Computer Program Functionality and Copyright

May 11, 2016

The advent of digitised property in our information society has raised critical challenges for the protection of intellectual property rights. As John Perry Barlow put it over 20 years ago in The Economy of Ideas (Wired, 1 March 1994), digitised property can be infinitely reproduced and instantly distributed anywhere around the world at no cost, which means traditional intellectual property law cannot protect it in the way it protects tangibly expressed intellectual property. This conundrum is very vivid when it comes to protecting software through copyright: the boundary between an idea and its expression and between a literal and a functional expression is blurred. Legislators and courts have thus spent years trying to determine which aspects of software can be protected by copyright, and have come to the conclusion that only the source code is protected, as a literal expression. The functionality of software cannot be protected by copyright, as it represents a mere idea. As such, any software developer can emulate existing software as precisely as possible, as long as he or she does not directly copy the original software’s source code. It is thus clear that the law has failed to provide software developers with any guarantee that no one will use their work to exploit their commercial advantage. However, this state of affairs can be praised as beneficial for competition and innovation, and thus for consumer choice. It prevents copyright protection from being extended to where it has never been before. It is also likely, and hoped, that the alleged downfall from this lack of legal protection – a disincentive to create new innovative software from scratch – will be resolved through platforms promoting mutually beneficial understandings between software developers.

Copyright law has protected the expression of ideas for centuries. It has been extended throughout its life to cover ideas as performed on stage, as recorded artistically, photographically, cinematically, etc. It has been effective because it does not protect the idea itself, but rather its expression – ie its physical manifestation. It is easy to control the dissemination of books, performances, and even DVDs. However, because digitised property can be infinitely reproduced and instantly distributed at no cost, copyright enforcement is much more difficult for content such as music, films or books in digitised form. As John Perry Barlow says, ‘without our old methods, based on physically defining the expression of ideas, and in the absence of successful new models for nonphysical transaction, we simply don’t know how to assure reliable payment for mental works’. When the need to protect software arose, copyright came naturally to the minds of legislators, as it was designed to protect ‘literary work’, which can easily include software code. Section 3(1) of the Copyright, Designs and Patents Act 1998 provides that ‘literary work’ includes ‘a computer program’. The problem, however, which has given rise to years of litigation over the copyrightability of software, lies in the fact that the valuable aspect of software to its developer is not its code – rather, it is the output that the code will display on the user’s computer screen.

This is why in the 1990s, UK copyright law protected the ‘look and feel’ of software – that is, the ‘user interface’ which it creates. The idea of the early cases (John Richardson Computers Ltd v Flanders (No. 2) [1993] FSR 497, IBCOS Computers Ltd v Barclays Mercantile Highland Finance Ltd [1994] FSR 275) was to protect the ‘functionality’ of software, to prevent someone from replicating the consumer experience without literally copying any of the code. These cases thereby provided very strong protection to software developers against non-literal copying.

However, the 2000s have seen the erosion of this protection, initiated in 2004 with the case of Navitaire Inc. v Easyjet Airline Co [2004] EWHC 1725 (Ch). Easyjet, the first defendant, had instructed the second defendant, BulletProof Technologies, to create software which would very closely emulate the functions of Navitaire’s software, to implement various upgrades. Importantly, BulletProof Technologies did not directly copy any of Navitaire’s code. Navitaire argued that BulletProof Technologies had infringed their copyright by copying command codes, screen displays, and the underlying business logic of their system – effectively the ‘look and feel’, or ‘functionality’ of the software. The judge, Pumfrey J, found infringement of copyright in the screen displays, which were considered to carry some artistic elements. However, he found that no copyright could exist in the command codes – which could not be considered ‘literary works’ – nor in the ‘business logic’ of the claimant’s system. Despite recognising (at [14]) that the functional element of software, ie ‘the manner in which a machine behaves under the control of a program’, is what really matters to a programmer, Pumfrey J refused to extend copyright protection to the ‘overall function’ of the interface, saying (at [130]): ‘it seems to me that to permit the “business logic” of a program to attract protection through the literary copyright afforded to the program itself is an unjustifiable extension of copyright protection into a field where I am far from satisfied that it is appropriate.’ Together with the later case of Nova Productions v Mazooma Games [2007] EWCA Civ 219, Navitaire thus established that the ‘look and feel’ of software cannot be protected by copyright – de facto stripping software developers of their right to control derivative works, a fundamental right afforded to copyright holders.

This may have come as a disaster to some – mostly big – software developers, since what they really care about selling is customer experience, not source code. However, these decisions are logically sound and must be taken as great news for innovation. The analogy that Pumfrey J used to reject Navitaire’s claims is telling of the underlying logic of his decision (at [127]): if one chef creates a pudding, he has copyright in the written recipe. If another chef manages to emulate this pudding and records his recipe for it, he cannot be said to have infringed the original recipe – otherwise this would stifle culinary innovation, as chefs would not be able to create variations on the pudding. The same is true of software: one should be able to put hard work into reproducing an existing software to eventually make it better or less costly. As Simon Stokes has argued after the Navitaire decision was rendered:  

[t]he law has to set boundaries between what is copyright infringement and what is not: go too far one way and innovation is stifled because the public domain of ideas is encroached upon; go too far the other way and copyright creators may be disinclined to create copyright works if the law does not adequately protect their works. Whilst this decision will alarm some in the IT industry, it is submitted that in this case the judge got the balance right.[1]  

Navitaire thus buried the protection of software through non-literal copyright infringement quite deep, and this should be praised for the sake of all future innovators and their customers.

The case of SAS Institute Inc. v World Programming Ltd [2010] EWHC 1829 (Ch) later supported Pumfrey J’s approach in Navitaire. In this case, WPL, the defendants, emulated the functionality of the claimant’s software in their own program, without accessing the software’s source code. Arnold J refused the argument of SAS that ‘it was a breach of copyright to study how a computer program functions and then to write a similar program to reproduce the functionality’.[2] The case was referred to the ECJ, who found that ‘neither the functionality of a computer program nor the programming language and the format of data files used in a computer program in order to exploit certain of its functions constitute a form of expression of that program’,[3] and thus are not protected by copyright. Thus anything which is not direct copying of source code is permissible, and copyright protection is thereby reduced only to protect software against software piracy (installing a copy of a software without authorisation). This decision can easily seem harsh: the claimant’s and the defendant’s programs were absolutely identical, and WPL’s differentiating factor was entirely based on price. However, it is entirely legitimate for WPL to be able to do this: in all industries, first movers have seen their innovations quickly copied by competitors, who were able to offer the exact same product at a lower price. Market advantage is often gained by doing the same things more efficiently, rather than doing different things. Moreover, allowing companies to emulate and improve on existing software allows the dismantlement of potentially harmful monopolies: in the case of SAS, WPL’s reproduction of the SAS software allowed a whole group of customers to move past their continued dependence on SAS created by the SAS monopoly over the ‘idea’ behind their software. We can use an analogy to imagine what would happen if the ideas behind software were protected by copyright: imagine that the ‘idea’ behind a car – four wheels carrying people around – had been copyrighted by its first inventor. For 70 years after his death, no company would have been able to produce cars, to improve on the original design and mechanism, nor to offer a lower price reflecting the evolution of technology. This would have been absurd. It would be just as absurd to afford copyright protection to the functionality of software – it would stifle innovation and reduce consumer choice to none by granting continued monopoly to first movers. Onslow and Jamal have explained the confusion which leads some to consider SAS as offering inadequate protection to software developers:  

Once it is established, as a technical fact, that the same behaviour of a computer may be achieved by two computer programs expressed completely differently from each other, the causative link between the expression of the claimant’s computer program and the expression of the defendant’s computer via the emulation is decisively broken. […] This important concept […] is sometimes not grasped to its full extent by lawyers, as distinct from programmers – there is a lingering suspicion among lawyers that the more closely the defendant emulates, the more likely it is that he must have reproduced lines of code.[4]  

The decision in SAS can thus be praised for its technical wisdom, and for avoiding a potentially heavy obstacle to innovation and competition in the software industry. Pamela Samuelson, writing before the SAS decision was rendered, and hoping that it would turn out in the way it did, supports this conclusion: ‘[t]he pro-interoperability rules of the [US and EU] jurisdictions have contributed to the phenomenal growth of this industry, allowing new entrants to the market for complementary and competing products. Consumers have benefited considerably from the availability of a wide range of interoperable information technologies.’[5]  

The future does not need look so gloomy for software developers, however. Although we are currently in a phase of extremely low software copyright protection, alternative solutions to protect one’s work do exist, and it is likely that others will arise in the future. First, when the computer program at stake is part of a broader invention with a real technical contribution, it can be protected through a patent. Although the Patents Act 1977, s 1(2) clearly states that a ‘program for a computer’ is not patentable, the European Patents Office has enlarged the protection to a large number of software types. But the best solution to software protection can be expected to come not from statutes or courts, but rather from what John Perry Barlow calls ‘uncodified or adaptive law’, emerging from the software industry itself. Just like the music and the film industries’ war against illegal peer-to-peer file sharing is coming to an end through the creation of services like Spotify or Netflix, the software industry could see the advent of private systems of licensing which would render the sharing of software and its re-use for improvement easier and more ‘ethically’ accepted. One can think of platforms such as Creative Commons, created by Lawrence Lessig, a fervent critic of the application of copyright law to software, as a glimpse of the potential solution. Such a system already exists for software: the GNU General Public License, as part of the F(L)OSS movement, allows software developers to attach a license to their software guaranteeing freedom to share and to change versions of a programme.[6] This movement came as a result of the recognition by most software developers that Open Source Software is much better for innovation, by allowing programmers to continuously add onto others’ work to achieve ever greater levels of performance. As Barlow convincingly argued over 20 years ago, ‘[f]amiliarity is an important asset in the world of information. It may often be true that the best way to raise demand for your product is to give it away.’ It will thus be highly beneficial for both the software industry and its clients to do away with legal barricades which stifle innovation and sharing of knowledge and resources. Software will evolve constantly into better forms, ever more suited to clients’ needs and desires.  

As a protector of the physical manifestation of ideas, traditional copyright law is ill-suited to protect content in a digitised form. Software protection is a case in point – copyright isn’t able to protect more than the software source code, meaning that anyone can emulate the functionalities of a software without infringing copyright. Although this has been criticised as removing any incentive to innovate by stripping software developers of their right to control derivative works, it is actually the most logical and beneficial outcome. Protecting anything more than the source code would be an over-extension of copyright to where it has never been before, and would hinder the development of ever better products. As John Perry Barlow had envisaged over 20 years ago, ‘[i]ntellectual property law cannot be patched, retrofitted, or expanded to contain digitised expression any more than real estate law might be revised to cover the allocation of broadcasting spectrum’. It is thus likely and hoped that software protection will develop from mutual understandings between software developers themselves, who come to realize that their industry is one where sharing and giving away is more beneficial to everyone than barricading and enclosing.

[1] Simon Stokes, ‘The Development of UK Software Copyright Law: from John Richardson Computers to Navitaire’ (2005) Computer and Telecommunications Law Review 129, 133.

[2] Andrew Murray, Information Technology Law (Oxford University Press, 2nd edition, 2013), 233.

[3] Case C-406/10 SAS Institute v World Programming Ltd [2012] 3 CMLR 4, para 39.

[4] Robert Onslow and Isabel Jamal, ‘Copyright Infringement and Software Emulation – SAS Inc v World Programming Limited’ (2013) European Intellectual Property Review 352, 353.

[5] Pamela Samuelson, ‘The Past, Present and Future of Software Copyright Interoperability Rules in the European Union and United States’ (2012) European Intellectual Property Review 229, 234.

[6] ‘GNU General Public License’ (GNU Operating System, 29 June 2007) <> accessed 2 April 2016.