For two days in mid-September of 1990, a steering committee operating under the auspices of the National Research Council (NRC) held a workshop on software intellectual property issues attended by some three dozen experts from government, academe, and industry. (An “expert” in this still-novel intersection between technology and law is simply someone who has an opinion and voices it often enough.) A little more than two months later, the committee held a larger forum (over 100 experts) on the same topic. Both events were chaired by Dr. Lewis Branscomb of Harvard. This book is a distillation of those proceedings, as well as of other materials submitted to the committee.

Early on, the committee decided that its goal was to facilitate discussion and to try to forge a common understanding of the issues, not to come out with a position on the issues. That wise decision prevented a potentially divisive politicization of the proceedings and allowed for a reasonably free exchange of ideas. It is that exchange of ideas that this book presents, organized into an orderly structure: the basic technological and legal context for software, the nature of relevant intellectual property laws, special attributes of software as intellectual property, current legal issues, and open agenda items. Since the authors, principally C. K. Gunsalus and Mark Bello, both acting as consultants to the Computer Science and Telecommunications Board (CSTB), have tried to give a fair rendition of the spectrum of opinion voiced during the workshop and forum, the book suffers slightly from an “on-the-one-hand-but…then-on-the-other” leitmotif. For the same reason, however, it provides a reasonably even-handed perspective on an important debate for software professionals.

That debate, as anyone in the programming community who is not living under a rock knows, is over whether the current intellectual property laws provide too much, too little, the wrong kind of, or both the right amount and the right kind of legal protection for computer programs. The authors cover such topics as whether one competitor should or should not be able to reverse assemble into pseudo-source code another competitor’s program available to the world in binary form only; whether one competitor should or should not be able to copy the user interfaces (or other interfaces) of another competitor’s program; whether the patent system is granting patents that are too broad or otherwise defective, and if so whether the trouble is with the patent law or simply with current procedures; whether the law is too uncertain to give guidance to software developers today; and whether any other legal framework could be less uncertain. In fewer than 100 pages of text, the authors set out a variety of views on these issues and more, as expressed by computer scientists, industry executives, venture capitalists, economists, lawyers, and others.

A reviewer is entitled to one or two quibbles, even with an exposition as balanced as the NRC’s, and I have two. First, the descriptions of the current state of the law tend to be slightly to substantially off the mark. For example, the core copyright principle that truly independent creation is not infringement even if the resulting works are identical is garbled into unrecognizability. The description of the issue in what the authors correctly call “the most influential decision handed down thus far on the scope of copyright protection for software,” Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., is badly distorted, apparently in an attempt to be neutral: the case, in which the defendant was found to have stolen a copy of the plaintiff’s source code and to have translated the source code into a language that compiled on a different computer, is described as simply posing the question “whether a business-managment system developed for a dental laboratory and written for an IBM PC computer was a copy of a predecessor system that had operated on an IBM Series/1 computer.” The patent law is described as not protecting incremental improvements to a patented invention, which will come as a great surprise to people not just in the software industry but in industry in general. Patents on incremental improvements to a base, patented invention are granted all the time, often to someone other than the inventor of the base invention.

My second quibble is that the authors emphasize reasons to worry about the current state of affairs, rather than reasons not to worry. While, unlike Voltaire’s Candide, I do not think this is the best of all possible worlds, no one has yet put forward a proposal for a new and different form of intellectual property protection for software that is either more predictable, more stable, better accommodating of both innovation and imitation, or more easily administered than the present system.

These two quibbles aside, this book more than meets its objective “to advance the state of knowledge and the quality of the public policy debate on intellectual property protection issues for software.” It is an important contribution to that debate and should be read by anyone interested in seeing all sides of the issue.