason to run as root:

    1) Your name is root and you're very surprised Linux was able to figure that out without you telling it.

    2) Your s,u, and d keys are broken.

  189. a few anecdotes by Fractal+Dice · · Score: 1

    For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.

    I see about one disaster a month where a user toasts their own files. Over the years I've seen three servers destroyed by an admin doing an rm -rf . in the wrong directory (and heard a number more). I've heard second-hand of a million-dollar typo of someone on an admin server with a whole division's NFS directories mounted root-writable.

    Statistically, I've concluded that any given person, regardless of intellegence, has a slightly more than one-in-a-million chance per day of doing a potentially machine-toasting typo.

  190. So, how easy is it? by SEE · · Score: 1

    Do you know how incredibly easy it is to have a Windows system infected?

    You know, I've been running Windows systems for thirteen years now, and DOS ones for five years before that, without a single virus, worm, trojan, spyware, or other infection. So my answer is, "it's not easy at all if you have a clue."

    Which, apparently, your friends and relatives don't have. They've been running IE and OE, I assume. They haven't had a hardware firewall between themselves and any IP connections, either. And they've run files they've downloaded or borrowed on a disk or found attached to their email without first having Norton or MacAfee look at them.

    So, my question, is it that they've ignored your warnings, or is it that you haven't given them the necessary clue?

    1. Re:So, how easy is it? by Greyfox · · Score: 1
      Oddly my DOS/Windows systems never got infected either. The only time I ever saw an infection was when the boss lent some installation disks out to one of his buddies (And NO he didn't ask me first) and they came back with a virus on 'em. I was pretty clueless back then too, but it was pretty easy to take care of.

      It's usually a matter of ignored warnings. Whenever anyone asks me anything I usually respond "You've been running Windows update and scanning for virusses like I told you to?" The answer's usually "No". Maybe I should pick up a cattle prod for repeat offenders... (I tell 'em to download Zonealarm, too.)

      It's readily apparent that people have no interest in learning even the basic steps they need to take to protect their personal information. Honestly I'm surprised that they change underwear or floss. I'm a proponent of some sort of license requirement to access the Internet (Works for Ham Radio anyway) to enforce at least a baseline level of cluefulness. I suppose that's an unpopular attitude to take, but I think it's the only way the Internet will be able to survive long-term.

      --

      I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

  191. you're wrong by ylikone · · Score: 1
    The main reason to not run as root on a single user system is that you don't want your box to end up a spam portal or hacking portal or whatever. Say you are running some web browser as root. Remote exploit for that particular web browser allows access to your machine, hacker gets in, sets up spam relaying and goes undetected. Hacker does not give a shit about deleting your data... he couldn't care less. Now, consider the same situation if you ran that particular web browser as a normal user. Hacker users remote exploit to gain access to your system, but his scripts fail to run because root privileges are required. Hacker moves on to an easier target.

    The reason Linux will never have the same problem with spyware, trojans and worms is because of most distros setting up the main account as a normal user, not root. So, once linux gets more popular and hackers start writing more tools to hack linux boxes, guess who the much easier target will be... people running Lindows.

    See, your rant was blown out of the water... you are just plain wrong.

    --
    Meh.
  192. Don't use root for normal use? by jgp · · Score: 1

    I can understand the confusion, but the full rule is: Don't use root for normal use drunk.

  193. Example by metamatic · · Score: 1
    For all the talk about it, I don't think I've ever actually known anyone to do the classic accidental rm -Rf / as root.

    My one-time boss did. He typed rm -rf /tmp/somedir, except his thumb accidentally bumped the space bar after the first slash and he didn't notice until after pressing return.

    After that, I got the job of sysadmin...

    --
    GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
  194. Mod parent... by planetoid · · Score: 0

    +5: The Only Post This Story Will Ever Need

    if such a mod exists...

    --
    Slashdot requires you to wait longer between hitting 'reply' and submitting a comment.
  195. okay - have to chime in by suezz · · Score: 1

    I would like to say I never had to be root to change my wallpaper

    Also - it is about data - if you have root privileges you can view everybodies data on the computer.

    That is not good - I think he is still living the the microsoft windows 95 world when there was no network and the pc was just that a PERSONAL computer. Today it is mutlfunctional device that needs privilage escalation and descalation.

  196. I'll tell you why by cluening · · Score: 1

    Way back when I was a high school student, the "system admin" at my school used to log in to the school's mail server as root and just leave his terminal window hanging open. One day I had somebody else distract him while I walked into his office, copied /bin/tcsh to my home directory, set the sticky bit on it, and left. Oh, the fun we had after that!

    Of course, he didn't learn...

    --
    Posted from the wireless couch.
  197. fat finger by Cheeze · · Score: 1

    fat finger something like:

    rm -rf /var/log/*.gz

    and turn it into:

    rm -rf / var/log/*.gz

    and you'll quickly find out why root is not to be used all the time.

    --
    Why read the article when I can just make up a snap judgement?
    1. Re:fat finger by fishbowl · · Score: 1


      > rm -rf /var/log/*.gz

      Here's the thing, though.
      That command shouldn't work as a regular user, either. So it's really "sudo rm /var/log/*.gz",
      the -rf is really the big problem here, not root.

      Also, this should be done as part of a find command, and even better, as an automated process that deletes the files after archiving them and validating the archive.

      Either your system and your data is important (or valuable), or it isn't.

      For most Linspire users, maybe "it isn't."

      --
      -fb Everything not expressly forbidden is now mandatory.
  198. Finish reading my post first - by arete · · Score: 1

    I SAID that sudo can do that. But Linux by default doesn't ship with that as "how you make yourself an admin user"

    Except Ubuntu apparently does, which rocks and I just didn't know. Hopefully this is the start of a trend - a trend I'm trying to push along a little bit.

    --
    Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
  199. unix is multi-user by vga_init · · Score: 1
    The guy has a point, and it's important to remember that the target user for his Linspire system (as mentioned in other posts) is someone who is used to running Windows as a single user with Admin priveleges.

    The strength of security levels that is provided by file permissions and non-root access shine forth in heavily multi-user environments for which unix was designed.

    When many people use the same system, we usually want to give them means to protect themsleves from each other, and the system administrator wants to protect the base system from the users. The idea is to make your environment safe from other users with legitimate access to the system.

    Single user machines found in most homes need only protect the user's data. As mentioned in other posts, the system really isn't all that importart--it's not a server system whose uptime is critical; if it breaks, pop in the install disk and fix it. What's really important are the person's files, which they themselves already have unrestricted access to. When security for their account is compromised by sneaky means, the data is vulnerable whether they are root or not.

    In any case the user always has the risk of contracting spyware or file damage, but non-root usage helps ensure that one user compromising himself has the least potential for leading to the compromise of other user accounts. If Billy Bob runs the wrong script, he may have to suffer data loss or theft, but the other users can breath easy because he is not root.

    On a home system, however, there is only Billy Bob, and there is nobody else for him to be protecting. If he runs as root, he seems no worse off than he was before.

    1. Re:unix is multi-user by Pastis · · Score: 2, Insightful

      On a home system, there's Billy Bob, Mis Billy Bob, Billy Bob Junior, and the occasional 10 persons a year that come and borrow your machine for 1 min to read their webmail, the ones that play music when you do parties, etc.. etc...

      Never assume that because you're the only one to use your machine today, that you're not going to let someone else use your machine tomorrow.

    2. Re:unix is multi-user by vga_init · · Score: 1
      I agree with you there, but wouldn't you also agree that Billy Bob family would do what most home users do and resort to collectively using a single user account?

      Yeah, it's definitely a good idea...I myself have individual user accounts set up for each family member on the computer we've got running in the kitchen, but I don't think this is common practice.

