decoration decoration

When you want to know more...
For layout only
Site Map
About Groklaw
Legal Research
ApplevSamsung p.2
Cast: Lawyers
Comes v. MS
Gordon v MS
IV v. Google
Legal Docs
MS Litigations
News Picks
Novell v. MS
Novell-MS Deal
OOXML Appeals
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v Novell
Sean Daly
Software Patents
Switch to Linux
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
IBM's Response to SCO's Memorandum Regarding Discovery - as text
Friday, June 25 2004 @ 07:13 AM EDT

Here is IBM's RESPONSE TO SCO'S MEMORANDUM REGARDING DISCOVERY. If you wish to read SCO's document this is answering, it is here , SCO's Memorandum Regarding Discovery.

We have Frank Sorenson to thank for transcribing this with me so quickly, and he did by far the bulk of it. Thank you, Frank.


Alan L. Sullivan (3152)
Todd M. Shaughnessy (6651)
[address, phone, fax]

Evan R. Chesler (admitted pro hac vice)
David R. Marriott (7572)
[address, phone, fax]

Attorneys for Defendant/Counterclaim-Plaintiff
International Business Machines Corporation








Defendant, Counterclaim-Plaintiff.



Civil No. 2:03CV-0294 DAK

Honorable Dale A. Kimball

Magistrate Judge Brooke C. Wells


Defendant/Counterclaim-Plaintiff International Business Machines Corporation ("IBM") respectfully submits this memorandum in response to Plaintiff/Counterclaim-Defendant The SCO Group, Inc.'s ("SCO") Memorandum Regarding Discovery.

Preliminary Statement

Nearly eleven weeks after IBM complied with this Court's order of March 3, 2004 (six weeks earlier than the Court-imposed deadline), and after IBM moved for partial summary judgment, SCO contends that the hundreds of millions of lines of source code and the nearly one million pages of paper IBM has produced are not enough. IBM has already produced—to the extent IBM has the information—the source code for every version and release of IBM's AIX and Dynix programs for the past five years. SCO now seeks production of all source code ever associated with AIX and Dynix since 1984, even if the code never appeared in an actual version or release of AIX or Dynix, and many millions of pages of additional documents concerning the development of the AIX and Dynix source code over the last 20 years. SCO's reqest is baseless and should be denied.

In it's "Memorandum Regarding Discovery", 1 SCO argues that it needs: (1) "all revision control system information (including documents, data, logs, files and so forth) for AIX, Dynix/ptx, ptx, and Dynix from 1984 to the present"; (2) "source code and log information for all interim and released versions of AIX, Dynix, ptx and Dynix/ptx from 1984 to present"; and (3) "all design documents, whitepapers and programming notes, created from 1984 to the present, related" to 22 AIX and Dynix/ptx functionalities. (SCO Mem. at 9.) This list goes well beyond what SCO sought in its motion to compel and seeks materials that are irrelevant and unnecessary to this case.

SCO already has all the materials that it needs to determine whether or not IBM violated its contractual obligations or infringed on SCO's purported UNIX copyrights by allegedly having improperly "dumped" source code into Linux: the source code for the UNIX programs at issue and the source code that IBM contributed to Linux. Indeed, SCO stated publicly, before it had any discovery from IBM, that SCO had all the evidence it needed to support its case against IBM. In addition, SCO has declared in this Court, without the benefit of any discovery, that companies similarly situated to IBM, Hewlett-Packard Co. ("HP") and Sun Microsystems, Inc. ("Sun"), are in compliance with their contractual obligations. The notion that SCO "needs" the mass of information that it now seeks is therefore not only unsupportable, but it is also contradicted by SCO's own statements and conduct.

Moreover, contrary to SCO's assertions, it is not a "simple matter" for IBM to produce the materials sought by SCO. In fact, the production of these materials would impose undue burden on IBM. The information SCO seeks amounts to millions of pages of documents created by the hundreds of individual programmers who worked on the AIX and Dynix operating systems over the past 20 years. It also includes approximately 2 billion lines of additional source code—the rough equivalent of approximately 40 million additional pages of paper. The enormous burden associated with the search for and production of these materials should be self-evident.

