Hacker News Reader

Comment by Osiris | Hacker News Reader

I've been thinking lately that with the number of patents in existence, the likelihood that a newly submit patent application is truly novel without any invalidating prior art is probably very low.

The USPTO should start each review under the premis that the application will probably be denied unless the application includes sufficient justification for it being truly novel.

If the USPTO started denying the majority of applications you'd, 1) start seeing less applications, and 2) applications would be of a higher caliber.

View on HN

USPTO denies over 90% of initial applications, and even with appeals factored in, less than 50% of applications end up getting approved (it was 48% in 2011).

The problem as I understand it from people who have worked in the industry, is that if you have the money, you can keep adjusting the application and resubmitting, and eventually the probability of getting the bogus patent approved approaches 1.

Some have argued that anyone applying for a patent should be required to have a prior-art search done and submit it to the patent examiner, just as a PhD candidate must do a literature survey as part of his (or her) dissertation work.

"I've been thinking lately that with the number of patents in existence, the likelihood that a newly submit patent application is truly novel without any invalidating prior art is probably very low."

A great scifi short story that examines this eventuality in the context of art & perpetual copyright : Melancholy Elephants, by Spider Robinson. http://www.baen.com/chapters/W200011/0671319744___1.htm

Except it's impossible to prove the absence of prior art, while proving the existence of it is pretty easy.

Yep. Finite set of prior art < Infinite set of everything else

Everything else is both finite and not clearly larger. Insofar as everything else might be infinite (e.g. Infinite multiverse) so might prior art.

Former examiner here. That's basically how it works. As an examiner you do your best to reject every application. There's no other way to go about it.

See: http://en.wikipedia.org/wiki/Charles_Holland_Duell

Reply on news.ycombinator.com