I did a small survey with James of publicly available government data to see what the obvious targets would be for an interesting mashup. It’s an interesting but very limited landscape.
One set we found was the incident reports of Dutch first aid services which are published at P2000 and parsed by the site Alarmeringen.nl into nicely Google Mapped displays with RSS feeds and SMS alerts.
What it didn’t have yet was a decent output to Twitter. So to prod things a bit I made twitter feeds for the four major cities in the Netherlands. Meet:
@al_amsterdam@pk2amsterdam1 @al_haaglanden@p2khaaglanden @al_rijnmond@p2krotterdam @al_utrecht@p2kutrecht
I think it’s easiest to attribute the quote to Matt Jones who in this presentation noted that the best people on Twitter aren’t in fact people. I agree but here the Tweets represent the aggregations of groups of people: the incident and its cause, somebody noticing it and dialing dispatch and finally the fire fighters, paramedics or policemen moving to the scene2.
It’s interesting imagining what has happened and where, but I don’t really see this feed changing people’s behaviour.
Ideally we would achieve a greater granularity intersecting your location with the path of the dispatch, so when you see a police car whizz by, you’d get a tweet later telling you why it was dispatched and what the result was.
As Pascal points out in his comment, this data was not supposed to be public at all but is de facto public because of technical architecture decisions. It is clear that these services have not been thought out with an internet enabled world in mind.
Given that I was more interested in this as a proof of concept than to make it a long lasting durable service anyway, I’ll take down my twitter feeds after I’ve given its subscribees time to switch to Remco’s accounts.