DAVID R. MARRIOTT
[address]

TODD SCHAUGHNESSY [sic]
CHRIS KAO
AMY SORENSON
[address]
February 6, 2004
10:00 a.m.

THE COURT: Thank you. 

Ladies and gentlemen, the record should reflect that I requested to meet with counsel in chambers for the purposes of determining those issues which would be addressed this morning, and I believe we have successfully identified how we are going to do that.

[For those speculating on the purpose of the meeting in chambers, this is the stated purpose, to get clear what would be argued and what didn't need to be. You'll note later SCO mentions that they goofed and filed an affidavit instead of a request for enlargement, and likely this is when it was brought to their attention by the judge.]

First it would be my request that we go forward to hear argument as to whether or not SCO has complied in accordance with the Court order of December 12th, and what if any measures need to be addressed or action taken with regard to that.

Secondarily, we will address SCO's motions for reciprocal discovery.

[SCO did, then, have an opportunity to address their Motion to Compel.]

So, given that circumstance then, Mr. Marriott, do you wish to go forward and address the issue of whether or not SCO has complied with the Court's order?

MR. MARRIOTT: Yes, Your Honor.

Thank you, Your Honor. The simple answer to Your Honor's question as to whether the SCO Group has complied with the Court's order is that the SCO Group has not complied, Your Honor, as we lay out in the submissions that we made to the Court yesterday afternoon.

[These submissions presumably are IBM's Report on SCO's Compliance and attachments.]

As Your Honor knows the Court ordered SCO by January 12th to provide documents responsive to IBM's document request and to provide full and detailed and complete answers to IBM's interrogatories. There is, I think, no dispute, Your Honor, that the SCO Group has not provided all of the documents that are responsive to IBM's discovery requests, and that is reflected in correspondence between counsel which is an attachment to our submission of yesterday.

Most notably perhaps, Your Honor, is that the SCO group has acknowledged that it has yet to produce documents from approximately 20 of the custodians of responsive documents, and to date in the case it is my understanding, Your Honor, that the company has produced documents from the files of only 20. So about half of the custodians have yet to have their documents produced in the litigation.

So, yes, Your Honor, there is no question that additional documents were produced. We do not dispute that. We appreciate that. But as to whether or not the SCO Group has complied with the order I think the answer as to documents is that it did not.

[Groklaw has posted SCO's Source Log of what it turned over so far and its Revised Supplemental Response to IBM's First and Second Interrogatories is here, referred to by Marriott.]

THE COURT: Assuming that I were to find that it had not, what are you suggesting should occur?

[ Here, the judge is giving IBM the opportunity to ask for a dismissal of some or all of SCO's claims.]

MR. MARRIOTT: I think what Your Honor should do in that regard is to, in consultation with SCO, determine by what date I hope in the reasonable and imminent future they can comply with the request and order them to do that by that date.

[Notably, IBM does not take the opportunity offered. Instead, they make clear that they intend to go forward, that they intend to let this case play out all the way. They want a definitive decision, I'm guessing, on the GPL, as well as a full clearing of their good name, not a TKO.]

Now, with respect to the interrogatories, Your Honor, as you know when we were last here we explained that as we understand and then understood the SCO case, their theory of the case was that IBM had taken code from Unix System Five and dumped that code into the Linux operating system. We asked the Court to require them to identify by file and line of code, what it is they say we took from Unix System V, and where it is exactly in Linux that they say that we put that. Your Honor ordered them to do that.

In response to that order, SCO does essentially three things. First, Your Honor, they abandon any claim that IBM misappropriated any trade secrets. They fail to cite a single trade secret allegedly misappropriated by IBM

Second, they fail to identify a single line of Unix System V code which IBM is alleged to have dumped into Linux. Third, what they do is they clarify their theory of the case. The theory of the case appears to be, Your Honor, from the supplemental submissions, not that IBM dumped code from Unix System V into Linux, but rather that IBM took code out of its flavor of Unix known as AIX and Dynix and dumped that code into the Linux operating system.

Now, specifically SCO identifies 17 files, parts of 17 files, which it says were improperly contributed. With respect to many of the lines of code in those 17 files they properly identify which line it is they say we took from AIX or Dynix and where it is they say we put it in the Linux operating system. With respect to many the disclosure is, I think, sufficient. There are, nevertheless, a number of files as to which they have not properly identified the lines of code which they say were misappropriated, and we would like to have them do that.

More fundamentally, Your Honor, we asked in our interrogatories in at least seven different spots for them to link up the AIX Dynix code which they say we dumped into Linux with the System Five code from which they say it is derived. The theory here appears to be, Your Honor, that IBM cannot properly contribute code from AIX or Dynix even if it is its own home-grown code, if it ever at some point in time touched the AIX or Dynix operating system.

The notion is, Your Honor, that somehow IBM is prohibited from disclosing that code because it derives it in some way from Unix System V. What we asked for in our responses is that they tell us, if that is the theory, exactly where it is in Unix System V that that code derives from. Now, if it is the SCO Group's position, Your Honor, that the 17 or so files which they say we dumped from AIX or Dynix into Linux do not derive from Unix System V, they are not derivative works of Unix System V, then they need merely tell us and much of our concern with respect to this issue will disappear. But it is not at least my understanding, Your Honor, that that is their position. Insofar as it is not their position we want to know exactly what line of code these 17 files, or whatever files in the future they identify, are supposed to have derived from.

[This is IBM trying to get SCO to state its case clearly. Is there System V code involved? If not, then IBM would like a clear statement to that effect. If so, they would like to know exact lines of code involved. If it's only a question of derivative code, they'd like to know exactly where System V appears in the derivative chain, that is, what code in System V is later showing up in AIX and then in Linux?]

In addition, Your Honor, the interrogatories that we propounded asked SCO, and I think in very clear terms, and this is in Interrogatories 12 and 13, to identify exactly what it is in Linux that they contend they have rights to. Irrespective of whether or not IBM is supposed to have contributed this code to Linux, and that matters not just for the case against I.B.M. but also for our counterclaims against the SCO Group.

[Here, IBM once again asks for what part of Linux SCO claims it owns. SCO has made an assertion, but it has said that because the code in question wasn't put in Linux by IBM, they don't need to answer. IBM says here, Sorry. We get to ask, because it has to do with our counterclaims, not just your theory of the case.]

We asked them to identify that, Your Honor, and what we have gotten is an answer that says, with respect to 17 files we own those and we may own some other ones, but there might be all kinds of other code in Linux to which we claim we have rights, but they won't tell us what that code is. We don't have a definitive statement as to this open operating system, Your Honor, which they have complete access to. We don't have a statement that says those are the lines of code that we own, and those are the only lines of code that we own.

Instead what we have is a statement that says we own these and we think we might own some other ones, and then we get a list of a score of companies which they say might have contributed code and, therefore, they may have additional rights in Linux, but they won't tell us what those rights are.

We asked, Your Honor, that they categorically tell us with respect to what they claim they have rights in in Linux. Did we or did we not infringe that? We have been told that we infringed some, but they will not and have not told us but you don't infringe the rest, and we think we are entitled to a statement as to an open operating system that either we infringed the code in Linux or we don't, and if we do exactly what code is it.

Furthermore, we asked that they tell us with respect to all of the code in Linux to which they contend they have rights, exactly whether they distributed the code or made it available over the internet or gave it to somebody else, and I think we get about two sentences which purport to describe the extent to which they have disclosed it, and we don't think that description, even with its reference to some invoices, is enough.

Finally, Your Honor, what I would say is that what is particularly troubling to us is that we are being told that there are 17 files from AIX or Dynix that we improperly contributed. And yet as Your Honor I believe is aware and as we lay out in our submission, the company CEO is publicly making statements to the effect that there are roughly a million lines of code to which IBM is tied, whatever exactly that means. We want to know, Your Honor, if there is anything other than those 17 files, which we're supposed to have done something with, what exactly is it.

That, Your Honor, is not an exhaustive recitation of the shortcomings in the response. Those are the most important ones. The other ones can be found in the correspondence with counsel.

I think what Your Honor should do with respect to the interrogatories is to order the SCO Group to again, within some, I hope, short-term time frame, provide the additional information which we have requested. Certainly with respect to the question of whether these lines of code tie to Unix System V, and if they contend that they do tell us in unequivocal terms that the files that we're said to have contributed do not derive from and are not derivative works of Unix System V.

Thank you, Your Honor.

THE COURT: Mr. Marriott, I am wondering if during the remainder of the hearing if you could perhaps ask someone with you to make a handwritten summary list of those things, specifically.

[This is a hint, to my way of thinking, that the judge likely intends to ask SCO to produce whatever is on the IBM list, in harmony with IBM's suggested resolution.]

MR. MARRIOTT: I can do that, Your Honor. Thank you.

THE COURT: Thank you.

MR. HEISE: Good morning, Your Honor.

THE COURT: Good Morning.

MR. HEISE: Mark Heise, Boies, Schiller & Flexner on behalf of the SCO Group.

With respect to this first issue of compliance with this Court's order requiring the Supplemental Interrogatories and Requests for Production, at the last hearing virtually the entire time was spent on the interrogatories so I am going to focus my attention on the interrogatories.

We filed our interrogatory answers that supplemented and they exhaustively detailed the improper contributions that IBM has made to Linux. On Monday IBM sent to me a letter detailing what they thought were deficiencies in it.

[Groklaw has posted that letter.]

And on Wednesday I responded to that, both of which are in the package that IBM provided to you I believe it was yesterday. In there we have detailed why it is what they are asking for in this next round of supplemental answers is, A, not what was asked for in the questions and, B, not at all appropriate in light of what this case is about.

[That letter is here. ]

When they filed their responses or their report on the compliance yesterday, it appeared to me that they were abandoning their nitpicking points that they raised in their letter, which I will be glad to address in detail because we have in fact answered those questions. I did detail that in our response dated February 4th.

THE COURT: You can choose to say whatever you wish within your time constraints.

[This appears to be the court gently telling him, If you wish to waste the time allotted to you this way, you are free to do so, but you may wish to think that play through, because if you run out of time before you address the matters we just agreed in chambers we'd cover here today, you won't get any more time.]

MR. HEISE: I understand, Judge. Thank you.

With respect to the overriding issue that IBM has presented to the Court, it is that SCO has somehow failed to identify line-for-line codes of System V code that was licensed to them that IBM has put into Linux. That is not and has not and will not be this case at this point.

As I started to allude to in chambers, Your Honor, the fundamental reason why that information is not relevant and is not provided is because of section 2.01 of the license agreement. In 1985 IBM made a commitment that they would get this operating system called Unix System V from AT&T and they would agree to those terms. One of the most significant terms of that is found in 2.01. It says that they have the right to use Unix System V and, in fact, they have more than that. They have the right to modify it and they have the right to create derivative works. I am reading from the bottom part of 2.01 that is in bold. The important limitation is that on that Unix System Five Code, on the modifications to it and the derivative works to it, they must treat it as part of the original software product.

[However, note the Novell letter of February 6 to SCO, which answers this point directly, saying this clause does not mean what SCO thinks it does.]

And then the license agreement is very detailed as to what they can or cannot do with the original System V code or their modifications or derivatives. Their modifications and derivatives are called AIX and Dynix. They are required to keep them for their own internal business purposes and keep them confidential and not give them away.

