I should remember to identify my literature review posts as you indicated above. I knew you had written about this in the past, I just couldn’t find it when I went looking.

What they told me they’re going to be doing(they haven’t started yet, so that’s why you don’t see anything), in the email they sent me, is parsing the COinS tags out of RSS feeds. They’ll be providing a COinS generator form and apparently expect people to fill out the generator form and then paste the tags for reviewed articles in their posts. This means that both COinS for the current post and COinS for reviewed articles will both be picked up, but AFAIK they aren’t expecting to find COinS referencing anything but reviewed articles, and for 90% of the blogs in their system this probably will be true. They’ll just have to make an exception for those of use not on scienceblogs.com. I kinda like being an edge case, anyways.

I’m guessing the idea here is just to get sciencebloggers, even the ones who’ve never typed an angle bracket in their whole life, to start including machine readable citation data in their posts. There are more extensible ways of doing this, and I think they’re aware of those, but are just trying to hack together something relatively novice-friendly that writers at scienceblogs.com will be able to use and that can help readers cut through the chatter a little.

One easy way to add semantics to a review is to link to the paper like normal and add rev=”review” to the attributes of the ‘a’ element. Postgenomic uses this to identify reviews of papers: http://www.postgenomic.com/wiki/doku.php?id=markup

There doesn’t yet seem to be anything uniquely machine-readable about the BPR3 marks, so I can’t see a way to aggregate them so far.

I also have a tendency to go off on a political tangent, and the way I deal with it is to keep it on its own blog.