Early in July, the European Union's parliament rejected, by an overwhelming 648 to 14 vote, a bill proposing a common framework for patenting of software across its 25 member countries. Given the vote, the EU stays with the status quo in which software patenting is still possible, though there are as many software patenting regimes in place as there are members. This near unanimity in favour of the status quo was unusual, especially because the debate on software patenting that preceded the vote was contentious to say the least.

In fact the long and contentious debate has helped clarify the case, (mostly) against and (less) for the provision of patent protection for software. The European debate began when, backed by big technology companies, the Commission of the European Communities prepared in 2002 a proposal for a directive of the European parliament and of the council on the patentability of computer-related innovations. A "computer-implemented invention", which is a rather convoluted term for software, is defined as any invention the performance of which involves the use of a computer, computer network or other programmable apparatus, and which has one or more prima facie novel features that are realised wholly or partly by means of a computer programme. A computer-implemented invention is patentable on the condition that it is susceptible of industrial application, is new and non-obvious, and makes a technical contribution. A "technical contribution" in turn is defined, quite ambiguously, as a contribution to the state of the art in a technical field which is not obvious to a person skilled in the art. The ambiguity surrounding the definition of what is patentable is thus obvious.

The Commission's case was straightforward. Software makes a significant contribution to GDP and employment in the Union and that contribution is expected to rise rapidly over the coming years if members are provided appropriate opportunities. Patent protection, that offers an inventor a protected right to exploit the invention without fear of competitive replication for a specified period of time, has proved its usefulness as an incentive to innovators to invest the necessary time and capital on new innovations. And the growing use of software patents in the US could put large technology firms in Europe at a disadvantage. Hence, there was need to harmonise and thereby strengthen the software protection regime in Europe.

However, the Commission's immediate ambition was presented as being more limited. It suggested that the prevailing regime relating to software protection in the EU was ambiguous for two of reasons. First, though under Article 52(2) of the European Patent Convention (EPC), programmes for computers "as such" are not inventions and are therefore excluded from patentability, a large number of patents for computer-implemented inventions had been granted by the European Patent Office (EPO) and by national patent offices. The EPO alone is reported to have granted more than 30,000 software patents. Secondly, even when statutory provisions setting out the conditions for granting such patents were similar, there were substantial differences in the implementation of these provisions as reflected in the case law of the Boards of Appeal of the European Patent Office and the courts of Member States.

In order to deal with these ambiguities, the proposed legislation sought to codify with the help of European case law the circumstances in which software can be considered as being more than a computer programme "as such". This was expected to provide information as to the nature of software innovations that can be patented and increase the volume of patenting. A second important objective of the directive was to harmonise the patent regime across the Union, so as to reduce the ambiguities resulting from differences in case law.

This effort of the EU to expand the software patenting regime was a major step forward given the fact that software patents were till recently not the norm and given the controversy which still surrounds the provision of patents for software. In fact, it was only in 1981 that the US Supreme Court ruled that software was patentable, at least when it functioned or operated to control an industrial process. However, for long after that, software was essentially considered to be non-patentable, since many computer programmes performed a range of more "intangible" functions such as word processing, accounting, graphics generation, and the like. Protection for such software was available only under the copyright convention.

However, there is a substantial degree of difference in the extent of protection afforded by copyrights and patents. Copyright traditionally protects only the "expression" of an idea, whereas a patent protects the idea itself. Thus, in the case of software, copyright protection extends only up to the original software code, which cannot be replicated or altered. But if an alternative code can be deployed to achieve the same end result, copyright is not violated. As a result courts have interpreted the extent of copyright protection for software by closely examining the details of a computer programme rather than the overall characteristics of the end product.

A classic instance of this was when Borland International included in its accounting software Quattro a set of Lotus 1-2-3 type menu commands, which allowed customers to easily switch from the latter to the former. Customers had the option of using Quattro's own menu commands or a virtually identical copy of the entire 1-2-3 menu tree. However, Borland did not copy any of Lotus' underlying computer code that delivered the functions performed by the spreadsheet. However, Lotus, as expected, chose to go to court on grounds of copyright infringement.

