I'd recommend PHP as an introductory language. It's freely available, is easily installable if you don't already have it (for Linux and Windows, and perhaps the Mac), gives the newbie immediate feedback, on Linux has a command line interpreter that doesn't require a web server (or anything else for that matter), has very simple syntax, and is an exceptionally easy language to grasp. It's the next best thing to the 80s when all home computers came with BASIC in the ROM.
"There's no way that the Linux kernel isn't in public domain."
Linux is not in the public domain. It is fully copyrighted and licensed under the terms of the GPL. That is in no way similar to being in the public domain. The GPL grants you the right to copy, modify, and distribute Linux provided you abide by the terms of the license. It's just like any other copyrighted software, except the license is very lenient and attractive to outside developers and to users.
"It was fine for a while, btu at some point someone told me I had to write it in cursive."
According to my business law instructor, your signature is any written pattern that you use to identify yourself. It can be as long or as short as you wish, but short and common written patterns are easier forge, hence you're advised to make your signature long enough to make illicit reproduction difficult. I suppose the rationale behind it is that long sequences of writing will probably provide enough distinctive character of the writer to make even good forgeries relatively easy to catch. Short patterns do not provide enough entropy for reliable comparisons.
Like many other people here, I use cursive now primarily for my signature (I'm 33). Even when I was compelled in my youth and to cursive extensively, my fine motor control was severely lacking. Hence my handwriting sucked the big cursive loop. Even my printing is somewhat messy and incongruous. I type everything. We're moving towards technology that will probably allow all manual input to be digital, and I won't becry the day when handwriting is relegated to historical enthusiasts.
It's completely irrelevent whether SCO put its own GPL notice on the code it contributed to the kernel. The original authors of the kernel code did put a GPL notice in the code, and SCO contributed additions to the kernel which SCO then proceeded to distribute. This clearly places SCO's kernel contributions under the GPL.
SCO has absolutely no valid argument. SCO willingly and repeatedly distributed its source code contributions under the terms of the GPL.
"This was true in the past, but today all nations that follow the Berne copyright convention everything created after April 1, 1989 is considered copyrighted (GPL or otherwise) whether it has a notice or not. "
That's not what the laughable dummies at SCO are arguing. They are now trying to say that the SCO employees who were selling SCO Linux, with the blessings of SCO top leadership, did not own the copyrights on the source code that SCO was distributing and therefore the GPL does not apply.
It's a totally brain-damaged argument, and I would be astounded to read it from anyone other than SCO. The argument is that employees do not have the right to sell and/or distribute their company's products even when the highest level executives, whose primary job is to authorize the sale and/or distribution of the company's products, demand that it be done as part of the company's core business!
So SCO's top bosses passed down the mandate to sell copies of SCO Linux, but the employees should not have done it because they don't personally own the copyrights on SCO's code.
This gets funnier everytime SCO tries to fumble for a defense.
Re:forging sender address
on
I, Spammer
·
· Score: 1
"BTW, I've always thought it funny that/. folks are so against spam, yet they're all for anonymity on the net. Weird."
How do you equate a desire to not have your personal life broadcast to the world with a desire to get bombarded with unwanted, bandwidth stealing, trash?
"But the ASP never copies the program to your computer's memory. The program sits on the server, does its thing, and the output is the only thing that gets transfered across the 'net."
Oops, you're right. I had forgotten that ASPs aren't just file servers.:)
I said the Walmart machines wouldn't run anything more than Win98.
When I last researched the Walmart machines, they were way below 1.2GH, 128MB, and 10GB for the $200 price tag. I don't remember the exact specs, but it was something like 500Mhz Via processor, 64MB RAM, 10GB hard drive.
If your given specifications are correct, then Walmart has greatly improved its offering. On the other hand, you won't be getting those specs for $200, Walmart or no Walmart. The motherboard+processor alone will run you about that much.
"Has it been revealed to the world yet exactly which code is in question?"
No, and for two primary reasons:
1) SCO's bluffing until the bitter end. There is no copyright infringement, and everyone in high level positions at SCO knows it. SCO's market exit strategy (to be bought by IBM) backfired in a huge way, and there is no way to repair the damage. If SCO backs off now, it will be destroyed by the SEC for gross negligence towards the shareholders. Everyone at SCO knows that IBM is going to destroy SCO, and are merely trying to hold off the inevitable for as long as possible.
2) There is no copyright infringement. When the trial starts (and it's a given that this will get to trial considering #1 above), SCO will have to produce something (and I guarantee it will be manufactured ala Microsoft). At that point, the jig will be up. SCO doesn't actually want to go to trial, but now the company has painted itself into a corner and will -have- to go to trial. This is a great big "OOPS!" on SCO's part.
"And what, exactly, happened to their statement that they weren't going after Redhat or Joe Linuxuser, but instead just IBM?"
SCO realized that IBM wasn't biting, and panicked. This is SCO's flailing around for anything and everything to halt its inevitable destruction.
I don't have to agree with everything (though I agree with most of it) to admire the man's vision and coherency. His answers were so clear and to the point, I can't believe they were anything but sincere.
"I'm not so sure that the lindows/microtel boxes are anything but cheap white boxes that people install their pirated copies of windows software on."
I doubt it. The Microtel/LindowsOS boxes are not powerful enough to run anything beyond Win98 comfortably (and I'm skeptical they'll run Win98 comfortably). And they are not powerful enough to play most of the Windows games -- most certainly not most 3D games.
These Walmart/LindowsOS machines are very entry level machines designed to run just the basic apps (email, web, word processing). The people who buy these machines are not likely to know how to install an operating system. Even if they know people who can, it's unlikely to happen simple because the machines are quite underpowered.
"Linux often requires higher permissions than it should, because it is coming from the perspective of a shared, multiuser environment, or it does not give the console user extra priviledges."
KDE 3.1 handles this nicely. For every given function that requires root privs, the user is required to enter the root password the first time the privs are required. The password window has a checkbox that allows KDE to remember the password for that given function. So if the user decides that requiring the root password for changing the clock is going overboard, he can set KDE to not require the password in the future.
"And, as you say, there's no need for root access for an email virus to spread."
So many people who use this argument are overlooking the obvious. An email virus requires an email program that will automatically execute attachments. There is no email program for Linux that does this. All Linux email programs require the user to:
1) Save the attachment to disk. 2) Change the file attribute(s) to executable (because Linux email programs never save files with the execute bit(s) set). 3) Expicitly run the program.
The reason Microsoft email viruses are so prevelant is because Microsoft's email program requires the user to:
1) Read the email. 2) There is no second step.
Running the system as root will in no way make a user susceptible to email viruses on Linux. However, running as root will greatly increase the damage done to the system if a user runs a malicious program.
Assume that I, under my user account "stormreaver", get tricked into performing the three steps needed to run a malicious binary emailed to me and it completely wipes out my home directory.
If I didn't have backups, which is foolish considering how cheap CD-R's are nowadays and how easy KDE 3.1 makes backups, it's true that I'm out of luck for my personal files. I have to remove my account (by becoming root), recreate my account, and start over. The rest of my system (games, utilities, etc) are all fine.
If I did have backups, which I have, then I delete my compromised account, recreate the user account, and copy my backed-up config files. It takes my (estimating) about ten minutes to get back to where I was (allowing time for reading hundreds of megabytes from the backup CDs). I'm then back in business. Since all my executables (minus my in-development binaries) were protected by root access, nothing was permenantly damaged. At most, I'm somewhat inconvenienced. I saved days of redownloading and installing large non-system apps and several hours of installing and tweaking my operating system (with the latter taking most of the time).
The superuser/regular-user distinction made in Linux is real protection against filesystem viruses because it eliminates the automatic propogation mechanism required by such viruses.
Only profound stupidity on the part of email software developers (and profound stupidity on the part of users running Outlook through WINE, though the threat is greatly reduced on Linux because most WINE installations don't have Microsoft address books) will present an email virus problem on Linux.
Running everything as root will still require active user participation to propogate any type of Linux virus. However, the damage done will be far more extensive.
None of this will matter, of course, if a regular user installs something by temporarily becoming root, and the install program then makes the software setuid root. However, the protections added by the superuser/regular-user distinction are significant.
KDE has almost nailed all circumstances where this is inconvenient and provides a simple way to temporarily become root to perform system-altering tasks. The only one remaining, that I encounter, is having to use the command prompt to su to root in order to install a game using Loki's installer.
"So let's say that this "leasing" is really taking place in the form of an Application Service Provider. Let's say then that (as per the GPL - flame me if I'm wrong, please) I modified OpenOffice or something and I'm offering it as an online service. But I'm not giving you the source. Technically I'm not engaging in redistribution of a modified version of something covered by the GPL, right?"
Recent court decisions have indicated that loading a program into memory creates a copy that is covered by copyright laws. Therefore I would think that transferring an in-memory version of the software as an ASP would constitute distribution covered by the terms of the GPL.
i dont know about you, but i don't like it when my ass engages in sexual activities..."
Everytime something "nerdy" appears on Slashdot, we get a whole chorus of people complaining that Slashdot users won't get laid. This happens every single time.
I can't help but think that this falls under the, "I think thou dost protest too much" category.
"Grokster does not operate a Supernode server or a server with IP addresses or any type of server that interfaces in any way with the operation of Grokster or FastTrack with the exception of ad serving via the Start page."
Do Grokster client have to route through the ad server before being allowed to operate normally? If so, then Grokster will lose the appeal as this is a central point of control. This will also demonstrate that, since the Grokster execs know that its service is being used to extensive copyright violations, the service operators can control who is allowed to use the service and eliminate those who violate copyrights.
If the clients are not required to pull ads for the normal course of operation, then the above doesn't apply. However (and I admittedly know nothing about Grokster), if Grokster is a for-profit company and makes money by selling ad space to users, and since the Grokster clients are closed source, I can't help but think that Grokster operates by requiring users to pull down ads from a central server in order to operate correctly.
