Software Patents Pending?

Robin Widdison
Director
Centre for Law and Computing
University of Durham [ 1 ]
Robin.Widdison@durham.ac.uk

This paper is a revised version of a paper presented at:
  The 15th Annual BILETA Conference
13-14 April 2000
University of Warwick.

Abstract

In two very recent decisions, the European Patent Office's Technical Board of Appeal has shifted its stance on software patents somewhat. This article explores the background to this shift and suggests that the more liberal attitude towards software patents enshrined in the Trade-Related Aspects of Intellectual Property Rights Agreement (TRIPS) is already being adopted ahead of formal amendment of the European Patent Convention and attendant national legislation. In the light of this altered attitude, the article goes on to consider what the legal, economic and political implications may be. In conclusion, the article puts forward a longer-term scenario for global software protection.

This is a Refereed Article published on 31st October 2000.

Citation : Widdison R, 'Software Patents Pending?', 2000 (3) The Journal of Information, Law  and Technology (JILT). <http://elj.warwick.ac.uk/jilt/00-3/widdison.html>. New citation as at 1/1/04: <http://www2.warwick.ac.uk/fac/soc/law/elj/jilt/2000_3/widdison/>

1. Introduction

1.1 Curiouser and Curiouser!

Software is the most changeable and mercurial of substances. It is like something out of Lewis Carroll's Alice in Wonderland . Before we explore this idea further, though, we need to be clear about what we mean by the term 'software'. Here, I use the term to refer solely to computer programs i.e. instructions that tell computer hardware what to do. I exclude for the purposes of this article data associated with the programs, physical media whether in the form of tapes or disks, and peripheral information such as help files, user manuals and other documentation.[ 2 ]

Our understanding of the role and importance of software has changed dramatically over time. Once, we lived in a hardware-orientated world. Computer programs were simply what you told your all-important computer to do. Typically, such programs were written by the hardware manufacturer and arrived bundled with the hardware. Next, software came to be seen as something separate from, albeit still subordinate to, hardware. Such 'unbundled' software could be bought from software houses that were independent of the hardware manufacturer. Later, programs were perceived as not only separate from hardware, but of equal importance to it. Now, we have reached the opposite pole to that from which we began. We live in a software-orientated world where hardware is little more than what you choose to run your favourite software on.

A substantial proportion of information can be digitised. It seems likely that soon, for example, any type of information that can be perceived, i.e. seen, heard, felt, smelt, or tasted, will be reducible to a collection of bits. Once in digital form, the information in question can be stored, processed and displayed by a computer. Furthermore, it can readily and rapidly be transmitted from one computer to any other computer regardless of distance, political frontiers, and physical obstacles. From a digital point of view, there is absolutely no distinction between text, sounds, graphics, photographs, music, animations, videos...and software. To include software in our list, however, may seem rather odd. Why is this so? Computer programs are, after all, only digitised information like the other examples. However, there is one vital difference between software and the rest. While text, sounds, graphics etc. are generally passive in nature, software, by contrast, is essentially active. It is information with attitude! As Hart reminds us, 'Computer programs are not only texts: they also behave'.[ 3 ] Software is simultaneously both form and substance, both symbolic and functional, it both 'is' and 'does'. For this reason, software is conceptually very challenging.

It is useful to make one other point about the curious nature of software at this stage. It is a superb mimic. In particular, it has an extraordinary capacity to emulate hardware, thus frequently making itself interchangeable with that hardware. So, any given solution to a problem may well be implemented as either a combination of hardware and software, or as software alone. The significance of this characteristic will become apparent later.

1.2 Intellectual Property Rights

Intellectual property law comprises a collection of rights granted over the creations of the human mind. Commonly, these rights give the creator a limited monopoly over the use of his/her creation for a prescribed period of time. Such intellectual property rights can usefully be divided into two great families of regimes - 'Two households, both alike in dignity...'! Broadly speaking, the first family protects rights in aesthetic creations while the second family seeks to protect rights in industrial property.

