notes from the Wed. IAB/IESG breakfast, fyi

[Sally Floyd <floyd@icir.org>]

I took some notes during the discussion Wed. morning of
the IAB Open Architecture Meeting for Mark, who couldn't
be at breakfast Wed. morning.  I am forwarding them just in
case they are of any use to anyone else.  There are no
accuracy guarantees...

- Sally

--------------------------

IAB/IESG breakfast:

Review of IAB Open Architecture Meeting:

Technically, we think it was a successful event, even thought
  it is not clear that anything new was learned.
  Other opinions:
Margaret: I thought the presentations were interesting.  She had
  hoped for more conclusion about an action plan for moving forward.
  If the IAB does nothing for followup, then it was a waste of time.
Mark will write a report.
There is also a plan for a moderated mailing list.
Research groups?
Margaret: I don't know if research groups are the right answer.
  It would be good if we could write down the common knowledge
  that some subset understands.  We could end up with a list
  of research questions, of pressing questions that are not
  research, etc.
Geoff:  This is a well hashed area, but we should publish a pointer
  to the literature.
Eric:  Most of the discussion was about addressing, but not about
  scoped addressing, and the scoped addressing hasn't been discussed
  so much.
Harald: I missed most of the meaning, but from what I heard back,
  there is a larger issue of how machine identifiers are used.
  A frustration from the application layer.  The net runs on
  DNS names.
Randy: Was it scoping in the micro sense instead of the macro sense?
Geoff: Nope.
Rob: Said last night, this is NSRG all over again. (Name Space Research Group.)
Leslie's metaphor:  There is an elephant here, not just tail and trunk
  and isolated parts.
Randy: Inside Multi6, there is a design team moving towards 8plus8.
  Tony Li, Mike O'Dell.
Allison: the relationship between scoping and 8plus8?
Provider independence.
Geoff:  If you unbind locators and identifiers, then this concern
  about provider independence goes away.  The mobile world deals with this.
  He sees a lot of value in the Multi6 design team.
Randy: Two different scoping issues.  
  The RFC 1918 scoping problem (private internets), and nimrod.