Oh I love it when people say innovation is stifled by patents, when it is self-evident that the products they are on about are available today on the market, and innovation rages on all the time.
Open source products are great but 9 out of 10 are imitating something that someonbody else had to invent first. If it's patented and that people can't just copy it for free, well that's what patents are there for.
Trolls and the rest are just part of the innovation ecosystem by which engineers and scientists get paid jobs. If you can understand Rambus wanting to monetize non-core patents, why condemn the company that provides them the route for them doing that? Each patent represents a past investment in R&D, and each future patent represents a future investment in R&D.
The basic issue is the USPTO cannot hire those "skilled in the art" for most fields, ie the examiners should, but cannot see what is *obvious*. This is especially true in the software/firmware field. It takes (my observation) about 12 years of schooling/practice to become "skilled in the art" of software. Firmware takes longer, sometimes, because of the restriction of resources (RAM, FLASH, etc).
The net effect is the flood of bad/silly/restricting patents. It's like going into a hardware store and patenting every nut, bolt, and washer in the store and everything that uses them. throw in "methods of applying chemicals to walls for the purpose of coloring and preserving surfaces" (ie paint and stains), and it becomes tragic to innovation.
Very large corporations use that very same logic after you demonstrate a new patent to them and the concept filters around their various offices. At that point you need a rich uncle to even get them to a court.
Iíve never given a patent to a Troll for less than it cost me to get the idea patented, but Iíve been tempted. Usually if you can tell a Troll where your patent is being illegally used, you can get at least enough to cover your costs.
Wow! We didn't even get into submarine patents! Fortunately, that one got resolved. There's hope, at least.
I like to use open source file formats in my designs. It keeps us and the customer out of gray areas. The documentation is nice to have, too.
fundamentals wrote: " put "A" in system "B". Neither "A" nor "B" can be patented any more, they have been around for a while, possibly patented sometime in the past. But nobody put "A" in "B" before, so the company who does it first tries to patent it and almost always succeeds.
Most of these type of patents should be rejected because they are obvious to an "expert in the field of invention", which the criterion of obviousness is supposed to be judged by. Unfortunately the patent examiners are very rarely experts in any field of invention, so they approve patents almost automatically."
This is utter nonsense. You have described the exact basis for a patent application rejection under 35 USC 103.
And you didn't even get it right. It's not "expert in the field of invention." It's a "person having ordinary skill in the art." Not an expert.
The patent system is broken because the Patent Office issues far too many obvious patents. There is an infinite variety of obvious patents, but the most common one goes as follows: I put "A" in system "B". Neither "A" nor "B" can be patented any more, they have been around for a while, possibly patented sometime in the past. But nobody put "A" in "B" before, so the company who does it first tries to patent it and almost always succeeds.
Most of these type of patents should be rejected because they are obvious to an "expert in the field of invention", which the criterion of obviousness is supposed to be judged by. Unfortunately the patent examiners are very rarely experts in any field of invention, so they approve patents almost automatically.
One way to solve this problem could be that the patent office should solicit opinions from a group of experts (who are not involved in the patent and who do not work for the same comany seeking to patent the idea) about the obviousness of the patent in question. Once that is done, I am quite sure that the number of patents issued will decrease by an order of magnitude. What is issued will be original and non-obvious. Obviously these type of patents should be enforced, and if they happened to be enforced by patent trolls, I'm OK with it.
What matters is not who owns the patent, but what the patent says, if anything at all. The problem is not in enforcing patents, but in enforcing patents which should have never been issued in the first place. Unfortunately, under the current system, going to court with the intent of invalidating a patent can be very very costly. The legal system (and the patent office) do not like to appear as fools for having issued a patent that should never have been issued. Because of that, they always weigh very heavily in favor of the patent holders and aginst challengers. The patent trolls know this very well and they take advantage of it.
It would be nice if the public could in effect be used. We have seen cases of this where the public is paid to find reasons why a patent should not be granted, uncover prior art etc. When a patent is published - make it open for external comment, review and challenge. Allow people to challenge the validity before it becomes a patent rather than after. Make sure that the patent search system is as good as it possibly can be and that it is easy to get copies of patents. That in my mind would be a good first step.
This can be taken even further. The person making the clothes may have come up with the idea first, then receive the nasty letter after the USPT office has granted a patent to someone else and sues you. You now have to pay to defend yourself and strike down the patent that should not have been issued in the first place.
Blog Doing Math in FPGAs Tom Burke 18 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...