Notes from the SCO conference call

Spefically, it's very plausible that both originated with BSD. I really hope this is what happened, and that SCO comes out looking moronically stupid.

I also won't be impressed by their under-NDA reviewers saying "yup, looks copied" until we know something about their credentials and what they saw. Unless it's substantial blocks of code, it's very easy to see the identical expression of small pieces of code written completely independently. E.g., if you're writing a for loop in C to sort an array using a basic selection sort algorithm, what are the chances that it will look identical to the code that I wrote to do the same thing, even if we never talked to each other? There's a non-trivial chance that we'll both use i and j as our index variables, for example, since so many people do that.

-Rob

One analyst says the source she saw included duplicated comments. That certainly implies some transfer one way or the other.

It should be easy to trace the ancestry of source code comments. If SCO wanted, they could pick just one of these comments (a couple of sentences would do), and ask everyone to look for them. There would soon be a table going back in time showing which versions had the comment. Then we'd know who was borrowing from whom.

But SCO must be terrified that any examples they pick will turn out show that UNIX borrowed from BSD or Linux, not the other way around. That's where the NDA helps them.

Also, since their goal is to destroy Linux, they want to say Linux is violating copyrights without giving the community a chance to fix it. Hiding the details behind the NDA lets them do that. They speak of only handfuls of lines of code that were copied. Those could be rewritten in days, and then you have a clean, court-approved Linux again. And SCO doesn't want that.

So why does the NDA specifically forbid anybody from revealing line numbers in the Linux source code? Linux is already open, so it couldn't hurt confidentiality, right?

Not so with SCO's UNIX code. As it is closed
source, how can one be sure that precisely
this code with precisely these comments were
in the source of UNIX version a.b.c? There
is no "central depository" for the sources of
private companies; I mean, a depository on a
national level, not controlled by companies.

Being not-a-lawyer, it seems to me that they
(SCO) have to show the whole source tree, the
compiler they used at that time, and that indeed
this source compiles to binaries that are
*bit for bit* identical to the binaries of the
UNIX release in question. Of course, this leaves
a question on the comments, since the compiler's
output does not depend on them at all.

SCO's evidence of origin and Function dictates form

What proof did SCO present for the origin of both fragments of source code?

Was the SCO code originally from old BSD or Linux?

Who put the SCO source into Linux? - Was put there by Old Novell/SCO/Caldera in the first place?

What proof did SCO provide to show that the person had access to SCO's Unix sources?

The latter question raises another issue. The similarity is just as likely to be due to both operating systems performing the same role. Form is often directed by the function it performs. Function and variable names are often dictated by the API and common terminology.

Both the current Linux and Unix kernel developers have attended the similar university courses and read the same publicly available documentation. The works of W. Richard Stevens are very influential as a reference toward modern Unix and Linux and have dictated the implentation of APIs and TCP/IP stacks in both.

Levels towards DEF-CON 5

I don't believe that SCO could hope to even start a court case against IBM until the copyright in question has actually been transfered to the SCO Group. Any chance that Novell will hang tough?

Effectively this still means is that, before the SCO Group can even begin to sue Linux developers,vendors and users, SCO has to :-
1) Sue Novell - to tranfer SCO ownership of the copyrights ( forgetting the patents ) AND
2) Prove in a court of law that the source code in question did in fact originate from AT&T,Novell or SCO itself and was not copied from old BSD or Linux, AND
3) Sue IBM - to prove that IBM planted Unixware IP in Linux in breach of their contract with IBM, AND
4) Break the same GPL license in court of law that Caldera has been collectively developing Linux under since the groups forming in 1994 ( which predates even Caldera's purchase of SCO and old SCO's Monteray deal with IBM ) and that SCO made the choice to sell and distribute the code under.

SCO has to achive all this before even beginning to go after other Linux distributions, developers and users.

For (2), IBM under the terms of reasonable disclosure would have the right to demand copies of the source code going back to Version 7. IBM could then disclose the amount of Old BSD and Linux source code that has slipped into SCO's non-Linux products.

For (3), Unless the SCO Group get (1) and (2) it cannot precede with (3). Jim Stallings, the general manager of IBM Corp.'s Linux business recently stated: "I believe I am correct in saying there are no violations on any intellectual property issues [with Unix and SCO Group] and we will continue to support our Linux customers. It will be business as usual." Frankly IMO, given the quality of IBM's legal team, SCO don't stand a chance.

For (4), the GPL has never been succesfully challanged and has been recognised in a federal court. Even if SCO break the GPL, it means that EVERY developer who contibuted GPL'ed content to the Caldera/SCO distributions and is used with the CalderaLinux or SCO Group's Unixs, can persue a case against the SCO Group using the same loopholes the SCO Group has opened. Either way the SCO Group is F****D.

GPL clauses 5,6 and 7

GNU GENERAL PUBLIC LICENSE [ with emphasis added ]

...
5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.
....

Also reread clause #7

7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.

Note the phrase "imposed on you". You cannot use paragraph seven as a right to deny an effective automatic patent( or trade secret etc ) unless imposed to do so by a third party.Since SCO claims ownership of the infringed IP in question, and Novell or IBM is highly unlikely to force SCO to stop distributing it in Linux, SCO cannot deny recipients full effective license to the IP,patents,trade secrets,copyrights in question.

If even one of the blocks in question (that SCO submits as evidence because it is in their source too) had this "feature", then they too would be infringing on a copyright. What would they do then?

I posted this previously with another article and was hoping to get some feedback..

Thanx McGeek

Posted Jun 6, 2003 23:31 UTC (Fri) by NerdlyMcGeek) (Post reply)

Darl McBride doesn't understand Linux (NewsForge)

And not much about anything else....

It is interesting to note that Lindows made an agreement with SCO for this supposed "IP" some time ago. But they somehow just recently discovered this "IP" problem? I am very interesed as to exactly when did SCO shake down Lindows? Might I humbly suggest that the Lindows folks be issued a sopeona to testify in this regard. Should they have made an agreement prior to what SCO "claims" they became aware of these purported violations they not only loose their case but could face criminal prosecution for fraud as a result of manipulating this information with criminal intent. In the eyes of the Courts World over, omission and lies equate to the same thing.

(-;

However, as an SCO licensee, would not IBM have the SCO source code and be able to do a full analysis as to the similarities and where exactly they occur? They most certainly have the computing horsepower to check it every way possible, and in a flat hurry. Assuming this to be true, we can safely assume that IBM knows of each and every match and similarity, and has already done considerable analysis on the results. Perhaps this accounts for their apparent unruffled confidence going into this legal battle. Something tells me that they know what SCO sees, and already know better than SCO where it all came from. 4.4BSDLite? FreeBSD? Caldera during the cooperative period? Hardware manufacturer driver code? Sample code from Knuth? The Dragon Book? Some issue of the ACM Proceedings? We may not know, but IBM probably already has the paper trail to prove the source for every similarity that good software tools could find. IBM is nothing if not thorough in legal matters. Thus, their confidence in the outcome. IBM knows, and is keeping that as a closely guarded secret to be released in due time.

This also explains SCO's strained unwillingness to point out in public the specific parts of the code which they accuse of being copied. The distributed institutional memory of the entire UNIX/GNU community will find the true source of any similarities in due course (and rather quickly). If the common source is from some published literature, it will be found. If it is from some third party, that will come out. Someone remembers, and the truth will come out. SCO needs the theater to push stock prices up, and so does not want an early resolution.

(By the way, could this access to the SCO source code apsect be another reason Microsoft recently bought the SCO license? They are most assuredly curious to know the outcome before the trial, and this would give them a leg up. This foreknowledge requires access to the SCO source code, which Microsoft has just purchased.)

