UK Rejects Software Patents That Don't Actually Describe Process
from the a-decent-test dept
One of the reasons that defenders of the current patent system often give for the importance of keeping the system, is the idea that it benefits society by revealing ideas that otherwise would be kept secret. That is, indeed, one of the benefits... if those ideas wouldn't have come out otherwise and if the publication of the patent actually revealed anything about the invention in question. All too often we've seen that this second part is missing. Patents describe something extremely broad and general, so as to reveal as little as possible, but make sure as much as possible infringes on it. It seems like it would be a reasonable "test" of the validity of a patent to see if someone could use just the patent itself to replicate the invention in question. It seems like some judges in the UK agree. The Court of Appeals in England has ruled against two software patents, specifically noting that someone who knows the basic subject area should be able to replicate what's described in the patent with just the patent alone -- and these patents fail that test. In fact, the ruling said: "that the patent was missing vital details, contained wrong equations, demanded a higher level of expertise than allowed and that it relied on material external to the patent." In other words, the patent itself did not actually explain the invention it was trying to cover, but rather hinted at what it might be. With the huge awards for patent infringement going around, an increasing number of patents seem to be written the same way -- so it would be nice to see more examinations of patents that look at this particular factor and whether or not the patent actually does try to explain the invention for others to use, or simply word it broadly enough to catch random infringers.