23 February 2009

Patently Not the Case

Professionals who work in the field of intellectual monopolies have a problem. Most of them are quite able to see there are serious problems with the system, but since their entire career has been built on it, they can hardly trash the whole thing. Instead, they not unreasonably try to come up with a "reasonable" compromise. Here's a good example:

how about a world in which there are fewer patents, but better ones — in the sense that they are more carefully examined, forced to comply more strictly both with legal criteria and market reality? That stricter compliance would make them more likely to be valid, in a world in which software inventions are under constant threat of being deemed retrospectively obvious, and useful to read.

But it's based on a false premise - that we actually *need* patents for business reasons:

Others praise patents, even for software applications. The patent protects investment that would not be directed at software development if that protection did not exist. The patent specification opens up inventions for everyone to read, thus enriching the state of the art and saving developers the need to reinvent the programmer's equivalent of the wheel.

This is patently untrue for the software world, which happily invested in software development for decades before software patent madness took over. Microsoft is actually a good example of a company that succeeded without feeling the need to try to patent its software, in the early days at least. And Bill Gates famously noted that the success of his company would actually have been impossible had software patents existed and been held by rival companies.

The second point is also manifestly untrue. Software patents are almost always totally generic - they do not "open up inventions for everyone to read": instead, they are written as opaquely as possible in the hope that they can be bent in court to apply to the most extreme situations. They do not "enrich the state of the art", because the whole purpose of gaining patents is to stop anyone using them for 20 years, by which time, technology has moved on so far that any content they originally had has been superseded.

There are simply *no* good reasons for software patents, and hence no justification for halfway houses, however reasonably framed, and however intelligent and reasonable the framer.

No comments: