Introduction 4
Copyright in computer software in the United Kingdom 4
The Copyright, Designs and Patents Act 1988 5
Infringing acts 8
Whelan Associates Inc v Jaslow Dental Laboratory Inc 11
Computer Associates Inc v Altai Inc 13
The aim of copyright protection for software 14
The mechanisms of software copyright protection 16
Copyright protection and reality 18
Conclusion 19
Bibliography 20
A B S T R A C T
The protection of software by copyright has developed from an uncertainty to a complete prohibition on copying and an extensive prohibition on adaptation, modification and distribution of both the code and preparatory work. This article examines the statutory regulation and judicial emergence of copyright in the United Kingdom and United States respectively and argues that the emergence shows major flaws, particularly in its attempt to distinguish between idea and expression. It is further argued that the current model of legal protection is at least partially unjustified in doctrinal terms, and that an altered model should be adopted, reflecting software’s technological nature. Finally, it is considered that there may be an alteration in copyright law to reflect current reality with regard to “home copying practices” and the actual necessity – as opposed to available luxuries – behind copyright protection.
Introduction
Copyright in computer software, although now considered a legal certainty, faces growing opposition in a world of broadband Internet connections and removable mass data storage media. The worldwide losses resulting from illegitimate copying were estimated at US$10.97 billion in 20011, and the Software & Information Industry Association (SIIA) warns of a “serious potential threat to the intellectual property providers”, stating that “hundreds of software and information companies lay off workers and cut back product lines”2 as a consequence of software piracy. Microsoft3 claims that some 10,000 jobs were lost in the District of Columbia and neighbouring states alone as a result of software piracy in 1997.
The industrial reaction to such statistics has been to hold them up to emphasize the need for far-reaching legal protection of computer software – in society’s own interests. However, this attitude suggests that the main difficulty lies in breach of copyright. This article examines the legal and economic justification of copyright protection in this area with a view to demonstrating that the attitude described above is too simplistic, and that an altered model of copyright protection may, after all, be desirable.
Copyright in computer software in the United Kingdom
The invention of the first digital computer is widely attributed to Konrad Zuse4, and marked the way in which computers would develop in the near future. For example, the ENIAC, an American invention of 1946, boasted some 19,000 vacuum tubes, required 200 kilowatts of power to operate, and was split into 30 separate units5. Such devices were rare, required an entire room to be set up, and could only be afforded by governments. With computer programs stored in the form of punch cards designed for specific machines, the danger of “piracy” was negligible6.
It was not until the mid-1970s7that personal computers8 became widely available and thus the question of legal protection became a serious issue. The Whitford Committee established in 1977 that “copyright law was unsatisfactory as regards computer programs.”9 There had previously been a general assumption on the part of many writers that copyright subsisted in computer programs by virtue of the general provisions of the 1956 Act. Computer programs, which they classified as literary works, however, were defined as including “any form of notation, whether by hand or by printing, typewriting or any similar process”10. The definition is not exhaustive, but it was thought by some that literary works probably needed to exist in a written or printed form, which left the question of magnetically stored software open, since this left the actual programs intangible, unless they were printed out. Even if printed out, however, there remained the problem of copying from tape to tape – in that instance, one cannot speak of a printout being copied. Some writers resorted to s49(4), which provided for the protection of literary works in “writing or some other material form”. Wilson11 points out, however, that this is not certain to include magnetically stored programs. Software companies were therefore hesitant to litigate on the basis of copyright infringement12, simply because it was not clear whether copyright even subsisted in programs13. Such hearings as took place were usually interlocutory, and judges were reluctant to come to any certain conclusions.
Thus it was not until the Copyright (Computer Software) Amendment Act 1985, which plainly stated that copyright did apply to computer programs, that any degree of legal certainty in this field existed. This led to the Copyright, Designs and Patents Act 1988, which reflects the current statutory law.
The Copyright, Designs and Patents Act 1988
The 1988 Act14 expressly includes computer programs within the definition of literary works (see s3(1)(b)) and overcomes the question of magnetic storage by stating that literary works must be “recorded, in writing or otherwise”15, and that the method and medium of storage do not matter16.
As is the case for all literary works, a computer program must be original to obtain copyright protection, i.e. it must originate with the author and there must have been “a modest amount of skill, labour or judgment”17 involved in its creation. This will, on the whole, not be problematic – most programs for which copyright protection is claimed in court are more than a few lines long18 and will invariably satisfy such requirements. The one exception is where a program’s expression is the inevitable result of its function19 – that is to say, where the program could not have found any other expression given the task it was to perform. This is generally only likely to be the case for microcode written as an instruction set for a particular chip, if at all.
Establishing that copyright subsists in a computer program will therefore not be an onerous task. When a program has been copied, however, has proven rather more difficult to discern, and it is helpful at this stage to consider what software actually comprises of.
Since a computer’s processor only “knows” of two commands – on and off (represented in binary as 1 and 0 respectively) – programming languages such as BASIC20, Python, Java, C, C++ or Assembler were developed to facilitate programming, given that it would be a near-impossible task to program purely in 1s and 0s. The result is that two forms of programming
code exist – the source code, being the code which the programmer writes, and the object code. The latter is produced as a result of compiling the source code, and can be passed on to the processor for execution (see Figures 1.1 and 1.2, overleaf).
Object code and source code – a program to convert .eml files to .txt files
Language: HiSoft BASIC 2
Author: W. White
OPEN "ram:temp" FOR INPUT AS #1
DO UNTIL EOF(1)
LINE INPUT #1, a$ : LOCATE 1,1 : PRINT a$;"***"
OPEN a$ FOR INPUT AS #2 : a$=LEFT$(a$,LEN(a$)-4)+".txt"