Steve Martin did the HTML for us, removing the line numbers for the sake of our readers who are blind and depend on screen readers. I am so grateful to them both for doing the heavy lifting for me.

In the first motion, we finally find out what "negative know how" means on Planet SCO. The second motion is the one about SCO's copyright infringement claim regarding IBM continuing to distribute AIX after SCO purported to terminate IBM's irrevocable and perpetual license for allegedly violating SCO's trade secrets, which SCO later admitted in open court Unix System V doesn't have.

Yes, it's all profoundly silly, but it's very interesting, and if I had to characterize the dominant theme of both motions, it would be JFS. IBM informs the court that JFS isn't from Unix. It doesn't come from AIX even. What IBM donated to Linux came from OS/2. So even if SCO had a trade secret or contract leg to stand on, it has no control or rights with regard to JFS.

IBM opens by explaining to the court what SCO's theory of the contract is, namely that if IBM wrote its own code, if it was ever put with Unix System V code in a distribution, IBM thereby lost control of it and SCO got to swallow it up, like some bottom feeding sea creature. Further, SCO got to gobble up IBM's modifications and derivative works too, even if IBM spent millions on development. And they call the GPL "viral"! SCO, by its theory, would control pretty much all the code in the known universe, no matter who wrote it, and most of the programmers in the world too. Who'd ever sign such a deal? Not IBM, for sure, as David Marriott demonstrates by playing clips from various depositions for the court.

SCO's Stuart Singer seems confused on some dates and some tech. He says this, in his opening:

The UNIX operating system developed by AT&T was its crown jewel. It is the operating system of choice for mission critical application, and it became in the 1990s the leading operating system worldwide. Now, major computer companies like IBM wanted to adopt that system for their own hardware. They had a choice. They could come up with their own operating system. In fact, IBM tried. It is called OS2. It failed. IBM, like others, struck a deal with AT&T. Give us access to your source code so that we can develop our own flavor of UNIX that will run better on our hardware.

But from the Comes v. Microsoft exhibits, we've learned that OS/2 certainly had not failed by 1985, the date of the IBM/AT&T agreement as it was still being developed and strategized over and released in 1990, so Mr. Singer has it backwards. So often, SCO's side makes these kinds of mistakes. I've often wondered if that is part of why they present absurd theories. He says this about derivative works:

What they could not do is exactly what IBM has done here, give away that head start by open sourcing of derivative technology that they would never have had if it weren't for access in the first place to System V, and to allow a jump-start to Linux, which has had a tremendous effect on SCO's business.

IBM didn't need Unix System V to write JFS. Do you need to have Unix System V to write RCU? Remember that IBM patented RCU. Marriott already told the court that *nothing* at issue here is from Unix System V:

Now, we'll come back to those technologies in detail, but let me make a few things absolutely clear about those technologies from the outset. None of that material, Your Honor, is UNIX System V source code, methods or concepts. None of that material is a modification or a derivative work of Unix System V. None of it was contributed to Linux by reference to UNIX System V, which this is original IBM work created independent of UNIX System V.

IBM's David Marriott explains the negative know-how business:

Lastly, Your Honor, the negative know-how category. This is a category of 11 items of supposedly misused information. We lay this out at tab 64. There are two basic brands of claim as it relate to these 11 items. One of them concerns experience. Here is what SCO's claim is as it relates to the experience. It says, as we show at tab 65, Judge, it says IBM has breached its contract by permitting IBM developers exposed to DYNIX PTX methods and concepts to contribute to Linux in the same area for each developer's work. So anybody that had any exposure to DYNIX can't work in connection with any other operating system in that area. That is the claim.

With respect to the negative know-how, which also appears at 65, the gist of the claim is that people who had some knowledge about DYNIX told people who were working on Linux not to do certain things in DYNIX that they don't think worked. SCO has identified 16 supposed wrongdoers as it relates to these 11 categories....

And for that, SCO wants billions in damages? First, Marriott points out, all those folks have provided declarations:

Here is what they said. They didn't make any contributions to the files or the directories listed, or they didn't base their contributions to the list of files on UNIX System V in making the contributions.

And second, the contracts don't speak about negative know how at all:

First, Your Honor, the agreements which SCO contends were breached, Your Honor, do not contain any provision which would allow it to preclude IBM employees from using their experience and their general know-how in working on projects other than the one on which they are presently working. There is no connection drawn in SCO's interrogatory answers or in its paper between the contract and this claimed notion of misuse of experience....

For two of these items, and, again, this is undisputed, for two of these items, Your Honor, 188 and 187, the technology which SCO alleges was misused didn't even exist in DYNIX. For another two of the items, 43 and 94, the material had nothing to do with DYNIX PTX. It was discoverable outside of IBM. For five of the items there is absolutely no evidence that the alleged wrongdoers had any experience in the technology area where they were supposed to have misused it in some other area. SCO admits that with respect to four of them. With respect to items 23 and 90, Your Honor, for two of the items the material was disclosed in marketing materials, and in footnote ten of SCO's opposition papers, it says that any such disclosure -- if the material is inactionable anyway. Finally, item 189, Your Honor, was based on knowledge available in public literature.

In fact, he continues, SCO already admitted that there are no trade secrets, and it dropped that claim, so where do they get off now trying to "protect" any supposed secret knowledge, whether negative or not? Not to mention that Magistrate Judge Brooke Wells told SCO to provide specificity, and to date that hasn't happened. For that reason alone, IBM says it deserves to prevail. Finally, SCO's theory reaches an absurd result, in that if their theory is accepted, programmers ever exposed to Unix System V would be pretty much unemployable in their field of skill and expertise from that day forward:

If SCO's theory is correct that anybody with any experience, not just with UNIX System V, but any modification or derivative work of UNIX System V, is severely constrained in their employment, and that is contrary to public policy as the cases in our papers make clear....

SCO's theory is, finally, inconsistent with IBM's rights of ownership. It admits that IBM owns the material at issue. It admits that ownership carries with it the exclusive right to distribute, and yet they take the position, Your Honor, that would nullify as a practical matter IBM's rights of ownership.

This is all truly absurd, the result, IBM says, of "a distorted reading of Section 2.01 of the agreement", none of which applies to the code at issue, since they are not "resulting material" as defined in the agreement. Finally, SCO's CEO Darl McBride, in his deposition, contradicted SCO's argument about controlling AIX, Marriott points out in arguing the second motion:

Mr. McBride, the CEO of the SCO group, Your Honor, testified in his deposition in a way that is completely contrary to the proposition that SCO can control anything and everything that is in the AIX operating system. He said the following, quote, I am sure there are things inside of AIX that were not derived from System V or from one of our contractual arrangements, that they would be free to do whatever they wanted.

Both JFS, developed for OS/2, and RCU, patented by IBM, would seem to fall under that very McBride umbrella. Marriott:

SCO's theory depends on the proposition that AIX is a derivative work of AT&T's UNIX System V software, for which we don't believe there is adequate evidence in the record, despite Mr. Singer's assertions to the contrary, but assume that it is, Your Honor, the JFS contribution came from IBM's OS2 operating system and it did not come from IBM's AIX operating system.

Not only that, but if SCO did have any rights in JFS, it handed them over to UnitedLinux. I'm afraid SCO's experts, once again, come off looking not so expert:

The individual identified by SCO as having made these supposedly improper contributions have offered their testimony, the people in the best position to know that it came from OS2 and not from IBM's AIX operating system. SCO's claim that it came from IBM's AIX operating system, Your Honor, is mistaken for the reasons which we set out at tab 13 of our book. It relies principally if not entirely upon the testimony of SCO's Dr. Ivie, and that analysis, Your Honor, relies upon evidence that was required to be put in SCO's final disclosures, and that it didn't put in those final disclosures, and that Magistrate Judge Wells struck at a hearing late last year, an issue in which SCO has taken an appeal.

The only evidence they purport even to offer, Your Honor, that could competently demonstrate that and they argue is that of Mr. Ivie. Mr. Ivie's testimony was untimely provided and it can't support the motion. In any event, Your Honor, Mr. Ivie's testimony is not competent evidence that JFS came from the AIX operating system as opposed to the OS2 operating system. Dr. Ivie says that there is a similarity between the JFS contribution and the JFS that is in AIX. Well, that is not a surprising similarity, Your Honor, because the JFS contribution in AIX came from IBM's OS2 operating system, so the similarity on which he relies for these opinions is not a similarity that has any probative effect. That is point one, Your Honor, as to why the JFS contribution allegations fail.

And whatever does SCO mean by calling Paul McKenney, as you will see, an "IBM secret programmer"? You see how SCO's theories over and over hit the rocks? Or as Marriott puts it, they fall flat? I've concluded that either SCO never understood the tech, or they are pretending not to, in hopes the judge won't know the difference. I don't know what to say about the experts, because without reading what they wrote, it's impossible to know, and I have found it best not to rely solely on SCO's characterizations of anything. I sincerely hope Judge Kimball follows that same careful course and checks every word for accuracy.

SCO's Mr. Singer, as always, is very persuasive, and if one knew nothing about the tech, he might prevail. But then he says such silly things sometimes, it undercuts his arguments, if you do know the tech. When you read his argument about what Linux needs to run, you'll guffaw. But as for any predictions, the attorneys refer to so many documents that we can't see, it's not actually reasonable to predict anything. Remember that all SCO needs to do is raise one pertinent fact in dispute. It's a very low bar. If you held a gun to my head and said I had to predict or else, I'd say if the judge understands the tech, he'll rule for IBM on the first motion, and the second motion is such a slam dunk, I don't see any way SCO can prevail.

Wait a sec. We have these transcripts *before* PACER does. Uh oh. Shoutout to the SCO Smear-Groklaw Brigade. How in the world could Groklaw have these documents before they are digitally available to the world? Did IBM send them to us?

No, you silly wabbits.

Here's how we do it. You start with a paralegal who has filed many court documents and thus knows how the system works. Add thousands of volunteers in every state willing to trudge to the courthouse. Pick a couple of volunteers in Utah, explain the system, and ask them to go to the courthouse, get to know the very pleasant clerks and court reporters, and pick up the documents. When necessary pay for them. These transcripts, for example, cost Groklaw $254, which the Groklaw community's generous contributions paid for. See the tip boxes? By the way, thank you, everyone.

It's a technique any journalist could follow, if they knew how and had volunteers in every state, like Groklaw does. There's really no mystery.

That's America, for you. The public gets to know what is going on in the legal process, even if they have no in with anybody, no pull, no inside track. And may I just mention here that from our experience, the court personnel in Utah are about as nice as they come? Of course, dealing with court clerks in New York City might have tilted my expectations of courtesy. You know who you are.

: )

But let me say this: if SCO or IBM or Novell ever wanted to send me free copies of their filings, as a journalist I'd be very grateful, and if it could be plain text, well, that would be heaven. And if SCO ever wants to donate money to Groklaw, hey, why not? Joke, joke.

Of course, the latest smears are not true, to be serious for a minute. Groklaw isn't a front for anybody. Groklaw is trying to speak as an independent voice for the one group that isn't directly represented in this litigation, the Linux and FOSS community. You know. The guys who wrote the code being fought over? Or we thought it was that code. It turns out it wasn't at all. SCO just got media interest by making such a claim. Now, in the end game, it's all about "negative know how" and JFS that IBM wrote itself and methods and concepts and various other litigious spinach. But at the beginning, the whole world thought it was about Linux, and if Groklaw had thought the parties could represent the community's interests fully, there would have been no need for Groklaw.

Chris made a PDF, showing the line numbers, for those who want to see it exactly as it is in the original form, and of course, as always, for anything that matters, that's the version to go by. As for perceived errors, it is what it is. We don't change so much as a comma in the transcripts, even if we see what we think is an obvious error. Our job is to present the document exactly as is, with the only exception being that we do remove the line numbering, in honor of our readers who are blind and need to use screen readers. If the numbers are left in, it's very hard for them to follow along. So what I'm saying is that there is no need to list corrections for material in the transcripts, unless you wish to present a factual correction for the record. My stuff is fully correctable, as always. I will just mention that on page 65, Marriott mispoke (or it was recorded incorrectly), when he says, "There is absolutely no evidence that IBM has continued to distribute AIX." Rather, if you recall from the declarations submitted as exhibits and IBM's brief, it's Dynix that IBM didn't continue to distribute and in fact had stopped distributing before SCO even raised the issue in the litigation. IBM still distributes AIX.

Remember that transcribers have a really hard job, having to capture every word in real time, and people sometimes talk very fast or not so clearly, and there are proper names that they must guess how to spell, if the lawyers don't think to tell them. I want to personally thank them for the job they do, and I'm very thankful that the US court system allows the public to follow cases that matter to them, even if we are nobody.

You'll notice they begin by talking about timing, taking some time from this hearing to add it to the next one. Yes, that means exactly what you think, that the lawyers practice in advance what they intend to say and time it, knowing they must stay within firm time constraints. Why do courts do that? Because otherwise, some lawyers will talk on and on forever. Actually, SCO's Brent Hatch goes a bit overtime on the second motion.

Speaking of long-winded, I'll have to pass on commenting on the other two transcripts today, and leave them for when and if I get another wind. This one has me tuckered out, even with Chris and Steve's wonderful help, and for sure, SCO got one thing right: the motion papers and briefs in this litigation are longer than War and Peace.

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

IN THE UNITED STATES DISTRICT COURT
DISTRICT OF UTAH
CENTRAL DIVISION

BEFORE THE HONORABLE DALE A. KIMBALL
------------------------------------
March 1, 2007

Motion Hearing

Ed Young
Court Reporter
[address]

1

APPEARANCES

2

March 1, 2007 2:30 p.m.

THE COURT: We're here this afternoon in the matter of SCO v. IBM, 2:03-CV-294.

For plaintiff Mr. Brent Hatch and Mr. Edward Normand and Mr. Stuart Singer.

MR. SINGER: Good afternoon, Your Honor.

THE COURT: Good afternoon.

For defendant, Mr. David Marriott and --

MR. MARRIOTT: Good afternoon.

THE COURT: -- Ms. Amy Sorenson, Mr. Michael Burke and Mr. Todd Shaughnessy.

MR. SHAUGHNESSY: Good afternoon, Your Honor.

THE COURT: I understand that you want more time on Wednesday. We can start at 2:00 on Wednesday, so you have the extra 20 minutes that somebody is asking for.

MR. SINGER: We appreciate that, Your Honor.

MR. MARRIOTT: Thank you, Your Honor.

THE COURT: 2:00 on Wednesday.

I have a jury out, and if a verdict comes in, then you can all go somewhere else while we get the jury back in to take the verdict. That has to take precedence. Sorry about that.

First we have IBM's motion on summary judgment on

3

SCO's contract claims, and SCO's motion for partial summary judgment or SCO's third cause of action for breach of contract. 45 minutes per side.

Who is going first?

MR. MARRIOTT: I believe that is me, Your Honor.

THE COURT: Go ahead.

No one is required to take all of the time asked for, but --

MR. MARRIOTT: Thank you.

THE COURT: You, of course, can.

MR. MARRIOTT: We have prepared a book of exhibits, if I may approach for the Court's convenience?

THE COURT: Sure.

MR. MARRIOTT: We have a copy for counsel.

THE COURT: Thank you.

MR. MARRIOTT: Referring Your Honor to tab one, if I may, of the book, there are four contracts that underlie SCO's claims for breach of contract: The IBM software agreement, the IBM sublicensing agreement, the Sequent software agreement, and the Sequent sublicensing agreement. And SCO has a separate claim or count with respect to each of those four contracts. IBM is, we respectfully submit, entitled to summary judgement with respect to each of those counts for at least four reasons which are set out at tab two of our book.

THE COURT: And they are also in your briefs.

4

MR. MARRIOTT: And they are also in our briefs.

What we would like to do, with the Court's permission, is focus in the few minutes that we have on the first three of those reasons. And then I would like to focus, Your Honor, on the particular technologies which are at issue on this motion, those things which SCO contends were contributed improperly by IBM to Linux. Those appear, Your Honor, at tab four of the book.

Let me make clear, if I may, from the outset, something about those four items of technology. First is the JFS contribution, which is a contribution SCO contends was made by IBM to Linux in violation of the IBM agreements. That is the first category.

The next three categories are the RCU contribution, the LTP contribution, and certain negative know-how and experience. The allegation by SCO is that those contributions were contributed by IBM to Linux in violation of the Sequent agreements. Now, we'll come back to those technologies in detail, but let me make a few things absolutely clear about those technologies from the outset. None of that material, Your Honor, is UNIX System V source code, methods or concepts. None of that material is a modification or a derivative work of Unix System V. None of it was contributed to Linux by reference to UNIX System V, which this is original IBM work created independent of UNIX System V. That brings me, Judge,

5

to the first of the four points I would like to underscore.

First, Your Honor, is that SCO can't establish a breach of contract and that is true for three reasons. The first reason is the plain language of these agreements forecloses SCO's theory. Second is that the parol evidence, the overwhelming parol evidence forecloses the theory. The third is that SCO's theory is patently unreasonable, as I intend to demonstrate and, therefore, untenable under the controlling cases.

As I said, there are four contracts at issue. Two of those are the sublicensing agreements. SCO has not offered in its papers a shred of evidence to demonstrate that IBM breached the sublicensing agreement. It has not identified which provision it breached. It has not identified any evidence of breach and it has not explained a breach. For that reason alone, the two counts related to those contracts, counts two and four, are disposed of without further analysis.

That leaves the remaining two claims concerning the software agreements. Your Honor, let me take the three points as to why summary judgment is appropriate as to each of those. First, the plain language. There is no dispute here that the contracts at issue concern software product, AT&T's UNIX System V material. There is also no dispute that IBM has not contributed to Linux any UNIX System V material. The entire case as it concerns SCO's contracts turns on the so-called

6

Section 2.01 of the agreement, the resulting material provision.

That section, however, Your Honor, speaks about the licensee having the right to prepare modifications and derivative works, provided the resulting materials are treated under the contract as confidential. You'll see that language at tab nine of the book. It is undisputed, as I said at the outset, that none of the four categories of material at issue is resulting material. None of it is a modification and derivative work of System V. For that reason alone the claim fails.

Now, faced, Your Honor, with that fact, SCO contends that Section 2.01 not only gives it rights with respect to UNIX System V, and modifications and derivative works, but anything that ever touches or is in any way associated with the modification and derivative works of UNIX System V. That we respectfully submit stretches the meaning and the language of Section 2.01 to an absurd degree.

If I may illustrate, Your Honor. If you imagine and if you look at the demonstrative to Your Honor's left and to my right, that is a depiction of the Linux operating system. You'll see a series of circles with particular chunks or pieces in it. Imagine you have a product, Your Honor, with 1,000 different pieces to it. Those pieces represent various components of the system. So imagine you have this and this is

7

Your Honor's product and you have 990 of these pieces, if there are 1,000, and they are Your Honor's. They are your own original work.

Well, imagine that you want to license from others additional material and to add them to your product, and you take the additional ten that make up the 1,000 and you license them to party A and party B and party C. Let's just assume that one of the items that Your Honor licenses from a third party is licensed from SCO. It is AT&T's UNIX System V software. Let's assume that the contracts mean what SCO says they mean. Your Honor, under SCO's theory of the case, not only is Your Honor -- this is your product which you have licensed other people's materials in part, ten of the 990, under SCO's theory, Your Honor, not only are you required to keep confidential the one part of the 1,000 that you licensed from SCO, you're required to keep confidential the entirety of the product, so as to keep confidential the one.

That does not make SCO's case, Your Honor, because IBM has neither disclosed the one, nor has it disclosed in entirety of the product. So SCO's theory goes to another level. It says you are required, owner of this product, the 1,000 items, to keep confidential anything and everything that is in it, even if you take it out, and even if you license it from a third party who says to you you can do with it what you want, and even if you take out the one item, Judge, and put it

8

on the shelf, that material and those 1,000 items, 990 of which are your original work under SCO's theory are controlled by them. You might own them, they say, but they control them.

If you take it out, take out your one and you put it in another product, under their theory they also control the other product, and whatever the other components are in that product. That, Your Honor, we would respectfully submit is not a reasonable construction of Section 2.01 of the agreement.

The second point I want to make concerns the parol evidence. The overwhelming parol evidence here, Your Honor, compels the conclusion that SCO's theory is not a reasonable construction of the agreement. By the plain language, Your Honor, parol evidence ought not to be considered. In the event that Your Honor considers it, we would submit, and it is overwhelmingly in favor of IBM's construction, there are ten individuals, and we called them involved persons in our papers, who were involved in the execution and the negotiations of these agreements, five of them from AT&T, three of them from IBM, two of them from Sequent. They appear on the scene before Your Honor. Those individuals have offered specific, and we believe undisputed testimony, that refutes SCO's theory of the case. If we may share with Your Honor some clips from their depositions.

Mr. Wilson.

(WHEREUPON, the following deposition clips were

9

played.)

MR. WILSON: The only part of the derivative work that would have to be protected under the software agreement would be that portion of the software product that would be contained in a derivative work.

To the extent that modifications of derivative work contained portions of the software product, they were to protect the software product portion under the terms of the license agreement, that portion which was theirs, whatever, they can do whatever.

The intent was to make sure that we protected the software product. To the extent that they used that software product and created works which were original works, that was up to them to do what they pleased with those things.

When you say those things, you mean that portion of a derivative work that they had developed and that in your view they owned; is that correct?

That's correct.

MR. MARRIOTT: That was the head of AT&T's licensing business.

(WHEREUPON, the following deposition clips were played.)

Mr. Wilson.

If they created a derivative work and the derivative work contained zero content of the software product, then they

10

could do as they wish. If that contained portions of the software product, then they had to abide by the terms and conditions of the agreement with regard to that portion that contained the derivative work, contained the software product. Our agreements only went to the software product itself, not to their original created works.

MR. MARRIOTT: This is the man for AT&T that signed --

(WHEREUPON, the following deposition clips were played.)

David Frasure.

With respect to the agreements that IBM entered into with AT&T, is it your understanding that AT&T through those agreements had any right to control IBM's use of source code that it developed on its own?

They had no right to control that software that was developed by IBM.

With respect to the agreements that Sequent entered into with AT&T, is it your understanding that AT&T through those agreements, gained any right to control Sequent's use of the source code that it developed on its own?

They did not gain any rights to control the software developed by Sequent.

Was it the case that from AT&T Technologies' perspective that the licensee could do whatever it wanted with

11

the source code it developed?

Yes.

Michael DeFazio.

The basic idea is that the AT&T USL, Novell code was owned and protected by Novell. The code developed by our licensees was owned and controlled by them. When the two went together to market our rules prevailed. When the two were separate our rules prevailed on ours, and their rules prevailed on theirs.

Would you agree, Mr. DeFazio, that the agreements did not and do not give AT&T, USL and Novell or any of their successors or assigns the right to assert ownership or control over modifications and derivative works prepared by its licensees, except to the extent of the original UNIX System V source code included in such modifications and derivative works?

That's correct.

Stephen Vukasonvich.

And any code that IBM modified, in your view, under this provision, IBM thereafter owned?

They owned any modification. They owned their code that they developed, and we owned our code.

Ira Kistenberg.

By the terms of Section 2.01, did AT&T intend to restrict Sequent's rights to disclose code that Sequent

12

developed on its own?

If Sequent developed it on its own, AT&T had no rights to it.

My question is simply whether it was AT&T's understanding that Sequent would own the modifications and the derivative works that they prepared based on UNIX System V.

They owned it to do what they want with it, yes.

Were licensees of AT&T free to use and to disclose the modifications or derivative works they created provided they did not use or disclose any portion of licensed UNIX System V source code?

Yes.

Thomas Cronan.

In my discussions with AT&T what they were trying to protect were several derivative works. They were trying to protect their own System V code. They felt like they had to protect anything that was shipped around with their code in it. If we separated out our cord from their code they didn't need to protect it. That was our discussion.

Richard McDonough.

Another huge issue for us was ownership of whatever we developed, and we wanted to make it clear that whatever we developed we owned. We weren't making any claim to the code that AT&T owned and developed itself, but we wanted to make certain that anything we or our subcontractors or anybody

13

working for us developed, we owned and had the right to use and license.

I would just say that there is no way on God's green earth that I would have entered into an agreement where somebody else owned what IBM was investing tens of millions of dollars in developing. An agreement never would have happened. So we had to make it clear that whatever IBM developed IBM owned.

Jeffrey Mobley.

My understanding was that we were free to do whatever we wanted to do with the products we were developing.

David Rodgers.

My understanding of the license is that the UNIX System V code had to be maintained as the AT&T private property and withheld from disclosure, but if there were other elements of the software product created by Sequent, that those were Sequent's to dispose of as it chose.

Roger Swanson.

Those parts of the source code that we wrote were not required to be kept confidential under the terms of the agreement, but we could do with them as we saw fit.

The aspects of the derivative works that we developed, we could choose to disclose or not disclose. We had the ownership, we had the control. That was my understanding.

Once, again, that's precisely what our understanding

14

was, that once we had made modifications, we still had an obligation to protect that part of the UNIX System V material, according to the restrictions of the agreement, but that work that we had developed ourselves, the source code that we developed, was not included as a part of those confidentiality restrictions.

(WHEREUPON, the deposition clips were concluded.)

MR. MARRIOTT: These witnesses, Your Honor, have offered substantial additional testimony, and it is in the papers, and we have provided CD's in the book, and we have provided Your Honor with three hours of the same, should you feel like you need to see that.

THE COURT: I appreciate that.

MR. MARRIOTT: I thought you might.

In an attempt to avoid summary judgment, Your Honor, SCO has pointed to the declarations of a number of witnesses. They appear at tab 13 of the book. These people have the use of the contracts, which they never communicated to IBM as they were not a part of the negotiations, and a number of them were not employed by AT&T at the time the agreement were executed. Their testimony, we submit, is no impediment to the entry of summary judgment. As overwhelming parol evidence, the testimony of those who negotiated and executed these agreements, who communicated their views to IBM, is we submit overwhelming in favor of IBM and against SCO's theory.

15

You looked like you had a question.

THE COURT: I don't.

MR. MARRIOTT: The third reason, Your Honor, why the contract claims fail is that SCO's theory is, as we show in the book at tabs 19 through 23, we think unreasonable as a matter of law. It is inconsistent with IBM's ownership rights, it is contrary to copyright law, it is against public policy, and it leaves in circumstances I'll explore later, an absurd result.

The second point, Your Honor, independent of what we believe is the undisputed evidence of no breach, that I want to emphasize here is that SCO is estopped from pursuing its theory of the case. For almost two decades following the execution of these agreements, some of these representatives of AT&T and USL and Novell and others, told IBM and other licensees that they could do as they wished with their own original works, so long as they protected AT&T's UNIX System V software.

Mr. Wilson, the head of the licensing business, said he on numerous occasions did that. Mr. Frasure, who executed the agreement on behalf of IBM, as you'll see at tabs 25 and 26 of your book, said, quote, that he personally, quote, assured licensees in no uncertain terms that they could do as they wished with their original works. He, quote, often told people that UNIX licensees could freely use and disclose their original or homegrown UNIX method, code and concepts.

Mr. Green, another AT&T representative, also at USL

16

and Novell, has testified that, quote, more times than I could remember, close quote, licensees were told that they could do as they wished with their original works.

Ms. Tilley of AT&T, then USL, then Novell, said that representatives of AT&T and USL and Novell, quote, consistently informed licensees that they owned their code and that they could do with it as they wished.

Mr. Crab, formerly of Santa Cruz and then of Caldera, said that while at Santa Cruz, Santa Cruz told licensees that they were free to do as they wished with their own code, modifications and derivative works, so long as they protected AT&T's System V source code.

AT&T and its licensees, Your Honor, took AT&T's licensees like IBM and Sequent, took AT&T at their word. They publicly disclosed their own original works in the time following the execution of these agreements. Examples of those disclosures are in the papers we have provided. There is a cart with books in front of Your Honor in which repeated disclosures were made over decades of the code and the methods and the concepts of AT&T's UNIX System V software. And, Your Honor, of the modifications and supposed derivative works of that. Hundreds of books have been written on the subject. Those disclosures are no different, Your Honor, than the kinds of disclosures that SCO has included in its final disclosures.

Neither AT&T nor its successors raised any objection

17

to those disclosures until this lawsuit became a glimmer in the eye of current management. SCO's former CEO, Ransom Love, has testified that, quote, after Caldera acquired ownership of UNIX code, and this is at tab 29, even though we were aware that IBM was disclosing homegrown code, we made a conscious decision to take no action against such disclosures. IBM, Sequent and other licensees no doubt reasonably relied on the repeated statements by AT&T and its successors that they could do as they wished with their own works so long as they protected AT&T. IBM and Sequent built businesses, Your Honor, on that proposition and invested hundreds of millions of dollars in the idea that they would actually control their own stuff.

Estoppel can, no question, be in certain circumstances a fact intensive inquiry. It is not always appropriate for summary judgment. Courts can and they do enter summary judgment on grounds of estoppel, and I would respectfully submit, Your Honor, that if there is a case for it, this is it.

Third point, Your Honor, the alleged breaches here have been waived. They have been waived for three reasons. They have been waived because the repeated statements of AT&T and its successors over nearly two decades not only amount to estoppel but they amount to waiver. I won't repeat that ground.

The second reason why there is waiver here, Your

18

Honor, is that Novell, the supposed predecessor interest to SCO, exercised its rights under an asset purchase agreement with the Santa Cruz Operation, Inc. to waive the alleged breaches here. I will come back to that.

The second item I would like to focus on, Your Honor, is SCO's own conduct which we believe results in a waiver of alleged breaches here.

First, Your Honor, with respect to Novell. Section 4.16B. If you look at tab 36 of the book, Your Honor, after the commencement of this case Novell exercised its rights under Section 4.16B of the asset purchase agreement to cause a waiver of the alleged breaches here. The asset purchase agreement says that seller, in this case it was Novell, shall amend -- in addition, seller in its sole discretion, buyer shall amend, supplement, modify or waive any rights under, or shall assign any rights to any SVRX license to the extent so directed in any manner or respect by seller, Novell.

In the event that the buyer fails to take such action, the seller, Novell, shall be authorized and is granted the right to take that action on the buyer's behalf. Well, it is undisputed here that Novell, following the commencement of this case, exercised its rights under 4.16B. In the dissent letter it asked SCO to waive, and SCO declined to waive and Novell took that action. The only issue is whether Novell has the right to do that.

19

I respectfully submit, Your Honor, that the plain language of Section 4.16B gives Novell that right. SCO has raised a number of arguments in opposition, some as to the parol evidence, and some as to the plain language of the agreement. We deal with those arguments in our papers and at tabs 38 and 39 of the book. Time won't allow me to repeat them here.

Let me just say this about SCO's argument. SCO's reliance on the parol evidence is we submit impermissible, Your Honor, because the plain language controls, and the parol evidence cannot be used to alter the plain language of the agreement. As to the plain language of the agreement, what SCO says in effect, Your Honor, is that Novell had waiver rights, but it doesn't have waiver rights to the agreement that are at issue in this case, the agreements that IBM is supposed to have breached. The waiver rights by the terms of Section 4.16B relate to SVRX licenses. The question is, are the agreements at issue here SVRX licenses? As we show at tab 40 in Your Honor's book, I submit there is no question that they are. SVRX stands, Your Honor, simply for System Y release. There is no question that the agreements IBM is supposed to have breached are SVRX licenses. They are agreements licensing System V releases.

In a letter from SCO's CEO to the CEO of Novell, following Novell's exercise of its rights under 4.16B,

20

Mr. McBride acknowledged that the agreements at issue in this case are SVRX licenses. He said, and I quote, in your June 9 letter, you, referring to Novell, attempt to assert claims on behalf of IBM and with respect to its SVRX licenses with SCO. In SCO's opposition papers, Your Honor, in connection with this exact motion, at paragraph 201, SCO acknowledges that the agreements at issue are SVRX licenses. It says, quote, effective June 13, 2003, SCO terminated IBM's SVRX license. Effective July 30, 2003, SCO terminated the Sequent SVRX license. Novell has waived the alleged breaches at issue.

The next waiver issue I would like to discuss, Your Honor, concerns SCO's conduct. This is summarized at tab 41. SCO shipped the exact four categories of code material we're talking about, Your Honor, in its own product. It shipped it in its United Linux product, and it shipped it in that United Linux product pursuant to the terms of the General Public License or the GPL. Each of those acts resulted in a waiver of the alleged breaches. Let me tell you why.

First, United Linux. If you look at tab 43 of the book, Your Honor, SCO was part of an initiative known as United Linux. As a part of that initiative SCO and its partners assigned any intellectual property rights they had, with the exception that is not relevant here, to the material that ended up in United Linux product. The material at issue here is in the United Linux product. If you take a look at, Your Honor,

21

tab 44 of the book, you'll see the language of the United Linux agreements. The United Linux Joint Development Contract says that all intellectual property rights in the software developed pursuance to the JDC, other than the preexisting technology and enhancements, shall be assigned by the members, which SCO is one, and any individual member to the LLC and shall be owned by the LLC.

Well, SCO's Linux IV product was, Your Honor, software developed pursuant to the JDC. We show that at tabs 44 and 46 of the book. Again, the items of information that is supposedly misused here was in that product. Any rights SCO had to the material at issue, Your Honor, and we respectfully submit they had none, but any rights they had were assigned by them to the United Linux LLC, and they are in no position now to assert any claim of breach with respect to it.

With respect to the GPL, the General Public License, again, SCO shipped the material at issue in products that were licensed under the GPL. What does the GPL say? The GPL says, among other things, that if you distribute copies of the program covered by the GPL product or for a fee you must give the recipients all the rights that you have. You must make sure that they too receive and can get the source code. You give them the right to make copies and to distribute verbatim copies. So the very material that supposedly represents the breach was put by them in a United Linux product and shipped

22

under the General Public License under those terms. That, too, Your Honor, represents a waiver.

Now, if I may, as time is short, with respect to the technology items at issue, and let me say this again, there are four of them. I would point you to tab 53 of the book where they are listed. The facts related to these four technology area are at tab 54 of the book. These four items of technology have one thing in common. That one thing is dispositive of SCO's claims. That one thing is that none of those four categories of material is resulting material. Under their theory of the case, Judge, IBM's liability depends at a minimum on it being resulting material. They are not, therefore, the claims fail. None of the four categories include System V methods or concepts, none of them are modifications and derivative works of UNIX System V. They are original IBM works, just as in my example of the 990 original works of Your Honor, and SCO has under the plain language of the agreement, we submit, no right to control them.

