decoration decoration
Stories

GROKLAW
When you want to know more...
decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books
Your contributions keep Groklaw going.
To donate to Groklaw 2.0:

Groklaw Gear

Click here to send an email to the editor of this weblog.


To read comments to this article, go here
Transcript of the March 7 Hearing in SCO v IBM
Thursday, March 15 2007 @ 02:43 PM EDT

Here is the transcript of the March 7th hearing in SCO v IBM, the last of the summary judgment hearings transcripts. Thanks yet again to Chris Brown for arranging to obtain the transcripts.

On this day, Kimball was quite busy. He heard several motions, all the ones left over from the first two hearings on March 1 and March 5:

  • IBM's Motion for Summary Judgment on its Claim for Declaratory Judgment of Non-Infringement (Tenth Counterclaim) (PDF) -- asking for a judgment that the Linux kernel does not infringe copyrights owned by SCO
  • IBM's Motion for Summary Judgment on its Claim of Copyright Infringment (Eighth Counterclaim) -- IBM's counterclaim regarding SCO's violation of the GPL and consequent copyright infringment -- (PDF)
  • SCO's cross motion in which it tries to say it never violated the GPL (if you spin the wording their way) (PDF) and
  • SCO's motion for Summary Judgment on IBM's Second, Third, Fourth, and Fifth Counterclaims (PDF) -- SCO's motion trying to get SCO off the hook for all the trash talk in the media.

On this day, we learn from IBM's attorney, David Marriott that the "mountain of code" SCO's CEO Darl McBride told the world about from 2003 onward ends up being a measly 326 lines of noncopyrightable code that IBM didn't put in Linux anyway.

On the other hand, SCO has infringed all 700,000 lines of IBM's GPL'd code in the Linux kernel.

SCO's GPL defense is of the lip-curling variety and quite funny. (You can find a few of the copyrighted Caldera and SCO contributions to Linux here and here, by the way, to help you to understand David Marriott's argument regarding SCO's switching-the-names game.)

And it's also quite amusing to watch SCO try to wriggle out of responsibility for all the trash talk its executives treated us to in its PR campaign.

First Motion Heard -- Is This All There Is?

In the first motion, we learn a bit more about those 326 lines of code. Of those 326 lines, most are comments, not code. Allegedly, those lines of code infringe 320 lines of Unix code. But they aren't copyrightable, IBM says, because they are dictated by externalities, they are unoriginal and they are merger material. Even if they were protected by copyright, those 320 lines don't result in substantial similarity between Linux and Unix.

More details: As for the 326 lines, 11 of 12 files are header files, which aren't copyrightable. Header files don't do anything, IBM's attorney David Marriott explains. You can't run a header file or execute a header file. Header files are just descriptive of how information is shared among the components of an operating system.

Now, the header files themselves are of three types, #define statements, structure declaration, and function prototypes. The first specifies abbreviations. 121 of the 326 lines are #define headers.

A #define statement is comparable to abbreviating the 10th Circuit Court of Appeals as 10th Cir. Cir stands for the name and 10 being the number given that particular court. In #define, you have something like #define EPERM. All that stands for is Lack of Permission, Marriott continues, so if a user tries to do something in Linux that the user lacks permission to do, the system will throw up that error. The E stands for error, PERM for permission, and 1 or whatever would be the number associated with that. That's what a #define statement is: like 10th Cir. How exactly does EPERM infringe on any rights of SCO, Marriott asks? SCO lists it, but without explaining how it's infringing. Is it the error message? Is it that there is a number associated with the error? Is it the name EPERM? Is it the name and number put together that infringes? SCO has never explained with any specificity, although the court ordered it to do so multiple times.

As for structure declarations, they just identify something. For example, it could provide identification of the type of computer system running, what the machine is, what version, release, type of machine.

Function prototypes specify what operations may be performed, using what inputs and producing what outputs. A find function, for example, does exactly what you'd imagine it would do. Same with a message send function. Of the 326 lines, 12 are function prototypes.

SCO's expert, Mr. Cargill, stated that choosing to list names in alphabetical order and to assign numbers in sequential order evidences creativity sufficient to satisfy him as to whether the code is original within the meaning of copyright law, Marriott argues, but that's wrong according to the US Supreme Court's decision in Feist. Looking at this code, where is there any cognizable originality? EPERM for permission error? That's no more "original" than saying 10th Cir. for the 10th Circuit Court of Appeals. What else would you call it? And how many ways are there to say that a computer system is Linux, version such and so, release such and so? That's all structure declarations do. And we're not even talking about lines of executable code, Marriott explains. These lines at issue are scattered about. Header files aren't executable anyway. The files are not copyrightable.

Worse, SCO claims control over code copyrighted by IBM, such as JFS, and others. SCO's own experts said SCO has no copyright infringement claim over those.

At the end of the day, after putting IBM to the expense of all that massive discovery, that's all there is. SCO can't prove it even has any copyrights to be infringed, nor has it provided any specifics on exactly how its rights have been infringed, even if it had the copyrights, and even if it hadn't transferred all its copyrights to UnitedLinux, which it did, Marriott concludes. This is all there is. SCO Linux 4 powered by UnitedLinux included all the 326 lines that are at issue.

IBM has a right to use all 326 lines. It has multiple licenses (5, to be exact), Marriott tells the court, the GPL for one. There was also a business agreement between Caldera and IBM that also gives IBM a license to use this code and to prepare and have prepared derivative works and to distribute and sublicense and to grant others rights with regard to all 326 lines. It was Caldera that approached IBM to enter into that business agreement, and significantly it was after Caldera knew about the results of the 1999 study by Santa Cruz that showed some similarities between Unix and Linux, similarities SCO found were understandable and acceptable, that Caldera asked IBM to enter into this agreement. That means Caldera, now SCO Group, is estopped from now pursuing any infringement action. That estoppel is on top of IBM obtaining a license by the agreement.

Not only that, but the same agreement provided IBM a warranty; SCO represented that the code does not infringe any intellectual property rights of any third party.

The GPL, of course, grants IBM legal permission to copy, distribute and modify the software. Finally, SCO was a Linux company that distributed this code for years and encouraged the world and its dog to copy, modify, distribute, sublicense, whatever, this code, so they are estopped from suing IBM for doing what the GPL license SCO distributed under said IBM could freely do.

Further, SCO sponsored the standardization of Linux, making Linux compliant with various standards, such as the POSIX.1 specification. The code it now wants to sue over is in Linux because of that standardization, the LSB. SCO was in fact the only notable supporter of the STREAMS material it now wants to sue IBM over. STREAMS isn't even in the kernel, but SCO wanted it included to support its Netware for Linux products, Marriott says. The Linux community didn't want STREAMS in the kernel, so SCO collaborated with others to make it available for download from a web site. Now it wants to sue people for infringing the very code it put there.

What about ELF? There was a 1997 meeting of a group called the 86Open group, and at that meeting Linus Torvalds proposed creating a new format for the existing ELF format in Linux. Santa Cruz opposed his proposal and insisted on a Linux ELF so programs could more easily run on SCO UNIX. So the ELF standard was adopted, at least in part because of SCO's effort. Now it wants to sue over ELF.

SCO Replies

What is SCO's reply? I read it all, and I don't see any of IBM's points directly responded to, except for saying in passing that the business agreement and the GPL don't give IBM the rights it says they do and a broad assertion that estoppel isn't appropriate, for this reason and that, but estoppel isn't at all necessary for IBM to prevail. It's the cherry on top.

The business agreement doesn't let IBM open source Unix "intellectual property", SCO says. I guess it wants the judge to just leap to the assumption that the 326 lines are Unix IP. Of course, the GPL *does* give IBM and the world the right to open source code. SCO's answer is that they never authorized that code to be under the GPL, because there's no copyright notice from them. Maybe because, as IBM showed already, SCO owns no copyrights on this code? You think? Actually, IBM's answer is better. It points out that SCO is just renaming Linux code and calling it Unix. But it is the same code, and the Linux code was indisputably released under the GPL. According to SCO itself, it's identical code. IBM disputes that, but that is SCO's whole claim, so that exact code, according to SCO's allegations, was, in fact, released under the GPL, with certain representations and promises attached, code Caldera shipped and made money from. Now it wishes to pretend, IBM says, that those representations and promises are of no consequence.

It should all go to a jury, SCO says, and it talks about "non-literal copying," which IBM later points out after SCO finishes talking would be in the bucket of allegations the court already told SCO it can't argue.

And SCO claims Linux can't run without STREAMS, but as IBM already explained to the judge, it isn't in there now, unless you want it to be, and Linux was running just fine when Caldera tried to persuade the Linux kernel maintainers to include STREAMS in the kernel, and failed to persuade them. Caldera put up that website, with others, to make STREAMS available, and now it wants to claim STREAMS is protectable and that you can't run Linux without it. So that is the level of SCO's answer.

Instead SCO raises other issues, mainly in the context that IBM's issues require a jury, not that SCO will win the points, but that they have raised enough questions to require a jury, and it argues various points of law, such as whether IBM can seek a judgment that Linux doesn't infringe for others or only for itself and who bears the burden of proof in the motion. SCO smears some folks, as usual, and then the attorney sits down.

Oh, one more thing. As for the licenses, SCO asks why IBM didn't present those licenses years ago, if they are so powerful. Here's IBM's answer, in part:

Your Honor, the explanation for that is we spent the last four years horsing around trying to figure out exactly what it was that supposedly IBM did.

The court itself told the parties not to bring substantive motions until after discovery was finished, Marriott reminds the judge, so how can SCO argue that "delay" in bringing a motion mentioning estoppel means IBM's motion should be denied?

Finally, SCO tries to argue that it wasn't the one who released certain code. Santa Cruz did it. But IBM just points out that SCO Group claims that Santa Cruz is a predecessor in interest, so it's bound by what it did:

And Mr. Singer suggests, Your Honor, that somehow SCO should not be held responsible for the acts of its predecessors because they didn't own the copyrights at the time or they didn't know what was going on. Well, Your Honor, the law is clear that a company is bound by the conduct of its predecessors. The law could not be more clear in that regard. And I point you to Page 38 of IBM's reply brief, Judge. Estoppel is a doctrine of equity. And this Court has ample authority under which it exercises its equity to preclude a party that for nearly a decade distributed code under the promise it can be used with all the rights that they had from turning around a decade later under new management from disregarding the representations and warranties made from strategic business agreement, from disregarding the principles set out in the GPL, from disregarding the licenses given under the SBA, from disregarding the licenses given under the UnitedLinux, and pretending as if it's such a surprise to find out that there's a theory in which they think now maybe they can get somebody to maybe pay them some money. Respectfully, Your Honor, if there ever were a case for estoppel, this is the case.

As for whether SCO's claim that things like EPERM are copyrightable, IBM's David Marriott ends his final statement like this, with a baseball analogy, where "the pitcher is number 1 and the catcher is Number 2, the first base is 3 and second 4 and 5 is third and 6 is the short, left is 7, 8 is center field and 9 is right field":

What SCO contends, Your Honor, let's talk about those. Your Honor, in effect what they have claimed is that the pitcher, player pitcher is 1, player catcher is 2, player first base is 3 and so on. That's what those #defines represent, associating a number and a shorthand for a position and claiming that somehow it renders Linux so substantially similar to UNIX that they enact and claim rights. And I respectfully submit, Your Honor, that the law is clear that short names of that sort and associating integers randomly with phrases like PP1 or EPERM-1 simply is not protectable under the doctrines laid out in our papers by Professors Kernagen and Davis in their expert reports. And in no case can it result when it's 320 lines of non-contiguous essentially random numbers with essentially shorthand phrases represents substantial similarity.

Now, Your Honor with respect to misuse, briefly again, the facts here are simple. They claimed rights to more than a million lines of code in Linux. At the end of the day, there's 326 lines of code in which they have rights, and they have sought to exert the supposed monopoly they have and copyrights they claim to have over technology plainly owned by others.

For the five reasons I set out, Your Honor, summary judgment respectfully should be entered in favor of IBM. Thank you.

And with that the arguments on the first motion come to an end.

Motions Regarding the GPL/Copyright Infringement

Now, on the IBM motion and SCO cross motion regarding IBM's copyright claims, the GPL matters, in contrast to SCO's alleged 326 lines of infringed code, IBM owns about 700,000 lines of code that SCO has infringed, Marriott states. IBM says it is entitled to summary judgment for the following reasons:

First, IBM owns the copyrights at issue. Second is that SCO copied and distributed the copyrighted works. And the third is that SCO had no permission to copy and distribute those works.

Why does SCO have no permission? Because it breached the GPL and thus lost the license it had. How did it breach? Here's one way:

SCO breached the GPL, Your Honor, by restricting redistribution rights. Section 6 of the GPL says -- this is at tab 18 of your book -- that each time you redistribute the program; that is, a program licensed under the GPL, or any work based on it, the recipient automatically receives a license from the original licensor to copy, distribute or modify the program subject to these terms and conditions and that you may not impose any further restrictions on the recipient's exercise of the rights granted herein.

And yet, Your Honor, subsequent to the commencement of this case, SCO attempted to do exactly that with respect to users of Linux, including IBM. For example, if you turn to tab 19 of the book, you will find an excerpt from SCO's web site in which it describes the license that it represented to the world was required if one wished to use Linux after it asserted that there was, in Linux, infringing UNIX code. The license is described in SCO's words, Your Honor, as SCO intellectual property license for Linux. SCO stated that the license was created because, in their words: Quote. Many customers are concerned about using Linux. The license, Your Honor, applies to the commercial use, by its terms, of a Linux operating system that contains the 2.4 or later version of the Linux kernel. And notably, for this purpose, and for Section 6 of the GPL, SCO said on this web site describing its Linux license that the license did not grant any distribution rights, despite the fact that the GPL expressly required that SCO was required to provide distribution rights as to any material received by it under the GPL and distributed by it under the GPL.

Now, there isn't just, Your Honor, a web site describing a licensing program. In fact, SCO entered licenses in the marketplace with a variety of companies including, for example, Everyone's Internet. ... That, Your Honor, represents an impermissible imposition of obligations under the terms of the GPL and in so doing so exceeded its rights under the GPL, and when they did that, Your Honor, they automatically, by the plain language of the GPL in Section 4 lost the right to continue to distribute the code that IBM gave them the right to distribute, but only on the terms of the GPL.

Additionally, IBM points out, SCO's license breached the GPL by restricting rights to source code, its license saying it granted the right to use SCO IP in Linux in binary format only.

SCO's turn

What can SCO possibly say in its defense? Once again, poor Edward Normand gets the worst assignment, or so it seems to me. It seems whenever SCO needs a lawyer to say or do something silly, they use him. The arguments on the GPL are, frankly, laughable, but he gives it his all. Here are the arguments:

1. That Linux is a derivative of Unix (which I must say hasn't been proven by a long shot) and so IBM can't enforce its copyrights in its copyrighted material in the "derivative work". It cites the following on that point:

Section 103(a) of the Copyright Act states: Protection for a work employing preexisting material of which the copyright subsists, does not extend to any part of the work in which such material has been used unlawfully.

Now, what does that mean? The Court's have told us. As one Federal Court has explained: The case law interpreting 103(a) supports the conclusion that generally no part of an infringing derivative work should be granted copyright protection. That is particularly true under the case law, Your Honor, where the preexisting material in the derivative work tends to pervade the entire derivative work.

We believe we have made that factual showing with respect to the UNIX System V material in Linux. The precedent makes clear that the relevant question is not, as IBM suggests in its brief, whether its copyrighted material is derived from UNIX System V. That's not the relevant question. The question is whether the IBM material is part of a derivative work of SCO's copyrighted technology in which SCO's UNIX technology tends to pervade, and it does.

Aside from not having proven any of that, and according to IBM in the first motion of the day not being able to, and quite aside from IBM holding copyrights on the code and being therefore presumed to have them appropriately, the case is now about SCO's Unix "tending to pervade Linux"? Leaving out of the discussion for the moment the fact that the court already threw out all claims that weren't specified by file, line, version, which obviously a claim of "tending to pervade" would be affected by, where has SCO ever demonstrated or even alleged this? Which code? All of it? All of Linux tends to be a bit like Unix? That's the beef? Are they kidding? Or bluffing? They can't seriously have sued over this, can they?

2. IBM comes to court with "unclean hands" so it can't enforce its copyrights with respect to damages, which IBM isn't moving for summary judgment on anyway. What do they mean? That IBM "hacked" SCO's website.

I think we can safely leave that in the Puh-lease folder.

3. That SCO hasn't breached or violated the GPL. The UNIX materials in Linux were never licensed under the GPL by SCO. The same blah blah that Marriott answered already in the first motion. Marriott, more patient than I am, answers in detail again:

Now, Mr. Normand suggests, Your Honor, and I think ultimately this motion depends upon the idea that SCO didn't breach the GPL as to the code that is at issue because it did not take a registration -- excuse me -- it did not take a slug like this and put it in the GPL on the exact code and say: This is UNIX code, and we're SCO, and we're giving you rights to do with it under the GPL. Your Honor, that is nothing more than a name game, than a labeling game.

If Your Honor would look at the chart over here, what SCO is essentially doing, Your Honor, is saying that this code, which we'll assume to be in Linux, this code is not subject to the GPL even though it shows up in Linux, even though there is a copyright designation on it, even though it says it's distributed under the GPL just because it doesn't say: This is SCO code, and it's UNIX code in Linux. The code in question, Your Honor, which they claim is not under the GPL is, they say, not under the GPL just because they haven't, they say, put on it a label that says: We're SCO. We are giving up this UNIX code.

The fact of the matter is, Your Honor, the code we're talking about is there. It bears a GPL notice. It was offered to Linux by the writer and creator of that code pursuant to the terms of the GPL. And whether or not -- I respectfully submit that SCO has, in fact, licensed that code, despite what Mr. Normand suggests. But the fact of the matter is: Whether or not it was licensed as a licensor, whether or not SCO was a licensor, they were a licensee. They got it just like they got the IBM code, the 700,000 lines of it, that bore the representation and notice that they were under the GPL, and only to be used under the GPL.

And now Mr. Normand says: Yeah, but that doesn't matter, Judge, because all of Linux is a derivative work of UNIX System V, and nothing there, under their argument, Your Honor, is at all under the GPL because they purport to own it all. Its all a derivative work. And no one put it in -- there is, to my knowledge, at least, no notice in the GPL that says -- in Linux, rather, that says: This is all SCO code, and we are giving it up under the GPL.

And that's essentially their argument, Your Honor. There is not a notice that says that. What there is in fact is the code with the GPL notice by the person who wrote it saying that it is offered under those terms. And all they are really saying, Your Honor, is because they claim there is infringement in that code, they say it's really UNIX code and so the appropriate notice had to be one that said this is actually UNIX code. And since they didn't do that, the GPL has no absolutely no application.

But if they were right, Judge, if they were right in that argument, then all anybody would ever have to do is to say: You know, what, I think Linux infringes and so it's an impermissible derivative work, and I am therefore no longer subject to the terms of the GPL, and I'll go off and sue you, and then I will continue to distribute it under the terms of the GPL, putting the notice on there and make my profits from it and pretend as if it applies to me selectively.

We're talking about the same code, Your Honor. It's a labeling game of calling it UNIX and saying it wasn't GPL because it doesn't say SCO UNIX. It's there. It has the label of the GPL on it. It's therefore distributed under the terms of it. That's the way they got it from IBM, Your Honor, when they got the 700,000 lines of code from IBM. And it was only on the terms and conditions of the IBM GPL license, the designation that we put there on our code, that they had permission to use it.

Now, as to that code, they have undertaken the position to assert rights as to that, that other people can't copy it, they can't distribute it, they can't use it, though it's our code with our copyright notice issued duly from the Copyright Office because they say it infringes. And that, Your Honor, respectfully, turns the GPL on it's head. And if it were read and construed in that way, it would be effectively a meaningless document.

Not that SCO wouldn't love that exact result, but courts tend not to reward such doings as this.

SCO seems unaware of it, but I think it admits to a GPL breach in its defense, claiming that it never put the GPL notice in SCO's Linux Server 4.0:

IBM specifies this program is free software. You can redistribute it and/or modify it under the terms of the general public license, and it includes at the bottom a copy of the general public license. But section zero required IBM to do this if they wanted to open source the product. There is no such language in the SCO Server 4.0 product that IBM refers to. In fact, SCO didn't even place that notice in the product. And that's undisputed from Mr. Nagel's testimony at SCO Exhibit 233.

Now, I want to be clear, Your Honor, about why IBM is wrong in claiming that SCO had licensed the UNIX material in Linux under the GPL when it distributed verbatim copies of Linux as part of SCO's Linux Server 4.0 And we explained this point at pages 8 to 10 of our brief. The point is that the GPL licensee does not, by distributing a verbatim copy of the GPL software, claim to have made any material in that software licensed under the GPL.

If I've understood what he's saying, that alone would be a violation of the GPL, I think, to redistribute GPL code without the GPL.

SCO also has a convoluted theory as to why it isn't in breach by attaching restrictions on top of GPL code, something about merely distributing verbatim code giving them an out. IBM responds:

The suggestion has been made that SCO never modified in any way the kernel, and that simply is not true. It isn't supported by the evidence that is set out in our papers. They took the kernel, Judge, of Linux, and they created, based on that kernel, their own product, SCO Linux 4, which was a modification and a derivative work, something new from the kernel. Now suddenly the concept of derivative work takes on a new meaning. When we are not arguing about IBM's contract motion, derivative work is suddenly a very narrow construct, and it's very hard to become a derivative work.

By SCO's own definition and by the definition under the law, Your Honor, there is no question that what they distributed as SCO Linux 4 was a modification subjecting them to the terms of Section 2-B of the GPL, which, contrary to what Mr. Normand suggests, makes perfectly clear that you can't charge fees of the kind they charged for this software.

And, moreover, Judge, the Sections of the GPL that are at issue, the ones I cited, Section 6, Section 1, those don't turn on having a particular modification. Those turn on not giving people who gave you code under the GPL, not giving others who received the code under the GPL, the rights to which they are guaranteed under the GPL.

SCO was a licensor and SCO was a licensee and had obligations if it was going to continue to distribute this code. No one made them distribute our code. No one made them sell it for a profit. They undertook that on their own, and when they did, they undertook it subject to the terms and the conditions of the general public license.

Then SCO presents some contract law arguments, but the GPL is a copyright permission license, so the argument that you can't repudiate unless you give notice is plainly not what the terms of the GPL say, and the GPL says if you redistribute, which SCO admits to, then you have accepted the terms of the GPL. There is no right under the GPL to rewrite the terms, which is what SCO tries here to do. On the GPL motions, you really should read IBM's Redacted Reply Memorandum in Further Support of its Motion for SJ on its Claim for Copyright Infringement (10th Counterclaim) [PDF], particularly beginning on page 11, where IBM answers SCO's nonsensical GPL misunderstandings. I am giving them the benefit of every doubt when I say misunderstandings. I don't actuallly believe 100% that any competent lawyer could get it so wrong and not realize it.

Final Motion - What He Said

Finally, SCO would like Mr. McBride and SCO to escape from responsibility for all the trash talk they spread from coast to coast and around the world. SCO says they have the right to say truthful things about what their case was about under the privilege and qualified privilege doctrines. Of course, their argument suffers from Darl's "mountain of code" hyperbole, which resulted in 4 telephone book sized scrapbooks of collected clippings from just one month of SCO PR, as IBM attorney Amy Sorenson points out:

But SCO's cases concerning this qualified privilege are summarized at tab 13 of our binder, and they make the point nicely that no Court has anticipated, much less held, that a qualified privilege to summarize one's claim to the media should be extended to provide immunity to a litigant that embarks on a nationwide negative publicity campaign about a competitor and its products and services, which has been waged for years, and which continues to the present day.

Instead, each of SCO's cases concern a single press release or a single press conference, and as a result, they offer SCO no support in its attempts to avoid liability for the wide-ranging misconduct that is at issue here.

For example, IBM Exhibit 147 is a letter written by Mr. McBride to Congress claiming that Open Source software and Linux are nothing short of a threat to our nation's economy and to our nation's national security. These are the kinds of statements that have nothing to do with the litigation. They far exceed any scope of any qualified privilege because they are in no way a mere summary, at one point in time, of its allegations that are at issue in the case.

In addition, IBM has provided numerous cases -- this is at tab 15 of our binder -- holding that this kind of conduct, where a litigant actively stimulates press coverage and wide publicity of a Complaint is beyond the pale of judicial privilege protection. Far from distinguishing them, SCO ignores them in its reply, and it has ignored them today.

This is, in effect, an argument about excessive publication, and if there has ever been a case for excessive publication, I would submit it is here, where SCO has made numerous, repeated statements to the media about IBM's products and services and its rights to own, control and profit from AIX and Dynix and sell and profit from Linux-related services.

As this Court has stated in a previous decision in the SCO vs. Novell case, the issue of whether there has been excessive publication is a question of fact and therefore not appropriate for resolution on summary judgment. Perhaps the exhibit that makes the point best about the volume of SCO's public statements here and the extent of its excessive publication, over publication and the fact that it has exceeded any qualified privilege and certainly any litigation privilege, is made at tab 14.

This excerpt is a Fortune Magazine article from May 17, 2004, only just one year into the litigation, and that article opens with the observation that in the ascetic waiting room of the SCO Group's Lindon, Utah headquarters, the only reading matter is a stack of beige telephone-book sized binders. They are volumes 1, 2, 3 and 4 of the company's press clippings for the previous month.

There is a difference, in other words, between a single press release announcing litigation and the reasons for it and a multi-year PR campaign such as SCO undertook. Here, IBM is arguing against SCO's summary judgment motion; IBM very much wants the jury to hear this part.

This transcript was provided in two separate PDFs, corresponding to the portion of the hearing before and after the recess, so the page numbers in this HTML version reflect the page numbers in the respective PDF files.

**********************************

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF UTAH, CENTRAL DIVISION

____________________________________
)
SCO GROUP, et al., )
)
)
)
Plaintiff, )
)
vs. ) Case 2:03-CV-294
)
)
INTERNATIONAL BUSINESS MACHINES )
CORPORATION, )
Defendant. )
)
_____________________________________ )

BEFORE THE HONORABLE DALE A. KIMBALL
MARCH 7, 2007
REPORTER'S TRANSCRIPT OF PROCEEDINGS
MOTION HEARING
DAILY COPY TRANSCRIPT

Reported by: KELLY BROWN, HICKEN CSR, RPR, RMR
REBECCA JANKE, CSR, RPR, RMR

1

APPEARANCES

FOR THE PLAINTIFFS: HATCH, JAMES & DODGE
BY: BRENT O. HATCH
MARK F. JAMES
Attorney at Law
[address]

BOIES, SCHILLER & FLEXNER
BY: EDWARD J. NORMAND
STUART H. SINGER
Attorney at Law
[address]
FOR THE DEFENDANT: CRAVATH, SWAINE & MOORE
BY: DAVID R. MARRIOTT

Attorney at Law
[address]

SNELL & WILMER
BY: AMY F. SORENSON
PETER H. DONALDSON
TODD M. SHAUGHNESSY
Attorney at law
[address]

2

SALT LAKE CITY, UTAH, WEDNESDAY, MARCH 7, 2007

* * * * *

THE COURT: We're here this afternoon in the matter of the SCO V. IBM, 2:03-CV-294. For plaintiff, Mr. Brent Hatch, Mr. Ted Normand, Mr. Stuart Singer. For defendant, Mr. David Marriott, Ms. Amy Sorenson, Mr. Peter Donaldson, Mr. Todd Shaughnessy, Mr. Joseph Kriesh and Mr. Jefferson Bell.

All right. We have IBM's motion for summary judgment on its claim for declaratory judgment of non-infringement.

Is that the one you want to do first?

MR. MARRIOTT: It is, Your Honor.

THE COURT: And are you arguing?

MR. MARRIOTT: I am.

THE COURT: And Mr. Singer?

MR. SINGER: I'll be arguing.

THE COURT: How long is this one supposed to take?

MR. MARRIOTT: I believe 45 minutes a side, Your Honor.

MR. SINGER: That's correct.

THE COURT: All right. Go ahead, Mr. Marriott.

MR. MARRIOTT: Thank you.

Your Honor, we have, as always, a book. If I may approach.

3

THE COURT: Yes.

MR. MARRIOTT: After promoting Linux for nearly a decade, Your Honor, SCO changed management and undertook a series of legal attacks against it. And despite the fact that it contributed Linux and it induced thousands to use it, it threatened users of Linux with infringement including IBM. And to put an end to the fear, the uncertainty and the doubt generated by those allegations, IBM filed a claim asking this Court to declare that the kernel of Linux does not infringe the copyrights purportedly owned by SCO.

Now, when IBM last moved for summary judgment on this issue, Your Honor, which motion Your Honor denied as premature, you observed that SCO had produced no competent evidence in support of its allegations of infringement. And now four years after the commencement of this suit, nothing has changed. There's no competent evidence to support the conclusion of infringement.

And as illustrated, Your Honor, at Tab 1 of our book, SCO's allegations and infringement failed for five independent reasons, any one of which is a basis for summary judgment.

Now, before I come to those, Your Honor, if I may, by way of background, let me say this. As you know, for years SCO pronounced that there were more than a million lines of code controlled by it in the Linux kernel, as we illustrated

4

in Tab 2. At the end of the day, however, when all is said and done, there are 326 lines of code in the Linux kernel which supposedly infringed SCO's alleged copyrights. That's it. 326 lines of code in the Linux kernel.

The allegedly infringing code is trivial in size, Your Honor, less than one 5,000th of a percent of the kernel. And as we illustrated at Tab 3 of the book, it is equivalent of one spectator in 20,000 in an arena the size of the former Delta Center.

The actual code, Judge, appears at Tab 4, Tabs 4 and 5 of the book and here on the easel to my left and Your Honor's right. And so that it's perfectly clear, Your Honor, IBM did not contribute a single one of these 326 lines of code to Linux. Most of it was there long before IBM had anything to do with Linux. And as I will show Your Honor, much of it is there because of SCO and its support for Linux.

Now, portions, Your Honor, of the lines of these 326 lines of codes are not actually coded, although they are comments. And if you look either in your binder or on the chart, you will see that that which is in yellow represents comments not code.

Now, as for the actual code, Judge, 11 out of the 12 files that are at issue here consists of what is called header file code. And, in fact, it's not executable file, executable code at all. It does essentially nothing in terms

5

of being executed. It is interface information, Your Honor. It describes how information is shared among the components of an operating system.

And as we show at Tabs 6 through 8 of the book, Your Honor, this header file code consists of header file information of three types. And just so that it's clear exactly what we're talking about, let me describe briefly what each of these three types is.

First type, Your Honor, are so-called #define statements. The second are structure declaration, and the third are function prototypes. Let me just say a word briefly about each those.

A #define, Your Honor, specifies a shorthand or abbreviations. It put it differently, it simply names and numbers associated with anything you might want. Anything that could happen in connection with a computer an error might have associated with it a name and a number. Of the 326 lines that are at issue, 121 of those lines are #define statements.

And to give Your Honor an example of what a #define statement is, consider, if you would, the abbreviation for the 10th Circuit Court of Appeals. 10th Cir. The Cir, Your Honor, is the name that would be assigned, and the 10 is the number that is assigned. If you want to take an actual example of the code, if you would look under the first file, the first line says, #define EPERM, Your Honor. And what that

6

says is that if there is a user that attempts to access a file to which that user doesn't have permission, the system will indicate there is an error, an error for lack of permission.

So the E for error, the PERM for permission and the number 1 is associated with that. That is what a #define statement is, Your Honor, much like Circuit 10th.

If you take a look at the next type of information in these files, structure declarations. They specify, Judge, how information is to be stored and how it is to be displayed. To put it differently, a structure declaration is a collection of related data values.

And of the 326 items that are at issue in this case, 164 lines of them are structure declarations. There's somewhere south of 20 or so structure definitions within those lines.

And to continue the case citation analogy, Your Honor, structure declaration is like the format for a case. If you had a case, its format would be name of case, reporter, court and date. That is in effect a structure declaration.

And if you want to take an actual example from this code, Your Honor, if you look at file Number 10 there, it provides that, one of the structure declarations there provides that when specifically identifying a computer system, you should say what the machine is, what version it is, what release it is and the type of machine on which it runs. Those

7

are examples of things that come in structure declarations. It's the operating system saying, what kind of machine am I writing on? Well, it's this kind of machine. It's that version. It's that release. That's what a structure declaration does, Your Honor. It allows users to understand how components of the operating system interrelate.

Finally, Judge, function prototypes. It specifies what operations may be performed, using what inputs and producing what outputs. And of the 326 lines of codes that are at issue, there are 12 function prototypes.

So if you were to go to a cite, Your Honor, like Westlaw and you were to try to use the find function, the find function which has you put in certain parameters in a case would allow you to find that case. And that is in effect what a function prototype is.

So if you look at the chart, the second blue line, Your Honor, Line 67 of file number 5, you'll see a function prototype called message send. And that says that if you provide the message recipient ID and the message and the length of the message, any extra information, like if you wanted it to be sent high priority, that will tell you whether the message was properly sent. And if it doesn't go through, an error will register. And you might get that like an EPERM number 1.

Now, finally, Your Honor, before coming to the five

8

bases of summary judgment, let me just underscore, that though this is IBM's motion, it is SCO as the party asserting infringement that bears the burden to establish that IBM's Linux activities have infringed the alleged copyrights. And at Tab 9 in the book we lay that out for the Court.

Point number one, Your Honor. Summary judgment should be entered in favor of IBM because SCO cannot establish unauthorized copying by IBM of copyrights owned by SCO. And there are three separate reasons why that's the case, and with Your Honor's permission I want to deal with just two of them today.

The first reason is, Your Honor, despite this Court's order that SCO tell IBM exactly what it is that IBM has supposedly done to infringe on these 326 lines of code, no information has been provided in SCO's final disclosures. And a second reason is, Your Honor, the copyrights at issue on this record we submit are owned by Novell, not by SCO. And third, Your Honor, SCO transferred any rights it had in the code that is at issue to UnitedLinux for reasons we discussed at the last hearing, and I'll reiterate here.

I'd like to, Your Honor, if I may take the first and third of those and to defer the second to another day. To the question whether Novell or SCO owned these copyrights will be addressed in the Novell litigation. It's briefed here, and I'm happy to answer any questions Your Honor has about it, but

9

I think in the interest of time I'll focus on 1 and 3.

That brings me to 1, Your Honor. As we will see at Tab 11 of the book, SCO has repeatedly by this Court and Magistrate Judge Wells been ordered to identify with specificity its allegations of infringement to tell IBM how it is supposedly IBM has infringed this code. And that isn't in the final disclosures or any of SCO's interrogatories responses provide that information. And if Your Honor looks at Tab 12 in the book, you will see in IBM's, its support of IBM's motion for summary judgment, we laid out the fact that SCO failed to provide this information.

And what we got in response, Your Honor, is in effect, not a demonstration of where the information supposedly was provided, but yet another statement that SCO was not obligated to provide the information provided.

As shown at Tab 13, again, Your Honor, this Court and Magistrate Judge Wells have been clear. If it wasn't described with specificity in the final disclosures, it's out of the case. And respectfully, no information has been provided to describe the allegations of infringement. And for that reason alone, Your Honor, summary judgment should be entered in favor of IBM.

Now we come to the third reason, I'm skipping the second, third reason why SCO cannot prove that IBM has infringed copyright owned by SCO, is that SCO transferred any

10

interest, any ownership interest it had in the allegedly infringed code to the UnitedLinux, LLC.

Let me walk Your Honor through that. As you'll recall from the last year in May of 2002, SCO partnered with other Linux distributors to form a joint venture called UnitedLinux. And it was formed to streamline, in effect standardize Linux distributions. Each member of that UnitedLinux effort signed an agreement under which they assigned their intellectual property rights in the joint development product to the UnitedLinux, LLC.

And as we discussed in previous arguments, Your Honor, SCO's SCO Linux 4 was the joint development product of UnitedLinux of SCO. And as illustrated at Tab 19, Your Honor, of the book you'll see that the joint development contract defines the UnitedLinux software to be, quote:

At any implementation of the UNIX operating system developed and integrated under any conditions of the JDC.

Both of the appendices of the JDC explain that to mean an implementation based on a modified version of the Linux kernel. And as we illustrate at Tab 20 of the book, Your Honor, SCO has admitted that its SCO Linux 4 was a product developed pursuant to the JDC. And in a press release of 19 November 2002, SCO stated, quote:

SCO Linux 4.0 is based upon UnitedLinux 1.0,

11

the core standard base Linux operating system, development in an industry initiative to streamline Linux development, and so on. As we show at Tab 21, Your Honor: SCO Linux 4 powered by UnitedLinux included the 326 lines that are at issue.

It follows, Your Honor, that the 326 lines at issue were in the software development pursuant to the JDC. And so unless the 326 lines of code were excepted from assignment, SCO has no interest or right in those and cannot pursue against IBM or any other claim for copyright infringement.

SCO had pursuant to these agreements, Your Honor, its list of excluded technologies, a specific list. And as is demonstrated at Tab 22, Your Honor, how their list, the successor of the SCO, did not include the 326 lines of code that are at issue here. And as a result, that code is not owned by SCO, and it may not pursue a claim against IBM again or any other for infringement.

Point 2, Judge. IBM has a license. In fact, it has multiple licenses to the 326 lines of code at issue, as we illustrate at Tab 25 of our book. Now, time won't allow discussion of each of the five licenses that we've briefed, and so with Your Honor's permission what I'd like to do is focus on the two of the licenses that have the broadest coverage, the two licenses that cover every single one of the 326 lines of codes that are at issue. And that's, one, the

12

IBM strategic business agreement between Caldera and IBM; and second, the GNU general public license of GPL. And again, I reiterate, all 326 lines of codes were included by SCO in Linux products as distributed under these licenses. So with that predicate, let me take each of these licenses in turn, Your Honor.

First, the strategic business agreement. As shown at Tab 26, that was an agreement between IBM and Caldera, and it included a license that provided as follows: Quote -- well, almost quote. It provided that Caldera would grant IBM a, quote:

Worldwide, perpetual, irrevocable, fully paid-up license to prepare and have prepared derivative works of free existing materials and to use, have used, execute, reproduce, transmit, perform, transfer, distribute and sublicense preexisting materials for their derivative works and to grant others rights under this subsection.

So what are preexisting materials, Your Honor, and their derivatives? If you turn to Tab 27, we follow the chain from one definition to the next that demonstrates that these 326 lines of code were in what was licensed to IBM.

As you'll see, Your Honor, preexisting materials are defined in relevant part as items contained within a deliverable. A deliverable is defined as, expressly

13

identified as a deliverable in the statement of work. The statement of work in term provides the deliverables include license works which are defined to include packaged license works, and packaged license works are further defined in Exhibit A, Your Honor, which describes the packaged license work as the OpenLinux products of SCO including eDesktop, eServer products. The SCO OpenLinux product line was later named, Your Honor, SCO Linux 4 with release of that product. But was in any event a derivative of SCO OpenLinux products as we show at Tab 28.

So SCO granted to IBM, Your Honor, a license to the 326 lines of codes at issue under the strategic business agreements between IBM and SCO. And again, every one of the lines of code at issue is in there.

Let me drop a footnote, if I may, and I'll come back to this. This same agreement, Your Honor, provided IBM a warranty with respect to these 326 lines of codes. It says:

SCO represented that they do not infringe any intellectual property rights of any third party.

They do not infringe any intellectual property rights of any third party. And at the time, Your Honor, by SCO's reasoning it was Santa Cruz -- again, we disagree with that -- but it was Santa Cruz that owned the copyrights that SCO now claims it can assert against IBM.

14

That brings me to the GPL. In addition to granting IBM a license under the SBA, Your Honor, IBM also has a license from SCO to these 326 lines under the GPL. The GPL in its preamble and elsewhere provides that if you distribute copies of a program licensed under the GPL, that you, quote, must give your recipients all the rights that you have.

It further provides in its preamble that a license gives you the legal permission as a recipient of code distributed under it to copy, distribute and modify the software. And again, in Section One and Section Two in Section Three, a license is given to the code distributed under the GPL to do exactly that, which purportedly SCO accuses IBM of doing, as I say precisely what IBM was supposed to have done has never been disclosed, Your Honor.

There is no dispute, I respectfully submit, that SCO's Linux products were distributed under the GPL. And if you take a look at Tab 32, you will see in SCO's own words the following statement:

All of SCO's Linux distributions prior to and after May 2003 were made under the GPL.

Therefore, IBM, Your Honor, has not one license, but two licenses to the material issue without regard to the three licenses that I'll leave to the briefing.

Third point, Your Honor. SCO is estopped from pursuing against IBM a claim for copyright infringement. SCO

15

has a long history, Your Honor, as a supporter of Linux, and it is that history that precludes precisely the conduct which indicates here, to sue for infringement relating to Linux. SCO was not only founded as a Linux company, Your Honor, it distributed and it profited from Linux including those lines of code for a very long time.

And as we demonstrate at Tab 43, SCO's various Linux products again included exactly this code, and SCO employees have testified to that effect. Ralf Flaxa, for example, who was head of SCO's development in Europe, testified, quote:

While employed at Caldera, I was aware that the allegedly infringing material was present in Linux. I know so because of my familiarity with Linux and also because Caldera is incorporated in its Linux products.

Now as I said, Your Honor, the code was distributed under the GPL. In addition to that fact that that grants IBM a license, it also has an affect with respect to estoppel, because again, the license provides that that which one gives to others under the GPL, one gives that person's rights in that material to those other persons.

So SCO wasn't just family in the Linux company. They didn't just distribute it under the GPL. They produced what they described as an award winning set of Linux products

16

that made it, in their words, a leader in the movement toward the adoption of Linux. And again, based on the exact code that is at issue. So if you take a look at Tab 46 of the book, you will see a list of some of the awards that Caldera won for its promotion and sponsorship of Linux. Linux Show's Best Distribution of the Millennium. Linux Journal's Product of the Year award and CNET Editor's Choice Award.

THE COURT: Where does this illustration come from in 46? This cup with all this money?

MR. MARRIOTT: Someone on our team made that up, Your Honor.

As a leader in the Linux community, SCO played an important role in the standardization of UNIX. And putting aside UnitedLinux, which I'll come back to, SCO was the first signatory of a document proposing the so-called Linux standards base. Santa Cruz also, SCO's purported successor, was in support of that. And the Linux standards base was an attempt to define the common core of components that represent the Linux operating system.

And the LSB at issue here, Your Honor, required inclusion in Linux of the code that we're talking about. SCO sponsored the standard. The standard required the code. And as SCO former CEO, Your Honor, Ransom Love, has testified, quote:

To facilitate the porting of Linux to

17

application written primarily for UNIX-based operating systems, Caldera, Inc., worked to make Linux products compliant with various UNIX standards, including the X/Open brand for UNIX 95, and the POSIX.1 specification.

So again the allegedly infringing is there in part because of the SCO sponsored the standardization of Linux.

Now, in fact, Your Honor, SCO was the only notable support of one of the items as to which they accused IBM of infringing. STREAMS material, which, in fact, isn't even in the kernel. But SCO required the use of this material as support for its Netware for Linux products. The Linux community opposed the inclusion of this material in the kernel. But despite that opposition, SCO collaborated with others to have it included, made it available for download from its website. And now that very same material that it accuses IBM and others of infringing is there because in effect SCO put it there, Your Honor.

Santa Cruz, SCO's purported predecessor, hosted and Caldera participated in a 1997 meeting to perform a group called 86Open, which we show at Tab 51. Mr. Torvalds proposed then creating a new application format to replace the existing ELF format that was implemented in Linux.

Santa Cruz opposed the proposal, however, and insisted instead on a Linux ELF so the programs could more

18

easily run on the SCO UNIX. The ELF standard was adopted at least in part because of SCO's effort. And it is that precise standard which SCO now contends represents an act of infringement in the Linux kernel.

In 1999, Santa Cruz commissioned a study to compare Linux to UNIX. And according to the chief SCO engineer, that study found certain similarities between UNIX and Linux. But it found that the similarities were understandable and acceptable. Management of the current SCO, however, has publicly taken the position that that memo showed there was something wrong with Linux. And despite their view, however, that something was wrong with Linux as indicated by that memo, both Santa Cruz, Your Honor, and SCO continued over the course of the following number of years supporting and promoting Linux despite the supposed concern convinced by this memorandum. As SCO's former CEO put it in 2001 when Caldera acquired certain UNIX assets, quote:

We did not care whether UNIX source code had been included in Linux improperly, and we did not at any point disclose that there might be any problem with Linux.

At the very same time, Your Honor, that Santa Cruz was conducting this analysis with Mr. Swartz, Caldera was approaching IBM about entering into this strategic business agreement that I discussed with Your Honor a few moment ago.

19

And that courtship resulted in, culminated in the execution of that strategic business plan. I explained previously how that agreement resulted in a license for IBM.

But independent of the license, Your Honor, and you'll recall I said I'd come back to this, there was a warranty and a representation made in that strategic business agreement that there was no infringing code of any third party in the Linux kernel. That representation was made by Caldera, by SCO's predecessor. It promised IBM, Your Honor, that there was no supposed infringement in there, and it also promised to hold IBM harmless and indemnify it against any third party intellectual property claims.

Now, rather than make a public issue of this, Your Honor, raises concerns, commence litigation, SCO undertook, instead, following that effort to initiate and to participate in and to champion the UnitedLinux effort. And Caldera and SCO were not simply participants in the UnitedLinux effort, Your Honor, they were champions of it. Ransom Love, the former CEO, has been described as the champion of the project, And Ralf Flaxa, a former Caldera employee, was described as coordinating the creation of this united standardized version of Linux. And again, everyone in aligned code is in their UnitedLinux product.

Courts, Your Honor, have estopped parties from asserting claims of discriminate based on showings far less

20

than that which is here. And respectfully, Your Honor, we would submit that the Court should exercise its powers to enter summary judgment in favor of IBM on grounds of estoppel.

That brings me to my next point, which is that SCO cannot establish, Your Honor, that there is substantial similarity between Linux and UNIX. There are two reasons here why summary judgment should be entered in favor of IBM. The first is that none of the supposedly infringed material, Your Honor, there were 320 lines of UNIX code supposedly infringed by these 326, and none of those 320 lines are protective by copyright. The second point, Your Honor, is even if they were, SCO couldn't just demonstrate that those lines of code result in substantial similarity as between Linux and UNIX.

Now let me briefly just take each of those points, Your Honor. As to protectability, in IBM's papers we laid out a variety of reasons as to why these 320 lines are not protected. And the law is pretty clear that the material is not protectable by copyright if it's dictated by externalities, if it's unoriginal or if it's mere merger material. As we laid out in our papers, Your Honor, those 320 lines of code are dictated by externalities, they are unoriginal and they are merger material.

And the notable point to make here, Your Honor, with respect to externalities, IBM put forward evidence that five separate externalities dictate that 320 lines of code.

21

SCO's experts failed to respond, Your Honor, in any meaningful way to two of them. And we respectfully submit that for that reason alone, those two externalities should be deemed admitted as against SCO. That's programming practice and industry standard.

As to originality, we likewise offer evidence that the code at issue lacks the requisite originality. And SCO disagrees with that, Your Honor, and has offered testimony of their expert Mr. Cargill. But the methodology on which Mr. Cargill relied to conclude that the material at issue is original is, we respectfully believe, wrong as a matter of law.

For example, Mr. Cargill stated, Your Honor, that the choices used to list names in alphabetical order and to assign numbers in a sequential order, evidences reference to the creativity to satisfy his standard to determine whether that matter is original within the meaning of the copyright law. And that is wrong for the reasons set out in our papers and the reasons set out in the Supreme Court decision in Feist and in 10th Circuit decision in Mitel.

And looking, Your Honor, at the code issue here, I would submit that there is no cognizable originality in this. Calling in error occurs when a person seeks to access a file for which they have no permission, EPERM, evinces no originality, anymore than saying that the 10th Circuit Court

22

of Appeals should be referred to at 10 Cir reflects any originality.

Finally, Your Honor, as to protectability, IBM put forward evidence that the material at issue is merger material. SCO failings to offer any meaningful response to that, Your Honor. And again even if it had, even if that which is offered by SCO's expert Mr. Cargill were sufficient to create a fact question, and we do not believe it is, but even if it were, Your Honor, that information is not information that was properly disclosed by SCO as it was required to be disclosed pursuant to Court's order.

And again, just looking at the material issue, Your Honor, there are only so many ways to say that a computer system is Linux, and it's this version, and it's that release. And that's what some of the structured declarations do here, Your Honor. There's only so many ways to put it, just like there are only so many ways to organize the information in a case name. Case, reporter, court and date.

Now, the second part of this, Your Honor, is that even if this material was protectable, and we submit it's not, SCO couldn't and we submit has not shown substantial similarity. Substantial similarity is not all about quantity to be sure. But quantity is an important nevertheless part of that as reflected by the 11th Circuit's decision in the Mitel case. And 320 lines, Your Honor, of the millions of lines of

23

code at issue here are not of significance.

And in any event, when you look to the qualitative aspect of this code, there is no more significance to the codes than there is when you look at it from a quantitative perspective. We are not talking, Your Honor, about 320 contiguous lines of executable code. We are talking about a scattered collection of code. None of the header files are executable, Your Honor. They are simply interface that provide information like if there's an error, let's call it EPERM and assign a number 1 do it. And that we respectfully submit is not protectable by copyright.

Finally, Your Honor, with respect to misuse, SCO has misused the copyrights at issue, and as a result of that misuse may not enforce them. SCO doesn't dispute that a copyright may not be enforced, Your Honor, if it has been misused. We think the facts here establish that. SCO claims control of more than a million lines of code supposedly dumped into Linux kernel. By contrast, there's 326 lines properly described in its final disclosures. And among the lines of code claimed by SCO as infringing, Your Honor, are files plainly and indisputably owned by IBM, like JFS, files plainly and indisputably owned by BSD and information which SCO's own experts concede they have no claims of copyright infringement.

For those reasons, Your Honor, respectfully summary judgment should be entered in favor of IBM. Thank you.

24

THE COURT: Thank you, Mr. Marriott.

Mr. Singer?

MR. SINGER: Your Honor if I might approach with some materials for the Court.

THE COURT: Sure.

MR. SINGER: And for counsel, as well.

Your Honor, if it may please the Court, this is a claim by one of the world's biggest enforcers of intellectual properties rights, IBM, which publicly boasts of annually earning over a billion dollars from enforcing its intellectual property portfolio to seek to extend its fight with SCO by seeking a judicial declaration that SCO conversely has no rights.

Now, as we listen to IBM's position and read their papers, it seems to be this. It seeks through its counterclaim to enforce and determine first that SCO has no copyrights, even though Mr. Marriott didn't want to argue that point. Yet, that the copyrights it does not own were assigned to a joint venture.

IBM seeks a declaration that UNIX copyrights don't cover any protectable expression and that they're basically worthless with respect to the issues as they relate to Linux. While at the same time, they are so important that they led to a series of no fewer than four licenses which were negotiated and apparently IBM claims reliance on the use of this

25

material.

And finally, IBM says that despite having obtained licenses, it could properly ignore the language and the limitations in those agreements and just rely instead on the fact that Caldera was involved in Linux activities, and so therefore anything IBM wanted to do was okay.

We submit just to state that contradictory refusing sets of arguments points out the fact that this is not appropriate for summary judgment.

Well, let's take a look at what IBM's Tenth Counterclaim really seeks. And I would ask Your Honor to turn to the small book, which are a set of our exhibits to Tab 1. And we ask, what does IBM's Tenth Counterclaim really seek? And I think that leads to a comparison of what IBM said when they filed their Tenth counterclaim back in 2004 and how they characterize it now.

If you look at that Tenth Counterclaim, Paragraph 171 particular, IBM said that it did not believe that its activities relating to Linux, not talking specifically about the Linux kernel, but it's activities relating to Linux including any use, reproduction and improvement of Linux do not infringe, induce infringement or continue to infringement.

Now today IBM says, we're just seeking a declaration from this Court that the Linux kernel, the core of

26

the operating system, does not infringe copyrights owned by SCO. Why the difference? The difference is so that IBM can stand up and seek to argue to the Court that only 326 lines of code are at issue. By defining the Linux kernel so narrowly, by relying on solution of items from our December disclosures, from assuming that everything in our expert reports concerning non-literal copying will be struck and upheld by this Court, by assuming that technologies that involve interaction between that kernel and the user space around it, such as STREAMS, such as ELF are out of case, they define it as 326 lines.

THE COURT: Let me ask you a question about that. Hypothetically, assume I don't reconsider my November 29th order and hypothetically assume that I uphold Judge Wells' November 30 order that was written on December 21st, what affect does that have on this motion?

MR. SINGER: Your Honor, everything in this book is still indicates --

THE COURT: This one.

MR. SINGER: The big book -- is still indicates assuming both of those orders were sustained. The code which is in this big book shows precisely with copying in red and lines tied to the items still in the case indisputably not struck by Judge Wells, the copy that occurred between Linux and our protected expression. They are in red. There are lines going from the left-hand to the right-hand side of the

27

page. They are divided into four categories, and there are thousands and thousands of lines.

We have put them in parallel in various places in the book, smaller book, but this shows why IBM is seeking to narrow the declaration it seeks from the Linux activities in general and its support of Linux and its copying of Linux because these things go along with Linux. Linux doesn't work without STREAMS. It doesn't work without ELF. It won't work without these header files.

By seeking to redefine their request for declaratory relief they're able to say only 326 lines are at issue, when in reality everything in this book, even assuming every order by Judge Wells is upheld, everything in this book is still in the case and it is still subject to this claim.

And we would submit it is meaningless, and I don't know why IBM would seek a declaration only on 326 lines when it doesn't eliminate the fear and uncertainty and doubt they talk about with respect to all the things around the kernel that is protected copyright and expression.

Your Honor, with respect to the burden of proof, this is IBM's action for declaratory judgment. And under 10th Circuit law, IBM has the burden of proof. If one turns to Tab 2, you see the Wuv's case, which is a District of Colorado district court decision, citing 10th Circuit case in Steiner Sales making it plain in this circuit, the plaintiff

28

in declaratory action carries the burden of proving its claims.

And you'll notice in the book which you received from IBM, they cite cases in the District of Maryland, the Northern District of California and the Eastern District of Pennsylvania, none, of course, which is in the 10th Circuit.

Now, I'd like to turn to the issue that Mr. Marriott did not address. I won't spend a lot of time on it, but they put a lot of time in their briefs on it, you've heard it about every opportunity, and it's a central issue in this case, and that's the issue of who owns the UNIX copyrights.

It is our contention, Your Honor, that the documents make clear that those were transferred as part of the sales of UNIX business from Novell to Santa Cruz in what we know as the asset purchase agreement. The agreement as Your Honor has heard clearly indicated the intent was to transfer that entire business. There was an assets schedule of assets to be transferred, which specifically said that all rights and ownership of UNIX and UnixWare including all versions, all technical installation, the source code, the documentation, all of sellers' rights under software development contracts, all those contracts, all of that is transferred.

The only reason we submit there could be any

29

question at all was because of the way in which an item in the excluded assets schedule was originally worded. It was worded in Item 5A of that schedule as excluding the copyrights, all copyrights and trademarks. That was in error. It didn't make any sense that you transfer the entire business of UNIX, you exclude the copyrights. And that was clarified in Amendment Number 2, which made clear that what Novell retained were copyrights and trademarks except for the copyrights and trademarks owned by Novell as of the date of the agreement required for SCO to exercise its rights with respect to the acquisition of UNIX and UnixWare technologies.

And I submit, Your Honor, it cannot be any real question you need the UNIX and UnixWare copyrights to run a UNIX and UnixWare business, the right to extend intellectual property to others, to enforce those intellectual property rights against infringements.

This amendment clarified what was transferred as of the time of the closing because then with this amendment there's no confusion as to what is covered on the schedule of transferred assets. It's all right, title and interest.

Now, we don't even think it's necessary to look to extrinsic evidence, and, of course, Mr. Marriott doesn't want to argue any of this. And I have to say I probably wouldn't want to argue a motion where witnesses on both sides agree that it was the intent to transfer the copyrights to

30

Santa Cruz.

And this is set forth in our book. It is included in summary form at Tab 11, and behind that, more lengthier quotes from each of these witness who have testified under oath both from Novell, the lead Novell negotiator, the senior business executive, the senior engineer as well as Santa Cruz that the copyrights were intended to be transferred.

Now, there are other extrinsic evidence of conduct set forth in our book that support that. There is simply no question we think here that there was a valid transfer, and those rights are owned by SCO. And, of course, it is the Novell trial the Court has set that will determine the ownership of those copyrights to the extent that Novell continues to assert that they did not transfer them despite the expressed intent of parties on both sides of that transaction.

Now, IBM contends that these copyrights that we did not own we nevertheless managed to assign to a joint venture called UnitedLinux. And before turning to the merits of that contention, I'd like to say a few words about the unusual strategic posture of it. Novell, as the Court will recall, moved last year to stay parts of the case involving Novell because their wholly-owned subsidiary SuSe Linux had initiated an arbitration in Europe against SCO where this issue is precisely what is to be determined under the arbitration.

31

What were the rights under the UnitedLinux agreement?

Now IBM with whom Novell has a joint offense agreement concerning these two cases comes in here and says, Judge, you should decide those issues as a matter of summary judgment in this case.

We don't think that is appropriate. And we think that if this is an issue which is going to be decided and the Court is going to look to it, decide it in a forum where parties directly involved are going to resolve these differences, as Novell urged this Court last year, and the Court agreed upon.

Now, it is clear that this is not a matter of summary judgment, and we think it is equally clear that the UnitedLinux agreement did not give up our intellectual property rights.

Your Honor, if you turn to Tab 15 we see that the key assignment of intellectual property rights was the intellectual property to be development pursuant to the JDC. You form a joint venture, and you say that we assign the rights which are going to come out of the work done by the joint venture. There was not an intention by participating in that to give up rights either to UNIX or to anything that might be sitting around in Linux as it went into that joint venture.

The term software is defined as implementation of

32

the Linux system to be developed through the joint venture. There is evidence in the record that the copyright in UNIX material was not developed pursuant to the joint venture. And, in fact, as a factual matter, during the joint venture, SCO was asked to contribute UNIX technology, and SCO refused. And that's Mr. Nagle's declaration, which is SCO's Exhibit 233.

Now, the preexisting material which Mr. Marriott refers to, which is listed on Exhibit C, and we have a copy of that behind our arguments slide at Tab 16, that is the only material which UnitedLinux of existing material from Caldera was authorized to use. And it did not include the System V UNIX code. Mr. Nagle says that in his declaration. Even Mr. Love, who is a witness for IBM, agrees with that. And therefore, it is not part of what either expressly in Exhibit C UnitedLinux had a right to use, nor is the issue here over material that was created in a joint venture.

Now, Your Honor, this is not the exact sequence in which Mr. Marriott dealt with these points, but I'd like to turn to the next issue, which is the question of whether we have put forth enough evidence to defeat summary judgment on the issue of infringement, that IBM by distributing and copying and inducing others to distribute and copy Linux have infringed our copyrights.

The very first argument that IBM makes in Tab 18 of

33

our binder relates to this, is that we cannot prove unauthorized copying. We didn't have a line, I suppose, in the December disclosures that says, IBM infringes our copyrights by copying and reproducing Linux. Now, that's what they say in their brief, and that's what they say in their papers. But listen to what IBM's counsel said to Your Honor two years ago, more than two years ago in September of 2004 when they were resisting discovery to the issue of their Linux activities. They said, quote:

It is not disputed that we copied Linux and we encouraged others to copy Linux. That's not in dispute. We admit that we copied. No discovery with respect to IBM's Linux activities is required.

And, of course, the record has Mr. Frye from IBM admitting the same thing. That constitutes infringement. And as the cases we set forth on the very next page at Tab 18 indicate, if Linux infringes UNIX, their activity in copying Linux is infringement. If Linux infringes our UNIX copyrights, their encouragement of third party copying and reproduction also infringement. So there's no issue here.

Now, the question is whether Linux is copied from UNIX in a way that violates our protectable rights. And that is inherently a factual matter for fact evidence and expert testimony. The 10th Circuit recognized that. We reproduced

34

this at Tab 19 when they said:

Whether the defendant copied portions of the plaintiff's program is a factual matter.

And it's the Gates Rubber case.

There are a lot of facts that indicate this. I mean, if one goes just to IBM's own documentary admissions, a couple of which we've put on this blow-up of Exhibit 276 where IBM internally says, Linux is derived from UNIX. UNIX was a pre-write of Linux. It is proved by this book, Your Honor, that these thousands of similarities did not occur by coincidence. This was copying and certainly at a minimum is a factual matter.

We have put in both in that book and here in four areas, system calls, ELF-related materials, STREAMS-related material and memory allocation material where there is no question, even under Judge Wells' order that those are indicates and we have appropriate claims.

Now, we do not agree with Judge Wells' order which limited our copyright case in that matter. At Tab 26, we point out the fact that in addition to copying literal code, code in that book, there was a copying of the non-literal protectable elements. The structure, the algorithms of UNIX. That didn't mean we contend to be in the December disclosures as specific material misused. If we had said they copied the Linux structure, IBM would move to strike it on the grounds

35

that we didn't cite line, version and code.

It is indisputably within the May 2005 expert report set forth which IBM ignores in its motion and IBM's motion is predicated on us not having any rights to talk about non-literal copying. Non-literal copying is recognized by the 10th Circuit in Gates and by every other case as a relevant inquiry where experts have to go through an abstraction process and determine what are the protectable elements, go through that and filter it of unprotectable material and then compare it.

IBM experts in IBM's motion have done none of that with respect to the non-literal issues. Just this chart which we produced shows the structure of a system calls between Linux 2.4 and UNIX System V Release 4 their similarity. That structure we contend is protected. We contend that the structure just as the STREAMS module, which is put forth at Tab 26, as well, is protected. Those should be issues in this case, Your Honor. Those are the issues we contend in this case.

Your Honor, I would like to say a few words about whether this is protectable expression. The Gates Rubber case was applied by our expert Tom Cargill. At Tab 28, we have an excerpt of Mr. Cargill's report, which recites the applicable 10th Circuit law, explains how he applied it going through these three steps. He specifically looked at whether or not

36

the expression being used was dictated by external standards which includes the filtration issue that Mr. Marriott talks about. It was his opinion that there was infringement.

You know, IBM knew we were going to rely on Mr. Cargill. They wait until their reply brief on this motion to take shots at Mr. Cargill saying he's not admissible, that he isn't using the right legal standard, and et cetera. If they continue to make that argument seriously and the way to do it and the time to do it we submit is a motion in limine or Daubert motion later in the case. If it is not on grounds coming out of a reply brief to all of a sudden saying to the Court without even giving any response to IBM should determine that Professor Cargill did not apply the right standard, and that when he says this is protectable expression, he somehow applied that test wrongly.

I'm not going to go through each of the tabs where we deal with the issue of whether this is Scenes a Faire material, whether or not there was enough originality in the expression to be protectable and whether it merged with abstract ideas. On each of those points, Mr. Cargill gives examples, and he indicates his opinion.

It is clear that you have a strong amount of protectable expression in software code, and we include, in fact, in these papers excerpts from an amicus brief that IBM filed in the Gates Rubber case saying as much, which we think

37

was consistent with Gates Rubber. It said that source code is generally protectable.

If one turns to Tab 33, Dr. Cargill's report is quoted there where he gives an example. First he says generally there are numerous ways to express the ideas embodied in the copied material. This is second page of Tab 33. And numerous ways to write code that performs the same task of copied material.

He gives an example of a fork being a system call, which could refer to a diverging path. It relays a certain creativity on behalf of the programmers. The system called nice is another example.

If you turn to the next page you see IBM's amicus brief being quoted as well as Gates Rubber decision. In that brief, IBM said it is not appropriate to conclude that a program's function -- which is really what Mr. Marriott spent a lot of time talking to you about, the function of a program. It is not appropriate to conclude that a program's function and its expression of that function are the same.

At bottom, Your Honor, these issues will have to be decided at trial including the issue of substantial similarity. You know, it's interesting that even if only 326 lines of code were at issue, it would still be a factual issue requiring trial on it.

If Your Honor turns to Tab 35.

38

We have at that tab the Dun & Bradstreet case from the Third Circuit, 27 lines copied out of 525,000 were held they could be substantial.

The US Supreme Court on the next page in the Harper & Row case said that 300 to 400 words cannot be deemed insubstantial. And they reversed the Second Circuit in that case for so holding.

The Dun and Bradstreet case on the next page is cited to point out that the real importance is determined qualitatively, not quantitatively. And that the information, while it was only a few lines far less than what we're talking about in this book, was highly critical. And there's no dispute on this record that if you took out of Linux the lines at issue, even just those 326 lines, let alone ELF and STREAMS and everything else that is at issue here Linux simply would not work.

At bottom, as the 10th Circuit said in Gates, the issue of substantial similarity is a classic jury question. It's not appropriate for summary judgment.

Your Honor, I would next like to turn to the issue of a license. And I'd like to begin by observing the interesting posture in which the issue of a license comes before the Court, because IBM's position today is that this case could be resolved simply by looking at the SBA, which IBM obviously was aware of as a party to it in 2003, or the GPL

39

public license, which Caldera they claim gave up its rights in.

And if that is true, Your Honor, where has IBM been with that argument since March of 2003 when this case was filed? If this case could have been resolved with respect to these issues simply by coming in here with a copy of the general public license or the SBA, why didn't they do that back in 2003? If that is true, why didn't they do it in 2004 when they made a summary judgment motion to you on this precise counterclaim and never raised either of those issues?

We submit the reason is because these licenses do not give IBM the rights that today they are asking you to find as a matter of law.

Now, I'd like to say a few words about the SBA license, the first that Mr. Marriott relies upon. Tab 37 in our binder and a couple tabs after that we discuss this agreement. This is an agreement when the Court reviews it is clear it is an agreement to allow distribution of certain products that are then specified in an Exhibit A to a standard of work. And that standard of work makes it clear that IBM is simply a conduit to end users for the distribution of certain Caldera products and is not being given any intellectual property rights to UNIX software, let alone to give away that UNIX software to others.

If one turns to the next page behind Tab 37, we

40

quote Section 8.3. Your Honor has been directed to part of Section 8.3 when IBM spoke. It was their Tab 26. You could compare Tab 26 which reproduces in part Section 8.3 as opposed to the entire section which in red it includes the material they do not quote. And that material makes clear that:

The providing party, which was the Caldera, Inc., will not include any preexisting materials in any deliverable unless they are listed in the relevant standard of work.

And when you turn to the standard of work here, nothing is listed here which gives them the right to distribute, to modify UNIX intellectual property or gives them rights to open source that through Linux or anything else. That document which, is IBM 467 says:

IBM and its authorized agents shall be a conduit through which Caldera sells, offers to sell packaged license work and preload license work.

And the only thing they're authorized to do is to preload, install and reproduce the preload license work on two particular platforms which are identified in this scope of work, and the master copies are to be used solely for purposes consistent with this standard of work. That is Section 3.0.

So at bottom, what this means is if SCO was coming in here and suing IBM for distributing Caldera products that were distributed under this standard of work, they would have

41

an offense. They could distribute that. But what this is not in any form of intellectual property license, any release of our rights, anything which gives them or anyone else the right to take UNIX intellectual property and distribute it to the world.

Your Honor, the other license, the GPL license, also fully known from 2003, never raised then, never raised in 2004. Today they suggest that all of IBM's Linux activity is insulated under the GPL. The GPL, we submit, requires a copyright holder to make a knowing, voluntary express surrender of its copyright rights to its software and then to effectuate that decision with a particular notice spelled out in the general public license, which did not happen here.

We have that at Tab 38. Section 0. And Mr. Normand in the argument that deals with the GPL which Your Honor will hear right after this will have more to say about this, but I'd like to say a few things about the GPL. Section 0 says that:

The GPL only applies to work if it bears a notice placed by the copyright holder saying that it may be distributed under the terms of the GPL.

SCO has never placed any language on either UnitedLinux product or SCO Linux Server 4.0 indicating that it was granting any license or rights under GPL or any open source license.

42

That's both in the record factually through Mr. Nagle. And while IBM makes generalized assertions that we distributed under the GPL, what, in fact, the record shows is you have products from three divisions. Two of them were Linux from UnitedLinux that were just passed off. Whatever license they came with came from them. And then there was a disk of proprietary material. There was no SCO license under the GPL Section 0 that was required. And, in fact, you will see during Mr. Normand's argument that IBM knows how that license looks for their material that they decided in knowingly and willingly to distribute under open source, you have that copyright authorization notice required under 0. It's set forth in the GPL how to do it. We reproduced those directions under Tab 38. And we've also put in IBM's copy of that.

Now, in addition, there is no question, and we put this forth at Tab 39, that Linux has no copyright attribution to SCO or to Santa Cruz in the materials distributed under the GPL. That in itself takes it outside that protection.

And I would note in connection with this that even Ransom Love admits that UNIX was not an open source by Caldera. Now, Mr. Love is someone who IBM submitted a declaration from and we submitted a declaration from. We disagree strongly with a lot of things Mr. Love said in that declaration. We also pointed out to the Court what Mr. Love

43

did not disclose, and that is he's a paid IBM consultant.

But even Mr. Love in his declaration, and I have this excerpted at the last page on Tab 39, even Mr. Love acknowledged that while Caldera thought about open sourcing UNIX assets, it never did know. And it certainly didn't do so under the GPL. And I submit this is why for the last three years of this litigation IBM hasn't run into court and say, GPL resolved these issues.

Your Honor, I would also not go into the details of the Spec 1170 license that they argue in their brief, other than to say it was not established it even covers the material here and was only in the material that create -- of license that created specification, not a license to use that in a commercial product that would compete with UNIX.

In addition, there is the TIS issue which is raised in their briefs which Mr. Marriott left to the briefs. I just want to point out that with respect to that which relates to the ELF, even Mr. Harold, who is an IBM employee, recognized and did not have the authority to that ELF code on the basis of that license. And that appears at Tab 43 where Mr. Harold in 1999 indicated that SCO was the only source supplier, that they wanted to deal with it for Linux, but they couldn't because of SCO's copyrights.

And, in fact, in addition, Mr. Cargill, and this is excerpted at Tab 44, shows that the Linux programmers went

44

beyond using what was ever in this specification. They went so far as to copy the actual code from the System V Release 4, and that wasn't authorized no matter how long it looks at the TIS license.

Your Honor, next I'd like to turn to the equitable defenses which IBM asserts are a basis for summary judgment and declaratory relief. I submit to you that IBM is probably the last party rather than the first party that should come into this Court and seek a determination of equitable estoppel which depends on the concealment of true facts from their knowledge and their blind reliance on false facts that SCO or Caldera or others are supposedly making or implying to them by their actions as to which they have no awareness of the truths. Perhaps equitable estoppel is an argument that a user out on the street might make, but to suggest that IBM of all parties doesn't have an awareness of this truth.

And the premise of this argument has to be we own the copyrights, that there is protectable expression in those copyrights that has been copied in the Linux, that they are infringing, that they don't have an express license, but nonetheless, they should have been allowed to do whatever they would and be immune from liability because they were watching Caldera and SCO distribute Linux.

And we submit that that totally perverts the idea of equitable estoppel and doesn't come close to the standard

45

required by law. And I note that IBM did not even mention equitable estoppel in its Tenth counterclaim. You can read that Tenth counterclaim. Didn't talk about it. It also wasn't raised in a motion for summary judgment that was raised before Your Honor back in May of 2004.

And I submit that the reasons are why equitable estoppel is seldom appropriate. At Tab 45, the Court is aware I think that equitable estoppel presents issues of facts and is one only a fact finder can draw. We believe that that's clear from those authorities, as well as the Deseret case. Under New York law, we turn to the last tab there, estoppel, we think this is true, generally under the common law requires concealment of material facts, the lack of knowledge by the party claiming estoppel, none of which IBM has come close to establish.

Moreover, and this is a point in our briefs, and it's the last point we have at Tab 45, equitable estoppel can't justify a broad declaration that our rights are unenforceable or that what other people have done is okay. It is a personal defense. And I think there's probably maybe one or two exceptions in patent cases where courts have rendered declaratory judgments based on equitable estoppel. But essentially that's an offense to be viewed in light of actions alleged of infringement at the time.

Now, that's the law. The facts here are such that

46

IBM is just jumbling together in their submission actions taken by Caldera, actions taken by SCO and leaving out an important issue, which is who owned the copyrights in question at what time?

This is a chart -- Your Honor will recall this chart from last week, because it is the same equitable estoppel argument. The copyrights are owned by Santa Cruz all during this period of time. Santa Cruz is not in the Linux business. They argue, well, there is a few people who attended an X/OPEN conference. Those people did so as individuals. Each of those sets forth in their submission of facts is disputed in our response.

Santa Cruz which owned the copyrights was not in the Linux business during the time of 2001 when IBM decided to embrace Linux, to form the Linux technology center, to encourage others to use Linux, to copy Linux, to contribute technology to Linux, that could not possibly then be based on reliance of any action by the copyright owner.

What Caldera, a company that did own the copyrights during this time, was doing with respect to Linux cannot be any more of a basis for IBM to suggest that the copyright owner doesn't care. But if I distributed Linux or you distributed Linux and IBM said they saw us distributing Linux so it must have been okay,

With respect to actions taken after the copyrights

47

that were transferred in 2001 to SCO, the evidence is clear that at that time several things happened. First of all, the business of SCO because of Linux being out there changed dramatically. The UNIX products went south, and you saw that chart back in the contract argument last week. At Tab 49, we point out case law to indicate that even if you had nothing more here a plaintiff's decision not to sue until infringement action become a competitive threat cannot give rise to an equitable estoppel defense.

Suit was brought in 2003. IBM points to the facts that, well, you still had code on the servers that might have been downloaded after we filed suit. But how could anyone rely on the fact that you could download code from a SCO server? They tried to discontinue their business without jeopardizing the customer relationships. But the relevant point for estoppel is IBM can't rely on any action after we actually sued them to suggest we're not enforcing our rights.

Now, if this was all there was, it would be enough for a factual issue with respect to estoppel. But there's a lot more. In addition to these facts is what IBM was doing. Santa Cruz saw IBM engaging in some Linux activities here during the time they were engaged in the joint venture called Project Monterey Your Honor heard all about on Monday. And they were concerned about this, and they asked IBM.

And what did IBM say? That's at Tab 47.

48

Mr. Michels, who is the Santa Cruz CEO, said that he and others at Santa Cruz informed IBM that they were concerned about IBM's announced support for Linux and how that might impact Project Monterey. And IBM's response was to emphasize that Linux was not being supported by IBM as a commercially hardened operating system and would not encroach on Santa Cruz' core markets or the markets targeted by Monterey, and that we need not worry about it.

This alone creates a sufficient factual issue where a jury can determine whether SCO and Santa Cruz and then later SCO had a right to rely on what IBM expressly told them when they raised the issue with IBM. Of course it turns out that IBM was working as fast as it could to bolster Linux' activity and ability to target the markets in which SCO contended.

Furthermore, estoppel requires reasonable reliance by the people at IBM making these decisions, and they have not said that they relied on SCO. And you will see no document in any of the files that is contemporaneous with the events in question where someone at IBM wrote and said, I think we can go ahead because Caldera is distributing Linux or because someone from Santa Cruz attends a conference and we are okay. You won't see that.

We asked Mr. Frye who is the head of the Linux technology center at his deposition about these issues, and his answers appear at Tab 48. He says he recalls no

49

conversations about SCO or any conversations about the rights that SCO and its predecessors held regarding Linux. There were no discussions within IBM about whether IBM's technical contributions would violate any third party rights. No conversations with Wladawsky-Berger, who was a person even on top of Mr. Frye.

So there's not any decision here either in documents or otherwise which says, we've looked at these sales and we know what we can do because of that. That is simply a position in this litigation. In fact, Mr. Harold's statement also is inconsistent with that.

They refer during argument to Mr. Love saying that we were prepared to give up rights. If you turn to the last slide in Tab 48, we point out first of all that what Mr. Love is being relied upon by IBM is actually disputed by Mr. Love's declaration for SCO. And remember, this is an IBM paid consultant. And he later said that Caldera team did not investigate the issue of whether intellectual property rights existed or it had been disclosures that violated those rights.

There are five other members of the Caldera board and senior management who dispute that the company ever made a decision that it didn't care about enforcing its intellectual property rights. If IBM wants to raise that type of estoppel issue at trial, they can do so, but there is certainly a serious factual issue concerning that.

50

Now, the last point which was raised by Mr. Marriott was copyright misuse, which in essence is dependant upon a lot of other arguments. We submit that like other equitable offenses is inherently factual nature and depends on the resolution of those issues.

SCO -- I'll just mention the one example Mr. Marriott mentioned. He said that it proves misuse by us invoking JFS as an example of material viewer asserting protection. As the Court knows from last week, there is a serious factual issue about JFS. Our experts contend that JFS was deprived from System V. In any event, it was part of the derivative system AIX on which there were proprietary rights.

There is no basis, certainly not as a summary judgment on a declaratory judgment to say that we are guilty of copyright misuse. What we're guilty of, Your Honor, is trying to enforce our intellectual property.

THE COURT: Thank you, Mr. Singer.

Mr. Marriott?

MR. MARRIOTT: Thank you, Your Honor.

THE COURT: Mr. Singer disagrees with you about the burden of proof.

MR. MARRIOTT: He does, Your Honor. Let me address that, if I may, briefly.

If you would look, please, at Tab 9 of our book. You'll see there laid out cases which make clear who bears the

51

burden of proof.

And Mr. Singer refers the Court to the Steiner case from the 10th Circuit, which is a decision from 1938, and the Wuv's International case from the District of Colorado, which is a decision from 1980. And if you look carefully at both of those cases, Your Honor, you'll see what the plaintiff was seeking in those cases was an affirmative declaration, not a declaration of non-infringement. And if you look at the more recent cases and the cases that address the precise question here, which is who bears the burden in an action seeking declaration of non-infringement, you will see that the overwhelming authority supports IBM's position in that regard.

Now, Your Honor, Mr. Singer said a number of things. Let me take some of his preparatory remarks first and then come to his arguments about our specific arguments.

