Motion for Preliminary Injunction.
Bristol Technology v. Microsoft, Case No. 398 CV 1657 (JCH).
Date: August 18, 1998.
Source: Bristol Technology.  Reprinted with permission.  This document has been edited for HTML, but not for content.

UNITED STATES DISTRICT COURT
DISTRICT OF CONNECTICUT

PUBLIC VERSION
BRISTOL TECHNOLOGY'S MEMORANDUM IN
SUPPORT OF MOTION FOR PRELIMINARY INJUNCTION

TABLE OF CONTENTS

I. INTRODUCTION

II. STATEMENT OF FACTS

A. Microsoft and Windows

B. Bristol

C. The Windows Programming Interface -- The Consummate Lever

D. The WISE Promise

E. The WISE Trap

F. The Fate Of The WISE

III. THE PRELIMINARY INJUNCTION STANDARD

IV. MICROSOFT HAS ABUSED MONOPOLY POWER IN THREE RELEVANT MARKETS

A. Elements Of The Monopolization Claim

B. Microsoft Has Monopoly Power In Three Relevant Markets

1. Microsoft Has A Monopoly Of The Relevant Market For Personal Computer Operating Systems

a. User Alternatives

b. Market Share

c. Structural Indications Of Market Power

(1) Network Effects

(2) Control Of The Dominant Programming Interface

(3) High Entry Costs And Insignificant Marginal Costs

(4) Public Perception Of Microsoft's Invincibility

2. Microsoft Has Monopoly Power In The Relevant Market For Operating Systems For Technical Workstations

a. User Alternatives

b. Market Share

c. Structural Indications Of Market Power

3. Microsoft Has Monopoly Power In The Relevant Market For Departmental Server Operating Systems

a. User Alternatives

b. Market Share

c. Structural Indications Of Market Power

C. Microsoft Has Engaged In Exclusionary Conduct To Maintain And Expand Its Power In The Relevant Markets

1. Microsoft Has Used Its Monopoly Power In The Desktop Market To Gain A Competitive Advantage In The Server And Workstation Markets

2. Microsoft Has Engaged In Exclusionary Conduct To Acquire And Maintain Monopoly Power In The Workstation And Server Markets

a. Product Quality

b. Impact on Competitors

c. Efficiency Justification

3. Microsoft Has Wrongfully Denied Bristol Use Of An Essential Facility

a. Windows Source Code Is An Essential Facility

c. Microsoft Has Denied Access To The Essential Facility

d. It Is Feasible For Microsoft To Give Bristol Access To The Essential Facility

V. MICROSOFT'S REPUDIATION OF THE WISE COMMITMENTS IS AN ATTEMPT TO MONOPOLIZE

A. Anti-competitive Conduct

B. Specific Intent To Monopolize

C. Dangerous Probability Of Success

VI. MICROSOFT'S CONDUCT HAS CAUSED ANTITRUST INJURY TO BRISTOL

VII. BRISTOL IS ENTITLED TO AN INJUNCTION UNDER THE CONNECTICUT ANTITRUST LAW

VIII. BRISTOL IS ENTITLED TO AN INJUNCTION UNDER THE CONNECTICUT UNFAIR TRADE PRACTICES ACT

IX. BRISTOL IS LIKELY TO PREVAIL ON ITS CLAIM THAT MICROSOFT IS ESTOPPED TO WITHHOLD WINDOWS SOURCE CODE FROM BRISTOL

X. MICROSOFT'S WRONGFUL CONDUCT IS CAUSING IRREPARABLE INJURY TO BRISTOL

XI. THE BALANCE OF HARDSHIPS TIPS OVERWHELMINGLY IN BRISTOL'S FAVOR

XII. THERE IS NO NEED TO POST A BOND TO SECURE A PRELIMINARY INJUNCTION

XIII. CONCLUSION

TABLE OF AUTHORITIES

Subsequent history citations appear only in this Table

Statutes

Conn. Gen. Stats § 35-27

§ 35-37

§ 42-1102-1100

Other

U.S. Department of Justice/Federal Trade Comm'n

Horizontal Merger Guidelines,

4 Trade Reg. Rptr. (CCH) ¶ 13,104

Restatement (Second) Contracts § 90

I. INTRODUCTION

A monopoly is like a cancer. It strangles healthy competition, and, unless attacked early, it becomes untreatable. Also like cancer, the treatment of monopoly can nearly kill the patient. Bristol Technology ("Bristol"), a thus-far successful, privately-held company, is in the unfortunate position of coming to this Court for relief from the predatory actions of a consummate monopolist. Unless interim relief is granted, the very act of bringing this action may destroy Bristol's ability to compete.

This case concerns events at the boundary between a monopolized market and adjacent markets where the cancer of monopoly is spreading. The monopolized market is the market for PC operating systems, in which Microsoft occupies an undoubted monopoly. See United States v. Microsoft Corp., 56 F.3d 1448, 1451 (D.C. Cir. 1995). The adjacent markets are the markets for operating systems for advanced workstations and "departmental servers". These adjacent markets appear, at present, to be divided between Microsoft and several versions of UNIX. But this appearance is misleading. Experts agree that Microsoft's monopoly has metastasized into the workstation and server markets, where Microsoft Windows NT ("NT") is growing far more rapidly than any other operating system.

The acts of monopolization at issue consist of arbitrarily withdrawing from a procompetitive arrangement -- the WISE program -- and denying access to an essential facility -- the source code for Windows operating systems. Microsoft deliberately made this facility essential for Bristol's survival; and just as importantly, Microsoft deliberately created market dependence on Bristol's access to this essential facility. Microsoft has reneged on its commitments under the WISE program. It has done so for the specific purpose of assuring its triumph in the markets for workstation and server operating systems. This is not only a clear case of monopolization under state and federal law. It is also a clear violation of the Connecticut Unfair Trade Practices Act ("CUTPA") and a case of promissory estoppel.

Bristol seeks an injunction, pending trial, preventing Microsoft from blocking Bristol's timely and complete access to Windows source code, the facility necessary to enable Bristol to remain in existence as a Microsoft competitor.

II. STATEMENT OF FACTS

A. Microsoft and Windows

The world knows who Microsoft is, and how control of the market for personal computer operating systems fell into its lap by accident. Starting with MS/DOS and evolving into Windows, Microsoft accounts for over 90% of the operating systems used in personal computers.

B. Bristol

Bristol was formed in 1991 by members of the Blackwell family to exploit Ken Blackwell's idea for a cross-platform development tool. The idea was to provide a way to take application programs written for the Windows operating system and convert them to run on UNIX. Putting it another way, Ken Blackwell wanted to create a program that would automatically translate source code written using the Windows programming interface for Windows so that it would work on the UNIX programming interface.

Using their savings and their own ingenuity, the Blackwells were able to develop a program (which they called Wind/U) that facilitated translation of Windows programs to UNIX. Soon after Bristol began to publicize Wind/U, Microsoft contacted Bristol and offered to help Bristol make its product better. Eventually, Microsoft invited Bristol to become a "partner" in WISE, "a licensing program from Microsoft to enable customers to integrate Windows-based solutions with UNIX and Macintosh systems." After much soul searching, but realizing that Microsoft's power left it with no alternative, Bristol entered into a WISE agreement with Microsoft in 1994.

