Groupchats #174
Labels
No Label
android
backend/service
bug
contribution welcome
documentation
duplicate
enhancement
good first issue
help wanted
i18n
Improvement
invalid
ios
moxdns
moxplatform
moxxmpp
moxxy_native
needs investigation
needs testing
question
release
ui
upstream
No Milestone
No project
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: moxxy/moxxy#174
Loading…
Reference in New Issue
There is no content yet.
Delete Branch "%!s(<nil>)"
Deleting a branch is permanent. Although the deleted branch may exist for a short time before cleaning up, in most cases it CANNOT be undone. Continue?
It's still unclear as to whether we should bet on MIX or just go with MUC. If we go with MUC, however, the issue would be how do we deal with clients that cannot be constantly online?
My opinion is that you should just go for MUC. Widespread implementation and deployment of MIX on the public network is not likely to happen any time soon and it still has unresolved issues. If you're working on a fresh closed system that maps well to MIX, it's a fine choice though.
I'm not sure I understand exactly what you mean by "how do we deal with clients that cannot be constantly online". There are a few aspects to this:
If you have any more specific questions, I can provide advice or pointers :)
By "cannot be constantly online" I mean the following example:
As far as I know, vanilla MUC would not send the message to the client's JID and thus, even with enabled push notifications, not trigger a notification. Reading the linked issue,
muc_offline_delivery
is what takes the worry off my mind but as that is not standard behaviour (or a standard XEP), it feels hacky.The thing with the list of affiliated users instead of the currently online users also seems very interesting. That's good to know.
Would existing MUCs be converted to MIX and function correctly in Moxxy? If so then wouldn't that be a non-issue?
There are a couple of solutions for this (e.g. XEP-0452), but you're right that we've not yet settled on one as a community. Both MUC and "Push 2.0" are on the agenda for the XMPP summit in a few weeks, so I'm hoping the community can come to some agreement on which solution to proceed with then.
MUCs won't just upgrade overnight, as that would lock out all the MUC clients (which is basically... all of them).
One approach would be a MIX<->MUC bridge on the server. I'm only aware of one, in Tigase, and I think it only works with MUCs hosted on the same server. I tried to implement one in Prosody, but there are various places where MUC and MIX are very different and hard to bridge transparently.
My opinion is that MUC already works, and we have been improving it continuously for a long time. There is no reason we can't continue to do so - we don't need to break everything and spend years transitioning to an entirely different group protocol with its own set of problems.
However, XEP-0452 does not provide the kind of "notification experience" users, expect, i.e. a notification for every message. That is why I feel like most of the "solutions" are more like workarounds until they are standardized and deployed in the wild.
I am curious to see what will come out of it.
Relevant first PR: moxxy/moxxmpp#46
Moxxy PR: moxxy/moxxy#300