Nice observations. In my experience, the most important aspect is whether people care about other people's take. At a minimum, care enough to maintain a respe

Nice observations. In my experience, the most important aspect is whether people care about other people's take. At a minimum, care enough to maintain a respected persona, but mainly, to be interested in hearing and adopting or shaping others' approaches. It's the collective production of a refined collective understanding. If the motivation instead is malice or one-upsmanship or networking or shaping product reputations, the dynamic is less interesting. As a corollary, I wonder if the governance aspect - of controlling club entry, exit, and discourse standards - has to be animated by the same incentives. My main application for AI would be analyzing/evaluating discourse to automate or facilitate governance. If it could work regardless of incentives, then there's be less need for exclusivity, and discourse would scale much better.