Couple of recommendations/thoughts on the roadmap thought process:
1) The Features and Ideas are quickly getting out of hand to track and duplicates are bound to creep up. I would recommend moving the feature requests onto a platform such as "
http://getsatisfaction.com/". I first learned about them with the Songbird project and it's a quick/clean/efficient way to track suggestions with the ability to vote features up or down so you'll know where users want the most. This should probably be done before this gets too out of hand.
2) The idea for a bounty was proposed earlier in a different post. At another open-source project I follow, and have mentioned here, (Funambol), they have a code-sniper project where bounties can be offered for the completion of certain portions of the project. This can be a separate, side section.
3) Once priorities can be established, then a clear direction for a roadmap can be established. By not having a roadmap, users will constantly be bugging this project for their improvements to be implements, and their frustration/impatience may bring them to move off this fantastic platform, rather than give it the support it deserves. Plus, having a roadmap will also give those who want to contribute a better sense of what is currently being worked on, and where they can offer their assistance.
Of course, doing something like that would require to start opening up the code more and moving to a branch/tree-like structure where there would be more people involved. BUT, it would greatly improve/enhance performance and functionality.
Anyway, those are my two cents.
I know that I've communicated with the admins via email previously and would be happy to offer my assistance in Project Management in any way that I can.
Marc K.