      In addition, even I don't maintain guest accounts for those occasional visitors--not even on my linux system. Usually I let them load up the web browser from my own account, though I am there to sit and moniter their usage personally (I'd feel very uncomfortable leaving someone alone with the machine).

  200. Where's my cluestick? by dheltzel · · Score: 0, Flamebait
    Somebody hand me a cluestick, because that guy really needs a whack with it.

    Seriously, how can the CEO of a Linux-oriented company say something like this. He's either:
    a) managed to avoid listening to his engineers,
    b) got them cowed into telling him only what he wants to hear, or
    c) he's found the most clueless engineers on the planet to work for him (did he hire people from Microsoft, or what?).

    I think he's hearing rumors that people are criticizing Linspire for running as root and they are thinking about changing that, but honestly can't figure out why it's better to not run as root. He wants a list of reasons for the marketing of their new "secure" version of Linspire.

    Well, I for one am not helping with this, he can do his own homework, I'm not letting him copy mine.

  201. Root, Shmoot by zo219 · · Score: 1

    After having run Mac OS Classic for since 1992 - and always on Powerbooks - until Jaguar, I have to say, Hey, I ran my own machine! I could do anything I wanted with it! It was my responsiblity!

    I understand the power of Root, but as writer who works on her very own Powerbook, I am Up To Here with permissions. They have no meaning in my life whatsover, except as the constant price I do pay for using OS X.

    What you have here, fellas, is - gasp - only a Point of View. Mere mortals have been in charge of their own machines, free to run riot through the OS thereupon. The world has not ended.

    Have a point of view, but for god's sake, know so. I don't screw around with the 'nix underpinnings (much) . . but it is my computer.

    1. Re:Root, Shmoot by Narchie+Troll · · Score: 2, Insightful

      In 1992, there wasn't a spam and malware epidemic crippling systems and networks across the globe.

      You're like someone in the 1980s complaining about not being able to have unprotected anonymous sex anymore. Times have changed.

  202. It's not just MS... by veg_all · · Score: 1

    Apple fucked up PNG support too. I've just spent the last three hours dealing with this major annoyance. The humorous upshot: I had already frozen the blended layers to each other so the png-pointing style sheet goes to IE and I had to make a special jpg-based one for safari. Sheesh!

    --
    grammar-lesson free since 1999. (rescinded - 2005)
  203. You are forgetting something... by Anonymous Coward · · Score: 0

    Running as root is unsafe GENERALLY.

    However, there are two scenarios that lessen this vulnerability tenfold

    1) with a strong enough root password, 128 characters, mixed case alphaneumeric for example, it would take either a GOD or a supercomputer hundreds of thousands of hours to break into your system, assuming your "normal" user accounts are set up similarly.

    2) assuming the above is true, WHO THE FUCK IS GOING TO EXPLOIT THE PROVERBIAL "vulnabilities/exploits".

    You guys are a bunch of fucking n00bs. get a life for fuck sakes.

  204. MOD PARENT UP by shadowbearer · · Score: 1


    That's the best way I've seen that said yet, and I'll add that there are many adults for whom it's just as dangerous, whether it's linux or windows.

    Robertson is talking out of his ass.

    SB

    --
    It's old. The more humans I meet, the more I like my cats. At least they are honest.
  205. I ignore ActiveX and I hope it goes away by Monx · · Score: 1

    Ignoring it and hoping it goes away is not the right answer, either

    It is for me. I use Macs and Linux. If someone wants to make money from me through my web browser, they had damned well better do it using technology I can use. I won't be your customer unless you support my platforms of choice.

    I used to be a web developer. If one of our clients wanted some fancy single-platform buzzword technology just because they thought it was cool, we'd ask them if they'd rather be "cool" or reach 5% more customers. Guess which one they all chose?

    If the X% of people use Mozilla, then that's all the more incentive to develop sites that work in Mozilla. I do not want anything to remove that benefit.

    1. Re:I ignore ActiveX and I hope it goes away by Anonymous Coward · · Score: 0

      "If the X% of people use Mozilla, then that's all the more incentive to develop sites that work in Mozilla. I do not want anything to remove that benefit."

      Only on Slashdot would Mozilla not supporting a web standard be considered a benefit.

    2. Re:I ignore ActiveX and I hope it goes away by Anonymous Coward · · Score: 0

      I wasn't aware that the W3C had endorsed ActiveX as a standard. Nor ANSI, OSI or even ECMA for that matter. You have an odd definition for the word "standard".

    3. Re:I ignore ActiveX and I hope it goes away by Monx · · Score: 1

      Only on Slashdot would Mozilla not supporting a web standard be considered a benefit.

      The Web is supposed to be platform agnostic. Any technology that runs on only one platform cannot be a web standard.

      Mozilla supports web standards. Not supporting ActiveX is part of supporting web standards. This is particularly true if you consider that support can also be defined as: To keep from weakening or failing; strengthen. By avoiding proprietary, single-platform technologies, Mozilla is making web standards stronger.

    4. Re:I ignore ActiveX and I hope it goes away by Anonymous Coward · · Score: 0

      ActiveX has been steered by The Open Group since Microsoft handed over control in 1996. See the COMsource page for details of the COM/DCOM implementation for UNIX, etc.

  206. MYOB by Anonymous Coward · · Score: 0

    I *am* root.

  207. Re:you're wronger, nya by Lord+Bitman · · Score: 1

    Firstly: That is a reason for others to discourage the use of root. That is not a reason for me, as a user, to not run as root. I am talking in terms of what the user cares about. There is no reason for the USER to be concerned about that.

    Secondly: That is also more an argument for external firewalls. You should never have your network protected only by rules on the box you're trying to protect.

    Thirdly: spammers don't require root. Specific spammers require root for specific transport methods due to specific outdated conventions which make no sense at all in a desktop environment.

    The thoughts themselves:
    Sure, you dont want someone to gain root access, but you don't actually want them to have any access at all. If due to a convention that you've compiled in, blocking root access means making a handful of specific exploits not be able to work once you've already been infected, that's not really such a great thing to shoot for- You're still just as infected, and the infection is what you (as a user) want to avoid. Who cares about the effect (as a user, not as a peer of the infected)?
    You're still just as fucked if someone gains access as non-root, even if a handful of people wont be interested. Hey, if I compiled in an option that said you needed to be logged in as micheal in order to access the ports I use for DCC send, I could block anything which uses those from being useful after I'm infected.. and if I block the ports I use for printing if you're not logged in as bixbie... and for mysql if you're not logged in as bilbzerobaggins...

    --
    -- 'The' Lord and Master Bitman On High, Master Of All
  208. Well, a good response. by ta+bu+shi+da+yu · · Score: 1

    But you fail to see that those developers won't develop using FireFox, they'll just keep using Internet Explorer, as it clearly satisfies their needs. Those same developers also won't develop ActiveX for FireFox, because they won't see why they should do this.

    --
    XML is like violence. If it doesn't solve the problem, use more.
  209. That should be non-READABLE by achurch · · Score: 4, Informative

    as per this comment below (just bringing it up to make it more obvious). chmod a-x /home keeps you from doing anything in /home or any subdirectory, but will let you list /home; chmod a-r /home keeps you from listing /home but will let you do stuff in /home/bob.

  210. MR companies never have any revenue by herbthebod · · Score: 1

    Michael Robertson is the king of bad ideas that don't generate any revenue. Linspire has a zero chance of creating any resonable revenue just like Robertson's MP3.com Robertson simply laughs all the way to the bank as his stocks are overvalued and he somehow convinces idiots to give him millions of dollars of seed money.

  211. It's been done. by carlmenezes · · Score: 1

    I've seen people do an rm -rf * in / when they THOUGHT they were chrooted in a safe directory. As it turned out, they had forgotten they were not.