There are four, as I said, Your Honor. The JFS contribution is at issue in the next motion, and with Your Honor's permission, I will argue that in that connection. Let me just focus on the three remaining ones, the RCU contribution, the LTP contribution, and the negative know-how experience. The RCU contribution is at tab 58. Putting aside the fact that it is not resulting material, and putting aside

23

the fact that SCO has waived any right to pursue it, SCO's claim with respect to the RCU contribution is barred by the statute of limitation. The statute of limitations under New York law, which controls, is six years. RCU was disclosed in a patent application in 1993 and the patent issued in 1995.

Under SCO's mistaken theory of the contract, Your Honor, IBM's filing of the patent application amounts to a disclosure. The statute of limitations ran before the commencement of this claim and the claim is barred.

In any event, the witnesses, the only individuals identified by SCO as having anything to do with that, and whose pictures and testimony appear at tab 60, have debunked SCO's claims.

The LTP contribution. Again, putting aside the fact that it is not resulting material, putting aside the fact of waiver and estoppel, SCO's claim with respect to the LTP contribution fails, Your Honor, because it depends on the proposition that the LTP contribution came from the DYNIX operating system, which SCO contends is a modification and derivative work. It didn't.

The LTP contribution was not part of the DYNIX operating system. The evidence, which we set out at tab 62 of the book, makes that perfectly clear. It was not a part of it and under their own theory it fails.

Lastly, Your Honor, the negative know-how category.

24

This is a category of 11 items of supposedly misused information. We lay this out at tab 64. There are two basic brands of claim as it relate to these 11 items. One of them concerns experience. Here is what SCO's claim is as it relates to the experience. It says, as we show at tab 65, Judge, it says IBM has breached its contract by permitting IBM developers exposed to DYNIX PTX methods and concepts to contribute to Linux in the same area for each developer's work. So anybody that had any exposure to DYNIX can't work in connection with any other operating system in that area. That is the claim.

With respect to the negative know-how, which also appears at 65, the gist of the claim is that people who had some knowledge about DYNIX told people who were working on Linux not to do certain things in DYNIX that they don't think worked. SCO has identified 16 supposed wrongdoers as it relates to these 11 categories.

Those individuals, Your Honor, as I will come to, have offered undisputed testimony that debunks SCO's claim. The claims as to these 11, Your Honor, fail for three quick reasons, and I will sit down.

First, Your Honor, the agreements which SCO contends were breached, Your Honor, do not contain any provision which would allow it to preclude IBM employees from using their experience and their general know-how in working on projects other than the one on which they are presently working. There

25

is no connection drawn in SCO's interrogatory answers or in its paper between the contract and this claimed notion of misuse of experience.

Second, Your Honor, though Magistrate Judge Wells allowed SCO to pursue, and declined to throw out in the discovery phase SCO's claim as to these 11 items, she made it perfectly clear in her order, which we have set out in your Honor's binder, that these claims were subject to being, in her judgment, at least, disposed of on summary judgment.

Your Honor ordered SCO to provide particularity as to these claims as did Magistrate Judge Wells. That has never been provided, and for that reason alone they are out.

Finally, careful consideration of these items shows that the claim falls apart. If you look at tab 70, Your Honor, you will see the pictures of each of the 16 individuals who are at issue and what it is they said in testimony that is unrebutted by SCO. Here is what they said. They didn't make any contributions to the files or the directories listed, or they didn't base their contributions to the list of files on UNIX System V in making the contributions.

You will want to look item by item. Your Honor, look at tab 71. For two of these items, and, again, this is undisputed, for two of these items, Your Honor, 188 and 187, the technology which SCO alleges was misused didn't even exist in DYNIX. For another two of the items, 43 and 94, the

26

material had nothing to do with DYNIX PTX. It was discoverable outside of IBM. For five of the items there is absolutely no evidence that the alleged wrongdoers had any experience in the technology area where they were supposed to have misused it in some other area. SCO admits that with respect to four of them. With respect to items 23 and 90, Your Honor, for two of the items the material was disclosed in marketing materials, and in footnote ten of SCO's opposition papers, it says that any such disclosure -- if the material is inactionable anyway. Finally, item 189, Your Honor, was based on knowledge available in public literature.

As is summarized in tab 72 of the book, SCO's negative know-how claims and its experience claims we think underscore the absurdity of the position. If SCO's theory is correct that anybody with any experience, not just with UNIX System V, but any modification or derivative work of UNIX System V, is severely constrained in their employment, and that is contrary to public policy as the cases in our papers make clear.

The only conceivable reason, the only conceivable reason why AT&T could have wanted to protect IBM, and in my example Your Honor's original works, was to protect the one item that you put in your product. SCO has acknowledged, Your Honor, and it did it early in the case, that there are no secrets in UNIX System V. They said that in open court and

27

they withdrew their trade secrets claim.

SCO's theory is, finally, inconsistent with IBM's rights of ownership. It admits that IBM owns the material at issue. It admits that ownership carries with it the exclusive right to distribute, and yet they take the position, Your Honor, that would nullify as a practical matter IBM's rights of ownership.

Thank you.

THE COURT: Thank you. Mr. Singer.

MR. SINGER: Thank you, Your Honor. We also have arguments books, if I might approach?

THE COURT: Please.

MR. SINGER: Your Honor, I would like to begin with why these contract claims matter. The UNIX operating system developed by AT&T was its crown jewel. It is the operating system of choice for mission critical application, and it became in the 1990s the leading operating system worldwide.

Now, major computer companies like IBM wanted to adopt that system for their own hardware. They had a choice. They could come up with their own operating system. In fact, IBM tried. It is called OS2. It failed. IBM, like others, struck a deal with AT&T. Give us access to your source code so that we can develop our own flavor of UNIX that will run better on our hardware. AT&T agreed but subject to very strong

28

restrictions, and not just on the original System V code, but on what they knew IBM and others would do with it, the so-called modified and derivative works.

They required that those modified and derivative works be treated just like the original software code in language that is as plain as can be in the relative agreements that are enforceable today. That is why we are here.

AT&T preserved its head start in developing UNIX while allowing companies to, for a royalty, develop their own systems that were UNIX systems that, of course, would pay royalties for the object code that would run the various machines. What they could not do is exactly what IBM has done here, give away that head start by open sourcing of derivative technology that they would never have had if it weren't for access in the first place to System V, and to allow a jump-start to Linux, which has had a tremendous effect on SCO's business. SCO was the inheritor of AT&T's UNIX business. SCO dominated with UNIX on the Intel chip platform market, and it had 80 percent of the market in the late 1990s.

It did ventures with IBM, as you'll hear more about this on Monday when we have our unfair competition argument, but a joint venture project, Monte Ray, to use SCO's know-how in that field to develop joint products. IBM then decided to abandon project Monte Ray, and instead to devote technology to Linux and to disclose proprietary UNIX technology. At tab one

29

we just touch on a little bit of that, which is the experts', Mr. Rochkind's and Mr. Ivie's conclusions that IBM's disclosures is what turned Linux into a strong enterprise system.

If you turn to the third page, in tab one you see a graphic where IBM's own document shows it directed the strength of AIX and PTX, which is DYNIX, into Linux as their proposed UNIX strategy. You have a summary of the vast amount of technology just on the issues which are still subject to the Court's order in the case, substantial amounts of technology that have been contributed. You have Mr. McKenney, who is an IBM secret programmer, who expressly stated in a document that we will mine key ideas from AIX and DYNIX PTX, and that this will make Linux more disruptive to proprietary operating systems like SCO's business. And the reason they can do this is because they have access to DYNIX PTX source code and access to top operating system experts in AIX and DYNIX.

The result, which is depicted graphically on the last slide in section one, is the rapid decline in SCO's open server UNIX new license revenue following those technologies being distributed in early 2000. That is why these contract claims matter a great deal.

I would like to address at the outset why a trial is needed, and why this is not a matter for summary judgment. One might think that that is obvious just from IBM's presentation.

30

There are hundreds of pages of briefs, reply papers and addendum that are longer, I think, than War and Peace. 40,000 pages of exhibits. We counted 100 plus new cases just cited in their reply brief. That does not sound like a summary judgment case to us, Your Honor. It does not sound like a case where you can look at the plain language and you can say SCO does not have a contract claim.

I think that it is also clear when we look at the extrinsic evidence that IBM is so heavily relying upon, and because you're talking about both things that occurred over two decades ago, in part, and because you're talking about terms that are not clearly defined, like what does someone mean when they say an original IBM work? What do they mean when they say, as you heard on the screen, that it was developed without relying on System V?

Let's take a look at some of those witnesses just at the front end, and their changes of opinion at different points in the case. If you look at tab two, the first slide is Otis Wilson. He says, as you heard, his beliefs as set forth in a declaration that IBM and Sequent are free to open source all of AIX and DYNIX except the original System V code. 14 years ago, however, he gave strikingly different and contradictory testimony in a case, USL versus Berkeley, where he said that anything that that university created with the exposure to licensed software based on, contained a part of, was a

31

derivative work and had to be treated as a licensed software, which is, of course, what the plain language of the agreement says.

He also wrote that in a 1987 document. He also confirmed that that testimony was correct at his deposition, and he acknowledged it was no different in intent between a standard operating license and the educational license at issue in that case.

On the next page we talk about Mr. Kistenberg, another one of the IBM acknowledged involved persons. He negotiated the Sequent deal on behalf of Sequent. They have a declaration from Mr. Kistenberg that said no one at AT&T had intended to assert control over the portion of derivative work that did not contain System V code. He told a different story in his deposition.

In fact, he said in his deposition that he told IBM's counsel that his understanding was that the UNIX software given to any of the licensees, that any time they used the source code to develop derivative products, that that was a part of the System V source code, and that they could not turn around and give it to X, Y, Z companies. He states that clearly at various places in his deposition. He says in his deposition that he would never have signed that declaration if he knew the use that IBM would try to interpret his words to mean.

Mr. Chatlos, in fact, did exactly that. When he was

32

confronted with a proposed IBM declaration following the meeting, he said he would sign one that actually reflected what he had discussed, but that the ones presented to him didn't do that and he refused to sign.

Mr. Bouffard had his declaration submitted by IBM and, again on the key point we're talking about, these are not collateral points, but it is the key points that IBM wants you to overturn the plain language of the agreement and go with extrinsic evidence on. They have a declaration from Mr. Bouffard. He gives a later declaration to SCO which clarifies that when he said there was no right to control or own the modifications, he meant that AT&T couldn't appropriate such material, and said to IBM give us the AIX and we want to sell it for our benefit. And that he did not mean that IBM had the right to disclose the protected added on material and other code that was in the derivative product. That is directly contrary to his declaration and the position that IBM takes in this case.

Just recently in the Novell case, he gave a deposition and that is the next page which is excerpted here, where he was asked by Novell's counsel as to how it came to be that he signed the declaration that said that ownership and control was with IBM, and he explained that his IBM declaration wasn't written by him, that they went around and around in negotiating the language, and that finally he was worn down.

33

He said it wound up being a negotiation of my words rather than a document of my words.

Your Honor, this and the other extrinsic evidence is exactly why we have trials. We want to subject these witnesses to the crucible of cross-examination, and then the truth we believe will emerge. That cannot be shortcut in this case.

I would also like to talk about the assertion that was made that the particular derivative works at issue in this case are just things that were dropped in and had no basis related to System V. That is manifestly not the case. I would like to move all the way to tab 50, if you would, in the binder which we provided.

This deals with AIX. It is an excerpt from the report of Mr. Evan Ivie, who was formerly the head of the computer science department at Brigham Young. He has worked on UNIX all the way back to AT&T. He is a distinguished expert in the field. He has studied this, and as his report and underlying testimony show, half of the JFS files were based on System V source code. If you turn to the second page there we go into a little more detail. Mr. Hatch in the argument later this afternoon will even have even more detail on this. He points out that in 1990 or 1991 the first JFS was based on the preexisting system that was derived from UNIX source code licensed from AT&T, and that approximately half of those files were based on UNIX System V source code. How does he know

34

that? Because if you go back into the discovery of that, they have origin codes where IBM is included within AIX indicating that those came from System V.

Now, there is a dispute as to whether or not IBM created the JFS system that was contributed to Linux in a clean room environment from OS2, their own operating system.

THE COURT: Hang on a second.

MR. SINGER: Do you need a moment?

THE COURT: Go ahead, Mr. Singer.

MR. SINGER: Your Honor, that is a disputed fact as to whether or not it came out of 0S2 because we have competent expert testimony saying it came from System V.

In addition, we have admissions, such as an IBM programmer, who said, and this is also at tab 50, that the System V file system is where this stuff, referring to JFS, where this originated. It couldn't be more black and white than that.

Mr. Davidson's declaration is also further support of the fact that JFS is not some system that just was dropped into AIX and had no relationship to System V and now they are seeking protection over it. It itself was derived from System V.

At tabs 51 and 52 we have a similar analysis with respect to the origins and the relationship of DYNIX RCU. These are not things that were just dropped in and that were

35

extracted to give to Linux. These were interwoven with the DYNIX operating system. DYNIX wouldn't operate without them, and those were operating in a DYNIX environment, and they are by every meaning of the word derivatives.

Now, Your Honor, I would like to go back now to the legal argument and begin with the fact that there are four claims. We disagree with IBM on the issue regarding sublicensing contracts. They had no separate agreement about it in their initial brief. They said, which we agreed, that they depend on the original underlying primary source code agreement and, therefore, if there is a breach of the software agreement, there is also a breach of the sublicensing agreement for distributing material in violation of the software agreement, and Section 3.03 of the sublicensing agreement makes that clear.

Now, there are four and only four legal arguments made by IBM in their summary judgment briefing, their initial briefing, not their reply brief. I will deal with those here. We urge the Court to disregard all the new and additional arguments put into the reply papers that were not raised in the initial papers.

Turning to the plain language, Section 2.01 says that the right being given to IBM includes the right to modify the software product and to prepare derivative works based on such software product provided the resulting material, which we

36

submit are the derivative works, are treated hereunder as a part of the original software product. That means treated just like the System V code.

What does that language not say? It does not say what IBM suggests. It does not say that IBM is free to take any part of this and distribute it to whomever they wish in source code form. The contract between what IBM's position in this case is, and what the agreement says, could not be clearer than just contrasting the two. In IBM's brief they say that IBM owns and is free to disclose any material that Sequent created so long as it does not contain UNIX System V material, and, as we have seen, that is simply not what 2.01 says.

Section 7.06, another part of that agreement, also makes it clear that the licensee has to hold all parts of the software product subject to this agreement and in confidence for AT&T, that it may distribute products only to third parties having licenses of equivalent scope, and that the licensee may obtain materials based on the software products subject to this agreement from such a third party, and use such materials, provided that the licensee treats such materials as if they were a part of the software product. Think about what that means. IBM is restricting itself to use materials based on a software product, not the software product as they say just System V, but just based on that from a third party, but can only use those materials if they treat them as part of the

37

software product itself. That is just like 2.01 suggests.

Now, what IBM does not spend any time on are the side letter and amendment X, which clarified and modified in certain respects the IBM AT&T agreement but not the Sequent agreement. We have dealt with that in our briefs because those support our position. They gave ownership rights over derivative works to IBM, but did not give them a release from the control rights and they could not disclose the source code to others. That was clear under 706(a) which appears in the '85 letter. These are by way of tab 8 in the book. The 1987 letter continues to protect all parts of derivative works. It indicated that a third-party breach of derivative works cannot do so if it is based on all or any portion of such a derivative work, which is inconsistent with the position IBM takes here today.

In 1996 they had an amendment X which had an illustration attached at 3.7, and that illustration was that if company A, a sublicensee, is a general computer system manufacturing system, and it said IBM may not distribute source copies to that company for purposes of making modifications to adapt the sublicensed product as a general operating system for the company's general computer hardware system. But yet they maintain here that they can open source that to the world to come up with competing operating systems. Their position simply cannot be squared with the plain language of the agreement.

38

There would be no purpose for the language we saw in 2.01 talking about the derivative works if all that was protected was the original course code. You wouldn't need that. The source code would then have its own protection. The only sense that makes is if the derivative works, the resulting work is protected.

Now, Your Honor, there is no factual dispute that AIX and DYNIX PTX are derivative works within the meaning of Section 2.01. At tab 10 in the binder we provide the expert testimony on that point, and admissions from both Sequent and IBM people that these are derivative operating systems based on System V. We would submit that the combination of that plain language, and the expert testimony together with the plain meaning of the Sequent agreement, is why we're entitled to partial summary judgment on Sequent on the ground, one, that Section 2.01 means what it says, that the derivative work has to protect like the original product and, two, that DYNIX is in fact a derivative of System V, because that is not disputed at either the expert level or the level of the admissions which we have put into the record and excerpted at tab 10.

Your Honor, for that reason we don't believe it is necessary to turn to parol evidence in this case. There is an integration clause that would even exclude doing so. However, the parol evidence is sharply disputed and it would require a trial. If one turns to tab 13, and I have already touched on

39

this subject with respect to certain declarations from the very witnesses who IBM relies upon, but beyond that there are over 20 witnesses, 20 witnesses who are involved with AT&T and Novell and Santa Cruz in the negotiations, in the administration, in the enforcement of these agreements, who contradict IBM's contract interpretation.

I don't have time to go through here now all of this testimony that is in the record and excerpted at tab 14, but it is directly on point.

Mitzi Bond, for example --

THE COURT: You said 14. Do you mean 13?

MR. SINGER: It is 13 that I am referring to right now, Your Honor.

THE COURT: Okay.

MR. SINGER: Behind tab 13, which is the chart, there are excerpts from each of this witnesses and declarations or depositions.

Just to touch on one, Mr. Guffey, who during the relevant time period, '80 to '85, was the head of the software services division which included UNIX, and he said I believe that the members of my division and other AT&T employees involved in licensing UNIX share the foregoing understanding because it was a common subject in training and discussion. The licensee was obligated to keep all parts of those modifications and derivative works confidential, including the

40

methods and concepts embodied in those modification and derivative works, just as the licensee was required to keep all parts of the UNIX software product confidential.

If one turns to Burn Levine, who was an attorney at the time involved with these agreements at AT&T, and he said nothing in the agreement reduced AT&T's protection for UNIX software, and that it was not limited to the disclosure of just little source code.

Beyond those witnesses, let's turn to some of the very witnesses who IBM relies upon. If we turn to tab 16, we have again Mr. Kistenberg who says exactly the opposite of what IBM relies upon in connection with the testimony that you have heard earlier from IBM. If we turn to tab 17 you have Mike DeFazio who was there and he supports IBM's position, but he concedes that he really relied on Martin Pfeffer who was the general counsel, who had primary responsibilities for supervising the drafting of these agreement, and Mr. Pfeffer supports our position.

You have at tab 18 Otis Wilson, and Mr. Wilson, as we have seen, contradicted his position that he takes now, closer to the time in the USL case, where he gave sworn testimony and where he said anything that created an exposure to the licensed software or was based on or was a derivative work had to be treated in that way.

We have seen with respect to David Frasure, tab 19,

41

another one of the individuals up on the scene, that he specifically said in the USL versus Berkeley case that the UNIX source code had been instrumental in its development, and that that is why it cannot be freely distributed by Berkeley.

Mr. Vuksanovich, the other gentleman up there at tab 20, says that if there is a single line of UNIX source code in a modification or derivative work, then that modification or derivative work has to be treated like the software product that has been licensed under the agreement. He agreed with that. So clearly this is a case where both the subsequent UNIX agreement and the testimony requires a trial.

One final point on that, which is at tab 23. That is that there are documents from IBM that are also inconsistent with what IBM maintains today. This was an analysis of these very agreements done by IBM. We put the whole document behind tab 23 for the Court to read. This analysis was done in 1989, and specifically concluded that all copies of the derivative works of UNIX source code must be treated in exactly the same manner as the UNIX source code as received from AT&T. Nowhere in this document is there any mention that they could extract parts of the modifications and the derivatives and do with them what they would. That is something that is a position taken in this litigation.

Your Honor, they talk about the parade of horribles that will occur if this interpretation is afforded. That is

42

simply not the case. There are only two derivative works at issue, AIX and DYNIX. IBM can use that technology to market product, and they can come up with their own independent operating system, and their employees are not restricted in their work, they simply can't open source the result of that work if it is based on System V.

So the consequences of agreeing with our interpretation is simply that our contract is upheld, we receive damages, we receive an injunction against further violations to prevent this head start from being dissipated and being given away to Linux. No other company that we're aware of has done what IBM had tried to do, even though there have been many licensees. None of them have come in here and open sourced their source code in the manner that IBM has to enhance Linux.

Your Honor, we have put in the book as well the various legal authorities as to why this interpretation is reasonable. There has been a dispute of experts between Mr. Willick, who they submit, and Mr. Popono, who is a professor at Harvard Business School, and at trial I think that will be interesting testimony, and certainly not something that the Court needs to deal with on summary judgment.

Now, with respect to estoppel, at tab 27 we include the fact that estoppel is an issue of fact that we would submit is inherently unsuited for summary judgment. The requirements

43

are that there be some concealment of a material fact by us, and some lack of knowledge regarding these contracts by IBM. That simply is not feasible in this situation.

First of all, you have an integration clause which said that the parties can only change the agreement by a signed agreement in writing. IBM knows that even if it were true that some people told them, well, disregard this and don't believe that, they knew that the only thing they could rely on was a change in writing, and that is in the agreement, and that we submit is why IBM went to the trouble of getting those side letters in 1985 and 1987 and negotiating amendment X in 1996. That is how these things are changed.

The case law we cite at tab 28 in our brief say that that is an integral matter for an estoppel claim. Certainly there can be no concealment of material facts when what allegedly is being concealed is the plain language of the contract. Now, beyond that there is no competent evidence that establishes that AT&T or any of its successors told IBM or Sequent that they could simply do as they wished.

That library shelf of books which was brought in here, there is nothing shown that the specific technology that is at issue in this case is disclosed there. Many of those, we would submit all of those, have copyright notices, so to the extent there is information it is still protected by our copyrights. They talk generally about System V but don't

44

disclose the information here. They are welcome to roll in that cart we submit to the jury in this case and make that argument, but that is certainly not a part of a summary judgment proceeding.

Now, as to the witnesses who they claim say that AT&T gave them these assurances to do as they wished, that is sharply disputed. Your Honor, turn to tab 30, if you would, in the binder. You see there over a dozen witnesses on those two pages that flatly dispute that they or anyone they worked with at AT&T or Novell or Santa Cruz ever told licensees that they could disregard the plain language of those agreements. At a minimum, even if that is a legally tenable argument for estoppel, it is subject to a factual dispute.

Your Honor, specifically, Doug Michels testified, and this is at tab 31, that when there was a concern at Santa Cruz that IBM announced support for Linux, he went to talk to IBM and they assured Santa Cruz, which is now SCO, that they would not commercially harden Linux, and that they would not substantially encroach on Santa Cruz's core market, so that there was nothing here we had to worry about. Now they claim that is estoppel and waiver. I submit that there cannot be any evidence of reliance because IBM gave those rights no consideration at all.

