Full opinion text
Opinion for the Court filed PER CURIAM. Table of Contents .44 I. Introduction.47 A. Background.47 B. Overview.48 II. Monopolization.50 A. Monopoly Power.51 1. Market Structure .51 a. Market definition.51 b. Market power.54 2. Direct Proof.56 B. Anticompetitive Conduct.58 1. Licenses Issued to Original Equipment Manufacturers.59 a. Anticompetitive effect of the license restrictions .60 b. Microsoft’s justifications for the license restrictions .62 2. Integration of IE and Windows.64 a. Anticompetitive effect of integration.65 b.Microsoft’s justifications for integration. CD CD 3. Agreements with Internet Access Providers. tr-CD 4. Dealings with Internet Content Providers, Independent Software Vendors, and Apple Computer. 5. Java. a. The JVM b. The First Wave Agreements. c. of Java developers. d. The threat to Intel. 6. Course of Conduct. C. Causation. III. A. Relevant Market. B. Barriers to Entry. TV. A. Under the Per Se Test. B. Per Se for this Case. C. On Remand. V.Trial and Remedy. A. Factual B. Trial Proceedings. C. Failure to Hold an Evidentiary Hearing. D. Failure to Provide an ^ E. Modification of F. On Remand. G. Conclusion. VI.Judicial Misconduct.. A. The District Judge’s Communications with the Press. ¿ B. Violations of the Code of Conduct for United States Judges. C. of D. Remedies for Judicial Misconduct and Appearance of Partiality. 1. 2. Review of Findings of Fact and Conclusions of Law. VII.Conclusion . PER CURIAM: Microsoft Corporation appeals from judgments of the District Court finding the company in violation of §§ 1 and 2 of the Sherman Act and ordering various remedies. The action against Microsoft arose pursuant to a complaint filed by the United States and separate complaints filed by individual States. The District Court determined that Microsoft had maintained a monopoly in the market for Intelcompati-ble PC operating systems in violation of § 2; attempted to gain a monopoly in the market for internet browsers in violation of § 2; and illegally tied two purportedly separate products, Windows and Internet Explorer (“IE”), in violation of § 1. United States v. Microsoft Corp., 87 F.Supp.2d 30 (D.D.C.2000) (“Conclusions of Law”). The District Court then found that the same facts that established liability under §§ 1 and 2 of the Sherman Act mandated findings of liability under analogous state law antitrust provisions. Id. To remedy the Sherman Act violations, the District Court issued a Final Judgment requiring Microsoft to submit a proposed plan of divestiture, with the company to be split into an operating systems business and an applications business. United States v. Microsoft Corp., 97 F.Supp.2d 59, 64-65 (D.D.C.2000) {“Final Judgment”). The District Court’s remedial order also contains a number of interim restrictions on Microsoft’s conduct. Id. at 66-69. Microsoft’s appeal contests both the legal conclusions and the resulting remedial order. There are three principal aspects of this appeal. First, Microsoft challenges the District Court’s legal conclusions as to all three alleged antitrust violations and also a number of the procedural and factual foundations on which they rest. Second, Microsoft argues that the remedial order must be set aside, because the District Court failed to afford the company an evi-dentiary hearing on disputed facts and, also, because the substantive provisions of the order are flawed. Finally, Microsoft asserts that the trial judge committed ethical violations by engaging in impermissible ex parte contacts and making inappropriate public comments on the merits of the case while it was pending. Microsoft argues that these ethical violations compromised the District Judge’s appearance of impartiality, thereby necessitating his disqualification and vacatur of his Findings of Fact, Conclusions of Law, and Final Judgment. After carefully considering the voluminous record on appeal — including the District Court’s Findings of Fact and Conclusions of Law, the testimony and exhibits submitted at trial, the parties’ briefs, and the oral arguments before this court — we find that some but not all of Microsoft’s liability challenges have merit. Accordingly, we affirm in part and reverse in part the District Court’s judgment that Microsoft violated § 2 of the Sherman Act by employing anticompetitive means to maintain a monopoly in the operating system market; we reverse the District Court’s determination that Microsoft violated § 2 of the Sherman Act by illegally attempting to monopolize the internet browser market; and we remand the District Court’s .finding that Microsoft violated § 1 of the Sherman Act by unlawfully tying its browser to its operating system. Our judgment extends to the District Court’s findings with respect to the state law counterparts of the plaintiffs’ Sherman Act claims. We also find merit in Microsoft’s challenge to the Final Judgment embracing the District Court’s remedial order. There are several reasons supporting this conclusion. First, the District Court’s Final Judgment rests on a number of liability determinations that do not survive appellate review; therefore, the remedial order as currently fashioned cannot stand. Furthermore, we would vacate and remand the remedial order even were we to uphold the District Court’s liability determinations in their entirety, because the District Court failed to hold an evidentia-ry hearing to address remedies-specific factual disputes. Finally, we vacate the Final Judgment on remedies, because the trial judge engaged in impermissible ex parte contacts by holding secret interviews with member's of the media and made numerous offensive comments about Microsoft officials in public statements outside of the courtroom, giving rise to an appearance of partiality. Although we find no evidence of actual bias, we hold that the actions of the trial judge seriously tainted the proceedings before the District Court and called into question the integrity of the judicial process. We are therefore constrained to vacate the Final Judgment on remedies, remand the case for reconsideration of the remedial order, and require that the case be assigned to a different trial judge on remand. We believe that this disposition will be adequate to cure the cited improprieties. In sum, for reasons more fully explained below, we affirm in part, reverse in part, and remand in part the District Court’s judgment assessing liability.. We vacate in full the Final Judgment embodying the remedial order and remand the case to a different trial judge for further proceedings consistent with this opinion. I. Introduction A. Background In July 1994, officials at the Department of Justice (“DOJ”), on behalf of the United States, filed suit against Microsoft, charging the company with, among other things, unlawfully maintaining a monopoly in the operating system market through anticom-petitive terms in its licensing and software developer agreements. The parties subsequently entered into a consent decree, thus avoiding a trial on the merits. See United States v. Microsoft Corp., 56 F.3d 1448 (D.C.Cir.1995) (“Microsoft I”). Three years later, the Justice Department filed a civil contempt action against Microsoft for allegedly violating one of the decree’s provisions. On appeal from a grant of a preliminary injunction, this court held that Microsoft’s technological bundling of IE 3.0 and 4.0 with Windows 95 did not violate the relevant provision of the consent decree. United States v. Microsoft Corp., 147 F.3d 935 (D.C.Cir.1998) (“Microsoft II”). We expressly reserved the question whether such bundling might independently violate §§ 1 or 2 of the Sherman Act. Id. at 950 n. 14. On May 18,1998, shortly before issuance of the Microsoft II decision, the United States and a group of State plaintiffs filed separate (and soon thereafter consolidated) complaints, asserting antitrust violations by Microsoft and seeking preliminary and permanent injunctions against the company’s allegedly unlawful conduct. The complaints also sought any “other preliminary and permanent relief as is necessary and appropriate to restore competitive conditions in the markets affected by Microsoft’s unlawful conduct.” Gov’t’s Compl. at 53, United States v. Microsoft Corp., No. 98-1232 (D.D.C.1999). Relying almost exclusively on Microsoft’s varied efforts to unseat Netscape Navigator as the preeminent internet browser, plaintiffs charged four distinct violations of the Sherman Act: (1) unlawful exclusive dealing arrangements in violation of § 1; (2) unlawful tying of IE to Windows 95 and Windows 98 in violation of § 1; (3) unlawful maintenance of a monopoly in the PC operating system market in violation of § 2; and (4) unlawful attempted monopolization of the internet browser market in violation of § 2. The States also brought pendent claims charging Microsoft with violations of various State antitrust laws. The District Court scheduled the case on a “fast track.” The hearing on the preliminary injunction and the trial on the merits were consolidated pursuant to Fed. R.CivP. 65(a)(2). The trial was then scheduled to commence on September 8, 1998, less than four months after the complaints had been filed. In a series of pretrial orders, the District Court limited each side to a maximum of 12 trial witnesses plus two rebuttal witnesses. It required that all trial witnesses’ direct testimony be submitted to the court in the form of written declarations. The District Court also made allowances for the use of deposition testimony at trial to prove subordinate or predicate issues. Following the grant of three brief continuances, the trial started on October 19,1998. After a 76-day bench trial, the District Court issued its Findings of Fact. United States v. Microsoft Corp., 84 F.Supp.2d 9 (D.D.C.1999) (“Findings of Fact”). This triggered two independent courses of action. First, the District Court established a schedule for briefing on possible legal conclusions, inviting Professor Lawrence Lessig to participate as amicus curiae. Second, the District Court referred the case to mediation to afford the parties an opportunity to settle their differences. The Honorable Richard A. Posner, Chief Judge of the United States Court of Appeals for the Seventh Circuit, was appointed to serve as mediator. The parties concurred in the referral to mediation and in the choice of mediator. Mediation failed after nearly four months of settlement talks between the parties. On April 3, 2000, with the parties’ briefs having been submitted and considered, the District Court issued its conclusions of law. The District Court found Microsoft liable on the § 1 tying and § 2 monopoly maintenance and attempted monopolization claims, Conclusions of Law, at 35-51, while ruling that there was insufficient evidence to support a § 1 exclusive dealing violation, id. at 51-54. As to the pendent State actions, the District Court found the State antitrust laws conterminous with §§ 1 and 2 of the Sherman Act, thereby obviating the need for further State-specific analysis. Id. at 54-56. In those few eases where a State’s law required an additional showing of intrastate impact on competition, the District Court found the requirement easily satisfied on the evidence at hand. Id. at 55. Having found Microsoft hable on all but one count, the District Court then asked plaintiffs to submit a proposed remedy. Plaintiffs’ proposal for a remedial order was subsequently filed within four weeks, along with six supplemental declarations and over 50 new exhibits. In their proposal, plaintiffs sought specific conduct remedies, plus structural relief that would split Microsoft into an applications company and an operating systems company. The District Court rejected Microsoft’s request for further evidentiary proceedings and, following a single hearing on the merits of the remedy question, issued its Final Judgment on June 7, 2000. The District Court adopted plaintiffs’ proposed remedy without substantive change. Microsoft filed a notice of appeal within a week after the District Court issued its Final Judgment. This court then ordered that any proceedings before it be heard by the court sitting en banc. Before any substantive matters were addressed by this court, however, the District Court certified appeal of the case brought by the United States directly to the Supreme Court pursuant to 15 U.S.C. § 29(b), while staying the final judgment order in the federal and state cases pending appeal. The States thereafter petitioned the Supreme Court for a writ of certiorari in their case. The Supreme Court declined to hear the appeal of the Government’s case and remanded the matter to this court; the Court likewise denied the States’ petition for writ of certiorari. Microsoft Corp. v. United States, 530 U.S. 1301, 121 S.Ct. 25, 147 L.Ed.2d 1048 (2000). This consolidated appeal followed. B. Overview Before turning to the merits of Microsoft’s various arguments, we pause to reflect briefly on two matters of note, one practical and one theoretical. The practical matter relates to the temporal dimension of this case. The litigation timeline in this case is hardly problematic. Indeed, it is noteworthy that a case of this magnitude and complexity has proceeded from the fifing of complaints through trial to appellate decision in a mere three years. See, e.g., Data Gen. Corp. v. Grumman Sys. Support Corp., 36 F.3d 1147, 1155 (1st Cir.1994) (six years from filing of complaint to appellate decision); Transamerica Computer Co., Inc. v. IBM, 698 F.2d 1377, 1381 (9th Cir.1983) (over four years from start of trial to appellate decision); United States v. United Shoe Mach. Corp., 110 F.Supp. 295, 298 (D.Mass.1953) (over five years from filing of complaint to trial court decision). What is somewhat problematic, however, is that just over six years have passed since Microsoft engaged in the first conduct plaintiffs allege to be anticompetitive. As the record in this case indicates, six years seems like an eternity in the computer industry. By the time a court can assess liability, firms, products, and the marketplace are likely to have changed dramatically. This, in turn, threatens enormous practical difficulties for courts considering the appropriate measure of relief in equitable enforcement actions, both in crafting injunctive remedies in the first instance and reviewing those remedies in the second. Conduct remedies may be unavailing in such cases, because innovation to a large degree has already rendered the anticompetitive conduct obsolete (although by no means harmless). And broader structural remedies present their own set of problems, including how a court goes about restoring competition to a dramatically changed, and constantly changing, marketplace. That is just one reason why we find the District Court’s refusal in the present case to hold an evidentiary hearing on remedies — to update and flesh out the available information before seriously entertaining the possibility of dramatic structural relief — so problematic. See infra Section V. We do not mean to say that enforcement actions will no longer play an important role in curbing infringements of the antitrust laws in technologically dynamic markets, nor do we assume this in assessing the merits of this case. Even in those cases where forward-looking remedies appear limited, the Government will continue to have an interest in defining the contours of the antitrust laws so that law-abiding firms will have a clear sense of what is permissible and what is not. And the threat of private damage actions will remain to deter those firms inclined to test the limits of the law. The second matter of note is more theoretical in nature. We decide this case against a backdrop of significant debate amongst academics and practitioners over the extent to which “old economy” § 2 monopolization doctrines should apply to firms competing in dynamic technological markets characterized by network effects. In markets characterized by network effects, one product or standard tends towards dominance, because “the utility that a user derives from consumption of the good increases with the number of other agents consuming the good.” Michael L. Katz & Carl Shapiro, Network Externalities, Competition, and Compatibility, 75 Am. Eoon. Rev. 424, 424 (1985). For example, “[a]n individual consumer’s demand to use (and hence her benefit from) the telephone network ... increases with the number of other users on the network whom she can call or from whom she can receive calls.” Howard A. Shelanski & J. Gregory Sidak, Antitrust Divestiture in Network Industries, 68 U. Chi. L. Rev. 1, 8 (2001). Once a product or standard achieves wide acceptance, it becomes more or less entrenched. Competition in such industries is “for the field” rather than “within the field.” See Harold Demsetz, Why Regulate Utilities?, 11 J.L. & Eoon. 55, 57 & n.7 (1968) (emphasis omitted). In technologically dynamic markets, however, such entrenchment may be temporary, because innovation may alter the field altogether. See Joseph A. Schumpeter, Capitalism, Socialism and Democracy 81-90 (Harper Perennial 1976) (1942). Rapid technological change leads to markets in which “firms compete through innovation for temporary market dominance, from which they may be displaced by the next wave of product advancements.” Shelanski & Sidak, at 11-12 (discussing Schumpeterian competition, which proceeds “sequentially over time rather than simultaneously across a market”). Microsoft argues that the operating system market is just such a market. Whether or not Microsoft’s characterization of the operating system market is correct does not appreciably alter our mission in assessing the alleged antitrust violations in the present case. As an initial matter, we note that there is no consensus among commentators on the question of whether, and to what extent, current monopolization doctrine should be amended to account for competition in technologically dynamic markets characterized by network effects. Compare Steven C. Salop & R. Craig Romaine, Preserving Monopoly: Economic Analysis, Legal Standards, and Microsoft, 7 Geo. Mason L. Rev. 617, 654-55, 663-64 (1999) (arguing that exclusionary conduct in high-tech networked industries deserves heightened antitrust scrutiny in part because it may threaten to deter innovation), with Ronald A. Cass & Keith N. Hylton, Preserving Competition: Economic Analysis, Legal Standards and Microsoft, 8 Geo. Mason L. Rev. 1, 36-39 (1999) (equivocating on the antitrust implications of network effects and noting that the presence of network externalities may actually encourage innovation by guaranteeing more durable monopolies to innovating winners). Indeed, there is some suggestion that the economic consequences of network effects and technological dynamism act to offset one another, thereby making it difficult to formulate categorical antitrust rules absent a particularized analysis of a given market. See Shelanski & Sidak, at 6-7 (“High profit margins might appear to be the benign and necessary recovery of legitimate investment returns in a Schumpeterian framework, but they might represent exploitation of customer lock-in and monopoly power when viewed through the lens of network economics - The issue is particularly complex because, in network industries characterized by rapid innovation, both forces may be operating and can be difficult to isolate.”). Moreover, it should be clear that Microsoft makes no claim that anticompetitive conduct should be assessed differently in technologically dynamic markets. It claims only that the measure of monopoly power should be different. For reasons fully discussed below, we reject Microsoft’s monopoly power argument. See infra Section II.A. With this backdrop in mind, we turn to the specific challenges raised in Microsoft’s appeal. II. Monopolization Section 2 of the Sherman Act makes it unlawful for a firm to “monopolize.” 15 U.S.C. § 2. The offense of monopolization has two elements: “(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, 86 S.Ct. 1698, 16 L.Ed.2d 778 (1966). The District Court applied this test and found that Microsoft possesses monopoly power in the market for Intel-compatible PC operating systems. Focusing primarily on Microsoft’s efforts to suppress Netscape Navigator’s threat to its operating system monopoly, the court also found that Microsoft maintained its power not through competition on the merits, but through unlawful means. Microsoft challenges both conclusions. We defer to the District Court’s findings of fact, setting them aside only if clearly erroneous. Fed R. Crv. P. 52(a). We review legal questions de novo. United States ex rel. Modern Elec., Inc. v. Ideal Elec. Sec. Co., 81 F.3d 240, 244 (D.C.Cir.1996). We begin by considering whether Microsoft possesses monopoly power, see infra Section II.A, and finding that it does, we turn to the question whether it maintained this power through anticompetitive means. Agreeing with the District Court that the company behaved anticompetitively, see infra Section II.B, and that these actions contributed to the maintenance of its monopoly power, see infra Section II.C, we affirm the court’s finding of liability for monopolization. A. Monopoly Power While merely possessing monopoly power is not itself an antitrust violation, see Northeastern Tel. Co. v. AT & T, 651 F.2d 76, 84-85 (2d Cir.1981), it is a necessary element of a monopolization charge, see Grinnell, 384 U.S. at 570, 86 S.Ct. 1698. The Supreme Court defines monopoly power as “the power to control prices or exclude competition.” United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 391, 76 S.Ct. 994, 100 L.Ed. 1264 (1956). More precisely, a firm is a monopolist if it can profitably raise prices substantially above the competitive level. 2A Phillip E. Akeeda et al., Antitrust Law ¶ 501, at 85 (1995); cf. Ball Mem’l Hosp., Inc. v. Mut. Hosp. Ins., Inc., 784 F.2d 1325, 1335 (7th Cir.1986) (defining market power as “the ability to cut back the market’s total output and so raise price”). Where evidence indicates that a firm has in fact profitably done so, the existence of monopoly power is clear. See Rebel Oil Co. v. Atl. Richfield Co., 51 F.3d 1421, 1434 (9th Cir.1995); see also FTC v. Indiana Fed’n of Dentists, 476 U.S. 447, 460-61, 106 S.Ct. 2009, 90 L.Ed.2d 445 (1986) (using direct proof to show market power in Sherman Act § 1 unreasonable restraint of trade action). Because such direct proof is only rarely available, courts more typically examine market structure in search of circumstantial evidence of monopoly power. 