1055 - SCO's Sur-Surreply Memorandum in Further Support of SCO's Motion to Amend its December 2005 Submission

1056 - SCO's Sur-Surreply Memorandum in Further Support of its Objections to the Magistrate Judge's Order on IBM's Motion to Confine

Yes, two sur-sur-replies in which SCO begs the court to let it use some more of its theories and evidence that it didn't reveal by the deadline, please. Except they aren't all that polite, I don't think. SCO wasn't obligated, it argues in 1056, to have its experts finish their analysis by December 22, 2005:

The Magistrate Judge's Order confuses the obligation to disclose misused material with the accumulation of supporting evidence and with the theories of literal and non-literal infringement that are naturally found in experts, and are clearly allowed by law in this Circuit.

Well, that's a winning argument: the judge got confused. Another argument is that the court's orders were not clear. Right. That's the ticket. SCO didn't know it was supposed to do what all those court orders said it should do.

What SCO wants is what it always wants, to get to do what it wants:

SCO respectfully submits, for the reasons stated above and in SCO's previous memoranda, that the Court should reverse the Magistrate Judge's Order and permit SCO to proceed with all theories of liability disclosed in its expert reports, including copyright infringement based on both literal and non-literal elements of the copyrighted works as well as the selection, arrangement, and coordination of elements in the copyrighted works.

This wouldn't be about trying to avoid having to pay for all those public comments by Darl about a mountain of copyright infringement, by any chance, would it?

I haven't yet had a chance to read them carefully, but I see one argument that is interesting. In 1055, SCO argues that even if IBM's motion to confine was rightly decided, and a sanction was appropriate back then, SCO's motion to amend should be granted because there's lots of time now before trial, since Novell is going first, so "any possible prejudice can be cured" while we all wait for Novell to be decided. What I haven't sorted out yet is who this memo goes to. Did SCO get permission to file two sur-sur-replies, one to each judge? I'll look into that.

You have to hand it to BS&F. They think of absolutely every possible argument. But the net effect, which they probably don't point out, would be that the court's many orders to put all allegedly infringed material on the table by the deadline would be flouted and SCO would be sanction-free and able to present all the materials it failed to present by the deadline. Somehow that doesn't sound quite right. Why would any court reward a party for disobeying its orders? The issue wasn't just prejudice to IBM, which remains, according to IBM. The issue is obeying court orders. Well, you argue what you think you can.

The docket also shows some new deadlines:

#1053 - Order Re Extension of Deadlines

#1052 - Order Granting Stipulation and Joint Motion for Extension of Time

The first Order, 1053, gives SCO until May 18 to file its Reply Memo in Support of its Motion for Reconsideration of the Order Denying SCO's Motion for Relief for IBM's Spoliation of Evidence and its Reply Memo in Support of its Objections to the Order Denying SCO's Motion for Relief for IBM's Spoliation of Evidence.

The second order, 1052, represents a change in the schedule:

Rule 26(a)(3) Disclosures are now due on June 13;
The deadline for exchanging jury instructions is now June 27;
Motions in Limine are now due on July 9

The May 25 deadline for responding to requests for admission stands unchanged. As you can see on the IBM Timeline page, these are not big changes. The Motions in Limine deadline, for example, was June 4, and now it's July 9. The jury instructions deadline was May 28; now it's June 27; and the Rule 26(a)(3) disclosure deadline was May 14 and now it's June 13. It probably just means everyone is scrambling to meet the deadlines, and there are a few items still up in the air, so they agreed to give each other a little more time.

Here is the complete Docket, with all the entries, and as you can see, there is a bit of a mixup, so it's possible that I'm missing something:

1048 - Filed & Entered: 05/10/2007
Terminated: 05/11/2007
Motion for Extension of Time
Docket Text: Stipulated MOTION for Extension of Time for Extension of Deadlines filed by Plaintiff SCO Group. (Attachments: # (1) Text of Proposed Order) Motions referred to Brooke C. Wells.(Normand, Edward)

1049 - Filed & Entered: 05/11/2007
Order on Motion for Extension of Time
Docket Text: DISREGARD - SEE ENTRY [1052] for correct order on motion. ORDER granting [1048] Motion for Extension of Time. Signed by Judge Dale A. Kimball on 5/9/07. (blk) Modified on 5/11/2007 SEE ENTRY [1052] for correct order on motion (blk).

1050 - Filed & Entered: 05/11/2007
Motion for Extension of Time to File Response/Reply
Docket Text: Stipulated MOTION for Extension of Time to File Response/Reply filed by Plaintiff SCO Group, Counter Defendant SCO Group. (Attachments: # (1) Text of Proposed Order) Motions referred to Brooke C. Wells.(Normand, Edward)

1051 - Filed & Entered: 05/11/2007
Order on Motion for Extension of Time to File Response/Reply
Docket Text: DISREGARD- WRONG ORDER ENTERED FOR THIS MOTION THIS MOTION IS NOT RULED ON YET. ORDER granting [1050] Motion for Extension of Time to File Response/Reply Replies due by 5/18/2007.. Signed by Judge Dale A. Kimball on 5/11/07. (blk) Modified on 5/11/2007 (BLK).

1052 - Filed & Entered: 05/11/2007
Order on Motion for Extension of Time
Docket Text: ORDER granting [1048] Motion for Extension of Time. Rule 26(a)(3) Disclosures due 6/13/07. Exchange of Jury Instructions due 6/27/07. Motions in Limine due 7/9/07. Signed by Judge Dale A. Kimball on 5/11/07. (blk)

1053 - Filed & Entered: 05/11/2007
Order on Motion for Extension of Time to File Response/Reply
Docket Text: ORDER granting [1047] Motion for Extension of Time to File Response/Reply re [986] MOTION FOR RECONSIDERATION BY THE MAGISTRATE COURT OF THE ORDER DENYING SCOS MOTION FOR RELIEF FROM IBMS SPOLIATION OF EVIDENCE. Replies due by 5/18/2007. Signed by Judge Dale A. Kimball on 5/11/07. (blk)

1054 - Filed & Entered: 05/11/2007
Modification of Docket
Docket Text: Modification of Docket: Error: Clerk initially docketed orders on wrong motions. Correction: Clerk has corrected this and re-entered the orders with the correct motions. They have been sent out to all counsel as usual. re [1051] Order on Motion for Extension of Time to File Response/Reply, [1053] Order on Motion for Extension of Time to File Response/Reply, [1049] Order on Motion for Extension of Time, [1048] Stipulated MOTION for Extension of Time for Extension of Deadlines, [1052] Order on Motion for Extension of Time, [1047] Stipulated MOTION for Extension of Time to File Response/Reply as to [986] MOTION FOR RECONSIDERATION BY THE MAGISTRATE COURT OF THE ORDER DENYING SCOS MOTION FOR RELIEF FROM IBMS SPOLIATION OF EVIDENCE JointlyStipulated MOTION for Extension of Time to File Response/Reply as to [986] MOTION FOR RECONSIDERATION BY THE MAGISTRATE COURT OF THE ORDER DENYING SCOS MOTION FOR RELIEF FROM IBMS SPOLIATION OF EVIDENCE Jointly. (blk)

1055 - Filed & Entered: 05/11/2007
Reply Memorandum/Reply to Response to Motion
Docket Text: REPLY to Response to Motion re [913] MOTION to Amend/Correct DECEMBER 2005 SUBMISSION (SCO's Sur-Surreply Memorandum) filed by Plaintiff SCO Group, Counter Defendant SCO Group. (Attachments: # (1) Exhibit A-B)(Normand, Edward)

1056 - Filed & Entered: 05/11/2007
Reply Memorandum/Reply to Response to Motion
Docket Text: REPLY to Response to Motion re [917] Plaintiff's MOTION for Leave to File OVER LENGTH MEMORANDUM IN SUPPORT OF ITS OBJECTIONS TO THE MAGISTRATE JUDGE'S ORDER ON IBM'S MOTION TO CONFINE No correct entry to link to:SCO's Sur-surreply re: Docket Entry 921 - MEMORANDUM IN SUPPORT re 899 Objection to Magistrate Judge Decision to District Court filed by Plaintiff SCO Group. filed by Plaintiff SCO Group, Counter Defendant SCO Group. (Normand, Edward)

Court clerks are used to complexity. But this case, as you can see, is very likely having the clerk going home and telling his or her significant other that this case takes the cake. No one can keep all the SCO twists and turns straight, not even the clerk, and apparently not even the parties. The date for filing three motions, finalized in docket number 1050 as the 18th, was the 25th, then the 18th, or vice versa. Everyone is having trouble keeping up, including me. I note in 1050, however, that SCO has until May 25 to file its reply in support of its Motion, which IBM opposed, to Deem a Prospective Third-Party Deposition in Related Litigation to be a Deposition Taken in this Case As Well. SCO spelled it Perspective, but we know what it meant.

Here is SCO's Sur-Reply Memorandum in Further Support of SCO's Motion to Amend its December 2005 Submission (PDF), as text.

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

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

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

Devan V. Padmanabhan (admitted pro hac vice)
DORSEY & WHITNEY LLP
[address]
[phone]
[fax]

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

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

Attorneys for Plaintiff, The SCO Group, Inc.

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF UTAH

(1)

TABLE OF CONTENTS

i (2)

TABLE OF AUTHORITIES

ii (3)

PRELIMINARY STATEMENT

IBM's Sur-Reply memorandum rehashes procedural arguments IBM has already raised in its continued efforts to avoid the merits of SCO's claims, and fails to distinguish or rebut the case law cited by SCO supporting the motion to amend. IBM attempts to show that it would be prejudiced if SCO's motion to amend were granted, but the time remaining before trial, which has not been set, would be more than sufficient to conduct limited additional discovery, if any were necessary, and would actually allow more time for such discovery than was contemplated under the initial scheduling order. Furthermore, the only reason such discovery is necessary at all is because IBM refused to depose SCO's experts regarding these issues and instructed its experts not to address these aspects of SCO's expert reports. Accordingly, IBM's claims of prejudice must fail.

The amendment SCO seeks would allow the parties to resolve whether the Linux operating system, used by millions across the world, infringes SCO's intellectual property rights (assuming, also, that the court rejects IBM's effort to artificially limit the scope only to technology wholly within the Linux kernel). This is a win-win scenario for IBM, which purported to seek resolution of the question of infringement on the merits when it filed its declaratory judgment action; for SCO, whose UNIX business has been crippled by the improvement of Linux with its own intellectual property; and for the public, which has a right to a resolution of this dispute on the merits. SCO respectfully requests that its motion to amend be granted.

1 (4)

ARGUMENT

I. IBM CONFLATES THE MOTION TO AMEND WITH SCO'S OBJECTIONS ON THE MOTION TO CONFINE, AND PROPOSES THE WRONG STANDARD FOR DECIDING SCO'S MOTION TO AMEND.

In its Sur-Reply, IBM has again conflated SCO's Motion to Amend with SCO's Objections to the Order Granting IBM's Motion to Confine, and this tactic continues to be misleading. While SCO contends that IBM's motion to confine was wrongly decided, SCO's Motion to Amend can and should be granted even if this Court concludes that the Magistrate Judge's order on the motion to confine was correct within the confines of the pretrial and trial schedule that was in effect when IBM initially brought its motion. What may be a justifiable decision, or even sanction, when trial is just months away, becomes insupportable when a longer period before trial is available and when any possible prejudice can be cured during that time. See, e.g., United States v. Golyansky, 291 F.3d 1245, 1250 (10th Cir. 2002); Hertz v. Luzenac Am., Inc., No. CIVA04CV1961TBCBS, 2006 WL 994431, at *21 (D. Colo. Apr. 13, 2006) (Ex. A); Scott v. IBM Corp., 196 F.R.D. 233, 247 n.9 (D.N.J. 2000); cf. Wolfson v. Lewis, 168 F.R.D. 530, 533 (E.D. Pa. 1996); Karbon v. Turner, No. 91-C-337, 1991 WL 319976 at *2 (E.D. Wis. Dec. 16, 1991) (Ex. B); Pub. Serv. Co. of N.H. v. Westinghouse Elec. Corp., 685 F.Supp.1281 (D.N.H. 1988).

Notwithstanding its Sur-Reply, IBM has made no effort to distinguish the above cases (discussed in SCO's Reply at 1-2), and has failed to explain why the Court should disregard the controlling law in Golyansky that the concept of prejudice does not encompass the expense of additional trial preparation and that prejudice should be found only where there is some evidence that a delayed disclosure impacted a party's ability to prepare for trial. Rather, IBM

2 (5)

cites (at 36 n.31) an older Tenth Circuit case and a Colorado district court case in an apparent attempt to show that the Court should find it was prejudiced simply by the inconvenience of having to conduct limited additional discovery before trial and respond to SCO's claims on the merits. The cases do not support that conclusion. In Jones v. Thompson, 996 F.2d 261, 264 (10th Cir. 1993), the Tenth Circuit plainly recognized that the plaintiffs' delay and noncompliance with court orders had prejudiced the defendants ability to prepare their defense, not simply additional expense in trial preparation. Id. ("At the end of it all the Defendants had still not been able to complete their deposition of George Jones nor even begin their deposition of Peggy Jones.")

