Via Mary Hoder , it appears that the entire blogosphere encompasses 6.5M blogs and that there are about 600,000 posts per day.
If you assume each post is 1k and that it all started in a flash 3 years ago, the entire blogosphere archive is only 540GB. Given that text compresses 80%, we are probably talking about storing the entire archive in ~100GB. (A 100GB HD today costs around $100). But, lets follow Mary’s logic in advocating PubSub’s business model and assume we only want 30 days (older stuff has probably reached google). Then you only need 20GB uncompressed and 5GB compressed. Downloading 600Mb on a 1Mbps link will take you 1.5 hours. Some people will object that the blogosphere is growing. I would respond that it is unlikely that the blogosphere is growing as fast as hard drives and bandwidth costs are shrinking.
What would you do with the entire blogosphere downloaded?
* blog surf faster!
* personalized searches e.g. find me this word in all blogs reachable from the set you read regularly
* ideology searches: find me this word in the right-wing blogosphere
You tell me?