2A Areeda et al„ Antitrust Law ¶ 531a, at 156; see also, e.g., Grinnell, 384 U.S. at 571, 86 S.Ct. 1698. Under this structural approach, monopoly power may be inferred from a firm’s possession of a dominant share of a relevant market that is protected by entry barriers. See Rebel Oil, 51 F.3d at 1434. “Entry barriers” are factors (such as certain regulatory requirements) that prevent new rivals from timely responding to an increase in price above the competitive level. See S. Pac. Communications Co. v.AT & T, 740 F.2d 980, 1001-02 (D.C.Cir.1984). The District Court considered these structural factors and concluded that Microsoft possesses monopoly power in a relevant market. Defining the market as Intel-compatible PC operating systems, the District Court found that Microsoft has a greater than 95% share. It also found the company’s market position protected by a substantial entry barrier. Conclusions of Law, at 36. Microsoft argues that the District Court incorrectly defined the relevant market. It also claims that there is no barrier to entry in that market. Alternatively, Microsoft argues that because the software industry is uniquely dynamic, direct proof, rather than circumstantial evidence, more appropriately indicates whether it possesses monopoly power. Rejecting each argument, we uphold the District Court’s finding of monopoly power in its entirety. 1. Market Structure a. Market definition “Because the ability of consumers to turn to other suppliers restrains a firm from raising prices above the competitive level,” Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210, 218 (D.C.Cir.1986), the relevant market must include all products “reasonably interchangeable by consumers for the sampurposes.” du Pont, 351 U.S. at 395, 76 S.Ct. 994. In this case, the District Court defined the market as “the licensing of all Intel-compatible PC operating systems worldwide,” finding that there are “currently no products — and ... there are not likely to be any in the near future — that a significant percentage of computer users worldwide could substitute for [these operating systems] without incurring substantial costs.” Conclusions of Law, at 36. Calling this market definition “far too narrow,” Appellant’s Opening Br. at 84, Microsoft argues that the District Court improperly excluded three types of products: non-Intel compatible operating systems (primarily Apple’s Macintosh operating system, Mac OS), operating systems for non-PC devices (such as handheld computers and portal websites), and “middleware” products, which are not operating systems at all. We begin with Mac OS. Microsoft’s argument that Mac OS should have been included in the relevant market suffers from a flaw that infects many of the company’s monopoly power claims: the company fails to challenge the District Court’s factual findings, or to argue' that these findings do not support the court’s conclusions. The District Court found that consumers would not switch from Windows to Mac OS in response to a substantial price increase because of the costs of acquiring the new hardware needed to run Mac OS (an Apple computer and peripherals) and compatible software applications, as well as because of the effort involved in learning the new system and transferring files to its format. Findings of Fact ¶ 20. The court also found the Apple system less appealing to consumers because it costs considerably more and supports fewer applications. Id. ¶21. Microsoft responds only by saying: “the district court’s market definition is so narrow that it excludes Apple’s Mac OS, which has competed with Windows for years, simply because the Mac OS runs on a different microprocessor.” Appellant’s Opening Br. at 84. This general, conclusory statement falls far short of what is required to challenge findings as clearly erroneous. Pendleton v. Rumsfeld, 628 F.2d 102, 106 (D.C.Cir.1980); see also Terry v. Reno, 101 F.3d 1412, 1415 (D.C.Cir.1996) (holding that claims made but not argued in a brief are waived). Microsoft neither points to evidence contradicting the District Court’s findings nor alleges that supporting record evidence is insufficient. And since Microsoft does not argue that even if we accept these findings, they do not support the District Court’s conclusion, we have no basis for upsetting the court’s decision to exclude Mac OS from the relevant market. Microsoft’s challenge to the District Court’s exclusion of non-PC based competitors, such as information appliances (han-dheld devices, etc.) and portal websites that host serverbased software applications, suffers from the same defect: the company fails to challenge the District Court’s key factual findings. In particular, the District Court found that because information appliances fall far short of performing all of the functions of a PC, most consumers will buy them only as a supplement to their PCs. Findings of Fact ¶ 23. The District Court also found that portal websites do not presently host enough applications to induce consumers to switch, nor are they likely to do so in the near future. Id. ¶27. Again, because Microsoft does not argue that the District Court’s findings do not support its conclusion that information appliances and portal websites are outside the relevant market, we adhere to that conclusion. This brings us to Microsoft’s main challenge to the District Court’s market definition: the exclusion of middleware. Because of the importance of middleware to this ease, we pause to explain what it is and how it relates to the issue before us. Operating systems perform many functions, including allocating computer memory and controlling peripherals such as printers and keyboards. See Direct Testimony of Frederick Warren-Boulton ¶ 20, reprinted in 5 J.A. at 3172-73. Operating systems also function as platforms for software applications. They do this by “exposing”—i.e., making available to software developers—routines or protocols that perform certain widely-used functions. These are known as Application Programming Interfaces, or “APIs.” See Direct Testimony of James Barksdale ¶ 70, reprinted in 5 J.A. at 2895-96. For example, Windows contains an API that enables users to draw a box on the screen. See Direct Testimony of Michael T. Devlin ¶ 12, reprinted in 5 J.A. at 3525. Software developers wishing to include that function in an application need not duplicate it in their own code. Instead, they can “call”—ie., use—the Windows API. See Direct Testimony of James Barksdale ¶ ¶ 70-71, reprinted in 5 J.A. at 2895-97. Windows contains thousands of APIs, controlling everything from data storage to font display. See Direct Testimony of Michael Devlin ¶ 12, reprinted in 5 J.A. at 3525. Every operating system has different APIs. Accordingly, a developer who writes an application for one operating system and wishes to sell the application to users of another must modify, or “port,” the application to the second operating system. Findings of Fact ¶ 4. This process is both timeconsuming and expensive. Id. ¶ 30. “Middleware” refers to software products that expose their own APIs. Id. ¶ 28; Direct Testimony of Paul