At 06:40 AM 3/29/2001, Donald E. Eastlake 3rd wrote:
>should. Should be 2 years, and we tend to run no better than 1. That
>constrains choice and that either increases price or decreases convenience.
While it contrains choice, it does not necessarily increase price or
decrease convenience. Many large conventions reserve space 5 or 10
years out. When you get down to 1 year out, facilities that are not
yet reserved are generally desperate and you can frequently negotiate
good deals.
What you say sounds entirely logical. Unfortunately reality intrudes on
the logic: we have been burned by the "too little choice" problem more
than once.
My reference to needing a 2 year advance did not spring from my own
theories, it was discussed previously (a couple of years ago, or more) and
was essentially a consensus view.
>Another is that the host is usually not skilled at the relevant technical
>details for a conference.
These hosts are technical companies. I'm not involved with the behind
the scenes part of the IETF meeting network.
Thank you for underscoring the problem: It exactly demonstrates the
failure to appreciate the convention-style networking is quite different
from building products or running services for long-term use. It is a
specialty. Added to that is that the IETF places significant strain on
many products.
>Re-use reduces learning curve and that reduces problems (and cost).
You can also get stuck in a rut where people are afraid of change.
Worrying about being stuck in a rut is entirely appropriate for an activity
that benefit from innovation. Our meetings have had the same style for
roughly 11 years. Our biggest change has been to move the plenary from the
end of the meeting week to Wednesday evening.
Your comment again exactly underscores the problem: We need to treat these
3-times-a-year, forever meetings as a utility, not an experiment.
>Three factors are more than enough the try to optimize.
>The rest need to be ignored.
I don't agree. I consider variety of location to be a viture in
itself.
Again I'll apologize for the misstatement. I also believe that changing
continents is appropriate.
As to having the specific cities vary, the problem is finding an argument
that justifies it for the week-long, working IETF event, especially given
how intense the working activities are.
d/
----------
Dave Crocker <mailto:dcrocker(_at_)brandenburg(_dot_)com>
Brandenburg InternetWorking <http://www.brandenburg.com>
tel: +1.408.246.8253; fax: +1.408.273.6464