Of course, the ad and spyware servers may not be hard coded in as a requirement for the P2P software to operate, and the P2P software may merely assume that the ad and spyware servers will not be firewalled off or otherwise blocked. If the operation of the software depends upon the ad and spyware servers being operational, then Grokster has already lost the appeal and will be guilty of contributory infringement just like Napster.
"so you get 4 and a half years to use the cartidge after you buy the thing."
No, you get 4.5 years from the date the cartridge was manufactured or 30 months (2.5 years) from the date the cartridge was inserted into the printer (whichever comes first). Since you can always remove and reinsert cartridges, you effective have 4.5 years from the date the cartridge was manufactured.
The company in the story buys cartridges in bulk, which means that some of them could easily sit in storage for half that time. If the cartridge sat on the store shelves for a while in addition to the time it sits in storage at the company, it's conceivable that cartridges could still be usable 4.5 years after the manufacture date.
I, on the other hand, have never had a cartridge that still had liquid (as opposed to solid) ink after a mere one year of storage, much less 4.5 years.
However, it must be possible, if not extremely common, otherwise HPaq would never have bothered with an expiration date.
Even when I was first learning Linux all those years ago, the command names, when they weren't self-evident, were a non-issue. Finding out what they did was a simple matter of "man [command]".
Your argument would be more convincing if you'd focused on the command parameters. Now some of those can be damn-near incomprehensible.
"It is amazing how much training it takes in UNIX to do something simple in Windows. For example, recursively searching through a subtree for some text in a file."
Yeah, "grep -R [sometext] *" is soooo hard. Try using Windows to remotely search five different computers, each in a different building (or perhaps in a different city), and then tell me how much easier Windows makes it. I used to work for Western Resources in Topeka, Kansas just a few years ago. We would have to drive three or four hours (one way!) just to install some damn Windows program in one of our remote offices because Windows didn't have even the most basic capacity for remote management. Very few things are made easier in Windows than in Linux.
And have you ever heard of "books"? They're like oral wisdom, but they're, like, written down and stuff.
"VT100 pops up its ugly head decades after it should have been killed."
When we have to remotely administer a computer through a low link, we certainly don't want some way to make the screen easy to read. Oh no, we'd much rather just see endless blocks of monotonous text. Or better yet, we'd rather three minutes per frame for the remote GUI display to finish. That makes life sooooo much easier.
"Dang X sucks. Bad. What a mess! "Motif Self-Abuse Kit" made me laugh because my brief experience programming Motif was one of the worst in my life."
News flash: Motif is not the X Window System. Motif is a horrid and ugly layer that sits on top of the X Window system and provides the most horrible excuse for a visual interface that I've ever seen (with GTK trailing by a not so wide margin). If Motif if your main experience with X11, then it's no wonder you have such a rancid opinion of X. The X Window system itself, though, is genius.
"I've never found a make that I liked."
We're in complete agreement here. The make command, while doing a great job internally, is a great example of how to drive away perpective developers. We certainly need to create a much higher level make tool that encompasses and hides the common operations that developers perform when building the projects:
1) Compile those files that are out of date. The make process should be able to work this out easily. files need to be recompiled, etc.
2) Have a simple command language that performs the common pre and post make tasks such as building RPMs.
The make command is one of my few really hated implementations. It's truly painful to use.
He said: "We're still finding issues. But all of the newer code has got to be ten or a hundred times better."
What it really means: "We were lying all those years we told you that we created top notch code. Now that nobody believes us anymore, we're starting another round. No really! This time we really do have top notch code. Just trust us. We wouldn't lie about something this important."
He said: "The [applications] that people make themselves tend to be better than the larger, all-encompassing applications."
What it really means: "We were lying all those years we told you that tight integration makes for better applications. Software written by people outside Microsoft, who make one application and make it work well, outperform the all-in-one software produced at Microsoft."
He said: "Personally, I'm against shoving things into the kernel. That was a very careful decision. We have a lot of parsing in there, and that opens you up to buffer overruns and attacks."
What it really means: "We know that our kernel is already badly broken because we keep stuffing shit in there, and we know that stuffing yet more shit in there is a really, really, really bad idea, but we're going to do it anyway."
He said: "That's something Linux has that's ahead of us, but we're looking at it. We will have a command line-only version, but whether it'll have all the features in is another matter."
What it really means: "We were lying all those years that we told you how bad the command line was, and how much more reliable a GUI makes a server. The Linux guys have been ahead of us for years, but we're just now willing to admit it (since nobody believes us anymore)."
He said: "It looks really good on PowerPoint! Reality is never quite as good."
What it really means: "We're gonna send out our Marketroids to sell you on a whole new batch of lies and PowerPoint slides. Rest assured that the actual product will suck badly."
I'm not a big game player, so I'm not that upset with the limited game selection on Linux. However, I do like to play them sometimes. I bought the tin-box Linux version of Q3A, bought UT and downloaded the Linux installer, and bought RtCW for Linux.
