I left my bank when it bragged about giving a mortgage to nine out of ten applicants

THE USPTO has been scrutinised here for a number of years primarily because of patent quality (or lack thereof). Few people can honestly say that it's hard to get granted a patent at the USPTO (there are like 10 million of them and one single company like IBM can gain 7,355 patents in just one year). Recent figures suggest that eventually 92% of applications are "successful", so what kind of quality control is that? Engineers at Sun once joked about how silly an application can be accepted and even competed over this as applicants (to see who gets the stupidest patent claims through the examiners). A lot of granted patents are simply dubious, but small companies would not be able to afford challenging them in court (or it would be a lot more expensive than simply settling).

"When the Government Accountability Office (GAO) asks for improvement in patent quality it also alludes to patent scope and software patents explicitly."Matt Levy, who habitually complains about patent trolls and even about software patents (in his more recent articles), has just mentioned the US Government Accountability Office's rant about US patent quality (which we first wrote about last week). He says [1, 2] that "[e]ven if you’re not a patent lawyer, you’ve probably noticed that patents have been in the news more. The growing problem of patent trolls, companies who make their money by suing other companies for patent infringement, has been the primary reason. Patent trolls now account for nearly two-thirds of all patent infringement lawsuits, draining billions of dollars away from productive companies."

According to this article (behind paywall) a "US High Court Restores Treble Damages For Patent Infringement," which means that things are not necessarily improving.

When the Government Accountability Office (GAO) asks for improvement in patent quality it also alludes to patent scope and software patents explicitly. Benjamin Henrion said that "improving quality does not mean much for the average programmer, just more spam." Well, in practice, raising the bar may mean that few software patents would be authorised at all (both at the courts and the patent office). That's definitely a step forward. IBM's Manny Schecter, a longtime proponent of software patents, was yammering that: "Backlog easy to measure, quality not, but..."

Backlog depends on the number of examiners or the lenience of examination. There are some correlations there and under David Kappos, who had joined from IBM, the USPTO tackled backlog by just granting lots of things faster (basically the same error Battistelli and his goons make at the EPO).

Here is a good article about the GAO report. It says: "The Patent and Trademark Office, at a time of rising lawsuits charging patent infringements, needs to improve quality and better monitor examiners’ work, a watchdog found.

"This is not a sign of innovation. Quite the contrary.""In a pair of reports and staff survey released on Wednesday, the Government Accountability Office said that patent examiners report being pressed for time as they process patent applications without a consistent and clear definition of quality.

"“District court filings of new patent infringement lawsuits increased from about 2,000 in 2007 to more than 5,000 in 2015, while the number of defendants named in these lawsuits increased from 5,000 to 8,000 over the same period,” GAO wrote in one report addressing quality in intellectual property protections."

This is not a sign of innovation. Quite the contrary. A patent trolls expert (he conducted academic studies on the subject), Professor James Bessen, said: "The GAO finds Patent Office issues 1000s of "unclear & overly broad" patents, causing excessive litigation" ("Stop issuing Software Patents & there wouldn't be a backlog" was one response to this).

"Most disputes concern new computer technology and software," according to another new article about the GAO study/reports. Just like in the EPO, the US office is "focusing too much on the timeliness of reviews, customer service and “process or production goals” rather than quality."

It's added that "[e]xaminers are rated largely on their production, auditors said, and they are given different times" and "GAO recommended that the agency define what is a good patent, update its performance goals for reviewers" (the same should be done at the EPO). ⬆