[manet] AODVv2 Decision

[Justin Dean <[e-mail]>]

Working Group,

I will not be passing AODVv2 out of the working group and to the IESG.
This was not an easy decision, nor one made lightly. However, despite many
attempts to draft this email otherwise, I found I could not justify
AODVv2’s release. AODVv2 is released to the document editors. The following
is the rationale for my decision.

There is insufficient consensus within MANET to make the passage
appropriate.  More importantly, perhaps, is that this lack of consensus is
justified by the significance of the outstanding issues.  These issues
include: security, metrics, multiple interface operation with shared ip
addresses, possible looping, non-symmetric links, ordering of processing
operations, lack of jitter specification, removal of timeouts, and gateway
limitations.  Many of these issues have been discussed at length, and have
apparent consensus solutions proposed and written up.  In fact, the
progress within the past month has been impressive.  Yet, to address these
issues, the underlying protocol specification operation has changed within
the last call and I do not believe the document is without error.  The lack
of reference implementation or analysis, in simulation or otherwise,
reinforces this concern.

Despite these flaws, I continue to believe AODVv2 is better specified in
many ways than AODV, as it acknowledges and attempts to solve issues AODV
does not even mention such as: multiple interfaces (beyond a passing
sentence), designed with security in mind, extendable with the use of
RFC5444, metric based routing built in, gateway support (although limited),
bi-directional link checking.  There is value in this draft and it is my
hope that it can be published in some form. I am personally willing to help
make that happen.

To the editors who worked so hard, you were given an impossible task: to
address WG concerns, form consensus and provide verifiable fixes within the
timeframe provided.  When issuing the last call and doing my review I was
not sure that it was impossible, and held out hope until the very moment of
this decision that it wasn’t.  I am sorry that I was mistaken.

The amount of work produced in the last month has been impressive but at
this time I’m not willing to pass it out of the working group.

Justin