* Christian Schudt <[email protected]> [2015-07-28 10:22]:
> I think this has already been discussed once, but wouldn't it be more
> intuitive to use IQ semantics for this instead of sending a message
> which confirms the (de)activation?

+1 for an IQ. Basically just what google:queue does, with a properly
declared namespace and its own XEP.

IMO, the discussion about a session forced to be CSI-active after
resumption is only a workaround due to the fact that (in the current
XEP) the CSI change nonzas can not be reliably tracked by means of
XEP-0198. I can not see a real reason for how a stream resumption (i.e.
due to a mobile/wifi change) should affect CSI state (which is related
to the user interacting with the client).

I think it would be much cleaner to bind the CSI state to the session
(thus keeping it on session resumption), to use IQs for changing it
(thus allowing XEP-0198 resumption/replay to properly take care of
possible indication losses), and possibly also to require the server
implementation to flush all "queued" data before the iq response.

Georg
-- 
|| http://op-co.de ++  GCS d--(++) s: a C+++ UL+++ !P L+++ !E W+++ N  ++
|| gpg: 0x962FD2DE ||  o? K- w---() O M V? PS+ PE-- Y++ PGP+ t+ 5 R+  ||
|| Ge0rG: euIRCnet ||  X(+++) tv+ b+(++) DI+++ D- G e++++ h- r++ y?   ||
++ IRCnet OFTC OPN ||_________________________________________________||

Attachment: signature.asc
Description: Digital signature

Reply via email to