    I've also seen that happen with scripts that had a lot of sed and awk in them and that basically had a few bugs in their regular expression implementations that caused a lot more to be removed than was necessary.

    I've also seen that happen with people who are too impatient when entering commands...they don't enter the entire wildcard expression and end up taking out whole chunks of the file system like /etc

    --
    Find a job you like and you will never work a day in your life.
  212. He's right by ChiralSoftware · · Score: 2, Insightful
    The Unix/Linux security model is terrible. Here are just a few of the most painful flaws:
    • Any process that is owned by a given user has all the authority that that user could have
    • Some executables allow a process to start other processes with root access. If there's an overflow in any of the numerous suid binaries, any process can use it to escalate.
    • The most dangerous operations, such as processing network data, require root privileges. I still think that "must be root to bind ports < 1024" is the #1 Unix/Linux security bug and we've been suffering with it for three decades.
    • There is a user (root) which can access everything in the system. There's no way to grant a program the capability to listen to port 80 without also granting it the capability to write raw blocks on the disk, access raw devices, access other users' files, etc. This is an absolute disaster. No ordinary web server needs the ability to write raw disk blocks, so it shouldn't have the capability to do it.
    So yeah, the Unix/Linux security model is such a disaster that he's right! On a single-user machine (such as a typical Linspire machine) the user isn't really any worse off running everything as root.

    What would be nice is if someone would actually fix the Unix/Linux security model one of these decades.

    I'm sure a lot of Unix old hands (perhaps complete with beards!) will dismiss what I'm saying as rubbish, but I also believe that just being an old Unixer doesn't give anyone any special understanding of security. The way to get a special understanding of security is to think about, and understand, some theory ideas like least-privilege, capabilities, compartmentalization, that kind of thing. All those are foreign to the traditional Unix world, which is based on users and permissions. The users-and-permissions model is the ROT-13 of security models.

    1. Re:He's right by asuffield · · Score: 1

      Any process that is owned by a given user has all the authority that that user could have

      Linux doesn't have this problem.

      The most dangerous operations, such as processing network data, require root privileges. I still think that "must be root to bind ports < 1024" is the #1 Unix/Linux security bug and we've been suffering with it for three decades.

      Or this...

      There is a user (root) which can access everything in the system. There's no way to grant a program the capability to listen to port 80 without also granting it the capability to write raw blocks on the disk, access raw devices, access other users' files, etc. This is an absolute disaster. No ordinary web server needs the ability to write raw disk blocks, so it shouldn't have the capability to do it.

      Or even this. Get with the decade. We're using Linux now, not SysV UNIX. There are at least three entirely different ways of doing things. You don't normally see people bothering to deploy this sort of stuff because most of the time it just doesn't matter. The basic unix privilege model is good enough for almost every scenario, and flexible enough to expand upwards on the occasions where you need something more.

    2. Re:He's right by Anonymous Coward · · Score: 0

      Actually Linux does pretty much have all those problems, but so does Windows.

      In _theory_ the Windows Administrator accounts aren't equivalent to SYSTEM, but since the Administrator is given the power to change people's privileges they can temporarily uprate their acccess and then downrate it later. They can do this programmatically.

      IF you are a big corporate installation with physical access control (ie armed guards surrounding a vault room with servers in it), and IF there's a dual sign-in procedure that effectively prohibits you from tampering with the logs and IF someone actually audits those logs properly at frequent intervals THEN you can't do this sort of thing without eventually being caught.

      So, yeah, on a tiny microscopic fraction of machines the complicated multi-tier privilege system provides slightly better security. In the real world, where the machine is sat under someone's desk, or in a room to which even the illegal immigrant cleaners have a key, and where the same guy administrates the machine, runs the backups, checks the audit logs (once in a blue moon) and so on, Windows security is no better than Unix, just pointlessly more complicated.

    3. Re:He's right by lems1 · · Score: 2, Insightful

      instead of complaining you should really be doing research, writing white-papers and/or implementing these functions in code.

      stop bitching and help us all. enlighten us with your wealth of wisdom: put it in code.

      --
      This sig can be distributed under the LGPL license
    4. Re:He's right by ajs318 · · Score: 1

      As well as file permissions similar to {but not quite the same as -- there were four user levels including system, and four operations including delete} those on Unix, VAX/VMS had Access Control Lists {ACLs} for privilege separation, allowing you to specify exactly who could access what resource, when, and where from. ACLs gave the administrator an amazing level of control.

      They also saw about as much "real life" use as a pay toilet in a forest.

      The Unix security model works because it's simple. It's good enough, for enough of the time.

      If you really want the Apache user to have access to port 80, just recompile your kernel. {I'm taking it on faith that you know how to do that, because of what you've said. You couldn't have formed that opinion independently without compiling a kernel at least once in your life.}

      --
      Je fume. Tu fumes. Nous fûmes!
  213. Come on, this is no where near insightful. by Some+Random+Username · · Score: 1

    I manage to use apt as a non-root user all the time, using sudo. Just like I do with every other package manager for every other unix OS I use. There is simply NO excuse to be running as root when sudo is so powerful and simple.

    1. Re:Come on, this is no where near insightful. by killjoe · · Score: 1

      We are talking about the people who use lindows here. And really they should not have to sudo either then you think about it. On the mac you can install software into your own home directory without any kind of hassle, you should be able to do the same with linux. It's something you can do with --with-prefix=~ if you were compiling by hand so I don't see why it should not be possible with a package system.

      --
      evil is as evil does
  214. Who cares? by Some+Random+Username · · Score: 1

    Hooray, I deleted the data in my home dir. Big deal, its backed up. It took a whole 8 minutes to restore. Somehow I think it would have taken me longer to reinstall my OS and all the applications I have installed. And if I hadn't been intentionally doing this just to see how long it takes to restore my data, I would have hit ctrl+c during the long period of "rm: blah: Permission denied" scrolling up the screen.

  215. Robertson is right by pvdl · · Score: 2, Interesting

    Actually, Robertson is right.

    He said "why is it more secure to not run as root. Nobody really has a good answer. They say "oh, yeah, it is!", but it really isn't. Here's why: What's the most important thing on your desktop? It's the data. If someone gets access to your libraries or whatever, who cares? Your data is the most precious thing on your computer. And whether you log in as root or log in as user, you have access to that data, technically anyone who's compromising your account has access to your data as well. "

    Obviously he is talking about single user computers, as most PCs are. If you have a single user computer, when your user account is penetrated, your root account is penetrated next time you su.

    The last step in a Linspire install, which apparently noone in this thread has done, is to set up user accounts for a multi-user system. If it is a single user system, there is NO additional security to setting up a user account.

    My data is the most important thing for me. I can reinstall Linux in 15 minutes, but my data is irreplacable.

    Peter

  216. heh by smash · · Score: 1
    Michael Robertson Says Root is Safe
    Ahh well in THAT case, I'll just abandon my decade of unix administration experience, plus 30 years plus of Unix heritage, and take his word for it :D

    smash

    --
    I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  217. wow..... by paulevans · · Score: 1

    It must be a happy happy place in Linspire-land.

    --
    "When I want your opinion, I'll give it to you." --leonstryker
  218. The Uptime Flap, almost 10 years now by istartedi · · Score: 1

    This reminds me of back in the day when everybody bragged on Linux uptime vs. Windows. The *NIX snobs still don't get the desktop after all this time. As long as a desktop stays up all day, that's all that matters because desktop users treat their PCs like TVs. Yes, 1990s Windows desktops often failed that test, but Linux users back then were comparing CLIs to desktops and bragging on uptimes.

    This root flap is the same sort of deal. Yeah, root can kill all the users on a box. Guess what? Granny is the only user on the box. If user Granny's files are dead, the whole box is dead as far as she's concerned, and if her kernel has been replaced by KERNEL.i0wn3du, she is just going to run the restore CD.

