Should Section "Approaches for moving followers and following" refer to the FEP with a test spec?

[bumblefudge]

We wrote this up as a testable migration proposal over on the FEPs repo - should we consider aligning the section "Approaches for moving followers and following" with it, or mentioning it as one possible way of doing things there? It feels like a big discussion, so probably best scheduled in advance and give people an email heads up to read it if they have opinions...

[lisad]

The LOLA spec already refers to FEP 7628 in section 7.2.1, which is normative.

The source server SHOULD offer UI for the user to choose to notify followers of a new location. This is done via the Move Activity type, defined in [ActivityStreams Vocabulary](https://www.w3.org/TR/activitystreams-vocabulary/#dfn-move) and described more for use to move Actors in the [Move Actor FEP](https://codeberg.org/fediverse/fep/src/branch/main/fep/7628/fep-7628.md).

The "Approaches for moving followers and following" section is part of the explanation preparing users with the overall picture to understand all the normative pieces.

With this as background, what's the goal in referring to FEP 0f2a? It actually suggests a slightly different way of notifying followers of a moved actor, via an Announce- whereas LOLA and FEP 7628 suggest using a Move.

I can schedule a discussion at any rate, flagging this personally for the next discussion.

[bumblefudge]

Hey! So, a few thing:

what's the goal in referring to FEP 0f2a?

Honestly, the intention of 0f2a was just to elaborate the behavior of 7628 more explicitly and make it more testable. The test-cases were unaffected by the Move/Announce debate, since they test the actors themselves, not the S2S messages, but you'll notice that in the in-FEP issue tracker there's still a to-do to add example Move and Announce messages 😅

It actually suggests a slightly different way of notifying followers of a moved actor, via an Announce- whereas LOLA and FEP 7628 suggest using a Move.

Short answer: yeah, no, Move makes more sense.

Long answer: I think I got a little carried away reading this SocialHub thread about Announces versus bare activities and overthought it. After stewing on it a bit, backwards compatibility is more important than consistency. Announce still makes sense for Tombstoning, but sending a Move instead of an Announce means the current Mastodon-API approach would require no changes where already implemented (i.e., add copiedTo in first push-mode Move activity from the old server, change it to movedTo upon confirmation, i.e. pull-mode Move from new server). Opened a PR here if you have the bandwidth to double-check that we're on the same page now.

[lisad]

Thanks @bumblefudge that looks good to me (I don't have a codeberg account to give it a thumbs up there but consider it reviewed if that helps). clears up the confusion!

[bumblefudge]

The confusion was entirely in my own head, I was so fixated on the end-state (what you get when you query the Actor) that I forgot the S2S messaging is what most FEPs are actually about in the first place 😅