The EC has commissioned a report on the impact of software patents on innovation, but is pressing ahead with plans to introduce the patents in Europe without waiting its conclusions.

A team of researchers at the University of Maastricht in the Netherlands began their three-year study last December, examining the legal, technical and economic effects of software patents on software innovation.

The study will take a broad view of the economic consequences. "We are doing a survey not so much on patents as on innovation, and not just software," said Rishab Aiyer Ghosh, program leader for studies of free and open-source software at the Maastricht Economic Research Institute on Innovation and Technology (Merit).

The study will not finish until late 2007, although the Merit team will release an interim report on its findings about patents: "We will be publishing something towards the end of this year," he said.

Even those interim findings on software patents may arrive too late to influence relevant European legislation, however.

While Ghosh's research is supported by the Commission's Directorate-General for the Information Society and Media, another branch of the Commission, the Directorate-General for the Internal Market, is pressing ahead with hotly contested legislation that will harmonise the way in which software can be patented throughout the 25 member states of the European Union. EU laws, or directives, take precedence over the national laws of EU member states.

Does it seem odd that the Commission would ask for a report on whether software patent legislation is good or bad for innovation, and then not wait for the answer? "It does," said Ghosh.

Today, applications for patents on software inventions are treated differently by national patent offices in the EU member states, and by the European Patent Office (EPO), which serves Bulgaria, Iceland, Liechtenstein, Monaco, Romania, Switzerland and Turkey in addition to most of the EU member states. Some of the countries (and the EPO) allow software to be patented on its own; others allow it to be patented as part of a machine - a computer-implemented invention - while others don't allow it at all.

The Commission wants to harmonise this treatment in member states by introducing a law on the patenting of computer-implemented inventions.

The draft law proposed by the Commission met stiff resistance in the European Parliament. Dissatisfaction with the project was so great that in February, Parliament called for the drafting process to start anew. However, the Commission got its way: the draft, heavily amended, made it through its first reading, and will shortly be presented for a second reading.

The Commission is interested in the economic role played by open-source software, and this week announced it will fund another study of the topic, Flossworld, to the tune of 663,000 euros. The two-year global study will look at the effect on regional economies of the use and development of free, "libre" and open-source software (FLOSS). The French word libre is used to distinguish the intended meaning of free - without restriction - from its other sense in English, without charge.

The study brings together EU researchers and others at partner institutions in Argentina, Brazil, Bulgaria, China, Croatia, India, Malaysia and South Africa. Those partners will receive EU funding; others involved in the project, in Japan, South Korea and the US, will have to provide their own.

Among the things Ghosh hopes Flossworld will discover is how developers around the world contribute to open-source projects. For example, he wonders, do programmers in, say, China spend most of their time localising code developed elsewhere, or are they making original contributions to projects? Such answers could be found in a study of source code being conducted in Madrid, he said. Flossworld will report back in April 2007.