SCO suspends, Gartner warns [LWN.net]

The SCO Group, it seems, has finally read the GPL; the company has announced that it has suspended shipments of its Linux distribution. It does not do, after all, to be claiming proprietary rights on code which has been mixed into a GPL-licensed product. SCO stands every chance of losing its right to distribute (at least) the kernel in any case; better to take the step ahead of time.

Of course, other interpretations are possible. The company's Linux shipments have, most likely, dropped to something approximating zero in any case. SCO, having lost in the Linux marketplace (even before the lawsuit) appears to wish to bring that whole market down in flames. It's hard to come up with another motivation for statements like:

The SCO Group, the owner of the UNIX operating system, today warned that Linux is an unauthorized derivative of UNIX and that legal liability for the use of Linux may extend to commercial users. SCO issued this alert based on its findings of illegal inclusions of SCO UNIX intellectual property in Linux.

SCO has also sent an unsettling letter to some 1500 companies worldwide.

As FUD (fear, uncertainty, and doubt) attacks go, it is hard to be less subtle than this. If you use Linux, SCO has just threatened to sue you. So much for them not having anything against the Linux community. (The company's posting of a page of quotations from "Linux leaders" - such as noted kernel hacker Richard Stallman - also gives a hint as to what their current attitude toward the community is).

SCO has also trotted out the Gartner Group to drive the point home.

System administrators must be admonished to submit open-source code to inspection for potential violation of patents. An open-source quality assurance process should determine and approve allowable code for production systems. Such efforts may slow adoption of Linux in high-end production systems of critical applications.

Of course, the SCO suit has nothing to do with patents, but it is time to adopt procedures which "may slow adoption" of Linux just a little bit. Of course, Gartner has no suggestions on how anyone might verify that a given chunk of code does not violate anybody's patents. To top it off, Gartner states "However, one thing is certain: The community process is fraught with risk to users." (The report does also note, for what it's worth, "In Gartner's opinion, SCO's claim that IBM misappropriated trade secrets from AIX will be difficult to prove...")

SCO's action, which was once presented as a simple contractual dispute between two corporations, has now been clearly exposed as an attack on Linux itself. At some point, however, SCO is going to have to stop talking and demonstrate some stolen source. If the company actually has something to show, it's past time to put some cards on the table. As it is, SCO gives the impression of trying to destroy the Linux community away with words that have little backing in the real world.

(Log in to post comments)

SCO suspends, Gartner warns

Posted May 15, 2003 2:14 UTC (Thu) by rknop (guest, #66) [Link]

SCO has gone from stupid to downright evil.

I hope that they get quashed, and fast. This is just ridiculous.

This kind of FUD has failed before, so perhaps we shouldn't be worried. The whole line that proprietary software is so much better becasue it insures security and intellectual property has been sold by Microsoft before, and it didn't do much good then. Of course, Microsoft could never say "it is OUR intellectual property", which SCO is saying (whether or not there is anything to it), so it is a new wrinkle.

Is SCO still a part of United Linux? If so, the rest of United Linux needs to kick them out stat. Until then, the rest of us need to boycott anybody in United Linux.

-Rob

SCO suspends, Gartner warns

Posted May 16, 2003 7:37 UTC (Fri) by jdthood (guest, #4157) [Link]

     Is SCO still a part of United Linux? If so, the rest
     of United Linux needs to kick them out stat. Until then, 
     the rest of us need to boycott anybody in United Linux.

No; just boycott SCO.

SCO suspends, Gartner warns

Posted May 15, 2003 4:20 UTC (Thu) by error27 (subscriber, #8346) [Link]

According to one interview on May 12th, Chris Sontag of SCO source was not aware of the implications of violating the GPL as late as May 12th (the interview has since been removed). If they had subscribed to lwn they would have known on May 8th. :)

A comment on osnews.com forum says that SCO contractors are claiming that SCO owns the intellectual property of all the files in /sbin/. While SCO admits that the source code for the Linux implementations is original, it holds that since SCO owns the copyrights to the original UNIX, it owns all the ideas.