    That's not to say that I agree whole-heartedly with the "root is safe" assertiong; but I understand where the guy is coming from. In a way the *NIXs are perhaps a victim of their own success in writing apps that are (usually) secure enough so that you can get away with running as root.

    --
    For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
  219. some desktop reasons not to run as root by thrift24 · · Score: 1

    Yes, your data may be deleted when you launch a nasty virus under your account, but when your 12 year old son is scouring the Internet for pr0n, your data is safe.

    The OS can't be mangled. So even if you lose your data, at least you can still perform whatever functions you need to perform under a seperate user. Or the kid from next door can come over, log on as the super user, rescue all your documents and mp3s, delete you user and home directory, add your user back, then move the data back....then you might not lose your data or your ability to use your computer.

    There are more options for data recovery. assuming a very annoying virus that just does 'rm -fR /' your links to the inodes are gone, not the data. Now if you were smart, you could just make an area available to the super user where all those indodes were hardlinked and could be restored. Of course there are ways around such things, like simply 'find | xargs tee', but at least there are other options available to secure files.

  220. wow by mkiwi · · Score: 1

    Oh dear God now you've done it....

  221. Good point, but there were alternatives by Anonymous Coward · · Score: 1, Insightful

    A better choice would have been binding a key on the local keyboard (a secure tty) to cause a root-privilege menu to come out over the current desktop (in such a way that no user-level window can obscure it).

  222. one eye blue, one eye brown? by Anonymous Coward · · Score: 0

    Is it just me, or is one of his eyes blue and the other brown. Probably had a play in Photoshop. Creepy.

    1. Re:one eye blue, one eye brown? by Anonymous Coward · · Score: 0

      Nah, I know a person whose eyes are the same. It's quite rare, I think.

  223. The high-res pictures by Anonymous Coward · · Score: 0

    Those are for use by the press. E.g. there's something in a magazine about Linspire, so they pick up a press-quality picture and make a 2 cm x 1 cm box with the guy's face in the page.

    Also, I'm convinced he has no clue. I will never go near "Linspire", ever. And I will not let anyone near me do the same.

  224. I'll bite... by afxgrin · · Score: 1

    What does ActiveX do that XPCOM and Java are incapable of performing?

    I'll play devil's advocate here, and say the first thing that comes to mind: It can't run all those ActiveX applications that already exist. Companies invest money into programmers who will maintain or program existing software and infrastructure.

    Maybe what's needed is an ActiveX type clone just to end this problem. Let's keep in mind that there are large numbers of IT workers who do not frequent this website regularly, and don't really care if XPCOM and Java are better - it still won't run the ActiveX applications they currently have.

    There are large numbers of people who like having LESS work to do when they are at work. If they need to hunt down and replace all their internal ActiveX applications, that will simply make more work for them. Plus, their PHBs will always see this as just costing more money.

    But, if this means swallowing some open source pride, and writing a GPL ActiveX clone just to edge out Microsoft elsewhere, then it's worth it. Of course it would never be installed by default, and of course there would be multiple pop-up windows telling you why this is a bad idea, but it should be available.

    If the strategy is played correctly, we could start slowly moving companies over to more OSS solutions and away from ActiveX.

    But, in the end, Compatibility is King, since your PHB sees this as a means of saving money in the short, and long-term. Anything that is seen as a non-drastic change to the infrastructure will always be regarded as a good thing as they view it as a means to save money. Think of all those computer "n00bs" who will still refuse to buy a Mac even though all their computer nerd friends tell them to buy one. (I have a PC - but I want my friends to start using Macs since I'm sick of fixing their PCs) But, I always hear the concern over compatibility, since they know not everything will work on a Mac. Yes, there are equivalent products, but they don't care since it won't run the "majority" of computer products already existing.

    Am I completely wrong? Please tell me, I know someone here will anyway. :-)

  225. Well... Even a Windoze could run... by Anonymous Coward · · Score: 0

    ...for years (5 or more) without any problems and with Internet connection when the user is not an Admin-user...

    Virus (wich involves any user activity) may be avoided or/and removed easily and so can Spyware...

  226. RedHat and that "rm" alias! by PowerBert · · Score: 1
    I blame RedHat.

    If you wanted to be safe and have it confirm your request before it does each and every action you shouldn't specify the 'force'


    If you want that behaviour you require the 'interactive' option. rm -r will not prompt for every action. unless you're on a RedHat box with rm aliased to "rm -i"

    That alias was the worst thing RedHat ever did. It didn't protect anyone, it just made the avergae joe too familiar with the force option. I haven't used RedHat since version 8.0, but the first thing I used to do was delete that bloody alias in my .bashrc

    I suspect most people find the force option a long time before they find the alias, but by that time using -f has become entrenched.
    1. Re:RedHat and that "rm" alias! by Anonymous Coward · · Score: 0

      The cp, mv and rm -i aliases only exist in root's .bashrc file, the average joe isn't supposed to be using root in the first place.

    2. Re:RedHat and that "rm" alias! by Anonymous Coward · · Score: 0

      That "alias rm rm -i" is a very old thing. I recall discussion about that when I was in university (early 90's).

      It was already universaly considered as a bad thing, as you start relying on it, and, one day, you end up on a system without the alias, and *blam* your files are gone.

    3. Re:RedHat and that "rm" alias! by Cro+Magnon · · Score: 1

      I prefer "rm -i" for multiple deletes, but I would be pissed off if I thought it did it automatically and got bit on Slackware, where it doesn't.

      --
      Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
  227. Root is safe? by Francis85 · · Score: 2, Funny

    Has netcraft confirmed this yet?

  228. ONE DAY by cfalcon · · Score: 1

    You will see computers that are just as "usable" as everyone keeps wanting. You'll be able to do all your things without thinking about a filesystem or files or anything like that. It'll work great...

    But it'll be so restrictive that it won't qualify as a general purpose computer in your eyes. It'll be good for companies, who can sell you ten computers instead of one, but that's it. Some will buy them. You will, if they become cheap, but you'll keep your gp machine.

    The elevator crack is also silly because an elevator goes up, down, and stays still. The operation can be more or less complex, but that's all it does. A computer is a canvas for a painter, a darkroom for a photographer, a pasteboard for a desktop publisher, a video editor, a storage device capabable of holding everything you have ever done or had the time to record or take a photo of, a typewriter for the novelist or a poet, a super fast and simple calculator for a child, a super fast and complex calculator for a scientist, a general purpose slave who has no complaints for a programmer, a spreadsheet for anyone running their own business or keeping track of their family finances, a limited tax attorney, a way of designing every physical thing from PVC pipes to rocket ships, and a robotic operator in charge of the construction of those things. It's a blazing testament to the will of man, a monument to tool use, and the best thing humanity has going for it since the opposable thumb.

    And all of that was BEFORE the web!

    1. Re:ONE DAY by Anonymous Coward · · Score: 0

      At first I thought your comment was just stupid. Then I got to the part about how computers are the best thing ever. At that point I realized that you're dangerously insane.

  229. It's a question of priorities by r_newman · · Score: 1

    It really is. Michael Robertson sees his Linspire Linux desktop as Linux for the masses. This is his number one priority over-riding all else. He probably knows, is probably well aware of every reason that it's inadvisable to run as root, and his technical staff are probably tearing their hair out now because they think he wasn't listening.

    I think people here are misunderstanding his motives. Linspire is not a server distribution. It's not even marketed as a multi-user desktop distribution. Linspire is marketed as a single-user PC OS, with great multimedia capabilites. It's designed to appeal to the Windows users.

    And if anything can break Microsoft's monopoly on the desktop it's what Linspire is trying to be; a freely distributable OS that matches and exceeds the functionality and user-friendliness of MS Windows.

