braden@VENERA.ISI.EDU (07/22/89)
At its July meeting, the Internet Activities Board made some modifications
in the management structure for the Internet. An outline of the new
structure is given below. These changes resulted from an appreciation
of our successes, especially as reflected in the growth and vigor of the
IETF, and in rueful acknowledgment of our failures (which I will not
enumerate). Many on these lists are concerned with making the Internet
architecture work in the real world. We believe the new structure will
provide a more effective organization and better leadership, to give us
all more leverage in solving present issues and evolving for the future.
Vint Cerf is unable to attend the IETF meeting next week, but he
will be present via video tape to elaborate on these matters, and
Dave Clark will be present in the flesh to answer questions.
Bob Braden, for the IAB
The Internet Activities Board and its Task Forces
July 1989
A. INTRODUCTION
The Internet, begun as a DARPA-funded experiment in internetworking, has
become the prototype national research network, forming a major fraction
of the networking infrastructure for the research community and,
increasingly, the government and business sectors. The TCP/IP technology
has been propagated even more widely into business and industry.
Networking and internetting are now entering a new era with the emergence
of extremely high speed packet switching technology and with the steadily
growing availability of OSI software. The decade of the 1990's is likely
to prove as revolutionary in network technology development as the
ARPANET was almost twenty years ago.
The Internet Activities Board (IAB) functions as a Board of Directors for
the Internet. The IAB itself set technical policy and standards for the
Internet protocols and architecture, and reviews the work of its two
major task forces: the Internet Engineering Task Force (IETF) and the
Internet Research Task Force (IRTF).
Representatives of the government agencies that sponsor important
segments of Internet research and infrastructure are members of the
Federal Research Internet Coordinating Committee (FRICC). The FRICC
is concerned with the present Internet and the future National
Research Network. The IAB and the FRICC work closely together.
B. THE IAB
The IAB is an independent committee whose members generally share a
long-term involvement in, and responsibility for, Internet design,
engineering, and management. The IAB members were chosen for the
specific roles they play ( e.g., the chairs of the IETF and IRTF), to
represent major groups in the Internet community (e.g, national
network, vendor, government, and international groups), and for specific
areas of expertise (e.g., security). They are deeply committed to
making the Internet function effectively and to evolving the Internet
to meet a large scale, high speed future. All IAB members are required
to have at least one other major role in the Internet community in
addition to their IAB membership.
The IAB chair serves for a period of two years. The current IAB chair
is Vint Cerf of NRI, who has been called the "father of the Internet".
In 1974, he co-authored with Bob Kahn the original paper on the
Internet architecture. As a program manager at DARPA he directed the
research effort that resulted in development of the TCP/IP protocol
suite, and directed the construction of the prototype of today's
Internet.
The IAB membership is currently as follows:
Vint Cerf - Chairman
Dave Clark - IRTF Chairman
Phill Gross - IETF Chairman
Jon Postel - RFC Editor
Bob Braden - Executive Director
Hans-Werner Braun - NSFNET Liaison
Barry Leiner - CCIRN Liaison
Dan Lynch - Vendor Liaison
Steve Kent - Security
The IAB typically meets four times a year, to supplement its work by
electronic mail. Whenever possible, and at least once a year, the IAB
plans to hold its meetings in conjunction with an IETF meeting.
In more detail, the IAB performs the following functions:
* Sets Internet Standards
* Manages the RFC publication process
* Reviews the operation of the IETF and IRTF
* Performs strategic planning for the Internet, identifying
long-range problems and opportunities
* Acts as external policy representative for the Internet
community.
* Resolves technical issues which cannot be treated within
the IETF or IRTF frameworks.
C. The Internet Engineering Task Force
The Internet has grown to encompass a large number of widely
geographically dispersed networks in academic and research communities.
It now provides an infrastructure for a broad community of interest.
Moreover, the family of Internet protocols and system components has
moved from experimental to commercial development. To help coordinate
the operation and management of the Internet, the IAB established the
Internet Engineering Task Force with the charter to:
1. Act as a clearinghouse to promote the exchange of information
within the Internet community. This community includes Internet
software and hardware developers, Internet operators and Internet
research and development groups.
2. Identify pressing and relevant short- to mid-range problem areas
and convene Working Groups to develop solutions. Working Groups
might deal with a wide range of Internet issues, such as operational
management problems or protocol enhancements that would improve
Internet performance or extend the range of application of the
architecture.
3. Report Working Group results and recommendation to the IAB and
to the Internet community at large.
The Internet Engineering Task Force is a large open community of network
designers, operators, vendors, and researchers concerned with the Internet
and the Internet protocol suite. The work of the IETF is governed by
a board known as the Internet Engineering Steering Group, or IESG. The
chairman of the IETF and of the IESG is Phill Gross of NRI.
The work of the IETF is performed by subcommittees known as Working
Groups. There are currently more than 20 of these. Working Groups tend
to have a narrow focus and a lifetime bounded by completion of a specific
task, although there are exceptions. The IETF is generally the major
source of proposed protocol standards, for final approval by the IAB.
The IAB has delegated to the IETF general responsibility for making the
Internet work, and, to the IESG in particular, responsibility for the
resolution of all short- and mid-range protocol and architectural issues
required to make the Internet function effectively.
The members of the IESG are, in addition to the IETF chairman, Area
Technical Directors (ATDs). Each ATD has primary responsibility for
one area of Internet engineering activity, and hence for a subset of
the Working Groups. The ATDs perform a technical management function
that is vital to the function and development of the Internet; they
are selected not only for their technical expertise but also for
their managerial skills and judgment.
For more information on the Internet Engineering Task Force, on attending
meetings and proposing or joining Working Groups, contact Phill Gross,
Corporation for National Research Initiatives (gross@SCCGATE.SCC.COM
pgross@NRI.RESTON.VA.US, 703-620-8990).
D. The Internet Research Task Force
The IAB is concerned with promoting research in networking, to develop
the technology that the IETF will need; this is the job of the Internet
Research Task Force, or IRTF.
In the area of network protocols, the distinction between research and
engineering is not always clear, so there will sometimes be overlap
between activities of the IETF and the IRTF. There is, in fact,
considerable overlap in membership between the two. This overlap is
regarded as vital for cross-fertilization and technology transfer. In
general, the distinction between research and engineering is one of
viewpoint and sometimes (but not always) time-frame. The IRTF is
generally more concerned with understanding than with products or
standard protocols, although specific experimental protocols may have to
be developed in order to gain understanding.
The IRTF is a community of network researchers, generally with an
Internet focus. The work of the IRTF is governed by a board known as
the Internet Research Steering Group or IRSG. The chair of the IRTF
and of the IRSG is Dave Clark of the MIT Laboratory for Computer
Science. Clark played a seminal role in development of the Internet
architecture, and for nearly 10 years guided the Internet development
and deployment as chair of the IAB and as "the Internet Architect".
As befits their research goals, the IRTF and IRSG are organized in a
much less formal manner than are their engineering counterparts. The
IRTF is organized into a number of Research Groups (RG's), and the
chairs of these RG's sit on the IRSG. These groups typically have 10
to 20 members, and each covers a broad area of research, pursuing
specific topics, determined at least in part by the interests of the
members as well as recommendations from the IAB and IETF.
===============================================================================joaquin@SRI-NIC.ARPA (07/25/89)
Dear IABers: Shakeups are always interesting, specially when your are not part of the bunch being shaken. I have a few questions: 1. Can we expect the IRTF to meet in a similar fashion as the IETF (maybe not as often, please)??? .....There is no question that the success of the IETF over the past couple of years is due to Phil's great job as a chair. However, I also think it is due in part to the fact that multiple working groups meet at the same time, which helps tremendously in fostering the creation of new working groups when they are needed, and in disseminating ideas. In addition, it is much easier for some of us to find funding to attend larger (fewer) meetings. 2. I think that part of the IRTF goals should be to serve as a clearinghouse of research ideas in internetworking. This function will be SOOOO important as we migrate to the future NREN!....and I don't see any better forum that can be used to foster internetworking research (the usual conferences cannot have a very clear focus over the years). 3. Should we have an Annual IAB Workshop (i.e., a joint meeting of the IRTF and IETF) once a year to bring together the latest accomplishements in our community regarding research results and engineering?......of course, with the best presentations during the year from both groups. 4. I am confused about the IAB structure (not that it matters to anyone, of course). Bob Braden explained what IRTF and IETF were meant to be, but what is the role of the other members? 