I wonder why IBM didn't file a motion to dismiss SCO's original lawsuit. Darl was all over that in the media. "They didn't file a motion to dismiss. Therefore they know they'll lose", or something to that effect.
What if this ESR tool runs and finds commonality, and the research shows that, in fact, SCO's rights were breached....
So, given that hypothetical, what would people here think? Would you forive SCO?
No. There is no reasonable possiblity that infringements within several orders of magnitude as the "millions of lines" SCO has claimed actually exist.
As sanctimonious and hypocritical as they've been about this whole thing, I expect SCO to continue the FUD and gain another few dollars on their stock price while the media reports ESR's findings support SCO (despite the fact that 99% of commonalities will be explained without infringement--a fact the media will ignore).
It doesn't really matter until it gets to court. And even if SCO succeeds in proving some infringement, do you think the judge will allow 3 billion dollars from IBM and $1400/CPU from all Linux users? Especially since SCO has unclean hands, and the GPL issue, and the fact that they probably steal GPL code themselves, etc. etc.
However, if code was diliberatly obfuscated, that would be very, very bad news for Linux. That shows that it was not an honest mistake, but the programmer knew something about the origins and they needed to be hidden.
It would be hard to prove that the differences between Unix and Linux amount to deliberate obfuscation.
SCO thought that Linux's BPF implementation was an obfuscated copy of the original, when in fact it's a clean room reimplementation based on published specifications.
In addition, natural evolution of code could be easily mistaken as "obfuscation" by someone who is not a kernel hacker.
In doing this we angered some in the Open Source community by pointing out obvious intellectual
property problems that exist in the current Linux software development model.
It's not obvious to me that there are intellectual property problems in the current Linux software
development model. In particular, there are no problems that don't exist in traditional closed
source development either. If anything, Linux developers have a greater incentive not to copy
code, because their work is available for public scrutiny. Had Microsoft, for example, "infringed"
on your copyrights, you wouldn't even know about it!
The second development was an admission by Open Source leader Bruce Perens that UNIX System V code
(owned by SCO) is, in fact, in Linux, and it shouldn't be there. Mr Perens stated that there is "an
error in the Linux developer's process" which allowed Unix System V code that "didn't belong in
Linux" to end up in the Linux kernel (source: ComputerWire, August 25, 2003).
You have misquoted Perens. He certainly did not claim that "Unix System V" code was in Linux. He
said there was code in Linux that is "very similar" to System V, but has a traceable open-source
origin:
The code is from Unix version 3, the oldest known version of Unix that still exists in
machine-readable form. The complete source for that system can be found here on the net. In
2002, Caldera released this code as Open Source, under this license. Caldera is, of course,
the company that now calls itself SCO. The license very clearly permits the Linux developers to use
the code in question.
It is also worthy of note that this code was removed from the kernel for technical reasons
before it was shown at SCOForum, and that, as part of the ia64 port, it was never even a
part of any major commercial Linux distribution.
To date, we claim that more than one million lines of Unix System V protected code have been
contributed to Linux
To arrive at a number remotely close to "one million lines", you must not be referring to Unix
System V code itself. This is probably why you used the phrase "Unix System V protected code".
You seem to think that NUMA, RCU, and JFS--technologies copyrighted and patented by IBM--somehow
now belong to SCO thanks to your contract with IBM. IBM disputes this, of course.
In any event, your contract claims give you absolutely no right to collect license fees
from end-users over these technologies. IBM holds the copyrights on the code; it was IBM's to give
to Linux. If doing so breached IBM's contract with you, you are within your rights to seek damages
from IBM. However, the contracts between SCO and IBM simply cannot apply to end-users who are not
a party to the contracts.
Bruce Perens addresses this issue here, at the end of same the article from
which you misquoted him earlier in your letter.
You draw some pretty serious conclusions from this one weak example of infringement: "In fact, this issue goes to the very heart of whether Open Source can be trusted as a development model for enterprise computing software." Closed source software is not immune from the same kinds of problems. Take a look at the recent patent lawsuits against Microsoft. Can Closed Source be trusted as a development model for enterprise computing software?
You mentioned that the aforementioned code copying issue damaged the open source community's
credibility, but I find it telling that you didn't mention the other code copying example
given at SCOForum. This example intended to show that SCO owns the Berkely Packet Filter code
(which is, of course, part of BSD), and that an implementation of the BP
We view IBM's counterclaim filing today as an effort to distract attention from its flawed Linux business model
This must be why IBM made three billion dollars last year and you lost 30 million.
It repeats the same unsubstantiated allegations made in Red Hat's filing earlier this week.
Ha! I think it's painfully obvious who has been making unsubstantiated allegations. The fact is, you claim Linux infringes on your IP and you haven't proven it, or even presented any evidence at all (other than doctored snippets of code taken out of context and shown under NDA)
If IBM were serious about addressing the real problems with Linux, it would offer full customer indemnification and move away from the GPL license. As the stakes continue to rise in the Linux battles, it becomes increasingly clear that the core issue is bigger than SCO (SCOX) , Red Hat, or even IBM.
Oh. This explains why Caldera distributed Linux under the GPL for a decade, and why their extortion^H^H^H^H^H^H^H Linux license disclaims all warranties...
The core issue is about the value of intellectual property in an Internet age.
And how to milk yours for all its worth while completely disregarding everyone else's.
In a strange alliance, IBM and the Free Software Foundation have lined up on the same side of this argument in support of the GPL. IBM urges its customers to use non- warranted, unprotected software.
The same software you sold for a decade. Or did you change your name to "The SCO Group" to deliberately try to obscure that fact?
This software violates SCO's intellectual property rights in UNIX,
This is highly debatable. I believe that's why you're being sued by two companies. Prove you're right and THEN collect license fees.
and fails to give comfort to customers going forward in use of Linux. If IBM wants customers to accept the GPL risk, it should indemnify them against that risk.
Or it should defend the GPL from parasites like SCO. I find IBM's countersuit quite comforting regarding my future use of Linux.
The continuing refusal to provide customer indemnification is IBM's truest measure of belief in its recently filed claims.
Oh, I see, since IBM decides to fight your false claims directly rather than wait for you to sue end-users, they must be tacitly acknowledging their flawed business plan. Whatever.
SCO has shipped these products for many years, in some cases for nearly two decades, and this is the first time that IBM has ever raised an issue about patent infringement in these products.
IBM and RedHat have shipped Linux 2.4 since 2000, and this is the first time SCO has ever raised an issue about copyright infringements in these products.
Furthermore, these claims were not raised in IBM's original answer.
But they were raised in IBM's answer to your amended claim. You started this fight. Turnabout is fair play.
SCO reiterates its position that it intends to defend its intellectual property rights. SCO will remain on course to require customers to license infringing Linux implementations as a condition of further use.
Translation: SCO will continue to make unsubstantiated allegations against Linux and Linux users in order to keep their stock value high long enough for the execs to sell off.
This is the best and clearest course for customers to minimize Linux problems.
No, that would have been to publicize your claims immediately so that they could be disproven and/or any infringing code could be removed from the kernel.
You can't compare the share prices of different companies directly. It doesn't mean anything (since all companies don't have the same amount of shares).
SCO's Market Value $ 143,952,530
IBM's Market Value $ 139,432,320,000
If there is SCO code in the kernel, then why doesn't someone remove it and replace it with non-SCO code?
Because SCO won't say what the disputed code is. RedHat alleges in its suit against SCO that there is no disputed code, and SCO is just raising allegations in order to FUD Linux and extort money from Linux users.
SCO's "SYSV for Linux" are libraries that people were using to run UNIX applications on Linux. SCO saw that people were pirating these libs from UnixWare, and so it offered to sell them so that people could run UNIX apps on Linux WITHOUT having to license all of UnixWare. I'm fairly certain that these libs were never included in any commercial Linux distro.
In any event, all this is completely irrelevant to SCO's current claims of infringement within the Linux kernel.
Blake Stowell on Moglen's statement that users don't need a "copyright license":
Copyright absolutely applies. For the same reason that a commercial user must have a valid licence to run Microsoft Word, a user must also have a valid licence to run our Unix source code.
It was my understanding that the Microsoft EULA is the "use" license for Word--Linux customers have made no such agreement with SCO.
In addition, Stowell admits that IBM holds the copyrights to the code in question (emphasis mine):
While IBM owns the copyrights on these derivative Unix programs, SCO owns the control rights to these and they cannot be contributed to open source. The contracts between IBM and SCO state all of this.