He said, Your Honor, at the outset that there was, suggested at least, that there was some sort of gainsmenship here that IBM had brought its claim seeking a declaration of infringement, seeking a claim as to anything and everything related to Linux, and then suddenly now years later has figured out that that wasn't such a good idea, and we should narrow the claim.

THE COURT: Narrow it to the kernel.

MR. MARRIOTT: Narrow it to the kernel. As if the kernel, Your Honor, isn't as Mr. Singer well knows what people

52

refer to when they talk generally about Linux. And I will refer Your Honor, if I may, to IBM's reply papers. In its motion for summary judgment more than two years ago just after we filed with the Court's permission in our motion -- our claim seeking declaration of non-infringement, we were clear, Your Honor, from the outset, Your Honor, and I refer you to Footnote 3 of our reply brief, August 23rd, 2004, where we said, quote:

Linux is susceptible to multiple meanings and can be used in different ways. For purposes of its Tenth counterclaim in this motion, IBM uses the term in its generally understood sense to refer to the core Linux code that is available at http/www.linux.org and is commonly known as the Linux kernel.

That was long before SCO finally disclosed in its final disclosures what supposedly is at issue in this case. The suggestion that we waited around to figure out what was going to be identified and then gerrymander the case and claims to fit it, Your Honor, is simply unsupported.

Mr. Singer complains that I didn't address in my argument a number of allegations of infringement, if I may, which he refers to in his tables including a particular, this one, Linux copies the overall structure of UNIX, Your Honor, that allegation of misconduct was not disclosed as it was

53

required to be disclosed in SCO's final disclosures. And Magistrate Judge Wells ruled that SCO shouldn't be allowed to proceed as to it. It is in effect not in the case. So, yes, it's true, I didn't address it in my opening argument.

THE COURT: What's the reach of your argument there? Are you saying there are basically claims relating to any non-literal copying or not in the case?

MR. MARRIOTT: Your Honor, what I'm saying that what SCO identified with specificity in its final disclosures is in the case as it relates to this Tenth Counterclaim. And on that issue and the Linux kernel, there were 326 lines of code.

Now, contrary to what Mr. Singer suggests, and this goes to the first of my original points, Your Honor, IBM does not now know and SCO has never disclosed precisely what it is that IBM is supposed to have done. As I said in my first point, Your Honor, SCO can prove that IBM copied protectable materials covered by their supposed copyrights. And Mr. Singer says in response to that, well, IBM knows full well that it copies Linux. And he points to statements, I suspect my statements, Your Honor, sometime ago saying that IBM copies them. Of course IBM copies them. That's not the point.

When you say that SCO has not disclosed as the Court required it to do how it is that IBM is supposed to infringed. That's what put IBM frankly in the impossible

54

position of being unable to properly prepare a defense because we weren't aware of what SCO contends.

And if I can take an example, Your Honor, I pointed the Court to the first line of code in the first file identified, the EPERM-1 file. SCO has never said what it is about that, Your Honor, that supposedly represents infringement. Is it that the computer sends out an error when someone tries to access a file to which they don't have permission? Is it that there's a number associated with the error? Is it the name EPERM? If the name had been in Linux used has no permission, would that present the problem? Is it the association of the name and number together? What if it were no permission 6 instead of EPERM-1?

Nowhere, Your Honor, despite the Court's orders did SCO ever disclose as the request in the orders made perfectly clear it was required to do how it is IBM is supposed to infringe? That's what was required for the final disclosures, not just a bunch of lines of code, if I may, the big book that was handed up, Your Honor, which by the way includes the same code again and again and again.

There were 326 lines of code which are identified, contrary to whatever impression this book intends to give. 326 lines of code were in the kernel. And while there was, in fact, a red line drawn suggesting that those lines of code were apparently similar to those, not a single piece of

55

additional information was given as to what it is that supposedly represented infringement, what it was about that that supposedly represented actual infringement.

The orders weren't complied with. IBM as a result was not in a position properly to prepare a defense, and that alone is basis for summary judgment.

Now, Your Honor, Mr. Singer refers to the licenses. And he says there's no merit to the argument about licenses. If there were any merits, says Mr. Singer, IBM would have raised this questions two years ago.

THE COURT: That's what he said.

MR. MARRIOTT: What's the explanation for that, says Mr. Singer.

Your Honor, the explanation for that is we spent the last four years horsing around trying to figure out exactly what it was that supposedly IBM did. And it wasn't until we got the final disclosures after who knows how many motions to compel that we were able to say, this is what's in the case. And we can figure out that, in fact, they distributed that code under the GPL, or it was in the UnitedLinux agreements. And then we can figure out, Your Honor, that we, in fact, had those licenses. And until SCO told us what was at issue in the case, it was a little difficult to make an argument to the Court that it was covered by License A or B or C or any of the other licenses that, in

56

fact, covered this exact code.

Now, Your Honor, Mr. Singer suggests with respect to the UnitedLinux agreements that it's untoward for IBM to be making arguments of this sort here when that issue is presented in the Novell matter. Well, Your Honor, the arbitration agreement -- first of all, the argument is a new argument today. It's not an argument that was raised in the briefs. And even if it had been raised in the briefs, Your Honor, the arbitration agreement that applied apparently as between SCO and Novell is not an agreement by which IBM is bound, and SCO is in no position to invoke or seek refuge behind that agreement when if it had wished to make an issue of the ownership of the code at issue, it could have done that four years ago when it brought a case predicated on the proposition that they owned the supposedly infringed code.

Now, Your Honor, with respect to the SBA license. If I may ask Jefferson Bell to put up another board. Mr. Singer suggests that IBM was merely a conduit for Caldera. And as a result of being a mere conduit for Caldera, it didn't really get a license of any consequence.

Well, Your Honor, it may well be that IBM was a conduit for Caldera. But the license that IBM got in connection with the strategic business agreement was in no way limited by some motion being a conduit. The language which I read to the Court which I won't repeat now could not be more

57

broad. A worldwide perpetual license to use the material in the product identified.

And if Your Honor follows the schedule down, these are the defined terms in it, and I'm not going to repeat them, but IBM's license is in the deliverable. And all you have to do is follow that right down, Your Honor, to the SCO Linux 4 product. And in that product, Caldera gave IBM a license to do exactly what IBM supposedly is now doing improperly with the code, the theory to which has never been disclosed.

Now, Your Honor, with respect to the GPL. The suggestion has been made in a classic name game that somehow the code which they distributed for more than -- well, I want to say more than a decade, Your Honor, for nearly a decade, the code that they distributed under the terms of the GPL was shipped with copyright notice on.

Now, they come in here today and say, Your Honor, you know, what's the big deal here? That's UNIX code. It's only Linux code that goes into -- it's the same code, Your Honor. They say it's the same code. All they've done is change the name and say, you know, it's the same code, they allege. And again, we don't in any way concede that it is. But they say it's the same code, and just they called it a different name and pretend as if the representations and the promises that were made by them in distributing that code for profit for a very long time are without consequence. And

58

that, Your Honor, respectfully is contrary to in more respects than I can list the purposes, the tenor of the GPL, the general public license.

Now, with respect to estoppel, Mr. Singer suggests again, this isn't raised -- this is raised for the first time now in this point in the proceedings, and that must somehow suggest that there is no merit to the argument, Your Honor. Again the appropriate time to bring a motion as to estoppel is when one understands what's in the case.

And finally now we understand what's in the case. And following the Court's directions as to when the motions should be submitted, we brought it. If you'll recall, Your Honor, that we have brought the original motion for summary judgment immediately after filing the case. And it was only because the Court directed, and we respected the direction, that no motion should be filed again until discovery was closed and consistent with the Court's schedule. That can hardly be used as a basis to argue that IBM's motion for estoppel should be denied.

Now Mr. Singer suggests, Your Honor, that IBM ought not be allowed to make any allegations here of estoppel because it's a bad actor, and it's supposedly done very bad things in connection with Project Monterey.

Well, Your Honor, that's absolutely wrong. But it's all entirely irrelevant because the Monterey allegations

59

to which Mr. Singer refers has nothing whatever to do with the 326 lines of code at issue. Nothing whatever to do with that, Your Honor.

And Mr. Singer suggests, Your Honor, that somehow SCO should not be held responsible for the acts of its predecessors because they didn't own the copyrights at the time or they didn't know what was going on. Well, Your Honor, the law is clear that a company is bound by the conduct of its predecessors. The law could not be more clear in that regard. And I point you to Page 38 of IBM's reply brief, Judge. Estoppel is a doctrine of equity. And this Court has ample authority under which it exercises its equity to preclude a party that for nearly a decade distributed code under the promise it can be used with all the rights that they had from turning around a decade later under new management from disregarding the representations and warranties made from strategic business agreement, from disregarding the principles set out in the GPL, from disregarding the licenses given under the SBA, from disregarding the licenses given under the UnitedLinux, and pretending as if it's such a surprise to find out that there's a theory in which they think now maybe they can get somebody to maybe pay them some money. Respectfully, Your Honor, if there ever were a case for estoppel, this is the case.

Now, with respect to the similarity and

60

protectability, if I can ask Jeff to put up the chart, in effect what Mr. Singer has said, Your Honor, what Mr. Singer -- you may or may not be a baseball fan.

THE COURT: I am.

MR. MARRIOTT: Your Honor, good. Then you know that the pitcher is number 1 and the catcher is Number 2, the first base is 3 and second 4 and 5 is third and 6 is the short, left is 7, 8 is center field and 9 is right field.

THE COURT: I do know that.

MR. MARRIOTT: Pardon, Your Honor?

THE COURT: I do know that. I've scored a few games even.

MR. MARRIOTT: Pardon?

THE COURT: I've scored them, you know.

MR. MARRIOTT: What SCO contends, Your Honor, let's talk about those. Your Honor, in effect what they have claimed is that the pitcher, player pitcher is 1, player catcher is 2, player first base is 3 and so on. That's what those #defines represent, associating a number and a shorthand for a position and claiming that somehow it renders Linux so substantially similar to UNIX that they enact and claim rights.

And I respectfully submit, Your Honor, that the law is clear that short names of that sort and associating integers randomly with phrases like PP1 or EPERM-1 simply is

61

not protectable under the doctrines laid out in our papers by Professors Kernagen and Davis in their expert reports. And in no case can it result when it's 320 lines of non-contiguous essentially random numbers with essentially shorthand phrases represents substantial similarity.

Now, Your Honor with respect to misuse, briefly again, the facts here are simple. They claimed rights to more than a million lines of code in Linux. At the end of the day, there's 326 lines of code in which they have rights, and they have sought to exert the supposed monopoly they have and copyrights they claim to have over technology plainly owned by others.

For the five reasons I set out, Your Honor, summary judgment respectfully should be entered in favor of IBM. Thank you.

THE COURT: Thank you, Mr. Marriott.

Now, Mr. Singer, you don't get to reply. Why are you standing up?

MR. SINGER: I was hopeful the Court might ask for a sur reply, but I understand I'm out of time.

THE COURT: Okay.

Now the next motions we have IBM's motion for summary judgment on its claim for copyright infringement, the Eighth counterclaim, and SCO's motion for summary judgment on IBM's Sixth, Seventh and Eighth counterclaim, they ought to

62

argued together, hadn't they?

MR. MARRIOTT: They should, Your Honor.

MR. NORMAND: Yes, Your Honor.

THE COURT: Mr. Normand and Mr. Marriott. How long are we supposed to take on those?

MR. MARRIOTT: I think, Your Honor, the agreement was no more than 15 -- was it 20 minutes?

MR. NORMAND: 20.

MR. MARRIOTT: It may be 20 minutes each.

THE COURT: 20 minutes each?

MR. MARRIOTT: Yes. I can be briefer than that.

THE COURT: And then we have SCO's motion for summary judgment on IBM's Second Third, Fourth and Fifth counterclaim. Who's arguing those? Mr. Hatch? Is that your hand behind that big board there?

MS. SORENSON: And I am, Your Honor.

THE COURT: And Ms. Sorenson?

MS. SORENSON: Yes.

THE COURT: And how long are we supposed to take on that per side?

MS. SORENSON: 20 minutes each, as well.

THE COURT: Let's take a 10-minute break, and we'll come back about 20 to and do these last two arguments, all right?

MR. MARRIOTT: Thank you, Your Honor.

63

MR. NORMAND: Thank you, Your Honor.

(Recess.)

64

STATE OF UTAH )
) ss.
COUNTY OF SALT LAKE )

I, KELLY BROWN HICKEN, do hereby certify that I am a certified court reporter for the State of Utah;

That as such reporter, I attended the hearing of the foregoing matter on March 7, 2007, and thereat reported in Stenotype all of the testimony and proceedings had, and caused said notes to be transcribed into typewriting; and the foregoing pages number from 1 through 64 constitute a full, true and correct report of the same.

That I am not of kin to any of the parties and have no interest in the outcome of the matter;

And hereby set my hand and seal, this ____ day of _________ 2007.

______________________________________ KELLY BROWN HICKEN, CSR, RPR, RMR

65

THE COURT: Who's going first on this one?

MR. MARRIOTT: I believe I am, Your Honor.

THE COURT: Go ahead. Now, you've got -- you have cross motions, so you're each going to reply?

MR. NORMAND: That's fine with me, Your Honor.

MR. MARRIOTT: Likewise. If I may, yet, another book.

THE COURT: Sure. What would we do without these books?

MR. MARRIOTT: Your Honor, there are 16 separate works that are at issue on IBM's copyright claim, as are illustrated at tab 1 of the book, and each of those works represents a contribution by IBM to Linux. Collectively, those works represent more than 700,000 lines of code. And that is in stark contrast to the 326 lines identified by SCO in its final disclosures. We think, respectfully, that IBM is entitled to summary judgment on this claim, Your Honor, for three reasons.

First. IBM owns the copyrights at issue. Second is that SCO copied and distributed the copyrighted works. And the third is that SCO had no permission to copy and distribute those works.

Let me now, if I may, take each in turn.

First, Your Honor, IBM owns the copyrights for

1

the reasons that are illustrated at tab 3. The holder of a copyright registration, the law is clear, is entitled to the presumption of ownership if it is timely registered. You will see those at tab 4 of the book, Your Honor. IBM timely registered statements of registrations for each of these 16 works, and the only argument made with respect to ownership made by SCO lacks merit as is set out in tab 6 of the book.

The second point, Your Honor, is that SCO copied and it distributed the 16 IBM works at issue, as we summarize at tab 7 of the book. As we show at tab 8, your Honor, there is in fact no dispute that they copied them over a substantial period of time. If I may, for example, in IBM's statement of undisputed facts in support of this motion, we asserted and supported with admissible evidence, the following proposition.

In November of 2002, SCO began distributing a product it called SCO Linux 4 powered by United Linux, which included the 2.4.19 kernel. SCO Linux Server 4 contained verbatim the entirety of each of the IBM copyrighted works listed in the table above, with the exception of IBM's Omni Print driver. SCO's response, Your Honor? Disputed to the extent the word "admitted" suggests that SCO has acknowledged any wrongdoing in its distribution, but deemed admitted, Your Honor, as to

2

whether in fact SCO copied and distributed those works.

And if you would look at paragraph 36 at tab 8, you will see that IBM likewise asserted and supported with admissible evidence the following proposition: Pursuant to its business strategy over time, SCO distributed copies of SCO Linux 4 and SCO Open Server 3.1.1 to customers.

In making and distributing each and every one of these copies, SCO literally copied the IBM copyrighted works. SCO's response? Undisputed.

Now, Your Honor, the third point. The third point is that SCO acted without permission of IBM in copying and distributing these works. SCO's only permission to copy and distribute the 16 works, the 700,000-plus lines of code, was pursuant to the terms of the General Public License, and if you look, Your Honor, at tab 12, you will see that SCO acknowledges that. And I will cut right to the case with paragraph 26, where we asserted and supported with admissible evidence the following: Quote. Each of the IBM copyrighted works is licensed to others, including SCO, only under the terms of the GPL or the LGPL. Close quote.

SCO's answer? Again, undisputed.

The GPL, Your Honor, could not be more clear in paragraph 4 of that document, that you, as a recipient of

3

materials licensed under the GPL, may not copy, modify or sublicense the program except as expressly provided in the GPL, and that if you do: Quote. Any attempt otherwise to copy, modify or sublicense or distribute the program is void and will automatically terminate your rights under this license.

And, Your Honor, SCO lost the permission that it had to copy and distribute the 700-plus thousand lines of code when it, one, repudiated its license under the general public license and when it, two, breached or acted in excess of the terms of the agreement when it attempted to impose restrictions and obligations on IBM and others that are not consistent with the terms of the GPL.

And let me now quickly take each of those two bases for finding lack of permission, Your Honor. Your Honor, the law is, I think, clear that a challenge to the viability and to the life of a license represents repudiation of the license. In the Dow Chemical case, the Federal Circuit held that a stated intention to, quote, the viability and life of the license itself, is a basis for repudiation of a contract.

And the Restatement Second of Contracts, likewise says: Quote. Language under a fair reading that amounts to a statement of intention not to perform

4

except on conditions which go beyond the contract constitute a repudiation. Close quote.

Now, SCO, in the course of this ligation, repeatedly, Your Honor, has repudiated the terms of the GPL. If you will take a look at tab 13 of the book, you will see that as early as August of 2003, it began declaring publicly and in this court that the GPL is unenforceable. In fact, it said, Your Honor, quote, that it is unenforceable, void and/or voidable; that it, quote, violates the U.S. Constitution, together with copyright, antitrust and export control laws; that it is unenforceable and inapplicable in litigation and that it is preempted under federal copyright law and unenforceable under state law.

Those arguments, Your Honor, those assertions, those representations, represent a repudiation in clear and precise terms of the GPL as a valid, viable license under which SCO could have continued to distribute and copy IBM's code.

Putting aside, if you will, repudiation, Your Honor, SCO lost the license, the right to use and distribute IBM's works, when it acted in excess of, when it, in effect, breached the terms of the general public license. A copy of the GPL was, of course, provided with our papers. There is another copy at tab 16 and with

5

some of the more important provisions highlighted, Your Honor. I would refer you, if I may, to tab 16 -- or tab 17, rather, of the book, where you will see a flow chart or diagram laying out several examples of SCO's breach of various provisions of the GPL, and I would like, if I may, to walk Your Honor through two or three examples of that.

First example is that SCO breached the GPL, Your Honor, by restricting redistribution rights. Section 6 of the GPL says -- this is at tab 18 of your book -- that each time you redistribute the program; that is, a program licensed under the GPL, or any work based on it, the recipient automatically receives a license from the original licensor to copy, distribute or modify the program subject to these terms and conditions and that you may not impose any further restrictions on the recipient's exercise of the rights granted herein.

And yet, Your Honor, subsequent to the commencement of this case, SCO attempted to do exactly that with respect to users of Linux, including IBM. For example, if you turn to tab 19 of the book, you will find an excerpt from SCO's web site in which it describes the license that it represented to the world was required if one wished to use Linux after it asserted that there was, in Linux, infringing UNIX code. The license is described

6

in SCO's words, Your Honor, as SCO intellectual property license for Linux.

SCO stated that the license was created because, in their words: Quote. Many customers are concerned about using Linux.

The license, Your Honor, applies to the commercial use, by its terms, of a Linux operating system that contains the 2.4 or later version of the Linux kernel. And notably, for this purpose, and for Section 6 of the GPL, SCO said on this web site describing its Linux license that the license did not grant any distribution rights, despite the fact that the GPL expressly required that SCO was required to provide distribution rights as to any material received by it under the GPL and distributed by it under the GPL.

Now, there isn't just, Your Honor, a web site describing a licensing program. In fact, SCO entered licenses in the marketplace with a variety of companies including, for example, Everyone's Internet. On March 1, 2004, SCO representatives gave Everyone's Internet a license that it described as a Linux license or a, quote, Linux IP license telling Everyone's Internet, Your Honor, that the license was required to use the so-called SCO IP on Everyone's Internet's Linux service.

Likewise, Your Honor, on August 10, 2003, SCO

7

granted a license to Computer Associates, which it styled as a UNIX license, but under which it purported to grant rights to Computer Associates to code in the Linux kernel. On October 14, another. On October 19, another, and so on. The particulars are laid out in our papers.

That, Your Honor, represents an impermissible imposition of obligations under the terms of the GPL and in so doing so exceeded its rights under the GPL, and when they did that, Your Honor, they automatically, by the plain language of the GPL in Section 4 lost the right to continue to distribute the code that IBM gave them the right to distribute, but only on the terms of the GPL.

One more example, Your Honor, and I will sit down.

(Someone sneezes.)

Bless you.

SCO breached the GPL, Your Honor, by restricting rights of those who received the GPL to source code. Section 1 of the GPL provides, Your Honor, and this is at tab 20, that, quote, you may copy and distribute verbatim copies of the program source code. The preamble of the GPL says that you must make sure that recipients receive and get the source code. Yet, Your Honor, when SCO undertook its licensing program, it restricted the rights of licensees to access, to copy and

8

to distribute the Linux source code received by them under the terms of the GPL.

So, again, turning, Your Honor, to tab 19 and to SCO's description of its Linux license, it says there, Your Honor, that SCO's -- SCO states that the license, quote, grants the right to use SCO IP in Linux in binary format only. That is not source code. They further say, Your Honor -- at page 2, SCO explains: Quote. The license does not grant any rights to use SCO IP in source format. Close quote.

Here, again, Your Honor, just a second example. There are more which I won't bother you with now that represent excesses by SCO of the rights and obligations that it had, having received code under the GPL. It sought further to restrict people's use of it. And it isn't just that description, Your Honor. In a press release of August 5, 2003, SCO announced that the license it was offering, quote, permits the use of SCO's intellectual property in binary form only as contained in Linux distributions.

It said that it's SCO's source division which was responsible for these licensing programs was to sell licensees the right to use, quote, SCO intellectual property contained in Linux in binary format only, not in source.

9

For those three reasons -- for those three reasons, Your Honor: IBM owns the copyrights. SCO copied the copyrighted material, and it did so after it lost permission to do so because its rights automatically terminated under the GPL when it exceeded the terms of the GPL and when it repudiated the GPL as a license, its only source of rights to copy and to distribute IBM's code.

Thank you, Your Honor.

THE COURT: Thank you, Mr. Marriott.

Mr. Normand.

MR. NORMAND: Hi. I have a book.

THE COURT: Sure.

MR. NORMAND: The usual binder.

THE COURT: Thank you.

MR. NORMAND: Your Honor, IBM's three counterclaims relating to the GPL are merely an effort by IBM to try to retaliate against SCO for its efforts to enforce it's intellectual property rights in it's UNIX System V technology, which no copyright owner ever contributed to Linux and which the GPL plainly does not preclude SCO from seeking to protect.

IBM's sixth, seventh and eighth Counterclaims all turn, as you heard Mr. Marriott say, on the assertion that SCO has breached the GPL or has repudiated the GPL.

10

The facts show that SCO has not breached or repudiated the GPL. In addition, there are two other reasons to grant SCO's motion and to deny IBM's motion that I will discuss today.

First. The evidence shows that Linux is an unauthorized derivative work of SCO's UNIX System V technology, as you heard Mr. Singer allude to, and therefore IBM cannot enforce its copyrights in its copyrighted material in the derivative work because that is a derivative work as to which SCO has the rights.

Second. IBM comes to the Court with unclean hands and therefore, as matter of law, is not entitled to the relief it seeks in its eighth Counterclaim. Now, although SCO has not moved for summary judgment on the question of damages, it is worth pointing out the context in which IBM brings these Counterclaims.

IBM's principal claim for damages is now that there are nominal damages available to it under the Copyright Act and that it should get its attorney's fees for pursuing these Counterclaims. Indeed, any claim that SCO has actually damaged IBM through its conduct in distributing Linux, beginning from October -- August, 2003, is illusory. IBM points to what can only be described as a very obscure instance in which one technically oriented person could access the Linux source

11

code through SCO's web site and to the fact that the third party company that hosts SCO's web site runs Linux.

As I said, we don't move for summary judgment on the question of damages, but these facts put into perspective the motivations behind IBM's pursuing these counterclaims.

First, your Honor, SCO has not breached the GPL and has not repudiated. As Your Honor has heard, IBM alleges that SCO breached the GPL by selling a license that allowed ends users to use SCO's UNIX material in Linux. IBM says that such licenses breached the GPL because the UNIX material had already been licensed under the GPL and therefore SCO is not permitted to put additional restrictions on it. The premise of the argument is flat wrong.

The plain language on the GPL and the undisputed facts show the UNIX material in Linux was never licensed under the GPL. Section zero of the GPL which we have set forth on the board closest to Your Honor contains an express requirement for a copyright authorization notice, which we quoted, as well, at tab 3 in the book and which we have placed on the board closest to the Court.

Section zero says: The GPL, quote, applies to any program which contains a notice placed by the

12

copyright holder saying it may be distributed under the terms of this general public license. The form of notice is attached to the end of the GPL itself, as we show at tab 5 and on the second board we put up before the Court. In addition, the web site describing the terms and operations of the GPL contains detailed instructions regarding, quote, how to use the GPL and states that quote, the process involves adding two elements to each source file of your program, a copyright notice and a statement of copying permission. The web site explains that the copying permission statement should come right after the copyright notices.

The GPL, thus, does not even apply to material that lacks the copyright authorization notice. The GPL makes the same point in another way, Your Honor. By its express terms, the GPL applies to the Capital P program referred to in the license. Absent the inclusion of the UNIX material by the UNIX copyright holder, let alone without the copyright authorization notice, the UNIX material was not part of the, quote, program by the express terms of the GPL. Accordingly, in imposing restrictions on the extent to which certain Linux users could copy and distribute the UNIX material, SCO was not restricting the use of the whole program under the GPL.

The GPL thus makes clear that the decision to

13

have copyrighted material licensed under the GPL must be the result of a knowing and deliberate decision, not the result of error or accident or misappropriation. The GPL gives everyone the choice of being able to decide to be a licensee under the GPL or of deciding to be a licensor of its own material under the GPL.

IBM says that we made a distribution, quote, under the GPL. That raises a semantic issue that was the distinction I just draw -- I just drew. We may have been a licensee under the GPL, turned around and distributed the GPL software verbatim, but we did not make a choice to become a licensor of our own material under the GPL.

IBM tries to downplay -- they don't even address it -- the clear requirements of Section zero, but they are fundamental to the whole spirit of free software. Free software invites copyright holders to decide whether to make their copyrighted material available under a free software license. The assumption is that the base or foundation of the free software product is material that was deliberately concertedly, legitimately placed into the free software product.

Section zero of the GPL captures that basic precept. Indeed, we cite in our briefs, Your Honor, that Section zero codifies the basic principle that an individual can only give away the rights that he,

14

himself, possesses. IBM acknowledges in its own brief that in a free software model, it is the quote, copyright owners, end quote, who choose to make their source code available under a free software license.

In fact, IBM knows that identification of the copyright owner is a necessary component of an effective GPL license, as demonstrated by the fact that IBM placed such notices on its own contributions to Linux. We have highlighted that language at tab 6 of the binder and on this board closest to me for the Court.

Here it is undisputed that none of the copyright holders of UNIX contributed UNIX material into Linux for distribution under the GPL, let alone the requisite written, affirmative copyright authorization notice. The former president and CEO of Caldera, Mr. Love, whom you have heard referred to today, has made the point expressly, and we quote that at tab 7, Your Honor. He says: After Caldera International acquired the UNIX assets from Santa Cruz, I and other officers of Caldera International considered donating the core UNIX source code that Caldera International had acquired to the Linux development effort or licensing it under another type of open source license as a way to leverage and possibly enhance our UNIX and Linux businesses. Caldera International, however, never made a decision to

15

open source those UNIX assets. If Caldera International had licensed UNIX openly, it would not likely have done so under the terms of the GPL but would have chosen some other more commercially friendly but industry accepted open source license that would have strongly positioned Caldera in the Linux market.

Mr. Love has also pointed out in his testimony in this case this although Caldera did decide to open source the so-called ancient UNIX technology the company expressly decided not to open source the UNIX System III and System V technology at issue in this case, and IBM knows that Caldera made that decision. We submit at tab 22, Your Honor, Caldera's letter to Linux enthusiasts specifically excluded UNIX System III and System V technology from its contribution of ancient UNIX technology to the open source community.

The UNIX material in Linux was, thus, never licensed under the GPL. No copyright holder ever submitted the copyright authorization notice and SCO therefore could not have breached the GPL by licensing the UNIX material to Linux end users as it did. IBM's only response this argument in its briefing is to cite a copy of the GPL contained in SCO Linux Server 4.0 and you heard Mr. Marriott refer to that in the first argument today, as evidence that SCO, quote, placed an effective

16

GPL license notice on Linux, but there was no copyright authorization notice specifying the contribution of UNIX or any other technology. It was just a copy of the GPL, and it makes -- Section zero of the GPL makes clear, Your Honor that it would not be enough just to include a SCO -- excuse me -- a GPL license in the product.

IBM specifies this program is free software. You can redistribute it and/or modify it under the terms of the general public license, and it includes at the bottom a copy of the general public license. But section zero required IBM to do this if they wanted to open source the product. There is no such language in the SCO Server 4.0 product that IBM refers to. In fact, SCO didn't even place that notice in the product. And that's undisputed from Mr. Nagel's testimony at SCO Exhibit 233.

Now, I want to be clear, Your Honor, about why IBM is wrong in claiming that SCO had licensed the UNIX material in Linux under the GPL when it distributed verbatim copies of Linux as part of SCO's Linux Server 4.0 And we explained this point at pages 8 to 10 of our brief. The point is that the GPL licensee does not, by distributing a verbatim copy of the GPL software, claim to have made any material in that software licensed under the GPL.

A GPL licensee does not claim to be a GPL

17

licensor of any of the material of the software merely by distributing verbatim copies of the software. IBM appears to argue that the result should be different here because SCO owned the UNIX technology at the time it distributed verbatim copies of Linux, but that fact does nothing to eliminate the requirement for the copyright authorization notice in the GPL, the requirement of a deliberate decision to subject one's code to the GPL under the plain terms of the GPL.

In addition, Your Honor, SCO has not breached the GPL in offering its UNIX license for several additional reasons. IBM first argues -- at least in the briefs they did -- that SCO breached Section 6 of the GPL. That's wrong. Section 6 states that 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 recipient's exercise of the rights granted herein.

Now, I want to spend some time on Section 6, as Mr. Marriott did. Section 6 applies on its face. Section 6 applies to restrictions placed by the licensee on third parties to whom the licensee has distributed the GPL program. IBM repeatedly quotes in its brief the

18

second part of Section 6 without quoting the first part which says that Section 6 applies each time you redistribute the program. The Section discusses the license that the quote recipient receives and provides that the licensee may not impose any further restrictions on the recipient's exercise of the rights granted herein.

SCO has never attempted to collect licensing or royalty fees from anyone who received a Linux distribution from SCO, as reflected in SCO Exhibits B and 49, testimony from SCO executives. SCO sold a UNIX license only to those who acquired a Linux operating system from someone else. SCO hasn't breached Section 6.

IBM contends that anyone who received a Linux distribution -- they contend in their briefs -- from SCO subsequently paid a fee for the use of SCO Linux. But that is wrong. SCO agreed to hold Linux customers harmless from any SCO intellectual property issues. SCO didn't provide a compliance license, contrary to what IBM asserts, to customers in consideration for the fees that they had already paid. The testimony from the SCO sales person on that issue was customers received a compliance license at no extra charge.

It bears emphasis, moreover, that IBM's reliance on Section 6 of the GPL suffers from the same fundamental defect, that the UNIX technology was never

19

licensed under the GPL, the point I started with. Section 6 says that the license granted from the licensee is from the, quote, original licensor. The UNIX material in Linux was not contributed to Linux by any UNIX copyright holder. As such, with respect to the UNIX material covered by SCO's licenses at issue here, there was no original licensor.

IBM admits as much in its reply brief. This is at page 9, where it acknowledges that Section 6 unambiguously provides that Linux licensees, quote, receive their license to Linux and the IBM copyrighted works from IBM itself and others who originally licensed Linux and its component parts under the GPL.

The fact is that no UNIX copyright holder ever did so for the UNIX technology at issue here. And Section 6 of the GPL makes this point expressly when it says: Quote. You may not impose any further restrictions on the recipient's exercise of the rights granted herein.

Where there was no original licensor of the UNIX technology, no such licensor had granted rights under the GPL so that there were no, quote, rights granted herein that SCO was placing any restrictions on through its SCO source licenses. IBM cites in its brief Section 2-B of the GPL in contending that SCO was

20

obligated to cause Linux, as a whole, to be licensed to all third parties and appears to argue that SCO breached that section by charging a fee for the use of Linux. Section 2-B states: Quote. You may modify your copy or copies of the program or any portion of it, thus forming a work based on the program and copy and distribute such modifications or work under the terms of Section 1 above on certain conditions, including that the modification be licensed at no charge.

On its face, then, Section B plainly applies only to a work that was modified by a licensee. It is Section 1 of the GPL that addresses the distribution of verbatim copies, and that section does not contain the prohibition on no charge licensing. Again, the web site addressing the GPL says, quote -- Section 2 says that modified versions you distribute must be licensed to all third parties under the GPL.

Now, the evidence shows that SCO did not modify Linux. IBM claims in its brief that SCO made certain modifications and additions to the Linux 2.4.19 kernel. IBM previously cited a product announcement for that proposition, but SCO showed that product announcement didn't indicate that SCO had modified Linux, and IBM now doesn't rely on the product announcement anymore.

SCO did not in fact make any modifications or

21

additions to Linux. SCO redistributed Linux 2.4.19 as is, on two disks received from the company SUSE. IBM cites the provision in its briefs in the United Linux joint development contract for JVC, but that provision identifies only modifications that SUSE was to make to Linux. SCO never made any modifications.

IBM interprets in its briefs Section 3 of the GPL to mean that there can be no charge for any distribution of the GPL software whether the person distributing the software has modified it or not, but Section 3 only imposes additional requirements on how distributions are made under Sections 1 and 2. It does not even suggest that the no-charge provision in Section 2, as to modified works, applies to Section 1, which pertains to verbatim distributions.

IBM argues that the GPL's express grant of permission to charge transfer and warranty fees under Section 1 should be read as a prohibition on all other types of fees. We Explain in our briefs why that is not sound canon of the construction of the GPL and made the point that Section 2-B, which prohibits licensing fees where the licensee distributes modified works, shows that the drafters knew how to write fee prohibition where that was intended.

You heard Mr. Marriott claim that SCO violates

22

Section 4 of the GPL. That's wrong as well. Section 4 states you may not copy, modify, sublicense or distribute the program except as expressly provided under this license. To the extent there is no breach of the other sections of the GPL, there is no breach of Section 4 because SCO's distributions of IBM's programs were as expressly provided under this license. Moreover, Section 4 only deals with copying and distribution of GPL-licensed works. It does not and cannot be read to constitute an agreement by a licensee to refrain from charging for other products in which that licensee holds a proprietary interest or to desist from enforcement of its IP rights.

Now, IBM has argued that SCO has repudiated the GPL. The record makes very clear that SCO did no such thing. We set forth the relevant law, Your Honor, at tab 16 and 17. Repudiation occurs when a party to a contract makes an overt communication or intention or action which renders performance impossible or demonstrates a clear determination not to continue with performance. A party repudiates a contract when it refuses to perform and communicates that refusal distinctly and unqualifiedly to the other party. There is no evidence that SCO has made any such refusal.

We cite the bulk of that evidence and testimony

23

in our briefs. I'll highlight some now, Your Honor. IBM suggests in its brief that Mr. McBride of SCO gave testimony constituting a repudiation in the course of his deposition. Under the Tenth Circuit law, however, repudiation requires more than statements that a party has misgivings about a contract. Mr. McBride's testimony, at most, reveals his misgivings about the GPL, misgivings which, by the way, an entire community of commentators share, which, as a matter of law, is not enough to support repudiation. In fact, Mr. McBride's testimony as to his views on the GPL neither demonstrates any intent to breach the GPL or demonstrates any determination not to continue with performance under the GPL.

Mr. McBride said, at deposition that to the best of his understanding, we haven't done anything that would violate the GPL. IBM even argued in its brief that SCO continues to repudiate the GPL now by disagreeing with IBM over its interpretation of the GPL. We cite cases, Your Honor, in our brief showing that IBM cannot boot strap a claim of repudiation onto the parties' disagreement over the meaning of certain provisions of the GPL. We cite those cases at tab 17.

Your Honor, the string running through all of these reasons for rejecting IBM's interpretation of the

24

GPL is that the license expressly gives everyone the choice of being either a licensee of the GPL software or a licensor of its own software subject to the GPL, and nothing in the GPL prohibits a party from licensing its own software or entering into settlements to resolve disputes regarding that software, and any such interpretation of the GPL would run counter to well established doctrines of public policy as we set forth in our brief.

Now, IBM spends some time in their opening argument on the -- I believe it was the Merrill Dow case. We address that case -- excuse me, Your Honor, Dow Chemical case. We address that case, Your Honor, in our opposition brief at pages 22 and 23. I won't dwell on that other than to say that that case affirms the rule that a repudiation occurs when one party refuses to perform and communicates that refusal distinctly and unqualifiedly to the other party. SCO has never said any such thing.

The second point, your Honor, regarding one of IBM's slides on this issue of breach. Tab 19 of IBM's binder shows that in its intellectual property license of Linux, SCO was seeking to protect its IP, not all of Linux. Language in the second full paragraph says: The license ensures that Linux end users can continue to run

25

their business uninterrupted, without misusing SCO's intellectual property.

And it goes on to say: This new SCO license is a binary right to use SCO intellectual property in the distribution of Linux.

Now, I said at the beginning of my argument, Your Honor, that there are two additional reasons for denying IBM's motion. First. Linux is an unauthorized derivative of SCO's UNIX System V technology. Now, the relevant case law there is set forth at tab 18. Section 103(a) of the Copyright Act states: Protection for a work employing preexisting material of which the copyright subsists, does not extend to any part of the work in which such material has been used unlawfully.

Now, what does that mean? The Court's have told us. As one Federal Court has explained: The case law interpreting 103(a) supports the conclusion that generally no part of an infringing derivative work should be granted copyright protection. That is particularly true under the case law, Your Honor, where the preexisting material in the derivative work tends to pervade the entire derivative work.

We believe we have made that factual showing with respect to the UNIX System V material in Linux. The precedent makes clear that the relevant question is not,

26

as IBM suggests in its brief, whether its copyrighted material is derived from UNIX System V. That's not the relevant question. The question is whether the IBM material is part of a derivative work of SCO's copyrighted technology in which SCO's UNIX technology tends to pervade, and it does.

The second reason for denying IBM's motion and granting SCO's, Your Honor, is that IBM comes to the Court with unclean hands. That doctrine closes the doors of the court to one in equity tainted in inequitableness or bad faith relative to the matter in which the plaintiff seeks relief.

In the copyright context, the doctrine applies when the plaintiff's transgression is of serious proportions and relates directly to the subject matter of the infringement action. The first basis on which IBM comes to the Court with unclean hands is that IBM's contributions to Linux violates those rights. That's the argument you heard last week, Your Honor. And in the Tenth Circuit, where a plaintiff interferes with a defendant's ability to comply with his or her responsibilities, a Court of equity will not turn a blind eye to the net effect on the parties' equitable relationship.

