Re: H2 or J

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

On Mon, Jun 09, 2003 at 08:36:11AM -0700, Christian Huitema wrote:
> I prefer H2. Well, in fact, I like the concise wording of J, but I
> prefer the outcome of H2, i.e., a candidate not explicitly confirmed or
> rejected is *rejected*. Confirm by default is a bad idea, and also a
> path to laziness. Reject by default sends the strong message to the
> confirming body that they have better get their act together, or
> something violently public will happen.

I agree with Christian's summary of the tradeoffs and with his
ultimate preference: H2.

I'll also add that if the default answer is "reject", then it's less
important as to exactly what the more ambiguous specification of
"reasonable time" is means, which is another advantage of H2.  

Regardless, it's pretty clear though that if the confirmation body
doesn't hand back an answer in time for the March IETF meetting, most
people would agree that this would be considered problematic!

						- Ted