India, bureaucracy, Bihar, 2003. By Jan fullscreen

Who In The World Is My Representative?

What if there was a resource that aggregated every government official from around the world down to the district level? If we’re going to be serious about government transparency and digital democracy, this is sorely needed.

Information about who is in power is notoriously hard to get. Sure, who the Presidents and Prime Ministers are is pretty obvious, but scratch that surface and the data becomes much more obtuse. Yet this is something that many, if not all, organizations can use to their advantage on a daily basis. Local empowerment means having a clear road to accessing these people. More often than not, who those people are remains a mystery. On the ballot, they’re just a series of names, next to a political party, with less clarity than one would find on their Facebook profile.

Creating such a tool is pretty straightforward. A database that serves as a sort of github meets DocumentCloud meets a Wiki. Most, of the code already exists and is open source. Add a clever name for the project, a nice skin and a few dollars to support the hosting it can be established quickly.

The hard part is actually putting together the information and maintaining it as the circumstances change. Harder still is to connect those politicians to maps, given the heavy gerrymandering that politicians and parties engage in.

With so many groups interested a solution such as this, it seems that this type of endeavor would be less of an undertaking from a specific organization, but rather a task best suited to crowdsourcing. Perhaps we could even call it gerrysourcing.

The US-based company MobileCommons has been a pioneer in demonstrating that data such as this can be enormously useful. During the anti-SOPA campaigning, they powered a click to call campaign where users could enter their phone number, street address and zip code. From there they’d receive an automated call from Tumblr CEO David Karp about the bill, and then be connected with their congressperson to voice their opposition. This generated 400,000 calls and helped to promote the groundswell that defeated the bill

This was powered by the open source Legislative Lookup tool they had developed to deal with the tricky task of connecting a caller directly to their elected representative without haveng the rely on the old fashioned and inefficient Congressional switchboard. Instead, the information is taken from a Postgres database with PostGIS extensions. The shapefiles are taken from

Burundi Provinces

Part of the problem is that the shapefiles of the districts change. So do the number of districts and the representatives of those districts. Any tool that sources from a database including the information needs to change as the database does. But the database needs to be kept up to date. For example, Burundi is divided into 17 provinces, 117 communes, and 2,638 collines (hills).¬†And this changes. The US National Geospatial-Intelligence Agency (NGA) has this geospatial intelligence and provides complete files of geographic names information covering countries or geopolitical areas. The files are not in a particularly accessible machine-readable format online, nor shapefiles. The result is that it’s currently easier to get detailed information on where Burundi exports to than political data:

To get the information, make sure it’s up to date, and to make it useful is a lot of work, but not if there are a lot of people working on it. And the payoff can be enormous. Applications can range from those like the movement to protect the internet by stopping SOPA, to more sophisticated election monitoring tools, to improving access to clean drinking water. And it’s easy to imagine how this can interface with such proposals as a github for legislation.

It’s time to redraw the maps and to make sure that we know who has been elected to hold responsibility for the slivers of land inside of those lines.