When you think about it, patent examiners make industrial policy that affects the entire United States for up to 20 years. To be fair, the USPTO takes that resp

When you think about it, patent examiners make industrial policy that affects the entire United States for up to 20 years. To be fair, the USPTO takes that responsibility pretty seriously. Still, examiners are human --- and a lot of them aren't experienced enough to have signature authority: their actions must be signed by their supervisors, who have to oversee multiple junior examiners. In many fields, the examining corps is pretty buried by the workload and doesn't have the time or other resources they'd like to have. [EDIT: In response to 'dkhenry, I agree, there's no excuse for the patent being discussed here.] This examiner burden is exacerbated by three things: First, patent examiners' performance is graded in part on the basis of a "count system," which provides at least some incentive for examiners to allow at least some claims [1]. Second, the statute mandates that a patent be issued unless the examiner can demonstrate that the application is not patentable. Third, there's no requirement that a patent applicant conduct any kind of patentability search. An applicant and his patent attorney must disclose any "material" prior art of which they're aware, but the applicant need not do any kind of literature search. [EDIT: When you apply for a patent, you pay filing fees in part to help cover some of the cost of having a patent examiner do a search.] Just imagine if a PhD candidate wasn't required to do a literature survey as part of her dissertation work, but instead her advisor and thesis committee were required to approve her degree unless they and their TAs could affirmatively demonstrate that her research was insufficiently novel. That's not unlike the way it works in the patent system. And now think of how much more national impact can result from the issuance of a patent compared to the issuance of a PhD degree. Inventors and patent attorneys tend to fiercely oppose any proposal that patent applicants be required to conduct prior-art searches, on grounds that it would increase the cost of a patent application. But if an inventor wants a national industrial policy to be made in his favor that will last for as much as 20 years, it doesn't seem per se unreasonable for society to require him to do some due diligence first. (Of course, as long as the law is what it is, patent applicants, including my own clients, will quite properly abide by the law as it is and not as some might think it ought to be.) Richard Stallman once asserted, in testimony at the USPTO [2], that: --snip-- Some years ago a professor I know patented Kirchoff's current law, which says that the electric currents flowing into a junction equal the currents flowing out. He did this to confirm, privately, his suspicion that the PTO could not handle the field of electronics. He never tried to enforce the patent which has since expired. I will disclose his name if you give assurances that he and his lawyer will not get in trouble for this. Kirchoff's laws were formulated in 1845. If the PTO couldn't understand electricity after a century, how can we expect it to understand software in another decade or two. (applause) --snip-- [1] http://www.ipwatchdog.com/2010/02/26/usptos-new-examiner-count-system-go-into-effect/id=9310/ http://www.ipwatchdog.com/2010/02/26/usptos-new-examiner-cou... [2] http://www.uspto.gov/web/offices/com/hearings/software/sanjose/sj_stallman.html http://www.uspto.gov/web/offices/com/hearings/software/sanjo...