Data federation and the police
Consolidation is not the answer
Last week's Bichard report into the mistakes made by various English police forces in the events leading up to the Soham murders focuses, at least in part, on the endemic deficiencies in data management procedures and policies.
There were multiple databases in use, both within individual police forces, and across police forces, and these heterogeneous systems could not talk to one another.
There is a single nationwide database for offenders, that is, those who have been found guilty. But each force has its own database or databases for suspects and others who have not been found guilty of any crime. The idea now is to combine all of these suspect databases into a single, central, database. However, reports suggest that this will take some years to complete and will cost nine figures. Worse, given the UK government's record on large-scale computer projects, this is likely to be (very) late and grossly over budget.
So, why adopt this approach? These disparate databases can be linked using data federation technology. Such a solution would provide real-time query capabilities for relevant personnel enquiring about particular individuals, which is presumably the point of the whole exercise.
More to the point, a federated solution would cost lots less, take much less time, could be implemented incrementally, and altogether makes much more sense. Certainly, there may be a case for consolidating some of the existing suspect databases - perhaps within particular police forces - but that doesn't mean that the whole lot should.
Of course, one of the arguments against a centralised solution (whether in a police force or not) is precisely about losing local control of data and police forces, in particular, have a long history of local independence and being (in theory at least) responsive to the local community. Yet I have seen or heard no representations from local police chiefs that a federated solution could preserve their independence while fulfilling national needs at the same time - although that may be because they have not been reported or I read the wrong newspaper - or it may be because they don't know enough about the subject to make such a suggestion.
Alternatively, they may simply know that this is the sort of thing that is going to be sewn up centrally. And if that's the case, then we will almost certainly go down the consolidation route. First, because that will generate more consulting revenues. Second, because government (or at least this government) likes to centralise things. This will be good for however gets the contract, and bad for everyone else. One day, maybe, it will work.
IBM gets hot and spicy
Grand Unification Theory
End of the road for EAI?
UK police lack e-crime savvy officers
UK.gov announces hi-tech elite police squad
Blunkett appoints police IT supremo
Hampshire police deploy e-warranting
Home Office to centralise police intelligence
Grampian police trial facial biometrics