RECENT DEVELOPMENTS IN COPYRIGHT AND PATENT PROTECTION OF COMPUTER SOFTWARE1

• Author: CHRISTOPHER LEE GEN-MIN
• Published date: 01 December 1994
• Citation: (1994) 6 SAcLJ 321
• Date: 01 December 1994

The law of Singapore has yet to deal with any of the new and difficult issues peculiar to the protection of intellectual property rights in computer programs, but it is probably only a matter of time before it does. When that time comes, we shall have the luxury of being able to learn from the experience of other jurisdictions. It is proposed in this article to briefly identify the major legal issues peculiar to the intellectual property protection of computer programs and consider developments over the last couple of years with an eye to the lessons that they hold for us. Constraints of space means that consideration will be limited to landmark cases or statutory interventions and trends in the law that have become discernible over that period in the United States, the United Kingdom and Australia,

I. INTRODUCTION

a. Computer Programs

Before considering developments in the law, it is necessary to introduce the subject matter of this article briefly. A starting point would be the definition in the Section 7 of the Copyright Act2:

“computer program” means an expression … of a set of instructions … intended … to cause a [computer] to perform [functions].

This definition would almost certainly drive a technically trained person to despair for its lack of technical precision, but it does indicate two very important things about computer programs:

1. (a) every computer program is essentially a set of instructions to the computer on which it is to run; and

2. (b) when a properly-written program is run on a computer, the computer performs functions. Without computer programs to direct the computer’s internal operations, the computer hardware is nothing but inert machinery.

These two features are the source of a great deal of conceptual and legal difficulty. Computer programs are ‘information’ (that is, sets of instructions)

and yet at the same time, they are akin to machine parts.3 Computer programs are intangible but are capable of leading to tangible physical results.

Although we tend to credit computers with almost superhuman ‘intelligence’, the truth is that the brains of a computer, its central processing unit (‘CPU’), is capable of performing only a few very basic logical operations. The magic behind the computer is that it can perform these operations at great speed under the control of a computer program. It is the manipulation of electronic imputs at great speed by the CPU and other hardware in accordance with the program that produces electronic outputs which may in turn result in the tangible physical results that we now take for granted. The detailed technical wizardry by which this is achieved lies beyond the scope of this article, and the writer’s ability to explain briefly, so we will have to take this effect as a given. Those who wish to explore the technical aspects of computers and computer programs further may wish refer to sources intended specifically to explain the basics of computer technology to the layman4.

b. Copyright

The law of Singapore and the other countries under consideration provides that the basic protection for persons who develop or exploit computer programs is through copyright. The choice of copyright is a pragmatic one; the other major forms of intellectual property protection — patents, the law of confidence and trademarks — are in many ways even less suitable for dealing with computer programs. This choice has been implemented in all the jurisdictions under review by simply treating computer programs as literary works5. This means that effectively, intellectual property protection of computer programs focuses on the ‘instructional’, textual aspect of programs rather than its ‘machine part’ aspect. On the surface, that is understandable. Every computer program is capable of taking a form very much like a traditional literary work; the instructions constituting the program may be printed out in ‘hardcopy’. Stored in a computer, those

instructions are little different from any word processor file which might contain a true literary work. The normal use of such a program simply involves its translation into electronic signals for electronic consumption by a computer, and all of this seems well within the ability of copyright law to cope.

At this level, copyright law is more than adequate for dealing with ‘piracy’ of computer programs — mass, unauthorised copying of entire programs for sale at prices that undercut legitimate copies. In this respect, computer programs are no different from other copyright works like compact discs and video cassette tapes since the copying involves taking the entire content of the work, or of a computer-usable translation of the work. The problems in this area are the same practical problems of enforcement and proof that afflict other copyright works and thus fall outside the scope of this article. The problems unique to the application of copyright to computer programs lie elsewhere.

At the risk of oversimplifying, the similarity between computer programs and normal literary works like novels goes beyond the fact that computer programs can take the form of words on a page. Besides the literal text of the instructions, which corresponds to the words of a novel, a computer program has a textual ‘structure’ which can be compared to the ‘plot’ of a novel. The sequence of instructions making up the program must be organised according to programming rules to achieve the ultimate purpose of the program, just as the words making up a novel have to comply with rules of grammar and logic in order to effectively communicate the meaning which the writer wishes to convey.

This apparent similarity is deceptive; copyright protection of the structure of computer programs presents a number of special difficulties which do not arise with traditional literary works. First, the structure of a computer program is likely to be dictated by external factors to a much greater extent than the plot of a novel6. Considerations like the function to be performed by the program and the dictates of the hardware on which it is to be run have a very significant impact on the form which a program’s structure ultimately takes. A computer program must be as ‘efficient’ as possible to be commercially viable; it must not occupy too much space in a storage medium like a computer hard disk, it must be able to lead to the computer’s performance of the appropriate functions at a reasonable speed given the limitations of the CPU and the random access memory (‘RAM’) which the user’s computer is expected to contain. Much of the commercial value of a computer program lies in its structure, since this determines how

well the program deals with these constraints to operate as the user would demand. In this regard, the text of the program is far less significant than its structure.

Also, much of the effort and cost of designing a computer program goes into the design of the program structure, and the writing of the literal text is again relatively insignificant in comparison. Non-literal copying of programs — that is, copying the structure of the program only — is thus a commercially viable activity because a programmer writing a program to perform a particular function could save himself much time and effort if he simply copied the structure of a successful program which performed a similar function. This would hold true even if he had to supply all the literal code himself. Copyright law is fundamentally flawed in its application to computer programs because it focuses on the literal text of the copyright work in question, which in the case of a computer program is often relatively less valuable than the structure.

While copyright does provide some protection for non-literal aspects of traditional literary works such as the plots of novels, its application to the non-literal aspects of computer programs is conceptually more difficult than for traditional literary works. It is a relatively simple matter for a judge faced with a claim that the plot of a novel has been copied to compare the two works and intuitively decide whether ‘copying’ of plot is so significant that it amounts to infringement. Such analysis is not possible with computer programs since few judges are in a position to perform that kind of intuitive analysis. Judges hearing claims for copying of the structure of computer programs are therefore faced with the problem of putting into clear words the appropriate legal analysis so that experts can reliably assist in the process.

The constraints affecting computer programs add another difficulty in relation to claims for infringement by copying of the structure of a program. Proof of copying of copyright works usually involves showing such close similarity between the alleged copy and the allegedly copied work as to make independent creation unlikely. But programs performing similar functions on similar systems will be subject to the same set of external constraints, so there is a greater likelihood of similarities between two such works, even if they were independently created. It would therefore be more difficult in practice to establish copying on the basis of similarities between two works which are subject to the same constraints since those similarities might also be a result of independent creation subject to the same external constraints. In a literary work, the problem would be comparable to trying to determine whether copying had taken place where two writers wrote descriptions of the same historical subject, with both under a very strict word limit as well as the same restrictions on the range of vocabulary that could be used in the description. Similarities between the two resultant works would not necessarily be probative of copying. If the constraints were severe enough that there was only a very limited

number of ways in which the subject could be described, a further question might arise: whether copyright could subsist...