SCO v. IBM Memorandum in Opposition to IBM's Motion by BenSkerrett

VIEWS: 27 PAGES: 139

									Brent O. Hatch (5715) Mark F. James (525) HATCH, JAMES & DODGE 10 West Broadway, Suite 40 Salt Lake City, Utah 84101 Telephone: (801) 363-6363 Facsimile: (801) 363-6666 Robert Silver (admitted pro hac vice) Edward Normand (admitted pro hac vice) Sean Eskovitz (admitted pro hac vice) BOLES, SCHILLER & FLEXNER LLP 333 Main Street Armonk, New York 10504 Telephone: (914) 749-8200 Facsimile: (914) 749-8300 Stephen N. Zack (admitted pro hac vice) BOLES, SCHILLER & FLEXNER LLP Bank of America Tower - Suite 2800 100 Southeast Second Street Miami, Florida 33131 Telephone: (305) 539-8400 Facsimile: (305) 539-1307 Attorneys for The SCO Group, Inc. IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF UTAH THE SCO GROUP, INC., Plaintiff/Counterclaim-Defendant, v. INTERNATIONAL BUSINESS MACHINES CORPORATION, Defendant/Counterclaim-Plaintiff. [FILED UNDER SEAL] MEMORANDUM IN OPPOSITION TO IBM’S MOTION FOR SUMMARY JUDGMENT ON SCO’S BREACH-OFCONTRACT CLAIMS

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

CONFIDENTIAL INFORMATION- SUBJECT TO COURT ORDER THIS ENVELOPE SHALL NOT BE OPENED AND THE CONTENTS SHALL NOT BE DISPLAYED, COPIED OR REVEALED EXCEPT BY COURT ORDER OR BY THE WRITTEN AGREEMENT OF THE PARTIES

Brent O. Hatch (5715) Mark F. James (525) HATCH, JAMES & DODGE 10 West Broadway, Suite 40 Salt Lake City, Utah 84101 Telephone: (801) 363-6363 Facsimile: (801) 363-6666 Robert Silver (admitted pro hac vice) Edward Normand (admitted pro hac vice) Sean Eskovitz (admitted pro hac vice) BOLES, SCHILLER & FLEXNER LLP 333 Main Street Armonk, New York 10504 Telephone: (914) 749-8200 Facsimile: (914) 749-8300 Stephen N. Zack (admitted pro hac vice) BOLES, SCHILLER & FLEXNER LLP Bank of America Tower - Suite 2800 100 Southeast Second Street Miami, Florida 33131 Telephone: (305) 539-8400 Facsimile: (305) 539-1307 Attorneys for The SCO Group, Inc. IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF UTAH THE SCO GROUP, INC., Plaintiff/Counterclaim-Defendant, v. [FILED UNDER SEAL] INTERNATIONAL BUSINESS MACHINES CORPORATION, Defendant/Counterclaim-Plaintiff. Case No. 2:03CV0294DAK Honorable Dale A. Kimball Magistrate Judge Brooke C. Wells MEMORANDUM IN OPPOSITION TO IBM’S MOTION FOR SUMMARY JUDGMENT ON SCO’S BREACH-OFCONTRACT CLAIMS

TABLE OF CONTENTS TABLE OF AUTHORITIES ...........................................................................................................v INTRODUCTION ...........................................................................................................................1 BACKGROUND .............................................................................................................................1 SUMMARY OF ARGUMENT .......................................................................................................4 SCO’S RESPONSE TO IBM’S “STATEMENT OF UNDISPUTED FACTS” AND STATEMENT OF ADDITIONAL MATERIAL FACTS.............................................................10 A. B. C. UNIX Was a Seminal Software Innovation That AT&T Diligently Protected Through Its Software License Agreements ...........................................10 The Software Agreements Provide Broad Protections for the UNIX Intellectual Property...............................................................................................12 Extrinsic Evidence Confirms the Broad Protection of the Software Agreements ............................................................................................................14 1. 2. 3. 4. 5. D. E. Protection of Modifications and Derivative Works ...................................15 Protection of Methods and Concepts .........................................................25 The Ownership Language of the IBM Side Letter.....................................30 The Effect of the IBM Side Letter on Sequent ..........................................34 The $ echo Newsletter ...............................................................................36

The Alleged Novell Waiver Is Contrary to the Entire Novell Asset Purchase Transaction .............................................................................................37 The Alleged Linux Waiver ....................................................................................43 1. 2. 3. 4. SCO’s Efforts to Protect Its Rights in UNIX.............................................43 SCO’s Distribution of Linux......................................................................43 SCO’s Suspension of Its Distribution of Linux .........................................44 SCO’s Protection of Information on Its Website.......................................45

F.

SCO’s Application Under Rule 56(f) ....................................................................46 1. IBM’s Contract-Interpretation Arguments ................................................46
i

2. 3.

IBM’s Breach of the Software Agreements...............................................47 IBM’s Waiver Arguments..........................................................................48

LEGAL STANDARDS .................................................................................................................48 ARGUMENT.................................................................................................................................49 I. THE PLAIN LANGUAGE, THE DISPUTED EXTRINSIC EVIDENCE, AND THE PURPOSE OF THE UNIX SOFTWARE AGREEMENTS PRECLUDE IBM’S CLAIM THAT ONLY LITERAL COPYING OF UNIX SOURCE CODE IS PROHIBITED ...............................................................................................................50 A. Section 2.01 of Both Software Agreements Plainly and Unambiguously Extends All of the Contractual Restrictions Beyond UNIX Source Code, to Any Modifications and Derivative Works.........................................................53 1. Section 2.01 of Both Software Agreements Plainly and Unambiguously Extends All of the Contractual Restrictions Beyond Source Code, to Any UNIX Modifications or Derivatives. .........54 Section 7.06(a)’s Confidentiality Requirement Precludes IBM’s Narrow “Literal Source Code Only” Claim...............................................57 IBM’s “Literal Source Code Only” Reading Conflicts with the Restrictions Imposed by the UNIX Sublicensing Agreements..................58 IBM’s Reliance on the IBM Side Letter Is Misplaced. .............................60

2. 3. 4. B.

IBM’s Appeal to Extrinsic Evidence for Its Summary Judgment Motion Is Meritless As a Matter of Law and Fact..................................................................64 1. 2. The Extrinsic Evidence Cannot Entitle IBM to Summary Judgment on the Meaning of Ambiguous Agreements ..............................................65 Numerous Material Questions of Fact Exist Concerning the Intent of the Parties to the UNIX Software agreements.......................................69 a. b. c. The software agreements protect against more than just literal copying of UNIX source code. ............................................72 AT&T did not intend to forfeit its contractual protections for UNIX methods and concepts....................................................79 The IBM Side Letter concerns ownership of modifications and derivatives only, and does not even purport to affect the use and disclosure restrictions in the software agreements. ....................................................................................85

ii

d.

Neither the IBM Side Letter nor later versions of the UNIX software agreements can or did alter the terms of Sequent’s software agreement ........................................................................89 The $ echo newsletter was not intended to have any effect on licensees’ rights or obligations under the UNIX licenses.........92

e. C. II.

IBM’s Restrictive Interpretation of the Software Agreements Is Commercially Unreasonable..................................................................................94

IBM’S “WAIVER” ARGUMENTS ARE MERITLESS ..................................................98 A. Novell Does Not Have Any Authority to Waive SCO’s Intellectual Property Protections Under UNIX Software Agreements .....................................98 1. 2. 3. B. IBM’s Interpretation of APA Section 4.16(b) Is Inconsistent with the Language, Purpose, and Intent of the APA..........................................99 IBM’s Out-of-Context Interpretation of the APA Is Commercially Unreasonable and Produces Absurd Results............................................105 The Extrinsic Evidence Available to Date Refutes IBM’s Interpretation of the Parties’ Intent ..........................................................108

SCO Has Not Waived Its Rights Under the Software agreements ......................110 1. 2. SCO Has Not “Waived” Its Contract Claims by Distributing Limited Linux Products ...........................................................................112 SCO Has Not Made “Publicly Available” Source Code That IBM Misappropriated .......................................................................................115

III.

THE COURT MAY ALSO DENY, OR CONTINUE, IBM’S MOTION UNDER FEDERAL RULE OF CIVIL PROCEDURE 56(f).........................................................116 A. The Outstanding Discovery That SCO Will Develop Bears Directly on IBM’s Contract-Interpretation Argument ............................................................117 1. 2. 3. 4. B. Depositions of IBM Declarants ...............................................................118 Additional Declarations IBM Has Refused to Produce ...........................118 Additional License-Related Documents ..................................................120 Programming-History Evidence ..............................................................120

SCO Is Also Entitled to Additional Discovery Concerning IBM’s “Waiver” Arguments ...........................................................................................122

iii

CONCLUSION............................................................................................................................123

iv

TABLE OF AUTHORITIES CASES 67 Wall St. Co. v. Franklin National Bank, 37 N.Y.2d 245 (1975) ........................................................................................................69 Alexander & Alexander Services, Inc. v. These Certain Underwriters at Lloyd's, London, 136 F.3d 82 (2d Cir. 1998).................................................................................................66 Alfin, Inc. v. Pac. Insurance Co., 735 F. Supp. 115 (S.D.N.Y. 1990).....................................................................................69 America Fidelity & Casualty Co. v. London & Edinburgh Insurance Co., 354 F.2d 214 (4th Cir. 1965) .............................................................................................67 America Maplan Corp. v. Heilmayr, 165 F. Supp. 2d 1247 (D. Kan. 2001)..............................................................................117 Atari, Inc. v. N. America Philips Consumer Electrics Corp., 672 F.2d 607 (7th Cir. 1982) .............................................................................................96 Bay Cities Paving & Grading, Inc. v. Lawyers' Mutual Insurance Co., 5 Cal. 4th 854 (Cal. Ct. App. 1993) .................................................................................108 Bennett v. U.S. Lines, Inc., 64 F.3d 62 (2d Cir. 1995) ................................................................................................114 Blackhawk-Central City Sanitation District v. America Guaranty & Liability Insurance Co., 214 F.3d 1183 (10th Cir. 2000) .........................................................................................48 Blumenfeld v. R. H. Macy & Co., 92 Cal. App. 3d 38 (Cal. App. Ct. 1979) .................................................................109, 110 Boston Concessions Group v. Criterion Ctr. Corp., 606 N.Y.S.2d 696 (App. Div. 1994) ..................................................................................98 Bouzo v. Citibank, 96 F.3d 51 (2d Cir. 1996) ..................................................................................................66 Boyer v. Board of County Comm'rs of County of Johnson County, 922 F. Supp. 476 (D. Kan. 1996).......................................................................................49

v

Brightway Adolescent Hospital v. Health Plan of Nev., Civ. No. 2:98CV0729C, 2000 WL. 3371084. S (D. Utah Sept. 20, 2000)............49, 50, 51 Bur Rigid Box, Inc. v. Travelers Prop. Casualty Corp., 302 F.3d 83 (2d Cir. 2002)...............................................................................................114 In re Caldor, Inc., 217 B.R. 121 (Bankr. S.D.N.Y. 1998)...............................................................................98 Cameron v. Frances Slocum Bank & Trust Co., 824 F.2d 570 (7th Cir. 1987) .............................................................................................70 Celotex Corp. v. Catrett, 477 U.S. 317 (1986)...................................................................................................49, 116 Coleco Industrial, Inc. v. Universal Studios, Inc., 110 F.R.D. 688 (S.D.N.Y. 1986) .....................................................................................119 Compagnie Financiere de CIC et de L'Union Europeenne v. Merrill Lynch, 232 F.3d 153 (2d Cir. 2000)...............................................................................................68 Cooper v. Mart Assocs., 225 Cal. App. 2d 108 (Cal. Ct. App. 1964) .....................................................................103 Coplay Cement Co., Inc. v. Willis & Paul Group, 983 F.2d 1435 (7th Cir. 1993) .....................................................................................65, 67 County of Marin v. Assessment Appeals Board, 134 Cal. Rptr. 349 (Cal. Ct. App. 1976) ..................................................................102, 103 Creative Computing v. GetLoaded.com LLC, 386 F.3d 930 (9th Cit. 2004)............................................................................................116 David v. City & County of Denver, 101 F.3d 1344 (10th Cir. 1996) .........................................................................................49 Davison v. Kless, 280 N.Y. 252 (1939) ........................................................................................................111 Diamond v. Mohawk Rubber Co., 33 F.R.D. 264 (D. Colo. 1963) ................................................................................119, 122 Dunlop-McCullen v. Pascerella, No. 97 Civ. 0195, 2002 WL. 31521012 (S.D.N.Y. Nov. 13, 2002) ................................114

vi

EOTT Energy Corp. v. Storebrand International Insurance Co., A/S, 45 Cal. App. 4th 565 (Cal. Ct. App. 1996) ......................................................................108 ESS & Vee Acoustical & Lathing Contractors, Inc. v. Prato Verde, Inc., 702 N.Y.S.2d 38 (App. Div. 2000) ..................................................................................114 Emerson Radio Corp. v. Orion Sales, Inc., 253 F.3d 159 (3d Cir. 2001)...............................................................................................67 EPN Ingenieria S.A. de C.V. v. General Electric Co., No. 92 Civ. 1563, 1996 WL 531867 (S.D.N.Y. Sept. 19, 1996) .......................................71 Farmland Industrial, Inc. v. Grain Board of Iraq, 904 F.2d 732 (D.C. Cir. 1990) ...........................................................................................67 Frontier Refining, Inc. v. Gorman-Rupp Co., 136 F.3d 695 (10th Cir. 1998) .........................................................................................119 G-I Holdings v. Baron & Budd, 213 F.R.D. 146 (S.D.N.Y. 2003) .....................................................................................118 Gamble, Simmons & Co. v. Kerr-McGee Corp., 175 F.3d 762 (10th Cir. 1999) ...........................................................................................64 Garza v. Maritime Trans. Lines, Inc., 861 F.2d 23 (2d Cir. 1988).................................................................................................56 General Motors Acceptance Corp. v. Clifton-Fine Central Sch. District, 85 N.Y.2d 232 (1995) ......................................................................................................111 Gilbert Frank Corp. v. Federal Insurance Co., 70 N.Y.2d 966 (1988) ......................................................................................114, 115, 116 Gomez v. America Electric Power Serv. Corp., 726 F.2d 649 (10th Cir. 1984) ...........................................................................................66 Goodman v. Resolution Trust Corp., 7 F.3d 1123 (4th Cir. 1993) ...............................................................................................67 Harm v. Frasher, 181 Cal. App. 2d 405 (Cal. Ct. App. 1960) .............................................................104, 107 Hartford Acc. & Indemnity Co. v. Wesolowski, 350 N.Y.S.2d 895 (1973)...................................................................................................66

vii

Heidi E. v. Wanda W., 620 N.Y.S.2d 665 (N.Y. App. Div. 1994) .......................................................................111 Heidlebaugh v. Miller, 271 P.2d 557 (Cal. Ct. App. 1954) ..................................................................................103 Hertzog. Calamari & Gleason v. Prudential Insurance Co. of America, 933 F. Supp. 254 (S.D.N.Y. 1996).....................................................................................69 Heston v. Farmers Insurance Group, 160 Cal. App. 3d 402 (Cal. Ct. App. 1984) .............................................................104, 105 Hicks v. City of Watonga, 942 F.2d 737 (10th Cir. 1991) ...........................................................................................48 Holt v. Wesley Medical Ctr., LLC, No. 00-1318-JAR, 2002 WL. 31778785 (D. Kan. Nov. 25, 2002)..................................117 HotSamba, Inc. v. Caterpillar Inc., No. 01 C 5540, 2004 WL 609797, at *5 (N.D. Ill. Mar. 25, 2004) ...................................62 Howe v. America Baptist Homes of the W., Inc., , 112 Cal. App. 3d 622 (Cal. Ct. App. 1980) .....................................................................105 I.M.S. Inquiry Management System, Ltd. v. Berkshire Information System, Inc., 307 F. Supp. 2d 521 (S.D.N.Y. 2004)..............................................................................116 Jacobson v. Sassower, 66 N.Y.2d 991 (1985) ........................................................................................................69 Jellinick v. Joseph J. Naples & Associates, Inc., 296 A.D.2d 75 (N.Y. App. Div. 2002) ..............................................................................52 K. Bell & Associates, Inc. v. Lloyd's Underwriters, 827 F. Supp. 985 (S.D.N.Y. 1993)...................................................................................112 Kaiser-Francis Oil Co. v. Producer's Gas Co., 870 F.2d 563 (10th Cir. 1989) .....................................................................................67, 68 King Airway Co. v. Routt County, No. 97 CJ C.A.R. 563, 1997 WL 186256, at *5 (10th Cir. Apr. 17, 1997) .....................117 Leibowitz v. Elsevier Sci., Ltd., 927 F. Supp. 688 (S.D.N.Y. 1996)...................................................................................112

viii

Leo F. Piazza Paving Co. v. Foundation Constructors, Inc., 128 Cal. App. 3d 583 (Cal. Ct. App. 1981) .....................................................................102 Liberty Mutual Insurance Co. v. Pine Bluff Sand & Gravel Co., 89 F.3d 243 (5th Cir. 1996) ...............................................................................................67 Madison-Oneida-Herkimer Consortium v. N. America Administrations, 765 N.Y.S.2d 184 (Sup. Ct. 2003)...................................................................................112 McDonnell Douglas Finance Corp. v. Pa. Power & Light Co., 858 F.2d 825 (2d Cit. 1988)...............................................................................................71 Merrill Lynch v. Adler, 6.51, N.Y.S.2d 38 (App. Div. 1996) ...........................................................................................55 Messina v. Mazzeo, 854 F. Supp. 116 (E.D.N.Y. 1994) ..................................................................................118 Met. Paving Co. v. City of Aurora, Colo., 449 F.2d 177 (10th Cir. 1971) ...........................................................................................66 Miller v. Wolpoff & Abramson, L.L.P., 321 F.3d 292 (2d Cir. 2003)...............................................................................................70 Moncrief v. Williston Basin Interstate Pipeline Co., 174 F.3d 1150 (10th Cir. 1999) ...................................................................................67, 68 Morrison,.340, F.2d at 432 (f) ..................................................................................................................117 Morrison Flying Serv. v. Deming National Bank, 340 F.2d 430 (10th Cir. 1965) ...........................................................................................49 National Envmtl. Serv. Co. v. Ronan Engineering Co., 256 F.3d 995 (10th Cir. 2001) ...........................................................................................71 Navillus Tile, Inc. v. Turner Construction Co., 770 N.Y.S.2d 3 (App. Div. 2003) ....................................................................................114 Nowak v. Ironworkers Local 6 Pension Fund, 81 F.3d 1182 (2d Cir. 1996)...............................................................................................65 Omni Quartz, Ltd. v. CVS Corp., 287 F.3d 61 (2d Cir. 2002).................................................................................................64

ix

Pac. Gas & Electric Co. v. G.W. Thomas Drayage & Rigging Co., 69 Cal. 2d 33 (1968) ........................................................................................................108 Price v. W. Res., Inc., 3232 F.3d 779 (10th Cir. 2000) .......................................................................................117 Procter & Gamble Co. v. Haugen, 179 F.R.D. 622 (D. Utah 1998) .......................................................................................117 Quark, Inc. v. Harley, 141 F.3d 1185, 1998 WL. 161035 (10th Cir. Mar. 4, 1998)............................................119 RMLS Metals, Inc. v. International Bus. Machine Corp., 874 F. Supp. 74 (S.D.N.Y. 1995).......................................................................................71 R/S Associates v. N.Y. Job Dev't Authority, 98 N.Y.2d 29 (2002) ..........................................................................................................64 Ramming v. Barnard, No. E030334,2002 WL. 393118 (Cal. Ct. App. Mar. 13, 2002) .....................................108 Readco, Inc. v. Marine Midland Bank, 81 F.3d 295 (2d Cir. 1996)...............................................................................................112 Reape v. N.Y. News, Inc., 122 A.D.2d 29 (N.Y. App. Div. 1986) ..............................................................................94 Robshaw v. Health Management, Inc., 470 N.Y.S.2d 226 (App. Div. 1983) ..................................................................................56 Rogers v. United States, 281 F.3d 1108 (10th Cir. 2002) .........................................................................................48 Ronnen v. Ajax Electric Motor Corp., 88 N.Y.2d 582 (1996) ........................................................................................................55 Ruttenberg v. Data Davidge System Corp., 626 N.Y.S.2d 174 (N.Y. App. Div. 1995) ...................................................................56, 66 SAS Institute, Inc. v. S&H Computer Svs., Inc., 605 F. Supp. 816 (M.D. Tenn. 1985).................................................................................96 Saint-Gobain/Norton Industrial Ceramics Corp. v. General Electric Co., 884 F. Supp. 31 (D. Mass. 1995) .....................................................................................119

x

Schering Corp. v. Home Insurance Co., 712 F.2d 4 (2d Cir. 1983) ............................................................................................69, 71 Securities Industrial Automation Corp. v. United Computer Capital Corp., 723 N.Y.S.2d 668 (N.Y. App. Div. 2001) .......................................................................111 Seiden Associates, Inc. v. ANC Holdings, Inc., 959 F.2d 425 (2d Cir. 1992)...............................................................................................65 So. Cal. Edison v. Super. Ct., 37 Cal. App. 4th 839 (Cal. Ct. App. 1995) ......................................................................109 State v. Home Indemnity, 66 N.Y.2d 669 (1985) ........................................................................................................65 T.W.A. Trading, Inc. v. Gold Coast Freightways, Inc., No. 2001-900 KC, 2002 WL 1311648 (N.Y. App. Term. Apr. 2, 2002) ........................111 Theofel v. Farey-Jones, 359 F.3d 1066 (9th Cir. 2004) .........................................................................................116 Tougher Heating & Plumbing Co. v. State, 73 A.D.2d 732 (N.Y. App. Div. 1979) ..............................................................................94 Trainor v. Apollo Metal Specialities, Inc., 318 F.3d 976 (10th Cir. 2002) ...........................................................................................52 United Rentals (No. America), Inc. v. Keizer, 355 F.3d 399 (6th Cir. 2004) .............................................................................................67 Viacao Aerea Sao Paulo, S.A. v. International Lease Finance Corp., 859 F.2d 924, 1988 WL. 103286 (9th Cir. 1988) ............................................................102 Wards, Inc. v. Stamford Ridgeway Associates, 761 F.2d 117 (2d Cir. 1985)...............................................................................................52 Westchester Resco Co., L.P. v. New England Reinsurance Corp., 818 F.2d 2 (2d Cir. 1987) ..................................................................................................69 Whelan Associates, Inc. v. Jaslow Dental Laboratoriess, Inc., 609 F. Supp. 1307 (E.D. Pa. 1985) ....................................................................................96 Wolf v. Suver. Ct., 114 Cal. App. 4th 1343 (Cal. Ct. App. 2004) ..................................................................108

xi

World Trade Ctr. Properties v. Hartford Fire Insurance Co., 345 F.3d 154 (2d Cir. 2003).............................................................................................108 STATUTES Fed. R. Civ. P. 56(c) ......................................................................................................................48 Fed. R. Civ. P. 56(f).....................................................................................................................117 MISCELLANEOUS E. Allan Farnsworth, Contracts § 7.14 (2d ed. 1990) ....................................................................67 Wright, Miller & Kane, Federal Practice and Procedure § 2730.1 (2d ed. 1983) .........................67 Wright, Miller & Kane, Federal Practice and Procedure § 2726 (2d ed. 1983) ............................70 IBM Dictionary of Computing, 18 (George McDaniel ed. 1994) ...........................................47, 51 Restatement (Second) of Contracts § 212(2) (1981) .....................................................................67

xii

INTRODUCTION This case arises out of International Business Machines Corporation’s (“IBM”) flagrant disregard for the contractual limitations to which it agreed when it licensed from The SCO Group’s (“SCO”) predecessor-in-interest AT&T one of the most valuable computer software products ever developed, the UNIX operating system. The contracts to which IBM and Sequent (which IBM now owns) agreed required, among other things, that AT&T’s licensees use UNIX only for their “own internal business purposes” and treat any “derivative works” they prepared on the basis of the UNIX software product in precisely the same manner as the licensed UNIX software itself. IBM has disregarded these plain contractual protections by essentially giving away to the world, for its own business purposes, products derived from UNIX - as open-source Linux code. While IBM denies liability in this case, internal IBM and Sequent documents repeatedly admit that AIX and Dynix, the products from which IBM contributed substantial code to Linux, are UNIX “derivatives.” IBM seeks to avoid accountability for its actions before a jury by a summary judgment motion that defies the plain language of the operative agreements, rests on extrinsic evidence from witnesses who have given sworn testimony directly contrary to IBM’s argument, and contends that SCO’s valuable UNIX rights have been waived for no consideration whatsoever. None of IBM’s arguments makes sense, none can survive scrutiny, and certainly none can support an entry of summary judgment. BACKGROUND In the 1980s, AT&T owned a seminal innovation - the UNIX operating system - that was heavily demanded in the computer industry for its unique and powerful capabilities. To satisfy

that demand, AT&T agreed to license UNIX to hundreds of the world’s most sophisticated technology companies, large and small, including IBM and Sequent. Under the UNIX software license agreements, AT&T gave its licensees special access to the UNIX software and even permitted them to rely on the innovations in UNIX to develop modifications and derivatives of the UNIX product. Under separate sublicensing agreements, AT&T allowed its licensees to sell derivative operating systems based on UNIX to end users, but permitted such products to be distributed only in object code format (which is not readable by human beings). AT&T understood that the special access it gave licensees to the UNIX software required strict protections for the valuable UNIX intellectual property. If the proprietary innovations in UNIX were not carefully safeguarded, then the value of the UNIX asset would be quickly eroded, and AT&T’s business based on that innovation (including through licensing and sublicensing arrangements) would be eviscerated. Although the law already provided AT&T with certain intellectual-property protections - including against literal and non-literal copyright infringement and for the trade secrets embodied in UNIX - AT&T sought to ensure that UNIX would be afforded substantial additional protections against uncompensated misappropriation and uncontrolled disclosure. Accordingly, AT&T’s legal department created a UNIX software license agreement that imposed strict requirements on licensees’ use, export, transfer, and disclosure of the UNIX software. Id. The software agreements extended those protections not only to the literal source code in which the UNIX innovations had been originally expressed, but also to the ideas, structures, sequences, organizations, methods, and concepts embodied in UNIX. AT&T insisted

2

that all of its licensees agree to core intellectual-property protections for the UNIX software, and all of them, including IBM and Sequent, agreed. The cornerstone of AT&T’s software license protections was Section 2.01, which (1) strictly limited licensees’ use of the UNIX “SOFTWARE PRODUCT” to their “internal business purposes” on designated computers and (2) permitted licensees to develop modifications and derivative works based on the original licensed product - “provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT.” In other words, if a licensee exercised its right under its software agreement to rely on the original UNIX product in preparing modifications or derivatives based on that software, then the licensee had to afford such “resulting materials” the same strict protections required by the software agreement for the original UNIX product itself. Thus, pursuant to Section 2.01, all of the use, transfer, export, and confidentiality restrictions that applied to the original UNIX product (and, of course, the literal source code in which it was written) also covered any such modifications or derivative works. For many years, IBM and Sequent relied on the access they had been given under their software agreements to develop products based on UNIX, which they named AIX and Dynix, respectively. Indeed, IBM and Sequent have each acknowledged, repeatedly, that AIX and Dynix were derived from the UNIX operating system. Nevertheless, in the late 1990s, after purchasing Sequent and inheriting its licensing rights and obligations, IBM shifted its business focus from UNIX to Linux (an “open source” operating system, in which the underlying source code is openly available to anyone). And, in order to accelerate the commercial appeal of Linux to businesses, IBM, in violation of its contractual obligations, began dumping large portions of code from AIX and Dynix into Linux.

3

IBM now attempts to evade responsibility for its conduct by imposing a new, and indefensible, gloss on the UNIX software license agreements. Specifically, IBM claims that it is entitled to summary judgment on SCO’s breach-of-contract claims because the IBM and Sequent software agreements protected the licensor only against the literal copying of UNIX source code. See IBM Mem. at 47-71. Even though IBM attempts repeatedly to characterize its position as protecting only its “own homegrown code,” in reality, IBM’s position on summary judgment is that it (and any other UNIX licensee) was free to do anything it wanted with any modification or derivative of the UNIX software (including copying, distributing, and even open-sourcing the product in source-code format) to the extent that such modification or derivative did not include literally copied UNIX source code. See Part I, below. SUMMARY OF ARGUMENT IBM’s argument cannot be reconciled with the plain language of the agreements, is disputed by ample extrinsic evidence that has been developed thus far in discovery, and defies commercial reason in light of the circumstances surrounding the execution of the software agreements. First, IBM’s “literal source code only” reading is indefensible as a matter of plain language interpretation of the software agreements. Those agreements clearly restrict use of the software product to IBM’s “own internal business purposes” and require, among other things, that “derivative works” based on the licensed UNIX software be subject to the same restrictions as the UNIX code itself. IBM’s position cannot be reconciled with the language of the software agreements. For example:
•

Because the literal UNIX source code is already plainly included in “SOFTWARE PRODUCT,” IBM’s constricted reading of Section 2.01 would effectively read out of the agreement the critical final clause of that provision i.e.,

4

“provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT’).
•

IBM’s contract-interpretation argument ignores entirely the language in Section 7.06(a) of the Sequent agreement that expressly extended the strict contractual non-disclosure requirements for the licensed UNIX product to “methods or concepts” embodied in such “SOFTWARE PRODUCTS.” In suggesting that AT&T somehow “clarified” these protections out of existence, IBM attempts to rely on language in an IBM side letter agreement (and in subsequently executed agreements with third parties) that expressly concerned only the “ownership” of modifications and derivatives and plainly did not in any way change the express use and disclosure requirements set forth in all of AT&T’s software agreements. And, in attempting to alter the terms of the IBM and Sequent software agreements at issue in this case based on contractual terms and other matters beyond the scope of the agreements themselves, IBM improperly disregards the merger clauses in each of those software agreements, which expressly provided that each licensee’s obligations would be governed exclusively by the terms of its software agreement with AT&T.

•

•

Thus, far from demonstrating IBM’s entitlement to summary judgment on SCO’s contract claims, the plain language of the operative software agreements precludes IBM’s contractinterpretation argument as a matter of law. See Part LA, below. Second, IBM purports to support its tortured reading of the contracts with witness declarations that it claims are so “one-sided” as to warrant judgment for it as a matter of law. See IBM Mem. at 66. But even accepting IBM’s disregard for the plain language of the software agreements, the Court may not resort to extrinsic evidence to resolve contractual ambiguities on summary judgment. Under well-settled New York and Tenth Circuit law, contractual ambiguities alone give rise to material questions of fact because they necessarily implicate the functions reserved for the fact finder, including weighing evidence, drawing inferences, and determining witness credibility. See Part LB-1, below. In any event, however, the extrinsic evidence on which IBM purports to rely is far from undisputed. Indeed, ample record evidence confirms the plain reading of the software

5

agreements and directly undermines IBM’s declarations on each and every material point for which IBM relies on them. The extrinsic evidence documented herein demonstrates, among other things, that:
•

the protections of the software agreements are not limited only to the original UNIX source code; AT&T never forfeited its contractual protections for methods and concepts; the “ownership” language that AT&T agreed to include in IBM’s side letter agreement was never intended to compromise AT&T’s protections for its intellectual property in UNIX and UNIX modifications/derivatives-, and the merger clauses in the software agreements were specifically designed to ensure that each licensee’s rights would be determined by that licensee’s individual written agreement, and not by the written agreements of any other licensee or by statements made by any party beyond the four corners of the written agreements.

• •

•

Moreover, the extrinsic evidence contradicting IBM’s position comes from a wide array of sources, including many of the same declarants on whom IBM purports to rely for its motion. As detailed below, IBM’s contract motion is directly contradicted by:
•

a declaration given by one of the very same witnesses (former AT&T account executive Ira Kistenberg) on whose prior declaration IBM attempts to rely for its motion; inconsistent deposition testimony given by the four IBM declarants whom SCO has thus far deposed; inconsistent prior testimony that two of IBM’s key declarants (former UNIX licensing supervisors Otis Wilson and David Frasure) gave concerning substantially similar provisions in another AT&T software agreement, approximately twelve years closer to the events at issue here and before IBM secured their legal representation in this case; contemporaneous documents written by IBM’s declarants, particularly including Otis Wilson, concerning AT&T’s intellectual-property protections under the software license agreements; declarations from two of the AT&T licensing attorneys (Martin Pfeffer and Burton Levine) who were primarily responsible for overseeing the drafting and enforcement of the contractual protections for the UNIX intellectual property;

•

•

•

•

6

•

declarations from two former AT&T contract managers (Mitzi Bond and Evelyn Davis) who worked in AT&T’s UNIX licensing group, handled numerous AT&T accounts, and, based on the training they received in the licensing group, routinely explained the core intellectual-property protections in AT&T’s software agreements to customers and prospective customers; and course-of-dealing evidence revealing that IBM and Sequent each treated their AIX and Dynix products as derivative works based on UNIX and, indeed, paid royalties to AT&T on that basis.

