Features

net.wars: Follow the database

Well, it's here: the British Government's draft ID cards bill. The consultation period ends July 20. Comment early and often.

The government's ideas on the subject don't seem to have changed much since the last consultation. It's what we in the "skeptical" world like to call the "ratchet effect". Any supporting evidence winds the belief tighter; evidence against has no effect. In this case, the belief is that national ID cards will end terrorism, put a spoke in the operations of the real-life equivalent of the Sopranos, stop drug dealers, kill off benefit fraud and "health tourism", and, doubtless, make everyone braised vegetables every morning for breakfast.

A few useful statistics came out at Wednesday's meeting, but in the best tradition of the ratchet effect these are not likely to unwind the belief mechanism at all. Blunkett told David Frost in April that 35 percent of terrorists use multiple identities. That mean that 65 percent do not. Of the 25 countries that have been most adversely affected by terrorism since 1986 (according to a Privacy International report, 20 have ID cards, a third of those incorporating biometrics. Plus, you know, if you want to stop terrorism now, a system that won't be fully functional until 2013 probably isn't going to help much.

Most benefit fraud does not rely on false identities, but on correctly identified people claiming more than they're entitled to. While it was certainly true that at one time people came to the UK for free NHS treatment, now we see some people leaving the UK to find faster treatment. Finally, organised crime ought to be capable of organising an identity-card forgery in a brewery.

But ID cards are, as Hitchcock would say, the McGuffin. The real story is the national database. The bill says registration will not be compulsory - at first - but makes it plain that the government will make every effort to extort registration from each and every British resident by designating such commonly sought-after documents as driver's licenses and passports as national identity documents. Apply or renew, and you'll get a national identity number. Then, once it's compulsory to have the ID card, it will become compulsory to produce the card in order to access public services. Though if you can't produce the card, that won't matter, because everyone will have readers, and you'll be able to give a fingerprint and have it checked against the database.

A layer up, the bill itself may be a McGuffin: the Children Bill specifies that all children will be issued at birth with an identification number. So even if everyone with a vote opposes the national identity cards bill, 20 years from now there'll be an entire generation of kids whose lives have already been tracked comprehensively, and to whom an additional statutory instrument continuing that database into their adult years may seem acceptable because it's familiar.

Andrew Grossman, a legal expert in personal status, says that bringing in such a database does indeed represent a profound change for Britain which, like Australia, Canada, and the US, doesn't know who its citizens are. "They think they know if you have a passport," he says. "But Belgium, the Netherlands, Switzerland, and a number of other countries know virtually every citizen they have." In most other European countries, you register locally and everybody knows who you are; identity theft is virtually impossible because everyone is registered and tracked. Britain, by contrast, focuses on entry via the ports.

You can see why the civil service keeps blowing the dust off the identity card proposals every time a new government comes in if that's the case, can't you? It must seem so unutterably messy to them.

The most important objections at the moment are the practical ones. The system is going to cost an insane amount to set up even if there are no technical hitches. As an IT project, the national database has a number of points against it: 1) it's a government project, and these have an average success rate of 34 percent; 2) nothing on this scale has been attempted before in any country; 3) the biometrics on which Blunkett is pinning all his hopes are relatively untested and unproven technology, at least in this application; 4) the system is going to have to be supported by tens of thousands of readers all over the country, and hundreds of thousands of people will have to be taught how to operate them correctly; 5) the database will represent the biggest possible target for criminals to hack into.

The government keeps citing opinion polls showing people are in favour of the ID card, but a lot of them are in favour the way people in my neighbourhood are in favour. "I think it's quite a good idea," said one of them just yesterday. "Even if you have to pay £77 for one?" "No." The notes accompanying the bill fudge this part by saying that the price of passports and driver's licenses would rise anyway - by US mandate, British passports will have to include a biometric in any case. (Hey, folks? You know how Britain isn't actually part of the US? You could Just Say No and withhold all your tourists.) So the government figures that the national identity card really adds only an extra £4 over what prices would reach anyway. If you believe this, I have this broken-down copy of London Bridge I'd be happy to sell you.