Indicating User Agent Capabilities in the Session Initiation Protocol (SIP)
RFC 3840

Note: This ballot was opened for revision 03 and is now closed.

(Allison Mankin) (was Discuss, Yes) Yes

(Jon Peterson) Yes

(Harald Alvestrand) No Objection

(Steven Bellovin) No Objection

(Margaret Cullen) No Objection

(Bill Fenner) No Objection

(Ned Freed) No Objection

Comment (2003-12-12 for -** No value found for 'p.get_dochistory.rev' **)
No email
send info
There seems to be some confusion in section 10.1 as to which tags are in
the sip tree and which are not. In particular, section 10.1 states that
all the tags in this section are in the sip tree. But the tags in sections
10.1.1 through 10.1.6 do not appear to be in the sip tree. I was thinking
the idea was that the sip. prefix was to be assumed, but then along came
sections 10.1.7 through 10.1.18 where the sip. prefix does explicitly appear.

Is it that the first six of these aren't in the sip tree and the remainder
are, or was the sip. prefix on the first six omitted in error? I kind of hope
it is the former, since I think the first six of these could have utility 
outside of SIP, but I won't object it is the latter.

I also note that the 10.1 level seems to serve little purpose since all of
the subsections of 10 are in it. Of course this would change if 10.1 was for
the registrations outside the sip. tree and 10.2 was for those in the sip.
tree.

Section 12.2 reiterates that all the registrations go in the SIP tree, BTW.

Nits:

  Really bad orphan on pages 14, 18 and 42. I'm sure the RFC Editor will fix 
    this.

T

(Ted Hardie) (was Discuss) No Objection

Comment (2003-12-16 for -** No value found for 'p.get_dochistory.rev' **)
No email
send info
I believe some of the registrations need re-wording before they are recorded with IANA, as they
appear to register sets as tokens, rather than registering the tokens, then using set syntax.  Section 10.1.16 says:

Summary of the media feature indicated by this tag: The set of URI schemes [10] that are 
supported by a UA.

I think what they intend to do is parallel to the feature tag registration of something like "paper
size" (see section 2.4 of RFC2534).  That registers individual tokens with typical values 
like "A4" and "B4".  That is then expressed using a set syntax like   ( paper-size=[A4,B4] ) 
(See RFC 2533 and the update  in rfc 2738).

There are several registrations like 10.1.16, and a quick pass through to confirm that they are
each registrations with a fvalue of token, expressable in sets is needed.  This should probably be done before AUTH48, though, since AUTH48 occurs after the IANA registration.

Sorry for not catching this earlier.

(Russ Housley) No Objection

(Thomas Narten) No Objection

(Bert Wijnen) No Objection

(Alex Zinin) No Objection