Speaking only for me ...
On Mar 14, 2015 3:10 AM, "John C Klensin" <john-ietf@xxxxxxx> wrote:
>
>
>
> --On Friday, March 13, 2015 21:16 -0500 Spencer Dawkins
> <spencerdawkins.ietf@xxxxxxxxx> wrote:
>
> > On 03/13/2015 03:33 PM, Stephen Farrell wrote:
> >> John,
> >>
> >> On 13/03/15 18:05, John C Klensin wrote:
> >>> However, the responses by a few IESG members to comments
> >>> about the IESG's exempting itself (and other parts of the
> >>> leadership [1]) from the provisions seems to call for
> >>> additional comment and probably a few more issues for the
> >>> list of topics needing addressing.
> >> I don't believe the above matches what the document says at
> >> all.
> >>
> >> The IESG didn't "exempt" itself from anything - the ombudsteam
> >> can very clearly exclude an AD from one or more meetings, just
> >> as with any IETF participant. (If that is not clear, it needs
> >> to be, but I think it is.)
> >
> > I agree with Stephen that this is what we _meant_.
> >
> > If it's not what the document says, please tell us, because
> > (as Stephen says) this needs to be clear.
>
> It is not what the document says. The document says
>
> "(The Ombudsteam can not impose that a Respondent who is
> in a IETF management position be removed from that
> position. There are existing mechanisms within IETF
> process for the removal of people from IETF management
> positions that may be used as necessary.)"
>
> That seems fairly clear to me although I should have been more
> clear about "the provisions" from which the IESG exempted
> itself. Given a document written written by two IESG members
> and extensively massaged by the IESG, it is hard to read it in
> any other way than "the IESG has exempted itself...". If the
> IESG sees itself as a team, which I hope it always will, the
> ability of the IESG Chair (sic) to remove an Ombudsperson
> without cause or explanation just adds to that perception.
Ok, that's fair. Thanks for the insight.
What I was worried about, was the alternative that it takes an entire Nomcom plus community input plus a confirming body to put someone in a Nomcom-reviewed position, but only an Ombudsteam to take someone out of that position.
For extra credit, for better or worse, it's pretty difficult to "pack" a Nomcom's voting membership.
For extra, extra credit, we have years of experience with Nomcoms, and we don't have that experience with an Ombudsteam.
So, speaking only for me, what I thought the final version of that text was doing, was protecting the community, not protecting the IESG.
I get that others might see that differently.
Spencer
> That statement also raises the "management position" issue on
> which Brian and I have commented. And the only "existing
> mechanisms" I know of amount to either recall (difficult,
> inappropriate, or useless for reasons I explained in that note
> plus the confidentiality problem Brian raised in his.
>
> Moving away from hair-splitting and to the bottom line...
>
> Because of the power and explicit and discretionary authority
> the community gives them, I think it is appropriate and probably
> necessary to hold ADs --and expect them to hold themselves-- to
> a higher standards than random IETF participants. Consequently,
> if the Ombudsteam has a discussion about a harassment claim with
> an AD and the issue cannot be explained by the AD to the
> Ombudsteam's satisfaction, I'd expect the AD to either correct
> the behavior forthwith or resign. If neither option is taken,
> if the community is really serious about eliminating harassment,
> I think the community wants the AD out of there and, for
> effective operation of the IESG, replaced quickly. Put
> differently, having such an AD linger on for many months (see
> Note 1), doing whatever he or she was doing with (as Brian has
> pointed out) full community knowledge of what is going on, is
> not in the best interest of the IETF.
>
> I'm also completely unpersuaded by "well, the AD can be excluded
> from meetings and/or a mailing list". Sorry, but the community
> simply doesn't need ADs who are forced to function without the
> full range of tools available to them. Any AD who is unable or
> unwilling to cease or correct harassing behavior after the
> various interventions called for by the document _and_ unwilling
> to resign should probably be removed for bad judgment and
> insensitivity to the needs of WGs and document processing in the
> community... and that should not require a 5 - 0 month (minimum)
> recall process.
>
> I don't know whether to describe reliance on conventional recall
> for these sorts of situations is "ineffective" or an indication
> that harassment by "management" is not taken seriously.
>
> Mumble.
> john
>
>
> Note 1: For the source of my "many month" and "5-6 month
> (minimum) assertions, consider the following steps to form a
> plausible estimate of how long the recall process takes:
>
> * 20 signatures must be collected (see my earlier note
> about that one)
> * Secretariat confirms the signatures and makes a public
> announcement
> * ISOC President appoints Chair (I assume no one is on
> standby and that, while it is not prohibited by RFC 7347
> or elsewhere, extensive consultation with IESG and/or
> IAH members about who should be chosen when there has
> already been a public announcement that a petition has
> been filed against one of their members would be
> unseemly, so getting this done, and done well, in a
> couple of weeks would be really fast).
> * A recall committee is created using the same rules as
> the Nomcom (See Section 7.3 of RFC 7437). That is,
> among other things, up to three months to select and
> organize that committee (Section 4.1 of 7437).
> * Then the committee needs to investigate and reach a
> decision, which presumably involves some effort,
> interviews, calls, etc. From a scheduling standpoint,
> one hopes that does not involve waiting until f2f
> interviews can be help at an IETF meeting, but nothing
> in 7437 seems to prohibit that.
> * If the person is removed, the normal vacancy-filling
> process applies.
>
> My crude guess is that, if the normal recall procedure is used,
> it takes a absolute minimum of five months from the point that
> someone decides a petition is appropriate to actually remove
> someone... and that is after complaint/ reporting, Ombudsteam
> investigation and discussion, attempts to get the behavior
> corrected, time to see if those attempts were effective, etc. I
> think that adds up to a very large fraction of a year, even
> without counting time to pick a replacement. It has got to be a
> major portion of the reason why the recall process has never
> been used.
>