First of all, let's look at the fact, and they talk about this in connection with waiver, with the distribution of

45

Linux, but who was distributing what when? When did IBM make their Linux decision? IBM decided to pursue Linux and to distribute technology beginning in '98, and they started the Linux Technology Center in '99, and the first disclosures were at the end of '99 and into 2000. At that time period, up to May of 2001, these copyrights and contracts were owned by Santa Cruz which was not in the Linux business. Caldera, which was in the Linux business, did not own them. Caldera acquired them afterwards, after they had already made this decision and gone forward.

But beyond that, Mr. Frye, who is the very head of the IBM Linux Technology Center, specifically denied that he ever gave any consideration to any of these issues with SCO. That is at tab 32.

Your Honor, in addition, at tab 34 you have Mr. Sandve's e-mail from IBM, and Mr. Sandve specifically states, because he was asked by one of his superiors, why can't we let you look at the AIX source code? He says it was because of the 5.3 source code license, and that it would take 50 to $80 million to buy it from SCO even if SCO would deal with it right now. All these assertions of estoppel and waiver are at a minimum factual issues which have to be decided at trial.

Very briefly with respect to the issue of Novell waiving our rights.

By the way, one last point on IBM's suggestion that

46

Mr. Love waived his rights and that SCO made a knowing waiver. Mr. Love, who IBM did not disclose to the Court, is a paid IBM consultant, he contradicted that in a subsequent declaration, saying that there was never an investigation while he was the C.E.O. of these property rights. We have submitted declarations, and this is at tab 40, from other members of the board of directors and management directly contradicting his testimony.

So all of these issues, and the United Linux issue they raise, and it is treated in the copyright argument which you'll hear later in the week, and they incorporated that in their briefs and we incorporated it in our briefs, and that will be later this week, and so will the issues concerning the other waiver argument they made about our distribution of Linux. At a minimum there are factual issues that preclude an estoppel and waiver argument.

I would like to talk about the issue of Novell's purported waiver of SCO's rights. This interpretation would make a mockery, Your Honor, of the very asset purchase agreement that was signed by the parties, where it gave the whole business to SCO. The Court will remember the argument about this in the Novell case a few weeks ago, and yet they maintain that they can come in here and they can prevent SCO from enforcing its intellectual property rights by a contorted interpretation of Section 4.16. First of all, this issue is

47

resolved in the Novell trial.

Secondly, their position is simply flat wrong. If one turns to tab 43, and Your Honor has seen this section, but 4.16 defines the SVRX licenses for purposes of this agreement under item six of the schedule. That schedule does not list these IBM agreements in item six. It lists them as a different item, item three. That is where the software and sublicensing agreements at issue here are listed. At a minimum, there is an ambiguity in that issue.

At tab 46 we present for the Court testimony from the lead negotiators and the businesspeople on both sides of the issue who agree that Novell had no ability to waive IP rights with respect to the source code license at issue here. That was limited to protect their binary royalties, which were bought out in 1996, and had no continuing role. Both Chatlook and Wilt, who were the negotiators for Novell and Santa Cruz respectfully, and the other individuals on these pages all agree, and more testimony is being developed every week in the Novell case supporting this position, that they had no ability to waive these rights.

That was made clear as well in an amendment in writing to the APA. It is amendment number two, which is at tab 44. It says that Novell has no right to increase any SVRX licensee's rights to source code, and may not prevent SCO from exercising its rights with respect to SVRX source code in

48

accordance with the agreement.

So their position, which may or may not be related to the fact that IBM paid Novell about $50 million right at about the time of this waiver, is not supported by the plain language. It is not supported by the extrinsic evidence. It is contrary to the entire purpose of the agreement. In any event, it will be resolved in the Novell trial.

Briefly on the statute of limitations, their final argument, that breach arose from open sourcing to Linux. That is not the same thing as a patent which restricts the use to IBM. Only IBM could use it. That would be consistent with our agreement because IBM has the right to use it. Open sourcing, which is what occurred during the period when the statute of limitations had not run, it occurred only a few years ago, that is what gives rise to our cause of action. IBM's patent, therefore, does not begin the statute of limitations. Their cases do not support that proposition.

Your Honor, at tab 49 we briefly point out what the Court is well aware of, the fact that you cannot raise new legal arguments in a reply brief for the first time. All of these arguments that we heard at the end of Mr. Marriott's presentation about this specific technology are not the four arguments listed in their reply brief. We are not charged with anticipating everything they might draw from all of the factual material.

49

So in addition to the fact that that is new, we have put in testimony related to each of these technologies at tabs 50, 51 -- these are all in the record -- 52, which indicates that there is a factual dispute at a minimum with respect to these technologies issues, rather than the legal issues which they base their initial legal brief on, but now presumedly they have found some reason to want to raise new issues in reply.

Your Honor, we have established, as I mentioned earlier, that our cross motion for summary judgment should be granted. Their only real argument with respect to that is that you should not issue partial summary judgment. They note that the courts are in disarray on that issue. We believe that the better authority is that you can and should issue such a partial summary judgment. Alternatively, under Rule 56(d) you're entitled, in fact, directed to make such findings if the facts are not in dispute.

I would like to save the balance of my time for my rebuttal on the cross motion.

THE COURT: Go ahead, Mr. Marriott.

MR. MARRIOTT: Thank you, Your Honor.

First, with respect to the plain language, the provision on which SCO relies, Your Honor, says simply that resulting materials are to be treated hereunder like software product. Whatever precisely that means, Your Honor, the categories of information that IBM is supposed here to have

50

improperly disclosed are not resulting materials. And, therefore, the limitation, whatever precisely it is, on materials covered by Section 2.01 is inapplicable.

Again, we're talking -- Your Honor, if I may approach with a chart?

THE COURT: You may.

MR. MARRIOTT: If this is the one item, Your Honor, that you licensed as your product as I described previously from SCO, and that is subject to the terms of their agreement, we didn't disclose that. Nor, Your Honor, did we disclose the derivative work. The allegation is that IBM took pieces out of its own original works and used them as it wished. Some of them got put in different products. That, they say, is a violation of the agreement.

That disclosure reveals nothing about this, and that disclosure does not compromise the entirety of the product so that there would be some reason to be concerned about that. Taking this out and putting it here, Your Honor, does not mean that that might be a derivative work, or it might not, and you have to look at that particular thing. Is that a derivative work? Under SCO's theory you might have an obligation to limit what you do. If what you take out is not a derivative work, Your Honor, then it is not subject to the provisions of 2.01 of the agreement.

2.01, if it means anything, Your Honor, it relates to

51

precise terms, resulting materials. The things which IBM has alleged to have taken out, that are not modifications or derivative works or resulting material, can't possibly have been distributed in violation of Section 2.01 of the agreement.

Parol evidence, Your Honor. Mr. Singer points to, as I suggested he would, a long list of individuals who he says have offered testimony to support SCO's case. He pointed to the testimony from Mr. Wilson who he says, for example, has offered contradictory testimony in a case in 1991. Your Honor, time won't allow for a line by line recitation of that. Let me, if I may approach, with one example.

In 1992 in the litigation to which Mr. Singer refers and to which he claims Mr. Wilson offered inconsistent testimony, Mr. Wilson offered testimony that is perfectly consistent with the testimony he and the other individuals who negotiated the agreement in this case have given. If you look, Your Honor, at page 47, he says, quote, we did not -- we, AT&T, did not want to have any rights or ownership to anything they created. And yet SCO says that while we, IBM, may own our stuff, they control it. Mr. Wilson said that in 1992 in the litigation in which Mr. Singer claims he offered inconsistent testimony.

If you turn to page 76 from that same case, Mr. Wilson said, quote, the intent is what I have stated many times earlier. In other words, the intent is such that we protect

52

our intellectual property, and assert no rights in the licensee's intellectual property. Yet SCO asserts the right forever to control IBM works original to it, the disclosure of which couldn't possibly disclose anything owned by SCO.

Your Honor, the parol evidence offered by SCO is no impediment to the entry of summary judgment, because the individuals upon which it relies didn't negotiate and execute the agreement. Your Honor can look at those and decide if there is any contradiction. That testimony is overwhelmingly in favor of IBM's construction.

To decline summary judgment here, Your Honor, is to basically say that a reasonable jury could project the testimony of the individuals who negotiated and executed the agreements on behalf of IBM, and accept in its stead the subjective understanding of individuals who might have been employed at AT&T along with hundreds of thousands of others, but who never communicated their subjective intent to IBM as part of any negotiations. Under New York law, Your Honor, that testimony is not capable of altering the plain language of the agreement of impeding the entry of summary judgment.

The reasonableness of SCO's claim -- Your Honor, if that operating system, if the DYNIX operating system, if the AIX operating system were a General Motors car, and if the chassis of that car, Your Honor, were licensed under the agreements as SCO interprets them, that SCO, Your Honor, would

53

not only control the car in its entirety, but IBM would be unable to take the dice off of the mirror of the car and do what it wanted with those. It would be unable to take the radio manufactured by Sony and put it in another car. It couldn't take glass manufactured by PPG and put it in another car. The design of the car could never be used in connection with any other vehicle.

Estoppel. The cases are clear, Your Honor, that estoppel can be entered at the summary judgment phase of the case. AT&T and its successors, again, as we laid out, said over the course of decades that IBM could do and other licensees could do as they wished. The mere fact that SCO has produced a number of witnesses who say they never said that, and they never heard that, does not make incompetent the sworn testimony of the numerous people laid out in our book.

The fact that these books, Your Honor, some of them may have a copyright notice on them does not in any way mean that there wasn't the disclosures of the supposed secrets which SCO claims it seeks to protect. The integration clause to which Mr. Singer refers, Your Honor, has no bearing on testimony as to statements made over the course of decades that followed the execution of the agreement.

Section 4.16B. Mr. Singer says that IBM's interpretation of 4.16B makes a mockery of the agreement. Your Honor, under that agreement Novell retained the right to 95

54

percent of the royalties with respect to the licenses. It is hardly a surprise that Novell would retain the right to waive or to supplement or to change conduct that SCO might engage in that could compromise Novell's interests. Mr. Singer refers to amendment number two and suggests that that somehow is inconsistent. Amendment number two, Your Honor, relates to prospective buyouts. There is no prospective buyout at issue here. There is no buyout at issue here. IBM's rights to continue to distribute AIX were already bought out.

United Linux. Mr. Singer didn't to my knowledge address United Linux, and it was raised in our papers, and we'll discuss it on the 7th, Your Honor, in a different context independent of what it means in the context of IBM's claim for a declaration of non-infringement, it precludes SCO's claims here. A general public license on that is similarly deferred. It has independent meaning here to which SCO has not responded.

As to the four categories of alleged or misused information, Mr. Singer suggested that that is somehow new in the reply papers, Your Honor. I respectfully refer the Court to the statement of undisputed facts in our opening papers. It is not new. It was laid out and supported there by undisputed testimony. If it is new, Your Honor, it is curious that Mr. Singer's binder would have included a set of materials that supposedly refute the testimony there.

These specific items at issue here, about which SCO

55

seems to skirt, Your Honor, are items which when examined preclude SCO's claims. Mr. Singer refers to the RCU contribution and he says it is not barred by the statute of limitations because the disclosure was in a patent application, and a patent application is subject to certain protections. Your Honor, this case is not about misusing a patent, it is about disclosure. Whether or not a person could practice an invention, which is set out in a patent, is irrelevant to whether the information which supposedly is secret, and supposedly had to be kept confidential, was out there sufficient to start the running of the statute of limitations. The claim as to RCU is barred.

Mr. Singer said nothing, Your Honor, as a consequence about the Linux technology project contribution, except to suggest that experts had dealt with it. The expert testimony on which Mr. Singer relies, Your Honor, is not only mistaken, but it is testimony that Magistrate Judge Wells precluded SCO from proceeding as to because it was not disclosed with particularity in the final disclosures.

