Slashdot Mirror
Could You Really Do Better than the USPTO?
lllama asks; "Is there such a thing as an obvious patent? Some ideas are so obvious that they needed to be stated before it's clear how obvious they are. Some are so blindingly simple that maybe a few dead monkeys might not have thought of them first. How would one go about deciding which ideas deserve a patent? Could/Should the patent office open itself up for public voting or moderation? Could a voting system affect how long a patent is grated for? Should the system be allowed to be swayed by public opinion? Should there be another tier of patents that protect IP but cannot be licensed out for money (BSD style patents?)? Given that simply hearing an idea affects how obvious it appears, could a system be devised that would allow an unbiased measure? An Obvious Quotient as such. Is there an obvious other word for 'obvious' that I could be using?"
You don't patent an idea...or at least you aren't supported to: you patent the execution of an idea - the 'how'.
Yes, the system should be allowed to be swayed by public opinion, or at least expert opinion: we make requests for comment to academia, for instance, all the time in plenty of areas of government, why not the patent system?
The Patent system was designed to reward innovation and further the public good. It's our way of conferring the 'king's monopoly' to those who do something useful. It's supposed to avoid uncreative people making cheap knock-offs of a good implementation and profiting that way, but, it's supposed to be only for a limited period of time, to encourage additional innovation. Given the speed of our economy, a 3-year patent cycle might be just the thing we need to boost it in the 'right way'.
Other words for obvious might be: trivial, uninteresting, derivative.
Final note: there's no reason why Patents have to be awarded so statically, or that the government couldn't charge a variable fee.
For instance, any product that is protected by patents could be required to pay a 5% patent-surcharge (minimum $1k/yr to keep the patent), instead of a static fee of a few hundred dollars.
Hire some qualified practicioners in each field (under strict NDAs). Give them a statement of the problem being solved. If they come up with the same solution within a few days, it's so blatantly obvious that the "inventor" is really trying to patent the problem instead of a solution.
That's a great idea! In fact, they could probably have something similar to the way TopCoder handles the architecture review boards. Basically they have lots of small component designs come in and need people to review them. These people to do the review are choosen at random from a pool of probably 30-40+ people. This could work in a similiar fashion by having a group of qualified people in each field and every day/week etc the USPTO office assigns one or two people a problem and see their solution in a few days. At TopCoder the people on the board are paid per-project so I imagine that this could be done the same way.
So, instead of necessarily hiring and paying full salaries to these people, they could just pay them on a per-project basis from a pool of qualified people that essentially do this in their spare time for extra cash. What a great idea, I think you should patent it!
Things you think are in the Constitution, but are not.
...and add:
All pre-existing hardware and software patents are revoked in 12 months. Anyone holding these patents is free to reapply (i.e. the existing model is screwed and bad patents have been awarded. Need to weed this out)
If it's an IT-related patent, maximum patent length is 3 years.
If patent enforcement creates a court-ruled monopoly situation, patent is revoked.
Patent queries relating to prior art are investigated as a priority. If existence of prior art is established, costs of investigation are borne by Patent Office. If not established, costs of investigation up to a certain level are borne by the person/corporation requesting the investigation.