RE: Sam's text and way forward on the last call of draft-farrresnickel-harassment-05.txt

[Spencer Dawkins at IETF <[e-mail]>]

Fwiw ...

On Mar 19, 2015 9:10 AM, "Adrian Farrel" <[e-mail]> wrote:
>
> I completely support the aims of Sam's text, but doubt the practicality.
>
> For example, if the Ombudsteam decides to remove someone from a WG chair
> position, how will this actually work? Will they tell the AD "this person
is no
> longer a WG chair" and will the AD be supposed to say "Ah, that was the
> Ombudsteam speaking, so I had better react by, for example, not instantly
> reappointing them"?
>
> The -05 text had...
>
>      The Ombudsteam can recommend, but 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.
>
> I stand by that text and I believe it achieves everything that is desired
> without:
> - needing any changes to existing processes
> - needing any more lapses in confidentiality than Sam's proposal.

So, I also read the -05 version, in addition to a few others.

What was confusing to me, in that text, was " the Ombudsteam can recommend
..."

I asked "recommends to whom?"

My understanding when we were noodling on -05 was that we were really
pushing on maintaining confidentiality, to the point that the Ombudsteam
would not tell a WG chair or individual AD that Spencer-not-an-AD needs to
be removed.

If we are ok with that happening, to the least extent possible, I will stop
asking snippy questions.

> It is, of course, vague. But let's look at some cases:
>
> 1. Document editor.
> "Hello WG chairs, we are the Ombudsteam and we ask you to remove Adrian
from
> editing draft-ietf-foo."
> The WG chairs will know what that means and don't need to know any more
details.
> What is the difference between this and Sam's proposal?
> - We do not need to update the text about who hires/fires document
>    editors
> - We have to trust WG chairs to understand that the Ombudsteam would not
>    ask if there wasn't good reason.
>
> 2. WG secretary.
> Ditto case 1
>
> 3. WG chair
> "Hello AD, we are the Ombudsteam and we ask you to remove Adrian from
chairing
> the FOO WG."
> The AD will know what that means and doesn't need to know any more
details.
> What is the difference between this and Sam's proposal?
> - We do not need to update the text about who hires/fires WG chairs
> - We have to trust ADs to understand that the Ombudsteam would not
>    ask if there wasn't good reason.
>
> 4. IESG, IAB, IAOC
> Here I think a little more is needed.

Oh, I'd agree. My thought was, that we shouldn't special-case more than we
must, so since we have a recall procedure, (try to) use it at least once
and see what breaks.

I saw a note, and I'd have to look at some email threads to see who else
was cc-ed, doing the math on how long it takes to run a recall through to
"Spencer is no longer an AD", and coming up with an estimate of 3-5 months.
My quick take was that the estimate was likely close.

But if that's the case, I'm not sure why fixing the recall process wouldn't
make about as much sense as special-casing anything here.

> Requiring a 20 person petition requires that 20 people know of the case
and
> possibly agree with the Ombudsteam.
> That is a possibly a step too far.
> I think we should allow
> - the Ombudsteam to directly petition the ISOC President
> - the fact of the petition to be published
> - the recall committee to listen to the Ombudsteam and trust them, maybe
>    asking for an outline, but not demanding full details
>
> I *do* agree that asking someone to resign is a good first step. I am
slightly
> suspicious that someone who harasses to the extent that this remedy is
judged
> appropriate would probably not choose to resign. But running code will
expose
> that.

Yes, and speaking as a retired process weenie, I'm not sanguine that we
would be coming up with more tweaks that will address the first problem we
encounter in practice. Running code good.

So what I'm wondering, is whether starting with either -05 or -06 is worse
than what we have in place now ...