•

Thus, the extrinsic evidence in the record precludes summary judgment for IBM not only because it contributes numerous genuine issues of material fact concerning IBM’s proposed contract interpretation, but also because it raises serious credibility issues. See Part LB.2, below. Third, IBM’s “literal source code only” claim should be rejected for the independent reason that it defies commercial reason and, therefore, cannot accord with the reasonable expectation of the parties in entering into the software agreements. Aside from the textual and evidentiary problems posed by IBM’s motion, its contract-interpretation claim would require the Court to conclude that:
•

AT&T devoted substantial resources to drafting, requiring all of its licensees to sign, and enforcing a software agreement that gave AT&T easily evadable protections that were substantially more limited than the protections AT&T would have had without any such agreement; AT&T’s software agreements thus gave AT&T fewer protections against misappropriation and uncontrolled disclosure by the sophisticated computer companies to whom AT&T provided special access to the entirety of its UNIX intellectual property than AT&T had even against third parties who had no such access; and AT&T thereby knowingly jeopardized the value of its UNIX asset and compromised its entire UNIX business even though AT&T was the exclusive owner of a seminal and unique innovation in UNIX for which there was high demand by at least hundreds of UNIX licensees who needed access to compete in the marketplace.

•

•

See Part I.C, below.

7

Beyond its misconceived contract-interpretation argument, IBM also seeks to preclude further scrutiny of its Linux-related conduct through two misguided “waiver” claims. See IBM Mem. at 71-76. These claims are meritless, and certainly cannot justify a conclusion that SCO knowingly and intentionally waived IBM’s contractual breaches as a matter of law. IBM first asserts that Novell waived SCO’s rights for it (against SCO’s will) under the UNIX software agreements when, in late 2003 and early 2004, after becoming IBM’s close ally in the Linux business, Novell purported to excuse IBM’s breaches of those agreements. IBM bases this argument on the untenable premise that even though Novell expressly transferred the entire UNIX business to SCO through a 199.5 Asset Purchase Agreement (“APA”), that same agreement gave Novell the unilateral, unfettered right to require SCO to change or to waive, at any time and for any reason, any of SCO’s rights (including SCO’s right to enforce its intellectual-property protections) under any software agreement for UNIX System Five. IBM’s argument should be rejected on several grounds, including:
•

IBM’s argument is precluded by the plain language of the APA. Premised entirely on a single sentence in a provision of the APA that was expressly designed to address the System Five binary royalty rights that Novell retained as part of the APA purchase price, IBM’s argument misinterprets “SVRX Licenses” in that sentence to cover broadly the software agreements containing the intellectual-property protections that Novell has purported to waive. Although “SVRX Licenses” is not defined, the APA contradicts IBM’s argument by expressly distinguishing between SVRX Licenses and such software agreements. The more reasonable reading of the term “SVRX Licenses” - to cover the product Schedules in which the binary royalty fees were enumerated - accords with the APA’s text and intent of permitting Novell to retain control over the management of its contractual right to the future System Five binary royalties. IBM’s interpretation would undermine the entire purpose of the APA and contradict many of its other key terms. If Novell had a veto right over the intellectual-property protections in the System Five software agreements, then Novell could destroy the entire value of the transfer to SCO of the UNIX assets, including by, among other things, requiring SCO to open source the entire UNIX operating system. Among other critical provisions, the APA’s provisions granting SCO full ownership rights in the UNIX intellectual property would be rendered meaningless. And Novell would even have the right to eliminate SCO’s
8

•

protections under the source-code license that SCO issued to Novell contemporaneously with the APA’s closing.
•

IBM’s argument completely ignores Amendment No. 2 to the APA, executed a year later, which makes perfectly clear that Novell had no “right to increase any SVRX licensee’s rights to SVRX source code” or to “prevent SCO from exercising its rights with respect to SVRX source code in accordance with the Agreement.” IBM’s argument is directly contradicted by sworn declarations from the principal negotiators on both sides of the NoveIVSCO UNIX asset transfer. This evidence verifies that the APA was never intended to give Novell any continuing right to compromise any of SCO’s intellectual-property protections under any UNIX licenses and, indeed, that Amendment No. 2 to the APA was intended to confirm that point.

•

See Part ILA, below. IBM’s second “waiver” claim - based on SCO’s purported implied waiver of its rights fairs no better than its first. See Part 11.13, below. Contrary to IBM’s suggestion, SCO has diligently and consistently protected its UNIX rights. Indeed, promptly after SCO learned (in early 2003) of IBM’s breach of its contractual obligations, SCO filed this lawsuit and terminated the IBM and Sequent software agreements. Having devoted substantial resources to developing its Linux-based products and facing continuing contractual obligations under existing distribution agreements, SCO shortly thereafter discontinued any further Linux-related activities, suffering a 5-10% loss of its revenues. Because SCO could not possibly have waived any of its rights before it knew of IBM’s breach, IBM’s “waiver” argument reduces to the claim that when SCO discovered IBM’s improper dumping of contractually protected materials into Linux, SCO was required either (1) to terminate its entire Linux business immediately, thereby forfeiting its entire investment in Linux and inflicting the maximum financial damage on itself and its customers; or (2) to release forever IBM from liability for the wrongful conduct that put SCO in that position in the first place. IBM’s audacious argument, unsupported by any authority,

9

certainly cannot support summary judgment against SCO under all the circumstances of this case. Finally, while numerous genuine issues of fact already exist to warrant the denial of IBM’s motion, the Court also has discretion under Rule 56(f) to continue or deny that motion based on its prematurity. IBM’s contract motion is one of three summary judgment motions that IBM has filed in this case, seeking summary adjudication of all the key issues while IBM continues to withhold the critical predicate discovery that SCO needs to develop the proof of its claims. See Part III, below. The ample textual and extrinsic evidence detailed below, while sufficient to defeat IBM’s motion at this juncture, also indicates a strong likelihood that SCO will develop through additional discovery further evidence to refute IBM’s motion. SCO’S RESPONSE TO IBM’S “STATEMENT OF UNDISPUTED FACTS” AND STATEMENT OF ADDITIONAL MATERIAL FACTS 1 A. UNIX Was a Seminal Software Innovation That AT&T Diligently Protected Through Its Software License Agreements 1. UNIX occupied an enviable market position when AT&T licensed System Five to

IBM, because UNIX was a seminal, highly demanded innovation in the computer industry. See e.g., Davis Decl. (11/4/04) ¶ 5 (Exh 4); Green Dep. (11/15/04) at 26-28 (Exh. 5). Licensees sought to license UNIX to copy, emulate, and build on the contents of its intellectual property so that licensees could “avoid having to spend substantial time and money developing their own software products from scratch.” Davis Decl. (11/4/04) ¶ 5 (Exh. 4) 2. In response to this demand, AT&T permitted licensees “special access” to the

UNIX source code and all of the intellectual property embodied in that code, and AT&T even

1

In addition to presenting additional material facts not addressed in IBM’s motion, SCO herein responds to and disputes IBM’s “Statement of Undisputed Facts,” paragraphs 64, 89, 103, 115, 122, 145, 153, 160, and 163 herein identify by paragraph number IBM’s specific assertions that SCO disputes.

10

permitted its licensees to rely on the UNIX software to develop products based on UNIX, including derivatives, modifications, “substantially similar” copies, and even literal copies of the UNIX code. Green Dep. (11/15/04) at 31, 48-49, 132, 135 (Exh. 5). 3. AT&T appreciated the risk inherent in letting “that kind of development process

go on with the valuable UNIX source code,” particularly behind licensees’ closed doors. Id. at 49. As a result, AT&T and its attorneys devoted substantial efforts to developing software license agreements that would protect against the misappropriation and disclosure of the UNIX intellectual property. AT&T diligently sought to prevent the intellectual property embodied in UNIX from reaching the open market as a royalty-free competitor to UNIX, thereby jeopardizing UNIX’s value as an exploitable asset. See id. at 37, 47-49. 4. AT&T required software license agreements and sublicensing agreements for any

licensee who wished to sublicense UNIX-based products. Through those agreements, AT&T strictly prohibited any licensee from distributing or otherwise disclosing any UNIX modifications or derivatives in anything but object code format - the binary format that is not readable by human beings. See, e.g., IBM Agreement § 1.04 (2/1/85) (Exh. 2); IBM Sublicensing Agreement (2/1/85) § 1.04 (Exh. 6). 5. AT&T understood that it was essential to safeguard “AT&T’s valuable UNIX

asset by protecting the entire chain of development that resulted from the special access and development rights AT&T afforded its licensees” under their software agreements. Levine Decl. (11/23/04) ¶ 6 (Exh. 7). 6. Accordingly, AT&T’s attorneys developed a license that was “specifically

designed to cast the widest possible net” of protections, “beyond the common law protections that would otherwise have been available (absent the agreement) for AT&T’s trade secrets and

11

‘know-how.”‘ Pfeffer Decl. (9/7/04) ¶ 9 (Exh. 8). 7. And AT&T’s account executives took a hard line in refusing to compromise the

core UNIX intellectual-property protections in AT&T’s license negotiations. See Davis Decl. (11/4104) ¶ 9 (Exh. 4) (explaining that AT&T was “adamant in refusing to compromise its core intellectual property protections” and that “it was not uncommon for Account Executives and Contract Managers to explain to potential licensees that they were not negotiable”). AT&T was vigilant in maintaining the value of its UNIX asset when it licensed its UNIX software. See Davis Decl. (11/4/04) 5 (Exh.. 4). B. The Software Agreements Provide Broad Protections for the UNIX Intellectual Property 8. The UNIX software agreements that AT&T entered into with each of IBM (in

February 1985) and Sequent (in April 1985) contain identical provisions governing the licensee’s rights to use the licensed software product and to prepare modifications and derivative works based on that software product. These provisions are contained in Section 2.01 of each agreement, which states: “AT&T grants to LICENSEE a personal, nontransferable and nonexclusive right to use in the United States each SOFTWARE PRODUCT identified in the one or more Supplements hereto, solely for LICENSEE’S own internal business purposes and solely on or in conjunction with DESIGNATED CPUs for such SOFTWARE PRODUCT. Such right to use includes the right to modify such SOFTWARE PRODUCT and to prepare derivative works based on such SOFTWARE PRODUCT, provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT.” IBM Agreement (2/1/85) § 2.01 (Exh. 2); Sequent Agreement (4/18/85) § 2.01 (Exh. 1). 9. Section 1.04 of the software agreements defines “SOFTWARE PRODUCT” as

“materials such as COMPUTER PROGRAMS, information used or interpreted by COMPUTER PROGRAMS and documentation relating to the use of COMPUTER PROGRAMS.” IBM Agreement (2/1/85) § 1.04 (Exh. 2); Sequent Agreement (4/18/04) § 1.04 (Exh. 1). It is

12

undisputed that “SOFTWARE PRODUCT” includes UNIX System Five source code. See IBM Mem. ¶ 57 10. Section 7.06(a) of the Sequent software agreement states in relevant part:

“LICENSEE agrees that it shall hold all parts of the SOFTWARE PRODUCTS subject to this Agreement in confidence for AT&T. LICENSEE further agrees that it shall not make any disclosure of any or all of such SOFTWARE PRODUCTS (including methods or concepts utilized therein) to anyone, except to employees of LICENSEE to whom such disclosure is necessary to the use for which rights are granted hereunder.” Sequent Agreement (4/18/04) § 7.06(a) (Exh. 1). 11. Section 7.06(b) of the software agreements states in relevant part:

“Notwithstanding the provisions of Section 7.06(a), LICENSEE may distribute copies of a SOFTWARE PRODUCT, either in modified or unmodified form, to third parties having licenses of equivalent scope herewith from AT&T (or a corporate affiliate thereof) for the same SOFTWARE PRODUCT, provided that LICENSEE first verifies the status of any such third party in accordance with specific instructions issued by AT&T.” IBM Agreement (2/1/85) § 7.06(b) (Exh. 2); Sequent Agreement (4/18/85) § 7.06(b) (Exh. 1). 12. Section 7.10 of the software agreements states: “Except as provided in

Section 7.06(b), nothing in this agreement grants to LICENSEE the right to sell, lease or otherwise transfer or dispose of a SOFTWARE PRODUCT in whole or in part.” IBM Agreement (2/1/85) § 7.10 (Exh. 2); Sequent Agreement (4/18/85) § 7.10 (Exh. 1). 13. Each of the software agreements contains a New York choice-of-law provision.

See IBM Agreement (2/1/85) § 7.13 (Exh. 2); Sequent Agreement (4/18/85) § 7.13 (Exh. 1). 14. Each of the software agreements also contains a merger clause:

“This Agreement and its Supplements set forth the entire agreement and understanding between the parties as to the subject matter hereof and merge all prior discussions between them, and neither of the parties shall be bound by any conditions, definitions, warranties, understandings or representations with respect to such subject matter other than as expressly provided herein or as duly set forth on or subsequent to the date of acceptance hereof in writing and signed by a proper and duly authorized representative of the party to be bound thereby. No provision appearing on any form originated by LICENSEE shall be applicable unless such provision is expressly accepted in writing by an authorized
13

representative of AT&T.” IBM Agreement (2/1/85) § 4 (Exh. 2); Sequent Agreement (4/18/85) § 4 (Exh. 1). 15. At the same time that IBM entered into its software agreement with AT&T, and a

couple of months before Sequent executed its software agreement, AT&T and IBM entered into a so-called side letter agreement (the “IBM Side Letter”). Exh. 3. 16. The IBM Side Letter left the language of Section 2.01 of the IBM software

agreement entirely intact, and states as follows: “Regarding Section 2.01, we agree that modifications and derivative works prepared by or for [IBM] are owned by [IBM]. However, ownership of any portion or portions of SOFTWARE PRODUCTS included in any such modification or derivative work remains with [AT&T].” IBM Side Letter (2/l/85) at 2 (Exh. 3). 17. The IBM Side Letter further amended Section 7.06(a) of the IBM software

agreement to read in relevant part: “LICENSEE agrees that it shall hold SOFTWARE PRODUCTS subject to this Agreement in confidence for AT&T. LICENSEE further agrees that it shall not make any disclosure of such SOFTWARE PRODUCTS to anyone, except to employees of LICENSEE to whom such disclosure is necessary to the use for which rights are granted hereunder. LICENSEE shall appropriately notify each employee to whom any such disclosure is made that such disclosure is made in confidence and shall be kept in confidence by such employee. Nothing in this agreement shall prevent LICENSEE from developing or marketing products or services employing ideas, concepts, know-how or techniques relating to data processing embodied in SOFTWARE PRODUCTS subject to this Agreement, provided that LICENSEE shall not copy any code from such SOFTWARE PRODUCTS into any such product or in connection with any such service and employees of LICENSEE shall not refer to the physical documents and materials comprising SOFTWARE PRODUCTS subject to this Agreement when they are developing any such products or services or providing any such service.” IBM Side Letter (2/l/85) at 4 (Exh. 3). C. Extrinsic Evidence Confirms the Broad Protection of the Software Agreements 18. IBM generally alleges that the restrictions of the UNIX software agreements were

so narrow as to cover only literally copied UNIX source code. See IBM Mem. ¶¶ 78-124. Several of IBM’s declarants, however, have given sworn testimony that squarely contradicts the
14

positions for which IBM cites their declarations. Other key witnesses involved with UNIX licensing confirm that, consistent with their plain language, the software agreements were intended to protect modifications of and derivative works based on UNIX as well as the ideas, structures, sequences, organization, methods, and concepts embodied in UNIX. 1. Protection of Modifications and Derivative Works a. 19. Ira Kistenberg

IBM relies on a declaration from Ira Kistenberg, the account executive who

negotiated Sequent’s UNIX license on behalf of AT&T, for its position that only copying of literal UNIX source code is protected by the agreements. See IBM Mem. ¶¶ 80-81, 84-85, 92; see also Kistenberg Decl. (6/24/04) ¶ 22 (Exh. 9) (“Our licensees, including Sequent, were free to use and disclose the modifications or derivative works they created, provided that they did not use and disclose any portion of the licensed UNIX System V source code.”). 20. In his more recent declaration submitted herewith, however, Mr. Kistenberg

clarified that the UNIX software agreements protect much more than literal source code. See Kistenberg Decl. (11/12/04) ¶ 5 (Exh. 10) (“AT&T intended that protected modifications and derivatives would include any product that contained any source code that had been copied verbatim from UNIX System V; any copied source code that was similar in substance to the original source code in UNIX System V; any structures, sequences, patterns, ideas, methods or concepts from UNIX System V; and any source code that the licensee developed with the benefit of exposure to the UNIX System V source code.”). b. 21. Otis Wilson

IBM relies on a declaration from Otis Wilson to support its position that only

copying of literal UNIX source code was protected. See IBM Mem. ¶¶ 79, 81-82, 84-87; see also Wilson Decl. (4/26/04) ¶¶ 12, 14 (Exh. 11).
15

22.

Mr. Wilson gave sworn testimony directly contrary to these statements, however,

when he was deposed in UNIX System Laboratories, Inc. v. Berkeley Software Design, Inc., Civil Action No. 92-1667 (DRD) (D.N.J.) (“BSD”), nearly twelve years closer to the events in question. Mr. Wilson testified that AT&T’s UNIX software agreements protected far more than just “UNIX System V code,” and that he had expressly communicated that view to AT&T’s licensees. See Wilson Dep. (12/10/92) at 128 (Exh. 12). 23. The BSD case involved a software agreement with substantially similar

intellectual-property protections as those found in the IBM and Sequent software agreements. Compare The Regents of the University of California Agreement (11/12/85) §§ 2.01, 7.06 (Exh. 13) with IBM Agreement (2/1/85) §§ 2.01, 7.06 (Exh. 2) and Sequent Agreement (4/18/85) §§ 2.01, 7.06 (Exh. 1); see also Kistenberg Decl. (11/12/04) ¶ 4 (Exh. 10) (explaining that UNIX commercial, educational, and governmental licenses all provided the same core intellectualproperty protections); Green Dep. (11/15/04) at 68, 240 (Exh.5) (same); Pfeffer Decl. (9/7/04) ¶ 5 (Exh. 8) (same); Bond Decl. (11/4/04) ¶ 10 (Exh. 14) (same); Davis Decl. (11/4/04) ¶ 7 (Exh. 4) (same). 24. Mr. Wilson further testified in BSD: “My understanding is that anything created

by the university with exposure to the licensed software, based on, contained, a part of, was a derivative work with regard to these documents and had to be treated as licensed software.” Wilson Dep. (12/10/92) at 51 (Exh. 12). 25. According to Mr. Wilson’s BSD testimony, AT&T’s software agreement covered

a licensee’s product that did not contain any UNIX source code if the licensee developed that product with exposure to the licensed UNIX product. See id., at 127-28. 26. In a letter sent to IBM in 1987, Mr. Wilson wrote: “We have sent a letter to all

16

our licensees who have also acquired sublicensing rights to address an issue that is of great concern to AT&T.” Letter from Otis Wilson to IBM dated July 27, 1987 at I (Exh. 15). He continued: “We want to stress at the outset that licensees may not distribute, or permit access to, any licensed source code software product (either in modified or unmodified form) to any third party at any time, or distribute sublicensed products based on such software products or its derivatives, unless specifically permitted by the license agreements.” Id.; see also Exh. 16 (similar letter from Mr. Wilson to Sequent). 27. In his recent deposition in this case, Mr. Wilson admitted that under Section 2.01

of the UNIX software agreement, modifications and derivative works were included in the definition of the protected software product, and thus had to be treated in the same way as the licensed software. See Wilson Dep. (6/10/04) at 220 (Exh. 17); see also, id. at 340-43. c. 28. David Frasure

IBM further relies on a declaration from David Frasure to support its claim that

the UNIX software agreements protected AT&T from only the literal copying of UNIX source code. IBM Mem. ¶¶ 79, 81, 83-85, 88; see also, Frasure Decl. (3/28/04) ¶¶ 12, 14 (Exh. 18). 29. Like Mr. Wilson, Mr. Frasure took a directly contradictory position in his 1992

BSD testimony. Mr. Frasure acknowledged then that modifications and derivative works were to be treated as licensed software under the software agreement. See Frasure Dep. (1218192) at 146 (Exh. 19). 30. Mr. Frasure described an occasion on which he and Mr. Wilson threatened

litigation to stop a licensee from distributing a derivative product, without regard for whether that derivative product contained any literal UNIX source code. In so doing, Mr. Frasure explained how the fact that the derivative product was based on UNIX sufficed to ensure that it was subject to the protections of the software agreements. See id. at 23; see also id. at 121.

17

31.

Mr. Frasure repeatedly testified to AT&T’s intent to protect the entire universe of

products that could be created as a result of UNIX exposure, regardless of whether the resulting product actually contained any UNIX code. See id. at 20; see also id. at 42, 166-68. 32. Mr. Frasure also testified about a meeting that he had attended with Otis Wilson,

Geoff Green (the sole attorney who has signed a declaration submitted with IBM’s motion), and representatives of one of AT&T’s licensees in late 1984. Id. at 96-97. Mr. Frasure recalled the concept of “mental contamination” that he and Mr. Wilson had used to describe the expansive scope of the protections provided under the software agreements. See id. at 18-19. Mr. Frasure also specifically reiterated that AT&T’s “exposure” concept was intended to cover products even if they did not contain the UNIX source code. See id. 33. Moreover, in his deposition testimony in this case, Mr. Frasure acknowledged that

the language of the software agreements does not support IBM’s position. After recognizing that modifications and derivatives of the original product are clearly protected by the agreements, Mr. Frasure qualified that protection by adding the caveat that such modifications and derivatives had to contain literal UNIX source code to be protected. When questioned about the basis for such a limitation, Mr. Frasure admitted that this position is not supported by the plain language of the software agreements. See Frasure Dep. (6/8/04) at 256-57 (Exh. 20). 34. Mr. Frasure also acknowledged that the creation of a “derivative work” does not

have to involve literal copying: “Derivative work means that it was based on Unix. It was, in other words it was derived from, from Unix. It also could be that it worked with Unix, but it may not have specific Unix code in it. But the software itself that would be written by a licensee could not have been written, if you will, without Unix existing.” Id. at 178 (emphasis added). 35. Further, in a declaration that he gave to IBM in October 2003 - almost six months

18

before to the March declaration on which IBM now relies - Mr. Frasure made similar statements that undermine his current position. In his original declaration, Mr. Frasure recognized that an entire modification or derivative work, and not literal lines of UNIX source code within that work, had to be protected under the software agreements. He explained that “licensees owned their modifications and derivative works and were permitted to use or disclose them as they might choose, so long as any modification or derivative work containing any part of a software product was treated the same as a software product under the license agreements.” Frasure Decl. (10/7/03) ¶ 12 (Exh. 21). d. 36. Geoff Green

IBM also cites the declaration of Geoff Green, who served as an AT&T attorney

in its UNIX licensing department, to support its claim that the UNIX software agreements protect against only the literal copying of UNIX source code. See IBM Mem. ¶¶ 80, 93; see also Green Decl. (6/17/04) ¶ 6 (Exh. 22). 37. During his recent deposition, however, Mr. Green admitted that modifications of

and derivative works based on the UNIX product were protected under the agreements by the same restrictions that governed the original licensed product, regardless of whether or not a licensee’s modification or derivative product contained any UNIX source code. Examining the Sequent software agreement, Mr. Green testified: “Q: And so any restrictions under the software agreement that applied to the original licensed software product, it was AT&T’s intent that the resulting materials would be treated in the same way? Yes. So that applies, then, to any derivatives or modifications that are based on the original software product? Under this provision, yes. Is there any requirement in the license agreement that such modifications or derivatives have to include literally copied source code from the original product? In this form of the agreement, no.” Green Dep. (11/15/04) at 114-15
19

A: Q: A: Q:

A:

(Exh. 5). 38. When asked if there was anywhere in the standard UNIX System Five software

agreement where “the protection for derivatives or modifications” is “limited to those derivatives or modifications that include literally copied source code,” Mr. Green responded: “No there isn’t. Yes, that is correct.” Id. at 117-18. 39. Mr. Green, whose job it was to “change the language to reflect the intent” of

AT&T, id. at 109, stated that “the language of the agreement should speak for itself as to - as to what we did to protect the software.” Id. at 161. 40. Mr. Green noted that “something that was based on the licensed product” or “a

variation of the product” would fall within the definition of a derivative work. Id. at 113. Mr. Green further made clear that literal copying is not required in order for a licensee to create a modification or derivative work subject to the protections of the UNIX software agreements under Section 2.01. See id. at 145. 41. Mr. Green agreed that one of the concerns that prompted the protection of UNIX

modifications and derivatives was AT&T’s concern that “it would be easy for a licensee to copy the intellectual property in UNIX without literally copying the source code.” Id. at 149-50. 42. Mr. Green recalled the concept of “mental contamination” articulated by Otis

Wilson and David Frasure, and explained that the UNIX licensing group, and particularly Messrs. Wilson and Frasure, were interested in protecting products created with the benefit of programmers’ exposure to UNIX, even if such products did not contain any literal UNIX source code. Id. at 82-84, 150-52. 43. Mr. Green stated that, as per the terms of the software agreements in 1985 (when

both the IBM and Sequent agreements were executed), modifications and derivatives that contained the “concepts, the ideas, the structure, the organization, the methods from the original
20

licensed product,” but did not contain the “source code in the original licensed product” would have been “covered by the agreement” and its various restrictions. Id. at 130-31. 44. Mr. Green further agreed that AT&T would “certainly” be interested “in products

that were developed with the benefit of UNIX even if the licensee were to go off and develop that product on its own based on that exposure,” claiming that AT&T “would be concerned about whether any of AT&T’s intellectual property was involved in the result.” Id. at 90-91. 45. Mr. Green was at a total loss to explain how IBM’s view of the contract would

not convert the final clause of Section 2.01 (“provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT’) into mere surplusage. See id. at 113-1.5, 124-25, 247-48. e. 46. David Rodgers

IBM further relies on the declaration of David Rodgers, Sequent’s former Vice

President of Engineering who executed Sequent’s UNIX software agreement, to support its claim that the license agreements restricted only the literal copying of UNIX source code. See IBM Mem. ¶¶ 79, 97-98; see also Rodgers Decl. (11/5/03) ¶ 8 (Exh. 23). 47. In his deposition in this case, however, Mr. Rodgers admitted that the software

agreements protected even modifications comprised of “completely new source code.” Rodgers Dep. (6/10/04) at 31-32 (Exh. S1). 48. Mr. Rodgers acknowledged that products created entirely by Sequent would be

covered by the software agreements if they were based on the UNIX product. Mr. Rodgers testified that only “work which had already been created by Sequent” before licensing UNIX and “work that in the future was created by Sequent, not based upon that source code, remained the property of Sequent.” Id. at 27 (emphasis added). 49. Regarding Dynix, Mr. Rodgers testified that “Dynix/ptx is almost certainly a
21

derivative work of Unix System V.” Id. at 138. 50. Further, Mr. Rodgers could not explain how IBM’s view of the contract would

not render the final clause of Section 2.01 (“provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT’) a nullity. See id. at 101. f. 51. SCO’s Declarants

The declarations of Martin Pfeffer, Burt Levine, Mitzi Bond, and Evelyn Davis

further establish that the UNIX software agreements protect much more than code literally copied from UNIX. 52. At the time of the signing of the IBM and Sequent software agreements, AT&T

General Attorney Martin Pfeffer had primary responsibility for drafting and enforcing the UNIX software agreements. See Pfeffer Decl. (9/7/04) ¶¶ 3-4 (Exh. 8); see also Bond Decl. (11/4/04) ¶ 5 (Exh. 14); Davis Decl. (11/4/04) ¶ 4 (Exh. 4); Frasure Dep. (6/8/04) at 63-64, 171 (Exh. 20). 53. According to Mr. Pfeffer’s sworn declaration, “it was AT&T’s intent to prevent

through its UNIX license agreements the unauthorized use and disclosure of more than just literally copied UNIX source code,” and the software agreements required “licensees to treat all materials resulting from any exercise of their ‘right to modify’ and ‘to prepare derivative works’ as if such materials had been ‘part of the original SOFTWARE PRODUCT.”‘ Pfeffer Decl. (9/7/04) ¶ 10 (Exh. 8). 54. Mr. Pfeffer explained that Section 2.01 “set forth the parties’ intent and agreement

that the ‘SOFTWARE PRODUCT’ licensed and protected under the terms of the license agreements included the full content of all of the ‘resulting materials’ created over time from the licensees’ exercise of their contractual ‘right to modify’ and ‘to prepare derivative works’ based on the original licensed material.” Id. ¶ 6. This protection extended to the entire chain of products derived from the original licensed materials. See id.
22

55.

Mr. Pfeffer did “not recall any instance during my tenure in which either AT&T

or USL agreed (in any license agreement or supplement or modification thereof) to reduce its protection under a UNIX license to prevent the unauthorized use or disclosure of only source code.” Id. ¶ 11. Any such change would have represented “a significant and material change to the standard terms of AT&T’s license agreement” that would have required his, or to a lesser extent Burt Levine’s, approval. Id. 2 Not only did Mr. Pfeffer confirm that no such change was ever made, but he did not even recall any such “proposed” modification to AT&T’s standard license agreement. Id. 56. Burt Levine, another former AT&T attorney, further refuted IBM’s “literal source

code only” position, specifically identifying its fatal textual error: “Section 2.01 certainly did not require that modifications or derivatives of the UNIX product contain any literal copying of UNIX source code in order to be protected. Indeed, there would have been no need for the last clause of Section 2.01 if it were so limited; the literal UNIX source code was already protected by virtue of its inclusion in the term “SOFTWARE PRODUCT.” Levine Decl. (11/23/04) ¶ 6 (Exh. 7). 57. According to Mr. Levine’s testimony, the “license agreements safeguarded

AT&T’s valuable UNIX asset by protecting the entire chain of development that resulted from the special access and development rights AT&T afforded its licensees under those agreements.” Id. 58. The sworn declaration of former AT&T Account Executive Mitzi Bond further

explains that the “strict contractual protections” of the UNIX license agreements “went beyond safeguarding just the source code in the original licensed UNIX product.” Bond Decl. (11/4/04) ¶ 7 (Exh.14). Ms Bond testified that she is “not aware of any instance in which AT&T agreed

2

Burt Levine confirms that he also never approved any such change, which would have been contrary to AT&T’s consistent and firm policy. See Levine Decl. (11/23/04) ¶ 8 (Exh. 7).

23

(in any license agreement or any supplement, modification, or side letter thereto) to reduce AT&T’s protection under a UNIX license so as to protect against the unauthorized use or disclosure of just source code.” Id. ¶ 9. 59. Ms Bond’s declaration cites her prior testimony in the BSD case in explaining

that “the agreement required licensees to ‘treat the derivative or modified work they had created with the same care they would our own software product, meaning adhering to the same terms and conditions that were here.’” Id. ¶ 11 (a) (Exh 14) (quoting Bond Dep (12/9/92) at 134 (Exh 24)). In defining the scope of modifications and derivatives, Ms Bond stated that “you can prepare derivative work and you don’t necessarily have to include code from our software product”; rather, a derivative was protected simply “based on the fact that it was based on our product.” Bond Decl. (11/4/04) ¶ 11 1(e) (Exh. 14) (quoting Bond Dep. (12/9/92) at 220-21 (Exh. 24)). 60. Ms Bond echoed the statements of Messrs Wilson and Frasure regarding exposure

to UNIX code. She testified that “if a licensee is exposed to our code and that licensee develops a modification, a derivative work, an application or any type of software product as a result of seeing our code, then that product should be treated the same as our software product.” Bond Decl. (11/4/04) ¶ 11(c) (Exh. 14) (quoting Bond Dep. (12/9/92) at 224 (Exh. 24)). 61. Ms Bond also stated that her “understanding of the license agreements had been

supported by conversations” she had with other account executives, including Chuck Greene, who had specifically advised her that “if they see our code and develop something, then they need to treat it like ours,” and IBM Declarant Steve Vuksanovich, who “‘was in agreement’ with ‘the fact that the licensee should treat that application the same as our software product.”‘ Bond Decl. (11/4/04) ¶ 11 (f) (Exh. 14) (quoting Bond Dep. (12/9/92) at 228-29, 231-32 (Exh. 24)).

24

62.

