John Rumm wrote:


... . Part of me thinks "well it would be rich pickings for anyone even
remotely into technology or IT for a good few years".

'Would' nothing: there are already flocks of well-suited conslutants
salivating all over this opportunity. It's instructive to read around
the relevant section of the Home Office website to see the range of
tenders, solicitations for advice, pilot proposals, yada yada, which our
favourite suppliers-of-IT-services-to-the-State are bidding for.

What is worse is you will have lots of technology providers egging them
on since it will sell product. If you are non technical it is easy to be
blinded by the science and convinced that whichever technology is being
hawked will solve all the problems.

Yurp - the Arthur C. Clarke aphorism applies: 'Any sufficiently advanced
technology is indistinguishable from magic'.

BTW, it is worth reading the Duagman papers on iris recognition; there's
solid science there, and the false-accept and false-reject rates for
that particular biometric can be taken low enough to make
population-wide recognition feasible; but it does require 'willing'
subjects, and AFAIK there hasn't been enough work on compatibility
across the whole ethnic spectrum. Oh, and in a further discomfiting
twist, John Daugman's approach to patenting the technology is a clear
societal good: by licensing only the 'whole package', he's effectively
making sure that implementations don't make obvious, crappy, technical
mistakes - for example, they must set the discrimination parameters to
reflect the population size. Of course, getting the technology right
doesn't mean that the surrounding processes for registration, use,
minimisation of unnecessary data retention, yada yada will be gotten right!

Stefek