Dec 13, 2009

Open Government Needs Data (not just provide it)

Flickr network universe 2005Last week was a "Big One" in dot-gov. The much anticipated Open Government Directive was announced. Peter Orszag, the Director of the White House Office of Management and Budget, summed it up.
The directive, sent to the head of every federal department and agency today, instructs the agencies to take specific actions to open their operations to the public. The three principles of transparency, participation, and collaboration are at the heart of this directive. Transparency promotes accountability. Participation allows members of the public to contribute ideas and expertise to government initiatives. Collaboration improves the effectiveness of government by encouraging partnerships and cooperation within the federal government, across levels of government, and between the government and private institutions.--Read more at the Open Government Blog
Out of the box, much of the specifics in the Directive was focused on transparency and the opening of public data sets. For example, in 45 days, three heretofore unreleased high value data sets need to be available on [Sunlight Foundation parsed the directive and published a timeline for agency requirements.]

Requirements for collaboration and participation are less specific. [See Nancy Scola's very good summary and analysis on the Directive content on Tech President.]

Getting ready for collaboration and participation is what I wanted to talk about.

There's been plenty of work on engaging with the public--from the March Open for Questions exchange fueled by people submitting questions and voting for the ones they wanted the President to answer to the development of the Directive itself.

Chris Brogan is a well-known new media marketing guy. He suggests that people try and think like "-ologists" and learn from -ologists (anthropologists, sociologists, psychologists) in their outreach. Check out his 45 second video.

It's critical for government to take a broad look at how we interact, how people interact, and what it might mean. Taking a structured analytical approach--dare I say scientific--becomes critical. Open government efforts need to be structured to include measurement and evaluation. Best practices can't be defined without understanding the variables and the inputs. Did we get a good result because we were lucky? Did we get a good result because we think we did? Or, did we get a bad result that we believe is good? Did we get a good result because we went into the experiment as an -ologist?

This post was inspired by a reference in my Sunday paper to "Web-Based Experiments for the Study of Collective Social Dynamics in Cultural Markets" (PDF). The authors, sociologists Salganik and Watts, provide some perspective on how the idea of popularity can influence what is popular. The researchers found that there is randomness in the creation of popularity within a network. A song, for example, becomes more popular as other people in the network show interest. In another network, a different song could be the winner. That's why it's hard to predict the next hot song, band, movie or toy. It's somewhat random and unpredictable.

The got me thinking about how important network dynamics are to spreading ideas, forming consensus and developing meaningful interactions. Meaningful for citizens as well as government. And, it got me thinking that people are already studying these patterns in other venues.

This isn't getting any easier.


  1. This is incredible dicey in the collaboration context. Say you want folks voting ideas up or down - if you jazz up design for a leading idea, you may bias future votes. Don't differentiate the design, you may get a participation drop-off from boredom. Expanding inputs is terribly important, and terribly difficult.

  2. @adriel yeah, not easy. But we need to up the value of best practices beyond what seemed to have "worked" to using data and social science to make decisions and design programs. If Amazon and Disney and Microsoft and Facebook are studying behaviors why should dot-gov go with gut?

    We should be able to use social sciences to better design collaboration systems. At least that's what I think. :-)

  3. That's a fine step and will be assumed to be a great move towards the final assessment.. if the projects are open to the public, the reviews will be faster and will help to make the things happen rapidly.. even rapid than before.

    cheap exotic limos


I hope that you will read and comment, ask questions and make suggestions. I just ask that you simply stay on topic, respect other people’s opinions, avoid profanity, offensive statements, illegal content, and other unpleasantries. Since this is my personal blog, I reserve the right to delete any comment.