Network Working Group S. Crocker
Request for Comments: 1396 Trusted Information Systems, Inc.
January 1993
The Process for Organization of Internet Standards
Working Group (POISED)
Steve Crocker, Chair
Status of this Memo
This memo provides information for the Internet community. It does
not specify an Internet standard. Distribution of this memo is
unlimited.
Abstract
This report provides a summary of the POISED Working Group (WG),
starting from the events leading to the formation of the WG to the
end of 1992. Necessarily, this synopsis represents my own
perception, particularly for the "prehistory" period. Quite a few
people hold strong views about both the overall sequence and specific
events. My intent here is to convey as neutral a point of view as
possible.
Background and Formation of POISED Working Group
The POISED WG resulted from two sequences of activity, both
intimately related to the growth of the Internet. During 1991, there
was great concern that the IP address space was being depleted and
that the routing tables were growing too large. Some change in the
IP addressing and routing mechanisms seemed inevitable, and it became
urgent to explore and choose what those changes should be. The ROAD
Working Group was formed to study the issues and recommend changes.
The ROAD group returned with a specific recommendation for the short
term, but did not reach a conclusion on a long term plan.
The Internet Engineering Steering Group (IESG) then formulated a plan
of action for further exploration of the issues and forwarded these
recommendations to the Internet Architecture Board (IAB). In June
1992, after the INET '92 meeting in Kobe, Japan, the IAB met and
considered the IESG's recommendations. After considering the IESG's
recommendations, the IAB felt that additional ideas were also
important, particularly some of the addressing ideas in the CLNP
protocol. The IAB communicated its concerns, and there was immediate
controversy along two dimensions. One dimension was technical: What
is the best course for evolving the IP protocol? How important or
useful are the ideas in the OSI protocol stack? The other dimension
Crocker [Page 1]
RFC 1396 Poised Report January 1993
was political: Who makes decisions within the Internet community?
Who chooses who makes these decisions?
As often happens during periods of conflict, communication suffered
among the several parties. The June communication from the IAB was
understood by many an IAB decision or, equivalently, a sense of the
decisions the IAB would make in the future. In contrast, many if not
all on the IAB felt that they were trying to open up the discussion
and their memos were intended as advice and not decisions. From my
perspective, this form of miscommunication was partly due to the
extended size of the Internet technical community. When the
community was much smaller, the IAB was in close contact with the day
to day workings of the technical groups. With the creation of the
IESG and area directorates, there are now two or three layers between
a working group and the IAB.
These matters came to a head during the Internet Engineering Task
Force (IETF) meeting in July in Cambridge, MA. It was made clear
that the consideration of changes to the IP protocol remained open.
Work on that topic has proceeded and is reported in the appropriate
forums. However, it became clear that it was necessary to examine
the decision process and the procedures for populating the IESG and
IAB. With respect to the procedures for selecting IAB and IESG
members, the procedures that were in place derived from the creation
of the Internet Society (ISOC) and the ISOC's sponsorship of the IAB.
These procedures had been developed during the early part of 1992 and
had been adopted by the ISOC during its meeting in Kobe in June.
Hence, as fast as the ISOC was building the framework for supporting
the Internet community, the community was questioning its structure
and processes.
Following the IETF meeting, Vint Cerf, Internet Society president,
called for the formation of working group to examine the processes
and particularly the selection process (Attachment 1). During
August, the working group was formed, I was asked to chair it, and a
charter for the WG was formulated (Attachment 2). (The acronym is
due to Erik Huizer and originally stood for The Process for
Organization of Internet Standards and Development. It was shortened
to fit into the space available on paper and in the IETF
Secretariat's database.)
Deliberations: August through mid-November
The formation of the POISED WG provided a forum for discussion of
process issues. An estimated 20 MB of messages filled up disks all
over the world. Much of this discussion was fragmented or focused on
narrow issues. The salient point that emerged was the need for a
well defined process for selecting leaders with explicit community
Crocker [Page 2]
RFC 1396 Poised Report January 1993
representation in the selection process. There was also substantial
discussion of the role of the IAB -- to what extent should it make
decisions and to what extent should it provide technical guidance? --
and the relationship between the IAB and IESG.
