There is a thoughtful article by Todd Bishop in the Seattle Post-Intelligencer on the Microsoft application to patent a computer method of conjugating verbs. It is thoughtful not only because Todd quotes my earlier post, and talked to Preston McAfee and Mike Masnick, but because he also talked to the people at Microsoft. So it makes sense to discuss: is this kind of patent a good idea or not? Microsoft argues that there is innovation that seems to largely revolve around switching from one language to another. No doubt compared to other computer programs this is an innovation. However, as Preston correctly remarks - it is a feature of every language textbook ever written. So does it make sense to allow every existing idea to be patented the first time it is ever done on the computer? Obviously the only "innovation" is in the specific computer code that achieve the purpose - but that isn't what is patented - anyone else implementing the idea would likely write somewhat different code, and probably wouldn't benefit that much from the "innovator"'s code. Moreover, the purpose of the patent system is to encourage innovation. On the one hand the cost of Microsoft's "innovation" is trivial - you and I could whip up some code in a few hours. On the other hand, the negative impact of the patent on innovation may be substantial - anyone who wants to write a computer translation program or textbook program now has to contend with the Microsoft patent - and how much worse if I take nouns? So regardless of whether Microsoft deserves to be a laughing stock for patenting such an obvious idea, issuing and enforcing patents like this cannot possibly be a good idea.