Initial Analysis of the IESG Review Process

[Eric Rescorla <ekr@rtfm.com>]

I've spent some of the past week analyzing the external face
of the IESG review process with a focus on measuring the
time to approval and its predictors. The results are
written up in a paper which can be found at:

	http://www.rtfm.com/iesg-review.pdf

I figured I'd send the I* an early copy before sending it to
the problem-statement list.

ABSTRACT
We report on an analysis of the IESG Review Process.  We studied all
drafts Last Called in the period of August 1998-March 2003 in order to
characterize what factors contribute to the outcome and duration of the
process. We find substantial variation in the duration of the review
process both between drafts (sd = 149 days) and between areas
(hazard ratio = 2-3). Finally, we find that the time to approval is
loosely correlated to the number of revisions that a document goes
through before revision but we did not find sufficient evidence to
conclude that the magnitude of changes between versions was correlated
to the time to approval when the number of revisions is controlled for.

Comments welcome.
-Ekr