The other case IBM cites, Hertz v. Luzenac Am., Inc., No. CIVA04CV1961TBCBS, 2006 WL 994431, at *21 (D. Colo. Apr. 13, 2006), strongly supports SCO's position. In that case, the plaintiff tried to prevent the defendant from using at trial certain damages material that had not been disclosed in its Rule 26 disclosures. The court concluded that the parties' failure to comply with its Rule 26 disclosure obligations was not substantially justified. The court nevertheless allowed the material to be used, noting that "the original trial date was vacated for reasons completely unrelated to the pending motions and alleged disclosure violations," and that the "absence of a fixed trial date affords a greater ability to cure any prejudice to the parties and minimizes the possibility of future disruption of the District Court's docket." Id. at. *21. The court further reasoned that, while "counsel should have been more diligent in providing the required disclosures, all parties are now on notice of the damages sought in this action." Id. These considerations are plainly applicable here. Regarding the language IBM quotes that delay and mounting attorneys' fees can constitute prejudice, the

3 (6)

court actually noted "that consideration carries less weight in this case given the attorney fees that must have been generated by the parties' vituperative wrangling through the discovery process." Id. Clearly, that observation applies here, where both parties have had to file numerous motions to compel. Moreover, SCO maintains, and has shown, that no delay would be occasioned by SCO's motion.

In addition, IBM continues to argue that SCO has to show "extremely compelling circumstances" for its motion to be granted. This is not correct. The impetus and sole basis for the Motion to Amend is the substantial change in trial schedule that has already occurred — independent of SCO's motion. It is this change that permits the motion to amend to be granted, without prejudice to any party, and without any change to the trial date. In addition, SCO has clearly shown "extremely compelling circumstances" supporting the amendment. Permitting the amendment is the only way to ensure that the question whether Linux infringes SCO's intellectual property will be resolved on the merits. Furthermore, as set forth below, IBM would have as much time to conduct limited discovery and analysis of the material as originally contemplated under the initial scheduling order. Finally, denial of the amendment would prejudice SCO because it would impede SCO's ability to seek redress for the infringement of Linux. These factors constitute the "extremely compelling circumstances" IBM contends are necessary.

II. SCO ACTED IN GOOD FAITH WHEN INTERPRETING THE SCHEDULING ORDER.

In an effort to show that SCO has acted in bad faith with regard to the material at issue in SCO's motion to amend, IBM points to the Court's prior ruling on IBM's Motion to Limit. However, a different question was at issue in the Court's resolution of IBM's Motion to Limit:

4 (7)

whether certain material disclosed in SCO's December Submission was identified with sufficient particularity, and whether SCO's failure to identify the material with more particularity constituted bad faith. The Court's finding regarding SCO's intent as to that different material has no bearing here. The Court could consistently find that SCO should have identified that different material with more specificity,¹ and also find that SCO's conduct with regard to the material at issue in this motion was in good faith. For instance, as discussed in SCO's opening brief and reply, as well as SCO's objections, even if the Court concludes that the material at issue in this motion should have been disclosed in SCO's December Submission, it could conclude that SCO's actions resulted from the reasonable confusion created by IBM's own representations regarding when expert analysis should be disclosed, and by the later deadline that was established for disclosure of expert analysis. Accordingly, IBM's attempt to bootstrap the Court's rulings on its Motion to Limit (which is still subject to a motion for reconsideration) should be rejected.

IBM further argues (IBM Sur-Reply at 30) that SCO could not have acted in good faith because it did not comply with "what the rules require." But that misses the point, and disregards the law cited by SCO. SCO contends here that its interpretation of what the rules required, even if ultimately not correct, was reasonable and in good faith. There is not a shred of evidence suggesting that SCO's conduct in this regard was bad faith or willful, or anything other than a different understanding than IBM now purports to have over what was required by the unique scheduling order entered in this case. In addition, the law favors disposition of cases on the merits, even if a party did not comply with rules, where, as here, sufficient time

5 (8)

exists before trial to cure any resulting prejudice. Both the Hertz case cited by IBM and the other cases that SCO cites support that proposition.

