It’s been a summer of drama in the U.S.: the battles around the debt limit and the S&P’s downgrading of the U.S.’s Triple A rating, all fueling citizens’ already fueled concerns about the economy and jobs and the nation’s future. Now, the controversial Super Congress is taking shape to cut discretionary funding and direct spending by $1.5 trillion through 2021.
In light of the roller-coaster ride, any takers to see what they can do with the Sunlight Foundation Sunlight Lab’s recently introduced Influence Explorer Text API? Another step in the Foundation’s mission to make government transparent and accountable, the API provides real-time textual analysis for the Foundation’s Inbox Influence plug-in that lets users see in their emails the political contributions behind names mentioned in these messages.
We had a chance to have an email conversation with Ethan Phelps-Goodman, Project Lead: Data Commons, at the Sunlight Foundation to better understand what the API can do.
Semantic Web Blog: With lawmakers tapped for the Super Congress, we’re starting to see stories about their political patrons, their special interest connections, and so on. How can the Influence Explorer API be leveraged to help the public better understand these individuals and their political backgrounds/agendas? And if you’ve been using it to that end in recent days, come across anything interesting the public should know about?
Phelps-Goodman: Influence Explorer can give a great deal of background on the fundraising activities of members of congress. A recent series of posts from The Atlantic is a great example. Our own Reporting group also did a series of influence profiles of the Super Congress. Even more interesting will be the lobbying activity around the Super Congress’ work. Tracking the money spent on lobbying will have to wait until the disclosure period ends at the end of the year. But you can get a head start on things now at our Lobbyist Registration Tracker, where new lobbying contracts are updated as they happen.
Semantic Web Blog: Can you talk a little about the data sources?
Phelps-Goodman: Very little of our data comes to us in what would be considered “modern” formats. The FEC, for instance, releases data in a custom text format from a COBOL database. Other datasets come to us as poorly formatted Excel files. Others have to be scraped from official websites. Part of the value we provide is simply exposing these datasets through modern technologies, with powerful search functions and easy to use formats. Another way we improve the usefulness of the data is through encoding the entities in the various datasets–the politicians, companies, lobbyists, etc.–in a way that lets you make connections between datasets. For example, with Influence Explorer a user can easily see when a company has received an earmark after lobbying heavily on a spending bill.
Semantic Web Blog: Any hooks to Linked Data, or any other ways you’re harnessing semantic web technologies?
Phelps-Goodman: One place we do interface directly with the semantic web is through Wikipedia. Where possible, the entities in our system have been linked with a Wikipedia bio. This lets users of the API pull in the whole range of linked data that’s accessible through DBpedia.
Also, since our data and code is all completely open, the bar for integrating our data into a Linked Data framework is very low. [Its APIs follow a RESTful philosophy and return results in JSON.] And all of our JSON APIs could be transformed into an RDF version of the API without very much work.
Semantic Web Blog: You have mentioned in your blog that there are lots of additional opportunities beyond the Gmail plug-in of Inbox Influence – other email plugins, news story plug ins, integration into blog engines and content management systems and so on. What’s the grand vision of impacting all these arenas?
Phelps-Goodman: The goal is to bring data on money and influence in politics into a context that’s relevant to people’s lives. So far we’ve done that with personal spending (Checking Influence), email (Inbox Influence) and news (Poligraft). There’s the immediate goal of getting the user to notice something interesting about a politician or organization. But there’s also the broader goal of increasing awareness in people’s minds of the scale of influence companies and special interests exert on the political process.
Semantic Web Blog: What third-party developers/apps have you already begun to see leveraging this? Are any of them doing a particularly good job not just of presenting the data dots, so to speak, but helping to put information in context for the public?
Phelps-Goodman: One good example of third-pary development is the TexLege iOS app, which is built in part on our Influence Explorer and OpenStates APIs. The app includes just about everything you could ask for, including campaign finance information, voting records, schedules and district maps. This is all packaged into a slick app that seems genuinely useful to people working in and around the Texas legislature. We’re looking forward to the app being released for more states.
Semantic Web Blog: As we get closer to 2012 and the election season, what are your expectations of interest among the development community growing – and how might the greater accessibility of this data to the average voter mean to politicians? Do you expect to see any significant impact from that influence this coming election, or is it more likely to happen down the road?
Phelps-Goodman: I expect developer interest, particularly among those working in journalism, to grow considerably in the next year as election coverage comes to dominate the news. Papers such as the New York Times and Washington Post experimented with interactive content in the 2010 elections, and I expect many more news organizations will join in in 2012.
While it’s important that what data we have will be more accessible than ever, it’s discouraging that the scope of data available is narrower than ever due to recent Supreme Court and FEC rulings. I expect that SuperPACs and anonymous donations will play a much larger role in the election than they have in the past, which is an obvious step backwards for election transparency.
- Les Echos Select TEMIS to Enrich and Link Content
- WordLogic Corporation Sees Further Validation of Its Predictive Technologies With New U.S. Patent Allowance
- Open Data Institute Chairman Nigel Shadbolt Knighted by Her Majesty
- In Defense of PRISM's Big Data Strategy