-
Notifications
You must be signed in to change notification settings - Fork 43
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Use DM to disable instead of blocking? #1335
Comments
If you are doing this, it may be a good idea to also have a (preferred) command that only deactivates the account on the Bluesky side. IIrc the PDS doesn't need to serve data for inactive accounts and the AppView won't show any (but will hold onto its internal cache). |
Hmm! I don't quite follow "only deactivates the account on the Bluesky side." As opposed to fully deleting? If so then yes, that's what we do now, in hopes that Bluesky team will eventually make the AppView able to undelete repos, #1130 / bluesky-social/atproto#2806 . |
I think currently you send a tombstone or something with a status set to deleted, or similar.. The gist is that (in my eyes) it'd be good if Bridgy Fed had an option to temporarily deactivate the bridged account without deleting any data – like the 'Deactivate my account' feature in the Bluesky settings that's distinct from deletion: This is already undoable and would be useful to turn off bridging when full distributed data deletion isn't a strong concern in the moment. The UX difference in this case is that Bluesky would say the account is deactivated, instead of calling it a "Deleted Account" or presenting a not-found page. Maybe this is also the action that blocking should take, if any, since it's easy to undo. |
On that note (tangentially related), (how) do you translate soft account deactivations and reactivations from ATProto into ActivityPub? I don't think an account- |
Ahhh! Yes that's what Bridgy Fed currently does. We're just waiting on Bluesky team to support re-activating accounts in their AppView.
We just send a new |
Delivery to the server may pick back up, but sending a Notably, this completely severs follow-relationships: The posts may still go to the federated feed, but the previous followers won't see them in their home timeline. This is very much not equivalent to Bluesky's 'deactivation', which is entirely non-destructive. |
Maybe on the main PDS, but not yet for federated PDSes, bluesky-social/atproto#2806 .
True! Apologies, I didn't mean that it had perfect parity with how Bluesky will hopefully be. |
Oh, I see. Sorry, I wasn't aware of that specifically (but should have inferred it).
I'm just wondering if there's a way to get closer to the intent (that doesn't have you send out a ton of But probably not and failing with lower-than-expected connectivity is indeed the 'safe' option in some ways. |
This seems lower priority now that we've switched tombstoning in Bluesky to deactivating instead, #1130 (comment), which can be undone. |
See #1440 (reply in thread): Mastodon doesn't track delivery of |
I think this issue doesn't actually fit the remaining open questions here. I'll file a new one based on #1335 (comment). (The other parts are either implemented, superseded or tracked elsewhere like in #1293.) |
Blocking is an awkward, surprising mechanism for disabling the bridge. Blocking can be undone in both the fediverse and Bluesky, but right now disabling the bridge for fediverse => Bluesky can't be undone, #1130. I get a steady trickle of users who do this, and don't realize it disables the bridge, and they're confused why it's no longer working for them.
One option would be to switch to using DMs instead. You can already DM
no
to the bridge user to disable the bridge. I don't advertise this, since multiple ways to do the same thing is generally confusing UX, but we could switch to this.Alternatively, we could just push to get Bluesky team to make repos undeletable on their end, #1330 , and then this wouldn't be a problem. That's arguably the better long term fix, but I'm not holding my breath on ETA on their end.
The text was updated successfully, but these errors were encountered: