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
SCO's Memo in Opposition to IBM's Motion to Strike Sontag's Declaration - as text
Monday, September 06 2004 @ 10:31 AM EDT

This is SCO's Memorandum in Opposition to IBM's Motion to Strike the July 12, 2004 Declaration of Christopher Sontag. At least, that is what they call it. It is supposed to be where SCO tells the court why the declaration should not be stricken. Instead, SCO mostly bursts out into song again on their theme of how badly they want and need to look at all the AIX and Dynix code since the beginning of time.

Now they say they never meant for Sontag to be offering an expert opinion. His declaration should be viewed as a kind of summary, as per Utah Rules of Evidence 1006. That rule says:

"Rule 1006. Summaries.

"The contents of voluminous writings, recordings, or photographs which cannot conveniently be examined in court may be presented in the form of a chart, summary, or calculation. The originals, or duplicates, shall be made available for examination or copying, or both, by other parties at a reasonable time and place. The court may order that they be produced in court."

The application SCO makes is that his declaration was mostly just a way to bring exhibits to the attention of the court:

"In many regards, Mr. Sontag's observations simply reflect a permissable summary of the documents referenced. See Utah R. Evid. 1006."

Does that makes any sense to you, or does it look more like clutching at straws?

The level of argument throughout is such things as IBM saying in one place that handing over the code would take many months and in another, many weeks. SCO pretends that is a contradiction. Many weeks could equal many months, of course, but SCO acts like they have caught IBM in a lie. It's ridiculous. Or even sillier, that it's an "admission" by IBM that it wouldn't be burdensome to ferret out all the code and comments SCO wants. IBM has already told the court it would be burdensome, so to argue that they have made an admission otherwise is pointless.

There is a convoluted argument that goes sort of like this: Sontag isn't an expert. Non-experts can testify as to facts they have personal knowledge of. They can review documents and testify as to their opinions on the documents. They cite a case where a witness was allowed to testify, based on a review of company documents, as to events that happened prior to his tenure. Therefore, Sontag's declaration should be allowed. What does that case have to do with Sontag reading IBM booklets about CMVC and then giving his non-expert testimony about CMVC? He's testifying as to his opinion of something he has never used or seen work, so what is the basis for his "summary" and his opinion? If he were now in Joan Thomas' job, maybe then he could read all that stuff and give his opinion of what happened in 1993, despite not having the job in 1993, but he's not an expert and he has no personal experience with CMVC, so ... well, you see what I mean. They drop a stitch in the logic, but just barrel ahead anyway. By their logic, anyone you dragged in off the street could testify to anything they read on the internet and then give their opinion on it. I think standards in a courtroom are higher than that.

I think they think so too, because they go on to try to qualify Mr. Sontag as an expert. He oversaw the work of hundreds of developers at Novell, they state, as if that will make him an expert in something as massive as CMVC. Then they quote from a case that says an expert witness can give opinions based on review of documents and experience in an industry, and so far so good, but not "flights of fancy, speculations, hunches, intuitions or rumors", but that's exactly what Sontag offers when he testifies that he doesn't believe CMVC can work the way IBM says it works. Without some kind of proof or specificity otherwise or experience with that precise type of system at least, what is that but speculation?

They go on and on about CMVC and how it violates all reason to think that it wouldn't have what they want it to have in it. But they want extremely old code, from a system that IBM has already indicated isn't arranged the way they want. It wasn't designed for lawsuit-readiness. And while SCO quotes IBM documents that say you can readily track "your application" or your "project", AIX isn't an application. Because I have no knowledge of such systems, I asked someone to explain to me in layman's terms the SCO arguments and to give me an opinion, since I knew he had several years' experience (including their use on large public sector projects) with various software version control systems. Here is what he explained to me:

"SCO seems to be relying on the judge not being an IT expert and is, I believe, mischaracterising the way VC (version control) systems are used in the real world for non-trivial applications.

"I have not personally used CMVC but have used CVS, PVCS and VSS, which are of similar vintage. They have differences; but the similarities, in terms of the overall way they hook into the development process, are much larger. So I am fairly confident that what I'm about to say is largely correct.

"1) Application tracking -- VC systems allow you to mark a version of your application with a label. This enables you to retrieve that old version at a later date. However, you can only easily retrieve a version that you took the time to label.

"Example: On Tuesday, I put a label in for my application called 'MyApp_Tuesday'. Over the next few days I continue hacking. The next week, the boss asks me for Tuesday's version of the app. No problemo! VC gives it to me. But if the boss asks for Wednesday's version of the app, things are harder, because I didn't bother to put in a label.

