Re: What is a "management position? [Last Call: <draft-farrresnickel-harassment-05.txt> (IETF Anti-Harassment Procedures) to Best Current Practice]

[John C Klensin <[e-mail]>]

--On Wednesday, March 18, 2015 12:52 -0500 Pete Resnick
<[e-mail]> wrote:

>...
> We really want the Ombudsteam to be as confidential as
> possible. That's certainly important for the Respondent, for
> whom public accusations can be destructive and could bring
> legal liability for both them *and* the IETF, but I think the
> most important reason for confidentiality is for the sake of
> the Subject: Being the recipient of sexual, racial/ethnic, or
> other sorts of harassment (in contrast to most other sorts of
> bad behavior) can be embarrassing and shaming for the
> recipient, and can make one the object of all sorts of
> horrible negative comments. Of course, it would be perfectly
> lovely if we lived in a world where comments along the lines
> of "she was asking for it" or "he's just being overly
> sensitive to any comment about race" or "she seemed to be
> happily participating in it" were not the first things out of
> people's mouths, but we don't live in that world. And it would
> be nice if people did not look upon the targets of harassment
> as victims to be pitied, but they do.

> Because of that, the best thing to do is to limit as much as
> possible the number of people who get exposed to the
> information about the incident. In the current text, even in
> the most extreme cases of remedies (e.g., removal from a
> meeting), the only people -- beyond the Ombudsteam -- that
> need to be made aware of the imposition of a remedy are people
> who are under contract to the IAOC/ISOC (i.e., the
> secretariat), which means that some level of confidentiality
> can be contractually imposed. As soon as we give the ability
> to the Ombudsteam to remove someone from an appointed position
> (NomCom-appointed or otherwise), members of the open volunteer
> community are going to have to be made aware of the existence
> of the remedy, whether it's the entire IESG ("This person can
> no longer be a chair for any WG"), the IAB ("This person can
> no longer be confirmed as an AD") or others. Maybe we want to
> go through the exercise of trying to figure out what sorts of
> confidentiality "envelopes" we want to have and examine how
> this document interacts with 2026, 2418, and 7437, but it's
> not going to be a pleasant exercise.

Pete,

This seems entirely reasonable and quite appropriately
idealistic, but it seems to me that there is a turning point
between "persuade offender to understand the problem and stop
offending" and "protect present and future victims from the
offender by restricting the offender's action, presence, and
behavior".   Once that corner has to be turned (and I hope that
will be rare or less often), measures that try to maintain
confidentiality about what is being done and why are likely to
be indistinguishable from "cover up", whether that is the intent
or not.

> The thing I like about the original text, and reason that I'm
> OK with in Sam's text, is that it limits how many people need
> to know the result; the Ombudsteam removes the person from the
> situation that is causing the problem (from a mailing list or
> a meeting in the extreme case), but can only tell the
> Respondent that it would be a good idea to resign at that
> point. If things go the right way, the person resigns and the
> outcome is identical to having the Ombudsteam remove them
> (with potentially less people having to be told).

If things go the right way, the Respondent corrects his or her
behavior, apologizes if appropriate, and never offends again.
By the time there needs to be even a discussion within the
Ombudsteam about removing someone from something, either things
have failed to go the right way and the issue has become about
protecting the community from expected future bad behavior or we
are entirely into the sort of punitive situation you are trying
to avoid.

>  If things go
> badly and they refuse to resign and instead make a stink, the
> outcome doesn't seem all that much worse than the stink such a
> person will make after the Ombudsteam summarily dismisses them
> from their position. And again, it avoids the Ombudsteam
> having to figure out who in the community needs to be told,
> and what particular details they need to know.

Pete, we've got a community that is pretty good at figuring
things out, even at getting sucked into conspiracy theories and
figuring things out that might not be true.  So consider two
scenarios which seem to be feasible:

(i) A WG Chair is removed.  When people ask what happened, the
relevant AD claims no knowledge of the issue or claims that the
reason the person stepped down is confidential and the
sitting/remaining co-chair claims ignorance (probably
truthfully, unless she was interviewed by the Ombudsteam).  The
former chair refuses to discuss the situation.  How long do you
think it takes for the speculation to start?  Do you think the
IETF community would believe "decided to spend more time with
his family" when the Respondent denies it?  Moreover what is the
risk of damage to the personal and professional reputations of
completely innocent people who step down as WG Chairs for
personal confidential reasons that have absolutely nothing to do
with harassment because the very existence of this policy and
procedures causes speculation and vicious rumors to start?
Forcing such people to publicly disclose the reasons why they
stepped down to defend themselves from such rumors would be,
itself, a form of harassment.  As an exercise, tell me who
should be a Respondent.

(ii) Someone is removed from a mailing list by the Secretariat.
The drop in participation is observed and people ask why.
Incorporating your first paragraph above by reference, the
Respondent answers questions by saying "that <expletive> Joe
Blow couldn't take a joke, accused me or harassment, and the
oversensitive and excessively politically correct Ombudsteam
kicked me off".  The Subject has just been outed, despite the
language in the spec about confidentiality.  The Ombudsteam is
going to do what?  Hint 1: Any response would be punitive, not a
remedy and corrective.  Hint 2: A recall of an IETF participant
who is not a member of the IAB/ IESG/ IAOC is a little hard. 

IMO, in either of those scenarios and remembering that removal
or blocking of any sort really are measures that gets applied
only after education and counseling have failed (or punitively),
it is not clear to me that attempts at confidentiality are worth
it... and that they might have bad as well as useful
consequences.

   john