We now have 11 months of aggregated data on UKCrimeStats – a special thanks again to our Chief Data Architect for working to make this available asap. We have some new upgrades coming through shortly which we’re excited about too – watch this space.
On a separate note – we are aware of – and will always be open about – some errors in the data (which we take from police.uk, purchase unchanged) which have still not being fixed in the latest despite multiple proddings of the NPIA and RKH – amongst the worst is probably Cheshire Police (who we don’t blame) – their ASB figures for the first 3 months – just look at the chart and Sussex neighbourhood totals for December 2010 and January 2011 – still unavailable. We track these and others on the Forum part of our website.
I’ll be adding some more posts on the forum about data I’ve found which look suspect very shortly. Notably;
The original open data philosophy is just get it out there as fast as you can, treatment no matter how flawed it is, with little or no regard to data governance. This is flat wrong. It is as academically misinformed as the “experts” who told us in the early 90s, all that counted regarding privatisation in the ex-Soviet Block was speed and a liberal market economy is just a series of institutions – central bank, regulators, etc. – as if trust and corruption were immaterial. This misconstrued blind faith has in the worst case, arguably turned Russia into a nationalist oligarchy rather than an outwardly open democracy.
We know at UKCrimeStats how expensive and time-consuming it is to clean up the crime data and we don’t receive a penny for it or even a word of thanks. I could understand that in the first couple of months that not everything was right with such a large dataset and have said so frequently. But we’re now 10 months in, they’ve had all the time in the world to sort it out and there are no excuses any more.
Open data from government should not be immune from commercial standards of accuracy – especially if you’re paid by the Home Office allegedly £300k to release it in a format for developers and have now started charging for advice on the api. Beat that for a vertically-integrated conflict of interest.
That’s why I am sceptical about the case for a £10 million Open Data Institute quango when just minor increases in data responsibility, resource and political will could lower the barriers to entry for developers and deliver the £6 billion open data economy (and possibly much more) the likes of Dr Rufus Pollock dreams about.