decoration decoration
Stories

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

Groklaw Gear

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


To read comments to this article, go here
SCO's Memorandum in Support of Motion to Compel - as text
Friday, September 23 2005 @ 05:20 PM EDT

Here's SCO's [REDACTED] Plaintiff's Memorandum in Support of Its Renewed Motion to Compel with exhibits, as text, thanks again to Steve Martin.

This is one of the three motions that will be heard on October 7. Yes, three. The court has just posted an amended notice about the hearing that reads in pertinent part:

AMENDED NOTICE OF HEARING ON MOTION re: [365] Renewed Motion by SCO Group to Compel Discovery, [503] MOTION to Compel discovery, [508] MOTION for Discovery

This Renewed Motion to Compel is about Linux materials. The others being heard are the old, old motion from December of 2004, SCO's Renewed Motion to Compel Discovery, which is about SCO claiming IBM is stonewalling them about documents SCO thinks should exist in the files of IBM's executives and Board of Directors and about a dispute over who should testify on several issues; and the third motion that will be heard is the newest one, where SCO asks to do 25 additional depositions.

At first, I couldn't figure out what the amended notice meant, because there have been so many discovery motions from SCO, it's a little hard to keep track. The numbers helped me, hopefully, to sort it out. Here's my understanding. #365 is the December motion, originally filed as a sealed document, and then in redacted form, with exhibits [PDF] filed separately; then there is #503, where SCO interprets Judge Wells' orders, particularly the January 18, 2005 Order, to define nonpublic Linux materials more broadly than IBM does. We'll see which interpretation Wells decides to go with. She did order IBM to turn over all nonpublic Linux materials, but what does that entail? And #508 is SCO's most recent request to take 25 additional depositions. So, Judge Wells has the three of them down on her calendar to be heard on the same day, which makes sense, since all three are SCO asking for more discovery.

SCO's most recent 10-Q also helps to figure out what is what. Here's how they characterize the first two of the three motions:

The parties have also now fully briefed the Company’s December 23, 2004 Renewed Motion to Compel Discovery, which seeks to compel IBM’s compliance with prior Court orders relating to IBM’s obligation (1) to produce all documents pertaining to Linux from the files of high-level IBM executives and board members; and (2) to produce witnesses to testify on several topics in two deposition notices that SCO has served on IBM. The Court has not set a hearing date for this motion.

On September 6, 2005, the Company filed a renewed motion to compel IBM to produce information relating to IBM’s work on the source code that IBM contributed to Linux where such information is in IBM’s hands and not publicly available.

That's SCO's position. Hopefully, IBM will be able to get it through Judge Wells' head at the October 7 hearing, if not SCO's attorneys' heads, that Linux is not developed in secret the same way proprietary Unix is developed, as I've explained before. SCO seems to me to greatly misunderstand IBM's role in Linux kernel development, as well as how the kernel is developed. I remember how hard it was for SCO to comprehend the GPL, and this reminds me of that same unwillingness or inability to learn.

Earlier, SCO's whining in discovery was all about AIX. Then it was all about Project Monterey. Now it's all about Linux. Evidently, they can't find what they told the world they already had mountains of, namely evidence that Linux in any way infringes on SCO's code. And now deadlines are looming. They have to finally tell us what IBM has allegedly done, what materials were infringed, very soon. So now, SCO interprets Judge Wells' January 18 order as saying that IBM must turn over all Linux programmer notes, such as were ordered regarding AIX and Dynix:

Based on SCO's review of the discovery to date, IBM persists in its refusal to provide SCO with all non-public information relating to its Linux contributions — particularly the development history (e.g., programmer's notes, comments, design documents, white papers, and iterations, revisions, and interim versions of code contributed to Linux) of any code copied into Linux.

Hang on, there, cowboy. Having a different understanding of what a judge's order means isn't a "refusal" to comply. IBM is allowed to disagree and let the judge decide. SCO has its view, but instead of just saying so, they accuse IBM of a "refusal". I really hate when they do that, and they do that all the time.

Further, they claim that their interpretation is what Wells ordered. I have just reread the January order, again, and the April 19, 2005 order, and while SCO *asked* for that, it isn't what the judge ordered, by my honest reading. IBM doesn't read the orders that way either. Here's the heart of IBM's position, found in a letter attached as Exhibit 2:

Second, with respect to IBM's Linux contributions, the Court expressly ruled in an Order dated March 3, 2004 (and reaffirmed in its April 19, 2005 Order) that IBM is not required to produce to SCO information concerning IBM's Linux contributions insofar as such information is publicly available. The Court only ordered IBM to produce "all non-public Linux contribution information". (April 19, 2005 Order at 5-6.) As detailed in Peter Ligh's July 5 letter to Ted, IBM has fully complied with that obligation.

If you read the January order too, you will understand IBM's position, I think. As usual, SCO characterizes IBM's position as playing games in discovery. But, while Judge Wells may ultimately decide to order further discovery, it's not a "refusal" on IBM's part here. It's a difference of opinion as to what the order requires. We've seen SCO playing the incremental discovery game for a long time, and their mock outrage gets old fast.

I suppose it might be authentic. But then it speaks to a different problem. It reminds me of liars, who invariably assume everyone else in the world lies too. When you tell a liar the truth, they never believe you, because they are always looking for hidden angles. Because they lie, they assume you do too. Two honest people talking have an easier time communicating.

