Re solutions - I'm inclined to agree (with Tomma and Gary) that a combination of ready-made open-source solutions might be the way to go if the project is to go ahead in the timeframe and at a low-cost.
I've been involved in building a community network from scratch (http://ki-work.com) and it's a full-time job for a few people. Far better to start small to test out uptake, then build up from there.
One open-source solution may come from http://elgg.org - I'm in the process of reviewing their platform for a separate project so will feedback how customizable it is later.
A really quick and easy start for a working prototype can be done for free with Ning (http://ning.com). This way we could get real feedback from a cross-section of TT users. As was mentioned on the requirements forum, usability will be a key issue as members will be all ages and levels of computer literacy. It's easy to get carried away with solutions that are pioneering and world-changing, forgetting who is actually going to use it.
Jim