First impressions

[Brian Kantor]

I remarked earlier this year that perhaps the Usenix conferences were
being held too often; the evidence at the time was the list of papers to
be presented at the Baltimore conference.  I believe that it simply
takes more than a few months to research and write a quality paper, and
if the conferences are held too often then the supply of good papers
will be drained too soon leaving few good ones for subsequent
conferences.

The San Diego conference people remarked that they'd have liked to have
had two paper tracks running at the conference but that they didn't get
enough good ones to do that.  Others have remarked that the San Diego
conference probably drained off the good papers for at least a year.
(The San Diego conference wasn't unique in this; I'm just using it as
an example because I live here.)

To worsen the situation, there are also special-interest symposia and
mini-conferences held throughout the year.

Twice-yearly Usenix conferences are traditional; I don't seriously
imagine that will be changed soon.  However, we might be able to find
some way of ensuring more worthwhile technical sessions and more good
papers.

One suggestion is to have more panel discussions; I've usually found them
to be more lively and they are often more informative than someone just
presenting research results.  Another might be to have sessions of
repeat papers, where papers selected from those presented at
Usenix-sponsored symposia and mini-conferences would be presented again
to the wider audience of the general conference.

Or perhaps there should be a series of half-day tutorials presented
throughout the week.

Waddayouthink?
	- Brian