"Normally, one only puts in labels for those versions of an app which are likely to be needed again in future. Those would be the versions which get released - either to the public or for some internal milestone. So, going purely on labels, the versions of AIX that IBM would be able to easily get hold of would be the released versions. But, if I understand correctly, IBM has already given SCO those; SCO wants the *intermediate* versions, which have no formal name - and likely, no label.

"SCO is using how easy it is for IBM to hand over what IBM has already handed over, as an argument that it must be equally easy for them to hand over what they haven't.

"2) Apps and Systems -- You are quite right that AIX isn't an application. Nor is it even a set of applications that were conceived and written at one time as one single project. It is a system that has grown and evolved. Over time, new parts and applications have been added, and other parts have been removed. There are parts of AIX which most likely are also present in other systems.

"The first point that arises from this is that the 'labels' in the VC system may not say 'AIX version 2' or similar. There may be a label on the foo application saying 'Foo v4' and a label on the Bar application saying 'Bar v6' and somewhere else *outside* the VC system, a table saying that in AIX v2 we have Foo v4 and Bar v6. That creates another level of difficulty in recreating an intermediate version.

"The second point is brought up by Joan Thomas herself: 'Paragraphs 15 through 30 of the Sontag Declaration are all based on Mr. Sontag's erroneous assumption that CMVC is organized in a way similar to that depicted by Figure 1 of the Sontag Declaration. In fact, CMVC's directory hierarchy is completely different than that posited by Mr. Sontag.'

"If we think about AIX for a moment, that system has a logical structure, rather like a company's organizational chart. Some modules are in charge of others, some talk to each other, etc., and you can express these relationships by the arrangement of the modules in the diagram.

"Point (i): the arrangement of the AIX source code on the disk need not, and for good reasons (include paths etc.) will almost certainly not, look anything like the logical arrangement of the AIX system.

"Point(ii): the arrangement of the modules in the VC database need not, and because of the evolutionary way AIX has developed probably will not, look much like the arrangement of the source code on disk.

"Although the CMVC page link in Argument, Section A, to says 'Within the family, the components are arranged in a tree-like hierarchical structure that reflects the structure of your organization and the structure of the projects you are working on,' this is is referring to a possibility, not a necessity. And indeed, when you have an evolving product, it is a positive *boon* that you can change the on-disk/logical structure of your application without being forced to change the layout inside the VC system.

"There are *two* levels of disconnect here. Therefore, the diagram of the logical structure of AIX is not necessarily of any use in locating the code on the VC system. If AIX were a single application that had been conceived, designed and written as a tidy single project by one team, then things might be different. But AIX is a complex operating system developed over many years by many teams in many buildings.

"But, you may ask, how can such a system work? Isn't it impossible to code efficiently if you can't get from 'logical unit' to 'VC file' easily?

"The answer is that a VC system is a working system, not an archival system. It isn't a glorified backup program. And as such, it is designed so that while some things (like complying with discovery requests) are very hard; those things are the things you don't need to do all the time. The things you do need to do day in, day out -- like making a bugfix in a source file and releasing it to the testing department -- are easy.

"My Conclusion

"It is possible to imagine a way in which AIX could have been stored in CMVC such that complying with an extensive discovery order would be made easier. But software projects are not usually managed with the assumption they will be the subject of lawsuits. Crucially, VC systems enable 'mixed-up' layouts, which make wholesale retrieval laborious but enable development to continue without interruption, to be used without loss of efficiency. So the argument that no company could operate with such a layout is simply incorrect, based on a lack of understanding of how VCs work in practice."

Of course, this isn't testimony in a court of law, but I know many of us are not programmers, including me, and we need help to follow some of the technical arguments.

SCO ends by asking the court to keep as much of Sontag's declaration as possible and throw out only the parts he has no personal knowledge of. That would be a lot, by IBM's reckoning. And they have come up with a new way to delay, I think. If the court declines to deny IBM's motion to strike, and I suspect they think that might happen, they ask that the court order the immediate deposition of Joan Thomas, so they can ask her all about CMVC. I am not clear why they can't just depose her anyway, regardless of what happens to this particular motion, but I could be missing something.

But under all that they say, there is the underlying theme that shows up in everything they do now, in every courtroom, regardless of their victim. It's all about their attempt to stretch copyright law to make it encompass methods and concepts according to SCO's lights. That's why they can't exhale until they get all that AIX code, and they as much as acknowledge that purpose. Here's why they say they want it:

"First, the discovery sought is indisputably relevant: it contains or will lead to the discovery of admissions impeaching the contract interpretations IBM offers in its pending summary judgment motion; it contains material showing that IBM depended on UNIX to develop products in ways that, under the contract, prevented the later contribution of all or any part of those products into Linux; and it will be necessary to test IBM's own claim to the Court (its Ninth Counterclaim) that, independent of any Linux issue, nothing in the entire history of AIX violated any SCO copyright (IBM's sophisticated central storage system, 'CMVC,' contains the entire development history that the IBM Ninth Counterclaim asks the Court to declare free of any violation of any SCO copyright)."

IBM's 9th counterclaim is about this:

"164. SCO has sued IBM claiming that IBM has infringed, induced the infringement of, and contributed to the infringement of, SCO's purported UNIX copyrights by, among other things, continuing to 'reproduce, prepare derivative works of, and distribute copyrighted UNIX materials through its activities relating to AIX and Dynix'. . . .

"167. IBM is entitled to a declaratory judgment pursuant to 28 U. C. Section 2201 that IBM does not infringe, induce the infringement of, or contribute to the infringement of any SCO copyright through the reproduction, improvement, and distribution of AIX and Dynix, and that some or all of SCO's purported copyrights in UNIX are invalid and unenforceable."

I can see why Microsoft might want to fund something like this methods and concepts attempted land grab, even if I can't approve. If SCO can place methods and concepts in software soundly and utterly under copyright law, Microsoft can control and kill off anything in GNU/Linux that acts like their software, and even more effective, SCO can block anything that acts like UNIX, and then what's left of GNU/Linux? All software tends to do the same tasks. So whoever started first wins, in such an IP universe. No one can follow on at all, because most if not all of the methods and concepts are quickly taken. And with the DMCA in their sling shots, they could shut down anybody they wish, like a huge, crazed RIAA. And then, voila! A monopoly on steroids, raking in the dough, and GNU/Linux dead and bleeding on the floor or exiled to the noncommercial space. I think that was the SCO strategy, and that's why they are so desperate to establish this methods and concepts idea and why they put it forth everywhere, even though so far, it's been on a slow boat to nowhere.

Thanks go to David Truog for transcribing.


Brent O. Hatch (5715)
Mark F. James (5295)
[address, telephone, fax]

Stephen N. Zack (admitted pro hac vice)
Mark J. Heise (admitted pro hac vice)
[address, telephone, fax]

Robert Silver (admitted pro hac vice)
[address, telephone, fax]

Frederick S. Frei (admitted pro hac vice)
Aldo Noto (admitted pro hac vice)
John K. Harrop (admitted pro hac vice)
[address, telephone, fax]

Attorneys for The SCO Group, Inc.









Case No. 2:03CV0294DAK

Honorable Dale A. Kimball
Magistrate Judge Brooke C. Wells

Preliminary Statement 1
Background 4
Argument 6
A.     IBM Admits That Complying with Its Discovery Obligations Is Not Burdensome 6
B.     The Sontag Declaration Is Based on Personal Knowledge 10
C.     If the Court Concludes That an Expert Opinion Is Required, Mr. Sontag Qualifies as an Expert 13
D.     IBM's Request to Strike the Entire Declaration Is Meritless 13
Conclusion 15

A.H.L. Inc. of Del, v. Star Ins. Co.,
10 F. Supp. 2d 1216 (D. Kan. 1998)
Burton v. R.J. Reynolds Tobacco Co.,
183 F. Supp. 2d 1308 (D. Kan. 2002)
Fenstermacher v. Telelect, Inc.,
21 F.3d 1121 (10th Cir. 1994)
10, 12
Fla. Farm Bureau Mut. Ins. Co. v. B & B Miller Farms. Inc.
No. 87-1021-C, 1991 WL 201188 (D. Kan. Sept. 17, 1991)
Kloepfer v. Honda Motor Co.,
898 F.2d 1452 (10th Cir. 1990)
S.E.C. v. Wolfson,
309 B.R. 612 (D. Utah 2004)
In re Tex. E. Transmission Corp. PCB Contamination Ins. Coverage Lit.
870 F. Supp. 1293 (E.D. Pa. 1992), affd, 995 F.2d 219 (3d Cir. 1993)
Trestle & Tower Eng'g. Inc. v. Star Ins. Co.,
13 F. Supp. 2d 1166 (D. Kan. 1998)
Visser v. Packer Eng'a Assocs. Inc.
924 F.2d 655 (7th Cir. 1991)
10, 12
Fed. R. Civ. P. 26(b) 7
Fed. R. Civ. P. 34(b) 8
Fed. R. Evid. 701 12
Report of the Special Committee for the Study of Discovery Abuse, Section of Litigation of the American Bar Association (1977) 9

Plaintiff The SCO Group, Inc. ("SCO") respectfully submits this Memorandum in opposition to defendant International Business Machines Corporation's ("IBM") Motion to Strike the July 12, 2004 Declaration of Chris Sontag (the "Motion").

Preliminary Statement

IBM's Motion is an attempt to prevent the Court from seeing (1) IBM public documents that contradict IBM statements to this Court, and (2) the rudimentary, common-sense observations of an experienced industry participant that IBM, one of the most sophisticated computer companies in the world, would never deliberately make its internal operations as inefficient as IBM now tells the Court they are. IBM tries to prevent the Court from examining these two categories of evidence because IBM recognizes they undermine IBM's overall attempt to hold back more evidence.

To place IBM's argument in context, it is necessary first to describe what SCO has shown in support of its discovery motion independent of Mr. Sontag's declaration. SCO's motion seeks to compel IBM to produce the source code of the multiple versions of the AIX and Dynix/ptx computer operating systems and related materials stored in IBM's and Sequent's sophisticated centralized storage systems. As shown in SCO's May 28 and July 12 briefs1, SCO is entitled to that discovery for two reasons—both of which would apply even if there were no Sontag declaration and even if IBM's Thomas declaration were uncontradicted:

First, the discovery sought is indisputably relevant: it contains or will lead to the discovery of admissions impeaching the contract interpretations IBM offers in its pending summary judgment motion; it contains material showing that IBM depended on UNIX to develop products in ways that, under the contract, prevented the later contribution of all or any part of those products into Linux; and it will be necessary to test IBM's own claim to the Court (its Ninth Counterclaim) that, independent of any Linux issue, nothing in the entire history of AIX violated any SCO copyright (IBM's sophisticated central storage system, "CMVC," contains the entire development history that the IBM Ninth Counterclaim asks the Court to declare free of any violation of any SCO copyright).

Second. IBM's admissions show that producing the discovery at issue would plainly involve no undue burden. As SCO's July 12 brief makes clear, this fact is independent of any disagreement between Mr. Sontag and Ms. Thomas, and would be so even if there were no Sontag declaration and all of Ms. Thomas' assertions were credited as true. IBM admits that the materials SCO seeks relating to Dynix can be produced without any delay and that those relating to AIX can be produced in a matter of "weeks" (a reversal of IBM's original claim to this Court it would take "many, many months"). As SCO has shown, black-letter law makes clear that this is not an "undue" burden.

SCO thus does not need the Sontag declaration to prevail on this motion. SCO offered the declaration to demonstrate something else - the general lack of credibility of IBM's claims about discovery. The Sontag declaration does this in two ways, neither of which remotely requires specialized expert qualifications, and both of which IBM does not want this Court to be able to examine.

First. IBM has made numerous statements in public documents that contradict its claims to this Court. IBM tells this Court that using its CMVC is so difficult that it would take "many weeks" to produce the information SCO has requested. But in public documents, IBM told prospective customers that that very same storage technology should be valued because of the ease and speed with which it allowed access to that information. Mr. Sontag's declaration makes those IBM public documents available to the Court and calls the Court's attention to various key paragraphs—a function for which expert qualifications obviously are not needed.

Second, if IBM was not being accurate in its public documents, but instead was accurate in its statements to this Court, that would require the conclusion that IBM—one of the most sophisticated computer companies in the world - has adopted a system that imposes enormous internal inefficiency on itself. Mr. Sontag, based on his experience gained through a lengthy career in the industry as a software designer, engineer and manager of very large software projects, observes that this is highly implausible, particularly given IBM's stature and internal business needs—as well as given the availability of well known "top-down" technology. None of these observations—rudimentary to anyone with a background in the industry—remotely requires "expert" knowledge.

On the other hand, IBM's expert, Ms. Thomas, never even tries to reconcile her testimony with IBM's public statements, nor does she offer any explanation for the inefficiency she claims IBM's system imposes on it.

The Sontag declaration may thus be split into two parts: documents and testimony. There is no legitimate claim that the documents are inadmissible. The portion of Mr. Sontag's declaration containing testimony is also admissible, particularly in light of the supplemental Sontag declaration, filed herewith, setting forth Mr. Sontag's qualifications to provide the rudimentary, common-sense industry perspective. If IBM wishes to focus on expertise, however, IBM should make its purported expert available for deposition so that she may be examined about, among other things:

  • how IBM's public claims about the speed and ease of accessing information in its tracking system can be reconciled with its claims to this Court about how difficult it is to access information through that system;
  • why the particular commands outlined in IBM's manuals that automate the collection of the information SCO seeks would somehow not function in this particular case;
  • why IBM would contradict its public claims and impose such internal inefficiency on itself;
  • why IBM first said its production would require "many, many months" but now concedes it will only take "many weeks".

As discussed in more detail below, IBM's motion should be denied.


