You may recall me thanking Robert Alexander for finding a typo in my feed. He mentioned that he had been crawling blogs using the <source:blogroll> element, along with those using '/.well-known/recommendations.opml', and has now mapped a fledgling "RSS Blogroll Network" based on sites using these.
He has a blog post RSS blogrolls are a federated social network which goes into a bit more detail.
I will have to think about adding detection to /reader and then how the found blogrolls could be presented or, maybe, I could use it to automatically discover new (to me) blogs or ones that I'm not subscribed to.
As Robert says, the value of any network grows with the number of nodes, so it would be great if this idea became more widespread. Any methods of improving blog discovery should always be encouraged.
