We have a clear consensus in the WG that "something needs to be done
about INFO."
In this week's SIP meeting, we discussed a draft by Eric Burger that
proposes formally restricting the usage of INFO to the existing RFCs
that exercise it.
Several people noted that just making this limitation in a void is
insufficient -- that we also need to explain the fundamental
architectural problem and give guidance on how to more effectively
solve the kinds of problems that people have been trying to solve
using INFO.
We also have a contingent that thinks that we should fix info by
adding a negotiation mechanism and registration procedure for INFO
usages, much as SIP events have been done.
Others think that rather than an extensible mechanism using one
message type (INFO), we should instead extend SIP as needed with new
message types, further blurring the line between protocol and
application (which I'll admit is already pretty blurry in SIP's
case). Note that this is a follow-on mechanism to the first step of
documenting NOT using INFO.
It seems to me that this means we need to clarify the SIP extension
model, which I'll come back to later.
At this week's meeting, there was a majority (at least a 2 to 1
ratio, and more by some counts) in favor of taking the path of Eric's
draft and declaring a formal moratorium on new usages of INFO. This
is pretty close to "rough consensus", but not so close that I was
willing to declare consensus without taking the question to the list.
So here's a proposal:
I suggest we accept the model of Eric's draft, but ask him to add
more depth to the architectural discussion and ask Christer to
contribute some text in this direction, as he has some very definite
ideas of what needs to be addressed. This leaves us with clear
direction on not extending INFO with new usages or a negotiation
mechanism.
Please discuss this on-list, especially if you object or have a
counter-proposal.
This leaves open the question of clarifying the extension model of
SIP, which I will raise in another thread to make following the
discussion easier.
--
Dean Willis
SIP co-chair
_______________________________________________
Sip mailing list https://www1.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
