I want to propose a hypothesis.
Suppose our new president -- let's just call him "Obama," for simplicity's sake :) -- gets serious about using the Internet as a tool of governance. So, he takes his email list and uses it to kickstart a new e-gov social network. In fact, McCain provides his email list, too. So, let's say we have five million on this social citizen's network. Let's say it prominently features blogs and forums. Let's say after two years there are 30 million registered users, and some good percentage of those are at least occasionally active. Of course, I'm making all of this up.
Now, the problem the Internet has faced almost from the beginning is how to scale conversations. We've solved it time after time, whether it's threading and forking Usenet discussions or Amazon's reviews of reviews. So, let's imagine that this new social network solves the problem through a combination of forking and a reputation system more or less along the Daily Kos lines.
So, 30 million people are engaged in vital conversations. Some people gain prominence in discussions on particular issues. The Obama administration notices this. The relevant government policy makers want to engage in these conversations, because otherwise the 30 million citizens feel like they're being ignored. The emergent discussion leaders become the online points of contact between the administration and the conversations, because that's how those conversations scale.
For example, let's say a participant known as PolarKing111 gains an enormous reputation because he writes about climate change so knowledgeably and passionately, because he engages with all sides in the discussion with respect, and because he's so good at representing all the various opinions. Administration officials engage with him on the site, often in a spirited back-and-forth. He ably represents the concerns emerging from the many discussions on the site. It's a public dialogue with just enough structure, one unlike any our democracy has seen.
Inevitably, one day in early 2011, the media will discover that PolarKing111 is a 15 year-old girl, but that's not my point. My point is that the emergent online discussion leaders play a role unprecedented in our democracy. They are not elected yet they represent us. They are not members of the government yet they directly affect government. They have some power but the power comes from an emergent process. We don't even have a word for this role.
Of course, I'm making all of this up. It's just an hypothesis. Yet, it's easy to imagine something like this happening, while it's simultaneously impossible to predict exactly what will happen. Nevertheless, there's a strong possibility that some form of e-gov social network will emerge, either from the government or from the people. This social network could create new roles or processes of democracy that could well turn out to be quite important.
But, just as Facebook can alter the nature of privacy by deciding whether or not to set a checkbox on or off by default, the roles and processes of this new layer of democracy will depend to a large degree on small decisions about how the software happens to work. If the designers of the network decide only to count the "thumbs up/down" votes of people who have logged in, or who use their real names, or who have posted more than five times, PolarKing111 might be out, or we might go from having 300 conversation leaders to having 3,000, or perhaps the process will tend to favor those who respond quickly and briefly as opposed to those who contribute longer posts. There are an indefinite number of variables and factors. Tinkering with any one of them could result in large changes. That's the nature of emergent systems.
The people's most direct interface to their democracy thus will be susceptible to big changes caused by tiny software decisions.
Personally, I think that's likely to be a good thing, especially if the designers engage in a continuous process of tweak-feedback-tweak. But, this is our democracy they'll be tinkering with. And, we hope, continuously improving.
Democracy mediated by software: Could be quite wonderful. It certainly will be at least a little bit scary.