Last year I bought a PS2 to satisfy my non-FPS gaming desires. Since then I've bought about eight games for it (GTA, GTA/VC, Spiderman, ATV Offroad-whatchamacallit, and a few others I can't recall right now). I mostly play them with several cheats enabled to bring them down to a level where they're fun. I'm not a gaming purist. I have games to pass some time and divert my attention for a while, not to forge a new lifestyle (which is why I won't buy or play NWN or participate in any MMORPG).
The point of all this? I'm more than willing to pay for quality games that I'll only play every now and then, and they don't necessarily have to be big blockbusters.
Two of the most enjoyable games I've been playing recently aren't even commercial. I've been playing FooBillard off and on for a few weeks, and I love the game (I'm not even a pool fan). Same thing with Kolf. They're quite simple (from a playability perspective) and a great way to recharge my brain after (or while) working.
High quality doesn't have to mean expensive or big name. Too many game designers are focusing heavily on visual effects and too little on re/playability. Castle Wolfenstein, Doom, and Quake were brilliant (I am quite impressed with Carmack and co.) in terms of both replayability (I still fire them up from time to time) and uniqueness.
But now everyone and his brother (and including Carmack and co.) keeps pumping these rehashes out. When I had first heard of RtCW, I couldn't wait to get my hands on it. After playing it for a while, though, it quickly dawned on me how repetitive and uninspired it was. It just just another Quake with different graphics and a new storyline. The gameplay was the same old stuff. Like Star Wars and Star Trek, it's time to let these franchises take a vacation.
I know that John, for example, is capable of great gameplay insights, and I'm still greatful that he is allowing for Linux versions of ID games (and releasing old games under the GPL!). But I just can't bring myself to buy yet another FPS or common rehash.
I remember from years ago, a company called MVP [Software | Games] made a very simplistic but incredibly fun game based on the 1991 gulf war. I had no money back then, but I would have bought the full version if I could have. Now that was different and fun. It was shareware, and would have been well worth the money.
Now that I've been gainfully employed for a few years, I have discretionary income that's looking for fresh outlets. If shareware game makers would use OpenGL/SDL and exercise creativity, I'd be happy to pay for their products. As it stands, games have become mostly monotonous and trite over the last couple years.
"I've never heard of VB or Office being considered part of the programming course on a university campus."
You've never been to Southwest Missouri State University, once ranked as the #1 CIS program in the country until the focus shifted to Microsoft vocational training. This illustrious career path split a brain-dead-easy 3-credits, one semester VB class into a 4-credits, 2 semesters, even-more-brain-dead assurance of computer illiteracy.
To add embarassment to insult, the AOS (Administrative Office Systems [or something like that]) associate degree program was folded into the bachelor of science CIS degree program. The result? A few extra classes on using Microsoft Office were added to the "required for graduation" list for incoming freshment (thankfully I had long since filed my degree program, so was not subject to the new requirements).
Unix systems were relegated to "this is what you need to type to make your Java applets readable by the instructor" prior to the AIX web server being replaced by NT (which, SURPRISE, quickly became extraordinarily unreliable). The rest of the campus quickly converted to all-Microsoft.
The end results (whether related or not, but I think they were) were that the SMSU CIS program completely dropped off the ranking charts around the country, and graduates frequently knew very little about computing in general and software development in particular (a primary focus of the CIS degree at SMSU).
I think I may have been in the last (or next to last) class that contained highly competent graduates. The vast majority of the graduates in my class that were highly competent were already (or nearly so) that way due to their own outside efforts (many times before even entering the University program).
I suspect that most CIS graduates at SMSU that known anything have gotten to that point on their own steam despite the best efforts of the university's administration.
Sometimes it's embarassing to admit that I got my degree there, but I point out to people that I was over halfway done before the school was bought by Microsoft and that most of what I learned was self-taught; the degree was just a formality I had to endure.
Recently on Slashdot, when I click on the links to story comments, I get taken to the comments and then immediately get directed to an ad page. This isn't a popup that I can filter out, but a forced redirect from Slashdot to the ad page. I have to click the back button to return to the comments page.
Is this a new (or old) Slashdot ad system? It is infuriatingly annoying.
"Can't ATI just release their drivers themselves?"
I was wondering the same thing. Are XFree86 drivers modular? If not, why not? And if not, then XFree86 is severely broken and badly needs to be redesigned to be modular.
If so, why does ATI wait for the core XFree86 team to (months and months later) even look at the possibility of including its drivers?
I'm not familiar with XFree86's internals, but my guess is the former (XFree's internal architecturer is broken). If it were the latter, then this whole thread would be brainlessly moot.
If a fork will improve the system, then I'm all for it.
Slashdot Mirror
I'd recommend PHP as an introductory language. It's freely available, is easily installable if you don't already have it (for Linux and Windows, and perhaps the Mac), gives the newbie immediate feedback, on Linux has a command line interpreter that doesn't require a web server (or anything else for that matter), has very simple syntax, and is an exceptionally easy language to grasp. It's the next best thing to the 80s when all home computers came with BASIC in the ROM.
"There's no way that the Linux kernel isn't in public domain."
Linux is not in the public domain. It is fully copyrighted and licensed under the terms of the GPL. That is in no way similar to being in the public domain. The GPL grants you the right to copy, modify, and distribute Linux provided you abide by the terms of the license. It's just like any other copyrighted software, except the license is very lenient and attractive to outside developers and to users.
"It was fine for a while, btu at some point someone told me I had to write it in cursive."
According to my business law instructor, your signature is any written pattern that you use to identify yourself. It can be as long or as short as you wish, but short and common written patterns are easier forge, hence you're advised to make your signature long enough to make illicit reproduction difficult. I suppose the rationale behind it is that long sequences of writing will probably provide enough distinctive character of the writer to make even good forgeries relatively easy to catch. Short patterns do not provide enough entropy for reliable comparisons.
Like many other people here, I use cursive now primarily for my signature (I'm 33). Even when I was compelled in my youth and to cursive extensively, my fine motor control was severely lacking. Hence my handwriting sucked the big cursive loop. Even my printing is somewhat messy and incongruous. I type everything. We're moving towards technology that will probably allow all manual input to be digital, and I won't becry the day when handwriting is relegated to historical enthusiasts.
"If you hire a dumpster, does that mean the stuff in the dumpster is PD?"
If you hire a dumpster for anything, it means you should probably stop licking those funny stamps.
"You misunderstand what is being said here."
It's completely irrelevent whether SCO put its own GPL notice on the code it contributed to the kernel. The original authors of the kernel code did put a GPL notice in the code, and SCO contributed additions to the kernel which SCO then proceeded to distribute. This clearly places SCO's kernel contributions under the GPL.
SCO has absolutely no valid argument. SCO willingly and repeatedly distributed its source code contributions under the terms of the GPL.
"This was true in the past, but today all nations that follow the Berne copyright convention everything created after April 1, 1989 is considered copyrighted (GPL or otherwise) whether it has a notice or not. "
That's not what the laughable dummies at SCO are arguing. They are now trying to say that the SCO employees who were selling SCO Linux, with the blessings of SCO top leadership, did not own the copyrights on the source code that SCO was distributing and therefore the GPL does not apply.
It's a totally brain-damaged argument, and I would be astounded to read it from anyone other than SCO. The argument is that employees do not have the right to sell and/or distribute their company's products even when the highest level executives, whose primary job is to authorize the sale and/or distribution of the company's products, demand that it be done as part of the company's core business!
So SCO's top bosses passed down the mandate to sell copies of SCO Linux, but the employees should not have done it because they don't personally own the copyrights on SCO's code.
This gets funnier everytime SCO tries to fumble for a defense.
"BTW, I've always thought it funny that /. folks are so against spam, yet they're all for anonymity on the net. Weird."
How do you equate a desire to not have your personal life broadcast to the world with a desire to get bombarded with unwanted, bandwidth stealing, trash?
"But the ASP never copies the program to your computer's memory. The program sits on the server, does its thing, and the output is the only thing that gets transfered across the 'net."
:)
Oops, you're right. I had forgotten that ASPs aren't just file servers.
"You don't think this will run Win 98?"
I said the Walmart machines wouldn't run anything more than Win98.
When I last researched the Walmart machines, they were way below 1.2GH, 128MB, and 10GB for the $200 price tag. I don't remember the exact specs, but it was something like 500Mhz Via processor, 64MB RAM, 10GB hard drive.
If your given specifications are correct, then Walmart has greatly improved its offering. On the other hand, you won't be getting those specs for $200, Walmart or no Walmart. The motherboard+processor alone will run you about that much.
"Has it been revealed to the world yet exactly which code is in question?"
No, and for two primary reasons:
1) SCO's bluffing until the bitter end. There is no copyright infringement, and everyone in high level positions at SCO knows it. SCO's market exit strategy (to be bought by IBM) backfired in a huge way, and there is no way to repair the damage. If SCO backs off now, it will be destroyed by the SEC for gross negligence towards the shareholders. Everyone at SCO knows that IBM is going to destroy SCO, and are merely trying to hold off the inevitable for as long as possible.
2) There is no copyright infringement. When the trial starts (and it's a given that this will get to trial considering #1 above), SCO will have to produce something (and I guarantee it will be manufactured ala Microsoft). At that point, the jig will be up. SCO doesn't actually want to go to trial, but now the company has painted itself into a corner and will -have- to go to trial. This is a great big "OOPS!" on SCO's part.
"And what, exactly, happened to their statement that they weren't going after Redhat or Joe Linuxuser, but instead just IBM?"
SCO realized that IBM wasn't biting, and panicked. This is SCO's flailing around for anything and everything to halt its inevitable destruction.
