Slashdot Mirror
No Business Like SCO Business
The SCO must go on. Informationweek has a roundup. News.com has some analysis of the legal case. SCO reiterates their threat to revoke IBM's license. Reader hobsonchoice sends a blurb: "Also more from analysts who saw SCO/Linux code comparisons under NDA. Bill Claybrook, of Aberdeen Group Inc., says SCO changed their story to him about whether they had any "direct evidence" that IBM copied any System V code into Linux.
Laura Didio of Yankee Group has answered some detailed questions about her code review process. Lastly Fujitsu Siemens have joined in the debate: they don't think SCO's case is going anywhere." One observer of the SCO case has compiled some notes about Caldera's active participation in the IA-64 project. And look on the bright side: if you follow the school of thought that all publicity is good publicity, at least this suit has gotten Linux mentioned in many places where it normally wouldn't be.
Perhaps your new to Slashdot and don't know about many of the exciting features we have to offer you!
Like the lameness filter for commenting, Anonymouse posting for when you're violating your NDA, or TURNING OFF PARTICULAR SUBJECTS.
You may have noticed these are all listed under Caldera.
You may be sick and tired of it, but I for one am curious to see how this works out in the end. Sure it's a bit over dramatized, but it relates to me.
Computational Madness in a round package.
if you follow the school of thought that all publicity is good publicity, at least this suit has gotten Linux mentioned in many places where it normally wouldn't be.
The entire reason this is being done is to plant FUD into the minds of IT and Business community about liablity in using Linux and Open Source alternatives. MS is behind it and they are doing what they do best. Planting FUD about a competitor.
On one hand you have to give them credit. No company in the history of mankind has ever done it better. On the other hand it is a low class, childish, unethical thing to do.
Don't have much time at the moment, but in case nobody has posted these yet:
NOTE: German Language
A German developer (who says that he didn't sign their NDA!) reports on SCO's "evidence". He says that he's seen 46 pages (not just 80 lines) but doesn't seem convinced.
In another article, Claybrook gives more details of how the story changed, and also remarks on some rather odd things about SCO's "evidence".
You could tell that even if she didn't admit it. Her response to the last quation just parrots SCO's convoluted logic.
Down towards the bottom.
t_t_b
I'm on PJ's "enemies" list! Are you?
I think he meant "Ostus Gustobus"
Which roughly translates to "they eat thier young" (or something close).
I really don't want to know the details.
Read, L
ftp://ftp.sco.com/pub/updates/OpenLinux/3.1.1/Serv er/current/
They seem to have their whole distro available.
http://www.osopinion.com/perl/story/21722.html
IF the offending code is actually from the Unix System V kernel dated in early 1980s (predating the first Linux kernel), and it is not from BSD, and if it is primarily in Linux kernel 2.4/2.5 and it wasn't put there by SCO/Caldera employees THEN what are the legal consequences for this lawsuit?
/ 13/10552207 51243.html
I strongly prefer that SCO lose this case but in light of what I'm hearing now, I can't help but wonder if they have a stronger case than we all imagined earlier. Just because we wish something to be true doesn't mean it has to be true.
Here are some quotes from the interview with Laura Didio of Yankee Group regarding this:
1. 'There are "bits and pieces" of copied material in Linux version 2.2, according to SCO. However, the vast majority of their claims centre around the later Linux 2.4 and 2.5 versions.'
2. 'The Yankee Group as well as the other analyst firms and members of the press, were only shown small portions of a few pieces of code. In my case, I saw Unix System V, version 4.1. Incidentally, this particular code is from the early 1980s, and hence predates Linus Torvalds' first Linux code.'
3. 'SCO hired three separate teams of code experts, including a group from the Massachusetts Institute of Technology. According to SCO, these groups all found code in Linux that purportedly originated in the Unix System V kernel and not BSD.'
I wanted to keep this post relatively short so I suggest you read the entire interview to get better context to the quotes and to get the full story.
Source:
http://www.smh.com.au/articles/2003/06
SysV R4.1 didn't come out until '92 or '93. She is on crack.
