Ned Freed wrote:
Ned,
Dave, I'm sorry, but it didn't show that at all. The specific problem
that
arose here WAS anticipated and analyzed and the correct thing to do in
this case
WAS determined and documented. See RFC 3797 section 5.1 for specifics.
I don't know how many ways I can say this, but 5.1 is irrelevant to the
problem I was concerned about, which is having the pool come out at the
same time as the results. That allows for mischief in many ways (not
that I'm accusing anyone of that). Under the circumstances I *still*
believe that the chair did the correct thing, and that his doing so has
ensured the integrity of the process.
First of all, as others have suggested, the problem with the proximity of the
list and result publication can be addressed trivially by having the
secretariat provide the list they received for vetting purposes as well as the
result they handed back.
What we had here was a foul-up between the secretariat and the chair,
largely based on timing and technical problems. There is no need for
the secretariat to post its input list (although I have no objections to
that). And as I wrote, in *this* case, which is what I am talking
about, given the problem, the chair did the right thing. The rule which
states that the chair will wait at least a week, is *almost*
sufficient. If the secretariat posts the list, it should also post the
date the algorithm should be run, and it should be responsible for
seeing that the message is properly distributed - which is really what
happened here. And even so, *other* problems can arise, and so I agree
with Dave that attempting to address every contingency is going to yield
to the qualifications for chair being an Internet lawyer, and nobody
should want that. Save the lawyers for the real elections :-(
As to whether the chair gamed the system in the way you described, it's
easy enough to determine: ask the secretariat if a message got stuck in
the queue. If it did, then he didn't. If it didn't then perhaps he
did. Of course who is to say the secretariat didn't collude? Well, at
some point we have to draw a line and simply believe they didn't.
Besides having the list visibly posted in advance is sufficient to
tackle this particular case.
Eliot
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf