Especially their statement that they could not reconcile moderation with free speech,

The way I’d reconcile them would be when everyone can override moderation for what they themselves want to see, but without that the common default is applied.

That would be what we’d see in Fediverse projects if people were acting in good faith, too.

I dunno, somehow the best approaches I can imagine are those that existed in Usenet before it went out of use for discussions. Except for news servers having to store too much, and for spam protection happening after it gets posted, bad results. So probably things like group membership and post limits and such from today would be useful.

But in general Usenet was the way. I won’t change my mind, because a few different systems converged on models similar to Usenet, that being itself, Fidonet, Frost and FMS in Freenet, boards in Retroshare, even frankly places like Reddit and Lemmy.

We need a Usenet 2.0, with some precautions from it turning into a place for bots and pirates, like the old one. IMHO. It can even use Fediverse identities (but preferably not, identities should be cryptographic and untied from instances ; or maybe an instance would only be needed when an identity is created and posted into the network, but then it can be banned\removed on that instance all they want, it’ll be fully usable).