    Give the guy a break. It might not be our cup of tea, but so what? We don't have to use it. We have a choice, unlike the Windows users.

    --
    Bzzzzzt..."AAAAaaaaarrrgh!!!" Thud.
  230. ...That's a good thing? by GotenXiao · · Score: 1

    "XP has a super-root account which nobody but MS has access to"

    So Microsoft left themselves a back door into *my* PC? This is exactly why the Blaster worm took off; because they left a back door entry into Remote Assistance.

    Besides which, why the hell should they have a username on my machine? Why can't I utilise my PC to the utmost?

    In any *NIX system root is GOD. There is very, very little that root can *not* do. Hence why it's very heavily recommended to use a seperate account for everday use, and just use su or sudo to install apps etc.

    Some people (yes, me included. I hang my head in shame :P) will accept the risks and run as root anyway. It's a conscious decision on my part, and I'm fully aware of the possible consequences. 'tis another reason why I'm muchos careful when running rm with -R...

    --
    Goten Xiao
  231. He just doesn't have the technology by nitecoder · · Score: 1

    He probably just doesn't have the code to run as a non-root user without requiring password. While his developers are working on it, he's blowing smoke and waving hands. "these are not the droids you are looking for" "who needs to run as non-root anyway". Meanwhile Windows is moving to a Least Priveleged User model, and for a good reason. I bet Linspire will shortly as well.

  232. root on a single user system isn't a big deal by OrangeTide · · Score: 3, Interesting

    When one RTFA they will notice that Robertson is talking about a desktop system. Having users log in as some root/admin account is not a big deal because the only thing valuable on that system is the data stored as the only user on their system. Obviously he's not saying "run apache as root". In fact he implies it would be a very bad idea to allow things like a webserver to have write-access to a user's data!

    Now if you are maintaining a multi-user system, root access is more powerful because it grants you full access to all user's information. Although these days a family computer has multiple accounts on it, Little Timmy and Mom's data is seperate. If Timmy downloads some malicious code in some new music sharing program that turns out to be a trojan, at least Mom's calendar, address book and tax information will be protected.

    Of course I'd recommend periodic backups to give you real data security. That's perhaps more important than the root/non-root issue.

    --
    “Common sense is not so common.” — Voltaire
    1. Re:root on a single user system isn't a big deal by Craster · · Score: 3, Interesting

      This is only true if it's only your data that you care about.

      A compromised system with a user running as root? Simplicity itself to set an SMTP daemon running and happily accepting and relaying mail.

      Welcome to the world of the Linspire open relay network.

    2. Re:root on a single user system isn't a big deal by PigleT · · Score: 1

      > Little Timmy and Mom's data is seperate.

      Not if you're root, it isn't.

      --
      ~Tim
      --
      .|` Clouds cross the black moonlight,
      Rushing on down to the circle of the turn
    3. Re:root on a single user system isn't a big deal by OrangeTide · · Score: 1

      You don't need to be root to be an spam zombie. You only need SMTP daemon to recieve spam. You can connect to other smtp daemons directly, or just use the one locally installed to send out mail. And user account on a compromised OpenBSD or Debian box can be used as a spam relay.

      --
      “Common sense is not so common.” — Voltaire
    4. Re:root on a single user system isn't a big deal by OrangeTide · · Score: 1

      That's the case for not using root. Sorry if I didn't make that very clear. I was arguing both the pros and cons of running a system as root. In a system where only one dumb user is using it, then root is fine. But a typical household has several dumb users sharing a computer.

      --
      “Common sense is not so common.” — Voltaire
  233. Don't try... by Anonymous Coward · · Score: 0

    Not kind.

    You should have used test and test2 as a subdirectory :
    your script will delete *.tar in the current directory.

    Well I don't have any anyway : only .tar.gz and .tar.bz2

    1. Re:Don't try... by ilikejam · · Score: 1

      Anyone who runs scripts found on places like Slashdot without understanding them deserves to have their .tar files deleted.

      --
      C-x C-s C-x k
    2. Re:Don't try... by Anonymous Coward · · Score: 0

      Some man pages are better than others. But I can see as how the first AC may have been confused. Some man pages are better than others.

      For instance this is more enlightening than the one that usually ships:

      -r Permits recursive removal of directories and their contents when the File parameter is a directory. This flag is equivalent to the -R flag.

  234. rm -rf /* by Gribflex · · Score: 1

    only slightly different than rm -rf *
    and way worse if you are running as root.

  235. The user doesn't have to know about it. by Some+Random+Username · · Score: 1

    You do realize the user isn't actually running apt themselves right? The graphical application that they use currently does "install command", it just needs changed to do "sudo install command". Users will never see or notice the difference, the lindows knobs just have to do their job instead of the current cop-out of making linux into a single user system.

    1. Re:The user doesn't have to know about it. by Anonymous Coward · · Score: 0
      I might be wildly off track here, but when you run a program using sudo, isn't it running as (wait for it)... root? sudo == "super-user do" IIRC. Correct me if I'm wrong.

      Giving inexperienced users permissions to run sudo is a bad idea - they'll just get used to the idea that whenever the machine tells them they can't do something, sudo it, not step back and think that maybe they don't actually need to do it that way, or that there's a reason why they aren't allowed to.

      Seriously, if I was trying to exploit a machine and gained access to a non-root account, one of the first things to try would be sudo.

      Check this out:
      phil@xerxes phil $ sudo bash
      Password:
      bash-2.05b# whoami
      root
      bash-2.05b#
      .. and the password it wanted there? MY password, not root's.
  236. Robertson WANTED to be /.'d by DenDave · · Score: 1

    People, move along now, this is a jest from Robertson to get our attention and start up a yes-no debate.. fugeddabatit...

    --
    -if at first you don't succeed, stay the heck away from paragliding.
  237. MarketSpeak by tacocat · · Score: 2, Interesting

    Obviously his answer is Market Force driven and non-technical. He ships as root, he doesn't want to sacrifice his products perception. He'll never say anything else.

    Would you expect the CEO of Exxon to openly state that there is something called Global Warming and it is necessary for everyone to stop driving gasoline powered cars?

    Certainly not until they have the answer. It maybe be the Linspire is working on changing this for real, but it won't be openly discussed.

    1. Re:MarketSpeak by argent · · Score: 1

      It maybe be the Linspire is working on changing this for real, but it won't be openly discussed.

      What is there to work on? He only needs to undo.

  238. his real goal in saying this by Anonymous Coward · · Score: 0

    hey kids, can you say:
    FLAMEBAIT!

  239. Re:Robertson is the "Billy Mays" of the Linux worl by davew2040 · · Score: 1

    I know the hardcore geeks feel differently, that's fine.

  240. Difference in focus, really by Moraelin · · Score: 1

    The /. crowd seems to assume that everything is a server. Even if it's their home computer, it's got to run Apache, MySQL, Squid, a mail server, etc.

    That is _not_ what Joe Average needs, however. For that matter, not what _I_ need.

    I explicitly do _not_ want a web server, database server, or any other goddamn server on my desktop machine. I explicitly don't need one, and I explicitly don't want one using up my RAM and CPU cycles. I'm pretty sure mom and dad don't either.

    So the whole "but what if someone uses a vulnerability in Apache?" is a moot point: they won't find Apache on my machine to start with.

    Also the whole idea why it's called a _Personal_ Computer (PC) is that I don't need, nor want, a multi-user bonanza on it. It's not some server where every Tom, Dick and Harry has their own separate account and their own separate data.

    The same, incidentally, applies to most family computers. Joe and Jill Average, and their 2.2 children, most of the time don't keep their files secret from each other. It's not like Jill's digital photos of trees and squirrels are some top secret.

    So all that someone could exploit is some program _I_ am currently running, as _my_ user. Period. And then it can erase my data.

    And that assessment is right: that's what's important on that system. The programs are the easy part: reinstalling the whole system and all the programs is a few hours exercise. Getting your own data back might not even be possible, short of having a time machine.

    I.e., for Joe and Jill Average, with a _desktop_ machine (not a server), it really makes zero difference whether they run as root or as some other user.

    --
    A polar bear is a cartesian bear after a coordinate transform.
  241. Well... by Etriaph · · Score: 1
    ...in about 95% of all cases I would say that running as root really isn't going to run you into trouble, but let's put this out there.

    Bob runs his system as root, and Susie doesn't. Bob downloads something from the internet which (*gasp*) has a virus attached to it which *can* affect a Linux system. Bob infects his whole system. Susie, who downloads the same file, infects her user account. Running as a user reminds you that you are not in control of the system, and to be in control you have to enter a password. This usually makes people more wary about what they do and do not do as root.

    Make sense?

    --
    "It's here, but no one wants it." - The Sugar Speaker
  242. What if your finger slips? by holy_calamity · · Score: 0

    "bash$ rm -r /"

    'nuff said

  243. Makes sense by Anonymous Coward · · Score: 0

    The average user wants to change computer setup for zxy piece of hardware doesn't want to bother with passwords. Hell what is the point of passwords anyway. Most passwords are easily cracked or sniffed. Dialogs / apps that need super user access to perform tasks and keep asking for higher privileges constantly just condition the user to enter the passwords automatically. If a popup in a browser asks the user for the root password I wouldn't mind betting that most uneducated users would quite happily enter it.

  244. ONE GOOD REASON by Anonymous Coward · · Score: 0

    rm -rf *

    DOHHHH!

  245. Some of you are missing the point by GauteL · · Score: 2, Insightful

    While I do agree that the most important data is what is stored in the $HOME directory, running everything as root, puts the OTHER users at risk and not just yourself.

    Some would say that this doesn't matter if you are a home user but even home users should (and often do) have different users for the different family members.

    If the 13 year old kid downloads lots of 31337 warez and gets a worm thrown in with it, this shouldn't affect dads documents, budget, tax stuff and credit card information.

    If you run each account as root, this is bound to happen sooner or later.

    The Right Thing [tm] to do is to make it easier for home users to live with security, rather than just remove security. OS X manages this decently, why can't Linspire?

  246. Firefox ActiveX Plugin by Millennium · · Score: 1

    A plugin that would let Gecko-based browsers use quite possibly the most fundamentally-flawed piece of software ever to have been written?

    Gee, who'da thunk it?

  247. Do you mean indicative? by Omni-Cognate · · Score: 1

    Endemic. I do not think it means what you think it means.

    --

    "The Milliard Gargantubrain? A mere abacus - mention it not."

  248. Re:This thread is scary! Here are the reasons... by Anonymous Coward · · Score: 0

    None of that stuff applies to a Lindows box. This isn't the mainframe days anymore, Lindows is not a mainframe.

    The "the point of why we do things a certain way" is that in the old days when things were different (eg 2 computers on the whole school campus) it used to make sense.

    "as root you can still get in and fix it, completely" - yeah, I'd like to see a Lindows user try and do that.
    Try put the CD in the drive, click reinstall.

    The only reason you thought up that's potentially arguable is not being able to run services below port 1000, specifically port 25, but does that really matter when the box can still connect to port 25, and accept incoming connections on higher ports? It can still be a zombie on user privileges.

  249. Selling Lindows by northcat · · Score: 1

    At the risk of getting modded as Redundant, I'll say it... Certain things can't be done as regular users and needs root privileges. Lindows (or Linspire or whatever) wants to dumb down things and so wants to avoid asking password prompts and such. Therefore Lindows either already logs in the main regular user as root or it wants to do it in the future (I don't know, I've never used it). Of course, pretty much every Linux newbie site and every Linux user says that doing normal things as root is bad and stupid. This makes Lindows look bad in the eye of its users and in the eyes of experienced people and the media. Therefore it wants to start spreading shit that root is safe and people who disagree are paranoid and idiots.

    (Although I don't like getting modded down, I do hope this post is redundant and everyone did know this point)

  250. Stupid comments like this won't hurt Linspire by BlakeCaldwell · · Score: 1

    Everyone here knows his was a stupid quote, but from a marketing standpoint, it's valid... None of us are going to buy Linspire -- we'll run our slack/deb/rh/fedora -- but, cheap newbies will, and to them, it's no different than running their existing OS, Windows 98...

    There's no chance Robertson actually means what he's saying though. The CEO of a linux company can't possibly be that stupid.

  251. For most people there are NO other users by Moraelin · · Score: 1

    Just for the sake of repeating myself, don't assume that every computer in the world is a server, and/or that everyone must be running a l33t multi-user system with 500 different accounts define.

    A normal user's home computer is a very different beast. Normal users don't measure their e-penis in number of uptime hours or number of l33t server processes on their system. Most of that l33t stuff that your average /. nerd is proud of, is just a waste of any normal user's time.

    The average normal user doesn't have a server or thousands of user accounts on his/her server. And has no intention of going that route. Now in that context:

    "Running as root is like pointing a loaded gun at everyone just in case they're a criminal."

    Yes, except there is noone else on that computer. So it's more like pointing around a gun in your own concrete basement, with noone else around. Whop-de-do, that must be sooo good a reason to not run as root. Not.

    "Running as root is like driving down the highway with your hood open and your oil cap off."

    Except it's on your own private strip of road, and noone else has any business to be on that road to start with. So the problem is?

    " Running as root is like posting to slashdot without reading TFA. :)"

    Except it's a post in your own private diary. So the problem is?

    --
    A polar bear is a cartesian bear after a coordinate transform.
  252. Install? by No+Such+Agency · · Score: 1

    Of course you can't INSTALL it without Admin, that's the whole point of Admin! You can still USE Firefox on a user account.

    --
    Freedom: "I won't!"
  253. He's full of shit by FreeUser · · Score: 1

    He's full of shit. Even Mac OS X, the quintessentially "easy to use, just works" OS, requires you to type in the administrator (similiar but not the same as root) password before installing new software or OS updates. Granted, they do give "administrative users" access to the /Applications directory, so the default non-root user can copy apps to /Applications in those cases where drag-and-drop is all you need, rather than an install wizard.

    The idea that everyone should "just run as root" is asinine and toxic. People like this, and the distributions they create implimenting this philosophy, will give Linux a bad name security wise, and probably become the Microsoft posterchild for how Linux doesn't measure up to its "hype." He'll give the disinformationists in Redmond something to hang their deceptive hats on, and damage the reputation of hundreds of distributions that are more secure than windows ever will be because they don't succumb to the least common denominator.

    --
    The Future of Human Evolution: Autonomy
    1. Re:He's full of shit by killjoe · · Score: 1

      "Even Mac OS X, the quintessentially "easy to use, just works" OS, requires you to type in the administrator (similiar but not the same as root) password before installing new software or OS updates."

      Only for OS updates (logical) or applications installed into a folder other then your home directory (also logical). There are a ton of Mac OS X apps which don't have an "installer" per se you simply drag the item to your home folder and use it without having to know the root password.

      --
      evil is as evil does
    2. Re:He's full of shit by FreeUser · · Score: 1

      Only for OS updates (logical) or applications installed into a folder other then your home directory (also logical). There are a ton of Mac OS X apps which don't have an "installer" per se you simply drag the item to your home folder and use it without having to know the root password.

      Yes, and that would work under Linux as well. Perhaps you skimmed over it, but I did point out exactly what you said, namely that anyone in the administrator group (the default non-root user, and any others you add) also has write permissions to the /Applications directory, so you can drag and drop those applications to the Applications folder and make them available system-wide without typing a password. But that is limited to the /Applications directory (and your home directory, of course), and to applications which can be copied and don't use an install wizard.

