IBM Ordered to Show More Code to SCO

editingwhiz writes "Bob Mims of the Salt Lake Tribune has the scoop straight from the courthouse steps: 'A federal magistrate has handed a partial victory to Utah's SCO Group, ordering computer giant IBM to turn over more of its Linux operating system-related program codes. U.S. Magistrate Brooke Wells' ruling, released just minutes after Salt Lake City's federal courthouse closed Wednesday, came in the Lindon software company's contractual suit stemming from Big Blue's alleged distribution of Linux applications purportedly tainted with SCO's proprietary Unix code.' If at all possible, SCO's going to be even more insufferable now -- it has a glimmer of hope."

Comments

Even more code?

[Mongo222]

How can you turn over "even more" of something that's already open sourced?

Re:Even more code?

[DjReagan]

Because the code they're being ordered to turn over is for AIX and Dynix, which isn't Open-Sourced.

Re:Even more code?

[DjReagan]

SCO are alleging that IBM contributed code from AIX and Dynix into Linux, and that this contravened the licensing agreement that IBM had for using Unix System V code, which is the basis for AIX and Dynix.

The code they are alleging was copied however, was written by IBM, but incoporated into AIX and Dynix before being put over to Linux. So what it comes down to, and what is being tested in court is whether the Sys-V contracts that mention control of code can be extended to other code that is later incorporated into IBM's derived product.

Re:Even more code?

[Anarchofascist]

IBM had already handed over all of AIX and Dynix. SCOG/Caldera's lawyers have the source code to every version of AIX and Dynix ever released.

NOW SCOG/Caldera's lawyers have told the judge - "nope we can't find any infringements here, we need more" and the judge has ruled that they can have every version of AIX and Dynix which ever existed in version control including all notes made by anyone who worked on the project.

This is just another delay... in part because IBM's defence was that producing all the version controlled source was too hard and would take to long. Now they have to work hard and take long or get branded as liars. They'll probably just run through source control, and for ever single commit they will burn a stack of source code CDs. I expect that will run to a few truckloads of CDs which SCO/Caldera will then decide will take years to sift through... and blah blah blah the glacial US legal process drags on until SCO runs out of money.

But never fear! Even if there is code there which ended up in Linux, they don't have a prayer. IBM owns a developers license to Unix (bought from AT&T) to modify Unix and sell it as its own product AIX. SCO (which bought the license rights) says the license also says that new code IBM has added to Unix to produce AIX cannot also be donated to Linux, but that's just plain nuts. If they write B and insert it into A, how can the license agreements with A prevent you from adding the same code to your own product C?

Re:Even more code?

[ThosLives]

This is one of my beefs with the "derivative works" clauses in many software products. There's this strange idea of what is and what isn't a derivative product.

What is seems is that IBM may have gotten System V from SCO and then added some features that were original creations of IBM (to create AIX/Dynix). IBM is then alleged to have given those new features to Linux. SCO is claiming they own those "new" features created by IBM because of some wording in the System V contracts.

Now, we can all reasonably conclude that this is stupid; if I buy a car and design a new fin for it, then sell the fin design to others, the car companies don't claim the fin design is their property. However, when we purchase a car we don't sign a "derivative works" contract with the car dealer or manufacturer.

The way I see it there are a couple of things that could happen. One is that the original contracts may be found to say that derivative works are property of SCO, in which case IBM did not have the right to give them away for use in Linux - and I would say IBM was dumb to sign such a contract and should legally pay. However, even if the contract was written for that scenario, the court might find that such a provision is "unreasonable" and is void (similar to the car example). Another way it could go is that IBM somehow took code that it did not originally create but thought it owned and contributed it to Linux. This is a little more tricky because it's basically an "oops, we didn't read the original contract correctly".

Out of those scenarios, the latter is the worst because it would mean that [Linux] is infringing. In the former case, either SCO owns the derivative works or doesn't. I would argue that, if SCO is found to own the derivative works, IBM charges them some large amount of money for the development of those new works. Probably to the tune that SCO is filing in the suit, since that is the apparent value on those features. After all, IBM did not develop those features gratis for the benefit of SCO!

All in all, I think all sides (SCO, IBM, the Linux crowd) are focusing on the wrong things here. Litigation never added value to society at all and sucks up resources. I definitely agree with the camp that thinks SCO would be better off devleoping product than litigating.

Re:Even more code?

[jeffasselin]

No, quite wrong.

If you yourself wrote product A and B, you can relicence your code for product C any way you want.

OTOH, if *I* take code from your product A to make product D, then yes product D has to be GPL.

Uhm...

[Black Parrot]

Doesn't SCO know about kernel.org?

SCO loses the waiting game

[bigtallmofo]

This is just another meaningless delay of the case. Such things are actually a victory for IBM. IBM can afford to have this drag on forever. SCO's days are numbered.

Now's the time to sell.

[AceCaseOR]

SCO's stock price is going up

Re:Now's the time to sell.

[metlin]

*adjusts tinfoil hat*

You know, you're probably joking but you do have a point.

Perhaps the whole point of this was to have an increase in SCOs stocks so that folks have the opportunity to dump and jump the ship.

These days when laws are written by the rich and bought over by the powerful, you never know =)

Nothing to see here really

[Psionicist]

To quote Groklaw:

Finally, it has arrived, Judge Brooke Wells' Order on SCO's Motion Re Discovery. It's annoying because she enables more delay, but other than that it is a pretty normal discovery order. SCO doesn't get access to CMVC, they do get more code and they get not all programmers' notes but some. She postpones any decision on production of documents from top managment. Keep in mind, she isn't the trier of fact. That is Judge Kimball. She is the Magistrate, so it's not her job to decide who is right or wrong. Her job is simply to make sure everybody's cards are on the table.

