There was a really interesting article in the New York Times yesterday that had a headline attention grabbing headline of Foundations Find Benefits in Facing Up to Failures. I had two reactions: "how refreshing" and "well, sure."

Sunlight deliberately set out to be experimental – to throw ideas and projects on the wall and see if they stuck and if they didn't, to stop and figure out why. This was key to our grant making strategy as well. When you hope to be on the cutting edge you expect some things not to work. So we figured we'd win some, lose some. That seemed right to us. Fascinating how risk averse some of the truly big foundations have been and unwilling to admit, until recently, that some things just don't work.

I've been reflecting what has been working well and what not so well at Sunlight because I've been asked to present at a conference next week in Boston — The Internet as a Public Good (co-convened by the Mozilla Foundation, Harvard Business School and The Berkman Center for Law and Society at Harvard Law School) a real-world case study where we took seriously the potential of the Internet for a project and then ran it right up to the edge of what's possible, but a project in which we had not quite gotten to our goal (a failure?). Obviously the point of this exercise at this upcoming meeting is to help the larger group look at what works and what doesn't work on the ‘Net and to try to contemplate what's next in the 'Net's innovation.

So my presentation is going to focus on how far the Internet-based feedback loop can be extended to effectively influence our elected representatives, using our "What's McConnell Hiding?" campaign as an example. We have found that sometimes collaborative online organizing has worked brilliantly to produce results — think how citizens, urged on by an active blogosphere drove the passage of the Coburn-Obama bill — and how in the case of the secret hold on legislation that would require Senators to post their campaign finance information electronically – it has stalled.

The ‘Net appears to work best as a system for focusing a disparate community of concerned individuals on a common goal. It's clear that people online enjoy the scavenger hunt game of digging out some kind of hidden fact and that the new age wisdom that "many hands make light work" definitely holds when you look at a political task, like gathering attention around a bill. But the feedback loop to incumbent politicians is still a weak one.

Is there a technology solution to this problem? I'll report back what those who attending the conference have to say.