On Mon, Nov 3, 2008 at 8:00 PM, Gregory Maxwell gmaxwell@gmail.com wrote:
I've not seen any concrete evidence that supports this. Traffic in meta areas doesn't always translate into real productive work.
In one of the emails I explained it through data from a couple of projects; without en.wp, of course, because statistics are too old. However, I don't think that behavior is significantly different than in French and Italian cases (according to the list activity).
There are no statistics for Commons [for which I know], except for the list, where communication decreases, too.
Besides that, the only case which doesn't support the relation between mailing list activity and project activity is a negative one. Japanese Wikipedians talk more at the list, but their activity at the project was in decrease for the period January-May 2008.
I can't speak for project that I don't frequently use, but EnWP and Commons haven't started imploding as far as I can tell.
The main reason for raising this question is exactly to prevent imploding. (BTW, we already have one implosion: The third Wikiquote, Polish, has less than 50 edits per day, the fourth, German, has somewhat more than 50 edits per day.)
Also, I would be happy if someone [like you] checks more systematically statistics. I have very rudimentary knowledge in statistics (yes, I know that curves of development are not not straight forward and that a couple of paths are possible), so someone with better knowledge in that field should check it. (And I think that the question is very important. I would be very happy to see that I am wrong.)
[1] - http://lists.wikimedia.org/pipermail/foundation-l/2008-October/046831.html