[Standards-JIG] Re: Jingle xeps

Matthew O'Gorman mogorman at astjab.org
Thu Oct 12 22:52:11 UTC 2006


it seems to me joe that you want to allow a fake redirect that the
server could control.  I don't think that is a good idea personally.
Anyways it isnt that much extra work for the client developers and
myself to add a responder field to the answer.  Just my opinion.

Mog

On 10/12/06, Joe Beda <jbeda at google.com> wrote:
> The idea here is provide for a scenario where there is intelligent call
> routing on the server.  Obviously not all servers will support this.
>
> (Again I don't think that this is necessary for the base jingle xep...)
>
> I would imagine that the client would do a disco to the server to discover
> if jingle routing is supported.  In that case, the client would call the
> bare jid instead of the full resource and the server would handle this in an
> intelligent way.  This is analogous to sending a message to the bare JID --
> it is up to the server to figure out the right resource (or resources) to
> send this to.  It is necessarily more complicated than the simple fire and
> forget message case since any broadcast results in parallel negotiations
> (dialogs).
>
> Joe
>
>
> On 10/12/06, Alexander Gnauck <gnauck at ag-software.de> wrote:
> > Joe Beda wrote:
> > > For all of the steps where Bob (bare) is involved, IQs go between
> > > Alice/home and Bob (bare) and between Bob (bare) and Bob/<resource>.  If
> > > Alice/home wants to keep state for each dialog that dialog needs to be
> > > identified.  The extra identification is the responder.
> >
> > this is an edge case. Normally we send IQ's to full Jids in XMPP. Who
> > handles the IQ to Bobs Bare jid? I think all current servers will reply
> > with an error message to such an IQ. Or did i get smth wrong here?
> >
> > Alex
> >
> >
>
>



More information about the Standards mailing list