Right then: I’ve made a tentative foray into the world of webtools for urban living: The Farringdon Lane Docking Station on Twitter. Now I need help making it better.
There’s a long established trend of inanimate objects being on Twitter, including Tower Bridge and The River Thames. In both cases, they update people on crucial aspects of their current status: for example, is the tide in or out? I felt it was time this particular bike rack joined the service – as a keen user of Boris Bikes (an ironic moniker given they were commissioned by Ken Livingstone) I often need to check the status of the rack outside the Free Word Centre (where I work).
The Farringdon Lane Docking Station is a popular rack, one that is usually either (a) completely full with no space to park a bike, or (b) completely empty with no cycles available to use. For that reason, I often find myself trying to check the status of the racks online or on the move. Unfortunately, the workflow required is relatively difficult, involving several steps through the website or iPhone apps like Fliplab’s London Cycle. This actually takes a fair few frustrating minutes via a 3G connection, which is no good when I am in a hurry and keen to make a quick decision about whether to take a tube train or a bike.
I thought I could solve this by creating a twitter account that automatically updates itself, whenever the status of the dock changes. That way, whenever I think I may need a bike, I can simply fire-up my twitter application of choice and look at the latest status of the docking station. I don’t have to load unnecessary information about the status of every other docking station.
Unfortunately, my coding skills are minimal and limited to simple PHP. I don’t have the wherewithal to pull data from the London Cycle Hire site using their API. I solved this by making a customised RSS feed using the Feed43 (Feed for free) service. This scrapes the cycle hire map page (which has the status of all the bike racks embedded into it). Then I used Twitterfeed to post the results into a customised twitter account. The results are below:
The problem with my system should be obvious! If the chain of data was linked together properly, then we should be able to see every single change in status, not a huge jump from 16|0 to 8|8. This is clearly happening because both Feed43 and Twitterfeed pull data a long but regular intervals, not on a second-by-second or minute-by-minute basis. This is useless for my purposes. How can I improve it?