neal
---------- Forwarded message ----------
Date: Tue, 23 Jun 1998 18:39:57 -0700 (PDT)
From: Tom Anderson <tom@emigrant>
To: acollins@cs, alec@cs, becker@cs, cardwell@cs, geigudr@cs, hoffman@cs,
karlin@cs, nitin@cs, savage@cs, tkl@cs, tom@cs, vahdat@cs, voelker@cs
Cc: bershad@cs, gaetano@cs, lazowska@cs, levy@cs, zahorjan@cs
Subject: goals of the retreat
Anna suggested that I outline the goals for the retreat talks,
particularly for the web cache people. I've already done this
verbally for the Detour folks.
The principal goal is that this provides an opportunity for you
to get feedback on your work. Realize that there will be a lot
of very smart people in the audience; if you do it right,
you'll get half an hour of 30 smart people engaging their brains
trying to think how you could do your part of the project better.
So remember to:
a. motivate your work (there will be a spectrum of backgrounds
from industry to profs to students, so don't assume everyone already
knows everything about your work)
b. provide enough background for the key design choices and results
so that everyone can understand (and disagree with!) them. You want
people to be able to say, you're on the right track, or you
aren't, which parts of your work are really important, and
which parts they think you should skip, here's some related
work you should know about, you missed some key fatal flaw
in your approach and/or measurements, etc.
Note that you should keep it high level -- you don't need
these people to do a design review of your data structures;
you do need to get them engaged in the intellectual problem
you are tackling.
c. outline future directions and priorities, again with a goal
towards getting useful feedback about those priorities,
and maybe get some ideas about different directions to take.
To repeat, the key goal is to get feedback. Feedback can only
help the quality of your work -- none of us have all the answers.
Realize that you'll be able to take credit for all the good ideas
you can get people to come up with on your behalf ;-)
You'll also get a lot of benefit from the practice of preparing
and giving talks on your work to outsiders, but particularly
at this stage of the project, polish is less important than ideas
and content.
Finally let me point out that most of the time at the retreat you'll
be in the audience, where your job will be to engage your brain,
think about the problem from the speaker's point of view, and give them
constructive feedback on their part of the problem. What's fun
about a retreat like this is the chance to kibitz, to see the connections
between different pieces of work, and to brainstorm new ideas.
tom