Vol. 7 No. 5 (May 1997) pp. 206-208.


Reviewed by Robert G. Brookshire, Department of Information and Decision Sciences, James Madison University.

INTERFACES ON TRIAL is a review of the issues, cases and legal history surrounding two aspects of computer software law: the protection of software interfaces under copyright and the permissibility of reverse engineering. These two issues are important to the international computer industry because they affect the ability of firms to develop products which interoperate with those of other companies.

As used in this book, the term "interoperability" has two dimensions: interchangeability, the capacity of one software product to substitute for another, and connectability, the capacity of one product to work with another. Microsoft Word and WordPerfect are both word processing programs, performing the same functions in the same environment, and are thus interchangeable. Both Word and WordPerfect are connectable to the Windows operating system but cannot substitute for it.

In order to develop a new word processing program for the Microsoft Windows environment, a programmer would need detailed knowledge of the interfaces between Windows and the computer hardware, and between Windows and other software. Microsoft might claim that this interface information was protected under copyright, and refuse to provide it, or provide it only for a fee. If the programmer was unable to get sufficient detail about the interfaces from Microsoft, she might try to discover the interface specifications through reverse engineering techniques. This could involve merely a detailed study of the way that Windows interacts with existing hardware and software. It might extend, however, to disassembly or decompilation, where the programmer converts the software from the executable version to a readable version using a special program. Microsoft might then claim that this activity violated its copyright. The programmer could counter that Microsoft is abusing the copyright law by attempting to prevent legitimate development of competing products.

Band, an intellectual property lawyer, and Katoh, a computer company executive with legal training, are not disinterested observers of the software industry. They have participated in some of the cases they describe, and are advocates of the position that interfaces are not protected by copyright and that reverse engineering, including disassembly, should be permitted. This bias informs their commentary on the legal proceedings they write about, giving the book a critical edge which enlivens what might otherwise be a pretty dry description of a fairly restricted area of copyright law.

Band and Katoh identify three positions in the interoperability debate. Those they call "ultraprotectionists" believe that software interfaces are protected and that disassembly of a computer program is a violation of the copyright. These include many of the largest computer firms like International Business Machines, Digital Equipment Corporation, and Apple Computer. At the other extreme, they locate a group of programmers and academics who believe that software should not be protected by copyright at all. This group's position is represented by the Free Software Foundation. They are thus able to position themselves between the two extremes, advocating copyright protection for some software elements but not for others. They are joined in this position by such firms as AT&T, Sun Microsystems, Groupe Bull, Oracle, and many intellectual property law professors. Band and Katoh have created something of a false trichotomy, however, as the Free Software Foundation and its allies' position are given fairly short shrift. The book instead concentrates on the debate between the ultraprotectionists and the authors' group, labeled the "interoperable developers."

After an introduction which lays out the scope of the book and frankly admits the authors' biases, Band and Katoh give a brief overview of computer hardware and software architecture, and a more detailed discussion of software interfaces and reverse engineering techniques. This presentation is aimed at the computer literate layman, and succinctly provides all the technical background necessary for understanding the issues. They also present a short history of the computer industry, concentrating on the rise of IBM and Microsoft and the attempts of firms to develop products which interoperate and compete with those of these companies. The economic and technical interests of the parties which form the foundation of the copyright debate are clearly delineated.

In their second chapter, Band and Katoh review the fundamentals of copyright and patent law as they apply to computer software. Here they give enough background so that readers without training in intellectual property law can understand the legal issues. They review the Copyright Act of 1976, the report of the National Commission on New Technological Uses of Copyrighted Works, and the Software Protection Act of 1980. These two introductory chapters set the stage for the heart of the book.

In their third chapter, Band and Katoh cover the debate over the protectability of interface specifications. They begin with Apple v. Franklin, which prohibited Franklin Computer Company from copying the Apple II operating system in its entirety in order to achieve compatibility with Apple's computers, and Whelan v. Jaslow, which established that the structure, sequence and organization of a computer program could be protected by copyright. In these two cases the courts established rules which resulted in software receiving more protection than many other copyrighted works such as novels. These first two rounds, then, favored the ultraprotectionist side. Later cases weakened this position, however, preparing the way for the decision which currently governs this area, Computer Associates v. Altai.

Computer Associates rejected "thick" protection of software, and set a standard procedure for identifying which elements of a computer program are protectable. In particular, elements necessary for compatibility with other programs and with computer hardware were judged not to be protectable, giving a boost to the interoperable developers' position. Band and Katoh describe this case in detail, reviewing briefs by the parties and amici as well as the decision. They then discuss subsequent cases, both in the U.S. and Canada, which used the procedure established in Computer Associates, including the background of Lotus v. Borland, which was pending as this book went to press.

The authors then turn to the debate over reverse engineering of software. This debate focused on the disassembly technique, specifically over whether disassembly was "fair use" of the original program, or if the readable version of the program created by disassembly was an unauthorized derivative work. Attempts to resolve this controversy through legislation were complicated by the involvement of the literary community, which had its own concerns about fair use unrelated to computer programs. Two cases involving video games, Sega v. Accolade and Atari v. Nintendo, established that disassembly was fair use if it was done for legitimate reasons and no other method of determining interface specifications was available.

In Europe, the interoperability debate was settled decisively through the passage by the European Union of a Directive on the Legal Protection of Computer Programs in 1991. This Directive states that the ideas underlying any element of a computer program, including its interfaces, are exempted from copyright protection. It also specifically permits disassembly of computer programs in order to achieve interoperability. Band and Katoh describe the legislative history and content of the Directive, and carefully analyze its provisions. They review how the Directive has been implemented in member states, and compare it to the law in the U.S., other European countries, and Australia.

Band and Katoh next turn to interoperability in Japan. In a 1985 amendment to its copyright law, Japan established a separate category of copyrightable works, "program works," for computer software. Copyright protection is explicitly not granted, however, to programming languages, rules or algorithms used in creating these works, so interface specifications are not protected. When the Japanese attempted to examine the issue of disassembly through the creation of a special committee, the U.S. government weighed in, effectively squelching the committee's work. Band and Katoh provide an interesting account of this incident, sharply criticizing the U.S. government's actions.

In the book's final chapter, the authors briefly review other issues affecting software copyright, including user interfaces, software patents, and the information infrastructure initiatives in the United States and Europe. These are factors that are likely to have important effects on the future development of interoperable software.

INTERFACES ON TRIAL is very clearly written. Band and Katoh are especially adept at making complex legal and technical issues understandable. Some parts of the book are a little repetitious, however, as the authors review in later chapters topics covered in detail in previous sections. Their aim may have been to have each part of the book stand on its own, but this means that the reader working her way through the entire book will find the same material covered more than once. All in all, though, INTERFACES ON TRIAL is interesting reading for anyone developing or considering the development of computer software, or for lawyers or others who want a better understanding of this aspect of intellectual property law.

Copyright 1997