Cool, @anth@a.9srv.net, thanks for the followup! I have to reread the original v2 in order to really follow your explanation, but that document seems to be offline at the moment. I’ll try again later. :-)

Righto @anth@a.9srv.net, v2 is up again for me:

Clients (and human readers) just assume a flat threading
structure by default, read things in order […]

I might misunderstand this, but I slightly disagree. Personally, I like to look at the tree structure and my client also does present me the conversation tree as an actual tree, not a flat list. Yes, this gets messy when there are a lot of branches and long messages, but I managed to live with that. Doesn’t happen very often. Anyway, just a personal preference. Nothing to really worry.

The v2 spec requires each reply to re-calculate the hash
of the specific entry I’m replying to […]

Hmmmm, where do you read that the client has to re-calculate the hash on reply? (Sorry, I’m probably just not getting your point here in the entire paragraph.)

Clients should not be expected to track conversations back
across forking points […]

I agree. It totally depends on the client.