Someone please ask SCO this:
Since IBM has the copyrights to the code in question, what recourse can SCO possibly have against end-users?
If the contracts forbid these "derivative works" from being contributed to open source, what recourse do you have against end-users now that they have been? You don't own the copyrights for such code, and end users are not party to your contract with IBM.
Or do you realize that you have no claim against end-users, and as such are inventing a new kind of intellectual property called "control rights"?
Blake Stowell on Moglen's statement that users don't need a "copyright license":
Copyright absolutely applies. For the same reason that a commercial user must have a valid licence to run Microsoft Word, a user must also have a valid licence to run our Unix source code.
It was my understanding that the Microsoft EULA is the "use" license for Word--Linux customers have made no agreement with SCO.
In addition, Stowell admits that IBM holds the copyrights to the code in question (emphasis mine):
While IBM owns the copyrights on these derivative Unix programs, SCO owns the control rights to these and they cannot be contributed to open source. The contracts between IBM and SCO state all of this.
Someone please ask SCO this:
Since IBM has the copyrights to the code in question, what recourse can SCO possibly have against end-users?
It looks to me like all they can do is go against IBM for breach of contract. But they're inventing a new kind of intellectual property, "control rights", which allow them to go against end-users, or so it looks to me.
First, the premise of your case--you claim that your UNIX on x86 market share was unfairly eroded because enterprise features which you claim as your IP (including NUMA and RCU) were misappropriated into the Linux kernel. This argument doesn't hold much water, for two reasons:
(1) These features were never present in your own UNIX offering. They were not even developed by SCO/Caldera; they were developed by Dynix, which is now a subsidiary of IBM. Because these features were developed for SVR5, you claim they are derivative works of SVR5 and therefore your own intellectual property. The legitimacy of this claim depends upon your contracts with IBM; it is not as black and white as you make it out to be. When asked whether the code supposedly copied from SVR5 originated in BSD, you respond that this is high end "enterprise" code which isn't
present in BSD--but it's not present in SVR5 either. Your claims on this matter are misleading.
(2) To state that Linux stole your market is preposterous, since you yourself were a Linux value-added reseller. In fact, you actively contributed to the development of enterprise features for the Linux kernel. You even cooperated with IBM in the Trillian Project (SMP on Linux). Your previous CEO, Ransom Love, spoke of unifying UNIX and Linux into a single platform. Now you turn and say that an enterprise-ready Linux took you completely by surprise, even
though you helped bring it where it is.
You portray yourself as a protector of intellectual property rights, but then you seek to wrest control of Linux from its creators on the basis of unproven allegations of copyright infringement. Your arrogance and hypocrisy know
no bounds. Linux development has been very transparent, as Linus Torvalds has said. If you were really interested in protecting the intellectual property of all parties involved, you would work with the kernel developers to find
out which parties contributed your intellectual property to the kernel and seek relief from them and/or allow the infringing code to be removed. Even if this would disable Linux SMP for a time, there are millions of Linux users running uniprocessor systems who were never infringing on your IP in the first place, and should not have to pay you a license fee. Furthermore, your move to collect license payments from Linux users without identifying specifically what they are licensing or even proving that you have a claim on Linux at all is fraudulent.
Your proposed Linux licensing program amounts to the wholesale theft of years of effort from thousands of Linux contributors. You have profited from their efforts for nearly a decade, and now you stab them in the back and bite
the hand that fed you. Since you could not compete in the marketplace, you resort to barratry, racketeering, and extortion.
Re:NOBODY has mentioned SCO being shutdown in Germ
on
SCO SCO SCO!
·
· Score: 2, Informative
I believe the injunction only applies to SCO's German subsidiary. Maybe this is why www.sco.de has been blank for several days.
In a recent article (http://www.computerworld.com/softwaretopics/os/li nux/story/0,10801,81709,00.html), you indicated that, should Linux be shown to contain SVR5 code, the open source community will not be afforded the opportunity to remove it--instead, you will seek damages, presumably from anybody and everybody that distributes Linux and/or uses it commercially (judging from your
letter). You rationalize this threat with the following analogy:
"It's sort of like somebody stealing your car, and you hunt them down and you find them, and they say you can have your car back, but there's no penalty for that. If there's no penalty for stealing property, then where are we?"
The problem with this analogy is that commercial Linux distributors and customers *have not*
stolen your intellectual property. They did not know when they adopted Linux that it
(allegedly) infringed on your IP. They have made no agreement with you. Your desire to extract
payments from these commercial Linux distributors and customers is extortion, pure and simple.
What's more, your obstinate refusal to make the allegedly copied code public makes it impossible
for new Linux customers--those who have not already "stolen your car", so to speak--to adopt a
platform that excludes those portions of the kernel that you claim as your own. I believe this
is by design. Since "hundreds of lines" in the 3-million-line Linux kernel may infringe on your
intellectual property, you wish to throw out the other 2.999 million lines so you may unfairly
eliminate your competitors.
(A more correct car analogy is this one: Suppose Ford Motor Company stole some trade secrets
from SCO Motors, Inc., and used them to build a superior automobile. Then SCO Motors sues Ford
AND everybody who drives a Ford. Would this be just?)
You also claim that the open-source software model is inherently flawed because there is no
"traffic cop" to ensure that submitted code is unencumbered. This is a ridiculous argument. As
the Microsoft-Timeline case shows, traditional closed-source software companies are not immune
from misappropriation of code. The only "problem" with open-source software is that anybody can
review it, looking for their own code, as SCO has done. Linux users do not have the same
luxury--they cannot, for example, examine SCO's source code to see if it contains any GPL code.
Furthermore, how would it be possible for any company, open-source or otherwise, to ensure that
newly submitted source code is not plagiarized? Suppose a former Microsoft employee gets a
programming job at Cisco, and "writes" new code that he actually stole from his former employer.
He removes all of Microsoft's copyright notices, of course. How can Cisco know that the code
was stolen from Microsoft? It can't--it does not have access to Microsoft's code to compare.
Similarly, the maintainers of the Linux kernel *could not have known* when your alleged SVR5
code was submitted into the source tree. The best you can hope for is for them to remove it
once they discovered it was plagiarized--but you won't accept that. Instead, you'll jump to the
conclusion that the entire open source software model is flawed and that everyone should just
use proprietary SCO software instead.
You have indicated that you're only "trying to take care of the shareholders, employees and
people who have been having their rights trampled on." What about the rights of everybody
except SCO? The vast majority of LDPs and Linux users have done you no wrong. You are within
your right to pursue damages against IBM for breach of contract, but you have no business trying to use this opportunity to unfairly destroy your biggest competitor.
Slashdot Mirror
I wonder why IBM didn't file a motion to dismiss SCO's original lawsuit. Darl was all over that in the media. "They didn't file a motion to dismiss. Therefore they know they'll lose", or something to that effect.
No. There is no reasonable possiblity that infringements within several orders of magnitude as the "millions of lines" SCO has claimed actually exist.
As sanctimonious and hypocritical as they've been about this whole thing, I expect SCO to continue the FUD and gain another few dollars on their stock price while the media reports ESR's findings support SCO (despite the fact that 99% of commonalities will be explained without infringement--a fact the media will ignore).
It doesn't really matter until it gets to court. And even if SCO succeeds in proving some infringement, do you think the judge will allow 3 billion dollars from IBM and $1400/CPU from all Linux users? Especially since SCO has unclean hands, and the GPL issue, and the fact that they probably steal GPL code themselves, etc. etc.
It would be hard to prove that the differences between Unix and Linux amount to deliberate obfuscation.
SCO thought that Linux's BPF implementation was an obfuscated copy of the original, when in fact it's a clean room reimplementation based on published specifications.
In addition, natural evolution of code could be easily mistaken as "obfuscation" by someone who is not a kernel hacker.
It's not obvious to me that there are intellectual property problems in the current Linux software development model. In particular, there are no problems that don't exist in traditional closed source development either. If anything, Linux developers have a greater incentive not to copy code, because their work is available for public scrutiny. Had Microsoft, for example, "infringed" on your copyrights, you wouldn't even know about it!