After a four-year battle in the courts the Supreme Court upheld a decision by the a First U.S. Circuit Court of Appeals denying copyright protection for the menu commands in the Lotus 1-2-3 electronic spreadsheet program. The First Circuit Court had assessed Lotus’ menu tree to be a "method of operation" that was expressly excluded from copyright protection since the relevant clause states: "[I]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle or discovery, regardless of the form in which it is described, explained, illustrated or embodied in such work." What is more the court declared: "The computer program is a means for causing something to happen; it has a mechanical utility, an instrumental role, in accomplishing the world's work. Granting protection, in other words, can have some of the consequences of patent protection in limiting other people's ability to perform a task in the most efficient manner. It is no accident that patent protection has preconditions that copyright protection does not -- notably, the requirements of novelty and non-obviousness -- and that patents are granted for a shorter period than copyrights."

In sum, if software had to be protected in the manner in which Lotus and others felt it should be, then a patent rather than copyright was the appropriate route. Since a patent can protect an idea or concept, the patent claim can be written to cover the so-called novel combination of elements or steps making up the patented system or process. This would make it difficult or impossible to write another computer programme that would offer a similar system or process without infringing the patent.
It was in the context of such judgements that a virtual movement to ensure patent protection for software began in the US. The result has been a flood of patent applications for software, resulting in the grant of what has been identified as "clearly invalid patents" for inventions that are either not new or where there is no significant inventive step. That tendency has not slowed. According to Randall Stross, a Silicon Valley analyst who recently wrote on the subject in the New York Times, around a year back Bill Gates announced Microsoft’s decision to raise the number patent applications it submitted annually to 3,000 from 2,000. This amounts to identifying close to 60 novel and non-obvious patentable ideas every week. Not surprisingly, the subjects of Microsoft’s applications were indeed flimsy and included titles like "System and Method for Creating a Note Related to a Phone Call" and "Adding and Removing White Space From a Document."

Stross is convinced that the 3,955 patents that Microsoft has already been issued should be struck down and the 3,368 patent applications it has pending should be summarily rejected. His view is shared by software major Oracle, which had in an officially declared policy opposed the patenting of software as it was inappropriate for industries in which innovations occur rapidly, can be made without a substantial capital investment, and tends to be creative combinations of previously-known techniques. In its view, patent protection was devised for engineering and mechanical technologies that are characterized by large "building block" inventions that can revolutionize a given mechanical process. On the other hand, "software, especially a complex program, seldom includes substantial leaps in technology, but rather consists of adept combinations of many ideas."

There are also a number of practical problems in implementing a software patent regime. As Oracle put it, "software patent examinations are hindered by the limited capability of searching prior art, by the turnover rate among examiners in the Patent and Trademark Office, and by the confusion surrounding novelty and innovation in the software arena. The problem is exacerbated by varying international patent laws, which both raise the cost and confuse the issue of patent protection."

What is more, filings of patent applications tend to be expensive, especially because of the large sums to be paid out in the form of lawyers fees. Further, patents for incremental innovation which is typical of the software industry entail the economic costs of identifying the patent holders and negotiating the necessary licences. Small firms therefore would be unwilling to apply for patents, even though many new software ideas emanate in small firms. Hence, patents for computer-implemented inventions might strengthen the market positions of the big players and result in the monopolization of an industry that technologically tends to be more competitive in structure. The net result would be a setback for innovation.

Based on these and other similar arguments, the advocates of open source software joined the debate on the EU software directive. The controversies that ensued generated the fear among the big software companies that they would lose even the protection they currently receive through the EPC and national patent legislation. So while small firms and the open source software movement wanted the EU bill rejected and used as an occasion to question the appropriateness of software patenting, the big firms wanted the bill dropped to prevent its use as the basis for a reversal of the existing level of software protection. In the event the bill was rejected in a vote which Josep Borrell, president of the EU parliament described as the most decisive majority vote in the history of the chamber. This near unanimity in rejection was surprising given the contentious nature of the debate that preceded it. There was no commonality of purpose, but differences on the direction software patenting should take ensured convergence around a negative verdict. Meanwhile, the software patenting regime in the EU continues as before despite the arguments advanced to show that the practice, like the EU directive, must go.

August 24, 2005.