Re: Nomcom candidate name announcement alternatives

[Theodore Ts'o <tytso@mit.edu>]

On Fri, Jul 19, 2002 at 10:26:19PM -0400, Thomas Narten wrote:
> There has been a repeated theme that when (say) 50 people are given
> some information, that information is no longer secret and you might
> as well make it public. I do not quite follow the logic here.
> 
> There were some 2001 persons at the Yokohama IETF (plus many more that
> didn't make it at all). 50 is a rather small percentage of that. This
> doesn't seem to me to automatically translate to "everybody finds
> out".

#1) There is a saying that the probability that the a secret will be
blown increases with the square of the number of people who know it.
Hence the saying, "three people can keep a secret, if two of them are
dead".

#2) Numerically 50 is a rather small percentage, but it's a very
significant 50.  How many people really get up and participate in a
conversation in a typical working group meeting?  How many people just
sit in a working group meeting and take notes?  What percentage of the
working group can say they have read *all* of the internet-drafts that
have come out?  What percentage of the wg makes useful comments on
drafts?  From my experience, it's a very small percentage indeed!

So the question isn't whether "everyone" finds out, it's whether the
most active and the folks who most care about the outcome of a
particular area director decision find out --- and the answer is those
fifty people cover a large percentage of those folks --- by design!
(The reason why all wg chairs and all I-D author's were chosen was
that it was an easy, algorythmic way of trying to get complete
coverage of that set of people --- and it's failing is that while it
probably gets over 50% of those folks, it doesn't get them all.)

						- Ted