Bilski Is(n’t) About Software Patents July 10, 2009
In Podcast 2, we spent a good amount of time talking about Bilski, a case the Supreme Court will hear next year about business method patents. Dominik went to some length to make sure that we knew that Bilski was not a software patent case, which is isn’t.
However, the standard in Bilski was recently applied to a software patent, and the judge found the patent invalid.
Bilski, and the fate of software and business method patents are in the hands of the Supreme Court now. We don’t expect a decision before December, at the earliest.
Related posts:
- Martin Goetz’s Perspective on Software Patents This afternoon I came across Martin Goetz’s guest article on...
- Bilski at the Supreme Court Today Today, the Supreme Court hears arguments on the Bilski case....
- Still Waiting on Bilski The new story about the Bilski case is how long...
- No Bilski Decision from SCOTUS Yet At Technically Legal we’re all waiting with bated breath for...
- Bilski: Business Method Patent Invalid Today the Supreme Court handed down its long awaited opinion...
Posted in: Links by Ben Snitkoff.
What would happen if someone decided to attempt to patent a device that took as input a computer program, used the Curry-Howard isomorphism to translate it into pure math, and printed it (or whatever is required to make it arguably patentable as a ‘transformative’ process)?
http://en.wikipedia.org/wiki/Curry_Howard
I have to wonder if the patent examiners’ heads might explode, given how I’ve seen patent lawyers arguing that software isn’t math, even though there’s an isomorphism and the proof is constructive (i.e. tells you how to translate between software and math).
That would actually be patentable (as long as it was novel and non-obvious), because it is a device. Bilski problems crop up when the thing you are trying to patent isn’t connected with any device.
The mere fact that it’s a device that takes an input and makes a tangible output would be sufficient to make it patentable subject matter.
Now, if I tried to patent just a computer program that applied the isomorphism and spat out the formula that probably wouldn’t be patentable. Though, it’s a tough call if that isn’t transformative (because it is very much transforming something) or of it’s just obvious to make a program that implements something like that.