Program Claims: Bans on Publication of Patent Descriptions

Patent Claims directed to a

"computer program product", "computer program", "data structure" or other information object, "characterised by that upon reading out the information ... happens",

were introduced in the USA in 1994, shortly after a USPTO hearing in which all major software companies except for Microsoft pronounced themselves against software patentability. In Europe, program claims were introduced in defiance of the written law by the European Patent Office in 1998, followed immediately by the UK Patent Office and later by some more national patent offices. In 2002 the European Commission proposed to reverse this decision.

The European Commission's Software Patent Directive Proposal's Article 5 does not allow program claims, i.e. claims of the form "computer program, characterised by that upon loading it into memory [ some process ] is executed". In justification, the European Commission said that it did not want "software as such" to be patentable. Yet the Commission's proposal speaks of "infringing programs" and ensures that at least the execution of a computer program on a general-purpose computer can be a patent infringement, whereas its publication or distribution may or may not be seen as an infringement, depending on how the courts apply the doctrine of "contributory infringement". Critics from all sides have found this inconsistent. The European Parliament's committees for cultural and industrial affairs (CULT and ITRE) voted for amendments which safeguard freedom of publication. Amendment ITRE-13 states that the publication of a program can never constitute a patent infringement. The Legal Affairs Commission (JURI), on the other hand, explicitely rejected ITRE-13 and instead voted for a "compromise amendment 1", which bashfully legalises program claims (with a pseudo-limitation: "only if ... [ the program claims are derived from process claims ]"). According to this "compromise" between the patent lobby and the patent lobby, software authors and Internet service providers (ISPs) can be sued for direct patent infringement everywhere in the European Union, as soon as they make a program text with the claimed features available somewhere on the Internet.

The European Parliament's Committe for Legal Affairs and the Internal Market (JURI), in its resolution of 2003/06/18, proposes to legalise program claims, i.e. claims to

This would make publication of many programs a direct patent infringement, thus creating additional risks of litigation for programmers, university researchers and Internet service providers.

Program claims also lead the patent system ad absurdum. Taken to the extreme of consistency, they mean that patent descriptions are patented: by publishing the full disclosure (which in case of software should contain some usable source code), the patent office would infringe on the program claim. This illustrates nicely why software cannot be patentable: patents are supposed to be deals between the patentee and the public, where information is disclosed and matter monopolised.

see JURI Art 5

The "consumer-protection" argument in favor of program claims is groundless and cynical. It is always possible for the end user of commercial software to hold the distributor liable by means of standard contract clauses, and this is what would usually happen, if software is patentable without program claims.

The real reason why industrial patent lawyers are pushing for program claims seems to be a symbolic one: the existence of program claims brutally and indecently documents that the border to patenting software as such has been crossed. Conversely, by refusing to grant program claims, the European Commission can appear as moderate.

However, in order to protect the freedom of publication, as guaranteed by Art 10 ECHR, the mere refusal of an indecent claim form is not enough. A more explicit affirmation of the freedom of publication, such as ITRE-13, is needed.

Ultimately, in another context, it must be made clear that information objects in whatever form are not patentable.

Interestingly, JURI did not vote on ITRE-13 at all, because Arlene McCarthy judged it incompatible with her program claim amendment COMP-1 and therefore "caduc" (obsolete) if COMP-1 was voted for, which it was. In other words: Arlene McCarthy admits that program claims are a direct negation of the freedom of publication. One could conceive of a scenario where program claims are really considered to be "only of declaratory nature" (as said in the justification of COMP-1) and freedom of publication is nevertheless maintained. Such scenarios are not rare in law. Yet for JURI the matter is clear: on the one hand side they downplay the impact of program claims by calling them "merely of declaratory nature", on the other their decision is built on the implicit assumption that program claims and freedom of publication are so completely mutually exclusive that ITRE-13 needn't even be put to vote.