Thanks,

Spencer

> Thanks,
> Adrian
>
> > -----Original Message-----
> > From: ietf [mailto:[e-mail]] On Behalf Of Jari Arkko
> > Sent: 19 March 2015 10:04
> > To: IETF Discussion List
> > Subject: Sam's text and way forward on the last call of
draft-farrresnickel-
> > harassment-05.txt
> >
> > I wanted to recap where we are with respect to the topic
> > of incidents handled by the ombudsteam affecting roles
> > that people have in the IETF.
> >
> > First off, I think we have broad agreement that we need
> > to deal with this better than version -06 of the document
> > does, and that misbehaving leadership needs to be
> > removed. The debate has been about the specific
> > mechanics of doing that, and clearly -06 was not up
> > to the task, as well as leaving a bad impression.
> > I'm sorry. We are now in the process of seeing how to
> > correct that.
> >
> > I would like to get to the specific proposals. Sam
> > suggested one way of dealing with this, copied below.
> >
> > I personally like this text. There are some variations of
> > the general approach, I think Pete argued for a similar
> > treatment of WG chairs and nomcom-appointed
> > positions. I could live with that as well.
> >
> > However, there is clearly another class of approaches
> > to solving this. We could specify the mechanics of
> > ombudsteam initiating or running the recall process,
> > or providing the ombudsteam the authority to terminate
> > appointments. I think this type of an approach is also
> > possible, but would tie into our nomcom and recall
> > processes in a quite close manner. The details would
> > have to be specified, and of course, the resulting system
> > should still be workable, safe, etc. from overall IETF
> > perspective. Does anyone have a proposal in this
> > space, or believe we should take this path?
> >
> > Or are there other approaches not listed in this
> > e-mail?
> >
> > So at this point I would like to ask if people are
> > comfortable with Sam's proposal or if other
> > proposals are forthcoming and/or people believe
> > that another approach is needed. Concrete
> > text proposals would be appreciated.
> >
> > Here's Sam's proposal:
> >
> > > I'd like to take a stab to see if I understand what we do have
consensus
> > > on:
> > >
> > > old:
> > > (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.)
> > >
> > > new:
> > > The Ombudsteam MAY ask a respondent to consider resigning from an IETF
> > > management position.  The Ombudsteam May remove a respondent from a
> > > working group  or document editor position.  While this document does
> > > not create additional procedures permitting a nomcom appointee be
> > > removed, the Ombudsteam can exclude a respondent from meetings and
> > > mailing lists and other activities, making it impossible for them to
> > > carry out their appointed tasks.
> > >
> > > Rationale  for the above:
> > >
> > > I think we should split handling of chairs and wg-level positions from
> nomcom
> > stuff.
> > > The discussion to date seems to have focused on nomcom-level
> > > appointments, and we apparently don't have consensus to make changes
to
> > > that in this document.
> > > However, I think we should carefully ask ourselves how we handle chair
> > > harassment.
> > > Recommending to the AD seems like the wrong approach.  The AD isn't
> > > going to be in a position  to know the facts, the AD is not going to
be
> > > trained in harassment.  As a manager I've sometimes been told by HR
that
> > > I had to take certain actions; sometimes I agreed, sometimes I wished
I
> > > had other options.  However, sometimes the interest of (in that case
the
> > > company, in this case the IETF) to avoid harassment are more important
> > > than an individual manager's preference.
> > > So, I'd like to float the idea that the Ombudsteam is in the best
> > > position to make harassment-related removals at that level.
> > >
> > > I've removed the sentence saying that the Ombudsteam cannot make
> > > leadership removals because it's too easy to read that as an
affirmative
> > > statement against leadership removals.  Instead, I've floated a
specific
> > > instantiation of the idea that the Ombudsteam does have the power to
> > > make it impossible for a leader to do their job.  I think it's
important
> > > to confirm we have consensus on this point.  It would be a huge mess
for
> > > the Ombudsteam to try and do that and to discover we didn't have
> > > community support for that.  In effect I'm arguing that it's important
> > > enough to make sure we're on the same page here that we float specific
> > > text for this issue and confirm it doesn't attract unresolvable
> > > objections.  Pete has said on-list and in private discussions that he
> > > believes their is support for the Ombudsteam choosing remedies like
> > > this.
> > >
> > >
> > >
>
>