Data is There to be Processed – But as Cheaply as Possible

Two recent news articles highlight issues with the database state and the fallacy of the “nothing to hide, nothing to fear” argument so often used to claim surveillance is not something the law abiding should be worried about.

The first was a report in the Guardian that 17 people had been mistakenly arrested, due to incorrect telephone information or Internet records being provided to the Police or other public bodies investigating crime. The other was an article resulting from a Daily Mail investigation concerning people having county court judgements awarded against them, without them even knowing that proceedings had been issued against them or anything about the court case.

In both these cases information has been processed on individuals without their knowledge and with not even basic checks on the accuracy of the information being made by the Police, security or court services.  This in itself scandalous given the serious consequences that both can and have occurred due to incorrect information; however, it also begs the question why such checks were not made?

Well what is often forgotten is that checking data requires resource (usually people) and this costs money; however, this cost can often offset any financial benefits that, for example, streamlining law enforcement by database may offer.  Thus, having to go through a data validation exercise in the above two scenarios will certainly significantly increase costs and may even impact on economic viability.  As an example, in a previous post we highlighted how many foreign drivers failed to pay automatic number plate recognition (ANPR) generated fines for the non-payment of tolls for the Dartford River Crossing and showed that even accounting for the loss of income from failure to pay tolls and fines, it was still considerably more cost effective to use ANPR than toll booths with their 100% toll collection rate.

Unfortunately, with the Investigatory Powers bill now almost law and the mass surveillance and data processing that will come with it, incidents like the above are almost certainly going to become routine as given the amount of data that will be generated, it will not be financially viable to undertake detailed validation of the data prior to use.