MINUTES FOR MARCH 5, 1996 IAB MEETING
(at Los Angeles IETF meeting)
PRESENT:
Brian Carpenter
Steve Crocker
Robert Elz
Elise Gerich
Phill Gross
Christian Huitema
Erik Huizer
Allison Mankin
Paul Mockapetris
Jon Postel
Yakov Rekhter
Chris Weider
Abel Weinrib
Lixia Zhang
New IAB members present:
Radia Perlman
Steve Bellovin
John Klensin
NEXT MEETING:
Teleconference Tuesday April 9, 10-12 Eastern Time.
ACTION ITEMS:
NEW ACTION ITEMS:
OLD ACTION ITEMS:
- Abel Weinrib: Develop a draft document outlining the rules, practices, etc.
for the Internet Research Task Force. Emphasize open publishing of results.
NOTES:
1. last call on any minutes still pending
2. review actions and drafts in progress
- "Principles of the Internet architecture" will be published as a draft, incorporating
changes.
3. administrivia
Wednesday dinner, fix meeting of new IAB to
elect chair, fix 1st teleconference, IAB voting rules)
Christian, Phill, Lixia, Paul, Steve are leaving...
thanks for the their years of great service!
Chair election meeting after Thursday plenary; Brian Carpenter is
willing to continue as chair; other candidates should make their interest
known to Abel Weinrib before Thursday.
IESG liaison--Yakov is willing to continue; other candidates should
make their interest known to Brian Carpenter before Thursday.
The IAB "voting rule" was not clear. After some discussion, the following
was agreed to as approximating the ideal of consensus while removing
single-individual veto power:
The IAB agreed that where the IAB takes a decision, that decision requires
the support of at least seven voting members of the board, and no more
than one dissenting vote.
This will be in effect until a revised version of RFC 1601 has been approved.
4. vote on new IETF chair and IESG slate
Both chair and slate were approved.
5. D.Perkins appeal - final preparation for open hearing
6. IAB/IRTF/IETF futures discussion, and preparation for open meeting
An opinionated member of the community recently said that while individual
members of the IAB add a lot of value to the IETF, the IAB as a body
contributes only marginally to the community. In addition, the IRTF is
invisible, and some people have a big problem with research groups being
closed. The second point led to a long and somewhat heated discussion about
whether research groups should be closed or open. Points made included that
closed groups run counter to the values of the community; that the IPng
directorate seemed to work even though it was closed, but it did publish
minutes and mailing list for interested people and was perceived as being
part of elected IESG; and that on the other hand, IRTF groups demand long
term attention, and have vague enough goals to benefit from having the group
closed.
It became apparent that there was general agreement that to be an effective
part of the community, the IRTF needs to institute regular reporting
of activities and progress of research groups. This could be accomplished
through regular posting of status and meeting reports to the Web, open
meetings of research groups at IETF meetings, etc.
In addition, it was agreed that there is value in having one or more
architectural groups that are concerned with broad architectural issues
that often cross IESG areas but are still critical for the future Internet.
One suggestion is to start out by putting in place a review process
with area directors.
7. Architecture issues
IAB and IPv6, IPv4 Balkanization (Lixia)
Lixia Zhang presented a short talk questioning why the IAB is not
taking on driving the transition to IPv6. She argued that the IPv4 address
space is running out *now*, not ten years
from now as indicated by some allocation curves; that the allocation
curves do not even closely reflect the true Internet growth; and that
a scarce address space is an unstable state to be in and we must move
out of it as soon as possible. NAT boxes may serve as temporary solutions,
but should not be the permanent architectural design.
Lixia argued that, because of the exponential growth of the Internet,
the earlier we transition to IPv6, the cheaper. Therefore, the IAB should
explain to the community the inevitability of the transition and the
danger for waiting too long. The two major reasons against promoting
IPv6:
- "IPv6 does not solve the routing table crisis", and
- "Wait 5 more years and then we will have IPv7 developed",
are both incorrect. For (1), we need a separate effort to handle the
current routing table issue; IPv6 has a potential to provide long term
solutions. And (2) is from some people who argue for fundamentally changing
the current datagram architecture (and hopefully IAB is not
buying into that).
Lixia closed with her recommendation for actions of the IAB:
- Publish an informational RFC to clearly endorse IPv6 and urge the community
to plan the transition ASAP.
- Solve the routing problem with a separate effort.
- Reiterate the principle that globally unique addresses are a fundamental
piece of the architecture.
After Lixia's presentation, there was a heated exchange on these topics.
The chair suggested that Lixia send the list of recommended actions
to IAB mailing list for future discussion at the next teleconference.
Is RSVP deployable?
Here, the issue is the difference in difficulty between intra-enterprise
vs. inter-enterprise deployment. A fundamental issue is the economics
that will apply to the full Internet.
It was proposed that the IAB review both the RSVP specification and
the Multimedia Multiparty Session Control (mmusic) Working Group's architecture
specification.
Future Meetings
Regular teleconference second Tuesday of the month at 10:00 AM Eastern Time.
These minutes were prepared by Abel Weinrib, weinrib@intel.com. An online
copy of these and other minutes are available online the
directory http://www.iab.org/documents/IABmins. Also, visit the IAB Web page at
http://www.iab.org/iab.
This page is maintained
by the IAB Executive Director for the IAB.