The problem with this for SCO is that the courts are going to look at the actual laws. With the existing laws, SCO has no case. SCO doesn't have any patents on the ideas, SCO doesn't own the copyrights (according to the contractors) and BSD has already been proven a legal UNIX reimplementation in court.

BTW. I read on LKML that SCO may have copied GPL code illegally in it's ext2 implementation. If someone had a copy of that, it would be interesting to type strings on the .o file and compare it with the Linux module.

SCO suspends, Gartner warns

Posted May 15, 2003 4:54 UTC (Thu) by tjc (subscriber, #137) [Link]

A comment on osnews.com forum says that SCO contractors are claiming that SCO owns the intellectual property of all the files in /sbin/.

It's hard to know what they're claiming to own, since they won't actually tell anyone, but it seems like about half of it is probably covered by standards like POSIX, FIPS, etc. What I would like to know: is SCO claiming that it's illegal to implement a POSIX compatible OS? That would raise the interest of quite a few people, I would think.

SCO suspends, Gartner warns

Posted May 15, 2003 5:29 UTC (Thu) by eru (subscriber, #2753) [Link]

> It's hard to know what they're claiming to own, since they won't actually tell anyone...

I wonder if it would be possible and sensible for some of the most affected stakeholders (like Red Hat and SuSE) to sue SCO in order to find out? These companies could credibly claim to be harmed by SCO allegations, and request the court to make SCO either prove its claims, or shut up.
A bit like the security researcher (forgot the name) who wanted a court to rule if publishing his research was infringing the DMCA.

SCO suspends, Gartner warns

Posted May 15, 2003 16:41 UTC (Thu) by Baylink (subscriber, #755) [Link]

Given the circumstances, yes, I would say that RedHat et al, and even IBM, were possible candidates for an Emergency Injunction, but it's questionable whether they'd get it.

A shame, really.

The surest way to make this go away, as Darl "God will roast their kernels in hell" McBride is no doubt aware, is for IBM to simply buy SCO, which they could almost certainly do out of pocket change. Is the GPL test case worth the hassle? Likely not; the issues are too muddy for my taste.

But maybe it's just me.

So many things are just me.

SCO suspends, Gartner warns

Posted May 15, 2003 5:33 UTC (Thu) by addw (subscriber, #1771) [Link]

If they want license fees from anyone that implements all/part/... of POSIX, I assume that they will want some from M$ - as that company has a POSIX module -- of does the M$ shareholder stake in SCO/Caldera give them a free to use ticket ?

SCO is still distributing the Linux kernel

Posted May 15, 2003 15:23 UTC (Thu) by proski (subscriber, #104) [Link]

There is no kernel source there. Note that the kernel packages have "nosrc" in them. The best thing I could find is this 2.4.13 kernel package which actually includes the sources.

SCO is still distributing the Linux kernel

Posted May 15, 2003 16:21 UTC (Thu) by metacircles (guest, #8895) [Link]

There is no kernel source there. Note that the kernel packages have "nosrc" in them.

If those packages contain kernel binaries and there's no corresponding source, that falls foul of the GPL section 3. Perhaps the kernel's copyright holders want to take this up with SCO?

SCO is still distributing the Linux kernel

Posted May 16, 2003 2:50 UTC (Fri) by proski (subscriber, #104) [Link]

I don't see how this would violate GPL. GPL doesn't require the sources to be distibuted electronically and be available to everyone. I think it's assumed that the users who need the update already have the source from the original distribution.

SCO is still distributing the Linux kernel

Posted May 16, 2003 9:36 UTC (Fri) by error27 (subscriber, #8346) [Link]

GPL doesn't require the sources to be distibuted electronically

True.

and be available to everyone.

If you distribute code with source then you do not have to make it available to everyone. However, if you do not distribute source then you have to make the code available to anyone who asks within three years. Any way, in this case they are distributing the object files to everyone from a public ftp server and so they have to make the source available to everyone.

I think it's assumed that the users who need the update already have the source from the original distribution.