Our legal dept has decreed that we shouldn't discuss the
whole SCO thing via email, but other than that, the company
I work for (about 150 on the fortune 500 list) is still
very aggressively considering Linux deployments for all
sorts of things.
Most people don't care about the SCO issue and those that
do are convinced that SCO has no case.
*sigh* back to work...
Even the developpers of the IBM journalled filesytem (JFS) don't have access at all to the AIX version of the same filesystem. So, I think SCO is just shooting itself in the foot.
In fact, they just try to delay customers' acceptance of Linux and try to find someone to pay for there lack of business intelligence.
Achille Talon
Hop!
SCO's stock has double in value since their Law Suit rampage.
There was no final judgement, but the judge did issue a written opinion in declining AT&T's request for a preliminary injunction. That's an indication that the judge thought that AT&T didn't have much chance of winning the case.
An indication like that can certainly encourage the plaintiff to see the wisdom of dropping the suit and seeing what kind of settlement they can get.
-- Alastair
Read rest of the artical here: http://mozillaquest.com/Linux03/ScoSource-19-Inju
And least there is a place where unfair competition acts are based on common sense.
IBM Patches for NUMA - http://www-124.ibm.com/linux/patches/?project_id=5 6
Read-Copy Update - http://lse.sourceforge.net/locking/rcupdate.html "Read-Copy Update was originally designed for DYNIX/ptx, a UNIX operating system from Sequent Computer Systems Inc., now a part of IBM."
IBM's original RCU Patch - http://lse.sourceforge.net/locking/rcu/patches/rcl ock-2.4.1-01.patch
Other kernel patches from IBM - http://www-124.ibm.com/linux/patches/?project_id=5 2
If there is any SCO code that IBM submitted to the Linux kernel, then it should be in those patches. That's if the statements by the analyst is true that SCO is claiming their code is specifically in the NUMA and RCU sections of the kernel.
There is an interesting quote in this interview from MozillaQuest Magazine back in March. http://mozillaquest.com/Linux03/ScoSource-07_Story 02.html
An article that talks about IBM's purchase of Sequent and NUMA. http://news.com.com/2100-1001-228275.html And another article about IBM and NUMA. http://news.com.com/2100-1001-233626.html?legacy=Go not unto/. for advice, for you will be told both yea and nay (but have nothing to do with the question)
No. They don't have NUMA to this day. UnixWare got SMP in '97, after Linux ('96.) They still support only 4 processors btw, although you'd never guess that from reading their complaint.
They just fed the poor woman whatever they could come up with to bamboozle her, obviously.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
Today I had the opportunity to have a look at the relevant code-sections. By a mistake on the part of the representing law office my colleague and I, in contrast to the 7 other people who received insight today, did not have to sign a NDA. Contrary to the examiners of Microsoft, who must keep silence obviously even with their own superiors, and are only allowed to report to their internal audit department.
Now, to the code:
Under notarial supervision 46 pages, each with half code from Linux (to a large extent printed posts directly from the Linux Kernel mailing list) were shown and the other half with listings from SCO. Whether it is actually SysV source is not possible to determine, since they are taken out of context. Also it is interesting to note that all dates were removed from both, even from the comments.
The comments are actually identical in parts, even some jokes are the same on both sides. Remarkable is however, that the code before the comments of the most matching parts is quite different. The fundamental structure of the relevant functions is similar, the concrete implementation however quite different. Variable names and functions names are different, loops are structured differently, conditions are implemented through chains of conditions or bit-patterns. In the end only one thing can be said with confidence: The functionality made available by the respective code sections is often alike, which was to be expected. In the concrete implementation however, there are so many differences that a proof of the same origin will be difficult, although not impossible.
The crucial point however is a function of the scheduler, which is about 60 lines, which is actually identical apart from small differences. There are also many matching comments. Comparable similarity exhibits only one routine of the memory management, however here only in the Linux version there are comments at all. Whether with this two matching sections a judicious proof can be provided, only a lawyer can tell. The vague similarities of other parts are insufficient in my opinion, since both were written based on the same standards so certain similiarites are to be expected.
