Talking heads

Last weekend, my friends and I participated at DeveloperWeek 2015 hackathon. That was our second hackathon (out of 4 people, 3 met at a previous hackathon at LinkedIn). 

This time, we were way better prepared in terms of organizing our work. Building on the previous experience, we have decided on the project and the technologies in advance. Our core technologies were Python and Django for the back end, and Angular.js for the front end. The description of the project can be found here.

We used the following APIs for the project:

- Sunlight API to look up candidates: https://sunlightlabs.github.io/congress/legislators.html
- Politifact PersonLookup API: http://www.politifact.com/api/statements/truth-o-meter/people/barack-obama/json/?n=1

- Politifact All People API: http://www.politifact.com/api/people/all/xml/

- IDOL on Demand: https://www.idolondemand.com/developer/apis/detectfaces#overview

IDOL on Demand provides facial recognition API, and we used it to calculate the location of a politician's mouth in order to attach a speech bubble to it. We intended to use D3 in order to generate the word cloud demonstrating the frequency of words that a given politician uses, but, unfortunately, we ran out of time.

In the end, I am pleased with what we achieved in such a short period of time. Each new hackathon provides a wealth of experience to build on in the future.