Okay, I'm trying to bring together some of my private bipartisan talking, and yesterday's Transparency Caucus, along with exploring networks of trust and reputation. Also, I've learned a lot about identity and accountability in my day job.
I hear both Republicans and Democrats talking about finding a new way to have Americans set the agenda for the forthcoming elections, with a focus on social media/networking, and they're serious.
Tools already exist for this like IdeaScale, pioneered in current Open Government efforts.
The hard part? That's preventing people from stuffing the virtual ballot box, that is, having people pretend to be many people by creating many identities to vote or comment. That's like the old "vote early, vote often" thing, but scalable into very large numbers.
We need to be reasonably confident that people are who they say they are, and they get only one vote.
Here's the deal: we have the tools for large scale agenda setting now, but if we're serious about 'em, we need to make sure that we minimize abuse. We can't fully prevent the problem, but we can reasonably control it using emerging tools. Since this isn't real voting, we can live with a reasonable solution soon, while getting a much better solution later.
There are at least two serious (though imperfect) large scale providers of identity, that do the job well enough for November.
That's Facebook and Google, via the profiles people already have. They really aren't perfect, and there are other providers, (I'd like to hear about them), but these will probably do the job. There are also standards to make profiles interoperate, which should also mean you own your own data.
So, if the online stuff will be seriously used in the elections: I propose that whatever tools are used, that they rely on identity provided by the best systems we have, and as much as possible, we own our own profiles.