IBM now concedes that compliance with SCO's discovery requests would not be unduly burdensome. IBM has never claimed that the Dynix materials maintained in Sequent's Revision Control System ("RCS") would be difficult to produce, and does not even address the portions of Mr. Sontag's declaration relating to RCS. (7/12/04 Sontag Dec. ¶¶ 36-39.) As for AIX, though IBM initially claimed that extracting the AIX materials from its CMVC version control system would take "many, many months" (2/06/04 Tr. at 37:10-14), it subsequently shortened that estimate to "many weeks" (6/23/04 Thomas Dec. ¶11) and, in its most recent filing, gives no concrete time frame at all. The simple fact is that if IBM can produce the materials, which are undisputedly relevant and discoverable, in a matter of weeks - even several weeks - it must do so. That simply cannot constitute an "undue burden" sufficient to relieve IBM of its obligation to produce relevant discovery, particularly when the materials were requested well over 60 weeks ago.

IBM's present claims also contradict its public statements that CMVC is designed to keep track of this material in an efficient manner. The idea that CMVC could fail so completely at its essential purpose is contrary to common sense. Mr. Sontag describes that contradiction, relying on IBM's statements to the Court and in public documents. IBM's declarant, Joan Thomas, claims Mr. Sontag's testimony "contains numerous factual errors." (8/4/04 Thomas Dec. ¶5.) Ms. Thomas utterly fails to point out these alleged errors, however, and instead expects the Court simply to accept her unsupported accusation:

      Since it is clear to me, from reading the Sontag Declaration, that Mr. Sontag does not have any personal knowledge of IBM's CMVC system, or of the time and effort involved in collection and producing the information sought by SCO, I will not attempt to itemize and correct all of the errors made in the Sontag Declaration. (Id.)2

More importantly, IBM's own documents, as well as fundamental reasoning that underlies CMVC in the first place, flatly contradict Ms. Thomas' testimony. In its user's guide to CMVC, entitled Did You Say CMVC? (the cited excerpts of which are attached to the 8/26/04 Sontag Declaration as Exhibit B), IBM describes the benefits of software configuration management (or "SCM"), which is the function CMVC performs:

      The primary benefit of SCM ensures that you can define and identify all elements comprising your application. It also ensures that you know exactly in which manner they are generated, preprocessed, compiled, linked or otherwise combined to form specific releases of your application and related documentation. SCM ensures that you keep a historical record of these release configurations along with the exact versions of all the components and the application itself at each release. This means that you can re-create exactly any previous release of your application, which exhibits a failing characteristic reported by end users.

(Sontag Dec. Ex. B at 2 (emphasis supplied).)

Thus, according to IBM, the primary benefit of CMVC is that it makes it possible to keep track of, and recreate as necessary, the exact elements, versions and components of any previous release of a software project, such as AIX. Ms. Thomas claims that the process of recreating and producing previous releases of AIX would require weeks of work by skilled engineers, even though IBM's own documentation indicates this as the automated function CMVC is designed to accomplish, precisely to avoid the need for manual re-creation. It simply strains credulity to assert, as Ms. Thomas does, that IBM, one of the most advanced technology companies in the world, would use a package that makes its "primary benefit" so difficult to achieve.

According to IBM's documents attached to Mr. Sontag's declaration (and contrary to Ms. Thomas' sworn testimony), every file on a CMVC server is associated with a particular product name (such as AIX) and is uniquely identified as such. The data is identified by project and releases can be easily extracted with a few commands. Were it otherwise, CMVC would not only fail to comply with basic industry standards, but would also be useless as a version management tool. Ms. Thomas' position therefore is not likely and is unpersuasive. Even were Ms. Thomas' position accurate, however, the discovery sought is highly relevant and, given the nature of this litigation and the vast resources of IBM, the discovery should be produced.


A.    IBM Admits That Complying with Its Discovery Obligations Is Not Burdensome.

IBM's Motion is a sideshow—an effort to shift focus away from its inexcusable failure to produce materials fundamental to SCO's case. IBM's strategy is two-pronged: by forcing the motion to compel in the first instance, IBM is attempting to duck its obligation to provide relevant discovery without offering any valid basis for doing so; by attacking Mr. Sontag, IBM hopes to distract the Court from the substance of the motion to compel. Simply put, IBM has not argued—because it cannot—that it would be "unduly burdened" by complying with SCO's discovery requests, the showing required by Fed. R. Civ. P. 26(b)(iii).

IBM's position regarding the alleged "burden" of providing discovery from CMVC has shifted over time. At first IBM told the Court that it would take "many, many months" to produce the information requested (2/06/04 Tr. at 37:10-14), then it stated that it would take "many weeks" to do so (6/23/04 Thomas Decl. ¶11), and now it takes no position at all on the time frame. IBM cannot possibly argue that a production measured in weeks could constitute an undue burden, particularly given the nature of this litigation, IBM's enormous resources, and in light of the fact that SCO requested this information over a year ago.

With respect to the volume of the discovery sought, IBM claims that it would amount to "millions of pages of documents" created by programmers and "40 million additional pages of paper" for additional source code. (IBM Motion at 3.) That statement is misleading. It might be true if IBM decided to print out relevant portions of its CMVC system and to provide that material in print-out format rather the original electronic format of all of CMVC as it is stored, configured, maintained and managed by IBM in the ordinary course of business. However, IBM admits that it is fully capable of producing the requested information in a compact, electronic format, exactly as maintained in the ordinary course of its business, as SCO has repeatedly requested (6/23/04 Thomas Decl. ¶¶ 10-11) and as the Federal Rules require.

In fact, SCO needs a faithful reproduction of CMVC in the precise format in which it is stored and maintained in the ordinary course of IBM's business so that discovery can be meaningful, accurate and not obfuscated by confusing production. With access to CMVC in electronic form, SCO will be able accurately to identify source code check-in dates, authors, comments, and original works that formed the basis of derived works in AIX. These issues are fundamental to its case, as IBM well knows, and would be lost (or at least hidden) in a massive "paper dump" of the kind IBM apparently contemplates. IBM is not entitled to block SCO's access to (and use, for purposes of this litigation, of) this critical information by producing it in a form that permits IBM to affect substantive issues in the case through discovery ploys.

SCO has made a straightforward discovery request for production of data from CMVC in its original format. IBM, however, at most wants to produce carefully selected sections of the original data in CMVC, without the original, configurable format of electronic data itself in an attempt to prevent SCO from accurately evaluating the way source code was checked into CMVC in the first instance, prevent SCO from analyzing and evaluating check-in dates or check-in comments, and make it far more difficult for SCO to trace IBM's own use of SCO's source code in AIX. That is precisely the kind of activity that Fed. R. Civ. P. 34(b) was designed to prevent. Rule 34(b) requires: "A party who produces documents for inspection shall produce them as they are kept in the usual course of business or shall organize and label them to correspond with categories in the request." The commentary to Rule 34 (b) states that this sentence was inserted in the 1980 Amendment to prevent the "obscuring" of the significance of documents produced.3 In short, a paper production would hide the traces of IBM's conduct that form the very basis for SCO's claims. As a matter of compliance with Rule 34(b), SCO is therefore entitled to full production of CMVC, in precisely the format and configuration it is stored, maintained and used by IBM in the ordinary course of its business.

IBM's own documents clearly state that it would not be limited in its ability to produce AIX source code:

  • IBM states: "CMVC is broad in its application, thorough in its implementation, and very flexible." (8/26/04 Sontag Dec. Ex. B (Did You Say CMVC?) at 65.)
  • IBM states: "The most important thing to understand about CMVC is that you do not need to understand all of it, before you begin using some of it... CMVC provides many mechanisms to help you accomplish your . . . goals, but it does not dictate exactly how you should use them, nor does it require that you use them all if you use only some of them." (Id.)
  • IBM states: "CMVC ensures that an audit trail is maintained for every file by identifying for any file change: when the change occurred, who was responsible, and why the file was modified." (Id, at 9.)
  • An IBM employee explained that the Version Control ("VC") function of CMVC "eliminates confusion when looking for a particular source code file belonging to a particular product." (7/12/04 Sontag Decl. Exh. 3 (Email from Adrian Mitu (IBM Canada), Aug. 19, 1993 (located at
  • Because CMVC data may be accessed remotely (id.), IBM could grant to SCO remote access to CMVC data to respond to SCO's document requests.

It is unclear how IBM can credibly attempt to argue that it faces an undue burden in producing the requested information. The Court should not be distracted by IBM's protestations concerning the Sontag declaration when the real issue is IBM's attempt to avoid its discovery obligations and its failure to establish that it would experience any burden in meeting those obligations.

B.    The Sontag Declaration Is Based on Personal Knowledge.

Mr. Sontag's declaration is based on personal knowledge, notwithstanding IBM's assertions to the contrary. It is not offered as an "expert" opinion and such an opinion is not required—the entire purpose of the declaration is to introduce certain documents and make certain common-sense observations about them based on Mr. Sontag's personal knowledge of the field. In many regards, Mr. Sontag's observations simply reflect a permissible summary of the documents referenced. See Utah R. Evid. 1006. Moreover, as set out in the additional declaration included herewith, Mr. Sontag has many years of experience supervising the source code control process.

