The GCC compiler is required for compiling the Linux kernel and user programs. You can replace just about every other piece of GNU software in a Linux system with better, non-GNU alternatives. Try it. You might like it.
Frankly, Stallman's ego is just too big for my limited system. Other than the gcc compiler and linker, I don't use GNU tools. I use ksh, busybox, KDE and man. Everything GNU gets its hands into turns to crap. See emacs, info, gcc before the egcs fork, erc. etc. etc.
I have no need for a GNU/Linux. My Linux system works fine by itself.
Sure, but that's because the license to the original (GPL'd) code only permits distribution under the GPL.
And the IBM contract seems to indicate that derivative works need to be treated the as if they were part of the original code. It's somewhat GPL-like in that way.
SCO seems to be arguing that the new code can only be distributed under SCO's terms, unlike GPL copyright holders who in general don't care what else you do with your code so long as you obey the GPL when distributing it with theirs
I don't see that there's any difference between the two alternatives. In both cases, the new code can only be distributed under the terms of the license which, in the case of SCO, seems to be no disclosure of derivative works whether you wrote it or SCO wrote it. Call it an anti-GPL license.
The complaint hinges on some very unusual wording in the contract. The analysis is complicated by the fact that SCO can't/hasn't made any copyright claims. SCO has to do a lot of handwaving about IP rather than make clear legal points under copyright or trade secret law.
Also in SCO's favor is the Utah venue. Having Sen. Hatch's kids for your lawyers probably goes a way even in the Federal district court in Utah.
Like you, my initial reaction was "Bullshit!" Now I'm not so sure.
What constitutes a derivative work in a software context has not been litigated, to my knowledge.
GPL advocates would consider code which was written to be linked into a GPLed work to be undeniably a derived work. Distributing a kernel with such code linked inside would undoubtedly get someone a nastygram from the copyright holder demanding the source to the addition. the case will hinges on what non-copyright obligations the section I quoted implies.
This case holds a lot of perils for free software. When a case gets in front of a judge and he's got to rule what some of these vague phrases mean, watch out. There could be all kinds of unintended consequences.
Basically, the contract with AT&T says that IBM will treat all derivative works as if they were part of the original codebase. While that dosn't grant copyright to AT&T/SCO, transfers of copyright must be in writing, it might constrain how IBM can disclose code which was originally part of a SysV derivative. This is undoubtedly where SCO thinks its case lies.
I think the question is was the code originally part of a Unix derived work? If so, then IBM/Sequent may have the obligation to treat the code as if it were part of the SysV code.
But the point that they haven't asserted any copyright claims is probably significant. IBM might be able to shunt aside any discussion of derivative works as not germain to the complaint.
First, on the issue of derivative works Section 2.01 of the first contract says (emphasis added):
AT&T grants to LICENSEE a personal, nontransferrable and nonexclusive right to use in the United STates each SOFTWARE PRODUCT identified in the one or more suppliments hereto, solely for LICENSEE'S own internal business purposes and solely or in conjunction with DESIGNATED CPUs for such SOFTWARE PRODUCT. Such right to use includes the right to modify such SIFTWARE PRODUCT and to prepare derivative works based on such SOFTWARE PRODUCT, provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT.
The case will likely center on the meaning of the phrase I have italicized above. It may very well mean that IBM can't take parts of derivative works that they added to SysV, like RCU, and release them under the GPL.
Then, on the issue of exports, another contract which grants IBM the rights to license the software to third parties says that they can only export the software under license from the US Government. So the distribution to Iran, North Korea, et. al. may be a contractual violation as well.
Your understanding of the USL/BSD case is faulty. While the court did find that BSD included a small amount of code from AT&T, they also found that AT&T took much more than that from BSD.
Your understanding of derivative works is similarly faulty. While it's true that the issue of what is a derivative work in software has never been litigated, it is not true that the owner of the original work owns the copyright to the derivative work. When a company, like IBM, buys the right to make derivative works, they own the copyright on the derivative work. Disney bought the rights to make films from A.A. Milne's Winnie the Pooh. The Milne heirs do not own the rights to those films, Disney does.
A company agrees to a settlement with the stipulation that they do not admit guilt. Theyimmediately come out with a press release denying guilt but saying they settled with the intent of putting it behind them.
Re:didn't read the article did ya'?
on
Latest SCO News
·
· Score: 1
You don't. But you may need to be a programmer to decide if the code went from SCO to Linux or the other way around or came from some other third source like BSD.
The Story on the NDA
on
Latest SCO News
·
· Score: 3, Informative
Don't they have annual required safety inspections? Just make the fee depend, in part, on the number of miles on the odometer and make it illegal to tamper with an odometer.
The vast majority of estate taxes are paid on unrealized capital gains which have never been taxed. It's not double taxation, it's first time taxation.
Dosn't work. The GPL isn't even very clear on whether
Generally, if a license is revoked, that revocation is prospective from the date of revocation and not retrospective. Someone can rely on having a valid license. As, at least until they filed a patent suit (and it appears they filed a breach of contract suit as they don't even own the patents in question) the license was valid.
Classless addressing is 10 years old. Go read about CIDR if you can still find any of theose ancient documents. There are no more class As. There haven't been for a decade. Any old Class As were chopped up into/9s,/10s... , and/26s ages ago.
Slashdot Mirror
C libraries are provided by the compiller. They provide an interface between the C library API and the native OS API. It's part of the toolchain.