Maritz ¶ ¶ 234-36, reprinted in 6 J.A. at 3727-29. Because of this, a middleware product written for Windows could take over some or all of Windows’s valuable platform functions—that is, developers might begin to rely upon APIs exposed by the middleware for basic routines rather than relying upon the API set included in Windows. If mid-dleware were written for multiple operating systems, its impact could be even greater. The more developers could rely upon APIs exposed by such middleware, the less expensive porting to different operating systems would be. Ultimately, if developers could write applications relying exclusively on APIs exposed by middle-ware, their applications would run on any operating system on which the middleware was also present. See Direct Testimony of Avadis Tevanian, Jr. ¶ 45, reprinted in 5 J.A. at 3113. Netscape Navigator and Java—both at issue in this case—are mid-dleware products written for multiple operating systems. Findings of Fact ¶ 28. Microsoft argues that, because middle-ware could usurp the operating system’s platform function and might eventually take over other operating system functions (for instance, by controlling peripherals), the District Court erred in excluding Navigator and Java from the relevant market. The District Court found, however, that neither Navigator, Java, nor any other middleware product could now, or would soon, expose enough APIs to serve as a platform for popular applications, much less take over all operating system functions. Id. ¶ ¶ 28-29. Again, Microsoft fails to challenge these findings, instead simply asserting middleware’s “potential” as a competitor. Appellant’s Opening Br. at 86. The test of reasonable interchangeability, however, required the District Court to consider only substitutes that constrain pricing in the reasonably foreseeable future, and only products that can enter the market in a relatively short time can perform this function. See Rothery, 792 F.2d at 218 (“Because the ability of consumers to turn to other suppliers restrains a firm from raising prices above the competitive level, the definition of the ‘relevant market’ rests on a determination of available substitutes.”); see also Findings of Fact ¶ 29 (“[I]t would take several years for middleware ... to evolve” into a product that can constrain operating system pricing.). Whatever middleware’s ultimate potential, the District Court found that consumers could not now abandon their operating systems and switch to mid-dleware in response to a sustained price for Windows above the competitive level. Findings of Fact ¶ ¶ 28, 29. Nor is mid-dleware likely to overtake the operating system as the primary platform for software development any time in the near future. Id. Alternatively, Microsoft argues that the District Court should not have excluded middleware from the relevant market because the primary focus of the plaintiffs’ § 2 charge is on Microsoft’s attempts to suppress middleware’s threat to its operating system monopoly. According to Microsoft, it is “contradict[ory],” 2/26/2001 Ct. Appeals Tr. at 20, to define the relevant market to exclude the “very competitive threats that gave rise” to the action. Appellant’s Opening Br. at 84. The purported contradiction lies between plaintiffs’ § 2 theory, under which Microsoft preserved its monopoly against middleware technologies that threatened to become viable substitutes for Windows, and its theory of the relevant market, under which middleware is not presently a viable substitute for Windows. Because middle-ware’s threat is only nascent, however, no contradiction exists. Nothing in § 2 of the Sherman Act limits its prohibition to actions taken against threats that are already well-developed enough to serve as present substitutes. See infra Section II.C. Because market definition is meant to identify products “reasonably interchangeable by consumers,” du Pont, 351 U.S. at 395, 76 S.Ct. 994, and because middleware is not now interchangeable with Windows, the District Court had good reason for excluding middleware from the relevant market. b. Market power Having thus properly defined the relevant market, the District Court found that Windows accounts for a greater than 95% share. Findings of Fact ¶ 35. The court also found that even if Mac OS were included, Microsoft’s share would exceed 80%. Id. Microsoft challenges neither finding, nor does it argue that such a market share is not predominant. Cf. Grinnell, 384 U.S. at 571, 86 S.Ct. 1698 (87% is predominant); Eastman Kodak Co. v. Image Technical Servs., Inc., 504 U.S. 451, 481, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992) (80%); du Pont, 351 U.S. at 379, 391, 76 S.Ct. 994 (75%). Instead, Microsoft claims that even a predominant market share does not by itself indicate monopoly power. Although the “existence of [monopoly] power ordinarily may be inferred from the predominant share of the market,” Grinnell, 384 U.S. at 571, 86 S.Ct. 1698, we agree with Microsoft that because of the possibility of competition from new entrants, see Ball Mem’l Hosp., Inc., 784 F.2d at 1336, looking to current market share alone can be “misleading.” Hunt-Wesson Foods, Inc. v. Ragu Foods, Inc., 627 F.2d 919, 924 (9th Cir.1980); see also Ball Mem’l Hosp., Inc., 784 F.2d at 1336 (“Market share reflects current sales, but today’s sales do not always indicate power over sales and price tomorrow.”) In this case, however, the District Court was not misled. Considering the possibility of new rivals, the court focused not only on Microsoft’s present market share, but also on the structural barrier that protects the company’s future position. Conclusions of Law, at 36. That barrier — the “applications barrier to entry” — stems from two characteristics of the software market: (1) most consumers prefer operating systems for which a large number of applications have already been written; and (2) most developers prefer to write for operating systems that already have a substantial consumer base. See Findings of Fact ¶ ¶ 30, 36. This “chicken-and-egg” situation ensures that applications will continue to be written for the already dominant Windows, which in turn ensures that consumers will continue to prefer it over other operating systems. Id. Challenging the existence of the applications barrier to entry, Microsoft observes that software developers do write applications for other operating systems, pointing out that at its peak IBM’s OS/2 supported approximately 2,500 applications. Id. ¶ 46. This misses the point. That some developers write applications for other operating systems is not at all inconsistent with the finding that the applications barrier to entry discourages many from writing for these less popular platforms. Indeed, the District Court found that IBM’s difficulty in attracting a larger number of software developers to write for its platform seriously impeded OS/2’s success. Id. ¶46. Microsoft does not dispute that Windows supports many more applications than any other operating system. It argues instead that “[i]t defies common sense” to suggest that an operating system must support as many applications as Windows does (more than 70,000, according to the District Court, id. ¶ 40) to be competitive. Appellant’s Opening Br. at 96. Consumers, Microsoft points out, can only use a very small percentage of these applications. Id. As the District Court explained, however, the applications barrier to entry