I don't have to agree with everything (though I agree with most of it) to admire the man's vision and coherency. His answers were so clear and to the point, I can't believe they were anything but sincere.
"I'm not so sure that the lindows/microtel boxes are anything but cheap white boxes that people install their pirated copies of windows software on."
I doubt it. The Microtel/LindowsOS boxes are not powerful enough to run anything beyond Win98 comfortably (and I'm skeptical they'll run Win98 comfortably). And they are not powerful enough to play most of the Windows games -- most certainly not most 3D games.
These Walmart/LindowsOS machines are very entry level machines designed to run just the basic apps (email, web, word processing). The people who buy these machines are not likely to know how to install an operating system. Even if they know people who can, it's unlikely to happen simple because the machines are quite underpowered.
"Linux often requires higher permissions than it should, because it is coming from the perspective of a shared, multiuser environment, or it does not give the console user extra priviledges."
KDE 3.1 handles this nicely. For every given function that requires root privs, the user is required to enter the root password the first time the privs are required. The password window has a checkbox that allows KDE to remember the password for that given function. So if the user decides that requiring the root password for changing the clock is going overboard, he can set KDE to not require the password in the future.
"And, as you say, there's no need for root access for an email virus to spread."
So many people who use this argument are overlooking the obvious. An email virus requires an email program that will automatically execute attachments. There is no email program for Linux that does this. All Linux email programs require the user to:
1) Save the attachment to disk.
2) Change the file attribute(s) to executable (because Linux email programs never save files with the execute bit(s) set).
3) Expicitly run the program.
The reason Microsoft email viruses are so prevelant is because Microsoft's email program requires the user to:
1) Read the email.
2) There is no second step.
Running the system as root will in no way make a user susceptible to email viruses on Linux. However, running as root will greatly increase the damage done to the system if a user runs a malicious program.
Assume that I, under my user account "stormreaver", get tricked into performing the three steps needed to run a malicious binary emailed to me and it completely wipes out my home directory.
If I didn't have backups, which is foolish considering how cheap CD-R's are nowadays and how easy KDE 3.1 makes backups, it's true that I'm out of luck for my personal files. I have to remove my account (by becoming root), recreate my account, and start over. The rest of my system (games, utilities, etc) are all fine.
If I did have backups, which I have, then I delete my compromised account, recreate the user account, and copy my backed-up config files. It takes my (estimating) about ten minutes to get back to where I was (allowing time for reading hundreds of megabytes from the backup CDs). I'm then back in business. Since all my executables (minus my in-development binaries) were protected by root access, nothing was permenantly damaged. At most, I'm somewhat inconvenienced. I saved days of redownloading and installing large non-system apps and several hours of installing and tweaking my operating system (with the latter taking most of the time).
The superuser/regular-user distinction made in Linux is real protection against filesystem viruses because it eliminates the automatic propogation mechanism required by such viruses.
Only profound stupidity on the part of email software developers (and profound stupidity on the part of users running Outlook through WINE, though the threat is greatly reduced on Linux because most WINE installations don't have Microsoft address books) will present an email virus problem on Linux.
Running everything as root will still require active user participation to propogate any type of Linux virus. However, the damage done will be far more extensive.
None of this will matter, of course, if a regular user installs something by temporarily becoming root, and the install program then makes the software setuid root. However, the protections added by the superuser/regular-user distinction are significant.
KDE has almost nailed all circumstances where this is inconvenient and provides a simple way to temporarily become root to perform system-altering tasks. The only one remaining, that I encounter, is having to use the command prompt to su to root in order to install a game using Loki's installer.
"So let's say that this "leasing" is really taking place in the form of an Application Service Provider. Let's say then that (as per the GPL - flame me if I'm wrong, please) I modified OpenOffice or something and I'm offering it as an online service. But I'm not giving you the source. Technically I'm not engaging in redistribution of a modified version of something covered by the GPL, right?"
Recent court decisions have indicated that loading a program into memory creates a copy that is covered by copyright laws. Therefore I would think that transferring an in-memory version of the software as an ASP would constitute distribution covered by the terms of the GPL.
"wont even get your ass laid
i dont know about you, but i don't like it when my ass engages in sexual activities..."
Everytime something "nerdy" appears on Slashdot, we get a whole chorus of people complaining that Slashdot users won't get laid. This happens every single time.
I can't help but think that this falls under the, "I think thou dost protest too much" category.
"Grokster does not operate a Supernode server or a server with IP addresses or any type of server that interfaces in any way with the operation of Grokster or FastTrack with the exception of ad serving via the Start page."
Do Grokster client have to route through the ad server before being allowed to operate normally? If so, then Grokster will lose the appeal as this is a central point of control. This will also demonstrate that, since the Grokster execs know that its service is being used to extensive copyright violations, the service operators can control who is allowed to use the service and eliminate those who violate copyrights.
If the clients are not required to pull ads for the normal course of operation, then the above doesn't apply. However (and I admittedly know nothing about Grokster), if Grokster is a for-profit company and makes money by selling ad space to users, and since the Grokster clients are closed source, I can't help but think that Grokster operates by requiring users to pull down ads from a central server in order to operate correctly.