The issue being raised here is discovery gamesmanship. Boies Schiller has been in the news of late themselves, due to what Boies Schiller describes as an "inadvertent" failure to tell clients that the Boies family has a financial interest in a document management company, Amici. That story isn't going away. If anything, it's taking on legs, with the New York Times now covering the story. And, by my reading, it's not so much the seriousness of the matter as it is because a number of law firms who have dealt with Boies Schiller are unhappy with their conduct, specifically in discovery. Some now are complaining, as the Times article puts it, "that Mr. Boies has used Amici to put opponents at a disadvantage while enriching his family." In essence, the firm is being accused of using Amici to play games in discovery:

Lawyers for L. Dennis Kozlowski and Mark H. Swartz, former top executives at Tyco International who now face years of prison time, have accused Mr. Boies, Tyco's lawyer, of engaging in "rank gamesmanship to the defendants' disadvantage" by withholding a million potentially subpoenaed documents from the defense until the trial of the two men was nearly over, and millions more until after their conviction.

Mr. Boies, in an interview yesterday, said one set of issues was worthy of concern - the concerns about the ties to Amici - but dismissed the rest as overzealous advocacy.

"There is a disclosure issue here," Mr. Boies said. "That's a real issue, that's an issue that's worth talking about. But what's happening is a lot of desperate people who are in real trouble are trying to use it as a hook."

That's one interpretation. But even desperate people in real trouble deserve to have access to materials in discovery that might help them at trial. That is, after all, SCO's position in its discovery arguments here. And our legal system is set up, presumably, to give a man a fair shake, based on all the evidence, not based on what games lawyers play. So, while I'm not saying SCO can't persuade the judge, it hasn't, in this document, persuaded me, and by my lights, it has used tactics that don't endear.

The article goes on to describe the Boies Schiller style:

Still, some lawyers who have faced him or others at his firm say that Mr. Boies's firm plays a special brand of hardball. "They have been getting away with a lot of this stuff," said Aaron Richard Golub, who defended a modeling agency against a suit filed by models who were represented by Boies, Schiller. "They just pay the fine, that's all."

You'll remember the class-action lawsuit in May, where the judge reduced Boies Schiller legal fees. David Boies comments on that matter in the Times article:

Mr. Boies said that while mistakes were made in that case, events there had nothing to do with those in other cases, including the more recent disputes with lawyers for Deloitte and lawyers for Mr. Kozlowski and Mr. Swartz.

"The magistrate decided in favor of the plaintiffs and that was affirmed by the district judge," Mr. Boies said, "and the district judge held that we should've produced more documents faster."

But according to an order by Magistrate Judge Henry B. Pitman, lawyers for the plaintiffs did not just move too slowly. They did not comply with court orders instructing them to answer written questions posed by the defendants' legal team, and Judge Pitman imposed a $25,000 penalty on them. In an order issued in June, he summarized the reasons: "Plaintiffs, by their counsel, repeatedly failed to comply with order directing that interrogatories be answered, provided contradictory sets of interrogatory answers with no explanation and provided interrogatory answers that were unquestionably designed to obfuscate rather than illuminate, notwithstanding an earlier order directing the interrogatories be answered."

See what I mean? Ironic, isn't it? Might it be, perhaps, that Boies Schiller just assumes that IBM is playing games in discovery, because Boies Schiller apparently does so? The fact that other law firms are now speaking out and complaining tells me that not all firms play such games in discovery. It also tells me that when you are merciless to others, which is what hardball is, later, when you slip and fall and need mercy -- and all humans do -- nobody feels like giving you any. That's just human nature. And that, in my opinion, may be why the Amici story is growing, instead of dying, the way I expected it to.

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

Brent O. Hatch (5715)
Mark F. James (5295)
HATCH, JAMES & DODGE
[address]
[phone]
[fax]

Stuart H. Singer (admitted pro hac vice)
BOIES, SCHILLER & FLEXNER LLP
[address]
[phone]
[fax]

Robert Silver (admitted pro hac vice)
Edward Normand (admitted pro hac vice)
BOIES, SCHILLER & FLEXNER LLP
[address]
[phone]
[fax]

Stephen N. Zack (admitted pro hac vice)
BOIES, SCHILLER & FLEXNER LLP
[address]
[phone]
[fax]

Attorneys for The SCO Group, Inc.

IN THE UNITED STATES DISTRICT COURT

FOR THE DISTRICT OF UTAH

THE SCO GROUP, INC.

Plaintiff/Counterclaim Defendant,

v.

INTERNATIONAL BUSINESS
MACHINES CORPORATION,

Defendant/Counterclaim-Plaintiff.

[REDACTED] PLAINTIFF'S
MEMORANDUM IN SUPPORT OF
ITS RENEWED MOTION TO
COMPEL


Case No. 2:03CV0294DAK
Honorable Dale A. Kimball
Magistrate Judge Brooke C. Wells

1

Plaintiff The SCO Group, Inc. ("SCO") files this Memorandum in Support of its Renewed Motion to Compel pursuant to Rule 37(b)(2) of the Federal Rules of Civil Procedure.

PRELIMINARY STATEMENT

More than two years after SCO filed this lawsuit, IBM still has not produced the compete development history of its Linux contributions, which stand at the center of this case.

In an attempt to limit its own discovery obligations and deny SCO the basic materials it needs to develop its claims, IBM has repeatedly tried to persuade the Court that SCO's case implicates UNIX and Linux source code and no more. SCO has by now repeatedly demonstrated the fallacy of IBM's position by establishing the importance of the development history of the code that IBM and third parties have copied into Linux.

The Court has agreed with SCO on this issue in three successive Orders. See Order of March 3, 2004, at 4-5; Order of January 18, 2005, at 9-12, 15-17; Order of April 19, 2005, at 4-6. In those Orders, pointing to the fundamental importance of the programming history of the code ultimately copied into Linux, the Court ordered IBM to produce all of its Linux-development information, including the code of IBM's AIX and Dynix operating systems and their development histories. Id.