That is in fact what they have done as we have set forth in the next page of the exhibit. They have taken their modification or derivative known as AIX or Dynix, and they have contributed it to Linux allowing Linux to now become an enterprise corporate use of this operating system. In the absence of that it wouldn't have gotten there, but there can be no question, and you have not heard IBM come up here and say, Judge, we have not contributed AIX and we have not contributed Dynix. They have in fact, and they have publicly said they have done that, and we have provided line-for-line copying of exact AIX and exact Dynix code. They are prohibited under this contract from doing so.

If Your Honor thinks of it as a ladder with the first ten steps being Unix System V, because let there be no doubt, there are over 1,000 files in AIX that are attributable to AT&T. So in that AIX core there is that platform, those first ten rungs of the ladder. What they have done is, they have created their flavor of their version of that operating system called AIX or Dynix, and that is now rungs 11 through 20. They are saying in rung 16 you're not showing me the Unix System V code. That is not what this case is about. This case is exactly about what is set forth in 2.01. You can't take the System V code and you can't take your derivatives or your modifications.

If they want to come in here and say, but those derivatives or modifications came from somewhere else and then were wholly created by us, then you know what? They have to prove it. It is not good enough for Big Blue to come in and say that.

THE COURT: Mr. Heise, I think you're arguing the merits more than the scope of this hearing.

[Another hint from the judge, which again is ignored by SCO.]

MR. HEISE: The reason I am maybe going more into the merits than I probably should in front of Your Honor is it directly ties into the adequacy of these interrogatory answers. The interrogatory answers detail exhaustively the contributions of AIX and Dynix that were made in there. There is no dispute about that.

They then in this letter that they wrote earlier this week said, Well, you didn't identify the line-for-line matching in every single place. There are two times when we did not do that in our answers to interrogatories. One is in table A of our interrogatories which we identified eight different files and we said the copying is complete throughout. We are not matching up the lines and I gave an example of that in the demonstrative aids when it says copying of Dynix slash into Linux, and you can see the red on the right is exactly the same as the red on the left, and that is line-for-line copying. So that is the one instance in our interrogatory answers where we admittedly said in there it is throughout. We are not identifying lines here.

The other place where we did not identify the line-for-line copying are for certain technologies known as Asychronous Input/Output and for Scatter Gather Input and Output. There is a very fundamental reason. Because to be able to do the line-for-line matching we need to have their source code. They have given us zero AIX and two CD's of Dynix.

THE COURT: But the requirement of the Court is that you provide those source codes.

MR. HEISE: I think there is a fundamental misunderstanding and let me explain why.

[SCO persists in trying to reargue the point the judge has already ruled on, namely that SCO must produce first and only after that can they demand that IBM produce discovery.]

With respect to these other technologies that they have publicly acknowledged that they have contributed, they have laid out how it is that they have contributed it, and it was a part of AIX or Dynix, and what they are saying is, Show us the lines. That is the equivalent of saying I am not
going to show you the book that contains all of these lines of code, therefore, all we can do is say it is from AIX or Dynix and you have said it is and we have identified how it is and why we believe it is in fact from AIX or Dynix. But to sit here and say to us when they have not given us their source code, and their source code is what is matched up --

THE COURT: This is about your response and compliance with the Court order.

MR. HEISE: I understand that.

We have given the technology based upon the information we have. The answers to interrogatories that they are complaining say, yes, but for those given technologies you have not identified the specific lines. What we have said in our answers to interrogatories is we can't identify those specific lines because it comes from your confidential code which we don't have access to yet.

THE COURT: Mr. Heise, this is the problem. The problem is that unless you identify those codes, which was required by the Court order --

MR. HEISE: Which we did.

THE COURT: -- then I.B.M. is not in a position necessarily to respond, the way I see it. So we are at an impasse and we can't be at an impasse and have the case remain at a standstill. That is why there is an order in place that SCO has been required to comply with, so that I might then address what IBM has to comply with.

MR. HEISE: But I'm trying to stay focussed on our compliance.

[He persists in trying to argue the same point, that her order was wrong and IBM should have to go first.]

I guess maybe a way to explain it, is in the technologies that they have contributed, let's say in rungs 15 and 16, that is not from us. That is not our Unix System V code. That is AIX or Dynix. We don't have that source code to be able to identify the lines, because they are quibbling about the fact that we have not identified the lines of a couple of technologies. We don't have the source code for 15 and 16. They do.

If they give it to us we'll supplement if further, but in the absence of that it is literally impossible to identify the lines. We have identified the technology, we just cannot identify the lines because we don't have their derivative modification source code. That is why and that is what I am trying to get across.

THE COURT: Well, you have made your point, I am just not certain I agree with it.
MR. HEISE: Fair enough.

As I detailed in the February 4th letter that Your Honor has, we have addressed everything. All we are talking about is whether it is a derivative or whether it is a modification. As you pointed out when I kept going down that path, that is the merits of the case. That is not appropriate on a discovery motion.

With respect to the production of documents, we have in this case gone to every office, gone into peoples' individual offices, gone into peoples' home offices, and we have gathered and collected more than a couple million pages of documents. We have produced over a million pages of documents. We have produced 400 million lines of Unix code, most of which I fail to see how it has any relevance. We have produced 300 million lines of Linux code, and we have gone through exhaustively to provide them with documents in the order they wanted it, and they wanted it from the top executives down. As we indicated earlier there were technical difficulties when going through it. Some of the third party vendors didn't process materials because they were so focussed on the other. We have made every effort to correct that.

[This is him saying we didn't produce everything. We did produce a lot, though. Meaningless, because the hearing is about whether they have produced everything.]

I understand from our discussions before that we should have filed a motion for enlargement rather than explain it by affidavit. I take full responsibility for that. That was error on our part.

[Another SCO legal goof.]

But we have literally undertaken these Herculean efforts to provide them with every document that we can get our hands on.

And throughout the course of this case, Your Honor, there will be more documents. There will be more documents by them and there will be more documents by us. That is just the nature of discovery.