The GPL calls for "a complete machine-readable copy of the corresponding source code". Patches probably don't qualify as complete. This is all coverred in the paragraphs 3a and 3b of the GPL.

SCO is still distributing the Linux kernel

Posted May 15, 2003 19:02 UTC (Thu) by lyda (guest, #7429) [Link]

sco is not violating the license of "the community." they're violating the license of linus torvalds and the other copyright holders.

it's up to them to sue sco. we can help - as can redhat, suse, the fsf and others.

obviously this assumes the kernel is the only infringing part. the more gpl software sco claims infringes, the more copyright holders can defend their work.

and they should sue sco. it's their license being violated.

SCO is still distributing the Linux kernel

Posted May 16, 2003 9:50 UTC (Fri) by error27 (subscriber, #8346) [Link]

If SCO distributes GPL software on its ftp server, the people who download the software are entitled to the source code. If SCO refuses to provide the source, then the users have the right to sue SCO to obtain it.

The copyright holders have a particular cause to sue, but the users have been wronged by SCO's license violation as well.

SCO suspends, Gartner warns

Posted May 15, 2003 7:02 UTC (Thu) by ekj (subscriber, #1524) [Link]

Gartner has no suggestions on how anyone might verify that a given chunk of code does not violate anybody's patents.

This is a big problem, much bigger than the SCO-lunatics really. There is no way for anyone to release any code legally. Not even in principle is it possible to ensure against patent-snags.

Even if you could (you can't !) hire the required people to rea dtrough all relevant patents and compare them to your work, this would bring you nothing at all, because it's perfectly possible that a patent for something you are doing was submitted 3 years ago, but will only surface 3 years in the future.

It's an interesting set of laws where the very act of publisinh anything, regardless of how careful you are and what precautions you take make you subject to possible liabilities is supposed to stimulate new ideas and creativity.

SCO suspends, Gartner warns

Posted May 15, 2003 7:43 UTC (Thu) by beejaybee (subscriber, #1581) [Link]

" System administrators must be admonished to submit open-source code to inspection for potential violation of patents. "

Anyone following GPL will of course make their source code available for inspection. What the purpose of the inspection might be is irrelevant.

IANAL but I would have thought it is the responsibility of a patent holder to check for violation of patents, not an person or group of people working independently.

The problem here is that patent offices are now permitting patenting of concepts which are effectively as basic & obvious as the number 3.The system is becoming as absurd as if Disney Corp were claiming copyright on the letter "e", so that no-one would be able to produce any work in any form containing the letter "e" without paying royalties to Disney.

BTW I'm really pleased that Disney are apparently likely to lose the marketing rights for Winnie the Pooh, estimated to be worth $6 billion. I hope the action succeeds.

Lawyers, lawyers....

Posted May 15, 2003 12:10 UTC (Thu) by alspnost (subscriber, #2763) [Link]

Once lawyers really do control society - something that appears to be happening rather rapidly, especially in the US - then we'll have a whole lot more to worry about than threats to Linux.

The everyone-loses (except the lawyers) parasitic effect on our economies and technological progress will probably do more to destroy our societies than any threat from fanatical terrorism. What the lawmakers and law practitioners are doing is akin to "legal", non-violent financial and intellectual terrorism against all of us.

I don't mean to sound sensationalist, but the things I read in my newspapers these days make me *shiver*....

False attribution to Richard Stallman?

Posted May 15, 2003 9:16 UTC (Thu) by stuart_hc (subscriber, #9737) [Link]

Following your link to the page of quotations from "Linux leaders"
http://www.sco.com/scosource/quotes_from_leaders.html
I saw that SCO had attributed the quote
"Linux is a copy of UNIX. There is very little new stuff in Linux."
to Richard Stallman.

A quick look through the Linux-kernel mailing list archive reveals that Larry McVoy made this statement on 4th Jan 2003. Perhaps the mistake SCO made was to interpret Richard Stallman's reply to Larry McVoy, in which McVoy's statement is quoted, as the source of the quotation.

