Re: Alternative (H2)

[Robert Elz <kre@munnari.OZ.AU>]

    Date:        Fri, 30 May 2003 10:30:42 -0400
    From:        RJ Atkinson <rja@extremenetworks.com>
    Message-ID:  <4ACB5D12-92AB-11D7-B79B-00039357A82A@extremenetworks.com>

  | We *do* care that the Confirming Body doesn't deadlock without
  | any decision,

No, we don't.  What we care about is getting a decision.  If there's
no decision, it is irrelevant whether it is because of deadlock, or
because they haven't even met once to consider the issue, or ...

You're focussing on solving a particular problem you saw, and not
looking at the broader issue.

  | That deadlock would have occurred under the rules of the previous 
  | Nomcom RFC

I agree that getting rid of the rule that is there now on how the
confirming body should act is a good idea.

  | So Harald's assessment is quite correct IMHO:
  | 	- As soon as IAB creates their own rules, the contingent rule
  | 	  no longer applies.

First, it isn't "the IAB", it is "the conforming body", and second,
we really should not care how they make a decision.

Eg: what would you do if the confirming body created a rule "when confirming
candidates from the IETF nomcom, three months must elapse for study before
a vote can be taken" ?

  | The anxiety you seem to be concerned about -- excessive legislation
  | of rules -- is actually *fixed* by H2, not made worse by it.

While in some cases that is what I worry about, in this case, that isn't
it at all.   It is that it is totally inappropriate for us to be making
any rules at all.

  | And as Scott Bradner correctly observed a week or so back, we've done
  | a tremendous amount of collective wordsmithing to end up with H2.  Yet
  | more wordsmithing of H2 is unlikely to help in any meaningful way.

I think all that has been based upon attempting to solve the wrong problem.
That is, it has all been aimed at finding a way to force an answer, one way
of the other, out of the confirming body (and particularly when that is the
IAB).   That's simply wrong.   We cannot make another organisation do
anything, or do it in a way we choose, just because we say so (for the
purposes of this discuccion, you should consider the IAB to be a totally
independent organisation to the IETF - more like the ISOC, the other
confirming body that these rules also apply to).

  | IMHO, we need to just adopt the H2 text as replacement for the current
  | broken text and move forward with identifying/resolving any other open
  | issues with this draft.

Once again, the right way to fix this, is to decide what should happen if
the confirming body says neither yes, nor no within a reasonable period.
It should be irrelevant to us why they got into that position (internal
deadlock, didn't feel like calling a meeting as all members were on
extended vacation, ...).   That completely solves the deadlock problem in
the correct way, and also provides for all kinds of other things that
might go wrong  (and after thinking about it, I think the answer has to
be "confirmed" if there's no explicit rejection, otherwise we have a
potential loop, where the confirming body never does anything, and a
default reject applies every time).

kre