As is alleged in IBM's counterclaims, we believe SCO is engaged in a scheme to create—and maintain—fear, uncertainty and doubt in the marketplace concerning Linux and IBM's products and services. SCO's failure to respond timely and properly to IBM's discovery requests is, as is explained in IBM's pending motion for summary judgment, part and parcel of this scheme. 2 We believe, respectfully, that SCO's renewed request for massive amounts of unnecessary discovery (already once rejected by this Court) is merely another aspect of SCO's strategy to prolong the resolution of this case. Accordingly, SCO's request for additional discovery should be denied.


I. Operating Systems

Several different operating systems are involved in this case. An operating system is a group of computer programs that allows a computer to function by performing basic tasks such as recognizing input from the keyboard, keeping track of files and controlling disk drives.

An operating system is often composed of thousands of individual files, each of which consists of many lines of source code. Source code is the text of a computer program's instructions, written in a standard programming language.

As with most computer programs, operating systems are developed and packaged as "releases" or "versions". For example, there are numerous different versions of Microsoft's "Windows" operating system, including, for example, Windows 3.1, Windows 95 and Windows 2000.

SCO claims to have acquired the rights to a number of revisions and releases of the UNIX operating system, which was originally developed by AT&T's Bell Laboratories. In particular, SCO claims to hold certain rights to the many releases of the UNIX System V operatin gystem, including UNIX System V Release 1.0, UNIX System V Release 2.0, UNIX System V Release 3.0, and UNIX System V Release 4.0, among others.

IBM developed the AIX operating system, and acquired the Dynix operating system developed by Sequent Computer Systems, Inc. ("Sequent"). There are numerous versions and releases of both AIX and Dynix.

Linux is an operating system that has been developed publicly over the Internet beginning in 1991. Unlike most operating systems, all of the source code for Linux is publicly available.

II. SCO's Claims and Discovery Requests

SCO has proclaimed publicly (and in this lawsuit) that (1) IBM has breached the license agreements it entered into with AT&T for the source code to certain releases of UNIX System V by contributing source code to Linux; and (2) the use of Linux infringes SCO's alleged copyrights in certain releases of UNIX System V (along with certain other releases of the UNIX software). In addition, under SCO's apparent interpretation of IBM's UNIX System V licenses—which is, we believe, wrong—IBM is prohibited from contributing any source code contained in IBM's AIX or Dynix programs to Linux, even if the code was developed independently by IBM.

SCO's first set of document requests called for production of only limited "versions and iterations" of source code for AIX and Dynix:

Request No. 2.
All versions or iterations of AIX source code, modifications, methods and/or derivative works from May 1999 to the present, including but not limited to version 4.3 and above.

Request No. 3
All versions or iterations of Sequent Dynix source code, derivative works, modifications and/or methods from January 1999 to the present.

SCO did not anywhere in its first set of document requests ask IBM to produce "all revision control system information" for AIX and Dynix, "source code and log information for all interim" versions of AIX and Dynix, or "all design documents, whitepapers and programming notes" for AIX and Dynix.

Contrary to SCO's assertion that "IBM has produced selected pieces of AIX and Dynix" (SCO Mem. at 6), IBM has produced all of the source code for each version and release of AIX (e.g., AIX 4.3.2, AIX 4.3.3, AIX 5.1.0, AIX 5.2.0, etc.) and Dynix (e.g., Dynix 4.2.4, Dynix 4.4.5, Dynix 4.5.0, Dynix 4.6.1, etc.) that was distributed after the dates specified in SCO's requests. 3 In other words, IBM fully responded to Request Nos. 2 and 3.