THE COURT: How much time do you need to provide these additional things that have yet to be supplied? And if I order an absolute strict compliance to the previous order, and/or some of the items that IBM is indicating, I want you to state for me a reasonable and rapid date on which those could be provided.

[Here is a strong hint what she plans to do: give them an Do It By This Date Or Else order. She has a number of Or Else options to choose from.]

MR. HEISE: With respect to the supplemental documents that have been collected and that we are trying to gather and provide to them, I would anticipate it being done in two weeks. But to give myself, so I don't have to come back before you and file a motion for enlargement, I would rather say four weeks and go with that.

THE COURT: All right. Do you have anything else?

MR. HEISE: With respect to our compliance, no, Your Honor.

As I said, you know, I am sure we'll be here talking about this document is missing and that document is missing. That is just the nature of the beast.

Thank you very much, Your Honor.

THE COURT: Thank you.

Mr. Marriott, do you have any response to this? And I have a question for you as well.

MR. MARRIOTT: I do, Your Honor.

I believe what Mr. Heise said, Your Honor, was that the reason that we had not been given all of the line-for-line match-ups that we had asked for is because the only way that they can do that is for us to give them discovery. When I stood, Your Honor, ten minutes ago and described to you the principal failing, it is not that they have not identified all of the lines of code in AIX and Dynix, which they say we dumped into Linux, it is the lines of code in Unix System V. That is the product they purport to have acquired from AT&T and it is in their possession, and there is no reason that they can't do that, or state categorically that it is not the case and that they are derivatives of System V. You didn't hear Mr. Heise say that.

That is my only response, Your Honor.

THE COURT: Mr. Marriott, my question for you is, do you acknowledge or not acknowledge that SCO is in substantial compliance with the previous order?

MR. MARRIOTT: Well, that is a very hard question, Your Honor. We were provided with a lot of documents and we were given certainly a lot more specificity than we had been provided previously.

The difficulty is that since in our judgment without getting to the merits, but in our judgment the question here is whether the code they say we have dumped into Linux can be linked to Unix System V. They have a different view. We won't argue the merits of that. Certainly we are entitled to discovery as to whether that is the case. I would refer Your Honor to our interrogatory numbers one, two, four, six, nine and 13.

In one we ask for confidential information misused. In two we ask for the nature and the source of the rights. In four we ask for the manner of misuse. In six the origin of the code and the products upon which it is based. I mean, the list goes on, Your Honor. One of which, in fact, 12 asks specifically whether the code was derived from Unix System V. So whether or not we have the same view on the merits as SCO as to the contracts, which clearly we do not, certainly we are entitled to discovery as to our understanding of the way the contract works.

[I am guessing that he here is referring to the list that the judge earlier asked them to write up for her.]

We have clearly asked for that. This is a case that to my mind is about whether Unix System V in one fashion or another, either directly or because some derivative of it has been dumped into Linux, has been adequately provided and we don't have that.

To be asked when they have given us a lot, Your Honor, it is --

THE COURT: Maybe that is a determination for me and not for you.

Do you have anything else on this?

MR. MARRIOTT: I do not, Your Honor.

THE COURT: Thank you.

I will take that under advisement, but let's go forward now on the issue of SCO's motion for discovery from IBM.

MR. HEISE: Thank you, Your Honor.

With respect to SCO's Motion to Compel, in this case IBM prior to the Court entering the stay on December 5th had produced approximately 150,000 pages. They have produced two CD's of Dynix source code and zero CD's of AIX source code.

First and foremost, they have repeatedly stated throughout this case that they would provide the AIX. and Dynix source code and we just have not gotten it.

THE COURT: Well, that may be the result of this order which said hold on, we are stopping this until those source codes were revealed.

[Again, SCO is back arguing that it should get AIX and Dynix code from IBM, and the court is again reminding them that they are not going to get it, until her order is fully complied with. This isn't saying that she has decided that point, though there have been some hints, but it is saying that IBM has a reason for not providing anything yet, since they were likely obeying the court's order, and they also are of the view that SCO has not yet provided all that was required under the order.]

MR. HEISE:. Your Honor, there had been numerous, repeated promises of delivery of that source code prior to December 5. It had nothing to do with the stay that this Court entered. Numerous times we were told we would get it. What we then were told is we can't provide it to you because we have not gotten these third-party notifications done. What that means is that within the source code, some third party also has their source code and they need to make sure that they are okay on that.

That is a process that has literally dragged out for months, and I am still getting contacted by third-party vendors of theirs that are saying, How can we work this out? I have immediately responded and worked it out and it is still not happening.

The other critical deficiency in the production of documents and interrogatory answers is that there is nothing from any of the highest levels of the company.

[This is a big hint, to me, that they are looking for something specific from "the highest levels of" IBM, something they didn't yet get.]

As you saw when IBM was filing their Motion to Compel, they kept asking for Darl McBride, the CEO, Chris Sontag, senior vice president, all of the top key people and kind of working their way down the ladder.

What we have gotten from IBM is working its way up the ladder, despite the fact that on October 28th and other occasions I have spoken with representatives of IBM and said we want the documents and materials from Sam Palmisano, from Irving Wladawsky-Berger, the key executives that are intimately involved in the Linux project.

[This hints to me that there is something they are looking for that they think they should find said or done by one of these two, conceivably by both. The second one is very possibly thrown in there as camouflage, so IBM will not figure out what it is SCO is looking for.]

In our reply memo in support of this Motion to Compel, we in fact provided an article from the New York Times where Mr. Palmisano is identified as the leader of moving IBM into the Linux movement. Mr. Wladawsky-Berger is a core, critical person and they are not mentioned in any of their interrogatory answers and we have gotten no documents from them.