Account Executive Evelyn Davis similarly has stated in her sworn declaration that

the “the license agreements effected these protections by restricting, among other things, the use, export, and disclosure of much more than just the literal lines of UNIX source code.” Davis Decl. (11/4/04) ¶ 6 (Exh. 4). According to Ms Davis, “the license agreements covered any works that resulted from our licensees’ exercise of their contractual right to prepare modifications or derivatives of the UNIX product.” Id. ¶ 6(a). 63. Concerning the scope of the protected modifications and derivatives, Ms Davis

also referred to the “exposure” concept, explaining that “the agreements required each licensee to treat any work prepared with the benefit of having been exposed to our product - regardless of how the licensee further developed or changed that product - as if it were part of our original licensed product.” Id. ¶ 6(c). 64. In light of the evidence that products resulting from a UNIX licensee’s exercise of

its right to modify and prepare derivative works based on UNIX were clearly protected by the software agreements, see ¶¶ 10, 19-67, above, paragraphs 55-56, 3 79-88 and 90-101 of IBM’s “Statement of Undisputed Facts” are therefore disputed. 2. Protection of Methods and Concepts a. 65. Otis Wilson

IBM relies Mr. Wilson’s recent declaration in denying that the UNIX software

agreements were intended to protect the methods and concepts embodied in UNIX System Five. See Wilson Decl. (12/11/03) ¶ 14 (Exh. 25). 66. Such testimony is inconsistent with Mr. Wilson’s contemporaneous statements

3

To the extent that IBM contends that the software products identified on the schedules attached to the software agreements were the only products protected by the agreements, ignoring Section 2.01’s independent protection for modifications and derivatives of such products, Paragraph 56 of IBM’s “Statement of Undisputed Facts” is also disputed.

25

and actions. Just before and well after the IBM Side Letter was signed in February 1985, for example, Mr. Wilson (on behalf of AT&T) repeatedly and expressly sought to protect UNIX “methods and concepts” pursuant to AT&T’s software agreements. See Exh. 26 at 2; see also Exhs. 27-29. b. 67. David Frasure

IBM similarly relies on the declaration of David Frasure in attempting to deny

protection for UNIX methods and concepts. See Frasure Decl. (3/28/04) ¶ 15 (Exh. 18). 68. Mr. Frasure flatly contradicted these statements in his 1992 BSD testimony, when

he testified that “the licensed software to me encompasses methods and concepts techniques,” so that “to the extent at some point [a licensee] gets involved with the source code, the I - and its methods and concepts, then I think there - there’s a restriction on its - on its use.” Frasure Dep. (12/8/92) at 45-46 (Exh. 19). 69. Mr. Frasure stressed that there is “more to the agreement “than the protection of

UNIX source code itself, and pointed to “other portions of paragraphs that cover methods and concepts and - and stuff.” Id. at 37. Mr. Frasure characterized these protections as a “key part of the agreement.” Id. at 147. 70. When asked about the types of products covered by AT&T’s software agreement,

Mr. Frasure explained that products containing UNIX methods and concepts were covered. See id. at 148. Mr. Frasure also acknowledged the application of the methods and concepts protection to work created by a licensee’s programmers. See id. at 147-48. 71. Mr. Frasure also expressed doubts about a licensee’s ability to develop genuinely

independent, but related products after exposure to UNIX. See id. at 169-70. 72. In response to questioning about what products could be disclosed under the

agreement, Mr. Frasure was hesitant to concede that any product remained unprotected, noting
26

that “if you develop your own system modifications and you use - and it works with the software that was licensed, there is a - there is a very good chance you are revealing methods and concepts just strictly through the interface technique that- that’s used.” Id. at 41-42. Mr. Frasure maintained this position even when asked about products that did not overtly disclose methods or concepts. Id. c. 73. David Rodgers

IBM further relies on the declaration of David Rodgers to deny that the UNIX

software agreements afforded any protection to methods and concepts. See Rodgers Decl. (11/5/03) ¶ 12 (Exh. 23). 74. In his deposition in this case, however, Mr. Rodgers acknowledged that the

software agreements protected UNIX methods and concepts. See Rodgers Dep. (6/10/04) at 159 (Exh. S1). Responding to a hypothetical concerning a product created by a licensee that contains “structures and sequences and organization as it appears in System V,” Mr. Rodgers conceded that “if the reason the similarity was there was because it was just copied, then yeah, I would agree that that would be subject to the constraints.” Id.; see also id. at 137. d. 75. Geoff Green

IBM also relies on the declaration of Geoff Green to deny that the UNIX software

agreements afforded any protection to methods and concepts. See Green Decl. (6/17/04) ¶ 6 (Exh. 22). 76. In his recent deposition, however, Mr. Green admitted that methods and concepts

were specifically protected by both the use and disclosure provisions of the software agreements. He stated that, as per the terms of the agreements in 1985 (when both the IBM and Sequent agreements were executed), modifications and derivatives that contained the “concepts, the ideas, the structure, the organization, the methods from the original licensed product,” but did not
27

contain the “source code in the original licensed product” would have been “covered by the agreement” and its restrictions. Green Dep. (11/15/04) at 130-31 (Exh. 5); see also id. at 141-43. 77. Further, Mr. Green contended that UNIX methods and concepts were included

under the definition of the “SOFTWARE PRODUCT” in Section 1.04 of the software agreements. Id. at 106. Addressing the specific time frame of April 1985, when the Sequent agreement was signed (months after the IBM Agreement was signed), Mr. Green acknowledged that the very definition of the “SOFTWARE PRODUCT” term used in Sections 2.01 and 7.06 of the agreements included much more than literal UNIX source code. See id. at 106, 147. 78. Mr. Green testified that the agreements’ protection of modifications and

derivatives extended to products created by employing UNIX methods and concepts, even if they did not contain any literal UNIX source code. See id. at 142-43. 79. Mr. Green farther testified that the IBM Side Letter’s version of Section 7.06

replaced the explicit “methods and concepts” language with another clause that was designed to protect against copying UNIX methods and concepts. See id. at 223; see also id. at 224 (noting again that the revised Section 7.06 was “another way of getting at the methods and concepts”). 80. This continued protection of methods and concepts in the IBM Side Letter is

“consistent” with the fact that AT&T specifically included language expressly protecting UNIX methods and concepts in its standard agreements for many years after the IBM and Sequent software agreements were signed. Id.; see also Sequent Agreement (4/18/85) § 7.06 (Exh. 1); Santa Cruz Agreement (5/6/87) § 7.05 (Exh. 30); Systech Agreement (1/4/89) § 7.05 (Exh. 31); Gates Rubber Agreement (3/1/90) §7.04 (Exh. 32); Information Foundation Agreement (11/24/92) § 7.05 (Exh. 33); Green Hills Agreement (6/21/94) § 7.05 (Exh. 34); Modcomp/Cerplex Agreement (3/28/96) § 7.05 (Exh. 35).

28

e. 81.

Ira Kistenberg

Ira Kistenberg has acknowledged that the software agreements protected against

copying “any structures, sequences, patterns, ideas, methods or concepts from UNIX System V” aside from the express protection for UNIX source code. Kistenberg Decl. (11/12/04) ¶ 5 (Exh. 10). 82. Mr. Kistenberg has further explained that the UNIX structures, sequences,

patterns, methods, and concepts were subject to 7..06(a), which prohibits the disclosure of those methods and concepts (even in modified form) “to anyone, except to the employees of the licensee to whom such disclosure was necessary to the use for which AT&T granted rights under the license agreements.” Id. ¶ 6. f. 83. SCO’s Declarants

The declarations submitted by SCO herewith support the plain reading of the

software agreements and the prior testimony of IBM’s declarants - that the software agreements specifically protected the methods and concepts that were embodied in the UNIX operating system, and that AT&T consistently protected those methods and concepts. 84. AT&T General Attorney Martin Pfeffer has testified: “AT&T intended to protect,

and through its standard software agreements expressly protected, its UNIX business by preventing anyone from using AT&T’s proprietary material in UNIX (including by literally or non-literally copying UNIX source code, disclosing methods or concepts from UNIX, and/or exploiting licensees’ access to the technology in UNIX) without paying UNIX license fees to AT&T.” Pfeffer Decl. (9/7/04) ¶ 10 (Exh. 8). 85. Mr. Pfeffer explained that the further specific protection of “methods or concepts”

in the Sequent software agreement was “specifically designed to cast the widest possible net in order to ensure protection for AT&T’s proprietary information, beyond the common law
29

protections that would otherwise have been available (absent the agreement) for AT&T’s trade secrets and ‘know-how.’” Id. ¶ 9. 86. Mitzi Bond confirmed that “the agreements protected all of the innovations

embodied in the licensed UNIX software, including, among other things, the structures, sequences, patterns, ideas, methods, and concepts.” Bond Decl. (11/4/04) ¶ 7(b) (Exh. 14). 87. Ms Bond recalled her testimony in the BSD case, in which she explained that to

violate “the UNIX license agreement ‘doesn’t necessarily mean you have to have copied exact code. You can also copy an idea or method or concept that you have presented in a different form.’” Id. ¶ 11(d) (Exh. 14) (quoting Bond Dep. (12/9/92) at 219 (Exh. 24)). 88. Evelyn Davis has stated that “the agreements protected all of the innovations

embodied in the licensed UNIX software, including, among other things, the structures, sequences, patterns, ideas, methods, and concepts.” Davis Decl. (11/4/04) ¶ 6(b) (Exh.4). As she has explained, “AT&T agreed to allow its licensees special access to its proprietary UNIX product by licensing the UNIX operating system to them, but complimented that special access with specially designed, and broad, protections in its license agreements that would safeguard the valuable asset.” Id. ¶ 5. 89. In light of the evidence that the methods, concepts, structures, organizations and

other intellectual property in UNIX were clearly protected by the software agreements, see ¶¶ 12-13, 69-96, above, paragraphs 79-89 and 90-101 of IBM’s “Statement of Undisputed Facts” are disputed. 3. The Ownership Language of the IBM Side Letter a. 90. Otis Wilson

IBM cites Otis Wilson in support of its claim that the IBM Side Letter, which

addresses a licensee’s ownership of modifications and derivative works, shows that AT&T’s
30

UNIX licensees were unrestricted in their right to use and distribute such works. See IBM Mem. ¶¶ 100, 107, 110, 113; see also Wilson Decl. (4/26/04) ¶¶ 19-20 (Exh. 11). 91. Contrary to his current declaration, at the time AT&T entered into its software

license and side letter agreements with IBM, Mr. Wilson explained that the licensee’s ownership rights did not eliminate the contracted use and disclosure restrictions. See Exh. 26 at 2 (differentiating between the “property rights” of a licensee and the continued restrictions on such property pursuant to the software agreement); Exh. 36 at 3 (same). b. 92. David Frasure

IBM similarly relies on the declaration of David Frasure in claiming that the IBM

Side Letter demonstrates AT&T’s intent to relinquish any control over the use and disclosure of derivative works. See IBM Mem. ¶¶ 100-01, 107, 109-10, 113; see also Frasure Decl. (3/28/04) 18 (Exh. 18). 93. In his deposition in the BSD case, however, Mr. Frasure made clear that the

contractual statement of ownership did not in any way alter the contractual restrictions on use and disclosure. Frasure Dep. (12/8/92) at 151-152 (Exh. 19). Mr. Frasure articulated that distinction numerous times in his BSD testimony. See id. at 166-67; see also id. at 157-59. 94. Mr. Frasure also used that distinction to explain the textual change, made later in

1985, to revise AT&T’s software agreement in order to reflect substantially similar ownership language from the IBM Side Letter. See id. at 103; see also id. at 111. 95. Moreover, in his testimony in this case, Mr. Frasure acknowledged the distinction

between AT&T’s ownership rights and AT&T’s use and control restrictions. See Frasure Dep. (6/8/04) at 49 (Exh.20). c. 96. David Rodgers

David Rodgers contradicted IBM’s other declarants in his recent deposition,
31

discussing what he characterized as the “open-ended” requirement in Section 2.01, which did not address ownership, but rather use and disclosure restrictions. See Rodgers Dep. (6/10/04) at 128 (Exh. S1). Mr. Rodgers also acknowledged that this understanding applied to a licensee’s own source code. See id. at 135. d. 97. Ira Kistenberg

IBM relies on the declaration of Ira Kistenberg in arguing that the language of the

IBM Side Letter meant that AT&T removed use and disclosure restrictions on licensees’ modifications and derivatives of UNIX System Five that did not contain literally-copied UNIX source code. See IBM Mem. ¶¶ 110, 113; see also Kistenberg Decl. (6/24/04) ¶ 18 (Exh. 9). 98. But Mr. Kistenberg’s more recent declaration recalled AT&T’s decision to

change the language of its agreements (months after the IBM and Sequent agreements had been executed) and clarified that the change related only to ownership, and did not affect the existing use and disclosure restrictions. See Kistenberg Decl. (11/12/04) ¶ 8 (Exh. 10). e. 99. Geoff Green

In his recent deposition, Mr. Green also recalled that AT&T distinguished

between ownership rights, on the one hand, and the contractual use and disclosure restrictions, on the other, and that he was “sure” he had conversations about that distinction with both Mr. Wilson and Mr. Frasure. See Green Dep. (11/15/04) at 300 (Exh. 5). f. 100. SCO’s Declarants

AT&T General Attorney Martin Pfeffer confirmed that AT&T’s ownership

language was not intended to release any licensee from the contractual restrictions on its use, copying, distribution, or disclosure of materials covered by the software agreements. Mr. Pfeffer recalled that in August 1985, AT&T added language to its software agreements “to address inquiries from certain of its licensees concerning the ownership of modifications and derivative
32

works.” Pfeffer Decl. (9/7/04) ¶ 7 (Exh. 8). 101. According to Mr. Pfeffer, the language that AT&T added to its agreements (which

is substantially similar, in relevant part, to the “Regarding 2.01” clause in the IBM Side Letter) “set forth the parties’ intent and agreement concerning only the ownership of modifications or derivative works, and that this language was not intended to change, and did not in fact change, AT&T’s right to control (for example, the use and disclosure) of such modifications and/or derivative works, as set forth in Section 2.01 and other provisions of the software license agreements.” 4 Id. 102. Moreover, Mr. Pfeffer did not recall during his tenure with AT&T and USL

(which lasted from 1962 through 1993) “any such instances in which AT&T or USL changed or agreed to change the contractual limitations on use and disclosure that were set forth in the UNIX license agreement.” Id. 103. In light of the evidence that the language in the IBM Side Letter addressed only

ownership of modifications and derivative works, and did not purport to affect the use and disclosure restrictions on those products, see ¶¶ 18, 97-112, above, paragraphs 100-01, 104-110 and 112-113 of IBM’s “Statement of Undisputed Facts” are disputed. To the extent that IBM maintains that the “clarification” of Section 2.01 related to anything other than a licensees’ ownership rights, paragraphs 18, 53, 108 of IBM’s “Statement of Undisputed Facts” are also disputed.

4

The particular similarity between AT&T’s change to the standard UNIX license agreement in late 1985 and the change enumerated in the IBM Side Letter concerned the addition of the ownership clause to Section 2.01: “AT&T-IS claims no ownership interest in any portion of such a modification or derivative work that is not part of a SOFTWARE PRODUCT.” See, e.g., IBM Side Letter (2/1/8.5) at 2 (Exh. 3); Santa Cruz Agreement (5/6/86) § 2.01 (Exh. 30). The standard software agreement also altered the text of the preceding sentence in Section 2.01, but that change was not included in the IBM Side Letter, and is thus irrelevant to both the Sequent and IBM agreements. See id.

33

4.

The Effect of the IBM Side Letter on Sequent a. Otis Wilson

104.

IBM relies on the declaration of Otis Wilson in arguing that “whether or not we

entered into a side letter or other agreements with our licensees to clarify the treatment of modifications and derivative works, or altered the language of Section 2.01, AT&T’s and USL’s intent was always the same.” IBM Mem. ¶ 112 (citing Wilson Decl. (4/26/04) ¶ 27 (Exh. 11)). 105. In his deposition testimony, however, Mr. Wilson admitted that side letters such

as the one IBM entered into with AT&T could not govern the terms of the AT&T-Sequent UNIX license. See Wilson Dep. (6/10/04) at 232-35 (Exh. 17). b. 106. Geoff Green

Mr. Green acknowledged that the merger clause contained in the Sequent and

IBM agreements was intended “to make clear that the agreement and its supplements constituted the entire agreement. That’s what the language says.” Green Dep. (11/15/04) at 93 (Exh. 5). 107. Applying the merger clause to the terms of the Sequent software agreement,

Mr. Green explained that Sequent’s rights and obligations would be governed only by the agreements that Sequent executed with AT&T. See id. at 95. 108. Mr. Green further testified that, while the software agreements sometimes

contained a “most favored nation clause,” such provisions were typically specific to the context of pricing, and he was not aware of any such clause relating to the software agreements’ intellectual-property protections. Id. at 96. In any event, Mr. Green acknowledged that no such clause appeared in the Sequent software agreement. Id. at 96-97. c. 109. David Rodgers

Mr. Rodgers testified about the written, countersigned process through which a

licensee or AT&T would change the terms of the written software agreements, directly
34

contradicting IBM’s position that the terms of the individual agreements could change even without the parties executing a new agreement or modification. See Rodgers Dep. (6/10/04) at 169-70 (Exh. S1). 110. Mr. Rodgers acknowledged that Sequent signed the “standard form agreement”

and that, aside from price, Sequent did not negotiate or change any terms from that standard agreement. See id. at 82-84. 111. With regard to the language in the IBM Side Letter that removed the explicit

reference to methods and concepts in Section 7.06, Mr. Rodgers acknowledged that no such revision was proposed or made in Sequent’s case. See id. at 146. d. 112. Ira Kistenberg

In his most recent declaration, Mr. Kistenberg rejected the notion that licensees

who were not party to side letters or subsequent agreements would receive the benefit of such language, despite IBM’s attempted reliance on his prior declaration to support such a claim, IBM Mem. ¶¶ 111, 113. See Kistenberg Decl. (11/12/04) ¶ 9 (Exh. 10). 113. Mr. Kistenberg also insisted that “the policy of the UNIX licensing group was that

any changes to a licensee’s software agreement had to be in writing. I never verbally agreed with any licensee to modify its license in any way without memorializing the modification in writing.” Id. ¶ 10. e. 114. SCO’s Declarants

AT&T General Attorney Martin Pfeffer confirmed that the standard merger

language in AT&T’s software agreements “set forth the parties’ intent and agreement that only the terms of each licensee’s written agreements with AT&T would govern those parties’ contractual obligations. In other words, any agreements (including any side letter agreements) that AT&T entered into with one of its UNIX licensees would have no legal effect on the
35

contractual obligations of AT&T vis-a-vis any of its other UNIX licensees.” Pfeffer Decl. (9/7104) ¶ 12 (Exh. 8). 115. In light of the evidence that the software agreements were designed to ensure that

only the terms of each individual licensee’s agreement would govern its licensing of UNIX, see ¶¶ 17, 114-24, above, paragraphs 97-98, and 111-13 of IBM’s “Statement of Undisputed Facts” are disputed. 5. The $ echo Newsletter a. 116. David Frasure

David Frasure acknowledged that the $ echo newsletter did not have any effect on

the provisions of the IBM and Sequent (or any other) software agreements. See Frasure Dep. (6/8/04) at 246-47 (Exh. 20). b. 117. Otis Wilson

When asked about the $ echo newsletter, Otis Wilson admitted that “our purpose

with the newsletter was just to provide information to our licensees, to keep them abreast of what was going on with the product.” Wilson Dep. (6/10/04) at 68 (Exh. 17). c. 118. Geoff Green

Geoff Green testified that the $ echo newsletter did not create binding letter rights

or obligations, and was not intended to modify any of the software agreements or side letter agreements. See Green Dep. (11/1.5/04) at 100-0 1 (Exh. 5). d. 119. Ira Kistenberg

Ira Kistenberg stated that the $ echo newsletter did not, and was not intended to,

create binding contractual rights. See Kistenberg Decl. (11/12/04) ¶ 8 (Exh. 10). e. 120. SCO’s Declarants

Mitzi Bond, who was the editor of the $ echo newsletter for the duration of its
36

existence, explained that the publication was purely informational, and that each individual licensee’s rights and obligations was governed by that particular licensee’s signed software agreement. See Bond Decl. (11/4/04) ¶ 4 (Exh. 14). f. 121. The Text of the $ echo Newsletter

The text of the $ echo newsletter itself confirms that the changes it “announced”

were not even intended to apply to licensees retroactively, but instead to licensees who executed new agreements, which a licensee could request from its AT&T account executive. See AT&T $ echo Newsletter (Aug. 1985) at 5 (“Contact your Account Executive for information or specimen copies.”) (Exh. 37). 122. In light of the evidence that the $ echo newsletter was never intended to, and

never did, have any legal effect on licensees (whose rights and obligations were governed only by the terms of their written agreements, and was created and distributed for informational purposes only, see ¶ 17, 126-32, above, paragraphs 114-124 of IBM’s “Statement of Undisputed Facts” are disputed. D. The Alleged Novell Waiver Is Contrary to the Entire Novell Asset Purchase Transaction 123. In 1995, Novell sold its entire UNIX-related business to SCO’s predecessor-in-

interest The Santa Cruz Operation, Inc. (also referred to herein as “SCO”). See Asset Purchase Agreement dated Sept. 19, 1995 between Novell and SCO (“APA”) Recital A, §§ 1.1(a), 1.3(a)(i), Schedule 1.1(a) (Exh. 38). 124. The available extrinsic evidence confirms that SCO obtained Novell’s UNIX

copyrights through the APA. See Chatlos Decl. (10/1/04) ¶¶ 6-12 (Exh. 39); Wilt Decl. (11/23/04) ¶¶ 5-10,12-16 (Exh. 40); Exh. S2 (E-mail from Ron Smith to Terry McKenna (11/3/97).

37

125.

Moreover, Amendment No. 2 to the APA, which was executed on October 16,

1996, confirms that the APA transferred to SCO “the copyrights and trademarks owned by Novell as of the date of the Agreement required for SCO to exercise its rights with respect to the acquisition of UNIX and UnixWare technologies.” APA Amendment No. 2 ¶ A (Exh. 41). 126. There was one limited exception to the wholesale transfer of Novell’s UNIX

business to SCO: Novell was to receive future royalties paid to SCO from the use of the UNIX product, in binary (or object) code format, by sublicensees of the UNIX System Five (“SVRX”) source code licenses. See APA §§ 1.2(a)-(b), 4.16(a); id. Schedule 1.1(b), Item VIII (excluding from asset sale “All right, title and interest to the SVRX Royalties, less the 5% fee for administering the collection thereof pursuant to Section 4.16 hereof”) (Exh. 38). Otherwise, pursuant to the other provisions in the APA, SCO received complete ownership and control of all UNIX source code, including the right to license the source code for SCO’s own benefit and the right to enforce intellectual-property protections against licensees of the source code. See Chatlos Decl. (10/1/04) ¶¶ 9-10 (Exh. 39); Wilt Decl. (11/23/04) ¶ 11 (Exh. 40); Sabbath Decl. (11/19/04) ¶ 6 (Exh. 42); APA Amendment No. 2 (Exh. 41). 127. The SVRX binary royalties were part of the consideration paid to Novell for the

transfer of the UNIX assets to SCO. See APA § 1.2(a)-(b) (Exh. 38). The parties agreed to this exception because SCO could not pay Novell the full purchase price for the UNIX assets up front; the future binary royalty stream bridged the price gap. See Chatlos Decl. (10/l/04) ¶¶ 7, 9, 1.3 (Exh. 39); Wilt Decl. (11/23/04) ¶¶ 6, 9, 10 (Exh. 40). 128. Section 4.16 of the APA governs Novell’s rights to the future SVRX binary

royalties. Section 4.16(a) sets out the procedures for SCO’s collection and payment to Novell of the royalties, and gave Novell the right to audit those collection efforts. See APA § 4.16(a)

38

(Exh. 38). SCO receives a fee for performing this administrative function. See id. (“In consideration of such activities described in the preceding sentence, Seller shall pay to Buyer within 5 days of receipt of SVRX Royalties from Buyer as set forth in the preceding sentences, an administrative fee equal to 5% of such SVRX Royalties.”). 129. Section 4.16(b) protects Novell’s interests in the future binary royalties by

providing in relevant part: “In addition, at Seller’s sole discretion and direction, Buyer shall amend, supplement, modify or waive any rights under, or shall assign any rights to, any SVRX License to the extent so directed in any manner or respect by Seller.” Id. § 4.16(b). 130. The text of Section 4.16 of the APA does not give Novell rights to veto or to

waive decisions made by SCO under the software agreements at issue here. Section 4.16 refers only to SVRX Licenses, not more broadly to software agreements. “SVRX Licenses” is not defined by the APA, but are referenced to in Section 4.16 as being listed in “detail” under Item VI of APA Schedule l.1(a). By contrast, UNIX “Software Agreements” are expressly covered by Item III.L to Schedule 1.1 (a) - an Item not referenced anywhere in Section 4.16. 131. Section 4.16(c) of the APA protects the assets transferred to SCO from

interference by Novell, by prohibiting Novell from promoting the sale of SVRX products: “Seller further covenants that immediately following the Closing Date neither it, nor any of its officers, directors or employees shall … take any material action designed to promote the sale of SVRX products.” Id. § 4.16(c). 132. Amendment Number 1 to the APA, effective as of the December 6, 1995 closing

of the APA, permits SCO to retain all of the SVRX royalties with respect to certain categories of SVRX licenses, including SCO’s own SVRX license. See APA Amendment No. I ¶ E (Exh. 43). 133. Amendment Number 2 to the APA, which was executed on September 19, 1995,

39

states: “This Amendment does not give Novell the right to increase any SVRX licensee’s rights to SVRX source code,” and that “Novell may not prevent SCO from exercising its rights with respect to the SVRX source code in accordance with the Agreement.” APA Amendment No. 2 ¶ B.5 (Exh. 41). 134. At the time that Novell purported to waive IBM’s breaches of the software

agreements, no binary royalty payments were being made or were expected to be made in the future, under the IBM or Sequent SVRX licenses. In Amendment X (Oct. 17, 1996) to IBM’s software license, IBM had agreed to pay SCO a nonrefundable fee of $10.125 million for a perpetual license to use the UNIX source code and to sublicense binary products; Amendment X thus effected a by-out of IBM’s future SVRX binary royalty obligations. Exh. 57. Similarly, by the time Novell purported to waive SCO’s source code rights with respect to Dynix (in February 2004), no further royalty payments were being made under Sequent’s license; the last such royalty payment to SCO (of only approximately $3,000) was made in the first quarter of 2003. Acheson Decl. (11/30/04) ¶ 3 (Exh. 44). 135. At the time it purposed to direct SCO to waive its rights, Novell was seeking to

promote business interests entirely separate from any rights it retained under the APA. In a series of deals struck between November 2003 and March 2004, Novell and IBM formed a very close Linux partnership: Novell acquired Linux distributor SuSE Linux, and IBM invested $50 million in Novell to finance part of the acquisition. See Press Release, Novell, Inc., Novell Finalizes IBM Investment (Mar. 3, 2004) (Exh. 45). 136. Under the IBM-Novell deal, Novell became a major Linux distributor, allowing

IBM to continue its official distance from the Linux distribution channel. IBM spokesman Mike Darcy stated: “IBM is not a Linux distributor and has no interest in being in that business.”

40

Hiawatha Bray, Novell to Buy SuSE Linux for $210M; Firm’s Commitment to Software Boosted, The Boston Globe, Nov. 5, 2003 (Exh. 46). 137. Nevertheless, IBM benefits significantly from the Linux distributions made by

Novell. Novell CEO Jack Messman explained the benefits of its partnership with IBM as follows: “IBM stands to gain a great deal of revenue by providing systems, peripheral devices, software and services for Linux.” Bob Mims, Novell, IBM and HP unite efforts to put Linux on top: Three companies unite to boost Linux and topple Microsoft, The Salt Lake Tribune, Mar. 25, 2004, at El (Exh. 47). 138. Novell and SCO entered into a Technology License Agreement

contemporaneously with the APA. As contemplated by Section 1.6 of the APA, the Technology License Agreement effected a “license back” to Novell of rights to use the UNIX System Five code. See Technology License Agreement (12/6/95) (Exh. 48); APA § 1.6 (Exh.38). 139. Under the Technology License Agreement, Novell was permitted to use UNIX

and UnixWare only for internal purposes and only in bundled products, and then only to the extent that no material part of the bundled product competed with SCO’s UnixWare business. See Technology License Agreement (12/6/95) (Exh. 48). 140. The chief negotiators for both Novell and SCO agree that neither SCO nor Novell

intended for the APA to grant Novell the sweeping rights over the assets it sold to SCO that IBM argues were granted. See Chatlos Decl. (10/1/04) ¶ 13 (Exh. 39); Wilt Decl. (11/23/04) ¶ 10 (Exh. 40); Sabbath Decl. (11/19/04) ¶ 5 (Exh. 42). 141. Ed Chatlos, who negotiated the APA for Novell, explained that “Paragraph 4.16

of the APA was specifically designed and intended to protect Novell’s retained binary product royalty stream” and that Novell retained no “right to waive, or to direct or require SCO to waive,

41

any of SCO’s source code rights, including under customer source code licenses.” Chatlos Decl. (10/1/04) ¶ 13 (Exh. 39). 142. Jim Wilt, who negotiated the APA for SCO, confirmed that Paragraph 4.16 of the

APA allowed Novell to manage royalty streams, but “not at the expense of SCO’s right to enforce its intellectual-property protections under any such licenses, and not to permit Novell to waive any of those protections.” Wilt Decl. (11/23/04) ¶ 10 (Exh. 40). Mr. Wilt also explained that Amendment No. 2 was intended to confirm the parties’ intent in this regard. See id. 143. Steve Sabbath, Santa Cruz’s former Vice President of Law and Corporate Affairs

and a participant in a number of the meetings and discussions with the chief negotiators and attorneys leading up to the APA, similarly refuted the suggestion that Section 4.16(b) gave Novell the right to waive any breach of the intellectual-property protections. According to Mr. Sabbath, IBM’s proposed interpretation would permit Novell to eviscerate the entire purpose of the asset-purchase transaction. See Sabbath Decl. (11/19/04) ¶ 5 (Exh.42). 144. Mr. Sabbath, who was involved in the discussions leading up to Amendment

No. 2 and who executed that amendment on SCO’s behalf, has also explained that: “Amendment No. 2, however, was intended to confirm, among other things, the parties’ intent that SCO would obtain ownership of the UNIX copyrights under the APA and that Novell had received no rights with respect to UNIX source code under the APA. Paragraph B.5 of Amendment No. 2 was specifically intended to make clear that Novell had no right to increase any SVRX licensee’s rights to SVRX source code, no right to grant any new SVRX source code licenses, and no right to prevent Santa Cruz from exercising the rights it obtained under the APA with respect to SVRX source code.” Id. ¶ 6. 145. In light of the evidence that Novell and SCO, in executing the APA, intended and

agreed to transfer to SCO complete ownership and control of Novell’s UNIX business, with the exception that Novell retained an interest in future binary-code royalties from the licensing of the SVRX product, see ¶¶ 134-40, above; and in light of the evidence that Section 4.16(b) of the

42

APA was intended only to reserve for Novell the right to protect its interests in the binary-code royalties and not to interfere with SCO’s otherwise unfettered the UNIX source code (including the right to enforce the intellectual-property protections in the UNIX software agreements), see ¶¶ 141-55 above, paragraphs 21-23, 25, 127-38 of IBM’s “Statement of Undisputed Facts” are disputed. E. The Alleged Linux Waiver 1. 146. SCO’s Efforts to Protect Its Rights in UNIX Other than through express written agreements and for due consideration, SCO

has never intended to waive and has always sought to protect and evidence its rights under the UNIX System Five (“SVRX”) source software agreements that SCO has administered and enforced since 1995. See Sontag Decl. (11/30/04) ¶ 3 (Exh. 49). 147. SCO did not know that IBM had contributed source code to Linux in violation of

its (and Sequent’s) SVRX software agreements until December 2002 or January 2003. Id. ¶ 5. 148. After filing suit against IBM for breaching the SVRX agreements, SCO further

demonstrated its intent to enforce its rights under those agreements by delivering a termination notice to IBM pursuant to Section 6.3 of the SVRX license. Id. ¶ 10. 149. SCO attempted to meet and confer with IBM, but IBM failed to cure its breaches

during the two-month period provided in SCO’s termination letter to IBM. Id. ¶ 11. Accordingly, effective June 13, 2003, SCO terminated IBM’s SVRX license; and effective July 30, 2003, SCO terminated the Sequent SVRX license. Id. SCO thus further demonstrated its intent to enforce its rights under those licenses. Id. 2. 150. SCO’s Distribution of Linux SCO copied, advertised, and distributed the Linux kernel and other related Linux

software for years before 2003. See Hughes Decl. (11130/04) 9 (Exh. 50).
43

151.

