This article is more than 1 year old

Security: Is technology saint or sinner?

Data pragmatism

If we continue in this way, we'll see more headlines where a child dies owing to information from one group not being available to another, to people who should be being tracked being lost due to insufficient data being available, to the continued billions of pounds being wasted in fraudulent claiming of benefits, of insurance claims, the booming black market economy and so on.

ID theft will continue to rise without any means of being able to prove irrevocably who we are, and that ID can be taken from us.

And for anyone who has had full ID theft occasioned against them, then all of a sudden, you really wish that you'd backed the implementation of ID cards, at least in a correct way.

(Please note that I am not backing the government's half-hearted, half-baked way of providing government-backed false IDs.)

To my mind, it's technology which can help us by ensuring sophisticated controls over access to data. We can design, say, a DNA database that is just that: a genetic fingerprint that is held against an identifier.

We could do the same for iris recognition and/or fingerprinting. Three different databases, none of which actually provides any information against named people.

To get onto these databases, you have to go through three different groups. Why? So that any chance of using insiders to create false IDs is minimised.

Any check against these databases would use full auditing. Any access to any field within the database is time stamped and stamped with an access code showing which user or body nominally accessed that field.

Security profiles then begin to take over; having verified that the DNA, iris and/or fingerprint are in each of the databases, what else do we need to do?

Do we need to be able to carry out another match to ensure that this person is who they are saying they are? Maybe a PIN or something similar? OK, a fourth database, maybe within the private sector.

Again, all that this has is the PIN against a unique identifier. We now have up to four pieces of unique data against four unique identifiers. In comes database number five: a correlation database of unique identifiers.

If all of these unique identifiers correlate as being from the same person, we can pretty much assume that we have a match. And at no stage have we had to go to a database that has any names or other personally identifiable information held within it.

However, if this is the police, ambulance or fire brigade, they may then need to go to a different database where such personal information is held.

Again, all fully audited against access type, named ID and, where necessary, correlated against biometric information of the accessing individual.

For the highest levels of information being held on us, we need the same sort of approach that we have for nuclear warheads being set off: a dual key system.

No single person should be able to access every last item about another without some balance being available.

For me, we have to look at data pragmatism. I want to be able to walk the streets without too much fear of aggravated assault against me, I want to be able to see my insurance premiums go down because thieves find it harder to get away with misdemeanours, I'd like to see my tax go down due to fraud being eradicated.

This won't happen unless we make the most of technology, but also use appropriate technology as the controls against inappropriate usage.

Copyright © 2006,

More about


Send us news

Other stories you might like