When I wrote the post about the government building APIs after my Rewired State demo it had to be carefully written and edited as I knew about two really cool APIs which were upcoming: The Guardian’s OpenPlatform and TfL’s tube station data.

First of all, hats off to TfL for releasing APIs. We can now build AccessCity the right way; making it an aggregation hub weaving together data from TfL and other public location sources we’re talking to and user driven experiences all keyed together with hopefully everlasting foreign keys from these providers. This block in the TfL data is hopefully a sign of good things

<station id=”1000004” type=”tube” xmlns=”“>

From a first glance they seem to have a firehose for all stations, would be nice to have a method where you could just get back a single station. That way it could really improve the response of the service you build on it, no need to get and parse a really quite large chunk of XML just to get the information for one station, giving the data in small pieces encourages good practice for developers of getting resources just in time and caching for short periods, I’ll have to try and write some sort of wrapper onto this stream object which will let me create a more fine grained cache so I can reduce the overhead of parsing the XML everytime I want some info on a station and then rebuild that cache regularly. It’s not a big one though and I’d much rather have the data from a reliable non-scraped source. A big win. Thank you again. 

Oh, and then there was Open Platform. I’m obviously biased as I’ve been lucky enough to be a small cog in this machine. I’ll be blogging more on this one later on today and about how we built ContentTagger and the integration with the Cass Sculpture Foundation site.