The e-mail on which SCO's expert, Mr. Rochkind, relies in saying that the Linux LTP contribution was actually from DYNIX is talking about a different set of LTP code than that which is at issue in this case. Mr. Rochkind's testimony is unsupportive of SCO's position.

Thank you, Your Honor.

56

THE COURT: Thank you.

Mr. Singer.

MR. SINGER: Your Honor, I suppose the short answer is that if DYNIX RCU does not matter, and if JFS does not matter and if they are like the dice on a car, then let's take them out. Let them go into Linux and take them out. Let them go into DYNIX and AIX and take them out and see what happens to those systems then. These are interwoven, as our experts have indicated, with the very operations of those systems, and those systems as a whole are derivative of and they would not exist but for System V.

Beyond that, we have shown direct links with respect to JFS to System V. I quoted Mr. Ivie's testimony and it is in record. I didn't hear Mr. Marriott say anything about it. It is interesting that JFS was said by IBM people at the time, and this is in the record, to be the most important contribution to Linux.

With respect to RCU, at tab 51 you have testimony from Mr. Rochkind, an acknowledged UNIX expert, showing that it is in Linux and that the DYNIX code is a derivative of System V, and that RCU is interwoven. So you can't simply take it out. By the way, he also noted in his report that Linus Torvell wrote that RCU was, quote, fundamental in summarizing Linux 2.543 which was the first version to have that contribution from IBM. He said the most fundamental stuff is

57

probably RCU and a program called Low Profile.

These are not dice hanging down on the dashboard, these are integral contributions that are integral to DYNIX and AIX, and they are derivative of our protected technology, and now they have been wrongly disclosed.

The second point which has been made is that the plain language does not support this, but they can't get around the fact that the plain language in both that agreement and the subsequent agreement never provide anywhere an invitation to IBM to disclose materials that they add to these derivative products that is not in someway based on these derivative products. What we have are snippets of testimony from these witnesses, that I submit really depend on what they mean when they are being asked about original software products or IBM's derivative products, and when you focus on the precise issue, as the testimony we presented here throughout, indicates that they say that it was the intent of AT&T to protect that. That testimony, if you need to get to extrinsic evidence, is fully admissible and at tab 14.

We submit that the cases from the Second Circuit, saying if there is ambiguity, the Court should accept any available extrinsic evidence, and that when you are dealing with form agreements and how they are interpreted generally is relevant, and these are form agreements, and that the course of conduct under them is a strong indication of intent, aside from

58

the fact that virtually everyone up there who they call and admit are involved persons have given contradictory testimony.

With respect to 4.16B -- well, first, with respect to the issue of estoppel, the integration clause does more, Your Honor, than simply say that parol evidence should be considered. This integration clause says any changes to the agreement must be made in writing. IBM knew that and they got changes in writing, they just didn't give them the rights, which IBM says that they have now, to do anything they wish with those products, and that is why you have not heard very much in this argument about those side letters or amendment X.

What that integration clause means is IBM cannot come into court years later and say, oh, we relied on the fact that someone else distributed a book that had a little bit of UNIX in it or something and, therefore, we have the right to disregard our contract. Or that we heard a licensee was told by someone that you could do it with a derivative product. That is directly in the face of the integration clause that says if you want a change, you get it signed in writing. Beyond that, there is a tremendous dispute where we have a dozen witnesses who were there that said those things were never said to licensees.

With respect to the Novell waiver, I didn't hear anything about the fact that 4.16B's definition of the SVRX license refers only to item six and that the IBM agreements in

59

issue here are in item three and are expressly not covered. That was made clear in amendment two. He says, well, amendment two is dealing with future buyouts. Amendment two was entered into at the same time in 1996 and IBM bought out its remaining royalty obligations. After that there was absolutely no interest Novell had with respect to how IBM acted under those agreements.

Last issue. There are 281 fact statements they make, 40,000 pages of exhibits, and we are not charged with responding to every legal argument in our opposition that might have been made from those. We entered with four legal arguments in the initial brief. The fact that they have not extracted three new arguments about RCU, about these tests and others, are not properly considered on this motion for summary judgment and they were not a part of their initial papers. And because we submitted as a part of the record all of our expert reports, it happens that there is a part of the expert reports that contradict those, and now they are wanting the Court on summary judgment to make rulings about the weight to be given to those expert reports, and we think that is purely inappropriate.

Your Honor, I heard virtually nothing about our cross motion with respect to DYNIX, and there is no dispute that DYNIX is, in fact, a derivative product of System V. If the Court agrees with us and the plain language of 2.01, then that

60

limited but appropriate partial summary judgment should be entered.

Thank you very much.

THE COURT: Thank you. Thanks to you both.

I'll take these contract motions under advisement. I have a jury I have to deal with. Realistically I think we're looking at about 4:30 to continue with our motion.

We'll be in recess on this matter.

(Recess)

THE COURT: Welcome back, everyone. Sorry about the delay.

We'll now take up IBM's motion for summary judgment on SCO's copyright claim.

Are you arguing this, Mr. Marriott?

MR. MARRIOTT: I am, Your Honor.

THE COURT: You asked for 30 minutes each on this, right?

MR. MARRIOTT: That sounds familiar, Judge.

THE COURT: Go ahead.

MR. MARRIOTT: On the grounds that IBM breached its contractual obligations, and this is IBM, Your Honor, not Sequent, SCO purported to terminate IBM's license to continue to use its AIX product, to distribute that product, and it demanded that IBM shut down its AIX business, which over the course of decades it has invested hundreds of millions of

61

dollars in. We declined to do that and SCO amended its complaint and asserted copyright infringement. That is the claim that is at issue with this motion.

There are five reasons why summary judgment should be entered in favor of IBM on this motion. One, SCO can't identify and prove unauthorized copying by IBM. Two, SCO can't establish a predicate breach of contract. Three, SCO cannot terminate and did not properly terminate IBM's license. Four, SCO can't prove that it owns the allegedly infringed copyrights. Five, SCO has misused those alleged copyrights. With Your Honor's permission I want to focus on just the first three of those. By the parties in agreement and by order of the Court, the remaining two, four and five, will be addressed at the hearing on May 7th.

With that, Your Honor, point one, SCO can't show unauthorized copying by IBM. This is summarized at tab two of the book which I would like to approach and provide Your Honor.

THE COURT: Sure.

MR. MARRIOTT: We have a copy for counsel.

As you know, Your Honor, IBM repeatedly asked over the course of this litigation for SCO specifically to identify the allegedly misused information, and the Court repeatedly ordered SCO to do that as we show at tab 3 of the book. In a December 2003 order, Magistrate Judge Wells ordered SCO to identify and state with specificity, and this is at tab three,

62

the source codes that form the basis of their action against IBM.

Magistrate Judge Wells further ordered that SCO provide detailed answers to IBM's interrogatories as set out and requested in the interrogatories. She said, for example, that SCO was to identify and state -- SCO was to respond fully and in detail as stated in IBM's first set. Interrogatory one said identify with specificity all of the confidential or priority information that plaintiff alleges or contends IBM misappropriated or misused.

Interrogatory four, likewise, Judge, asked that SCO describe in detail the date of any alleged misuse or misappropriation, and the specific manner in which IBM is alleged to have engaged in the misuse or misappropriation. Magistrate Judge Wells reiterated that order in March of 2004, and then Your Honor set interim and final deadlines for final disclosures, and at this point SCO was required finally to identify with specificity the allegedly misused information.

THE COURT: And you say they have not done that?

MR. MARRIOTT: They have not done that, Your Honor.

If you turn in the book to tab four, you will see that in connection with our summary judgment papers, Your Honor, we set out in paragraph 69 the following. Dispute the Court's orders, SCO has never described by version final line of coding any material allegedly infringed by IBM's

63

post-termination AIX and DYNIX activity. Moreover, SCO has declined to provide full and detailed responses to IBM's interrogatories directed at SCO's allegations of unauthorized copying.

In response, Your Honor, SCO does not dispute that. It says instead, simply, that, in effect, that it was not required to do that. The Court has since made it abundantly clear in a series of orders that, in fact, SCO was required to do that, Your Honor, and it has still never done that. For that reason, alone summary judgment on this claim should be entered in IBM's favor. The orders of the Court were clear that neither party could proceed with respect to any material that wasn't identified as directed by the Court, and SCO has not done that and the claim, Your Honor, should go for that reason alone.

Point two, SCO can't establish a predicate breach of contract, as we summarize at tab seven of our book. SCO's copyright claim here, Your Honor, depends on whether it can show that IBM breached one of its licensing agreements with AT&T. It is on that basis that SCO purports to terminate IBM's license. If there is no breach of contract, no predicate breach, then the copyright claim fails as a matter of law. The problems with SCO's contract claims have been discussed at length in the papers and in the argument, and I don't intend to repeat all of those here, but --

64

THE COURT: Good.

MR. MARRIOTT: Your Honor will recall that I said by way of footnote in connection with the past argument that I would dwell in this argument on the JFS language.

THE COURT: I do remember that.

MR. MARRIOTT: The one of the four not addressed there.

Your Honor, SCO's contract claim involves, as I have previously indicated, four contracts. Two of them are for IBM and two for Sequent. The Sequent contracts are irrelevant to this predicate breach of contract. The Sequent contracts were about the distribution of DYNIX, not AIX. There is absolutely no evidence that IBM has continued to distribute AIX. This is all about the IBM agreements with AT&T and not the Sequent agreements.

As I previously indicated, SCO has offered no evidence and Mr. Singer did not point, Your Honor, to a shred of evidence that IBM breached the sublicensing agreement. SCO's entire case depends upon the IBM software agreement as it relates to this purported cause of action for copyright infringement. In its final disclosures, Your Honor, the only contribution identified as having been made to Linux in violation of SCO's rights is the JFS contribution. This claim turns entirely on that.

So with that, Your Honor, let me suggest that there

65

are at least two reasons, and one of them has a lot of sub reasons, why SCO's claims with respect to JFS fail as a matter of law. The first of those reasons, Your Honor, is that SCO's allegations of breach with respect to JFS simply lack merit, and the second is that the alleged breach, even if it were a breach, is immaterial as a matter of law and, therefore, can't substantiate the kind of breach necessary to establish a breach of contract.

Let's take the first of those, SCO's allegations with respect to JFS lack merit. That is true, Your Honor, for at least six reasons, and I am going to quickly run through those.

THE COURT: They are at tab ten, right?

MR. MARRIOTT: They are at tab ten, Your Honor. Apparently the binder is not as difficult to follow as --

THE COURT: Not if I can figure it out. That is right.

MR. MARRIOTT: The first reason, Your Honor, the JFS contribution did not come from AIX, it came from IBM's OS2 operating system. SCO's theory depends on the proposition that AIX is a derivative work of AT&T's UNIX System V software, for which we don't believe there is adequate evidence in the record, despite Mr. Singer's assertions to the contrary, but assume that it is, Your Honor, the JFS contribution came from IBM's OS2 operating system and it did not come from IBM's AIX operating system.

66

The undisputed evidence shows that. If you look at tab 11, Your Honor, you will see evidence to this effect. The individual identified by SCO as having made these supposedly improper contributions have offered their testimony, the people in the best position to know that it came from OS2 and not from IBM's AIX operating system. SCO's claim that it came from IBM's AIX operating system, Your Honor, is mistaken for the reasons which we set out at tab 13 of our book. It relies principally if not entirely upon the testimony of SCO's Dr. Ivie, and that analysis, Your Honor, relies upon evidence that was required to be put in SCO's final disclosures, and that it didn't put in those final disclosures, and that Magistrate Judge Wells struck at a hearing late last year, an issue in which SCO has taken an appeal.

