Re: INFO Framework: Tags

"DRAGE, Keith \(Keith\)" <[email protected]>
Newsgroups gmane.ietf.sip
Message-ID <[email protected]>
You seem to be arguing that packages themselves may need to have option
tags, rather than the extension itself, which does not answer the
question I asked. 

What I asked was whether there was a need to have an option tag
specifically for the info-package extension.

There were no option tags for pre package info behaviour, so there is no
information to tell you that if I don't support the new extension that I
supported the previous incarnation of INFO. So I do not see how that can
be used for fallback.

regards

Keith

> -----Original Message-----
> From: Dean Willis [mailto:[email protected]] 
> Sent: Thursday, November 20, 2008 6:24 PM
> To: DRAGE, Keith (Keith)
> Cc: Christer Holmberg; Elwell, John; Eric Burger; SIP List
> Subject: Re: [Sip] INFO Framework: Tags
> 
> 
> The option tag is used in OPTIONS responses (and in Supported 
> header fields in other responses) to inform a correspondent 
> that the node in question supports this extension, 
> potentially leading to a decision as to which behavior to 
> apply to future requests.
> 
> It is also used in Require header fields of requests to cause 
> the request to be rejected by any UAS that does not 
> understand the extension.  This is potentially critical for 
> applications that find themselves completely dependent on the 
> extension.
> 
> One might ask whether this is critical; after all the INVITE 
> exchange will tell us what info packages (if any) the far end 
> is willing to use. What is the difference between not using 
> INFO packages and not supporting them?
> 
> The answer is in the fallback to pre-package INFO behavior. A 
> node supporting INFO packages will, IIRC (or at least "if I 
> get my way") , always use packages when sending INFO. A node 
> not supporting INFO packages is likely to send old-style 
> un-negotiated INFO, meaning you might need to be prepared 
> with that behavior. Being informed (by
> Supported) which state applies might be useful.
> 
> --
> Dean
> 
> 
> 
> On Nov 20, 2008, at 10:44 AM, DRAGE, Keith (Keith) wrote:
> 
> >> Option-tag:
> >> -----------
> >>
> >> I strongly think we should define an option tag for info packages.
> >
> > I think you need to provide more than this in terms of 
> explanation.  
> > What
> > interoperability issues do you think is solved by providing 
> an option 
> > tag for the extension itself, as opposed to individual packages?
> >
> > Give us some valid scenarios.
> >
> > regards
> >
> > Keith
> >
> >> -----Original Message-----
> >> From: Christer Holmberg [mailto:[email protected]]
> >> Sent: Thursday, November 20, 2008 1:12 PM
> >> To: DRAGE, Keith (Keith); Elwell, John; Eric Burger; SIP List
> >> Subject: RE: [Sip] INFO Framework: Tags
> >>
> >>
> >> Hi,
> >>
> >> Feature-tags:
> >> -------------
> >>
> >>> I agree with John here.
> >>>
> >>> I would include an informative reference to RFC 3427 
> indicating that
> >> any draft that requires option tags for a package
> >>> would need to have the status required for option tag 
> registration.
> >>
> >> One option would have been to define an info feature tag which can 
> >> list package value, and each package could then specify a value.
> >>
> >> E.g sip.info="dtmf-packkage, isup-package"
> >>
> >> But, I am ok with John's proposal also, but I think we should add 
> >> some text as suggested by Keith.
> >>
> >>
> >> Option-tag:
> >> -----------
> >>
> >> I strongly think we should define an option tag for info packages.
> >>
> >> Regards,
> >>
> >> Christer
> >>
> >>
> >>
> >>
> >>
> >>
> >>> -----Original Message-----
> >>> From: [email protected] [mailto:[email protected]] On
> >> Behalf Of
> >>> Elwell, John
> >>> Sent: Thursday, November 20, 2008 1:16 AM
> >>> To: Eric Burger; SIP List
> >>> Subject: Re: [Sip] INFO Framework: Tags
> >>>
> >>> Eric,
> >>>
> >>> If an individual package requires an option tag, that
> >> option tag could
> >>
> >>> be defined in the RFC that defines the package, which would
> >> need to be
> >>
> >>> standards track. I don't think we need to say anything about this.
> >>>
> >>> Similar situation for media tags, except that RFC would not
> >> need to be
> >>
> >>> standards track.
> >>>
> >>> John
> >>>
> >>>> -----Original Message-----
> >>>> From: [email protected] [mailto:[email protected]] On
> >>> Behalf Of
> >>>> Eric Burger
> >>>> Sent: 20 November 2008 00:48
> >>>> To: SIP List
> >>>> Subject: [Sip] INFO Framework: Tags
> >>>>
> >>>> Do we want to specify media tags to indicate package support?
> >>>>
> >>>> Do we want to specify SIP option tags to require Info
> >>> Package support?
> >>>>
> >>>> Comments welcome.
> >>>>
> >>> _______________________________________________
> >>> 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
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.