There is some kind of harmonic convergence going on with regard to social network analysis of blogs, blogrolls, etc.
Dave Winer and Jon Udell are co-conspiring and apparently there are ways to set this up. I'm waiting for a simple recipe for a best practice that will allow me to become part of the the data-pool.
Auto-discovery of mySubscriptions.opml. Things are really moving along! I just spoke with Dave Winer. It turns out that, by enabling uploading of mySubscriptions.opml, there already is a canonical XML representation of my feeds, suitable for harvesting. This is unrelated to radio.macros.blogroll, it turns out. ...
Discovering my weblog's neighborhood
Noted that Jon Udell is coming up to speed on the <link>s stuff. I'm starting to work on the other side, we have the content-side fixed, now I'm going to write a harvester, that I'm tentatively calling blogNeighborhood, which crawls the subscriptions list and blogroll three levels deep and builds a ranked list of sites that are nearby. I'm sure Jon and others will find this useful. I'm looking forward to running it myself.
(It's time for a set a glossary and cookbook, gentlemen. Or better, make a tool that says "click here to add this feature to your weblog.")
Meanwhile, I've had some nice interaction with Jon and (via Jon, I think) with Andy Edmonds of SurfMind.
It is precisely this kind of idea flow (e.g., from me to Jon to AndyEd to thee) that I'm interested in mapping. The tools are rapidly emerging, but we haven't done it yet. And Blogroll-diffusion is probably not going to be the same thing. The spread of Blogroll technology and of links to an individual's blog chart diffusion of a blogger's influence. This is is related to (but not the same as) an idea's spread and development from person to person and from context to context.
I think I have a pretty complete picture in my head of what I (we?) are after, but I'm not good enough with enough of the technologies to make it happen.
Perhaps I should dummy up a description....