Of course, the ad and spyware servers may not be hard coded in as a requirement for the P2P software to operate, and the P2P software may merely assume that the ad and spyware servers will not be firewalled off or otherwise blocked. If the operation of the software depends upon the ad and spyware servers being operational, then Grokster has already lost the appeal and will be guilty of contributory infringement just like Napster.
"so you get 4 and a half years to use the cartidge after you buy the thing."
No, you get 4.5 years from the date the cartridge was manufactured or 30 months (2.5 years) from the date the cartridge was inserted into the printer (whichever comes first). Since you can always remove and reinsert cartridges, you effective have 4.5 years from the date the cartridge was manufactured.
The company in the story buys cartridges in bulk, which means that some of them could easily sit in storage for half that time. If the cartridge sat on the store shelves for a while in addition to the time it sits in storage at the company, it's conceivable that cartridges could still be usable 4.5 years after the manufacture date.
I, on the other hand, have never had a cartridge that still had liquid (as opposed to solid) ink after a mere one year of storage, much less 4.5 years.
However, it must be possible, if not extremely common, otherwise HPaq would never have bothered with an expiration date.
"A) Cryptic Command Names. Still there in Linux"
Even when I was first learning Linux all those years ago, the command names, when they weren't self-evident, were a non-issue. Finding out what they did was a simple matter of "man [command]".
Your argument would be more convincing if you'd focused on the command parameters. Now some of those can be damn-near incomprehensible.
"It is amazing how much training it takes in UNIX to do something simple in Windows. For example, recursively searching through a subtree for some text in a file."
Yeah, "grep -R [sometext] *" is soooo hard. Try using Windows to remotely search five different computers, each in a different building (or perhaps in a different city), and then tell me how much easier Windows makes it. I used to work for Western Resources in Topeka, Kansas just a few years ago. We would have to drive three or four hours (one way!) just to install some damn Windows program in one of our remote offices because Windows didn't have even the most basic capacity for remote management. Very few things are made easier in Windows than in Linux.
And have you ever heard of "books"? They're like oral wisdom, but they're, like, written down and stuff.
"VT100 pops up its ugly head decades after it should have been killed."
When we have to remotely administer a computer through a low link, we certainly don't want some way to make the screen easy to read. Oh no, we'd much rather just see endless blocks of monotonous text. Or better yet, we'd rather three minutes per frame for the remote GUI display to finish. That makes life sooooo much easier.
"Dang X sucks. Bad. What a mess! "Motif Self-Abuse Kit" made me laugh because my brief experience programming Motif was one of the worst in my life."
News flash: Motif is not the X Window System. Motif is a horrid and ugly layer that sits on top of the X Window system and provides the most horrible excuse for a visual interface that I've ever seen (with GTK trailing by a not so wide margin). If Motif if your main experience with X11, then it's no wonder you have such a rancid opinion of X. The X Window system itself, though, is genius.
"I've never found a make that I liked."
We're in complete agreement here. The make command, while doing a great job internally, is a great example of how to drive away perpective developers. We certainly need to create a much higher level make tool that encompasses and hides the common operations that developers perform when building the projects:
1) Compile those files that are out of date. The make process should be able to work this out easily.
files need to be recompiled, etc.
2) Have a simple command language that performs the common pre and post make tasks such as building RPMs.
The make command is one of my few really hated implementations. It's truly painful to use.
He said: "We're still finding issues. But all of the newer code has got to be ten or a hundred times better."
What it really means: "We were lying all those years we told you that we created top notch code. Now that nobody believes us anymore, we're starting another round. No really! This time we really do have top notch code. Just trust us. We wouldn't lie about something this important."
He said: "The [applications] that people make themselves tend to be better than the larger, all-encompassing applications."
What it really means: "We were lying all those years we told you that tight integration makes for better applications. Software written by people outside Microsoft, who make one application and make it work well, outperform the all-in-one software produced at Microsoft."
He said: "Personally, I'm against shoving things into the kernel. That was a very careful decision. We have a lot of parsing in there, and that opens you up to buffer overruns and attacks."
What it really means: "We know that our kernel is already badly broken because we keep stuffing shit in there, and we know that stuffing yet more shit in there is a really, really, really bad idea, but we're going to do it anyway."
He said: "That's something Linux has that's ahead of us, but we're looking at it. We will have a command line-only version, but whether it'll have all the features in is another matter."
What it really means: "We were lying all those years that we told you how bad the command line was, and how much more reliable a GUI makes a server. The Linux guys have been ahead of us for years, but we're just now willing to admit it (since nobody believes us anymore)."
He said: "It looks really good on PowerPoint! Reality is never quite as good."
What it really means: "We're gonna send out our Marketroids to sell you on a whole new batch of lies and PowerPoint slides. Rest assured that the actual product will suck badly."
