How might we make it so it doesn't matter if user behavior is inconsistent?
As people say I don’t want my tool to tell me how to think, they each seem to have their own idiosyncratic style, and that style may change over time as apps with continued user involvement are responsive to increasing skill levels over time. This means that a tool for thought designed for scaling synthesis would need to be able to account for inconsistencies in structure across a users personal style over time as well as between users.