[case caption omitted]
MICROSOFTS MEMORANDUM IN OPPOSITION TO
The antitrust claims asserted by Bristol are nothing more than transparent "cover" for Bristols request that this Court intervene in what were ongoing contract negotiations and write and award Bristol a contract on terms it could never obtain at the bargaining table. Indeed, by this action Bristol seeks to obtain a much better deal than what its chief competitor, Mainsoft Corporation ("Mainsoft"), agreed to in recent negotiations with Microsoft, and thus to obtain a significant competitive advantage over Mainsoft.
There is no conduct implicating the antitrust laws even involved in this case. Bristol asserts a Section 2 "monopolization" claim, arguing (1) that Microsoft has "refused" to supply Bristol with source code (Microsofts confidential intellectual property), (2) that this "refusal" threatens Bristol with """"imminent" destruction as an on-going business, and (3) that if Bristol goes out of business, competition between Microsoft and companies that sell UNIX operating systems might be adversely affected in the markets for operating systems for technical workstations and operating systems for servers -- markets in which Bristol is not present -- thereby (4) perhaps facilitating Microsofts threatened monopolization of those markets.1
Not one of these assertions is true or could be supported at a hearing (if one is deemed necessary) with evidence. Moreover, even if all of them were true, Bristol would still be required to prove that Microsoft engaged in some predatory, anticompetitive conduct, which Bristol cannot do. There is simply no basis for any of Bristols claims.
In September 1994, Bristol and Microsoft entered into a contract pursuant to which Microsoft was obligated -- for a period of three years -- to provide Bristol with "source code" for certain Microsoft operating systems. All agree that Microsofts contractual obligation to provide source code to Bristol expired in September 1997, and that the source code to be supplied under that contract was limited to now-old versions of Windows operating systems. Indeed, Bristol does not even attempt to assert any breach of contract claim.
Even before the September 1997 expiration date, Microsoft and Bristol began negotiating the terms of a new contract that, if executed, would provide for Microsofts delivery to Bristol of source code for new Microsoft operating systems, namely Windows NT 4.0 and Windows NT 5.0 (which was and still is under development). Contrary to the entire premise of Bristols claims -- i.e., the assertion that Microsoft has "refused" to provide new source code to Bristol -- Microsoft has repeatedly offered to give Bristol source code for Windows NT 4.0 and Windows NT 5.0 pursuant to a new contract. Bristol declined those offers, choosing instead to negotiate through litigation.
Bristol now seeks the extraordinary remedy of a mandatory injunction: it asks the Court to order Microsoft to turn over its source code for Windows NT 4.0 and NT 5.0 --Microsofts confidential trade secrets and copyrighted property of huge value to Microsoft -- even before a trial on the merits. (All agree that source code for NT 4.0 and NT 5.0 was never part of the parties original 1994 agreement.) Bristol also asks the Court to write a new software licensing contract giving Bristol more favorable terms than it has so far negotiated, and more favorable than the terms to which Bristols competitor Mainsoft willingly agreed after arms-length negotiations with Microsoft.
The preliminary injunction Bristol seeks will not preserve the status quo; it will radically alter it. The Second Circuit has repeatedly cautioned that such injunctions should be granted in only the rarest of cases. In addition, Bristol faces no imminent or irreparable harm in the absence of an injunction. The state of affairs today is exactly the same as it has been for a long time; Bristol concedes that it has received no new Microsoft source code since no later than the first quarter of 1997.
Indeed, Bristols own documents show that this litigation is a strange form of brinkmanship that Bristol has been
Bristols delay in seeking any redress strongly refutes its claim for a preliminary injunction.
Moreover, Bristols claim that it will go out of business if it does not receive Microsofts NT 4.0 and NT 5.0 source code immediately, before a trial on the merits, is demonstrably false. Even after Bristol commenced this suit, it announced to the public at large that the lack of access to Microsofts new source code should not adversely affect Bristols business until well into 1999 or 2000 and that Bristol can vigorously compete with Mainsoft even in the absence of any new Microsoft code. Notwithstanding all that, Bristol now makes the incredible claim that a preliminary injunction is justified not because of anything Microsoft has done or is about to do -- since Microsoft continues to be willing to give Bristol source code for Windows NT 4.0 and NT 5.0 under the terms of a new license -- but because Bristols very filing of this action may itself have adverse effects on Bristols business. Even leaving aside the fact that Bristols public statements refute any such notion, such a self-inflicted injury cannot form the basis for an injunction, and the courts have so held.
All of Bristols claims are legally insufficient. Bristol cannot show any injury to competition -- which is required for each of its federal and state antitrust claims -- because Mainsoft (Bristols primary competitor) and others provide the same service to the market that Bristol provides. Thus, even if Bristol were to close its doors tomorrow, Bristols customers and others in the software industry would have abundant alternative means for running applications on both Windows and UNIX platforms. This means that Bristols presence is not even remotely essential for sellers of UNIX operating systems to compete with Microsoft. In any event, competition between UNIX systems and Windows systems would be very robust even without the existence of any cross-platform tools. Bristol simply cannot demonstrate any injury of the sort that the antitrust laws seek to prevent, and it also lacks antitrust standing to assert its claims.
Moreover, Bristols antitrust claims are meritless on other grounds as well. Bristol relies primarily on three narrow and largely discredited theories -- "monopoly leveraging," "refusal to deal" and denial of an "essential facility" -- and seeks to stretch them far beyond their previous applications. First, Bristol cannot point to any affirmative Microsoft misconduct -- such as tying, predatory pricing or contracts in restraint of trade -- that constitutes "leveraging" of monopoly power or attempted monopolization. Bristols so-called "refusal to deal" claims are legally deficient because, among other things, Microsoft has not refused to deal; it has offered NT 4.0 and NT 5.0 source code to Bristol, but Bristol (unlike Mainsoft) wants to get that Microsoft property on terms Microsoft is not willing to accept. Bristols attempt to rely on the "essential facility" doctrine fails as well. That doctrine, which was designed for certain types of monopolies not present here (e.g., publicly subsidized monopolies or monopolies resulting from conditions of nature or geography), is inapplicable because, among other reasons, Bristol is not a competitor of Microsoft.
Nor can Bristol properly rely on principles of promissory estoppel to force Microsoft to turn over its NT 4.0 and NT 5.0 source code on terms dictated by Bristol. Bristols 1994 contract unambiguously provided that Microsoft was obligated to deliver source code for specified older Windows products for only three years, until September 1997. The contract was clear on its face with respect to that issue, and it also contained an "integration" clause and a "no oral modification" clause. Extra-contractual "assurances," especially of the general and vague nature that Bristol refers to here, cannot form the basis for an estoppel claim as a matter of law. Even if that were not the case, it would have been inherently unreasonable for Bristol to rely on such "assurances" in view of the express terms of an integrated contract that permitted only written modifications signed by both parties.
Bristol has alleged twelve antitrust claims (under the Sherman Act and the Connecticut Antitrust Act), a claim under the Connecticut Unfair Trade Practices Act, and a promissory estoppel claim. Bristol asserts that Microsoft has a monopoly in the "market" for personal computer ("PC") operating systems, but this "market" is not directly relevant to Bristols claims. Instead, Bristol alleges that its Section 2 claims pertain to the purported "markets" for operating systems for (a) "technical workstations" and (b) "departmental servers." In those "markets," according to Bristol, Microsoft has "market" shares of approximately 30% and 45%, and UNIX operating systems have shares of about 70% and 22%, respectively. (Declaration of Richard N. Langlois, executed on August 17, 1998 ("Langlois Decl."), Tables 2, 3.) The competition allegedly injured by the Microsoft conduct at issue is competition in these two markets between Microsofts NT operating systems and a variety of UNIX operating systems; Bristol does not allege any antitrust injury to competition by Bristol or any product it makes.
Of course, Bristol does not make or sell operating systems itself. (See Bristols Memorandum in Support of Motion for Preliminary Injunction, dated August 18, 1998 ("Bristol Mem.") at 44-45.) Bristol makes and sells, among other things, a software product called "Wind/U." Wind/U is a "cross-platform tool" that enables software developers who choose in the first instance to write computer programs for a Microsoft operating system (such as Windows 95 or Windows NT) to "port" (or translate) those programs to specific UNIX operating systems. The buyer of Bristols product thus need not write its software program from scratch a second time. Microsoft does not compete in Bristols market -- that is, Microsoft does not sell a "cross-platform tool" that enables software developers to port programs (also known as "applications") written for Windows to UNIX.