Federal Rule of Evidence 701 provides: "If the witness is not testifying as an expert, the witness' testimony in the form of opinions or inferences is limited to those opinions or inferences which are (a) rationally based on the perception of the witness and (b) helpful to a clear understanding of the witness' testimony or the determination of a fact in issue." Thus, testimony may be given by a witness with personal knowledge of the subject matter of the testimony or in the form of opinions or inferences rationally based on perception. See Fenstermacher v. Telelect. Inc., 21 F.3d 1121 (Table), 1994 WL 118046, at *5 (10th Cir. Mar. 28, 1994); see also Visser v. Packer Eng'g Assocs., Inc., 924 F.2d 655, 659 (7th Cir. 1991) ("'[P]ersonal knowledge' includes inferences—all knowledge is inferential—and therefore opinions. But the inferences and opinions must be grounded in observation or other first-hand personal experience. They must not be flights of fancy, speculations, hunches, intuitions, or rumors about matters remote from that experience.") (citations omitted). "A witness may give an opinion when he has personal knowledge of the facts." Kloepfer v. Honda Motor Co., Ltd., 898 F.2d 1452, 1459 (10 th Cir. 1990). "If a witness is not testifying as an expert, his testimony need be rationally based on his perception and helpful to determination of a fact in issue to be admissible." Id.

As Mr. Sontag explicitly states, his declaration is based upon his review of documents, which is a permissible basis upon which to submit an affidavit. See, e.g., In re Tex. E. Transmission Corp. PCB Contamination Ins. Coverage Lit., 870 F. Supp. 1293, 1304 (E.D. Pa. 1992) (reliance on review of documents to acquire knowledge and vouching for truth of statements satisfied personal knowledge requirement for witness to testify about events that occurred before his term of responsibility), aff'd, 995 F.2d 219 (3d Cir. 1993); Fla. Farm Bureau Mut. Ins. Co. v. B & B Miller Farms. Inc., No. 87-1021-C, 1991 WL 201188, *11 (D. Kan. Sept. 17, 1991) (witness statement based on review of documents admissible). The Sontag declaration is based in large part on the numerous publicly available documents related to CMVC that IBM has published and that are posted on the Internet. See e.g., Sontag Decl. ¶¶ 5-12.

In additional to his review of IBM documents, and as set forth in his supplemental declaration, Mr. Sontag has an academic background in relevant areas of Computer Science and many years of experience supervising the source code control process. While at Novell, one of the positions that he held was director of program management. In that position he had overall responsibility for the development and release of the NetWare 4.0 product—an enormous project that included over 500 software developers, testers and documentation writers. Mr. Sontag was very familiar with the source control system used at Novell and implemented a number of stringent source lock-down procedures. From 1996 until 2000, as CTO of a company that he co-founded, Mr. Sontag had overall responsibility for software development, technical strategy, intellectual property and information systems as well as general executive management, and he led the evaluation and selection process of the source control and source management system that was used by the development team.

The law is clear that Mr. Sontag is permitted to give opinions and make inferences based on his review of the documents and his experience in the industry. See Fenstermacher, 21 F.3d 1121 (Table), 1994 WL 118046, at *5. Far from "flights of fancy, speculations, hunches, intuitions or rumors," Visser, 924 F.2d at 659, Mr. Sontag's declaration is based on his experience and his review of IBM's own documents.

Equally misleading is IBM's argument (to which it devotes minimal attention) that Mr. Sontag is not qualified as an expert. Although Mr. Sontag could certainly qualify as an expert in this field, that is a red herring: The subject matter of his declaration does not require an expert opinion. The Federal Rules of Evidence permit him to provide testimony in the form of opinions or inferences here, where they are rationally based on his perception (resulting from his experience in the industry and review of IBM's documents) and are helpful to a clear understanding of the determination of this discovery dispute. See Fed. R. Evid. 701. 4

C.    If the Court Concludes That an Expert Opinion Is Required, Mr. Sontag Qualifies as an Expert.

Although Mr. Sontag need not be qualified as an expert because his declaration is based on personal knowledge, Mr. Sontag in any event qualifies as an expert for purposes of resolving the discovery dispute currently before the Court.

To permit expert testimony, the Court must determine whether the proposed expert witness is qualified by "knowledge, skill, experience, training or education" to render an opinion. The dispositive question with regard to qualification is whether the opinion is "within the reasonable confines" of the expert's subject area. Burton v. RJ. Reynolds Tobacco Co., 183 F. Supp. 2d 1308, 1313-1314 (D. Kan. 2002). As discussed above, Mr. Sontag has many years experience, both academic and practical, in this area. He has been responsible for supervising the source code control process on complex projects involving hundreds of developers at Novell, and has had related experience at his own company, where he had overall responsibility for software development, technical strategy, intellectual property and information systems as well as general executive management. In those capacities he has led the evaluation and selection process of the source control and source management system used by the large development teams and he has first-hand, personal knowledge of the functions and features of those systems. There can be little dispute that Mr. Sontag possesses the knowledge, skill and experience to properly render an opinion here.

