For years, the big blocker to the adoption of fingerprint-based security was its association with criminals - for most people, the image it first brought to mind was a villain having his fingertips rolled on an ink-pad by a police officer.

That's changed, though. With so many laptops and other devices including fingerprint readers as standard, it seems we're all now used to being fingerprinted even when we know we're innocent.

Across the Atlantic, what used to be the Land of the Free fingerprints all its visitors - very politely, it must be said - and on this side of the pond, BAA wanted to fingerprint all travellers changing flights at Heathrow T5, although that plan got kicked out by the UK Information Commissioner.

Still, I felt a faint sense of unease at the news the supermarket chain Budgens is fingerprinting employees at six of its stores, with plans to roll the scheme out over many more.

The motive is entirely understandable - it wants to eliminate "buddy-punching", which is not as the term might suggest, some new variant on happy-slapping, but the age-old practice of clocking in your absent mates on their behalf.

That a problem exists here is undeniable - the company running two of the Budgens stores taking part in the pilot claims it is saving £10,000 a year. My question is whether (and when) the use of technology - and in particular this technology - is the right answer.

Researchers have demonstrated the ability to fool biometric systems with fake fingerprints for instance. In addition, it's known that not everyone has usable prints, and even if we do, they can be made useless by accidental damage.

Oh, and then there's the general problem facing biometrics which is that if someone does crack the system and impersonate you, changing your fingerprints - or your irises, or whatever - is just a tad more complex than changing your password.

Then again, maybe Budgens and its IT supplier Nivid are on the right track here - even if they're probably wrong in claiming that "Finger prints never lie". Maybe a clocking-on system for the dozen or so staff of a shop is the right place to use this kind of technology. The opportunity for gain isn't great enough to justify a major criminal attack, and the organisation is small enough to deal with any individual problems that might arise.

Now, if someone were proposing applying similar technology to something really valuable and important, something that would justify the effort and expense of cracking it - something like, say, oh I don't know, perhaps an access-all-areas-style National ID Card - we'd laugh the fools out of court.

Well, wouldn't we?