Yet IBM has construed the Court's Orders to cover programming-history information only for AIX and Dynix — not Linux. In its Motion for Reconsideration of the Court's January 18 Order, IBM complained that the Court had not specifically ordered IBM to comply with SCO's requests for Linux-development information.(1) See IBM Motion for Reconsideration of

2

the Court's January 18 Order at 3 n.4. Rejecting IBM's argument, the Court ruled again that its "prior orders" had directed IBM to provide "ALL non-public Linux contribution information," and reordered IBM to comply within 75 days. April 19 Order at 5.

Having now failed to provide all such information, even while it claims to have produced it, IBM apparently seeks to relitigate this seemingly closed issue yet again. IBM produced AIX and Dynix programming history pursuant to the Court's January and April 2005 Orders. But IBM has not produced any significant programmer's notes, comments, design documents, white papers, and iterations, revisions, or interim versions of its Linux contributions from IBM's multi- billion dollar Linux Technology Center, through which it has made "the vast majority of those contributions." Letter dated October 10, 2003, from T. Shaugnessy to M. Heise, at 3 (Exh. 1).

SCO asks the Court again to compel IBM immediately to produce the complete history of IBM's Linux-development effort with respect to all contributions that IBM has made to Linux, up to the present, and to consider the propriety of imposing sanctions against IBM for its continued and persistent refusal to provide such Court-ordered and plainly relevant information.

3

ARGUMENT

I. THE COURT HAS REPEATEDLY ORDERED IBM TO PRODUCE LINUX-DEVELOPMENT INFORMATION IN RESPONSE TO SCO'S DISCOVERY REQUESTS

A. SCO Has Repeatedly Requested All Documents Concerning IBM's Linux-Development Work

In its First Request for Production of Documents and First Set of Interrogatories (June 24, 2003), SCO requested all of IBM's Linux-contribution information:

  • Interrogatory No. 5 — "Identify all IBM or Sequent personnel that work or worked on developing source code, derivative works, modifications or methods for AIX, Dynix, and Linux, specifying for each person their precise contributions to each."

  • Document Request No. 11 — "All contributions made without confidentiality restrictions by IBM or anyone under its control including, but not limited to, source code, binary code, derivative works, methods, and modifications to Open Source Development Lab, Linus Torvalds, Red Hat or any other entity."

  • Document Request No. 29 — "All documents concerning any UNIX source code, derivative works, modifications or methods disclosed by IBM to any third party or to the public."

  • Document Request No. 32 — "All documents concerning any UNIX source code, derivative works, modifications or methods found in Linux, open source, or the public domain."

  • Document Request No. 35 — "All documents concerning any contributions to Linux or to open source made by IBM and/or Sequent."

SCO's First Request for Production of Docs. and First Set of Interrogs. (6/24/03) at 8, 10, 13.

In moving to compel a complete response to Interrogatory No. 5 and Document Request No. 11, SCO noted IBM's failure to respond to "the most important part of the interrogatory" — the request for the specific files and development work that each IBM programmer contributed to the development of AIX, Dynix, and Linux. See SCO Mem. in Support of SCO Motion to Compel, at 9, 13. In support of its request, SCO showed that the 15-year development process

4

that separated IBM's licensing of UNIX and IBM's ultimate misuse of that licensed code implicates the code's entire development history — from UNIX, to AIX and Dynix, to Linux. Therefore, SCO argued, IBM "should be required to identify and produce all of its contributions and development work in Linux." Id. at 9 (underline added).

In addition to detailing the outstanding Linux-development discovery in its November 2003 Motion to Compel, SCO filed its Memorandum Regarding Discovery in May 2004, which outlined for the Court the specific IBM materials that SCO sought. SCO again demonstrated the relevance of the code's programming history with respect to SCO's analysis of IBM's Linux- development efforts. See SCO Mem. Re Disc. at 8-9.

Most recently, when IBM again sought to limit its discovery obligations by asking the Court to reconsider the January 18 Order, SCO explained:

"In many instances there may well have been a development process — including one conducted by IBM or Sequent programmers immersed in SCO's proprietary UNIX — between the selection of AIX or Dynix material for Linux and the actual contributions to Linux. SCO requires access to that development history (including both code and related documentation) for exactly the same reason this Court has held that SCO needed access to the material evidencing the developers and development process of Dynix and AIX themselves."
SCO Opp. to IBM Motion for Reconsideration of the Court's January 18 Order at 9 n.4 (emphasis added).

B. IBM Resisted Producing Documents Concerning Its Linux-Development Work

IBM has persistently denied SCO this discovery. IBM first argued that producing the development history of its Linux contributions was too difficult, objecting to SCO's requests as overbroad and unduly burdensome. See IBM's Responses to SCO's First Set of Interrogs. and Doc. Reqs. (8/13/03) at 5, 21-26. In response to SCO's Motion to Compel, IBM again refused to

5

comply with SCO's request for IBM to "identify and produce all of its contributions and development work in Linux." IBM Opp. to SCO Motion to Compel at 3 (emphasis in original). IBM maintained that SCO first should be required to identify particular lines of code that IBM misused, and only then would IBM "provide SCO with the development history of that code to the extent possible." Id.

In addition to arguing that the requested production was too burdensome, IBM tried to convince this Court that any "development history of the code IBM in fact contributed to Linux is therefore irrelevant." IBM Response to SCO Mem. re Disc. at 14 n.5. At hearings in September and October 2004, IBM argued that AIX and Dynix programming history was entirely irrelevant to this case, since "this discovery is not in any way necessary to prove a claim for a public breach of contract." Tr. (10/19/94) at 57. IBM claimed:

"It's about Unix and Linux. The only road map that counts, Your Honor, is what's in Unix and what's in Linux. And that's the comparison. And you can compare Microsoft Word source code or AIX source code or Dynix source code all you want. It is entirely irrelevant to the question of whether Unix is substantially similar to Linux."
Tr. (9/15/04) at 116. The Court rejected IBM's argument and (re)ordered IBM to produce the outstanding discovery. See January 18 Order at 9-12, 15-17. In response, IBM again recast its argument. IBM argued that — while specifically ordered to produce all programmer's notes, design documents, and code versions and changed for AIX and Dynix — it was not obligated to produce any programming history for its Linux contributions. In fact, IBM complained in its most recent motion that "having been given an inch," SCO sought "to take a mile" by suggesting that the Order "requires IBM to detail its Linux contributions." IBM Motion for Reconsideration of the Court's January 18 Order at 3 n.4.

