Talk:Raising examination standards wouldn't fix much
 Minor Counterpoint
[ User:Ciaran says: I've moved these three paragraphs from the article to here. The first is very broad for this article - it's a total reform of the patent system. There might be a good suggestion in there but I can't work on it right now and I want to submit this page to Slashdot soon. The second and third paragraphs, at a glance, seem offtopic for this article.] Ciaran 13:21, 4 November 2010 (EDT)
One way to raise the bar "sufficiently" would require adherence to keep annually awarded patents to something like a handful (perhaps not far off par with Nobel Prize awards). This could be accompanied with required narrowness (eg, not over fundamental algorithm or observation but a very precise advanced one). Additionally, have this accompanied with a shorter duration period (no more than half the current duration, though ideally just a few years). And tack on as well other restrictions, such as that this could not form part of a significant standard unless very generous terms were allowed (eg, compatibility with the licenses of most FOSS) and/or that noncommercial or research use would be allowed. In short, with enough limitation to software patents, the few granted could become fairly innocuous overall yet allow those with reservations about abolishing the system feel like they still can get some sort of lottery ticket or backup to where they fear a free market might fail.
Partial countering the counterpoint is a quote from Why_abolish_software_patents: "because going from source code to running binary is effectively a trivial step, applying patents to the running machine wholly pre-empts all the more balanced protections and safeties from copyright law."
Arguably, avoiding copyright denial would apply 100% of the time, but it is a possibility that courts would want to set some limitations if enough evidence was presented for some specific case or other.
 What got changed in the 2010 guidelines?
Can someone explain what effect the 2010 guidelines have?
It's quite possible I misinterpreted the situation.
I read this article: , which concludes that with these guidelines will make it easier for patent applicants to pass the obviousness test (see the last sentence of that article).
I thought it did this by deleting four of the six tests (not explicitly but implicitly). This sentence is one example of where I got this impression:
- In the 2010 Guidelines, however, only two of the new tests for obviousness appear to survive.
If we ignore that phrase, I see another possible reading: there's no change in the four tests which weren't mentioned, but obviousness standards have been lowered because the two tests which are discussed are weakened.
Any insights into the effects of the 2010 guidelines? Comments on my interpretation? Thanks. Ciaran 15:00, 7 November 2010 (EST)
- Nothing substantive was changed. This is just an update to inform examiners of some Federal Circuit cases discussing obviousness that were decided since the 2007 Guidelines were published. The 2010 Guidelines Update does not supersede or replace the 2007 Guidelines. From the 2010 Guidelines: "Thus, Office personnel may use this 2010 KSR Guidelines Update in conjunction with the 2007 KSR Guidelines (incorporated into MPEP §§ 2141 and 2143) to provide a more complete view of the state of the law of obviousness." The blog's analysis of the 2010 Guidelines is almost willfully misleading. It takes quotes out of context to make it seem like the 2010 guidelines are eliminating tests, when it clearly is not. Another excerpt from the 2010 Guidelines, this time given in full: "Although the other rationales discussed in the 2007 KSR Guidelines are not the focus of separate discussions in this 2010 KSR Guidelines Update, it will be noted that obviousness concepts such as applying known techniques, design choice, and market forces are addressed when they arise in the selected cases." Again, it's clear that the 2010 Guidelines leave the tests from the 2007 Guidelines intact. They are only not mentioned in the 2010 Guidelines because there aren't any important new cases to discuss that use those tests. --184.108.40.206 21:38, 7 November 2010 (EST)
- I'd like to get to the bottom of this, so I've re-added the claims to the article so that the discussion can continue. I think the Patently-O quote is important, because Dennis Crouch is someone whose analysis is very trustworthy. If he says this will change how patent examination is done, that carries weight. It's unfortunate that he doesn't give any comment about how this will change examination.
- It also seems logical that these Guidelines will have an effect. That's usually the point of publishing something (unless you it was to say "Hey look, the courts confirm that we've been doing it right all along" - which doesn't seem to be the message in this case). Ciaran 11:52, 8 November 2010 (EST)
Because the blog's analysis is wrong, I'm deleting it from the page. The tests not specifically mentioned in the 2010 Guidelines are still in force. Further, Kappos has made it clear in public discussions that he is interested in allowing high-quality claims, not allowing anything and everything. The policy goal is that examiners and applicants should work together to quickly agree on what is actually patentable instead of taking an adversarial approach where the applicant asks for everything and the examiner has to cut the claims down. It should also be noted that rejections have also increased under Kappos. The change is one of increased total throughput, not so much the proportion of allowances to rejections. --220.127.116.11 21:45, 7 November 2010 (EST)