On December 4, 2003, one day before this Court entered a stay of discovery as to IBM, SCO served its second set of document requests on IBM requesting additional source code:

Request No. 57.
All source code for AIX and Dynix since 1985 including all instructions, information used and all documentation relating to the use of AIX and Dynix, including but not limited to, all development notes.

After the Court lifted the stay, and on the schedule agreed to by the parties, IBM on April 19, 2004 objected to producing the additional source code requested by SCO on the grounds that SCO's request is "overbroad, unduly burdensome and seeks information that is irrelevant and not reasonably calculated to lead to the discovery of admissible evidence". SCO never moved to compel the production of documents responsive to its Request No. 57.

The materials SCO seeks in its "Memorandum Regarding Discovery" are not limited to source code contained in actual versions or releases of the AIX and Dynix operating systems. Rather, SCO seeks all of the source code contained in any "interim version" of AIX and Dynix, which would contain, among other things, drafts of source code and source code that may have been contemplated for inclusion in a version or release but was not actually included in the program. SCO also seeks any and all information relating to the development of the source code for AIX and Dynix, including any written revision histories and logs of AIX and Dynix, any design documentation for AIX and Dynix, and any notes written by the programmers who created the code in AIX and Dynix.

Put differently, if AIX and Dynix were sets of encyclopedias that were written by hundreds of contributing authors, SCO is not content with every edition of the encyclopedias published since 1999 (the date given in SCO's first set of document requests). Instead, SCO seeks, in addition to every single edition published between 1984 and the present, all of the thousands of interim drafts of every entry in the encyclopedias ever created during the last 20 years; a history of every revision ever made to every entry of the encyclopedias during the last 20 years; and all notes ever taken by the hudreds of individual authors who ever worked on the encyclopedias during that 20-year period. IBM believes that such requests are overbroad and unduly burdensome, and intended primarily to lay the groundwork for further delay.

For purposes of comparison, it bears mention that SCO has not come anywhere near producing the types of information it now asks IBM to produce. In response to IBM's requests for the production of source code for SCO's UNIX programs, SCO produced the source code for some (but not all) versions and releases of the UNIX software. SCO did not produce any source code contained in any "interim versions" of the UNIX software. SCO did not produce any revision control system information for any versions or releases of the UNIX software. SCO did not produce any design documentation or programming notes for any versions or releases of the UNIX software. If the information SCO seeks was necessary to this case (and it is not), then the informatino it has declined to produce would be no less necessary. SCO has represented to this Court that its production to IBM is full and complete. SCO has never indicated, and presumably does not believe, that SCO's own production of source code consists of only "selected pieces" and "selected snapshots" of the UNIX software. (SCO Mem. at 6-7.)

III. IBM's CMVC System

SCO asserts that IBM's Configuration Management Version Control ("CMVC") system should contain all of the information that SCO seeks and that it is therefore "a small matter" for IBM to produce it. (SCO Mem. at 15) That is simply incorrect. IBM's CMVC system contains certain information relating to AIX; it contains no information relating to Dynix. Moreover, as is explained in the Declaration of Joan Thomas ("Thomas Decl."), dated June 23, 2004 (attached hereto as Exhibit A), the CMVC system is used to keep track of code for many programs other than AIX (but not Dynix). (Thomas Decl. ¶ 7.) Identifying and segregating the approximatedly two billion lines of source code that are related to the AIX operating system from the other millions of lines of source code present in CMVC is not a "small matter". 4 (Id. ¶ 8.)

Extracting the relevant AIX operating system source code files from CMVC requires that persons familiar with the entirety of the AIX operating system engage in a time-consuming, multi-step process. (Thomas Decl. ¶ 8.) First, an engineer would have to review the files stored on the relevant CMVC server and identify—component by component—which of the hundreds of thousands of source code files are part of the AIX operating system and which are not. (Id. ¶ 9.) Second, a computer program, called a "script", would have to be written in order to map the identified AIX components to specific AIX source files, estimated to be in the hundreds of thousands. (Id.) Third, one or more knowledgeable persons would have to confirm that the source files identified by the script are in fact part of the AIX operating system. (Id.) Fourth, the list of source file names and identifiers would have to be correlated to their corresponding source code revision files—known as Source Code Control System ("SCCS") files. (Id.) These hundreds of thousands of SCCS files are the ones that contain the complete file development history (since 1991) for each of the corresponding source code files in the AIX operating system. (Id.) Finally, the SCCS files would have to be extracted and copied onto appropriate storage media. (Id.)

It is impossible to say for sure in advance how long the entire process would take to complete. But it is estimated that the process would surely take many weeks. (Thomas Decl. ¶ 11.) Insofar as SCO claims to be able to use the high-capacity storage tapes that are standard throughout the computer industry, downloading the information SCO seeks into SCO's preferred format would likely require many additional weeks of work. (Id.)

In addition, to produce all design documentation, whitepapers and programming notes for each of the 22 broad areas of functionality in AIX and Dynix requested by SCO, IBM would be required to collect and review an enormous volume of hardcopy material. We estimate that IBM would be required to collect and review approximately 10 million pages of programming notes for AIX alone and to identify, collect, and review documents from each of the hundreds of programmers who have worked on particular functionalities within AIX in the past 20 years. (Thomas Decl. ¶¶ 12-13.) At the very least, this process by itself would take months to complete. (Id. ¶ 14.)



IBM has already provided SCO with an enormous library of source code. As stated, IBM has provided SCO with all versions and releases of AIX and Dynix source code distributed during the period from 1999 to the present. That source code, consisting of approximately 250 million lines of code, represents the equivalent of more than 3 million pages. IBM provided this information nearly six weeks before the Court ordered it to do so. After eleven weeks and IBM's filing of a summary judgment motion, SCO now says it needs more.

According to SCO, it requires this additional code and other materials because it is necessary: (1) to "respond further to IBM's discovery requests"; (2) to "gather evidence relevant to IBM's theory of SCO's contract claim"; and (3) to "gather relevant evidence in support of SCO's theories of its contract case". (SCO Mem. at 2.) None of these arguments, two of which SCO never mentioned in its motion to compel and are raised here for the first time, has merit.

A. SCO Does Not Require Additional Information to Respond to IBM's Discovery Requests

SCO first contends that it requires additional information "[t]o respond further to IBM's discovery requests". According to SCO, it needs more information from IBM specifically to respond to IBM's assertions of "discovery misconduct and delay by SCO". (SCO Mem. at 2.)

None of the information SCO seeks is necessary for SCO to respond to IBM's discovery requests or to demonstrate that it has complied with the Court's orders. IBM's discovery requests ask SCO to describe SCO's claims and alleged rights and to disclose the "mountains" of evidence that SCO has described in its public statements. Not even the code IBM has already produced is necessary for SCO to provice the information IBM seeks. SCO must know what its claims and rights are (more than one year into this litigation) and if it can publicly discuss its alleged evidence, there is no reason SCO cannot produce the alleged evidence for IBM.

Moreover, SCO does not need any additional information to comply with the Court's order directing SCO to identify the specific lines of code from UNIX System V from which IBM's contributions to Linux are allegedly either copied or derived. SCO has, and has had since it commenced this lawsuit, access to the relevant information—the source code for UNIX System V and the source code for IBM's contributions to Linux. If the code that IBM contributed to Linux is in fact copied or derived from UNIX System V, then by doing a simple code comparison, SCO should be able to identify the code in UNIX System V without any difficulty. If the code IBM contributed to Linux is not copied or derived from any code in UNIX System V, then SCO should just say so, rather than seek irrelevant and unnecessary discovery, the production of which would merely impose burden and delay.

