Better (and More) Congressional Member Information

Over the summer, as lawmakers in Washington, D.C., haggled about the budget, energy, immigration and other issues, we've been at work upgrading the information about members contained in the Congress API.

Part of that process involved adding new elements to some of our member responses, but just as important is that with this update, the API is now part of a larger congressional data infrastructure effort that extends outside The Times.

First, we've added more detail to the member and member list responses, which now include a broader range of social media identifiers and other data that will help make it easier to connect to other sources of information. In particular, member responses now include Twitter and Facebook account names as well as the Facebook “id” used by the Graph API.

We've also added more details about lawmakers' websites, including the address and the URL of the RSS feed, if one is present. There are a number of official congressional sites that do not use RSS, and if you want to retrieve press releases from them (and those sites with feeds), we have released a Ruby gem that does just that.

In response to user requests, we've also added a new response for current members of the House or Senate. These two responses, detailed in the member list responses, include members who are serving in a given chamber as of the date of the request. For earlier congresses, this response provides the “final” membership list for that chamber and congress.

There is even more at work underneath the hood of our congressional data. The bulk of our data comes from official sites run by the House of Representatives, the Senate and the Library of Congress. And we're not the only people who are interested in compiling that data: organizations such as GovTrack and the Sunlight Foundation have been retrieving congressional data for years.

Instead of each maintaining separate code for fetching this data, last year Josh Tauberer of GovTrack, Eric Mill of Sunlight and I created a single repository for code that we'd all work on and try to incorporate into our organizations. The “unitedstates” organization also hosts other federal government data collection efforts, but our initial efforts focused on legislative data produced by the Library of Congress.

While we've all incorporated at least some of this data into our respective systems, at The Times we've most heavily relied on the member data assembled in another repository under the project. Using a single source helps us to avoid duplicative work as well as inconsistencies between sites and services.

Finally, in response to a request posted in our forums, the Congress API now supports JSONP requests. As always, let us know if you have questions or features you'd like to see.