I respect both Pete and Paul's position here, but I believe this
frustration is endemic to our efforts rather than specific to how the
working group meeting agendas are set. I also believe that the
frustration is worth the result.
One of the things which sets the IETF apart from other efforts to
produce standards is the breadth of perspective it brings to the
problems which it chooses to tackle. Where it is fairly easy to get a
group of like-minded people together to tackle a specific network
problem, it is much harder to get a group of people together who can
see how the proposed solution will impact the other parts of the
network. From my perspective, one of the chief values of the IETF
face-to-face meetings is the opportunity they present to have folks
from different parts of the Internet engineering community provide
input into the solutions which have been proposed by the communities
of interest which the working groups represent. Sometimes that
results in those communities of interest growing, as individuals
recognize their need to participate. Sometimes a new perspective
cogently expressed at a face to face meeting is all it takes to move a
working group in a new direction.
In either case, without the participation of the larger IETF community
in the working group meetings, those perspectives do not get expressed
to the working group at early stages of the process. That means much
more work for the IESG in managing the inter-area issues when drafts
are ready to move forward. It can also mean delay as things which
might have been caught early have to be unraveled after other elements
of the design depend on them.
Whether the agenda of a working group takes the shape of draft
presentations or issue lists, I believe it is important for working
groups to be open to new voices during the IETF meetings, for they
don't have that many other opportunities to hear them. This certainly
engenders frustration when the new voices rehash old problems, and it
requires skill on the part of the presenters and chair to keep the
resurgence of old problems from eating all the time. It also competes
with the use of the meetings to handle pressing technical issues which
benefit from focused face-to-face work. Balancing those competing
interests, again, requires work and skill on the part of the chair.
As thankless and unsung as that work often is, it is worth it. We may
not get perfect standards from the process, but we do get engineering
solutions which do a pretty good job of balancing the needs of the
different parts of the network infrastructure.
I believe we can all agree that we need better ways of scaling the
input to IETF working groups. I hope we can also agree that we need
them because we need to retain the skills and perspective of our
participants, not simply because some group sizes are unwieldy or some
meeting resources constrained.
At 11:20 AM -0600 12/19/00, Pete Resnick wrote:
How about it? Other chairs wish to join me in this mission?
Yup. As someone who chaired a meeting where we had three
presentations on three drafts that had already been on the list, and
the discussion was all around topics that could have been brought to
the list, I share your frustration.
--Paul Hoffman, Director