AustLII Home | Databases | WorldLII | Search | Feedback

Journal of Law, Information and Science

Journal of Law, Information and Science (JLIS)
You are here:  AustLII >> Databases >> Journal of Law, Information and Science >> 1994 >> [1994] JlLawInfoSci 8

Database Search | Name Search | Recent Articles | Noteup | LawCite | Help

Duigan, Alicia --- "Why the Copyright of 'Look and Feel' is Not Applicable to Australian Law" [1994] JlLawInfoSci 8; (1994) 5(1) Journal of Law, Information and Science 78

Why the Copyright of 'Look and Feel' is Not Applicable to Australian Law

ALICIA DUIGAN[1]

Abstract

This paper outlines the developments which have led to the protection of look and feel being rejected by Australian legislators. Australia's Copyright Law Review Committee has produced a draft report on computer software protection, where it was decided that the principles expounded in the American cases did not apply to the Australian situation.[2] This was partly because of the inherent flaws in the reasoning displayed by the American judges. The American position has however recently shifted with the decisions of Computer Associates v Altai and Apple v Microsoft, where the Courts expressly rejected the former approach to applications for protection of 'look and feel' of the computer screen, which suggests the draft report is correct in excluding protection for look and feel. The other decisive factor is that Australia's computer industry is occupied with standardisation and compatibility as opposed to America where the computer industry is innovative and world leaders in their field.

1. Introduction

The protection of 'look and feel has been a contentious issue for many years, with widely diverse views being expressed on whether copyright law should afford such protection. The 'look and feel' of the computer program may be defined as its user interface. The 'look' of a computer program is the display generated when a computer executes a program. The 'feel' of a computer program is its function and overall purpose which is in part determined by the structure of the program and the sequence of commands it contains. Computer programs are regarded as having the same feel if their overall pattern and theme is the same. Thus look and feel incorporates two aspects of a program's behaviour for the user: visual and operational.

With the increasing proliferation of personal computers the user interface has become the single most important factor in marketing software. The user interface may be described as the graphic screen displays and imagery which appear when using the computer to make it easier to use...(the user interface) presents the functional capabilities of the computer to the user and provides the means of communication and control...(the user interface) provides the window through which users pass messages to the computer and view what happens in response.[3] It is one of the most important facets of mass marketed software because it is the feature given the most weight by potential purchasers.[4] Also, it serves to identify a particular creator or software manufacturer.

As software developers continually increase the sophistication and thereby the cost of their user interfaces, the desire to protect them from being copied and distributed has led them to seek legal protection. Five main reasons are advanced why the design and engineering of the user interface is now emphasised in computer systems development:

1) Advances in computer science and electrical engineering have placed better technology at the disposal of those designing and building user interfaces,

2) With better interfaces for non programmers, the user interface has become a crucial issue in marketing computer software,

3) The user interface typically accounts for at least forty percent and sometimes much more, of the computer instructions written in building a commercial application program such as text editor, statistical program, or accounting package,

4) The user interface is a growing concern to computer system developers because standards are being built around them,

5) The cost of learning a software package is substantial.[5]

Thus there is increasing pressure on law-makers to provide protection for the user interface,

"In today's ultra-competitive market, software developers have invested a great deal of time and money in developing the user interface...With this increased investment in resources, software developers have sought legal protection for their creations."[6]

The issue law-makers then face involves a balance of two competing public policy arguments, intellectual property protection and competition. More precisely, the need to balance the encouragement of innovation and the reward of authors for their labours versus the policy of preserving competition. The usual justification by government for intellectual property rights is that intellectual creations benefit society generally, and therefore protection is justified to reward creators and provide an incentive to others to create. It has been argued that the economic incentive provided by protection beyond that which already existed in contract or trade secrets, was unnecessary for the development of the software industry and remains unnecessary for its present or future health. Software producers would reply that their claim for protection against piracy rests on the character of software, and intellectual property generally as a public good.

Like other intellectual creations, the problem is that without specific legal protection, appropriation by others cannot be prevented. This results in a market failure leading to reduced capital investment in software production, economic inefficiency, and just plain unfairness, which can and should be protected by specific legal protection. The appropriate legal protection would be some form of intellectual property scheme, which in effect enforces the appropriation of the public good as private property which can then operate in a market system. There are fears that if copyright protection for computer programs is extended to include interfaces, the potential monopolisation of interfaces which will ensue may curtail competition in the computer market and inhibit the development of inter-operative systems. Yet the law recognises that original and creative work will not be encouraged if the product of an inventor or author may be used or exploited by those who have put neither time nor money into the work of creation or of development. So the prevailing attitude will influence the degree of protection software developers can attain for their user interfaces.

Also important is the fact that any decision about intellectual property law is also a decision about the grounds for competition, and will thereby inevitably shape the future of the software industry. Narrow protection (for example protecting only literal source code), for instance, would permit more latitude in creating functionally equivalent programs. This in turn would lead to a marketplace with more software clones and likely shift the basis of competition from functionality to other factors like price, quality of implementation, user support, and so on. Broader support, by contrast, would likely lead to an increased emphasis on innovative functionality in second comers: being competitive would require functionality sufficiently better that it could win converts from established alternatives. So economic considerations will also have a bearing on the decision whether to protect the screen or not.

The scope of protection provided for user interfaces will have a major impact on the economic development in this software field, including considerations of standardisations and compatibility. The primary objectives in designing user interfaces are the user's ease of learning and use. Through studying how users work with different devices or languages and how they interpret different messages and symbol systems (words or icons), user interface designers can improve the effectiveness of the user's interaction with the computer.[7] Once users have mastered one way of coping with a computer they may be loathe to change and it is exactly this 'look and feel' or sequence, structure and organisation of the original program which is likely to be copied by competitors. This factor has practical implications for software developers in that compatibility and standardisation may well be just as important as protection of the user screen. Thus economic policy plays an important role in the applicability of copyright law to the look and feel of a user interface.

To understand how to best protect the intellectual property of user interfaces, lawyers need to understand the development of interface technology and human factors engineering. Lack of knowledge may lead to the situation where copyright is granted on a false understanding of the workings of a computer. Making good laws to protect the intellectual property of user interfaces requires that lawyers, judges, and legislators understand how good user interfaces are developed.

2. United States Developments

In the United States, under the Copyright Act of 1976, computer programs are copyright as literary works and the screen displays created by the programs are copyright as audiovisual works. This separation of copyright for computers has led to the emergence of two streams of computer copyright cases which have combined to form the basis for 'look and feel' litigation: the literary work and audiovisual work streams. The literary work stream was extended to cover the structure, sequence and organisation of a computer program with the decision of the Third Circuit in the case of Whelan v Jaslow. [8] It was decided categorically that structure was a proper subject matter for copyright. The judges did not use the term 'look and feel' but commentators have accorded Whelan landmark significance because it adopted the approach that the structure, sequence and organisation of a computer program deserves copyright protection. However, upon analysis of the reasoning of the judges in Whelan , it appears to have been based on a misunderstanding of the workings of a computer.

From a report by a computer scientist [9], we find that a program is not only a text; a second fundamentally important property of a program, is its behaviour. It is crucial to recognise that these two views of the program are quite distinct. In particular, they each have their own structure, sequence and organisation.

"This is reflected in a useful piece of standard computer science jargon: Programmers talk about both the static structure and the dynamic structure of programs. The static structure is the organisation of the program as a textual document (the program-as-text view); the dynamic structure is its behaviour, what it does when it actually runs (the program-as-behaviour view). It is quite possible to write any number of differing versions of a section of computer code in such a manner, that they will all perform exactly the same function in so far as the user is concerned. The crucial observation here is that we can quite literally switch the sequence of two routines in a piece of code, but the behaviour will be completely unchanged. By that I mean that there is no way that you as a user could determine which of the two programs you were using." [10]

This demonstrates that the two aspects of a program - text and behaviour - are only loosely connected: we can make significant changes to one without affecting the other. Hence the structure, sequence and organisation of the text of a program is distinct from the structure, sequence and organisation of the behaviour of the program. This distinction was not drawn by the judges in Whelan and they failed to advert to the distinction at all. This tends to indicate a lack of technical know-how on the part of the judiciary and undermines the strength of Whelan's authority.

The independence of program as text(static structure) from program as behaviour(dynamic structure) has several significant consequences. The first consequence is that it makes no technical sense to talk simply about the "structure" of a program, because the term is ambiguous, and the distinction matters.

