Network Working Group E. Huizer
Request for Comments: 2031 SURFnet ExpertiseCentrum bv
Category: Informational October 1996
IETF-ISOC relationship
Status of this Memo
This memo provides information for the Internet community. This memo
does not specify an Internet standard of any kind. Distribution of
this memo is unlimited.
Abstract
This memo summarises the issues on IETF - ISOC relationships as the
have been discussed by the Poised Working Group. The purpose of the
document is to gauge consensus on these issues. And to allow further
discussions where necessary.
Introduction
The Internet Engineering Task Force (IETF) is the body that is
responsible for the development and maintenance of the Internet
Standards. Traditionally the IETF is a volunteer organization. The
driving force is dedicated high quality engineers from all over the
world. In a structure of working groups these engineers exchange
ideas and experience, and through discussion (both by e-mail and face
to face) they strive to get rough consensus. The engineers then work
on building running code to put the consensus to the test and evolve
it into an Internet Standard.
The growth of the Internet has also led to a growth of the IETF. More
and more people, organizations and companies rely on Internet
Standards. The growth of responsibility as well as amount of
participants has forced the IETF to more and more structure its
processes. Non technical issues, such as legal issues, liaison issues
etc., have become an undesirable but a seemingly unavoidable part of
the IETF organization. To address these issues the IETF established
the Poised95 working group. The working group is now trying to
structure and document the IETF processes in such a way as to keep
the maximum flexibility and freedom for the engineers in the IETF to
work in the way the IETF has always been most successful, and to
honour the IETF credo: "Rough consensus and running code".
One of the more obvious recommendations that came out of the Poised
WG was to move all non technical issues that can be moved safely, to
another related organization. The Poised WG finds that the Internet
Huizer Informational [Page 1]
RFC 2031 IETF-ISOC Relationship October 1996
Society (ISOC) is the obvious choice for this task. A straw poll at
the open plenary session of the IETF in december 1995 in Dallas
clearly confirmed this notion.
However, since this is an issue that is crucial to the functioning of
the IETF as a whole it is necessary to get a broad (rather than a
rough) consensus on this issue. At the same time it is necessary to
clearly indicate the extend of the relationship between the IETF and
ISOC. So both the IETF participants and the ISOC board of trustees
get a clear picture on the division of responsibilities.
The details of the Poised WG recommendations on the IETF - ISOC
relationships can be found in the appropriate places in a series of
Poised documents in progress: - The IETF Standards Process - The IETF
organizational structure - The IETF charter - The Nomcom procedures -
The Appeals procedures
The current document is meant to summarize the Poised WG
recommendations in order to gauge the consensus. This document does
not have, and is not intended to get, a formal status. The current
and upcoming working documents of the Poised WG will become the
formal documents. Readers who are interested in the nitty gritty
details are referred to these working documents of the Poised WG.
Main boundary condition
The IETF remains responsible for the development and quality of the
Internet Standards. The ISOC will aid the IETF by facilitating legal
and organizational issues as described below. Apart from the roles
described below, the IETF and ISOC acknowledge that the ISOC has no
influence whatsoever on the Internet Standards process, the Internet
Standards or their technical content.
All subgroups in the IETF and ISOC that have an official role in the
standards process should be either:
- open to anyone (like Working Groups); or
- have a well documented restricted membership in which the
voting members are elected or nominated through an open process.
The latter means that within the IETF the IAB and the IESG need to be
formed through a nomination process that is acceptable to the IETF
community and that gives all IETF participants an equal chance to be
candidate for a position in either of these bodies. For the ISOC this
means that the Board of Trustees should be elected by the ISOC
individual membership, where all individual members have an equal
vote and all individual members have an equal opportunity to stand as
a candidate for a position on the Board of Trustees.
Huizer Informational [Page 2]
RFC 2031 IETF-ISOC Relationship October 1996
ISOC will, like the IETF use public discussion and consensus building
processes when it wants to develop new policies or regulations that
may influence the role of ISOC in the Internet or the Internet
Technical work. ISOC will always put work related to Internet
standards, Internet technical issues or Internet operations up for
discussion in the IETF through the IETF Internet-drafts publication
process.
The legal umbrella
To avoid the fact that the IETF has to construct its own legal
structure to protect the standards and the standards process, ISOC
should provide a legal umbrella. The legal umbrella will at least
cover:
- legal insurance for all IETF officers (IAB, IESG, Nomcom and WG
chairs);
- legal protection of the RFC series of documents; In such a way
that these documents can be freely (i.e. no restrictions
financially or otherwise) distributed, copied etc. but cannot
be altered or misused. And that the right to change the document
lies with the IETF.
- legal protection in case of Intellectual property rights disputes
over Internet Standards or parts thereof.
The standards process role
ISOC will assist the standards process by
- appointing the nomcom chair
- approving IAB candidates
- reviewing and approving the documents that describe the standards
process (i.e. the formal Poised documents).
- acting as the last resort in the appeals process
Security considerations
By involving ISOC into specific parts of the Standards process, the
IETF has no longer absolute control. It can be argued that this is a
breach of security. It is therefore necessary to make sure that the
ISOC involvement is restricted to well defined and understood parts,
at well defined and understood boundary conditions. The Poised WG
attempts to define these, and they are summarised in this document.
There are three alternatives:
- Do nothing and ignore the increasing responsibility and growth; the
risk here is that the IETF either becomes insignificant, or will be
suffocated by US law suits.
Huizer Informational [Page 3]
RFC 2031 IETF-ISOC Relationship October 1996
- The IETF does everything itself; this keeps the IETf in control,
but it would distract enormously from the technical work the IETF
is trying to get done.
- The IETF finds another organization than ISOC to take on the role
described above. But why would another organization be better than
ISOC?
All in all a certain risk seems unavoidable, and a relationship with
ISOC, under the restrictions and boundary conditions as have been
described above, seems more like an opportunity for the IETF than
like a risk.
Acknowledgement and disclaimer
The author is chair of the Poised 95 WG. The author has tried to
summarise e-mail and face to face discussions in the WG. All the good
ideas in this paper are the result of the WG, all the mistakes and
errors are probably due to the author or his lack of command of the
American language as well as the American legal system.
The author is a member of the Internet Society.
Author's Address
Erik Huizer
SURFnet ExpertiseCentrum bv
P.O. Box 19115
3501 DC Utrecht
The Netherlands
Tel: +31 302 305 305
Fax: +31 302 305 329
E-mail: Erik.Huizer@sec.nl
Huizer Informational [Page 4]