Let us look more closely at the first family. Authors of aesthetic creations such as literary works, musical compositions, dramatic scripts, works of art and films are all protected by copyright. In Europe, such protection typically extends for a period of 70 years after the death of the author. Also protected through the copyright regime are producers of derivative works, essentially technologically advanced ways of embodying aesthetic creations. Derivative works include sound recordings, terrestrial or satellite broadcasts, and cable programmes. These works are generally protected for a period of fifty years from creation. A related right protects investment in the contents of databases for a period of fifteen years. What is the social purpose of this type of protection? It is argued that its main purpose is to encourage and reward original, creative works.

Turning now to the second family - industrial property. The main social purpose here, by contrast, is said to be the protection of investment in useful new technology, thus providing an incentive to finance research and development. So, for example, developers of novel, inventive, industrial products and processes are protected by the patent regime for renewable periods up to a maximum of twenty years. So too, creators of industrial designs are protected by registered or unregistered design right for various periods. A third branch of this family, albeit with a rather different social purpose, protects both owners of registered trademarks and geographical indicators such as 'champagne'. This third type of industrial property can be protected indefinitely.

1.3 Which Family for Software?

If we have to determine which of the two families of intellectual property rights - copyright or industrial property, is more appropriate to protect software, we immediately hit a problem. As we have already seen, software is both symbolic and functional in nature. In symbolic, 'textual' mode, programs should surely be protected by copyright as a type of aesthetic creation such as a literary work. However, in functional, 'behaving' mode, it appears more appropriate to protect them as a species of industrial property. From a detached and theoretical point of view, it would seem to follow that software ought to be protected by both families. How, in fact, is it protected in Europe? We will look at that question next.

2. The 'Either/Or' Mindset

2.1 Protection by Copyright

When the issue of intellectual property protection for software first emerged as one of key importance, the dominant intellectual property mindset was exclusionary. New technologies were to be protected either by the copyright family or by the industrial property family.[ 4 ] At the time, this made sense. After all, at this early stage nothing had been developed which had a strong claim to belong to both intellectual property families simultaneously. So, it seemed to follow naturally from this stance that computer programs should be protected either as aesthetic creations or as industrial property, but not both. Which, then, was the more appropriate type of protection? With only two possible options, it is not surprising that there were two main schools of thought. The first school favoured protecting software as an aesthetic creation by means of copyright. The second school, by contrast, preferred protecting software as a type of industrial property. From the outset, it seems to have been taken for granted by this second school that the right approach would be to protect software as a patentable invention.

Over time, the copyright family began to emerge as the preferred means of protection in Europe. Why should this be so? Here are four suggested reasons:[ 5 ]

• When clients displayed software to lawyers, they tended to show it in symbolic rather than functional mode. A source-code listing looks like a literary work and this may well have coloured early legal understanding of the concept of software.
   

• Generally, copyright protection required no registration. Thus, there would be no need to develop or adapt an examination system in order to subject software to demanding, high-threshold tests for registration.[ 6 ]
   

• Copyright had long before become a conceptual dustbin for new technological forms of representation as they were created.[ 7 ]
   

• Copyright had developed a high degree of global harmonisation at an early stage. From the outset, software was clearly seen as a phenomenon of international significance.

On this basis, it appeared logical to classify source code, i.e. the symbolic representation of software in human-readable form, as a type of aesthetic creation such as a literary work. Object code, the symbolic representation of computer programs in machine-readable form, created more difficult conceptual problems. Insofar as object code could be seen as a translation of source code into another language, however, the fact that copyright readily protected translations from e.g. English into French seemed to provide a sufficiently close analogy to allow a machine-readable version of a computer program to be protected in the same way as the human-readable version.

If the European solution to the either/or conundrum was to be protection by means of copyright, it followed that the other leading contender - the possibility of patent protection, should be suppressed. This was effectively achieved at continental level by means of the European Patent Convention 1973 (The Convention), a provision designed to harmonise patent law in much of Europe and to facilitate the acquisition of a bundle of European national patents by means of a single application. By virtue of article 52(2) of the Convention, computer programs were deemed not to be inventions and therefore were unpatentable. As signatory states to the Convention were required to ensure that their domestic law was in step with the terms of the Convention, this exclusion percolated down into national laws.[ 8 ]

The rationale that was adopted for excluding software from the patent regime can perhaps be gleaned from the company that it keeps in the Convention. Computer programs are lumped together with 'schemes, rules and methods for performing mental acts, playing games or doing business' all of which are also deemed to be unpatentable. Why should this be so? The exclusions appear to be based on the notion that all the methods in question are abstract, intellectual constructs rather than useful, 'down-to-earth' products or processes. As Gall argues:

While such abstract constructs might well qualify for copyright protection once set down and recorded in some way, they were all deemed to be unpatentable. Interestingly, though, the exclusions from patentability were not absolute. Article 52(3) provided a loophole. It declared that the exclusion from patentability applied only to software 'as such'. So, for example, provided that a patent application was for something more than, or different from a computer program as such, the application would be considered in the normal way. Unfortunately, though, the key expression 'as such' was not further explained in the Convention. So it was unclear, at first, how large the loophole was.

2.2 VICOM and the Aftermath

In due course, the loophole came to be considered by the Technical Board of Appeal of the European Patent Office (EPO) in VICOM's Application .[ 10 ]That case concerned hardware with a computer program which, together, would digitally process images in a sophisticated way. It was conceded by the applicant, however, that the hardware used could just as easily be a conventional computer. None-the-less, the Board stated:

Looked at holistically, i.e. hardware and software combined, the Board was of the view that the claimed invention escaped the 'as such' exception and so was potentially patentable. There was a difficulty, however. In arriving at this result, the Board substituted one obscure concept, 'technical contribution', for another , 'as such'. Subsequently, the European Patent Office (EPO) itself tried to clarify the meaning of this new concept in its official guidelines. An early version of the guidelines gave the following examples:

The underlying thinking seemed to be that where software had an impact on the external world, e.g. an industrial machine or process controlled by a program, then the whole system, taken together could potentially be patentable. This seems both clear and logical, half an answer to the question of the patentability of software. But, what about the other half of the answer? Could a program that only had an internal effect on a computer be patentable? The guidelines went on to say:

This extract seems to be rather less helpful. Indeed, the continuing uncertainly over the patentability of software with an internal effect on a conventional computer led to a prolonged paper chase through the decision making bodies of both the EPO and signatory states in search of a meaning for this new concept.[ 13 ]

2.3 The IBM Twins

Twin decisions of the EPO Technical Board of Appeal, both involving applications by IBM,[ 14 ] may now have taken us a long way towards the other half of the answer, at least within the framework of the Convention as it stands at present. In relation to software making a technical contribution to the internal workings of a computer, the Board has in these cases made it clear what sort of technical effect it has in mind. Mere changes in the physical state of hardware by e.g. the manipulation of electrical currents are not enough. What is required is something else, 'a further technical effect'. What does this third obscure concept mean? The Board goes on to give examples:

Where are we now? The overall effect of both VICOM and the IBM twins may well be to complete a de facto reversal the software exclusion enshrined in article 52 of the Convention. Tapper certainly thinks that this is so. He comments: 'It is astonishing that the explicit provision of the Convention denying protection to computer programs have been read into virtual non-existence...'.[ 16 ] The original stance taken by the Convention seemed to be that software as such was presumed to be unpatentable unless the presumption was rebutted by evidence of something more. Now, it appears that the modern position is this. Software is presumed to be patentable unless that presumption is rebutted by evidence that the application is for software in isolation, a computer program as such . While this may appear to be no more than a change of emphasis, as any practising lawyer will know, it is an enormously important one. What is the basis for this assessment of the current situation? I believe that the EPO is now adopting an inclusionary mindset. I shall now examine why this should be so.

