Hi, On Sat, Apr 26, 2025, at 12:57, Goffi wrote: > I don't see that it's needed to check for other replies. If two people start > a > reply at the same time, there will be two threads with the same parent, in > the > UI this appears like a tree, i stays readable. This is what happens on sites > like Reddit on ActivityPub (but on Mastodon, the tree doesn't appear in > comments, making it very confusing IMO).
Never used those services, but the normal flow i imagine is, im in a support channel, someone asks a question "How do i do X?", now 2 people start to type, and create two separate threads with the answer, now a 4. person wants to join, then the client needs to decide to reply to one of the threads. Also from the GUI side its unclear how should handle this, should i hide the fact that there are 2 threads? Should i show 2 threads? But then obviously both users wanted to answer the same person, i would say it would be the expectation that this conversation is wrapped in a single thread and not multiple. But maybe i envision this worse than it is in practice. > > This comes at the cost of losing the feature to signal that a message > > should > > not be used for a thread. > > How do you signal that a message should not be used for a thread, and why > would you do that? > > From a UX point of view, I think that it would be bad that some messages can > be replied in a thread, and some other can't, the end-user would not > understand why sometime the button appears, and sometimes not. I just wanted to note what the implications are of the design decision. I could imagine someone wanting in their channel that no thread is created under a bot message. I dont see this as a show stopper, just want to prevent that in a year we need to add <no-thread> for some reason, when we could have it now naturally by not including a thread. As for the moderation XEP, i definitely think the moderation and als retraction XEP should exclude the <thread> element from being removed, from the message. Regards Philipp
_______________________________________________ Standards mailing list -- [email protected] To unsubscribe send an email to [email protected]
