[Foundation] Recent confusion about the JSF, JINC and the IET F

Tony Bamonti tbamonti at jabber.com
Fri Aug 2 10:25:35 CDT 2002


JSF Members,

I am responding to this message as a representative of Jabber, Inc. and one
of the people responsible for the IETF initiative.  I am reasonably new to
the company (coming on 3 months)and have responsibility for representing
Jabber, Inc. management in driving open alliance and standards initiatives
for the Jabber protocol/technology.  I have had limited participation with
the JSF to date, primarily observing mailing list and conference activity,
learning about its organization, structure and dynamics and trying to
understand how I can contribute to what appears to be a healthy and growing
open source and development community.

Rob has done a commendable job of articulating a number of important issues
and questions that are receiving alot of attention among the JSF membership.
Many of which are directed towards JINC.  I have attempted to address these
questions in-line below (TB), but these are important issues that deserve
on-going dialogue.  I encourage additional discussion either openly on this
list or privately by contacting me as indicated below.


Sincerely,

Tony Bamonti
Jabber, Inc.
tbamonti at jabber.com
303/308-3655

-----Original Message-----
From: Robert Norris [mailto:rob at cataclysm.cx]
Sent: Wednesday, July 31, 2002 11:32 PM
To: members at jabber.org
Subject: [Foundation] Recent confusion about the JSF, JINC and the IETF


Hi all,

This email is in response to a recent discussion in jdev. If you're not
already aware of it, you should checkout the links that stpeter has
posted here:

  http://www.saint-andre.com/blog/2002-07.html#2002-07-31T22:02

Now, the recent debate appears to have been sparked by the recent IETF
BOF on the XMPP protocol drafts. Below I've outlined my perception of
the membership at large. This doesn't necessarily reflect my own
opinions on the topic, which I have noted at the end of this message.


So, the way things look to me:

 - The JSF membership, in general, understands the role of the JSF to be
   an overseer of management and development of the protocol. What this
   means, for most people, is that the JSF effectively "owns" the
   protocol.

(TB:  I would argue that as an "Open" protocol, no one "owns" the Jabber
protocol.  The JSF is chartered with guiding the evolution and management of
the protocol.)

 - The core Jabber protocols (XMPP) have been offered to the IETF.
   However, the membership was not formally informed of that submission,
   nor were they necessarily in agreeance with it.

(TB:  The protocol was submitted to the IETF well over a year ago, prior to
the formation of the JSF, so there is some precedence and the membership
knew about this.  There is some truth to the statement that there wasn't
"formal" communication of this recent effort within the JSF
membership....although, the submissions and the Jabber BOF were discussed in
the Foundation meeting on 7/10/02 prior to the BOF.....at that time there
was no apparent dissention from anyone present in the meeting.  However, in
hindsight, we could have done more to communicate the effort to the broader
JSF membership.  There were some timing  issues that complicated broad
communication and debate....but this is no excuse, we need to do a better
job of making sure the broader JSF membership was aware of the effort and I
will work on ensuring this happens in the future.)

The confusion has been furthered by a couple of lines in section 6 of
the minutes of the IETF BOF minutes (available at
http://www.jabber.org/ietf/bof-minutes.txt):

Q: Are you comfortable with the notion of giving up control of the
   technology in your flagship project?
A: Yes, this has sign off at the highest levels of managment in Jabber,
   Inc.
  

Q: Will you support the resulting specification?
A: Yes, this has sign-off at the highest levels of managment in Jabber,
   Inc.

(TB:  Both of these questions need to be considered in the context of the
situation.  I was not at the BOF and Joe Hildebrand is currently out of
town, so I will try to explain to the best of my knowledge and let him
clarify later if necessary.  Joe was on the podium and was forced to answer
these questions.  He was at the BOF as a Jabber, Inc. representative.  While
he is also member of the JSF, he was not there in a capacity to represent
the JSF membership.  He (astutely I might add) answered the question with
the authority he had, as a JINC representative......he did NOT say that
Jabber, Inc. had the "authority" to give up this control.....he only said
that the highest levels of management at Jabber, Inc. had signed off on
it........Joe answered in the only way he could to not jeopardize the
objective of the BOF and at the same time not misrepresent himself as
speaking for the entire JSF community.)

>From this, the questions that many people are asking seems to be:

1. Who (JSF or JINC) offered the protocol to the IETF?

(TB:  Since no one "owns" the protocol, no one can really "offer" it.  It
was "submitted" for consideration by several individuals who admittedly all
work for JINC but many of whom are also JSF members. The submittal was not
represented as coming from a single organization.)

2. If it was the JSF, why was the membership not consulted?

(TB:  It wasn not positioned as a submittal specifically by the JSF.
However, as mentioned above, there could have been been better communication
of the effort to the membership.)

3. If it was JINC, why did they do it? Especially since, as a JSF member
   (and the major provider of funds to the JSF) they understand its
   role.

(TB:  It also was not positioned as a submittal specifically by JINC.
Individuals from JINC, members of the JSF and others (e.g. France Telecom,
BellSouth) contributed to this effort, believing it was the right thing to
do for the protocol and the greater Jabber community.)

4. Is there a difference between the Jabber protocol that the JSF has
   been charged with managing, and the XMPP protocol that was offered to
   the IETF?

(TB:  There is no difference today.  IF the IETF approves an XMPP working
group, I would expect significant participation and contribution by JSF
members to achieve IETF ratification.  And by doing so, retain a high level
of influence on the protocol now and in the future.  The JSF will also
continue to be responsible for developing and managing protocol extensions
as this is not part of the IETF charter.  No one would benefit from
diverting the efforts.)

5. If so, exactly who (currently) controls the XMPP? And, where is the
   line between XMPP and Jabber?

(TB:  See above.)


Now, the situation (as I understand it) is that the IETF process was
begun before the formation of the JSF (and technically, even further
before that - the protocols were offered to the IETF several years ago).

(TB: true)

The IETF is not some huge mega-entity, that will wrest control of the
protocol away from the users and developers (ie us). Instead, the XMPP
Working Group (if one was formed), is responsible for the
standardisation of the protocol, and that working group is made up of
interested individuals, much like the JSF membership in that regard.

(TB:  There is no formal IETF membership.  There is a governing body that
stewards the development of proposed standards and makes final decisions on
ratification of standards.  But the real work of any working group is done
by individuals interested in promoting the development and standardization
of a particular technology.  I would hope and expect that constituency to be
largely comprised of JSF members.)

XMPP is the core Jabber protocols - streams, messages, IQs, presence,
rosters and auth. That is the only thing that an XMPP WG would be
responsible for. Definition of protocol extensions is a function
performed by any interested parties, including the JSF.

(TB: Right....the XMPP WG would be focused on refining the "core" protocol
to meet the IETF IMPP and CPIM standards primarily for use as an INSTANT
MESSAGING standard.  It is unlikely the IETF will do much at all in the area
of protocol extensions (except for possible areas of security).  The JSF can
and should still be the primary entity responsible for defining and
approving extensions.......if the XMPP goes to RFC, these extensions can be
brought to the IETF for future consideration/inclusion in another WG....or
not.)

If my understanding is correct, we genuinely have nothing to fear from
the IETF. Representation in the IETF means that large vendors are more
likely to support the protocol, and we can continue with the JEP process
to make whatever extensions we like.

(TB: Agreed. Ratification as an Internet standard will definitely increase
Jabber's adoption  and visibility in the broader Internet and commercial
markets.  Responsibility for extensions will fall on those most interested
in evolving Jabber beyond what it is today (i.e. the JSF.)

Rob.

-- 
Robert Norris                                       GPG: 1024D/FC18E6C2
Email+Jabber: rob at cataclysm.cx                Web: http://cataclysm.cx/


Tony Bamonti
Jabber, Inc.
tbamonti at jabber.com
303/308-3655



More information about the Members mailing list