III. IBM WOULD NOT BE PREJUDICED BY THE PROPOSED AMENDMENT.

A. IBM Overstates the Scope of Discovery and Expert Analysis Needed To Rebut SCO's Evidence.

IBM asserts that allowing SCO's motion to amend would require extensive additional discovery and expert analysis. Indeed, IBM claims that SCO's expert reports "challenge essentially every file in Linux," that IBM would have to conduct an "arduous" "line-by-line" analysis of "over one hundred thousand files" of Linux source code, and that such discovery and analysis would take "at least an additional year." (IBM Sur-Reply at 30, 32-34.) IBM would like to convince the Court that, in order to defend effectively against the evidence of infringement presented in SCO's expert reports, IBM would have to analyze every single file in Linux with excruciating detail. This is far from true. Indeed, this file-level myopia is misguided at best and disingenuous at worst.

6 (9)

B. IBM Exaggerates the Scope of Material at Issue in SCO's Proposed Amendment.

IBM has significantly exaggerated the scope of material at issue in SCO's Motion to Amend in order to create an appearance of prejudice. Dr. Cargill's copyright infringement analysis does not, as IBM contends (IBM Sur-Reply at 17), "challenge essentially every file in Linux." Rather, Dr. Cargill's expert reports clearly show that Linux has copied the selection, arrangement, and coordination of the following literal and non-literal UNIX elements: (1) the UNIX structure embodied in 96 copied system calls, (2) the combination of 8 characteristics that make up the UNIX file system, (3) the UNIX Streams module, and (4) the ELF format. This material does not implicate "every file in Linux." (Cargill Expert Report dated May 15, 2006 at 16-17, 23-36, 41-52, 57-58, 63-81 (11-11-06 Decl. of Brent Hatch, Ex. 274 [DEN 876]); Cargill Expert Report dated August 28, 2006 at 20-22, 31-34 (11-11-06 Decl. of Brent Hatch, Ex. 276 [DEN 876]).) It is neither necessary nor useful to perform a micro-grain, line-by-line, file-by-file analysis of Linux to try to rebut the validity of SCO's non-literal infringement evidence. Essentially, SCO's non-literal infringement evidence shows that, even if the lines and files of source code in Linux were not copied from UNIX,² the structural organization of such lines and files was copied from UNIX.

The facts here are analogous to a photo mosaic, in which hundreds or thousands of individual photographs are organized in a manner that conveys a larger image. SCO's evidence shows that the overall image conveyed by Linux is copied from UNIX, even if the individual photographs that make up the mosaic are not. In response, IBM's position is to (1) ignore SCO's

7 (10)

argument that the overall image was copied from UNIX, (2) insist that SCO is actually challenging each individual photograph used to make up the overall image, and (3) argue that analyzing each individual photograph would require extensive additional discovery. In fact, an analysis of each individual photograph would be irrelevant to SCO's claims of infringement.

In support of this code-level misdirection, IBM argues (IBM Sur-Reply at 34) that analysis of the non-literal infringement shown in SCO's expert reports "ultimately requires the same arduous code comparison as is required in order to respond to SCO's literal claims" because "just as a novel's plot can always be found in the novel's text, so too are the more abstract elements claimed by SCO embodied in the system's code." It is true that non-literal elements of UNIX and Linux are embodied in literal code. But this does not mean that IBM must perform a line-by-line analysis of over 100,000 files of source code in Linux in order to rebut SCO's expert reports. This is true for several reasons.

As described above, even if IBM could prove that every line of code in Linux was "independently created," which it cannot, that would not show that the non-literal organization of Linux was independently created. IBM's argument that an "arduous" literal-level analysis is necessary to rebut SCO's non-literal infringement evidence is no different than claiming that every grain of sand in a sand castle must be analyzed to show that the overall design was not taken from a pre-existing sand castle, or that each word of a book must be analyzed in order to show that chapter order and headings were not taken from a pre-existing book. Such analysis is neither necessary or helpful. Non-literal infringement is a well established aspect of copyright law that IBM should expect to address in connection with its defense of Linux.

