- 1 -

Question(s): / 14/15 / Meeting, date: / Holmdel, 24 – 28 January 2005
Study Group: / 15 / Working Party: / 3
Source: / ITU-T SG15, Q.14/15
Title: / Liaison Statement to IETF on MPLS and GMPLS Change Process
LIAISON STATEMENT
To: / Loa Andersson & George Swallow
(Co-chairs of the MPLS Working Group of the IETF)
Adrian Farrell and Kireeti Kompella
(Co-chairs of the CCAMP Working Group of the IETF),
Alex Zinin and Bill Fenner (Routing Area Directors of the IETF)
Scott Bradner (IETF/ITU-T Liaison Coordinator)
Approval: / ITU-T SG15 Q.14/15 Interim Meeting
For: / Action
Deadline: / 14 March 2005
Contact: / Hing-Kam Lam
Lucent Technologies
USA / Tel: +1 732-949-8338
Fax: +1 732-949-1196
Email:

Thank you for your liaison sharing the proposed IETF (G)MPLS change process. We are concerned with the direction taken in this draft, which to us appears in contradiction with the spirit of the previously agreed “Procedures for handling liaison statements to and from the IETF” (internet draft-baker-liaison-statements-04.txt). Over the past few years, we have jointly cultivated a cooperative relationship between Q14/15 and ccamp to facilitate the application of GMPLS protocols to ASON. We view this draft as threatening to undermine this relationship.

Q14/15 members welcome a process that facilitates productive collaboration with IETF when a new application or requirement for (G)MPLS is identified. We expect that such a process would include: (1) enabling IETF expert assistance in identifying the best solutions to new applications, features, and/or requirements, which may result in the development of (G)MPLS extensions, (2) supporting documentation of such new applications and extensions in a central repository, and (3) ensuring that code points are assigned in a coherent manner through IANA to avoid collisions where different extensions might otherwise use the same code points.

We are concerned that the process outlined in draft-andersson-rtg-gmpls-change-00.txt will hinder, rather than encourage, collaboration with ITU-T. Our most serious concerns are related to the way in which this draft deals with new applications or requirements identified by external standards organizations.

  • The process does not take into account the fact that an ITU-T liaison statement represents the consensus view of another Standards Development Organization (SDO) and not just the view of a single (or a few) individual(s).
  • As currently phrased, the process lends itself to: (1) risking greater or indefinite delay, and/or (2) potentially not addressing requirements published by ITU-T.

This can result from individual disagreements or individual lack of interest in the problem space, etc. We are sure you are well aware of the path such discussions can take.

We would like to see a stronger commitment within the process to focus on meeting the requirements stated by ITU-T fully and in a timely manner, and without being blocked by non-consensus opinions.

  • The "response" to an internet draft, at least in the short term, is typically an email discussion. This is not something that can be meaningfully used as an input document by ITU-T. It would also be inappropriate for ITU-T to try to draw its own conclusions about what the consensus response of IETF is. There is clearly a danger that we might misinterpret the true IETF consensus due to the views of a vocal minority on the email list. Any judgment regarding the IETF regarding the consensus response to ITU-T input is best made by IETF itself, and then sent to ITU-T as a specific reply liaison statement.
  • There is no way to put a deadline on an internet draft. When ITU-T sends something to IETF, we generally request a reply by our next meeting so that we will know how to proceed. (Of course the deadline is only the date by which a response is expected - ITU-T has no authority to compel any particular standards action by IETF in a particular timeframe. But at least the response will let us know the standing of any particular request, e.g., whether IETF understands and agrees with our statement of a problem, what involvement IETF plans to have in development of a solution, the timeframe in which IETF expects to take action, or some counter proposal.)

It is our view that adoption of this proposed process, as it relates to ITU-T, will send a misleading message to the industry at large, and risks creating a result counter to that desired by the IETF. Creating impediments to applying IETF protocols to the application space of another standards development organization would result in a proliferation of (G)MPLS "like" protocols instead of a coherent suite of protocols with broad applicability. We view such a result as detrimental to the IETF and the industry at large, and would like to draw your attention to these concerns.

An electronic copy of this liaison statement can also be found atftp://sg15opticalt:/tsg15opticaltransport/COMMUNICATIONS/index.html

______