A new report for the Joseph Rowntree Reform Trust by a very credible largely Foundation for Information Policy Research (FIPR) team that combines engineers, lawyers, software developers, and political scientists, has concluded that a quarter of the UK public-sector databases are illegal under human rights or data protection law. It also looks at UK involvement in some European database projects and finds all of them questionable too.
The report rates the 46 databases on a traffic light system – green, amber, red – and argues that those rated ‘red’, in particular the National Identity Register and the Communications Database, and are simply unreformable and should be scrapped. This is massively important because it is based not simply on a financial, political or even an ethical position, but on the database projects’ respect for existing law. They are simply illegal, and not just massively expensive, morally questionable or politically undesirable. In fact, a quarter of all the databases were found to contravene the law and more than half were ‘problematic’ (i.e. open to challenge in court) . All of those rated ‘amber’ (29 databases) the authors argue, should be subject to independent review.
There are a number of other major recommendations, including the reassertion of the necessity and proportionality tests contained in DP law, citizens should anonymous rights to access data, more open procurement of systems, and better training processes for civil servants. The most important and radical measures proposed, and entirely correctly in my view, are those concerning the location of data and the whole nature of UK IT development. For the former, the report recommends that the default location for sensitive personal data should be local, with national systems kept to a minimum – this appears to be rather like the ‘information clearing house’ system as opposed to central databases, that we proposed in our Report on the Surveillance Society, but better worded and justified! In the latter case, the authors simply note that fewer than 30% of government IT projects succeed at a cost of 16Bn GBP per annum and that there should never be a general and aimless government IT program, rather there should only ever be specific projects for clearly defined and justified (proportional and necessary) aims.
It is an excellent report and probably unanswerable in its logic. Tellingly, The Guardian report contains no response from any government minister…
ubisurv 