I see the problem really as abuse of the system and inadequate resources. I doubt that there are very many government institutions that are not somehow broken. That's just a fact of life. Our best attempt to control that to date is the system of free elections.
The abuse of the system part is really disheartening and I certainly don't have an answer to that. We get that same kind of abuse with car title loan companies, for profit schools and a lode of other places.
I would think that technology would be able to help with the resources issue. Given how much of the world's knowledge-base is connected to the Internet, being able to electronically qualify or disqualify a patent application I can see automated validation as being very achievable. It would not surprise me at all if 90% of all patent applications could be managed automatically. If that were the case, a lot of resources would be freed up to deal with the difficult patents.
This is not a question of stealing clothes. It's more a situation where you make your own clothes, and then receive a nasty letter from an NPE insisting that you pay them for using a technique that you discovered yourself without any knowledge whatsoever of the NPE's government-granted monopoly.
In a fair world, reinventing something would give you immunity from a lawsuit, or better yet be proof that the original patent is obvious and thus invalid, but that's not The Way of the World these days.
It is easy to say that the patent system is broken and that NPE's are just a burden, but I am not seeing any 'solutions' that address the core problems.
The reality is that, in the long term, technology development takes time and effort invested by many skilled individuals -- and those skilled individuals need to be able to pay for their own living expenses. Corporations generally cannot continue to pay salaries unless they gather funds somehow. 'Practicing entities' gather funds by selling products that leverage lots of technology from multiple sources -- some internal and some external. 'Nonpracticing entities' are effectively 'malls' that don't personally produce goods, but facilitate 'producers' gathering desirable/required technology components.
As in the case with malls, some are better, more ethical, cheaper, more efficient than others. However I would not advocate that we abolish malls and only allow manufacturers to sell products -- which is the PE/NPE (patent troll) argument. I would also not advocate that we steal our clothes because we find that to be simpler and cheaper -- general position against patents. In general, I would argue against any store being allowed to charge for air and/or water. There are examples, though, where it makes sense to allow stores to even sell air (e.g. compressed air) and/or water (e.g. bottled water).
By all means, encourage people to contribute goods and technology that they have legally acquired/produced -- just don't expect and/or limit people to only giving things away.
Obviously the USPTO needs to undergo a reform to prevent examiners from granting invalid patents in the first place. A change in the law could limit the rights of NPE's vis a vis original patent holders. It's also worth noting that a troll, unlike a practicing entity, has nothing to gain from fighting a patent in court, so will prefer to reach a settlement. Risking a court judgement not in their favour is a net loss.
I share Brian's frustration but I just can't see the light at the end of the tunnel, yet. It's not just a US problem, either, becuase what happens there has very strong global influence.
The patent system was meant to help innovation but its purpose was forgotten along the way and now it's just a bad caricature of itself. It's ripe for reform. Ideally, it should bring benefits to the whole industry and society while not forgetting about the developer, of course.
What about some sort of global cooperation between countries?!
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 ...