On Mon, 2011-09-12 at 15:18 +0100, Dave Cridland wrote: > On Thu Sep 8 12:35:58 2011, Ralph Meijer wrote: > > XEP-0060 was written with XMPP Core as the basis, not requiring or > > assuming XMPP IM. This means that it is not required for a > > Publish-Subscribe service to require knowledge about how bare JIDs > > and > > full JIDs interact in certain environments. Client JIDs are just one > > subset of possible JIDs, others include MUC rooms, servers, other > > components. > > Yes, all true, but the various forms of jids are heirarchical to some > degree, even without IM. > > That is, any jid with a resource part is subordinate to the same jid > with the resource part removed. > > It would be reasonable to have this semantic used by XEP-0060.
I didn't mean to imply that we should not have to consider the connection between full JIDs and their bare JIDs. E.g. node affiliations are always based on the bare JID, so this semantic is already assumed and encouraged. My point was rather that the pubsub service should not be required to have knowledge about the delivery mechanics when it would process subscriptions and send notifications to full and/or bare JIDs. Rather it would just store and process the subscriptions as requested. It should not do some fancy unsubscription if a subscription for a bare JID comes in when a subscription for an associated full JID was already in place. Or at least not for generic implementations. -- ralphm