From 1994 until 1997, Bristol bet its future on WISE, and convinced important customers to do likewise. In reliance on WISE, Bristol irrevocably gave up its independent design and development, and created a valuable and effective tool based on use of the promised Windows source code. Indeed, WISE is an acronym for "Windows Interface Source Environment." (Emphasis supplied.)

Beginning in 1997, Microsoft cut off Bristol's timely access to the Windows source code. Bristol has grovelled at Microsoft's doorstep over the intervening year, offering any reasonable accommodation to avoid the damage to its business and reputation that will inevitably result from filing this action. All that Microsoft has offered was a poison pill, certain to destroy Bristol, although perhaps more slowly.

C. The Windows Programming Interface -- The Consummate Lever

People choose computers on the basis of the application programs available for that product, and they recognize (at least implicitly) that application programs require a particular operating system in order to work properly. Thus, it is common for people to talk about preferring "Apple," but choosing "IBM" because there are more and cheaper programs available on the IBM-compatible computers. What this really means is that there are more and cheaper programs available on computers that support the Windows programming interface. A visit to a computer superstore provides this point. The section for Windows programs is huge. There are thousands of application programs offered by hundreds of suppliers. The section for Macintosh programs is small by comparison, and a consumer looking for UNIX application programs will find very little.

Since it is obvious that the owners of Windows-based application programs would be delighted to sell their products to Macintosh or UNIX users if they could, there must be a reason that all the Windows application programs are not available on Apple or UNIX. There is, and this reason is not hard to understand: Each operating system has its own programming interface. A modern operating system supports literally thousands of different "calls" for services. An application program may use any of these calls at any time. Each call in the application program must conform exactly to the formats and protocols specified as the application program interfaces for the operating system. Even the slightest deviation will cause the application to "crash."

Since a programming interface is like a language in its own right, it is difficult for programmers to learn and use two different programming interfaces with equal proficiency, and, even if that proficiency could be acquired, it would take a great deal of effort to translate a program written for one programming interface so that it properly invokes with a different programming interface. Accordingly, programmers are careful to chose the programming interface that will represent the best market opportunity for their product. As Kenneth Arrow, the Nobel prize-winning economist, testified in United States v. Microsoft, "The larger the installed base of a particular operating system, the more likely it is that independent software vendors will write programs to run on that operating system." 56 F.3d at 1452.

Once programmers have written a product using the chosen programming interface, they will translate their code to a second programming interface only if the profit in doing so is greater than they could expect from other work. As the shelf space in the average computer store indicates, translation of an application program from Windows to another programming interface has been the exception not the rule.

Microsoft controls the Windows programming interface. While this interface is developed in consultation with others, Microsoft decides what the Windows programming interface will be, and how it will change. This control of the Windows programming interface gives Microsoft enormous power to influence the supply of application programs that will run on Windows operating systems, and enormous power to influence the supply of application programs that will run on its competitors' operating systems. If Microsoft elects to make it easy to translate programs from the Windows programming interface to competitors' programming interfaces, this will increase the choices available to computer users. If Microsoft elects to make it difficult to translate programs from the Windows programming interface to the interface used by competitive systems, that restricts consumer choices.

As explained more fully below, in this case, Microsoft deliberately used its control of the Windows programming interface to leverage its monopoly of personal computer operating systems into a commanding position in the workstation and server operating systems markets.

D. The WISE Promise

Despite Microsoft's total dominance of the market for PC operating systems, there are a few niches where consumers have consistently chosen other operating systems. Two of these niches are the operating systems for technical workstations ("workstations") and the operating systems for departmental servers ("servers"). In fact, five years ago, Microsoft products represented only a tiny fraction of the operating system software in machines used for either of these applications. The great majority of such machines ran under some version of UNIX. As a result, in these two narrow areas, the most popular application programs were written to the UNIX programming interface.

Recognizing that Windows was at a disadvantage because the better application programs for workstations or servers used the UNIX programming interface, Microsoft drew on its control of the market for desktop PCs to sell programmers -- even UNIX programmers -- on using the Windows programming interface as a "universal API". Specifically, Microsoft promised to "remove the need for developers to learn multiple application program interfaces" and "enable developers to write to Windows APIs and use the resulting applications on Macintosh and various UNIX systems." WISE was the name of this promise.

The advantages of this strategy for Microsoft were significant. Absent WISE, writers of the best UNIX applications would continue to use the UNIX programming interface, and Microsoft's product, NT, would have a harder time breaking into the server and workstation markets. With WISE, NT would be more attractive to workstation and server customers previously loyal to UNIX. As Microsoft explained to Bristol, "[w]e realize that the battle of the APIs is more important than the battle of products at this time, and what you [Bristol] are doing only furthers the cause of the Windows API as the solution for writing applications for the greatest number of users."

Had WISE been an honest commitment, it would have offered advantages to program developers, too. First, because more programmers are "fluent" in the Windows programming interface, developers of server and workstation applications could draw on a larger pool of talent. Similarly, more and cheaper development aids are available for users of the Windows programming interface than any other. Finally, and most importantly, Microsoft's control of the desktop directly affects operating system requirements in the adjacent server and workstation markets. Since servers would almost certainly be "networked" with desktop PCs using Windows, parts of the Windows programming interface must be supported by a server in any event. Getting into the Windows world provided software developers with a hedge against being left behind in the future. In short, WISE (as advertised) offered a "win-win" opportunity for an application developer working in the server or workstation marketplace.

The key to the WISE promise was the assurance that Microsoft would make source code available to developers of cross platform tools. This assured server and workstation companies that they could use of the Windows programming interface and still deliver a top-quality product to their UNIX customers. And this promise was credible only if it extended into the future. A one-time or temporary bridge between Windows and UNIX would not have been acceptable. Microsoft recognized this and told the market:

Companies can adopt a solution that will evolve along with future versions of the Windows family... Microsoft is committed to providing WISE licensees with future versions of Windows family source code.... Customers can be confident that their investments today will continue to realize benefits well into the future.

E. The WISE Trap

Shortly after Bristol first published its plans for Wind/U, Microsoft, the industry giant, sought out tiny Bristol, and expressed an interest in assisting Bristol with its product. After intervening discussions and appraisal of Wind/U's quality, Microsoft invited Bristol to become a WISE "partner." Dazzled by the compliment and intimidated by the risk that it would be left behind otherwise, Bristol accepted this invitation in 1994. As promised in the WISE Mission Statement, Microsoft committed to supply Bristol with access to Windows source code -- all the source code. Bristol was licensed to use this code to support the Windows programming interface on non-Windows platforms. Bristol was also licensed (at a fee) to translate necessary segments of the Windows source code in order to support Windows features not readily available in UNIX products.

Although not the only WISE partner, Bristol came to be recognized as the leading provider of a bridge from Windows to UNIX. With the advantage of the "source environment" designed by Microsoft, Wind/U became a powerful tool to achieve Microsoft's publicly-touted objective to "remove the need for developers to learn multiple application program interfaces (APIs)." The quality of Wind/U convinced customers to build their product strategy on Wind/U, either to "port" Windows applications to UNIX, or to "back port" UNIX applications after writing them initially using the Windows programming interface. These commitments were based on Bristol's proven performance, but current performance alone could not alone make the sale. Customers needed assurance that Wind/U would be just what it claimed to be over the long haul. Customers got that assurance, not from Bristol, but from Microsoft. As Bill Gates emphasized to a convention of UNIX programmers:

[W]e work with [WISE partners] to make sure they've got the very latest Windows API technology. Bristol and Mainsoft also provide source and binary compatibility, and again that's a close relationship where it's just not some old version of Windows, it's the very latest.