Microsoft and Bristol are thus not competitors. Bristol nevertheless claims that it "facilitates" competition between Windows NT and UNIX operating systems because, without Bristols "porting" program, operating systems customers might not buy UNIX for fear that programs written for Windows NT would not be available on UNIX. This makes no sense for at least five reasons: (1) Mainsoft, Bristols primary competitor, provides the same "portability" service that Bristol sells, meaning that if Bristol goes out of business tomorrow, software written for Windows NT could still be ported to UNIX; (2) many software developers write for both Windows NT and UNIX in the first instance, and do not use any "cross-platform tool" at all; (3) "emulator" products allow Windows-based applications to run on UNIX by a different method than that employed by Bristol and Mainsoft; (4) Sun Microsystems, a large Microsoft competitor, offers software developers the option of writing programs in a new language -- Java -- and, by doing so, their programs will run on both Windows NT and UNIX operating systems; and (5) software developers are free to, and often do, write programs directly for UNIX alone. Any one of these factors establishes that there can be no injury to competition even if Bristol goes out of business.
Bristol was formed by Keith Blackwell, Jean Blackwell (his wife) and Ken Blackwell (Keiths brother) in 1991. Shortly thereafter -- before it had any association with Microsoft or any access to any of Microsofts proprietary source code -- Bristol developed Wind/U and brought it to market. (Declaration of W. Keith Blackwell, executed on August 18, 1998 ("Keith Blackwell Decl."), ¶¶ 29-30.) Wind/U is a "cross-platform tool" designed to translate applications written for Microsoft operating systems (such as Windows NT) to run on several types of "UNIX" operating systems. (See Bristol Mem. at 4-5.) 3
In 1993 and 1994, Microsoft began to negotiate agreements with various software companies, pursuant to which Microsoft would license certain "source code" 4 for particular Windows operating systems to assist those companies in developing or enhancing cross-platform tools. Microsoft referred to this licensing program as the Windows Interface Source Environment ("WISE"). (Bristol Mem. at 5.)
On September 21, 1994, Microsoft and Bristol executed a WISE license (the "1994 Contract") in which Microsoft agreed, in exchange for royalties, to provide Bristol with source code for certain identified Microsoft operating systems, as well as various updates and specified new versions of those operating systems.5 The 1994 Contract unambiguously provided
that Microsofts obligation to deliver that source code to Bristol expired on September 21, 1997 (although Bristol was entitled to continue thereafter using code provided during the three-year period, and was obligated to continue paying royalties with respect to its use of that code). (1994 Contract (Martin Aff., Ex. 1) at §§ 3(b), 7(a).)
As both sides agree, the contract provided that, with respect to Windows NT, Microsoft would only give Bristol the source code for certain versions of Windows NT 3. (See 1994 Contract (Martin Aff., Ex. 1) at §§ 1(w), 3(b).) As Keith Blackwell
The 1994 Contract contained the following "integration" and "no oral modification" clauses:
Upon execution by both parties, this Agreement shall constitute the entire agreement between the parties with respect to the subject matter hereof and merges all prior and contemporaneous communications. It shall not be modified except by a written agreement signed on behalf of Bristol and Microsoft by their respective duly authorized representatives.
(1994 Contract (Martin Aff., Ex. 1) at § 23.)
Microsoft also entered into a WISE license with Mainsoft in 1994. (See Martin Aff., Ex. 5.) The Mainsoft contract provided that Microsoft would provide a similar set of source code to Mainsoft, and also for a period of only three years. (Mainsoft 1994 contract (Martin Aff., Ex. 5) at §§ 1(z), 3(b).) Robert Kruger, the Microsoft employee who negotiated the WISE licenses, testified that Microsoft did not wish to favor either Bristol or Mainsoft over the other in these licenses. (Deposition of Robert Kruger (Martin Aff., Ex. 6) at 201-02.) Bristol itself recognized that it was in its interest to sign a WISE license, because doing so
Pursuant to the 1994 Contract, Microsoft provided source code for the identified Windows operating systems to Bristol. Microsoft delivered the last of that source code to Bristol no later than early 1997. (Keith Blackwell Decl. ¶¶ 61, 75.)
The parties began to negotiate a new license agreement in early 1997, this one meant to cover source code for Windows NT 4.0 and NT 5.0, which
was not covered by the 1994 Contract.
During the same period, Microsoft was also discussing a new agreement with Mainsoft. As it did in 1994, Microsoft desired to maintain consistency between the two deals. Microsoft was not willing to provide all of the source code for the new Windows NT 4.0 and NT 5.0 products to either Bristol or Mainsoft. (As one might expect, Windows NT 4.0 offers, and NT 5.0 (which is not yet on the market) will offer, many more features and functions than the older operating systems covered by the 1994 Contract.) In November 1997, Keith Blackwell, Bristols President, assured Microsoft that Bristol would not
This Bristol proposal was then used by Microsoft -- -- as the basis for Microsofts negotiations with Mainsoft as well. (Martin Aff., Ex. 10.)
On August 25, 1998, Microsoft and Mainsoft executed a new WISE license. (See Martin Aff., Ex. 11.)
On August 18, 1998 --
-- Bristol filed this action and its preliminary injunction motion. On that very day, before knowing of the litigation, Microsoft Senior Vice President Jim Allchin again sent a letter to Bristol seeking a face-to-face meeting to work out the terms of a new agreement. (Martin Aff., Ex. 17.) And despite the anti-Microsoft publicity campaign that Bristol has launched and the burden that this litigation has already imposed -- particularly in light of the approaching trial in the Department of Justices unrelated case against Microsoft -- Microsoft is still willing to offer Bristol the same material terms as those in its 1998 contract with Mainsoft. (Martin Aff., Ex. 18.)
Instead, Bristol seeks a better deal, representing to the Court that, unless it gets Microsofts NT 4.0 and NT 5.0 source code immediately (and on the terms Bristol dictates), it may "go under." Of course, Bristol can save itself from extinction merely by agreeing to the terms that its competitor Mainsoft has found acceptable. More fundamentally, Bristols tale of threatened imminent destruction is untrue.
For example, Bristol has insisted to the Court that, without a preliminary injunction pending trial on the merits,
On August 26, during the first conference with the Court, Bristols counsel represented: "When we filed the lawsuit [on August 18] . . . Bristols business came to a halt within 24 hours. . . . [I]n the week[ ] since we filed the lawsuit, Bristols business has been in a state of paralysis." (August 26, 1998 Tr. at 21-22.) Counsel further urged the Court to schedule a prompt hearing on the preliminary injunction motion "before our client goes under." (Id. at 35.)
Counsels statements to the Court were incorrect. In fact, as Bristol has represented on its Internet website, Bristols business has been largely unaffected by this suit. Bristol there says that its customers products -- both those "shipping with Wind/U and new customer applications under development using Wind/U" -- are not affected by Bristols lawsuit. (Martin Aff., Ex. 19 at 2.)
Bristol has also stated that its "strategic relationships" with IBM, Compaq and other companies "will not be affected," and that its Wind/U product "was recently upgraded" and "will continue to evolve" hereafter. (Martin Aff., Ex. 19 at 2-3 (emphasis added).) At deposition earlier this week, Keith Blackwell affirmed under oath that
Moreover, Bristol has also acknowledged that
and has publicly assured its customers that
Bristols website confirms that there is no imminent crisis for Bristol if it cannot now obtain Windows NT 5.0 source code; it says that "Windows NT 5.0 does not have a projected ship date at this time, and is expected to be delayed well into 1999 (possibly 2000)." (Martin Aff., Ex. 19 at 2.) This is why Bristol informed customers in late August that
Thus, the assertions by Bristol in its papers and by its counsel at the August 26 conference that Bristol faces irreparable injury -- namely, its imminent destruction as a going concern -- are also false. As Bristol told its employees after it filed this suit
And, remarkably, just three days ago, Keith Blackwell testified unequivocally that,
In order to obtain a preliminary injunction, Bristol must demonstrate that it will suffer "irreparable harm" in the absence of an injunction and "either (1) a likelihood of success on the merits of its case or (2) sufficiently serious questions going to the merits to make them fair ground for litigation and a balance of hardships tipping decidedly in its favor." USA Recycling, Inc. v. Town of Babylon, 66 F.3d 1272, 1280-81 (2d Cir. 1995). The award of a preliminary injunction is an extraordinary remedy and will not be granted absent a clear showing that the moving party has met its burden of proof. See, e.g., Beech-Nut, Inc. v. Warner-Lambert Co., 480 F.2d 801, 803 (2d Cir. 1973).