After several weeks of discussion, Carl Malamud and I attempted to
capture the main elements of the discussion by presenting a specific
proposal for the reorganization of the entire structure. The main
elements of the proposal were:
o Retention of the WG and area structure now in place within the
IETF.
o Replacement of the IAB and IESG by two boards, one devoted to
technical management and one devoted to oversight of the
process.
o Well defined terms for members of both boards.
o Selection by committees with input from the community.
This proposal was technically radical in the sense that it proposed
new structures to replace existing structures instead of proposing
changes within the existing system. The proposal focused all further
discussion and set the stage for the fall IETF in mid-November in
Washington D.C.
November IETF meeting
By virtue of the intensity of interest throughout the community, the
POISED WG was one of the focal points of the IETF meeting. The
schedule included a plenary session Tuesday morning to present the
current state of the POISED WG discussions, a formal POISED WG
session Tuesday afternoon and an open IESG meeting Thursday evening
devoted to the POISED issues. The formal schedule was only the tip
of the iceberg; numerous meetings took place over breakfast, lunch
and dinner, in the halls and off in the corners. The more active
participants probably had a dozen or more separate meetings on this
over the three most active days, Tuesday, Wednesday and Thursday.
Amidst all this frenetic activity, remarkable progress occurred at
two key points. At the Tuesday afternoon POISED WG meeting, Lyman
Chapin, IAB chair, Phill Gross, IETF chair and IAB member, and other
IAB members proposed changes within the existing IAB/IESG structure
which converged with the process management elements of the Malamud-
Crocker proposal. The key point was that all processing of standards
actions, including the final decision to advance a specification
along the standards track, would be made by the IESG. This change in
Crocker [Page 3]
RFC 1396 Poised Report January 1993
the process shortens the decision cycle and brings it a step closer
to the working group. Convergence on this key point obviated a
radical proposal and signaled the building of a consensus on how the
standards process should evolve. Over the next two days attention
then turned to the selection process.
As indicated above, there was a strong feeling in the community that
the IAB and IESG members should be selected with the consensus of the
community. A natural mechanism for doing this is through formal
voting. However, a formal voting process requires formal delineation
of who's enfranchised. One of the strengths of the IETF is there
isn't any formal membership requirement, nor is there a tradition of
decision through votes. Decisions are generally reached by
consensus with mediation by leaders when necessary.
Various formulas were considered, and the one that emerged was that
IAB and IESG members would be selected by a nomination and recruiting
committee. The committee is to consist of seven members from the
community, with non-voting representatives from the IAB and IESG and
a non-voting chair provided by the ISOC. The seven members are to be
volunteers, with selection by lot if there are more than seven
volunteers. The only requirement for volunteers is they must have
attended two IETF meetings. This requirement is designed to ensure
the nomination committee has some familiarity with the Internet
community and the standards process.
IAB and IESG members are to serve two years. Half of each body is to
have terms starting in odd years, and half is to have terms starting
in even years. Selections to the IESG have to be ratified by the
IAB, and selections to the IAB had to be ratified by the ISOC. In
the event that the nomination committee is unable to reach a
consensus on a single candidate for each position, it may forward
multiple nominations to the ratifying body, and the ratifying body
will select the candidate.
In addition to this selection process, a recall mechanism was
outlined using a similar scheme. The ISOC is to supply an ombudsman
who will field complaints after all oversight processes have been
exhausted. If the ombudsman is unable to resolve a complaint after a
cooling off period, a recall committee, selected at random among
volunteering community members, will consider the matter. A two
thirds vote by the committee is necessary to remove someone.
This proposal was formulated and circulated during Wednesday and
Thursday and presented at the IESG open plenary. In contrast to the
extraordinarily contentious open IESG meeting in Cambridge, this
meeting was characterized by a strenuous effort by numerous people,
representing diverse points of view, to reach consensus on this
Crocker [Page 4]
RFC 1396 Poised Report January 1993
proposal, and the meeting ended with a distinct decision to proceed
on this basis. Given the strong consensus that emerged at that
meeting, the group decided to implement the selection process by the
next IETF meeting, with the new IESG and IAB members to begin their
terms at the termination of the IETF meeting in March.
On Friday, the IAB and IESG met jointly to determine what to do next.
Both groups agreed to implement the change in processing standards
actions quickly and cooperatively and to identify the positions which
are open for selection. Within a couple of weeks, the IAB finished
processing the standards actions in its queue, and IESG began to
handle standards actions on its own.
December ISOC meeting
The Internet Society Trustees met December 10 and 11 at CNRI in
Reston, VA. The process and organization of the IAB and IETF was one
of their major concerns. A session of the Trustees at 3:00 p.m. EST,
December 10, was broadcast via the Internet. It was not clear how
many people listened, but Geoff Huston, Internet Trustee, was spliced
in separately from Australia.
At this session, I presented the POISED WG results deliberations and
asked on behalf of the IETF that the Trustees approve the selection
process described above. For the long run, a new charter is needed.
Given the very compressed schedule for these activities, there has
not been time to draft and refine a new charter, so the Trustees were
asked to approve the general direction of the reorganization of the
IAB and IETF and give temporary approval to the selection process in
order to permit the first round of selections to proceed.
The Trustees expressed strong approval for the work of the POISED WG
and general approval for the direction of the effort. One area of
concern for the Trustees is the legal liability of the Internet
Society regarding decisions the IESG might make in the future. The
Trustees made it quite clear that they are not inclined to
micromanage the IETF process, but they do feel compelled to
understand the legal issues and help construct a charter which is
consistent with their responsibilities as Trustees.
The session adjourned with agreement to proceed on the current course
and for the IETF, IAB and ISOC Trustees to work together to draft the
appropriate charter.
Crocker [Page 5]
RFC 1396 Poised Report January 1993
Future activities
Both the IESG and IAB have selected the positions which must be
filled through the new selection process. As I write this, Vint Cerf
has been working to find a chair for the nominations committee, and
the process should move forward during January and February.
Communications on the details of the nomination process will be
published on the IETF mailing list and possibly other forums. As
described above, the selection process should be complete well before
the next IETF meeting, and preferably by the end of February.
The other open agenda item is the draft of a new charter for the IAB
and IETF and adoption of the charter by the ISOC. This is the next
order of business for the POISED WG.
ATTACHMENT 1: Message from Vint Cerf to the IETF, et al.
To: pgross@nis.ans.net
Cc: iab@isi.edu, iesg@NRI.Reston.VA.US, ietf@isi.edu,
internet-society-trustees:;@IETF.NRI.Reston.VA.US,
internet-society-members:;@IETF.NRI.Reston.VA.US,
isoc-interest@relay.sgi.com
Subject: Request for Recommendations
Date: Wed, 12 Aug 92 21:46:48 -0400
From: vcerf@NRI.Reston.VA.US
Message-Id: <9208122146.aa10744@IETF.NRI.Reston.VA.US>
To: IETF Chairman
CC: IAB, IETF, IRTF, ISOC Trustees, and ISOC Members
From: President, Internet Society
Subject: Request for Recommendations on IAB/IETF/IRTF/ISOC Procedures
Introduction
In June, 1992, The Internet Society Board of Trustees received a
proposed charter for an Internet Architecture Board to be created
within the Internet Society. The new Internet Architecture Board was
envisioned to be made up of the then current members of the Internet
Activities Board. The functions of the IETF and the IRTF would be
brought under the umbrella of the Internet Society as elements of the
newly chartered Internet Architecture Board. The Trustees voted to
accept the proposed charter, which is contained in RFC 1358.
In the period between June and July, a great deal of discussion
occurred both by email and at the IETF meeting in Boston about the
Crocker [Page 6]
RFC 1396 Poised Report January 1993
procedures by which the IAB, IETF, and IRTF should work together
under the general umbrella of the Internet Society. Among the many
points raised, three seemed particularly important to consider:
1. Procedures for making appointments to the Internet Architecture
Board.
