Campaign against software patents - CatalÓ
The European Parliament Legal Affairs and Internal Market (JURI) committee 2 voted on Tuesday, June 17th 2003 on its report on the COM(2002) 923directive, proposed by the European Commission. This directive would legalise software patents in Europe, with disastrous consequences.
The Foundation for a Free Information Infrastructure published the committee report4on which we comment here, and a collection of statements from politicians, business people, professionals and jurists condemning it5. There is more detailed information on the amendments that were tabled in JURI6and on what could have come out of the voting, and also an attempt at debate about the rapporteur's views7.
This document is also available as
Some of the contributions that were accepted in JURI may offer Europe positive provisions:
Since member states' patentability laws are already unified, it would seem that the very existence of a directive is not necessary. What the directive would do is mainly putting European courts as the highest court for all of the EU. The directive is therefore an attempt to unify any diverging interpretations of the unified law, which follows the European Patent Convention (EPC8). Such unification, on the other hand, would also be accomplished by the Community Patent, to be enacted soon, and would be an advantage only if the directive contents brought clarity and good measures for the EU economy and values.
Amendment 20, taken from the Industry Committee opinion9, may prevent software patents from being used to control communication protocols and data formats. This would avoid some of the negative effects of software patents on the information society, like monopolies based on forbidding compatible software and blocking of standards, which could otherwise lead to exclusion in the information society. Still, negative effects of software patents on freedom of expression for programmers, juridical certainty for entrepreneurs and developers would remain, as would obstacles to innovation and the negative economic effects of the monopolies resulting from software patents, but this is a necessary and important amendment.
Amendment 7 introduces a recital saying ``If the contribution to the state of the art relates solely to unpatentable matter, there can be no patentable invention irrespective of how the matter is presented in the claims.'' This is reasonable; it reflects a wish to prevent patent applicants from circumventing the limits to patentable subject matter by careful application wording. There is also talk of avoiding patentability for business methods and of clarification of the current law instead of change. Unfortunately, the rest of the JURI report does not fulfil these intentions.
The good points seem outweighed by the bad points, which would lead to unlimited patentability and lack of clarity, no matter the amount of good will declared:
The whole directive relies on the word ``technical''. It talks of technical effects, technical contributions and technical problems, but a definition of what is ``technical'' and what isn't is nowhere to be found. This term is very ambiguous if left undefined, and is commonly used to refer to accounting techniques, writing techniques, or marketing techniques. Amendment 12 from the Culture committee 10offered a useful definition of technical contribution, and Amendments 41, 43, 45, and 46 in JURI 11restrictively stated that in the sense of the directive, ``technical'' meant involving use of controllable forces of nature for an overseeable result, and clarified that data processing is not technical in that sense. Unfortunately, there wasn't enough support for these amendments in JURI.
The EPC defines innovations that are not inventions, and therefore not patentable. For instance, programs for computers, presentations of information, business methods, and mathematical methods are not inventions; they are explicitedly excluded subject matter. To assess whether something is patentable it is necessary to test that the innovation is in patentable subject matter, that it is new, that it is inventive (non-obvious), and that it has industrial application.
The JURI report only requires novelty, inventiveness and industrial applicability, not patentable subject matter. In fact, when defining ``computer-implemented invention'', a term contradictory with the EPC, it states that an invention may have features realised wholly by computer programs. JURI pretends they are not patenting all software but only proposes 3 limits instead of 4, removing the test that excluded programs for computers.