3. The 'And' Mindset

While Europe wrestled with concepts like 'as such', 'technical contribution' and 'further technical effect', the position in other key parts of the developed world evolved in a rather different way. In the United States, for example, the authorities seemed to have moved much more rapidly from the exclusionary, either/or mindset to an inclusionary, 'and' mindset. As a result, software patents have been available much more readily in the United States than is the case in Europe for some time [ 17 ].This would be only of passing interest if it were not for the fact the United States is:

It seems likely that the United States used its dominant position to have a decisive influence over key provisions of the Trade-Related Aspects of Intellectual Property Rights Agreement 1994 (TRIPS), an agreement designed to form a basis for the global harmonisation of intellectual property laws. At first glance, however, TRIPS looks rather tame in that it appears to adopt a traditionally exclusionary approach. The agreement expressly states that software is copyrightable. Article 10(1) states:

When we come to the section on patentability, though, Article 27(1) says:

From our study of the European Patent Convention, we might well expect the above mentioned paragraphs 2 and 3 to contain an exclusion for software. However, when we read these paragraphs, we find that they only exclude immoral inventions, medical treatments and plant or animal varieties. There is absolutely no mention of software. Indeed the declaration in Article 27(1) that 'patents shall be available for any inventions...in all fields of technology...' clearly supports a view that computer programs have the same potential for patentability as any other product or process. In essence, TRIPS turns out to be the expression of a strongly inclusionary mindset.

This inclusionary TRIPS approach is now set to dictate the global nature of software protection, at least for many years to come. It is already having a significant effect at least with regard to software patents. In the twin IBM decisions, The EPO Technical Board of Appeal accepts with obvious regret that - as the EPO itself is not a signatory to TRIPS - it cannot unceremoniously dump the Convention-based software exclusion on the spot. The Board is, however, clearly straining at the leash to minimise the effect of the exclusion within the terms of the Convention against the day when the software exclusion is removed from the Convention in line with TRIPS expectations. Consider this comment by the Board:

The Board itself was ultimately constrained by the terms of the Convention. However, pressure is now growing to bring the Convention into line with TRIPS by amending Article 52(2)[ 20 ].The most striking change, of course, would be to take software out of the excluded category and place it firmly in the main stream of potentially patentable inventions. That is not all, though. The European patent has a younger sister - the Community patent. The Community patent was born in Luxembourg in 1975 [ 21 ].While the European Patent Convention enables inventors to obtain a bundle of national patents by means of one application, the Community patent regime, by contrast, is designed to provide a single, unified European Union-wide patent. However, like Sleeping Beauty, the Community patent has been in a deep coma for decades. It has never been implemented. Recently, though, there have been moves by the EU Commission to awaken the Community patent scheme [ 22 ]. If this does occur, it now looks highly likely that this regime too will treat software as potentially patentable in line with the TRIPS formula [ 23 ]. Is it only a matter of time, therefore, before the European approach to the protection of software is fully inclusionary - until both the copyright and patent regimes are freely available to authors or owners of new and inventive computer programs. What are the implications of adopting the inclusionary mindset? We shall examine this question in the next section.

4. The Implications

4.1 Legal Implications

First, what are the legal implications of removing the bar on software patentability? On the plus side, the law will be free to respond appropriately to the dual nature of software, turning its back on the conceptual quagmire that currently surrounds the issue. There will be no more need to wrestle with obscure concepts like 'as such', 'technical contribution' and 'further technical effect'. Another major change is that the EPO will no longer have to live with a glaring logicality. As I mentioned earlier, one of the striking characteristics of software is that it can mimic hardware. As Hart has pointed out:

In the twin IBM decisions, the Board echoes these sentiments:

It seems very likely now that this anomaly will be swept away. Yet another benefit will be that the EPO will be cured of its schizoid stance of prima facie denying the patentability of software while, at the same time, actually granting many thousands of applications in respect of computer programs.[ 26 ]

From a legal perspective, though, it is not all good news. As Tapper points out, there is a risk that patent offices may be deluged with claims for software patents [ 27 ]. However, it is not just quantity that is likely to put a strain on patent examiners. Given the rapidly increasing complexity of computer programs and the burgeoning state of the art, examiners my find themselves increasingly unable to cope adequately with qualitative questions such as novelty and inventive step as well. In practical terms, the system may simply fail to cope. We will return to this problem shortly.