gives consumers reason to prefer the dominant operating system even if they have no need to use all applications written for it: The consumer wants an operating system that runs not only types of applications that he knows he will want to use, but also those types in which he might develop an interest later. Also, the consumer knows that if he chooses an operating system with enough demand to support multiple applications in each product category, he will be less likely to find himself straitened later by having to use an application whose features disappoint him. Finally, the average user knows that, generally speaking, applications improve through successive versions. He thus wants an operating system for which successive generations of his favorite applications will be released — promptly at that. The fact that a vastly larger number of applications are written for Windows than for other PC operating systems attracts consumers to Windows, because it reassures them that their interests will be met as long as they use Microsoft’s product. Findings of Fact ¶ 37. Thus, despite the limited success of its rivals, Microsoft benefits from the applications barrier to entry. Of course were middleware to succeed, it would erode the applications barrier to entry. Because applications written for multiple operating systems could run on any operating system on which the middle-ware product was present with little, if any, porting, the operating system market would become competitive. Id. ¶ ¶ 29, 72. But as the District Court found, middle-ware will not expose a sufficient number of APIs to erode the applications barrier to entry in the foreseeable future. See id. ¶¶ 28-29. Microsoft next argues that the applications barrier to entry is not an entry barrier at all, but a reflection of Windows’ popularity. It is certainly true that Windows may have gained its initial dominance in the operating system market competitively—through superior foresight or quality. But this case is not about Microsoft’s initial acquisition of monopoly power. It is about Microsoft’s efforts to maintain this position through means other than competition on the merits. Because the applications barrier to entry protects a dominant operating system irrespective of quality, it gives Microsoft power to stave off even superior new rivals. The barrier is thus a characteristic óf the operating system market, not of Microsoft’s popularity, or, as asserted by a Microsoft witness, the company’s efficiency. See Direct Testimony of Richard Schmalensee ¶ 115, reprinted in 25 J.A. at 16153-14. Finally, Microsoft argues that the District Court should not have considered the applications barrier to entry because it reflects not a cost borne disproportionately by new entrants, but one borne by all participants in the operating system market. According to Microsoft, it had to make major investments to convince software developers to write for its new operating system, and it continues to “evangelize” the Windows platform today. Whether costs borne by all market participants should be considered entry barriers is the subject of much debate. Compare 2A Areeda & Hovenkamp, Antiteust Law § 420c, at 61 (arguing that these costs are entry barriers), and Joe S. Bain, BaReiees to New Competition: Their Character and Consequences in Manufacturing Industries 6-7 (1956) (considering these costs entry barriers), with L.A. Land Co. v. Brunswick Corp., 6 F.3d 1422, 1428 (9th Cir.1993) (evaluating cost based on “[t]he disadvantage of new entrants as compared to incumbents”), and Geokge Stiglee, The Organization of Industry 67 (1968) (excluding these costs). We need not resolve this issue, however, for even under the more narrow definition it is clear that there are barriers. When Microsoft entered the operating system market with MS-DOS and the first version of Windows, it did not confront a dominant rival operating system with as massive an installed base and as vast an existing array of applications as the Windows operating systems have since enjoyed. Findings of Fact ¶ ¶ 6, 7, 43. Moreover, when Microsoft introduced Windows 95 and 98, it was able to bypass the applications barrier to entry that protected the incumbent Windows by including APIs from the earlier version in the new operating systems. See id. ¶ 44. This made porting existing Windows applications to the new version of Windows much less costly than porting them to the operating systems of other entrants who could not freely include APIs from the incumbent Windows with their own. 2. Direct Proof Having sustained the District Court’s conclusion that circumstantial evidence proves that Microsoft possesses monopoly power, we turn to Microsoft’s alternative argument that it does not behave like a monopolist. Claiming that software competition is uniquely “dynamic,” Appellant’s Opening Br. at 84 (quoting Findings of Fact ¶ 59), the company suggests a new rule: that monopoly power in the software industry should be proven directly, that is, by examining a company’s actual behavior to determine if it reveals the existence of monopoly power. According to Microsoft, not only does no such proof of its power exist, but record evidence demonstrates the absence of monopoly power. The company claims that it invests heavily in research and development, id. at 88-89 (cit~ ing Direct Testimony of Paul Maritz ¶ 155, reprinted in 6 J.A. at 8698 (testifying that Microsoft invests approximately 17% of its revenue in R&D)), and charges a low price for Windows (a small percentage of the price of an Intel-compatible PC system and less than the price of its rivals, id. at 90 (citing Findings of Fact ¶ ¶ 19, 21, 46)). Microsoft’s argument fails because, even assuming that the software market is uniquely dynamic in the long term, the District Court correctly applied the structural approach to determine if the company faces competition in the short term. Structural market power analyses are meant to determine whether potential substitutes constrain a firm’s ability to raise prices above the competitive level; only threats that are likely to materialize in the relatively near future perform this function to any significant degree. Rothery, 792 F.2d at 218 (quoting Lawrence Sullivan, ANTITRUST § 12, at 41 (1977)) (only substitutes that can enter the market “promptly” should be considered). The District Court expressly considered and rejected Microsoft’s claims that innovations such as handheld devices and portal websites would soon expand the relevant market beyond Intel-compatible PC operating systems. Because the company does not challenge these findings, we have no reason to believe that prompt substitutes are available. The structural approach, as applied by the District Court, is thus capable of fulfilling its purpose even in a changing market. Microsoft cites no case, nor are we aware of one, requiring direct evidence to show monopoly power in any market. We decline to adopt such a rule now. Even if we were to require direct proof, moreover, Microsoft’s behavior may well be sufficient to show the existence of monopoly power. Certainly, none of the conduct Microsoft points to — its investment in R&D and the relatively low price of Windows — is inconsistent with the possession of such power. Conclusions of Law, at 37. The R&D expenditures Microsoft points to are not simply for Windows, but for its entire company, which most likely does not possess a monopoly for all of its products. Moreover, because innovation can increase an already dominant market share and further delay the emergence of competition, even monopolists have reason to invest in R&D. Findings of Fact ¶ 61. Microsoft’s pricing behavior is similarly equivocal. The company claims only that it never charged the short-term profit-maximizing price for Windows. Faced with conflicting expert testimony, the District Court found that it could not accurately determine what this price would be. Id. ¶ 65. In any event, the court found, a price lower than the short-term profit-maximizing price is not inconsistent with possession or improper use of monopoly power. Id. ¶ ¶ 65-66. Cf. Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263, 274 (2d Cir.1979) (“[I]f monopoly power has been acquired or maintained through improper means, the fact that the power has not been used to extract [a monopoly price] provides no succor to the monopolist.”). Microsoft never claims that it did not charge the long-term monopoly price. Micosoft does argue that the price of Windows is a fraction of the price of an Intel-compatible PC system and lower than that of rival operating systems, but these facts are not inconsistent with the District Court’s finding that Microsoft has monopoly power. See Findings of Fact ¶ 36 (“Intel-compatible PC operating systems other than Windows [would not] attract[ ] significant demand ... even if Micosoft held its prices substantially above the competitive level.”). More telling, the District Court found that some aspects of Microsoft’s behavior are difficult to explain unless Windows is a monopoly product. For instance, according to the District Court, the company set the price of Windows without considering rivals’ prices, Findings of Fact ¶ 62, something a firm without a monopoly would have been unable to do. The District Court also found that Microsoft’s pattern of exclusionary conduct could only be rational “if the firm knew that it possessed monopoly power.” Conclusions of Law, at 37. It is to that conduct that we now turn. B. Anticompetitive Conduct As discussed above, having a monopoly does not by itself violate § 2. A firm violates § 2 only when it acquires or maintains, or attempts to acquire or maintain, a monopoly by engaging in exclusionary conduct “as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.” Grinnell, 384 U.S. at 571, 86 S.Ct. 1698; see also United States v. Aluminum Co. of Am., 148 F.2d 416, 430 (2d Cir.1945) (Hand, J.) (“The successful competitor, having been urged to compete, must not be turned upon when he wins.”). In this case, after concluding that Microsoft had monopoly power, the District Court held that Microsoft had violated § 2 by engaging in a variety of exclusionary acts (not including predatory pricing), to maintain its monopoly by preventing the effective distribution and use of products that might threaten that monopoly. Specifically, the District Court held Microsoft liable for: (1) the way in which it integrated IE into Windows; (2) its various dealings with Original Equipment Manufacturers (“OEMs”), Internet Access Providers (“IAPs”), Internet Content Providers (“ICPs”), Independent Software Vendors (“ISVs”), and Apple Computer; (3) its efforts to contain and to subvert Java technologies; and (4) its course of conduct as a whole. Upon appeal, Microsoft argues that it did not engage in any exclusionary conduct. Whether any particular act of a monopolist is exclusionary, rather than merely a form of vigorous competition, can be difficult to discern: the means of illicit exclusion, like the means of legitimate competition, are myriad. The challenge for an antitrust court lies in stating a general rule for distinguishing between exclusionary acts, which reduce social welfare, and competitive acts, which increase it. From a century of case law on monopolization under § 2, however, several principles do emerge. First, to be condemned as exclusionary, a monopolist’s act must have an “anticompetitive effect.” That is, it must harm the competitive process and thereby harm consumers. In contrast, harm to one or more competitors will not suffice. “The [Sherman Act] directs itself not against conduct which is competitive, even severely so, but against conduct which unfairly tends to destroy competition itself.” Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 458, 113 S.Ct. 884, 122 L.Ed.2d 247 (1993); see also Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 225, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993) (“Even an act of pure malice by one business competitor against another does, not, without more, state a claim under the federal antitrust laws.... ”). Second, the plaintiff, on whom the burden of proof of course rests, see, e.g., Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 763, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984); see also United States v. Arnold, Schwinn & Co., 388 U.S. 365, 374 n. 5, 87 S.Ct. 1856, 18 L.Ed.2d 1249 (1967), overruled on other grounds, Cont’l T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977), must demonstrate that the monopolist’s conduct indeed has the requisite anticom-petitive effect. See generally Brooke Group, 509 U.S. at 225-26, 113 S.Ct. 2578. In a case brought by a private plaintiff, the plaintiff must show that its injury is “of ‘the type that the statute was intended to forestall,’ ” Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 487-88, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977) (quoting Wyandotte Transp. v. United States, 389 U.S. 191, 202, 88 S.Ct. 379, 19 L.Ed.2d 407 (1967)); no less in a case brought by the Government, it must demonstrate that the monopolist’s conduct harmed competition, not just a competitor. Third, if a plaintiff successfully establishes a prima facie case under § 2 by demonstrating anticompetitive effect, then the monopolist may proffer a “pro-competitive justification” for its conduct. See Eastman Kodak, 504 U.S. at 483, 112 S.Ct. 2072. If the monopolist asserts a procompetitive justification — a nonpretextual claim that its conduct is indeed a form of competition on the merits because it involves, for example, greater efficiency or enhanced consumer appeal — then the burden shifts back to the plaintiff to rebut that claim. Cf. Capital Imaging Assocs., P.C. v. Mohawk Valley Med. Assocs., Inc., 996 F.2d 537, 543 (2d Cir.1993). Fourth, if the monopolist’s pro-competitive justification stands unrebutted, then the plaintiff must demonstrate that the anticompetitive harm of the conduct outweighs the procompetitive benefit. In cases arising under § 1 of the Sherman Act, the courts routinely apply a similar balancing approach under the rubric of the “rule of reason.” The source of the rule of reason is Standard Oil Co. v. United States, 221 U.S. 1, 31 S.Ct. 502, 55 L.Ed. 619 (1911), in which the Supreme Court used that term to describe the proper inquiry under both sections of the Act. See id. at 61-62, 31 S.Ct. 502 (“[WJhen the second section [of the Sherman Act] is thus harmonized with ... the first, it becomes obvious that the criteria to be resorted to in any given case for the purpose of ascertaining whether violations of the section have been committed, is the rule of reason guided by the established law....”). As the Fifth Circuit more recently explained, “[i]t is clear ... that the analysis under section 2 is similar to that under section 1 regardless whether the rule of reason label is applied.... ” Mid-Texas Communications Sys., Inc. v. AT & T, 615 F.2d 1372, 1389 n. 13 (5th Cir.1980) (citing Byars v. Bluff City News Co., 609 F.2d 843, 860 (6th Cir.1979)); see also Cal. Computer Prods., Inc. v. IBM Corp., 613 F.2d 727, 737 (9th Cir.1979). Finally, in considering whether the monopolist’s conduct on balance harms competition and is therefore condemned as exclusionary for purposes of § 2, our focus is upon the effect of that conduct, not upon the intent behind it. Evidence of the intent behind the conduct of a monopolist is relevant only to the extent it helps us understand the likely effect of the monopolist’s conduct. See, e.g., Chicago Bd. of Trade v. United States, 246 U.S. 231, 238, 38 S.Ct. 242, 62 L.Ed. 683 (1918) (“knowledge of intent may help the court to interpret facts and to predict consequences”); Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585, 603, 105 S.Ct. 2847, 86 L.Ed.2d 467 (1985). With these principles in mind, we now consider Microsoft’s objections to the District Court’s holding that Microsoft violated § 2 of the Sherman Act in a variety of ways. 1. Licenses Issued to Original Equipment Manufacturers The District Court condemned a number of provisions in Microsoft’s agreements licensing Windows to OEMs, because it found that Microsoft’s imposition of those provisions (like many of Microsoft’s other actions at issue in this case) serves to reduce usage share of Netscape’s browser and, hence, protect Microsoft’s operating system monopoly. The reason market share in the browser market affects market power in the operating system market is complex, and warrants some explanation. Browser usage share is important because, as we explained in Section II.A above, a browser (or any middleware product, for that matter) must have a critical mass of users in order to attract software developers to write applications relying upon the APIs it exposes, and away from the APIs exposed by Windows. Applications written to a particular browser’s APIs, however, would run on any computer with that browser, regardless of the underlying operating system. “The overwhelming majority of consumers will only use a PC operating system for which there already exists a large and varied set of ... applications, and for which it seems relatively certain that new types of applications and new versions'of existing applications will continue to be marketed....” Findings of Fact ¶ 30. If a consumer could have access to the applications he desired — regardless of the operating system he uses — simply by installing a particular browser on his computer, then he would no longer feel compelled to select Windows in order to have access to those applications; he could select an operating system other than Windows based solely upon its quality and price. In other words, the market for operating systems would be competitive. Therefore, Microsoft’s efforts to gain market share in one market (browsers) served to meet the threat to Microsoft’s monopoly in another market (operating systems) by keeping rival browsers from gaining the critical mass of users necessary to attract developer attention away from Windows as the platform for software development. Plaintiffs also argue that Microsoft’s actions injured competition in the browser market — an argument we will examine below in relation to their specific claims that Microsoft attempted to monopolize the browser market and unlawfully tied its browser to its operating system so as to foreclose competition in the browser market. In evaluating the § 2 monopoly maintenance claim, however, our immediate concern is with the anticompetitive effect of Microsoft’s conduct in preserving its monopoly in the operating system market. In evaluating the restrictions in Microsoft’s agreements licensing Windows to OEMs, we first consider whether plaintiffs have made out a prima facie case by demonstrating that the restrictions have an anticompetitive effect. In the next subsection, we conclude that plaintiffs have met this burden as to all the restrictions. We then consider Microsoft’s proffered justifications for the restrictions and, for the most part, hold those justifications insufficient. a. Anticompetitive effect of the license restrictions The restrictions Microsoft places upon Original Equipment Manufacturers are of particular importance in determining browser usage share because having an OEM pre-install a browser on a computer is one of the two most cost-effective methods by far of distributing browsing software. (The other is bundling the browser with internet access software distributed by an IAP.) Findings of Fact ¶ 145. The District Court found that the restrictions Microsoft imposed in licensing Windows to OEMs prevented many OEMs from distributing browsers other than IE. Conclusions of Law, at 39-40. In particular, the District Court condemned the license provisions prohibiting the OEMs from: (1) removing any desktop icons, folders, or “Start” menu entries; (2) altering the initial boot sequence; and (3) otherwise altering the appearance of the Windows desktop. Findings of Fact ¶ 213. The District Court concluded that the first license restriction — the prohibition upon the removal of desktop icons, folders, and Start menu entries — thwarts the distribution of a rival browser by preventing OEMs from removing visible means of user access to IE. Id. ¶ 203. The OEMs cannot practically install a second browser in addition to IE, the court found, in part because “[p]re-installing more than one product in a given category ... can significantly increase an OEM’s support costs, for the redundancy can lead to confusion among novice users.” Id. ¶ 159; see also id. ¶217. That is, a certain number of novice computer users, seeing two browser icons, will wonder which to use when and will call the OEM’s support line. Support calls are extremely expensive and, in the highly competitive original equipment market, firms have a strong incentive to minimize costs. Id. If 210. Microsoft denies the “consumer confusion” story; it observes that some OEMs do install multiple browsers and that executives from two OEMs that do so denied any knowledge of consumers being confused by multiple icons. See 11/5/98 pm Tr. at 41-42 (trial testimony of Avadis Tevanian of Apple), reprinted in 9 J.A. at 5493-94; 11/18/99 am Tr. at 69 (trial testimony of John Soyring of IBM), reprinted in 10 J.A. at 6222. Other testimony, however, supports the District Court’s finding that fear of such confusion deters many OEMs from pre-installing multiple browsers. See, e.g., 01/13/99 pm Tr. at 614-15 (deposition of Microsoft’s Gayle McClain played to the court) (explaining that redundancy of icons may be confusing to end users); 02/18/99 pm Tr. at 46-47 (trial testimony of John Rose of Compaq), reprinted in 21 J.A. at 14237-38 (same); 11/17/98 am Tr. at 68 (deposition of John Kies of Packard Bell-NEC played to the court), reprinted in 9 J.A. at 6016 (same); 11/17/98 am Tr. at 67-72 (trial testimony of Glenn Weadock), reprinted in 9 J.A. at 6015-20 (same). Most telling, in presentations to OEMs, Microsoft itself represented that having only one icon in a particular category would be “less confusing for endusers.” See Government’s Trial Exhibit (“GX”) 319 at MS98 0109453. Accordingly, we reject Microsoft’s argument that we should vacate the District Court’s Finding of Fact 159 as it rela