The only evidence they purport even to offer, Your Honor, that could competently demonstrate that and they argue is that of Mr. Ivie. Mr. Ivie's testimony was untimely provided and it can't support the motion. In any event, Your Honor, Mr. Ivie's testimony is not competent evidence that JFS came from the AIX operating system as opposed to the OS2 operating system. Dr. Ivie says that there is a similarity between the JFS contribution and the JFS that is in AIX. Well, that is not a surprising similarity, Your Honor, because the JFS contribution in AIX came from IBM's OS2 operating system, so the similarity on which he relies for these opinions is not

67

a similarity that has any probative effect. That is point one, Your Honor, as to why the JFS contribution allegations fail.

Point two. For this I refer Your Honor to tab 14 of our book. SCO's claim fails unless the JFS contribution was resulting material. Their entire theory depends upon, we think, a distorted reading of Section 2.01 of the agreement. But in any case, it depends upon the JFS contribution being resulting material. As I said in the prior argument, it is not. The undisputed evidence demonstrates that there is no UNIX System V method, code or concept in that JFS contribution, and that it was created independent of UNIX System V, and the people who created it, Your Honor, the people who supposedly made the contributions have offered testimony to that effect.

The only thing on which SCO relies in this connection is the testimony of Mr. Ivie, which testimony again, Your Honor, Magistrate Judge Wells could not properly be relied upon because it was not properly disclosed in the final disclosures.

Point three. As I said in the last argument, Your Honor, IBM owns the JFS contribution. There is not any dispute about that. IBM owns it. IBM has copyrights in that distribution, and SCO concedes that, as you see, at page 15 of the book. It further concedes in its opposition papers that, quote, under the Copyright Act, copyright ownership consists of exclusive rights to, among other things, reproduce, prepare derivative works, and distribute a work. IBM owns the

68

contributions and the notion that SCO can control it effectively guts IBM's rights of ownership. If IBM's rights of ownership mean anything, Your Honor, it means that IBM has the right to do what a copyright holder under the copyright laws has the right to do.

Mr. McBride, the CEO of the SCO group, Your Honor, testified in his deposition in a way that is completely contrary to the proposition that SCO can control anything and everything that is in the AIX operating system. He said the following, quote, I am sure there are things inside of AIX that were not derived from System V or from one of our contractual arrangements, that they would be free to do whatever they wanted. That appears, Your Honor, in Mr. McBride's deposition which is IBM Exhibit 330, page 231, lines 18 through 23.

Fourth point, Your Honor. Novell waived SCO's right to terminate IBM's license pursuant to 4.16B of the asset purchase agreement. We discussed this briefly in the last argument. As we discussed there, the APA expressly gave Novell the right to waive alleged breaches with respect to SVRX licenses. The only issue, Your Honor, is whether the licenses in question are SVRX licenses. There is absolutely no question that they are. Mr. Singer made reference to a schedule in which he says there is some lack of reference to the IBM agreements being SVRX licenses. Let me walk you through that, Your Honor, if I may.

69

If you don't have the book from the last argument, it is at tab 40 of that book, and I have a copy to hand up.

THE COURT: I might have it.

MR. MARRIOTT: At tab 40 of that book, Your Honor, we walked through why it is that the licenses in question here are SVRX licenses. Let's take that given Mr. Singer's focus on that issue. The first sentence of the asset purchase agreement, 4.16A, states that SVRX licenses are those licenses listed in detail under item six of Schedule 1.A hereof. Item 1.A of the asset purchase agreement provides a list of the SVRX licenses that related to various UNIX System V software releases, including System V releases 2.0 and so on. As well as, quote, all prior UNIX system releases and versions preceding UNIX System V release 2.0.

The supplements, Your Honor, to the IBM Sequent agreements which are at issue in this motion identify the licensed software product as consisting of various UNIX System V releases. Again, as I said in the last hearing, Mr. McBride in a letter to the C.E.O. of Novell conceded that the licenses at issue here are SVRX licenses. In SCO's own opposition papers in connection with this case acknowledge that the licenses at issue are SVRX licenses. There is, we respectfully submit, no genuine issue on that question, and as a result the only real argument that SCO has made in opposition with respect to 4.16B falls flat.

70

Fifth point, Your Honor, with respect to the JRS contribution. SCO assigned its rights to the JFS contribution to United Linux, as is illustrated at tab 19 of the book. We discussed this briefly in connection with the prior motion hearing, Your Honor. But the flaws in SCO's case are particularly pronounced as they relate to the JFS contribution.

Again, as a member of the United Linux initiative, SCO assigned all rights, all intellectual property rights that it had with the exception of those specifically carved out to the United Linux LLC. SCO's product -- SCO's Linux IV included the JFS contribution. The JFS contribution was not on the list of exclusions and, in fact, Your Honor, the JDC itself specifically refers to JFS as a part of the joint development product. And then SCO in its product announcement for its United Linux product touted the product as including the JFS contribution. Any claim to the JFS contribution, Your Honor, is gone pursuant to SCO's assignment of rights under the United Linux agreement.

Sixth point. SCO licensed the JFS contribution and the GPL, the General Public License. This is the same basic point as before. Again, the contribution was included in their product, and the GPL is clear as to what it means, the license in the GPL, and any claim as to that is gone.

Now, I said there were two points with respect to -- two general points with respect to JFS. The first is that

71

SCO's allegations related to JFS lack merit. That is the sixth point I just listed. There is an additional point, Your Honor. That is that even if there were merit to SCO's allegations of breach relating to the JFS contributions, the alleged breach is material. Assuming SCO has the right to terminate, as we'll talk about in connection with my next point, it does not, but assuming that it did have that right, the right to terminate applies only with respect to breaches that are material breaches of the agreement.

The IBM side letter, which Mr. Singer suggests in the last argument IBM had ignored, expressly says that breaches can be used as a basis for termination only if they are material breaches. In the case law, Your Honor, which we lay out at tab 31 of the book, indicates that breaches that are sufficient to permit termination must be material breaches. A material breach is a breach that frustrates the core of the contract. The cases describe it that it goes to the very purpose or the root of the agreement.

The JFS contribution here, Your Honor, could not possibly have gone to the root of an agreement between IBM and AT&T in 1985 that concerned the protection of AT&T's UNIX System V software. The JFS contribution, Your Honor, again, which is owned by IBM and copyrighted by IBM, represents 01 percent of the Linux Kernel. There are as we show at tab 40 -- at tab 34 of this book, a large number of file systems, as much

72

as Mr. Singer suggests in the last argument that the JFS is the next greatest thing to sliced bread, that there were 40 some file systems in the Linux operating system, Your Honor. The JFS contribution could not possibly be considered a material breach of the contract, especially when IBM owns it and when there is no UNIX System V code in it, and when the protections of the software agreement as between IBM and AT&T, if they meant anything, were about ultimately protecting AT&T's UNIX System V source code. They have conceded there is no UNIX System V -- there is no trade secret in UNIX System V. And yet the contribution of IBM's own original work could go to the root of an agreement that was about protecting not IBM's original works, but the UNIX System V software? The arguments, Your Honor, which SCO makes in this regard are dealt with in our papers, and they are dealt with at tab 35 of the book if Your Honor wishes to look at them there.

The third point and final point that I wish to make this afternoon, is that SCO cannot establish a predicate -- rather cannot establish that it properly terminated IBM's license. That is true for two reasons. The first reason, Your Honor, is that IBM has pursuant to an amendment to its original agreement with AT&T a perpetual and irrevocable license. That is point one. Point two, Your Honor, is that even if under the language of the original agreement the license could be terminated, SCO failed to abide by the requirements for

73

termination. I will talk Your Honor through those.

Let me take now, if I may, each of those in turn. First the irrevocable and perpetual license. Referring Your Honor to tab 37 of the book, the plain language of amendment X granted IBM a perpetual and irrevocable license. Amendment X says, and I quote, IBM will have the irrevocable, fully paid up perpetual right to exercise all of its rights under the agreement. The meaning of the term irrevocable and perpetual is no mystery. They are clear and they are unambiguous, and some of the definitions of those terms, Your Honor, appear at tab 38 and tab 39 of the book, from a variety of dictionaries.

For example, irrevocable is defined to mean impossible to retract or revoke, that which cannot be abrogated, annulled, or withdrawn, not revocable, irreversible, final, unmodifiable, indistinguishable, unalterable, immovable.

THE COURT: I see one that says lasting for eternity. Are you claiming that here?

MR. MARRIOTT: I like that idea, Your Honor, lasting for eternity. A lasting irrevocable license that lasts for eternity, in a sense, Your Honor.

Similarly, with respect to perpetual, which is actually the definition for lasting for eternity, it is also defined, Your Honor, as continuous, without interruption, everlasting, eternal, lasting or destined to last forever.

Accordingly, Your Honor, on the plain language of the

74

agreement IBM has a perpetual and irrevocable license. It does not have a terminable license as SCO suggests, as is required for it to have terminated IBM's license, which is the predicate to this claim of breach of contract.

Now, the second point here, Your Honor, is that even if the license were revocable, even if it were not perpetual, and even if SCO could do as it purports to have done here to terminate it, there are under the terms of the agreement before it was amended, to give IBM an irrevocable and perpetual license, requirements that have to be satisfied. Prior to being able to terminate IBM's license, SCO had to give IBM, and we lay this out at tab 45, SCO had to give IBM notice, it had to give IBM an opportunity to cure, and it had to exercise its good faith best efforts to avoid termination.

As shown at tab 46, Your Honor, the case law in New York which controls this agreement is clear that where there are provisions of this kind, that the plaintiff must satisfy the requirements to provide notice and cure and an opportunity for cure and meet its duty of good faith best efforts to resolve the agreement short of termination before it can in fact terminate. SCO couldn't satisfy any of those three, Your Honor, and for that reason, summary judgment should be entered in IBM's favor as well.

Let me take those each briefly. Notice, SCO's notice letter, Your Honor, which was filed with its complaint in this

75

action, which we have attached in the book at tab 48, accused IBM of improperly disclosing, of misappropriating even SCO's trade secrets. Well, again, Your Honor, as I have now said at least twice, SCO has conceded that there are no trade secrets in UNIX System V. It made that concession in open court after it purported to terminate IBM's license. The notice letter says you misappropriated our trade secrets, stop or we're going to terminate your license.

It then admits after it has terminated IBM's license that there are no trade secrets in UNIX System V. It withdraws its claim for trade secret misappropriation.

Opportunity to cure. Because it never disclosed with any meaningful particularity what it was it was complaining about, Your Honor, IBM was never given an opportunity to cure the alleged breach. In fact, Your Honor, if I may approach --

THE COURT: You may.

MR. MARRIOTT: Thank you.

Following SCO's letter of March 6th in which it indicated that it was going to terminate IBM's license, IBM sent SCO a letter and said, well, what is it that you contend we did? Please tell us what it is you claim that we need to do to cure this alleged breach. The response that we received, Your Honor, from Mr. McBride, the C.E.O. of the company said, quote, if you would like further written information regarding IBM's past and continuing violations, we need more information

76

from you. So rather than provide IBM a meaningful opportunity to cure, Your Honor, we were told that before we would learn anything more about what we supposedly had done, we would need to tell SCO what it is that we, in fact, had done.

Finally, Your Honor, with respect to good faith and best efforts, again, as you know, and I reluctantly repeat what has been said so many times before, IBM has repeatedly asked SCO in this litigation what it is that we supposedly did. SCO has repeatedly refused to provide IBM that information and, instead, Your Honor, played what I think is a game of where is the pea? It has required motion after motion to figure out what exactly it was that IBM supposedly did.

It was only after IBM filed motions to compel that we finally learned something of consequence about the JFS contribution, about which so much still remains a mystery. It simply cannot be, Your Honor, that SCO provided by way of its notice letter, proper notice, a notice 100 days before the supposed termi