Another bit from ArXiV: Learning and innovation expand cooperative network topologies
. From the abstract:
...long-term learning and innovation, when acting together, extend the range of network topologies enabling the development of cooperation at a wider range of costs and temptations
It's very difficult to translate this to other kind of social networks such as Facebook, since the topology in these networks seems to be fixed; it's very difficult to drop a connection. But how can you move it to blog social networks, for instance? Learning
can be assimilated to exploitation (keep linking what you already know) and exploration (look for new blogs to link). The authors say that
innovation makes the level of cooperation less dependent on the actual network topology
That is, if blogs explore more thoroughly, the power law structure of the network, usually dominated by a few A-listers, have much less influence (presumably making the A-list more dynamic). Besides, long-term learning promotes cooperation: if "cooperation" is "mutual linking", that means that if you research links for your posts, and reciprocate to those that link to you, the network will ultimately be more stable.
Of course, translation can't be performed straight away, and specific situations and dynamics of the blosphere would make a major difference. In any case, read the paper (which, BTW, has a curious structure).