Mr. Charles F. Rule

December 12, 2001

Mr. Chairman and members of the Committee, good morning. It is a pleasure to appear before you today on behalf of Microsoft Corporation to discuss the proposed consent decree or Revised Proposed Final Judgment (the "PFJ") to which the U.S. Department of Justice and nine of the plaintiff states have agreed. As this committee is aware, I am counsel to Microsoft in the case and was one of the principal representatives for the company in the negotiations that led to the proposed consent decree.

The PFJ was signed on November 6th after more than a month of intense, around-the-clock negotiations with the Department and representatives of all the plaintiff states. The decree is currently subject to a public interest review by Judge Kollar-Kotelly under the Tunney Act . Because we are currently in the midst of that review and because nine states and the District of Columbia have chosen to continue the litigation, I must be somewhat circumspect in my remarks. However, what I can -- indeed, must -- stress is that, in light of the Court of Appeals' decision last summer to "drastically" reduce the scope of Microsoft's liability and in light of the legal standards for imposing injunctive relief, the Department and the settling states were very effective in negotiating for broad, strong relief. As the chart in the appendix depicts, ever since the Department and the plaintiff states first filed their complaints in May 1998, the case has been shrinking. What began with five claims, was whittled down to a single monopoly maintenance claim by a unanimous Court of Appeals. Even with respect to that surviving claim, the appellate court affirmed Judge Jackson's findings on only about a third (12 of 35) of the specific acts which the district court had found support that claim.

Given that history and the law, there is no reasonable argument that the PFJ is too narrow or that it fails to achieve all the relief to which the Department was entitled. In fact, as these remarks explain, the opposite is true -- faced with tough, determined negotiators on the other side of the table, Microsoft agreed to a decree that goes substantially beyond what the plaintiffs were likely to achieve through litigation. Quite frankly, the PFJ is the strongest, most regulatory conduct decree ever obtained (through litigation or settlement) by the Department.

Why then, one might ask, would Microsoft consent to such a decree? There are two reasons. First, the company felt strongly that it was important to put this matter behind it and to move forward constructively with its customers, its business partners, and the government. For four years, the litigation has consumed enormous resources and been a serious distraction. The constant media drumbeat has obscured the fact that the company puts a premium on adhering to its legal obligations and on developing and maintaining excellent relationships with its partners and customers. Litigation is never a pleasant experience, and given the magnitude of this case and the media attention it attracted, it is hard to imagine any more costly, unpleasant civil litigation.

Second, while the Department pushed Microsoft to make substantial, even excessive concessions to get a settlement, there were limits to how far the company was willing or able to go (limits, by the way, which the Department and the settling states managed to reach). Microsoft was fighting for an important principle -- the ability to innovate and improve its products and services for the benefit of consumers. To that end, Microsoft insisted that the decree be written in a way to allow the company to engage in legitimate competition on the merits. Despite the substantial burdens the decree will impose on Microsoft and the numerous ways in which Microsoft will be forced to alter its conduct, the decree does preserve Microsoft's ability to innovate, to improve its products, and to engage in procompetitive business conduct that is necessary for the company to survive.

In short, at the end of the negotiations, Microsoft concluded that the very real costs that the decree imposes on the company are outweighed by the benefits, not just to Microsoft but to the PC industry and consumers generally.

The Court of Appeals' "Road Map" for Relief

In order to evaluate the decree, one must first appreciate the history of this case and how drastically the scope of Microsoft's liability was narrowed at the appellate level. When this case began with the filing of separate complaints by the Department and the plaintiff states in May of 1998, it was focused on Microsoft's integration of browsing functionality called Internet Explorer or IE into Windows 98, which the plaintiffs alleged to be an illegal tying arrangement.

The complaints of the Department and the states included five separate claims: (1) a claim under section 1 of the Sherman Act that the tie-in was per se illegal; (2) another claim under section 1 that certain promotion and distribution agreements with Internet service providers (ISPs), Internet content providers (ICPs), and on-line service providers (OSPs) constituted illegal exclusive dealing; (3) a claim under section 2 of the Sherman Act that Microsoft had attempted to monopolize Web browsing software; (4) a catch-all claim under section 2 that the alleged conduct that underlay the first three claims amounted to illegal maintenance of Microsoft's monopoly in PC operating systems; and (5) a claim by the plaintiff states (but not part of the Department's complaint) under section 2 that Microsoft illegally "leveraged" its monopoly in PC operating systems. As discovery got underway, the case dramatically expanded as the plaintiffs indiscriminately began identifying all manner of Microsoft conduct as examples of the company's illegal efforts to maintain its monopoly. But then, the case began to shrink.

" In response to Microsoft's motion for summary judgment, the district court dismissed the states' Monopoly leveraging claim (claim 5).

" After trial, Judge Jackson held that the plaintiffs failed to prove that Microsoft's arrangements with ISPs, ICPs, and OSPs violated section 1 (claim 2).

" Judge Jackson did, however, conclude that the plaintiffs had sustained their claims that Microsoft illegally tied IE to Windows (claim 1), illegally attempted to monopolize the browser market (claim 3), and illegally maintained its monopoly (claim 4), basing his decision on 35 different actions engaged in by Microsoft.

" In a unanimous decision of the Court of Appeals sitting en banc, the court reversed the trial court on the attempted monopolization claim (claim 3) and remanded with instructions that judgment be entered on that claim in favor of Microsoft.

" The unanimous court also reversed Judge Jackson's decision with respect to the tie-in claim (claim 1). The appellate court held that, in light of the prospect of consumer benefit from integrating new functionality into platform software such as Windows, Microsoft's integration of IE into Windows had to be judged under the rule of reason rather than the per se approach taken by Judge Jackson. The Court of Appeals refused to apply the per se approach because of "our qualms about redefining the boundaries of a defendant's product and the possibility of consumer gains from simplifying the work of applications developers [by ensuring the ubiquitous dissemination of compatible APIs]." The court's decision did allow the plaintiffs on remand to pursue the tie-in claim on a rule of reason theory; however, shortly after the remand, the plaintiffs announced they were dropping the tie-in claim.

" With respect to the only remaining claim (monopoly maintenance - claim 4), the Court of Appeals affirmed in part and reversed in part the lower court and substantially shrank Microsoft's liability. After articulating a four-step burden-shifting test that is highly fact intensive, the appellate court reviewed the 35 different factual bases for liability and rejected nearly two-thirds of them.

¢ In the case of seven of those 35 findings (concerning such conduct as Microsoft's refusal to allow OEMs to replace the Windows desktop, Microsoft's design of Windows to "override the user's choice of a default browser," and Microsoft's development of a Java virtual machine (JVM) that was incompatible with Sun's JVM), the appellate court specifically reversed Judge Jackson's decision.

¢ The Court of Appeals dismissed sixteen of the remaining findings by reversing Judge Jackson's holding that Microsoft had engaged in a general "course of conduct" that amounted to illegal monopoly maintenance -- the so-called "monopoly broth" theory.

¢ With respect to the remaining twelve findings (concerning such things as Microsoft's refusal to allow PC manufacturers (OEMs) to remove end-user access to IE, Microsoft's exclusive arrangements with ISPs, and its "commingling" of software code to frustrate OEMs ability to hide access to IE), the court did affirm Judge Jackson's findings as not being "clearly erroneous." And even as to those twelve, a number were practices -- for example, the arrangements with ISPs -- that Microsoft had already ceased.

As a result, when the case was remanded to the district court and reassigned to Judge Kollar-Kotelly, four-fifths of the original claims were all but gone. With respect to the sole surviving claim, nearly two-thirds of the supporting findings had been rejected by the Court of Appeals. In the words of the Court of Appeals, its decision "drastically altered the scope of Microsoft's liability."