8 (11)

IBM's argument is especially disingenuous because SCO has identified the precise literal and non-literal elements it claims are infringed by Linux. IBM further characterizes SCO's expert reports as challenging "essentially every line" of Linux, but that is not true. As discussed above, SCO's expert reports identify as the infringed material the selection, coordination, and arrangement of (1) the UNIX structure embodied in 96 copied system calls, (2) the combination of 8 characteristics that make up the UNIX file system, (3) the UNIX Streams module, and (4) the ELF format. (Cargill Expert Report dated May 15, 2006 at 16-17, 23-36, 41-52, 57-58, 63-81 (11-11-06 Decl. of Brent Hatch, Ex. 274 [DEN 876]); Cargill Expert Report dated August 28, 2006 at 20-22, 31-34 (11-11-06 Decl. of Brent Hatch, Ex. 276 [DEN 876]).) To the extent IBM would have to analyze any literal source code, such analysis would be limited to (1) the 96 identified system calls embodying the copied UNIX structure, (2) the source code implementing the 8 identified UNIX file system characteristics, (3) the source code implementing the ELF format, and (4) the source code implementing the UNIX Streams module. In addition, multiple system calls, ELF, and multiple stream files are indisputably included in SCO's December Submission, and IBM must deal with those aspects of SCO's infringement claim in any event.

Because these infringed elements are precisely defined in SCO's expert reports, SCO's statements from 2004 suggesting that all Linux and UNIX source code should be analyzed — on which IBM relies so heavily — are inapplicable to the current situation. (See IBM Sur-Reply at 34-35.) These statements were made at a time when the scope of the infringement had not been determined. Because the scope of infringement has been narrowed to precisely identified elements, IBM would only have to analyze only those elements. Furthermore, there is no need

9 (12)

to conduct discovery on every Linux developer, as IBM claims (IBM Sur-Reply at 33 n.26), when SCO has identified the specific material on which its claims are based.

1. IBM exaggerates the scope of analysis required.

IBM also errs in purporting to describe the analysis necessary to perform on the material identified as infringing in SCO's expert reports. IBM lists eleven characteristics that it would purportedly have to analyze in relation to each individual file in Linux if it were forced to rebut the evidence of infringement in SCO's expert reports. (IBM Sur-Reply at 31-32.) However, IBM's experts did not even analyze all of these characteristics with regard to individual files that were explicitly listed in SCO's December Submission. Rather, IBM simply performed an analysis on a representative line or file and applied its conclusions to all lines or files of that type. For example:

• IBM's experts analyzed a single structure declaration, a single #define statement, and a single function prototype, and then applied conclusions regarding originality to all "such" structure declarations, #define statements, and function prototypes (material contained in header files). (Kernighan Expert Report dated May 18, 2006 at 13-15 (09-25-06 Decl. of Todd Shaughnessy, Ex. 213 [DEN 774]).) This analysis is also applied to the ELF-related source code. (Id. at 23).

• IBM's experts made blanket idea/expression and merger conclusions for header files in general, rather than analyzing individual lines of code or individual header files. (Id. at 20.)

• Although IBM says it evaluated the "significance" of code in relation to System V as a whole, its expert reports show no evidence of this evaluation other than a quantitative analysis, which does not require analysis of individual files and is not time-consuming. (Id. at 29-31; Kernighan & Davis Expert Report dated July 17, 2006 at 41 (09-25-06 Decl. of Todd Shaughnessy, Ex. 214 [DEN 774]); Kernighan & Davis Expert Report dated August 28, 2006 at 11-18 (09-25-06 Decl. of Todd Shaughnessy, Ex. 215 [DEN 774]).)

• In other words, IBM's lawyers now argue that IBM would have to undertake an "arduous" line-by-line analysis of every file in Linux with regard to numerous characteristics in order to rebut the evidence in SCO's expert reports, because this is

  10 (13)

  the only way that IBM can show it was prejudiced, and avoid having to rebut SCO's evidence on the merits. However, IBM's experts did not even undertake this individualized analysis with regard to the code that was identified in SCO's December Submission and that they knew they had to analyze.

IBM also exaggerates the non-literal analysis it would have to perform by listing duplicative and unnecessary types of analysis. For example, IBM lists (IBM Sur-Reply at 32-33) "[a]nalyz[ing] the history of the particular file system technology," "[r]esearch[ing] texts relating to the technology," and "[t]rac[ing] the creation of the allegedly infringing elements" as separate endeavors to give the appearance of burdensome work, but these are simply different ways to characterize the same analysis. Also, it is not necessary to analyze the originality of each file system characteristic mentioned in SCO's expert reports, because SCO's claim is that the combination of non-literal elements infringes. This is simply another example of IBM's focus on individual elements when SCO is claiming infringement of the non-literal combination of elements.

