[Jingle] Typo and possible clarification in XEP 0166

Justin Uberti juberti at google.com
Tue Apr 21 18:03:19 CDT 2009


"Roughly equivalent" is interesting terminology for a standards document,
but overall it looks good.
Also I'm not sure this is a MUST, perhaps there are applications that will
want to preserve the double-busy that you would get in the telephone world,
so maybe SHOULD is more appropriate.

On Tue, Apr 21, 2009 at 9:49 AM, Peter Saint-Andre <stpeter at stpeter.im>wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 4/21/09 10:38 AM, Peter Saint-Andre wrote:
> > On 4/20/09 5:10 PM, Justin Uberti wrote:
> >
> >> On Mon, Apr 20, 2009 at 3:05 PM, Robert McQueen
> >> <robert.mcqueen at collabora.co.uk <mailto:robert.mcqueen at collabora.co.uk
> >>
> >> wrote:
> >
> >>     Peter Saint-Andre wrote:
> >>     > No existing session
> >>     >
> >>     >     If there is no existing session and both parties
> >>     simultaneously send
> >>     > a Jingle session-initiate message, the action with the lower of
> >>     the two
> >>     > session IDs MUST overrule the other action, where by "lower" is
> meant
> >>     > the session ID that is sorted first using "i;octet" collation as
> >>     > specified in Section 9.3 of RFC 4790 [27].
> >
> >>     Presumably Justin's issue is just about people audio calling each
> other
> >>     at the same time. So, this should probably contain wording about the
> >>     same content type(s). If I send you a file when you're calling me,
> >>     neither should fail. I almost wonder if this shouldn't be in the RTP
> >>     description XEP, given I can't think why the semantics are otherwise
> >>     desirable, and I do also wonder whether these semantics are specific
> to
> >>     RTP calling clients and should be carefully worded to avoid making
> >>     certain implementations (which do trunking or other call-switching
> >>     activities?) XEP-incompliant by design.
> >
> >>     Regards,
> >>     Rob
> >
> >
> >> I mostly agree. I think the key thing here is that the conflict only
> >> occurs when the sessions are logically similar. This is the case for
> >> simultaneous audio or video calls, but could also be the case for other
> >> things like an invitation to play a game, establish a direct connection,
> >> etc. So I don't totally buy that this should go into the RTP XEP.
> >
> > You're both right. The text about conflicts on session-initiate needs to
> > make clear that this is about "similar" sessions (where by "similar" we
> > mean that they have roughly the same contents). Let me see if I can come
> > up with some good wording about that.
>
> How about this?
>
>   If there is no existing session and both parties simultaneously send
>    a Jingle session-initiate message with a content-type that is roughly
>   equivalent (e.g., each message requests initiation of a voice call),
>    the action with the lower of the two session IDs MUST overrule the
>    other action...
>
> Peter
>
> - --
> Peter Saint-Andre
> https://stpeter.im/
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.8 (Darwin)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>
> iEYEARECAAYFAknt+Q8ACgkQNL8k5A2w/vxAtQCg2wkXchhiZgDGk/Hht0sLtSIP
> BzoAoN9uPVgDDDPItKS33zb6Coq/XF4Z
> =QDsK
> -----END PGP SIGNATURE-----
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.jabber.org/pipermail/jingle/attachments/20090421/12e0430b/attachment.htm>


More information about the Jingle mailing list