By the EU’s customary standards, the row over the patentability of computer-implemented inventions is being fought in Brussels and Strasbourg with unusual tenacity.

Patentability of...what? Ok, let’s start again: interpreting the words is a big part of this spat. “Patentability of computer-implemented inventions (CII)” is EU-speak for inventions which depend on computers, and therefore computer software, for their operation. The argument, which has been running for several years now, concerns an EU Commission-inspired proposal to harmonise the way CIIs are dealt with across the EU.

Proponents of the proposed directive argue that without a “coherent system” of patent protection, in the words of Alain Pompidou, president of the European Patent Office, the EU risks “splendid isolation” among countries with advanced technical capabilities.

Opponents claim that, as drafted, the directive has loopholes which could lead to - as has already happened in the US - the patenting of software and business methods, which they say will act as a drag on innovation and competition.

They argue that existing copyright provisions, which protect the detail - the order of the words in a novel, for example - of a piece of work but not the basic concept - the novel itself - are adequate. They are concerned that if it is possible to patent the concept of a piece of software, it will stifle innovation and give large organisations, who can afford the cost of patents and the lawyers to protect them, an advantage over smaller, more entrepreneurial companies.

It’s a textbook example of the way the ludicrously simple can, to outside observers anyway, seem simply ludicrous.

The basic problem is nothing to do with patenting software: it is that inventions involving computers are handled differently by different patent offices across the EU leading to the possibility of anomalies and even disruption of trade. Everybody agrees that the situation should be remedied. But how? And how, in any case, do you define a computer-implemented invention?

The story to date is that a draft directive prepared by the Commission (or, more correctly, its technical specialists) was substantially amended by the European Parliament in September 2003. The Parliament, greatly impressed by the arguments of the anti-software patent lobby, stripped out most of the patent protection for software-related inventions incorporated in the draft.

By May 2004, the EU Council of Ministers agreed a “common position” on the directive, a draft to which all the member states could subscribe. At that point the newly elected European Parliament asked the Commission to withdraw the draft and start again from scratch, a measure to which the Commission would not agree.

The “common position” - essentially the original draft with all the patent protection reinstated - has now gone back to the Parliament for a second reading. A decision is expected before the Parliament rises for the summer recess. Pessimists say it is unlikely the Parliament will agree to the common position this time, possibly leading to conciliation talks. And that would not be good, according to Mark MacGann, director general of the European Industry Association for the IT industries (EICTA), who says it leaves too much to horse-trading behind closed doors.

The campaign against the directive has been mounted by the Open Source community, developers who believe that programs can be created more effectively if authors are allowed freely to modify and distribute source code on a world-wide basis. (Open source code may be free, but to be useable by most companies it has to be packaged as a product for which a charge is made.)

Mr Pompidou says that the groups leading the campaign, including the Foundation for a Free Information Infrastructure (FFII) and NoSoftwarePatents.com, who agreed to combine their interests earlier this year are driven by “entirely noble” aims. But they may be aiming at the wrong target. Last year EICTA commissioned Prof Robert Lind, formerly of Cornell University and now a senior associate with the London-based consultancy Charles River Associates, to conduct an independent analysis of the situation.

Although Prof Lind was commissioned by an element of the pro-directive lobby, his report supports the Commission’s case. He writes: “It should be noted that the Open Source movement has grown and flourished in the current environment of patents for CII. This group has based its opposition to patent protection primarily on the argument that patents stifle innovation. While this view does have some academic support, in fact it is an unproven hypothesis.”

Prof Lind says, essentially, that the software industry makes little use of patent protection while other technological industries - telecommunications, or automotive engineering - believe that stripping away such protection would be harmful. He broadly concludes , that the “common position” should form the basis of the directive, arguing for well funded provisions to monitor the system: “It is important to put to rest, once and for all, the notion that allowing patents on CII will ultimately lead to the patenting of discoveries, mathematical methods, scientific theories, aesthetic creations schemes, rules and methods for performing mental acts, playing games, or doing business and presentation of information”.

If he is right, the opposition should let the Commission get on with it. But the Commission has a duty to ensure nothing slips by on the blind side and should spell out the mechanism it intends to use. As for the definition of a computer-implemented invention? Well, another day, another column...

Get alerts on Front page when a new story is published

Copyright The Financial Times Limited 2019. All rights reserved.
Reuse this content (opens in new window)

Comments have not been enabled for this article.

Follow the topics in this article