4.2 Economic Implications

Let us now turn to consider some of the economic implications of the coming change. One of the key issues for intellectual property law is how to strike the optimum balance between, on the one hand, the rights of the owners of existing property and, on the other hand, the interests the population in general, and would-be creators in particular. It is widely accepted that the public interest is best served by granting existing owners a monopoly over e.g. their aesthetic works or inventions to encourage them to invest time, money and effort in creating, making and selling them. However, it is equally widely accepted that such monopolies must be so restricted in terms of extent, breadth and duration that the general population are not unfairly exploited and would-be creators are not stifled. How will the widespread availability of software patents effect this balance?

From one point of view, the effect may not be very great at all. Surely the software industry is much like any other industry? And anyway, removing restrictions on the potential patentability of computer programs is doing no more than taking away one fence from a lengthy steeplechase. Software, like all other inventions, will still have to jump all the other fences to complete the race. Applicants will have to establish, for example, that their software is novel, that it involves an inventive step and that it has industrial application. As we have seen, though, Tapper is concerned that patent examiners may well be unable to cope adequately with the new demands placed on them. He expresses the fear that the result may well be a swing from automatic rejection of software patent applications to automatic acceptance.[ 28 ]

From another point of view, though, software patents may seriously distort and disrupt the existing balance between the proprietary rights of owners and the interest that the general population has in the rapid development and ready availability of software. Bear in mind that, as we have seen, software is simultaneously a symbolic, aesthetic work and a functional invention. As the former, it is already fully protected by copyright. As the latter, it will soon be fully protected by the patent regime. As there is no proposal to do away with copyright protection, the result is that software will be doubly protected. Who will receive the benefits of this double protection? Despite arguments that the overhaul of the European patent regimes in general will remove a brake on the competitiveness of European companies [ 29 ], as far as computer programs are concerned, most of the world's major software houses are now American. Surely, there must be real grounds to fear that, if European companies lost out in the past because of inadequate software protection, they may now be about to lose out again, this time because of too much software protection! As D'Amico reports:

Certainly, if we witness a manic land grab by the existing powerful software houses, followed up by well-financed, aggressive policies designed to defend those territories, it is hard to see how either the general population will benefit or would-be creators will be encouraged. On the contrary, there must be real cause for concern that tightening patent laws will stymie those wishing to create new computer programs. They will live in constant fear of being sued for violating patents [ 31 ]. Quite conceivably, such threats and actions by American software houses will foster in a massive growth of the black market in computer programs, but this is hardly a satisfactory solution. Is it now too late to heed the warnings?

4.3 Political Implications

Finally, what of the wider political implications? As the Internet speeds up our move towards a global economy, the worldwide harmonisation of key aspects of trade law is no doubt essential. Certainly, TRIPS was designed with that end in mind and any attempt to hold out against harmonisation in this or any other area of trade law would now seem positively antediluvian. However, accepting the inevitability of harmonisation on this basis in the short term does not entail acquiescence in any particular regime indefinitely. Is a harmonised regime based on readier availability of software patents the best long-term solution for Europe and for the world community as a whole?

Another thought about the political implications of TRIPS is this. Are we, as Tapper alleges, the victims of American economic imperialism?[ 32 ] There would certainly seem to be at least one precedent for such a situation. The US Semiconductor Chip Protection Act 1984 only granted protection to US nationals and nationals of such other states as offered equivalent protection. Not surprisingly, this had the effect of bouncing the European Union and its member states into adopting a US-style regime in a short space of time.[ 33 ]

5. Conclusion: A 'Neither/Nor' Mindset?

Removing illogicalities and difficult concepts from the law is certainly desirable. Constructing a system that protects computer programs in both symbolic and functional mode is undoubtedly necessary. Achieving global harmonisation of software protection is absolutely crucial. Severe, long-term distortion and disruption of the balance between the rights of owners of existing property, and the interests of the population in general and of would-be creators in particular could, however, prove disastrous.