5. Why is "Security" not part of IRTF or IETF concerns (or both)? why not have "Privacy" or "Billing" there too? 6. Why do we have liasons for the NSFNET and vendors, but not for the academia at large, standards groups, research centers, or the government (i.e., FRICC and DCA)? This becomes particularly important when one looks at the IAB functions!!! At a time when the government is pushing for OSI standards, it is hard for me to see why vendors and NSFNET have direct participation in the IAB, while organisms like NIST and ANSI (for example) don't. I'd really like to have some folks from the standars groups collaborating with the Internet very closely, so that developments in the IETF and IRTF can impact the standarization process "from within" (e.g., X500 and the handling of distributed queries) and viceversa. 7. What is meant by " [The IAB] Resolves technical issues which cannot be treated within the IETF or IRTF frameworks."????? 8. What is meant by "[The IAB] Acts as external policy representative for the Internet community."???? who is the target audience of the IAB? 9. What is the detailed structure of the IETF and IRTF?????? Which are the ATDs in IETF? which are the RGs of the IRTF? inquiring minds want to know! Looking forward to tomorrow. Confusedly yours, JJ
CERF@A.ISI.EDU (07/26/89)
1. IRTF will meet on a schedule to be determined by Dave Clark. He is just as interested as you are in reducing the number of meetings we all have to attend. 2. IRTF will have access to the same clearinghouse mechanisms that the rest of the IAB system has (RFCs, Internet-Drafts Directory, etc.) 3. Annual IAB meeting... In some sense, the Interop meetings seem to fill a part of that need. And the IETF meeting minutes are reported quarterly. I'm not sure we need another meeting... 4. The IAB itself is responsible for technical policy, for setting the objectives for IRTF and IETF (which are reduced to practice by the IRSG and IESG), dealing with external liaison (e.g. with FRICC, CCIRN, RARE, NIST, ...), breaking logjams impeding progress on standards for the Internet, and so on. The members of IAB proper have other roles in the Internet community, as should be obvious from the list that Braden provided. 5. Security has become such a visible issue for the Internet (see GAO report and recent hearings on the Hill) that I wanted top-level attention to policy matters. There may, indeed, continue to be an IRTF group focused on security/privacy chaired by Steve Kent - but this remains to be seen as Dave Clark works out the structure of IRTF. 6. IAB has liaison with FRICC in several ways. I am generally invited to FRICC meetings except when they are for gov't only; Phill Gross runs the special FRICC engineering planning group; FRICC is invited to IAB meetings ex officio. IAB has had NIST visitors and will be working out specific IETF and possibly IRTF working group/research group joint efforts with them. Network management and white pages registrations are examples of areas where we expect some specific work. It is not clear that we need a NIST representative on the IAB if we can work out specific working group interactions. Collaboration with ANSI and IEEE is under consideration but, again, I suspect that we may find this working better through specific IETF working groups or IRTF research groups because ANSI and IEEE are not easily representable by a single individual. Rather, they are represented by each working group on particular topics, which matches our IRTF and IETF structure. 7. IAB will resolve standards matters when they cannot be resolved within the IETF or IRTF. 8. IAB will represent the Internet community and its standards, when necessary, before the Defense Protocol Standards Steering Group, ANSI, NIST, CCIRN, FRICC, RARE, and so on. 9. IETF and IRTF structures are being worked out in detail and will be reported by Phill Gross and Dave Clark as soon as these details are settled. Vint Cerf Chairman IAB
patterso@hardees.rutgers.edu (Ross Patterson) (07/27/89)
Bob, The IAB's position in the pre-NSF Internet was pretty clear, given DARPA's and DCA's position on policy matters, and the fact that they funded the whole kit and caboodle. With the advent of the NSFNet Backbone, and soon the Federal Research Internet, things seem a little cloudy. From where does the IAB draw its authority in this modern world? Is it still formally authorized by the funing agencies, or has it evolved into an independent group of network gurus, leading rather than commanding? In short, is there anything special about the IAB, or could someone form a competeing group and offer their own view of the world now? Ross Patterson Rutgers University P.S. No, I haven't got any plans. Just curious, thanks. RAP
CERF@A.ISI.EDU (07/29/89)
Ross, Bob may already have responded (I persist in going through my mail sequentially), but here is my two cents. The IAB works very closely with the FRICC which is the government group which sponsors most of the Internet backbone and deals with international connections to the Internet. The government supports the IAB and its subsidiary groups, IETF and IESG, and many of the members of these groups continue to be supported individually by government research grants and cooperative agreements. Things haven't really changed all that much from the days with DARPA sponsored the whole thing - except that in place of DARPA, the FRICC acts to set government policy and to fund the R&D and infrastructure work necessary to keep the Internet healthy and relevant. Vint Cerf