The Relevance of the Drastic Narrowing of Liability

The Court of Appeals' decision makes clear the critical significance of the drastic reduction in the scope of Microsoft's liability in terms of the relief to which the plaintiffs are entitled. As the court noted in instructing the lower court on how the remand for remedy should be handled,

"A court . . . must base its relief on some clear 'indication of a significant causal connection between the conduct enjoined or mandated and the violation found directed toward the remedial goal intended.' 3 PHILLIP E. AREEDA & HERBERT HOVENKAMP, ANTITRUST LAW 653(b), at 91-92 (1996). In a case such as the one before us where sweeping equitable relief is employed to remedy multiple violations, and some -- indeed most -- of the findings of remedial violations do not withstand appellate scrutiny, it is necessary to vacate the remedy decree since the implicit findings of causal connection no longer exist to warrant our deferential affirmance. . . . In particular, the [district] court should consider which of the decree's conduct restrictions remain viable in light of our modification of the original liability decision."

At the time Judge Kollar-Kotelly ordered the parties into intensive negotiations, she clearly recognized the importance of the drastic alteration to the scope of Microsoft's liability. The judge informed the government that its "first and most obvious task is going to be to determine which portions of the former judgment remain appropriate in light of the appellate court's ruling and which portions are unsupported following the appellate court's narrowing of liability." The judge went on to note that "the scope of any proposed remedy must be carefully crafted so as to ensure that the enjoining conduct falls within the [penumbra] of behavior which was found to be anticompetitive." The judge also stated that "Microsoft argues that some of the terms of the former judgment are no longer appropriate, and that is correct. I think there are certain portions where the liability has been narrowed."

Before discussing the negotiations and the decree itself, I would like to make three other points about the crafting of antitrust remedies that also are relevant to considering the relief to which the plaintiffs were entitled. First, the critics of the PFJ routinely ignore the fact that the Department has long acknowledged that Microsoft lawfully acquired its monopoly position in PC operating systems. Indeed, the Department retained a Nobel laureate in the first Microsoft case in 1994 to submit an affidavit to the district court opining that Microsoft had reached its position in PC operating systems through luck, skill, and foresight. It is true of course that Microsoft has now been found liable for engaging in conduct that amounted to illegal efforts to maintain that position; however, there is precious little in the record establishing any causal link between the twelve illegal acts of "monopoly maintenance" and Microsoft's current position in the market for PC operating systems. Thus, contrary to the critics' overheated rhetoric, there is no basis for relief designed to terminate an "illegal monopoly."

Second, decrees in civil antitrust cases are designed to remedy, not to punish. All too often, the critics of this decree speak as though Microsoft was convicted of a crime. It was not. This is a civil case, subject to the rules of civil rather than criminal procedure. To the extent the plaintiffs tried to get relief that could be deemed punitive, that relief would have been rejected.

Third, a decree must serve the purposes of the antitrust laws, which is a "consumer welfare prescription." I realize we are in the "season of giving," but an antitrust decree is not a Christmas tree to fulfill the wishes of competitors, particularly where that fulfillment comes at the expense of consumer welfare. Calls for royalty-free licensing of Microsoft's intellectual property, or for imposing obligations on Microsoft to distribute third party software at no charge, or for Microsoft to facilitate the distribution of an infinite variety of bastardized versions of Windows (and make sure they all run perfectly) are great for a small group of competitors who know that such provisions will quickly destroy Microsoft's incentives and ability to compete (not to mention violate the Constitution's proscription against "takings"). Such calls, however, are anathema to consumers' interests in a dynamic, innovative computer industry. Twenty years ago, my old boss and antitrust icon, Bill Baxter, warned about the anticompetitive consequences of antitrust decrees designed simply to "add sand to the saddlebags" of a particularly fleet competitor like Microsoft. It's a warning the courts would certainly heed today.

To their credit, the negotiators for the Department and the settling states understood these three fundamental antitrust principles. While we may have had to remind the other side of these principles from time to time, we did not have to negotiate for their adherence to them. Taxpayers and consumers can be proud that their interests were represented by honorable men and women with the utmost respect for the rule of law. For others to insinuate that, by agreeing to a decree that honors these three fundamental principles, the Department and the settling states "caved" or settled for inadequate relief is as offensive as it is laughable.

The Negotiations

It is against the background I have sketched that, on September 27th, Judge Kollar-Kotelly ordered the parties into intensive, "around the clock" negotiations. Microsoft had already indicated publicly its strong desire to try to settle the case, and so it welcomed the judge's order. As has been widely reported, all the parties in the case took the court's order very seriously. Microsoft assembled in Washington, D.C., a core team of in-house and outside lawyers who have been living with this case for years, and who spent virtually all of the next five weeks camped out in my offices down the street. Microsoft's top legal officer was in town during much of the period directing the negotiations. Back in Redmond, the company's most senior executives devoted a great deal of time and energy to the process, and we were all supported by a large group of dedicated lawyers, businesspeople, and staff.

From my vantage point, the Department and the states (at least those that settled) made an equivalent effort. As the mediator wrote after the process ended, "No party was left out of the negotiations. ... Throughout most of the mediation the 19 states (through their executive committee representatives) and the federal government (through the staff of the antitrust division) worked as a combined 'plaintiffs' team." Jay Himes from the office of the New York Attorney General Eliot Spitzer and Beth Finnerty from the office of the Ohio Attorney General Betty Montgomery represented the states throughout the negotiations, putting in the same long hours as the rest of us. At various points Mr. Himes and Ms. Finnerty were joined by representatives from other states, including Kevin O'Connor from the office of Wisconsin Attorney General James Doyle.

The negotiations began on September 28th and continued virtually non-stop until November 6th. During the first two weeks, we negotiated without the benefit of a mediator. As they say in diplomatic circles, the discussions were "full and frank." The Department lawyers and the state representatives in the negotiation were extremely knowledgeable, diligent, and formidable.

Microsoft certainly hoped to be able to reach a settlement quickly and before a mediator was designated. However, the views on all sides were sufficiently strong and the need to pay attention to every sentence, phrase, and punctuation mark so overwhelming that reaching agreement proved impossible in those first two weeks. Eric Green, a prominent mediation specialist, was appointed by the court and with the help of Jonathan Marks spent the next three weeks helping the parties find common ground. As Professor Green and Mr. Marks wrote after the mediation ended,

"Successful mediations are ones in which mediators and parties work to identify and overcome barriers to reaching agreement. Successful mediations are ones in which all the parties engage in reasoned discussions of issues that divide them, of options for settlement, and of the risks, opportunities, and costs that each party faces if a settlement isn't reached. Successful mediations are ones in which, settle or not, senior representatives of each party have made informed and intelligent decisions. The Microsoft mediation was successful."

Working day and night virtually until the original November 2 deadline set by the judge, Microsoft and the Department agreed to and signed a decree early on November 2. The representatives of the states also tentatively agreed, subject to an opportunity from November 2 until November 6 to confer with the other states that were more removed from the case and negotiations. During that period, the states requested several clarifying modifications to which Microsoft (and the Department) agreed. From press reports, it appears that during this period the plaintiff states also were being subjected to intense lobbying by a few of Microsoft's competitors who were desperate either to get a decree that would severely cripple if not eventually destroy Microsoft or at least to keep the litigation (and the attendant costs imposed on Microsoft) going. Notwithstanding that pressure, New York, Wisconsin, and Ohio -- the states that had made the largest investment in litigating against Microsoft and in negotiating a settlement -- along with six other plaintiff states represented by a bipartisan group of state attorneys general signed onto the Revised PFJ on November 6.

The Proposed Final Judgment

Throughout the negotiations, Microsoft was confronted by a determined and tough group of negotiators for the Department and the states. They made clear that there would be no settlement unless Microsoft went well beyond the relief to which, Microsoft believes, the Court of Appeals opinion and the law entitles the plaintiffs. Once that became clear, Microsoft relented in significant ways, subject only to narrow language that preserved Microsoft's ability to innovate and engage in normal, clearly procompetitive activities. Professor Green, the one neutral observer of this drama, has noted the broad scope of the prohibitions and obligations imposed on Microsoft by the PFJ, stating during the status conference with Judge Kollar-Kotelly that "the parties have not stopped at the outer limits of the Court of Appeals' decision, but in some important respects the proposed final judgment goes beyond the issues affirmed by the Court of Appeals to deal with issues important to the parties in this rapidly-changing technology."

I do not intend today to provide a detailed description of each provision of the PFJ; the provisions speak for themselves. It may come as something of a surprise in light of some of the uninformed criticism hurled at the decree, but one of Microsoft's principal objectives during the negotiations was to develop proscriptions and obligations that were sufficiently clear, precise and certain to ensure that the company and its employees would be able to understand and comply with the decree without constantly engendering disputes with the Department. This is an area of complex technology and the decree terms on which the Department insisted entailed a degree of technical sophistication that is unprecedented in an antitrust decree. Drafting to these specifications was not easy, but the resulting PFJ is infinitely clearer and easier to administer than the conduct provisions of the decree that Judge Jackson imposed in June 2000.

If, as one might suspect would be the outcome in a case such as this, the PFJ were written to proscribe only the twelve practices affirmed by the Court of Appeals, the decree would be much shorter and simpler. The Department and settling states, however, insisted that the decree go beyond just focused prohibitions to create much more general protections for a potentially large category of software, which the PFJ calls "middleware." But even these expansive provisions to foster middleware competition were not sufficient to induce the Department and the states to settle; rather, they insisted that Microsoft also agree to additional obligations that bear virtually no relationship to any of the issues addressed by the district court and the Court of Appeals. And lastly they insisted on unprecedented enforcement provisions. I will briefly describe each of these three sets of provisions.

1. Protections for "Middleware"

The case that the plaintiffs tried and the narrowed liability that survived appellate review all hinged on claims that Microsoft took certain actions to exclude Netscape's Navigator browser and Sun's Java technology from the market in order to protect the Windows operating system monopoly. The plaintiffs successfully argued that Microsoft feared that Navigator and Java, either alone or together, might eventually include and expose a broad set of general purpose APIs to which software developers could write as an alternative to the Windows APIs. Since Navigator and Java can run on multiple operating systems, if they developed into general purpose platforms, Navigator and Java would provide a means of overcoming the "applications barrier" to entry and threaten the position of the Windows operating system as platform software.

A person might expect that a decree designed to address such a monopoly maintenance claim would provide relief with respect to Web-browsing software and Java or, at most, to other general purpose platform software that exposes a broad set of APIs and is ported to run on multiple operating systems. The PFJ goes much further. The Department insisted that obligations imposed on Microsoft by the decree extend to a range of software that has little in common with Navigator and Java. The decree applies to "middleware" broadly defined to include, in addition to Web-browsing software and Java, instant messaging software, media players, and even email clients -- software that, Microsoft believes, has virtually no chance of developing into broad, general purpose platforms that might threaten to displace the Windows platform. In addition, there is a broad catch-all definition of middleware that in the future is likely to sweep other similar software into the decree.

This sweeping definition of middleware is significant because of the substantial obligations it imposes on Microsoft. Those obligations -- a number of which lack any correspondence to the monopoly maintenance findings that survived appellate review -- are intended to create protections for all the vendors of software that fits within the middleware definition. Taken together, the decree provisions provide the following protections and opportunities:

" Relations with Computer Makers. Microsoft has agreed not to retaliate against computer makers who ship software that competes with anything in its Windows operating system.

" Computer Maker Flexibility. Microsoft has agreed to grant computer makers broad new rights to configure Windows so as to promote non-Microsoft software programs that compete with features of Windows. Computer makers will now be free to remove the means by which consumers access important features of Windows, such as Internet Explorer, Windows Media Player, and Windows Messenger. Notwithstanding the billions of dollars Microsoft invests developing such cool new features, computer makers will now be able to replace access to them in order to give prominence to non-Microsoft software such as programs from AOL Time Warner or RealNetworks. (Additionally, as is the case today, computer makers can provide consumers with a choice --that is to say access to Windows features as well as to non-Microsoft software programs.)

" Windows Design Obligations. Microsoft has agreed to design future versions of Windows, beginning with an interim release of Windows XP, to provide a mechanism to make it easy for computer makers, consumers and software developers to promote non-Microsoft software within Windows. The mechanism will make it easy to add or remove access to features built in to Windows or to non-Microsoft software. Consumers will have the freedom to choose to change their configuration at any time.