6

C. The Court Has Repeatedly Ordered IBM To Produce Its Linux-Development Work

After considering both parties' initial motions to compel on these issues, the Court ordered IBM "to provide further responses to SCO's interrogatory numbers two, five, and eleven," specifying that IBM must also provide "to SCO any and all non-public contributions it has made to Linux." Order (3/3/04) at 4-5 (emphasis added).

Almost one year later, ruling on SCO's Renewed Motion to Compel, the Court again ordered IBM to supplement its response to Interrogatory No. 5. See Order (1/18/05) at 15-16. In doing so, the Court settled an issue that had pervaded discovery disputes: Could IBM withhold from SCO everything but source code, or was IBM also obligated to provide SCO with the complete programming history of the allegedly misused code? Citing the "general principle that 'at the discovery stage, the concept of relevance should be construed very broadly,'" the Court ordered IBM to provide the complete development history of the code in question, including "programmer's notes, design documents, white papers, the comments and notes made by those who did the changes," in addition to "the changes to the code" and "the names and information concerning those individuals who made the changes." Id. at 9, 15-16.

In its most recent Order, the Court addressed IBM's unilateral attempt to "modify its discovery obligations" by limiting the scope of Linux-development work that it was required to produce. April 19 Order at 5. While the Court deferred issuing a "premature" ruling because IBM had not yet completed, and could still cure any deficiency in, the required production, the Court nevertheless rejected IBM's position, warning that "prior orders make it clear that IBM is to provide ALL non-public Linux contribution information." Id. (emphasis in original). The Court also reiterated that the Linux-development-work "production is to be specific in nature"

7

and is "inherent within the court's previous orders because it would be considered 'non-public' Linux information that is available to IBM." Id. at 6.

D. IBM Has Not Produced All Documents Concerning Its Linux-Development Work

The Court's April 2005 Order directed IBM to complete its production by August 1. That deadline has passed, and IBM has recently confirmed its view, both in writing and in conversations with counsel for SCO, that IBM has produced all of the "non-public Linux contribution information" that it is obligated to produce. See, e.g., Letter dated July 19, 2005, from T. Shaughnessy to B. Hatch, at 1 (Exh. 2). Based on SCO's review of the discovery to date, IBM persists in its refusal to provide SCO with all non-public information relating to its Linux contributions — particularly the development history (e.g., programmer's notes, comments, design documents, white papers, and iterations, revisions, and interim versions of code contributed to Linux) of any code copied into Linux.(2) To illustrate:

  • Although IBM has (inadvertently, it appears) produced numerous Linux source-code files with its CMVC production, IBM has not produced the many more files from the Linux development tree that (it stands to reason) must also be stored in CMVC.(3) Under the Court's orders, IBM must produce all non-public Linux-development information stored in CMVC or in any other revision control system.

    8

  • Although IBM contributed the AIX Journaling File System (JFS) to Linux, IBM's CMVC production contains no source code concerning the origin or development of the JFS Linux patch, including the programming history necessary to substantiate IBM's claim that it independently developed JFS code as part of its OS/2 operating system.

  • Although IBM is a founding member and active participant in the Open Source Development Lab (OSDL), IBM's production does not contain any code or internal e-mails concerning Linux-development work in conjunction with the OSDL. Fewer than 40 documents in IBM's May and July 2005 productions even reference the OSDL.

Aside from the conspicuous absence of significant Linux-development programming history, materials that IBM has produced confirm the existence of discovery that IBM has withheld. For example:

  • REDACTED

  • REDACTED

    9

  • IBM's CMVC production includes database entries for at least 450 JFS-related files but does not contain the files themselves. Instead, IBM has replaced them with a "non-responsive" placeholder. In correspondence with IBM counsel, SCO requested that IBM produce these and other missing files, but IBM responded that the files "are neither part of nor are related to the AIX operating system, and thus were properly withheld from production."(7) Contrary to IBM's view, materials related to JFS are obviously responsive to SCO's requests for Linux contributions information, and must be produced pursuant to the Court's orders.

  • REDACTED

CONCLUSION

This Court has repeatedly held that Linux-development materials are relevant and has ordered IBM to produce them. SCO respectfully requests that the Court again compel IBM's immediate compliance with the Court's orders to produce all documents concerning IBM's development work for Linux with respect to all contributions that IBM has made to Linux, up to

10

the present, and that the Court consider the propriety of imposing sanctions against IBM for its continued and persistent refusal to provide such Court-ordered and plainly relevant information.

DATED this 2nd day of September, 2005.

Respectfully submitted,

HATCH, JAMES & DODGE, P.C.
Brent O. Hatch
Mark F. James

BOIES, SCHILLER & FLEXNER LLP
Robert Silver
Stuart H. Singer
Stephen N. Zack
Edward Normand

By (signature of Brent Hatch)

Counsel for The SCO Group, Inc.