2. IBM's exaggeration is shown by the discovery schedule.

IBM's dramatic exaggeration of the time and effort needed to rebut SCO's evidence of infringement assumes that the original discovery calendar was entirely flawed. The original discovery schedule called for IBM to submit its first expert reports five months after submission of SCO's December Submission. IBM's current claim that it would need at least an extra year to analyze the material in SCO's expert reports amounts to a post-hoc attempt to amend the discovery schedule to its benefit.³ Furthermore, nearly a full year has passed since IBM received Dr. Cargill's first expert report, which contained a precise description of SCO's

11 (14)

infringement theory and the evidence on which it is based. As described more thoroughly below, any possible prejudice or delay that IBM now faces, after receipt of Dr. Cargill's expert report, is the result of IBM's conscious, tactical decision to ignore the content of that report — even refusing to depose him on contents IBM contended should not have been included. If IBM had dealt with the evidence presented instead of manufacturing prejudice to gain a tactical advantage, this would be a non-issue.

Moreover, IBM ignores SCO's argument in its reply (at 4-5) that, even if IBM were to start the analysis now of the material SCO seeks to add to its December Submission, IBM would have more time than originally allowed under the scheduling order that was in place at the time of the December Submission. Accordingly, there is no possible prejudice to IBM.

C. Any Prejudice Is of IBM's Own Making.

Almost a full year has passed since IBM received a copy of Dr. Cargill's first expert report. This is more than twice the time between SCO's filing of its December Submission and the parties' submission of initial expert reports. This is time in which IBM could have analyzed the material contained in SCO's expert reports. Instead, hoping to obtain a discovery windfall and exclude SCO's evidence of infringement, IBM has tactically refused to analyze such material. Any delay caused by allowing consideration of such material now is due to IBM's year-long refusal to address such material.

IBM argues that, although trial date is not set, the discovery required to address the material in SCO's expert reports would delay any contemplated trial date and a new round of briefing would be required to address such material. (IBM Sur-Reply at 35-36.) First, as described above, the discovery required to address such material is not nearly as extensive as

12 (15)

IBM makes it out to be and the original discovery calendar provided only five months from the time of the December Submission to undertake such analysis. IBM chose not to take any discovery on the material it challenges, told its experts not to analyze such material, and disregarded SCO's arguments relating to such material as a tactical measure so that it would be able to avoid dealing with such material. To now claim that it would be prejudiced by having to undergo a new round of briefing because it made a conscious decision to ignore the material in SCO's expert reports is simply complaining about the bed IBM made for itself.

IBM also claims that the prejudice caused by any delay will be incurable, because such delay will perpetuate "fear, uncertainty, and doubt" regarding Linux, "relitigating" issues will be expensive, and IBM relied on SCO's identification of 326 lines "of Linux." Any such problems are the direct result of IBM's own actions. First, SCO specifically identified over 1,000 lines of Linux kernel and non-kernel source code as infringing in its December Submission. IBM, not SCO, unilaterally limited its expert reports and briefing to 326 lines of source code residing within the Linux kernel. Accordingly, IBM has no basis for claiming that it relied on SCO's supposed identification of 326 lines of source code in its December Submission. IBM's unilateral decision to ignore the thousands of lines non-kernel Linux source code identified in SCO's December Submission, not any action by SCO, has ensured that the "fear, uncertainty, and doubt" surrounding the Linux operating system will remain after resolution of IBM's motion for summary judgment on its tenth counterclaim, because the majority of the infringing material will remain unadjudicated. Finally, any expense incurred by IBM due to "relitigating" this matter is directly caused by IBM's tactical refusal to rebut SCO's evidence and arguments that Linux infringes non-literal elements of UNIX when it had the opportunity to do so.

13 (16)

IBM also exaggerates the extent to which this issue would impact summary judgment briefing. The arguments raised by IBM in its briefing on copyright issues would also apply to the material at issue in this motion. Any additional briefing focused on new evidence purporting to contravene the evidence in SCO's expert reports could do no more than raise new disputed issues of material fact that preclude summary judgment. See Repp v. Webber, 132 F.3d 882, 890-91 (2d Cir. 1997). IBM cannot legitimately claim prejudice based on the time needed to produce briefs that would have no effect on the resolution of IBM's motions for summary judgment.⁴ Moreover, IBM had received all the material at issue here well before it filed its summary judgment motions. However, if, after analysis, IBM felt it necessary to bring an extremely narrow summary judgment motion on limited issues raised only by the material at issue in SCO's motion, there would be time to do so before trial.

IV. SCO WOULD BE PREJUDICED IF NOT PERMITTED TO AMEND ITS DECEMBER SUBMISSION.

Notably, in its opposition and again in its sur-reply, IBM has not attempted to argue that SCO would not be prejudiced if its Motion to Amend is not granted. Rather, IBM essentially contends that the prejudice to SCO is warranted and should be inflicted, even though denial of the motion may preclude a full resolution on the merits of SCO's claims that Linux infringes its intellectual property. In its sur-reply, IBM argues (at 22 n.22) that SCO would simply be losing the ability to conduct a trial by ambush. This allegation could not be further from the truth. As discussed, SCO reasonably believed (and maintains) that its interpretation of the

14 (17)

scheduling order was correct, and that IBM already knew in any event that SCO was contending that the structure of Linux infringes its UNIX copyrights (a fact IBM has not denied). The prejudice to SCO arises from the fact that IBM would have substantial portions of its timely disclosed expert analysis excluded based on the fact that SCO reasonably interpreted the requirements of the scheduling order, and complied with what it believed those requirements to be — even though more than enough time exists before trial to correct the mistake and avoid prejudice to either party.

IBM claims that SCO is "eliminating" one of the categories of discovery sanctions from the Court's arsenal. This is not so. When considering what sanction to impose, one of the fundamental considerations in the Tenth Circuit and elsewhere — regardless of whether the party was at fault — is the amount of time before trial, because that determines whether any prejudice can be cured. This motion is predicated on the fact that, in light of the adjustment of the trial schedule, any prejudice that might have existed can now be cured.

V. THE MATERIAL AT ISSUE WAS FULLY DISCLOSED IN SCO'S EXPERT REPORTS.

IBM incorrectly argues that "the material at issue was not fully disclosed in SCO's expert reports." (IBM Sur-Reply at 38.) This is, first of all, an entirely new issue raised in the Sur- Reply. The basis for IBM's complaint in its Motion to Confine, which gave rise to the instant motion, was that the SCO's expert reports contained information beyond the December Submission, not that material in the expert reports was insufficiently identified. Furthermore, IBM's argument ignores SCO's representations regarding the Rochkind material at issue. IBM claims that a segment of material in Rochkind's report was not sufficiently identified (IBM Resp. at 33), yet SCO has explicitly stated that it does not allege misuse of such material (SCO Reply

15 (18)

at 7). Such material is simply supporting evidence that a particular Linux contributor, identified in the December Submission as contributing specifically identified testing technology, did in fact work in this area of technology for IBM. Although IBM argues that "there would be no reason for SCO's motion" if such material were not allegedly misused material (IBM Sur-Reply at 38), this is not true. Because of the nature of the Magistrate Judge's order, and IBM's interpretation thereof, SCO needs clarification that material that is not alleged to be misused, but does support a finding of misuse, will not be excluded from this case, and therefore seeks to add that material to its December Submission.

In short, IBM has everything it needs in SCO's expert reports to conduct whatever analysis it deems necessary to respond to the material SCO is seeking to add to its December Submission. Moreover, IBM had a full and complete opportunity to cross examine SCO's experts on these issues at deposition, and would have time to conduct limited additional discovery before trial.

CONCLUSION

SCO respectfully submits, for the reasons stated above and in SCO's previous memoranda, that the Court should grant SCO's Motion to Amend and permit SCO to proceed with all theories of liability disclosed in its expert reports, including copyright infringement based on both literal and non-literal elements of the copyrighted works as well as the selection, arrangement, and coordination of elements in the copyrighted works.

16 (19)

DATED this 11th day of May, 2007.

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

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

DORSEY & WHITNEY LLP
Devan V. Padmanabhan

Counsel for The SCO Group, Inc.

By: /s/ Edward Normand

17 (20)

CERTIFICATE OF SERVICE

Plaintiff/Counterclaim-Defendant, The SCO Group, Inc., hereby certifies that a true and correct copy of the foregoing SCO'S SUR-SURREPLY MEMORANDUM IN FURTHER SUPPORT OF SCO'S MOTION TO AMEND ITS DECEMBER 2005 SUBMISSION was served on Defendant/Counterclaim-Plaintiff, International Business Machines Corporation, on this 11th day of May, 2007, via CM/ECF to the following:

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

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

/s/ Edward Normand

18 (21)