But in terms of going to the specifics of the request for production, we have asked for in items two and three of our requests for production, all AIX and Dynix versions and iterations. As I said, we have gotten zero from AIX and we have gotten two CD's of Dynix. What was laid out in IBM's response to this Motion to Compel, in part, was that would be unduly burdensome. At the last hearing they told you that that could be up to 40 million pages of code and how could we possible undertake that extravagant exercise to get that?

In the limited discovery that we have gotten from them it is clear why no affidavit or no supporting proof was given as to this and why it is allegedly burdensome.

If I may hand this to Your Honor?

THE COURT: Okay.

MR. HEISE: What I am handing you is a document from IBM that has been marked as confidential. It is regarding an item called the C.M.V.C. which stands for Configuration Management Version Control. As you can see, Your Honor, it says in the beginning it is used by the AIX development organization, and through the highlighted portions of the document it identifies that configuration management is a process of identifying, managing and controlling software modules as they change over time.

In other words, so that we would be able to get every version, every iteration, and that version control is the storage of multiple versions in a single file along with information about each version. Then it gives a simplified description at the bottom saying what it basically does is it boils down to that all levels of all files are stored on a central server and are available for viewing and/or updating by those with proper authority.

They can get us the AIX. It is clear as a bell we are entitled to it and they said they would give it to us and we just have not gotten it.

With respect to request or production number 11 and interrogatory number five, they are directed towards all of IBM's contributions to Linux. From AIX to Dynix, anything that you have done, any work that you have done for Linux, provide it to us. With respect to the request for production the response IBM has made is, quote, IBM has made a lot of contributions so it is going to be a daunting task. IBM has made a lot of contributions. That is not a reason why they are not required to produce them.

That is a core issue to this case, as I kind of went off track before under 2.01. What did you do with this material that we said that you were not allowed to make public? They are required to identify that. And what is a critical followup to the production of all of what they have produced is Interrogatory 11, which specifically requires that they identify who worked on it and what they did. They say there are hundreds of people and that is an onerous task. Well, it is a critical task. That is exactly what this interrogatory is designed to do, is for us to know who at IBM worked on it and what they did. Because at the end of the day if we get a list of their 300 people that they have identified already or approximately in that range, and a person has made one contribution over here and this other person has made 50 contributions, the deposition is more likely to be taken of the person who has made 50.

If I just get a random list of names I have no idea of how to weigh who it is I should be focusing my attention on. That is why it is critical that the interrogatory be answered fully and completely.

With respect to Interrogatory number two, we asked for all persons with knowledge. They limited it in their answers to just IBM people. They voluntarily have agreed that they will in fact provide the identity of all persons with knowledge and with information in this case. The only thing I would reiterate here is it has to be inclusive. They can't exclude top management because they are very important executives. Sam Palmisano is a critical witness in this case. The fact that he is the CEO of IBM does not make him somebody who is not to be put on this list.

[ I'd say this is an indication of who it is they specifically think they have something on.]

There is that New York Times article that was attached to our reply memo, it identifies and there was a ten-page report that he and Mr. Wladawsky-Berger and a couple of others put together in deciding whether IBM should shift gears and go to Linux. We don't have that ten-page report and it is a critical document. Those are the things that we have asked for. We have had specific conversations with Christine Arena at Cravath, asking specifically for Mr. Palmisano stuff, for Mr. Wladawsky-Berger, Paul Horn, Nick Bowen, those peoples' information. We have not gotten it.

Throughout these they have not provided the contact information so that we would not be able to locate these people, and that is just clearly information that needs to be put in there.

The final point is more of a housekeeping matter, and that is in the production that we have received to date, we will get a CD and it will say there are two documents on it. The two documents will be 4,000 pages long. Clearly that is not the case. When SCO has been producing CD's it has identified where each document begins and ends. We have asked them, you have to identify where the documents begin and end. Put a source log with the CD. Otherwise it is impossible to know how these documents were kept in the ordinary course of business as is required under Rule 34(b).

Certainly on some documents you can figure it out and match it up and see where it begins and ends, but we can't be left to the guessing game. It is a technical issue but it is something that can presumably be corrected, and it certainly needs to be done on a going-forward basis.

That is the gist of our Motion to Compel, Your Honor. I appreciate your time this morning.

THE COURT: Thank you.

Mr. Marriott.

MR. MARRIOTT: Thank you, Your Honor.

The SCO Group propounded 57 document requests and/or interrogatories, Your Honor. 52 document requests and there were five interrogatories. SCO's Motion to Compel concerns only six of those requests, three document requests and three interrogatories.

[Here, my feel is that he is indicating that no Motion to Compel seems needed, since IBM has already provided the bulk of what was requested and has agreed to give over most of the rest. So he is narrowing down the scope and also showing no intransigence on IBM's part. Motions to Compel are normally only appropriate if there is some evidence of a lack of cooperation.]

The requests, Your Honor, break down into roughly four categories. There are, I would submit really, only two issues that deserve argument, that is argument as to two categories of the four. That is because if Your Honor looks at our Opposition to their Motion to Compel, I think in part this is a motion that makes much ado about nothing, because we either have indicated that we will provide or have provided much of the information requested.

For example, Mr. Heise makes reference to desiring to know the identity of the people who have contributed in some way to AIX or Dynix. Well, there is provided as an exhibit to our response, Your Honor, a list of about 8,000 people who made contributions. So the notion that somehow we have not done that is absolutely incorrect.

Let me focus, if I may, Your Honor's attention on the two points on which I think we do have a genuine dispute. The SCO Group has asked for the production of all Dynix and AIX code during the relevant period, every iteration, every version known to man or woman. They have also, Your Honor, and this is the other category, asked for every contribution that IBM has ever made of any kind, however irrelevant to this case, to Linux or to the open source community. Let me take each of those in turn.

