Clearly this is a security vulnerability, and why the Guest account ships in the disabled state. It would be very nice if Windows would warn you when you enabled it, and made an attempt to explain the implications of doing so.
Why have the guest account at all?
When I inherited an existing Exchange 5.x server at my last contract, I was totally unaware of the guest account. A few months into the contract, I found "guest" on a security audit and still couldn't figure out why in the world this would be a good thing -- even if you used a relay to handle external email and took Exchange out of the external loop, it just seemed like a really bad idea.
Are people this lazy that an account like this is practical? That's the only reason I see for it at all.
I did read the article and am fully aware of it's implications. However... SHUT UP... I'm trying to get them to upgrade!:) SHHHH
No kidding. As a former Exchange admin, POP/SMTP/... support -- or at a bare minimum an upgrade to Exchange 2000 -- is exactly what I do want so I can stop using that damn Outlook Web Access (OWA).
I've asked multiple times if they have plans for any upgrade -- I've sent links to alternatives, asked if Exchange 2000 was planned -- and get no response from corporate except "only the Outlook desktop client is supported". (Exchange 2000 is supported by Ximian's plugin, though Exchange 5.x is not.)
I want to use Evolution where all my other mail is, and not muck around with file format converters and OWA is a real weak client app.
The workarounds -- file converters and exporters -- only help with scraping out what is on the servers and don't help with making it dynamic.
Even using Outlook under Wine -- something I've not tried -- would still be second rate. No vFolders; why bother? It's such a pain to drag and drop mail between folders and filters only move, delete, or duplicate -- causing a long term mess.
That means that everyone dealing in leads makes less money, but the spammers make more. That would squeeze everyone, until the only ones making money in mortgages are spammers. This would result in rich spammers, plowing more money into spam.
The lead business is much less efficient than you think, with hundreds/thousands of buyers and sellers, so if one company dumps the lead broker, another one will pick up their leads. The leads are mostly unpriced, and buyers are chasing lead sources.
If that's the case...why not just generate false leads and skip the middle spamming step?
You mean phone book as in the phone books that have huge-ass ads all over the insides of them?
Yes. Adds aren't bad, as you likely agree (?), it's the positioning of them that is drastically different when compared between phone books and newspapers vs. MSN search. (also agree?)
MSN sucks. Don't use it. Wow, problem solved.
This is contrary to the 95,000,000 search results for linux that are reported in the article. What's the deal? Maybe an ISP cache of google that's outdated?
I get 95M total, though even Google limits it to 44 pages of results (keep clicking on the last page shown). The total: 866 sites shown. The rest are marked as likely duplicates.
Well, this isn't for supercomputers, though for clusters I did find that about 90% ran some form of Linux, with about 3-4 each for Windows NT, Solaris, BSD, Tru64, and Other.
So, if I understand you - which I probably don't - Microsoft is a Bad, Evil Thing for being biased.... in favor of it's own products? So... they own a search engine and that bias shows.
Think "newspaper" or even "phone book" and go through that thought process again.
Strange. I certainly don't want to defend MSN search, but it seems to intentionally limit the response set for common words. Hence, low numbers of responses for "linux", "apache", and "windows". But if you search for "apache server", there's about 5 million responses on MSN search.
It's the helpfulness of Clippy... without Clippy! Egads, they *are* smart!
The cry of the buggy whip manufacturer...with the gaul to make it look noble or even patriotic.
It annoys me that Linux developers try to compete with commercial companies.
Most programs are customized for a specific set of users, not off the shelf. Open Source fits right in with that and helps *MY* business.
If you don't want competition, then don't ignore those markets in the first place! The vast majority of projects I see are started because there *isn't* a commercial company in that segment or the existing one is unresponsive or even abusive.
I propose the following "constitution" for open source developers:
1) I will freely license my code for use in commercial products (ie, use BSD license not GPL.)
Is it your code?
2) If a commercial equivalent exists for what I'm developing, I will not try to market it as a replacement for the commercial product.
See earlier comment on commercial companies.
3) My software will not be targeted at the average consumer (read: no easy to use UI, no easy installation process).
Excuse me? This is the biggest complaint people have about OSS, and when it delivers...you don't like it?
So basically, you can develop research software, specialized software, etc. But please, if no-one buys MS Office and d/ls openoffice instead, innovation in word processors and spreadsheets will stop. We do not want this.
Knoppix is a good starter distribution. Since it boots and runs entirely off of the CD, it is low fuss and can be used as a crude way to see if your hardware is supported.