You know that libc is maintained by a Red Hat employee?
The GCC compiler is required for compiling the Linux kernel and user programs. You can replace just about every other piece of GNU software in a Linux system with better, non-GNU alternatives. Try it. You might like it.
Frankly, Stallman's ego is just too big for my limited system. Other than the gcc compiler and linker, I don't use GNU tools. I use ksh, busybox, KDE and man. Everything GNU gets its hands into turns to crap. See emacs, info, gcc before the egcs fork, erc. etc. etc.
I have no need for a GNU/Linux. My Linux system works fine by itself.
Sure, but that's because the license to the original (GPL'd) code only permits distribution under the GPL.
And the IBM contract seems to indicate that derivative works need to be treated the as if they were part of the original code. It's somewhat GPL-like in that way.
SCO seems to be arguing that the new code can only be distributed under SCO's terms, unlike GPL copyright holders who in general don't care what else you do with your code so long as you obey the GPL when distributing it with theirs
I don't see that there's any difference between the two alternatives. In both cases, the new code can only be distributed under the terms of the license which, in the case of SCO, seems to be no disclosure of derivative works whether you wrote it or SCO wrote it. Call it an anti-GPL license.
The complaint hinges on some very unusual wording in the contract. The analysis is complicated by the fact that SCO can't/hasn't made any copyright claims. SCO has to do a lot of handwaving about IP rather than make clear legal points under copyright or trade secret law.
Also in SCO's favor is the Utah venue. Having Sen. Hatch's kids for your lawyers probably goes a way even in the Federal district court in Utah.
Like you, my initial reaction was "Bullshit!" Now I'm not so sure.
What constitutes a derivative work in a software context has not been litigated, to my knowledge.
GPL advocates would consider code which was written to be linked into a GPLed work to be undeniably a derived work. Distributing a kernel with such code linked inside would undoubtedly get someone a nastygram from the copyright holder demanding the source to the addition. the case will hinges on what non-copyright obligations the section I quoted implies.
This case holds a lot of perils for free software. When a case gets in front of a judge and he's got to rule what some of these vague phrases mean, watch out. There could be all kinds of unintended consequences.
...look for another post of mine in this story.
Basically, the contract with AT&T says that IBM will treat all derivative works as if they were part of the original codebase. While that dosn't grant copyright to AT&T/SCO, transfers of copyright must be in writing, it might constrain how IBM can disclose code which was originally part of a SysV derivative. This is undoubtedly where SCO thinks its case lies.
I think the question is was the code originally part of a Unix derived work? If so, then IBM/Sequent may have the obligation to treat the code as if it were part of the SysV code.
But the point that they haven't asserted any copyright claims is probably significant. IBM might be able to shunt aside any discussion of derivative works as not germain to the complaint.
The case will likely center on the meaning of the phrase I have italicized above. It may very well mean that IBM can't take parts of derivative works that they added to SysV, like RCU, and release them under the GPL.
Then, on the issue of exports, another contract which grants IBM the rights to license the software to third parties says that they can only export the software under license from the US Government. So the distribution to Iran, North Korea, et. al. may be a contractual violation as well.
Your understanding of the USL/BSD case is faulty. While the court did find that BSD included a small amount of code from AT&T, they also found that AT&T took much more than that from BSD.
Your understanding of derivative works is similarly faulty. While it's true that the issue of what is a derivative work in software has never been litigated, it is not true that the owner of the original work owns the copyright to the derivative work. When a company, like IBM, buys the right to make derivative works, they own the copyright on the derivative work. Disney bought the rights to make films from A.A. Milne's Winnie the Pooh. The Milne heirs do not own the rights to those films, Disney does.
A company agrees to a settlement with the stipulation that they do not admit guilt. Theyimmediately come out with a press release denying guilt but saying they settled with the intent of putting it behind them.
You don't. But you may need to be a programmer to decide if the code went from SCO to Linux or the other way around or came from some other third source like BSD.
From Linuxjournal
...are largely going down to defeat now that the public has caught on to what's been going on. See Ed Felton's S-DMCA Topic page
Don't they have annual required safety inspections? Just make the fee depend, in part, on the number of miles on the odometer and make it illegal to tamper with an odometer.
The vast majority of estate taxes are paid on unrealized capital gains which have never been taxed. It's not double taxation, it's first time taxation.
A photograpoher knows which of his photographs are worth protecting and which aren't.
The Berne Convention requires a minimum 50 year term of copyright and no formalities. That's what you get here.
If you're willing to pay the fee, you can get more time, but the minimum term offered is not a violation of Berne.
Dosn't work. The GPL isn't even very clear on whether
Generally, if a license is revoked, that revocation is prospective from the date of revocation and not retrospective. Someone can rely on having a valid license. As, at least until they filed a patent suit (and it appears they filed a breach of contract suit as they don't even own the patents in question) the license was valid.
Or at least vim does.
Classless addressing is 10 years old. Go read about CIDR if you can still find any of theose ancient documents. There are no more class As. There haven't been for a decade. Any old Class As were chopped up into /9s, /10s ... , and /26s ages ago.
Now take your hands off the mouse and step away from the keyboard.
Microsoft Accuses Open Source Programmers of Peeking at Private members.
...How to Write Code Guaranteed to Work Only with Microsoft's Implementatyion of .NET and Not Mono.
The models were announced some time ago.