With respect, Your Honor, to the request that we produce every conceivable version or iteration of AIX, the only theory, Your Honor, disclosed by the SCO Group as to how it is that IBM breached its contracts with AT&T, is that somehow IBM has disclosed code from AIX and Dynix into Linux. Having production of every iteration and version of AIX and Dynix is entirely irrelevant, Your Honor, to the determination as to whether or not those products are improperly contributed. The theory is they are somehow derived from Unix System V.

[Here he is using logic, which is always refreshing, pointing out that SCO says they need every version of IBM's code since the founding of the world, but to prove what? As he points out, their theory is that any contribution from AIX or Dynix is improper, so what does it matter which line of code is involved? IBM isn't saying they never made such contributions, so what does SCO need the code for?]

If I may, Your Honor, referring to the SCO Group's exhibits in connection with this hearing, if you look at page 2, which they call defendant's improper contributions to Linux, and they are not numbered, Your Honor, but I believe it is page 2, substantive page two.

Their theory, Your Honor, is that we have taken some code from here and we have dumped it into Linux. One does not need to know and does not need to have production of every version and iteration of AIX and Dynix in order to figure out whether the contribution of these 17 files is somehow improper.

One determines, Your Honor, whether a contribution is a derivative work of Unix System V, and therefore -- under their theory -- improper, simply by comparing the 17 files that were disclosed that they have identified to Unix System V, to determine whether they are a derivative of Unix System V. If they are, then under their theory there might be a problem. Under our theory, and we have a different theory from them, but under their theory there might be a problem. But one does not need the code at this level to figure that out. The case law is absolutely that you figure out whether a work is a derivative work by comparing level C here to level A, not by looking at the millions of lines of source code that they want at this level.

THE COURT: What case are you relying on for that proposition?

MR. MARRIOTT: I would refer the Court, for example, to the Computer Associates decision out of the Second Circuit, which is one of the leading cases on copyright infringement, wherein the nature in which a derivative work is determined is laid out. If you will permit me --

THE COURT: That is fine. You can provide it at a later time.

MR. MARRIOTT: Put aside the case law, Your Honor, and just look at the next page in the SCO Group's book.

[Note that he is here using SCO's own graphic against them.]

This is the page in which they say others have complied with the requirements of licensing agreements. What you see is the SCO Group reflecting itself at the top and representing that they have relationships with HP and IBM and Sequent and Sun, indicating, according to this chart, that somehow Sequent and IBM have improperly made contributions, the 17 files we referred to.

Then referring to HP and Sun and saying HP and Sun have not made any contributions to be concerned about; they have complied with the licensing agreement. How do they figure that out, Your Honor, that HP and Sun complied? I would submit to you that HP and Sun have not produced millions of lines of source code to them so that they can do this comparison that they represent is so critical to the Court. That is absolutely not the case.

They have figured it out and they have reached the conclusion that they are willing to announce publicly, that these other companies don't somehow infringe their rights without reference, I would submit, to a single line of the code from HPUX or Solaris. If they have used that code, then they should have produced it in discovery to us because our request would have called for it and we don't have it.

[This is a slam dunk argument: why do they need us to provide every version of our code to see if we complied, when they didn't require that from HP and Sun? If they need our code to determine that point, they should need it from HP and Sun as well, because they are also under the same kind of license and they have "derivative" versions of Unix System V, if you follow SCO's logic.

If that is not enough, Your Honor, take a look if you would, please, at a document which I can only describe generally

[A reader points out likely he is talking about the document carefully, because it is marked confidential, and so he is trying to avoid saying out loud what it is.]

because it has been marked by the SCO Group as --

May I approach, Your Honor?

THE COURT: Certainly.

MR. MARRIOTT: This document has been marked as "confidential" so I'm limited in what I can say about the document, Your Honor.

But if you will look at this you'll see that it is a letter from SCO to HP. If you take a look at the last paragraph of the document, Your Honor, you'll see that in order to reach the conclusion they reach here, they didn't rely upon, and I would submit that they won't tell you that they did, the production of millions of lines of code from HP or from Sun or from anybody else. That is because that code is not required for them to reach the conclusions they reach as to why it is IBM has supposedly breached its agreement with them.

Their own documents, Your Honor, damn the notion that they somehow require the production of millions and millions of lines of source code in order to figure out whether IBM has breached its agreements with AT&T.

Your Honor, we are nevertheless, however irrelevant I believe that the production of this code is, we are nevertheless prepared to produce a substantial amount of code. We have produced already significant lines of code from Dynix, and we are prepared to produce, and the reason we have not produced it, by the way, Your Honor, is because you ordered us not to. You put in place a stay and the production of the code that Mr. Heise complains about would have put us in violation of the order of the Court. We thought it prudent not to do that.

[Here, he reinforces her earlier speculation that they didn't produce more discovery because of the order, not because of intransigence.]

If you look, Your Honor, at what we are willing to produce, it is a substantial amount of code. We either have produced or will produce three million pages of paper of source code. That isn't every conceivable iteration of these products. It is, however, about 232 products.

If I may approach?

[Note how respectful he is at all times, showing her the respect her position calls for.]

THE COURT: Certainly.

MR. MARRIOTT: Now, again, I think the production of this material is entirely uncalled for, Judge, but we are prepared to do it to put to rest this notion that somehow IBM is somehow hiding the ball with respect to the production of source code. This amounts to well over 100 million lines of source code and we are prepared to produce that. We said we were prepared to produce that in our opposition papers. This is the releases of AIX and Dynix and the released products during the relevant time periods that they are concerned about.

What we are not willing to do, Your Honor, is to produce every conceivable draft and iteration and version of this stuff that might exist in the files of the company that has more than 100,000 employees, with respect to products that were developed over decades, and as to which 8,000 different individuals worked on.

To state it, Your Honor, is to express its absurdity. We are willing to produce far more than they ever had. I would submit again that they had no lines of code from HP and Sun and they were able to reach the conclusion that they are perfectly compliant. We are willing to give them more than 100 million lines of code but that is not good enough. What they say they have to have is every single conceivable version of a product worked on by thousands of people, and if there is a draft of this line or that line, if it exists in this database to which Mr. Heise refers, he wants us to dump on him a database that would be, I would submit among other things, horribly burdensome to do and for which there is simply no cause.

We are prepared to produce these lines of code, Your Honor, and that ought to be enough, I think, for any case, certainly in view of what seems to be sufficient with respect to other persons who were in no different situation for this purpose than is IBM. That is the first category, Your Honor, and there is no reason for an order to compel of any kind. As we said in our opposition papers, we will provide them with the information and we are happy -- we are not happy, Your Honor, we are willing to do that.

With respect to the next category that I think merits mention, they have requested -- again I think in an overreaching fashion -- for every conceivable contribution that IBM has said to have made or may have made to Linux or to the open source community. We have already produced or agreed to produce approximately three million pages of paper. We have produced documents from approximately 90 separate custodians, Your Honor, located in various parts of the country.

We have not withheld from the production of those materials a contribution that a person may have made to the open source community. We are not running through and pulling out contributions. What we are saying is that it is entirely unrealistic and uncalled for and unduly burdensome to expect that we would produce every conceivable contribution. Why is that the case? That is the case because as you may remember from our last hearing before Your Honor, when we handed Your Honor the Open Source Development Lab chart which shows the way in which Linux was developed, the contributions to Linux, Your Honor, are public. This is a public affair. They know what they are.

Anyone can find out what they are simply by looking in the public record. There is no reason to have us have to run around and interview hundreds of people to figure out whether they may have made a contribution, whether it may still be in their files, when all the SCO Group has to do is get on the internet and find the contributions. How after all did they get the information that they provided in response fo the interrogatories? They got it off the internet.

Your Honor, I would, if I may, again wish to show you two documents which I am constrained from describing in any detail because they are marked as confidential under the protective order.

May I --

THE COURT: Certainly, and you need not ask.

MR. MARRIOTT: Thank you.

These, Your Honor, are e-mails produced in the litigation by the SCO Group. If you look at them what you will see it is and, again, I will not describe them in any great detail, you will see that they fully understand that the documents that they are interested in about IBM's contributions are available on the internet. They can get them and find them for themselves.

I don't have much more to add, Your Honor, than that I think it a silly exercise to require us to produce to them that which is already publicly available. Indeed, that is a proposition without support in the case law. Rule 26 itself expressly provides, and I quote, and this is 26(b)(2), the frequency or extent of use of the discovery methods otherwise permitted under these rules shall be limited by the Court if it determines that the discovery sought is obtainable from some other source that is more convenient, less burdensome and less extensive.

There are two decisions, the Ebers case from the District of North Dakota, which is 2003 WestLaw, 22097788, in which the Court there grants a protective order for discovery that sought information that was available to the public by calling the court, for example. In another case, American Medical Systems versus National Union Fire, which is 1999 WestLaw, 562738, where the court denied a request of a party to compel discovery with respect to documents that were available under F.O.I.A.

These documents are publicly available and we shouldn't have to run around collecting them. There is ample support for the proposition, Your Honor, that nebulous requests for all kinds of contributions can't be the basis of massive discovery.

THE COURT: Assume, Mr. Marriott, that I am going to require IBM to comply in some fashion. What period of time reasonably after receipt of continued discovery that I may order to be supplied by SCO will it take?

[This is an indication that she likely will ask them to produce something, but since IBM has already offered to provide something, what she will do is detail exactly what.]

MR. MARRIOTT: Are you referring specifically to the request for production of contributions, Your Honor?

THE COURT: Yes. Well, just everything.

MR. MARRIOTT: With respect to the code, we are in a position to produce the code within 14 business days of the lifting of the stay.

THE COURT: All right.

MR. MARRIOTT: By the way, when I say produce the code I am referring to the code on this list of 232 products, not every conceivable iteration known to man or woman. That would take many, many months and I don't even want to think how long it would take to compile that information.

[Here, he is letting the judge know that it would delay the case unduly were she to order them to produce every version of AIX and Dynix and every contribution everyone has ever made to Linux from IBM. He's offering to provide in two weeks what he said they are willing to provide.]

But as to this list, hundreds of millions of lines of code, this can be done within 14 business days, Your Honor, of the lifting of the stay.

With respect to the contributions, all conceivable contributions, it would take months to identify, collect, review for privilege, send to a vendor and get produced onto a CD that information. It would be done I think at extraordinary expense.

THE COURT: Thank you, Mr. Marriott. Mr. Heise?

MR. SHAUGHNESSY: Your Honor, may I just give you the cite to the case you asked for?

[Notice it isn't Mr. Heise here speaking. I don't know what that means.]

THE COURT: Certainly. Why don't you just hand it to --

MR. MARRIOTT: This is probably a better cite than the Computer Systems case, although that is a leading case on copyright infringement.

MR. SHAUGHNESSY: It is a case from the Ninth Circuit, Litchfield versus Spielberg, 736 F2nd, 1352, a 1984 case from the Ninth Circuit.

THE COURT: Thank you.

MR. HEISE: Thank you, Your Honor. Very briefly because I know you have a busy day.

The vast, vast majority of what you heard was argument relating to the merits of this case and not about what is at the core of any inquiry on a motion to compel, which is: Are we entitled to relevant information, information that tends to prove a fact one way or the other? IBM has spent the better part of today trying to say that AIX and Dynix are not derivative works. They say that and then they say we are not going to give you any proof to be able to prove otherwise.

So if they are going to come up here and say AIX and Dynix are derivative works, that may obviate the need for some of this discovery. But in fact we have asked them in a request for admission, which we have provided to you as a part of our notice of filing, admit AIX and Dynix are derivative works of Unix System V. They say we don't know what you mean by AIX, and we don't know what you mean by derivative works, and we don't know what you mean by Unix System V and there are multiple versions so we can't answer this so, therefore, it is denied.

They can't talk out of one side of their mouth and say it is not a derivative work, and then turn around and talk out of the other side of their mouth and say we are not going to give you any of the source code for you, SCO, to be able to disprove our contention that in fact AIX and Dynix are derivatives and modifications of System V.

[Note that the Novell letter of February 6, this same day, addresses this issue in a new and devastating way. Presumably, SCO didn't yet know about the Novell letter. ]

We know from the little bit of discovery that we do have that there is over 1,000 files of AT&T that are within AIX. That is going to make it a modification for a derivative work. They are not entitled to continue to sit here and say, one, it is not a derivative work and, two, we are not going to give you any evidence to be able to disprove that.

With respect to this contention regarding how did they know this regarding Hewlett Packard and that Hewlett Packard complied with the license agreement? Well, there is a fundamental difference between Hewlett Packard and Sun and virtually every other licensee out. They have not all gone around and said, great news, we are taking our derivative work, our modifications, and we are contributing it to Linux. H.P. has specifically not done that.

These other companies are setting up Chinese walls and they are not taking that modification or that derivative work, and in the case of HP, it is called HPUX, and they are not taking bits and pieces of HPUX and dumping it into Linux. That is what IBM is doing and they are not allowed to do that under the terms of their agreement. It is a very simple proposition as to why that statement can be made comfortably by the company.

With respect to this notion that they don't have to identify their contributions to IBM because it is public, not every contribution that they have made is public. Not everything that they have done to put into Linux is public.

[Actually, even contributions that don't make their way into the kernel are also a matter of public record.]

Unless somebody is going to come up here and say that, and maybe a way to limit it, is show us everything that is not on the web site. But the fact that some of the information is public does not make it a complete disclosure of everything that IBM did. IBM is obligated in this case to answer this very straightforward question that goes to the core of this case: What contributions did you make to Linux? What work did you provide to Linux?
At the end of the day they can say, do you know what? This thing we did here, that is not a violation of the agreement. This thing we did over here, that is not a violation of the agreement. But until we see what it is that they are acknowledging and that they must under the rules of discovery, then we are entitled to that information. But they don't get to just say some of this may be public and, therefore, we don't have an obligation to respond.

They have got all of these versions and iterations on a central server. They make it available to all of their employees. I fail to understand how it can be on a central server at IBM available to all IBM employees to track all versions and all iterations of AIX, but we can't have access to that in their responses to litigation. It is not what the rules provide. They have got easy access. There has been no affidavit or other evidence of the allegedly burdensome nature of this. In fact, this document belies such an argument.

As a result we are clearly entitled to the information that we have asked for, and particularly the contributions and the source code that they have agreed to give us, and they have to have these employees identify which employees made which contributions to this, so that when discovery progresses we don't look at a list of 300 or 8,000 and have to guess which ones we should start with.

Thank you very much for your time, Your Honor.

THE COURT: Mr. Marriott?

MR. MARRIOTT: May I just briefly respond, Your Honor?

THE COURT: Certainly.

MR. MARRIOTT: I don't think, Your Honor, that we have suggested this morning that we are giving them no source code. What we have said is we are going to give them hundreds of millions of line of source code. So I think it is inaccurate to say or suggest that they should somehow figure this out without the production of any source code.

As to the notion that HP is somehow different, Your Honor, it is a matter of public record that HP makes significant contributions to Linux. Under Mr. Heise's theory, Your Honor, there is absolutely no way that he could know whether these contributions were proper or improper or from their Unix product or not, unless under his theory he had all of their source code. So it is impossible to distinguish HP under some notion that somehow they are not making contributions to Linux. It simply is not true, and there would be no way under his theory for him to know whether or not the contribution was a problem unless he had the millions of lines of source code which he has not been provided, which he didn't tell you he has been provided but which we have said we are willing to provide to them.

The C.M.V.C. database, Your Honor, is not a database that can simply be produced, Your Honor, and turned over. It is not a database that concerns solely AIX code. It concerns code well beyond the AIX code base.

Moreover, the notion that somehow we are unallowed to contend that it would be burdensome for us to comply with these requests because we have not submitted an affidavit is entirely inconsistent with the law that governs in this circuit.

I would refer the Court to the Aikens decision at 217 F.R.D. 533, the Bradley decision at 2001 WestLaw, 1249339, and the Pulsecard case at 1996 WestLaw, 397567. Affidavits are not required, Your Honor, to show overbreadth or undue burden where the details are provided in the briefs or that the overbreadth is obvious.

I would submit to Your Honor that asking us to produce what would amount to a billion lines of code, if we were to produce every conceivable iteration, is on its face overly burdensome.

Thank you, Your Honor.

THE COURT: Counsel, while it is somewhat unusual in a discovery matter to take something under advisement, I think that based upon the somewhat complex nature of the requests that I will issue a written order as to both of the issues before the Court. We'll try to do so within the next week.

Mr. Marriott?

MR. MARRIOTT: May I inquire, Your Honor, you had asked whether I could have someone prepare a summary of our --

THE COURT: Yes.

MR. MARRIOTT: It is not very pretty, and if I might, we have -- I just didn't bring it -- a prettier version of this which I would be happy to send to you this afternoon, or I can hand you this.

THE COURT: I am happy to take that, but if you'll give that to Ms. Pehrson she can make a copy of that for Mr. Heise and for me.

MR. HEISE: The only thing I was going to suggest, is their criticisms and our responses are laid out in the letters of January 3rd and February 4.

[This is a reminder to her to please look over and review the matters addressed in their exhibits, when making her decisions.]

THE COURT: I understand that. All right.

With that, we'll be in recess on this matter and we will get that order out as quickly as possible. Thank you.