Internet Info 1997 December/Internet_Info_CD-ROM_Walnut_Creek_December_1997.iso/ietf/93nov/iia-minutes-93nov.txt

Text File | 1994-02-08 | 10KB | 255 lines

CURRENT_MEETING_REPORT_

Reported by Luc Boulianne/McGill University

Minutes of the Integrated Information Architecture BOF (IIA)


Introduction - Phill Gross

Phill Gross took a few moments to introduce the reasons behind the
creation of this group:


     A year or so ago, the IESG was hit all at once with the
     creation of a large set of working groups in the general areas
     of network information discovery, retrieval, and user
     information handling.  It became apparent that many of these
     working groups were related, or should be.  There appears to be
     two ways in which the IETF operates:  top-down and bottom-up.

      1. Top-down:  (or pro-active) such as the IPNG.
      2. Bottom-up:  the usual way things are done.  Usually the
         `right' things come out of this approach.  And yet, it
         would appear that sometimes, the area directors are still
         needed for pro-active planning.

     When the working groups were chartered, they were made jointly
     part of an ``Internet Information Architecture'' (IIA)
     activity.  The expectation was that these groups would work
     together, as well as on their own primary foci, and would do so
     under the joint supervision of the User Services and
     Applications Areas.


Phill suggested that the area directors now write a new overview of the
IIA, providing a framework only.  Because of the importance of this
issue, Phill suggested that the IESG request a working group be charged
to create an IIA architecture framework definition citing as an example:
IPNG (Allison Mankin and Scott Bradner).


Summary of the Issues - John Klensin

Working groups were formed, work was done and documents began to appear.
Some concluded that there was a lack of coordination among the working
groups, but that the current meeting is an effort to reconcile this
lack.

IIA is comprised of several working groups, overlapping work with an
overlapping cast of characters.  The working groups should be
coordinated technically, but it often appears that they are not.

Characteristics of the IIA working groups and their membership:

   o Very expert in several types of work.

   o There is, however, some evidence that, in protocol design areas,
     they may be moving out of their depth or succumbing to the Not
     Invented Here syndrome.

   o There are interactions with the ``real world'' that one must
     consider, e.g., librarians and other information specialists,
     external standards.

   o Most of the groups seem to have nearly the same membership, with
     topics and issues flowing back and forth between them.


Finally, there were these questions to ask the group:

   o Is the current model as efficient as possible?  If it is not, what
     can be done to improve things?

   o Is there a structural way of going about this?

   o What about working group functional boundaries?

   o What is the definition of a functional boundary?

   o What can be done to not break anything that is now working, while
     we try to increase efficiency and productivity?


A suggestion was made by the group that multiple solutions to this
problem (i.e.  working groups) which have trivial differences, should be
merged into a homogeneous solution.  This would help to avoid diluting
the merged efforts.


User Services Area - Joyce K. Reynolds

Joyce believes that it is important to make sure there is communication
between areas.  A meeting of the User Services Area Council (USAC) was
held on Tuesday evening.  USAC observed that developers and users are
well represented in these gatherings, but operators (information
providers) are not.  The following items are lacking:


   o Tools for maintaining information
   o Support tools
   o How does one share information
   o An adequate level of cooperation
   o An adequate level of operational effectiveness


Working Group Chair Input

   o It was suggested that a small group of closely interacting people
     could keep an eye on working groups in different IETF areas.  For
     example, ``multicasting'' could be used for information
     directories.  A mechanism is needed to promote this inter-area
     information diffusion.

     Possible functions of this group of inter-working group
     communicators might be to monitor the minutes of various working
     groups or to bring up ideas, when appropriate, in other working
     groups.

   o Some resistance towards inter-working group communicators was felt.
     It was pointed out that the working group is the only place for
     this kind of discussion.  The working group is there a) to
     Review/Refine and b) for the Community Buy-in.

   o The Applications Area Directorate (APPLES), and its mailing lists,
     should attempt to provide a pro-active exchange of information.

   o There is an overlap among working groups and this might be
     wasteful.

   o There appears to be a lack of coordination within the IETF. Yet,
     this work requires that the IETF do research that breaks new
     ground.  This means that some lack of focus may be required to get
     the best ideas out.

   o The example of Data Elements (IAFA) was brought up to support the
     need for communication among the working groups.  Data elements are
     required by many working groups, and it is a very difficult problem
     in itself.

   o It was suggested that working group chairs get together at the
     start of each IETF and discuss the ``Hot Topics.''  The chair would
     then bring these to the attention of their working groups.


General Discussion

   o There was some concern that extra (needless) bureaucratic structure
     might be created.

   o Multiple groups developing different solutions to a common problem
     is not a bad thing.

   o Sometimes the expertise to solve a problem is in another group, and
     both its availability and the knowledge itself are unknown to the
     group that needs it.

   o Possibly, vision documents could be drafted to help guide and
     measure work done.

   o It was felt that a ``Hot Topics'' discussion might not work, as
     there is so little time as it is.  Many felt it would be worth the
     time.

   o Possibly, Five Year Planning documents could help guide working
     groups.  Some thought that the IAB does this to some extent.  Some
     felt that this would inhibit better solutions.

   o It was suggested that working group tutorials be set up to bring
     new members up to speed, including those from other working groups
     who want to take advantage of local expertise.  Possibly, the old
     tradition of two minute introductions to all working groups could
     be restored to the opening plenary.

   o It was pointed out that Area Reports are written, but few people
     take the time to read them.  It was also pointed out that these
     Area Reports do not point out the ``Hot Spots''; the IESG might be
     overloaded already.

   o A road map showing the interrelation of the various groups might
     prove to be useful.

   o The IETF is getting larger and larger.  Communications is becoming
     a problem.


Conclusion

Using the APPLES mailing list for further discussion, it should be
determined 1) how to improve communications, and 2) what structure might
work to propagate this newly acquired information.


Attendees

Steve Alexander          [e-mail]
Harald Alvestrand        [e-mail]
Jules Aronson            [e-mail]
Luc Boulianne            [e-mail]
James Conklin            [e-mail]
John Curran              [e-mail]
Alan Emtage              [e-mail]
Urs Eppenberger          [e-mail]
Sallie Fellows           [e-mail]
Jill Foster              [e-mail]
Paul Francis             [e-mail]
Ned Freed                [e-mail]
Kevin Gamiel             [e-mail]
Tony Genovese            [e-mail]
Judith Grass             [e-mail]
Terry Gray               [e-mail]
Phillip Gross            [e-mail]
Martyne Hallgren         [e-mail]
Deborah Hamilton         [e-mail]
Alf Hansen               [e-mail]
Roland Hedberg           [e-mail]
Marco Hernandez          [e-mail]
Russ Hobby               [e-mail]
Jeroen Houttuin          [e-mail]
Tim Howes                [e-mail]
Richard Huber            [e-mail]
Christian Huitema        [e-mail]
Erik Huizer              [e-mail]
Steve Kille              [e-mail]
John Klensin             [e-mail]
Jim Knowles              [e-mail]
Barry Leiner             [e-mail]
Ben Levy                 [e-mail]
Clifford Lynch           [e-mail]
Glenn Mansfield          [e-mail]
April Marine             [e-mail]
Larry Masinter           [e-mail]
Mitra                    [e-mail]
Keith Moore              [e-mail]
Robert Moskowitz         [e-mail]
Chris Newman             [e-mail]
Masataka Ohta            [e-mail]
Lars-Gunnar Olsson       [e-mail]
Scott Paisley            [e-mail]
Rakesh Patel             [e-mail]
Pete Percival            [e-mail]
Marsha Perrott           [e-mail]
Karen Petraska-Veum      [e-mail]
Cecilia Preston          [e-mail]
Joyce K. Reynolds        [e-mail]
Srinivas Sataluri        [e-mail]
Richard Schmalgemeier    [e-mail]
Rickard Schoultz         [e-mail]
Henry Sinnreich          [e-mail]
Mark Smith               [e-mail]
Karen Sollins            [e-mail]
Chris Weider             [e-mail]
Jackie Wilson            [e-mail]