Typing indicators are perhaps the more obvious of the two affordances I'm interested in. I think the primary purpose is to support turn-taking in an environment where we don't have the same sort of cues to understand when someone's done talking or has finished their point.
Of course some of this is linguistic, and there are plenty of models that develop a kind of ruleset to describe how turn-taking works in conversation.
I think it's worth thinking about how to build a robust turn-taking mechanism to support productive discussions. Robust could mean typing indicators, or it could mean something a bit more structured, or even looser. I'm not sure yet. Regardless, once designed this is something that I'd like to do some more formal testing around.