After Bristol had given up its independence, after customers had staked their business on the WISE promise, Microsoft sprung the trap. For more than a year, Microsoft has refused to provide Bristol "with future versions of Windows family source code." Microsoft now claims the right to provide just part of the programming interface, and to price a license in a way that makes it impossible for "[c]orporate standards committees [to] choose what they consider to be the best, lowest-cost productivity tools..." Now, Microsoft wants

to replace WISE with WOE ("Windows Owns Everything"), leaving Bristol and Bristol's customers stranded where Microsoft wants them to be -- with no choice except to offer only Windows application programs.

This is a clear breach of Microsoft's assurances. It is a clear act of monopolization.

F. The Fate Of The WISE

The need to file this action may, by itself, have done irreparable injury to Bristol. Program development is a forward-looking, what-have-you-done-for-me-lately business. When customers have doubts about the ability of a supplier to meet commitments, they do not wait for confirmation. They head for the lifeboats.

REDACTED

REDACTED

III. THE PRELIMINARY INJUNCTION STANDARD

To meet the traditional standard for issuance of a preliminary injunction, Bristol must "demonstrate irreparable harm and a likelihood of success on the merits." Jolly v. Coughlin, 76 F.3d 468, 473 (2d Cir. 1996).

Alternatively, Bristol is entitled to a preliminary injunction if it demonstrates irreparable injury and "sufficiently serious questions going to the merits of its claims to make them fair ground for litigation, plus a balance of the hardships tipping decidedly in favor of [the party requesting the preliminary relief]." Acquaire v. Canada Dry Bottling Co., 24 F.3d 401,409 (2d Cir. 1994); see also NAACP v. Town of East Haven, 70 F.3d 219, 223 (2d Cir. 1995).

As shown the following sections, the record submitted with this motion more than meets either of the foregoing standards, based on claims of monopolization under federal and state law, attempted monopolization under federal and state law, violation of the CUTPA and promissory estoppel.

IV. MICROSOFT HAS ABUSED MONOPOLY POWER IN THREE RELEVANT MARKETS

A. Elements Of The Monopolization Claim

Bristol will prove at trial and has offered substantial proof in this motion (1) that Microsoft has a monopoly of three relevant markets, and (2) that Microsoft has maintained its monopolies through exclusionary behavior and use of its monopoly power in the desktop operating system market to gain a competitive advantage in other markets. United States v. Grinnell Corp., 384 U.S. 563, 570-71 (1966); Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d 90, 97 (2d Cir. 1998); Berkey Photo Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir. 1979).

Bristol will show, in addition, that it has standing to complain of Microsoft's exclusionary conduct, and that it is threatened with irreparable injury in its business or property if Microsoft's conduct is allowed to continue.

B. Microsoft Has Monopoly Power In Three Relevant Markets

1. Microsoft Has A Monopoly Of The Relevant Market For Personal Computer Operating Systems.

Definition of the relevant markets is the first step in analysis of a monopolization claim. Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d at 268. A relevant market has a product dimension, i.e., all those products and services that belong in the relevant market ("product market"); and a geographic dimension, i.e., the area of effective competition ("geographic market").

a. User Alternatives

Product market analysis begins with the product or products over which the defendant is alleged to have monopoly power, and asks what alternatives consumers would have if the defendant arbitrarily raised its prices above a competitive level. United States v. Grinnell Corp., 384 U.S. at 571, U.S. Dept. of Justice/Federal Trade Commission, Horizontal Merger Guidelines § 1.1 (1992), 4 Trade Reg. Rptr. (CCH) ¶13,104. All those products to which consumers could reasonably switch in such event should be included in the relevant product market. Nifty Foods Corp v. Great Atl. & Pac. Tea Co., 614 F.2d 832, 840-41 (2d Cir. 1980). Where, however, alternative products would require consumers to accept substantial costs, delays or retraining, such products should not to be included. Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. 451, 473, (1992). Applying these criteria, there is clearly a relevant market consisting of operating systems for personal computers or "PCs". As a practical matter, there are, at present, only two kinds of PCs: so-called "IBM-compatible" computers (more properly called "WinTel" computers) and versions of the Apple Macintosh family. Larger computers are too expensive (and generally too complicated) to be an alternative for PC-class computers.

b. Market Share

According to reliable industry data, Microsoft currently supplies 95% of all operating systems used in desktop computers, and of this total, 94% run under one of the Windows family of operating systems. This is more than sufficient to raise an inference of monopoly power. Eastman Kodak Co. v. Image Tech. Services, Inc., 504 U.S. at 481; Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d at 99.

c. Structural Indications Of Market Power

In addition, there are structural characteristics of this market which make Microsoft's dominance particularly difficult to combat. These include the network effects of software compatibility, control of the dominant programming interface, high entry cost and insignificant marginal costs, and public perception of Microsoft's invincibility.

(1) Network Effects.

People tend to buy the most popular computer programs because this makes it easier for each user to share files and know-how, and gives assurance that the product will survive over time. Consumer action based on such considerations are called "network effects." The Government's economist in United States v. Microsoft defines network effects as a case where "each person's benefit from using a product or technology increases with the number of other people who also use that product or technology."

Economists are in agreement that computer software markets are strongly affected by network effects. This is true of end user decisions, and, as explained herein, it is also true of the decisions of application program developers. In fact, the interaction of consumer adoption of one operating system on the development of more and more good application programs has been called as the "software feedback loop."

Microsoft's extreme monopoly share in personal computer operating systems creates extreme network effects. As more users opt for Windows application programs, there is a greater and greater motivation to use those programs. This, in turn, motivates program developers to design additional programs that work with Windows.

Even if a Microsoft competitor offers a superior product on a one-to-one basis, consumers will not switch to the competitive operating system when the second system will not run that favorite application. Likewise, the Microsoft competitor would have a difficult or impossible time persuading the owners of such popular applications to "port" their products to a different operating system. Thus, although MacOS is the second most popular PC operating system, only a few very strong application programs have been "ported" from Windows to Macintosh.

(2) Control Of The Dominant Programming Interface.

Closely related to network effects is Microsoft's control of the programming interface necessary to write a compatible application program. This means that a would-be operating systems competitor, even if it could support translation of popular applications to its own programming interface, could not keep pace with Microsoft product changes until it had timely access to information about intervening changes in the Microsoft programming interface. Since considerable time is required to develop new features and functions in a program, the lead time advantage that Microsoft enjoys as a result of its control of the dominant programming interface creates a further barrier to entry by new competitors.

(3) High Entry Costs And Insignificant Marginal Costs

A modern operating system contains millions of lines of code. Windows NT 5.0, for example, is said to include more than 30 million lines. Development of NT 5.0 apparently entailed the work of over 1500 people, and will require approximately three years' time to develop assuming it is shipped to customers next year, as announced. NT 5.0 is huge: The cost to Microsoft of this behemoth will exceed $270 million.

A would-be competitor entering the field would need to make an even greater investment because Microsoft was able to build on its experience in writing previous Windows versions. It is doubtful that any firm other than Microsoft could attract the capital or the skilled programmers to develop a program of equivalent scope.

In contrast to the enormous cost of entry, the cost of making copies of a program for sale is very low, to the point of being inconsequential. This enables the established operating system supplier like Microsoft to reduce its prices to virtually nothing if faced with competition. The mere perception that this can be done discourages entrepreneurial investment in a competitive product and creates a barrier to new entry.

(4) Public Perception Of Microsoft's Invincibility

Since the introduction of the IBM Personal Computer, attempts to compete against Microsoft in the operating system business has been unsuccessful. Dynamic Research, the originator of personal computer operating systems, was unable to compete with Microsoft in the DOS era. IBM, previously thought to be unbeatable, was unable to make a go of OS/2. Vendors of some of the leading application programs (e.g. Lotus123) tried to compete against Windows by adopting graphic interfaces of their own. Each such effort failed, in most cases with the result that the contestant became an insignificant factor in a field where it had been a leader.

Capital markets, employees and customers are aware of this history, and observe the extent to which Microsoft has become even stronger since those victories. This results in a perception that a competitor, even one with a temporary advantage over Microsoft, will not be able to succeed in the long run. This perception creates barriers to entry affecting access to capital, ability to recruit qualified personnel, and ability to sell product.

2. Microsoft Has Monopoly Power In The Relevant Market For Operating Systems For Technical Workstations.

a. User Alternatives

As explained above, technical work stations are powerful desktop computers capable of exotic applications like computer simulations, or high quality computer animation. Until recently, the only microcomputers powerful enough to run these applications used Reduced Instruction Set or RISC processors. With one, fairly recent, exception, Windows did not run on such processors. As a result, there were no Windows application programs available to run technical jobs. Accordingly, the various "flavors" of UNIX competed with each other in the workstation market.

Currently, however, more powerful machines running Windows or Windows NT (see n.2 above) have become available, and Windows-based application programs suitable for the technical requirements of this field have been introduced. Since this change, UNIX and Windows NT are, for most users of technical work stations, the only practical alternatives. Thus, operating systems for technical work stations constitute a relevant product market for antitrust purposes.

b. Market Share

The growth of NT sales into this market segment has been dramatic. Windows now operates 43% of installed workstations. Windows substantially outsells any single version of UNIX, and its sales continue to grow.

Generally, market share below 50% is not indicative of monopoly power, but this is not an absolute. In some market conditions, monopoly power may be present even where present market share seems small. United States v. Columbia Steel Co., 334 U.S. 495, 528 (1948); Hayden Publishing Co. v. Cox Broadcasting Corp., 730 F.2d 64, 69 n.7 (2d Cir. 1984); Broadway Delivery Corp. v. United Parcel Service, 651 F.2d 122, (2d Cir. 1981). Such conditions are surely present in the workstation operating system market.

First of all, Microsoft's present share of the workstation OS market has grown from nothing overnight, and continues to grow dramatically. Second, no single competitor has as much as half the share that Microsoft has already achieved. Thus, Microsoft's relative power is great. See Pacific Coast Agricultural Export Assn v. Sunkist Growers, Inc., 526 F.2d 1196, 1204 (9th Cir. 1975). More important, Microsoft has overwhelming power in the adjacent market for PC operating systems, and its power in that market gives it significant power when competing with customers for technical workstations. Further, the significant barriers to entry previously discussed apply equally to the server market. United States v. AT&T, 524 F.Supp. 1336, 1447 (D.D.C. 1981). Finally, application program dependencies largely "lock in" existing customers, limiting the ability of other products to compete. Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. at 473. Considering the growth trend and the structure of this peculiar market, Microsoft's share is indicative of monopoly power.

c. Structural Indications Of Market Power

As in the PC operating system market, any firm competing with Microsoft is at a significant structural disadvantage. There are huge barriers to entry. Microsoft's power in the PC market gives it power in the workstation market, and consumer perception of Microsoft's invincibility means that even when a competitor has a superior product, consumers are afraid to purchase for fear that the supplier will not survive in the long run. Because of these factors, Microsoft's present share is indicative of market power in the workstation market.

3. Microsoft Has Monopoly Power In The Relevant Market For Departmental Server Operating Systems.

a. User Alternatives

Servers are the computers used to supply common services to a desktop PCs tied together in a network. Such computers require operating systems, but the operating systems used in common desktop computers (like Windows95) do not have enough function or reliability to meet user requirements for servers used in typical business networks. Conversely, the operating systems used in larger computers, while they will provide adequate server function, are overkill because, in most cases, the investment in the hardware necessary to use those operating systems cannot be justified for server uses alone. Thus, there is a distinct market for operating systems suitable for running small to medium computers when they are used as "servers" in a network of desktop PCs.

b. Market Share

As with workstations, until quite recently, the machines commonly used as departmental servers would not run Windows. UNIX was the dominant operating system. Within the past five years, the availability of high-powered Intel processors has changed this market. Windows NT now operates an estimated 19% of all servers, and an amazing 49% of new shipments of server software. Variations of UNIX, taken together, continue to be used in about 23% of servers, although the UNIX share would be less if figures were available that just those configurations for which NT is a viable alternative. Measured by these statistics, UNIX and Windows NT account for the overwhelming majority of server operating systems, with NT growing and UNIX shrinking in relative share.

As with workstations, although Microsoft's share of the total installed base is below 50% at this time, this does not preclude a finding of monopoly power. Hayden Publishing Co. v. Cox Broadcasting Corp., 730 F.2d at 69 n.7.

c. Structural Indications Of Market Power

All of the structural advantages discussed above -- network effects, control of the programming interface, low marginal cost and market confidence -- reenforce Microsoft's burgeoning power in the departmental server marketplace.

C. Microsoft Has Engaged In Exclusionary Conduct To Maintain And Expand Its Power In The Relevant Markets

1. Microsoft Has Used Its Monopoly Power In The Desktop Market To Gain A Competitive Advantage In The Server And Workstation Markets.

It is an act of monopolization to use "market power attained in one market to gain a competitive advantage in another." Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d at 268; Eastman Kodak Co. v. Image Technical Services, Inc, 504 U.S. 451, 480 n. 29 (1992). There is no reasonable doubt that Microsoft is guilty of monopoly leveraging as defined in Berkey Photo.

As already shown, Microsoft deliberately and strategically used the control of the Windows programming interface to gain the competitive advantage in the workstation and server markets. This control flows directly from Microsoft's monopoly of the PC operating system market. Over and over again, Microsoft's literature for WISE stressed the "availability of more than 10,000 high-quality, inexpensive, shrink-wrapped Windows-based applications" as the motive for making the Windows programming interface the "universal" interface. By so doing, Microsoft was able persuade UNIX programmers to use the Windows programming interface, and thus to minimize the mini-network effects that might have been something of a bulwark against the onslaught of Windows NT into the server and workstation markets.

While this might have been proper if WISE had been operated in the unbiased manner that Microsoft advertised, in fact, WISE was not used this way. Microsoft has withheld access to Windows source code when making that code available would enhance the viability of UNIX. Microsoft has provided source code only where it benefits Windows NT as against UNIX. This is a clear case of a use of monopoly in one market to foreclose competition in another.

Great Western Directories, Inc. v. Southwestern Bell Telephone Co., 63 F.3d 1378 (5th Cir. 1995), opinion modified in part on other issues 74 F.3d. 613 (5th Cir. 1996) is directly in point. In Great Western, independent publishers of yellow page directories sued the monopoly telephone company for failing to provide timely listing information on commercially reasonable terms. The evidence indicated that access to such information was "vital to the publishing industry," and that "without sharing this updated information with competing directory publishers, telephone companies are able to leverage their monopoly position in the telephone service area into the competitive directory market." 63 F.3d at 1386. The court held that by increasing prices and imposing unreasonable restrictions on access to the listing information, the telephone company had committed an act of monopolization. Accordingly, the district court had properly ordered the defendant to give the plaintiff timely access to its listings on commercially reasonable terms.

The remedy Bristol seeks here is precisely analogous to the remedy in Great Western. Bristol seeks access to a body of information controlled by the monopolist that is essential to creation of cross-platform tools. Microsoft has denied Bristol access to all such information for over a year, and now offers only those pieces that will enhance Microsoft's advantages over UNIX. These practices have had and will have the effect of smothering competition in workstation and server markets.

2. Microsoft Has Engaged In Exclusionary Conduct To Acquire And Maintain Monopoly Power In The Workstation And Server Markets.

Separately, Microsoft's conduct violates §2 of the Sherman Act under the doctrine of Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585, 596 (1985). In that case, the Court held that the defendant had breached a duty not to withdraw from cooperative arrangements solely for the sake of destroying competition. 472 U.S. at 605. These were the facts. During the years when there had been competition between operators of the four ski mountains in Aspen, all had considered it beneficial to offer a combination ticket that allowed skiers to use the facilities of the different resorts on successive days. When Aspen Skiing Company acquired monopoly power in the market, however, it discontinued the multi-mountain ticket, and refused to cooperate in any arrangement that would allow the one remaining competitor to sell a ticket for use of Ski Co.'s facilities.

The Court held that "[i]f a firm has been 'attempting to exclude rivals on some basis other than efficiency,' it is fair to characterize its behavior as predatory." Id. The Court considered the impact of the defendant's abandonment of the cooperative multi-mountain ticket from the standpoint of the competitor, consumers and on the defendant itself. It found that the conduct was harmful to consumers and competitors, and beneficial to the defendant only because it would assure an unchallenged monopoly. A similar analysis of the facts of this case leads to the conclusion that Microsoft's treatment of Bristol is unlawful.

a. Product Quality.

Microsoft's own words concede that the WISE world was a better deal for consumers and program developers. Thus, Microsoft summed up the benefits of WISE for consumers as follows:

• Users can buy inexpensive shrink-wrapped software off the shelf and use it on UNIX or Macintosh systems

• Users have a wider choice of software on different platforms; developers have a larger market.

• ...

• Corporate purchasing can set cross-platform standards based on Windows-based applications, thereby reducing support headaches while increasing purchasing power. (Emphasis in the original.)

The benefits of WISE for developers of application programs are similarly significant:

If you are developing applications simultaneously for the Windows operating system family, UNIX and Macintosh platforms, WISE software development kits (WISE SDKs) will help make your job easier by significantly reducing development and maintenance time. WISE SDKs removes the need for developers to learn multiple application program interfaces (APIs). With WISE SDKs, developers can write for a standard, consistent and well-documented set of APIs and leverage their solution for the Windows family, UNIX and Macintosh platforms.

The world of WOE offers none of these benefits. Microsoft is its only beneficiary.

b. Impact on Competitors.

REDACTED

REDACTED

 

c. Efficiency Justification.

Last of all, the Court in Aspen Highlands considered whether the defendant's refusal to cooperate was warranted by any efficiency, and found none.

To Bristol and publicly, Microsoft has asserted that it is free to do whatever it pleases because Windows source code is Microsoft's intellectual property. This argument might have appeal if Microsoft had simply stood on its intellectual property rights and allowed market forces to take their course, but this is not what Microsoft did. WISE was Microsoft's invention -- an entirely voluntary offer to allow the use of its intellectual property. This offer was made, as Microsoft's agents then admitted, to help Microsoft in the "battle of the APIs." As the Court said in Aspen Highlands:

[T]he monopolist did not merely reject a novel offer to participate in a cooperative venture that had been proposed by a competitor. Rather, the monopolist elected to make an important change in a pattern of distribution that had originated in a competitive market and had persisted for several years.

472 U.S. at 603.

Microsoft cannot rely on intellectual property rights to explain an arbitrary change in "a pattern of distribution" it originated for its own advantage. See Data General Corp. v. Grumman Systems Support Corp., 36 F.3d 1147, 1188 (1st Cir. 1994). (distinguishing Aspen on the ground that Ski Co. had discontinued a program in which it had taken part prior to acquisition of monopoly power in the relevant market.) See Image Tech. Services, Inc. v. Eastman Kodak Co., 125 F.3d 1195, 1219-20.

In summary, all of the characteristics that made the defendant's conduct predatory in Aspen Highlands are present here, but the consequences of that conduct is immeasurably more malignant. Far beyond taking some of the luster off an expensive ski vacation, Microsoft's conduct damages fundamental markets affecting the entire economy. Bristol is likely to prove at trial that Microsoft has monopolized by refusing to deal in order to extinguish competition. See also Image Tech Services, Inc. v. Eastman Kodak Co., 125 F.3d at 1209.

3. Microsoft Has Wrongfully Denied Bristol Use Of An Essential Facility.

Every student of antitrust knows about the "essential facilities" doctrine. United States v. Terminal Railroad Assn., 224 U.S. 383 (1912); Delaware & Hudson Ry. Co. v. Consolidated Rail Corp., 902 F.2d 174, 179-80 (2d Cir. 1990); Twin Laboratories, Inc. v. Weider Health & Fitness, 900 F.2d 566, 568-69 (2d Cir. 1990); Hecht v. Pro-Football, Inc., 570 F.2d 982, 992 (D.C. Cir. 1997). The elements of an essential facilities case have been defined as follows:

(1) control of the essential facility by a monopolist;

(2) a competitor's inability practically or reasonably to duplicate the essential facility;

(3) the denial of the use of the facility to a competitor; and

(4) the feasibility of providing the facility.

Delaware & Hudson Ry. Co. v. Consolidated Rail Corp., 902 F.2d at 179

Each of these elements is established in the evidence submitted with this motion.

a. Windows Source Code Is An Essential Facility

Source code is the only complete and reliable specification of the entire programming interface of an operating system. In order to provide a cross-platform product, therefore, timely access to source code for the entire programming interface is essential. And this access must include new and changed interfaces as an operating system evolves. Not surprisingly, therefore, Microsoft told Bristol from the beginning that access to Windows source code would be essential to providing cross platform development tools bridging the Windows programming interface to the UNIX interface. Obviously, Microsoft knew what it was talking about, and had the power to make its predictions come true. Every company that tried to provide cross platform development tools for Windows without access to source code has failed.

It is ironic that the first "essential facility" case also involved a bridge and this case is about a bridge. See United States v. Terminal Railroad Assn, 224 U.S. 383. In Terminal Railroad, the bridge was physical, and geography made it essential. Here, the bridge is intangible, and it is essential because the Windows monopoly means that there is no other practical way to "port" application programs originally written for the Windows programming interface so that they will run on UNIX or other operating systems. Unlike Terminal Railroad, however, this is not a case of simple denial of access to the only available bridge. This is a case where Microsoft encouraged Bristol and its customers to use the bridge, and pulled it up only after they had crossed over to the other side.