As stated in IBM's motion for partial summary judgment, we do not believe that SCO has complied with this Court's discovery orders. SCO tacitly concedes as much in arguing that it needs more information from IBM to respond to discovery requests with respect to which SCO has twice certified to the Court that it already provided true, complete and detailed responses. We believe that SCO's non-compliance with the Court's orders entitles IBM to summary judgment on at least one of the claims in suit. SCO appears to disagree that it has failed to comply with the Court's orders, and we presume that in responding to IBM's summary judgment motion SCO will undertake to explain to Judge Kimball why it believes it has complied. However, the information SCO now seeks is entirely irrelevant to whether SCO complied with the Court's prior orders. Either it did comply or it did not. No amount of further discovery will shed light on this question. Thus, SCO's claim that it needs more discovery from IBM to show that SCO did not violate the Court's prior orders is meritless.

B. SCO Does Not Require Additional Informatino to Show That The Use of Linux Infringes SCO's Copyrights

SCO next contends that it needs more information from IBM to "gather evidence relevant to IBM's theory of SCO's contract claim", which SCO describes as a theory that "in order for SCO to succeed on its contract claims, SCO must prove copyright infringement". (SCO Mem. at 3.) SCO therefore claims that it needs the information to show that the code IBM is alleged to have contributed to Linux infringes SCO's purported copyrights. (SCO Mem. at 4.) Here again, SCO is wrong.

To prevail on a claim of copyright infringement, a claimant must show: (1) ownership of a valid copyright, and (2) copying by the defendant, without authorization, of protected elements of the work that are original. See Jacobsen v. Deseret Book Co., 287 F.3d 936, 942 (10th Cir. 2002); Gates Rubber Co. v. Bando Chem. Indus., Ltd., 9 F.3d 823, 831 (10th Cir. 1993). If SCO cannot adduce evidence sufficient to show both ownership of a valid copyright and copying of protected components of the work that are original, then SCO's claims of infringement must fail.

To establish that IBM "copied" protected elements of SCO's alleged copyrights, one critical element SCO must show is that code in Linux is "substantially similar" to code in the allegedly copyrighted work (here, the UNIX software), so that the "copying" of Linux could be said to constitute "copying" of the UNIX software. See Jacobsen, 287 F.3d at 943. This necessarily requires SCO to identify the precise lines of Linux code in which it claims rights, and the precise lines of code in the UNIX software from which SCO alleges the Linux code is copied or derives.

SCO does not need any additional codee to perform this analysis. In order to prevail on a copyright infringement claim, SCO must show that the code IBM contributed to Linux (which SCO has in its possession or has ready access to) is "substantially similar" to code contained in its UNIX software (which SCO has in its possession). The history of every change made to the code before IBM contributed it to Linux is irrelevant to the question of whether the contributed code, as it appears in Linux, is substantially similar to any code in the UNIX software.5

In a similar circumstance, when plaintiff in a copyright infringement action opposed defendant's motion for summary judgment on the theory that it required additional discovery regarding the computer program at issue, the court denied plaintiff's request. See Gemisys Corp. v. Phoenix Am., Inc., 186 F.R.D. 551, 566 (N.D. Cal. 1999). In Gemisys, the plaintiff, Gemisys, had licensed a software program, PMIS, to the defendant, Phoenix American. Id. at 554. Phoenix American subsequently developed and introduced its own, competing software product, S.T.A.R. Id. at 555. Gemisys sued Phoenis American, alleging trade secret, copyright infringement, unfair competition and breach of contract claims. Id. at 553. Phoenix American moved for summary judgment on all claims. Id. Gemisys requested that the Court deny the defendant's summary judgment motion under Rule 56(f) on the grounds that Gemisys needed additional discovery "to ask questions about the design, structure, and development of the S.T.A.R. program", whose code Gemisys already possessed. Id. at 565-66. In denying Gemisys' request for additional discovery and granting summary judgment of noninfringement to the defendant, the court noted that "with respect specifically to Gemisys' claim for copyright infringement", plaintiff did not explain "how the disclosure of the S.T.A.R. [program's] design and development documents . . . would provide evidence essential to the motion for summary judgment, namely, proof of substantial similarity". Id. at 566.

