The concept of citizenship has evolved to include those who, by choice or profession, engage in high-security transactions, and those who will choose to live by their own means, without the need for involvement with the State, other than to record their birth and death. In between the two extremes, ordinary citizens may require more than one "identity", as Crompton talks about it. There is nothing wrong with the person who has access to top secrets being free to wander down to the shops in his pajamas to buy a loaf of bread, and not have to be bothered by proof of identity.

It isn't commerce that will test the limits of society's acceptance of robust proof of identity. The financial institutions are trialling two-factor authentication, that requires users to carry a hardware token. Commerce will move quickly, because they deal with people's assets, and the users will not think to use the dog-ate-it excuse.

It will be a long time before we demand the same level of security for someone to saw off grandma's gangrenous foot, as we seek to ensure no-one fraudulently gets at our cash. I walked along a public corridor in one of esteemed public hospitals the other day. A chap in operating theatre attire cruised past, on the way to have a gasper. I hope he was careful with the flame because his rosy demeanour and the trail of fumes suggested he was on about 0.3 breath alcohol. He was a porter, not a surgeon, but he still has access to critical functions and facilities. What am I getting around to? We would like to be assured that all healthcare workers are not impaired by alcohol. And substance abuse is a particular problem where personnel have access to potent drugs. It is in the interests of the professional groups, and work units, to be able to monitor their own performance and behaviour. If they wanted to do random drug and alcohol screens, they would need very good identity systems. After all, the anaesthetist who has been helping himself for twenty years has pretty good tricks, and it's safe to assume will exploit every loophole. You cannot sack someone of that calibre, without a foolproof system. The systems of trust, voice, signature and passwording in today's health systems are not good enough.

The IT sector will provide the tools, but it will be up to niche sectors of society to show how ID management will help them do a better job. A smart government would lead by working with highly motivated professional groups, and simultaneously invest in testing a range of options. A dumb government will threaten ordinary citizens with having to carry eminently lose-able chunks of plastic, to get access to the services that are theirs by right.

When all is said and done, it may be that the weight of the databases and their management overheads, along with the fragilities of biometrics, will kill off the argument for a centralised citizens register, that goes beyond the simple Danish system. Final proof of identity has to rely on DNA, doesn't it? So, I would like to see the costings for a system that uses DNA. We might only need to have 10% of citizens in it (selected by occupation), but it doesn't follow that putting the remaining 90% in it will only cost x10. More likely, getting the final 20% in, with the same level of reliability, will cost an awful lot more than it is worth.

But the boys will have to play with the toys, first. They can't help it.

Edward - is ACS on to this? - http://csrc.nist.gov/piv-project/

The article in Washington Post
http://www.washingtonpost.com/wp-dyn/articles/A35071-2004Dec29.html
suggests 2 million will need the cards.

The public meeting on Jan19 will be fun.