5 ms·
That is definitely a good idea, but what about the patents that are so blatantly vague or obvious, that many people are unknowingly violating them. It seems it
by phene 14y ago
That is definitely a good idea, but what about the patents that are so blatantly vague or obvious, that many people are unknowingly violating them. It seems it's easier to prove non-infringement rather than prior art with most software patents.
- gbhn 14y agoThe patent grant puts the burden of proof on the alleged infringer, on the theory that a granted patent has been pre-vetted for all the right criteria. But when I look at patent litigation, it seems (and perhaps this is a selection bias problem) that in most cases, about 90% of the patent claims get immediately thrown out. Doesn't it seem that the presumption that the patent grant is a strong enough process to force the burden of proof onto the infringer is just a social mistake? If the burden of proof shifted to the patent holder, it would totally change patent litigation, while maintaining all the abilities we applaud about the patent system -- the ability of a patent holder to sue an infringer who stole their idea, and get compensated.
- regularfry 14y agoPatent claims get thrown out because that's how patents are designed. The idea is to break your invention down into an onion of claim layers so that when it comes to court, if a claim is thrown out as being partially or totally invalid, it's just like peeling as small a layer off the onion as possible. The point is that you can't be sure exactly where the court are going to stop peeling, but by breaking it down into layers you can be more sure that there'll be something useful left when they do, and you minimise the risk of their going too far. I don't think that strategy is dependent on which direction the burden of proof lies.