"Second, with all due respect for an otherwise carefully argued position, Whelan v Jaslow runs afoul of this distinction: The litany of "structure (or sequence or organisation)" is flawed by the failure to distinguish between the static and dynamic views of a program. As a consequence the terms are inherently ambiguous, in the important manner noted above. Indeed, the decision trips coming out of the starting gate, in note 1:" [11]
"We use the terms "structure", "sequence" and "organisation" interchangeably when referring to computer programs, and we intend them to be synonymous in this opinion."[12]

This is, alas, exactly wrong technically: it is precisely because a program is not only text, but also behaves, that these terms are not synonymous.[13] These criticisms have serious ramifications for the future application of the look and feel cases, for if the seminal case upon which future cases are based is fundamentally flawed in its reasoning it follows that such flaws might be common to all subsequent cases. In addition Whelan has been criticised as not maintaining a distinction between the idea and its expression and thus contradicting one of the fundamental concepts of copyright protection. The position critics take, is that the protection afforded in Whelan extends too far, to protect the underlying function of the program and thus erodes the idea/expression dichotomy.

Whilst it may be argued that Whelan is applicable in Australia, the lack of a separate category for audiovisual copyright in the Copyright Act appears to suggest that the audiovisual cases are not applicable to Australia. However, this position has changed since the case of Broderbund,[14] which represents the union of the two streams of cases, in that the Court applied the literary stream cases to audiovisual copyright. The audiovisual copyright has been determined by the court in Stern v Kaufman [15] to protect an identical display, even where the program code is different, but where there is any interaction by the user of a computer, which changes the display, this will not invalidate the copyright.

The Court held that the reasoning of Whelan applied to audiovisual copyright as well, when it said that,

"Whelan thus stands for the proposition that copyright protection is not limited to the literal aspects of a computer program, but rather extends to the overall structure of a program, including its audiovisual displays."[16]

Softklone [17] then rejected the broad reasoning of Broderbund, and expressly distinguished between the copyright of the program and the screen displays, ultimately relying on the fact that an identical screen could be created by separate and independent programs, and thus could not be regarded as a copy of the program. The court here tended to focus on the process or manner in which the screen operated, and not on the function of the underlying program, as Broderbund had done. They held that the process or manner by which the screen operated was the idea, while the arrangement of information in the status screen was an expression of that idea. The court therefore greatly narrowed the concept of what constitutes expression in computer software screen displays. They found that Broderbund's analysis was flawed, because it could find infringement in the case of two entirely different programs with entirely different purposes, possessing one substantially similar screen display. In such a situation copying is not likely to have occurred, and the similarity of the screen display is likely to have arisen from other factors. Interestingly, the court rejected the defendant's argument based on the importance of standardisation in the computer industry and so policy appears to have played a role in the decision of the court. When looking at the issue of standardisation later, it will become apparent that standardisation is not as important in the American market as it is to Australian software developers.

Then came the equally important case of Lotus v Paperback [18] where the District Court for the District of Massachusetts ruled that the structure, sequence and organisation of the menu command system is a copyrightable non literal element of the computer program. This case has been criticised on the basis that the judge did not hold that there was any copyright in the 'look and feel' of the program. 'Look and feel' as a separate type of protected expression is impossible to define, and leaves too much to the judge's discretion. Second, the judgment concentrated on the evidence which Lotus adduced to prove that Paperback had copied the audio-visual works which made up the user interface. It did not examine the interface as a whole, but each element separately. Third, the court took the approach that where the expression was dictated by external constraints, similarity would not amount to evidence of copying. However, it also held that where the constraint was the success of the plaintiff's program in the market, that was not an external constraint permitting the defendant to use similar expression.

"Many software developers are unhappy with the decision, arguing that if they are unable to use industry standard user interfaces the software industry will stagnate. Intellectual property lawyers, on the other hand, should be pleased. The purpose of the intellectual property laws is to protect originality and innovation. Had the case been decided in the defendant's favour, it would have meant that, in the United States at least, original software would lose much of its protection if it was so original that the majority of users adopted it."[19]

The development of the look and feel cases in America has received a fatal blow in the recent decision of the District Court in Computer Associates v Altai. [20] The court applied the extrinsic-intrinsic test to determine substantial similarity. The extrinsic test inquires whether there is sufficient similarity to conclude that the infringer used the copyrighted work, Judge Sparr explained, while the intrinsic test inquires whether a lay observer, without expert testimony, would find illicit copying or unlawful appropriation.

The court in Computer Associates v Altai expressly rejected the approach of Whelan v Jaslow by viewing a program as not only text but also as behaviour, and concluded that the behaviour aspect of a program is ineligible for protection as a process. The court also rejected Whelan's analysis that a program's overall purpose can be formulated as the program's idea. Judge Sparr focused on the program as text, in either source or object code, and found that only structure which is text, and not behaviour, is protectable. This was earlier recognised by Davis when he said,

"Programs are not only text; crucially, they also behave, and it is their behaviour that is most central to these discussions. Because they are both text and behaviour, creating programs is simultaneously an act of authorship and an act of invention; programs are both engineered devices and literary works."[21]

Thus the court has then acted upon the criticisms of Davis and other commentators in criticising the reasoning of the judges in Whelan. It further found that the look and feel of the complexity level, according to the court, is closer to the idea end of the idea-expression continuum and is thus excluded from protectability. The court felt it was more productive to look at behaviour and engineering calculation modules which are more expression than idea.

"This court is of the opinion that it is far preferable, especially in an area of legal and technological sophistication as complex as the area of copyright protection, to draw upon a larger arsenal of facts in order to design or derive the appropriate legally significant facts. Once these are gathered and expert testimony is heard, the court can then analyze which portions of the program, according to the expert testimony, infringe the protected expression."[22]

The case leaves the state of look and feel in America and its potential for protection under copyright unclear. Obviously the structure, sequence and organisation test of Whelan no longer applies and arguably the look and feel is an idea and therefore not capable of protection. The court appears to have significantly narrowed the scope of protection available to software producers in the realm of copyright law.

"They (the courts) have also emphasised that while the copyright system provides an incentive for creative activity, its ultimate purpose is to serve the public interest by promoting broad availability of new creative works. Doctrine and policy required that copyright protection for computer programs be confined narrowly."[23]

The faint hope that there may still be protection for 'look and feel' in American copyright law would surely have been quashed when the recent Apple v Microsoft [24] decision was handed down. In his ruling in that case, Judge Walker said,

"The court believes this is a fundamental misunderstanding of the law."

Implicit in the judgment, is a rejection of the notion that the total concept and feel of the Macintosh graphic user interface is protectable expression. Rather, unprotectable elements of the expression should be identified and excluded prior to application of the substantial similarity test, appears to be what the court has proposed. Thus the court is clearly separating the idea from its expression, and in doing so have found that the concept of look and feel is not protectable, if based solely on unprotectable elements. Judge Walker insisted on examining each feature for copyrightability. In rejecting the 'look and feel' approach Judge Walker explained that,

"the elements of such an arrangement served a purely functional purpose in the same way that the visual displays and user commands of the dashboard, steering wheel, gear shift, brakes, clutch and accelerator served as the user interface of a car. Purely functional items or arrangements were wholly beyond the realm of copyright;...Moreover the similarity in competing products of such functional elements of a user interface or their arrangement did not suggest unlawful copying, but standardization for functional reasons."[25]

Besides being applied in Apple v Microsoft, Computer Associates has also been applied and followed in a number of other jurisdictions. In particular, the judgments delivered in John Richardson Computers Ltd [26] in the UK and the Carolian Case[27] in Canada appear to have followed Computer Associates.

In John Richardson Computers Ltd Justice Ferris rejected the argument that any general rule as to the means by which copying of a computer program might be established by past judgments.[28] Justice Ferris adopted the abstraction-filtration-comparison test from Computer Associates v Altai [29] however, he adapted it by interpreting it himself and firstly looking at the program as a whole and deciding if there is any substantial copying. By adopting such an approach, Justice Ferris tended to undermine the effectiveness of the Computer Associates test [30] by deciding early on that the program as a whole was copyright protectable. "Once this decision had been made it was small wonder that an attempt to carry out the processes of abstraction, filtration and comparison came to grief." [31] It appears then that, although approving of the abstractions test in Computer Associates, Justice Ferris seemed to have some difficulty in applying the test to the facts before him and the case cannot be seen as a direct application of the test.

