[Foundation] quick IETF report

Peter Saint-Andre stpeter at jabber.org
Wed Nov 19 11:32:36 CST 2003


Hi JSFers,

Here is a quick report on our IETF progress, based on attendance by me
and Joe Hildebrand at IETF 58 last week in Minneapolis:

1. The "core" and "im" drafts are effectively done. Aside from one
outstanding issue regarding stringprep profiles, all feedback has been
incorporated into the latest candidate documents available at our IETF
page: <http://www.jabber.org/ietf/>

2. A Working Group Last Call (WGLC) will be issued shortly for our 
other two Internet-Drafts (CPIM mapping and end-to-end encryption). 
These docs exist mainly to conform to other IETF protocols. Given that 
WGLC usually lasts a week or two, and that the subsequent IETF Last Call
usually lasts another two weeks, I would say that we will be finished 
with these documents in January (to be realistic on timing).

3. It seems likely that our IETF Area Director (Ted Hardie) will present
all four I-Ds simultaneously for a vote by the Internet Engineering
Steering Group (IESG), which plays a role in the IETF similar to that of
the Jabber Council in the JSF. If so, I would expect all four of our
Internet-Drafts to be voted to Proposed Draft in January.

4. If we have dependencies on other Internet-Drafts that are not yet
RFCs (this is certainly the case for the CPIM docs, which have been
approved but do not yet have RFC numbers, probably because they in turn
are dependent on other documents), then our I-Ds may not receive RFC
numbers for a while after they are approved. Sometimes this can take a
few months.

5. There was discussion at the XMPP WG meeting last Friday about
possible other work that could be pursued either within the XMPP WG or
in a follow-on working group (I would prefer the latter). Some possible
items include pubsub (though that would probably force us to also submit
service discovery and perhaps x:data) and MUC. To my mind, no one has
yet made a strong argument that taking these protocols to the IETF would
be a good thing, since we would lose change control and doing so would
introduce a level of protocol churn that was appropriate (I think ;) for
our core protocols but not for more application-oriented protocols.
Naturally we need to have discussion about this in the JSF (probably
more discussion than we had about our initial work with the IETF).

6. We also had some discussions with people from the SIMPLE WG about
some ways to seek more seamless interoperability between SIMPLE and
Jabber. The discussions were not conclusive. We agreed to continue
talking about finding ways to at least make sure that IM addresses are
consistent (and even that may be difficult, since SIP addresses are not
internationalizable as far as I know), and I said that I would explore
the use of the IETF's RPID format for extended presence in Jabber (since
we need to figure out extended presence anyway, and I would prefer not
to reinvent the wheel if we don't need to -- though of course we may
need to in the end). But there is no grand unified theory of IM in the
works, no hybrid protocol, no great merger (as the press would have you
believe).

7. The chatrooms hosted at ietf.xmpp.org were quite popular, and we
heard reports of people downloading Jabber clients simply in order to
join these chatrooms. Special thanks to David Sutton for code changes 
and configuration help, and to Peter Millard for admin assistance. The
folks at the IRTF (the research counterpart to the IETF) have asked me
if we can host rooms for them as well, so we plan to get that service
running soon. BTW, as you might imagine many IETFers are quite conscious
of security, so in general they used only clients that support SSL and
end-to-end encryption. We received feedback from some of the security
gurus there, and hope to get them to comment on security proposals
within the JSF going forward.

For the record, minutes and logs of the XMPP WG meeting are here:

http://jabberstudio.org/pipermail/xmppwg/2003-November/001725.html
http://www.xmpp.org/ietf-logs/xmpp@ietf.xmpp.org/2003-11-14.html

HTH,

Peter

-- 
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.php




More information about the Members mailing list