Keeping up with the Windows programming interface is like manning a dike in a flood, the job of maintenance and repair is never ending. The real boundaries of the programming interface change continually, and the only way to keep track of these changes is to have timely access to source code. Microsoft's publications do not suffice, because they do not contain complete or accurate descriptions of all the interfaces. Microsoft recognized this when it named the program Windows Interface Source Environment. (Emphasis supplied.)

Reliance on anything other than up-to-date source code would leave Bristol customers exposed to bugs, malfunctions, and failures that would drive users of UNIX versions of the customer's application programs into the Windows fold. In addition, the incorporation of Windows source code into Wind/U has become essential to Bristol's ability to compete against the Microsoft juggernaut. As explained above, the programming interface define the way that application programs can "call" upon the operating system to perform a desired function. An increasing number of the "calls" recognized at the Windows programming interface have no matching function in UNIX. Prior to WISE, Bristol used various techniques to "map" Windows calls to UNIX. After WISE got underway, Bristol, with Microsoft's encouragement, simply modified Microsoft Windows source code to implement calls unknown to UNIX. Bristol, of course, pays Microsoft a fee for use of this source code.

Now that Bristol has staked its business on Microsoft's good faith, it is commercially impossible for Bristol to double back to the pre-WISE status quo. It is equally impossible, in real terms, for Bristol to survive as a competitor if it is frozen in place. Bristol's customers did not adopt Wind/U as a one-time expedient, they made an investment in a strategic product in one of the fastest moving markets in history. They accepted Wind/U with Microsoft's assurance that "[c]ustomers can be confident that their investment today will continue to realize benefits well into the future."

Through the WISE program, Microsoft aggressively positioned the Windows source code as an essential facility of competition for companies whose products use the Windows programming interfaces. United States v. Terminal Railroad Assn, 224 U.S. 383 (1912); Ottertail Power Co. v. United States, 410 U.S. 366 (1973). Microsoft has an obligation to make access to this facility available on reasonable terms.

b. It Is impossible for Bristol To Duplicate The Microsoft Windows Source Code.

The impossibility of duplicating Microsoft Windows source code is self-evident. First of all, Microsoft has exclusive control of the programming interface. Second, the content of the programming interface, is, in a real sense, arbitrary. Any number of alternative expressions could be used, but once a particular format is adopted, that exact format must be followed or "calls" in the application program will not function. Bristol has to know and support each of these formats or Wind/U-enabled application programs will not run on UNIX.

Third, untimely access to the essential facility is equivalent to no access. If Bristol were required to write its own call-supporting code for each bell or whistle Microsoft adds to the Windows programming interface, not only would the cost be prohibitive, but Bristol's product would be obsolete by the time it reached the market.

The infeasibility of duplication in this case is the inevitable and intended consequence of Microsoft's deliberate intervention in Bristol's independently-established business. Bristol had invested in development of its own facility at a time when market condition made that development feasible. It is speculative how Bristol's product would have evolved absent WISE, but Microsoft convinced Bristol to displace the only viable alternative means of serving its customers. Under those circumstances, Bristol cannot put Humpty Dumpty back together again. In this respect the present case is somewhat like Intergraph Corp v. Intel Corp., 3 F.Supp.2d 1255 (N.D. Ala. 1998).

In Intergraph, Intel persuaded Intergraph, a customer, to abandon the development and manufacture of its own microprocessor, and purchase Intel Pentiums instead. Later, Intel got into a dispute with Intergraph over patent rights, and refused to sell Intergraph or to supply information about future products. Although Intel's conduct had far less market significance than Microsoft's behavior here, the court held that Intel had abused its monopoly power, saying:

[B]efore it was induced by the representations of Intel CEO Andy Grove to abandon its own Clipper technology, Intergraph competed successfully in the relevant market for high-end microprocessors. Because of the conduct of Intel, inducing the plaintiff to abandon the Clipper, Intergraph has been eliminated as a competitor in the high-end microprocessor market and is now locked-in to Intel as its sole source of CPUs and relevant technical data.

3 F.Supp.2d at 1265.

c. Microsoft Has Denied Access To The Essential Facility.

Microsoft has refused to license Bristol to use Windows source code in the manner promised with the WISE charter. It will provide only those parts of Windows that suit Microsoft's purposes, and then only on price terms that are economically infeasible. In Delaware & Hudson Ry. Co. v. Consolidated Rail Corp., supra., 902 F.2d at 179-80, the court held that an offer to provide an essential facility at a prohibitive cost amounts to a refusal to deal at all.

d. It Is Feasible For Microsoft To Give Bristol Access To The Essential Facility.

The best evidence that Microsoft can feasibly and profitably provide Bristol access to current and updated WISE materials is the fact that it willingly obligated itself to do so in the 1994 WISE Agreement, and thereafter did so. Since 1994, Bristol's performance in developing cross-platform solutions for Windows application programs has won Microsoft's praise. Microsoft's abandonment of WISE is not caused by infeasibility, but by Microsoft's realization that WISE was all too feasible. Like Aspen Skiing Company or Southwestern Bell, Microsoft will not tolerate an arrangement that is mutually beneficial. It wants only those cross-platform tools as will augment Microsoft's market power.

As can be seen, all four elements of an essential facilities claim are established on the present record. Moreover, that Bristol's exclusion from WISE adversely affects not simply Bristol, but competition in general, is demonstrated in the preceding section, and thus, the requisite harm to competition is unquestionably present here. Twin Laboratories, Inc. v. Weider Health & Fitness, 900 F.2d at 568.

V. MICROSOFT'S REPUDIATION OF THE WISE COMMITMENTS IS AN ATTEMPT TO MONOPOLIZE.

Proof of attempted monopolization under § 2 of the Sherman Act requires a showing of (1) anticompetitive or exclusionary conduct, (2) specific intent to monopolize, and (3) dangerous probability that the attempt will succeed. Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d at 99-100; Volvo North America Corp. v. Men's International Professional Tennis Council, 857 F.2d at 73-74; National Ass'n of Pharmaceutical Manufacturers, Inc. v. Ayerst Laboratories, supra., 850 F.2d at 915; International Distribution Centers, Inc. v. Walsh Trucking Co., Inc., 812 F.2d 786, 790 (2d Cir. 1987). All three elements necessary to establish Microsoft's refusal to license WISE technology as an actionable attempt to monopolize are present here.

A. Anti-competitive Conduct

As demonstrated at length above, Microsoft's refusal to license is both anticompetitive and exclusionary. Bristol, application programs developers, end users, and manufacturers of competing operating systems are all adversely affected and limited in their ability to compete.

B. Specific Intent To Monopolize

The requisite specific intent may, and in this case should, be inferred from Microsoft's conduct. Volvo North America Corp. v. Men's International Professional Tennis Council, 857 F.2d at 74; National Ass'n of Pharmaceutical Manufacturers, Inc. v. Ayerst Laboratories, 850 F.2d at 915. This inference does not depend on conduct alone. Microsoft's blatant effort to control the end use of Wind/U and explicit statements that "Bristol should have known all along that Microsoft would begin to restrict [Bristol's] access to new Windows technology as soon as Windows NT was better established" is direct proof of intent to monopolize.