D.    IBM's Request to Strike the Entire Declaration Is Meritless.

Although Mr. Sontag's declaration is entirely proper and should not be stricken, in the event that the Court finds that some portion of Mr. Sontag's declaration should not be considered, the proper remedy is not the sweeping suggestion by IBM that the entire declaration be stricken. Rather, if the Court concludes that it will not consider some part of the declaration, the proper approach is to strike only that portion, not the entire declaration. See, e.g., Trestle & Tower Eng'g. Inc. v. Star Ins. Co., 13 F. Supp. 2d 1166, 1167 n.l (D.Kan. 1998) ("The court shall not strike the affidavit, but shall simply disregard those portions of the affidavit not based on personal knowledge"); A.H.L. Inc. of Del. v. Star Ins. Co., 10 F. Supp. 2d 1216, 1217 n.l (D. Kan. 1998) (same). Here, IBM seeks to strike Mr. Sontag's entire declaration without even specifying the particular paragraphs of Mr. Sontag's declaration IBM suggests are flawed.

As noted above, Ms. Thomas refuses to state what parts of Mr. Sontag's declaration she rejects; rather she states only that the assumptions underlying Mr. Sontag's top-down approach are flawed. (8/4/04 Thomas Decl. ¶ 5.) Given that Mr. Sontag's discussion of this approach constitutes less than one-quarter of his declaration (at most 10 paragraphs of 46), it is difficult to see how IBM can argue that, even if it is flawed (which it is not), that somehow poisons the entire declaration. IBM says nothing, for example, about Mr. Sontag's testimony relating to the structure of CMVC files, the ability to provide remote access, IBM's contradictory public statements about CMVC's functions, the ability to provide discovery in electronic form, or the RCS system. All of these areas IBM simply ignores.


SCO respectfully requests, for the reasons set forth above, that the Court deny IBM's Motion to Strike the Declaration of Chris Sontag or, in the alternative, order the immediate deposition of Joan Thomas on the issues herein.

DATED this 26th day of August, 2004.

Brent O. Hatch
Mark F. James

Robert Silver, Esq. (admitted pro hac vice)
Stephen N. Zack (admitted pro hac vice)
Mark J. Heise (admitted pro hac vice)

Attorneys for The SCO Group, Inc.

1 See SCO's May 28, 2004 Reply Memorandum in Support of its Motion to Amend the Scheduling Order and July 12, 2004 Reply Memorandum Regarding Discovery.

2Ms. Thomas does not know Mr. Sontag and has never spoken to him about his professional background or knowledge of CMVC or similar systems. She is testifying here to her opinion—that which is "clear" to her "from reading the Sontag Declaration." Although Ms. Thomas' conclusions happen, in this case, to be wrong, they are a perfectly valid subject for testimony. By IBM's logic, however, Ms. Thomas' entire declaration should be stricken—she neither has, nor claims, any personal knowledge of Mr. Sontag's state of mind whatsoever.

3Commentary to the 1980 Amendment to Rule 34(b) states: "The Committee is advised that, 'It is apparently not rare for parties deliberately to mix critical documents with others in the hope of obscuring significance.' Report of the Special Committee for the study of Discovery Abuse, Section of Litigation of the American Bar Association 22 (1977). The sentence added by this subdivision follows the recommendation of the Report."

4As noted above, IBM states that Mr. Sontag's declaration is "riddled with inaccuracies," yet cites only Ms. Thomas' declaration, which mentions only a single point of disagreement with Mr. Sontag and explicitly refuses to enumerate what those "inaccuracies" might be. See S.E.C. v. Wolfson, 309 B.R. 612, 629-30 (Bankr. D. Utah 2004) (declining to strike affidavit in the absence of specific examples of any claimed inaccuracies or evidentiary problems).


Plaintiff, The SCO Group, hereby certifies that a true and correct copy of the foregoing MEMORANDUM IN OPPOSITFON TO DEFENDANT INTERNATIONAL BUSINESS MACHINES CORPORATION'S MOTION TO STRIKE THE JULY 12, 2004 DECLARATION OF CHRIS SONTAG was served on Defendant International Business Machines Corporation on this 26th day of August, 2004, by U.S. mail, postage prepaid, addressed to:

Alan L. Sullivan, Esq.
Todd M. Shaughnessy, Esq.
Snell & Wilmer L.L.P.
Copy to:
Evan R. Chesler, Esq.
Cravath, Swaine & Moore LLP

Donald J. Rosenberg, Esq.

Attorneys for Defendant/Counterclaim Plaintiff IBM Corp.


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 )