Article 4 states that the technical contribution ``shall be assessed by considering the state of the art and the scope of the patent claim considered as a whole, which must comprise technical features, irrespective whether or not such features are accompanied by non-technical features.'' That is, the claim needs to include something not previously known, and it has to include something technical. It is not necessary that the new knowledge contributed be technical, because what is compared is ``the scope of the claim as a whole'', and this whole can include non-technical features which represent the ``inventive'' step. In other words, even if a program for computers is considered non-technical, one could monopolise the use of a program by claiming a generic computer (technical, but not new) programmed with the program (new, but not technical), because there is no requirement that the contribution be technical, only that the claim contains a contribution and includes technical things. This is a move to remove the patentable subject matter requirement pretending that it is moved to another requirement. Amendment 5 also says ``If no technical problem is present, then the invention cannot be considered to make a technical contribution'', but this avoids the problem of whether the solution is technical. You can solve technical problems in new ways with old technical solutions and new non-technical features that do not deserve patents because they don't require the same investements as new technical solutions.
Amendment 18 allows claims for programs ``on its own, on a carrier or as a signal''. This does not make any more programs patentable. However, it increases liabilities for distributors of software. The Commission proposed to allow claims for programmed computers and running program processes, which is bad enough for software users and sellers of computers or other devices with preinstalled software. By allowing program claims, web sites offering software for downloaand software distributors move from contributory infringement to direct infringement. Therefore, program claims increase the legal uncertainty for Internet Service Providers (ISPs) and program resellers. The Commission proposal would only have scared away their customers or made them demand protection against liabilities from infringing on patents in their software. Conceptually and symbolically, introducing program claims is like announcing clearly that programs for computers will be patentable as such from now on.
Besides measures which are either good or bad, the report contains provisions that are in contradiction with each other or with other law.
Amendment 3, recital 7a, says ``The aim of this Directive is not to amend the European Patent Convention''. This is consistent with the intention to clarify and harmonise, instead of changing the rules unnoticedly. However, Amendment 25 asks the European Commission to report in three years of ``in what respects it may be necessary to prepare for a diplomatic conference to revise the European Patent Convention''.
It is obvious that the report, as well as the CEC proposal, contradicts the EPC, which says that programs for computers are not inventions and therefore cannot be patented as such. Amendment 18 allows ``A claim to a computer program, on its own, on a carrier or as a signal''. This contradicts article 52 EPC by removing the test for whether the claimed innovation is an invention or not. The amendment justification m clear that a `` computer program is a preferred embodiment of the protected invention'', as if the lists of non-inventions in EPC did not exist.
Article 3 has been removed. It said that ``computer-implemented invention is considered to belong to a field of technology'', which would make them patentable by the TRIPS agreement. Simultaneously, however, a new clause has been added to recital 12: ``a computer-implemented invention belongs by virtue of its very nature to a field of technology''. That is, the error has not been corrected, only moved around.
The report includes unfounded assertions, like recital 16, defending a false protectionism: ``The competitive position of European industry in relation to its major trading partners will be improved if the current differences in the legal protection of computer-implemented inventions are eliminated and the legal situation is transparent. With the present trend for traditional manufacturing industry to shift their operations to low-cost economies outside the European Union, the importance of intellectual property protection and in particular patent protection is self-evident.'' Allowing companies from other continents, which already hold the majority of software patents improperly granted by the EPO, to not only compete on cost, but also acquire software patents in the EU to exclude our own companies from our own market is just the opposite of what European enterprises need. European patents will not stop software development outside of Europe, where they have no force. What they will do is to increase the legal costs in the European market, which is not the best way to compete with low-cost economies. While the USA has had legal software patents and the EU has not, USA research and development intensity has decreased by 10 to 15 %12whereas the OECD13 has invested more in software R&D than has the USA. The JURI report would move this competitive advantage to developing countries without software patents, which is a form of solidarity quite destructive for Europe.
The report is too contradictory for any political view and fails in the stated aims of harmonisation and clarification, leaving for plenary almost all the work of amending the directive to come up with a clear and useful one. Since the JURI report disregards the opinions of the Culture and Industry committees, the Economic and Social Council of the EU, all economic studies, a petition by leading European computer scientists, another petition by 150000 citizens and 400 organisations and businesses, and much more opposition, we can only expect it enjoys little success in plenary.
In case the European Parliament plenary approved the JURI report, the consequences for Europe would be disastrous14: