Thanks to a scan of my Netvibes feed page I ran across a very interesting item called Are Software Patents Evil? by Paul Graham. This is not one of those brief stream-of-consciousness blog entries, this is a serious piece that reviews and comments on the serious issues surrounding patents, software patents in particular.

Graham starts by commenting on the current state of software patents. After stating ".. if you're against software patents, you're against patents in general." He continues:

I think the problem is more with the patent office than the concept of software patents. Whenever software meets government, bad things happen, because software changes fast and government changes slow. The patent office has been overwhelmed by both the volume and the novelty of applications for software patents, and as a result they've made a lot of mistakes. ... The most common is to grant patents that shouldn't be granted.

The "poster boy" for "patents that shouldn't be granted" is the Amazon "one click order" patent. That's one that makes a lot of software engineers roll their eyes. The problem is, as with other areas of intellectual property law, there's always an opposite view held strongly -- and loudly -- by another interest group. The arguments tend to be either very legalist on the one hand, or very philosophical on the other.

Graham touches on some of these arguments but does not succumb to them. He gives a good rundown of the major issues involved with software patents and, even though he has a definite point of view, I think his piece deserves a read by anyone interested in how software patents are created and used.

A refreshing things about Graham's article is the focus on how patents are used in the "warfare" of business. Graham describes how some major companies probably have many more software patents than they'll ever enforce -- but they hold on to them just in case. He seems to be saying that, despite the "temporary monopoly" given by a patent, the reality is that in many cases these "temporary monopolies" are not being enforced, and this is A Good Thing, otherwise business could grind to a halt, legal bills would mount, and companies would turn to secrecy, not patents, to protect their innovations.

Graham also lists a series of reasons why software patents are not as important as you might think, mainly because software is so complex and the boundaries associated with the rules and transactions that describe software use are so difficult to define sometimes. He views this as a good thing, and I'm inclined to agree.

That's not to say that the process by which software patents are reviewed and issued shouldn't be improved, it should. An interesting and creative example of a change to the process is the Peer to Patent project. But I think that Graham's piece is a refreshing change from the Byzantine religious and philosophic arguments that intellectual property issues often stimulate.