Although Bristol cannot satisfy even these requirements, its burden is heavier in this case. The Second Circuit has made it clear that a higher standard applies where, as here:
"(i) an injunction will alter, rather than maintain, the status quo, or (ii) an injunction will provide the movant with substantially all the relief sought and that relief cannot be undone even if the defendant prevails at a trial on the merits." Tom Doherty Assocs., Inc. v. Saban Entertainment, Inc., 60 F.3d 27, 33-34 (2d Cir. 1995) (emphasis added). Although only one of these characteristics triggers the higher standard, the injunction Bristol seeks has both.
First, the preliminary injunction requested by Bristol would radically alter the status quo. There is no dispute that Microsofts obligation under the 1994 Contract to provide old versions of source code expired in September 1997, and that the contract did not cover source code for Windows NT 4.0 or NT 5.0 (which is what Bristol now seeks). (See pp. 8-10, supra.) Yet the proposed injunction would compel Microsoft to turn over to Bristol those confidential, copyrighted trade secrets before the merits of the case are decided. 7 Accordingly, Bristols injunction can be granted "only upon a clear showing that [Bristol] is entitled to the relief requested, or where extreme or very serious damage will result from a denial of preliminary relief." Wali, 759 F.2d at 1025 (emphasis added) (internal citations and quotations omitted); see also Topps Chewing Gum, Inc. v. Major League Baseball Players Assn, 641 F. Supp. 1179, 1190 (S.D.N.Y. 1986) (mandatory injunctions "are not issued in doubtful cases").
Second, the requested preliminary injunction would give Bristol all the ultimate relief sought in the case, and could not be undone if Microsoft prevails at trial. The Second Circuit has observed that the paradigmatic example of such an injunction is one seeking "the disclosure of confidential information." Tom Doherty Assocs., 60 F.3d at 35. Bristols request is exactly that. Further, Bristol acknowledges that once it receives Microsofts source code, its product will become "irretrievably locked into dependence on" the code. (Keith Blackwell Decl. ¶ 60.) Simply stated, if Microsoft prevails on the merits, it will be impossible for Bristol to extract those portions of its product that are based on Microsofts source code. Moreover, once the Microsoft code is revealed to Bristol, there is no way to erase Bristols knowledge of it. FMC Corp. v. Taiwan Tainan Giant Indus. Co., 730 F.2d 61, 63 (2d Cir. 1984) ("A trade secret once lost is, of course, lost forever.") Thus, for this reason as well, Bristol must satisfy the "higher standard of substantial, or clear showing of, likelihood of success" on the merits. Tom Doherty Assocs., 60 F.3d at 35; accord Koppell v. New York State Bd. of Elections, __ F.3d __, 1998 WL 546604, at *1 (2d Cir. Aug. 28, 1998).
II. BRISTOL IS NOT FACED WITH IMMINENT IRREPARABLE INJURY IN THE ABSENCE OF A PRELIMINARY INJUNCTION.
The Second Circuit has observed that a showing of imminent, irreparable harm is "the single most important prerequisite for the issuance of a preliminary injunction." Reuters Ltd. v. United Press Intl, Inc., 903 F.2d 904, 907 (2d Cir. 1990) (internal citations and quotations omitted). Bristol cannot show that it will imminently suffer any irreparable injury if the requested preliminary injunction is not granted. Indeed, Bristols own conduct and statements conclusively establish that it will not.
First, Bristol acknowledges that it has received no source code from Microsoft since
the first quarter of 1997. (Keith Blackwell Decl. ¶¶ 61, 75.) Nothing has changed in
this respect over all that time. Yet Bristol waited these 18 months -- with no access to
any of the new Microsoft source code that Bristol now says is crucial to its survival --
before seeking its "emergency" relief. Moreover, unlike the usual preliminary
injunction plaintiff who seeks judicial relief immediately upon learning of the threat of
imminent injury, Bristol
This alone demonstrates the lack of any imminent, irreparable injury. See, e.g., Citibank, N.A. v. Citytrust, 756 F.2d 273, 276-77 (2d Cir. 1985) (preliminary injunction vacated because delay in bringing suit "tends to indicate at least a reduced need for such drastic, speedy action," and "suggests that there is, in fact, no irreparable injury"); National Football League Players Assn v. National Football League Properties, Inc., No. 90 Civ. 4244, 1991 WL 79325, at **3-4 (S.D.N.Y. May 7, 1991) (delay "demonstrates that [movant] is not facing the kind of irreparable damage that is required before a preliminary injunction will be issued"); Capital Cities Communications v. Twentieth Century Fox Film Corp., No. N-83-464, 1983 WL 3309, at *5 (D. Conn. Oct. 18, 1983) (delay "undercuts the sense of urgency" and "suggests that there is, in fact, no irreparable harm").
In addition, to sustain a claim that it is about to go out of business, Bristol "must offer detailed information about its finances" and a "quantitative analysis of [its] lost business" to show that its inability to continue as a going concern is imminent; mere conclusory statements of the sort offered by Bristol are not sufficient. Paramount Brands, Inc. v. Peerless Importers, Inc., No. CV-92-5190, 1992 WL 355601, at *5 (E.D.N.Y. Nov. 19, 1992) (citing Auto Sunroof of Larchmont, Inc. v. American Sunroof Corp., 639 F. Supp. 1492, 1494 (S.D.N.Y. 1986) (Weinfeld, J.)); accord Anti-Monopoly, Inc. v. Hasbro, Inc., No. 94 Civ. 2120, 1994 WL 202730, at **1-2 (S.D.N.Y. May 23, 1994); National Football League Players Assn, 1991 WL 79325, at *4 (movant "has not shown that it is facing imminent financial ruin").8
As shown above (pp. 12-14, supra), Bristols own statements contradict any assertion that Bristols survival depends on obtaining immediately new Microsoft source code. For example, Keith Blackwells declaration shows that, during the past 18 months that Bristol has been without any new code, Bristol has prospered. (See Keith Blackwell Decl. at 42.) Indeed, Bristol stated publicly that it "ended Q1 1998 with the best first quarter in the companys history." (Martin Aff., Ex. 23.) Furthermore, Keith Blackwell testified that he expects Bristols revenues for the third quarter (ending September 30, 1998) to be similar to Bristols second quarter 1998 revenues. (Keith Blackwell Dep. (Martin Aff., Ex. 3) at 132-33.)
Bristol has also made it clear that there is no urgency to its request for source code because, as Keith Blackwell said, "NT 5.0 is the only code that matters to our customers" (Martin Aff., Ex. 20 at 14), and NT 5.0 will not even be on the market until "well into 1999 (possibly 2000)." (Martin Aff., Ex. 19 at 2.) In fact, even without any new Microsoft source code, Bristol
Finally, Bristol is not a one-product outfit. As Bristol says on its website, Bristols products other than Wind/U -- HyperHelp, Xprinter, and Jprinter -- "are not affected" by this lawsuit or Bristols lack of access to Microsoft source code.9 (Martin Aff., Ex. 19 at 3.)
Most importantly, even if Bristol was able to establish that it cannot survive any longer without Microsofts new NT 4.0 or NT 5.0 source code, it would still not be entitled to a preliminary injunction because it is entirely within Bristols power to save itself; all it needs to do is accept Microsofts existing offer. Bristol concedes that this is true; although it says its business would suffer if it accepted Microsofts existing offer, it admits that this would occur "more slowly." (Bristol Mem. at 6.) In this regard, Judge Conners decision in Topps Chewing Gum is particularly instructive. In that case, a baseball card manufacturer (Topps) brought an antitrust action against the Major League Baseball Players Association ("MLBPA"), seeking to force the MLBPA to enter into a contract on Topps preferred terms. The court rejected Topps motion for a preliminary injunction because (among other reasons) Topps had failed to demonstrate irreparable harm. 641 F. Supp. at 1190-92. Judge Conner explained that "while [certain] aspects of the MLBPAs proposed deal may not be entirely satisfactory to Topps, the offer does remove the probability that Topps will suffer irreparable harm. . . . Thus, it appears that Topps can easily avoid the irreparable harm it claims it will suffer by accepting the offer the MLBPA has made. . . . Under these circumstances, it is inappropriate for the Court to order extraordinary relief." Id.
The same situation is presented here. Although Bristol may not like the terms of the contract proposed by Microsoft, it can avoid any irreparable harm by accepting those terms. (See Bristol Mem. at 6.) Further, Bristols argument that Microsofts offer is commercially unreasonable -- an argument that, even if true, does not satisfy the irreparable harm requirement in any event -- is not credible given Mainsofts agreement to very similar terms. See Topps, 641 F. Supp. at 1191.10
Wholly aside from the lack of merit of Bristols theories on monopolization and attempted monopolization, Bristol has not -- and cannot -- demonstrate any antitrust injury or antitrust standing. These two fundamental points completely doom Bristols antitrust claims.
As the Supreme Court explained in Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977), an antitrust plaintiff "must prove more than an injury causally linked to an illegal presence in the market." Instead, the Court held:
Plaintiffs must prove antitrust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants acts unlawful. The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation. It should, in short, be "the type of loss that the claimed violations . . . would be likely to cause."
Brunswick, 429 U.S. at 489 (quoting Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 125 (1969)).
As Bristol suggests, though never explains (see Bristol Mem. at 44), there are two often confused concepts within this requirement. Bristol must demonstrate both "antitrust injury" in the sense of harm to competition (i.e., "injury of the type the antitrust laws were intended to prevent") and "antitrust standing" in the sense that it is the proper party to bring such a claim. See Triple M Roofing Corp. v. Tremco, Inc., 753 F.2d 242, 247 (2d Cir. 1985) ("Antitrust standing and its terminological cousin, antitrust injury, are often confused.") Bristol completely fails to discuss antitrust injury, and its attempt to demonstrate antitrust standing is unavailing.
1. Antitrust Injury.
It is fundamental that the antitrust laws were enacted for "the protection of competition, not competitors." Brown Shoe Co. v. United States, 370 U.S. 294, 320 (1962). Here, although Bristol -- which is not a Microsoft competitor -- may suffer economic harm (as a result of its own choice not to sign a contract with Microsoft), there can be no harm to competition because Mainsoft and other companies provide the same capabilities as Bristol; they enable software developers who write programs ("applications") for Windows to "port" those applications to UNIX.
Thus, even if Bristol were correct that tools such as Wind/U are important to UNIXs ability to compete with Windows NT -- an assertion that Bristol will be unable to establish and which is, indeed, demonstrably incorrect -- Mainsofts product accomplishes the same objective.
Moreover, there are many other alternatives available to software developers for writing to both Windows NT and UNIX. First and most fundamentally, many software developers write applications themselves for both Windows and UNIX environments, without the use of any "cross-platform" tool, and of all the applications that run on both Windows and UNIX -- including those that employ a cross-platform tool -- the majority do so without any cross-platform tool like Bristols Wind/U. (See Kruger Dep. (Martin Aff., Ex. 6) at 190-95; Share Dep. (Martin Aff., Ex. 24) at 30-32.)11 Moreover, a new computer language called "Java," which is licensed by a large Microsoft competitor (Sun Microsystems), advertises that developers can "write once, run anywhere" (i.e., develop an application on Java that will run on both Windows NT and UNIX). (Kruger Dep. (Martin Aff., Ex. 6) at 188-90; Share Dep. (Martin Aff., Ex. 24) at 24-25, 27, 32-33.) There are also "emulators" available that allow Windows applications to run on UNIX. (Kruger Dep. (Martin Aff., Ex. 6) at 188; Share Dep. (Martin Aff., Ex. 24) at 18-21.) Finally, there are a number of companies that offer cross-platform tools that give software developers the option of writing applications for UNIX and then porting them to run on Windows operating systems. (Kruger Dep. (Martin Aff., Ex. 6) at 186-88; Share Dep. (Martin Aff., Ex. 24) at 25-27.)
Thus, contrary to Bristols statement that its customers will have "no choice except to offer only Windows application programs" (Bristol Mem. at 15), Bristols Wind/U product is merely one option of many in a thriving competitive market. (See Kruger Dep. (Martin Aff., Ex. 6) at 185-98; Share Dep. (Martin Aff., Ex. 24) at 26-27, 32-34.) Indeed, Bristol essentially concedes that neither it nor Wind/U is indispensable to applications developers or to competition between Windows NT and UNIX;
Accordingly, even if Bristols business declines or disappears, there would be no detrimental impact on competition between Windows NT and UNIX.
In this respect, Bristol is similar to a terminated distributor claiming antitrust violations by a manufacturer who has chosen a different distributor. Such claims have repeatedly been rejected because, although the terminated distributor undoubtedly suffers harm, its injuries are not related to any damage to competition. See, e.g., Electronics Communications Corp. v. Toshiba America Consumer Prods., Inc., 129 F.3d 240, 244-46 (2d Cir. 1997); Trans Sport, Inc. v. Starter Sportswear, Inc. 964 F.2d 186, 190-91 (2d Cir. 1992) ("we see no traces of the burdensome effect generally associated with illicit monopolist activity" based on defendants choice to limit number of retailers).
In addition to the absence of any harm to competition, Bristol does not have "antitrust standing" to challenge Microsofts conduct. In Associated General Contractors of California, Inc. v. California State Council of Carpenters, 459 U.S. 519, 535-45 (1983), the Supreme Court identified several factors that may be used to determine whether a plaintiff should be permitted to assert a private antitrust cause of action. One of the most important factors is whether the plaintiff is a consumer or competitor "in the market in which trade was restrained." Id. at 539. See generally 2 Phillip Areeda & Donald F. Turner, Antitrust Law ¶ 340 (1978) ("Areeda & Turner") ("indirect" competitors generally lack antitrust standing).
Bristol alleges that the relevant competition here is between Microsoft operating systems and UNIX (and other) operating systems. (See Bristol Mem. at 9-10.) Bristol does not sell an operating system, and it has no plans to do so. (See Bristol Mem. at 44-45.) Thus, Bristol does not and cannot claim that is a competitor of Microsoft in any of the operating system "markets" it identifies. This alone warrants dismissal of Bristols antitrust claims on standing grounds. See Associated General, 459 U.S. at 539; Automated Salvage Transport, Inc. v. Wheelabrator Environmental Sys., Inc., __ F.3d __, 1998 WL 515702, at *21 (2d Cir. Aug. 20, 1998) (standing requires that plaintiff "be a participant in the same market as the alleged malefactors").12
Bristols claimed status vis-à-vis Microsoft is analogous to that of a competitors supplier. It does not offer an alternative operating system, but rather helps to "supply" Windows-based applications that will run on UNIX operating systems by enabling software developers to port their Windows applications to UNIX. (See Bristol Mem. at 13-14, 44-45.) However, as the Second Circuit explained in Crimpers Promotions Inc. v. Home Box Office, Inc., 724 F.2d 290 (2d Cir. 1983), suppliers such as these do not have antitrust standing. Id. at 294 (citing 2 Areeda & Turner ¶ 340(e) at 217); accord Billy Baxter, Inc. v. Coca-Cola Co., 431 F.2d 183, 187-89 (2d Cir. 1970) (standing denied to plaintiff that licensed information and supplied ingredients to competitor of alleged monopolist); 2 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law¶ 375d (1995) ("Areeda & Hovenkamp") (antitrust victims suppliers generally lack antitrust standing).
Bristol attempts to overcome its lack of antitrust standing by advancing a strained, and incorrect, reading of the Second Circuits decision in Crimpers. In that case, defendants HBO and Showtime were networks that purchased movies from independent producers and then sent them to cable television operators around the country. 724 F.2d at 291. The plaintiff (Crimpers) produced a cable television trade show designed to bring together independent producers and cable operators. Id. The Second Circuit held that Crimpers had standing to challenge the defendants organized boycott of the trade show because Crimpers was a direct competitor of the defendants: "Crimpers was not just a supplier of a competitor, to whom standing is generally denied. It was endeavoring to forge a link in the chain of programming, to wit, direct contact between program producers and cable stations, that would compete with defendants in their role as middlemen." Id. at 294 (citing 2 Areeda & Turner § 340(e) at 217 (1978)); see also Areeda & Hovenkamp ¶ 375d (recognizing that Crimpers "competing alternative" to defendants service justified standing). Here, by contrast, Bristol does not compete with Microsoft; at most, it merely assists the ability of UNIX operating systems to compete with Microsoft systems.13
By Bristols theory, Microsofts "refusal" to extend a license to Bristol (at least on the terms Bristol demands) may cause damage to competition between Windows NT and UNIX operating systems. Thus, only sellers of rival operating systems -- not Bristol -- might have antitrust standing to assert the claims at issue here.
Bristol relies on three theories for its monopolization claims under Section 2 of the Sherman Act: monopoly leveraging, refusal to deal and denial of an essential facility.14 There are two elements of a Section 2 monopolization claim: "(1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident." United States v. Grinnell Corp., 384 U.S. 563, 570-71 (1966). Bristol cannot establish either of these elements.
The only "markets" posited by Bristol that are of any relevance to Bristols claims are those for technical workstation and departmental server operating systems, assuming that such separate markets even exist.15 Microsoft does not have monopoly power in these "markets." Even Professor Langlois, Bristols economist, asserts that for 1997, Microsofts market share of new operating systems for the technical workstation and departmental server "markets" was 27.8% and 43.7%, respectively, and its share of the server installed base (the only installed base figures provided) was only 15.5%. (Langlois Decl., Tables 2-4); see, e.g., AD/SAT v. Associated Press, 920 F. Supp. 1287, 1300 (S.D.N.Y. 1996) ("[m]any courts in the Second Circuit have found market shares of less than 50% or even 60% insufficient to support" even a showing of "dangerous probability" of acquiring a monopoly). Moreover, the projections about future market share upon which Professor Langlois relies -- which are of questionable reliability (see Schmalensee Decl. ¶ 30) -- do not suggest that UNIX is in danger of being displaced.16 Bristol agrees: its website endorses the view that "UNIX is still growing at a significant rate" and will continue to grow at least "through the year 2002" and even "well beyond that." (Martin Aff., Ex. 19 at 3, 5.)
To the extent that Microsoft is gaining market share in the workstation and server operating systems "markets," that gain -- according to Professor Langlois own reasoning -- must be attributable to the considerable benefits that Windows NT offers over UNIX and other systems. According to Professor Langlois, "tipping" and "network effects" would prevent Windows NT from gaining any significant market share from established UNIX systems unless Windows NT offered "considerably higher functionality." (Langlois Decl. ¶¶ 11, 13 (original emphasis); see also Schmalensee Decl. ¶¶ 21-29.)
Microsofts lack of monopoly power in either workstation or server operating systems alone dooms Bristols claims. Yet the Court need not rely solely on that ground to find that Bristol is not likely to succeed on the merits of its monopolization claims, because Bristol also cannot satisfy the second element of the claims -- often referred to as predatory, "exclusionary" or "anticompetitive" conduct. See, e.g., Areeda & Hovenkamp ¶ 650a at 66; Trans Sport, Inc. v. Starter Sportswear, Inc., 964 F.2d 186, 188 (2d Cir. 1992), affg, 775 F. Supp. 536, 541 (N.D.N.Y. 1992). As the Court appreciates, merely obtaining or maintaining a monopoly does not violate the antitrust laws. Rather, the alleged monopolist must take some impermissible, predatory action to obtain or perpetuate its monopoly. Bristol cannot point to any such conduct by Microsoft.
Instead, Bristol seeks to turn the antitrust laws on their head: it insists that Microsoft is required to take affirmative steps to assist its competitors (the sellers of UNIX operating systems) by disclosing its most valuable trade secrets for use on those platforms through Bristols product. According to Bristol, if Bristol goes out of business, UNIX systems will be disadvantaged in the competition with Windows NT. But the law does not require Microsoft to facilitate any such competition. In Olympia Equipment Leasing Co. v. Western Union Telegraph Co., 797 F.2d 370 (7th Cir. 1986), cert. denied, 480 U.S. 934 (1987), Judge Posner, writing for a unanimous Seventh Circuit, addressed this very principle:
Today it is clear that a firm with lawful monopoly power has no general duty to help its competitors, whether by holding a price umbrella over their heads or by otherwise pulling its competitive punches. "There is a difference between positive and negative duties, and the antitrust laws, like other legal doctrines sounding in tort, have generally been understood to impose only the latter."
Id. at 375-76 (citations omitted). The Second Circuit has ruled to the same effect: "Antitrust law, however, does not require one competitor to give another a break just because failing to do so offends notions of fair play." Twin Labs., Inc. v. Weider Health & Fitness, 900 F.2d 566, 568 (2d Cir. 1990) (citing Olympia Equipment).
Bristol nevertheless argues that Microsofts conduct is unlawful because Microsoft first offered, and then withdrew from, the "procompetitive" WISE program. (See Bristol Mem. at 2, 34.) Although this is quite obviously incorrect on the facts -- since Microsoft has licensed its new source code to Mainsoft and has offered the same deal to Bristol -- Bristols legal argument was rejected by the court in Olympia Equipment:
If a monopolist does extend a helping hand, though not required to do so, and later withdraws it as happened in this case, does he incur antitrust liability? We think not. Conceivably he may be liable in tort or contract law . . . . But the controlling consideration in an antitrust case is antitrust policy rather than common law analogies. Since Western Union had no duty to encourage the entry of new firms into the equipment market, the law would be perverse if it made Western Unions encouraging gestures the fulcrum of an antitrust violation.
Olympia Equipment, 797 F.2d at 376 (emphasis added); accord Trans Sport, 775 F. Supp. at 542, affd, 964 F.2d 186 (2d Cir. 1992).
These basic principles preclude Bristols Section 2 monopolization claims. Moreover, an analysis of Bristols three theories of improper antitrust conduct -- monopoly leveraging, refusal to deal, and essential facility -- also reveals that each of them is completely inapplicable here.
Bristol relies on dicta from Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir. 1979), to support a claim for "monopoly leveraging." (See Bristol Mem. at 29-31.) The "monopoly leveraging" doctrine has been widely criticized, see, e.g., ABA Section of Antitrust Law, Antitrust Law Developments, 282-85 (4th ed. 1997) ("Antitrust Law Developments"); 3A Areeda & Hovenkamp ¶ 776b7 at 244 ("[w]e suggest a narrow reading of" Berkey Photo), and has been largely rejected by other courts of appeals.17
The Second Circuit has also recognized the limited nature of the Berkey Photo discussion.18 In Twin Labs., the Second Circuit rejected a monopoly leveraging claim, referring to the statements in Berkey Photo as "dictum (since the plaintiff in Berkey Photo failed to allege an applicable violation) and [because] Berkey Photo involved a tying claim . . . ." 900 F.2d at 570; accord Berkey Photo, 603 F.2d at 275 ("This rule is linked to the prohibition against tying arrangements in the sale of goods and services.") There is no allegation here of any tying by Microsoft.19
In any event, even if a "monopoly leveraging" claim does exist in this Circuit, Bristol needs to show that Microsoft took some affirmative action to use its alleged monopoly power in one market (here, the PC operating systems "market") to gain advantage in other markets (the "markets" for technical workstation and/or server operating systems). See Twin Labs., 900 F.2d at 570; Berkey Photo, 603 F.2d at 275. Bristol fails even to assert any such Microsoft conduct. The closest Bristol comes to doing so is to state that Microsoft uses "control of the Windows programming interface" for PC operating systems "to gain the competitive advantage in the workstation and server markets." (Bristol Mem. at 29.) Although Bristol does not explain how Microsoft does so, Bristols theory appears to be that the popularity of Microsofts Application Programming Interfaces gives Microsoft an advantage in the "markets" for workstation and server operating systems. Putting aside for the moment Bristols lack of standing to make such an argument so unrelated to its business, this theory was rejected in Berkey Photo itself:
[A] large firm does not violate § 2 simply by reaping the competitive rewards attributable to its efficient size, nor does an integrated business offend the Sherman Act whenever one of its departments benefits from association with a division possessing a monopoly in its own market. So long as we allow a firm to compete in several fields, we must expect it to seek the competitive advantages of its broad-based activity -- more efficient production, greater ability to develop complementary products, reduced transaction costs, and so forth. These are gains that accrue to any integrated firm, regardless of its market share, and they cannot by themselves be considered uses of monopoly power.
Berkey Photo, 603 F.2d at 276. Microsoft has not used its alleged monopoly power to prevent any software developer from writing to any competing operating system; any developer can write to UNIX without Microsofts help. Bristols complaint that Microsofts integrated business gives it an advantage -- whether true or not -- simply does not make out an antitrust violation.
In any event, perhaps the most fundamental factual premise of Bristols theory is false. The Second Circuit has stated that, assuming a "monopoly leveraging" claim does exist, such a claim "requires tangible harm to competition." Twin Labs., 900 F.2d at 571. As demonstrated above (pp. 21-22, supra), Bristol cannot show any harm to competition because, even if Bristol were to disappear, Mainsoft provides the very same service as Bristol, and software developers have many other options as well for having their applications run on UNIX.
2. Refusal To Deal.
Microsoft has not refused to deal with Bristol. Rather, as Bristol acknowledges, Microsoft has offered a deal to Bristol (both prior to and during the pendency of this action), but Bristol apparently does not find the terms satisfactory, even though Mainsoft did. This alone compels dismissal of Bristols "refusal to deal" claims.20
Even if that were not so, Bristol cannot maintain a "refusal to deal" claim for an independent reason: Microsoft has a legitimate business justification (rather than any unlawful, anticompetitive objective) for dealing as it has with companies like Bristol. In considering the "refusal to deal" theory, courts have consistently recognized the Supreme Courts insistence in United States v. Colgate & Co., 250 U.S. 300, 307 (1919), that "[i]n the absence of any purpose to create or maintain a monopoly, the [Sherman] act does not restrict the long recognized right of a trader or manufacturer engaged in an entirely private business, freely to exercise his own independent discretion as to parties with whom he will deal."21 See, e.g., Aspen, 472 U.S. at 602; Trans Sport, 964 F.2d at 189; accord 3A Areeda & Hovenkamp ¶ 770e at 169-72 (advising that the "refusal to deal" theory should not be extended to the unilateral action of a single firm.)
A private entitys business discretion to deal with whomever it wishes cannot be abridged where the refusal to deal is predicated on protection of intellectual property rights. In dismissing a "refusal to deal" claim on summary judgment, the Second Circuit has held that a party may freely choose not to deal with another so that it can protect, and enhance profits on, its intellectual property. See Trans Sport, 964 F.2d at 190 (Starters "policy allows it better to identify and combat counterfeit Starter goods" and "to select only those retailers willing to make an economic investment in Starter products"); accord Data General Corp. v. Grumman Systems Support Corp., 36 F.3d 1147, 1187 (1st Cir. 1994) ("an authors desire to exclude others from use of its copyrighted work is a presumptively valid business justification for any immediate harm to consumers"); SCM Corp. v. Xerox Corp., 645 F.2d 1195, 1206 (2d Cir. 1981) ("where a patent has been lawfully acquired, subsequent conduct permissible under the patent laws cannot trigger any liability under the antitrust laws"); Image Technical, 125 F.3d at 1216-19 (agreeing with Data General presumption and stating "[w]e find no reported case in which a court has imposed antitrust liability for a unilateral refusal to sell or license a patent or copyright"); In re Independent Service Organizations Antitrust Litig., 989 F. Supp. 1131, 1134 (D. Kan. 1997) ("where a patent or copyright has been lawfully acquired, subsequent conduct permissible under the patent or copyright laws cannot give rise to any liability under the antitrust laws"); Corsearch, Inc. v. Thomson & Thomson, 792 F. Supp. 305, 329 (S.D.N.Y. 1992) (holder of copyright on computer database did not violate Sherman Act by denying competitor continued access to database).
Thus, although it has not even done so, Microsoft could lawfully refuse altogether to license its confidential copyrighted source code to Bristol.
Bristols attempt to invoke the "essential facility " doctrine is also misguided. That doctrine has been employed in only the most rare of cases. Essential facilities claims almost always fail, precisely because they offend the basic principle that a person or company is entitled to use its own property as it sees fit. See 3A Areeda & Hovenkamp ¶ 771c at 176 ("Lest there be any doubt, we state our belief that the essential facility doctrine is both harmful and unnecessary and should be abandoned."); see also International Audiotext Network, Inc. v. American Telephone & Telegraph Co., 62 F.3d 69 (2d Cir. 1995) (essential facility doctrine not applicable where plaintiff was in essence seeking to force AT&T to joint venture its promotional service); Illinois Bell Telephone Co. v. Haines & Co., 905 F.2d 1081 (7th Cir. 1990) (claim failed where defendant offered access on terms not shown to be discriminatory), vacated on other grounds, 499 U.S. 944 (1991); City of Malden, Missouri v. Union Electric Co., 887 F.2d 157 (8th Cir. 1989) (no "denial" of access because defendant was only negotiating with the plaintiff for a reasonable rate of compensation); Phillip Areeda, The "Essential Facility" Doctrine: An Epithet in Need of Limiting Principles, 58 Antitrust L.J. 841 (1989).
The "essential facility" doctrine was designed for, and is usually only successful with, particular types of monopolies not present here (e.g., monopolies that result from nature or geography, or are publicly subsidized). See Twin Labs., 900 F.2d at 569; Trans Sport, 775 F. Supp. at 542 ("As has been noted previously, no cases have been found in which a manufacturer was held liable for refusing to let go of its natural monopoly over distribution, i.e., for refusing to open up an integrated wholesale distribution system."). Indeed, in Twin Labs., the Second Circuit cited with apparent approval a "leading antitrust commentator [who] would limit the analysis to facilities that are a natural monopoly, facilities whose duplication are forbidden by law, and perhaps those that are publicly subsidized and thus could not practicably be built privately." 900 F.2d at 569 (quoting Areeda & Hovenkamp, Antitrust Law ¶ 736.2 at 680-81 (1988 supp.)).
Even if the "essential facility" doctrine does apply in this Circuit -- and the Twin Labs. decision suggests that any such application should be narrowly limited -- it cannot apply here for several reasons. First, the doctrine requires that the aggrieved party be a competitor of the alleged monopolist. Twin Labs., 900 F.2d at 568-69 (utilizing four-part test of MCI Communications Corp. v. AT & T Co., 708 F.2d 1081, 1132-33 (7th Cir. 1983), which includes "denial of the use of a facility to a competitor"); accord Bristol Mem. at 36. As previously discussed (pp. 23-24, supra), Bristol is not a Microsoft competitor. Bristol admits, as it must, that it does not sell a competing operating system, and Microsoft does not sell any product that "ports" Windows applications to UNIX. (See Bristol Mem. at 44-45;
At most, even by Bristols theory, Bristol merely facilitates competition between Microsoft operating systems and UNIX operating systems. (Bristol Mem. at 45 ("Bristol thereby enhances UNIX and makes it more competitive.").) However, many companies "enhance" or facilitate competition by UNIX and could never sensibly be called Microsoft competitors. According to Bristols theory, all software developers that write applications for UNIX would be competitors of Microsofts operating systems, because their applications run on UNIX and thus help UNIX compete with Windows NT. And if Bristols "competitor" theory were correct, it would be directly inconsistent with Bristols monopolization theories: Microsoft could not possibly be a monopolist in any operating system "market" because its market share among all of those "competitors" would be extremely small. (See Schmalensee Decl. ¶¶ 14-17.)
The illogic of Bristols "essential facility" theory is even more apparent when analyzed from the standpoint of true Microsoft competitors -- competing operating systems and competing application programs. Developers of operating systems have no need for Microsoft source code -- their entire goal is to make a superior product to Windows NT. Similarly, as Bristol concedes, software developers can write computer applications themselves that run on any Microsoft operating system (including Windows NT) and on UNIX. (See Declaration of Ken Blackwell, executed on August 17, 1998 ("Ken Blackwell Decl.") ¶¶ 19, 24, 40.) Thus, Bristols argument is not that any actual competitor needs Microsoft source code; instead, Bristol argues that Microsoft must enhance UNIXs ability to compete with it by enabling Bristol to assist software developers in writing applications that can run on both Windows and UNIX. Whatever the vitality of the essential facility doctrine, it cannot possibly be stretched this far.22
At bottom, if Bristols essential facility argument makes anything clear it is this: even if Microsofts NT 4.0 and 5.0 source code is essential to Bristols survival (which it is not, as explained below), it is not at all essential to the survival of competition between Windows and UNIX operating systems for Bristol to have it. Bristols claim thus fails, as the Second Circuit made clear in affirming the dismissal of the plaintiffs essential facility claim in Twin Labs.: "A particular plaintiffs plight is relevant only as it bears on market effects. The existence of a facility, even if essential in a technical sense, does not constitute an antitrust violation if the plaintiff cannot allege harm to competition." 900 F.2d at 568-69 (internal citation and quotations omitted). As previously discussed (pp. 21-22, supra) Bristol cannot demonstrate that any harm to competition will result if Bristol does not get the Microsoft code it seeks.
Finally, Bristol cannot establish, as it must, that it is "essential" to obtain access to Microsoft source code. Bristol acknowledges that, prior to obtaining any access to Microsoft source code, it developed its Wind/U product (see Bristol Mem. at 4-5; Keith Blackwell Decl. ¶¶ 30-33; Ken Blackwell Decl. ¶ 60), and had a growing company based on that product (Keith Blackwell Decl. ¶¶ 31-33). Indeed, Bristol is even
Accordingly, Microsoft source code is not "essential" to Bristol. See Twin Labs., 900 F.2d at 568 ("At the very least, a plaintiff must demonstrate that duplication of the facility would be economically infeasible and that denial of its use inflicts a severe handicap on potential [or current] market entrants."). Bristol cannot cite any authority for its wild assertion that an intellectual property product can become an "essential facility" merely because Bristol made the knowing choice to become dependent on it.23
As Bristol acknowledges, "to demonstrate attempted monopolization a plaintiff must prove (1) that the defendant has engaged in predatory or anticompetitive conduct with (2) a specific intent to monopolize and (3) a dangerous probability of achieving monopoly power." Spectrum Sports, 506 U.S. at 456. (See Bristol Mem. at 42.) For the reasons described above, Microsoft has not engaged in any "predatory or anticompetitive conduct" (as both a matter of law and fact), and thus cannot be guilty of attempted monopolization. See Olympia Equipment, 797 F.2d at 373 ("Conduct lawful for a monopolist is lawful for a firm attempting to become a monopolist."); 3A Areeda & Hovenkamp ¶ 806d.
Even if this were not true, Microsofts "intent" in offering the particulars of the deal it has offered to Bristol was two-fold: (1) to achieve a reasonable return on its intellectual property while retaining necessary protections of that property, and (2) maintaining a consistent licensing structure for both Mainsoft and Bristol so that neither has an unfair advantage over the other. Far from being an "intent to monopolize," this intent is perfectly consistent with the antitrust laws. See Trans Sport, 964 F.2d at 190 (protection of intellectual property is a "valid business rational[e]").
Finally, Microsoft does not have a "dangerous probability of achieving monopoly power" in either of the "markets" for departmental server operating systems or technical workstation operating systems. Even Bristols experts data indicates that UNIX operating systems will
Bristol itself also endorses projections that UNIX will grow "through the year 2002," and Bristol has stated that it "anticipate[s] [the growth of UNIX] to be strong well beyond that." (August 30, 1998 excerpt from Bristol website (Martin Aff., Ex. 19) at 5.) This hardly shows a "dangerous probability" that Windows NT will eliminate UNIX as a competitor any time in the near future. See, e.g., Twin Labs., 900 F.2d at 570 ("We have held that a 33% market share does not even approach the level required for dangerous probability of success."); AD/SAT, 920 F. Supp. at 1300 ("Many courts in the Second Circuit have found market shares of less than 50% or even 60% insufficient to support an attempted monopolization claim.").
To succeed on a claim under CUTPA, Bristol has to demonstrate, among other things, that Microsoft acted in contravention of Bristols contractual rights. Shawmut Bank Conn., N.A. v. L & R Realty, Nos. 523134, 522814, 1995 WL 384661, at **32-33 (Conn. Super. June 20, 1995), revd on other grounds, 46 Conn. App. 443, 699 A.2d 297 (1997), revd on other grounds, __ A.2d __, 1998 WL 436816 (Aug. 4, 1998). Bristol does not and could not assert that Microsoft has any obligation to provide Windows NT 4.0 or 5.0 source code pursuant to the parties 1994 Contract. Indeed, the history of negotiations between the parties demonstrates that
For this reason alone, Bristols CUTPA claim cannot succeed and should not have been brought.
Bristol argues that Microsoft is estopped from "withholding" Windows NT 4.0 and 5.0 source code from Bristol because Microsoft made purported "assurances" years ago that it would continue to provide source code for new products to WISE licensees (which Microsoft in fact did while obligated to under the 1994 Contract). (Bristol Mem. at 47-50.)
To assert a claim for promissory estoppel under Connecticut law, Bristol must show " a clear and unambiguous promise;  a reasonable and foreseeable reliance by a party to whom the promise is made; and  an injury sustained by . . . reason of his reliance." Dacourt Group, Inc. v. Babcock Indus., Inc., 747 F. Supp. 157, 161 (D. Conn. 1990). Bristols claim fails all of these requirements.
First, in its most recent decision in this area, the Connecticut Supreme Court recognized the importance of closely guarding against claims of promissory estoppel that do not rely on "the existence of a clear and definite promise." DUlisse-Cupo v. Board of Directors of Notre Dame High School, 202 Conn. 206, 213, 520 A.D.2d 217, 221 (1987). In that case, the court held that a teacher did not have a promissory estoppel claim against her school because the schools written notice that "[a]ll present faculty members will be offered contracts for next year" was not a sufficiently definite promise to support a claim. 202 Conn. at 214-15, 520 A.D.2d at 221-22. Other courts, relying on this decision, have rejected promissory estoppel claims based on insufficiently definite promises. See Redgate v. Fairfield Univ., 862 F. Supp. 724, 730-31 (D. Conn. 1994) (promises of "employment for ten or twenty years and for as long as you want" are "insufficiently promissory and insufficiently definite" for promissory estoppel claim); Dacourt Group, 747 F. Supp. at 161 (defendants statement (among others) that it "would provide a keep-well agreement" did not form basis for claim of estoppel); Daley v. Gaitor, 16 Conn. App. 379, 385 n.6, 547 A.2d 1375, 1378 n.6 (1988) (pronouncement that "at least 50 percent of all persons" will be allowed to take test is insufficiently "definite or promissory").
These courts have reasoned that, to form the basis for a promissory estoppel claim, a promise must manifest a "present intention . . . to undertake immediate contractual obligations" and must contain "the material terms that would be essential to" the contract; a mere "expectation of a future contract" is not sufficient. DUlisse-Cupo, 202 Conn. at 214-15, 520 A.D.2d at 221-22; accord Redgate, 862 F. Supp. at 731; Dacourt Group, 747 F. Supp. at 161 ("[a]t the time these alleged promises were made, material terms remained unresolved").
Here, there is no doubt that Microsofts alleged "assurances" cannot satisfy these standards. Not one of the Microsoft statements cited by Bristol (see Bristol Mem. at 12-14, 48-49) ever manifested a "present intention" to enter into a new contract with Bristol or any other specific WISE licensee years later. To the contrary, all of the statements merely showed Microsofts recognition that it was under multi-year contracts with both Mainsoft and Bristol. The statements were made years ago, while the 1994 Contract was being performed, yet Bristol now seeks to rely on them long after Microsofts obligation to provide specified source code to Bristol under the 1994 Contract has, as Bristol acknowledges, expired. Moreover, none of the statements mentions any terms of a new contract, much less the material terms (which Bristol apparently refuses to accept).
Second, Bristol cannot, as a matter of law, demonstrate reasonable and foreseeable reliance by Bristol on any Microsoft "assurance." The law of Connecticut is clear that, when a written contract delineates the rights of the parties (here, a three-year term for Microsoft to provide source code for operating systems other than Windows NT 4.0 and 5.0), those are the only rights that the party can reasonably rely upon. Lark v. Post Newsweek Stations Conn., Inc., No. 94-0705326, 1995 WL 491290, at *3 (Conn. Super. Aug. 9, 1995). This is reinforced by the integration and no oral modification clauses in the 1994 Contract. Shawmut, 1995 WL 384661, at **21, 28-29 ("to use the doctrine of promissory estoppel . . . to enforce the oral promise . . . would contradict the parol evidence rule insofar as it would vary the terms of an integrated written agreement"). Thus, as a matter of law, Bristol cannot assert that Microsoft is estopped from "withholding" source code when a written contract between the parties provides that Microsofts duty to provide source code exists only between September 1994 and September 1997, and only for earlier versions of Windows NT not here at issue. If Bristol did anything in reliance on Microsofts alleged extra-contractual "assurances" -- and it does not really say what it did -- it was inherently unreasonable given the explicit three-year term.
Finally, Bristol does not identify any injury it incurred as a result of Microsofts purported "assurances." In his declaration, Keith Blackwell states that Bristol has become "locked into dependence" on Microsoft source code by virtue of entering into the WISE license it signed in 1994. (Keith Blackwell Decl. ¶¶ 43, 58-60.) But if that is so, it was the 1994 Contract and not Microsofts "assurances" that caused Bristols reliance and whatever (undescribed) subsequent detriment that resulted.
Even leaving aside the lack of any "serious questions going to the merits [of Bristols claims] to make them fair ground for litigation," USA Recycling, Inc., 66 F.3d at 1280-81, the balance of hardships weighs heavily in Microsofts favor, not Bristols. Bristol seeks the Courts intervention to obtain extremely valuable intellectual property from Microsoft, while it is undisputed that judicial intervention would be unnecessary if Bristol agreed to a deal similar to the one reached by its competitor Mainsoft.
Over the years, Microsoft has spent huge amounts of money developing the NT 4.0 and NT 5.0 operating systems that are in dispute in this case. As opposed to more traditional manufacturing companies, the most important and valuable asset a software company like Microsoft has is its intellectual property, and Microsoft has gone to great lengths to protect its confidentiality. Microsoft has copyrighted the source code for all of its operating systems and, pursuant to Copyright Office procedure, see 37 C.F.R. § 202.20(c)(2) (1996), deposited only a small percentage of that code so that the remainder can be kept as Microsofts trade secrets. As explained in the Declarations of Dan Neault ("Neault Decl."), Microsofts Senior Group Program Manager, and James Allchin ("Allchin Decl."), Microsofts Senior Vice President, submitted in another litigation (and which Bristol attached as Exhibits 4 and 7, respectively, to the Declaration of Anthony L. Clapes), Microsoft protects this intellectual property with "special vigor" (Allchin Decl. ¶ 3), and licenses it only under very prescribed conditions (Neault Decl. ¶¶ 27-28).
Windows NT 4.0 and 5.0 are two of Microsofts most important products. Bill Gates has frequently said that "Microsoft has bet its future on Windows NT 5.0." (Neault Decl. ¶ 3.) The source code for these two products is a huge quantity of information: although hard to visualize, the two programs now amass roughly 80 million lines of code -- approximately 20 times the amount of code licensed under the 1994 Contract.
The development effort for Windows NT 5.0 (which has not yet been commercially launched) will be the most extensive in the companys history. (Allchin Decl. ¶ 3; Neault Decl. ¶ 3.) As of December 24, 1997, Mr. Allchin estimated that the cost of the development of Windows NT 5.0 alone was $270 million and growing. (Allchin Decl. ¶ 5.)
Forcing Microsoft to provide such a valuable asset to Bristol -- a company with which Microsoft has had rocky relations in the past -- could be disastrous. As even Bristol concedes (see Keith Blackwell Decl. ¶ 60), once the information in that source code is revealed, it is nearly impossible to simply "return." See FMC Corp., 730 F.2d at 63 ("A trade secret once lost is, of course, lost forever.") It is for this reason that courts are extremely resistant to granting preliminary injunctions that order the disclosure of proprietary, confidential information. See, e.g., Anacomp Lockheed Missile & Space Co. v. Hughes Aircraft Co., 887 F. Supp. 1320, 1323-25 (N.D. Cal. 1995); Bell Atlantic Business Sys. Servs., Inc. v. Hitachi Data Sys. Corp., 856 F. Supp. 524, 525 (N.D. Cal. 1993).
Moreover, it is extremely troublesome for an adversary in litigation to be given access to such proprietary materials. See Corenswet, Inc. v. Amana Refrigeration, Inc., 594 F.2d 129, 134 n.3 (5th Cir. 1979) ("The Court should not be called upon to weld together two business entities which have shown a propensity for disagreement, friction, and even adverse litigation."); Lanvin Inc. v. Colonia, Inc., 739 F. Supp. 182, 196 (S.D.N.Y. 1990) ("a movant does not establish that the balance of hardships tips decidedly in its favor if the relief it requests would force its former supplier to endure an unsatisfactory relationship with a licensee it no longer trusts"); cf. House of Materials, Inc. v. Simplicity Pattern Co., 298 F.2d 867, 871 (2d Cir. 1962) ("the bringing of a lawsuit . . . may provide a sound business reason for the manufacturer to terminate their relations"). It is even more worrisome when that company says it could go out of business soon, and thus may have little incentive to maintain the tight controls on Microsofts intellectual property that are necessary.
Finally, Bristols assertions of hardship are premised on self-serving naivete. Bristol knew full well that it only had a three-year agreement, and it made a reasoned business judgment to take that deal. It should not now be heard to complain of hardship when it is in its current situation only because it walked away from the bargaining table rather than taking new source code on the terms that Microsoft offered. Nor should Bristols self-inflicted hardship form the basis for awarding it a contract that would put its chief competitor, Mainsoft, at a significant competitive disadvantage. (See Share Dep. (Martin Aff., Ex. 24) at 192.)
Rule 65(c), Fed. R. Civ. P., mandates that "[n]o restraining order or preliminary injunction shall issue except upon the giving of security by the applicant."24 The Second Circuit has given district courts discretion to dispose of the bond requirement only where (1) the defendant would not be harmed by the injunction (or is a governmental entity in a public interest case), or (2) the court issues the injunction solely to preserve its jurisdiction. See Doctors Assocs., Inc. v. Distajo, 107 F.3d 126, 136 (2d Cir. 1997); Pharmaceutical Society of State of New York, Inc. v. New York Dept of Social Services, 50 F.3d 1168, 1174-75 (2d Cir. 1995). Neither situation is applicable here.
Microsoft will clearly suffer a substantial risk of financial harm if an injunction issues. Bristol seeks a mandatory injunction requiring that Microsoft provide to Bristol all the source code to Windows NT 4.0 and 5.0, two of Microsofts most important operating systems and the lifeblood of its business. The value of that source code is virtually immeasurable. Once disclosed, there is no way to erase Bristols knowledge of it, FMC Corp., 730 F.2d at 63, and Microsoft could lose its ability to protect this valued property. If Microsofts source code were improperly disclosed, it could destroy the value of Microsofts operating system business, both now and for the future.
Bristol argues that it should be excused from the bond requirement because its prosecution of this action vindicates an important public interest. In purported support of that argument, Bristol cites a line of cases involving indigent plaintiffs and public interest claims, neither of which is applicable here. (See Bristol Mem. at 53.) Indeed, Bristol does not and cannot cite any authority for the remarkable claim that its private antitrust action -- designed solely for its own pecuniary and competitive interest to garner a better deal than the one that its primary competitor, Mainsoft, has reached -- vindicates an important public interest.25
In fact, the public interest supports denial of Bristols motion. The strict protections afforded to intellectual property and trade secrets by copyright, patent and other statutes, and by common law, reflect the strong and longstanding public policy of encouraging and rewarding innovation. Microsoft submits that, in order to avoid the risk of undermining this important public policy, it should be a rare case indeed in which a defendant is ordered to give its virtually priceless, effectively unreturnable, confidential intellectual property and trade secrets to a litigation adversary even before the merits of the case are decided. This is not such a case, especially given that Bristol cannot come close to making the "clear showing" of likelihood of success on the merits that is required of a plaintiff seeking a preliminary, mandatory injunction ordering the disclosure of confidential, proprietary information. See Tom Doherty, 60 F.3d at 35.
In the unlikely event that the Court is inclined to issue a preliminary injunction here, Microsoft respectfully requests an opportunity to make an additional submission to the Court addressing and substantiating the amount of the bond that Bristol should be required to post, as well as appropriate restrictions that the Court should impose on Bristols use of Microsofts source code during the pendency of this action.
Bristol cannot satisfy any of the requirements for the extraordinary relief of a preliminary, mandatory injunction forcing Microsoft to turn over, even before a trial on the merits, the crown jewels of its business -- Microsofts Windows NT 4.0 and 5.0 source code containing confidential trade secrets and copyrighted intellectual property. To summarize:
Simply put, Bristol does not have a leg to stand on here. Its motion for a preliminary injunction should be denied, and the Court may do so even without conducting any evidentiary hearing.
SULLIVAN & CROMWELL
DAY BERRY & HOWARD LLP
Counsel for Microsoft Corporation
September 25, 1998
[Certificate of Service omitted]