We have a good law already, but some patent courts are not respecting it. However they are under strong and growing public pressure to return to a sound practise, in conformance to the law. The pressure will grow if we do not pass a directive. It can grow more quickly if we pass a good directive. The only way for the European patent establishment (i.e. the national patent administrators who sit on the EU Council's patent policy working party and on the Administrative Council of the European Patent Office) to stabilise its bad practise is by passing a bad directive.

We can accept almost any directive, as long as it consists only of clear and simple exclusions from patentability.

Art 52(2) EPC consists of such exclusions. It says in clear and simple terms what is not an invention in the sense of patent law.

We can not accept inclusive language, such as

where X is not in turn delimited in clear and simple exclusions. Council-style language of the sort is even less acceptable.

Any directive that contains statements of the above type can not claim to be intended to "harmonise the status quo" or to "prevent a drift toward patentability of business methods".

First things come first. Below we try to outline the exclusions on patentability and patent enforcability in their order of priority, as they build on one another. There is little sense in scoring points on the second if the first is not assured.

Our constituents' basic interest is to keep the software free from patents, regulated by copyright only. I.e. even if there are patents on the much cited "anti-lock braking system", "washing machine", "intelligent vacuum cleaner" etc, they must apply only to the makers and users of the devices, not to people who create or provide software (= control logic, similar to user manuals) for these devices.

Our first and most basic demand is "no program claims, no direct or indirect infringment by software distribution". The European Parliament has fulfilled this demand with Art 5A. Unfortunately the Council, by introducing program claims in their Art 5(2), has torn down the only bridge to constructive negotiations that the European Commission ever built. If they insist on keeping it torn down, that was it.

We can not accept restrictions on the use of equipment that consists of general-purpose computers only. Patent claims of whatever form (process, device, system and method, ...) are unacceptable when the contribution to the prior art consists of pure data processing (i.e. instructions for operation of general purpose data processing equipment).

The Parliament has solved this problem by a clear and simple exclusion statement in the spirit of Art 52 EPC:

In a similar article, it has specified some items that do not constitute a "technical contributions": e.g. "improvement of calculation efficiency".

Given that, according to Art 27 TRIPs, "patents shall be available for inventions in all fields of technology, provided that they are .. susceptible industrial application", the terms "technology", "technical", "fields of technology", "invention" and "industrial" need to be defined at the legislative level. Because of the inclusive nature of the TRIPs statement, it is now no longer adequate to let caselaw define these terms. The European Parliament has provided the needed definitions in its first reading.

The Parliament has also clarified the inclusive term "computer-implemented invention", which was imposed by the European Patent Establishment without any legitimate need. The EP definition must be retained, but the term should be discarded as far as possible. The directive title should be amended to something more straightforward, such as "on the limits of patentability with regard to data processing and its fields of application".

An attorney may lay claim to "a computer, characterised by that it orders windows on a screen according to scheme X". In this case, the real "invention" (= "contribution") is the "scheme X", and such schemes are not inventions according to Art 52 EPC.

Under the new formalistic approach^[1] introduced by the European Patent Office (EPO) in 2000, the "invention" always consists in the "claim as a whole", e.g. in the above example the "computer", which is of course technical and therefore patentable, whereas the term "technical contribution" might seem to refer to what was previously called the "invention". However, even the term "technical contribution" has come to be used in counter-intuitive ways which few outside the EPO understand. E.g. in the above example, there might be a "technical contribution in the inventive step", because the "technical problem" of "improving efficiency in the use of space on the screen" is solved during the "inventive step" from the "closest prior art" to the "invention".

This obscure reasoning of the EPO is not used in general patent law, but only in the case of "computer-implemented inventions". It emerged in the late 1990s as a scheme for twisting the existing law so as to allow software patents. By writing such reasoning into a directive, the EU would authorise arbitrary practises which only the EPO itself can control. Moreover, by mingling the question of "technical contribution" with that of "inventive step", the EPO's new doctrine makes it nearly impossible for most national patent offices, which do not conduct prior art searches, to reject software or business method patents. Finally, the new doctrine introduces a sui generis software patent law, which does not correspond to the general principles of patent law as used in other fields, and therefore, as the US government and some European law scholars have warned, could be seen as a breach of the TRIPs treaty.

The Council's attempts to call any claim object an "invention", to equate "invention" with "the claim as a whole", or to distinguish between "invention" and "technical contribution", must be rejected. As stressed by the EPO's Examination Guidelines of 1978, the examiner "should disregard the form or kind of claim and concentrate on the content in order to identify the novel contribution which the alleged invention claimed makes to the known art. If this contribution does not constitute an invention, there is not patentable subject matter." It would be enough to reconfirm this principle. The exact methodology of how the examiner identifies the contribution/invention needn't be formalised. If it is to be formalised, then this can be done only in the way proposed by the Parliament in the first reading, i.e. "identify the technical contribution by subtracting all non-technical and non-novel features from the claimed object".

Art 30 TRIPs requires clarification with regard to interoperability at the legislative level, not at the caselaw level.

Commission and Council pretend, e.g. by Art 6, that they want to guarantee freedom of interoperation in the spirit of the 1991 software copyright directive.

Indeed there is a need of clarification, and the Parliament has given the right answer: Art 6a. The CULT/ITRE/JURI version of this proposal, which was endorsed by the Luxemburg delegation in the Council, serves the purpose as well as the plenary version. In both cases, some wording details can be improved.