Seen some conversations recently about taking a general discussion thread here onto discord/matrix for "real time chat".
It then struck me, as someone who's been on lemmy since before the Reddit API migration ... that lemmy used to be more "real time" than it is now with the front-end receiving updates over websockets.
Coupled with the "chat" sort for comments (which is buggy I think), you could turn any post into a live chat.
Obviously you wouldn't want too many of these as they burden the backend. But it could be a nice feature, using mostly old lemmy tech (?), to allow selected posts to become "live chats".
It would probably make sense to add time limits for how long this can be on for, and maybe to add limits for how many posts per community ... all configurable by admins. But also it could make mega-threads and free-form discussions much more dynamic and attractive here.
EDIT:
There could be both user-specific and post-specific modes for this too.
Any particular user could be able to turn on chat mode for them, so that comments are flattened and updates happen automatically, but just for them. Limiting this in someway on a user based would make sense.
Then a particular post could be put into "chat mode", such that everybody who opens the post does so in "chat mode" automatically, unless they opt out. Again, limitations on how many posts and for long they stay in "chat mode" make sense here.
Any chance the relevant incident could be unpacked and used as a demonstration of how these changes would alter the outcome or encourage a different outcome?
As someone who only saw pieces of it after the fact, I am potentially in the dark here about the purposes and context of these changes.
That being said, from what I did see, it seemed very much like an instance admin imposing themselves and their superior power on a community when there were probably plenty of other more subtle action that could have been taken, where subtlety becomes vital for any issue complex and nuanced enough to be handled remotely well. I'm not sure I'm seeing any awareness of this in this post and the links provided.
For instance, AFAICT, the "incident" involved a discussion of if or how a domestic cat could eat a vegan diet. Obviously that's not trivial as they, like humans, have some necessary nutrients, and AFAICT the vegans involved were talking about how it could be done, while the admin involved was basically having none of that and removed content on the basis that it would lead to a cat dying.
And then in the misinformation link we have:
In the context of cats and their food ... which "living beings" are being harmed and who is encouraging or discouraging this harm?
Whether you're vegan or not, this seems to me formally ambiguous and on the face of it only enshrines the source of the conflict rather than facilitating better forms of communication or resolution (perhaps there are things in the by-laws I've missed??).
Two groups can have exactly the same aim and core values (reduce harm to living beings) but in the complexity of the issue come to issue a bunch of friendly fire ... that's how complex issues work.
So, back to my original question ... how exactly would things be done better?