The past week has delivered yet another example that our patent system is horribly broken. On April 9, a jury found that Microsoft infringed on Lucent/Alcatel touch-screen patents. But wait! On April 13, news stories reported that a March 26 re-examination by the US Patent Office said aspects of the patent were "an obvious variation of know-how that was public."Geez, thanks a lot Mr. Patent Examiner. I don't know all the ins and outs of examining patent claims and issuing a patent, but it seems like you should figure out if the claims are obvious before issuing the patent! Once a company is granted a patent, they can -- and often do -- wield it like a club, selectively bopping either their competitors or any company with deep pockets. Microsoft, of course, has deep pockets, so lots of patent holders can find something Microsoft is doing that could infringe on their patent -- especially a patent that shouldn't have been issued.

Once a patent is issued, it's much more difficult to revoke. Armed with a patent, the holder can start the slash-and-burn process of getting royalties and concessions from the "infringers" until someone can convince the Patent Office to re-check their work. That doesn't stop the fun completely though; during the re-review process, the patent is generally assumed to be valid. Imagine how much money has been spent already on this crazy process. Lawyers, of course, will walk away from this experience much richer regardless of the outcome.

One of the weaknesses in the current system is that an invention is supposed to be considered worth of a patent only if it is not obvious to a person having ordinary skill in the art. However, the examiner may not meet that requirement well enough to be able to judge whether an invention is patentable! Professor Dan Wallach wonders whether whether we should also try to get juries of "technical peers", but acknowledges that it might be hard to find experts without a conflict of interest.

The jury trial is late in the process, after the patent has been granted. Catching bogus patents earlier would be better. One heartening development in the process has been the Peer-to-Patent Project, which opens the patent review process to the Internet. Companies, academics, and other interested parties can see the patent applications and provide examples of prior art or opinions on the invention's obviousness. We need more true peer review to ensure that "obvious" inventions don't turn the patent process into a winner-takes-all lottery.