Sunday, June 04, 2006
patent reform without changing anything
There has been a lot of talk on the web about fixing the patent system for software. And I think there is work to be done there. But a lot of the complaints people have could be fixed without changing anything.
Timliness. If you submit a patent now it takes a couple years to show up as published, and another couple years to be granted. Meanwhile the date it is valid to goes back to the date it was filed. Making it worse people can intentionally cause delays that keep a patent valid but unpublished for a lot longer. If we make this process reasonably fast at least we will be able to look at the patents before we infringe them.
Prior art. Slashdot mentions this over and over and over for some of the scarier patents out there, and their analysis it rubbish. But the fact remains that there is a whole lot of prior art that anybody in the field would be aware of that gets totally overlooked in granting many patents.
Non-Obviousness. It's really like this test isn't applied anymore to patent applications. Part of the reason is that with hindsight brilliant ideas often seem obvious. But if a person in the field says, "that's how I would have solved that, and here's why any college freshman would have solved it the same way" and the you give the problem to some college freshman and they indeed do solve it the same way, well throw that patent out.
How you solve all three of these is simply to pay your patent clerks more, so you get better ones, and to hire many more of them. Heck, activly recruit. Be at the college job fairs. If we had an adaquate number of well trained patent clerks the system would be 90% fixed.
Timliness. If you submit a patent now it takes a couple years to show up as published, and another couple years to be granted. Meanwhile the date it is valid to goes back to the date it was filed. Making it worse people can intentionally cause delays that keep a patent valid but unpublished for a lot longer. If we make this process reasonably fast at least we will be able to look at the patents before we infringe them.
Prior art. Slashdot mentions this over and over and over for some of the scarier patents out there, and their analysis it rubbish. But the fact remains that there is a whole lot of prior art that anybody in the field would be aware of that gets totally overlooked in granting many patents.
Non-Obviousness. It's really like this test isn't applied anymore to patent applications. Part of the reason is that with hindsight brilliant ideas often seem obvious. But if a person in the field says, "that's how I would have solved that, and here's why any college freshman would have solved it the same way" and the you give the problem to some college freshman and they indeed do solve it the same way, well throw that patent out.
How you solve all three of these is simply to pay your patent clerks more, so you get better ones, and to hire many more of them. Heck, activly recruit. Be at the college job fairs. If we had an adaquate number of well trained patent clerks the system would be 90% fixed.
Comments:
Post a Comment
