Alan Burlison was kind enough to send me details of the membership out of auth, which gives a breakdown of roles by collective. A little massaging later, and I got:
Leaders: mode = 2 median = 3 average = 5.6 max = 166 total = 2209 Affiliates: mode = 0 median = 0 average = 7.3 max = 328 total = 1179 Developers: mode = 0 median = 0 average = 0.4 max = 15 total = 98 Participants: mode = 0 (3 is almost as high, though) median = 12 average = 25.5 max = 359 total = 10095 (For clarification, Developers and Affiliates are the same thing named differently in different collective types.) I throw these numbers out there for discussion, but my impression here is that the numbers aren't those of a well structured community. We have far too many leaders, and far too few affiliates/developers. Now, some of that is structural history: granting of electoral rights equated to leader status; we've required a minimum quorum of leadership so that smaller communities tend to be all leaders. And we haven't, historically, made much use at all of the affiliate/developer tier. (In the constitutional model, the number of contributors is very small - it's really the same issue.) Breaking the leader/voter tie should eliminate some of the extreme cases; the new powers of the affiliate/developer tier should make that a more attractive role for the majority. Our attempted revisions of the constitution are, in part, based on the idea that there is a progression through the roles. Is the current balance of [website, ignoring the electorate] roles something we should concern ourselves with? If so, should we let it even itself out on its own, or should we be looking to provide guidance and encouragement to the community to better manage the roles? -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
