I see some of the issues with long term retention of editors being that we have not figured out a method to effectively deal with sockpuppets. We ban people but they just simply come back under a new user name and continue more or less where the left off. Would it be possible to flag new user names from well known sock puppet IPs for further investigation? In other words dealing with the problem up front rather than waiting for another editor to notice that a sockpuppet has returned (which might take a year or two and lots of disruption to the community).

Another issue I see is the blocking of long term content contributors in good standing without warning. If someone losses their cool (in other words insults another editor) and the first action is to block them this is bad for the long term stability of the community. If someone has donated thousands of hours of time they deserve a warning and at least a chance to change their behavior before the "ban hammer" is dropped. I have proposed this sort of change here http://en.wikipedia.org/wiki/Wikipedia_talk:Blocking_policy#RfC:_Warning_bef...

More oligarchy is not the solution. We have some oligarchy (arbcom) and it is of questionable effectiveness ( often unable to pick up sock puppets and therefore unable to enforce bans, does not take content issues into consideration and thus IMO comes to often to questionable conclusions ). What we need to do is reach out to those who write content to see what they need. While some of them are involved in these sorts of discussion may of them are too busy, will writing content.