REVERSE ENGINEERING OF COMPUTER PROGRAMS — RETHINKING ITS PROHIBITION

• Author: NG-LOY WEE LOON
• Date: 01 December 1994
• Citation: (1994) 6 SAcLJ 131
• Published date: 01 December 1994

INTRODUCTION

The setting up of an advisory committee of experts by Japan’s Agency of Cultural Affairs in July 1993 to study potential changes to Japanese copyright law has raised much furore in the American quarters.1 At the heart of this dispute between the world’s leading software producer and the ““copycat” nation”2 is the fear that the advisory committee will recommend a legalisation of reverse engineering of computer software, which the US would view as a “Japanese plan to weaken copyright protection for computer programs and help Japanese companies catch up in software, an industry that the Americans have dominated”.3

The US companies like IBM are particularly concerned about a common form of reverse engineering known as “disassembly” or “decompilation”. In this process, the zeros and ones of the computer program’s binary code (the “object code”) are transformed into a human-readable language (the “source code”), thereby allowing software engineers to study the interface specifications of the program so as to understand what makes it tick. With such information, new and competing or compatible software can be created, thus encouraging competition which can only benefit consumers. Or so the proponents of reverse engineering argue. Yet it is equally important that adequate legal protection is granted to the creators of computer programs so that they can recoup the investments put into research and development of new programs, in order to encourage further innovation. It is precisely for this reason that, in many countries, creators of computer programs are granted extensive protection within the copyright regime, the scope of which can prohibit reverse engineering of computer programs.

The question that faces the Japanese advisory committee is really how to strike the right balance between these two apparently conflicting interests. This concern is no less that of Singapore’s; it is crucial for the development of the software industry here to appreciate the extent to which it is allowed by the Singapore Copyright Act 19874 to engage in reverse engineering of computer programs first published in the US or created by American citizens or residents, since such computer programs enjoy copyright protection in

Singapore.5 The aim of this article is to examine the approaches taken by the American courts and by the English legislature and their courts in their attempts to resolve this issue, and then to move in closer home to look at the present position of Singapore on decompilation of computer programs. During this exercise, the author hopes to show that the copyright regime can be utilised to promote innovation in the software industry without unduly stifling competition.

COPYRIGHT PROTECTION FOR COMPUTER PROGRAMS IN THE U.S. AND IN THE U.K.

In these jurisdictions, computer programs are protected within the copyright regime as a type of “literary works”,6 if they are “original”.7 The requirement of “originality” in copyright law is far from the stringent concepts of novelty and inventiveness in patent laws; it is satisfied so long as the author8 of the work has expended some independent effort and skill towards its creation.9 Therefore, it is not difficult for a computer program to attract copyright protection. The scope of protection includes the exclusive right to prevent another from:

1. (a) reproducing or copying the computer program;10 and

2. (b) adapting the computer program or making a derivative work thereof.11 In the UK, such adaptation right is statutorily defined

1. to include the act of translating the computer language or code in the program into a different language or code.12

Such exclusive rights relate not only to the whole of the computer program, but also to “a substantial part”13 of it.

As briefly explained above, the process of decompiling a computer program necessarily involves copying the object code of the computer program and converting it into its source code in order to obtain the information necessary for the development of new competing or compatible products. Therefore, decompilation of a computer program, the preliminary step in the development of new computer products, prima facie infringes the copyright in the decompiled computer program, if it was done without the authorisation of the copyright owner. Since the copyright monopoly in a computer program lasts for the lifetime of its author plus 50 years thereafter,14 it becomes impossible for a commercial rival to compete because he is not likely to get the authorisation of the right owner to decompile the computer program while it is commercially viable to do so, given the rapidity with which computer technology becomes obsolete. In fact, the act of decompilation can also amount to a breach of the licence granted by software houses who are known to impose an express term prohibiting such activity, in the licence agreement attached to every copy of the computer program they sell.

Such is the full impact of protecting computer programs within the traditional copyright framework. Did the legislators of copyright laws envisage that the monopoly to be enjoyed thereunder could be used as a weapon to stifle competition in the software industry? It is submitted that this effect could not have been intended. The granting of copyright may undeniably be to secure an economic reward to the author, but this must be but one of the purposes of the copyright scheme, the other being the promotion of works of intellect for the general benefit of the public. In fact, the latter is considered to be the primary purpose of copyright in the US. The federal power conferred by the US Constitution on Congress to enact copyright and patent legislations is:

To promote the progress of science and useful arts, by securing for limited time to authors and inventors the exclusive right to their respective writings and discoveries.15

The US Supreme Court in Twentieth Century Music Corp. v. Aiken16 has elaborated the dual purpose of copyright as follows:

Creative work is to be encouraged and rewarded, but private motivation must ultimately serve the cause of promoting broad public availability of literature, music, and other arts. The immediate effect of our copyright law is to secure a fair return for an ‘author’s’ creative labour. But the ultimate aim is, by this incentive, to stimulate artistic creativity for the general public good.17

An absolute prohibition against decompilation of computer programs under copyright laws runs counter to this primary purpose of copyright, because others would be prevented from further exploiting the “science and useful arts” concealed within a copyrightable computer program and adding to the pool of such “science and useful arts” for the benefit of the public in the form of new (whether compatible or competing) computer products. Further, it is incongruous that patent law, which protects novel and inventive ideas and principles, should compel full disclosure of the idea or principle in the claims and specifications so as to allow the public access to such information, while copyright law which protects works, which need only satisfy the must less stringent requirement of “originality”, should allow the functional elements of the works to be hidden from the public’s study.

A greater objection becomes apparent from the foregoing: the owner of copyright in the computer program is being allowed to exercise a monopoly over functional principles which constitute the idea(s) underlying the program. This is a result which could not have been intended by legislators when they brought computer programs within the ambit of copyright protection, for the axiomatic principle of copyright law is that copyright protects not ideas, concepts or principles, but the form or expression thereof.18

The question of what should be the proper scope of copyright protection for computer programs confronted the US Court of Appeals recently in Sega Enterprises Ltd. v. Accolade Inc..19 The plaintiffs Sega Enterprises

Ltd (“Sega”), a Japanese corporation, developed and marketed video entertainment system, including the “Genesis” console. The defendents Accolade Inc. (“Accolade”), another manufacturer of computer entertainment software, wished to develop game cartridges that were compatible with the Genesis console as well as with other computer systems. Accolade explored the possibility of entering into a licensing agreement with Sega but abandoned such efforts because of Sega’s requirement that it be the exclusive manufacturer of all games produced by Accolade. Accolade had no other choice but to decompile Sega’s video game programs in order to obtain the source code which would reveal the requirements for compatibility with the Genesis console. Having thus obtained the information concerning the interface specification of the Genesis console, Accolade then relied on such information to develop the game program “Ishido” which could be used with the Genesis console as well as with Macintosh and IBM personal computer systems. As Accolade’s “Ishido” was not substantially similar to Sega’s programs, Sega’s complaint was not that Accolade had reproduced its copyrighted works in Accolade’s end product “Ishido”, but that the very act of decompiling their programs in itself was an infringing act. A classic example of a right owner taking advantage of the all-embracing effect of his copyright to aid him to monopolise a certain segment of the software market.

The US Court of Appeal had to agree with Sega that such “intermediate copying done by Accolade…falls squarely within the category of acts that are prohibited by the statute”,20 in particular the exclusive rights to reproduce the work in copies and to prepare derivative works based on the copyrighted work. Yet the Court of Appeal concluded that Accolade was not liable, while acknowledging that:

We are not unaware of the fact that to those used to considering...