Slashdot Mirror
"Stolen" SCO Linux Code Snippets Leaked
stere0 writes "An article (in German) published on the German IT news site Heise includes two pictures (1, 2) of the "stolen" source code SCO claims to be theirs. Part of the first screenshot has been scrambled, the font has probably just been changed to Symbol; can anybody decipher it? I searched for the code snippets on Google. The code does indeed come from the kernel; the photographs show what seems to be lines 88-102 and 109-123 of /arch/ia64/sn/io/ate_utils.c from the 2.4 kernel tree. " Update: 08/19 16:39 GMT by M : LWN has a nice piece tracing the origins of the disputed code, and showing that SCO is simply lying.
If so, how can SCO demand that we give them money for code that's distributed but that 99% of linux users ARE NOT USING?
This is exactly why they want you to sign your life away by signing a NDA before they will show you the code. They want to use this to bludgeon people into settling BEFORE IT GETS TO COURT . They are not interested in legitmately rectifying the situation.
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
this is code that was contributed by Caldera employees and thus released under full SCO Group knowledge to Linux..
So where is the magical proof that McBride keeps claiming that he has?
I smell a fraud lawsuit against McBride on the basis of both Federal and State BlueSky Laws on the basis on making false factual public statements that investors relied upon to buy SCO Group stock..
and Boise should know better than to perpuate false information about the laws and regs on software copyrights!
Don't Tread on OpenSource
My reaction is "so what." I wouldn't be surprised if you saw those same lines in NT. They probably originated in BSD as so many others have stated and will continue to state. If it is true Caldera sent an employee or two to IBM to help *beef up* Linux, then that would be a valid explanation as to why the code is the same. SCO is Caldera and they cannot deny that no matter how many times they change their corporate name. They put the lines in there and they distributed the offending versions of Linux under the GPL. Just because they are no where as successful as RedHat or SuSE gives them no rights to try to weasel out of it now... When will SuSE, Xandros, and Lindows join the RedHat lawsuit against *Caldera*???
"Right now, somewhere in this world, Scott Baio is plowing a woman he doesn't love," - Peter Griffin, *Family Guy*
Obviously they were silly enough to believe that since they had every legal right to copy it, they didn't need to hide the copying.
No one expects the spanish inquisition!?
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
It is so great that everyone here in the /. community is so on top of this. It's great that so many of you know where to look to find the true origins of the "stolen" code, that by today's evidence, is obviously not stolen.
However, this is not yet the time to celebrate. SCO is claiming 829,000 lines of code was "stolen" from SMP code alone. Of course this is probably ridiculous, but a screen shot of some comments from the late 70's only shows that those particular comments were not stolen.
There is still a lot of work to do. Mr. McBride is creating so much work because for each claim of copyright, the onus is going to be on the linux community to find the origins and prove the allegations wrong. SCO is only going to present SCO code that was supposedly 'written' before the linux code. Their entire offense is going to rest solely upon the fact that they have a plaintext file with an earlier date than the linux kernel's corresponding code file.
The work is going to be on our backs to locate even older code that SCO's predecessors used to write SYS V. I would raise the bar as well and go so far as to attempt to show that SCO's code was itself misappropriated.
We are just now starting to see how much work we have in front of us, and believe me, that mountain of work is only going to get larger. But, as with the development of linux itself, there are millions of developers across the globe that will be able to find evidence to refute each and every one of their fraudulent and baseless claims.
Anyone ever given serious thought that perhaps the SCO code was lifted from the Linux source?
Or maybe both came from a third party. Especially considering that the Linux version of the code is marked "Copyright (C) 1992 - 1997, 2000-2002 Silicon Graphics, Inc."
