Re: Capability-based routing in SIP

On Wed, 6 Jan 1999, Dean Willis wrote to the IETF mmusic mailing list:

[...]
> To summarize the issues:
> 
> 1) Capability-based routing seems to require parsing the SDP to decide
> what the session type is. This is problematic for fast setups.
> 
> 2) There seems to be no defined concise way to describe endpoint
> capabilities.

The IETF Content Negotiation (conneg) WG is working on just that. Archives
available from <http://www.imc.org/ietf-medfree>, and there are several
drafts available (draft-ietf-conneg*):

  "The ietf-medfree mailing list is to discuss negotiating
   elements of the presentation of documents that are not naturally
   captured by the MIME Media Type."

> 3) The SIP REGISTER method does not seem to support capability
> information.
> 
> Am I missing something obvious here, or do we really need a richer
> capability description and handling mechanism to handle these scenarios?
> Once we start talking about an even richer set of messaging sessions --
> presence, instant-messaging, and so on, I see the problem becoming even
> thornier.

The basis for CONNEG's work has been fax (specifically T.37 / RFC2305 --
store-and-forward fax using SMTP) and the intent is the same expression
syntax can be used for other protocols as well.

-Dan Wing