You have misquoted Perens. He certainly did not claim that "Unix System V" code was in Linux. He said there was code in Linux that is "very similar" to System V, but has a traceable open-source origin:
The code is from Unix version 3, the oldest known version of Unix that still exists in machine-readable form. The complete source for that system can be found here on the net. In 2002, Caldera released this code as Open Source, under this license. Caldera is, of course, the company that now calls itself SCO. The license very clearly permits the Linux developers to use the code in question.
It is also worthy of note that this code was removed from the kernel for technical reasons before it was shown at SCOForum, and that, as part of the ia64 port, it was never even a part of any major commercial Linux distribution.
To arrive at a number remotely close to "one million lines", you must not be referring to Unix System V code itself. This is probably why you used the phrase "Unix System V protected code". You seem to think that NUMA, RCU, and JFS--technologies copyrighted and patented by IBM--somehow now belong to SCO thanks to your contract with IBM. IBM disputes this, of course.
In any event, your contract claims give you absolutely no right to collect license fees from end-users over these technologies. IBM holds the copyrights on the code; it was IBM's to give to Linux. If doing so breached IBM's contract with you, you are within your rights to seek damages from IBM. However, the contracts between SCO and IBM simply cannot apply to end-users who are not a party to the contracts.
Bruce Perens addresses this issue here, at the end of same the article from which you misquoted him earlier in your letter.
You draw some pretty serious conclusions from this one weak example of infringement: "In fact, this issue goes to the very heart of whether Open Source can be trusted as a development model for enterprise computing software." Closed source software is not immune from the same kinds of problems. Take a look at the recent patent lawsuits against Microsoft. Can Closed Source be trusted as a development model for enterprise computing software?
You mentioned that the aforementioned code copying issue damaged the open source community's credibility, but I find it telling that you didn't mention the other code copying example given at SCOForum. This example intended to show that SCO owns the Berkely Packet Filter code (which is, of course, part of BSD), and that an implementation of the BP
The correct posting is:
In Soviet Russia, the LIGHT BULB changes YOU.
Here or Here
Look here to find the same comment used in SCO's slideshow in a 1984 Usenet posting: http://tinyurl.com/khnj
So SCO's new story is that, even though IBM owns the relevant copyrights, IBM didn't have the right to donate NUMA and RCU to Linux.
Well, IBM did.
SCO has no grounds to charge end-users license fees for using this code, as SCO owns no relevant copyrights or patents.
SCO is welcome to seek damages for breach of contract against IBM, which of course they are. This is their only recourse, however. Good luck.
You can't compare the share prices of different companies directly. It doesn't mean anything (since all companies don't have the same amount of shares). SCO's Market Value $ 143,952,530 IBM's Market Value $ 139,432,320,000
Because SCO won't say what the disputed code is. RedHat alleges in its suit against SCO that there is no disputed code, and SCO is just raising allegations in order to FUD Linux and extort money from Linux users.
In any event, all this is completely irrelevant to SCO's current claims of infringement within the Linux kernel.
It was my understanding that the Microsoft EULA is the "use" license for Word--Linux customers have made no such agreement with SCO.
In addition, Stowell admits that IBM holds the copyrights to the code in question (emphasis mine):
Someone please ask SCO this:
Since IBM has the copyrights to the code in question, what recourse can SCO possibly have against end-users?
If the contracts forbid these "derivative works" from being contributed to open source, what recourse do you have against end-users now that they have been? You don't own the copyrights for such code, and end users are not party to your contract with IBM.
Or do you realize that you have no claim against end-users, and as such are inventing a new kind of intellectual property called "control rights"?
It was my understanding that the Microsoft EULA is the "use" license for Word--Linux customers have made no agreement with SCO.
In addition, Stowell admits that IBM holds the copyrights to the code in question (emphasis mine):
Someone please ask SCO this:
Since IBM has the copyrights to the code in question, what recourse can SCO possibly have against end-users?
It looks to me like all they can do is go against IBM for breach of contract. But they're inventing a new kind of intellectual property, "control rights", which allow them to go against end-users, or so it looks to me.
(1) These features were never present in your own UNIX offering. They were not even developed by SCO/Caldera; they were developed by Dynix, which is now a subsidiary of IBM. Because these features were developed for SVR5, you claim they are derivative works of SVR5 and therefore your own intellectual property. The legitimacy of this claim depends upon your contracts with IBM; it is not as black and white as you make it out to be. When asked whether the code supposedly copied from SVR5 originated in BSD, you respond that this is high end "enterprise" code which isn't present in BSD--but it's not present in SVR5 either. Your claims on this matter are misleading.
(2) To state that Linux stole your market is preposterous, since you yourself were a Linux value-added reseller. In fact, you actively contributed to the development of enterprise features for the Linux kernel. You even cooperated with IBM in the Trillian Project (SMP on Linux). Your previous CEO, Ransom Love, spoke of unifying UNIX and Linux into a single platform. Now you turn and say that an enterprise-ready Linux took you completely by surprise, even though you helped bring it where it is.
You portray yourself as a protector of intellectual property rights, but then you seek to wrest control of Linux from its creators on the basis of unproven allegations of copyright infringement. Your arrogance and hypocrisy know no bounds. Linux development has been very transparent, as Linus Torvalds has said. If you were really interested in protecting the intellectual property of all parties involved, you would work with the kernel developers to find out which parties contributed your intellectual property to the kernel and seek relief from them and/or allow the infringing code to be removed. Even if this would disable Linux SMP for a time, there are millions of Linux users running uniprocessor systems who were never infringing on your IP in the first place, and should not have to pay you a license fee. Furthermore, your move to collect license payments from Linux users without identifying specifically what they are licensing or even proving that you have a claim on Linux at all is fraudulent.
Your proposed Linux licensing program amounts to the wholesale theft of years of effort from thousands of Linux contributors. You have profited from their efforts for nearly a decade, and now you stab them in the back and bite the hand that fed you. Since you could not compete in the marketplace, you resort to barratry, racketeering, and extortion.
I believe the injunction only applies to SCO's German subsidiary. Maybe this is why www.sco.de has been blank for several days.
"It's sort of like somebody stealing your car, and you hunt them down and you find them, and they say you can have your car back, but there's no penalty for that. If there's no penalty for stealing property, then where are we?"
The problem with this analogy is that commercial Linux distributors and customers *have not* stolen your intellectual property. They did not know when they adopted Linux that it (allegedly) infringed on your IP. They have made no agreement with you. Your desire to extract payments from these commercial Linux distributors and customers is extortion, pure and simple.
What's more, your obstinate refusal to make the allegedly copied code public makes it impossible for new Linux customers--those who have not already "stolen your car", so to speak--to adopt a platform that excludes those portions of the kernel that you claim as your own. I believe this is by design. Since "hundreds of lines" in the 3-million-line Linux kernel may infringe on your intellectual property, you wish to throw out the other 2.999 million lines so you may unfairly eliminate your competitors.
(A more correct car analogy is this one: Suppose Ford Motor Company stole some trade secrets from SCO Motors, Inc., and used them to build a superior automobile. Then SCO Motors sues Ford AND everybody who drives a Ford. Would this be just?)
You also claim that the open-source software model is inherently flawed because there is no "traffic cop" to ensure that submitted code is unencumbered. This is a ridiculous argument. As the Microsoft-Timeline case shows, traditional closed-source software companies are not immune from misappropriation of code. The only "problem" with open-source software is that anybody can review it, looking for their own code, as SCO has done. Linux users do not have the same luxury--they cannot, for example, examine SCO's source code to see if it contains any GPL code.
Furthermore, how would it be possible for any company, open-source or otherwise, to ensure that newly submitted source code is not plagiarized? Suppose a former Microsoft employee gets a programming job at Cisco, and "writes" new code that he actually stole from his former employer. He removes all of Microsoft's copyright notices, of course. How can Cisco know that the code was stolen from Microsoft? It can't--it does not have access to Microsoft's code to compare.
Similarly, the maintainers of the Linux kernel *could not have known* when your alleged SVR5 code was submitted into the source tree. The best you can hope for is for them to remove it once they discovered it was plagiarized--but you won't accept that. Instead, you'll jump to the conclusion that the entire open source software model is flawed and that everyone should just use proprietary SCO software instead.
You have indicated that you're only "trying to take care of the shareholders, employees and people who have been having their rights trampled on." What about the rights of everybody except SCO? The vast majority of LDPs and Linux users have done you no wrong. You are within your right to pursue damages against IBM for breach of contract, but you have no business trying to use this opportunity to unfairly destroy your biggest competitor.