Government Transparency is almost as big a buzzword as Web 2.0 lately. And for that matter, you would think the two would go hand-and-hand. With all of the killer technology available in our information age, one would think that it should be relatively straight forward to get a picture of what our government does in return for our tax dollars and votes.

The sad truth is that it is extremely difficult to decipher anything that goes on in Washington, from laws that are passed to how tax dollars are spent.

Jason recently pointed out an article on Why Congress needs version control. I think its a fantastic idea, and certainly serves as an example of how today’s technology can be applied to the governmental processes. However, I think we can go well beyond that.

I’ve spent the last several months scouring the net for what information is publicly available and from where. There are a number of government sites that make pieces of information available, and private sites that take those pieces of raw data and try to make something intelligent out of it.

For example:

  • The Federal Elections Commission (FEC) makes available the donor records of everyone that contributes more than $200 to a politicians campaign. But who wants to download all that data and crunch it through excel? I tried, trust me its interesting stuff but tough to glean anything useful from it.
  • GovTrack.US does a fantastic job of tracking every bill that hits the floor of the House and the Senate, complete with voting records of politicians. They even make it available in RSS feeds. This is cool, but as a standalone tool you can’t figure out much more than how many Post Offices Congress has named this year.
  • Maplight is trying to marry these two concepts together and tie voting records with PAC contributions. That’s an admirable effort and I think with a little data modeling and consolidation we could use this as a starting point to gain some very useful knowledge.

The examples go on and on, and if you have some good ones, I’d love to see them. But the point is when you step back from all of this it starts to resemble a corporation, or “the enterprise”, with multiple data points and silo vendors trying to address individual application needs.

Enter Business Intelligence.

First we start with a data mart. We identify all the publicly available data points (there are far more available than I mentioned above), and we create a data model that does some cool things like create unique identifiers for congressmen and candidates and Bills that hit the floor, etc. It will take some leg work to get all of the ETL loads, or automated data pulls, set-up. But as the sites mentioned above have demonstrated this is not insurmountable.

The next step is the presentation layer. Once we have all the data and its modeled appropriately, the sky is the limit here. If built correctly, every U.S. citizen could have an almost real time dashboard on Congress, or run an ad hoc report on budgetary spending, or create a scorecard on their local Congressman, or… (more on these in a later post).

I think the key to this is to treat it like an open source project. So perhaps MYSQL and something like Pentaho are in order. This would keep commercial conflicts out of the equation and make people feel like they are more apart of the process.

I don’t see Washington creating a tool like this anytime soon, therefore we’ll have to take a step forward in doing it ourselves. Any takers?