Will Peers Process Patents Perspicaciously?

Well, once you get on one of those alliteration kicks, it's hard to stop. This is a story about a big problem with the U. S. patent system, which is of concern to any engineer whose work is valuable enough to patent. And, about one small attempt to make it better.

For some years now, there has been general agreement that the patent system has major flaws. Basically, it's too easy to get a bad patent, and too easy to clog the legal system with patent lawsuits that never should have been started in the first place, based on overly-broad patents that never should have been issued. Partly because it's so easy, more patents are being filed every year, but the U. S. Patent and Trademark Office (USPTO) can't keep up—it now takes an average of more than two years to get a patent. And since many technologies such as software engineering come up with a whole new generation of products every few months or so, the patent system starts to look like a glacier stuck up on a mountain while a flood of water rushes by in the valley.

Part of the problem is that there aren't enough good patent examiners. Those are the government folks who pass judgment on whether a patent should be granted or not. The ideal patent examiner has advanced degrees in both law and a technical field, plus the patience and incorruptibility of a good detective. Such people have never been easy to find, and attracting them with a government pay scale is even harder. Faced with the rising flood of patents, patent examiners nowadays err on the side of generosity, allowing all sorts of patents through which in more rigorous days would have been tossed out. But to toss out a patent you need a good reason such as a citation of "prior art," and apparently doing a thorough job in that area is simply not something the patent office can handle very well anymore.

A recent news article highlights an attempt to improve the situation with something called Peer-to-Patent, a collaboration between the USPTO and New York Law School professor Beth Noveck. She has set up a website at which ordinary citizens (you or I included) can review selected patent applications, read and interpret the claims, cite prior art, and in short, pretend you are a patent examiner. If the "community" of volunteer examiners votes to forward your citations to the patent office, one of them may make a Top Ten list that actually gets used in the patent, if it gets issued, or more likely denied if your prior-art citation is a good one.

I viewed the little video on the site that gives an overview of the process. While it puts the best face on the matter, even my passing familiarity with patents (I have managed to obtain a couple over the years) tells me that to do a good job on just one application would require a good bit more work than it takes to do your average income-tax form, if not more. When I read about the Peer-to-Patent idea, my first question was, "Why would anybody bother to donate several hours of their highly marketable expertise to such a thing?" and after looking at the website, my first question remains unanswered.

As a practical matter, the only people I can imagine who would want to fool with this and devote the serious amount of work it would take, would be rivals of the inventors who made the original application, who are of course highly motivated to see it fail. If you translate this idea to a more familiar setting, I think you can see its problems better. Suppose you sue your neighbor for building a corner of his garage on your property. And suppose the judge in the civil suit, instead of hearing testimony from duly sworn-in experts such as surveyors and land-title experts, opens a website, posts the records of the case on line, and invites all and sundry to make comments, without even requiring them to give their real names. (The Peer-to-Patent website doesn't require real names, although it is recommended that you not hide behind an alias.) Who is the person most motivated to get online and trash your side of the case? Your neighbor, of course, or maybe his lawyer.

The analogy is not exact, but it does seem to me that by asking for "volunteers" to put in such a large amount of work—effort that the government can't seem to be able to hire on the open market—the site automatically selects only for the people who have the greatest motivation to criticize an application—that is, rivals of the original applicant who would dearly love to see it fail. And maybe that's exactly what Prof. Noveck is trying to do. But if that's the case, it seems more than a little hypocritical to just pretend that the volunteers are random, public-spirited citizens whose only motivation is the honor of having one of their prior-art citations selected for use by the USPTO. I mean, wouldn't that send you into orbit for weeks? Maybe there are some people like that, but I'm not optimistic that they'll be coming out of the woodwork to make the Peer-to-Patent idea succeed.

You have to give the USPTO and the New York Law School credit for trying something. The other day I heard a review of a rather cynical book by a fellow who says that the real motivation for Republicans who gain control of the federal government is to make it run so badly that people will lose faith in it, and not complain when it eventually withers away to the small government of many decades ago. I sincerely hope he's wrong about that, since if he's right we have been watching institutionalized hypocrisy in action for a long time. But weird ideas like this Peer-to-Patent business make me wonder. Maybe I'm wrong and Peer-to-Patent will be the answer to many of the USPTO's problems. But we'll have to wait a while to see.

Sources: The Associated Press article on Peer-to-Patent by Joelle Tessler was carried by many papers, including the Baltimore Sun on Sept. 15 at http://www.baltimoresun.com/technology/bal-patent0915,0,1444023.story. The USPTO's main website is http://www.uspto.gov, and the Peer-to-Patent website is http://www.peertopatent.org.