Guilty until proven innocent?

[FortKnox]

IANAL, far from, actually, but shouldn't it be the other way around? Shouldn't SCO be releasing code to an independent party to determine if its copyright has been breeched? Or will they keep requesting more code and fish around for something they can 'try to claim' is a copyright violation?

meanwhile...

[kidlinux]

Has SCO been ordered to show any code of theirs that they claim has been infringed upon?

So IBM turns over some code, and SCO says "yep, all of that is an infringement, pay up!" How do we know otherwise?

It's simple

[$RANDOMLUSER]

SCO couldn't find anything conclusive in the 900 milion lines of code they've already gotten from IBM, so they have to fishing again.

Uhh Darl, you're fifteen minutes are up.

Re:It's simple

[Zocalo]

so they have to fishing again.

Actually, when it's a scam like this, it's usually spelt "phishing".

Re:IBM should just pay them off

[chris09876]

Paying them off would admit that SCO's got a case. By fighting them (which they definitely have the resources to do), they're sticking up for the entire community. Way to go IBM!

It's not about linux kernel

[Baramin]

from TFA :

Wells refused to grant SCO complete review of all of the IBM programs it listed, but threatened to grant "unfettered access" in the future if IBM fails to provide all data - including approximately 2 billion lines of code - from its AIX and Dynix systems

If Ya Want It...

[Bios_Hakr]

I'd pipe the code into a *.doc file and make the font about 48ish tahoma script.

Then print it on standard paper.

Hey, SCO, you might wanna bring a truck.

Re:On the bright side...

[TubeSteak]

I kinda agree with the AC, SCO is just getting a little more time before they will have to pony up some information.

btw- Magistrate != Judge
I'm not sure Judge Kimball would have agreed to SCO's motion, but I'll give the Magistrate credit for this zinger:

To the extent possible the parties are to agree upon the 3000 designated individuals. However, based on the record before the court it appears that IBM has better access to information regarding who made the most changes or significant contributions to AIX and Dynix. Therefore, IBM is to file an affidavit detailing the process by which the 3000 were chosen. Once again by requiring this, the court seeks to circumvent the rote objection by SCO alleging that they did not get enough information.

No Copyright Code in Linux

[RichMan]

SCO has already failed to produce any real evidence of Copyright code in Linux. Nothing produced by SCO from any of this evidence can change that. Everyone using Linux is safe.

This request is to support SCO's weird derivative tale that despite AT&T contracts saying IBM was free to develop code, Novell waiving the rights, and testimony from various people that IBM is bound by contract such that any code that touches SCO code is still controlled by SCO.

The Chinaman is not the issue, Darl!

[WormholeFiend]

Why would SCO need to analyse IBM's code, if the reason for the lawsuit is in the first place because SCO knows IBM infringed on SCO's code?

I mean, in order to know there is infringement, you'd have to have already figured out what part of the code was "stolen", right?

Re:The Chinaman is not the issue, Darl!

[arkanes]

The argument is something like this: SCO has inherited certain control rights over UNIX code. AIX and Dynix are based on UNIX. We think that IBM donated parts of AIX and Dynix to Linux that they shouldn't have. But because we don't know whats in AIX and Dynix, we need to get the source for all of those to compare to Linux and see what matches. Furthermore, because we already did this and nothing matched, we need to get all the intermediate versions of AIX/Dynix, because some programmer might have written something that never actually was released, and THAT then got copied into Linux.

Yes, it sounds ridiculous. And note that for it to make even the semblance of sense you have to totally ignore all the public claims SCO has made about knowing that there's copied code.

Some Encouragement for SCO (and something bizarre)

[judmarc]

The magistrate (who is presiding over pre-trial discovery, not the trial, where a judge will be in charge) noted that the contract between SCO and IBM may well give SCO more rights to the code than copyright law does. Thus IBM might be liable to SCO for putting code in Linux that doesn't violate any copyright SCO may have. IOW, Linux would be OK, IBM would not.

Another theory advanced by SCO that the magistrate did not dismiss out of hand is the possibility that internal IBM AIX and Dynix code changes may show a path between SCO code and Linux. IBM has been contending that if a bare comparison between Linux and SCO code shows no obvious copying, anything IBM did internally is irrelevant.

The bizarre bit is a footnote paying tribute to New York Giants football player Mel Hein that (to mix sports metaphors) comes from way out in left field. It's also ungrammatical ("He played 15 seasons going 60 minutes a game without nearly any rest"), which I'm definitely not accustomed to seeing in Federal court orders.

Re:As far as I can tell...

[andycat]

SCO's argument is far more bizarre than that. They claim that since AIX and Dynix might have touched UNIX code once upon a time, all code ever written for AIX and Dynix since then must be covered by the contract terms of IBM's license for UNIX. Those terms talk about "methods and concepts" as protected entities. IBM used some of those allegedly protected methods and concepts when, say, they moved JFS over to OS/2, then implemented the Linux version of JFS using that as a reference. It's not about AIX. They're trying to argue that Linux inherited from OS/2 inherited from AIX touched UNIX, so (of course) they own the things in Linux.

Their theory of derived works is totally at right angles to reality. That isn't the way it works. However, Judge Wells is not permitted to smack them down. That falls to Judge Kimball (the trial judge) and, perhaps, a jury.

This is an annoying delay but really isn't going to change the outcome of the case.