Continuing the discussion from Reputation Systems:
So as we were discussing reputation systems, I threw out an idea for a privacy management system. Basically this system would classify your personal data, contact info, true social, work/business, financial, political, medical, or any other kind of category.
The default on the safe network is true privacy, with no on seeing your data. But in practice you want people to be able to access your data in a private way, family, friends, the professionals who do specific things for you (doctors, brokers, lawyers, etc). The trouble is that in the conventional system we often have no idea what types of data people are gathering on us, what the specific content of that data is, and we certainly don’t have any control over who has access to the data.
Now, I suppose that someone on the Safe network could keep track of every document, every piece of information that they shared with someone else, and then if the relationship terminated or changed, go through and change the sharing settings on ever specific document. But realistically, I think that we need to give people a way to manage classes of information.
This could be supplemented with a client-side watcher, perhaps based on the deduplication protocols, which scans all outgoing data and says Hey! This document which you are sending to so-and-so looks like it contains X personal data, which goes beyond the level of trust/sharing which you have authorized this user to have. Are you sure you want to send this?
This system would track who has access to what at any given time. This would be enormously helpful in giving teeth to lawsuits for invasion of privacy, because its so difficult to prove who had access to what at specific times.
Any other ideas for such a system? Can anyone tell me to what extent SAFE already does, or intends to do this?