Matthew Toseland

5 Lower Grove Road

Bristol BS16 2BS

toad@amphibian.dyndns.org

28 July 2006

Software Patents again: the London agreement, the IPRED2 and the EPLA.

Dear Lord Sainsbury,

Thank you for your past correspondence and work on the software patents issue. I was at the 14/12/2004 meeting, and one of the technical contribution workshops. Unfortunately I must now write to you on closely related matters. Firstly, according to the press the UK, France and Germany are considering the proposed London agreement, which would allow patents to be filed in the UK in French or German. I object most strongly to this. It is difficult enough for me, as a small independent software developer, to identify relevant patents if they are written in “English”, but forcing small businesses to read patents written in foreign languages is unreasonable. Any savings to the patent author from not having to translate it into all three languages are offset by increased costs and increased legal uncertainty on the part of everyone else affected by the patent. But if the directive on criminal measures aimed at enforcing intellectual property rights (IPRED2) goes through as well, the situation in the UK will become untenable for all but the largest players in the software industry.

That is my second concern. IPRED2 would make “intentional” patent infringement “on a commercial scale” a criminal offence; also aiding and abetting, attempting and inciting, and not just patents but all IP. This directive is grossly unreasonable. It's one thing to risk a civil suit when your business becomes big enough to be a target (at which point you might be able to settle). It's quite another thing to risk jail time, judicial winding up, or a ban on commercial activities and state aid!

This is a risk that all software developers, apart from those working for the largest players, face: It is likely that sooner or later I will unwittingly violate a software patent. Firstly, because many patents are trivial, or are necessary for using standard file formats. Secondly, even though a technique does not immediately appear trivial, in many cases it will have been routinely reinvented; this is the nature of computer science. All but the first person to file a patent on the technique now become criminals. Since a substantial software product will use thousands of techniques, it is impractical to conduct an exhaustive patent search.

Lets say I'm contracted to build an e-commerce site. This may not be innovative in itself, but a specific web-shop may well be innovative. I happen to know of a group of trivial patents which cover web-shops (http://webshop.ffii.org). I believe they will not stand up in court if challenged. However if the IPRED2 passes, I would be staking not a royalty, not even my business, but my freedom and my right to do business (as well as my right to legal aid) on that belief! What am I supposed to do? Martyrs make poor businessmen – most will just leave the field, or the country.

Worse still, according to the Free Software Foundation Europe, “intentional” in criminal law does not necessarily require knowledge of the patent. “On a commercial scale” is no help either; anything commercially useful is “on a commercial scale”. The IPRED2 would have a range of other chilling effects; the Foundation for Internet Policy Research has a page on this: http://www.fipr.org/copyright/ipred2.html.

Finally, the European Patent Litigation Agreement may legalise software patents without approval from the European Parliament, creating a new system of courts based on faulty EPO case law which allows both patents on computer programs and “business methods”. EPO rulings frequently widen the scope of parentability (for example, the Hitachi judgement).

Yours sincerely,

Matthew Toseland

Chief developer, Freenet Project Inc.