We Are All Patent Reviewers

There are many areas of government over which Obama has pledged to wave the magic wand of reform. And given the severity of the current economic crisis, fixing the system by which the US awards patents and trademarks has presumably been placed on the back burner. If that's the case, it's a shame, because the patent system's problems have reached tragi-comic proportions. As of early 2008 (when I last researched the matter for my book) it took an average of 2.5 years for an application to work its way through the USPTO, which has a backlog of more than 1 million applications. The 5,500 examiners themselves are notoriously underpaid and overworked, and can only allot twenty hours to review even the most abstruse patents. Further hampering the process, reviewers rarely posses expertise in computer science and aren't allowed to consult the Internet in their research, relying instead on the USPTO's own database of prior art. The result? An increase in undeserving patents and a Gordian Knot of conflicting claims. Unsurprsingly, patent litigation has doubled since 1990, at an average cost of $2 million per lawsuit.

In my book I chronicled how Beth Noveck, a professor at New York Law School, was trying to harness the network--which is to say, all of us--to improve the process. She launched Peer-to-Patent in June of last year, in cooperation with the USPTO, as well as such companies as IBM and Microsoft, whose outsized patent portfolios exercise a magnetic force for frivolous litigation. How it works: Post applications to the public, and let interested parties submit prior art and comment on the claims. The wisdom of crowds will do a better job of determining validity, the logic goes, than a single reviewer. I'm a fan, as I make clear in my book, but to judge by the number of participants (2352 members have submitted 274 instances of prior art), it's failed to gain the level of traction that would--and I'm speculating here--persuade the USPTO to adopt Noveck's methodology across the board.

Now a new model has emerged: On Monday a company called Article One Partners hung up its shingle. It's like Peer-to-Patent, in that it relies on the collective intelligence of the public. It's unlike Peer-to-Patent in that it offers money--up to $50,000--for anyone that can provide documentation that an existing patent either is or isn't valid. My colleague Julian Sanchez at Ars Technica (now part of the Wired empire) notes that this isn't entirely original, and perhaps Article One--which has filed for patent protection--is guilty of patent infringement itself. That's not quite right--in fact, Article One is working with Noveck. It will pay Peer-to-Patent contributors if they submit prior art that is accepted by the non-profit.