Fascinating, thanks for the pointers.

I'll have to look at them further, but what I think they might lack from what I posted is a notion that different users will value some gray area arguments differently. The system should preserve those differing value systems in the arguments constructed and use them to build upon the webs of trust. For example, user bob may find it axiomatic that "healthcare is a right". User jeff will typically downvote arguments that assume that right and not want to build his framework on top of it.