Reverse Engineering and the Rise of Electronic Vigilantism: Intellectual Property Implications of "Lock-Out" Programs

* J.D., Harvard Law School, 1991. Visiting Assistant Professor of Law, University of Pittsburgh School of Law. I would like to thank Anthony Clapes, Robert Clinton, Susan Freiwald, Sondra Hemeryck, Mark Lemley, Eileen Mullen, Pamela Samuelson, Eugene Volokh, and Lloyd Weinreb for their insightful comments and suggestions, Donald Cohen and Brian Wilson for discussion of technical and scientific issues, and Page Barnes for manual labor. I served as law clerk to Judge Reinhardt, who wrote the Ninth Circuit's opinion in Sega, and assisted him with the opinion's preparation. However, the views expressed here are my own.

[p*1092]

Nearly twenty years ago, Congress officially extended copyright protection to computer programs. ⁿ¹ Five years later, the Supreme [p*1093] Court issued a decision that definitively established computer programs' eligibility for patent protection. ⁿ² The two developments had very different trajectories; the debate over patent protection was long, hard-fought, and occasionally acrimonious, ⁿ³ while the extension of copyright protection was accomplished by committee and consensus, almost as an afterthought. ⁿ⁴ The developments were similar in one respect, however. Both Congress and the Supreme Court treated computer programs as autonomous intellectual products, intended for use on a stand-alone basis in the same manner as a copyrighted book or a patented industrial apparatus. ⁿ⁵ Today, in contrast, it is evident that the value of a computer program to its users depends heavily on its compatibility, or interoperability, with a particular computer system and with other programs. ⁿ⁶ Whether interoperability-related issues should affect copyright and patent treatment of computer programs, and if so, how, are among the decade's most hotly debated legal questions. ⁿ⁷

For creators of computer programs, achieving interoperability with particular computers and operating systems is necessary for commercial survival. ⁿ⁸ Interoperability has also become a watchword for consumers who seek applications programs that will operate on their [p*1094] existing computer systems, or who may base selection of new systems on the applications programs available. Manufacturers of computer systems and operating systems have responded in a variety of different ways to program developers' demands for access to interoperability-related information. Some have made program interface specifications and protocols freely available to applications developers. ⁿ⁹ Others have licensed the rights to create compatible programs to third parties, although some withhold complete technical information on interoperability requirements from their licensees. ⁿ¹⁰ Still others, chiefly manufacturers of specialized computers designed to serve industry-specific customer bases, have attempted to keep their systems completely proprietary. ⁿ¹¹ As a result of the frequent unavailability of interoperability-related information through ordinary market channels, "reverse engineering" of interface specifications for proprietary and quasi-proprietary systems has become common. In particular, many third-party software developers have come to rely on a method of reverse engineering known as "disassembly" or "decompilation," which parses the binary object code in which computer programs are distributed into higher-level, human-readable commands. ⁿ¹²

The rise of reverse engineering by third-party software developers in turn has led some computer manufacturers to seek technological protection against unwanted competitors. ⁿ¹³ Within the video game industry, several system manufacturers have developed specialized "lock-out" programs that limit access to their hardware to program disks or cartridges that contain the "key." ⁿ¹⁴ Lock-out programs are [p*1095] designed to exclude all "unauthorized" programs, and to make reverse engineering more difficult. ⁿ¹⁵ However, lock-out programs, like other computer programs, also can be reverse engineered. Lock-out programs therefore complicate, but do not defeat, third-party research and development efforts. Ultimately, neither technological nor market solutions have enabled computer manufacturers to prevent determined competitors from creating amd marketing compatible programs. As a result, computer manufacturers and software developers have sought recourse under the copyright and patent laws. They have argued that both the reverse engineering process and the subsequent creation of compatible programs that include "keys" to their systems infringe their intellectual property rights.

Reverse engineering of interface specifications and use of the information gained through reverse engineering to create a compatible program raise novel questions in the overlapping realms of copyright law, patent law, and public policy. Over the past few years, there has been an abundance of scholarship dealing with the appropriate scope of copyright and patent protection for computer programs. ⁿ¹⁶ This Article approaches those problems from a slightly different perspective, focusing on the discrete problem of lock-out programs. The choice of lock-out as a paradigm for exploring the interoperability question and the contours of copyright and patent protection of computer programs [p*1096] is informed by two considerations. First, for purposes of the interoperability inquiry, lock-out programs represent an extreme; they are discrete, self-contained modules that are highly innovative in design, yet that serve no purpose other than to regulate access to a computer or computer operating system. Copyright and patent analyses of the lock-out problem highlight a fundamental tension between intellectual property rights and considerations of public access, and so afford a useful vehicle for examining the scope of copyright and patent protection for computer programs generally. Second, lock-out may well become a defining technology of the coming "Information Age." Pundits have prophesied a "set-top box" in every home that affords a gateway to an "information superhighway" where goods and services may be purchased and information accessed. ⁿ¹⁷ Whether or not the manufacturer of the set-top box will be able to exclude unauthorized purveyors of goods, services, and information will significantly affect both the structure of the emerging market in information services and the nature of individual participation in that market. ⁿ¹⁸

The purpose of this Article is twofold. First, I argue that neither the copyright laws nor the patent laws preclude duplication of protected program features, including "lock" and "key" features, to whatever extent necessary to achieve full compatibility with an unpatented computer system. Second, and more generally, I address inconsistencies and conceptual flaws in the current understanding of copyright and patent protection for computer programs that emerge during the first inquiry, and propose doctrinal modifications to resolve them. Although computer programs have been protected by both copyright and patent regimes for years, the precise contours of the protection these regimes afford remain unsettled. For that reason, some scholars, computer lawyers, and computer industry professionals have urged the adoption of sui generis protection for computer programs, ⁿ¹⁹ but the question of sui generis protection may have become [p*1097] largely irrelevant. The United States has convinced many other countries to follow its lead in extending both copyright and patent protection to computer programs and is unlikely to change course. ⁿ²⁰ For better or worse, it seems we are stuck with the existing modes of intellectual property protection for computer programs. However, this Article argues that certain adjustments to the copyright and patent doctrines governing the protection of computer programs are necessary if the intellectual property laws are to continue to serve both their new and their traditional functions.

Part I of this Article describes the facts and outcomes of two recent cases: Sega Enterprises Ltd. v. Accolade, Inc. ⁿ²¹ and Atari Games Corp. v. Nintendo of America, Inc., ⁿ²² both of which involved attempts to enforce intellectual property rights in lock-out programs. The remainder of the Article takes those cases as a starting point for discussion of the interoperability question and what it reveals about the scope and structure of copyright and patent protection for computer programs. Parts II and III explore the copyright implications of reverse engineering interface specifications and lock-out programs and of using the information gained thereby to create and market a compatible program. Part II focuses on the copyright issues resulting from intermediate copying during the reverse engineering process. Part III considers whether the reverse engineer may create a program that duplicates the "key" to the "lock" and other functional features of interoperability-related routines. Part IV addresses issues bearing on the validity of a lock-out patent. Finally, Part V considers whether, in light of the analyses in Parts II, III, and IV, attempts to enforce patents and copyrights against competitors who crack the code for a lock-out program constitute patent or copyright misuse. The Article concludes with some general reflections on the efficacy and viability of the copyright and patent models for intellectual property protection of computer programs.

[p*1098]