Consider the possible reactions to such an imbalance. There is a spectrum of such scenarios. At the benign end, we might see a huge growth in communal software, programs developed and enhanced by many different programmers and offered cheaply, or completely free to users. An existing example is Linux the alternative to the Microsoft Windows family of operating systems [ 34 ]. Rather less benign, though, would be a massive expansion of the existing black market. In this scenario, legal protection of software is blatantly ignored. Technological protection is a challenge rather than an obstacle. Suppressive action against black market operators may, as William Gibson predicts in his writings, enhance the image of those operators so that they are increasingly seen as cyberspace guerrillas fighting in the cause of freedom. Driving black market activities into exile is hardly possible in the era of mass interconnectivity that is the Internet. Operators will simply relocate to unregulated parts of the world, and set up software havens. Then what? Economic sanctions, blockades, military strikes, or even software wars?

I believe that it is not a good idea to develop a global regime that overprotects software owners. Is there an alternative? Certainly. It could involve a further shift beyond the current inclusionary mindset towards a 'neither/nor' mindset. Computer programs, at least, would be protected neither by copyright law nor by patent law. Instead, we could renew our search for a third way - a sui generis regime [ 35 ]. Such a regime might be broadly similar to that adopted by the European Union for the protection of databases [ 36 ] albeit with a few key differences. What might such a regime involve? To open the debate, I propose the following scheme, which I argue strikes a fair balance between the interests of existing owners, would-be creators and the public at large:

• Existing owners of computer programs who could demonstrate that their software was substantially their own intellectual creation would be entitled to protection. No further account would be taken of such issues as novelty, inventiveness or industrial application.
   

• Protection would extend to every aspect of programs - from initial design algorithms, to code listings, and beyond to all functional characteristics.
   

• The duration of legal protection would be appropriate to the useful life of a computer program - say a maximum of five years, renewable annually. A new version of a program that involved a substantial change might quality for a new term of protection.
   

• During the period of protection, other would-be creators could decompile a program for the limited purpose of studying and understanding the concepts, processes and techniques used.
   

• During the period of protection, would-be creators could demand a licence of right in respect of part or all of a protected program. The fee for such a licence would then be agreed by the parties or, in default, by a suitable independent body.

At present, it may seem that TRIPS is the very last word on global software protection. However, I believe that we will eventually see it as no more than the first word [ 37 ]. It is true that sui generis protection of computer programs is an unfashionable topic, particularly since the advent of TRIPS [ 38 ]. In time, though, growing distortions and disruptions in the global software market will force TRIPS signatory states to come together again in order to search for a software protection regime that is not only applicable world-wide, but also seen to be both even-handed and workable on that same global scale.

Footnotes

  1 . I would like to thank both Professor Ian Lloyd of Strathclyde University and Alan Story of Kent University for their very helpful comments on an earlier draft of this article.

2 . See e.g. Millard C, 'Chapter 4: Copyright' in Reed C (editor), Computer Law (London: Blackstone, 3 ^rd edition) at 112 .

3 . Hart R, 'The Case for Patent Protection for Computer Program-Related Inventions' [1997] 13 Computer Law & Security Report 247.

4 . Demonstrated by the exclusion of aesthetic creations from patentability in article 52(2) of the European Patent Convention 1973 and mirrored in The Patent Act 1977 s 1(2).

5 . Compare Christie A, 'Designing Appropriate Protection for Computer Programs' [1994] 11 European Intellectual Property Review 486.

6 . Indeed in 1970, the Banks Committee (Cmnd. 4407) expressed considerable concern at the practical difficulties that patent examiners would face if they were required to apply standards of novelty and inventive step to software.

