Liaison statement
Response to your liaison dated 31st October 2016

Thank you for your liaison message of 31st October this year addressed to the
IESG and OPS Area Directors.

Since your liaison, the Layer 2 VPN Service Model (L2SM) working group has been
formed by the IESG. This happened right after the L3SM working group succeeded
in its goal of describing an L3VPN service as a data model by documenting it in
the YANG modeling language. You can see the new working group's charter at
https://datatracker.ietf.org/wg/l2sm/charter/ and the product of the L3SM
working group at https://datatracker.ietf.org/wg/l3sm/documents/

The L2SM working group met for the first time at IETF-97 in Seoul during
November and had a lively discussion about its work including the issues of
terminology, functional boundaries, and potential overlap with or use of MEF
work. You can see the meeting materials and draft minutes at
https://datatracker.ietf.org/meeting/97/materials/ by searching for "L2SM". 
Both Mahesh Jethanandani and Scott Mansfield were in the room during the
meeting and can probably give you their views on what happened.

L2SM will follow the same path as was taken for L3SM, and is tasked to do
something similar to L3SM.

Your liaison message is most helpful in pointing out and supplying copies of
existing and on-going MEF work related to the task of the L2SM working group,
and the fact of you making available log-in details and material for your
work-in-progress is very much appreciated.

You asked a specific question in your communication:

> Is the focus of L2SM about the internal network technologies and
> implementations as opposed to "characteristics of the service as
> discussed between the operators and their customers"?

The answer is the latter (although it should be noted that there is some debate
about whether "operator" or "service provider" is the correct term).

Through the discussion during the meeting and on the list, we believe a
suitable point of differentiation between MEF and L2SM is that MEF has worked
on CE-based models while L2SM will work on PE-based models. PE-based models
necessarily include a number of technology-specific details and will focus on
the networking aspects of the service while still keeping the realisation of
the service within the provider's network as technology agnostic.

Furthermore, while the MEF clearly embraces all aspects of the LSO architecture
including billing, SLA, order management, life-cycle management, etc, the L2SM
working group is tasked to do something much smaller: a self-contained service
YANG module similar to tat produced for L3SM. While it would be nice to fit the
L2SM work into the MEF LSO architecture, it appears that L2SM follows a
different architectural approach, and that there is no value in trying to "fit
a square peg in a round hole" by trying to force a match of the functions
across the two approaches. This does not invalidate either approach, but only
observes that they are different.

The working group has so far been clear in its intent to, where possible and
consistent with its own work, re-use by reference any relevant existing
material such as that developed by the MEF Forum. So far it has identified
SLAs, QoS, and Billing as components that may very well be candidates for such
re-use.

At the moment, the working group is at a very early stage. We welcome 
participation by all interested parties, but in particular by service providers
and service customers since they are the principal users of the YANG model
under construction. Participation is open to all via the mailing list
https://www.ietf.org/mailman/listinfo/l2sm.

We wish you well with your January meeting.

Qin Wu and Adrian Farrel
L2SM Working Group Chairs