While agreeing wholeheartedly with Paul Jones's reply to Francois Menard I side
with Tom's in his conclusion that the Megaco/H.248 collaboration has been a
success. I would go further, and say I think it has been an outstanding
achievement, considering the odds that were stacked against it.
I would list among these:
1. Deep philosophical differences of approach between the two Standards bodies
in the way a Standard emerges and the decision processes involved,
2. Strongly held technical beliefs on both sides; arising partly from 'religion' and
partly from conflicting strategic objectives,
2. Vested commercial interests on the part of some of the key players,
3. A marketing campaign going on in the background, aimed at replacing H.323
with SIP/SDP,
To talk of mistrust and suspicion is to grossly simplify the situation.
Having said all this, I would like to thank and congratulate all involved for bringing
this effort to such a successful conclusion. In particular, I think we all owe a
debt of gratitude to Tom for the way he has negotiated all these obstacles with
patience and good humour and for the way he has straddled two very different
cultures and two very different processes.
I would also like to thank John Segers for the long hours of frustrating work
involved in producing a well written and well structured document from the
amorphous scratchings he inherited. Its a thankless task being an ITU editor
and I think John deserves a special thanks, particularly for the 48 hours or more
that he went without sleep during the meeting to get the final version of the
document ready.
Despite all these problems, I believe we will see a lot more of this type of co-
operation. Why, because its the only approach that makes sense. The market
demands interoperability in the area of VoIP and this can only be achieved by
collaboration between the two Standards bodies involved.
Going forward, I would suggest that an MOU is drawn up early in the process not
at the end. A detailed MOU is not something to apologise for, its just good
business practice, and a practice that results in the minimum of
misunderstandings and mistrust.
Finally, let's not let the process(es) on each side get in the way of progress. I
think the less process involved the better. A document should be approved when
consensus agree its ready for approval. Two sets of beaurocratic hurdles don't
help matters, they just slow things down.
Mike Buckley
____________________ Begin Original Message
___________________________
Date: Sun Feb 27 19:43:19 -0500 2000
From: internet!NORTELNETWORKS.COM!taylor (Tom-PT Taylor)
Subject: In Defense Of Cooperation
To: internet!MAILBAG.INTEL.COM!ITU-SG16
Content-Type: multipart
Content-Length: 7439
Paul Jones used the term "H.248/Megaco fiasco", and Dale Skran was quite
intemperate in his criticism of the joint effort when he spoke in Geneva. I
was not in a position to raise much of a defense there, because it would
have conflicted with Glen's positioning of the H.248 work as stable. Now
that that issue is water under the bridge, I would like to suggest that it
really has been quite a successful project. I don't think we should go at
it in quite the same way in the future, but we should not come away totally
discouraged about the prospects for IETF/ITU-T cooperation.
Here are bases for my conclusion:
We have been technically successful, and what we have created is better than
it would have been if created in either body separately. To be specific, we
retained a great deal of the thinking and experience which went into MGCP,
but escaped the limitations MGCP's connection model imposed on the use of
different bearer types and multiple media.
There is a perception that the IETF decision model is lacking because so
much had to be done at the ITU-T meetings. However, the grueling effort put
in at the latter resulted partly because of fundamental conflict to which
Q.13-14 has been unaccustomed (Monterey and much of Red Bank in particular)
and partly because there was so much to cover. This may have given the
impression that the IETF didn't do anything on their watch, but we did work
through a lot of material. Moreover, both Q. 14/16 and Megaco made a host
of decisions which have shaped the final protocol.
I have pleaded guilty to a failure to complete IETF process in timely
fashion. There was a brief window of opportunity in early December when it
seemed that the list was quieting down. If I had made Last Call at that
point, the long list of issues which was contributed into the Geneva meeting
would instead have been forced into the open at an early enough point that
IETF process could still have been completed in time for the Geneva meeting.
I consider this to be a tactical failure, in that real work was done in the
intervening month and a half, and the protocol specification is the better
for it.
We have come to a point where SG 16 is thoroughly mistrustful of the IETF as
a partner, and has shown this distrust by negotiating a contract for further
progress on H.248/Megaco down to the last detail. The pity is that I could
see that SG 16 was overly anxious to do its part, from Santiago onwards. We
were not really ready for determination at that point, and we were not ready
for decision coming out of Red Bank. I can understand that SG 16 feels
betrayed because the IETF was not finished in time, but the fact is that the
job was bigger than the schedule allowed for.
This is why, if I am ever involved in a project like this again, I will
insist that the IETF and ITU-T working together create an IETF Proposed
Standard before determination ever happens. The opportunity for false
expectations and artificially tight deadlines will thus never arise, and
both sides will be the happier for it.
Tom Taylor
Advisor -- Emerging Carrier IP Standards
E-mail: taylor(a)nortelnetworks.com (internally, Tom-PT Taylor)
Phone and FAX: +1 613 736 0961
