Sumana Harihareswara, 22/04/2013 21:02:
A few people were passing around a paper that "could be potentially very useful for EE. tl;dr: just look at figure 8." -Giovanni Luca Ciampaglia
http://cs.stanford.edu/people/jure/pubs/language-www13.pdf
Steven Walling wrote:
Fascinating. Someone should _definitely_ run a similar study of how it takes editors to pick up markers of in-group Wikipedia jargon, if they haven't already. A lexicon would be fairly easy to develop.
The point however is different: «A way to summarize this finding is to say that users generally die “linguistically old” (i.e., at a stage when they have relatively little reaction to linguistic change), no matter if they contribute relatively few posts to the community, or if they are heavy contributors». This means that, by checking if they're getting linguistically old, we could predict which users (old or new) are dangerously approaching their wikilifecycle end. What to do with this information would be another challenge, but it would be interesting to find some project where the local coaching group would be interested in having such lists/data to work on. Of course the Germans come to mind, but a smaller wiki may be more viable to start with.
Nemo
Tilman Bayer added:
See also this paper: http://www.mpi-sws.org/~cristian/Echoes_of_power.html (reviewed here: https://meta.wikimedia.org/wiki/Research:Newsletter/2012/January#Admins_infl... )
Hope this is useful!