Getting the Most Out of Your Workshop

The short version is:

* Good ideas are born out of bad ideas.  Tell us your bad ideas.
* Don't scare people off by being too harsh - we need their bad ideas.
* Turn bad ideas into good ideas by pointing out holes - and fixing them.
* Avoid talking about stuff that can be done over email.

The Linux File Systems Workshop is a rare opportunity for some really
creative and exciting discussion about file system ideas.  We have a
chance to talk to each other real-time, in person, with whiteboards
and hand-waving and everything - much higher bandwidth and lower
latency than mailing lists.  But if we use the workshop to have a
mailing list-style discussion, we might as well stay home and save
ourselves the hassle of going through airport security.

How can we get the most out of this workshop?  First, our goal is to
brainstorm and discuss interesting new file system ideas suitable for
the next decade's file storage needs.  We'll start with some data on
hardware trends and existing file system techniques, identify the
problems we have to fix, brainstorm new ideas, and winnow out the good
ideas.  At the end of the workshop, we should have 3-5 promising
approaches that various people are interested in experimenting with.

The key to coming up with these 3-5 good ideas is to propose 500 bad
ideas, 1% of which turn out to be good ideas in disguise.  Bad ideas
are key to good ideas.  Sometimes your bad idea has a problem which
you don't know how to solve but someone else will instantly see the
solution that turns it into a good idea.  Sometimes figuring out the
reasons why a bad idea is bad will help you understand what a good
idea should look like.  And sometimes sheer random luck (such as
someone sneezing and misunderstanding what you said) will turn a bad
idea into a good idea.

Bad ideas can only turn into good ideas if people aren't afraid to
tell us about their bad ideas.  This is the hardest part about
actually coming up with new ideas - fear of looking stupid.  We have
to work hard to not make people feel stupid or they won't tell us
their bad ideas.  Tips for making this work better:

* Never say "Your idea is stupid."  You have to say _why_ you think
  someone else's idea is stupid.  When you are specific about what's
  wrong with it, it's easier to turn it into a smart idea.

* Feel free to say "I have a stupid idea."  Then tell us your stupid
  idea and why you think it's stupid.  What's the worst that could
  happen - we agree with you?

Turning bad ideas into good ideas has two parts.  One, pointing out
the holes in someone's bad idea.  Two, coming up with solutions to
those holes.  If you are only doing the first part, we'll never come
up with any good ideas.

We won't have any formal limitations on what we talk about at the
workshop, but we would prefer talking about wild departures from the
status quo rather than incremental improvements to existing code.
This is simply because it's much easier to talk about incremental code
changes over email than wild new ideas.  We have whiteboards, let's
make sure we're using them.

Finally, have fun!  We're awesome amazing suave file systems
developers, out on the town in wild, crazy Portland, Oregon!!!  Party
on!!!