C. SCO Does Not Require Additional Information to Support Its Contract Claims.

SCO's third argument for more information is that it needs "[t]o gather relevant evidence in support of SCO's theories of its contract case against IBM". (SCO Mem. at 2.) Again, SCO does not require any more information than IBM has already produced.

Despite the allegations of its complaint (Second Am. Compl. ¶ 110 (SCO's first)), SCO apparently no longer contends that IBM breached its contractual obligations to SCO by "dumping" UNIX System V code into Linux. Instead SCO contends that IBM breached the agreements by contributing portions of its own software, AIX and Dynix, to Linux. Thus, the only "contract" discovery to which SCO could possibly be entitled is discovery relating to its contention that IBM contributed code from AIX and Dynix into Linux.

As we explained in opposing (and the Court appeared to acknowledge in denying) SCO's motion to compel, SCO does not require any additional discovery to prove the alleged breach (on any viable theory). That is because the alleged breach necessarily involves allegations of public misconduct by IBM (i.e., the alleged "dumping" of code into Linux). IBM's contributions to Linux are publicly available and are therefore available to SCO and anyone else with Internet access. To the extent IBM has any code which it unsuccessfully attempted to contribute to Linux, and which is not publicly available, IBM has produced that code (pursuant to the Court's order dated March 3, 2004). Therefore all the alleged evidence SCO could possibly require is available to SCO, despite SCO's contrary contention.

The single case cited by SCO in support of its request is inapposite. In Dynamic Microprocessor Assocs. v. EKD Computer Sales, 919 F. Supp. 101 (E.D.N.Y. 1996), the plaintiff sought production of the source code for specific versions—Versions 3 and 4—of defendants' "pcAnywhere" product, not all iterations of all source code ever created for possible inclusion in "pcAnywhere". Id. at 103. IBM has already produced the source code for every version and release of AIX and Dynix distributed in the past five years. There is no reason SCO also needs all of the code that never ended up in any of those versions of AIX and Dynix.

The fact that SCO does not need the information it now seeks—and, indeed, that it does not need the hundreds of millions of lines of code that IBM has already produced—in order to show the alleged breach, is plain from SCO's conduct with respect to HP and Sun. HP and Sun are parties to licensing agreements like the agreements IBM is alleged to have breached; they market and sell software products similar to IBM's AIX and Dynix products; and they have publicly disclosed and/or contributed code to Linux. As the Court may recall, SCO featured both HP and Sun in the exhibits it produced to the Court at the February hearing. (See Ex. B.)

According to SCO, HP and Sun, unlike IBM, have honored (not breached) their agreements with AT&T. SCO reached the conclusion that neither HP nor Sun has breached its contractual obligations to AT&T without reference to a single line of source code from either HP or Sun. This is because it is not at all necessary for SCO to have the source code of AT&T's licensees to determine whether they breached their contractual obligations. Just as SCO did not require source code from HP and Sun to determine whether they breached their obligations, it does not require any from IBM—certainly SCO does not require any more than the hundreds of millions of lines of code IBMhas already produced.

If SCO's allegations and public statements are believed, then it had a "mountain" of evidence that IBM has breached the contracts even before it filed this suit and received from IBM the hundreds of millions of lines of code that IBM has already produced. 6 Moreover, SCO states in its discovery memorandum, as it has repeatedly elsewhere, that it can prove IBM breached the agreements based upon the evidence SCO has to date. (SCO Mem. at 13.) If that is true, then SCO does not require any more code or other information from IBM.


In addition to being irrelevant and unnecessary, the code and other information sought by SCO can be produced by IBM only at great cost. SCO essentially asks IBM to identify and collect more than approximately 2 billion lines of source code, representing the rough equivalent of more than 40 million of pages of paper, and to collect more than an an estimated 10 million pages of additional documents -- stretching back 20 years -- from the hundreds of programmers who worked on developing AIX or Dynix. Producing these materials is not, as SCO contends, "a trivial task" (SCO Mem. at 16) or "a small matter". (Id. at 15.) Rather, it is a huge effort and, in light of the irrelevance of the information to this case, clearly unduly burdensome.

