Maybe there should be some burden of proof that many people have tried to solve the problem at hand but not come up with the solution in the patent application. For instance, if you invent a new sorting algorithm today, it would be reasonable to patent it; whereas if you invent a new voice activation scheme for a smartphone, it would be unreasonable to patent it because the whole field is relatively new.
This touches upon the biggest problem I have with the patent system: The bar is very low on the "novel and non-obvious" criteria. If that were used properly, there would be fewer patents, and they would be higher quality.
Yes, yes, yes!! I think this is the smallest change that could be made to the current system to make it sane.
The burden of proof needs to be on the patent holder to show, by a preponderance of the evidence, that they invented something nonobvious. That's a pretty low bar, but would require them to present some evidence, which they currently don't have to do.
In my limited experience with both successful and unsuccessful patent applications, the first response from the patent office is often a "nonfinal rejection" due to obviousness (section 103 rejection). In response, you should present evidence that the invention is not obvious. Here are examples of types of evidence that the USPTO finds compelling:
- commercial success (not due to better marketing).
- failed attempts by other people to solve a longstanding problem.
- awards or publications in the scientific literature.
- the invention involves a combination of components that produce an unexpected result.
Considering some of the patents I have seen, I would like to know what evidence of nonobviousness could have been presented for them. Perhaps one useful change would be to require that evidence to be part of the public record.
In any case, as I understand, once the patent has been issued, the courts generally defer to the PTO on the obviousness question. I think plaintiffs in infringement cases should have to demonstrate nonobviousness again in order to proceed. While it would be better for the PTO to be more of a stickler on this point, the fact is that we don't give them the kind of resources they would need to do as thorough a job as they should do on every patent.
For example, the PTO could maintain a team of engineers specifically for obviousness testing. When an application comes in, they would pose the problem the patent claims to solve to a selected panel of maybe 4 or 5 of them, who would have a couple of days to sketch some possible solutions. If they come up with the same idea as in the patent, that should pretty much kill it.
But as I say, we don't spend enough money on the PTO for them to do that. Since we don't, having the courts re-evaluate each patent on this point seems the only solution.
