Slashdot Mirror
Culture Clash: SCO, OpenLinux, Linus And The GPL
hobsonchoice writes "SCO has issued a letter saying SCO Linux customers won't be sued. The same does not seem to apply if using a non-SCO distribution such as RedHat." LightSail points to the SCO letter itself, and raises an interesting point: "If they approve the use of 'their' IP in Linux in a single kernel, then the GPL holds that IP SCO allows to be used by a select few must be freely released to any and all. It appears that all Linux users everywhere were just given a license to continued use of Linux even if SCO would win their suit with IBM." And Haikuu writes "eWeek recently posted an interview conducted by e-mail exchange with Linus Torvalds regarding his recent move to the OSDL and the SCO suit."
No, emphatically not.
GPL applies to copyright'ed materials only. If SCO have other form of IP protection (such as patent, or, as they in fact claim, trade secret) the the GPL does not even interact with it. And see below.
Further, it is possible for SCO to liscence their copyrighted code such that all thier customers may use it. That does not make it GPL'd.
The GPL only applies upon redistribution - it is quite valid for me to link in code written under any sort of liscence to the Linux kernel. However, I may not freely redistribute it unless I can meet all the restrictions on it. From the GPl, section 7:
Granted, it is talking principly in terms of patent liscencing, but that's inconsequential.
As a case in point, conside the NVIDIA binary drivers. If you have an NVIDIA card, you have a liscence to use that copyrighted code. You do not have a liscence to re-distribute NVIDIA's code. Yet you may link the two systems together, just fine, provided you don't try to redistribute the combined work.
Now, SCO redistributing binaries from their ftp site, _after_ they make claims about thier code being in Linux is a whole different kettle of fish. That's a different issue however.
"If they approve the use of 'their' IP in Linux in a single kernel, then the GPL holds that IP SCO allows to be used by a select few must be freely released to any and all. It appears that all Linux users everywhere were just given a license to continued use of Linux even if SCO would win their suit with IBM."
.. the software was licensed under GPL. Will SCO be able to _change_ the license terms (if not any existing SCO linux customer can cut & paste and recontribute their code back to Linux). Especially after issuing this statement of saying they will ensure their customers are unharmed. I would consider being revoked and then given shitty license terms as harmed. Dont know about the courts.
Unfortunately, this may be false (IANAL etc.) because SCO can say that "credit" their liability. That is, if the court finds that their IP was stolen and all Linux users must pay $1000 to SCO, SCO can simply credit their own customers as having "paid up" as it were. Now the interesting issue is this
They said they won't sue people using SCO Linux, not that it was ok with them for the code to be used. There's a difference. If I jaywalk and don't get charged, it's not that jaywalking is not illegal, it's that the law was not enforced. Big difference.
It would be, except that they're still distributing said product.
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
head? I'm sick of hearing this.
It does NOTHING to take away ANYONE's rights under copyright.
Free software authors, like ANY authors, have the right to choose how their works are licensed, and under what terms derivitave works can be made... and that choice includes saying "Anyone can use this as long as they abide by the GPL".
You are always free to contact the copyright holder of any work available under the GPL and request licensing under different terms, you know. Think the author of a GPL work is perhaps a bunch of people? That depends on the project.. often copyrights are assigned to the project leader... and depending on how contributions are made, that may still be the case despite having lots of authors.
I'm really unclear on what aspect of the GPL needs to be "challenged". If the GPL does not apply, then copyright law forbids the things people are doing with those works the authors placed under the GPL. It's not a use license, but if you don't accept it, the law is clear: you don't have the right to do certain things with that code.
SO unless the argument is "THE gpl doesn't apply, so all works available under the GPL are actually in the public domain" there is no argument.
IT doesn't have to be tested in court... copyright is already testd in court, and the authors of any GPL software are free to sue anyone who is not abiding by the license, and hence, has NO right to do what they are doing.
Since no one came up with any better research i thought i would share this, which was the only substantial discussion i was able to find of RCU patent & copyright history:
http://lwn.net/Articles/36164/
What have you got to say for yourselves now, dumbasses? This may not contain the exact stuff that they're all worked up about (although it sounds like they want to claim RCU entirely), but it is a patch for source that does contain the offending material and therefore a derivative work.
Read parent again. Internal use means they don't give/sell/license the product to anyone, so they are not obligated to provide the code to anyone. Not a violation of the GPL.
SCO is now saying they have no intention of suing Linux distributors!
From http://www.computerwire.info/brnews/6FF3308412856B 4D80256D4E005D45FA
(Last time I pasted in a URL, it inserted a space. So if you can't connect, look for spaces)
However, they still appear to want "licensing fees"
Also, notice here that they are now saying the "infringing code" is from AIX and Dynix:
In other words, SCO now seems to be saying that the so-called infringements are from AIX and Dynix instead of System V.
What they still aren't saying is those so-called infringements are of code written by Sequent and IBM, owned by Sequent and IBM, copyrighted by Sequent and IBM, and is theirs to do with as they wish.
While AIX and Sequent Dynix are derivative works of System V, they have a long way to convince me that the IBM/Sequent modifications are derivative works of System V.
It's a contract dispute with IBM based on definitions that have extremely interpreted to SCO's benefit and in ways that I suspect have rarely, if ever, been interpreted before.
Also in the article:
This brings up the obvious questions:
1) Are they talking about other flavors of UNIX that have IBM/Sequent code? For that matter, are there other flavors of Linux have the IBM/Sequent code?
or
2) Have they found other so-called infringements apart from the IBM/Sequent code? Are they claiming, as it sometimes seems, that if System V and another UNIX has a few identical lines of code, there is necessarily an infringement regardless of the actual source of the code?
The source is still on the server, only its location is obfuscated a bit. It seems to me that the move to remove the kernel is just an issue of trying to appear that they've been irreparably damaged. I'd offer to diff from the mainstream kernel myself, but I'm on my laptop which has a tiny, tiny amount of storage space.
Marxism is the opiate of dumbasses