Forgot your password?

Comment: Re:This is not a new or unique problem (Score 1) 124

Keep a pool of about two months worth of patents that might meet the bar of being granted. Then reject the rest.

I don't understand what that means; I'm not certain you do either. First of all, til SCOTUS or the USPTO say otherwise, all patent applications either meet the requirements or don't. How exactly is the USPTO supposed to play favorites? Hire a contractor to pick "good" patents"? Plus there's the other problem: unless you get about 147 other countries to go along with this plan, PCT patents applied for through WIPO internationally are enforceable in the US.

Comment: Re:Next = Examiner Competence in their field (Score 3, Insightful) 124

How did you phrase the claim? Regardless, it varies a lot from one section to the next based on the job market for the applicants. Biology/Pharma/Biotech/O-chem has had huge layoffs over the last 15 years, so there are plenty of PhDs with years of industrial experience, patent writing experience and perhaps a JD for each examiner position that opens up. Software and engineering, not so much.

Comment: Re:This is not a new or unique problem (Score 4, Insightful) 124

Though it is complicated by the government service issue, there are ways to measure performance...

- Salt the case load with fictitious, bogus applications intended to be declined. In fact, this can both detect work that is disingenuous, and start applying some quality checks. Applications that are so flawed as to be obvious can be expected to fall through as approved if examiners are just phoning it in.

- Break up the review process, no insight into the next step for any examiner. At some point, some examiners will be doing too little work to keep up, or the backlog will inspire some investigation. Perhaps.

- This is an oldie. Full tracking of the examiner's work, down to the keystroke.

- Even older, time to put up the performance chart. Peer pressure will probably not work in Civil Service, but it's a valiant try nonetheless.

Now, the real trick is how to measure performance. That scares me.

- The first would waste 1 or 2 days of patent examiner time per bogus application. On top of that, each bogus application could only be used a handful of times before they were known throughout the section, so the ~$10-50K it would take to hire patent agents to create each bogus application would add up quickly. Copypasta applications would be too easy to spot, as would ones that are quickly thrown together. A 500 page patent application from a pharma is not something you just whiz together overnight. A better bet might be to just assign the same application to two examiners 5% of the time and then compare the first office actions they submit. That would work so long as examiners can't compare their caseloads ...

- Breaking up the review process would mean that many patent examiners would have to familiarize themselves with each patent as opposed to just one. That could easily double or triple the man hours necessary to examine each patent, and significantly delay each office action. Which is why I suggested only duplicating the effort 5% of the time.

- Full tracking of their workload would be a PITA all around. A program that monitors how many hours are actually spent on each portion of the examination, as well as when those hours are spent, would be good enough.

- Performance charts, monitoring, rank and yank, etc: might work in biotech and organic chemistry sections where there is no shortage of qualified applicants, but in IT if you make the working conditions any more unpleasant everyone will just head off to the private sector. If you want examiners willing and able to work like patent attorneys, you'll have to compete on the patent attorney pay scale ($250k per year and up).

Comment: Re:It is not just the "extra" channels... (Score 1) 108

by pepty (#47897261) Attached to: Verizon Working On a La Carte Internet TV Service

The problem with "a la carte cable channels" is the presumption that people want channels. They want shows that are suited to their tastes. The center of value is the program, that's what brings in the people, but due to underlying economics, the center of costs remains the channel, this Netflix must offer its subscription on a channel-wise, take-it-all or leave-it-all basis.

Increasingly I'd say that's more true of cordcutters than cable subscribers. Some folks want a continuous background of thematically related stuff they can vaguely watch for hours and hours at a time: a channel. The underlying economics you mentioned make that much harder for Netflix or anyone else whose content delivery costs are directionally proportional to the amount of content consumed per customer. Cable companies traditional delivery costs were pretty much the same whether you left the TV on 24/7 or just watched 12 episodes of The Wire per year.

Comment: Re:What they don't tell you (Score 1) 588

by pepty (#47848971) Attached to: Low-Carb Diet Trumps Low-Fat Diet In Major New Study
The links for soy protein don't say much. The first is a single case study based on a guy drinking 3 quarts of soy milk a day. The second concludes:

Only one study has examined soy and whey protein supplementation together in conjunction with resistance training. Kalman et al. [46] reported that after 12 weeks of supplementation with soy there were no significant differences between groups for total testosterone, free testosterone, and SHBG. This study expanded upon the limited prior research examining the effects of soy and whey protein supplementation on testosterone, SHBG, and cortisol responses to an acute bout of resistance exercise. Contrary to popular misconceptions, soy protein supplementation does not appear to hinder anabolic signaling postexercise by means of eliciting increases in estradiol concentrations. However, our main findings demonstrate that 14 days of supplementation with soy protein does appear to blunt serum testosterone. In addition, whey might influence the response of cortisol during an acute bout of resistance exercise by also blunting its normal increase. Further research will need to explore a possible interaction effect on sex hormone binding globulin.

Comment: Re:And don't forget mercury in the CFLs... (Score 1) 173

by pepty (#47847977) Attached to: Surprise! More Than Twice As Much Mercury In Environment As Thought

There is nothing anti-science about having slaves and

Of course that is anti-science. Any economist would tell you that it is much cheaper to just get rid of minimum wage and overtime laws and then put your workers in a company town than it is to import, purchase and maintain slaves.

Comment: Re:its not about the education (Score 1) 161

by pepty (#47844249) Attached to: Getting Into College the Old Fashioned Way: With Money
Connections, yes, but also peers. Are you surrounded by incredibly smart and ambitious people in your school? Do they have Incredible work ethics, creative ideas, already planning the companies they want to start? Or are you surrounded by people who are going to college because, um, it's what you do after high school and before real life?

Some folks will do a lot or a litte no matter what their environment is, others will take a lot of cues from the folks around them.

Comment: Re: What the heck? (Score 4, Funny) 354

by pepty (#47841401) Attached to: DMCA Claim Over GPL Non-Compliance Shuts Off Minecraft Plug-Ins

As I understand it Bukkit is a mod licensed under the GPL. The Minecraft server is proprietary. They don't share any code, so individually they're not derivates of anything. CraftBukkit combines the server code with the mod code. This is illegal both ways - the server license doesn't let you link to Bukkit, the Bukkit license doesn't let you link to the proprietary server.

You either just described what the argument is about in an accurate, succinct and understandable way (a first for this thread) or you are completely wrong in a succinct and understandable way. You should really do technical writing or PR on the side, depending on which is true.

Comment: Re: What the heck? (Score 1) 354

by pepty (#47841341) Attached to: DMCA Claim Over GPL Non-Compliance Shuts Off Minecraft Plug-Ins

He rightly claims that Bukkit is distributed in violation of the GPL, because it contains parts of Minecraft server that aren't licensed under the GPL, and that license violation is grounds for the DMCA takedown /p>

OK, I'm confused. Bukkit contains parts of Minecraft server or reverse engineered parts of Minecraft server?

It is not best to swap horses while crossing the river. -- Abraham Lincoln