ASON reqts - Moving on?

["Adrian Farrel" <afarrel@movaz.com>]

Can we back off a bit and just look at the ASON requirements draft?

Having had a quick chat with a couple of the authors (not all) we are in
agreement that this draft is to state clearly the requirements of the ASON
architecture as applicable to GMPLS signaling within the context of the IETF.

The motives are to
- make those requirements accessible to the IETF community
   in a form and format with which they are familiar
- provide a basis for determining whether current protocol
  solutions fully address the requirements.

We believe the former is a necessity notwithstanding the prior work. At the very
least it rehashes old ground with which the informed reader should already be
familiar. At best it puts added clarity and context on top of the existing work.

The second point is the one that might give rise to debate in the future. That
is, once the requirements have been exposed in a way we understand, we may
discover that the existing work is exemplary or we may determine there are holes
to be plugged. Views on this are premature until the requirements draft is solid
(although the prematurity of views will in no way prevent them from being
expressed).

As has been pointed out by several people, there are areas in the current draft
which need clarification and extension. There have also been some topics raised
which the authors need to examine - prinicpally whether certain requirements
*are* requirements and whether they have been omitted.  This is good, and why
the mailing list exists.

My personal view is
- the authors need to do more work on the draft
- it would help widen the discussion of the draft outside the IETF if it were a
WG draft
- the material is relevant to the WG
- it may be a repeat of work that is already done, but we can't tell until we've
done it.

Kireeti, what are the next steps?

Adrian