I need help: a website strategy and design issue

I want to build a political website whose role will be to act as a communications nexus around which the organization can evolve - but, as it turns out, I don't know how to do this, so I'm asking for help.

As regular readers know I'm interested in political stuff as well as technology - and now I'd like to put those two together to develop an Alberta political site.

As it turns out, however, I don't know how to do this - more precisely: I believe that the web can be used to build the kind of long term member support this would need, but don't know how to actually make use of that capability.

My initial thought was that I needed simply to create a site supporting concurrent news and opinion programming with lots of discussions tied to specific themes like "medicare", "policing", or "the environment" and just let things build from there. Geeklog's software seems perfect for that, is free, installs easily on Solaris, and scales well enough to support noted geeklog users like groklaw.

Unfortunately, there are two distinct kinds of problems. First, the external ones: most notably the fact that the people I most need to appeal to and involve don't typically hang out on the web - and those who do use the web often tend to be the least conservative and most opportunistic among them. Given highly visible leadership and the consequent publicity, we can expect many of these people to check out the site and any processes embedded in it; but only once, and then tentatively. Worse, they're predictably going to feel uncomfortable about their own abilities to work with computers - meaning that it has to look very simple, very non threatening, and somehow welcoming.

Internally, the two big problems are tightly inter-related: long term audience retention and the step from communications to actions.

You'd think this wouldn't be much of a problem, but most conservative sites across the United States have failed to extend their audience beyond core groups and (at least to my knowledge) only Gingrich's American Solutions site has come close to providing a clear route to bridging the gap between talking and doing.

I've had access to some interesting, but sketchy, demographic and behavioral data on this - and most of it leads me to a disquieting conclusion for conservatives trying to use the web for political purposes. I think the data points at a sharp behavioral demarcation between the two camps: specifically that seeing the donation of money to the cause as personal action toward meeting social responsibilities is consistent with underlying leftist beliefs about society and the donor's position as a leader/driver in it, but inconsistent with conservative beliefs about social coercion and the importance of personal responsibility for actions.

Piling it high and deep right? I know; but there's an immediate consequence: a request for donor cash builds audience loyalty at a leftist sight like the daily kos but destroys it at a right wing site like HumanEvents.com - and that, of course, has obvious, direct, consequences for what my site should try to do, how it should go about doing that, and where the money for doing it is coming from.

The related problem involves audience retention when specific discussions reach consensus. What typically happens in the more heated debates is that the perceived "losers" turn off, the winners get bored, and the whole effort becomes pointless until someone else heats up the war of words again and the cycle restarts - and the smarter and more involved your audience, the worse this problem gets. There's just an inherent contradiction here: the site's goal is to achieve resolution through consensus building and action - but resolutions, when achieved, put the site out of business.

So, at this point, what I'm thinking is: copy some ideas from linkedIn to give people personal and financial incentives for joining, re-invent some structural and audience involvement ideas from groklaw, and copy the solutions focus and local action ideas from Gingrich.

Better, I think than my first ideas - but maybe you guys can help? Assume high visibility leadership - and then tell me how you think this can be done. Please?