How many other simple mistakes has SCO made...

False attribution to Richard Stallman?

Posted May 15, 2003 15:56 UTC (Thu) by jdthood (guest, #4157) [Link]

I am unable to find the McVoy posting you mention, but I took a closer look at SCO's quotation and noticed that this one is different from the others in not naming Stallman on the line following the quotation. It seems that this is a quotation randomly lifted from linux-kernel (... as it happens, from a severe critic of Stallman ...) for its useful sound bite qualities.

The other quotations are very misleading when presented out of context. Stallman does not in the least condone illegal copying, but the purpose of the second quotation seems to be to give the impression that he does. The third quotation is equally misleading. To the general public, 'hack' means 'crack', but we know that this is not what Stallman means when he uses the word.

SCO is engaging in anti-Linux propaganda. Disgusting.

For your reference, I append how the "Richard Stallman" section appears on SCO's webiste at 15:44 UTC on 15 May 2003.

Richard Stallman

"Linux is a copy of UNIX. There is very little new stuff in Linux."
Linux kernel forum

"I consider the law prohibiting the sharing of copies with your friend the moral equivalent of Jim Crow. It does not deserve respect."
Richard Stallman, Free as in Freedom, Richard Stallman's Crusade for Free Software: O'Reilly (2002) at p. 72

"The whole GNU project is really one big hack. It's one big act of subversive playful cleverness..."
Richard Stallman, Revolution OS (DVD)

False attribution to Richard Stallman?

Posted May 15, 2003 16:00 UTC (Thu) by dbreakey (guest, #1381) [Link]

Somehow, I think that SCO knows perfectly well that it's an incorrect attribution and simply doesn't care. Those pages have absolutely no legal weight, so the only purpose to them is to try and carry public opinion over to their side.

Pathetic, if you ask me…

False attribution to Richard Stallman?

Posted May 15, 2003 18:52 UTC (Thu) by lyda (guest, #7429) [Link]

therefore richard stallman should sue them for libel. while sco can afford a huge lawsuit, the free software community can affod a lot of small lawsuits.

False attribution to Richard Stallman?

Posted May 15, 2003 21:42 UTC (Thu) by dododge (subscriber, #2870) [Link]

Here's the original by Larry McVoy, and the followup by RMS. RMS just has horrible quoting style (or lack thereof) so without seeing his message in thread context it's not very clear that he isn't making this statement himself. The lack of a "GNU/" is at least an indicator to experienced RMS-readers that something is up :-)

Re: SCO suspends, Gartner warns

Posted May 15, 2003 17:16 UTC (Thu) by gleef (guest, #1004) [Link]

He's certainly a noted hacker, he wrote the core of Emacs and GCC. I would think that if he were a "kernel hacker" he would have spearheaded writing the kernel for GNU/Hurd rather than waiting for others to do so, and it would be further along by now. He's not mentioned in the CREDITS or in any source file in the Linux Kernel source (v2.4.20), so he's not a Linux Kernel Hacker, though he does post on the Linux Kernel mailing list on occasion.

RMS: Hacker, not Kernel Hacker

SCO suspends, Gartner warns

Posted May 15, 2003 18:50 UTC (Thu) by lyda (guest, #7429) [Link]

i'm not certain, but i think "noted kernel hacker richard stallman" was sarcasm.

SCO suspends, Gartner warns

Posted May 28, 2003 7:28 UTC (Wed) by rletson (guest, #11488) [Link]

I don't know if it's been posted prior but Eric S. Raymond is taking affidavits.

"If you have ever had read access to proprietary Unix source code (not just binaries and documentation) under circumstances where either no non-disclosure agreement was required or whatever non-disclosure agreement you had was not enforced."

This is in preparation to help fight the SCO vs. IBM lawsuit. He has already collected around 60 affadavits of people that had access to source code outside a non-disclosure (which by law would nullify the trade secret of the source) but is looking for more.

please see:
http://www.catb.org/~esr/nosecrets/

--rletson
Ric Letson