SCO makes much out of the fact that AIX source code is stored on a computer server, using a system known as Configuration Management Version Control ("CMVC"). SCO concludes from this that the source code "should be easily available for downloading and production to SCO". This is not so. In fact, providing SCO with the billions of lines of source code and revision control information it seeks would require a very substantial effort. It is impossible to know for certain how long it would take IBM to provide the information SCO seeks, but we estimate that it would require many weeks (indeed, likely months) of effort, especially if SCO cannot utilize the data tapes that are industry-standard for storing and transferring source code. 7 (Thomas Decl. ¶ 11.) Moreover, SCO ignores entirely the fact that CMVC does not include any Dynix code. (Id. ¶ 5.)

The source code for the AIX operating system, as stored on CMVC, is not something that can be downloaded with a simple click of a button. As noted above, the CMVC server that contains source code for the AIX operating system also contains a large amount of other material, including source code, for hundreds of other IBM products and programs. (Thomas Decl. ¶ 7.) Extracting the relevant AIX operating system source code files from CMVC requires that engineers familiar with the entirety of the AIX operating system engage in a time-consuming, multi-step process. (Id. ¶¶ 8-9.) The process, depending on the storage media that is used, is estimated to take many weeks to complete. (Id. ¶ 11.)

SCO similarly makes light of the heavy burden involved in producing "design documents, whitepapers and programming notes" for each of the broad areas of AIX functionality identified by SCO, claiming without support that "the burden on IBM to produce these materials is negligible". (SCO Mem. at 9 n.8.) As with the production of the source code and revision control system information sought by SCO, producing the design documents, whitepapers and programming notes would be extremely burdensome for IBM, and, as discussed above in Section 1, would yield only materials that are irrelevant and unnecessary. For example, IBM, with respect to AIX alone, would have to collect and review an estimated approximately 10 million pages of programming notes for AIX to determine what records exist for the particular functionalities identified by SCO. (Thomas Decl. ¶ 12.) IBM would further have to interview and collect many pages of documents from the hundreds of persons that worked on each of those functionalities over the past 20 years -- a process that would take months on each of those functionalities over the past 20 years -- a process that would take months to complete at the very least (Id. ¶¶13-14), and involve not just the hundreds of IBM employees that will need to be interviewed, but many thousands of hours of attorney time as well.

As stated, none of the materials SCO asks for is necessary for the prosecution of this case. The only purpose served by the production of the material SCO seeks would be to impose undue burden on IBM and to lay the groundwork for further delay. SCO should not be allowed to impose undue burden on IBM, especially when it required no discovery from either Sun or HP to reach conclusive determinations as to those companies' liability.


For the foregoing reasons, the Court should deny SCO's request for the materials sought in its May 28, 2004 Memorandum Regarding Discovery.

DATED this 23rd day of June, 2004.


Alan L. Sullivan
Todd M. Shaughnessy

Evan R. Chesler
David R. Marriott

Attorneys for Defendant/Counterclaim-Plaintiff
International Business Machines Corporation

Of counsel:

Donald J. Rosenberg
Alec S. Berman
[address, phone]

Attorneys for Defendant/Counterclaim-Plaintiff
International Business Machines Corporation

1 SCO's Memorandum Regarding Discovery is cited herein as "SCO Mem. at _____."

2 Although this Court has twice ordered SCO to do so, SCO continues to fail to comply in full with IBM's discovery requests. After all this time, SCO still has not matched with specificity the lines of Linux code allegedly at issue in this case to the lines of UNIX code from which the Linux code is allegedly copied or derived. Rather than burden the Court with another motion to compel, IBM moved for summary judgment on its counterclaim seeking a declaration of noninfringement of SCO's purported copyrights in certain UNIX software. IBM is entitled to summary judgment on that claim because, even putting aside the fact that SCO failed to comply with the Court's orders, SCO has failed to adduce (and cannot adduce) evidence sufficient to sustain its allegations.