" Internal Interface Disclosure. Even though there is no suggestion in the Court of Appeals' decision that Microsoft fails to disclose APIs today and even though the Court of Appeals' holding on monopoly power is predicated on the idea that there are tens of thousands of applications written to call upon those APIs, Microsoft has agreed to document and disclose for use by its competitors various interfaces that are internal to Windows operating system products.

" Relations with Software Developers. Microsoft has agreed not to retaliate against software or hardware developers who develop or promote software that competes with Windows or that runs on software that competes with Windows.

" Contractual Restrictions. Microsoft has agreed not to enter into any agreements obligating any third party to distribute or promote any Windows technology exclusively or in a fixed percentage, subject to certain narrow exceptions that apply to agreements raising no competitive concern. Microsoft has also agreed not to enter into agreements relating to Windows that obligate any software developer to refrain from developing or promoting software that competes with Windows.

These obligations go far beyond the twelve practices that the Court of Appeals found to constitute monopoly maintenance. One of the starkest examples of the extent to which these provisions go beyond the Court of Appeals decision relates to Microsoft's obligations to design Windows in such a way as to give third parties the ability to designate non-Microsoft middleware as the "default" choice in certain circumstances in which Windows might otherwise be designed to utilize functionality integrated into Windows. As support for his monopoly maintenance conclusion, Judge Jackson had relied on several circumstances in which Windows was designed to override the end users' choice of Navigator as their default browser and instead to invoke IE. The Court of Appeals, however, reviewed those circumstances and reversed Judge Jackson's conclusion on the ground that Microsoft had "valid technical reasons" for designing Windows as it did. Notwithstanding this clear victory, Microsoft acceded to the Department's demands that it design future versions of Windows to ensure certain default opportunities for non-Microsoft middleware.

2. Uniform Prices and Server Interoperability

Nevertheless, agreeing to this wide range of prohibitions and obligations designed to encourage the development of middleware broadly defined was not enough to get the plaintiffs to settle. Instead, they insisted on two additional substantive provisions that have absolutely no correspondence to the findings of monopoly maintenance liability that survived appeal.

" Uniform Price List. Microsoft has agreed to license its Windows operating system products to the 20 largest computer makers (who collectively account for the great majority of PC sales) on identical terms and conditions, including price (subject to reasonable volume discounts for computer makers who ship large volumes of Windows).

" Client/Server Interoperability. Microsoft has agreed to make available to its competitors, on reasonable and non-discriminatory terms, any protocols implemented in Windows desktop operating systems that are used to interoperate natively with any Microsoft server operating system.

In the case of the sweeping definition of middleware and the range of prohibitions and obligations imposed on Microsoft, there is at least a patina of credibility to the argument that the penumbra of the twelve monopoly maintenance practices affirmed by the Court of Appeals can be stretched to justify those provisions, at least as "fencing in" provisions. There is no sensible reading of the Court of Appeals decision that would provide any basis for requiring Microsoft to charge PC manufacturers uniform prices or to make available the proprietary protocols used by Windows desktop operating systems and Windows server operating systems to communicate with each other. Nevertheless, because the plaintiffs insisted that they would not settle without those two provisions, Microsoft also agreed to them.

Before turning to the enforcement provisions of the PFJ, I want to say a word about the few provisos included in the decree that provide narrow exceptions to the various prohibitions and obligations imposed on Microsoft. Those exceptions were critical to Microsoft's willingness to agree to the sweeping provisions on which the plaintiffs insisted. Without these narrowly tailored exceptions, Microsoft could not innovate or engage in normal procompetitive commercial activities. The public can rest assured that the settling plaintiffs insisted on language to ensure that the exceptions only apply when they promote consumer welfare. For example, some companies that compete with Microsoft for the sale of server operating systems apparently have complained about the so-called "security carve-out" to Microsoft's obligation to disclose internal interfaces and protocols. That exception is very narrow and only allows Microsoft to withhold encryption "keys" and the similar mechanisms that must be kept secret if the security of computer networks and the privacy of user information is to be ensured. In light of all the concern over computer privacy and security these days, it is surprising that there is any controversy over such a narrow exception.

3. Compliance and Enforcement

The broad substantive provisions of the PFJ are complemented by an unusually strong set of compliance and enforcement provisions. Those provisions are unprecedented in a civil antitrust decree. The PFJ creates an independent three-person technical committee, resident on the Microsoft campus, with extraordinary powers and full access to Microsoft facilities, records, employees and proprietary technical data, including Windows source code, which is the equivalent of the "secret formula" for Coke. The technical committee provides a level of technical oversight that is far more substantial than any provision of any other antitrust decree of which I am aware. At the insistence of the plaintiffs, the technical committee does not have independent enforcement authority; rather, reports to the plaintiffs and, through them, to the court. The investigative and oversight authority of the technical committee in no way limits or reduces the enforcement powers of the DOJ and states; rather, the technical committee supplements and enhances those powers. Each of the settling states and DOJ have the power to enforce the decree and have the ability to monitor compliance and seek a broad range of remedies in the event of a violation.

Microsoft also agreed to develop and implement an internal antitrust compliance program, to distribute the decree and educate its management and employees as to the various restrictions and obligations. In recent years, Microsoft has assembled in-house one of the largest, most talented groups of antitrust lawyers in corporate America. They are already engaged in substantial antitrust compliance counseling and monitoring. The decree formalizes those efforts, and quite frankly adds very substantially to the in-house lawyers' work. As we speak, that group, together with key officials from throughout the Microsoft organization, are working to implement the decree and to ensure the company's compliance with it.

As with the substantive provisions, Microsoft agreed to these unprecedented compliance and enforcement provisions because of the adamance of the plaintiffs and because of the highly technical nature of the decree. Microsoft, the Department, and the settling states recognized that it was appropriate to include mechanisms -- principally, the technical committee -- that will facilitate the prompt and expert resolution of any technical disputes that might be raised by third parties, without in any way derogating from the government's full enforcement powers under the decree. Although the enforcement provisions are unprecedented in their stringency and scope, they are not necessitated or justified by any valid claim that Microsoft has failed to comply with its decree obligations in the past. In fact, Microsoft has an exemplary record of complying with the consent decree to which the company and the Department agreed in 1994. In 1997, the Department did question whether Microsoft's integration of IE into Windows 95 violated a "fencing in" provision that prohibited contractual tie-ins, but Microsoft was ultimately vindicated by the Court of Appeals. Microsoft has committed itself to that same level of dedication in ensuring the company's compliance with the PFJ.

Conclusion

The PFJ strikes an appropriate balance in this complicated case, providing opportunities and protections for firms seeking to compete while allowing Microsoft to continue to innovate and bring new technologies to market. The decree is faithful to the fact that the antitrust laws are a "consumer protection prescription," and it ensures an economic environment in which all parts of the PC-ecosystem can thrive.

Make no mistake, however, the PFJ is tough. It will impose substantial new obligations on the company, and it will require significant changes in the way Microsoft does business. It imposes heavy costs on the company and entails a degree of oversight that is unprecedented in a civil antitrust case. For some competitors of Microsoft, however, apparently nothing short of the destruction of Microsoft -- or at least the ongoing distraction of litigation -- will be sufficient. But if the objective is to protect the interests of consumers and the competitive process, then this decree more than achieves that goal.

Finally, for all those who are worried about the future and what unforeseen developments may not be covered by this case and the decree, remember that the Court of Appeals decision now provides guideposts, which previously did not exist, for judging Microsoft's behavior, and that of other high technology companies, going forward. Those guidelines, it is true, are not always easy to apply ex ante to conduct; however, now that the Court of Appeals has spoken, we all have a much better idea of the way in which section 2 of the Sherman Act applies to the software industry. In short, what antitrust law requires of Microsoft is today much clearer than it was when this case began. We have all learned a lot over the last four years, and Microsoft has every incentive to ensure that history does not repeat itself.