Current SCO management does not appear to understand software development particularly well. (The modern business schools seem to teach that a good manager can manage absolutely anything - a naked lie, but a lie which many of us have witnessed firsthand - i.e. the PHB.) There has arisen a generation which knows finance better than technology,. and this crowd is in control. As such, the possibilities of what actually happened are not apparent to them, and they can only see one-way transfers of their precious code to the unworthies, and not that it well could have occurred the other way around. The fact that CVS repositories could actually prove such a thing has escaped their notice. Discovery should be a hard time for this crowd.

All we see now is posturing and bluff. But in this battle I would say that IBM is actually the cat and SCO the mouse. IBM now gets to choose whether to go for the quick kill, or play with their mouse for while first. SCO will go down - and will not need to be bought off to do so. It is merely a matter of what torment they will endure in the process.

I now see a second motivation (beside its wanting to appear
"open") for Microsoft to decide to "share" its source code
with some of its customers. Anyone who has seen that code
is now contaminated -- they can no longer safely write free
software!

My conversation with will be very short. It will begin with, "I went to my
local MicroCenter and bought Mandrake PowerPack 9.0. I didn't know it was stolen!"

Once SCO gets their $1 billion, are they going to make an Ashcroft-style sweep of all MicroCenters, CompUSA's, and the like to locate every last Linux customer? Oh, and that thick book of Linux source I saw in the MicroCenter Books Department: are they going to send out secret agents to storm every buyer of THAT tome? After all, it IS a book of Trade Secrets!

This could be a BIG opportunity for the FreeBSD people to polish up their distribution, get some more documentation out, and help replace all those cute penguins out there with cute red daemons.

In spite of the interlocutory injunctions issued by district courts in Bremen and Munich, a German SCO web site still displays the prohibited statement that Linux contains illegally obtained SCO property.

"For them to ignore the court orders is outrageous. Evidently they intend to continue their strategy of intimidating GNU/Linux users and the Linux community," said Michael Kleinhenz, spokesman for the LinuxTag association, about the behavior of SCO's German subsidiary.

After a first injunction against SCO had been obtained by the system supplier univention GmbH in Bremen on May 28th, 2003, the software and consulting company tarent GmbH obtained another injunction in Munich's first district court on June 5 in spite of a caveat filed by SCO.

"The Munich I district court determined that SCO had not plausibly demonstrated that the Linux kernel violates SCO's rights," said attorney Dr. Till Jaeger of the firm of Jaschinski Biere Brexl, which represents tarent GmbH in this matter. Because SCO continued to publish the disputed claims, tarent GmbH requested that the company be fined. "We can't simply allow them to confuse GNU/Linux users, causing damage to Linux businesses," said tarent's Managing Director, Elmar Geese. "We hope our initial success will help to demonstrate that Linux can be used dependably. As a company that profits from Free Software, we feel it is our duty to take action against SCO."

Klaus Knopper, developer of the GNU/Linux distribution Knoppix, criticized the strategy behind SCO's claims: "What kind of respect is SCO showing for the intellectual property of the many Linux developers who have worked to develop an operating system that is free and open to all -- on which SCO's own Linux distribution was based, by the way? You can read on the SCO website that they have used and continue to use other GPL software in their products," Knopper added.

source: http://linuxtag.de

My understanding is that a large amount of the system 5 code is out in the wild, or that at least IBM, Novell and others have a copy of it, so would not a simple shell script that parses the system 5 code, extracts all the comments, and then searches for them in the Linux kernel code be easy to implement?

Why has this not been done? Am I missing something?

Take a look...

http://uptime.netcraft.com/up/graph/?mode_u=off&mode_w=on&site=www.sco.com&submit=Examine

It seems to me that they are evidence against intentional copying from SCO. Someone knowingly stealing code and trying to cover his tracks would probably change or eliminate all comments first, because they are the easiest to modify without taking any risk that the functionality of the code itself breaks. So whoever copied the code (in either direction, or from some third source) probably thought it was OK.