I'm not sure I agree that patents really help progress.
As a founder of several small companies, I have a somewhat different perspective. FWIW, I have only applied for (and received) 1 patent. It was for work I did when I was 20. Since that time, I try to protect some ideas as trade secrets instead.
First, I don't think its hard to have many patentable ideas. The hardest part is posing a problem that needs to be solved. Most of the patentable ideas that I have came from looking at a new application. I remember a Saturday afternoon where 3 of us where brainstorming on a project where I think we came up with maybe 20 patentable ideas (or at least candidates). It wasn't that these ideas were earthshaking. We were looking at an application that only a few companies were thinking about and one was our customer, another was their partner. When I see someone with 100 patents, I don't see someone who is exceptionally brilliant (might be - might not be). I see someone who works for a large company.
The problem with patents is that they are very expensive to obtain and incredibly expensive to defend. They are mostly used by large companies to monopolize markets. The large companies cross license to each other and keep everyone else out. If I had to defend a patent against a large company, I would have to sell my company or at least the rights to the patent to someone else. The cost of winning would be prohibitive and losing would be worse.
Maybe the only reason for a small company to pursue patents is if they want to sell out to a larger entity (harvest).
Software patents are the worst. Most prior art is not clearly exposed to the general public.
I'm not sure where the partitioned convolution part of this thread started out. I have read the Microsoft partitioned convolution patent and I can't tell what they actually think they patented. Most of their claims are already expressed in the prior art they list.
The idea that someone should be able to hold the world hostage for 17 years is crazy. That is forever in today's technological world. If I have a patent, there is no specific requirement for me to do anything with the patent at all. I can just sit on it and theoretically freeze the idea for the length of the patent. I don't have to let anyone else license the idea, etc.
Of course, non obvious ideas as Nigel mentions can be expensive to create. The catch is that 99% of the patents I have seen are not that special. They are awarded by reviewers who probably don't know what prior art exists outside the patent record and almost certainly haven't been long time product designers in the field they are reviewing. More likely, they are 20 somethings that went to college and just got a degree.
Anyway, this is just another rant. Nothing is likely to change. Al Clark Danville Signal On 1/28/2011 3:47 PM, Nigel Redmon wrote:
I've been on a number of patent cases (as software expert, sometimes electronics), big players, on both sides... First, patents are important, and help progress. Non-obvious advances often come from expensive and lengthy research. Imagine a situation where company A invests in research, and makes a breakthrough. Imagine companies X, Y, and Z, who wait for researchers to make a breakthrough, then flood the market with products that compete with company A. Company A must recoup their investment, so it's unlike they can compete with the research-free companies, so they go out of business. Who spends the big research money next time? Of course, the problem is that patents are being given to trivial inventions. They are also often given to things that have been in use for some time, but never patented; although this falls under "prior art", and should invalidate the patent, good luck on coming up with the funds to fight the patent if it's owned by a major company.
-- dupswapdrop -- the music-dsp mailing list and website: subscription info, FAQ, source code archive, list archive, book reviews, dsp links http://music.columbia.edu/cmc/music-dsp http://music.columbia.edu/mailman/listinfo/music-dsp