SCO did not know that any of the features mentioned in SCO’s advertisements

(with the sole exception of the journaling file system support, or “JFS”) had been contributed by IBM to Linux. Sontag Decl. ¶¶ 6-8. As to that one exception, SCO marketed JFS as “developed by IBM,” but did not know that JFS was derived from SCO-licensed code or that IBM had contributed it in breach of the software agreements. Id. ¶ 8. 152. Indeed, SCO did not know that any of the advertised features had been derived

from SCO’s proprietary software licensed to IBM or had been contributed to Linux by IBM in violation of IBM’s agreements with SCO. Id. ¶ 17-8. 153. In light of the evidence that SCO had no knowledge that the features mentioned in

its advertisements, with the exception of JFS, had been contributed by IBM, and that with regard to JFS, SCO had no knowledge that JFS was derived from UNIX and contributed to Linux by IBM in breach of the UNIX software agreements, see ¶¶ 150-52, above, paragraphs 144-49, 154 of IBM’s “Statement of Undisputed Facts” are disputed. 3. 154. SCO’s Suspension of Its Distribution of Linux SCO marketed and licensed Linux Server 4.0 for only a few months, from

November 19, 2002, until May 14, 2003. Hughes Decl. (11/30/04) ¶ 2 (Exh. 50). 155. After filing suit against IBM, SCO considered whether to continue selling and

marketing Linux-related products, including SCO Linux Server 4.0. Sontag Decl. (11/30/04) ¶ 12 (Exh.49). 156. An important issue SCO considered was its obligations to existing SCO

customers. Id. ¶ 13 SCO took the view that SCO’s customers were entitled to order SCO’s products and updates from SCO for a period of time after becoming customers 157. SCO decided that the most sensible solution was to suspend its sale and marketing

of its Linux-related products (which SCO did effective May 14, 2003), and to continue to allow
44

SCO’s pre-existing customers to order such products. Id. ¶ 14. Since May 14, 2003, SCO has not entered into any further obligations to sell Linux Server 4.0 or OpenLinux 3.1.1. Hughes Decl. (11/30/04) ¶ 3 (Exh. 50). 158. By suspending the sale of its Linux-related products, including the operating

system, services, support, professional services, education, and layered applications, SCO loss approximately 5-10% of its revenues. Sontag Decl. (11/30/04) ¶ 15 (Exh. 49). From May 14, 2003, until May 31, 2004 (when SCO last sold a unit of Linux Server 4.0), SCO sold 83 units and had 79 units returned, for a gross revenue of $1,849. Id. After August 5, 2003, there were 401 sales of SCO OpenLinux 3.1.1, for total gross revenue of $50,796. During the same period, 22 copies of this product were returned, so net revenue was $50,025. Hughes Decl. ¶ 5. 159. In taking into account the foregoing considerations and reaching the foregoing

decisions, SCO never intended to waive its right to enforce its SVRX software agreement against IBM and Sequent. Sontag Decl. (11/30/04) ¶ 13 (Exh. 49). 160. In light of the evidence that, contrary to IBM’s assertions, SCO suspended its

distribution of Linux shortly after filing this action in May 2003; that SCO did not enter into any new contracts to sell SCO Linux Server 4.0 after May 14, 2003; and that SCO’s post-May 14, 2003 sales were made to honor pre-May 14, 2003 contractual obligations, see ¶¶ 154-59, above, paragraphs 150-152 of IBM’s “Statement of Undisputed Facts” are disputed. 4. 161. SCO’s Protection of Information on Its Website Nor has SCO provided public access to Linux through its website. To the

contrary, in order to preclude public access to SCO Linux Server 4.0 Server files, SCO provided customers who purchased this product with a password to enter at a log-in screen on SCO’s website, so that only those customers would have access. Sontag Decl.. ¶ 19. 162. Between October 31 and December 1, 2003, IBM repeatedly accessed the SCO
45

log-in site but did not obtain access to the Linux Server 4.0 files. Id, ¶ 25. After news of a bug in the website’s security system was reported on internet websites, IBM exploited the bug to bypass the security system, hacked into SCO’s website, and downloaded the very files IBM has now attached to its motion. Id. ¶¶ 22-27. 163. In light of the evidence that SCO never made its Linux distributions publicly

available on its website, see ¶¶ 161-62, above, paragraph 153 of IBM’s “Statement of Undisputed Facts” is disputed. F. SCO’s Application Under Rule 56(f) 164. IBM mischaracterizes SCO’s contract claims, and further complains that SCO has

not produced proof of its claims, IBM Mem. ¶¶ 29-48, when IBM has refused to provide the discovery essential to developing that proof. Based on the evidence that SCO has thus far discovered even despite IBM’s delay, SCO reasonably believes that additional discovery will bring to light additional facts giving rise to triable issues regarding the interpretation of the UNIX software agreements, IBM’s breaches of those agreements, and IBM’s two “waiver” theories. 1. 165. IBM’s Contract-Interpretation Arguments SCO is entitled to further discovery regarding the contract interpretation

arguments IBM advances in this motion, particularly in light of the fact that many of IBM’s declarants - including Otis Wilson, David Frasure, Geoff Green, Ira Kistenberg, and David Rodgers - have given testimony that is directly contrary to the declarations on which IBM relies. See ¶¶ 20-65, 78-101, 108-122, 126-136, 138-141, above. Given the conflicts in the testimony, SCO reasonably can expect that other depositions will likewise call into question IBM’s other declarations.

46

2. 166.

IBM’s Breach of the Software Agreements SCO has already identified tens of thousands of lines of literally-copied code that

IBM has dumped directly from AIX and Dynix into Linux. See SCO’s Revised Supplemental Responses to IBM’s First and Second Set of Interrogatories, at Tbls. A-K; April 19 Letter from B. Hatch to T. Shaughnessy, at Exhs. B-C. 167. There is ample evidence that IBM and Sequent themselves viewed AIX, Dynix,

and even Linux as UNIX derivatives. See, e.g., IBM Dictionary of Computing, 18 (George McDaniel ed. 1994) (Exh. 51) (defining IBM’s “AIX operating system” as “IBM’s implementation of the UNIX (trademark of AT&T Bell Laboratories) operating system.”); IBM’s Mills Sets Sights on Middleware, Linux, Computerworld, Oct. 24, 2003 (Exh. 52) (“We took the Unix System V kernel and we made modifications”); IBM Small and Medium Business Advantage Frequently Asked Questions, available at http://www1.ibm.com/partnerworld/pwhome.nsf/weblook/smb offerings_ sol_linux_faq.html (“What is the difference between UNIX and Linux? Linux is derived from UNIX but is developed through an open source community of programmers.”) 5 (Exh. 53). 168. SCO has previously demonstrated in detail to this Court and to the Magistrate

Court how the programming-history discovery SCO has long sought (and IBM continues to withhold) will permit SCO to show specifically how the source code that IBM contributed to Linux was covered by the IBM and Sequent software agreements. SCO adopts those
5

See also E-mail from Bill Sandve to Kim Tran, dated Jan. 22, 2002 at 2 (Exh_ 53) (“AIX was derived from System V.”); Agreement for Licensing of AIX Source Code and Related Products between Argus Systems Group, Inc.. and IBM ¶ 8.4 (Exh, S4) (“AIX is derived from software under license from SCO.”); IBM Royalty Statement, June 30, 1987 (Exh. 55) (stating that AIX is “derived from System V”); IBM/Supplier Technical Services Agreement at 1 (Exh. 56) (“‘AIX Operating System’ or ‘AIX’ shall mean the UNIX operating system that operates on the Power, Power PC, Power 2, Power 3, and Power 4 architectures or derivative or follow-on architectures irrespective of the names of such architecture.”); Rodgers Dep. (6/10/04) at 138 (Exh. S1) (“Dynix/ptx is almost certainly a derivative work of Unix System V); IBM Linux Strategy Presentation (Exh..57) (“remember, Linux is derived from UNIX”).

47

memoranda and arguments by reference. 3. 169. IBM’s Waiver Arguments With respect to IBM’s claim that Novell has waived SCO’s intellectual-property

rights, SCO intends to take the depositions of other participants in the negotiations of the 1995 APA and of the October 1996 Amendment No. 2 thereto. See Normand Decl. (11/30/04) ¶ 14. The declaration of Edward Normand is being submitted together with SCO’s Rule 56(f) motion. 170. SCO reasonably expects that the testimony of the other participants will be

consistent with the declarations that SCO has submitted from the chief negotiators (Ed Chatlos, Jim Wilt, and Steve Sabbath) and will further confirm that the parties never intended for Novell to have the light to waive, or to direct or require SCO to waive, any of SCO’s intellectualproperty protections under any UNIX software agreements. Id. LEGAL STANDARDS The standards prescribed for summary judgment under Federal Rule of Civil Procedure 56 are stringent. “Summary judgment should not be granted unless the evidence, viewed in the light most favorable to the party opposing the motion, shows there are no genuine issues of material fact and the moving party is due judgment as a matter of law.” Blackhawk-Cent. City Sanitation Dist. v. Am. Guar. & Liab. Ins. Co., 214 F..3d 1183, 1188 (10th Cir. 2000); see Fed. R. Civ. P. 56(c). It is axiomatic that the “moving party carries the burden of showing beyond a reasonable doubt that it is entitled to summary judgment,” and that “the court must review the record in the light most favorable to the opposing party.” Hicks v. City of Watonga, 942 F.2d 737, 743 (10th Cir. 1991). Thus, the Court “must resolve factual disputes and draw inferences” in favor of the non-moving party, Rogers v. United States, 281 F.3d 1108, 1113 (10th Cir. 2002), and the Court may not “act as the jury and determine witness credibility, weigh

48

the evidence, or decide upon competing inferences.” Boyer v. Bd. of County Comm’rs of County of Johnson County, 922 F. Supp. 476, 484 (D. Kan. 1996). In short, summary judgment may not be granted unless “the uncontroverted material facts establish that the moving party is entitled to judgment as a matter of law.” David v. City & County of Denver, 101 F.3d 1344, 135.5 (10th Cir. 1996) (emphasis added). The Court also has the authority under Rule 56(f) to deny or continue a summary judgment motion where “the nonmoving party has not had an opportunity to make full discovery.” Celotex Corp. v. Catrett, 477 U.S. 317, 326 (1986). Summary judgment is improper “at a stage of the case when all of the facts and circumstances material to a just determination of the controversy [are] not before the court.” Morrison Flying Serv. v. Deming Nat’l Bank, 340 F.2d 430, 432 (10th Cir. 1965). Thus, to invoke Rule 56(f), the nonmoving party need only show “the probable facts not available,” and the steps taken to obtain those facts, and “how additional time will enable him to rebut the movant’s allegations of no genuine issue of fact.” Brightway Adolescent Hosp. v. Health Plan of Nev., Civ. No. 2:98CV0729C, 2000 WL 3371084.S (D. Utah Sept. 20, 2000) (Exh. A). ARGUMENT IBM’s motion for summary judgment on SCO’s contract claims - the heart of SCO’s case - is based on (1) IBM’s proposed interpretation of the operative IBM and Sequent software license agreements and on (2) IBM’s claims that Novell (“on SCO’s behalf’) and SCO itself have waived IBM’s breaches of those agreements. IBM’s arguments, however, rely on contract interpretations that cannot be squared with the language of the relevant agreements, the intent of the parties to those agreements, or commercial sense. Moreover, substantial evidence developed

49

to date precludes not only IBM’s reliance on its disputed extrinsic evidence of the contracting parties’ intent, but also IBM’s waiver claims. I. THE PLAIN LANGUAGE, THE DISPUTED EXTRINSIC EVIDENCE, AND THE PURPOSE OF THE UNIX SOFTWARE AGREEMENTS PRECLUDE IBM’S CLAIM THAT ONLY LITERAL COPYING OF UNIX SOURCE CODE IS PROHIBITED The software agreements at issue mean what they say: the license is limited to “internal use,” confidentiality requirements must be observed, and these (and other) restrictions apply to modified and derivative products. Despite the plain language of the software agreements, IBM argues that “The IBM and Sequent Software Agreements cannot, as a matter of law, be construed to prohibit IBM’s use and disclosure of code in AIX or Dynix, like the code that IBM allegedly contributed to Linux, that does not contain any UNIX System V code.” IBM Mem. at 63; see also id. at 48 (arguing for “only one conclusion - that IBM owns and is free to disclose any source code that it or Sequent creates, so long as that code does not contain UNIX System Five code,).\ 6 IBM downplays the vast sweep of its position by repeatedly asserting that it seeks only to protect its “own homegrown code.” IBM Mem. at 2, 23, 47-8, 55, and 64. But to IBM, “homegrown code” means “source code that is written by IBM and does not include UNIX System V code.” Id. at 2. What IBM calls “homegrown” is not homegrown at all. IBM’s argument reduces to the astonishing claim that under the very software agreements designed was to protect the UNIX intellectual property, IBM (and all other UNIX licensees) obtained, among other things, the right to misappropriate for its own unfettered use, control, copying, export, and disclosure any non-original source code that they derived, modified, or copied in substantially

6

50

similar form from UNIX (including any source code employing UNIX ideas, methods, concepts, structures, sequences, and organization) so long as IBM did not copy the UNIX source code verbatim. See id. at 51-52 (acknowledging that under IBM’s contract interpretation, UNIX licenses would not even protect against non-literal copying prohibited by copyright law). 7 IBM also mischaracterizes SCO’s contract theory by stating that SCO’s claims “rest entirely on the proposition that ‘[t]he AIX work as a whole and the Dynix/ptx work as a whole are modifications of, or are derived from [UNIX] System V.”‘ IBM Mem. ¶ 62. As SCO has previously explained, the language of the software agreements does not require SCO to show that AIX and Dynix are “as a whole” modifications of or derived from the licensed UNIX product, because those agreements clearly prohibited IBM from making particular contributions to Linux of code sections that were themselves modified or derived from UNIX. See, e.g., SCO’s Reply Memorandum Regarding Discovery (7/12/04) (“Reply Mem. re Disc.”) at 11-14; SCO’s Mem. in Support of SCO’s Mot. to Enforce Sched. Order (9/08/04) at 22. Moreover, IBM and Sequent have repeatedly described and treated the AIX and Dynix programs (and, indeed, Linux itself) as UNIX derivatives. See, e.g., IBM’s Mills Sets Sights on Middleware, Linux, Computerworld, Oct. 24, 2003 (Exh. 52) (“We took the Unix System V kernel and we made modifications”); IBM Dictionary of Computing, 18 (George McDaniel ed., 1994)

7

Although IBM’s motion is premised on its “literal source code only” interpretation of the software agreements, IBM also suggests in passing that SCO’s claims fail under “even the most expansive” and “broad reading” of the agreements. IBM Mem. at 51-52. What IBM incorrectly characterizes as the broadest possible reading of the software agreements, however, would merely provide the same protection as is already provided by copyright law i.e., against literal and non-literal copying). See id. at 52 n.8. IBM thus disregards again the breadth of the contractual protections in this case, which cover, among other things, all works resulting from exposure to the original licensed UNIX product as well as the methods, concepts, and other intellectual property embodied in the licensed software. IBM’s motion does not- and cannot - even attempt to challenge SCO’s claims under this well-supported interpretation of the operative agreements. While IBM has refused SCO the programminghistory discovery that would permit SCO to demonstrate precisely how IBM’s Linux contributions can be traced back to the licensed UNIX software, the evidence that IBM has produced thus far contains repeated admissions by IBM and Sequent that the AIX and Dynix products from which IBM dumped its substantial Linux contributions are, in fact, derivatives of UNIX. See ¶ 167, above.

51

(Exh. 51) (defining IBM’s “AIX operating system” as “IBM’s implementation of the UNIX (trademark of AT&T Bell Laboratories) operating system.”); IBM Small and Medium Business Advantage Frequently Asked Questions, available at http://www-l.ibm.com/partnerworld/ pwhome.nsf/weblooklsmb_offerings_sol_linux_faq.html (“What is the difference between UNIX and Linux? Linux is derived from UNIX but is developed through an open source community of programmers.”) (Exh. 53). 8 In any event, IBM, as the movant, bears the burden of demonstrating that its proffered interpretation of the software agreements is correct as a matter of law. Trainor v. Apollo Metal Specialities, Inc., 318 F.3d 976, 979 (10th Cir. 2002). To prevail on its motion, IBM must establish that its proposed “literal source code only” reading of the software agreements is the only reasonable interpretation of the contract language. See Jellinick v. Joseph J. Naples & Assocs., Inc., 296 A.D.2d 75,78-79 (N.Y. App. Div. 2002) (affirming denial of summary judgment because “the moving party has the burden of establishing that its construction of the agreement is the only construction which can fairly be placed thereon”); Wards, Inc. v. Stamford Ridgeway Assocs., 761 F.2d 117,120-22 (2d Cir. 1985) (reversing summary judgment because material issue of fact exists for trial unless “the moving party can establish that contractual language is not susceptible of at least two fairly reasonable meanings”). IBM’s motion does not - and cannot - carry that burden. As shown below, neither the plain language of the software agreements, nor the extrinsic evidence developed to date, nor
8

See also E-mail from Bill Sandve to Kim Tran, dated January 22, 2002 at 2 (Exh. 53) (“AIX was derived from System V.”); Agreement for Licensing of AIX Source Code and Related Products between Argus Systems Group, Inc. and IBM ¶ 8.4 (Exh. 54) (“AIX is derived from software under license from SCO.”); IBM Royalty Statement, June 30, 1987 (Exh. 55) (stating that AIX is “derived from System V”); IBM/Supplier Technical Services Agreement at I (Exh. 56) (“‘AIX Operating System’ or ‘AIX’ shall mean the UNIX operating system that operates on the Power, Power PC, Power 2, Power 3, and Power 4 architectures or derivative or follow-on architectures irrespective of the names of such architecture.”); Rodgers Dep. (6/10/04) at 138 (Exh. S1) (“Dynix/ptx is almost certainly a derivative work of Unix System V); IBM Linux Strategy Presentation (Exh. 57) (“remember, Linux is derived from UNIX”).

52

IBM’s appeal to “general principles of public policy,” IBM Mem. at 48, supports IBM’s attempt to rewrite the contractual protections to cover only literally copied UNIX source code. Indeed, each of these considerations requires the denial of IBM’s motion. A. Section 2.01 of Both Software Agreements Plainly and Unambiguously Extends All of the Contractual Restrictions Beyond UNIX Source Code, to Any Modifications and Derivative Works IBM first attempts to support its contract-interpretation argument based on a purported appeal to the “plain and unambiguous terms” of AT&T’s standard software agreement (which was executed by both IBM and Sequent) as well as the Side Letter Agreement that AT&T entered into with IBM (but not with Sequent). IBM Mem. at 49. As discussed below, however, the operative agreements plainly, unambiguously, and specifically protect against the misappropriation and unauthorized disclosure of any “resulting materials” from licensees’ contractual right under the agreements to “modify such SOFTWARE PRODUCT and to prepare derivative works based on such SOFTWARE PRODUCT” (hereinafter, “UNIX Modifications and Derivatives”), and the Sequent agreement specifically requires confidential treatment of any UNIX “methods or concepts.” Moreover, neither of the agreements mention - or, according to the plain reading, permit - a limitation on the licensor’s rights to protection against only literal copying of UNIX source code. Accordingly, there is no merit to IBM’s argument that it is entitled to summary judgment on the ground that these documents unambiguously resolve, in its favor, the contract interpretation issues on either, much less both, of the IBM and Sequent software agreements. If anything, however, the text of those agreements demonstrates, as a matter of law, that they protect far more than mere UNIX source code.

53

1.

Section 2.01 of Both Software Agreements Plainly and Unambiguously Extends All of the Contractual Restrictions Beyond Source Code, to Any UNIX Modifications or Derivatives.

Consistent with their purpose of safeguarding the valuable UNIX intellectual property AT&T’s software agreements imposed strict limitations on licensees’ use of the software products licensed under the agreements. The agreements require, among other things, that:
•

each licensee use the licensed “SOFTWARE PRODUCT” “solely for LICENSEE’S own internal business purposes and solely on or in conjunction with DESIGNATED CPUs for such SOFTWARE PRODUCT,” Sequent Agreement (4/18/85) § 2.01 (Exh. 1), IBM Agreement (2/1/85) § 2.01 (Exh. 2); each licensee hold the “SOFTWARE PRODUCTS subject to this Agreement in confidence for AT&T,” Sequent Agreement (4/18/85) § 7.06(a) (Exh. 1), IBM Side Letter (2/1/85) § 9 (Exh. 3); and no licensee “sell, lease or otherwise transfer or dispose of a SOFTWARE PRODUCT in whole or in part,” Sequent Agreement (4/18/85) § 7.10 (Exh. 1), IBM Agreement (2/1/85) § 7.10 (Exh. 2).

•

•

SCO claims that by dumping portions of its UNIX-based programs into Linux - specifically, portions of IBM’s AIX and Sequent’s Dynix programs - IBM breached these provisions of the IBM and Sequent software agreements. The only issue presented by IBM’s contract-interpretation motion is the scope of the materials protected by the above-described provisions. The plain language of the provisions belies IBM’s contention that the restrictions apply only to the literal source code of the licensed UNIX product. In addition to expressly protecting the original licensed “SOFTWARE PRODUCT” itself which indisputably included the UNIX System Five source code 9 - the software agreements plainly and unambiguously extended the protections for the “SOFTWARE

9

Each of the software agreements defines “SOFTWARE PRODUCT” as follows: “materials such as COMPUTER PROGRAMS, information used or interpreted by COMPUTER PROGRAMS and documentation relating to the use of COMPUTER PROGRAMS.” Sequent Agreement (4/18/85) § 1.04 (Exh_ 1); IBM Agreement (2/1/85) § 1.04 (Exh. 2).

54

PRODUCT” to products that licensees developed through their access to the original licensed product (and source code). Section 2.01 of both the IBM and Sequent software agreements - the cornerstone “grant of rights” found in all of the agreements - specified that each licensee’s “right to use” the licensed product “includes the right to modify such SOFTWARE PRODUCT and to prepare derivative works based on such SOFTWARE PRODUCT, provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT.” (Emphasis added). 10 In direct contrast to IBM’s claim that the software agreements protect against only a licensee’s literal copying of IBM source code, Section 2.01 requires that “resulting materials” be treated “as part of’ UNIX for purposes of the contract. Section 2.01 plainly protects the “materials” that “result[]” from (1) the licensee’s exercise of its “right to modify” the original “SOFTWARE PRODUCT” (which includes the UNIX System Five source code) and (2) the licensee’s right “to prepare derivative works based on such SOFTWARE PRODUCT.” As discussed below, IBM’s proposed construction of the software agreements violates well-settled principles of contract interpretation requiring (1) that no contractual provisions be rendered superfluous, and (2) that contractual provisions are read so as not to conflict with one another. See Ronnen v. Ajax Elec. Motor Corp., 88 N.Y.2d 582, 589 (1996) (New York Court of Appeals has “long and consistently ruled against any construction which would render a contractual provision meaningless or without force and effect”); Merrill Lynch v. Adler, 6.51 N.Y.S.2d 38, 39 (App. Div. 1996) (stating that “an agreement should be read to give effect to all

10

Section 2.01 of the IBM agreement (but not the Sequent agreement) was amended by the IBM Side Letter to include language concerning the “ownership” of derivative works. Because that provision does not affect the plain reading of Section 2.01 set forth above and, in any event, does not affect Sequent, it is addressed in Parts LA.4. & I.B.2.c., below.

55

of its provisions and to render them consistent with each other”) 11 ; Robshaw v. Health Mgmt., Inc., 470 N.Y.S.2d 226, 227 (App. Div. 1983) (employing “general rule that a contract should be interpreted to avoid inconsistencies and to give meaning to all of the terms”). 12 First, if IBM’s interpretation were correct, then the requirement that “resulting materials” he “treated as part of the original SOFTWARE PRODUCT” would be rendered superfluous, because UNIX source code is already clearly and indisputably contained within the original SOFTWARE PRODUCT. In other words, there would simply be no reason for the contract to require that source code contained in “resulting materials” be treated “as part of the original SOFTWARE PRODUCT” because such source code is already covered by the contractual protections afforded to the original SOFTWARE PRODUCT. 13 Second, IBM’s reading also fails to account for the protection under Section 2.01 for both modifications of the licensed UNIX product and “derivative works” based on that product. A modification plainly denotes that a programmer changed or altered portions of a product, but the separate protection under the agreement for “derivative works” clearly reflects an intent under Section 2.01 to cover products that were based on, but did not necessarily include, the original licensed source code. IBM’s reading would render meaningless the term “deriviative works.” If

11

The software agreements contain a New York choice-of-law provision. See Sequent Agreement (4/18/85) § 7.13 (Exh. 1), IBM Agreement (2/1/85) § 7 13 (Exh.. 2). Accord Garza v. Mar. Trans. Lines, Inc., 861 F.2d 23, 27 (2d Cir. 1988) (suggested contract interpretation that would render “at least one clause superfluous or meaningless” should be “avoided if possible”); Ruttenberg v. Davidge Data Svs. Corp., 626 N.Y.S.2d 174, 178 (App. Div. 1995) (citing “rule of construction that a contract shall be read so as to give effect to each and every term”). At their depositions, none of the witnesses from whom IBM has submitted declarations were able to reconcile IBM’s “literal source code only” argument with this glaring textual problem. See Wilson Dep. (6/10/04) at 241, 264-6.5 (Exh. 17); Frasure Dep. (6/8/04) at 256-57 (Exh. 20); Rodgers Dep. (6/10/04) at 101 (Exh. S1). Indeed the only attorney on whom IBM has relied, Geoff Green, was at a complete loss to explain how IBM’s view of the contract would not convert the final clause of Section 2.01 (“provided the resulting materials are treated hereunder as part of the original SOFTWARE PRODUCT”) into mere surplusage. See Green Dep. (11/15/04) at 111-12, 121-22, 241 (Exh. 5).

12

13

56

IBM were correct in contending that the software agreements protected only literal UNIX source code, there would be no reason to delineate an additional - let alone two additional - categories of products that are distinct from the actual code, but were required to be treated the same as the original licensed product. Third, IBM’s “literal source code only” argument is textually unsupportable for the additional reason that it is incompatible with Section 2.01’s language requiring derivatives or modifications to be treated “as part of” the original SOFTWARE PRODUCT. By definition, treating such materials “as part of the original SOFTWARE PRODUCT” must mean adding something to the “original SOFTWARE PRODUCT” that was not already there. In other words, because source code is already contained in the “original SOFTWARE PRODUCT,” IBM’s reading of Section 2.01 (i.e., that it covers only UNIX source code) would provide nothing to add (i.e., nothing to treat “as part of”). 14 2. Section 7.06(a)’s Confidentiality Requirement Precludes IBM’s Narrow “Literal Source Code Only” Claim.

Even beyond the plain meaning of Section 2.01, Section 7.06(a) of the operative IBM and Sequent software agreements includes a specific provision requiring the licensee to hold the “SOFTWARE PRODUCTS subject to this Agreement in confidence for AT&T.” 15 IBM Agreement (2/1/85) § 7.06 (Exh. 2); Sequent Agreement (4/18/85) § 7.06 (Exh. 1); IBM Side

14

Moreover, IBM’s reading of the contractual treatment of “SOFTWARE PRODUCT” under the software agreements runs into further conflict with Section 7.06(b) of the IBM and Sequent agreements, which states that “LICENSEE may distribute copies of a SOFTWARE PRODUCT, either in modified or unmodified form, to third parties having licenses of equivalent scope herewith from AT&T (or a corporate affiliate thereof) for the same SOFTWARE PRODUCT, provided that LICENSEE first verifies the status of any such third party in accordance with specific instructions issued by AT&T.” IBM Agreement (2/1/85) § 7.06 (Exh. 2) (emphasis added); Sequent Agreement (4/18/85) § 7.06 (Exh. 1). That the agreements expressly recognize that “SOFTWARE PRODUCT” covered by the agreements may be in “modified or unmodified form” further supports SCO’s reading of those agreements to cover not only the source code, but also any UNIX modifications and derivatives. This language was included in the original IBM and Sequent software agreements, as well as in the IBM Side Letter.

15

57

Letter (2/1/85) (Exh. 3). The Sequent agreement further specifies that the licensee “shall not make any disclosure of any or all of such SOFTWARE PRODUCTS (including methods or concepts utilized therein) to anyone, except to employees of LICENSEE to whom such disclosure is necessary to the use for which rights are granted hereunder.” Id. (emphasis added). 16 Section 7.06(a) of the software agreements thus not only expressly extended the requirement of confidentiality to all aspects of the “SOFTWARE PRODUCTS” (which, by virtue of Section 2.01, must be treated identically to any modifications or derivatives based on such SOFTWARE PRODUCTS), but also specify (in the case of the Sequent agreement) that protection of the SOFTWARE PRODUCTS includes protection of the “methods or concepts utilized therein.” Id. Again, a plain reading of these provisions is directly at odds with IBM’s tortured construction. 3. IBM’s “Literal Source Code Only” Reading Conflicts with the Restrictions Imposed by the UNIX Sublicensing Agreements.

Beyond the plain language of Sections 2.01 and 7.06, further textual evidence undermining IBM’s “literal source code only” theory is found in the plain language of the UNIX sublicense agreements that AT&T executed with IBM (contemporaneously with its software agreement and side letter in February 1985) and Sequent (in January 1986, more than nine months after Sequent executed its software agreement).

16

The IBM Side Letter amended the language of the standard software agreement by removing “all parts of the” before the above-quoted “SOFTWARE PRODUCTS subject to this Agreement” language and by replacing the “methods or concepts” clause with an alternative formulation of IBM’s obligations under Section 7.06(a). IBM Side Letter (2/1/85) (Exh 3). The significance - or, more accurately, lack of significance - of those contractual amendments to IBM’s current motion is addressed in Parts I.A.4. & I.B.2.c., below.

58

Under those separate, but fully integrated, 17 sublicensing agreements, IBM and Sequent were permitted to distribute “SUBLICENSED PRODUCTS” - defined to include “COMPUTER PROGRAMS in object-code format based on a SOFTWARE PRODUCT subject to the software Agreement,” IBM Sublicensing Agreement (2/1/85) ¶ 1.04 (Exh.6); Sequent Sublicensing Agreement (4/18/85) § 1.04 (Exh. 54) - under certain conditions and in exchange for sublicensing fees and royalties. Id. § 4.01(a). Thus, under the sublicensing agreements, AT&T licensees were permitted to distribute, for a fee, products “based on a SOFTWARE PRODUCT subject to the Software Agreement” (which, pursuant to Section 2.01 of the software agreements, included any modifications or derivatives of the SOFTWARE PRODUCT), but only in object code format - the binary format that is not readable by human beings and, therefore, does not disclose any UNIX intellectual property. Significantly, the plain terms of the sublicensing agreements contain no suggestion or requirement that such “SUBLICENSED PRODUCTS” had to be based on the literal UNIX source code; in fact, the sublicensing agreements do not contain any language whatsoever on which IBM could even argue that the agreements limited the coverage of “SUBLICENSED PRODUCTS” to those “based on” the literal UNIX source code. In light of this plain reading of the sublicensing agreements, IBM’s “literal source code only” reading of the software agreements makes no sense as a matter of contract construction. Under IBM’s view of the agreements, a software licensee is free to distribute any product that is “based on” the UNIX operating system within the meaning of Section 2.01 - but that does not contain any literal copying of UNIX source code - in unprotected source-code format without charging, any sublicensing fees. But under the sublicensing agreements, that same product could

17

The sublicense agreements are integrated with the software agreements, see IBM Sublicensing Agreement (2/1/85) ¶¶ 3-4 (Exh. 6); Sequent Sublicensing Agreement (1/28/86) ¶¶ 3-4 (Exh. 54), and are thus required to be read together with the provisions of the software agreements.

59

only be distributed in object code format and at a charge to the licensee. IBM would thus have the sublicensing agreements require licensees to pay for distributions in restricted object code of the very same products that the licensee could distribute in unrestricted source code without charge under the software agreements. Accordingly, consideration of the interplay between the UNIX software agreements and the plain language of the sublicensing agreements further confirms what the plain text of the software agreements already establish: AT&T sought to protect its UNIX asset by restricting use and disclosure of products based on the UNIX intellectual property, whether or not such products contained literally copied UNIX source code. 4. IBM’s Reliance on the IBM Side Letter Is Misplaced.

Contrary to IBM’s claim, the Side Letter that it entered into with IBM (but not Sequent) neither nullified the plain language of Sections 2.01 and 7.06 nor otherwise reduced the contractual obligations (of IBM, Sequent, or the hundreds of other UNIX licensees) to the protection of only the literal UNIX source code. While IBM attempts to argue that its Side Letter accomplished this dramatic change merely by “clarifying” the agreements, IBM Mem. at 54, IBM’s attempted use of that side letter is misguided for several reasons. First, IBM ignores entirely that the plain language of the provision on which it relies concerned the ownership of derivatives and modifications, and not the licensor’s contractual protections concerning their use and disclosure: “Regarding Section 2.01, we agree that modifications and derivative works prepared by or for you are owned by you. However, ownership of any portion or portions of SOFTWARE PRODUCTS included in any such modification or derivative work remains with us.” IBM Side Letter (2/1/85) at 2 (Exh. 3) (emphasis added). Plainly, this language did not compromise the licensor’s right to control the use and disclosure, among other things, of such modifications or derivative works, as clearly set
60

forth in the software agreements - particularly, Sections 2.01 and 7.06(a). As the licensor, AT&T acknowledged in the IBM Side Letter that it would not claim ownership of modifications or derivative works that had been prepared by a licensee, but it never agreed to give away its right to protect its intellectual property by ensuring that licensees would not use or disclose works that were copied (even non-literally) from its innovations. Second, IBM’s suggestion that the “Regarding 2.01” clause of the IBM Side Letter somehow “clarified” out of existence most of the contractual protections afforded by the UNIX software agreements is incomprehensible. A contractual provision is not clarified by being rendered meaningless. As discussed above, Section 2.01 plainly covers far more than mere literal source code, and, for the reasons previously discussed, IBM’s “literal source code only” reading of Section 2.01 renders both superfluous and illogical the express contractual protection requiring “resulting materials” to be “treated as part of the original SOFTWARE PRODUCT”. Third, if the intent of the IBM Side Letter was to cut back the Section 2.01 protections to mere literal source code, as IBM contends, that would have been exceedingly easy to accomplish as a drafting matter. Instead of leaving the “resulting materials” clause of Section 2.01 entirely in place, the IBM Side Letter could have simply added the word “not” before “treated as part of the original SOFTWARE PRODUCT.” With that simple change, IBM could have effected the very change that it now reads into the side letter - economically, easily, and clearly. Without such a change, however, the plain language of the software agreements precludes the retroactive change that IBM is attempting to make. The language of the IBM Side Letter plainly addresses the ownership of modifications and derivative works, and does not in any way compromise the strict prohibitions on improper use and disclosure of such works. Ownership and use rights are not synonymous in the law, and

61

property rights - particularly intellectual-property rights - are often bifurcated along those lines. See, e.g., HotSamba, Inc. v. Caterpillar Inc., No. 01 C 5540, 2004 WL 609797, at *5 (N.D. Ill. Mar. 25, 2004) (discussing software license agreement that allowed licensee to create derivative works and provided that licensee owned any derivative works it created, but prohibited licensee from transferring any such derivative works to third parties) (Exh. B). Contractual language addressing ownership should not be understood to define the contracting parties’ rights with respect to use, particularly where, as here, the contract unambiguously distinguishes those rights, and addresses them separately and differently. IBM also misinterprets the IBM Side Letter when it suggests that the amendment to the language of Section 7.06(a) somehow “clarified” all of the licensor’s contractual protections out of existence. 18 Nothing in this part of the letter modifies, or purports to modify, any of the restrictions contained in Section 2.01. In addition, the IBM Side Letter specifically reiterated that IBM was required to accord confidential treatment to all SOFTWARE PRODUCTS including, pursuant to Section 2.01, all modifications and derivatives that were “based on” such SOFTWARE PRODUCTS. IBM Side Letter (2/1/85) at 4 (Exh. 3).

18

The IBM Side Letter Agreement amended Section 7.06(a) of the IBM software agreement to read, in relevant part, as follows: “LICENSEE agrees that it shall hold SOFTWARE PRODUCTS subject to this Agreement in confidence for AT&T. LICENSEE further agrees that it shall not make any disclosure of such SOFTWARE PRODUCTS to anyone, except to employees of LICENSEE to whom such disclosure is necessary to the use for which rights are granted hereunder. LICENSEE shall appropriately notify each employee to whom any such disclosure is made that such disclosure is made in confidence and shall be kept in confidence by such employee. Nothing in this agreement shall prevent LICENSEE from developing or marketing products or services employing ideas, concepts, know-how or techniques relating to data processing embodied in SOFTWARE PRODUCTS subject to this Agreement, provided that LICENSEE shall not copy any code from such SOFTWARE PRODUCTS into any such product or in connection with any such service and employees of LICENSEE shall not refer to the physical documents and materials comprising SOFTWARE PRODUCTS subject to this Agreement when they are developing any such products or service or providing any such service.” IBM Side Letter (2/1/85) at 4 (Exh. 3).

62

Indeed, although the IBM Side Letter would “Amend Section 7.06(a)” of AT&T’s standard agreement (i.e., the agreement that it entered into with Sequent) to permit IBM to engage in “developing and marketing products or services employing ideas, concepts, know-how or techniques relating to data processing embodied in SOFTWARE PRODUCTS”), it expressly conditioned this permission on, among other things, IBM’s not “copy[ing] any code from such SOFTWARE PRODUCTS into any such product or in connection with any such service.” Id. In light of the fact that Section 2.01 required UNIX modifications and derivatives to be treated as if they were part of the “SOFTWARE PRODUCTS,” even IBM’s Side Letter still required IBM to accord confidential treatment to the code contained in UNIX modifications and derivatives. Thus, there is no merit to IBM’s suggestion that the IBM Side Letter somehow unambiguously removes, even for just IBM, the software agreements’ use and disclosure restrictions on UNIX modifications and derivatives, and somehow cuts back the protections under those agreements to cover only the literal copying of UNIX source code. Finally, the IBM Side Letter would not entitle it to summary judgment for the independently sufficient reason that the side letter did not cover Sequent. In fact, although the Sequent software agreement was entered into just a couple of months after the IBM agreement and side letter, none of the language on which IBM (incorrectly) relies from its side letter was included in the Sequent agreement. And there is no textual support whatsoever for IBM’s suggestion that Sequent and other licensees somehow received the benefit of amendments to IBM’s software agreement even though no such language had been included in their written contracts. IBM Mem. at 53-54.

63

To the contrary, each of the UNIX software agreements (including the Sequent agreement) contained clear and unambiguous merger language that specifically ruled out afterthe-fact “me too” arguments such as those advanced here by IBM (on Sequent’s behalf): “This Agreement and its Supplements set forth the entire agreement and understanding between the parties as to the subject matter hereof and merge all prior discussions between them, and neither of the parties shall be bound by any conditions, definitions, warranties, understandings or representations with respect to such subject matter other than as expressly provided herein or as duly set forth on or subsequent to the date of acceptance hereof in writing and signed by a proper and duly authorized representative of the party to be bound thereby. No provision appearing on any form originated by LICENSEE shall be applicable unless such provision is expressly accepted in writing by an authorized representative of AT&T.” IBM Agreement (2/1/85) ¶ 4 (Exh. 2) (emphasis added); Sequent Agreement (4/18185) ¶ 4 (Exh. 1) (emphasis added). IBM ignores this merger clause in its strained attempt to export its misreading of the IBM Side Letter over to the Sequent software agreement. 19 Accordingly, there is no merit to IBM’s claim that it is entitled to summary judgment on its contract-interpretation argument as a matter of law. To the contrary, the Court should grant summary judgment for SCO on its plain-reading interpretation of the software agreements at issue. B. IBM’s Appeal to Extrinsic Evidence for Its Summary Judgment Motion Is Meritless As a Matter of Law and Fact IBM also improperly attempts to base its summary judgment on extrinsic evidence. IBM Mem. at 56-67. As an initial matter, “a court may not admit extrinsic evidence in order to determine the meaning of an unambiguous contract.” Omni Quartz, Ltd. v. CVS Corp., 287 F.3d 61, 64 (2d Cir. 2002); R/S Assocs. v. N.Y. Job Dev’t Auth., 98 N.Y.2d 29, 33 (2002) (declining to consider extrinsic evidence where contract was clear and unambiguous); accord Gamble,

19

To support its claim that the IBM Side Letter bears on the Sequent agreement, IBM even goes so far as to rely on contract language that was not included in either of the IBM or Sequent agreements. See IBM Mem. at 54 (citing IBM Exhibit 39, a software agreement between AT&T and The Santa Cruz Operation).

64

Simmons & Co. v. Kerr-McGee Corp., 175 F.3d 762, 767 (10th Cir. 1999) (“If the contract is unambiguous its language is the only legitimate evidence of what the parties intended, and we will not rely on extrinsic evidence to vary or alter the plain meaning.”). Thus, in light of the plain reading of the software agreements set forth above, IBM’s appeal to extrinsic evidence should be rejected outright. In addition, even if the Court were to conclude that relevant language in the operative agreements is ambiguous, IBM still would not be entitled to summary judgment because (1) under controlling authority, any contractual ambiguity that must be resolved by resort to extrinsic evidence necessarily gives rise to a jury question and precludes a resolution of the ambiguity on summary judgment; and (2) the sharply disputed extrinsic evidence in this case could not possibly support IBM’s requested summary judgment. 1. The Extrinsic Evidence Cannot Entitle IBM to Summary Judgment on the Meaning of Ambiguous Agreements

Courts applying New York law, as well as the Tenth Circuit, 20 have repeatedly held that “in a contract dispute, summary judgment may be granted only where the language of the contract is unambiguous.” Nowak v. Ironworkers Local 6 Pension Fund, 81 F.3d 1182, 1192 (2d Cir. 1996). Thus, under well-settled, controlling authority, courts may not resort to extrinsic evidence in order to resolve contractual ambiguities as a matter of law. See State v. Home Indem., 66 N.Y.2d 669, 671 (1985) (holding that “the resolution of the ambiguity is for the trier of fact”); Seiden Assocs., Inc. v. ANC Holdings, Inc., 959 F.2d 425, 428 (2d Cir. 1992) (holding

20

IBM asserts that New York law governs the Court’s interpretation of the license agreements. IBM Mem, at 49 n.7. The argument implicates conflict-of-law analysis, but raises a false conflict, because as the above-cited cases demonstrate, the result will be the same whether the Court is guided by New York contract law or Tenth Circuit precedent (which applies state law to decide the issue and therefore also points to New York law). Moreover, it is not proper under federal procedural law to use extrinsic evidence to resolve a contractual ambiguity on summary judgment. Coplay Cement Co., Inc. v. Willis & Paul Group, 983 F.2d 1435, 1438 (7th Cir. 1993).

65

that “the meaning of the words becomes an issue of fact and summary judgment is inappropriate”); Hartford Acc. & Indem. Co. v. Wesolowski, 350 N.Y.S.2d 895, 898 (1973) (if the “determination of the intent of the parties depends on the credibility of extrinsic evidence or on a choice among reasonable inferences to be drawn from extrinsic evidence, then such determination is to be made by the jury”); 21 accord Gomez v. Am. Elec. Power Serv. Corp., 726 F.2d 649, 651 (10th Cir. 1984) (holding that “in an ambiguous contract, if the intent of the parties is disputed, a genuine issue of material fact exists which cannot be determined summarily by the court”); Met. Paving Co. v. City of Aurora, Colo., 449 F.2d 177, 181 (10th Cir. 1971) (contractual ambiguities must be resolved by the finder of fact). This rule reflects the Court’s obligation under Rule 56 to construe all evidence on a motion for summary judgment - including ambiguous contractual terms - in the light most favorable to the non-moving party. Thus, if a contract is ambiguous, there necessarily exists a genuine issue of fact concerning the parties’ intended meaning: “If the court must resort to extrinsic evidence to ascertain the correct and intended meaning, of a term, material questions of fact necessarily exist. If the language is susceptible to different reasonable interpretations, and where there is relevant extrinsic evidence of the parties’ actual intent, then the contract’s meaning becomes an issue of fact precluding summary judgment.” Alexander & Alexander Servs., Inc. v. These Certain Underwriters at Lloyd’s, London, 136 F.3d 82 (2d Cir. 1998) (emphasis added). Because ascertaining the meaning of an ambiguous contract necessarily involves drawing inferences from, and determining the credibility of, extrinsic evidence, see Hartford, 350 N.Y.S.2d at 898, permitting a court to resolve such matters on a motion for summary judgment

21

See also Bouzo v. Citibank, 96 F.3d 51, 58 (2d Cir. 1996) (“We have repeatedly held that, in a contract dispute, summary judgment may be granted only where the language of the contract is unambiguous.”); Ruttenberg v. Data Davidge Sys. Corp., 215 A.D.2d 191, 19.3 (N.Y. App. Div. 1995) (“When the meaning of the contract is ambiguous and the intent of the parties becomes a matter of inquiry, a question of fact is presented which cannot be resolved on a motion for summary judgment.”) (internal quotation and citation omitted).

66

“would greatly curtail the prerogatives of juries and trial judges.” Coplay Cement Co. v. Willis & Paul Group, 983 F.2d 1435, 1439 (7th Cir. 1993) (Posner, J.) (recognizing that “to infer meaning from disparate bits of evidence, some textual, some testimonial, none contested in themselves but the aggregate forming a confused mosaic” is “a task more appropriate for a trier of fact”). In arguing that courts “routinely” rely on extrinsic evidence to resolve contractual ambiguity on summary judgment, IBM Mem. at 62, IBM fails entirely to address this controlling authority. Instead, IBM first cites to three cases from other jurisdictions. 22 It is equally true that even other jurisdictions “routinely” will not consider extrinsic evidence to resolve contractual ambiguity on summary judgment. 23 Moreover, the Tenth Circuit cases on which IBM relies are plainly inapposite. See IBM Mem. at 56, 66 (citing Moncrief v. Williston Basin Interstate Pipeline Co., 174 F.3d 1150 (10th Cir. 1999); Kaiser-Francis Oil Co. v. Producer’s Gas Co., 870 F.2d 563 (10th Cir. 1989)). In Moncrief, the district court had completed a full trial on damages, enabling the Tenth Circuit to review a complete factual record on liability. See 174 F.3d at 1173. After assigning error to the district court’s finding that the contract was unambiguous, the appellate court declined to remand

22

United Rentals (No. Am.), Inc. v. Keizer, 355 F.3d 399 (6th Cir. 2004) (applying Michigan law); Farmland Indus., Inc. v. Grain Bd. of Iraq, 904 F.2d 732 (D.C. Cir. 1990) (applying District of Columbia law); Goodman v. Resolution Trust Corp., 7 F.3d 1123 (4th Cir. 1993) (applying Maryland law). See, e.g., Emerson Radio Corp. v. Orion Sales, Inc., 253 F.3d 159, 163-64 (3d Cir. 2001) (“It is hornbook law that if the relevant terms in a contract are ambiguous, the issue must go to a jury.”); Coplay, 983 F.2d at 1438 (“if other evidence needs to be consulted interpretation is for the trier of fact”) (citations omitted); Liberty Mut. Ins. Co. v. Pine Bluff Sand & Gravel Co., 89 F.3d 243, 247 (5th Cir. 1996) (stating district court erred by granting summary judgment on ambiguous contract); Am. Fid. & Cas. Co. v. London & Edinburgh Ins. Co., 354 F.2d 214, 216-17 (4th Cir. 1965) (“since the existence of ambiguous terms creates an issue of fact the granting of summary judgment for either side is improper”); see also e.g., Restatement (Second) of Contracts § 212(2) (1981) (“A question of interpretation of an integrated agreement is to be determined by the trier of fact if it depends on the credibility of extrinsic evidence or on a choice among reasonable inferences to be drawn from extrinsic evidence.”); Wright, Miller & Kane, Federal Practice and Procedure: Civil 2d § 2730.1, at 2737.5 (2d ed. 1983); E. Allan Farnsworth, Contracts § 7.14, at 538-39 (2d ed. 1990).

23

67

the case and instead resolved the ambiguity on its own. In so doing, the court expressly relied on the “unusual procedural posture” of the case. See id. (“Because of the unusual procedural posture of this issue - an issue decided on summary judgment, yet about which the parties introduced evidence and testimony at trial - we have before us all of the arguments and evidence relevant.”). 24 Thus, although the appeals court in Moncrief acted as the fact finder “in the interest of judicial economy,” see id. at 1154-55, 1174, that case cannot be read to set forth a district court’s authority to resolve contractual ambiguities as a matter of law based on extrinsic evidence - much less extrinsic evidence that has not been fully developed. 25 Kaiser-Francis did not even involve the resolution of an ambiguity through the use of any extrinsic evidence; the case involved a suit to enforce a take-or-pay provision in a sales contract. The Tenth Circuit first affirmed the district court’s finding that the contract was not ambiguous, see 870 F.2d 565-66, and then, as an independent matter, ruled that the district court had properly granted summary judgment on the issue of the seller’s insecurity regarding the buyer’s performance, id. at 568-69. As with Moncrief, IBM has selectively lifted language from the opinion, ignoring the context that makes it plain that the case in no way departs from the settled

24

IBM’s quotation from Moncrief omits the court’s acknowledgement that the sole basis for its ruling was the “unusual procedural posture of this case”; recognized that “in most such cases” the court “must remand. .. for findings regarding the interpretation of the ambiguous contract”; and prefaced the very language from the opinion on which IBM relies with the express caveat that in that case “all the extrinsic evidence [was] in.” 174 F.3d at 1173, 1174. IBM cites (at page 66 of its brief) a similar case from the Second Circuit, Compagnie Financiere de CIC et de L’Union Europeenne v. Merrill Lynch, 232 F.3d 153 (2d Cir. 2000), which also rested on a unique procedural posture. In that case, the court emphasized that, because the summary judgment motion at issue “followed a bench trial, the trial record was before it.” Id. at 161. The court also observed that neither party had suggested the existence of additional evidence not already submitted at trial: “The parties confirmed at oral argument that the record was complete with regard to extrinsic evidence supporting their interpretations.” Id. In any event, unlike the New York state and Tenth Circuit cases cited above, cases such as Compagnie have no binding effect in this case.

25

68

Tenth Circuit law, discussed above, that (like New York law) prohibits a court from resolving contractual ambiguities on summary judgment by resort to extrinsic evidence. 26 2. Numerous Material Questions of Fact Exist Concerning the Intent of the Parties to the UNIX Software agreements

If even relevant to IBM’s motion, the extrinsic evidence of the parties’ intent is sharply disputed and leaves unresolved many questions of fact central to IBM’s motion. First, the very declarants on whom IBM relies in its motion have offered conflicting and inconsistent testimony that directly undermines IBM’s “literal source code only” theory. For example, while IBM relies on a declaration that former AT&T account executive Ira Kistenberg executed in June, Mr. Kistenberg’s more recent sworn statement, dated November 12, contradicts BM’s position. SCO has also had the opportunity thus far in discovery to depose four of the declarants on whom IBM heavily relies in its motion - Otis Wilson, David Frasure, Geoffrey Green, and David Rodgers. In each of those depositions, the witnesses contradicted the

26

In addition, IBM’s reliance on the doctrine of contra proferentem - which IBM claims makes summary judgment “especially appropriate in this case” - is entirely misplaced. IBM Mem. at 66-67. Contra proferentem “is applied only in cases where there is an ambiguity in some terminology of the policy [contract] which cannot be resolved by the presentation of extrinsic evidence.” Alfin, Inc. v. Pac. Ins. Co., 735 F. Supp. 115, 121 n.5 (S.D.N.Y. 1990). Accordingly, courts have rejected the application of the doctrine in the summary judgment context. See, e.g., Schering Corp. v. Home Ins. Co., 712 F.2d 4, 9-10, (2d Cir. 1983) (refusing to apply contra proferentem because “the party opposing summary judgment propound[ed] a reasonable conflicting interpretation of a material disputed fact”). Moreover, contra proferentem applies where one of the contracting parties had no voice in the selection of the contract language, and thereby protects the weaker of two contracting parties with unequal bargaining power. See 67 Wall St. Co. v. Franklin Nat’l Bank, 37 N.Y.2d 245, 249 (1975) (holding that contra proferentem benefits parties “who had no voice in the selection of [the agreement’s] language”); Hertzog. Calamari & Gleason v. Prudential Ins. Co. of Am., 933 F. Supp. 254, 2.5859 (S.D.N.Y. 1996) (noting that “the parties were in parity with each other” and thus “the contra proferentem maxim has no proper office to perform.”). The cases IBM cites in support of its contra proferentem argument make this very point. In Jacobson v. Sassower, 66 N.Y.2d 991, 993 (1985), the court noted that an ambiguous contract must be construed against the drafter “and favorably to a party who had no voice in the selection of its language,” and emphasized the special concerns governing attorney fee arrangements. (emphasis added). In Westchester Resco Co., L.P. v. New England Reinsurance Corp., 818 F.2d 2 (2d Cir. 1987), the court emphasized the particular rule in New York law that ambiguous insurance policies are to be construed against the insurer. Id. at 3. None of these concerns exist in this case, which indisputably involved arms-length deals between sophisticated and well-represented parties.

69

arguments that IBM now advances in reliance on the witnesses’ earlier declarations. 27 And the prior inconsistent testimony of Otis Wilson and David Frasure concerning substantially similar contractual provisions - given in the UNIX System Laboratories, Inc. v. Berkeley Software Design, Inc. (“BSD”) case, some twelve years closer to the events in question and before IBM secured their legal representation in this case - further contradicts the declarations that IBM has submitted from those witnesses. 28 This inconsistent testimony alone is sufficient to defeat summary judgment, not only because it shows IBM’s position to be incorrect and raises disputes as to genuine issues of material fact, but also because it undercuts the credibility of IBM’s declarants. The extrinsic evidence would necessarily require the Court to make credibility determinations, draw inferences, and weigh the evidence - functions reserved for the finder of fact. See, e.g., Miller v. Wolpoff & Abramson, L.L.P., 321 F.3d 292, 307 (2d Cir. 2003) (reversing grant of summary judgment because the plaintiff had not had opportunity to discover facts relevant to credibility of defendants’ affiants); Cameron v. Frances Slocum Bank & Trust Co., 824 F.2d 570, 575 (7th Cir. 1987) (“‘Clearly, if the credibility of the movant’s witnesses is challenged by the opposing party and specific bases for possible impeachment are shown, summary judgment should be denied and the case allowed to proceed to trial.”‘) (quoting 10A C. Wright et al., Federal Practice and Procedure § 2726, at 115-16 (2d ed. 1983)).

27

Aside from his prior inconsistent testimony, Mr. Frasure’s credibility is clouded by his recent felony conviction for insurance fraud, a serious crime of dishonesty. Frasure Dep. (1218/92) at 116-19 (Exh. 19). In the BSD case, Mr. Wilson and Mr. Frasure testified about a UNIX educational license agreement that included intellectual property protections nearly identical to those at issue in this case. See The Regents of the University of California Agreement (11/12/85) §§ 2.01, 7.06 (Exh. 13); IBM Agreement (2/1/85) §§ 2.01, 7.06 (Exh. 2); Sequent Agreement (4/18/8.5) §§ 2.01, 7.06 (Exh. 1); Wilson Dep. (12/10/92) at 58, 74 (Exh. 12). Indeed, all three types of UNIX software agreements - educational, commercial and governmental - were designed to provide the same core intellectual property protections. See, e.g., Kistenberg Decl. (11/12/04) ¶ 4 (Exh. 10); Green Dep. (11/15/04) at 68, 240 (Exh. 5); Wilson Dep.. (12/10/92) at 57 (Exh. 12); Pfeffer Decl. (9/7/04) ¶ 5 (Exh. 8); Bond Decl. (11/4/04) ¶ 10 (Exh. 14); Davis Decl. (11/4/04) ¶ 7 (Exh. 4).

28

70

Second, the sworn declarations submitted herewith – from former AT&T attorneys Martin Pfeffer and Burt Levine as well as former AT&T contract managers Mitzi Bond and Evelyn Davis – further undermine IBM’s argument and support the plain reading of the software agreements set forth above. Mr. Pfeffer and Mr. Levine were heavily involved in the drafting and enforcement of the standard UNIX software agreements, 29 and Ms Bond and Ms Davis were charged (together with others, such as Mr. Kistenberg) with communicating the terms of the standard agreements to licensees and prospective licensees. 30 Third, contemporaneous course-of-dealing evidence, including documents executed by the very declarants on whom IBM relies confirms the breadth of the software agreements’ protections and contradicts the current position of IBM and its declarants.

29

Indeed, not only did Mr. Pfeffer have supervisory authority over the attorneys in AT&T’s UNIX licensing department, including Burt Levine and Geoff Green (the one attorney upon whose declaration IBM does rely), but Mr. Pfeffer also had “ultimate approval” of “the documents when they were revised.” Frasure Dep. (6/8/04) at 177 (Exh. 20); see also Pfeffer Decl. (9/7/04) ¶ 4 (Exh. 8); Levine Decl. (11/23/04) ¶ 4 (Exh. 7). Various members of AT&T’s UNIX licensing department- including IBM’s declarants - came to their understanding of the standard software agreement protections based on training from the AT&T attorneys. See, e.g., Frasure Dep. (6/8/04) at 163-64, 171 (Exh. 20) (stating that “my understanding comes from various conversations with the attorneys that work with our organization” explaining that this was due in part to the obvious reason that there were “legal concepts and issues” at the center of those agreements); Bond Decl. (11/4104) ¶ 6 (Exh. 14) (identifying Marty Pfeffer and Burt Levine as the licensing group’s “principal legal contacts”); Davis Decl. (11/4/04) ¶ 4 (Exh. 4) (same). Courts routinely consider testimony from witnesses who were involved in drafting standard contract language, even if they did not sign or were not involved in the negotiation of the particular contract at issue. See, e.g., Schering v. Home Ins. Co., 712 F.2d 4, 9-10 (2d Cir. 1983) (denying plaintiff’s motion for summary judgment solely on basis of affidavits of two drafters of standard contractual clause); see also McDonnell Douglas Fin. Corp. v. Pa. Power & Light Co., 858 F.2d 825, 833 (2d Cit. 1988) (considering affidavits of two attorneys at contracting party’s outside law firm who were not involved in drafting contract at issue but had “worked to develop similar clauses in similar transactions”). This evidence is relevant if it provides context for the communications and writings exchanged between the parties. See Nat’l Envmtl. Serv. Co. v. Ronan Eng’g Co., 256 F.3d 995, 1003 (10th Cir. 2001). Thus, a party’s standard practice is relevant to a determination of that party’s intent in adhering to or deviating from that practice in a particular case. Accord EPN Ingenieria S.A. de C.V. v. Gen. Elec. Co., No. 92 Civ. 1563 (KMW), 1996 WL 531867 (S.D.N.Y. Sept. 19, 1996) (considering defendant’s standard contract, which plaintiff signed but defendant did not agree to in this case, as evidence of defendant’s intent) (Exh. C); RMLS Metals, Inc. v. Int’l Bus. Mach. Corp., 874 F. Supp. 74, 77 (S.D.N.Y. 1995) (considering defendant’s standard contract that was not adopted by the parties as evidence of defendant’s intent). IBM’s stated position that the UNIX software agreements were generally intended to contain the same core intellectual property protections only reinforces the plain relevance of the testimony from the attorneys charged with the drafting and enforcement of the standard software agreements. See IBM Mem. at 54, 64 & ¶ 123.c

30

71

The more detailed examination of the extrinsic evidence below confirms that SCO’s plain reading of the contractual language is correct and that, at a minimum, there are many issues of material fact that preclude summary judgment for IBM. a. The software agreements protect against more than just literal copying of UNIX source code.

IBM relies on a lengthy presentation of statements from witness declarations in an attempt to support its audacious claim that the software agreements protect against only the literal copying of UNIX source code. IBM Mem. ¶¶ 79-99, pp. 57-62. These declarations not only run into direct conflict with the plain language of the operative software agreements, see Part I.A.1 above, but also contradict ample other record evidence in this case - including testimony from the very same declarants on whom IBM relies. Thus, IBM’s extrinsic evidence on this pivotal point in its motion is neither undisputed nor credible. First, the testimony of IBM declarant Ira Kistenberg, the AT&T account executive who negotiated Sequent’s UNIX license, highlights the disputed extrinsic evidence concerning the contract-interpretation issue in this case. Although IBM attempts to rely on Mr. Kistenberg’s June 24 declaration for its “literal source code only” argument, IBM ¶ 80 (citing Kistenberg Decl. (6/24/04) ¶ 22 (Exh. 9)), Mr. Kistenberg’s more recent sworn declaration plainly contradicts IBM’s position. In his declaration of November 12, Mr. Kistenberg specifically identifies three distinct categories of intellectual property – in addition to the literal UNIX source code – that AT&T intended to protect through its standard software agreements (including the Sequent agreement): “AT&T intended that protected modifications and derivatives would include any product that contained any source code that had been copied verbatim from UNIX System Five; any copied source code that was similar in substance to the original source code in UNIX System V; any structures, sequences, patterns, ideas, methods or concepts from UNIX System V; and any source code that the licensee developed with the benefit of exposure to the UNIX System V source code.”
72

Kistenberg Decl. (11/12/04) ¶ 5 (Exh. 10). Second, while IBM relies on the declaration of David Frasure to support its position, IBM ¶¶ 83, 88 (citing Frasure Decl. (3/28/04) ¶¶ 12, 14 (Exh. 18)), Mr. Frasure’s prior testimony in the BSD case supports the opposite position. In that testimony, Mr. Frasure actually corrected the very mistake that IBM makes in this case, in response to a question from BSD’s counsel: “Q: Would that - could you tell me the components of what your understanding in May ‘85 would have been included in the licensed software? If you recall what your understanding was in 1985. Licensed software would have been … Would it have been the source code? … would have been the source code or some portion of it, with modifications made to it by the licensee. Where do - where do you get the understanding that a modification to the source code falls within the definition of licensed software? It says licensee agrees that any modifications or derivative work prepared by it that contains any licensed software … Okay. … shall be treated … Okay. … as licensed software.” Frasure Dep. (12/8/92) at 146 (Exh. 19) (alterations in original).

A: Q: A: Q: A: Q: A: Q: A:

Mr. Frasure also testified in the BSD case that AT&T intended to protect the entire universe of products that could be created as a result of exposure to UNIX, regardless of whether the resulting product actually contained the UNIX source code: “if you’ve got a group of people over here developing a set of source code and they have never seen the Unix source code, they’ve never been too exposed to it and the [sic] develop a product completely on their own, then that’s one thing. But if they’re developing a product with the benefit of Unix or perhaps they have used it for - for a number of years, ten years, and then they think they’re going to go off and develop something on their own that’s an operating system that may look like Unix, we had - we expressed our concern that - that we had an interest in that product.” Id. at 20 (emphasis added). Mr. Frasure recalled a meeting in late 1984 – roughly contemporaneously with the signing of the IBM and Sequent agreements – in which he and two other IBM declarants, Otis

73

Wilson and Geoff Green, explained to another AT&T licensee that AT&T’s “exposure” concept was intended to cover products even if they did not contain the UNIX source code: “Q: A: And if your code was not part of the product? Well, if it was used as part of the development – I really need to be careful here on words, I guess. If the source code, the Unix source code was was required, was used to generate the enhancement, was required to have the – the rest of the enhancements work, then we had an interest in it. It’s been a long time. I’m not sure of the right - the right key words to use, but we went through those discussions with them and what we felt the, you know, the agreement said. We also discussed contractor provisions which allowed a licensee to contract with someone to develop software and then when that development was done everything had to come back to them and we expressed concern I guess with – Otis [Wilson] and I used the term mental contamination, that if you had been exposed to the source code and its methods and concepts, even though you give something back to the the licensee, there was - there was concern there that someone could go off on their own and develop what they thought was their own product but really using the methods and concepts and techniques that were in the product that they had previously used.” Id. at 18-19 (emphasis added).

Third, IBM’s reliance on a recent declaration from Mr. Wilson, see IBM ¶¶ 82, 86 (citing Wilson Decl. (4/26/04) ¶¶ 12, 14 (Exh. 11)), is similarly undermined by Mr. Wilson’s prior BSD testimony, in which he affirmatively rejected IBM’s restrictive interpretation of the AT&T software agreements. Just like Mr. Frasure, Mr. Wilson invoked “exposure”: “You said if they had developed some software with exposure to the licensed software would it be subject to the AT&T agreement. I’m saying absolutely that would be the case. That’s why they always used both. They were careful if they had any exposure to the licensed software, it was a given that you had to have a license from AT&T.” Id. at 127-128 (emphasis added). Moreover, in letters that Mr. Wilson wrote to licensees – including to IBM on July 27, 1987, he clearly stated that AT&T intended for its software agreements to cover not only “any licensed source code software product (either in modified or unmodified form)” but also to prohibit them from distributing “sublicensed products based on such software products or its derivatives, unless specifically permitted by the license agreements.” Exh. 15 (emphasis added); see also Exh. 16 (similar letter from Mr. Wilson to Sequent).
74

Fourth, Geoff Green, the only licensing attorney on whom IBM even attempts to rely, IBM ¶ 80 (citing Green Decl. (6/17/04) ¶ 6 (Exh. 22)), squarely contradicted IBM’s position at his deposition. Mr. Green first confirmed that modifications and derivatives were the “resulting materials” referred to in Section 2.01 of the software agreements that had to be treated the same as the original “SOFTWARE PRODUCT.” He then testified that copying literal source code was only one way of creating a “derivative work” based on the UNIX product: “Q. A. What was AT&T’s intent with respect to the meaning of the term ‘derivative works’? Something that was based on the licensed product, and that would be considered to probably be in a variation of the product or would somehow include the product or part of the product.” Green Dep. (11/15/04) at 113 (emphasis added) (Exh. 5).

Indeed, Mr. Green admitted that Section 2.01 of AT&T’s standard agreement did not contain any limitation like the one that IBM claims here: “Q. Is there any requirement in the license agreement that such modifications or derivatives have to include literally copied source code from the original product? In this form of the agreement, no.” Id. at 115. 31

A.