7 . However, Christie points out that software actually has very little in common with the other new technological forms of representation that have been shoehorned into copyright protection.

8 . The United Kingdom Parliament enacted the Patent Act 1977 which mirrored the Convention. In particular s 1(2) of the Act reproduced the Convention's exclusion of software patents.

9 . Gunter Gall, Director of Legal Affairs at the European Patent Office, quoted in Hart R, 'The Case for Patent Protection for Computer Program-Related Inventions' [1997] 13 Computer Law & Security Report 247 at 250.

10 . [1987] 2 EPOR 74.

11 . VICOM's Application [1987] 2 EPOR 74 at 80-81.

12 . European Patent Office Guidelines , C-IV, 2.3 xii.1994.

13 . See e.g. Lloyd I, Information Technology Law 2 ^nd ed. (London: Butterworth, 1997) 280-289.

14 . Case T1173/97 and Case T0935/97. The latter has been reported in full at [1999] RPC 861.

15 . IBM's Application [1999] RPC 861 at 871.

16 . See comment to IBM's Application [1999] Masons Computer Law Reports 280

17 . Lloyd I, Information Technology Law 2 ^nd ed. (London: Butterworth, 1997) 295

18 . The Berne Convention for the Protection of Literary and Artistic Works 1886 (revised in 1971).

19 . IBM's Application [1999] RPC 861 at 868.

20 . See, for example, The EU Commission's Green Paper on the Community patent and the patent system in Europe (1997) 16-17.

21 . The Community Patent Convention 1975.

22 . See The EU Commission's Green Paper on the Community patent and the patent system in Europe (1997).

23 . Patents: Commission outlines ambitious series of measures (1999) at: europa.eu.int/comm./dg15/en/intprop/indprop/99/htm

24 . Hart R, 'The Case for Patent Protection for Computer Program-Related Inventions' [1997] 13 Computer Law & Security Report 247.

25 . IBM's Application [1999] RPC 861 at 874.

26 . Cohen L, 'The Patenting of Software' [1999] 16 European Intellectual Property Review 607.

27 . See his comment on IBM's Application [1999] Masons Computer Law Reports 280

28 . Again, see his comment on IBM's Application [1999] Masons Computer Law Reports 280

29 . For example, see The EU Commission's Green Paper on the Community patent and the patent system in Europe (1997) 1.

30 . D'Amico M, ' Europe considers loosening software-patent rules' ( 8 July 1999) IDG News Service.

31 . E.g. Stallman R, 'Saving Europe from Software Patents' in May 16, 1999 at http://features.linuxtoday.com/stories/5960.html

32 . A view he expresses in his comment on IBM's Application [1999] Masons Computer Law Reports 280.

33 . Lloyd I, Information Technology Law 2 ^nd ed. (London: Butterworth, 1997) 390-1. Contrast this situation with that of the regulation of trans-border data flows, where the European Union has been attempting to pressure the US into adopting conforming measures.

34 . See Stallman R, 'Saving Europe from Software Patents' in May 16, 1999 at http://features.linuxtoday.com/stories/5960.html

35 . The Latin term sui generis literally means 'of its own kind'. In the context of intellectual property rights, it refers to the creation of a special, tailor-made category of protection for the subject matter in question as opposed to simply including it within one of the existing regimes such as copyright or patent law.

36 . See Council Directive 96/9/EC on the legal protection of databases.

37 . Compare the current debate over the protection of digitised information in general e.g. Copyright and Related Rights in the Information Society - Proposal for a Directive (1997) at http://europa.eu.int/comm/internal_market/en/intprop/intprop/1100.htm and Amended Proposal For Directive On Copyright And Related Rights In The Information Society (1999) at http://europa.eu.int/comm/internal_market/en/intprop/intprop/copy2.htm

38 . See e.g. Christie A, 'Designing Appropriate Protection for Computer Programs' [1994] 11 European Intellectual Property Review 486 at 487.