As I mentioned in my proposal, I expect to rely heavily on Sunlight Labs for my basic congressional data. Specifically I will probably use the Sunlight Congress API. Another option from the same source is to download the latest API dump. This might provide better performance and might ultimately be more manageable since I can just dump it fairly easily into an SQL format.
To access the data in the API, I plan on leaning heavily on the Sunlight Labs PHP Library. Not only does it have methods and classes for Sunlight but also for some great data from OpenSecrets, such as lobbyist details, rep’s financial transactions, and various fundraising and campaign contributor information. What a boon!
Sunlight Labs also provides access to their Dumbone API, which they developed for their free “Congress” Android app. This looks especially useful for finding information on particular legislation or policy.
Also, GovTrack has a huge pile of source data. It might be a little more difficult to sift through but may ultimately prove to the best data source available.
I will spend the next week determining how to organize this data and also determining how to tackle the problem of historical data… It’s unclear how easy it will be access through the API, and if it isn’t, I either have to abandon the historical perspective or identify an alternative data collection method.