Two outstanding contributions about the way the blogosphere works:

1) Eszter Hargittai posts a summary of her research into the viability of Cass Sunstein's republic.com hypothesis -- that the Internet fosters cyberbalkanization -- by analyzing link structures in the political blogosphere. Her preliminary findings:

Overall, it would be incorrect to conclude that liberal bloggers are ignoring conservative bloggers or vice versa. Certainly, liberal bloggers are more likely to address liberal bloggers and conservative bloggers are more likely to link to conservative bloggers. But people from both groups are certainly reading across the ideological divide to some extent.

Two other interesting findings: balkanization is not increasing over time, and -- sorry, I can't resist this one -- "We found that about half of the [cross-ideological] links represent what we classify as strawman arguments. The liberal bloggers in our sample are more likely to engage in such cross-linking than the conservative bloggers."

2) Carl Bialik has a great piece in the Wall Street Journal (no subscription required) that looks behind the numbers floated around with regard to the number of blogs out there and how blog traffic is measured. These paragraphs might make some blog triumphalists pause a bit before declaring the death of dead tree media:

Advertisers may not be happy with [standard blog counters], since they count total visits, and not the "unique visitor" figure that is the standard currency for many kinds of online advertising (advertisers don't want to pay twice to reach the same reader). "That's a big issue," Henry Copeland, founder of Blogads.com, told me at a conference last week. "We're very aware that's a flawed number."

...ComScore Media Metrix and Neilsen//NetRatings are the sources most often used by online advertisers to track unique visitors. Neither tracks blogs as a matter of course, though comScore did look up traffic for 13 prominent blogs in April, upon my request (I picked ones from the top of the various rankings). Just five met the company's minimum threshold for statistical significance of about 150,000 monthly visitors. Media and gossip site Gawker had the most, with 304,000 unique visitors. The others that cleared the cut: Defamer (287,000), Boing Boing (250,000), Daily Kos (212,000) and Gizmodo (209,000). Among those that didn't were prominent political blogs Instapundit, Power Line and Eschaton. (I asked NetRatings about the same 13 blogs, and it had reportable data only for Defamer, Daily Kos, Boing Boing and Gizmodo -- and the sample sizes didn't meet standards for statistical significance.)

ComScore and NetRatings both recruit panels of online users who agree to install software that monitors their behavior. The companies use sampling techniques similar to those of political pollsters.

By point of comparison, comScore says the New York Times's Web site had 29.8 million unique visitors in April.