A better example of an adoption of Justice Sparr's abstractions test is the Carolian Case. In that case Justice O'Leary first noted that,

"Some problems in programming do not lend themselves to multiple solutions. Such problems can be solved in only one or two ways, no matter how many different programmers tackle them." [32]

He felt that as the same programmer had written both programs the similarities between the two could be accounted for by reasons other than copying. Justice O'Leary then reviewed the US cases and found that,

"[S]ome method must be found to weed out or remove from copyright protection those portions which, for the various reasons already mentioned, cannot be protected by copyright. After the portions that are not copyrightable have been filtered out, there may or may not be any kernels or golden nuggets left to which copyright can attach." [33]

Thus, whilst not expressly adopting the abstractions test from Computer Associates Justice O'Leary appears to find it a useful tool in determining whether copyright has been breached.

The changes in the approach of the courts, in the US and elsewhere appears to spell the end of copyrightability for the 'look and feel' of user interfaces. In the New York Times on April 16, 1992, it was said by Esther Roditti Schacter, lawyer and publisher of the Computer Law and Tax Report that the decision demolishes look and feel. She felt that the concept of look and feel had blurred the distinction between patents which cover ideas, and copyright which cover the expression of ideas. "The idea that you could have a copyright as broad as look and feel implied gave you a monopoly based just on appearance," she added, saying that was beyond the intended scope of the copyright laws. Many other commentators have found that Computer Associates has decreased the level of protection available to software developers for the 'look and feel' of their user interfaces. In particular Peter Stone and Peter Groves noted that,

"Fundamentally, the change in approach consists in rejection of a broad, overall assessment of substantial similarity (in terms, for example of 'look and feel' ), and insistence on dissection of the plaintiff's program, with a view to elimination of its unprotectable features (in terms of idea, merged idea and expression, and unoriginality ), before the comparison between the programs is made." [34]

Thus Computer Associates v Altai has effectively reversed the broad protectionist approach which had been accepted in previous Circuit judgments.

3. The Australian Position

Australian courts to date have not had to consider the question of whether copyright exists in the computer screen display. This prompts the question of whether legislators should pre-empt such a case by providing legislation to cover the eventuality. At present a computer program is defined under the Copyright Act 1968 as a literary work. This legislative attempt to provide protection for computers has been criticised as a stop-gap measure resulting from the unsuccessful litigation in the case of Apple Computer [35].

Dealing with the case of Apple Computer Inc v Computer Edge Pty. Ltd. it was said by Allan Bell,

"The comments of the court suggest that the issue of copyright in regard to software will remain unclear until such time that there is a greater understanding in the community about computers and computing in general."[36]

In 1984 the Copyright Act was thus amended to take account of computer programs and protect them under copyright law. Under the Copyright Act s31 it is an infringement of copyright to reproduce a literary work in a material form or to make an adaptation of that work. Only one case has since ruled on these amendments, that of Autodesk.[37] That case is not relevant to the present discussion as it concerned computer hardware, not software but does highlight the relative infrequency of computer copyright cases in Australia.

Recently the Copyright Law Review Committee, acting on a reference by the Attorney-General, considered the adequacy and appropriateness of copyright protection for computer programs. The Committee published a draft report for public comment in June 1993. In that report the CLRC concluded that Australian copyright law should not be extended to include 'look and feel'.

"It is the Committee's view that the 'user interface' or 'look and feel' of a computer program is properly characterised as behaviour rather than expression capable of copyright protection. In the Committee's view the need for standardisation and for development of efficient user interfaces outweighs the need to grant to authors of computer programs copyright protection of the 'look and feel' of a program's behaviour. The Committee recommends that no amendments should be made to the Act to secure protection for the 'look and feel' of computer programs." [38]

Many of the reasons for the Committee's conclusion concern maintaining the distinction between the idea and the expression of the idea; protecting development in the industry; in the interests of standardisation and the behavioural aspects of the look and feel of the user interface. The Committee started from the realisation that the nature of computers and the tendency of users to want standard features led to the situation where the "possibilities of copying, inadvertent or deliberate, by a competitor seem to be very great." [39]

Firstly the behavioural aspects of the computer were held to be outside the realm of copyright law and thus to provide protection would tend to erode the idea/expression dichotomy. The Committee "notes the behavioural nature of 'look and feel ' and considers that it is inappropriate for copyright protection to extend beyond expression of an idea to the behavioural aspects of computer programs."[40] So the Draft Report has highlighted the need to maintain the distinction between protection for expression of ideas in a material form and the nonprotectable elements of the behavioural aspects of programs. It was this analysis which was lacking in the earlier US cases, and which the latter cases of Computer Associates and Apple v Microsoft sought to correct.

Another factor which was often lacking from the Circuit judge's analysis, but which is pertinent to the Australian situation is a discussion of the need for standardisation in the computer industry. The report stated that,

"[T]he Committee notes the dangers to the industry as a whole in encouraging diversity for diversity's sake, where the resulting variations do not result in the most efficient and acceptable 'user interface' features being used or developed, simply because the most efficient and user friendly features have copyright protection." [41]

It is clear that the Committee has considered the importance standardised features are to users and that the awarding of copyright protection to one software developer would be tantamount to awarding them a monopoly as other developers could no longer adopt the same features. The Committee felt that standardisation and the need for competition in the form of advances and improvements in the efficiency of the screen display outweighed the need to reward a computer programmer for their innovations. This was felt to be in the "public interest" and the welfare of the public in general was held to be of greater economic importance than the welfare of select interest groups such as software developers. [42]

"[P]rotecting software structure, sequencing commands and language could rapidly exhaust alternatives and lead to more inefficient structures...due to the evolutionary nature of interface development, interface copyright will actually retard progress." [43]

Thus the CLRC has conclusively sided with those who advocate no protection for the 'look and feel' of computer screen displays, which given the inherent defects in the granting of copyright to such an area of computer law, appears to be the best approach. As such Australia will afford developers no such protection for the look and feel of their screen displays. However, the true extent of protection for computers in Australia is still left in a state of flux as it is unclear whether the existing law will cover a computer screen display.

"The Committee's conclusion that the 'look and feel' of computer programs amounts to behaviour and as such should not be protected by copyright in the program and its view that such copyright should not extend to the 'look and feel', does not dispose of the question of what protection, if any, should be afforded to screen displays." [44]

Section 10 of the Copyright Act 1968 defines a computer program as "an expression...of a set of instructions intended...to cause a device having digital information processing capabilities to perform a particular function." The Explanatory Memorandum said of this definition,

"The phrase "expression...of a set of instructions" is intended to make clear that it is not an abstract idea, algorithm or mathematical principle which is protected, but rather a particular expression of that abstraction. The word "set" indicates that the instructions are related to one another rather than being a mere collection."[45] Within that definition no mention is made of the computer screen display and so first it needs to be determined exactly what is a computer screen display.

A computer screen display may be defined as "the temporary presentation of visual information, text or graphics which exist as electrical impulses in a computer's memory."[46] There is usually no hard copy and the screen display may take a number of forms, ranging from artistic screens to text and graphics. The protection afforded to a computer screen display may well depend on which form the figures on the screen take.

To be capable of copyright protection, a computer screen display must be characterised under s32 as an "original" work made by a "qualified person" and the screen must be sufficiently fixed in a material form. In looking at originality and the processes behind the computer screen display, Jack Brown came to the following conclusion,

"I think it is a self-evident proposition that a computer human interface, with it (sic) heavy dependence on graphic design and animation and other motion picture techniques, is a highly creative enterprise, with many ways for the interface designer to accomplish his objective of communicating with the user."[47]

It is clear then that it is recognised within the industry that creative programs do exist and that computers are capable of being authors. The main hurdle is that there is no audiovisual copyright in Australia per se. In Australia, the two strongest possibilities for protection of computer screen displays are as artistic works or as literary works. Looking at each of these categories separately will enable us to determine the extent of protection for the user interface in Australia.