In regard to the identical comments however, I'm unable to find an explanation. That would have to be examined again more closely in any case, in particular with included dates. Because only with them a breach of copyright could actually be proven. In regard to the discussions about the parts of Linux distributed by SCO/Caldera under the GPL, it has to be said that no court ever actually confirmed the validity of the GPL.
If it is found to be valid however, SCO can only use the parts of Linux it didn't publish and wasn't involved in developing. This is another difficulty I see in the coming trial.
But since the original, unpatched Linux source hasn't been attacked, but only modifications included in different distributions, it has to be determined if they possibly own the rights to the said parts, either direct or indirect e.g. by buyouts or "all-inclusive-deals". Chances for an actual trial aren't very good, since most of comparable cases settled out of court.
This is however only my personally view, relevant in the end is only the decision of of the representing lawyers of both sides and/or eventually the responsible court.
My german is a little rusty,
It looks like he's talking about the memory manager and the scheduler for the most part. Based on the interview on CNET I read today it would seem that is corroborated by the SCO execs who are telling us EXACTLY where the problems lie. They are so damn stupid.
Today I had the opportunity to look at the [incriminating?] code snippets.
Through a mistake of the representing Law firm, my colleague and I did not have to sign an NDA, unlike the other 7 representatives that also were viewing the code today. This is in stark contrast to the Microsoft representatives who apparently even had to maintain silence with their own supervisors, and were only allowed to report back to their internal [Legal?] department.
Now to the code itself:
46 pages each containing one half Linux code (largely printed posts out of the linux-kernel lists) and one half listings from SCO were presented under legal supervision. It is therefore not possible to tell whether this actually comes from SysV-Sources, as they are taken out of context. Also interisting is that all dates are taken out of both parts, even out of the comments.
The comments are in fact identical in places. Even some of the jokes are the same on both sides. What is apparent, though, is that in the most similar places the preceding source code is quite different. The basic structure of the affected functions are similar, but the concrete implementation is quite different. Variables and function names are different, loops are structured differently, conditions run on chained conditionals or bitmaps. All in all only one thing is sure: the functions presented in the code-snippets were often the same, which was to be expected, though.
In the concrete implementation there are so many differences, however, that a proof of the same origin will be hard to construct, albeit not impossible.
However one function of the scheduler presents a [breaking point?], as except for minor differences it is identical. In this case there are also a whole row of matching comments.
Only one routine of the Memory Management offers comparable similarity. In this case, however, only the linux version has comments.
Only a lawyer could safely judge whether these two similarities alone provide proof enough for a verdict. The vague similarties in other parts are, in my opinion, insufficient, since a certain similarity is to be expected as both pieces are based on the same standards. On the other hand, I have no clue where the identical comments in different code could come from. In any case, this should be researched more closely, especially with the dates restored. Only with these would a copyright infringement be provable.
As to the discussion of the piece of Linux sold by SCO/Caldera itself under the GPL, one has to take into account that no Court has commented on the enforcability of the GPL yet...
[Sorry ran out of time. I will try to get to the rest later. Perhaps someone else can translate the rest.]
Here you will find the pdf of the Linux Kernel Internals, authored by Tigran Aivazian (tigran@veritas.com). Now, he has been submitting patches to the kernel for a long time.
He submitted patches for (among others)
Microcode updates
...)
iBCS patches
kgdb patches
Linux Implementation of SCO UnixWare BFS
and I'm sure a lot more, across a wide range of kernel versions (2.2/2.3/2.4
Why does this matter? Well his email used to be tigran@ocston.org. odd domain name, try reversing it.search and look at the first two results, then look here for more info about the first entry.
Before that his email was tigran@sco.org, but he
got a little paranoid
about it.
Searching google brings up patches supplied by him throughout the whole development cycle of 2.3/2.4 and more. He is directly connected to the author of the LKP on SCO Unix, draw your own conclusions here.