My Observer column for today:
Buried away in George ‘President’ Bush’s new budget proposals are some interesting measures to boost the US Patent and Trademark Office. According to a ‘ComputerWorld’ report, the proposals would boost the number of patent examiners at the agency by nearly a third, with 950 examiners added to a staff of 3,200. James Rogan, Director of the patent office, said that the increase was needed to help it process patent applications before the technologies mentioned in them become obsolete. The office currently receives about 350,000 applications annually and takes over two years on average to process each one.
Under the Bush budget plan, the patent office (which is funded by the income it gets from patent fees) would get a 21% increase in funding — to more than $1.3 billion — mainly by allowing it to keep 100% of the fees it collects, something its critics have been requesting for many years. The budget proposal also calls for a surcharge of 19% on patent applications — a move that, if Congress approves, would raise an extra $45 a year.
The intention of the Bush proposals is to streamline the patent-granting process so that applicants will not have to wait two years before they can run gibbering to venture capitalists waving their newly minted patents. If this is indeed the outcome of the Bush proposals, then they could turn out to be an unmitigated disaster. For the problem we have — at least in the areas of computer software and e-commerce — is not that US patents are granted too slowly, but that they are granted too easily, and with too cursory a scrutiny of their originality. The result has been an avalanche of daft patents, some of which may eventually strangle the online economy.
The classic examples are the ‘business process’ patents which the US Patent Office granted to companies like Amazon and Priceline. Amazon holds at least two such patents: one covers its ‘affiliates’ scheme (whereby websites which hyperlink book references to the Amazon catalogue receive a percentage of any resulting sales revenue); the other relates to its ’1-click ordering’ system (where users can ask Amazon computers to [OE]remember[base '] their credit card and address details, enabling them to buy goods simply by clicking on a single button. This is such a good wheeze that an Amazon.co.uk employee once told me that she had had to switch it off on her machine, lest she spent more on merchandise than she was paid by the company!)
Clever though these patents may be, they aren[base ']t terribly original. The affiliates scheme is just a variation on what back-scratching businesses have done for centuries. And the 1-click ordering wheeze could have been programmed by a chimpanzee. So neither is an ‘invention’ in the sense of something that is fundamentally different or novel.
The most pernicious feature of business process patents, however, is not their lack of originality, but the way they can stifle economic development. In an environment like the Web [^] the very essence of which is easy linking from site to site [^] affiliate marketing is an obvious way to do business, enabling sites to boost revenue by supporting one another. But if Amazon holds the patent on this, nobody will be able to apply this simple technique without obtaining prior permission from (and of course paying a royalty to) Amazon. By allowing commercial adventurers to patent business ideas, the US Patent Office has inadvertently given them a potential choke-hold on the online future.
In part, this happened because the Office was overwhelmed by the dot-com frenzy and lacked the technical skills needed to the originality of technological ideas. The new money coming from the Bush budget could remedy these deficiencies, but only if it is used to scrutinise patent applications more rigorously. Will it be? Watch this space.