11

CERTIFICATE OF SERVICE

Plaintiff The SCO Group, Inc. hereby certifies that a true and correct copy of the foregoing Plaintiff's Memorandum in Support of SCO's Renewed Motion to Compel was served by U.S. mail or hand delivery on Defendant International Business Machines Corporation on the 6th day of September, 2005:

By U.S. Mail:

David Marriott, Esq.
Cravath, Swaine & Moore LLP
[address]

Donald J. Rosenberg, Esq.
[address]

By Hand Delivery:

Todd Shaughnessy, Esq.
Snell & Wilmer LLP
[address]

(signature)

12

EXHIBIT 1

13

[Snell & Wilmer letterhead]

October 10, 2003

VIA FACSIMILE AND U.S. MAIL

Mark J. Heise
BOIES, SCHILLER & FLEXNER, LLP
[address]

Re: SCO v. IBM / IBM v. SCO — IBM's Discovery Responses

Dear Mark:

We have reviewed your letter of September 22, 2003, relating to our telephone call on September 18, 2003. This letter responds thereto.

In answer to your questions about when SCO can expect to receive documents responsive to specific requests, IBM has produced, and will continue to produce, responsive, non-privileged documents as they are collected, reviewed, and processed.

I. General Objections

General Objection Nos. 1-3, 5, 7, 10. I believe my September 15 letter to you, and your September 22 letter to me, are consistent. Please advise if you disagree.

General Objection No. 4. As stated in my September 15 letter to you, IBM has attempted to collect, where appropriate, documents dating as far back as January 1, 1985. If there are specific categories of documents for which you would like us to search for older documents than that we are producing, please let us know so that we can discuss. Our accompanying letter regarding SCO's discovery responses details the IBM document requests for which we request that SCO perform a broader search.

General Objection No. 6. As I made clear in subsequent telephone calls and correspondence, your identification of terms such as NUMA, RCU and SMP is insufficient to particularize the code and other alleged trade secrets which SCO alleges IBM has wrongfully contributed to Linux. As you know, we have moved to compel specific answers to IBM's interrogatories and we understand that SCO intends to

14

supplement its answers to interrogatories. We do not believe this issue can be resolved until we have adequate answers to our interrogatories, and a detailed and specific list of the trade secrets or confidential information that SCO contends IBM has misappropriated.

General Objection No. 8. I believe my September 15 letter to you, and your September 22 letter to me, are consistent — as we discussed, IBM is not, as a general matter, attempting to collect documents that are generally available to the public, but to the extent they are among the documents we have collected, we are not withholding them on this basis.

General Objection No. 10. I believe my September 15 letter to you, and your September 22 letter to me, are consistent. Please advise if you disagree.

General Objection No. 11. As we discussed, and as my September 15 letter explains, IBM has attempted to discern the documents SCO seeks, and to locate those documents for production. As my letter also confirms, IBM will produce responsive, non-privileged documents despite this objection. The issue with respect to Intel processors is addressed in Response to Request Nos. 40-41 below.

General Objection No. 15. I believe my September 15 letter to you, and your September 22 letter to me, are consistent. Please advise if you disagree.

General Objection No. 17. This issue is resolved by the entry of the Stipulated Protective Order.

General Objection No. 18. Agreed.

General Objection No. 19. During the telephone call, we confirmed that we were not withholding documents on the basis that the documents relate to prior versions, releases and updates of AIX or Dynix, or on the basis of code being a derivative of AIX or Dynix, rather than AIX or Dynix themselves. However, since SCO has not yet specified the code that it alleges IBM has wrongfully contributed to Linux, we remain unclear as to the scope and meaning of SCO's requests for "modifications", "methods", and "derivative works". As indicated in General Objection No. 6 above, we do not believe that this issue can be resolved until we have SCO's supplemental answers to interrogatories identifying with specificity the trade secrets or other confidential information at issue. With respect to the other issues you raise here, see our Response to Request Nos. 2 and 3 below.

General Objection No. 20. The definition of "IBM" for purposes of our responses includes officers. We are not aware of any directors who might have relevant

15

information. We agree, for purposes of the discovery requests at issue, to interpret "IBM" to include Sequent, to the extent Sequent materials are in IBM's custody, possession, or control.

General Objection Nos. 21-22. See our response re: General Objection Nos. 6, 19 above.

II. Document Responses

Response to Request No. 1. Responsive, non-privileged documents are being produced, subject to our agreement on General Objection Nos. 1-3, 5, 7, and 10 set forth above.

Response to Request Nos. 2 and 3. My e-mail to you of October 8 addressed the open issues regarding these requests. As detailed in that note, IBM will commence production of source code for the AIX and Dynix base operating systems once the process for notification of third parties (as specified in Paragraph 10 of the Stipulated Protective Order) is exhausted.

Response to Request Nos. 4-6, 7-9. Agreed; subject to General Objection Nos. 22 and 6, responsive, non-privileged documents will be produced.

Response to Request No. 10. I believe my September 15 letter to you, and your September 22 letter to me, are consistent. Please advise if you disagree.

Response to Request No. 11. Your request, even as narrowed, remains overly broad and unduly burdensome. Until SCO specifies the wrongful contributions IBM has allegedly made to Linux in violation of SCO's alleged confidentiality rights, it is very difficult to make a reasonable assessment as to the proper scope of discovery in this case. In the absence of that clarification, we have nevertheless attempted to conduct a reasonable search for documents that relate to IBM's open-source contributions to Linux. The vast majority of those contributions are made through the LTC; the OSSC is the corporate clearinghouse for those contributions. Our searches to date have thus included individuals in both of those groups, as well other potential sources of documents relating to IBM's contributions to Linux that have come to our attention. We are not limiting our searches to any particular geographic area — indeed, they have already included individuals residing in Beaverton, OR, Austin, TX, and a variety of other IBM locations. Our efforts to identify and collect documents responsive to this request are continuing, and we believe will be facilitated by adequate answers to our interrogatories.

16

Response to Request Nos. 12-14. IBM stands by its objections to these requests. We agreed that we will provide documents sufficient to identify the third parties, if any, to whom IBM provided UNIX, AIX, or Dynix source code.

Response to Request Nos. 15-17. Both parties have agreed to exchange supplemental answers containing names of employees and general subject matter of knowledge on October 10, 2003.

