Conversation
Notices
-
@mmn @pettter @takecherryakenji i think i agree with @gargron here. last time i checked, i remember thinking gs needed work on conversations
- Pi Rupe repeated this.
-
@mmn @pettter @takeFrankerZakenji @gargron e.g. i think gs doesn't store the reply-to uri if the notice replied to doesn't exist locally (yet)
-
@mmn @pettter @takekiwiakenji @gargron imo conversation_id in gnusocial is only a good idea if it speeds up db-stuff locally
-
@hannes2peer conversation_id is in place to prevent one of the most annoying things we had to deal with back in 2011 when most of the fediverse ran on statusnet: whenever someone deleted a post, the entire conversation tree would crumble and going into context would just show loose messages, or an individual post. It has a good reason to exist and should not be removed unless you want to go back 6 years in development. While it's not an issue for networks focused in ultra-casual, meaningless interaction, it's highly consequential for any node that values extensive and meaningful discussion, and for any implementation of GNUsocial in a work environment where easily following conversations is required.
-
@nerthos yes you are right, i wasn't thinking!
-
@nerthos > go back 6 years in development.
https://github.com/tootsuite/mastodonRDN's Lucifer likes this. -
@nerthos Oh, and federations didn't federate from GS to Mastodon a while back, either, so maybe that was another one of his workarounds. Not sure about the status of that now.
-
@takemangoakenji Honestly I'm a bit conflicted on whether it'd be better to keep being semi-compatible or whether Mastodon would be better off as a proper branch with it's own networking rather than being loosely compatible with the standing network.
takeshitakenji likes this. -
@nerthos With the work that @lambadalambda and @dtluna are doing, our side of the fediverse may move on to a better protocol that Gargron incidentally doesn't want to support.
-
@takegrapeakenji I'm not aware of what they're working on. IMO it's best to stick to classical OStatus as it's supported by basically every network.
-
@nerthos OStatus is a pretty bad protocol, Mastodon events notwithstanding, hence the push.
-
@takeFluffle Puffakenji Moving from it would cause a mastodon 2.0 where again older networks would not be compatible. I tend to prioritize sticking to used protocols unless they're unsustainable.
-
OStatus is a minimum viable protocol. It is up to us to work to extend it in sensible ways. Evan tried to do that, but some of its component protocols were based at Google, which lost its desire to interoperate once GPlus launched.