Face it, that has yet to be proven. Even if the screen shots provided are correct, it has yet to be determined who put those comments in each code and when. SCO could have just as easily inserted them in their code at the time because it was easier than developing it themselves. Or perhaps they inserted the code intentionally so that later they could say "See? It's the same." Or maybe SCO contributed the code to their Linux distribution? Or, yes, perhaps someone took it from SCO inappropriately and inserted it in Linux--in which case THAT PERSON (or company) should be SCO's target, not Linux and Linux users worldwide.
A reasonable advocate would be working on a method to right now to find coders who have NEVER seen either the SCO code, the licensed IBM code or the stolen Linux code and begin a process of writing true black-box replacements.
And I'm sure that as soon as SCO acts reasonably and friggin' tells the world what sections of code they have a beef with, that's exactly what will happen regardless of whether SCO's claims are valid or not.
Stolen? Stolen from where? Showing two identical blocks of code in two different OSes proves nothing. SCO has to prove that it is the rightful copyright holder of that code *and* it has to somehow weasel out of its release of that code in the Linux kernel under the GPL. If that code originated in Linux first, SCO is out of luck. If that code originated from a third party and was taked by both Linux and SCO, SCO is out of luck.
But once we see what code is in question, finding the original, rightful copyright holder is the easy part. And if the holder isn't SCO, SCO is out of luck. That's why SCO has been so afraid to show it in public.
Ita erat quando hic adveni.
That would explain the "register" variables. That keyword has been ignored by compilers for a long time, and so when you see it in code, it is almost always old code, copied from somewhere.
Call me paranoid....but this might be a trick by SCO to probe the defences of the Open Source community, by having us do their historical code research for them, gratis.
What do I mean? An example.
I used to be a magician - a classic trick in the magicians arsenal is called the "sucker trick"
In the sucker trick one does a seemingly stupid trick. As people start to think they have figured it out, the bright (and loud) ones start yelling how they think it works. Then, PRESTO, the real trick is revealed!
IF you do it right, people are amazed and impressed, and more importantly, you have identified the hecklers in the audience, who often remain quiet the rest of the show out of embarrassment.
I know this sounds paranoid, and you might think ol' Darl is no magician, but he has conjured ~ 20X increase in SCO "worth", from an essentially worthless company.
Just a thought.
I think, therefore I thought.
Not exactly!
It may show that there is identical code in Linux and Unix, but that in no way "proves" the code was stolen from the latter! The code may have come from BSD, it may have been stolen from Linux and copied into Unix, some of it may be OEM code that was released by a hardware vendor to many platforms with the same comments but slightly different actual code. There is no way possible to determine any of this with what pathetically little has been shown.
Which returns to the point that most here have. If this the all they can show - they've got crap for a case! If they have some "smoking gun" type example, then show it so the matter can be resolved. Using "smoke and mirrors" to extort money from Linux users is NOT an acceptable tactic.
I AM, therefore I THINK!
And so they think that everyone else is too. It's the classic syndrome called "projection".
The SystemV code shown is mroe recent than the Linux code, with added comments. No-one, ever, removes comments when copying code.
All their presentation shows is that the two functions have a shared pedigree, and this code is so old that the pedigree can be found in at least two books, and multiple versions of Unix.
SCO are lying, thieving, scurilous rumour mongers and sadly getting much too much attention.
Which makes me think: could the whole thing be simply intended to distract our attention from something else happening...? It is a classic ploy.
Ceci n'est pas une signature
Don't forget SCO's Linux licensing program.
The code which SCO showed here does not appear to have been donated by IBM. In the Linux kernel it was marked with an SGI copyright.
The fact that IBM donated code to Linux may (or may not) give SCO a case against IBM. However, since that code was not written by SCO, for SCO to claim that that code gives SCO any IP rights to Linux is very tenuous.
SCO is showing this example of direct copying from Unix to Linux to show that SCO has IP rights to Linux, thus justifying their Linux licensing program.
Mind you, since this code has already been removed from the Linux kernel, it looks like it's not going to help the Linux licensing program much. Of course, SCO claims to have other examples. They're probably worth about as much as this one.