Even if you don't 'switch', it has saved me and others quite a bit of time in emergencies or diagnosing hardware problems. (Simple example: Mouse stops working under Windows. Boot Knoppix. If the mouse works, it's Windows...if the mouse fails, it's the mouse.)
(SNIP! from previous post on/.);
[Knoppix] requires no commitment, and can be handy for fixing problems and testing out machines even if you don't use it as a Linux desktop. The 11/03/2003 release is a bleeding edge release and has OpenOffice 1.1, though the 09/24/2003 release should be more stable.
Knoppix is only slightly lightweight, and you don't compromise much by using it except for the fact it's a boot CD and isn't as snappy as a hard drive based one. You can even test out your TV card and other hardware if you want. 256MB RAM is not required but is a realistic minimal amount for full use of what's on the disk.
I prefer to go through the difficult installation process Debian is known for - I know what hardware I have and can update drivers in the kernel if necessary, manually.... The disadvantage with installer is that users generally become lazy because of the very nature of an installer.
Lazy, hell. You don't really believe this, do you?
I don't go to every machine I manage, I use shell scripts. When the machine boots, init configures the system. Hardware configuration is part of the entire scheme. If it fails, the user (not an admin) should then get someone else to fix it as it's not thier job to know how. If the hardware configuration software is worth it, there should be few situations where it does indeed fail. Kudzu (Red Hat's) is damn good. If the Debian folks want to reinvent the wheel, they can.
Getting the proper modules loaded automatically is exactly the kind of task that software does well. Looking up hardware details and slogging through kernel notes is an entirely automatable process...and automation is why we have computers in the first place.
I used to fiddle around with modules every time I upgraded the kernel -- either from source or from a new distribution. Kudzu (also used in Knoppix BTW) does an amazing job of auto configuration...so why not use it or something like it?
Would you use a boot CD like Knoppix if you had to configure the modules and other drivers each time you went to a new machine? It would take the joy out of it, making you do the work a computer is entirely capable of doing.
It doesn't make you any less special that the system figures out something that you also can figure out. Yes, experts should know how the system works. Tinker with modules.conf if you like. I personally would like to fiddle with other things beyond the base hardware configuration since I already know how it works.
That said, if you're a professional let me put in a plug for InstallBase. This is a TK-based, cross-platform installation program; Solaris, Windows, and Linux. It provides a good balence between simple and detailed configuration, as well as a silent mode. Currently, I'm using it to bring sanity and automation to a mismanaged network.
Here's something you likely agree with. The network management document I'm writing says -- up front -- installation is not running an install program. I'm a strong believer that If you don't know what the answer should be, using a computer to tell you is an act of trust in something that has proven itself untrustworthy; it is foolish.
InstallBase (the tool) is used becuase it meets the goal of automation, though to use it or any other tool properly you have to know exactly what it is you want it to do. That takes concerned effort. The result eliminates needless work and inconsistant human mistakes that happen when each machine is managed a little differently. (If done wrong, you get consistant mistakes...so, there you go!)
We can generate more than enough wealth to keep everyone happy, and I don't buy that nonsense that if people aren't paid they won't work (explain OSS to me then).
Currently -- not in Star Trek fantasy land -- there are enough resources to make every person on the Earth reasonably comfortable and well fed. Yet, there are starving and homeless people. That's a fact.
Do you feed everyone you are aware of who is presently starving? Send them clothes, shelter? Are you unable to? Are you going to sell your computer and keep to the bare necessities to help others starting today? No more gadgets, no new car, no fancy house.
Collectively -- through government taxes or private institutions -- people with resources could handle the necessities and then some for those without...yet, this does not happen.
That's a fact, not a theory.
OSS is a special case; once something has been created, the distribution effort is all that remains. Neither are necessarily trivial, though you don't have to burn CDs and put them in boxes every time someone asks for the software.
If you want to keep up to date on the conference and other happenings, but don't have time to attend or even slog through the mailing list(s), this is a good place to go.
Over the past few days, each summary has been updated, so if you haven't visited here before, stop reading Slashdot and go there!
I use to be all for Linux conquering the OS market but as time went on I came to understand that there's some desktop real estate that Linux shouldn't want to own. That segment is the home user that *thinks* they know what they're doing but really only know just enough to make a mess.
Exactly. A little bit of knowledge -- and the arogance that comes with it -- is a very dangerous thing. These people are alies I don't want.
Hope you find these comments contructive - they are not meant to assault.
You also have to make it painless to do things like install/remove software and install/remove drivers.
To install software, click on the package you want to install or drop in a CD built for your distribution.