3 IBM produced the source code for numerous versions and releases of Dynix on December 4, 2003. The next day, this Court stayed discovery. The discovery stay was lifted on March 3, 2004. The next day, March 4, 2004, IBM produced virtually all of the source code ordered by the Court to be produced in its March 3, 2004 order. IBM completed its production of source code six days later on March 9, 2004, almost six weeks before the court-imposed deadline of April 19, 2004.

IBM produced AIX source code to SCO on industry-standard data tapes, the same format IBM uses to distribute this source code to original equipment manufacturers and software vendors. IBM had previously advised SCO in October 2003 that IBM would produce the AIX source code on such industry-standard tapes, to which SCO raised no objection. Following IBM's production of the source code on these data tapes, SCO complained inexplicably that these industry-standard tapes were "non-usable". As a courtesy, IBM on March 23, 2004 (nearly four weeks before the April 19 deadline, and more than nine weeks prior to SCO's filing of its memorandum asking the court to order IBM to produce more code) sent to SCO 53 CDs containing the AIX source code that had previously fit onto 8 data tapes.

4 In addition, as Dynix was developed by Sequent, a company that IBM acquired, source code for Dynix is not contained in the DCMVC system. Accordingly, a separate search would need to be conducted for the type of interim code and revision histories SCO is seeking.

5 This is true even if the code IBM allegedly contributed to Linux had at one point in time been derived from code in the UNIX software. A derivative work based on an original work may be so transformed that it no longer is a derivative work and no longer infringes any copyright of the original work. See Castle Rock Entm't, Inc. v. Carol Publ'g Group, Inc., 150 F.3d 132, 143 n.9 (2d Cir. 1998)("Indeed, if the secondary work sufficiently transforms the expression of the original work such that the two works cease to be substantially similar, then the secondary work is not a derivative work and, for that matter, does not infringe the copyright of the original work."); Bucklew v. Hawkins, Ash, Baptie & Co., LLP, 329 F.3d 923, 930 (7th Cir. 2003) ("In some cases, however, though derivative in a literal sense, [a work] is so utterly transformed as to bear no traces of the original; and then there is no infringement."). The development history of the code IBM in fact contributed to Linux is therefore irrelevant to the inquiry of whether such code in its current form is substantially similar to any code in the UNIX software, the only relevant concern here.
See (video recording of a July 22, 2003 interview of SCO CEO Darl McBride) ("During the period of time shortly after filing the lawsuit . . . we turned three different teams of code programmers loose on the codebases of AIX, UNIX and Linux. . . . So the three teams came back independently and validated that there wasn't just a little bit of code showing up inside of Linux from our Unix intellectual property base. There was actually a mountain of code showing up in [Linux]".); see also Lisa Bowman, "SCO Puts Disputed Code in the Spotlight", CNET, Aug. 18, 2003, available at quoting McBride as saying SCO's code-comparison experts "have found already a mountain of code"); Maureen O'Gara, "SCO Files for AIX Injunction Against IBM",, June 16, 2003, available at (quoting McBride as saying that SCO has found "a truckload of code at the high end" from Sequent).

7 IBM maintains revision control information for the AIX operating system since 1991 on a CMVC server. IBM does not maintain revision control information for AIX source code prior to 1991, and thus, cannot produce that information to SCO. (Thomas Decl. ¶ 5.)


I hereby certify that on the 23rd day of June, 2004, a true and correct copy of the foregoing was sent by U.S. Mail, postage prepaid, to the following:

Brent O. Hatch
Mark F. James

Stephen N. Zack
Mark J. Heise

Robert Silver

Kevin P. McBride

Exhibits/ Attachments to this document have not been scanned.

Please see the case file.

  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 )