4. Artistic Works

This is defined in s10 of the Copyright Act to include inter alia a painting or drawing, whether the work is of artistic quality or not, and a work of artistic craftsmanship. There is no definition of "paintings" in the Copyright Act, and so the courts have turned to the ordinary meaning of a painting, but have not restricted it to mean only the use of paint. "It could be argued that there is little difference between a computer generated picture and the same picture which is created by the application of pencil to paper, and that therefore a computer generated picture is an artistic work."[48] A United Kingdom decision, which dealt with the issue of what constitutes a painting found that "a painting is not an idea; it is an object; and paint without a surface is not a painting."[49] This comment focuses on the problem which is raised in relation to the computer screen display: is it sufficiently fixed in a material form within the definition in the Act to be covered under the Copyright Act ? This problem will be dealt with later in the essay.

Drawing is defined in the Act to include "a diagram, map, chart or plan." This definition is not exhaustive and the courts generally rely on the word's ordinary meaning. It is possible to view a computer screen display as a drawing because it is a surface upon which an arrangement of lines can be displayed in the form of "artistic" images such as icons. However, it is unclear whether the courts would be willing to interpret the word "drawing" in such an expansive manner. The other alternative is to try and classify a computer screen display as artistic craftsmanship. In dealing with this category it is accepted that to be artistic the works must have aesthetic appeal.[50] Furthermore the test is the object which the author had in mind when creating the work. This test was expounded in Cuisenaire [51] where Pape J said,

"The emphasis is thus upon the object of the author in creating the work, rather than the reaction of the viewer to the completed work."[52]

If this test is applied to computer programmers, not their audience it is arguable that all computer generated images have aesthetic appeal, if that was what motivated the author to produce them. This is true, for arguably an image which is appealing to the programmer may look mundane or ordinary to the user, but it is a subjective test that is used, and so it is to the mind of the programmer that we turn. Once again the main hurdle is whether the courts would favour such a broad interpretation of the provisions of the Copyright Act .

5. Literary Works

This category is defined in the Act to include:

"a) A table, or compilation, expressed in words, figures or symbols,(whether or not in a visible form) and

b) a computer program or compilation of computer programs."[53]

In order to fall within the definition of literary work, the work must show sufficient skill and effort to effect a conveying of information or pleasure, or instruction in the form of literary enjoyment.[54] It is clear that a computer screen display is not a computer program, and so we must look for protection under the heading literary work in its own right. It is possible that the courts would be willing to extend protection under this heading, given their willingness to do so in the past. The courts have favoured a broad interpretation of the term 'literary work', for it has been held to cover football coupons[55], newspaper columns of births and deaths[56] and catalogues of spare parts.[57] The key to protection here, is that the Act provides for copyright protection, whether or not the compilation or words, figures or symbols are visible, and so the scope for protection appears to be very broad, which along with the courts willingness to provide a wide protection appears to be good news for software developers.

An argument to encompass the user interface would be along the lines, that a computer screen display which comprises text, graphs or tables or compilations which exhibit some degree of coherence or order which has been imposed upon it by the author of the underlying program should be protected. However, it needs to be noted that recent developments in the United States have indicated that standard features which are widely used have lost their right to copyright protection, so this is bad news for the protection of features such as pull-down menus, icons and menu screens.

6. Is a Computer Screen Display made by a Qualified Person?

This is required by s32 of the Act and is defined in s32(4) as the author of the work who is an Australian citizen, an Australian protected person or a person resident in Australia. If it is accepted by members of the computer industry that computers are actually authors (that creative programs exist) then it is clear that the screen display is created by the computer and not a 'natural person' as required by the Act. This brings into play the whole issue of Artificial Intelligence and is beyond the scope of this paper, but it suffices to say that there are those in the computer industry, Jack Brown among them, who believe that creative programs exist, which, if accepted then ends the possibility of protection for computer screen displays in Australia.

No case law on this point exists in Australia, but the case of Express Newspapers [58] offers some guidance as to the position in United Kingdom. There it was held that the computer was no more than a tool which produced the work in accordance with the computer programs written by the author. If this case is accepted in Australia, there is room for argument that copyright protection should extend to computer screen displays for the computer programmer as a natural person is the author. However, if the argument that creative programs exist is accepted in relation to expert systems, where it can be argued that, although the computer is activated by human intervention, it is the computer itself which determines the output in the form of the screen display and not the operator or the programmer.[59] So protection in this area may well depend on the type of program for which protection is sought. The possibility for protection, however, cannot be ruled out.

7. Is a Computer Screen Display Sufficiently Fixed in a "Material Form"?

This is probably the biggest hurdle which has to be overcome in Australia before protection for the screen will be awarded. The requirement here, is that the screen must be in a material form, which is defined to include any form of storage from which the work can be reproduced. A screen display may be presented long enough to take a photograph of it or to effect a screen dump into a file or a printed hard copy. Thus it is in a form which can be reproduced. The real problem is the transitory nature of a computer screen display, for once the screen is turned off the screen's contents disappear. So if storage connotes a degree of permanency the screen display will not qualify.

"On the other hand, the definition of "material form" is inclusive only and it is therefore arguable that the definition can be broadly interpreted, to embrace other less permanent forms of fixation."[60] Whether the Australian courts are willing to accept such a broad interpretation of the statute is however uncertain. The CLRC Draft Report "notes that such an icon or even a complete screen display can adequately distinguish a computer program because the program can be written such that the icon is always on the screen when the program is in use."[61] Thus it appears that whilst the 'look and feel' of the user interface will be uncopyrightable in Australia, the screen display itself may still be afforded some copyright protection.

8. Practical Considerations

It is useful to step beyond the realm of theoretical analysis and look at some of the practicalities surrounding the issue of user interface protection, particularly in relation to Australia. Such practical matters incorporate the reasons for user interface similarities, standardisation, compatibility and cloning.

Turning first to the reasons for user interface similarities, we find many reasons in practice why their 'look and feel' may appear the same as another software producers. First, the organisation, structure and sequence of the program and its user interface may be similar because there are a limited number of ways to perform the task...The more programs for a particular application that are built, the more a common understanding of the most efficient structure begins to emerge.

The instructions, commands, menu languages, and displayed instructions may be similar because of the common jargon and functions of an application. For similar reasons the macro commands and syntax may not differ markedly across programs. Frequently repeated task sequences are logical candidates for aggregation into a macro command.

Textual screen displays will be similar when a standard layout of information is used for performing a particular task. Further, the contents and sequence of screen displays will be similar when users perform a standard sequence of actions on a standard representation of the information when performing the task manually.[62] All of the reasons why similarities may arise, have caused difficulties for judges in formulating tests to apply, to determine if infringement has occurred. This is the reason why the substantial similarity test keeps altering, and why the extrinsic-intrinsic test was developed. It is clear that for infringement to be found, there needs to be some form of linkage between the two pieces of software, for otherwise the idea/expression dichotomy is lost, and the similarity may be due to one of the above reasons.

The difficulties we face, and they are numerous, arise most fundamentally from the character of software as a technology. As computer scientists learn early in their education, hardware and software are essentially interchangeable. More precisely, they are what we might call behaviourally interchangeable: any behaviour we can accomplish with one we can also accomplish with the other. Any software program can be mimicked, exactly, by a piece of hardware, and the behaviour of any piece of digital electronic hardware can conversely be imitated exactly by a program.

There are two important consequences of this interchangeability. If hardware and software are behaviourally interchangeable, the choice of which to use in any given circumstance becomes what is termed an "engineering decision." That is, the choice will have no impact on what we can do, only how it will get done. The decision will of course impact things like the cost of the final product, ease of modification, and its speed. So we may well choose between them for reasons of cost, speed, or modifiability, but their fundamental capabilities are identical.

The second consequence arises when intellectual property law comes into the picture. While hardware and software are interchangeable in the technical world, notice the enormous difference in the variety of intellectual property protection available depending on which of those we chose. If the device is implemented in software, copyright protection will hold, yet the identical device implemented in hardware will qualify for the far stricter protection offered by the patent regime.

A further fact of the interchangeability of hardware and software means that whilst the underlying machinery is different, the look and feel of the user interface is the same, and this is an issue the American courts have failed to address by not recognising the distinction or clarifying where the line should be drawn. This brings up the issue of standardisation which is complementary to the issue of compatibility, so these can be dealt with together.

Why should user interfaces be copied, or appear to be copied? First, the similarity in user interfaces may be derived from the underlying application, and what is being copied is the manual sequence for performing the task rather than a unique interface organisation. Second, users want to count on a standard vocabulary and standard functions, so that they do not face extensive learning and unlearning with each new software package they embrace. Since users want to minimise the learning they must undergo, employing a non-standard interface language can become a significant barrier to market entry, once users have already learned one interface vocabulary. As a result many interface techniques have become informal industry standards.

The most pernicious argument in favour of permitting a substantial degree of copying in order to create compatible programming, is that we should encourage the adoption of industry-wide uniform de facto standards. There are two responses to that argument:

First, it seems perverse to deprive the creator of a successful computer program or interface of the benefit of copyright protection because he was the first to create a successful popular program or interface of that type. The basis of the laws is that copyright protection is appropriate to reward and thereby stimulate endeavour. Even if it is assumed that, in a given case, copyright is not necessary to induce initial creation, the harm done in stultifying the development of competing, superior works may be enormous.

Second to permit copying in order to establish de facto standards runs a serious risk of losing the benefits that flow from continuous improvements. As one interface designer stated the point,

"'Fixed Standards can seriously impede the evolution of new applications. Software development is still an art: the great breakthroughs are made by creative designers, not appliers of formulas and rules...'"[63]

The argument that providing copyright protection encourages improved versions of the software, may be countered by the argument that this will also happen if no copyright protection is afforded. A more serious difficulty is the lack of incentive to create and market new products. This argument can also be countered, for a innovator in the marketplace always captures a high percentage of the consumer market, and those who copy will capture a portion of the market, but it is the first who is on the marketplace, that captures the substantial share.

Moreover, user interfaces constitute another product of new technologies that raises the question of whether and how much protection will promote or impede progress and innovation. This may include consideration of factors like standardisation and compatibility. User interfaces assume an increasing importance in marketing software. Widespread acceptance and training costs create the danger of a "lock-in". In addition, innovation in the field of intellectual property is based on building on the efforts of others. Extended protection may multiply development costs and erect additional barriers to market entry by the need to circumvent protected elements.

Farrell argued that certain traditional features of copyright law, make it plausible that using copyright to protect computer software, may unduly harm the delicate process by which compatibility is achieved.[64] Products are "compatible" in a broad sense if they work together easily. Standardisation means making products similar enough to be compatible. Standardisation of computer software facilitates the formation and use of computer networks, the transfer of files among users and across applications, and savings in training costs.

Standards also have other, indirect, benefits. For instance, the more people use a given computer operating system, the more software is likely to be written for that system, and the more choice and competition those users will have available. In addition, entry, competition and innovation may be easier if a competitor need only produce a single better component, which can then hook up to the market range of complementary components, than if each innovator must develop an entire "system."

Standardisation has costs as well as benefits. One cost is the potential loss of variety. Although standards may contain "options" they inevitably constrain design. A second, dynamic cost of standards is that, despite their advantages for innovation they can also retard innovation. Once standardisation is achieved, it can be hard to break out of that standard and move to another, even if the original is no longer the most suitable.

Worse yet, some users may have invested a considerable amount in training themselves to use the existing standard, or in organising their files or even their business practices to conform to the existing standard. They will be reluctant to abandon that investment simply because other users think a switch would be desirable. So they dig in their heels and refuse to budge. Then the new users must choose between the benefits of compatibility and the benefits of the new standard, and often compatibility wins out-even when it should not. eg QWERTY keyboard.

The benefits discussed above directly affect the users of the technology. In a competitive market, a vendor's profits may depend more on the value of its product relative to a rival's product than on the absolute value of its product. In that case, a vendor will rationally favour standards only if the value of its own product is enhanced at least as much as is the value of its rival's. The lack of standards can "lock in" users to a single vendor, creating a degree of what has been called "ex post monopoly."

One apparently commonplace practice in the industry is the 'cloning' of programs. This involves the use of key elements or the main features of a program in another program, although no literal copying is involved. Those who support cloning say that it provides competition within the industry as well as raising the industry's standards as clones often improve upon the original programs. The result of protecting structure, it is argued, is that monopolies over ideas will be created. It also prevents new software from taking advantage of older techniques and stifles the growth of business.[65]

All of the above practical considerations need to be analysed before one can make a judgment about the applicability of American law to Australia. This is something which the American judges seem to have failed to do, perhaps because of the differences in the two markets, and thus severely weakens the argument for American protection of look and feel to be adopted in Australia.

9. The Idea/Expression Dichotomy Dilemma

One of the fundamental concepts underlying copyright is that it does not protect ideas, only the formal expression of those ideas.[66] This has long been accepted as correct and is known as the 'idea/expression' dichotomy.

" It is of the essence of copyright that it does not protect ideas, but only the way in which those ideas are expressed...It is for this reason that objective similarity is the test of infringement: if the same idea is expressed in a different way, it is not an infringement of another expression of that idea."[67]

A substantial difficulty emerged in the law of copyright for computer software, however, when it was accepted that copyright in a work can be infringed not just by direct literal copying of the precise program used, but by less precise copying or adaptation. If copyright can be infringed by something less than wholesale copying of the precise mode of expression, how does the Court determine where the idea ends and expression begins?

In the United States, the emergence of the look and feel cases has brought the whole idea/expression dichotomy to the forefront of discussions. The courts approach essentially extended protection beyond the literal elements of the program to the organisation and structure of the software. To avoid protection of the idea rather than its expression the early courts, such as Whelan have sought to characterise what is essential to the purpose of a program as the idea and what is non-essential as its expression. Thus taking from a program those elements which are essential to its purpose or function is deemed to be taking the idea and not an infringement. Taking anything which is unnecessary or non-essential to the program's purpose or function is deemed to be the taking of the expression of the idea and would constitute infringement. One problem with this approach is that if the function of a program is interpreted broadly many aspects of computer programs will be afforded protection as they will be regarded as inessential to the program's purpose. This could result in protection of the idea, which is contrary to the underlying philosophy of copyright.

"Some commentators consider that the "look and feel" line of cases has extended copyright protection too far. It would certainly seem that the idea/expression dichotomy was not maintained either in Whelan or in Softklone , where status screen outputs were deemed to be separate works. Indeed the decision at first instance in Autodesk has been described as "look-and-feel with knobs on" for failing to maintain a clear distinction between idea and form."[68]

So the idea/expression dichotomy is very important for copyright law, and is one of the reasons underlying many of the criticisms of the American cases.

It would be a mistake to view ideas and their expression as separated discrete categories, and to imagine that it is possible to forensically determine whether a programming element belongs in one category or the other. Rather, the distinction is better conceived as a continuum which runs from a pure intellectual concept through to its literal reduction to basic code of bits, or "0s" and "1s". In between is a grey area of programming elements which are part ideas, part expression, or part commonly used programming vocabulary. Where the line is drawn in that grey area is a policy and not a legal question. A balance has to be struck between an individual developer's rights to commercially exploit his or her intellectual product, and the social values in the circulation of new developments and the achievement of interoperability of computers.

"Drawing the line too liberally in favour of copyright protection would bestow strong monopolies of specific applications upon the first to write programs performing those applications and would thereby inhibit other creators from developing improved products. Drawing the line too conservatively would allow programmers' efforts to be copied easily, thus discouraging the creation of all but modest incremental advances."[69]

Thus again we come to the conclusion that policy plays an important role in providing copyright protection for the user interface. With knowledge of the difficulties in providing protection for the user interface under copyright law be can now turn to the case law and see how the judges have dealt with these problems.

10. Does Copyright Provide the Best Protection?

It is useful to determine whether copyright provides the best protection for the user interface, or whether other protection is also necessary.[70] With extension of the software industry into mass market sales, where there is no privity of contract and secrecy is hard to maintain, attention has shifted to areas offering stronger protection: patents and copyright. Because the requirements of novelty and non-obviousness for patent protection would only rarely be satisfied by computer programs, attention focused on the copyright system.

"First, since copyright is concerned primarily with the use and flow of information and information-based products and services, it is the area of intellectual property law that will be most affected by advances in communications and information technologies. Second, it is to copyright rather than to other provisions that the creators, developers, producers and distributors of new information technologies are looking in their efforts to gain legislative protection for their works."[71]

There have been a number of criticisms of the use of copyright as a means of protection for computer software and in particular the look and feel of the user interface. These have ranged from the feeling that ordinary commercial and contractual remedies are sufficient to deal with any problems to concerns about the length of protection which copyright offers. However, although other forms of protection are available in Australia through the Trade Practices Act and trade secret laws, the relative ease and speed with which one can obtain copyright protection and its low cost make it one of the best forms of protection for the user interface of a computer.