Response to Request No. 18. We have agreed to produce documents, subject to our agreement concerning General Objection Nos. 1-3, 5, 7, and 10 set forth above, and with the understanding that Mr. Palmisano is among those whose files were reviewed. I believe that resolves this issue. If not, please advise.

Response to Request Nos. 19-25. I believe my September 15 letter to you, and your September 22 letter to me, are consistent. Please advise if you disagree.

Response to Request Nos. 26-27. As we discussed, IBM stands by its objection that these requests, as phrased, are overbroad and unduly burdensome. We have agreed, however, to provide supplemental answers to interrogatories identifying the names of employees and general subject matter of knowledge on October 10, 2003.

Response to Request No. 28. We have agreed that IBM will search for responsive, non-privileged documents from IBM employees with significant involvement in technical, business development, and contractual aspects of Project Monterey. During our call, we agreed that it would be unduly burdensome to require IBM to search for this information from every employee who had any involvement in Project Monterey.

Response to Request Nos. 29-31. With respect to these requests, General Objection Nos. 6 and 19 remain unresolved. We have agreed, however, to produce non-privileged documents identified through a reasonable search sufficient to identify Unix, AIX, or Dynix source code disclosed by IBM to a third party or the public.

Response to Request Nos. 32-34. Agreed.

Response to Request No. 35. We confirmed that we will conduct a reasonable search and produce, from the files of LTC, OSSC, or other personnel who may have relevant information, responsive, non-privileged documents relating to Unix, AIX or Dynix source code contributed by IBM to open source, or relating to open source contributions by IBM to Linux.

17

Response to Request No. 36. We had previously agreed to provide organizational and/or personnel charts for particular units involved in AIX, Dynix, and Linux development or maintenance. Historical organizational and/or personnel charts for Project Monterey, to the extent such documents still exist, will be captured in our production in response to your Request No. 28. We are not familiar with the "Project Gemini" referenced in your letter; please provide clarification.

Response to Request No. 37. Agreed.

Response to Request Nos. 38-39. With your clarification and limitations, I believe we are in agreement concerning the documents IBM agrees to produce.

Response to Request Nos 40-41. These requests remain overbroad and unduly burdensome despite your proposed limitation. For example, the Dynix/ptx operating system ran only on Intel processors. Your request for "all documents concerning IBM's use of [Dynix] on Intel processors" would literally call for the production of nearly every single document in the company relating to Dynix/ptx. Please clarify the types of documents you are looking for so that we can have a meaningful discussion about how to narrow these requests.

Response to Request No. 42. See Response to Request No. 11 above.

Response to Request Nos. 43-52. Agreed.

III. Interrogatory Responses

We agreed to exchange supplemental answers to interrogatories identifying the names of employees, including officers, with a general description of the subject matter of their knowledge on October 10.

Very truly yours,
(signature of Todd Shaugnessy)
Todd Shaughnessy

cc: Brent Hatch
David Marriott
Peter Ligh

18

EXHIBIT 2

19

[Snell & Wilmer letterhead]

July 19, 2005

VIA FACSIMILE AND U.S. MAIL

Brent O. Hatch
HATCH, JAMES & DODGE, P.C.
[address]

Re: SCO v. IBM; IBM v. SCO

Dear Brent:

I write in response to Ted Normand's July 14, 2005 letter to David Marriott.

First, Ted's concern that IBM has withheld pre-1991 AIX source code is unfounded. To the extent there is AIX source code in CMVC that was written prior to 1991 and maintained in CMVC, we have produced it. We have repeatedly searched for, but have been unable to find, any pre-1991 AIX source code or revision control information other than that which may be in CMVC. Please provide us with the basis for Ted's statement: "Our information is that IBM does possess the source code for all versions of AIX prior to 1991". If you have any specific information about where source code for versions of AIX prior to 1991 are located within IBM, let us know and we will follow up on it.

Second, with respect to IBM's Linux contributions, the Court expressly ruled in an Order dated March 3, 2004 (and reaffirmed in its April 19, 2005 Order) that IBM is not required to produce to SCO information concerning IBM's Linux contributions insofar as such information is publicly available. The Court only ordered IBM to produce "all non-public Linux contribution information". (April 19, 2005 Order at 5-6.) As detailed in Peter Ligh's July 5 letter to Ted, IBM has fully complied with that obligation.

The issues you raise under the headings of "Approximately 245,325 Missing Files", "Approximately 152,887 Un-Catalogued Files", and "Removal of CMVC Change-Log History" require more investigation and analysis. Because some of the people at IBM who were involved in prepairing the CMVC data for production to SCO several months ago are currently on vacation, I am unable to provide a response to those issues at this time. We will provide you with responses to those issues as soon as we can.

20

Very truly yours,

(signature of Todd Shaugnessy)
Todd M. Shaughnessy

TMS:dw
cc: Edward Normand
David Marriott
Peter Ligh
Amy Sorenson

21

EXHIBIT 3

(FILED UNDER SEAL)

22

EXHIBIT 4

(FILED UNDER SEAL)

23

EXHIBIT 5

(FILED UNDER SEAL)

24

EXHIBIT 6

(FILED UNDER SEAL)

25

EXHIBIT 7

(FILED UNDER SEAL)

26

EXHIBIT 8

(FILED UNDER SEAL)

27

EXHIBIT 9

28

[Boies Schiller & Flexner letterhead]

July 14, 2005

By Facsimile and First-Class Mail
David Marriott, Esq.
Cravath, Swaine & Moore LLP
[address]

Re: SCO v. IBM, Civil No. 2:03CV-0294DAK

Dear David:

I write to identify deficiencies in the production of the CMVC (AIX) database (the "CMVC/AIX Production") that IBM has made in response to the Magistrate Court's Order Regarding SCO's Renewed Motion to Compel Discovery dated January 18, 2005 (the "January 18 Order").

