Great inventions are driven by good science and technology, but great patent applications succeed by exploiting arcane patent law. This is a paradox of the patent system, and is one reason why it is in danger of being overwhelmed by thousands of complex applications, each trying to stake out the biggest possible piece of the intellectual-property pie.

Both the US and European patent organizations are seriously overstretched: the US Patent and Trademark Office (USPTO) says it has more than a million applications in its backlog; and examiners at the European Patent Office in Munich went on strike last week to protest against their workloads.

Also last week, the USPTO held a meeting in Alexandria, Virginia, to examine a novel approach to the problem. The Community Patent Review project (http://dotank.nyls.edu/communitypatent) tries to place the patent process firmly back in the hands of scientists and engineers by asking them for feedback on pending applications.

Under the proposed system, researchers would volunteer to be informed whenever patent applications in their areas of expertise are published. They could then use an electronic bulletin-board to post any prior publications that might be relevant. They could also rank other postings in order of interest. Ideally, the result would be a well-ordered list of publications that the patent examiner could use to determine whether the application is truly innovative.

The idea was the brainchild of Beth Noveck of New York Law School and has won forthright backing from IBM. The computer corporation, which obtains more patents than any other company, says its support is part of a broader push for patent reform. Now the USPTO is pondering a pilot project of the idea. IBM would volunteer some of its own applications for initial review.

patent peer review may be able to clean up some of the spurious claims of those looking to profit from vaguely worded applications.

Unsurprisingly, perhaps, some of the patent lawyers who attended the USPTO meeting were unimpressed with the idea. They pointed out grounds for legal challenges to the result of the suggested process: if patent examiners ignored some of the comments they received, for example, they might be accused of “inequitable conduct”, which in turn might invalidate the patent. Additionally, comments posted on the site could run foul of current law, which limits communication between patent examiners and outside sources while an application is under review.

Then there is the question of who would bother to participate in the process. For the most part, researchers take part in scientific peer review because the community expects it. No one expects anyone to review patents, and it is conceivable that the only people who come to the site would be rival companies, their lawyers, and cranky ‘inventors’ hocking cold-fusion reactors and perpetual-motion machines.

But similar pitfalls have not entirely undermined ‘open’ review systems in other spheres, and the USPTO should at least give the idea a try. Scientific peer review is often credited with keeping research honest, and patent peer review may be able to clean up some of the spurious claims of those looking to profit from vaguely worded applications. And even if it fails, the pilot could assist in the search for fresh approaches for dealing with the backlog.