I'm not a big game player, so I'm not that upset with the limited game selection on Linux. However, I do like to play them sometimes. I bought the tin-box Linux version of Q3A, bought UT and downloaded the Linux installer, and bought RtCW for Linux.
Last year I bought a PS2 to satisfy my non-FPS gaming desires. Since then I've bought about eight games for it (GTA, GTA/VC, Spiderman, ATV Offroad-whatchamacallit, and a few others I can't recall right now). I mostly play them with several cheats enabled to bring them down to a level where they're fun. I'm not a gaming purist. I have games to pass some time and divert my attention for a while, not to forge a new lifestyle (which is why I won't buy or play NWN or participate in any MMORPG).
The point of all this? I'm more than willing to pay for quality games that I'll only play every now and then, and they don't necessarily have to be big blockbusters.
Two of the most enjoyable games I've been playing recently aren't even commercial. I've been playing FooBillard off and on for a few weeks, and I love the game (I'm not even a pool fan). Same thing with Kolf. They're quite simple (from a playability perspective) and a great way to recharge my brain after (or while) working.
High quality doesn't have to mean expensive or big name. Too many game designers are focusing heavily on visual effects and too little on re/playability. Castle Wolfenstein, Doom, and Quake were brilliant (I am quite impressed with Carmack and co.) in terms of both replayability (I still fire them up from time to time) and uniqueness.
But now everyone and his brother (and including Carmack and co.) keeps pumping these rehashes out. When I had first heard of RtCW, I couldn't wait to get my hands on it. After playing it for a while, though, it quickly dawned on me how repetitive and uninspired it was. It just just another Quake with different graphics and a new storyline. The gameplay was the same old stuff. Like Star Wars and Star Trek, it's time to let these franchises take a vacation.
I know that John, for example, is capable of great gameplay insights, and I'm still greatful that he is allowing for Linux versions of ID games (and releasing old games under the GPL!). But I just can't bring myself to buy yet another FPS or common rehash.
I remember from years ago, a company called MVP [Software | Games] made a very simplistic but incredibly fun game based on the 1991 gulf war. I had no money back then, but I would have bought the full version if I could have. Now that was different and fun. It was shareware, and would have been well worth the money.
Now that I've been gainfully employed for a few years, I have discretionary income that's looking for fresh outlets. If shareware game makers would use OpenGL/SDL and exercise creativity, I'd be happy to pay for their products. As it stands, games have become mostly monotonous and trite over the last couple years.
From the Google translation:
"With such Deal jobs would be at one blow the most powerful figure in the international music business."
Pucker up, Steve.
"I've never heard of VB or Office being considered part of the programming course on a university campus."
You've never been to Southwest Missouri State University, once ranked as the #1 CIS program in the country until the focus shifted to Microsoft vocational training. This illustrious career path split a brain-dead-easy 3-credits, one semester VB class into a 4-credits, 2 semesters, even-more-brain-dead assurance of computer illiteracy.
To add embarassment to insult, the AOS (Administrative Office Systems [or something like that]) associate degree program was folded into the bachelor of science CIS degree program. The result? A few extra classes on using Microsoft Office were added to the "required for graduation" list for incoming freshment (thankfully I had long since filed my degree program, so was not subject to the new requirements).
Unix systems were relegated to "this is what you need to type to make your Java applets readable by the instructor" prior to the AIX web server being replaced by NT (which, SURPRISE, quickly became extraordinarily unreliable). The rest of the campus quickly converted to all-Microsoft.
The end results (whether related or not, but I think they were) were that the SMSU CIS program completely dropped off the ranking charts around the country, and graduates frequently knew very little about computing in general and software development in particular (a primary focus of the CIS degree at SMSU).
I think I may have been in the last (or next to last) class that contained highly competent graduates. The vast majority of the graduates in my class that were highly competent were already (or nearly so) that way due to their own outside efforts (many times before even entering the University program).
I suspect that most CIS graduates at SMSU that known anything have gotten to that point on their own steam despite the best efforts of the university's administration.
Sometimes it's embarassing to admit that I got my degree there, but I point out to people that I was over halfway done before the school was bought by Microsoft and that most of what I learned was self-taught; the degree was just a formality I had to endure.
Recently on Slashdot, when I click on the links to story comments, I get taken to the comments and then immediately get directed to an ad page. This isn't a popup that I can filter out, but a forced redirect from Slashdot to the ad page. I have to click the back button to return to the comments page.
Is this a new (or old) Slashdot ad system? It is infuriatingly annoying.
"Can't ATI just release their drivers themselves?"
I was wondering the same thing. Are XFree86 drivers modular? If not, why not? And if not, then XFree86 is severely broken and badly needs to be redesigned to be modular.
If so, why does ATI wait for the core XFree86 team to (months and months later) even look at the possibility of including its drivers?
I'm not familiar with XFree86's internals, but my guess is the former (XFree's internal architecturer is broken). If it were the latter, then this whole thread would be brainlessly moot.
If a fork will improve the system, then I'm all for it.