C. Dangerous Probability Of Success.

While Microsoft's shares in the markets for advanced workstation operating systems and servers is sufficient, along with the applicable structural factors, to support a finding of monopoly power, those shares and their growth trends are clearly enough to show that there is a dangerous probability that Microsoft will extend its monopoly into the server and workstation markets. Multistate Legal Studies, Inc. v. Harcourt Brace Jovanovich, 63 F.3d 1540, 1554 (10th Cir. 1995); Indiana Grocer Co. v. Super Valu Stores, Inc., 647 F.Supp. 254, 258 (S.D. Ind. 1986).

Thus, Bristol is likely to prevail at trial on its claim that Microsoft's conduct, if not monopolization, is an attempt to monopolize the server and workstation markets. On that independent basis, Microsoft's wrongful conduct should be enjoined pending trial.

VI. MICROSOFT'S CONDUCT HAS CAUSED ANTITRUST INJURY TO BRISTOL

Antitrust standing boils down to a question of whether the plaintiff has suffered antitrust injury and is the logical person to enforce the antitrust laws in the particular case. Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104, 111 n.5 (1986). Clearly, Bristol has suffered and is threatened with exactly the type of injury that the antitrust laws were designed to prevent, and is, therefore, the logical person to sue.

Bristol is a direct competitor of Microsoft. Although Bristol does not make and sell an operating system, the software that Bristol sells enables users to avoid switching from UNIX to Windows. Bristol thereby enhances UNIX and makes it more competitive. In this respect, Bristol's market posture is analogous to the plaintiff in Crimpers Promotions, Inc. v. Home Box Office, Inc., 724 F.2d 290 (2d Cir. 1983). There, organizers of a trade show designed to enable movie producers to deal directly with cable TV operators had standing to sue middlemen who arranged a boycott of the show. The promoters were competitors of the middlemen because if their shows were successful, the middlemen would lose business. In just the same way, Wind/U makes it possible for users to stick with UNIX when otherwise they would have to switch to Windows or NT. When Bristol makes a sale, it forestalls a sale of Windows.

VII. BRISTOL IS ENTITLED TO AN INJUNCTION UNDER THE CONNECTICUT ANTITRUST LAW

Microsoft's conduct violates the Connecticut Antitrust Law to the same extent that it violates federal. Conn. Gen. Stats. §35-27; Shea v. First Federal Sav. & Loan Assn., 184 Conn. 285, 303; 439 A.2d 997,1006 (1981).

Bristol is Connecticut-based and a Connecticut employer. The effects of Microsoft's conduct will have their most immediate impact in Connecticut. Connecticut has a clear and fundamental interest in having its antitrust laws vigorously enforced. Shea, 184 Conn. 285, 439 A.2d 997, 1002 (1981). Particularly as regards development of high tech industries in Connecticut, Microsoft's conduct threatens as much or more damage to Connecticut as to the national economy. Accordingly, Bristol is entitled to an injunction under the monopolization prohibitions of the Connecticut Act. Conn. Gen. Stats §35-34.

VIII. BRISTOL IS ENTITLED TO AN INJUNCTION UNDER THE CONNECTICUT UNFAIR TRADE PRACTICES ACT

The Connecticut Unfair Trade Practices Act ("CUTPA"), Conn. Gen. Stat. § 42-110(a-q), states a public policy in favor of "remedying wrongs that may not be actionable under other bodies of law." Omega Engineering, Inc. v. Eastman Kodak Co., 908 F. Supp. 1084. 1099 (D. Conn. 1995). The Connecticut Supreme Court recently reiterated the test for a violation of CUTPA. Under a CUTPA claim, a court must consider:

(1) [W]hether the practice, without necessarily having been previously considered unlawful, offends public policy as it has been established by statutes, common law or otherwise -- whether, in other words, it is within at least the penumbra of some common law, statutory, or other established concept of unfairness; (2) whether it is immoral, unethical, oppressive or unscrupulous; (3) whether it causes substantial injury to consumers [competitors or other businessmen]. . . All three criteria do not need to be satisfied to support a finding of fairness. A practice may be unfair because of the degree to which it meets one of the criteria or because to a lesser extent it meets all three."

Fink v. Golenbock, 238 Conn. 183, 215, 680 A.2d 1243, 1260-61 (1996) quoting Cheshire Mortgage Svc., Inc. v. Montes, 223 Conn. 80, 105-06, 612 A.2d 1130 (1992); see also Boulevard Associates v. Sovereign Hotels, Inc., 72 F.3d 1029, 1038 (2d Cir. 1995).

CUTPA specifically authorizes courts to grant injunctive relief. See Conn. Gen. Stat. § 42-110g.

Microsoft, as demonstrated above, has engaged in a continuous pattern of conduct that is unfair under all three of the CUTPA tests. First, as shown at length above, Microsoft's conduct conflicts with long-established principles of state and federal antitrust law discouraging the maintenance and expansion of monopoly power. Further, as shown below, the cutting edge of Microsoft's wrong doing violates a fundamental principle of fairness expressed in the doctrine of estoppel. This is not a matter of penumbras. This is a simple question of dealing honestly with the public. Second, there can be little question that Microsoft's behavior is both unethical and oppressive. "Bait and switch" is a byword for unethical market practices, yet that is just what WISE has become. And to deliberately destroy a fledgling business and leave its customers in the lurch is plainly oppressive. Thus, it follows that Microsoft's conduct is damaging to those businesses who have relied on Wind/U in light of Microsoft's representations, damaging to Bristol and its employees and damaging to public users who purchased Wind/U enabled programs. In short, the evidence makes out a clear case of an unfair practice with the meaning of CUTPA.

IX. BRISTOL IS LIKELY TO PREVAIL ON ITS CLAIM THAT MICROSOFT IS ESTOPPED TO WITHHOLD WINDOWS SOURCE CODE FROM BRISTOL

In addition to its antitrust claims (which are of greatest public interest), Bristol asserts that Microsoft is estopped to withhold source code from Bristol on the theory of promissory estoppel. This theory relies on the same facts, but focuses on Bristol's reasonable reliance on Microsoft's public and private assurances as distinguished from the market and strategic effects of Microsoft's behavior. In plain language, even apart from monopoly, the law requires businessmen to keep their word. The legal term to describe this common-sense requirement is promissory estoppel. Restatement (Second), Contracts, § 90.

There are two elements to a claim for promissory estoppel:

"Under our well-established law, any claim of estoppel is predicated on proof of two essential elements: the party against whom estoppel is claimed must do or say something calculated or intended to induce another party to believe that certain facts exist and to act on that belief; and the other party must change its position in reliance on those facts, thereby incurring some injury."

Connecticut National Bank v. Voog, 233 Conn. 352, 366, 659 A.2d 172 (1995); accord, Wellington Systems, Inc. v. Redding Group, Inc., 49 Conn. App. 152, 162 (1998); Redgate v. Fairfield University, 862 F.Supp. 724, 730-31 (D.Conn. 1994).

As demonstrated above, Microsoft made public and private promises, and gave assurances that it would make Windows source code available to "remove the need for developers to learn multiple application program interfaces." In particular, Microsoft "committed to providing WISE licensees with future versions of Windows family source code, thereby continuing to maximize application compatibility and performance for today's and tomorrow's applications." (Emphasis supplied.) In response to inquiry from a Bristol customer, Microsoft volunteered that "Microsoft is not in a position to arbitrarily terminate the source code support to Bristol and Mainsoft." Thus, wholly apart from the WISE contract, as such, Microsoft did and said things calculated to induce action in reliance by Bristol and its customers.

It has also been demonstrated that Bristol relied on Microsoft's promised support. Bristol based the entire future of Wind/U on access to and licensed use of Windows source code. Nor can it be denied that Bristol's reliance was reasonable. Microsoft still assures users that "customers can be confident that their investments [in WISE SDKs] will continue to realize benefits well into the future." See D'Ulisse-Cupo v. Board of Directors of Notre Dame High School, 202 Conn. 206, 213, 520 A.2d 217, 218 (1987); Kimberly-Clarke Corp. v. Dubno, 204 Conn. 137, 148, 527 A.2d 679 (1987).

Finally, there is no doubt that Bristol has been damaged by Microsoft's doublecross. Its good will with customers is in peril of destruction. It has fallen behind in the necessary task of keeping Wind/U up-to-date. Its revenues and profits have been reduced and there is a serious risk that it may lose gifted employees.

These facts establish a classic promissory estoppel. Courts have regularly recognized the availability of injunctive relief in actions for promissory estoppel. In Concourse Ticket Agency v. Kraft, 3 Mass. L. Rptr. 446 (Mass. Sup. Ct. 1995); Robbins v. Kristofic, 434 Pa. Super. 392, 643 A.2d 1079, app. denied, 539 Pa. 654, 651 A.2d 541 (1994); Levitt Homes, Inc. v. Old Farm Homeowner's Ass'n, 111 Ill.App.3d 300, 315, 444 N.E.2d 194, 204 (1982); Wizard Programming, Inc. v. Superstar/Netlink Group LLC, ___ F.Supp. ___ , 1997 WL 644082 (D.Conn. Oct. 10, 1997), at WL p. 7.

X. MICROSOFT'S WRONGFUL CONDUCT IS CAUSING IRREPARABLE INJURY TO BRISTOL

REDACTED

XI. THE BALANCE OF HARDSHIPS TIPS OVERWHELMINGLY IN BRISTOL'S FAVOR.

The balance of hardships is absurdly one sided. While Bristol is faced with extinction, a preliminary injunction would cause Microsoft no injury whatever. Microsoft has had dozens of Windows source code licensees, many of whom pay nothing at all for their use of that code. Bristol does not seek a royalty-free license. The injunction sought would require Bristol to pay the rates fixed in the original WISE agreement, despite the fact that those rates had the effect of limiting Bristol's business opportunities. Moreover, if at trial, it were shown that the rates paid were insufficient, Microsoft could be compensated in money for the shortfall. In either case, the injunction would represent profitable revenue for Microsoft, and the only "loss" would be the inability to use monopoly power to stifle competition. Under these circumstances, a preliminary injunction ought to be granted.

XII. THERE IS NO NEED TO POST A BOND TO SECURE A PRELIMINARY INJUNCTION.

In this Circuit, the district court has "wide discretion to set the amount of a bond, and even to dispense with the bond requirement." Doctor's Associates, Inc. v. Distajo, 107 F.3d 126, 136 (2d Cir. 1997); Doctor's Associates, Inc. v. Stuart, 85 F.3d 975, 985 (2d Cir. 1966); Ferguson v. Tabah, 288

F.2d 665, 675 (2d Cir. 1961). The Court should exercise its discretion here to dispense with any bonding requirement for three fundamental reasons.

First, Bristol's prosecution of this action vindicates a strong public policy, competition in one of the most vital sectors of the American economy; indeed, Bristol proceeds here as a private attorney general. 15 U.S.C. Sec. 15; Conn. General Stats. 35-27, Twin Laboratories, Inc. v. Weider Health & Fitness, 900 F.2d 566, 568 (2d Cir. 1990). Where the plaintiff vindicates an important public interest, it is appropriate to waive the bond requirement. Pharmaceutical Society of the State of New York, Inc. v. New York Department of Social Services, 50 F.3d 1168, 1174-75 (2d Cir. 1995 ); Temple University v. White, 941 F.2d 201, 219 & n.26 (3d Cir. 1991).

In Pharmaceutical Society, a for-profit group of pharmacists had brought suit to ensure Pennsylvania's compliance with the federal Medicaid reimbursement statutes. The Third Circuit affirmed the trial court's grant of an injunction without the posting of a bond because the suit vindicated national healthcare policy:

Here, the Society sued to ensure that pharmacies providing Medicaid services would receive full compensation for those activities. To the extent that pharmacies do not receive full compensation, some are likely to drop out of the Medicaid program and reduce the number of providers of health care to poor individuals. And, if they continue as providers, they will simply pass on the costs of the uncollected co-payments to the public at large. Thus, the Society's litigation was indeed pursued to enforce 'public interests' rising out of a comprehensive health and welfare statute, and the district court's waiver of the bond requirement was proper.

50 F.3d at 1175 (emphasis supplied.)

Second, where the enjoined party is unlikely to sustain any harm in the absence of a bond, none is required. Doctor's Associates, Inc. v. Distajo, 107 F.3d at 136; Doctor's Associates, Inc. v. Stuart, 85 F.3d at 975. Not only will Microsoft not be pecuniarily harmed by continuing to license Bristol Windows source code, it will actually earn reasonable licensing fees.

Third, where the bond requirement would create severe hardship, the Court may dispense with that requirement. Crowley v. Local No. 82 Furniture & Piano, 679 F.2d 978, 1000 (1st Cir. 1982), rev'd on other grounds, 467 U.S. 526 (1984). While hardship may be less of a factor in a commercial context, here it is both present and acute. As a result of Bristol's dependence on Microsoft's Windows source code and Microsoft's exploitation of that fact through withholding code updates and the code for new operating systems, Bristol's earnings have dwindled. Bristol could not bear the financial burden of posting a bond in any significant amount. Particularly where the public interest is being vindicated, the courts have refused to permit a larger adversary to deny his opponent access to the courts and relief because of financial weakness. See Pharmaceutical Society of the State of New York, Inc. v. New York State Department of Social Services, 50 F.3d at 1175.

XIII. CONCLUSION

For all the foregoing reasons, Microsoft should be ordered, pending trial, to make any and all source code for any Windows operating system, including but not limited to, Windows NT 4.0 and 5.0 available to Bristol for the purposes contemplated in the WISE program and the 1994 Bristol-Microsoft WISE agreement.

DATED: August 18, 1998

By:______________________________

JOHN L. ALTIERI, JR. (ct 06183)
O'Melveny & Myers LLP
c/o Bristol Technology, Inc.
39 Old Ridgebury Road
Danbury, CT 06810-5113
(203) 798-1007

Patrick Lynch, Esq.
James V. Selna
Carl R. Schenker, Jr., Esq
O'Melveny & Myers LLP
400 South Hope Street
Los Angeles, CA 90071
(213) 430-6000

Anthony L. Clapes, Esq.
Technology Law Network
3077 Ka'ohinani Drive
Honolulu, HI 96807
(808) 595-0128

ATTORNEYS FOR PLAINTIFF,
BRISTOL TECHNOLOGY, INC.