Weiyu Zhang

This afternoon, sitting on the small deck of Gecko, surrounded by the warm and humid air, hidden in the mild noise, I got myself ready for the long-delayed conversation. Three law professors, Lawrence Lessig, Cass Sunstein, and Yochai Benkler, are together to talk about blogs...through their writings. :)

Here is my record of this conversation. Lessig is L, Sunstein is S, Benkler is B and me is me.

S: Blogs are an unlikely venue for Habermasian public sphere because of fragmentation and polarization.

L: Whether blogs democratize should be examined within the constraint of their codes / architecture.

S: Alright. The codes for blogs are like-minded groups that are isolated from each other.

B: No. It is not true. The architecture of blogs is the power law distribution in general and the long tail distribution within like-minded groups.

Me: What is the power law distribution?

B: You guys should read Science and Nature. It means that most people still visit a few superstar websites. So fragmentation is not a problem.


Me: Even though people go to the same websites such as google.com, they could selectively choose information that echo their opinions. At the level of individual exposure, it is still fragmented.

S: I agree.

Me: But selective exposure itself has to be examined rather than being assumed. The first step of assuming preference for the like-mind might be wrong.

B: I agree. You have to provide empirical data.

Me: How does the long tail distribution help to prevent polarization?

B: It means no superstars can totally dominate a small world. Many low end sites are still connected to each other if you look at a smaller scale cluster.

Me: OK. So it prevents domination or centralization in small clusters. But how does it prevent fragmentation and polarization?

B: Well, it is actually that there is no fragmentation so there is no polarization.

L: The distributions you talked about could be changed, do you know? The codes that determine the dynamics are open to changes.

B: Oh yes. That is why we should pay close attention to how policy regulates the codes.

L: Yes, the constraints of law take advantage of codes to make cyberspace more regulatable. The government can interfere with the formation of the two distributions you mentioned.

B: But so far the government has a harder time to control cyberspace than mass media.

L: Are you sure? The government enjoys controls that they cannot have before. For example, it can even censor private communication.

Me: Probably we should not only use mass media as the benchmark. Other communication modes such as interpersonal comm. should be used as reference point, too.