2. Procedures for resolving disagreements among the IETF, IESG,
and IAB in matters pertaining to the Internet Standards.
3. Methods for ensuring that for any particular Internet Standard,
procedures have been followed satisfactorily by all parties
so that everyone with an interest has had a fair opportunity
to be heard.
Effective discussion of these issues requires the availability of an
open venue, an opportunity to comment by email, and the possibility
of conducting face-to-face meetings. Recognizing this, the IAB has
recommended that the most effective means available to the Internet
Community for gathering recommendations for refining our existing
procedures is to request that an IETF Working Group be formed.
I am pleased to make this request of the IETF to create such a WG.
It would be extremely helpful to have recommendations on the three
topics listed above by the first of December, 1992, in time to be
presented at the Internet Society Board of Trustees meeting scheduled
for December 10th.
With respect to point 1 above, the current procedures for making IAB
appointments are contained in the IAB charter (RFC 1358). RFC 1310
contains a description of the current process by which Internet
Standards are achieved. Points (2) and (3) could be discussed in the
context of the present procedures described in RFC 1310, which
recognizes a number of open issues in an appendix.
The Internet Society Trustees hope that members of the IAB, IRTF,
IESG, and IETF, as well as general members of the Internet Society,
will take the time to share their thoughts in the proposed working
group forum. It is vital to the future of the Internet that these key
groups work together well. Certainly, the Internet Society Trustees
are committed to doing everything possible to facilitate effective
procedures and to preserve and enhance the special spirit of the
Internet Community which has created and maintains the health and
growth of the global Internet.
Vint Cerf
President
Internet Society
Crocker [Page 7]
RFC 1396 Poised Report January 1993
Process for Organization of Internet Standards (poised)
Charter
Chair(s):
Steve Crocker <crocker@tis.com>
Mailing lists:
General Discussion:poised@nri.reston.va.us
To Subscribe: poised-request@nri.reston.va.us
Archive: nri.reston.va.us:~/poised/current
Description of Working Group:
The goal of this working group is to examine the Internet
standards process and the responsibilities of the IAB, with
attention to the relationship between the IAB and IETF/IESG.
The need for this working group was suggested during
discussions at the July 1992 IETF. This led to a request
from the Internet Society president to form such a working
group.
The WG will consider the following matters:
1. Procedures for making appointments to the Internet
Architecture Board.
2. Procedures for resolving disagreements among IETF, IESG
and IAB in matters pertaining to the Internet Standards.
3. Methods for assuring that for any particular Internet
Standard, procedures have been followed satisfactorily by
all parties so that everyone with an interest has had a
fair opportunity to be heard.
The WG will begin with a review of the procedures for making
IAB appointments as documented in RFC 1358 and a review of
the standards-making process documented in RFC 1310.
The WG has a goal of issuing a final report in time for IESG
consideration and publication as an RFC before the
ISOC Board of Trustee's meeting in December 1992. Given the
compressed timescale, the WG will conduct most of its
deliberations by electronic mail on the POISED WG mailing
list. There will also be a preliminary report and
discussions at the November 1992 IETF meeting in
Crocker [Page 8]
RFC 1396 Poised Report January 1993
Washington, DC.
This will be a normal IETF WG, i.e. the mailing list and all
discussions will be completely open.
Goals and Milestones:
Sep 92 Gather initial set of issues and write a preliminary
report.
Oct 92 Post as an Internet Draft the initial recommendations to
the ISOC Board.
Nov 92 Open discussion and presentation of the work of the POISED
Working Group at Washington D.C. IETF meeting.
Dec 92 Submit the recommendations document to the IESG for posting
as an Informational RFC. This document will be
subsequently transmitted to the ISOC Board.
Crocker [Page 9]
RFC 1396 Poised Report January 1993
Security Considerations
Security issues are not discussed in this memo.
Author's Address
Stephen D. Crocker
Trusted Information Systems, Inc.
3060 Washington Road
Glenwood, MD 21738
Phone: (301) 854-6889
Email: crocker@TIS.COM
Crocker [Page 10]