As an initial matter, IBM has not produced any AIX source code prior to 1991. Todd Shaughnessy has asserted that "CMVC has been used in AIX development since 1991. Other than the AIX source code stored in CMVC, IBM does not maintain revision control information for AIX prior to 1991." Affidavit of Todd Shaughnessy (May 3, 2005) ("Shaughnessy Aff.") ¶ 7. That might be true, but irrespective of the availability of revision control information prior to 1991, SCO has asked IBM to produce AIX code prior to 1991, in whatever format IBM has stored such code, and the Court has ordered IBM to produce it. Our information is that IBM does possess the source code for all versions of AIX prior to 1991. Please let me know by July 19 whether IBM will produce all code, notes and revision history for AIX prior to 1991.

Similarly, SCO asked IBM to identify the contributions that IBM has made to Linux, the specific identity of the contributors, and the specific contributions that each contributor made. Although IBM claims to have identified IBM's non-public contributions to Linux and the contributors (as set forth in Peter Ligh's July 5 letter to me), IBM has not identified the specific contributions that each contributor made. Nor has IBM identified the specific contributions that each contributor made with respect to IBM's public contributions to Linux. Under the Magistrate's March 2003, January 2005, and April 2005 Orders, IBM must produce that category of information. Please let me know by July 19 whether and when IBM will identify the specific IBM contributions (both public and non-public) to Linux that each contributor has made.

In addition, based on our review of the CMVC/AIX Production and as detailed below, we disagree with Mr. Shaughnessy's assertion that the CMVC/AIX Production contains "one hundred percent (100%) of the source code in CMVC that is part of or

29

related to AIX (including the operating system itself, development tools, documentation, and test programs)." Shaughnessy Aff. ¶ 10.

Approximately 245,325 Missing Files

We have identified approximately 245,325 entries in the CMVC database for which no file exists in the file trees associated with AIX (the "Missing Files").(1) Most of the Missing Files are marked "non-responsive." Since "non-responsive" is not a term used in software programming, we infer that IBM removed the files marked "non-responsive" because IBM concluded that the files were not responsive to the January 18 Discovery Order. Please promptly advise me if our conclusion is inaccurate. Notwithstanding the designation of "non-responsive," a large number of the file names associated with the Missing Files relate to AIX kernel code, AIX shared libraries, and AIX compilers — source code that clearly is at issue in the litigation.(2) It appears, for example, that at least 450 of the Missing Files relate to the Journaling File System (JFS).(3) As SCO has explained to the Court, IBM contributed JFS to Linux without modification, and JFS is one of the components of AIX centrally at issue in this litigation.

Mr. Shaughnessy has stated that in preparing its CMVC/AIX Production, IBM "removed the contents of the source files and programmer's notes that did not relate to AIX." Shaughnessy Aff. ¶ 18. As the foregoing files are part of or relate to AIX, we hereby request that you promptly produce all of the Missing Files and any and all other files related to AIX otherwise not produced. We also request that you produce the log history showing the "contents" of the source files and programmer's notes that IBM removed from the CMVC/AIX Production so that we may verify the extent to which AIX-related files may have been removed. The Court ordered IBM to produce "ALL AIX information" on CMVC, January 18 Order at 10; our request for the log is a logical and appropriate extension of IBM's discovery obligation.

30

Approximately 152,887 Un-Catalogued Files

We have identified approximately 152,887 files that do appear in the AIX file tree, but do not have a corresponding indexed entry in the CMVC database (the "Un-Catalogued Files"). IBM uses the CMVC database entries to provide ready access to, and navigation of, the files in CMVC. Without the corresponding CMVC database entries, it is impossible to access or review the Un-Catalogued Files in the way that IBM has maintained them in the ordinary course of business. In addition, barring an exhaustive manual examination of these files, it is very difficult for us to determine to which AIX components each file belongs, and it is usually impossible for us to know the name of each file. We therefore request prompt production of the CMVC database entries that correspond properly to the 152,997 Un-Catalogued Files, and any and all database entries related to AIX otherwise not produced.

Removal of CMVC Change-Log History

Upon exit of any command-line shell in AIX, the system apparently retains an annotated history of commands run by the user. We have reviewed the user history archive of the copy of AIX produced in the CMVC/AIX Production and have identified (among others) the following commands that IBM appears to have run in preparing the CMVC/AIX Production:
  • rm bash_history. This command removes the history of past events. Such history is useful for repeating past commands. We request that you promptly produce all shell history resulting from your preparation of the CMVC/AIX Production so that we may identify more precisely the acts taken to prepare the CMVC database for production. This request includes the bash_history for all root users as well as all administrators.

  • rm log. This command removes the log file. The log demonstrates a proper history of commands run against the operating system. We request that you promptly produce all log files related to your preparation of the CMVC/AIX Production so that we may identify more precisely the acts taken to prepare the CMVC database for production. This request includes the log file for all root users as well as all administrators.

Thank you for your prompt attention to these issues. As I am sure you appreciate, IBM's complete production of all of the AIX-, CMVC-, and Linux-related material that

31

SCO has requested and that the Court has repeatedly ordered IBM to produce is a prerequisite to SCO's compliance with the Court's revised scheduling order.

Sincerely,
(signature of Edward Normand)
Edward Normand

cc: Brent Hatch
Todd M. Shaughnessy

32

EXHIBIT 10

33

(Snell & Wilmer letterhead)

August 8, 2005

VIA FACSIMILE AND U.S. MAIL

Brent O. Hatch
HATCH, JAMES & DODGE, P.C.
[address]

Re: SCO v. IBM; IBM v. SCO

Dear Brent:

I write to follow up on my July 19, 2005 letter to you, responding to Ted Normand's July 14, 2005 letter to David Marriott.