Here SCO determined to offer its UNIX licenses

27

because IBM and others had misappropriated SCO's proprietary source code into Linux. Indeed, among the material that IBM has appropriated and contributed into Linux is one of very the programs, as Mr. Marriott said, the Journaled File System or JFS, for which IBM claims copyright in its motion. IBM says in response to this subject that it did not improperly contribute that material into Linux. That is a hotly debated fact issue.

In addition, Your Honor, SCO provided its customers who purchased SCO Linux Server 4.0 with a password to enter a log-in screen. You've seen this in the brief, the issue of IBM hacking into the SCO web site. In addition to the password requirement that IBM ignored in accessing the web site, SCO had clearly stated on its web site that the files were only for SCO's customers. IBM attempted to use those files previously as evidence in support of it's eighth Counterclaim and in support of the previous motion for summary judgment.

The Computer Fraud and Abuse Act precludes such hacking and makes it illegal. IBM responds by insisting that it no longer relies on the evidence that it obtained through hacking into that web site and that it relied on its previous motion for summary judgment. We submit that misses the point. IBM can't insulate itself from its prior conduct by claiming to forego the fruits of that

28

conduct. Now, by improperly and illegally obtaining that evidence for its claims, IBM comes to the Court with unclean hands.

At bottom, Your Honor, IBM argues that even though neither SCO nor any other UNIX copyright holder ever decided to license any of its UNIX System V technology under the GPL, and even though the companies, of course, never submitted the requisite copyright authorization notice and even though nothing in the GPL precludes a copyright owner from enforcing its rights in material contributed by someone other than the copyright owner; notwithstanding all those facts, according to IBM, SCO violated the GPL when it charged people for the right to use SCO's UNIX technology in Linux.

We submit that IBM's interpretation is wrong and therefore the Court should grant SCO's motion and deny IBM's. Thank you, Your Honor.

THE COURT: Thank you, Mr. Normand.

Mr. Marriott.

MR. MARRIOTT: Thank you, Your Honor. Several points. With respect to ownership. The suggestion has been made that IBM lacks ownership of the 16 copyrighted works at issue despite the fact that it has a timely-filed registration statement from the Copyright Office under which it has and is entitled to a

29

presumption of ownership because, as Mr. Normand puts it, Linux is an unauthorized, unlawful derivative work of UNIX. There is no evidence, Your Honor, in the record, no evidence no SCO's final disclosures which would support that proposition, and there is no evidence to overcome the presumption that IBM is the author and the appropriate copyright holder of those copyrights.

Now, Mr. Normand suggests, Your Honor, and I think ultimately this motion depends upon the idea that SCO didn't breach the GPL as to the code that is at issue because it did not take a registration -- excuse me -- it did not take a slug like this and put it in the GPL on the exact code and say: This is UNIX code, and we're SCO, and we're giving you rights to do with it under the GPL. Your Honor, that is nothing more than a name game, than a labeling game.

If Your Honor would look at the chart over here, what SCO is essentially doing, Your Honor, is saying that this code, which we'll assume to be in Linux, this code is not subject to the GPL even though it shows up in Linux, even though there is a copyright designation on it, even though it says it's distributed under the GPL just because it doesn't say: This is SCO code, and it's UNIX code in Linux. The code in question, Your Honor, which they claim is not under the GPL is, they say, not

30

under the GPL just because they haven't, they say, put on it a label that says: We're SCO. We are giving up this UNIX code.

The fact of the matter is, Your Honor, the code we're talking about is there. It bears a GPL notice. It was offered to Linux by the writer and creator of that code pursuant to the terms of the GPL. And whether or not -- I respectfully submit that SCO has, in fact, licensed that code, despite what Mr. Normand suggests. But the fact of the matter is: Whether or not it was licensed as a licensor, whether or not SCO was a licensor, they were a licensee. They got it just like they got the IBM code, the 700,000 lines of it, that bore the representation and notice that they were under the GPL, and only to be used under the GPL.

And now Mr. Normand says: Yeah, but that doesn't matter, Judge, because all of Linux is a derivative work of UNIX System V, and nothing there, under their argument, Your Honor, is at all under the GPL because they purport to own it all. Its all a derivative work. And no one put it in -- there is, to my knowledge, at least, no notice in the GPL that says -- in Linux, rather, that says: This is all SCO code, and we are giving it up under the GPL.

And that's essentially their argument, Your

31

Honor. There is not a notice that says that. What there is in fact is the code with the GPL notice by the person who wrote it saying that it is offered under those terms. And all they are really saying, Your Honor, is because they claim there is infringement in that code, they say it's really UNIX code and so the appropriate notice had to be one that said this is actually UNIX code. And since they didn't do that, the GPL has no absolutely no application.

But if they were right, Judge, if they were right in that argument, then all anybody would ever have to do is to say: You know, what, I think Linux infringes and so it's an impermissible derivative work, and I am therefore no longer subject to the terms of the GPL, and I'll go off and sue you, and then I will continue to distribute it under the terms of the GPL, putting the notice on there and make my profits from it and pretend as if it applies to me selectively.

We're talking about the same code, Your Honor. It's a labeling game of calling it UNIX and saying it wasn't GPL because it doesn't say SCO UNIX. It's there. It has the label of the GPL on it. It's therefore distributed under the terms of it. That's the way they got it from IBM, Your Honor, when they got the 700,000 lines of code from IBM. And it was only on the terms and

32

conditions of the IBM GPL license, the designation that we put there on our code, that they had permission to use it.

Now, as to that code, they have undertaken the position to assert rights as to that, that other people can't copy it, they can't distribute it, they can't use it, though it's our code with our copyright notice issued duly from the Copyright Office because they say it infringes. And that, Your Honor, respectfully, turns the GPL on it's head. And if it were read and construed in that way, it would be effectively a meaningless document.

Now, with respect to unclean hands, the argument has been made here, as it's been made before, that IBM ought not be allowed to enforce its rights in these 700,000 lines of code because somebody at IBM got on to a web site that SCO put out there on which they were able to access the code. There was absolutely no impermissible, improper hacking. And despite the use of that term, there is not a whit of evidence to suggest that anything was done improper in the taking of that code.

And, in any event, Your Honor, this motion isn't based on the taking of the code down from the web site. SCO has admitted in its interrogatories, it has admitted, Your Honor, in its responses to our requests

33

for admissions, and it's admitted in response to IBM's motion that the code at issue was in fact code they copied.

So the unclean hands argument presumably is an argument that says IBM shouldn't be allowed to argue that we copied it because they figured out that we copied it by improperly hacking into our web site. There was no hacking and, in any event, there is no dispute they copied it. You didn't hear any argument that SCO didn't actually copy these 770-plus thousand lines of code. What you heard is that it is not a problem because we didn't actually reach the GPL because we didn't label it UNIX, Your Honor, rather than Linux. That's a labeling game, and it has no basis in the construction of an agreement like the GPL, which is about promoting free software.

The suggestion has been made that SCO never modified in any way the kernel, and that simply is not true. It isn't supported by the evidence that is set out in our papers. They took the kernel, Judge, of Linux, and they created, based on that kernel, their own product, SCO Linux 4, which was a modification and a derivative work, something new from the kernel. Now suddenly the concept of derivative work takes on a new meaning. When we are not arguing about IBM's contract

34

motion, derivative work is suddenly a very narrow construct, and it's very hard to become a derivative work.

By SCO's own definition and by the definition under the law, Your Honor, there is no question that what they distributed as SCO Linux 4 was a modification subjecting them to the terms of Section 2-B of the GPL, which, contrary to what Mr. Normand suggests, makes perfectly clear that you can't charge fees of the kind they charged for this software.

And, moreover, Judge, the Sections of the GPL that are at issue, the ones I cited, Section 6, Section 1, those don't turn on having a particular modification. Those turn on not giving people who gave you code under the GPL, not giving others who received the code under the GPL, the rights to which they are guaranteed under the GPL.

SCO was a licensor and SCO was a licensee and had obligations if it was going to continue to distribute this code. No one made them distribute our code. No one made them sell it for a profit. They undertook that on their own, and when they did, they undertook it subject to the terms and the conditions of the general public license.

Finally, Your Honor, with respect to

35

repudiation. Mr. Normand says that all Mr. McBride did in his deposition was express some misgivings about the GPL and that surely that can't amount to repudiation. Well, Your Honor, they did a little bit more than express misgivings. They asserted in a pleading signed and submitted under Rule 11 defenses that asserted and provided interrogatory answers which at least were supposed to have been sworn. I don't know, as I stand here, whether they were sworn, that said that the GPL is unenforceable, that it's void, that it's voidable, that it violates the antitrust laws, that it violates the export control laws. How could you be more clear, Your Honor? It's unconstitutional. And they then said preempted by federal copyright law and unenforceable under state law.

If that isn't a repudiation, I don't know what it is. Thank you, Your Honor.

THE COURT: Thank you.

Mr. Normand, anything else?

MR. NORMAND: I'll be brief, Your Honor, because I understand I went over. First, as we cite in our briefs, our discovery responses made clear that we distributed the IBM copyrighted material properly, subject to and pursuant to and consistent with the GPL. There is no question that we didn't repudiate it in our

36

discovery responses, if read fairly.

Second, Your Honor, the factual record shows we made no modifications to SCO Linux 4.0. The record makes that clear. Principally, I would like to spend a minute, Your Honor, on counsel's failure to respond to Section zero. Counsel has no answer for Section zero. Counsel describes this as a name game but has no answer as to why Section zero would be the very first substantive section of the GPL, and he doesn't dispute the clear language of the section.

The labeling game is that to say we, under the GPL, supposedly made our material subject to the terms of the GPL. We did not. We distributed a verbatim copy of the GPL without making the required decision to put in a copyright authorization notice for either UNIX or Linux any material and subject that material and surrender rights to that material pursuant to the terms of the GPL.

Mr. Marriott's only argument was to say it's an absurd argument because we are supposedly arguing that every single file or subset of files would have to have the authorization notice. That's not the point. There is no authorization notice at all in any file. Thank you, Your Honor.

THE COURT: Thank you, Mr. Normand.

Let's move to SCO's motion for summary judgment

37

on IBM's second, third, fourth and fifth Counterclaims.

Mr. Hatch. You told me earlier today that you have lost your voice. That would certainly be a first.

MR. HATCH: This would be the first I've spoken in a day and a half, and my partners have enjoyed it immensely.

Your Honor, in its Counterclaims, IBM has painted S-C-O, SCO, as a company whose sole business plan is pursuing sham litigation to create issues about IBM and its Linux business. At tab 1 we produce a quote from their brief which says: With no other prospects, SCO shifted its business model again, this time undertaking to create fear, uncertainty and doubt in the marketplace from which it hopes to profit in regards to IBM's products and services.

Simply not true. The facts show that SCO was simply trying to run its business and to produce products and to protect its intellectual property rights. Now, it was discussed, I think, Monday by Mr. Normand and by Mr. James that by 2002 SCO was facing extreme downturn in its business due to the actions of IBM pulling out of Project Monterey, and in breach of its contracts with SCO, donated SCO's intellectual property rights to Linux.

Now, in the face of this business downturn, Mr. McBride was hired to rejuvenate SCO because he had

38

had experience in selling products in a business model similar to SCO's, not because he had litigation experience. Now, John Terpstra -- and it's interesting Mr. Terpstra was the so-called Linux guru at Caldera. Many called him a Linux evangelist. And he testified at tab 2 that when Mr. McBride came to the company, he wasn't looking for litigation, he said he was asked to identify what technologies that we had that we could build profitable business products and services upon.

That's very different than the picture IBM paints. New business. New products. Not litigation. In fact, Mr. Terpstra, this Linux evangelist, further told Mr. McBride, and this is actually a fairly ironic story, Your Honor, that several companies had actually wanted to use SCO's UNIX-based technologies to run applications in Linux and that could be a very profitable business model for SCO.

And at tab 3 he indicated that, I briefed Mr. McBride to the effect that a number of customers had asked us for licensing to legitimize using UNIX libraries with Linux programs, and that we had a revenue opportunity in doing that.

Now, based on that opportunity, SCO put together a licensing program for its libraries and approached Hewlitt Packard and other large companies to

39

confirm whether it would be a good idea and they would find support in the industry. Not surprisingly, the program was well received and would give an opportunity to protect SCO's intellectual property and give them a profitable business while enhancing Linux by making more UNIX applications available because, without applications, an operating system isn't really worth much. Everyone wins in such a situation. There was only one company that balked. IBM.

In fact, you will recall yesterday Mr. James referencing a meeting where IBM and their lawyers used four-letter words and went ballistic about the fact that SCO was going to do that. We reproduced that at tab 4. We also heard during the course of the argument on Monday, as well, the meetings Mr. McBride had with Karen Smith in the aftermath of that where she threatened to shut down SCO's businesses, and those are at tab 4 of the SCO summary judgment hearing binder on IBM's motion to dismiss on the interference claims.

Now, we now know that SCO -- that IBM did not want SCO letting others to know that you had -- to believe that you had to have a license to use Linux products. IBM's plan was to start putting these products into Linux. SCO was attempting to work with the Linux community on these issues while IBM was just looking out

40

for IBM.

Now, that gives us a context and brings us now to the public statements that IBM has based their claims on. First I would like to point out that it is very clear from the record, I think, that IBM's own people don't believe these claims. They don't believe that IBM is damaged by them or that they are legitimate. In fact, I always like to see what people say to the marketplace, and I think that's a good source instead of what their lawyers are telling them to say. At tab 5, there's a quote here, which is Jim Stallings. He was the general manager of Linux for IBM, so he was arguably one of, if not the most important people knowledgeable on the Linux business that IBM said.

At tab 5 he said -- this is in April of 2003, so this is about a month and a half after the initial lawsuit was filed. He said: We haven't seen any evidence that customers are concerned about making a decision against IBM or counter to any proposal that they have on the table from IBM. We talk to our customers on a regular basis. They know what's in the newspaper. I can't comment on the lawsuit itself, but the market certainly hasn't slowed down, from what I can see, because of the lawsuit. From a sales standpoint, we certainly haven't seen anything. As a matter of fact,

41

customers, the ones that are reading and listening and understand it, they are more enthused by the opportunity. We haven't seen anything in the marketplace that's negative.

So IBM, through their own words, are saying that they are not affected, and there is no basis for these claims.

At tab 6, Mr. Stallings again says the same thing, and it's interesting to note that this is now June of 2004, a year and a half after the Complaint. And he says: Despite SCO's suit and transmission of demand letters -- now, this transmission of demand letters is an issue in this -- despite those, business is actually accelerating, and Linux is the fastest growing operating system. It doesn't sound like they are being damaged.

Now, this is from Mr. Stuart Cohen. Now, he is not at IBM, but he is the chief executive of Open Systems Development Laboratory, which is funded by IBM and works in close proximity with them. What did he say? SCO's lawsuit has accelerated the use of Linux and is one of the best things that has ever happened to the operating system.

Now, the date on that is March, 2005. So we now are two years past the Complaint, and, interestingly, all that is well past the date of all of the comments and

42

the public statements that are at issue in these four counterclaims of IBM; in other words, nearly a year past the last -- reasonably the last of those, there is -- if anything, IBM's business is growing. And we saw a chart, I think during Mr. James' argument on Monday, that would indicate that.

These are things that these people said and these IBM employees and managers said publicly and not before the Court. Now, let's look at the actual public statements. Now, all the public statements complained of are not actionable because of -- for three reasons. They are either absolutely privileged, they are subject to a qualified privilege or, in most cases, they are simply true.

Under absolute privilege, a party to a private litigation is absolutely privileged to publish even defamatory matter concerning another and communications preliminary to a proposed judicial proceeding or in the institution of or during the course and as part of a judicial proceeding in which he participates if the matter has some relation to the proceedings. And that's the Price vs. Armor case in Utah, but I think it's consistent with the law in almost any jurisdiction.

That absolute privilege applies to pleadings, statements in court, demand letters, settlement letters,

43

cease and desist letters and letters threatening litigation

So at tab 10 is some of the allegations from paragraphs 57, 58 and 79. These are all allegations in the Second Amended Counterclaim of statements that apparently have somehow damaged IBM. The first one is from the March 7 Complaint. The second one is -- in 2003. The second one is from the Amended Complaint and the third one is from a copyright infringement suit against another Linux user, so another lawsuit that is not involved here directly -- actually it's probably from at least two other SCO lawsuits.

These are classic, absolute privileged comments. They are in pleadings, and IBM has not given the Court anything that would work as an exception from that rule.

Now, at 10 -- at 11, one of the statements that IBM has complained about is that Mr. McBride had stated that IBM had taken our valuable trade secrets. And they go at length in their brief to say that there are no trade secrets at issue in this lawsuit. The problem with that is there were trade secrets that were at issue in the lawsuit when the Complaint was initially filed. As a matter of fact, in the original Complaint at paragraph 108, it says that IBM, through improper means, acquired

44

and misappropriated SCO's trade secrets for its own use and benefit for use in competition with SCO in an effort to destroy SCO.

So Mr. McBride is, and from this particular quote --

THE COURT: You argue that there is an absolute privilege because he is summarizing a pleading?

MR. HATCH: Summarizing a pleading. And as a matter of fact, look at the date on this, Your Honor. It's the day of the Complaint. So this isn't like several years later, as the IBM brief makes it appear, that Mr. McBride was saying this at some point when this was not in the lawsuit. So, again, we have another statement that is absolutely privileged.

Now we go to tab 12. And here counsel for SCO -- and I have it on good authority that it was not me -- asserted at a hearing on December 5 that SCO will file a copyright infringement action as against IBM. Again, this was a statement made in open court and is subject to the absolute privilege.

Your Honor, I think there is absolute -- if there is anything in this particular motion, it is very clear that we are entitled to partial summary judgment, at the very least, on all of the statements that are subject to these absolute privileges. There is no action

45

there. But we believe, of course, that it doesn't end there.

The qualified privilege, of course, applies to statements to the public by parties, counsel or witnesses that provide a basic description or summary of the allegations of the Complaint. Buckley vs. Fitzsimmons is a U.S. Supreme Court case that states that. Muirhead vs. Zucker and others, we have cited in our brief. Advising the media that a lawsuit has been filed and basic descriptions of the allegations are allowed, as well as are statements that are made that are believed to be true and made good faith.

Now, I'd like to go through -- and I've tried my best, Your Honor, to put each of these statements in some type of a subcategory, and so I've made a slide on each one of these that are roughly equivalent so we don't have to go through each statement because we don't have time for that.

At tab 16, the quotations that IBM quibbles with all go to what has become somewhat infamous in this case, the hundreds of lines of code, numbers of lines of code, very extensive, gargantuan and similar statements that are alleged in -- that are set forth in IBM's opposition brief. There are a couple of things here, Your Honor, based upon as you know the arguments and

46

particularly the contract summary judgment arguments that you heard on Thursday is there is -- there is a big dispute between the parties as to what is -- what IBM is contractually obligated to and whether or not the AIX, and essentially Linux, are derivative works that IBM has breached its contract with regards to.

Now for IBM, though, to say that we have somehow -- we have somehow made an actionable statement by reciting what is truthfully in the lawsuit by saying that our belief that there are large, gargantuan amounts of code in these derivative works and to try and obfuscate the issue by saying there is only 326, that is only pointing out that there is a huge dispute, factual dispute between the parties on the contract claims and the copyright claims, but it certainly doesn't make it actionable because SCO would have a good faith basis for making those.

Secondly. SCO has submitted the expert reports of Mr. Rochkind and Mr. Ivey. Both of those individuals show that those large numbers of code have been taken in the derivative works and providing yet again another good-faith basis. And even to the extent that Judge Wells has excluded some of that testimony, that doesn't change the fact that those establish a good-faith basis for making those claims at the time because they

47

summarize -- as they summarize what's in the lawsuit.

