In theory, the approach to patentability should have been the same in both patent offices. They were applying the same statutory approach to patentability, and to the exclusions to patentability set out in article 52(2) of the European Patent Convention (EPC). However, on the question of the patentability of computer programs there has been significant divergence. It has taken nearly 30 years for the UK courts to reach conformity with the EPO approach and to allow patents to be granted for certain computer programs on media. That we have conformity is due to the crusade of one man, Keith Beresford, the senior partner and owner of Beresford & Co, European patent attorneys.
The problem arose because the EPC and the derivative UK Patents Act excluded from patentability “methods of doing business and computer programs ‘as such’”. It took a number of stages to get a consistent interpretation of what was allowed. The problem arose because a computer program is often a convenient way of implementing a set of instructions which produce a technical advance (technical effect in patent-speak). Early on it was decided both in the EPO and in the UK courts that where the computer program is used to implement a method which produces a technical advantage, the process would be patentable.
An early example was provided in Vicom [1986], EPO technical board of appeal, where the invention was to obtain a substantial increase in the processing speed of a computer chip. However, where there was no technical effect, and the computer program was doing no more that running a program on a computer, patentability was refused. An example was provided by Gale’s Application [1991], where the claims were to better the method of calculating a square root embodied in a read-only memory (ROM).
Starting in 1988, the EPO in two IBM cases, especially IBM/Computer Program Product held that computer programs which produced a technical effect would be allowable and eventually allowed claims to such computer programs on media (Microsoft/Data Transfer [2006]). The IPO and UK courts lagged behind. The IPO steadfastly refused all claims to a computer program, whether or not when run it had an inventive technical effect.
The position in the IPO has now conformed to the position in the EPO, more or less in two cases. In Aerotel v Telco Holdings: Re Macrossan’s Application [2007], the court of appeal held that the IPO was wrong to exclude from patentability all computer programs; only excluded were those that did not cause a technical effect, such as an ability to do something which had not been possible before or a significant improvement to an existing process, albeit that the process was run by a computer program.
Recently, in Astron Clinica and others v the Comptroller General of Patents [2008] Kitchin decided that the patent office was wrong to exclude from patentability claims to the computer program on media where the computer program is otherwise patentable. In doing so he held that there was no express exclusion to such claims in the EPC, nor was there an express prohibition in the court of appeal decision in Aerotel.
The decision is of great practical importance, since what is often sold is a computer program or a download over the internet. It can be sold either as a discrete item or as embodied in someone else’s program. Previously such claims would not be upheld in court if granted by the EPO, or even granted by the IPO. This left inventors relying on method claims and contributory infringement and resulted in exported programs being unprotected. That hiatus has now been closed, and not before time.
The UK courts have traditionally taken a harder line to patentability than their continental counterparts. This has been especially the case on software-driven inventions and second and subsequent generation pharmaceutical patents. Let us hope that the Astron decision is the start of a rebalancing; but I will not hold my breath.
Larry Cohen is head of European IP at Latham & Watkins in London.