5 ms·

This might be naive on my part but it seems like we could help the situation quite a bit, at least for software patents, by just reducing the amount of time for

This might be naive on my part but it seems like we could help the situation quite a bit, at least for software patents, by just reducing the amount of time for which new software patents are valid. It's simple, easy to understand and implement, more consistent with the speed at which the software market evolves, would reduce the perceived value of bogus patents making trolling less likely, free up technology sooner for everybody to benefit and still offer some protection to innovators. Right now they're good for between 14 to 20 years (1) which seems too long to me. I'm not sure what the right number is, maybe between 5 - 10? Even that seems long, but then again I don't have anything innovations to patent so I'm probably biased. (1) http://patents.stackexchange.com/questions/312/how-long-are-patents-valid http://patents.stackexchange.com/questions/312/how-long-are-...

transfire 12y ago
It's not a good idea. Ideally rights should be for as long as the life of the inventor. Instead of a cut-off time, I've heard people espouse a system of valuation and, in some cases, taxation, so that over time a patent becomes increasingly affordable for others to buy out and put into the public domain.
- AnimalMuppet 12y ago
  What is the basis for your "should" here?
- rhino369 12y ago
  >Ideally rights should be for as long as the life of the inventor. I'm fairly pro patent, but I disagree. Unlike copyright where there is an almost unlimited amount of expression available, some inventions are just something people would independently come up with. Nobody would organically and independently come up with Mickey Mouse or Sherlock Holmes. But different teams would have invented the BJT transistor if Bardeen, et al. never existed. Patents length acknowledges that by not giving exclusive rights forever. It's more of a challenge prize than a pure ownership of the idea indefinitely.
- gombotzd 12y ago
  i have to disagree to your statement, which misses foundation. First copyright is always a life long idea, but the focus of patents have always been to incentivice long term research efforts that have budget constraint by having a mid-term monopoly problem today: a lot of trivial design-patents and even patents that are used in the wars between big corps. as one of my advisor board state - screw that patent, we cannot defend it anyway Behind a lot of those patents there is no research in the sense but it rather came out of "usual" engineering. i think that the proof of non-trivial has to be a part of the application and is better suited than discussion about life-time - because patents could still have a strong point in basic-research Problem is what to do with the already existing bogus patents - any suggestions?
  - btilly 12y ago
    First copyright is always a life long idea... Not true in the USA until Disney started twisting the law. This idea has a longer history in Europe. See http://www.arl.org/focus-areas/copyright-ip/2486-copyright-timeline#.VOvVLUjmv0g http://www.arl.org/focus-areas/copyright-ip/2486-copyright-t... for a more detailed history.
- frederickf 12y ago
  Ideally from the perspective of the inventor :) But that's an interesting idea. In that system how is the valuation of an innovation determined?
- sigzero 12y ago
  I would think 7-10 year as a MAX to ensure that they could profit.
- pickle27 12y ago
  I had an idea similar to this where when filing for a patent you include an estimate of R&D costs. If the patent is awarded then its lifetime is related to how long it takes to get a reasonable payout on the R&D investment. This way big pharmaceuticals put a couple million into a new drug and they get to profit for a few years. An Apple engineer makes slide to unlock in a couple days and given the number of iphones sold the patent is over quite quickly. tl;dr a salary cap for patents based on R&D cost
raquo 12y ago
3-4 year limit on software patents would certainly help. It puts the incentives closer to where they should be, and is a straightforward fix too, unlike revamping / abolishing the whole patent system. It certainly doesn't fully address the fundamental problems of the system, but it could be a reasonable first step.

transfire 12y ago

It's not a good idea. Ideally rights should be for as long as the life of the inventor. Instead of a cut-off time, I've heard people espouse a system of valuation and, in some cases, taxation, so that over time a patent becomes increasingly affordable for others to buy out and put into the public domain.

AnimalMuppet 12y ago
What is the basis for your "should" here?
rhino369 12y ago
>Ideally rights should be for as long as the life of the inventor. I'm fairly pro patent, but I disagree. Unlike copyright where there is an almost unlimited amount of expression available, some inventions are just something people would independently come up with. Nobody would organically and independently come up with Mickey Mouse or Sherlock Holmes. But different teams would have invented the BJT transistor if Bardeen, et al. never existed. Patents length acknowledges that by not giving exclusive rights forever. It's more of a challenge prize than a pure ownership of the idea indefinitely.
gombotzd 12y ago
i have to disagree to your statement, which misses foundation. First copyright is always a life long idea, but the focus of patents have always been to incentivice long term research efforts that have budget constraint by having a mid-term monopoly problem today: a lot of trivial design-patents and even patents that are used in the wars between big corps. as one of my advisor board state - screw that patent, we cannot defend it anyway Behind a lot of those patents there is no research in the sense but it rather came out of "usual" engineering. i think that the proof of non-trivial has to be a part of the application and is better suited than discussion about life-time - because patents could still have a strong point in basic-research Problem is what to do with the already existing bogus patents - any suggestions?
- btilly 12y ago
  First copyright is always a life long idea... Not true in the USA until Disney started twisting the law. This idea has a longer history in Europe. See http://www.arl.org/focus-areas/copyright-ip/2486-copyright-timeline#.VOvVLUjmv0g http://www.arl.org/focus-areas/copyright-ip/2486-copyright-t... for a more detailed history.
frederickf 12y ago
Ideally from the perspective of the inventor :) But that's an interesting idea. In that system how is the valuation of an innovation determined?
sigzero 12y ago
I would think 7-10 year as a MAX to ensure that they could profit.
pickle27 12y ago
I had an idea similar to this where when filing for a patent you include an estimate of R&D costs. If the patent is awarded then its lifetime is related to how long it takes to get a reasonable payout on the R&D investment. This way big pharmaceuticals put a couple million into a new drug and they get to profit for a few years. An Apple engineer makes slide to unlock in a couple days and given the number of iphones sold the patent is over quite quickly. tl;dr a salary cap for patents based on R&D cost

raquo 12y ago

3-4 year limit on software patents would certainly help. It puts the incentives closer to where they should be, and is a straightforward fix too, unlike revamping / abolishing the whole patent system. It certainly doesn't fully address the fundamental problems of the system, but it could be a reasonable first step.