Re: [Iasa20] Barry Leiba's Discuss on draft-ietf-iasa2-rfc4071bis-08: (with DISCUSS and COMMENT)

[Barry Leiba <[e-mail]>]

(I'm not sure when nor why the IESG was removed from the distribution
list (Bob didn't do it), but I've added it back on; please leave it
there.)

> Another alternative not discussed is to change so it has to be the IETF Chair.   No exceptions.

It's correct that we haven't discussed it here, because THAT was
certainly discussed and decided in the working group, and relitigating
that would be inappropriate.

It sounds to me from the discussion here that my option (2) is a
fairly minor change and something we can live with and easily confirm
working group consensus on.  I'm happy to propose text, as I'm the one
who started this up; I'm also happy if the editors go off and do it.

Barry

On Tue, Apr 9, 2019 at 11:06 AM Barry Leiba <[e-mail]> wrote:
>
> Hi, Joe.
>
> >> What I see -- and, so far, no comments have shown otherwise -- is that
> >> the term period was well considered, but, in general, the effect that
> >> that period has on the ability of the IESG to actually have a
> >> meaningful choice in whom we select... was not.  It *appears* to me
> >> that the alternatives for the IESG were included grudgingly, where at
> >> least a few people just wanted it to be the IETF Chair ex officio, and
> >> that's it.  My issue here is that if we're given a choice it has to be
> >> a real choice, not a Hobson's choice.
> >
> > I still don't get the concern here, Barry, apologies. Is the sole
> > concern that if an AD is not on the same cycle as the Chair then the
> > IESG may be constrained in picking that person? I don't think that's
> > the case; even if that person cycles out of the AD role they can still
> > serve as the IESG-appointed member. It would be less of an effective
> > liaison with IESG since that person would not be deep in the mix for
> > that last year. Do we need to specify that this IESG-appointed LLC
> > director must be in cycle with the chair (since the text says this is a
> > two-year appointment and having someone not on IESG for their last year
> > doesn't seem ideal).
>
> My concern is that the IESG is limited in whom it can choose, and it's
> not clear to me that that was intentional and that the working group
> actually considered these issues.
>
> As I said in the message you quoted, I don't agree that the current
> text allows the IESG to pick an AD, have that AD leave the IESG in a
> year, and simply have that AD continue as the IESG-appointed Director.
> As I read the text, the IESG would have to run a selection process,
> which might or might not result in the same person being selected to
> continue.  Further, should that happen then the IESG-appointed
> position would become out of sync with the IETF Chair cycle, and that
> could cause unexpected and undesirable results down the line.
>
> OK, so let me try to lay out what I think we need to do to clear up
> this situation.  The point is that I want the document to be clear
> about what the choices really are -- because I don't think it is --
> and to be sure the working group has consensus on the result --
> because I think this is too significant a change to be simply
> editorial.
>
> I suggest the following alternatives (one could come up with others, I'm sure):
>
> 1. Change the term for the IESG-appointed Director to 1 year, with
> language strongly recommending multiple terms for that member (likely
> using the word "exceptional" to describe a single one-year term).
> This would also mean changing the term limit for that role to be
> specified in number of years, rather than number of terms.
>
> 2. Make it clear that once the IESG appoints an AD as Director, that
> person will serve her full term (modulo resignation, and see below),
> regardless of whether she remains an AD.  It sounds like this is the
> best way to make everyone on this discussion thread happy, but I think
> we still need to go to the working group for consensus on it.
>
> 3. Move the word "exceptionally" earlier in the sentence, to make the
> selection of anyone other than the IETF Chair exceptional.  It would
> be helpful, should the working group choose this, to also add text
> making it explicitly clear that this is meant to be an IETF Chair ex
> officio position.  I gather this is what some on this thread want;
> it's my personal last choice... but in any case, I'd want to see
> working group consensus for the choice.
>
> I'll also note that, as we've mentioned, it's possible for a Director
> to resign mid-term, and if we simply appoint another AD for a two-year
> term we might get that position out of sync with the IETF Chair cycle.
> So I think it's important to also add text specifying that a mid-term
> replacement will serve:
> * if the remainder of the term is more than a year, through the end of
> that term, or
> * if the remainder of the term is a year or less, through the end of
> that term plus another term.
>
> Does this make sense to everyone?
>
> Barry