23 juni 2005

Open and free (The Flemish Beerdrinker)

Richard Stallman, founder of the Free Software Foundation (free as in freedom), urges for action against the new European software directive. Shortly the European Parlement will probably vote to allow patents on software. He draws an analogy with novels. While copyrights only covers the details of a novel and only restricts copying, patents covers idea’s. It gives the author a monopoly on the idea’s embodied in his novel. Literary progress would stall, because other novels could infringe on many different patents at once. The same we see in software where one programme can infringe many patents (Linux infringes 283 different US software patents). And there are companies that do no other thing then to produced lawsuits against patent infringment. I believe Stallman has a point. Patents are too broad based and are awarded too easily. Government should be very carefull with granting monopolies. But he does not prove that we have to ditch the patent system, maybe reform is enough. Adam Jaffe and Josh Lerner suggest as much:

It is surely true that there was software innovation before software patents were widely used. As with all other technologies, it is unlikely that software development would grind to a halt without patent protection. And it is also true that software innovation is a highly cumulative process. But the reality is that virtually all innovation is a highly cumulative process, and the patent system has been struggling with the tradeoffs that implies for a long time. The relevant question is: on balance, would a properly administered regime of software patents foster innovation, by allowing parties that make true breakthroughs a measure of protection to reduce the risks of commercializing that development? As with business methods, we haven’t had a test of such a system because the PTO has failed to implement the requirements of novelty, nonobviousness and enablement. If the overall patent system were reformed as we have proposed, the only software that would be patentable would be that which truly represents a non-obvious step forward, and the implementation of which is laid out in some detail. Granting patents of this sort would not stop others who wish to work within the open-source paradigm from doing so, and would not prevent open-source advocates from arguing their case and trying to convince computer users not to buy patented software. It may be that the advantages of open-source development are sufficient that many or most software developers would choose to forego patents and work within the open-source paradigm instead. If the PTO were doing its job properly, any software that is developed and published freely by open-source advocates or other programmers could never itself be patented, because no one could ever claim novelty in having created it. So a properly functioning patent system is not inconsistent with a vibrant open-source software movement. The real enemy of open-source software—and software innovation more generally—is poor implementation of software patents, not the concept. The real question is whether a programmer who has a truly new software invention ought to have the option of patenting it rather than making it open-source. No one has put forth a convincing argument why that choice should not lie with the innovator, rather than being made for the entire industry as a matter of law or policy.

Free as in freedom of choice. Law should not allow broad patents but neither should the law limit the freedom of choice of the innovator. So Stallman should not take his crusade too far.

Next we have Michael Eisen, a computational biologist, and founder of the Public Library of Science. He is launching three new "open access" scientific journals, because he believes that the published results of publicly founded research should be freely available for everyone. He talks about "forces of darkness" that are the scientific publishers.

Apart from the hyperbole i do agree with Eisen’s message. It won’t do to let the publishers make money thanks to publicly funded research. That is called corporate wellfare, and wellfare is not destined for rich corporations.

<<Oudere berichten     Nieuwere berichten>>