This is my attempt to do something cool with DC's bike share data.
Need a sane way to test on smaller chunks of the data. For the smaller data: (not a random sample) head -n 1000 < data/... > test/data/... And need to link the test/src to the src (want them to be copies)
Basic ideas: -make a network of stations: stations as nodes and the rides as links -happiness around bike riding: cool, but hard -have time of bike returns, location of the station returned -look at happiness of tweets near that?