[lyndonh]

The author doesn't understand the patent system.

The government, via the patent office, is selling a monopoly on technology. If someone writes a frivolous application the patent office usually takes the money and grants it.

Does the author think the patent office gives refunds ? There is no incentive for them to stop granting bad patents. While this situation persists, it makes financial sense to apply for a patent if you think you can get it.

It's a self perpetuating cycle; it's easy to get a patent -> many people apply -> there are too many applications to examine thoroughly -> more people apply -> etc.

Killing patents, as the author suggests, is a waste of time. The USPTO is not going to be more rigorous when checking patent applications. The proper solution is to make it very easy to strike down bad patents and stop accepting them in the first place.

[pbhjpbhj]

>There is no incentive for them to stop granting bad patents. //

The problem actually doesn't lie with the USPTO specifically, it lies with the entire legal system.

The patent is the applicant's. They should need a patent that's guaranteed as much as possible to be valid. It should be that them getting a patent that's not valid means they can't use that patent for anything. They should therefore be closely allied with the patent examiner who is trying to "grant patents with a high degree of validity" (that's the UKIPO patent groups motto, or it was several years ago). They also want to get the broadest monopoly possible, that's the counterweight, but if it's granted so broadly as to be easily anticipated by a document that a skilled proponent in the field can find in 15 minutes of searching then the patent should be worthless.

All those shoulds depend on the legal system. Can a malevolent patent holder badger people and profit without a valid patent. If they couldn't then you'd be fine as they'd always want the best out of the examiner - make sure they' know all relevant prior art, make sure there's enough support in the description, etc. - to be as sure as possible they had a valid and hence useful patent to litigate with.

You have to make it highly costly to attempt to litigate with an invalid patent. Make the game-theoretic outcome favour working for good patents rather than just getting any rubbish patent and profiting from it.

[blablabla123]

You are implying that the US patent office is a profit oriented organization. Even if this is practice, it should be possible to fix that, no?

Also I disagree with that it's a self perpetuating cycle. It is only self perpetuating if your reaction to more applications is being less rigorous. Almost every organization in the world, no matter if it is profit or non-profit, gets more rigorous the more * applications come in.

Somehow you are right but I think the root cause is something else.

[lyndonh]

> Almost every organization in the world, no matter if it is profit or non-profit, gets more rigorous the more applications come in.

They get more money from processing more patents. They don't get more money by making sure the patents are better quality.

[14113]

It is - as someone else in the thread mentioned it makes a profit from each patent granted.

[justincpollard]

"The proper solution is to make it very easy to strike down bad patents ..." isn't this exactly what the author is trying to do? The author states that he uses StackExchange to "respond to requests for prior art that invalidate an overly-broad patent." Sounds like you two agree.

[lyndonh]

The author thinks he is "helping" the USPTO - to stop bad patents. But he is doing nothing of the sort. The USPTO are part of the problem.

Do you really think that you can strike down a bad patent by simply sending some links for prior art to the USPTO ? Someone who applies for a patent must pay a fee to have it examined by an expert. I'm pretty sure that they're going to have to spend some time re-examining the patent and the new material to decide if it's relevant or not. I don't think that's going to be free.

[frobozz]

> There is no incentive for them to stop granting bad patents.

I disagree.

a: If they are currently overrun with more work than they can handle, they can remedy this by refusing patents, thereby discouraging frivolous applications. Having done so, they can become a more efficient organisation.

b: Fees for continuations, amendments and extending response deadlines can be larger than the fee for simply issuing the patent.

c: As a Government entity, they have an interest in making as much money as possible change hands (and therefore attract tax). One way to do this is to occupy the applicants' lawyers with remedial work.

[lyndonh]

a. They get money for each patent examined. If they spend more time scrutinizing it, their profit is reduced.

At their annual review, one measure of how hard an examiner worked is probably based on how many patents they dealt with.

b. As an examiner you can send a patent back for a clarification or ask for more details. That's why as an applicant you pack in as many obfuscated details as possible, without giving the game away.

Also, this: http://yro.slashdot.org/story/14/03/02/214237/inventor-has-w...

c. The money goes to the lawyers. It's not an incentive for the examiners.

[frobozz]

a. They get money for: 1) Each patent claim (a patent is made up of a number of claims); 2) Each patent issued; 3) maintenance fees on granted patents; 4) Any extensions or continuations requested by the applicant; 5) Any post-rejection amendments.

At their annual review, one measure of how hard an examiner worked is probably based on how many requests they dealt with appropriately, whether that's 5 requests for 1 patent, or 1 request each for 5 patents.

Using a measure like the one you described is like using number of Lines of Code in a programmer's annual review.

It is in no one's interest for an examiner to simply stamp "PASSED" on any application that arrives on his desk.

b. Examiners can send it back for clarification, but if the applicant needs more than the allotted time to clarify, then that attracts a fee.

c. Part of the money goes to the lawyers, some of the money goes to IRS. It is a weak incentive, but not no incentive at all.

[noisy_boy]

Or maybe make the patent processing fee non-refundable (if it already isn't).