First, with respect to the issues raised under the heading "Approximately 245,325 Missing Files," Ted's letter does not identify, and we have been unable to determine, how he arrived at the number 245,325. However, as I stated in my May 3, 2005 declaration, we did not produce to SCO any components in CMVC that are unrelated to AIX or its code, internal design, or methods, such as code and information relating to IBM hardware, firmware, manufacturing-related components, and middleware and other software applications. Components that were withheld were replaced with a "non-responsive" placeholder. As we have repeatedly informed you, the CMVC server at IBM that contains source code and information related to AIX also contains a large amount of source code and other material that is neither part of, nor related to, AIX. Furthermore, directory names such as "/src/bos/kernel/base" that are recorded in the CMVC database are directory names that do not necessarily correspond to the AIX operating system. We have investigated each of the specific "non-responsive" components identified in Ted's letter, and have confirmed that these components are neither part of nor related to the AIX operating system, and were thus properly withheld from production. If you have a reasonable basis to believe that any other specific "non-responsive" files contain information related to the AIX operating system, please identify these files and why you believe they may be related to AIX. As for Ted's request that we provide a "log history showing the 'contents' of the source files and programmer's notes that IBM removed from the CMVC/AIX Production," IBM is not obligated to produce such information. Nor, in any event, have we maintained such a comprehensive "log history."

34

Second, with respect to the issues raised under the heading "Approximately 152,887 Un-catalogued Files," Ted's letter does not identify, and we have been unable to determine, how he arrived at the number 152,887 (or, as stated later in the letter, 152,997). Based on Ted's description of these "Un-Catalogued Files," however, it appears that he is referring to instances in which there is a source code file without any corresponding entry in the CMVC database describing that source code file. As you may know, CMVC consists of two separate layers: the underlying source code files (SCCS files), and a database directory of those files. We believe that during normal development activity, operations such as creating a new file and then undoing the create can leave behind an SCCS file, with no corresponding information in the CMVC database directory. In preparing the CMVC server for production, we used the database to identify which CMVC components were neither part of nor related to AIX, and removed the corresponding source files. We believe the "Un-Catalogued Files" likely include both AIX-related files, and non-AIX-related files. However, because the "Un-Catalogued Files" did not have database entries allowing us to determine whether they were or were not part of or related to AIX, we did not delete any of the underlying source code files, and produced the entirety of these "Un-Catalogued Files" to SCO. In short, the CMVC data we produced was, if anything, overinclusive.

Third, with respect to Ted's request that IBM produce all shell histories and log files related to our preparation of the CMVC server, IBM does not have a comprehensive collection of the shell histories and log files. In any case, IBM is not obligated to produce such information, nor was it ordered by the Court to do so.

Very truly yours,
(signature of Todd Shaughnessy)
Todd M. Shaughnessy

TMS:dw
cc: Edward Normand
David Marriott
Peter Ligh
Amy Sorenson

35

EXHIBIT 11

(FILED UNDER SEAL)

36


Memorandum Footnotes:

  1. The January 18 Order to produce all versions of AIX and Dynix code only highlights the flaw in IBM's argument. AIX and Dynix are themselves relevant to this litigation because they are stages in the development of Linux code. By ordering IBM to produce "all versions and changes to AIX and Dynix," January 18 Order at 9-10, the Court essentially ordered IBM to produce the development history for these interim stages in the development of Linux. If the Court saw fit to order that level of development information, then it certainly follows that the Court's Orders cover programming histories for the most proximate stages in Linux development.

    (Back to the main text)

  2. In addition, IBM has reduced the amount of publicly available Linux-contribution information by removing its Linux-patch contributions from its website (http://www-124.ibm.com/linux/projects/) some time between February 24 and March 8, 2005 ("the removal date"). IBM's obligation to provide SCO with accurate and up-to-date Linux-contribution information is inherent to SCO's requests and to this Court's orders. As a result, any Linux patches that IBM submitted after the removal date constitute non-public Linux-contribution information that IBM must produce pursuant to the Court's orders. Nonetheless, IBM's completed production does not contain source code or development history for any patches IBM contributed to Linux after the removal date.

    (Back to the main text)

  3. IBM has also withheld the index entries (from the CMVC database) that describe and correspond to the produced Linux files. These entries are necessary (among other things) to access and review the code files in the manner that IBM maintained them in the ordinary course of business.

    (Back to the main text)

  4. REDACTED

  5. REDACTED

  6. REDACTED

  7. Letter dated July 14, 2005, from E. Normand to D. Marriott, at 2 (Exh. 9); letter dated August 8, 2005, from T. Shaughnessy to B. Hatch, at 1 (Exh. 10).

    (Back to the main text)

  8. REDACTED


Exhibit 9 Footnotes:

  1. The file trees associated with AIX in CMVC are /family/aix/vc/ and /family/aix/admin/.

    (Back to Exhibit 9)

  2. For example, the file marked as "Non-responsive5742" corresponds to 252 files in the directory src/bos/kernel/base/. The file marked "Non-responsive575" corresponds to 32 files in the directory src/box/kernel/fs/. The file marked "Non-responsive5760" corresponds to 3 files in the directory /src/box/kernel/init/. The file marked "Non-responsive5761" corresponds to 3 files in the directory src/box/kernel/lib/licsys/, and 4 files elsewhere in src/bos/kernel/lib. The file marked "Non-responsive5765" corresponds to 41 files in the directory src/bos/kernel/mm/. The file marked "Non-responsive5768" corresponds to 23 files in the directory src/bos/kernel/lib/libcnet/.

    (Back to Exhibit 9)

  3. The Missing Files related to JFS were identified by a search for files whose names contain the characters "jfs".

    (Back to Exhibit 9)


  View Printable Version


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

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