or maybe more general #Fediverse crowd: Given some communication over here recently, I wonder whether it's possible to compose posts that are visible to followers only but for which each of my follower is able to see each response and able to interact with each other person responding there no matter whether these people follow each other too?
@z428 look, if you want all the contacts just to be able to answer each other or see each others answers, you have to tell every server and other user who is allowed and should see the answer of every other participant.
So you have to technically group and share this "virtual group with all participants" to everybody else in the group. Otherwise nobody knows who to send replies, except to you.
Even this might be possible in other protocols, you have to trust every other involved node, that it respects your intention to privacy and keep the closed list that you set initially.
@hoergen Hmmm, maybe I'm all off here, but ... this somehow feels like a problem that has been solved in the past, and be that as simple as in e-mail: Sending messages to a wide load of recipients with all of them in Cc (because I want to discuss issues with all of them), anyone who responds to any of these messages will send this response to all of the original recipients, and that is intentionally and expected to be this way. If someone modifies the set of recipients - fine, of course everyone's able and allowed to do that, that's a conscious decision, but it's not, like "no matter what - your response will only be seen by those people in that recipient list that have you in their address book"... . Maybe this analogy is a bit difficult, but at least that's a kind of behaviour that, for an addressing as generic as "Following", would somehow be not all too much off.
Serious question, no offense or provocation intended: With this stuff being baked into Hubzilla and, apparently, also design-wise into Bluesky / AT, can anyone out here involved with the #ActivityPub specification process outline why nomadic / easily portable identity isn't built-in here by design? Looking at the (to-be-expected) dynamics of instances going up and down, blocking each other or moving to newer, different pieces of software, this seems an absolutely obvious requirement, so I wonder why this has been left out of the standard / spec?