When asked later if the 1985 standard UNIX System Five software agreement contained language limiting “the protection for derivatives or modifications” to “those derivatives or modifications that include literally copied source code,” Mr. Green further conceded: “No there isn’t. Yes, that is correct.” Id. at 117-18 (emphasis added). And Mr. Green further agreed that copying literal source code was not the only way for a licensee to create a protected modification or derivative product:

31

At the time he gave this testimony, Mr. Green was examining the Sequent software agreement. With respect to whether AT&T’s intent under the software agreement should be gleaned exclusively from the software agreements themselves, Mr. Green, who maintained that it was his job to “change the language to reflect that intent,” agreed that “the language of the agreement should speak for itself as to - as to what we did to protect the software.” Green Dep. (11/15/04) at 109, 161 (Exh. 5).

75

“Q.

A.

If the first derivatives embodied methods and concepts but not literally copied source code, it needs to be treated under 2.01 as if it were part of the original software product, correct? Right.” Id. at 145 (emphasis added). 32

Fifth, notwithstanding IBM’s attempt to rely on the declaration of David Rodgers, the former Sequent Vice President of Engineering who executed Sequent’s UNIX license, IBM ¶ 79 (citing Rodgers Decl. (11/5/03) ¶ 8 (Exh. 23)), Mr. Rodgers contradicted IBM’s position at his deposition. Mr. Rodgers conceded that the UNIX software agreements did not require any literal UNIX source code to be present in a modification or derivative work in order to be covered, acknowledging that even “completely new source code” would be protected as well. Rodgers Dep. (6/10/04) at 32 (Exh. S1). Mr. Rodgers later reiterated that new code created entirely by Sequent would still be protected under the software agreements, if it was simply based on the UNIX product. Mr. Rodgers explained that only “work which had already been created by Sequent” prior to licensing UNIX and “work that in the future was created by Sequent, not based upon that source code, remained the property of Sequent.” Id. at 27 (emphasis added). Mr. Rodgers admitted that Sequent’s Dynix product was subject to this protection, by later agreeing that “Dynix/ptx is almost certainly a derivative work of Unix System V.” Id. at 138. Sixth, the declaration of AT&T General Attorney Martin Pfeffer, who “participated in, supervised, and approved the legal department’s drafting of the terms of the written license agreements (and any modifications thereto),” Pfeffer Decl. (9/7/04) ¶ 4 (Exh. 8), confirms that “it was AT&T’s intent to prevent through its UNIX license agreements the unauthorized use and

32

With respect to the scope of the software agreements, Mr. Green’s admitted in his deposition that he would have liked more “time to really sit down and analyze the question” he was asked “in terms of everything that there is in the agreements and the history and leading up to it and all the rest of it.” Id. at 228. He conceded that “this is too complicated to come to snap judgments without much more consideration and without knowing much more about the facts of the particular case to be able to come to a conclusion.” Id. at 230.

76

disclosure of more than just literally copied UNIX source code.” Id. ¶ 10. As Mr. Pfeffer explained Section 2.01 of the software agreements as follows: “This language set forth the parties’ intent and agreement that the ‘SOFTWARE PRODUCT’ licensed and protected under the terms of the license agreements included the full content of all of the ‘resulting materials’ created over time from the licensees’ exercise of their contractual ‘right to modify’ and ‘to prepare derivative works’ based on the original licensed material.” Id. ¶ 6 (emphasis added). This protection was designed to protect an entire chain of products derived from the original licensed materials: “Under Section 2.01, if a licensee created a modification or derivative work based on the original licensed product, then the agreement treated the ‘resulting’ work as if it had been part of the original SOFTWARE PRODUCT, and any further modifications or derivatives of that ‘resulting’ work would be treated in the same manner.” Id. Furthermore, Mr. Pfeffer debunks the claim that the software agreements’ protection of more than literal UNIX source code was ever subject to negotiation: “I do not recall any instance during my tenure in which either AT&T or USL agreed (in any license agreement or supplement or modification thereof) to reduce its protection under a UNIX license to prevent the unauthorized use or disclosure of only source code.” Id. ¶ 11 (emphasis added). Any such change would have represented “a significant and material change to the standard terms of AT&T’s license agreement” that would have required Mr. Pfeffer’s, or to a lesser extent Burt Levine’s, approval. 33 Id. Seventh, the declaration of Burt Levine, another former AT&T attorney, further refutes IBM’s “literal source code only” position, specifically pointing out its obvious textual problems: “Section 2.01 certainly did not require that modifications or derivatives of the UNIX product contain any literal copying of UNIX source code in order to be

33

Burt Levine confirms that he had also never approved of such a change, which would have constituted a material change contrary to AT&T’s consistent and firm policy. See Levine Decl. (11/23/04) ¶¶ 10, 12 (Exh. 7).

77

protected. Indeed, there would have been no need for the last clause of Section 2.01 if it were so limited; the literal UNIX source code was already protected by virtue of its inclusion in the term “SOFTWARE PRODUCT.” Levine Decl. (11/23/04) ¶ 6 (Exh. 7). Mr. Levine’s testimony also confirms that the “license agreements safeguarded AT&T’s valuable UNIX asset by protecting the entire chain of development that resulted from the special access and development rights AT&T afforded its licensees under those agreements.” Id. Eighth, the sworn declarations of former AT&T Account Executives Mitzi Bond and Evelyn Davis further undermine IBM’s “literal source code only” claim. Both Ms Bond and Ms Davis agree that the “strict contractual protections” of the UNIX software agreements “went beyond safeguarding just the source code in the original licensed UNIX product.” Bond Decl. (11/4/04) ¶ 7 (Exh. 14); see also Davis Decl. (11/4/04) ¶ 6 (Exh. 14). 34 In fact, Ms Bond testified that “the license agreements covered any works that resulted from our licensees’ exercise of their contractual right to prepare modifications or derivatives of the UNIX product.” Bond Decl. (11/4/04) ¶ 7(a) (Exh. 4); see also Davis Decl. (11/4/04) ¶ 6(a) (Exh. 4). Confirming the scope of protections against products created by licensees with the benefit of exposure to UNIX, as enumerated by Mr. Wilson and Mr. Frasure in their BSD testimony, Ms Bond explained: “The agreements required each licensee to treat any work prepared with the benefit of having been exposed to our product – regardless of how the licensee

34

Unlike Messrs. Wilson and Frasure, Ms. Bond’s declaration is consistent with the testimony that she gave in the BSD case. In that case, Ms. Bond testified that “you can prepare derivative work and you don’t necessarily have to include code from our software product” because it was protected simply “based on the fact that it was based on our product.” Bond Decl. (11/4/04) ¶ 11(e) (Exh. 14) (citing Bond Dep. (12/9/92) at 220-21 (Exh. 24)) (emphasis added). Ms. Bond also testified in BSD about AT&T’s protection of products created with the benefit of exposure to UNIX: “If a licensee is exposed to our code and that licensee develops a modification, a derivative work, an application or any type of software product as a result of seeing our code, then that product should be treated the same as our software product.” Bond Decl. (11/4/04) ¶ 11(c) (Exh. 14) (citing Bond Dep. (12/9/92) at 224 (Exh. 24)). Ms. Bond noted that the core intellectual property protections were substantially similar for both AT&T’s commercial (IBM and Sequent) and educational (BSD) UNIX licenses. Bond Decl. (11/4/04) ¶ 10 (Exh. 14).

78

further developed or changed that product – as if it were part of our original licensed product.” Bond Decl. (11/4/04) ¶ 7(c) (Exh. 14) (emphasis added); see also Davis Decl. (11/4/04) ¶ 6(c) (Exh. 4). Ninth, IBM’s and Sequent’s conduct pursuant to the terms of their sublicensing agreements also undercuts IBM’s position that only literal UNIX source code - and not derivative products based on UNIX - is protected by the software agreements. As explained above, the sublicensing agreements restricted use of the same derivative products as the software agreements, except that the protected sublicensed products were in object code format (not readable by people), while the software agreements protected source code products. Both IBM and Sequent paid royalties to AT&T and its successors for many years pursuant to the terms of their sublicensing agreements. See, e.g., Acheson Decl. (11/30/04) ¶ 3 (Exh. 44). Moreover, IBM and Sequent explicitly recognized that their respective UNIX flavors – AIX and Dynix – were derivatives of UNIX. See ¶ 167, above. By thus acknowledging that AIX and Dynix were UNIX derivatives as defined by the sublicensing (and thus software) agreements, and by further treating those products as derivative products under the sublicensing agreements, the course-of-conduct evidence further undermines IBM’s new position that the AIX and Dynix products are not covered by the software agreements. b. AT&T did not intend to forfeit its contractual protections for UNIX methods and concepts.

In addition to the express contractual protections for modifications and derivatives under Section 2.01, the Sequent software agreement expressly protected from disclosure any “methods or concepts” contained in the original licensed product. Faced with this additional plain-reading impediment to its “literal source code only” theory, IBM attempts to fall back again on witness declarations to suggest that by 1985, AT&T did not really intend to protect the methods and

79

concepts in UNIX. IBM Mem. ¶¶ 79-99, pp. 57-62. Again, however, IBM’s declarations contradict the plain contractual language and are sharply disputed by other extrinsic evidence including, again, the inconsistent testimony of IBM’s own declarants. First, IBM relies on Mr. Wilson’s recent declaration to deny that AT&T sought to protect the methods and concepts embodied in UNIX System Five and to contend that AT&T was “not aware of any particular ‘methods or concepts’ that needed to be protected.” Wilson Decl. (12/11/03) ¶ 14 (Exh. 25). 35 Contrary to IBM’s argument, however, Mr. Wilson’s own contemporaneous writings to UNIX licensees shows that he sought to protect UNIX “methods and concepts” pursuant to AT&T’s software agreements. For example:
•

In his letter to Digital Equipment Corporation (“DEC”) dated Sept. 5, 1984, Mr. Wilson wrote: “Regarding Section 2.01 of the referenced Software Agreement, you do not have all right, title and interest in derivative works based on a SOFTWARE PRODUCT if such derivative works include any of our code or embody any of the methods and concepts used in a SOFTWARE PRODUCT.” Exh. 36 at 2 (emphasis added). In proposed (and later executed) letter agreements dated June 18, 1990, June 29, 1990, and August 3, 1990, Mr. Wilson stated to other licensees: “You will not provide access to any copy of the source code of the SOFTWARE PRODUCT (including methods and concepts contained therein), in whole or in part, to anyone other than your organization’s employees who have a need to know.” Exhs. 2729 (emphasis added).

•

Second, notwithstanding IBM’s attempted reliance on its recent declaration from David Frasure, see Frasure Decl. (3/28/04) ¶ 15 (Exh. 18), Mr. Frasure’s 1992 testimony in the BSD case - which involved a software agreement signed in November 1985 -acknowledged that “the licensed software to me encompasses methods and concepts techniques,” so that “to the extent at some point [a licensee] gets involved with the source code, the I – and its methods and concepts,

35

Mr. Wilson’s position that AT&T was not aware of any methods or concepts that could be protected is additionally entirely belied by the fact that the language explicitly protecting UNIX “methods and concepts” remained in the standard UNIX license agreements for more than a decade after the IBM and Sequent agreements were executed.

80

then I think there – there’s a restriction on its – on its use.” Frasure Dep. (12/8/92) at 45-46 (Exh. 19) (emphasis added). Mr. Frasure stressed that there is “more to the agreement” than literal UNIX source code and proceeded to point to “other portions of paragraphs that cover methods and concepts and – and stuff.” Id. at 36-37 (emphasis added). Mr. Frasure characterized these protections as a “key part of the agreement.” Id. at 147. When asked about the types of products covered by AT&T’s software agreement, Mr. Frasure testified: “Q. So licensed software - your understanding of licensed software in the 1985 time period in the System Five license agreement could include source code of AT and T’s; is that correct? Yes. Okay. And it could include object code of AT and T’s as well? Yes. Okay. And it could also include documentation of AT and T’s; is that correct? Yes. And it could also include methods and concepts of AT and T’s included in their System Five product; is that correct? Yes.” Id. at 148.

A. Q. A. Q. A. Q. A.

Mr. Frasure also acknowledged the application of the methods and concepts protection to work created by a licensee’s programmers: “if someone has been exposed to the source code and has used the source code and then goes off on their own and wanted to develop a product that subconsciously or through rote or whatever the words are developed a product that contained the methods and concepts, techniques that we’re using, then it’s – I don’t know what the right word to use – understanding, interpretation or what – that - that AT and T would have a – an interest in that product.” Id. at 147-148. Third, although IBM also attempts to rely on Geoff Green’s cursory declaration to support its position, see Green Decl. (6/17/04) ¶ 6 (Exh. 22), Mr. Green admitted at deposition

81

that, when the Sequent agreement was signed, AT&T protected “methods and concepts” 36 under the use and disclosure provisions of the software agreements: “Q. If a licensee has access to UNIX code and develops a modification or a derivative under this Sequent agreement, but does not include the literally copied source code but does include methods, concepts, structures, organizations, other covered intellectual property, I think we agreed that that derivative or modification would be covered by the license agreement, correct? At which point in time. Under this Sequent agreement? Yes. Okay. And then that derivative or modification work under this agreement would be restricted to the licensee’s internal business purposes, used for internal business purposes? Yes. And the transfer restrictions under the agreement would apply? Yes. And the confidentiality provisions of the agreement would apply? Yes. And the export requirements would apply? Yes.” Id. at 142-43 (objection omitted, emphasis added).

A. Q. A. Q.

A. Q. A. Q. A. Q. A.

Mr. Green also agreed that UNIX methods and concepts were included under the definition of the “SOFTWARE PRODUCT” in Section 1.04: “Q. So at that time, for example, at the time the Sequent agreement was executed, SOFTWARE PRODUCT would have included not just the literally copied source code, but also the methods and concepts and the other intellectual property in the UNIX program? Yes.” Id. at 147; see also id. at 106.

A.

Mr. Green further testified that AT&T’s protection of methods and concepts extended even to the IBM software agreement. Although the IBM Side Letter did not contain the explicit methods and concepts language in Section 7.06, it replaced that language with other language that was designed to protect against copying of UNIX methods and concepts: “Q.
36

Is it true that as of February 1985 that AT&T was willing to relinquish any

When referring to the protected UNIX “methods and concepts,” Mr. Green included “any kind of know-how or structure or sequence or organization” from the UNIX product. Green Dep. (11/15/04) at 132 (Exh. 5).

82

A.

of its protections for any of its licensees with respect to methods and concepts? It may not have been. I think, refreshing myself from this review we’ve done in the last few minutes, it may very well be that the language further down in 7.06(a) relating to the contamination issue was part of the negotiation of removing the methods and concepts language because that was in a way getting at the same thing, another way of getting at the same thing.” Id. at 223.

When asked to elaborate, Mr. Green explained that AT&T included its protection of methods and concepts in all of its agreements - including IBM’s agreement as revised by the side letter: “Q. You were saying it’s another way of doing the same thing or something to that effect. My question was: By doing the same thing or ensuring the same protection, what you meant was, and correct me if I’m wrong, was making sure that the intellectual property other than just the literally copied source code was protected from misappropriation by AT&T’s licensees? Yes, it was another way of getting at the methods and concepts. And is that consistent with the fact that AT&T did not remove methods and concepts from its standard agreement for many years after the IBM side letter? I think it is.” Id. at 223-24 (emphasis added). 37

A. Q.

A.

Fourth, IBM’s reliance on David Rodgers’s declaration is misplaced, see Rodgers Decl. (11/5103) ¶ 12 (Exh. 23), in light of his inconsistent deposition testimony, in which he recognized that in addition to the protection of “actual System V source code” there was also “the methods and procedures issue as well.” Rodgers Dep. (6/10/04) at 137 (Exh. S1) (emphasis added). Responding to a hypothetical question concerning a product created by a licensee that

37

In fact, the change to the agreements that Mr. Green referenced never did occur. The UNIX standard software agreement did undergo several changes since the IBM and Sequent agreements were signed in 1985, but the protection of methods and concepts was never removed. See Sequent Software Agreement (4/18/18) § 7.06 (Exh. 1); Santa Cruz Software Agreement (5/6/87) § 7.05 (Exh. 30); Systech Software Agreement (1/4189) § 7.05 (Exh. 31); Gates Rubber Software Agreement (3/1/90) § 7.04 (Exh. 32); Information Foundation Software Agreement (11/24/92) § 7.05 (Exh. 33); Green Hills Software Agreement (6/21/94) § 7.05 (Exh. 34); Modcomp/Cerplex Software Agreement (3/28/96) § 7.05 (Exh. 3 5). In light of Mr. Green’s testimony that the language of the agreements reflected AT&T’s intent, Green Dep. (11/15/04) at 104, 142 (Exh. 5), it is fair to conclude, contrary to IBM’s argument, that AT&T never intended to give up its methods and concepts protections under the license agreements. See Pfeffer Decl. (9/7/04) ¶¶ 10-11 (Exh. 8); Levine Decl. (11/23/04) ¶¶ 11-12 (Exh. 7).

83

contains “structures and sequences and organization as it appears in System V, Mr. Rodgers acknowledged that “if the reason the similarity was there was because it was just copied, then yeah, I would agree that that would be subject to the constraints.” Id. at 159. Fifth, Ira Kistenberg’s recent declaration acknowledges that the software agreements protected against copying or disclosing “any structures, sequences, patterns, ideas, methods or concepts from UNIX System V” aside from the express protection of UNIX source code. Kistenberg Decl. (11/12/04) ¶ 5 (Exh. 10). Sixth, AT&T General Attorney Martin Pfeffer stated: “AT&T intended to protect, and through its standard license agreements expressly protected, its UNIX business by preventing anyone from using AT&T’s proprietary material in UNIX (including by literally or non-literally copying UNIX source code, disclosing methods or concepts from UNIX, and/or exploiting licensees’ access to the technology in UNIX) without paying UNIX license fees to AT&T.” Pfeffer Decl. (9/7/04) ¶ 10 (Exh. 8). Mr. Pfeffer explained the reason for the inclusion of this further specific protection of “methods or concepts”. “I know that this language was specifically designed to cast the widest possible net in order to ensure protection for AT&T’s proprietary information, beyond the common law protections that would otherwise have been available (absent the agreement) for AT&T’s trade secrets and ‘know-how.’” Id. ¶ 9 (emphasis added). Seventh, Mitzi Bond and Evelyn Davis each confirmed that AT&T software agreements “protected all of the innovations embodied in the licensed UNIX software, including, among other things, the structures, sequences, patterns, ideas, methods, and concepts.” Davis Decl. (11/4/04) ¶ 6(b) (Exh. 4); Bond Decl. (11/4/04) ¶ 7(b) (Exh. 14). Indeed, Ms Bond’s declaration echoes her testimony in the BSD case, in which she explained that to violate “the UNIX license agreement ‘doesn’t necessarily mean you have to have copied exact code. You can also copy an idea or method or concept that you have presented in a different form.’” Bond Decl. (11/4/04) ¶ 11(d) (Exh. 14) (citing Bond Dep. (12/9/92) at 219 (Exh. 24)).
84

c.

The IBM Side Letter concerns ownership of modifications and derivatives only, and does not even purport to affect the use and disclosure restrictions in the software agreements.

IBM also attempts to rely on witness declarations to argue, contrary to the plain language of the IBM Side Letter, that the side letter somehow limited all of the contractual protections under both agreements to literal copying of UNIX source code. IBM Mem. ¶¶ 100-12. As explained below, however, the extrinsic evidence concerning the intended effect of the “ownership” language in the IBM Side Letter is sharply disputed. 38 Indeed, the testimony of IBM’s own declarants, in addition to the testimony of other declarants in this case, supports the plain reading: that the IBM Side Letter’s addition of “ownership” language to Section 2.01 of IBM’s software agreement did not affect the software agreements’ use and disclosure restrictions on derivatives and modifications of the UNIX product. 39 First, AT&T General Attorney Martin Pfeffer explained AT&T’s decision to add language (in late 1985) to its standard agreement that in relevant part, as IBM has acknowledged, is substantially similar to the “Regarding 2.01” clause in the IBM Side Letter. 40 See IBM Mem. ¶ 110. But according to Mr. Pfeffer, the language was not intended to effect anything like the change that IBM attempts to ascribe to it: “In 1985, AT&T added language to its software license agreements to address

38

Moreover, the extrinsic evidence discussed in the next section undermines any attempt by IBM to carry over any language from the IBM Side Letter to the Sequent agreement. IBM makes the same mistake of ignoring this crucial distinction when it cites language from an irrelevant SCO document that addresses a licensee’s “ownership interest” in a derivative work to support IBM’s argument concerning a licensee’s freedom to “use and disclose” those works. See IBM Mem. at 55. The particular similarity between AT&T’s change to the standard UNIX license agreement in late 1985 and the change enumerated in the IBM Side Letter concerned the addition of the ownership clause to Section 2.01: “AT&T-IS claims no ownership interest in any portion of such a modification or derivative work that is not part of a SOFTWARE PRODUCT.” See, e.g., IBM Side Letter (2/1/8.5) at 2 (Exh. 3); Santa Cruz Agreement (5/6/86) § 2.01 (Exh. 30). The standard license agreement also altered the text of the preceding sentence in Section 2.01, but that change was not included in the IBM Side Letter, and thus entirely irrelevant to both the Sequent and IBM agreements.

39

40

85

inquiries from certain of its licensees concerning the ownership of modifications and derivative works. Later that year, the following language was added to AT&T’s standard license agreement: ‘AT&T-IS claims no ownership interest in any portion of such a modification or derivative work that is not part of a SOFTWARE PRODUCT.’ Based on my personal knowledge and professional experience at AT&T and USL (including my extensive experience in this area as described above), I know that this language set forth the parties’ intent and agreement concerning only the ownership of modifications or derivative works, and that this language was not intended to change, and did not in fact change, AT&T’s right to control (for example, the use and disclosure) of such modifications and/or derivative works, as set forth in Section 2.01 and other provisions of the software license agreements.” Pfeffer Decl. (9/7/04) ¶ 7 (Exh. 8) (emphasis added). Second, although IBM attempts to rely on Mr. Frasure’s recent declaration to support its misreading of the IBM Side Letter, IBM ¶ 107 (citing Frasure Decl. (3/28/04) ¶ 18 (Exh. 18)), he too contradicted IBM’s position in his BSD testimony. When Mr. Erasure was asked in the BSD case about the AT&T contractual language that was substantively similar to the “Regarding 2.01” provision in the IBM Side Letter, he testified: “Q. So that if an AT and T licensee developed a software that was a modification or derivative work which did not contain source code, object code, documentation or methods and concepts of AT and T’s, AT and T did not assert an ownership interest in that product. Is that your understanding - was that your understanding in May of 1985? It has no ownership in that portion of the product. Okay. But it doesn’t mean that it’s not part of the licensed software, I don’t think. I think there’s two - two things here that we’re talking about, licensed software and - and ownership, which I think are two completely separate subjects.” Frasure Dep. (12/8/92) at 151-52 (Exh. 19) (emphasis added).

A. Q. A.

Mr. Frasure reiterated that distinction numerous times in his BSD testimony: “I still feel like we’re talking about - about two separate things here. One is ownership and one is what is the licensed product.” Id. at 166-167. Furthermore, Mr. Frasure used that fundamental distinction to explain the textual change, made later in 1985, to revise AT&T’s software agreement in order to reflect substantially similar ownership language from the IBM Side Letter:

86

“All we were trying to do is to clarify that there was - they [the licensees) owned the - the modifications, any - any derivative or enhancements that they made to the product, but it was also going to be treated as part of the product.” Id. at 103. 41 See also id. at 157-59 (reiterating distinction between “completely separate” “clear and distinct” subjects of “ownership” and “licensed software,” and explaining that “if you generate the code, add the code to it then I think, clearly the software agreement and the side letter said that you owned that. That didn’t say that it was not part of the – the software licensed product...I’m saying that the ownership belongs clearly to the originator of - of that code, but it - it could be construed to be part of the software product.”). In fact, Mr. Frasure reiterated this crucial distinction even in his deposition testimony in this case: “While we asserted no ownership of the source code, that was written, at one point there was possible consideration depending upon if methods and concepts were used in the development of this source code that AT&T may have some claim to the control of that software, but not the ownership of that software.” Frasure Dep. (6/8/04) at 49 (Exh.20) (emphasis added). Third, IBM’s reliance on Mr. Wilson’s recent declaration to support its misreading of the IBM Side Letter, IBM Mem. ¶ 107 (citing Wilson Decl. (4/26/04) ¶¶ 19-20 (Exh. 11)), is belied by Mr. Wilson’s prior BSD testimony, which concerned a November 1985 software agreement that contained the revised ownership language upon which IBM now relies. 42 See The Regents of the University of California Agreement (11/12/85) § 2.01 (Exh. 13). Mr. Wilson’s testimony about the protections of both modifications and derivatives and methods and concepts, without any need for literal copying of UNIX source code, demonstrates that AT&T’s position regarding

41

The substantial similarity concerned only the ownership clause added to the standard license agreements in late 1985. Mr. Frasure’s BSD testimony concerns the same agreement, and thus similarly confirms that the new language in Section 2.01 plainly relates to ownership alone.

42

87

the use and disclosure protections of the software agreements was unaffected by the new ownership clause upon which IBM now hopes to rely. See ¶¶ 21, 23-24, above (detailing the scope of the protections about which Mr. Wilson testified in his BSD deposition). IBM’s reliance on Mr. Wilson’s recent declaration is further undermined by contemporaneous letters sent by Mr. Wilson to licensees that acknowledged the distinction that IBM now ignores between licensees’ “ownership” of modifications and derivative works, on the one hand, and the “use, copy, distribute, or disclosure” restrictions on such works, on the other. In Mr. Wilson’s letter to DEC dated September 5, 1984, for example, he explained: “If such derivative work does include any of our code or embody any of our methods and concepts you may have a property right in such derivative work to the extent of any modifications that you have added, but the exercise of that property right is subject to the terms of the Software and Sublicensing Agreements, including to the restrictions on the use of the SOFTWARE PRODUCT (for example, it must be kept in confidence).” Exh. 36 at 2 (emphasis added). Mr. Wilson reiterated this statement in a later letter to DEC on February 21, 1985 - twenty days after AT&T entered into the software agreement and side letter with IBM. See Exh. 26, at 3. Fourth, IBM also attempts to rely on Mr. Kistenberg’s first declaration to support its argument, IBM Mem. ¶ 110 (citing Kistenberg Decl. (6/24/04) ¶ 18 (Exh. 9)), but Mr. Kistenberg’s more recent testimony undermines any such attempt to ignore the plain contractual language: “During my tenure at AT&T, an issue arose concerning the ownership of modifications and derivative works that had been prepared by UNIX licensees. AT&T subsequently changed Section 2.01 of its standard agreement to state its position that AT&T would not claim ownership of any source code that a licensee developed entirely on its own. This ownership provision did not reduce or compromise the restrictions on use and disclosure that were set forth in the license agreements”. Kistenberg Decl. (11/12/04) ¶ 8 (Exh. 10) (emphasis added). Fifth, IBM’s dubious position is further and directly undercut by the deposition testimony of its own declarant, David Rodgers (from Sequent):
88

“Q.

A.

So then what you understood on Section 2.01 was that it was not discussing ownership but, instead, was stating that the right to use includes the right to modify and to prepare derivative works, providing the resulting materials are treated confidentially? Is that what you’re telling us? Yes.” Rodgers Dep. (6/10/04) at 128 (Exh. S1) (objection omitted, emphasis added).

Sixth, IBM declarant Geoff Green confirmed that AT&T distinguished between “ownership, on the one hand, and use and disclosure on the other.” Mr. Green testified that he was “sure” he had conversations about that distinction with both Mr. Wilson and Mr. Frasure. Green Dep. (11/15104) at 300 (Exh. 5). d. Neither the IBM Side Letter nor later versions of the UNIX software agreements can or did alter the terms of Sequent’s software agreement

The extrinsic evidence also does not support IBM’s attempt to alter the meaning of the Sequent agreement by importing terms from the IBM Side Letter (executed two months before the Sequent agreement) and other AT&T software agreements (executed more than four months after the Sequent agreement) that were not included in, and in some instances are (at least according to IBM’s attempted construction of those terms) directly contradicted by the Sequent agreement. IBM Mem. ¶¶ 100-12. Indeed, IBM’s attempt to rewrite and reinterpret the Sequent agreement in light of such terms is squarely rejected by ample extrinsic evidence proving that the parties intended for their respective obligations to be governed only by the terms of their respective software agreements. First, while IBM attempts to rely on Mr. Wilson’s declaration to suggest that the terms of the IBM Side Letter automatically applied to every licensee, IBM Mem. ¶¶ 111-12 (citing Wilson Decl. (4/26/04) ¶ 27) (Exh. 11), Mr. Wilson admitted at his deposition that AT&T held licensees only to the terms of their respective software agreements and that Sequent’s obligations were governed exclusively by its own written software agreement:
89

“Q.

A. Q.

A. Q. A.

Is it your testimony that a party can be bound to an amendment to a software agreement without having given written authorization to the amendment? No. I take it, based on that answer, Mr. Wilson, that Sequent was not bound by or a party to any side letter entered into by IBM and AT&T? Q. Am I correct about that? Sequent had agreements directly with AT&T. That’s what they were – that what they were bound by. Not by any other licensee. And any agreements between Sequent and AT&T were governed only by the express agreements assented to in writing by those parties; correct? That’s correct.” Wilson Dep. (6/10/04) at 232-35 (Exh. 17) (attorney colloquy omitted).

Second, IBM declarant Geoff Green confirmed at his deposition that the merger clause in the Sequent and IBM agreements was intended “to make clear that the agreement and its supplements constituted the entire agreement. That’s what the language says.” Green Dep. (11/15104) at 93 (Exh. 5). Mr. Green explained the implications for Sequent: “Q. A. Q. Was it AT&T’s intent that Sequent’s rights and obligations would be governed only by the agreements that Sequent executed with AT&T? With respect to UNIX’s - the UNIX’s software covered by the agreement, yes. And is it also fair to say that AT&T’s intent was that the rights and obligations of Sequent would be governed only by what was contained within the written agreements that Sequent executed? Yes. As opposed to oral representations? Yes.” Id. at 95 (objections omitted). 43

A. Q. A.

Third, former Sequent Vice President David Rodgers contradicted IBM’s position at his deposition, testifying about the process through which AT&T and Sequent would change any terms of the written software agreements:

43

Mr. Green further testified that, while the software agreements sometimes contained a “most favored nation clause,” such provisions were specific to the context of pricing, and he was not aware of any such clause relating to the intellectual property protections in the license agreements. Id. at 95-96. In any event, Mr. Green confirmed that the Sequent agreement contained no such provision and that Paragraph 12 of the IBM Side Letter (the “most favored nation clause”), was “not intended to provide IBM with most favored nation with respect to intellectual property protections.” Id. at 221-22.

90

“Q.

A.

Having had the opportunity to review Exhibits 101, 102, and 103, does this refresh your recollection at all as to written correspondence being the manner in which changes or clarifications to the various agreements would occur; namely, they would be done in writing and countersigned by Sequent or somebody at Sequent? If there was a material change, if it was an increment of rights or content.” Rodgers Dep. (6/10/04) at 169-70 (Exh. S1) (objection omitted, emphasis added) 44

Fourth, Ira Kistenberg’s most recent declaration contradicts IBM’s position: “Although AT&T generally employed standard license agreements, to the extent that any contractual language differed, AT&T always intended for the rights and obligations of each licensee to be governed by that licensee’s particular written agreement and side letter(s), and not the contents of any other licensee’s written agreement or side letter(s).” Kistenberg Decl. (11/12/04) ¶ 9 (Exh. 10). As Mr. Kistenberg has also testified, “the policy of the UNIX licensing group was that any changes to a licensee’s software agreement had to be in writing. I never verbally agreed with any licensee to modify its license in any way without memorializing the modification in writing.” Id. ¶ 10. Fifth, AT&T General Attorney Martin Pfeffer affirmed that the merger clause in the AT&T software agreements was intended to prevent the terms of any agreement between AT&T and one licensee from affecting the rights and obligations of any other licensee: “This language set forth the parties’ intent and agreement that only the terms of each licensee’s written agreements with AT&T would govern those parties’ contractual obligations. In other words, any agreements (including any side letter agreements) that AT&T entered into with one of its UNIX licensees would have no legal effect on the contractual obligations of AT&T vis-a-vis any of its other UNIX licensees.” Pfeffer Decl. (9/7/04) ¶ 12 (Exh. 8).

44

Mr. Rodgers also admitted that Sequent signed the “standard form agreement” and that, aside from price, Sequent did not negotiate or change any terms from that standard form agreement. Id. at 82-84 (Exh. S1).

91

e.

The $ echo newsletter was not intended to have any effect on licensees’ rights or obligations under the UNIX licenses

IBM further purports to rely on “documentary evidence” - exclusively in the form of an informational newsletter that AT&T sent to subscribing licensees months after the IBM and Sequent agreements at issue in this case were executed. IBM ¶¶ 115-23, pp. 63-64. The content from $ echo on which IBM relies, however, does not in any way support IBM’s reading of the agreements. To the contrary, as with the amended standard language that the newsletters announced (and that were not made to IBM’s standard agreements until many months after the IBM and Sequent licenses at issue here were executed), the $ echo newsletters on which IBM relies spoke only to the issue of the ownership of modifications and derivative works. For the reasons discussed above, neither the newsletter nor the related contractual language concerned the use and disclosure obligations imposed under the software agreements, and those sources certainly do not support IBM’s contention that the software agreements at issue here were intended to protect against only literal copying of UNIX source code. See Parts I.A.4. & I.B.2.c, above. The extrinsic evidence in the record further belies IBM’s attempt to import into the IBM and Sequent software agreements any subsequent announcements in $ echo. First, Mitzi Bond, who served as the editor of the $ echo newsletter for the duration of its existence, has explained: “AT&T used the newsletter to disseminate information to subscribing licensees, including about the requirements of the UNIX software agreements. The newsletter was never intended or designed to have any legal effect, on our licensees or on their license agreements with AT&T.” Bond Decl. (11/4/04) ¶ 4 (Exh. 14) (emphasis added). Indeed, IBM’s position here is contradicted by AT&T’s intent with respect to the newsletter: “To the contrary, if for, example, the $ echo newsletter announced language that AT&T intended to include in its standard agreement, such language would
92

become part of a licensee’s agreement with AT&T only if that licensee executed an agreement containing that language. As the license agreements themselves stated, each individual licensee’s right and obligations were governed by that particular licensee’s signed license agreement.” Id. (emphasis added). Second, even IBM’s own declarants David Frasure, Otis Wilson, Ira Kistenberg, and attorney Geoff Green have each acknowledged that the $ echo newsletter was not intended to have any effect on the actual provisions of the IBM and Sequent software agreements. David Frasure went so far as to acknowledge that the newsletter should not even be used in trying to interpret the meaning of the software agreements: “Q. It’s your testimony that the Dollar Echo document shouldn’t be used in trying to construe the meaning of agreements between AT&T and licensees?” Yes. I agree with that.” Frasure Dep. (6/8/04) at 246-47 (Exh. 20) (objections omitted).

A.

See also Kistenberg Decl. (11/12/04) ¶ 8 (Exh. 10) (“I recall that in a $ echo newsletter AT&T announced that it would be coming out with a clarification to the software agreement relating to the question of ownership. The $ echo newsletter did not, and was not intended to, create binding contractual rights; only the written agreements themselves could do that.”); Green Dep. (11/15/04) at 100-01 (Exh. 5) (testifying that he did not believe that it was “AT&T’s intent that statements made in the $ echo newsletter created binding letter rights or obligations with respect to the recipients of the newsletter” or that the newsletter was “intended to modify any of the license agreements or side letter agreements”); Wilson Dep. (6/10/04) at 68 (Exh. 17) (testifying that “our purpose with the newsletter was just to provide information to our licensees, to keep them abreast of what was going on with the product” (emphasis added)). Third, the text of the $ echo newsletter itself confirms that the changes it “announced” were not even intended to apply to licensees retroactively, but instead to licensees who executed new agreements, which a licensee could request from its AT&T account executive. See AT&T

93

$ echo Newsletter (Aug. 1985) at 5 (“Contact your Account Executive for information or specimen copies.”) (Exh. 37). C. IBM’s Restrictive Interpretation of the Software Agreements Is Commercially Unreasonable IBM’s proposed reading of the software agreements also defies commercial reason that it cannot represent a reasonable reading of the agreements. As IBM itself acknowledges, if a contract interpretation attributes to the parties an intent that defies commercial reason in light of the circumstances of the contract’s formation, “such a construction can be rejected in favor of one which would better accord with the reasonable expectations of the parties.” Reape v. N.Y. News, Inc., 122 A.D.2d 29, 30 (N.Y. App. Div. 1986); see also Tougher Heating & Plumbing Co. v. State, 73 A.D.2d 732, 733 (N.Y. App. Div. 1979) (rejecting reading of contract as unfair and unreasonable interpretation of “the parties’ purposes in making the contract”); IBM Mem. at 67 (collecting cases). Aside from its textual and evidentiary infirmities, IBM’s argument for summary judgment on the ground that the UNIX software agreements protect against only the literal copying of UNIX source code is particularly tenuous when viewed in light of the circumstances at the time the agreements were executed in 1985. UNIX (and its owner at the time, AT&T) occupied an enviable market position when System Five was licensed to IBM and Sequent. As Evelyn Davis, UNIX Account Executive at the time, has explained: “In UNIX, AT&T had developed a seminal innovation that was in high demand in the computer industry. Prospective licensees (including leading computer hardware manufacturers) desired to license and use AT&T’s UNIX operating system to avoid having to spend substantial time and money developing their own software products from scratch. In response to that demand, AT&T agreed to allow its licensees special access to its proprietary UNIX product by licensing the UNIX operating system to them, but complimented that special access with specially designed, and broad, protections in its license agreements that would safeguard the valuable UNIX asset.” Davis Decl. (11/4/04) ¶ 5 (Exh. 4) (emphasis added).
94

Indeed, IBM’s own declarant, former AT&T licensing attorney Geoff Green, confirmed in his deposition that the UNIX operating system was “an important innovation” that “AT&T spent quite a significant and substantial amount of resources on” and that was “in high demand among licensees or potential licensees because of its capabilities.” Green Dep. (11/15/04) at 25-28 (Exh. 5). In response to this demand, AT&T permitted licensees “special access” to the UNIX source code and all of the intellectual property embodied in that code, Davis Decl. (11/4/04) ¶ 5 (Exh. 4); Green Dep. (11/15104) at 294-95 (Exh. 5), and AT&T even permitted its licensees to rely on the UNIX software to develop products based on UNIX, including derivatives, modifications, substantially similar copies, and even literal copies of the UNIX code. See Green Dep. (11/15/04) at 48-49, 125, 132-36 (Exh. 5); Davis Decl. (11/4/04) ¶¶ 5-6 (Exh. 4); Bond Decl. (11/4/04) ¶ 7 (Exh. 14). AT&T “appreciated the risk” inherent in letting “that kind of development process go on with the valuable UNIX source code,” particularly behind the licensees’ closed doors. Green Dep. (11/15/04) at 47-49 (Exh. 5). So AT&T and its attorneys devoted substantial efforts to developing software license agreements that would protect against the prospect of licensees permitting the intellectual property embodied in UNIX to reach the open market as a royalty-free competitor to UNIX, thereby jeopardizing UNIX’s value as an exploitable asset. See id. at 37, 47-49. Moreover, in those same software agreements and the separate sublicensing agreements that AT&T required of any licensee who wished to sublicense UNIX-based products, AT&T prohibited any licensee from distributing or otherwise disclosing any UNIX modifications or derivatives in anything but object code format - the binary format that is not readable by people, AT&T Sublicensing Agreement with IBM (2/1/85) ¶ 1.04 (Exh. 6); AT&T Sublicensing Agreement with Sequent (1/28/96) ¶ 1.04 (Exh.54).

95

Although the copyright law recognized protections for non-literal derivatives of copyrighted work, 45 and the common law protected trade secrets and know-how, 46 AT&T understood that it was “essential to safeguard the valuable UNIX asset by protecting this entire chain of development” that could result from the special access and development rights that AT&T afforded its licensees under their software license agreements. Levine Decl. (11/23/04) ¶ 6 (Exh. 7). Accordingly, AT&T’s attorneys developed a license that was “specifically designed to cast the widest possible net” of protections, “beyond the common law protections that would otherwise be available (absent the agreement) for AT&T’s trade secrets and ‘knowhow.’” Pfeffer Decl. (9/7/04) ¶ 9 (Exh.8). And AT&T’s contract managers took a hard line in refusing to compromise the core UNIX intellectual-property protections in AT&T’s license negotiations. See Davis Decl. (11/4/04) ¶ 9 (Exh. 4) (explaining that AT&T was “adamant in refusing to comprise its core intellectual property protections” and that “it was not uncommon for Account Executives and Contract Managers to explain to potential licensees that [these core protections] were not negotiable”). Contrary to these facts, IBM would now have this Court rule as a matter of law that through its software agreements, AT&T obtained almost no protection at all - a competent programmer can easily evade literal-copying restrictions by paraphrasing the text of the restricted source code while copying the non-literal operative elements. Davidson Decl. (11/23/04) ¶ 12(c)

45

See, e.g., SAS Inst., Inc. v. S&H Computer Svs., Inc., 605 F. Supp. 816, 830 (M.D. Tenn. 1985) (finding copyright violation on the basis of substantial similarity of program’s “organization and structural details” and specifically rejecting the proposition that such similarity could only be demonstrated in “specific lines of code”); Whelan Assocs., Inc. v. Jaslow Dental Labs, Inc., 609 F. Supp. 1307, 1321 (E.D. Pa. 1985) (noting substantial similarity of two programs in holding that “direct evidence of copying is not required”); Atari, Inc. v. N. Am. Philips Consumer Elecs. Corp., 672 F.2d 607 (7th Cir. 1982) (“When the alleged infringing work is not a duplicate of the copyrighted work, a determination of whether protected elements have been taken requires a close analysis of the difference and similarities between the two works.”). See Pfeffer Decl. (9/7/04) ¶ 9 (Exh. 8); Green Dep. (11/15/04) at 44-45 (Exh. 5).

46

96

(Exh. 58). Such toothless protections for the UNIX intellectual property would have jeopardized the UNIX asset itself and compromised AT&T’s ability to exploit that asset (through licensing fees, royalties, or otherwise). Furthermore, under IBM’s reading, when AT&T permitted licensees to rely on its seminal innovation for the price of a license fee, it granted those licensees the unfettered right to distribute and disclose their modifications and derivative works in source code format - even though the sublicensing agreements required that such works be distributed only in object code format. IBM would also have the Court rule that through its software agreements, AT&T agreed and insisted that all of its licensees agree - to reduce AT&T’s pre-existing legal protections. Indeed, according to IBM’s argument, AT&T granted its licensees - the parties who were in a position to abuse their “special access” and extinguish the value of AT&T’s innovation by misappropriating the UNIX intellectual property - more lenient restrictions on use and disclosure than the law already imposed on third parties who had no such access to the code. Stated another way, according to IBM, AT&T’s own software agreements provided AT&T with the least protection against the very parties who posed the greatest risk to its valuable UNIX rights. That AT&T would abdicate such control over UNIX for the price of a license fee defies commercial reason, particularly in light of the fact that AT&T had exclusive ownership and control of a breakthrough innovation that was in high demand, and IBM and Sequent were among hundreds of licensees who needed access to that code to compete in the marketplace. Accordingly, IBM’s restrictive reading of the software agreements is unreasonable in light of the circumstances surrounding the formation of those agreements, and IBM’s argument for summary judgment based on its proffered contract interpretation should be denied for this additional, independently sufficient reason.

97

II.

IBM’S “WAIVER” ARGUMENTS ARE MERITLESS IBM further argues that Novell has waived SCO’s rights under the UNIX software

agreements on SCO’s behalf, and that SCO has waived those rights itself. It is well-settled, however, that waiver is ordinarily a question of fact and, as such, is “generally ill-suited for summary adjudication.” In re Caldor, Inc., 217 B.R. 121, 133 (Bankr. S.D.N.Y. 1998) (applying New York law); see also Boston Concessions Group v. Criterion Ctr. Corp., 606 N.Y.S.2d 696, 697 (App. Div. 1994) (“[U]nder New York law, the establishment of a waiver, requiring the intentional relinquishment of a known right, is ordinarily a question of fact which precludes summary judgment.”). And neither of IBM’s two meritless “waiver” arguments comes close to warranting a departure from that rule in this case. A. Novell Does Not Have Any Authority to Waive SCO’s Intellectual Property Protections Under UNIX Software Agreements In its first “waiver” argument, IBM attempts to argue that Novell waived SCO’s protections under the UNIX software agreements against SCO’s will. IBM’s argument is predicated on its fundamental misinterpretation of the Asset Purchase Agreement (“APA”) between Novell and SCO’s predecessor-in-interest The Santa Cruz Organization, Inc. (hereinafter “SCO”). Even though Novell sold its entire UNIX business to SCO under that agreement - specifically including all of the UNIX source code, UNIX licensing rights, and UNIX System V agreements - IBM claims that Novell retained the unfettered “right to waive any rights under the UNIX System V licenses.” IBM Mem. at 72 (emphasis added). Basic principles of contract construction, commercial reason, and the undisputed evidence available to date (including sworn testimony from the principals on both sides of the APA transaction) flatly contradict IBM’s position.

98

1.

IBM’s Interpretation of APA Section 4.16(b) Is Inconsistent with the Language, Purpose, and Intent of the APA

In the APA, Novell and SCO agreed that: “It is the intent of the parties hereto that all of the Business and all of Seller’s backlog, if any, relating to the Business be transferred to Buyer.” APA § 1.3(a)(i) (Exh. 38) (emphasis added). The APA defined the “Business” as “the business of developing a line of software products currently known as Unix and UnixWare, the sale of binary and source code licenses to various versions of Unix and UnixWare, the support of such products and the sale of other products which are directly related to Unix and UnixWare.” APA, Recital A. Section 1.1 (a) thus transferred to SCO “all of Seller’s right, title and interest in and to the assets and properties of Seller relating to the Business (collectively the ‘Assets’) identified on Schedule 1.1 (a) hereto,” APA § 1.1 (a) (Exh. 38) (emphasis added), which assets specifically included, among other things:
•

“All rights and ownership of UNIX and UnixWare including but not limited to,” among other things, “all versions of UNIX and UnixWare and all copies of UNIX and UnixWare (including revisions and updates in process),” “all technical, design, development, installation, operation and maintenance information concerning UNIX and UnixWare, including source code,” and “UNIX Source Code Products,” “Binary Product Releases,” and “Products Under Development,” APA, Schedule 1.1 (a) § I (Exh. 38) (emphasis added); “All of Seller’s claims arising after the Closing Date against any parties relating to any right, property or asset included in the Business,” id. § II (emphasis added); “All of Seller’s rights pertaining to UNIX and UnixWare under any software development contracts, licenses and any other contracts to which Seller is a party or by which it is bound and which pertain to the Business (to the extent that such contracts are assignable),” id. § III (emphasis added); All UNIX “Software and Sublicensing Agreements - This includes the source code and sublicensing agreements that Seller has with its OEM, End User and Educational customers,” id. (emphasis added); and

•

•

•

99

•

“All contracts relating to the SVRX Licenses” -including for all versions of UNIX System Five that had been licensed to IBM and Sequent, id. § VI (emphasis added). 47

As part of the consideration that SCO paid Novell for the entire UNIX business, Novell was permitted to continue receiving royalties paid by UNIX System Five (“SVRX”) licensees for their distribution, pursuant to sublicensing agreements, of binary (or object) code versions of System Five. See APA §§ 1.2(a)-(b), 4.16(a); id. Schedule 1.1(b), Item VIII (excluding from asset sale “All right, title and interest to the SVRX Royalties, less the 5% fee for administering the collection thereof pursuant to Section 4.16 hereof.”) (Exh. 38). 48 Although Novell thus had a limited financial stake in the future royalties paid under the SVRX licenses, SCO obtained through the APA all rights to the UNIX intellectual property and all rights to develop the UNIX business based on that property. 49 Section 4.16 of the APA, titled “SVRX Licenses,” governed Novell’s SVRX binary royalty interests. Section 4.16(a) set out the procedures for SCO’s collection and payment to Novell of the royalties, and gave Novell the right to audit those collection efforts.
47

This Court has previously determined that the plain language of the APA does not permit summary judgment against SCO on the question of whether the UNIX copyrights transferred to SCO under the APA. See Order of 6/9/04 at 14-15. Accordingly, IBM cannot (and does not) base its instant motion on Novell’s purported ownership of the UNIX copyrights. Indeed, the extrinsic evidence to date demonstrates that both sides of the Novell-SCO asset transfer understood that SCO was obtaining the UNIX copyrights through the APA. See Chatlos Decl. (10/1/04) ¶¶ 6-12 (Exh. 39); Wilt Decl. (11/23/04) ¶¶ 5-10,12-16 (Exh. 40); Sabbath Decl. (11/19/04) ¶¶ 4-6 (Exh. 42). Apparently, even IBM internally acknowledged that Novell had transferred ownership of the UNIX copyrights to SCO. See Exh. S2 (internal IBM e-mail: “The requirement for an equivalent license can only be satisfied by a source code license from SCO or prior copyright owner for the code (Novell/AT&T).”) Because the Court must construe the evidence in the light most favorable to SCO, the Court must resolve in SCO’s favor (if it deems it relevant) the issue of the ownership of the UNIX copyrights. The parties agreed to this exception because SCO could not pay Novell the full purchase price for the UNIX assets up front; the future binary royalty stream bridged the price gap. See Chatlos Decl. (10/1/04) ¶¶ 7, 9, 13 (Exh. 39); Wilt Decl. (11/23/04) ¶¶ 6, 9, 10 (Exh. 40). By the time of the APA closing, SCO’s business did not contemplate any additional significant sales of SVRX source-code licenses, and the remaining interest in that particular part of the UNIX business consisted primarily of collecting binary royalties attributable to sub licensed object-code product. Wilt Decl. (11/23/04) ¶ 11 (Exh. 40). Of course, because the System Five software included substantial intellectual property that SCO was using in later versions of its UNIX and UnixWare products, SCO had a strong continuing interest in protecting that property under the existing SVRX software agreements. Id.

48

49

100

Section 4.16(b) set forth legal protections for Novell’s future binary-royalty interests. And Section 4.16(c), in turn, protected SCO’s future UNIX business by prohibiting Novell from promoting the sale of SVRX products. Based on its misinterpretation of a single sentence in Section 4.16(b), 50 removed entirely from context, IBM attempts to argue that the APA gave Novell the unilateral and unfettered right to require SCO to change or to waive, at any time and for any reason, any of SCO’s rights (including SCO’s right to enforce its intellectual-property protections) under any UNIX System Five agreement. This reading of the APA - on which IBM’s entire Novell-related waiver argument rests - cannot entitle IBM to summary judgment for a number of textual reasons. First, IBM misinterprets the term “SVRX License” in Section 4.16. In suggesting that Novell had the authority to waive SCO’s intellectual-property protections under the SVRX software agreements, IBM necessarily sweeps such agreements under the definition of “SVRX Licenses”; it is the software agreements that specify the restrictions on licensees’ source code rights. 51 The term “SVRX Licenses” is not defined anywhere in the APA, 52 but a plain reading of the agreement reveals IBM’s contract interpretation to be wrong. Specifically, the included “Assets” schedule to the APA (Schedule 1.1(a)) expressly refers to the UNIX software agreements (in paragraph III.L) separately from the SVRX Licenses (in paragraph VI). Moreover, to the extent that the software agreements themselves may have been included within

50

That sentence states: “In addition, at Seller’s sole discretion and direction, Buyer shall amend, supplement, modify or waive any rights under, or shall assign any rights to, any SVRX License to the extent so directed in any manner or respect by Seller.” APA § 4.16(b) (Exh. 38). In purporting to waive IBM’s breaches on SCO’s behalf, Novell expressly referenced SCO’s rights under the UNIX software agreements. See IBM Mem., Exhs. 41, 43. Indeed, although SVRX Licenses is capitalized like the other defined terms in the APA - such as “Business,” “Shares,” “Assets,” “Excluded Assets,” “Assumed Liabilities,” and “Unassumed Liabilities” - unlike those other terms, SVRX Licenses is not defined in the APA. Section 4.16 refers to SVRX licenses “listen in detail under Item VI of Schedule 1.1(a).” The software agreements are not listed in detail or otherwise under Item VI, but rather in Item IV.L. of the Schedule - which is not referenced in Section 4.16 at all.

51

52

101

“All contracts relating to the SVRX Licenses” phrase in paragraph VI, the use of that broader phrase to describe the “Assets” included in the APA transfer belies any suggestion that the term “SVRX Licenses,” standing alone - as it does in Section 4.16(b) - includes such agreements. These considerations - and the fact that the APA specifically refers to “royalties, fees, and other amounts due under all SVRX Licenses,” Section 4.16(a) - strongly support a narrower reading of “SVRX Licenses,” to refer just to the SVRX product Schedules that, unlike the software agreements themselves, identify all such amounts. See, e.g., Exh. 56 (IBM System Five Schedule). Furthermore, this more precise reading is consistent with the role of Section 4.16 to facilitate the management of Novell’s binary royalty interests and, unlike IBM’s reading, does not (as described below) run into direct conflict with the purpose of the APA and many of its other key provisions. Accordingly, IBM’s argument is not supported by the language of the provision on which it relies. Second, IBM’s interpretation of that single sentence in Section 4.16(b) should be rejected because it would defeat the whole purpose of the APA. See e.g., Leo F. Piazza Paving Co. v. Found. Constructors, Inc., 128 Cal. App. 3d 583, 591 (Cal. Ct. App. 1981) (“The words, phrases and sentences employed are to be construed in light of the objectives and fundamental purposes of the parties to the agreement”); County of Marin v. Assessment Appeals Bd., 134 Cal. Rptr. 349 (Cal. Ct. App. 1976) (“A contract entered into for the mutual benefit of the parties is to be interpreted so as to give effect to the main purpose of the contract and not to defeat the mutual objectives of the parties; language which is inconsistent with the objective of the contract shall be rejected.”); Viacao Aerea Sao Paulo, S.A. v. Int’l Lease Fin. Corp., 859 F.2d 924, 1988 WL 103286, at *4 (9th Cir. 1988) (reversing summary judgment because “the language of a contract

102

is to be construed in light of the purposes of the parties to the agreement”) (Exh. D). 53 As IBM attempts to read it, Section 4.16(b) of the APA would give Novell the unfettered right to, among other things:
•

require SCO to permit any and all SVRX licensees to copy, distribute, export, or even open source the licensed UNIX source code (and all of the protected elements contained in that code) - the very value of the assets SCO had acquired without any protection or compensation for SCO; and direct SCO to waive any of its rights to enforce any licensee’s material breach of any SVRX license - even where such rights have no bearing whatsoever on Novell’s future binary royalty stream or relate to an SVRX license under which there are no future binary royalties at all.

•

IBM thus argues that in a single sentence of Section 4.16(b), SCO gave Novell the authority, for any reason or no reason at all, to eviscerate the entire value of the UNIX assets that SCO had acquired through the APA even though Novell’s sole continuing interest under the APA was limited to the SVRX binary royalties. That implausible construction of Section 4.16(b) cannot be accepted as a matter of well-settled contract interpretation law. Third, IBM’s interpretation of Section 4.16(b) runs into direct conflict with other key provisions of the APA. As a corollary to the requirement that contractual provisions are not be construed as meaningless, 54 it is well settled that a contract which confers certain rights or benefits in one clause will not be construed in other provisions completely to undermine those rights or benefits. See County of Marin, 134 Cal. Rptr. at 353-55 (rejecting interpretation of clause that would have permitted party to unilaterally deny the other party the bargained-for benefit provided for in another clause of the contract); Cooper v. Mart Assocs., 225 Cal. App. 2d

53

Paragraph 9.8 of the APA provides that the APA “shall be governed by and construed in accordance with the laws of the state of California” (Exh. 38). See, e.g., Heidlebaugh v. Miller, 271 P.2d 557, 559 (Cal. Ct. App. 1954) (“The court will if possible give effect to all parts of the instrument and an interpretation which gives a reasonable meaning to all its provisions will be preferred to one which leaves a portion of the writing useless or inexplicable.”).

54

103

108, 114-16 (Cal. Ct. App. 1964) (holding that contract’s waiver clause would not be read to nullify the right to sue provided in another clause, because a “contract is to be construed as a whole,” with “each clause helping to interpret the other”). IBM’s proposed reading of the second sentence of Section 4.16(b) must fail because it would completely undermine numerous other detailed provisions of the APA. For example:
•

Section 1.1 and Schedule 1.1(a) grants to SCO “All rights and ownership of UNIX and UnixWare including but not limited to” the “source code” to the “UNIX Source Code Products” (including UNIX System Five and all prior and subsequent UNIX and UnixWare products. But if Novell had the right to change or cancel any of the intellectual-property protections in the SVRX software agreements, this express asset-transfer provision would be rendered hollow for the same reasons that IBM’s contract interpretation would undermine the purpose of the APA itself. Section 4.16(a) provides that in consideration for collecting binary royalties and ensuring proper payment of such royalties, “Seller shall pay Buyer within 5 days of receipt of SVRX Royalties from Buyer.” But if Section 4.16(b) were as broad as IBM contends, then Novell purportedly could have vitiated this provision by simply requiring SCO to direct licensees to make their royalty payments directly to Novell. Section 4.16(c) states: “Seller further covenants that immediately following the Closing Date neither it, nor any of its officers, directors or employees shall … take any material action designed to promote the sale of SVRX products.” APA § 4.16(c). But if Section 4.16(b) were as broad as IBM contends, then Novell purportedly could have directed SCO to change its SVRX licenses in any of several ways designed to promote the sale of SVRX products, including by requiring SVRX licensees to promote such products.

•

•

Fourth, IBM completely ignores the important language in Amendment No. 2 to the APA, which is directly relevant to IBM’s argument and expressly confirms the error in its contractual interpretation. See, e.g., Harm v. Frasher, 181 Cal. App. 2d 405, 412-13 (Cal. Ct. App. 1960) (“It is a general rule that several papers relating to the same subject-matter and executed as parts of substantially one transaction, are to be construed together as one contract.”); Heston v. Farmers Ins. Group, 160 Cal.App.3d 402, 417 (Cal. Ct. App. 1984) (“The two documents are interrelated and must be read together for purposes of interpretation.”). Paragraph
104

B.5 of Amendment No. 2, which was executed on October 16, 1996, begins: “This Amendment does not give Novell the right to increase any SVRX licensee’s rights to SVRX source code.” APA Amendment No. 2 ¶ B.5 (Exh. 41) (emphasis added). 55 If Section 4.16(b) of the APA were as broad as IBM contends, then Novell would have already had the right to direct SCO to amend or supplement any SVRX software agreement in order to increase a licensee’s rights to SVRX source code. The first sentence of Paragraph B..5 to Amendment No. 2 confirms the parties’ understanding that Section 4.16(b) did not have the broad scope that IBM now proposes. SCO and Novell also expressly agreed in Amendment No. 2 that “Novell may not prevent SCO from exercising its rights with respect to SVRX source code in accordance with the Agreement.” Id. ¶ B.5 (emphasis added). This plain language further confirms that, contrary to IBM’s argument, Novell had no right to waive SCO’s protections for the SVRX intellectual property under the IBM and Sequent software agreements. It also highlights the crucial distinction - which must underlie any reasonable interpretation of the APA - between Novell’s limited binary royalty rights and SCO’s broad source code rights. This plain language of Amendment No. 2 forbids the very action that Novell seeks to take in preventing SCO from exercising its rights with respect to SVRX source code - and is in itself more than sufficient reason to deny IBM’s motion. 2. IBM’s Out-of-Context Interpretation of the APA Is Commercially Unreasonable and Produces Absurd Results

IBM’s proposed interpretation of the APA must be rejected for the additional reason that “the court shall avoid an interpretation which will make a contract extraordinary, harsh, unjust, inequitable or which would result in absurdity.” Howe v. Am. Baptist Homes of the W., Inc.,
55

Amendment 2 also confirmed that the APA had transferred to SCO “the copyrights and trademarks owned by Novell as of the date of the Agreement required for SCO to exercise its rights with respect to the acquisition of UNIX and UnixWare technologies.” Id. ¶ A.

105

112 Cal. App. 3d 622, 627 (Cal. Ct. App. 1980); see also IBM Mem. at 67 (citing similar New York authorities). As demonstrated above, IBM’s argument would give Novell an unfettered right to destroy SCO’s Tights in UNIX and UnixWare at any time after the APA closing. Thus, under IBM’s view, in exchange for 17% of SCO’s outstanding stock, SCO received only phantom paper rights because, under one sentence of Section 4.16(b), Novell could at any time change the SVRX software agreements in any way - including by open sourcing all of the UNIX product and thereby destroying the value of the assets transferred under the APA - as Novell alone saw fit. Moreover, the undisputed circumstances in which Novell purported to direct SCO to waive its rights against IBM and Sequent only underscore the absurdity of IBM’s argument. At the time that Novell purported to waive breaches by IBM and Sequent of the software agreements, neither of those parties were paying, or were expected to pay, any future binary royalties under their SVRX licenses. 56 Accordingly, when it purported to direct SCO to waive its intellectual-property protections to the UNIX System Five source code, Novell could not have been even arguably seeking to protect its interests in the future SVRX binary royalties. Indeed, in none of its letters to SCO did (or could) Novell even claim that it was acting to protect any binary royalty stream from IBM or Sequent. 57

56

In Amendment X (Oct. 17, 1996) to IBM’s software license, IBM had agreed to pay SCO a nonrefundable fee of $10.125 million for a perpetual license to use the UNIX source code and to sublicense binary products; Amendment X thus effected a buy-out of IBM’s future binary royalty obligations (Exh. 57). Similarly, by the time Novell purported to waive SCO’s source code rights with respect to Dynix (in February 2004), Sequent was no longer paying royalties, having made its last royalty payment to SCO of only approximately $3,000 in the first quarter of 2003. See Acheson Decl. (11/30/04) ¶ 3 (Exh. 44). To the contrary, the evidence shows that at the time it directed SCO to waive its rights, Novell was improperly seeking to promote business interests entirely separate from any rights it retained under the APA. In a series of deals struck between November 2003 and March 2004, Novell and IBM have formed a very close Linux partnership: Novell acquired Linux distributor SuSE Linux, and IBM invested $50 million in Novell to finance part of the acquisition. Press Release, Novell, Inc., “Novell Finalizes IBM Investment (Mar. 3, 2004) (Exh. 45). Under the IBM-Novell deal, Novell became a major Linux distributor, allowing IBM to continue its official 106

57

Furthermore, IBM’s proposed reading of the APA would render meaningless the Technology License Agreement that the parties entered into contemporaneously with the execution of the APA. See, e.g., Harm, 181 Cal. App. 2d at 412-13 (“Where two or more written instruments are executed contemporaneously, with reference to each other, for the purpose of attaining a preconceived object, they must all be construed together, and effect given if possible to the purpose intended to be accomplished.”). As Section 1.6 of the APA contemplated, that agreement effected a “license back” to Novell of the right to use, among other things, the UNIX System Five code. See Exh. 48. Under the Technology License Agreement, Novell was permitted to use the licensed code only for internal purposes and only in bundled products (and then only to the extent that no material part of the bundled product competed with SCO’s UnixWare business). See id. According to IBM’s untenable reading of the APA, however, Novell would have the right to waive, on SCO’s behalf, even Novell’s own strict obligations with respect to SVRX code under the Technology License Agreement. By waiving such obligations, Novell could have even given itself the right to continue its UNIX business in competition with SCO - effectively nullifying yet another fundamental purpose of the APA. In light of all of these perverse consequences, IBM’s proposed reading of the APA should be rejected as absurd, commercially unreasonable, and contrary to the reasonable expectations of the parties.

distance from the Linux distribution channel. IBM spokesman Mike Darcy stated: “IBM is not a Linux distributor and has no interest in being in that business.” Hiawatha Bray, Novell to Buy SuSE Linux for $210M: Firm’s Commitment to Software Boosted, The Boston Globe, Nov. 5, 2003 (Exh. 46). Thus, any purported “waiver” of SCO’s rights by Novell must be seen for what it is: an attempt by IBM and Novell to exploit a woefully misguided misreading of a single sentence in the APA in order to promote IBM and Novell’s new joint business interests in Linux. Nevertheless, IBM benefits significantly from the Linux distributions made by Novell. Novell CEO Jack Messman explained the benefits of its partnership with IBM as follows: “IBM stands to gain a great deal of revenue by providing systems, peripheral devices, software and services for Linux.” Bom Mims, Novell, IBM and HP unite efforts to put Linux on top, Three companies unite to boost Linux and topple Microsoft, The Salt Lake Tribune, Mar. 25, 2004, at El (Exh. 47).

107

3.