On further investigation, it appears the author is none other than Ken Thompson. See V5/usr/sys/ken/malloc.c.html for further details.
Of course, Ken might have lifted this from even earlier sources.
Regards,
--
*Art
And you really think that every user of Linux, every vendor and every company should bet that all 890,000+ lines of code come from 1979 or earlier? Do you really think UNIX Version 7 in 1979 had a NUMA implementation?
And has SysV or any version of UnixWare / OpenUnix had a NUMA implementation? As far as I know the answer to that question is a big, fat *no*. This seems to be the crux of the SCO headfake: It isnt' SCO code to begin with. Apparently, most (all?) of the code in question is IBM's (by SCO's own admission). If IBM submitted it to the Linux kernel it isn't exactly "stolen" since its hard to steal something that was given to you.
Of course, SCO doesn't frame it in those terms. They *may* have licensing rights over certain code assuming that a) the code in questions is deemed by a court to be derivitive of SysV code and b) the licence IBM and AT&T signed governing the SysV code is binding (in the way SCO claims it is), but the NUMA, RCP, etc. implementations are most certainly not their code.
*If* there actual SysV code found in Linux (that is copyrighted SCO/AT&T/Whoever) in Linux then they still have a bunch of problems. 1st being they seem to have released all the old legacy stuff under a BSD license, not to mention the whole AT&T vs. BSD which pretty much kills any of their claims.
To sum up: if its old code chance are SCO has no claim to it due to the AT&T case and the fact they BSD'ed a lot of stuff. If its new code chances are its not SCOs to begin with.
I find it sad to see how many here call sharing code for stealing. Without sharing code, there can be no further progress on computer science. Instead of having ad-hoc solutions, it can evolve into a fully fledged engineering science. But only if people can collaborate on standards and further its progress instead of being busy putting up tool-booths for inventing the inevitable.
You never drive over a bridge proprietary to BigCorporation(R)(TM)(C). You drive over an assembled construction errected by standardized plans, tools and mass. Instead, we have a mad goldrush that sinks the economy through the floor.
Sad.
http://www.debunkingskeptics.com/
In my school, the people in CS who got the 3.75+ GPAs were the professional students, who were only willing to learn what would help them on the exams, but not anything useful. If it wasn't going to help them on their exam, or help them complete projects. they simply didn't want to know about it.
The "real" geeks who really cared about CS, didn't always score quite so high, but they had a passion for computers, and therefore learned things outside the curriculum and picked up more useful skills, tended to spend their free time "tinkering", and therefore their grades in other requisite Liberal Arts courses may have suffered a bit.
At one point, we had a professor for an "Operating Systems" course, who had lots of real world experience, and his teaching style was less academic and more focused more the real-world. This drove the "3.75+ professional students" crazy. They didn't know how to study for his course, because they actually had to think in ways they weren't used to. His course threatened their GPAs, so they protested. The "geeks" loved his course and got straight A's in it. Too bad the instructor was a bit of a push-over on grading, and ended up bending to the other students' demands, and ended pushing up their grades more than they deserved.
I'm not saying that everyone who has a high GPA is this kind of student. I'm just saying I wouldn't decide who to hire based on GPA alone, from on my personal experience.
By reading this sig, you agree to the terms of my sig license.
I don't even want to try and figure out the web of licences, contracts, and original sources for this code.
Is it just me, or is anyone else getting the impression that it's corporate coders working for proprietary software companies whose coding practices are sloppy and reckless about intellectual property, and not us long-haired hippie commie free software freaks?
Proud member of the Weirdo-American community.
Also don't forget the shorts. At some point shorts have to buy the stock back, which can boost the price of the stock. I don't think short covering is a big part of the current holding value of SCOX but it does tend to mitigate the downramp a bit when downramps happen.
In short, SCOX is a highly speculative and volatile stock and any simplistic view of cause and effect in the matter is no more accurate then rolling the dice.