> -----Original Message-----
> From: Dean Willis [mailto:[EMAIL PROTECTED] 
> Sent: 20 November 2008 22:47
> To: Paul Kyzivat
> Cc: DRAGE, Keith (Keith); SIP List; Elwell, John; Christer Holmberg
> Subject: Re: [Sip] INFO Framework: Tags
> 
> 
> On Nov 20, 2008, at 4:08 PM, Paul Kyzivat wrote:
> 
> >
> >
> > Dean Willis wrote:
> >
> >> If I don't support info-packages, then I MIGHT support 
> old-info. If  
> >> I do, then I MIGHT understand an INFO (or the legacy set), and I  
> >> MIGHT send you one (from the legacy set). But I'm very, very  
> >> unlikely to understand any of the new CID-indirection-to-select-a- 
> >> body, multiple-body stuff, so don't send it!
> >
> > Dean,
> >
> > You don't need an option tag for this. After the invite, I can't  
> > send you an info package unless you have provided a 
> Recv-Info for it.
> 
> Okay, I'm willing to believe that.
> 
> What if I have an application that really, really requires info- 
> packages, so I want the call to fail if you can't handle them?
> 
> Would this arise, or would it always be preferable to complete the  
> INVITE exchange, then tear the call down once I find out that 
> you don't?
[JRE] This might arise on a per-package basis, in which case the spec
that defines that package can also define the option tag. I can't see it
arising in a package-independent context, so we don't need an option tag
for the extension as a whole.

PLEASE try to keep this extension as simple as possible - this option
tag / media feature tag business adds complexity. Are we really trying
to dissuade people from implementing this extension?

John
_______________________________________________
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

Reply via email to