"The scepticism which was widely expressed about the suitability of copyright protection for software has been replaced with a general acceptance at the international level that copyright is the way to go. Debate has shifted to how to tailor the traditional notions of copyright to the specific features of computer software."[72]

It is accepted then that copyright law, as opposed to patent or commercial remedies, currently provides the best protection for the user interface. Whether this remains the case in light of American case law developments remains to be seen. One other important introductory area, before we move to the substantive part of the paper, concerns the distinction, in copyright law, between an idea and the resulting expression of that idea.

11. Conclusion

Whilst, ideologically and based on policy grounds, protection for the look and feel of the computer screen display in Australia seems reasonable, a number of factors serve to dispel the notion. These are:

1. It seems that many of the problems inherent in the judgments of the look and feel cases stem from a basic misunderstanding of the workings of computers. It appears 'look and feel' has been accepted as a single term. However, it is possible to create a computer program which "looks" the same as another but with a different feel, such that,

a) Text is different,

b) Visual behaviour is identical,

c) Operational behaviour is different.

Such possibilities have been ignored by judges and show that protection was afforded by the judiciary at a time when there was little know-how in the workings of computers. The ramifications of judges ignoring the differences between a program's behaviour and its resulting look and feel have been outlined above and call for a critical analysis of the US cases which have developed the notion of look and feel.

2. The fact that hardware and software are interchangeable has also been ignored by judges.

3. Look and feel tends to delineate the idea/expression dichotomy.

4. There are many reasons for user interface similarities.

5. Australia is not a major innovator in the software industry, Australian software manufacturers tend to produce compatible software to operate with those programs brought out by its competitors. Therefore to copyright the look and feel to the extent to which it was given in Lotus would be to undermine the Australian software industry. Given Australia's current economic climate such a move is both unnecessary and unwarranted.

"Protection of the non-literal elements of a program would be particularly unsuited to the infant Australian software industry. While the Australian industry is undoubtedly vigorous and creative, the reality is that most significant developments occur overseas and many Australian software houses are engaged in developing software which can operate in conjunction with software and with standards developed by the large foreign developers."[73]

6. America, with many of the world's leaders in the software industry, Apple v Microsoft and Broderbund to name a few, has more of a vested interest in protecting the look and feel of the computer screen display. This could be why judges interpreted their statutes more broadly and also why protection was extended on policy grounds, ie. the economic power of the developers to lobby government.

7. America has recently been moving away from such broad protection, as shown in the cases of Computer Associates and Apple v Microsoft.

8. America has audiovisual copyright and Australia has none.

Given these factors, it becomes obvious that no new legislation should be introduced to protect look and feel in Australia, nor should existing statutes be interpreted broadly to provide protection. We should learn from the American experience that look and feel protection is beyond the scope and intention of copyright law and has no place in Australian copyright law.

12. Appendix One: Does Copyright Provide the Best Protection?

The critics start with the thesis that '[c]omputer programs are a technology' and 'too much [a part] of a machine to fit comfortably into the copyright system' although the argument also is made that 'they are too much of a writing to fit comfortably into the patent system.'[74] There are several reasons however why copyright is particularly attractive to producers of computer programs. Because it allows for non-contractual control by securing an intangible intellectual property right, it has potential to control distribution in a mass market. Protection is not conditional on satisfying high standards of inventiveness or even of originality, provided a program is actually created independently of its author, and copyright has an effective scheme of international protection.

Clapes felt that under the American constitutional mandate, no real change in legal protection seems appropriate or necessary. In America software is protected by: patents and trade secrets for innovative developments, copyright to some extent (prohibiting "free riders"), and trademark law when potential public confusion as to the origin of the software is present. However, American and Australian copyright laws are different and so we must turn to other commentators to determine the issue conclusively.

It should be noted however that copyright is not the only possible means of protection for programs. Other methods, technical, commercial and legal, are available and are used, but all have some disadvantages. Overall the best method of protecting any particular item of software always depends upon selection from the range of available measures. Non-legal methods are: copy-inhibiting devices-liable to be cracked and prevents legitimate buyer making back-up copies, provide only object code versions to buyers-prevent non-expert users from reading but can have copies made and the object to source code. Commercial tactics: regular release of improved versions of the program-only for complex programs with a long lifespan.[75] None of the non-legal devices is entirely satisfactory, so legal means of protection are sought.

In Australia, the Government accepts that software should be dealt with under copyright law "since the essential need is for protection against copying". [76] The Green Paper accordingly proposes that new legislation should explicitly provide that computer programs will attract protection under the same conditions as literary works. Intellectual property in Australia can be protected by copyright, patents and confidential information(trade secrets). The law recognises that original and creative work will not be encouraged if the product of an inventor or author may be used or exploited by those who have put neither time nor money into the work of creation or of development. But the law also recognises that the public interest will not be served by the creation of long-term monopolies and for this reason time limits are imposed on the rights conferred in this way.

It is noted by Taggart, that copyright offers significant protection to the software developer because it does not have to be registered(therefore no cost is involved and time-consuming litigation is avoided), and anyone who copies, duplicates or otherwise infringes may be civilly or criminally liable.[77]

"Copyright law provides the best currently available means of encouraging software development and public disclosure because such protection is easily obtained at a minimal cost."[78]

In particular copyright is suited for mass-produced and mass-marketed products and immediately follows any enhancement of the work. There is a limited monopoly on copyrightable software and therefore an incentive to publish and possibly distribute it, which leads to less inhibition on the exchange of information and the growth of competition in that area of the industry. Within the context of ordinary legal measures, copyright, patent and confidential information (trade secret) offer a reasonable scope of protection for computer hardware and software. Protection may also be obtained under the Design Act for hardware and under the Trade Practices Act for both hardware and software.

As technology becomes more economically important there will be increasing pressure on the courts to provide adequate protection for technological developments. One of the biggest problems currently is the inconsistency of protection offered in various countries....There is obviously a need to access ideas to provide encouragement for future technological development, yet this should not be at the expense of potential economic profit. Computer technology is a very recent development, perhaps it needs its own set of rules.

Many software developers rely on copyright protection as the most accessible form of legal protection because of the limited availability of other forms of legal protection for software. The extent to which copyright law protects more than the literal instructions (code) of a program, however, is uncertain, and software developers presently cannot rely on copyright protection alone. Greater certainty in the scope of copyright protection is necessary to promote the continued growth of the software industry.

In turning to other areas of the law for protection however, it needs to be recognised that trade secret protection is based upon restrictions on the flow of information and is gradually being regarded as an anachronistic method of maintaining the protection required for the ever-increasing flow of new technological ideas. So while it offers good protection per se to hardware and software, trade secrecy has limited application due to the inherent problems of maintaining its level of protection in face of the many potential and often inadvertent or accidental hazards which could render the protection useless.

Given the rather favourable state of precedent on pre-emption, software industry interest in trade secret protection will continue. In light of all the positive precedents on dual protection, those placing primary reliance on trade secrets will increasingly seek to enhance protection by also relying on copyright. Moreover, those who rely primarily on copyright will more frequently attempt to cover the perceived "holes" by securing desirable elements of trade secret protection. So whilst trade secret laws will protect computers to some extent it is the favourable aspects of speed and cost which make copyright the area we must turn to examine the extent to which the user interface is protected.

13. Appendix Two: The Idea/Expression Dichotomy Dilemma

Much uncertainty surrounding the legal protection for computer programs is a result of the fundamental copyright principle that copyright law protects only the expression of an idea, and not the underlying idea itself. The means to separate idea from expression is the test whether the idea can be expressed in a variety of ways or in only one or a few ways in which case the expression is said to merge with the idea. This ensures an individual's right to profit from the product of his or her intellectual efforts while also enriching the store of ideas which is available to all.