The Extrinsic Evidence Available to Date Refutes IBM’s Interpretation of the Parties’ Intent

The discussion above demonstrates that IBM’s proposed reading of the APA should be rejected as a matter of law. At the very least, however, IBM cannot establish its entitlement to summary judgment on its claim that Novell waived SCO’s rights because the language of the APA (including the scope of the undefined term “SVRX Licenses” in Section 4.16(b)) and of Amendment No. 2 is at least ambiguous. 58 And while IBM has not offered any extrinsic evidence to support its reading of Section 4.16(b), consideration of the undisputed extrinsic evidence of the contracting parties’ intent with respect to those two documents further demonstrates not only that the parties never intended for Novell to have the broad waiver rights that IBM claims, but also (under applicable California law) that the APA and Amendment No. 2 are at least ambiguous with respect to that issue. See Pac. Gas & Elec. Co. v. G.W. Thomas Drayage & Rigging Co., 69 Cal. 2d 33, 37 (1968) (“The test of admissibility of extrinsic evidence to explain the meaning of a written instrument is not whether it appears to be plain and unambiguous on its face, but whether the offered evidence is relevant to prove a meaning to which the language of the instrument is reasonably susceptible.”); Wolf v. Super. Ct., 114 Cal. App. 4th 1343, 1351, 1357 (Cal. Ct. App. 2004) (“Where the meaning of the words used in a

58

The “absence of a [contract] definition, though perhaps not dispositive, might weigh, even strongly, in favor of finding an ambiguity, for example, when the term in question has no generally accepted meaning outside the context of the [contract] itself.” Bay Cities Paving & Grading, Inc. v. Lawyers’ Mut. Ins. Co., 5 Cal. 4th 854, 867 (Cal. Ct. App. 1993). Indeed, California courts routinely deny summary judgment where an undefined relevant term is subject to more than one reasonable interpretation. See, e.g., Wolf v. Suver. Ct., 114 Cal. App. 4th 1343, 1357 (Cal. Ct. App. 2004) (reversing summary judgment because “the only way to construe the meaning of the term” was by considering extrinsic evidence); EOTT Energy Corp. v. Storebrand Int’l Ins. Co., A/S, 45 Cal. App. 4th 565, 574, 578 (Cal. Ct. App. 1996) (undefined term “certainly presents a problem [precluding summary judgment] if the term is reasonably susceptible to more than one meaning”); cf. Ramming v. Barnard, No. E030334,2002 WL 393118, at *5 (Cal. Ct. App. Mar. 13, 2002) (“It cannot seriously be contended that the [agreement] is not ambiguous. For example, the use of the undefined term ‘sale proceeds’ is susceptible to many meanings[.]”) (Exh. E); accord World Trade Ctr. Props. v. Hartford Fire Ins. Co., 345 F.3d 154, 190 (2d Cir. 2003) (summary judgment was properly denied because meaning of undefined term was “an open question as to which reasonable finders of fact could reach different conclusions”).

108

contract is disputed, the trial court must provisionally receive any proffered extrinsic evidence which is relevant to show whether the contract is reasonably susceptible of a particular meaning”; indeed, “it is reversible error for a trial court to refuse to consider such extrinsic evidence on the basis of the trial court’s own conclusion that the language of the contract appears to be clear and unambiguous on its face.”); So. Cal. Edison v. Super. Ct., 37 Cal. App. 4th 839, 848 (Cal. Ct. App. 1995) (“Whether the contract is reasonably susceptible to a party’s interpretation can be determined from the language of the contract itself or from extrinsic evidence of the parties’ intent.”); see also Blumenfeld v. R. H. Macy & Co., 92 Cal. App. 3d 38, 45 (Cal. App. Ct. 1979). The chief negotiators for both Novell and SCO agree that neither SCO nor Novell ever intended for the APA to give Novell the right to waive SCO’s protections for its intellectual property under the SVRX software agreements. The sworn declaration of Novell’s own principal negotiator, Ed Chatlos, explains: “Paragraph 4.16 of the APA was specifically designed and intended to protect Novell’s retained binary product royalty stream. Based on the foregoing, including my understanding of the parties’ intent, I do not believe Novell has any right to waive, or to direct or require SCO to waive, any of SCO’s source code rights, including under customer source code licenses.” Chatlos Decl. (10/1/04) ¶ 13 (Exh. 39). Jim Wilt, who negotiated the APA for Santa Cruz, concurs in his sworn declaration: “Paragraph 4.16 of the APA pertains to the binary royalty income stream that Novell retained through the APA. The parties agreed to the language in Paragraph 4.16(b) in order to allow Novell to manage that royalty stream within the operation of SCO’s customer source code licenses - not at the expense of SCO’s right to enforce its intellectual property protections under any such licenses, and not to permit Novell to waive any of those protections. I have reviewed Amendment No. 2 to the APA and believe that the language therein confirms that intent. In light of my intent, and based on my understanding of the parties’ intent, I do not believe that Novell had or has any right to waive, or direct or require SCO to waive, any of its intellectual property rights or protections.” Wilt Decl. (11/23/04) ¶ 10 (Exh. 40).

109

And Steve Sabbath, Santa Cruz’s former Vice President of Law and Corporate Affairs and a participant in a number of the meetings and discussions with the chief negotiators and attorneys leading up to the APA, has similarly declared: “I understand that IBM has argued that Section 4.16(b) of the APA gave Novell the right to require Santa Cruz to waive any breach of the intellectual property protections provided in the SVRX licenses. That argument is contrary to the intent of Paragraph 4.16(b) as I understood it. Indeed, Santa Cruz would never have agreed to give Novell the right under the APA to waive such protections under the SVRX licenses because such a right could have eviscerated the entire purpose of the APA and the value of the assets transferred to Santa Cruz under the APA.” Sabbath Decl. (11/19/04) ¶ 5 (Exh. 42). With respect to Amendment No. 2 to the APA, Mr. Sabbath, who was involved in the discussions leading up to that amendment and who executed the amendment on SCO’s behalf, has further testified: “Amendment No. 2 arose as a result of a dispute between Novell and SCO concerning Novell’s attempt to execute, on Santa Cruz’s behalf, a royalty buy-out with IBM. That dispute was ultimately resolved through an amendment to IBM’s SVRX license that was jointly executed by Santa Cruz, Novell, and IBM. Amendment No. 2, however, was intended to confirm, among other things, the parties’ intent that SCO would obtain ownership of the UNIX copyrights under the APA and that Novell had received no rights with respect to UNIX source code under the APA. Paragraph B.5 of Amendment No. 2 was specifically intended to make clear that Novell had no right to increase any SVRX licensee’s rights to SVRX source code, no right to grant any new SVRX source code licenses, and no right to prevent Santa Cruz from exercising the rights it obtained under the APA with respect to SVRX source code.” Id. ¶ 6 (emphasis added). Accordingly, for all of the above reasons, there is no merit to IBM’s claim that Novell had the authority to waive IBM’s breaches of the intellectual-property protections in the SVRX software agreements. B. SCO Has Not Waived Its Rights Under the Software agreements In its second “waiver” argument, IBM contends that “SCO’s acts and conduct are plainly inconsistent with an intention to assert a breach of contract claim against IBM based on the code

110

allegedly at issue.” IBM Mem. at 74. IBM’s “implied waiver” claim fails on both the wellestablished law and the disputed facts. “Waiver requires the voluntary and intentional abandonment of a known right which, but for the waiver, would have been enforceable.” Gen. Motors Acceptance Corp. v. Clifton-Fine Cent. Sch. Dist., 85 N.Y.2d 232, 236 (1995); see also Davison v. Kless, 280 N.Y. 252, 261 (1939) (“Whether an alleged waiver is express or implied, it must be intentional.”). In addition, ‘where a waiver is not express, but found in the acts of a party, summary judgment is not appropriate. An implied waiver is invariably a matter of intention and, therefore, an issue of fact.” McDarren v. Marvel Entm’t Group, Inc., No. 94 Civ. 0910 (LMM), 1995 WL 214482, at *5 (S.D.N.Y. Apr. 11, 1995) (Exh. F) (emphasis added). 59 IBM’s argument for implied waiver is thus untenable as a matter of law. Nor is there any factual basis for any claim that SCO has voluntarily and intentionally abandoned its right to pursue IBM’s breach of contract. SCO promptly filed this lawsuit, on March 6, 2003, shortly after learning of IBM’s breach, and SCO has always diligently protected and defended its rights under the UNIX System Five (“SVRX”) agreements. See Sontag Decl. (11/30/04) ¶¶ 3, 9 (Exh. 49). Indeed, IBM does not even allege (let alone present any evidence) that SCO failed promptly to pursue its rights after learning that IBM had improperly dumped source code into

59

The cases IBM cites do not suggest a different rule and are based on facts that are irrelevant to those presented here. In Secs. Indus. Automation Corp. v. United Computer Capital Corp., 723 N.Y.S.2d 668 (N.Y. App. Div. 2001), the court found a waiver in light of the uncontradicted evidence and under the UCC rule that “where a signed agreement excludes modification except by a signed writing, which was the case here, an attempted modification that does not meet the writing requirement or satisfy the Statute of Frauds may still operate as a waiver.” Id. at 669. In T.W.A. Trading, Inc. v. Gold Coast Freightways, Inc., No. 2001-900 KC, 2002 WL 1311648 (N.Y. App. Term. Apr. 2, 2002) (Exh. G), the court applied the “well settled” rule that “a shipper’s unqualified and unconditional acceptance of a check that was collected by its carrier in payment of merchandise, contrary to the shipping instructions, ratifies the carrier’s conduct, and the shipper thereby waives any claim it may have against the carrier for breach of contract.” Id. at *1. In Heidi E. v. Wanda W., 620 N.Y.S.2d 665 (N.Y. App. Div. 1994), the plaintiff indisputably had waived the defendant’s right to maintain the confidentiality of plaintiff’s medical records “by authorizing her friend to inquire about her condition and concerns.” Id.

111

Linux. In light of SCO’s diligent enforcement efforts, the Court could not possibly conclude, as a matter of law, that SCO has waived its contract claims. IBM nevertheless contends that SCO’s “implied waiver” of its contract rights may be found based on (1) SCO’s sales of certain Linux products that contained source code improperly contributed by IBM; and (2) SCO’s steps in making some of that same code purportedly “publicly available on the Internet.” IBM Mem. at 74-75. These “waiver” claims are meritless. 1. SCO Has Not “Waived” Its Contract Claims by Distributing Limited Linux Products

Insofar as SCO’s Linux sales are concerned, IBM’s argument reduces to the contention that SCO waived its contract claims by continuing to “distribute” its Linux Server 4.0 and OpenLinux 3.1.1 products after SCO learned of IBM’s breaches, 60 because SCO could not have possibly waived any of rights before it knew of IBM’s conduct. See K. Bell & Assocs., Inc. v. Lloyd’s Underwriters, 827 F. Supp. 985, 989 (S.D.N.Y. 1993) (“A proper claim of waiver should state that the waiving party voluntarily and intentionally relinquished a right, with knowledge of all the facts and circumstances that would constitute the entitlement to that right.”). 61 Although it is true that SCO copied, advertised, and distributed the Linux kernel and other related Linux software for years before 2003, see Hughes Decl. (11/30/04) ¶ 9 (Exh..50); during that time, SCO did not know that IBM had contributed source code to Linux in violation of its (and Sequent’s) SVRX licenses. Sontag Decl. (11/30/04) ¶ 5 (Exh. 49). SCO did not know that any of the features mentioned in SCO’s advertisements (with the sole exception of the

60

IBM contends generally that SCO’s “distributions of Linux” contain “code upon which SCO’s claims are based,” IBM Mem. at 75, but presents evidence concerning only SCO’s Linux Server 4.0 and SCO’s OpenLinux 3.1.1 products. Further, evidence that SCO “should have known” of its claim or of SCO’s “negligence,” “oversight,” or even “thoughtlessness” would not suffice to show SCO’s waiver of its rights. Leibowitz v. Elsevier Sci., Ltd., 927 F. Supp. 688, 705 (S.D.N.Y. 1996); Madison-Oneida-Herkimer Consortium v. N. Am. Admins., 765 N.Y.S.2d 184, 190 (Sup. Ct. 2003); accord Readco, Inc. v. Marine Midland Bank, 81 F.3d 295, 303 (2d Cir. 1996).

61

112

journaling file system support, or “JFS”) had been contributed by IBM. Id. ¶¶ 6-8. And, as to that one exception, although SCO marketed JFS as “developed by IBM,” SCO did not know that JFS had been derived from SCO-licensed code or that IBM had contributed it in breach of its contractual obligations. Id. Indeed, SCO did not know that any of the advertised features had been derived from SCO’s proprietary software licensed to IBM or had been contributed to Linux by IBM in violation of IBM’s agreements with SCO. Id. The same day SCO filed this suit, SCO sent a notice of termination to IBM’s Chief Executive Officer, explaining that IBM’s right to use or distribute any software product based on UNIX System Five, including AIX, would be terminated on June 13, 2003, unless IBM cured those breaches. Sontag Decl. (11/30/04) ¶ 10 & Exh. C (Exh. 49). SCO sent a similar notice to IBM regarding Sequent, and Dynix/ptx, on May 29, 2003. Sontag Decl. (11/30/04) ¶ 10 & Exh. D (Exh. 49). SCO attempted to meet and confer with IBM, but IBM failed to cure its breaches during the 100-day period provided in SCO’s termination letter to IBM. Sontag Decl. (11/30/04) ¶ 11. Sequent also failed to cure its breaches during the two-month period provided in SCO’s termination letter to Sequent. Id. Accordingly, SCO terminated IBM’s contract rights effective June 13, 2003, and terminated Sequent’s contract rights effective July 30, 2003. Id. & Sontag Decl. Exhs. C, D. After filing this suit, SCO carefully considered whether to continue selling its Linux products. Sontag Decl. (11/30/04) ¶ 12 (Exh. 49). SCO considered its obligations to its existing customers. Id. ¶ 13 SCO decided that the appropriate solution was to suspend its sale and marketing of its Linux products (effective May 14, 2003), but to continue to permit SCO’s pre-existing customers to order Linux products (which IBM appears to describe as the “distribution” of those products).

113

Id. ¶ 14. 62 SCO did not enter into any further obligations to sell Linux Server 4.0 or OpenLinux 3.1.1 after May 14, 2003. Hughes Decl. (11/30/04) ¶ 3 (Exh. 50). 63 This evidence precludes any finding that SCO “intended” - impliedly or otherwise - to “waive” any of its contract rights. Waiver “should not be lightly presumed.” Gilbert Frank Corp. v. Fed. Ins. Co., 70 N.Y.2d 966, 968 (1988). Instead, “the intent to waive a right must be unmistakably manifested, and is not to be inferred from a doubtful or equivocal act.” Navillus Tile, Inc. v. Turner Constr. Co., 770 N.Y.S.2d 3, 5 (App. Div. 2003); accord ESS & Vee Acoustical & Lathing Contractors, Inc. v. Prato Verde, Inc., 702 N.Y.S.2d 38, 39 (App. Div. 2000). In filing its suit for breach of contract and terminating the IBM and Sequent software agreements, SCO clearly manifested its intent to enforce those rights. It is illogical to suggest, as IBM does (without any authority), that a clear and unequivocal intent to waive certain rights can be inferred from acts taken after the filing of a suit to protect those rights. 64 That SCO continued

62

Contrary to IBM’s assertion (at paragraph 153 of its Statement of Undisputed Facts), SCO issued a press release on April 14, 2003, announcing the proposed release of “SCO Linux Server 4.0 for the Itanium Processor Family,” but never released or sold the product. Sontag Decl. (11/30/04) Exh. E (Exh. 49). By suspending the sale of its Linux-related products - including the operating system, services, support, professional services, education, and layered applications - SCO lost approximately 5-10% of its revenues. Sontag Decl. (11/30/04) ¶ 15 (Exh. 49). In contrast, from May 14, 2003, until May 31, 2004 (when SCO last sold a unit of Linux Server 4.0), SCO sold 83 units and had 79 units returned, for a net revenue of $1,849. Hughes Decl. (11/30/04) ¶ 15 (Exh. 50). Since May 14, 2003, SCO has sold 403 units of OpenLinux 3.1.1 and had 51 units returned, for a net revenue of $49,097. Hughes Decl. ¶ 5. IBM’s failure to argue that it ever believed (let alone acted to its detriment on any such belief) that SCO intended to waive its contract rights further highlights the conceptual defect in its “waiver” argument. Given that SCO’s prosecution of this lawsuit patently precludes any finding that SCO has intended to “waive” its contract claims, IBM’s argument (such as it is) is more properly framed under New York law as a claim that SCO’s conduct should constitute an equitable estoppel. See, e.g., Bur Rigid Box, Inc. v. Travelers Prop. Cas. Corp., 302 F.3d 83, 95 (2d Cir. 2002) (explaining difference between grounds for alleged waiver and estoppel). But IBM could not prove any such claim without demonstrating that it reasonably relied to its detriment on SCO’s conduct. See id.; accord Dunlop-McCullen v. Pascerella, No. 97 Civ. 0195 (PKL) (DFE), 2002 WL 31521012, at *14 (S.D.N.Y. Nov. 13, 2002) (Exh. H). IBM’s inability even to allege such reliance or detriment may explain why it seeks to manufacture its instant “waiver” argument. In addition: “Whether equitable estoppel applies presents an issue of fact.” Bennett v. U.S. Lines, Inc., 64 F.3d 62, 65 (2d Cir. 1995); accord Dunlop-McCullen, 2002 WL 31521012, at *15 (Exh. H) (“Estoppel is usually a question of fact inappropriate for summary judgment.”).

63

64

114

to honor pre-existing obligations to its own pre-existing customers does not remotely permit the inference that SCO intended to discontinue its enforcement of its contract rights. IBM’s argument amounts to the astonishing claim that its breaches of its contractual obligations created a situation in which SCO had only two choices when it learned of those breaches: (1) abandon its customers and its Linux business overnight, thereby inflicting the maximum damage on those customers and itself as a result of IBM’s breaches; or (2) immunize IBM for its improper conduct. IBM does not, because it cannot, cite any authority even suggesting that IBM, through its own improper conduct, could put SCO to such a choice. In light of all the circumstances - and particularly viewing those circumstances in the light most favorable to SCO - the Court could not possibly infer, as a matter of law, “a clear manifestation of intent” on SCO’s part “to relinquish the protection” of the software agreements. Gilbert Frank, 70 N.Y.2d at 968. 2. SCO Has Not Made “Publicly Available” Source Code That IBM Misappropriated

SCO provided customers who purchased Linux Server 4.0 with access to the product (including source code that is part of the Linux 2.4 kernel) through a confidential and individualized password for use at the log-in screen to SCO’s website. Sontag Decl. (11/30/04) ¶ 19 (Exh. 49). SCO made and will continue to make that source code available in that way through December 31, 2004, in light of SCO’s continuing contractual obligations. Id. ¶ 17. In its Product Announcement for Linux Server 4.0 (dated November 19, 2002), SCO promised to offer purchasers the “SCO Linux Update Service” for twelve months, including “Access to an up-to-date repository of UnitedLinux and other updates for their system.” Id. Exh. E. When a purchaser registered at SCO’s website for the SCO Linux Update Service, the purchaser was

115

informed of its entitlement to “download and apply the latest updates” to its SCO Linux Server. Id. Between October 31 and December 1, 2003, IBM repeatedly accessed the SCO log-in site but did not obtain access to the SCO Linux Server 4.0 files. Id. 25. After news of a bug in the website’s security system was reported on the internet, however, IBM exploited the bug to bypass the security system, hacked onto SCO’s website, and downloaded the very files that IBM has now attached to its motion. Id. ¶ 25-27. 65 Again, in view of all the circumstances, especially as they are construed in the light most favorable to SCO, the fact that IBM impermissibly accessed source code that SCO made available only to its pre-existing customers - and only pursuant to what SCO believed were its contractual obligations to those customers - cannot remotely establish SCO’s intentional waiver of its breach-of-contract claims against IBM as a matter of law. III. THE COURT MAY ALSO DENY, OR CONTINUE, IBM’S MOTION UNDER FEDERAL RULE OF CIVIL PROCEDURE 56(f) For all of the reasons demonstrated above, the law and record evidence developed to date are more than sufficient to require the denial of IBM’s motion; numerous genuine issues of material fact already exist. The Court may also continue or deny IBM’s motion under Rule 56(f), however, because SCO “has not had an opportunity to make full discovery.” Celotex Corp. v. Catrett, 477 U.S. 317, 326 (1986).

65

Section 103 0(a)(2)(C) of Title 18 of the United States Code, known as “The Computer Fraud and Abuse Act,” makes it a felony for an unauthorized user to access a company’s website. See, e.g., Creative Computing v. GetLoaded.com LLC, 386 F.3d 930, 933-35 (9th Cit. 2004); Theofel v. Farey-Jones, 359 F.3d 1066, 1078 (9th Cir. 2004); I.M.S. Inquiry Mgmt. Sys., Ltd. v. Berkshire Info. Sys., Inc., 307 F. Supp. 2d 521, 523-24, 526 (S.D.N.Y. 2004) (citing cases); see also SCO’s Mem. in Opp. to IBM’s Motion for Summary Judgment on Its Eighth Counterclaim (Nov. 30, 2004).

116

Because summary judgment “is drastic and should be applied with caution to the end that litigants will have a trial on bona fide factual issues,” Morrison,.340 F.2d at 432, Rule 56(f) “should be applied with a spirit of liberality” as “an additional safeguard against an improvident or premature grant of summary judgment.” King Airway Co. v. Routt County, No. 97 CJ C.A.R. 563, 1997 WL 186256, at *5 (10th Cir. Apr. 17, 1997) (Exh.1). Factors that favor 56(f) relief include the moving party’s exclusive control of information essential to oppose the motion, Price v. W. Res., Inc., 232 F.3d 779, 783 (10th Cir. 2000), and the fact the nonmoving party has been delayed in obtaining such information, Am. Maplan Corp. v. Heilmayr, 165 F. Supp. 2d 1247, 1254-5.5 (D. Kan. 2001). That relevant discovery remains outstanding also strongly militates for Rule 56(f) relief. See, e.g., Holt v. Wesley Med. Ctr., LLC, No. 00-1318-JAR, 2002 WL 31778785, at *1 (D. Kan. Nov. 25, 2002) (Exh. J) (pending, unanswered discovery requests); Procter & Gamble Co. v. Haugen, 179 F.R.D. 622, 633 (D. Utah 1998) (finding motion for summary, judgment “premature” where evidence had not been “fully developed”); Am. Maplan Corp., 165 F. Supp. 2d at 1254 (noting “the discovery period in this case has not yet closed”). A. The Outstanding Discovery That SCO Will Develop Bears Directly on IBM’s Contract-Interpretation Argument In its prior memoranda, supporting affidavits, and oral arguments in this case, 66 SCO has detailed its diligent efforts to obtain the discovery summarized below, which SCO reasonably believes will further bear out additional genuine issues of material fact precluding summary judgment on IBM’s contract-interpretation argument.

66

To avoid burdening the Court with repetition of its prior arguments, SCO hereby incorporates those prior submissions and arguments by reference, specifically including the memoranda, affidavits, and argument concerning SCO’s opposition to IBM’s motion for summary judgment on its Tenth Counterclaim, SCO’s 56(f) motion concerning the same, and SCO’s motion to enforce the scheduling order, which were all submitted to this Court, as well as the submissions concerning SCO’s memorandum regarding discovery and renewed motion to compel, the discovery motions currently pending before Magistrate Judge Wells. See Normand Decl. (11/30/04) ¶ 4.

117

1.

Depositions of IBM Declarants

SCO has not yet deposed several of the witnesses whose declarations IBM submitted with its summary judgment motion. If the Court were to decide that it can even properly consider extrinsic evidence on IBM’s motion, then the Court must consider all such evidence. It is well-settled that untested affidavits may not warrant summary judgment. See, e.g., G-I Holdings v. Baron & Budd, 213 F.R.D. 146,149 (S.D.N.Y. 2003) (denying summary judgment on the ground that “untested affidavits are not acceptable”); Messina v. Mazzeo, 854 F. Supp. 116, 141-42 (E.D.N.Y. 1994) (holding that “it cannot be said with certainty that a deposition” by nonmoving party would not “uncover material issues of disputed facts,” leading the court to rule that “summary judgment is premature”). These principles apply with even greater force here, where SCO has already discovered substantial evidence undermining the credibility of IBM’s key declarants. In light of all the evidence that already materially contradicts IBM’s contract interpretation, SCO reasonably expects that the depositions of IBM’s remaining declarants will give rise to further genuine issues of material fact. Normand Decl. (11/30/04) ¶ 5. Moreover, SCO reasonably expects that those depositions will permit SCO to identify other contract witnesses, including former IBM and Sequent witnesses who can provide admissions concerning the parties’ intent in executing the software agreements. Id. ¶ 6. 2. Additional Declarations IBM Has Refused to Produce

At the time IBM filed its motion for summary judgment on its contract claims, IBM had in its possession testimony that squarely contradicts its motion. SCO has learned that, in addition to the prior testimony of Messrs Wilson and Frasure in the BSD case, IBM has had in its exclusively possession a declaration from Martin Pfeffer, the former AT&T General Attorney,

118

who has directly refuted IBM’s interpretation of the software agreements. See Pfeffer Decl. (9/7/04) ¶ 13 (Exh. 8). In previously withholding witness declarations, 67 IBM has taken the position that any such declarations constitute protected attorney work product. IBM’s selective disclosure of its declarations, however, has effected a waiver of any purported work-product protections that may have otherwise attached. See Quark, Inc. v. Harley, 141 F.3d 1185, 1998 WL 161035, at *3 (10th Cir. Mar. 4, 1998) (Exh. K) (“The district court’s ruling that the attorney-client and work product protections are waived by selective disclosure comports with the law of this circuit.”); Coleco Indus., Inc. v. Universal Studios, Inc., 110 F.R.D. 688, 691 (S.D.N.Y. 1986) (selective disclosure “constitutes an implied waiver of all work product relevant to the same issue,” because the work-product “privilege impedes a fair adversarial presentation of the facts when it is involved to avoid disclosure of anything other than a necessarily biased presentation of the relevant facts”); Frontier Ref., Inc. v. Gorman-Rupp Co., 136 F.3d 695, 704 (10th Cir. 1998) (“[A] litigant cannot use the work product doctrine as both a sword and a shield by selectively using the privileged documents to prove a point but then invoking the privilege to prevent an opponent from challenging the assertion.”). Indeed, IBM’s reliance on selectively disclosed declarations to support its motion for summary judgment is particularly unfair. 68 In light of IBM’s motion, SCO has reiterated its request for all affidavits, declarations, and witness statements in IBM’s possession, custody or control concerning any subject matter on

67

IBM has held back witness statements until IBM used them with its dispositive motions, or else first produced them at the beginning of a declarant’s deposition to have the witness adopt the declaration as his testimony. See, e.g., Saint-Gobain/Norton Indus. Ceramics Corp. v. Gen. Elec. Co., 884 F. Supp. 31, 33 (D. Mass. 1995) (“Courts have recognized that it would be fundamentally unfair to allow a party to disclose opinions which support its position and to simultaneously conceal those that are unfavorable or adverse to its position.”); Diamond v. Mohawk Rubber Co., 33 F.R.D. 264, 268 (D. Colo. 1963) (ordering defendant to disclose witness statements where defendant enjoyed unfair advantage by withholding facts essential to preparation of plaintiffs case).

68

119

which IBM has submitted declarations in support of its summary judgment motions, including any declarations concerning the meaning, intent, or interpretation of the IBM and Sequent software agreements. Exh. 55. IBM not only has declined to produce any such material, but has not even responded to SCO’s letter requesting it. Particularly in light of the evidence that SCO has developed in discovery - and IBM’s withholding of its declaration from Mr. Pfeffer - SCO reasonably expects that production of IBM’s witness declarations on the subject matter of IBM’s motion will lead to further evidence raising issues of fact precluding summary judgment. Normand Decl. (11/30/04) ¶ 9. 3. Additional License-Related Documents

On October 8, 2004, SCO subpoenaed Novell and Paul, Hastings, Janofsky & Walker LLP (former counsel to the plaintiff in the BSD case) for documents concerning that case and the former AT&T software license agreements. After producing two boxes of documents concerning the BSD case on November 11, on November 15, Novell informed SCO counsel in New York that documents related to the software agreements would be available for inspection in Provo, Utah, on Friday, November 19. Accordingly, SCO has not yet been able to review those materials. Based on the evidence that SCO has developed thus far, however, SCO again reasonably believes that the documents that Novell has only recently made available will bring to light further evidence contradicting IBM’s contract-interpretation argument. Normand Decl. (11/30/04) ¶ 10. 4. Programming-History Evidence

SCO has already obtained through discovery IBM’s admission that AIX is “derived” from and “based on” UNIX System Five, and Sequent’s admission that Dynix is a “derivative work” of UNIX System Five, and IBM’s admission that Linux is derived from UNIX. Moreover, SCO has demonstrated in detail to this Court and to the Magistrate Court how the
120

discovery SCO has long sought will further permit SCO to show that the particular source code that IBM contributed to Linux was covered by the broad protection provided by the software agreements. See, e.g., Hr’g Tr. (10/19/04) at 8-16, 70-72 (argument before Magistrate Court concerning relevance of outstanding programming-history discovery to SCO’s contract claims). In the affidavit of Michael Davidson, SCO further illustrates how IBM’s withholding of evidence concerning the programming history of AIX and Dynix has precluded SCO from proving with specificity how programs that IBM contributed to Linux are derived from and based on programs in UNIX. SCO can demonstrate, for example, that the Journaling Filing system (“JFS”) in Linux, which originated in UNIX, is derived from the JFS in versions of AIX that IBM has produced; but due to IBM’s failure to produce critical predicate discovery, SCO cannot yet show precisely how the JFS in earlier versions of AIX were derived from the JFS in UNIX. See Davidson Decl. (11/22/04) ¶¶ 10-52 (Exh. 58). Even counsel for IBM has appeared to acknowledge the relevance of the outstanding programming-history discovery to SCO’s contract claims. At the September argument before this Court on IBM’s motion for summary judgment on IBM’s Tenth Counterclaim, counsel for IBM argued to the Court that the outstanding programming-history evidence is not discoverable with respect to IBM’s copyright claim even though the “road map” provided by the outstanding discovery would permit SCO to focus its copyright investigation by tracing IBM’s copyright infringement from the original UNIX product to the materials in AIX and Dynix that IBM ultimately dumped into Linux. Putting aside the illogic of IBM’s position on the copyright discovery issue, however, what is important for present purposes is that in attempting to argue that such discovery was not relevant to IBM’s copyright claim, IBM’s counsel conceded the important distinction between that claim and SCO’s contract claims:

121

“And the only road map offered, Your Honor, is a road map that relates only to AIX and Dynix, most of which you heard in SCO’s recitation of the discovery that it needs and what it would do with that discovery, that merely conflates SCO’s contract claims with the copyright claims.” Hr’g Tr. (9/15/04) at 123 (emphasis added). B. SCO Is Also Entitled to Additional Discovery Concerning IBM’s “Waiver” Arguments In addition, substantial outstanding discovery concerning IBM’s “waiver” claims would permit SCO to obtain further evidence demonstrating the existence of genuine issues of material fact precluding summary judgment. First, with respect to IBM’s claim that Novell has waived SCO’s intellectual-property rights, SCO intends to take the depositions of other participants in the negotiations of the 1995 APA and of the October 1996 Amendment No. 2 thereto. See Normand Decl. (11/30/04) ¶ 11. SCO reasonably expects that the testimony of the other participants will be consistent with the declarations that SCO has submitted from the chief negotiators (Ed Chatlos, Jim Wilt, and Steve Sabbath) and will further confirm that the parties never intended for Novell to have the right to waive, or to direct or require SCO to waive, any of SCO’s intellectual-property protections under the SVRX software agreements. Id. Second, with respect to IBM’s claim that SCO has waived its own rights, SCO intends to take depositions of, among others, the IBM employees responsible for accessing SCO’s password-protected website - the sole ground for IBM’s unfounded contention that SCO made certain source code available to “the public.” Id. ¶ 12. SCO reasonably expects that testimony to confirm SCO’s evidence, submitted herewith, that SCO did not make the source code on its website available to the “public,” but rather that IBM obtained access to that code only by improperly by-passing SCO’s log-in procedure and hacking onto SCO’s website. Id.

122

Accordingly, independent of the grounds for denying the motion set forth in Parts I and II above, IBM’s motion also fails under Rule 56(f). CONCLUSION SCO respectfully submits, for the reasons set forth above, that the Court should deny IBM’s motion for summary judgment on SCO’s contract claims.

123

DATED this 30th day of November, 2004.

HATCH, JAMES & DODGE, P.C. Brent O. Hatch Mark James BOLES, SCHILLER & FLEXNER LLP Robert Silver (admitted pro hac vice) Steven N. Zack (admitted pro hac vice) Edward Normand (admitted pro hac vice) Sean Eskovitz (admitted pro hac vice) Attorneys for The SCO Group, Inc.

124

CERTIFICATE OF SERVICE Plaintiff, The SCO Group, Inc. hereby certifies that a true and correct copy of SCO’S MEMORANDUM IN OPPOSITION TO IBM’S MOTION FOR SUMMARY JUDGMENT ON SCO’S BREACH-OF-CONTRACT CLAIMS was served on Defendant International Business Machines Corporation on this 30th day of November, by depositing it in U.S. Mail, first class, postage prepaid, to their counsel of record as indicated below: Evan R. Chesler, Esq. Cravath, Swaine & Moore LLP Worldwide Plaza 825 Eighth Avenue New York, NY 10019 Donald J. Rosenberg, Esq. 1133 Westchester Avenue White Plains, New York 10604 Alan L. Sullivan, Esq. Todd M. Shaughnessy, Esq. Snell & Wilmer L.L.P. 15 West South Temple, Ste. 1200 Gateway Tower West Salt Lake City, Utah 84101-1004 Attorneys for Defendant/Counterclaim Plaintiff IBM Corp.

By:


								
To top