On Jul 21, 2011, at 08:50, Kevin Smith wrote: > "A <message/> with a <gone > xmlns='http://jabber.org/protocol/chatstates'/> chat state. In this > case, the client MUST consider the conversation ended. Any further > correspondence MUST be sent with a new threadID and to the bare JID." > > I'm not convinced that this is the right place for threadID stuffs. We > have a thread practices XEP for that. (By 'not convinced', I mean > 'convinced otherwise than'.
That's a decent point, and I can remove mention of threadID from this spec. > > I'm also not convinced by the <gone/> handling. A conversation ending > doesn't mean that this is no longer the most appropriate resource to > find the contact at - which is the point of the locking. > <delicious_hyperbole>When I'm having a conversation with my wife and > we jump from one topic to another it's very rare for me to run out of > the room and search the house for her before deciding she's still in > exactly the same place and continuing the conversation > there</delicious_hyperbole>. > So, then my question is, how many dopplegangers does your wife have? (-: The only reason we really worry about locking is if there's multiple resources in play. For your analogy, there would have to be more than one of your wife at home, and their consciousness can fluidly shift between each instance, you actually would have to go hunt her down to make sure you're talking to the right one (-: - m&m
smime.p7s
Description: S/MIME cryptographic signature