      --
      The Future of Human Evolution: Autonomy
    3. Re:He's full of shit by killjoe · · Score: 1

      Right. All I am saying is that linux needs an easy way to install packages under your home directory without needing the root password.

      --
      evil is as evil does
  254. Not exactly... by dreadclown · · Score: 1

    Compuserve did not "hold the patent", Sperry (by then Unisys) did. Compuserve merely licensed the patent. See, for example, amongst many other web resources, http://www.kyz.uklinux.net/giflzw.php and http://lpf.ai.mit.edu/Patents/Gif/Gif.html Just being pedantic, but the idea of Compuserve ever coming up with something patentable was mind-boggling...

  255. Once again... by yoshi_mon · · Score: 1

    There is a huge difference between mission critical software and end user systems.

    I have never, and will never I'm quite sure, see any deaths related to some end users Dell system kill them because Outlook Express crashed on them.

    Once again, your comparing apples and oranges.

    --

    Really, I know what I'm doing...Ohhhh, look at the shiny buttons!
  256. Running everything root on a desktop? by Anonymous Coward · · Score: 1, Interesting

    Michael Robertson's market is rather different from the typical Linux market. He's trying to sell an end user commodity.

    The end user does not give a fuckola about permissions, user management, and the meaning of the word "root". Insecure? Yeah, a little.

    If a regular user runs a malicious program, they've already risked all of their own data. The system itself is "safe", but many of the reasons people 0wn Windows boxes can be satisfied just by having user privileges. It can be used as a spam conduit. It can be used in a DDoS attack. It can give the keys to someone else so they can try a local exploit to gain root, or it may have a set of local exploits built in to elevate to root right there.

    Running any malicious code represents some kind of compromise. The argument for running it as a non-privileged user vs. root user is just one about dampening the impact, but just slightly.

    On the other hand, running everything as root makes the end user experience a lot more comfortable. Security is inconvenient.

  257. he's a lame fuck by Anonymous Coward · · Score: 0

    He's such a lame fucker. He doesnt know a shit of
    security. Even a kid nowadays has an answer of why running a root is unsafe. damn bitch

  258. You Made The Choice by Makarakalax · · Score: 1

    You have to make choices in life. Your company chose to go the Microsoft route, and now they are a little stuck. There are many reasons that OpenOffice etc. don't support these MS solutions. Yes a few of them are a little pathetic (eg. "I aint supporting no MS shit!"). But this is just how it is.

    So you choose the MS way, the Open Source way, or perhaps a company that has a solution and doesn't believe in lock-in.

    Once you've chosen you have to eat what you've picked; there's no point whinging about it afterwards.

  259. No, it's exactly as secure as I think by arete · · Score: 1

    No, it's exactly as secure as I think - and I even said that it wasn't a perfect implementation in my post - I just think it's a brilliant model.

    The "right" answer to your complaint is to make the default admin account that is admin but isn't wheel - it'll just make you sudo more things. I agree with you about this detail, but it's an implementation detail, not a flaw in the model.

    Also, of course, the OSX firewall is a packet filter (pf, I believe) and a GUI. The GUI might not set the filter to block UDP but the underlying filter definitely does. So you definitely could fix this without installing any software. Still an Apple failure? Definitely. Better than the Windows equivalent structure? Also definitely.

    --
    Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
  260. Bullshit. by schon · · Score: 1

    don't assume that every computer in the world is a server, and/or that everyone must be running a l33t multi-user system with 500 different accounts define

    Nice straw man. You shouldn't assume that most computers are only used by one person.

    Most people live in a household with *MORE THAN ONE PERSON*, and when these people have a computer, it will *MORE OFTEN THAN NOT* be shared with the other members of the household.

    You don't need 500+ people on a system for it to be multi-user. You only need *TWO*. That's it - just two people.

    Ask someone "hey, you won't mind if your brother accidentally deletes all your files, right?" And you'll get a "of course I mind."

    The argument that data is more important than the system isn't relevant to most computers - it only matters if you can guarantee that *NOBODY* else will be using your computer.

    Now, in *YOUR* case, that may be true, but don't assume that because it's true in your case, that it's true in everybody's case.

  261. Only Worrying About the Worst Case by Delos · · Score: 2, Interesting

    The stupidity of this position is very easy to explain. He's claiming that the worst thing (losing user data) is the only thing to worry about. Since non-root doesn't prevent that, let's get rid of it.

    To use his own analogy, if the worst thing that can happen in a car is to run into a wall, then why have door locks? Whether you have locks on the door or not, you're still going to die. And they make it hard to get into the car, so let's get rid of them.

  262. rm -rf *? by Epsillon · · Score: 1

    What is this fixation with this command? I can think of a much more widespread example of why running as root is not secure: Anything you launch runs with root privileges. That means a fast-track route for anything even slightly malicious (worms, trojans, spyware) to your kernel. Please don't tell me they don't exist; if this running as root becomes widespread, think of the 1337-ness factor of being the VX-er who trashed a few thousand Linux boxen.

    If you think about the "root" username in terms of Australian slang, you'll get the idea: You *can* root your system using it in many ways that aren't immediately apparent. Try, for example, chmoding /var/log 400. Try chowning someone's home directory. Try deleting the kernel from /boot. Try changing root's shell and then deleting the shell binary (or one of its dependencies if it's dynamically linked) and dropping to single-user mode.

    It looks to me like a case of SELECT FROM Linspire_employees WHERE CLUE > 0
    No records found.

    No offence to anyone working for Linspire, but if you don't tell your boss he's a jerk and needs a good dose of clue stick #10, you're as clueless as he is.

    Between this little statement and Linus having a go at Tridge, I'm becoming more and more pleased with my own choice of open source OS. Please tell me these guys are going to stop before they destroy any good that they have done. If it wasn't for Linux, we'd have hardly any applications to run.

    --
    Resistance is futile. Reactance buggers it up.
  263. Easy by vargasmas · · Score: 0
    Who has he been asking? Maybe he needs to hire better IT people to advise him? A CEOs knowledge level is only as good as the knowledge level of the people who advise him. We wouldn't expect the CEO of GM to be able to gap a spark plug or tune an engine!

    Notice what he says: "Nobody really has a good answer." That should be your first clue that someone is falling down on the job and not giving the boss the right advice. Or maybe someone is afraid to tell the boss something he might not want to hear.

    The comment "I defy anybody to tell me why is it more secure to not run as root." is another way of telling me to send him my resume. It's also a opportunity for a smart consultant out there to land themselves a nice contract.

    And "I would imagine a few Slashdotters would dispute that." tells me he is smart enough to get the user community talking about this problem and seeing what solutions come out of the woodwork. Once he hears a good idea, it's time to get "his people working on that" and come out with a new product without paying royalties for the idea. Just another way to do low-cost research. Not the first time it's been done.

    Just my 3.14 cents worth.

  264. Root isn't bad... by Foolomon · · Score: 2, Funny

    ...it's evil. :D

    A looooong time ago in a galaxy far, far away when I was in college, there was a graduate student (Bill Crossman, if I recall, was his name) who one day decided to clear out his home directory before packing up to go home for the summer. So he typed "rm *" at the shell command line and thought all was good.

    He forgot he had logged in as root.

    From that point on he was known as "arr emm star Crossman" :D

  265. The "find" command by Mr+Guy · · Score: 1

    Learning it is the happiest you can be in linux when not surfing for porn.

    find /. -type f -name *.tar -exec rm {} \;

  266. Ignorance. Sheer ignorance. by Millennium · · Score: 2, Insightful

    The problem with this statement is that the man's idea of security is too simplistic. He sees the only security threat worth worrying about as stealing, altering, or erasing the user's data. Certainly this is a valid concern, and if it truly were the only thing to worry about on the Net then his complaint would be valid. Where it falls on its face, however, is the plain and simple fact that this is not the only thing to be concerned about, security-wise.

    Nowadays, many malware authors don't actually care much about any data that's stored on disk. It's the data you enter every day, often without realizing it, that's [i]really[/i] interesting. What sites you visit, what ads you respond to, and such: these can be gleaned from history files with some success, but by the time you get that data it's already out of date. Getting it in realtime is better, and this is what spyware does.

    Even this, however, is not the only reason malware exists. Very often, what malware authors want isn't even your data; they want your computer itself. That's what zombie networks are, essentially. This allows The Bad Guys (be they crackers, spammers, or whatever) to make use of your machine to perform their nefarious activities, and the hell of it is that they can do it in ways that make it look like you're the culprit.

    Of course, even this doesn't cover everything. Adware doesn't usually bother to collect data (though it can), and often doesn't act as a zombie: it's just there to shove even more ads in your face. Yes, this is more annoying than destructive, but it's still malicious.

    The point of all this: Data may be king, but a king is nothing without his court. That's the problem with this man: he's too ignorant to see any security problems outside of data theft. As a result, he advocates irresponsible computing, seeing no harm not because there isn't any harm -for there is- but because his concept of harm is not broad enough.

  267. Cars vs computers by Cro+Magnon · · Score: 1

    With cars, there's a clear middle ground. I don't know shit about fixing the damn thing, but I know when to get the oil changed and I know to take it to the shop when a light comes on.

    With computers, you have people who think any idiot should be able to use it without "oil changes" (security updates), and gear-heads who think you should be able to build a computer from scratch before you're allowed to use it. Both are wrong.

    --
    Slow down, cowboy! It has been 4 hours since you last posted. You must wait another few hours.
  268. Windows has this by zakharin · · Score: 0

    Yes, it is very difficult to configure, but the system policy and other aspects of NT security provide extremely fine-grained control over what users, applications, machines, etc may or may not do.

    Now if only it were pre-configured out of the box. But of course the minute Microsoft were to announce that any third-party application must be validated by MS to determine its default security rights, everyone would complain that MS gets to decide which programs make it to market.

    Also, .NET allows applications to demand, request, or refuse permissions, so if the developer chooses to, the program will say upon startup "I am an E-Mail program and want to access nothing but my data folder." Then, (provided there are no holes in .NET) if there is a security hole in the app that would allow an attacker to, say, connect to IRC, the .NET security model will refuse this connection.

  269. NOT OUR JOB by randomencounter · · Score: 1
    To fix Microsoft's broken crap for them.
    They want the big bucks for their "professional" OS, they accept the responsibility for making it work.

    If I want to "take steps to fix it", I'll run Gimp on Linux where it just works, and I can fix it (or pay someone to) when it doesn't.

    --
    Forget diamonds, copyright is forever.
    1. Re:NOT OUR JOB by Anonymous Coward · · Score: 0

      Newsflash, dipshit: Corel isn't a Microsoft product.

      Want to know of another product that doesn't work properly without Admin priveleges? It's Thunderbird. Enjoy choking on your own self-righteous bullshit.

    2. Re:NOT OUR JOB by randomencounter · · Score: 1
      So? If Thunderbird doesn't work without admin priviledges under Windows then maybe it can't.

      Which makes it who's fault again?

      Enjoy choking on your own smug sanctimonious bullshit, coward.

      --
      Forget diamonds, copyright is forever.
    3. Re:NOT OUR JOB by Anonymous Coward · · Score: 0

      Gee, Adobe Photoshop runs without Admin priveleges. Outlook Express runs without Admin priveleges. Opera, Putty, Winamp, hell even Virtual PC and Bochs run without Admin priveleges.

      But yeah, it's Microsoft's fault that Corel and Thunderbird are written poorly. Try paying attention and learn a few things before you blame Microsoft in your ignorance.

      Oh, and I'm a coward? For not creating a bullshit pseudonym? You really are clueless.

  270. Installation != Common Use by Dimensio · · Score: 1

    Installing software more often than not alters the Windows OS configuration by way of registry writes (and changes to Program Files). Requiring Administrator rights for that is perfectly reasonable and often preferrable. However, once installed, software like Firefox should not require Administrative rights to run.

    So, to recap, software installers make fundamental OS configuration changes and thus are fine when restricted to Admin users. The software that they install, however, should be accessable by anyone unless specifically denied by the system Admin unless the software is specifically designed to alter hardware or OS config.

  271. Street cred by billcopc · · Score: 1

    I would like to bring attention to this man. Michael Robertson was the founder of MP3.com, and consequently is the man who single-handledly drove it into the ground and sold out to Universal. This guy is a joke, plain and simple. A stubborn fool who is desperate to sell another big success like MP3.com was in its prime.

    Leave him and his Linspire alone. He doesn't "get it", he's just good at telling outrageous stories in an attempt to gain mindshare.

    --
    -Billco, Fnarg.com
  272. Nothing to add... by alfino · · Score: 1

    The thread covers most of the reasons why working as root is just bad and dangerous. The two most important for me are typos on /bin/rm calls and unaudited software.

    So far I didn't know what to think of Robertson. Now I am confident he's an idiot.

    --
    echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck
  273. Are you really as dumb as you sound? by Some+Random+Username · · Score: 1

    The debate at hand is wether to have them always being root for everything all the time without even logging in, or being a normal user and have apps that need root privileges run sudo in the background. Of course sudo lets you be root, for single commands as needed. Beats the fuck out of leaving these inexperienced users just running as root all the time doesn't it? You aren't just wildly off track, you are completely and hopelessly lost.

  274. Root terminal by PCM2 · · Score: 1

    I hear Ubuntu praised a lot for its approach, but it doesn't really do much to discourage users from running shell commands as root. Every user has access to the "Root Terminal" application, right in the Gnome menus. All it takes is your password (that's your own password, assuming you were the default user created during install, because that user is automatically placed on the sudo list) and you're free to do whatever you want as root. Get into lazy habits (i.e. choosing "Root Terminal" instead of the regular terminal when you want to do system maintenance tasks that might require root privileges) and the supposed no-root-account protections of Ubuntu are out the window.

    --
    Breakfast served all day!
  275. /. once again proving... by arete · · Score: 1

    /., once again proving the hypothesis that the best way to get a lot of information is to say something wrong on the internet.

    Thanks, all : )

    --
    Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
  276. Re:only multiuser! by malfunct · · Score: 1
    This doesn't mean that a script from the web won't do it for them. If they don't know they need the permission, often they don't need the permission and shouldn't have it.

    Case and point, I set my wife as a lua user on our windows xp box at home and she still hasn't noticed any difference.

    --

    "You can now flame me, I am full of love,"

  277. I did it by spitzak · · Score: 1

    On a NeXT machine. The NeXT desktop was dropping .desktop and other hidden files and directories all over the place and it had left some in root that I wanted to get rid of. Since they were directories and there was nothing else starting with dot, I typed "rm -R /.*". Stopped it after I noticed it was still churning after a minute, but by then the system was trashed.

  278. Re:uh... by fvwmfan · · Score: 1

    The answer is that
    1. hitting Ctrl-C stops it, but what is gone is gone (I don't know about restoring from inodes or whatever), and
    2. how much you loose, and what gets lost depends on how quickly you realise, and what you have placed in that directory.

    The problem is that you are never really sure what got lost! What did I have in that root directory? Didn't I put something important there 3 months ago when I was doing ... ? Luckily, I kept backups :( .

    I have done this. It is all too easy, even if you are consistently careful.