Remove/install drivers: Why bother now? This was a problem a few years ago, but not now. If you have the hardware, it'll be identified. If not, the module (aka 'driver') just won't load.
Nvidia has an installer that works well enough. Use it.
Root/user seperation and the pain of logging in: The password tool in Red Hat / Fedora works well enough; login 1x, and for a few minutes have full root access from the desktop. I'd be surprised if other distributions don't have this feature, though I haven't looked over the past year.
Odd crashes: Agreed. Specific programs bork for no good reason on occasion.
Polish: Agreed, though except for OSX, there seems to be quite a lack of polish for nearly every OS out there.
User adoption: If the OS is installed, people will use it. If not, they won't all the sudden decide to switch unless they highly motivated and willing to commit to deal with any issues that come up. My little sister had no problem, and she's at best "challenged".
Tech widgets: Yep, though they don't deal with those things elsewhere so there's no motivation (or reason) to do it under an X desktop.
The business people don't make choices like that. They make the decision that they want a free OS installed on the computers, then hire (or outsource) someone that reads/. to choose the best distro for their operation.
Oh, to live in your world! (No, really, where is this place? Tell me! Please!:( )
I'm constantly told exactly what software to use by managers who like to micromanage. Only seeing a nitch, filling it, and then announcing that the problem/issue is solved do I get to have any choice.
And, unfortunately for the proponents of a lot of the distros out there, one of the main features the technical people look for is ease-of-use... and that's the area that Linux fails WAY behind Windows.
Some distros do make it a bit too difficult, and there are annoying inconsistancies. (Current gripe: file managers in KDE/Gnome can't share, CLI over the network does not match what you see in the GUI, though the *exact* same set of problems exists in Windows.)
I find familiarity and fear are the biggest obsticals. Ease of use, while critical, is way down the list. It doesn't matter how easy something is to use if you never look at it or make it more complex than it is.
To be any good at using Windows or a flavor of Unix you have to put in substantial amounts of work. Everything else is just clicking on a dialog and guessing. In the grand scheme of things, software is really just an arrangement of bits and bytes -- Windows, prop-Unix, or open-Unix -- and one package doesn't really make a difference. Run Windows? Know the registry. Run Unix? Know basic scripting and the file system.
I'm not an admin right now, though the chief admin where I am is starting to get a clue that CLI != evil, and that it has it's place just as GUIs do. I've been carefully working on him for the past 6+ months, and he's reciently warmed to my advice realizing that I'm attempting to help not muck with his network or do an end run around him.
That is one of the killer components that is missing. Get people to use it at home and the office will stand a better chance.
Check this thread for some comments on this subject.
Bottom line: Linux has tons of games, though there will always be that one must have game that doesn't work on whatever platform you choose (example: a game that is exclusive to a specific console).
in fact, most "Linux" hackers/users have Windows installed on some partition or another system for gaming.
I feel special...I don't!
Linux will not take over, it will be what it has always been, and also-run, just like every other OS. It's now an additional choice. No OS will ever dominate.
I thought the same thing about Windows 1.24. Funny how things turn out, eh? I hope that people will not think "Windows = computer" though that will probably go on for most folks for a few years.
That said, bits are bits. As long as computers are changing, the interfaces will change...so folks may as well get used to the fact that they won't look and have the same menus from year to year...no matter what OS you're using.
Slashdot Mirror
Humans did solve the puzzle...we just used a better tool.
We will soon be replacing our market analysits and physicians with programmers!
Maybe market anaylists and physicians will learn more programming?
Why have the guest account at all?
When I inherited an existing Exchange 5.x server at my last contract, I was totally unaware of the guest account. A few months into the contract, I found "guest" on a security audit and still couldn't figure out why in the world this would be a good thing -- even if you used a relay to handle external email and took Exchange out of the external loop, it just seemed like a really bad idea.
Are people this lazy that an account like this is practical? That's the only reason I see for it at all.
No kidding. As a former Exchange admin, POP/SMTP/... support -- or at a bare minimum an upgrade to Exchange 2000 -- is exactly what I do want so I can stop using that damn Outlook Web Access (OWA).
I've asked multiple times if they have plans for any upgrade -- I've sent links to alternatives, asked if Exchange 2000 was planned -- and get no response from corporate except "only the Outlook desktop client is supported". (Exchange 2000 is supported by Ximian's plugin, though Exchange 5.x is not.)
I want to use Evolution where all my other mail is, and not muck around with file format converters and OWA is a real weak client app.
The workarounds -- file converters and exporters -- only help with scraping out what is on the servers and don't help with making it dynamic.
