FCW Insider: How to improve the virtual town hall

President Obama generated a lot of interest with his virtual town hall on March 26 (see FCW's story here). People were clearly intrigued by the idea of a president answering questions submitted through the White House Web site. All told, 92,937 people submitted 103,996 questions across eleven categories.

In addition to submitting questions, people were invited to vote on questions submitted by others, with the questions receiving the most votes rising to the top of the list.

This was a tantalizing taste of e-democracy on a national level and, no doubt, a sign of things to come. Which is why I think it is worthwhile, now that the buzz is gone, to stand back and ask, But was it any good?

Specifically, I asked that question (in so many words) of Kim Patrick Kobza, president and chief executive officer of Neighborhood America, which develops enterprise social software for business and government (Earlier this year, Kobza wrote a column in FCW on the role of public comments; you can read that here).

Here is what Kobza had to say:

The fact of engagement is always a positive and it is good to see the administration and government trying some new things. That being said whitehouse.gov is not an example of a model for public engagement.

The good

The application of whitehouse.gov is useful in the sense that it at least provides a public perception that the questions being answered are “real questions”. Not media questions. Positives include:

The opportunity for a small group of citizens to be heard.
Well defined participation guidelines.
The creation of clear expectations.
A good balance on attribution and identity.

All of these attributes are very well done.

The not so good

That being said, the application falls far short of what is possible. This is true for several reasons:

One way to think about it is in the context of network science. Think about citizens as a big network. That network can create value in many ways and for many purposes. Is identifying “questions” to ask the President compelling? Is that an appropriate purpose for a network – one capable of doing really big things? Would you hold a meeting to decide what questions a President or elected official should answer by the group? And does the application build strong network value – even if the purpose is appropriate? Not really. Here is why.

The application is poorly structured. It doesn’t create the conditions that would enable creation of network value from the citizens participating.

Lack of compelling purpose. The citizen communication is limited to two methods:(1)  “idea creation” and (2) to what is essentially voting on ideas (binary choices up and down). There is no story – no opportunity for citizen narrative that builds on the wisdom of each other. Notice the difference in the scenarios below:

* Scenario 1: “Tell me what you think” [results in an unmanageable amount of data, not well categorized, characterized, and classified].

* Scenario 2: Do you approve of the choice – if so why? Do you disapprove of the choice, if so why? Do you have any comments that would make the choice better if approved? Do you have any other comments? [

* Scenario 3: Scenario 2 plus a give and take between and amongst peers. This was the application used for Flight 93 which is basically an alternative analysis of a limited number of choices, and narrative that provides logic to enable agencies to make choices that blend alternatives.

The whole point of questions is not to find the question that the whole group wants to ask and that is predictable – but to enable cognitive outliers to ask the unpredictable question – to promote ways of thinking about problems (and solutions) that are uncommon. The application promotes “group think”. See Infotopia by Cass Sunstein.

Fatal flaw: Where the application breaks down is the volume of solution possibilities – it is not credible. Think about it this way. Most choices in policy are essentially an “alternatives analysis” -- a choice between different but well defined solution possibilities. What is happening in whitehouse.gov is that the public is being asked to create and compare an almost infinite number of choices. Who could possibly read 53,808 questions? 53,808 alternatives? Most people understand that it just isn’t credible to think that that is possible. So the application creates an illusion of meritocracy when in fact the statistical significance of any one person’s participation is negligible. Put in a network context, it treats choices as a mechanical data management exercise, rather than as one that develops and advances logic to why choices might be made. There is no exchange that takes place between members within the network (citizens) and that limits the network value (intelligence) being generated.

Another thought on design. In my view, when building citizen participation, and more importantly, in discovering great ideas it is important that these types of sites build social attention. Social attention comes from novelty, uniqueness, and experiences that are fun and that promote learning. The design of this site again is not interesting. It might be argued it was interesting enough to gain 54,000 ideas. The other side of that argument is that many congressmen receive tens of thousands of emails a week. This application has not upped the standard or participation.

The value test

At the end of the day let’s ask one question – Does the application accomplish anything truly remarkable?  Do you come away from the experience of the application saying – that is a better result than we otherwise could have achieved using traditional means of participation? Was there anything in the questions promoted that was novel, unexpected, and unique? Did we enable participation by citizens who otherwise would have been excluded? Would the application appeal to a broad range of citizens?

These things being said, no, I believe that this is not a good example of how social networking adds value to public processes. In fact, in some ways these methods hurt more than help because they set the wrong citizen expectations. They are not our best effort so to speak. So citizens will ultimately be discouraged because the application as used, will not achieve results. When we don’t get results in anything in life we lose trust. See Steven Covey, The Speed of Trust.

The path ahead

The thought process of submitting questions and letting the administration choose questions is a good one – standing alone. Trying to add meritocracy features to an almost unlimited amount of data diminishes the value and credibility of the system.

However, for real policy choices, there is a big opportunity to use citizen networks to add transformative value in a social collaboration. But to do this requires a much deeper appreciation for the complexities of network behavior. As in any complex system, small changes make big differences.

How to improve:

• Using social tools for appropriate purposes (purposes where social collaboration in network can provide significant improvements in marginal value).
• Define different but limited choices. Choices may be different but manageable. Enable choices between 4 or 5 solutions. [i.e. Do not create thousands of choices]
• Do not overly segment topics. Over segmentation in categories confuses most citizen users because they see multiple categorizations as adding complexity.

What is missing:

• A compelling purpose.
• The ability to combine meritocracy features with a citizen’s ability to add narrative to solutions. We learn from stories, not from data. Citizens contribute to solutions by telling stories. Narratives contextualize the citizen inputs.
• The ability for citizens to connect with agencies and with each other. Networks are only valuable with member exchange.
• Ultimately, an ability for management of structured feedback – a way to respond to citizens, to let them know that they have been heard and are valued. For instance social collaboration and CRM integrations.
• Visual interest that builds social attention.
• The ability to publish ideas into other social networks so as to build network value by tapping into multiple networks – share ideas that enable ideas to become “viral” which builds citizen inclusion.  

The Bottom Line

I think that the dominant theme is that citizen networks are complex – especially in contributing to public policy outcomes. Success depends upon much more than data management and mashing up technology features. We spend a lot of time studying what works and what doesn’t, and have had experiences to validate over 10 years. That has built an appreciation for how hard it is to get this right.

What we showcase at a Whitehouse.gov should be a shining example of our best efforts. It should be state of the art and based on our deepest knowledge and experience. It should not be, and does not have to be, based on experimentation.  That is my first impression of the site. Can and should be much better.

Here's how to connect with Kobza:
http://www.linkedin.com/in/neighborhoodamerica
http://twitter.com/kpkfusion
http://www.inflectionbykim.com (my podcast series)