So, not only do we believe they are true, but there is a good-faith basis for those.

On the 17th tab it says -- they give a quote and they have complained several times in their brief that we have said that IBM doesn't own the AIX code. Well, a couple of things here. One is that it's an incomplete statement of what Mr. McBride actually said. He actually said that SCO owned the source code in UNIX and the right to that system, implying that it refers to a control disclosure.

And, Your Honor, I would like to pass up this as Exhibit 601. This is, again, marked 6. This is quoting, and you will see I've underlined so we can go through it quickly. What Mr. McBride clearly -- it's very clear here that he isn't saying what IBM says in its brief, but he is talking about the ability to control, under SCO's contract with IBM, the derivative code.

It says: SCO is in the enviable position of owning the UNIX operating system.

So the bottom line -- so that's the operable words here is that SCO owned the source code in UNIX and has a right to that operating system.

A little lower it says: The UNIX contracts held by SCO were extremely powerful, and one of the

48

remedies under the contract is that we have the ability to revoke the AIX license.

That comes up in the comment they complain about later. The interesting thing to me here, and I've underlined the part about SUN Microsystems because they apparently were contacted by the same reporter and immediately understood that what was being talked about was the right to the derivative of that right. And SUN Microsystems, whose Solaris operating system is based on UNIX, quickly -- moved quickly Thursday to assure its customers that its licenses were all in order. Now, SUN is someone that has a license with SCO and so they understood that their derivative works required a license.

And as a part of a series of licensing agreements, SUN acquired rights to make and ship derivative products based on the intellectual property in UNIX. This forms the foundation for the Solaris operating system that ships today.

So, IBM wants to take partial quotes and take them out of context when the reality is people in the industry, based on the article that they are citing, understood 100 percent that what they are talking about is the rights in the derivative work.

At 18, and we've already alluded to this a

49

little bit, Mr. McBride has said -- is purported to have said: We do have pretty strong ownership of the AIX code base.

Again, it talks about the base, where it was derived from. That's a big issue in this lawsuit. Certainly if it stays an issue in this lawsuit, it can certainly not be bad faith to have made that type of a statement.

Nineteen. Same type of thing, talking about entire programs, because we are talking about derivative actions.

Twenty. We are the source of AIX and all it comes from. Again, this is another quote a little bit out of context. And, again, here it talks about the full quote in the case. AIX, HP, UX, Solaris, these are all companies that have licenses with us for their derivative works.

Your Honor, these are not actionable because they are true. People understood them, and at the very least, they clearly are issues that are -- everybody is debating and having thrown out of this suit and thus are a basis for a good faith statement for making those.

Now a lot of these -- almost all of these things will ultimately turn in the lawsuit on whether AIX is a derivative of UNIX, yet IBM claims that -- as I've

50

heard over the last three days, IBM claims its only obligation under the AT&T source license was to keep the UNIX source code confidential. That's it, that they could do whatever they wanted with AIX and Dynix, even though they were derivative works.

I'd like to show the Court another exhibit. Now, Mr. Singer touched briefly on this document on Monday, I believe. This is an official IBM report from 1989. Notably, this is four years after IBM claims to have been assured that their AIX did not need to remain confidential. Now, May Cherry, the person writing the report, is an IBM employee in software contracts and licensing procurement groups.

MS. SORENSON: I'm sorry, Your Honor, the document is marked confidential.

THE COURT: If it's confidential then people who can't see confidential documents have to leave. It is marked confidential, so if you're not cleared for confidential documents, you will have to leave the courtroom.

MR. HATCH: IBM has the ability to waive it at this point, if they want.

THE COURT: Apparently they are not, or they would not have raised it. That would be my guess.

MS. SORENSON: If you can argue the point

51

without quoting from the document, I think that's how we proceeded in the past. If you can't, then I think we would be forced to empty the courtroom for those that don't have confidential clearance.

MR. HATCH: I don't think any of the highlighted parts would be giving up any confidential information.

MS. SORENSON: That's an analysis on your part. I guess we don't agree. You can either quote from the document or not quote from the document. If you can proceed by not quoting, then --

MR. HATCH: I can point -- I can certainly point to those parts.

THE COURT: I can read it.

MR. HATCH: You can read it, yeah.

THE COURT: I can see what you've got highlighted.

THE HATCH: All right.

THE COURT: Stay in.

MR. HATCH: As you can see, the document is talking about derivative works, and it's talking about AIX essentially being a derivative work. They understood this. Their contract people understood this. This is an internal document that they are giving to their people. They are telling them how they have to treat it, and it's

52

exactly how we have said, in response to the contract motion in particular, that they had to treat it. This is their people essentially writing a brief for SCO, and they are saying because we make some of these statements, we are in bad faith.

THE COURT: What else do you want to tell me? You've used your time. I assume you want a little for reply?

MR. HATCH: Yeah. If you could give me just a second, your Honor. Just, I wanted to point out on this third tab it indicates who it is held in confidence for. And it doesn't say for IBM. It says they understand who it was for, and it says AT&T, in that instance, the predecessor of SCO. It's interesting to me, and I'll just -- if I give you one last thing. This is why there are factual disputes in the other motions. This Exhibit 134, Your Honor.

THE COURT: The one we were just talking about?

MR. HATCH: 134 is the one we were just talking about, that they claim was confidential, and that's fine. Because it supports our position wholly, they declare it irrelevant and hearsay. That's why they claim there are no factual disputes in the other motions, and that's why they want to be able to say here that factual, true, honestly-held opinions that have been said to the press

53

are now somehow actionable under their four causes of action, and it simply doesn't provide a basis for that, especially if SCO believes only the same thing that IBM's employees believe about these.

Nothing could be more clear. Therefore, Your Honor, SCO's motion -- I'll leave the arguments to the damages, which are fairly clear in our brief. I would only point that at tabs 22 and 23, you will see that their damages expert is giving kind of a general, you know, repeat of an employee of IBM, just general things. He basically just copies what he says without making any analysis whatsoever of the actual damages -- that there are actual damages in the case.

Their motion should fail for that reason as well. So, based on all the privileges, the lack of damages and the good faith made of these, Your Honor, we think the motion should be granted as to all four causes of action.

THE COURT: Thank you, Mr. Hatch.

Ms. Sorenson.

MS. SORENSON: May I approach? We also have a brief binder.

THE COURT: Yes. Thank you.

MS. SORENSON: Good afternoon, Your Honor. In spite of what SCO's briefs and arguments suggest, IBM's

54

Counterclaims here do not concern solely statements that are pertaining to or arising out of or made within the course of a judicial proceeding. Far from it. As you can see at tabs 1, 2, 3, 4 and 5 of our binder, IBM's actual Counterclaims are for violation of the Lanham Act, for disparagement of IBM's products and services, for common law unfair competition, for misappropriation of AIX and Dynix and Linux-related products and services, for intentional interference as a result of SCO's misrepresentations to IBM's prospective customers and for unfair and deceptive trade practices under New York's statutory law.

These Counterclaims in fact concern SCO's campaign to create fear, uncertainty and doubt about IBM's products and services, and during which campaign SCO has disparaged IBM, has made false and misleading statements to the press and to IBM's customers and abused and misused its alleged intellectual property rights.

In its briefs, SCO makes no arguments as to any of the elements of these Counterclaims. Instead, they make only three arguments, which are summarized at tab 6. They are as follows: SCO claims, as Mr. Hatch variously argued today, that it is shielded from liability for its misconduct under the judicial proceedings privilege, or under a qualified version of that privilege, to summarize

55

the contents of your pleadings to the media. SCO further argues that good faith is an affirmative defense to IBM's Counterclaims and that SCO has undisputedly established such good faith for purposes of summary judgment, and SCO claims that IBM has failed to offer any evidence of damages flowing from these claims. Each of these arguments fails.

First. SCO's entire argument depends on an improper and unduly narrow construction of IBM's Counterclaims. First SCO implies that IBM's Counterclaims are barred by the litigation privilege because they only or primarily concern SCO's conduct or statements during or in the course of these proceedings. That is false. In fact, IBM's Counterclaims are supported by extensive evidence of SCO's non-litigation misconduct, including statements made by their officers and directors in press releases, statements made to analysts, interviews with print and internet journalists, and at live presentations at SCO forum events.

This evidence -- some of it anyway -- is set out in detail at pages 6 through 13 of IBM's opposition brief, and a few of these examples are excerpted at tab 9. This evidence, taken as a whole, shows that SCO engaged in nothing short of a PR campaign against IBM's products and services over a period of years that was

56

waged through misstatements to the media and to the public and which took place literally from coast to coast.

The fact that any one -- the fact that there are certain allegations regarding SCO's allegations in this case, complaints that they have filed, that is not a basis, a primary basis for liability on these claims. Really, that is a predicate, a way of showing how SCO's misstatements about its claims about IBM's products and services have no basis. It's a way of showing how their misrepresentations to the press have not matched up and will never match up with what is being brought out in the litigation.

This campaign of public statements continues to the present day. It is not limited to press releases or interviews given in March of 2003, as Mr. Hatch argued. And this is shown in a March 1, 2007 article which is at tab 11 of the binder in which Mr. McBride characterizes IBM as nothing less than the enemy and suggests that IBM's strategy is to ensure that this case is not decided on the merits.

The result of all of the foregoing evidence, which is set forth in great detail in IBM's opposition brief and papers is that the litigation privilege, which only protects statements such as the arguments of counsel

57

or the testimony of witnesses made during or in the course of a judicial proceeding, so long as they sufficiently relate to that proceeding, simply does not apply to IBM's Counterclaims as a whole, which allege misrepresentations and disparagement of IBM's products and services.

SCO's inaccurate construction of the Counterclaims at issue here as concerning only SCO's conduct and statements in just this litigation creates another fatal problem for its motion. It's SCO and not IBM that has the burden of demonstrating that there is no disputed material fact in existence here regarding the affirmative defense of privilege. And the authorities for that are set forth at Tab 12.

Rather than coming forward in their motion and in their brief with evidence that no disputed material fact exists regarding application of a privilege to all of the misconduct at issue, the misstatements at issue and IBM's Counterclaims, SCO relies on only four brief assertions in its statement of undisputed facts that simply state that SCO had a good-faith basis for bringing certain of its claims in this litigation.

Those four assertions in its brief don't even approach establishing the absence of a genuine issue of material fact that a privilege protects all of the

58

misconduct at issue. I believe this Court has used the phrase, "the plethora of public statements SCO has made in this case" in connection with our previous declaratory judgment motion. So the four assertions do not even approach establishing the absence of a genuine issue of material fact the privilege protects all of the plethora of public statements disparaging IBM's products and services which are at issue in IBM's Counterclaims.

Mr. Hatch brought up, in his argument, the fact that Mr. McBride made statements about the fact that IBM has not been a respecter of SCO's UNIX System V trade secrets and that IBM has violated its trade secrets by dumping UNIX System V into Linux. In fact, you know, they only concede that UNIX System V later has no trade secrets and contains no trade secrets in it.

SCO's statement of undisputed facts doesn't address this aspect of IBM's Counterclaims at all, and it doesn't establish that a privilege applied to such conduct. Instead, SCO wants to argue that because they cannot show that every single misstatement has to do with a statement by a witness or an attorney in the course of this particular litigation, they instead argue that a qualified privilege to make statements to the press regarding the allegations in one's pleadings should apply.

59

But SCO's cases concerning this qualified privilege are summarized at tab 13 of our binder, and they make the point nicely that no Court has anticipated, much less held, that a qualified privilege to summarize one's claim to the media should be extended to provide immunity to a litigant that embarks on a nationwide negative publicity campaign about a competitor and its products and services, which has been waged for years, and which continues to the present day.

Instead, each of SCO's cases concern a single press release or a single press conference, and as a result, they offer SCO no support in its attempts to avoid liability for the wide-ranging misconduct that is at issue here.

For example, IBM Exhibit 147 is a letter written by Mr. McBride to Congress claiming that Open Source software and Linux are nothing short of a threat to our nation's economy and to our nation's national security. These are the kinds of statements that have nothing to do with the litigation. They far exceed any scope of any qualified privilege because they are in no way a mere summary, at one point in time, of its allegations that are at issue in the case.

In addition, IBM has provided numerous cases -- this is at tab 15 of our binder -- holding that this kind

60

of conduct, where a litigant actively stimulates press coverage and wide publicity of a Complaint is beyond the pale of judicial privilege protection. Far from distinguishing them, SCO ignores them in its reply, and it has ignored them today.

This is, in effect, an argument about excessive publication, and if there has ever been a case for excessive publication, I would submit it is here, where SCO has made numerous, repeated statements to the media about IBM's products and services and its rights to own, control and profit from AIX and Dynix and sell and profit from Linux-related services.

As this Court has stated in a previous decision in the SCO vs. Novell case, the issue of whether there has been excessive publication is a question of fact and therefore not appropriate for resolution on summary judgment. Perhaps the exhibit that makes the point best about the volume of SCO's public statements here and the extent of its excessive publication, over publication and the fact that it has exceeded any qualified privilege and certainly any litigation privilege, is made at tab 14.

This excerpt is a Fortune Magazine article from May 17, 2004, only just one year into the litigation, and that article opens with the observation that in the ascetic waiting room of the SCO Group's Lindon, Utah

61

headquarters, the only reading matter is a stack of beige telephone-book sized binders. They are volumes 1, 2, 3 and 4 of the company's press clippings for the previous month.

That's IBM Exhibit 373. SCO also claims good faith, that it's a substantive defense to each of the causes of action pleaded by IBM, and it further claims that IBM has no evidence that SCO acted in bad faith. SCO's good faith arguments fail for three reasons as well. These are summarized at tab 16. In the interest of time, I will address only the first and third of these.

First, again, as SCO concedes, good faith is an element of the affirmative defense of the qualified privilege it relies on. And SCO bears the burden of proof on this issue. SCO must and has not shown that no disputed material fact exists, that the affirmative defense of privilege protects all of the misconduct at issue in IBM's Counterclaims. SCO has not even approached doing so here for the reasons I summarized earlier.

As one example, as argued by Mr. Marriott earlier today in the first motion, SCO, despite its numerous public statements that are summarized I think at tab 2 of our declaratory judgment binder, regarding

62

gargantuan amounts of code, truckloads of code, limitless amounts of code and the tips of the iceberg, has in fact identified 326 lines of allegedly infringing UNIX System V code ultimately.

No SCO expert, which is what they rely on in their brief, can justify SCO's public statements about mountains of code in the face of this evidence.

Third. Even if SCO had introduced something other than conclusory assertions of good faith based in large part of the opinions of their experts, IBM has not merely raised an inference of SCO's bad faith in response here. There is, instead, in the record in our summary judgment brief and papers overwhelming evidence of bad faith to the extent that SCO knew its statements were false and its conduct unjustified.

This is laid out in detail in IBM's brief at page 26 and at paragraphs 10 through 28 of IBM's statement of facts. And that evidence alone is more than sufficient to dispense with SCO's arguments regarding good faith on summary judgment. Because of the volume of that evidence and in light of the time, I would simply state here that SCO cannot show good faith here and, in fact, SCO's bad faith in making the statements it has made regarding its alleged rights to own and control AIX, Dynix and parts of Linux, and regarding IBM's products

63

and services, is amply demonstrated by the fact the statements SCO makes, the extrajudicial statements SCO makes about owning AIX and ownership of the code base represent a radical about face from the conduct and statements of its predecessors who repeatedly and over the course of decades confirmed IBM's and other licensees' rights to profit from, own, control and disclose as they wished code they wrote themselves, such as AIX and Dynix.

And I would refer the Court to the library cart of UNIX System V internals books that were here in the courtroom last week disclosing those internals and discussing them as well as for licensees' sublicensed products.

And as this Court has correctly observed, and as I think Mr. Hatch's arguments explaining away Mr. McBride's statements to the press and explaining away his statements about owning the AIX code base, I think that those make the point amply that state-of-mind evidence does not lend itself to resolution on summary judgment. And for this reason, as well, this motion should be denied.

Finally. SCO makes a variety of arguments that IBM has sustained no damages and that it has adduced no evidence of damages in support of these Counterclaims.

64

What Mr. Hatch's arguments amount to, really, are a quibbling with the strength of IBM's damage evidence. And that, of course, again, is not a basis for summary judgment. The fact of the matter is IBM has a minimal burden to produce some evidence of damages, and IBM has more than done so, including by setting forth in its expert reports of Professor Kearl and its rebuttal report of Professor Kearl, the testimony of William Sandby that if the SCO actions have reduced the marketplace adoption of Linux by 20 percent, as suggested by Darl McBride in an interview; again, a public statement that SCO would like to now disavow, then the proportional impact on IBM would be $607.2 million in lost profits.

And Mr. Hatch, in his brief, made a few points about certain IBM witnesses' statements to the press at various points in time, most of them fairly early, about the fact that they weren't seeing much impact on Linux adoption as a result of the lawsuit. However, we have introduced evidence through our expert, relying on the testimony of SCO's own expert witnesses, their methodology, and SCO's own CEO about uptake -- about impact of Linux uptake as a result of SCO's claims. And the number is nothing short of, then, $607 million.

There is no argument here at all that IBM has not adduced sufficient evidence of damages on either

65

Linux, in the context of Linux products and services and sales or in the context of AIX. And that information and that evidence is summarized in tab 24 of our binder.

In addition, at tab 25 of the binder, IBM has proffered competent evidence through its expert and fact witnesses that it has suffered in excess of $40 million in attorney's fees and lost employee time in responding to SCO's statements and SCO's suit.

SCO makes some arguments in its brief that, again, not to quibbling with how this is broken out, and in fact that is not the obligation here at this point on summary judgment. As Mr. Singer argued earlier today, if there's an issue to be raised there at all, it's better raised, I think, in a motion in limine or in a special verdict form or jury instruction.

If the Court has no questions, I will rest.

THE COURT: Thank you. I don't have any.

MS. SORENSON: Thank you.

THE COURT: Thank you, Ms. Sorenson.

Mr. Hatch.

MR. HATCH: Your Honor, the only thing that was raised was tab 11, and I just would indicate this is just more of the same. The part that was underlined in mine that she's saying was so horrible that Mr. McBride said, it says: There is no doubt that our enemy -- that would

66

be IBM, if anybody is not sure of that -- is well-funded and has deep pockets.

Now, anybody would be really hardpressed to stand here before me and tell me that that is not true, especially given our wonderful audience in the courtroom.

THE COURT: Some of the audience is yours, isn't it?

MR. HATCH: Yes, indeed, Your Honor. And I'm happy to have them say in public that we are well-funded as well. The point is: These are not actionable. They know it. These are brought to sidetrack the litigation, and the motion should be granted.

THE COURT: All right. Thank you. Thank you all. I'll take these motions under advisement and get a ruling out when I can. It's been so good to spend these three afternoons with you all.

MR. MARRIOTT: Thank you, Your Honor.

(Whereupon the proceedings were concluded.)

REPORTER'S CERTIFICATE

STATE OF UTAH )

67

) ss.
COUNTY OF SALT LAKE)

I, REBECCA JANKE, do hereby certify that I am a Certified Court Reporter for the State of Utah;

That as such Reporter I attended the hearing of the foregoing matter on March 7, 2007, and thereat reported in Stenotype all of the testimony and proceedings had, and caused said notes to be transcribed into typewriting, and the foregoing pages numbered 65 through 131 constitute a full, true and correct record of the proceedings transcribed.

That I am not of kin to any of the parties and have no interest in the outcome of the matter;

And hereby set my hand and seal this 8th day of March, 2007.

_______________________________
REBECCA JANKE, CSR, RPR, RMR

68


  View Printable Version


Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details )