Even using Outlook under Wine -- something I've not tried -- would still be second rate. No vFolders; why bother? It's such a pain to drag and drop mail between folders and filters only move, delete, or duplicate -- causing a long term mess.
The lead business is much less efficient than you think, with hundreds/thousands of buyers and sellers, so if one company dumps the lead broker, another one will pick up their leads. The leads are mostly unpriced, and buyers are chasing lead sources.
If that's the case...why not just generate false leads and skip the middle spamming step?
Yes. Adds aren't bad, as you likely agree (?), it's the positioning of them that is drastically different when compared between phone books and newspapers vs. MSN search. (also agree?) MSN sucks. Don't use it. Wow, problem solved.
Not a problem for me.
I get 95M total, though even Google limits it to 44 pages of results (keep clicking on the last page shown). The total: 866 sites shown. The rest are marked as likely duplicates.
Here is the query I ran (top 256 clusters).
Think "newspaper" or even "phone book" and go through that thought process again.
It's the helpfulness of Clippy ... without Clippy! Egads, they *are* smart!
Well, to get the ball rolling, here is a query on the top 500 supercomputers using Microsoft Windows. Corrections and insight are appreciated.
That said, for what is provided, the Earth Simulator seems to be the current king by about 2x. (Corrections appreciated.)
Why did I have a flashback to Joe Isuzu?
Most programs are customized for a specific set of users, not off the shelf. Open Source fits right in with that and helps *MY* business.
If you don't want competition, then don't ignore those markets in the first place! The vast majority of projects I see are started because there *isn't* a commercial company in that segment or the existing one is unresponsive or even abusive.
1) I will freely license my code for use in commercial products (ie, use BSD license not GPL.)
Is it your code?
See earlier comment on commercial companies.
Excuse me? This is the biggest complaint people have about OSS, and when it delivers...you don't like it?
You had me going there...good troll.
Even if you don't 'switch', it has saved me and others quite a bit of time in emergencies or diagnosing hardware problems. (Simple example: Mouse stops working under Windows. Boot Knoppix. If the mouse works, it's Windows...if the mouse fails, it's the mouse.)
(SNIP! from previous post on /.);
[Knoppix] requires no commitment, and can be handy for fixing problems and testing out machines even if you don't use it as a Linux desktop. The 11/03/2003 release is a bleeding edge release and has OpenOffice 1.1, though the 09/24/2003 release should be more stable.
Knoppix is only slightly lightweight, and you don't compromise much by using it except for the fact it's a boot CD and isn't as snappy as a hard drive based one. You can even test out your TV card and other hardware if you want. 256MB RAM is not required but is a realistic minimal amount for full use of what's on the disk.
Tip: If you don't get a desktop, or the monitor doesn't sync up properly, use the option screen=800x600;
More details here if you want to tweak it more. It supports USB and floppy drives to store desktop information and program settings.
Lazy, hell. You don't really believe this, do you?
I don't go to every machine I manage, I use shell scripts. When the machine boots, init configures the system. Hardware configuration is part of the entire scheme. If it fails, the user (not an admin) should then get someone else to fix it as it's not thier job to know how. If the hardware configuration software is worth it, there should be few situations where it does indeed fail. Kudzu (Red Hat's) is damn good. If the Debian folks want to reinvent the wheel, they can.
Getting the proper modules loaded automatically is exactly the kind of task that software does well. Looking up hardware details and slogging through kernel notes is an entirely automatable process...and automation is why we have computers in the first place.
I used to fiddle around with modules every time I upgraded the kernel -- either from source or from a new distribution. Kudzu (also used in Knoppix BTW) does an amazing job of auto configuration...so why not use it or something like it?
It doesn't make you any less special that the system figures out something that you also can figure out. Yes, experts should know how the system works. Tinker with modules.conf if you like. I personally would like to fiddle with other things beyond the base hardware configuration since I already know how it works.
That said, if you're a professional let me put in a plug for InstallBase. This is a TK-based, cross-platform installation program; Solaris, Windows, and Linux. It provides a good balence between simple and detailed configuration, as well as a silent mode. Currently, I'm using it to bring sanity and automation to a mismanaged network.
Here's something you likely agree with. The network management document I'm writing says -- up front -- installation is not running an install program. I'm a strong believer that If you don't know what the answer should be, using a computer to tell you is an act of trust in something that has proven itself untrustworthy; it is foolish.
InstallBase (the tool) is used becuase it meets the goal of automation, though to use it or any other tool properly you have to know exactly what it is you want it to do. That takes concerned effort. The result eliminates needless work and inconsistant human mistakes that happen when each machine is managed a little differently. (If done wrong, you get consistant mistakes...so, there you go!)
Currently -- not in Star Trek fantasy land -- there are enough resources to make every person on the Earth reasonably comfortable and well fed. Yet, there are starving and homeless people. That's a fact.
Do you feed everyone you are aware of who is presently starving? Send them clothes, shelter? Are you unable to? Are you going to sell your computer and keep to the bare necessities to help others starting today? No more gadgets, no new car, no fancy house.
Collectively -- through government taxes or private institutions -- people with resources could handle the necessities and then some for those without...yet, this does not happen.
That's a fact, not a theory.
OSS is a special case; once something has been created, the distribution effort is all that remains. Neither are necessarily trivial, though you don't have to burn CDs and put them in boxes every time someone asks for the software.
Also, OSS is a fact, not a theory.
That is odd...though I regularly get modded down too for no good reason. (tossing out +1 Karma Bonus)
If you want to keep up to date on the conference and other happenings, but don't have time to attend or even slog through the mailing list(s), this is a good place to go.
Over the past few days, each summary has been updated, so if you haven't visited here before, stop reading Slashdot and go there!
Linux Kernel
Wine
GNU Enterprise
KDE
Exactly. A little bit of knowledge -- and the arogance that comes with it -- is a very dangerous thing. These people are alies I don't want.
You also have to make it painless to do things like install/remove software and install/remove drivers.
To install software, click on the package you want to install or drop in a CD built for your distribution.
Remove/install drivers: Why bother now? This was a problem a few years ago, but not now. If you have the hardware, it'll be identified. If not, the module (aka 'driver') just won't load.
Nvidia has an installer that works well enough. Use it.
Root/user seperation and the pain of logging in: The password tool in Red Hat / Fedora works well enough; login 1x, and for a few minutes have full root access from the desktop. I'd be surprised if other distributions don't have this feature, though I haven't looked over the past year.
Odd crashes: Agreed. Specific programs bork for no good reason on occasion.
Polish: Agreed, though except for OSX, there seems to be quite a lack of polish for nearly every OS out there.
User adoption: If the OS is installed, people will use it. If not, they won't all the sudden decide to switch unless they highly motivated and willing to commit to deal with any issues that come up. My little sister had no problem, and she's at best "challenged".
Tech widgets: Yep, though they don't deal with those things elsewhere so there's no motivation (or reason) to do it under an X desktop.
Oh, to live in your world! (No, really, where is this place? Tell me! Please! :( )
I'm constantly told exactly what software to use by managers who like to micromanage. Only seeing a nitch, filling it, and then announcing that the problem/issue is solved do I get to have any choice.
And, unfortunately for the proponents of a lot of the distros out there, one of the main features the technical people look for is ease-of-use... and that's the area that Linux fails WAY behind Windows.
Some distros do make it a bit too difficult, and there are annoying inconsistancies. (Current gripe: file managers in KDE/Gnome can't share, CLI over the network does not match what you see in the GUI, though the *exact* same set of problems exists in Windows.)
I find familiarity and fear are the biggest obsticals. Ease of use, while critical, is way down the list. It doesn't matter how easy something is to use if you never look at it or make it more complex than it is.
To be any good at using Windows or a flavor of Unix you have to put in substantial amounts of work. Everything else is just clicking on a dialog and guessing. In the grand scheme of things, software is really just an arrangement of bits and bytes -- Windows, prop-Unix, or open-Unix -- and one package doesn't really make a difference. Run Windows? Know the registry. Run Unix? Know basic scripting and the file system.
I'm not an admin right now, though the chief admin where I am is starting to get a clue that CLI != evil, and that it has it's place just as GUIs do. I've been carefully working on him for the past 6+ months, and he's reciently warmed to my advice realizing that I'm attempting to help not muck with his network or do an end run around him.
Yep, it wasn't clear. Sorry. I should have said "I run Linux games and don't dual boot into Windows."
As for your other points, I didn't forget them and mostly agree.
Check this thread for some comments on this subject.
I feel special...I don't!
Linux will not take over, it will be what it has always been, and also-run, just like every other OS. It's now an additional choice. No OS will ever dominate.
I thought the same thing about Windows 1.24. Funny how things turn out, eh? I hope that people will not think "Windows = computer" though that will probably go on for most folks for a few years.
That said, bits are bits. As long as computers are changing, the interfaces will change...so folks may as well get used to the fact that they won't look and have the same menus from year to year...no matter what OS you're using.