As some of you may know, I’ve been taking Computer Science 101 on Udacity, which is an online university which I think is fucking awesome. Their structure of 1-5 minute youtube videos with quizzes shoved in the middle totally works for me compared to stuff like Coursera and MIT Open University. But hey, what doesn’t work for me might work for you and vice versa.
Anyway, I’ve been relearning python (which I took in “real university”) and it’s been pretty interesting. Udacity’s first unit is actually way more difficult than the entirety of my CS class, which I found pretty surprising. The main thing that Udacity is really clear about is that programs are really just problem solving broken down into very basic steps. That’s probably worth the price of admission alone (which is $0.00).
Late one night while watching a Canucks game, I found myself having a problem that I always seem to have which is whether the goal that was let in Roberto Luongo was actually his fault, or if my bias was blinding me. Then it hit me. What if we could create an app that would go on twitter and tell me whether people online thought it was his fault? And that’s exactly what we did. After signing up for the Twitter API key, we got to work hammering on the Tweepy module and started to build code on previously documented code. It only took about an hour to build the first alpha version of the code, but holy shit it was the most exhilarating thing I’ve ever done on a computer.
We’re going to improve the algorithm that calculates the mood of twitter, but for the time being the code has spoken: