For Christmas a couple years ago I got my parents a laser pointer, wrapped in a printout of the U.S. patent on a ``method of exercising a cat''. This patent claimed a number of inventions related to wiggling a dot on the floor for a cat to chase around, all in lawyerly language with numbered drawings.
I'm listed as coauthor on a patent application that isn't much better, imho -- and worse in that it was meant seriously. I considered quitting my job rather than sign it, but fell for all the usual excuses: they'd just go ahead without me, and sure it's kind of bogus but they need it to defend against other bogus patents, and, hm, it'd be kind of neat to have any patent at all. It'd even me up with my online ex, for one thing. (She'd also thought her idea was just ``ordinary junk'' until her employer decided to ennoble it.)
So I'm not in a great position to complain about this recent patent
from Sun. I won't explain here why it's an especially awful one, but since it is, I'm especially happy to be able to invalidate it: there's prior art in section I.2 of Henry Massalin's ``Superoptimizer: a look at the smallest program'' in ASPLOS II (1987), page 125. I'm not impressed with Sun: the prior-art search consisted of a couple minutes looking for a well-known paper among printouts on my bedroom floor.
This passage has me particularly steamed:
While embodiments and applications of this invention have been shown and described, it would be apparent to those skilled in the art that many more modifications than mentioned above are possible without departing from the inventive concepts herein. The invention, therefore, is not to be restricted except in the spirit of the appended claims.
Which all sounds nice and professional until you notice that the `inventive concept' is just three lines of assembly code, and the ordinary skill they're coopting to fence in their property is the sort that should have made it unclaimable.
There's a twist: the original discovery was made by Massalin's computer, not Massalin himself -- he'd written a program to search for clever solutions to programming problems. So, if one of its findings could get patented, why not start a company, buy some serious
computing power (Massalin did his work with a 68020 back in the 80s), and crank out more of the same? Well, that's what John Koza is doing
-- which I used to think was cool. Now I wonder if it isn't yet another bug in the patent system: anything a computer can just barely discover today, its successor of the next year or two could discover routinely -- but patents extend way longer than that.
Something else interesting: what got pages of description in a patent merited just one short paragraph in an appendix to an academic paper.