3 ms·

I would also like to see a “cold room” exception. If you can show you invented the idea using completely independent thought, you don’t violate the patent and t

I would also like to see a “cold room” exception. If you can show you invented the idea using completely independent thought, you don’t violate the patent and the patent is invalidated. This seems to make a lot of sense at first glance. But how do you ever prove that you were unaware of the patented invention? I propose a different solution. The burden of proof should be on the patent applicant to show nonobviousness, rather than the burden being on the PTO to show obviousness, and objective evidence should be required. The kinds of objective evidence that could be supplied to show nonobviousness have already been delineated by the Supreme court: () commercial success () long-felt but unsolved needs () failure of others For software, if you managed to get a peer-reviewed paper accepted, I think that could also count as evidence of nonobviousness -- just how strong would depend on the prestigiousness of the conference or journal in question. Or maybe you could point to a passage in a textbook or someone else's published paper to the effect that the problem you are solving had been open for some time.

AnthonyMouse 14y ago
>() commercial success () long-felt but unsolved needs () failure of others I don't know how appropriate these are for software. The second one would almost always be impossible to show simply because of the recency and rate of change of the industry as a whole: How do you show "long" anything when the market segment has only existed for five years? Likewise failure of others; most of the problems haven't even existed long enough for anyone to have seriously tried and failed to solve them. And then commercial success goes the other way. Because software is so complicated, it infringes a zillion patents by accident, so is all you need to show commercial success that Android or iOS is allegedly infringing and you're done because they're commercially successful? Or do you have to show that the patent was the main cause of the success, which will then go back to almost never being the case? I suppose setting a nearly impossible standard that would result in a de facto abolition of software patents is one way to fix it, but I would much prefer just actually abolishing them. The alternative is to give the Federal Circuit yet another opportunity to read language broadly and expand the scope of patentability beyond what is reasonable.
- ScottBurson 14y ago
  > most of the problems haven't even existed long enough for anyone to have seriously tried and failed to solve them Bingo. Currently, it's my impression, a lot of patents are being granted for completely uncreative solutions to novel problems. New technology (multi-touch screens, for example) creates new technical possibilities; along with those possibilities come a bunch of new problems. People are coming upon these problems, solving them quite straightforwardly, and then patenting the solution. It is exactly this I want to put a stop to, because I think it's this that is doing so much mischief. Your second point is well taken. I think you'd have to show that the patent was a primary cause of the success -- which, as you say, is rarely the case. In any case, commercial success depends on a number of factors, and the courts would have to take that into account when weighing its probative value. I don't think it's quite "de facto abolition", but yes, it is both my belief and my intention that this change in the rules would rule out most current software patents. The few that remained would be for truly novel technology that would solve problems that were well known to be difficult. And those, I think, deserve to be patentable.
monochromatic 14y ago
The whole point (or at least one of the big ones) of getting patents is to cover people who *didn't" copy. Otherwise, copyright is often enough.

AnthonyMouse 14y ago

>() commercial success () long-felt but unsolved needs () failure of others I don't know how appropriate these are for software. The second one would almost always be impossible to show simply because of the recency and rate of change of the industry as a whole: How do you show "long" anything when the market segment has only existed for five years? Likewise failure of others; most of the problems haven't even existed long enough for anyone to have seriously tried and failed to solve them. And then commercial success goes the other way. Because software is so complicated, it infringes a zillion patents by accident, so is all you need to show commercial success that Android or iOS is allegedly infringing and you're done because they're commercially successful? Or do you have to show that the patent was the main cause of the success, which will then go back to almost never being the case? I suppose setting a nearly impossible standard that would result in a de facto abolition of software patents is one way to fix it, but I would much prefer just actually abolishing them. The alternative is to give the Federal Circuit yet another opportunity to read language broadly and expand the scope of patentability beyond what is reasonable.

ScottBurson 14y ago
> most of the problems haven't even existed long enough for anyone to have seriously tried and failed to solve them Bingo. Currently, it's my impression, a lot of patents are being granted for completely uncreative solutions to novel problems. New technology (multi-touch screens, for example) creates new technical possibilities; along with those possibilities come a bunch of new problems. People are coming upon these problems, solving them quite straightforwardly, and then patenting the solution. It is exactly this I want to put a stop to, because I think it's this that is doing so much mischief. Your second point is well taken. I think you'd have to show that the patent was a primary cause of the success -- which, as you say, is rarely the case. In any case, commercial success depends on a number of factors, and the courts would have to take that into account when weighing its probative value. I don't think it's quite "de facto abolition", but yes, it is both my belief and my intention that this change in the rules would rule out most current software patents. The few that remained would be for truly novel technology that would solve problems that were well known to be difficult. And those, I think, deserve to be patentable.

monochromatic 14y ago

The whole point (or at least one of the big ones) of getting patents is to cover people who *didn't" copy. Otherwise, copyright is often enough.