DNA BOF

["James Kempf" <kempf@docomolabs-usa.com>]

Basically, I agree with Mark's report.

For this WG to be successful, I think they will need to focus on stuff they
can do something about and avoid stuff they can't, like wireless link
protocol peculiarities. The stuff they can do involves improvements to IPv6
Neighbor Discovery to decrease the amount of latency involved in detecting
that the routing needs to change when a subnet change occurs and in
configuring on a new subnet (like DAD). The IPv4 work has apparently moved
to DHC, so is not relevent to this WG.

There was a question about whether the work on optimistic DAD should move to
the IPv6 WG. In principle, this is a management issue for the IESG, but
considering that the IPv6 WG is on track to close down at some point, it
probably would make some sense to have DNA do the work, separate from the
base ND/DAD spec. There are also some arguments in the other direction, of
course.

At the same time as this WG shouldn't get involved in wireless link
peculiarities, I think the IETF needs to consider working more closely with
IEEE for 802.11, similarly to the way we've worked with 3GPP in the past.
IEEE has a track record of making decisions that have not been helpful in
solving the problems or providing the right tools for the IP layer.

            jak