* Goffi <[email protected]> [2021-11-10 10:39]: > It should be spelled out clearly how items are returned (even if > unordered, bug again I don't think that it's a good idea to have > unordered items by default on pubsub).
Updating the spec won't update implementations, so unless you also add a new '60 feature that servers can announce, the client still can't make assumptions regarding the ordering, no? And if solving this problem involves touching both client and server implementations anyway, can't we take the chance and move PubSub users towards RSM, in order to avoid redundant mechanisms for retrieving item sets? I.e., keep '60 as-is, except for referring to '59 + '413, and update '413 to be a plain '59 extension? > Also implementing XEP-0413 imply to implement "creation" and > "modification" order in both directions, some implementations may want > to keep it simpler. Maybe have the server announce supported order critera? I.e., something along the lines of: urn:xmpp:order-by:1@http://jabber.org/protocol/pubsub#creation urn:xmpp:order-by:1@http://jabber.org/protocol/pubsub#modification In theory, it might even be cleaner to specify order criteria in '60/'313/whatever is using '413? So '413 would end up being really generic, without specifying anything PubSub/MAM-specific? (And maybe the <order/> element could become a child of the RSM <set/> element?) Holger _______________________________________________ Standards mailing list Info: https://mail.jabber.org/mailman/listinfo/standards Unsubscribe: [email protected] _______________________________________________
