Flickr - © Big Fat Rat

This study analysed a selection of threads from the Metafilter community blog to determine how it related to previously proposed characteristics of Communities of Practice — namely the nature of knowledge sharing; “Legitimate peripheral participation” (how newbies become accepted); Warrants (relevance) and Identity/Reputation.

They put forward evidence for each of these processes within the community blog. On knowledge, selected excerpts relate to the nature of the initial post (should be fresh and thought provoking) and the quality of the resulting discussion. On warrants and LPP, they note that — despite the system being reasonably open — the old timers having developed methods for discouraging new posters that don’t seem to add value, and encouraging those that do.

One of the interesting points for me was the idea of signal to noise and the fact that one person’s signal is another’s noise — even the old timers seeming to acknowledge that the noise is part of the acceptable baggage of the community.

The authors offer some recommendations on community design, though most of these are not particularly new — ie. the need for the technology to fit the community, not vice versa and the need for moderation in some shape or form. Also, the utility of identifying members (length of membership & contribution) to confer some sort of authority.