"It is a familiar copyright principle that substantial similarity does not require a finding of infringement where there are only a few ways of expressing an idea and, indeed, in the theoretical case '[w]hen the idea and expression coincide, there will be protection against nothing other than identical copying of the work.'"[79]

While the ideas/expression dichotomy is readily stated and understood in the abstract, it is difficult to determine where exactly to draw this line through the complex hierarchy of programming elements which make up a computer program. In distinguishing between the idea and the expression in a computer program, courts and commentators have recognised that limiting copyright protection only to the programming code in a computer program will not ensure proper monetary incentives for program developers. Consequently, courts have begun to protect additional elements in computer programs such as the structure and organisation of programs. Although courts recently have begun to address the issue of protectable expression in computer programs, the scope of protection for computer programs remains undetermined. The so-called "look and feel" cases apply to software the notion of generalised(rather than specific) similarity which is familiar in connection with novels and plays. Insteads(sic) of looking for similarity of plot, incident and characterisation, they look for similarity of aspects of the screen appearance and user interface - the look and feel of the programs. These can be similar without there being a corresponding similarity of the code which comprises the programs.

Following the lead of Whelan, SAS, Johnson and Williams courts, the federal judiciary should continue to expand the scope of protection for computer programs to encompass programs' user interface, structure, and organisation, and thereby promote Congress' desire to foster creative and intellectual achievements in computer programming through the copyright laws. The judicial extension of copyright protection to these elements of computer programs is consistent with the historical application of copyright law to new technologies.

In Whelan the court defined the expression of the idea in a computer program as "the manner in which the program operates, controls and regulates the computer in receiving, assembling, calculating, retaining, correlating, and producing useful information either on a screen, print-out or by audio-communication'. The unprotected idea was no more than "the mere idea or concept of a computerised program for operating a dental laboratory." While this reflects current law, this expansion of protection has been heavily criticised both as extending protection too far to the underlying function of a program and as to the danger of manipulating the scope of protection resulting from the definition of the program's function.

A substantial difficulty emerged in the law of copyright for computer software, however, when it was accepted that copyright in a work can be infringed not just by direct literal copying of the precise program used, but by less precise copying or adaptation. If copyright can be infringed by something less than wholesale copying of the precise mode of expression, how does the Court determine where the idea ends and expression begins?

The scope of protection provided for user interfaces accordingly will have a major impact on the economic development in this software field, including considerations of standardisations and compatibility. Policy consideration suggest a rather limited scope of protection. In copyright law, as the main type of protection, this concerns the delineation between the idea or contents and the expression as well as the exclusion of functional elements that limit the possible range of expression. Based on an understanding of a user interface a as work separate from the program, the copyright principles discussed in American law considerably limit the protection available for user interfaces.

If the unprotected idea is defined in this limited and generalised fashion, then it necessarily follows that almost every element of the program implementing the idea will be copyrightable expression. The extraordinary depth of protection which flows from this approach is well illustrated by the recent decision in Lotus v Paperback where his Honour held that the overall structure, the order of commands in each menu line, and the choice of letters or words to represent commands, were all "capable of being expressed in many if not an unlimited number of ways, and that the command structure of Lotus 1-2-3 is an original and non-obvious way of expressing a command structure.

"The lesson of the Lotus decision is that programmers should not mimic common elements of a successful program, even if it has become the de facto industry standard."[80]

It becomes apparent then that the idea/expression dilemma has serious repercussions for the industry as a whole and needs to be clarified at the outset.

The exact line of demarcation remains unclear, however. This was noted by Judge Keeton when he said that drawing the line between idea and expression tends to be ad hoc in character.[81] Thus, the use of trash can icons may be regarded as one of several possibilities of expressing the delete function. It may also be regarded as an idea, however, leaving the possible number of expressions very limited.

Other considerations have to be included. There is only a limited number of input methods and users get accustomed to specific commands. The requirement of inputting the same data in a logical way may necessitate the selection of similar input data and sequence of screens and limit the possible range of variations, perhaps to the field of decorative graphics.

Moreover, "human factor analysis" has established itself as a scientific discipline analysing the user/computer-communication and Menell mentions several design objectives that are widely being pursued and may limit the variety of expressions. All this may mitigate in favour of very limited copyright protection for user interfaces. As we have already found in favour of copyright as founding the best type of protection for the user it remains to be seen if the idea/expression distinction has been upheld.


[1] Tutor in law, University of Tasmania at Launceston; B.Ec. LLB (Hons) University of Tasmania. I would like to thank Daniel Hunter and Anne Fitzgerald for their suggestions and advice in writing this paper.

[2] Copyright Law Review Committee, Draft Report on Computer Software Protection, A.R. McLean Printing, ACT, 1993.

[3] Curtis, B., "Engineering Computer "Look and Feel": User Interface Technology and Human Factors Engineering" Jurimetrics Journal 30 (1989) pp76-77.

[4] The Copyright Law Review Committee noted from the submissions it received that the user interface is "the distinguishing feature as far as most, if not all, users are concerned."

[5] Ibid pp52-53.

[6] Rudnick, R.E., "Manufacturers Technologies, Inc v Cams, Inc.: A False Hope for Software Developers seeking Copyright Protection for their Generated Screen Displays" Rutgers Computer and Technology Law Journal 17 (1991) p211.

[7] Curtis, B., op. cit., pp 76-77.

[8] Whelan Associates, Inc v Jaslow Dental Laboratories, Inc 609 F Supp 1307. Elaine Whelan, a software contractor created a program for Jaslow Dental Laboratory which managed the accounting, inventory and billing for dental laboratories. Whelan retained copyright in the program and registered it with the Copyright Office as a literary work. The program ran on an IBM minicomputer and Jaslow translated the program to run on the IBM personal computer. Jaslow sued Whelan under State law for trade secret appropriation and Whelan retaliated in copyright. The court found infringement on the basis that the structure, sequence and organisation is copyrightable and that substantial similarity of these features is sufficient to constitute a breach of copyright.

[9] Davis, R., "Intellectual Property and Software; The Assumptions are Broken" Artificial Intelligence Laboratory, USA, 1991.

[10] Ibid, p113.

[11] Ibid, p115.

[12] Ibid, p114.

[13] Ibid, pp113-115.

[14] Broderbund Software, Inc v Unison World, Inc 648 F Supp 1125. The plaintiff was the copyright owner of a menu driven program, originally designed for use on Apple computers, that permitted its users to create customised greeting cards, signs, banners and posters which had allegedly been infringed by an IBM-version of the defendant. The court found infringement on the basis of the similarities in the sequencing of the screens and their layout, the choices presented to the user, in short the "total concept and feel" of the program.

[15] Stern Electronics Inc v Kaufman [1982] USCA2 54; 669 F 2d 852. In this case Omni, a co-defendant, sold games which were identical to Stern's copyright game called Scramble. Stern had registered audiovisual copyright and not literary work copyright because a given display can be generated by numerous different computer programs. The court decided that there was infringement and importantly that any interaction by the user of a computer which changes the display will not invalidate the copyright.

[16] Whelan, op. cit., 1307.

[17] Digital Communications Associates, Inc v Softklone Distributing Corporation 2 USPQ 2d 1385. This case concerned the purely textual status-screen of a command driven data communications program. The upper half of the screen reflected the present status of the operation parameter while the user could change the settings by inputting one of the commands listed in a window in the lower half. The status screen was registered as a compilation of program terms. The defendants produced a similar program with a mostly identical screen and a largely identical command language. The court concluded that several aspects of the status display such as highlighting of command symbols, underlining, or the use of brackets were not dictated by functional considerations, and were therefore copyrightable subject matter.

[18] Lotus Development Corp v Borland Int'l Inc, DC Mass, No. 90-11662-K 7/31/92. Lotus publishes an electronic spreadsheet program for personal computers called Lotus 1-2-3 and alleged that the defendant Borland breached copyright with its Quattro spreadsheet program. The Quattro menus differed from those of 1-2-3 in content and appearance, but the program can also operate in a mode that imitates the 1-2-3 menu system and so the court found infringement.

[19] Editorial, Computer Law and Practice, The Journal of Computer and Communications Law, Vol. 7, Nov.-Dec. 1990.

[20] Computer Associates Int'l Inc v Altai ( Gates Rubber Co v Bando American Inc) DC Colo, No. 92-s-136, 6/24/92. Judge Sparr found 1) Identical constants and near verbatim install files 2) Substantially similar menus, formulas, data flow, control flow and behaviour 3) Similar structure and organisation in design models and 'look and feel' in complexity level. 4) Different source code and programming style.

He found that the only explanation for findings one and two were that copying was evident. Furthermore he found that while the appearance of the screens was different the content and method of proceeding through calculations were quite similar, as was the overall operation of the two programs.

[21] Davis, op. cit., p101.

[22] Computer Associates Int'l Inc v Altai DC Colo, No. 92-s-1136, 6/24/92.

[23] Fitzgerald Anne, "Square Pegs and Round Holes: Recent US Developments in Copyright Protection for Computer Software" Journal of Law and Information Science, The Faculty of Law, University of Tasmania, Vol. 4, No. 1, 1993 p154.

[24] Apple Computer, Inc v Microsoft Corporation and Hewlett-Packard Company, 1992 U.S. Dist. LEXIS 5986. Apple sued Microsoft on the basis that Windows 2.03 program and Hewlett-Packard's New Wave copied the "look and feel" of the Macintosh's visual displays. The graphic user interface generated by the Macintosh system software consists of windows, icons, pull-down menus, and other images or visual displays projected on the computer screen. The defendants relied on lack of originality as such to warrant copyright protection.

[25] Stone Peter and Groves Peter, "Recent American Case Law on Copyright in Computer Programs" [1993] 9 CLSR p103.

[26] John Richardson Computers Ltd v Flanders and Chemtec Ltd ChD 19 February 1993. The case concerns a former employee (Flanders) and past consultant of JRC who modified a program of JRC's and having formed his own company, Chemtec, proceeded to market it in the UK. JRC brought an action for breach of copyright as they felt Flanders' program was "a blatant copy of all JRC features, but attempts had been made to camouflage this by using drop down menus and boxes".

[27] Delrina Corporation carrying on business as Carolian systems v Triolet Systems Inc. and Brian Duncombe (1993) 47 Canadian Patent Reports (3d) 1. Duncombe was an employee of Carolian and in the course of his employment he developed a computer program called Sysview. Upon leaving the employment of Carolian he founded his own company and marketed a program called Assess 3000 which, whilst similar in function, appearance and operation to Sysview was written in a different computer program language. Carolian's application for an injunction to restrain Duncombe from marketing Assess 3000 was denied by Justice O'Leary due to the non-literal copying of the program. Justice O'Leary set out the following principles:

1. An author has no copyright in ideas or information, but only in his expression of them.

2. Copyright subsists in original literary works. there is no copyright in what the author has copied from something already in the public domain or from a work in which another holds the copyright.

3. Even if the expression originated with the author, the expression of the idea is not copyrightable if the expression does no more than embody elements of the idea that are functional in the utilitarian sense.

4. If an idea can be expressed in only one or in a very limited number of ways, then copyright of that expression will be refused for it would give the originator of the idea a virtual monopoly of the idea.

5. Copyright does not subsist in any 'arrangement, system, scheme, method for doing a particular thing, procedure, process, concept, principle, or discovery, but only in an author's expression of them...therefore a particular expression of a procedure for solving a problem may be protected by copyright but the procedure as such cannot be protected by copyright.'

[28] Bull, G. "UK Case Commentary: John Richardson Computers Ltd v Flanders and Chemtec Ltd [1993] 9 CLSR p107.

[29] In Computer Associates three steps are recommended in the application of the 'abstractions test':

1. Discovery of the plaintiff's program's abstraction levels;

2. Filtration of these abstractions in order to discover a 'core of protectable material', such process of filtration to identify and exclude:

a) Elements dictated by efficiency ie. only one way of expressing an idea;

b) Elements dictated by external factors eg. particular facts can only be described in a particular way;

c) Elements taken from the public domain eg. expressions standard or commonplace in the computer industry and;

3. Comparison of what is left of the plaintiff's program following stages 1 and 2 with the program said to be an infringement of the plaintiff's program.

[30] Bull, G., op. cit., p 109.

[31] Ibid, p112.

[32] Kyer, C. Ian, "Canadian Court Adopts US Approach to 'Look and Feel'" [1993] 9 CLSR p174.

[33] Ibid, p174.

[34] Stone Peter and Groves Peter, op. cit., p99.

[35] Apple Computer Inc. v Computer Edge Pty. Ltd. (1983) 50 ALR 581. In this case Beaumont J held that computer programs were not classifiable as "literary works" under the current legislation.

[36] Computer Law in Australia, A Collection of Essays, edited by John Taggart, Legal Books Pty. Ltd., 1989, p20.

[37] Autodesk, Inc and Anor v Martin Peter Dyason and Ors (1992) 66 ALR 233. Autodesk owns the copyright in Autocad, a program to assist in the drafting of architectural and engineering plans and devices. To prevent unauthorised copying of the program Autodesk produced a lock which is a physical device that must be plugged into the serial port of the computer if the program is to be operated. The defendant Peter Kelly, an electronics designer developed an alternative device to the AutoCad lock which was marketed by Dyason under the name "Auto-Key". Kelly did not dismantle the lock to find how it works but observed its operation and developed the Auto-Key from those observations. The Full Court of the Federal Court found that the AutoCad hardware lock was protectable under the Copyright Act, not as a computer program but as an integral part of a set of instructions written into the AutoCad program itself.

[38] Copyright Law Review Committee, Draft Report on Computer Software Protection, A.R. McLean Printing, ACT, 1993, p6.

[39] Ibid , p105.

[40] Ibid , p120.

[41] Ibid , p119.

[42] Ibid , p120.

[43] Ibid , p118.

[44] Ibid , p121.

[45] Waters and Leonard, Op. Cit., p126.

[46] Mackay, J. and McLachlan, J., "Are Computer Screen Displays and their "Look and Feel" Capable of Copyright Protection?", Blake Dawson Waldron Reporter, p28.

[47] Brown, J.E., "The current status of copyright protection for computer software and some patent protection parallels" Computer Law and Practice, No. 17, Vol. 6, 1990, p175.

[48] Mackay, J. and McLachlan, J., op. cit., p29.

[49] Merchandising Corporation of America Inc v Harpbond Ltd [1983] FSR 32 at 46.

[50] Merlet v Mothercare plc (1984) 2 IPR 456.

[51] Cuisenaire v Reed [1963] VicRp 96; [1963] VR 719.

[52] Ibid p730.

[53] Copyright Act s31.

[54] Exxon Corporation v Exxon Insurance Consultants Ltd [1981] 2 All ER 495.

[55] Ladbroke Football Ltd v William Hill Football Ltd [1964] 1 All ER 465.

[56] John Fairfax & Sons Pty Ltd v Australian Consolidated Press [1960] SR (NSW) 413.

[57] Purefoy Engineering Co Ltd v Sykes Boxall & Co Ltd [1955] 72 RPC 89.

[58] Express Newspapers plc v Liverpool Daily Post & Echo [1985] FSR 306.

[59] Mackay, J. and McLachlan, J., op. cit., p33.

[60] Ibid, p34.

[61] CLRC, op. cit., pp122-123.

[62] Curtis, op. cit., p75.

[63] Ibid, p77.

[64] Farrell, J., "Standardization and Intellectual Property" Jurimetrics Journal 30 (1989) p36.

[65] Copyright Law Review Committee, p14.

[66] See Appendix Two.

[67] Burnside, op. cit., p142.

[68] McKeough and Stewart, Intellectual Property in Australia, Butterworths, Sydney, 1991, p177.

[69] Menell "Scope of Copyright Protection for Programs" 41 Stan L Rev (1988) pp1047-48.

[70] See Appendix One

[71] Clapes, A., Software, Copyright and Competition, Greenwood Press Inc., Connecticut, 1989, p14.

[72] Waters, P. and Leonard, P., "The lessons of recent EC and US developments for protection of computer software under Australian Law" The Australian Computer Journal, Vol. 22, No. 4, Nov. 1990, p123.

[73] Waters and Leonard, op. cit., p128.

[74] Brown, op. cit., p171.

[75] Gaze, B., Copyright Protection of Computer Programs, The Federation Press, Sydney, 1989, p16.

[76] Lahore, Dworkin and Smyth, Information Technology: The Challenge to Copyright, Sweet and Maxwell, London, 1984 p103.

[77] Computer Law in Australia, op.cit., p27.

[78] Rodan, A.G., "Protecting Computer Software: After Apple Computer, Inc. v Franklin Computer Corp., Does Copyright Provide the Best Protection?" Intellectual Property Law Review (1986) p435.

[79] Brown, op. cit, p171.

[80] Waters and Leonard, op. cit., p125.

[81] BNA's Patent, Trademark and Copyright Journal, News and Comment, p363.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/JlLawInfoSci/1994/8.html