The launch of http://data.gov.uk (runs on Drupal), a central place for making uk government datasets available to the public can only be a good thing. It will prove useful for geeks making mash-ups, but also hopefully help journalists, policy-makers and the public make informed judgements about everything from road safety to public expenditure.
However, as with most datasets, they provide "old news". Data collection is not a trivial process, and then the data has to be validated before it is published. This all takes time. And by the time the data is published, it is often way out of date. This uncertainty is a real problem when you want to make a decision about what you should do in the future.
Take for example the data released on accidents involving cyclists that was released earlier this year. It suggests that probably the most dangerous part of my cycle commute into central London is at the side-junctions on Uxbridge Road. But that was in 2007. What is happening now the road layout has changed, and even more strangely shaped cycle lanes have been added?
This problem is particularly hard to overcome in the public policy sphere, where the data has enormous social value, but no immediate financial value. But what if this data was collected and published by live on the web?
The technology already exists - almost all websites these days are database powered, and could be used to serve up live data as it is entered. In the long run, it would probably prove much cheaper too (this is speculation - no data is available), as the data collection process would be crowd-sourced.
We still couldn't predict the future, but we'd see the impact of a good policy much more quickly, and we'd at least know what was happening now. Even if it would make us all slaves to the numerati, it is surely better than making decisions based on guesswork?
