With applogies to Lessig, code is law and law is code.  I'm not sure he endorses the flipped part of that, but it works for me.  The app we envision could be an invasion of privacy if carried out with the ethic of the typical social media provider, and therefore all the code and systems need to be designed for security from the start.  We want the power of the connected network, while keeping the intimacy of the first one or two degrees of separation.  The data will be a sharing commons of knowledge, but it will respect the privacy of the producer networks who will want share information that is private within a network of friends and collaborators.  Sometimes it may be a lot of gossip, but even that can be useful in bonding a community of action.  Leaders and stewards of the commons will help turn interest into compassion, and rivialry into support.

It will help to consider this from the perspective of user stories.  Imagine our twenty-first century version of a precinct worker.  He does not work for a machine and get a patronage job in return, she is a neighbor who wants to contribute to building a stronger democracy, and maybe eventually makes some money or get a job with the skills and contacts created along the way.  She probably has a smart phone, but even if not a paper and pad for notes and a computer at home are enough.  With paper, I print out the voter/residence lists, or with my app on the smart-phone I just walk and as I pass each house I see the information I need to know if it is time for a contact, and when I go up to the door I'm shown more detail while I ring the bell and wait.  I have a nice conversation with my neighbor while taking notes and checking boxes for follow up.  Maybe they need help getting to the poll, or with an absentee ballot, maybe they will help getting out the vote later, or come to a community meeting.  After a day or afternoon of working, I can see a summary of what I produced, how many new/updated voter entries, how many more active commitments, what percentage of by area is covered, how are my neighboring captains doing, and so on.

If I am gathering this kind of data for a candidate or party, I am going to be particularly interested in how they vote, and here we are discussing a system the first values participation.  This is not to say that we are not interested in that information, this is a question of what we place value on.  The generic voter conversation is about engagement in democracy, not about who you will vote for.  On the other hand, donors and workers will often want to express their support for particular candidates and issues.  Rather than propose a solution, I am pointing to questions the network of workers and the stewards of the commons will have to answer so that they can be implemented in code.  Maybe only the local participants can see certain data, or they can have it presented as aggregated data for further agregation.  This can serve to demonstrate on-the-ground value to the local networks where it can be mobilized for positive participation investment where a candidates message is most effective, but not so much for negative and anonymous campaign tactics.

More important than aggregating poling data is aggregating actions and people.  When we ask instead about what concerns people we can create opportunities for the people themselves to aggregate in neighborhood rallies about issues of local concern.  At those rallies we can do the work of P2P Democracy at a larger scale.  We would have the opportunity to make presentations and gather for workshops that can spread the skills and experience from the current core group and recruit others into action.  Coming from shared concerns and building shared language and practice will grow the network.

This leads to another natural application connected to the contact app, a shared community calendar that marks the rallies and maps the issues and people engaged in them.  When someone comes to the website and sees events in another community, they can also see offers to bring a rally or a workshop to them and help them bootstrap a new node in the network that is taylored to their own community and concerns.