>what happens if an application includes in a multiple packaged INFO >request a body part the Sever does not understand? >how is the server able to indicate that only a part of the INFO request >was not understood ?
THAT is what you ARE supposed to use the C-D header for. If handling=required, and the receiver doesn't understand it, rejected the message shall be... Regards, Christer Jeroen van Bemmel wrote: > A possible reason to want multiple packages per INFO request, is that > it allows one to construct applications that rely on "all or nothing" > semantics for correlated events. > > For example, if one would use SIP to implement an application that > lets one post a small icon on a blog, and requires an authorization > token to do that, both would have to arrive together. Another example > could be an electronic variant of a postage stamp (i.e. some token > representing some monetary value), when using two different INFOs one > might fail to arrive where the other would (i.e. content without > payment, or payment without content). > > It has to do with certain atomic, transactional behavior and > consistency of the user experience > > Regards, > Jeroen > > Paul Kyzivat wrote: >> I also agree it is too extreme to restrict to one package per dialog. >> But as I stated earlier, I'm fine not defining multiple packages per >> INFO. >> >> Paul >> >> Christer Holmberg wrote: >>> Hi, >>>> I don't have a problem agreeing with that. >>>> >>>> Note that buried somewhere in this thread was a question of whether we >>> had a use case for multiple packages per dialog, or can we simplify >>> even >>> further. >>> >>> I don't think we should go that far, because that could become very >>> restrictive. >>> >>> For example, assume I want to use INFO packages e.g. for DTMF during >>> the >>> call setup, and then other INFO package(s) for something else during >>> the >>> call. >>> >>> Regards, >>> >>> Christer >>> >>> >>>> -----Original Message----- >>>> From: Christer Holmberg [mailto:[EMAIL PROTECTED] >>>> Sent: Thursday, October 23, 2008 10:51 AM >>>> To: Elwell, John; Dean Willis; DRAGE, Keith (Keith) >>>> Cc: SIP IETF; Eric Burger; Paul Kyzivat >>>> Subject: RE: [Sip] draft-ietf-sip-info-events-00: multiple packages >>>> per INFO >>>> >>>> >>>> Hi, >>>> >>>> I agree with John. Let's keep it simple. If allowing multiple packages >>> >>>> in a single INFO causes issues, let's forget about it. >>>> >>>> The whole idea with this is to allow people using INFO to do so in >>>> an easy and standardized way, so let's not shoot ourselves in the >>>> foot with complexity. >>>> >>>> Regards, >>>> >>>> Christer >>>> >>>> -----Original Message----- >>>> From: Elwell, John [mailto:[EMAIL PROTECTED] >>>> Sent: 23. lokakuuta 2008 12:30 >>>> To: Christer Holmberg; Dean Willis; DRAGE, Keith (Keith) >>>> Cc: SIP IETF; Eric Burger; Paul Kyzivat >>>> Subject: RE: [Sip] draft-ietf-sip-info-events-00: multiple packages >>>> per INFO >>>> >>>> In reply to this whole thread, please bear in mind that we had lots of >>> >>>> discussion about whether it would be worthwhile defining this new INFO >>> >>>> mechanism, since existing applications are unlikely to change and >>>> the best we can hope for is that new applications will exploit the >>>> new mechanism. Therefore we want to keep the mechanism as simple as >>>> possible. The complexities of matching body parts to header fields, >>>> dealing with cases where only some of the packages are understood, >>>> etc. >>>> are hardly likely to persuade people to implement the mechanism. >>>> Please keep it simple. >>>> >>>> John >>>> >>>> >>>>> -----Original Message----- >>>>> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On >>>> Behalf Of >>>>> Christer Holmberg >>>>> Sent: 23 October 2008 08:17 >>>>> To: Dean Willis; DRAGE, Keith (Keith) >>>>> Cc: SIP IETF; Eric Burger; Paul Kyzivat >>>>> Subject: Re: [Sip] draft-ietf-sip-info-events-00: multiple >>>>> packages per INFO >>>>> >>>>> >>>>> Hi, >>>>> >>>>>>> Why does putting two different packages in the same INFO >>>>> work better >>>>>>> than two different INFO messages each with their own >>>>> package usage? Is >>>>> >>>>>>> there a desirable relationship that can be implemented >>>>> between the two >>>>> >>>>>>> that we would otherwise lose? >>>>>> We have one package per NOTIFY. Let's stick with one package >>>>> per INFO, >>>>> unless we want to go back to using mime-types as the only >>>>> distinguisher of packages. >>>>> >>>>> I raised that issue in another e-mail. >>>>> >>>>> But, never the less, I have no strong feelings on the single >>>>> versus multiple package issue. >>>>> >>>>> Regards, >>>>> >>>>> Christer >>>>> >>>>> >>>>> _______________________________________________ >>>>> Sip mailing list https://www.ietf.org/mailman/listinfo/sip >>>>> This list is for NEW development of the core SIP Protocol Use >>>>> [EMAIL PROTECTED] for questions on current sip Use >>>>> [EMAIL PROTECTED] for new developments on the application of sip >>>>> >>> >> _______________________________________________ >> Sip mailing list https://www.ietf.org/mailman/listinfo/sip >> This list is for NEW development of the core SIP Protocol >> Use [EMAIL PROTECTED] for questions on current sip >> Use [EMAIL PROTECTED] for new developments on the application of sip >> > _______________________________________________ > Sip mailing list https://www.ietf.org/mailman/listinfo/sip > This list is for NEW development of the core SIP Protocol > Use [EMAIL PROTECTED] for questions on current sip > Use [EMAIL PROTECTED] for new developments on the application of sip > _______________________________________________ Sip mailing list https://www.ietf.org/mailman/listinfo/sip This list is for NEW development of the core SIP Protocol Use [EMAIL PROTECTED] for questions on current sip Use [EMAIL PROTECTED] for new developments on the application of sip
