On 3/18/14 11:54 AM, Michael StJohns wrote:
The IETF is a non-membership standards body.
Exactly right. That's why the IEEE has a lot easier time having a policy: With members, you can ask them to formally sign up to certain standards and they can be formally removed from membership or otherwise sanctioned (even monetarily if they wanted) if they don't behave within those standards. And investigating such behavior can be done (by professional staff of the organization) in a private fashion. We can have standards for participating in a mailing list or showing up for meetings, but sanctioning poor behavior is somewhat trickier. (Not impossible; just trickier.) And as you said, our openness rules make it hard for our normal leadership (who are all part of the volunteer participants) to address these issues other than in public.
We are perfectly within our rights to deal with public disruptions of the standards process, and we have procedures in place to do just that with respect to all IETF controlled mailing lists. We are also perfectly within our rights to exclude people who do not conform to accepted behavior norms - again - in public - at IETF meetings as a matter of safety, decorum or protecting standards production. It's less clear that we can exclude someone from attending a meeting solely based on bad behavior on a mailing list. [I've been told for example that one mailing list participant I consider somewhat disruptive was quite respectful in person]. The key word here is "public".
Yep. And we have processes right now that deals with public disruptions of the standards process. Chairs can deny people the floor in a face-to-face meeting and leadership can remove posting privileges to our mailing lists when people are disruptive of the standards process. But because of our openness rules, we have never had rules that allowed anyone to be excluded from quietly sitting in a room if they were not *directly* disruptive. And "directly" is part of the problem:
Where I think the document goes off the rails is that it provides an IETF cause of action for private interactions with a related confidential investigation and resolution process...
The problem with harassment is that I can harass someone in the hallway of an IETF meeting and create an environment for the target of that harassment that is disruptive for that individual, but perhaps not disruptive in a way that a chair could reasonably be asked to enforce. And making the issue public can *increase* the level of disruption for the individual target in question.
For example: If someone is extremely sexually aggressive toward me in the hallway before a meeting, they may have quite effectively disrupted my ability to participate in the meeting. But there was no direct disruption of the meeting. Even if I went to the chair, the most the chair can really do is not grant the floor to my harasser, but it's going to be pretty difficult (and not currently part of our procedures) to ask that person to leave the room. Furthermore, even if it were possible to kick the harasser out, a public airing of the incident could *increase* the disruption to my participation: Bringing up topics of a sexual nature publicly can be terribly embarrassing, and these topics are often met with dismissiveness (e.g., "Oh, he only made a pass at you; take it as a compliment and ignore it"). And as I said in an earlier message, harassers who might otherwise be willing to cooperate in the resolution of an incident if things were kept private might instead issue denials and claims of "simple misunderstandings" if the incident needs to be publicized, again because discussions of sexual topics can be embarrassing, and because publicity about them can escalate to legal and employment problems. If I, as a target, am forced to bring these things out in public, I may choose to simply not participate rather than have to have the discussion publicly. That may also cause others similarly situated to me to choose not to participate if the environment is seen as problematic.
Similar stories could be told regarding racist or other discriminatory sorts of interactions, including the same sorts of discomfort about discussing these things publicly. I would be thrilled to live in a world where harassment could always be made public without embarrassment or shame for those who get harassed. But I don't think we're there yet. Sure, some sorts of harassment are overt and public and can be handled by our current public procedures regarding disruption. Many sorts of harassment are not.
So I think the ability to have an (exceptional) process for dealing with incidents privately is necessary. I hope that we can come to consensus that having some set of folks in place for these (hopefully exceedingly exceptional) circumstances, who (in the most exceptional of these circumstances) have the power to quietly tell someone that they can't even stay at the meeting, is an exception to our openness that we can live with.
IANAL, but, I would expect that any attempt to impose IETF sanctions due to private interactions based on a confidential investigation could leave the IETF vulnerable to being sued. AIRC, the IETF has an insurance policy that indemnifies various IETF related folk against being sued for actions related to the standards activities - it would be interesting to get a read of the underwriter as to whether the policy would cover non-public harassment related actions without a substantial IETF node.
My understanding is that this has been discussed with the insurance folks at ISOC and that we are on reasonably safe ground so far.
pr -- Pete Resnick<http://www.qualcomm.com/~presnick/> Qualcomm Technologies, Inc. - +1 (858)651-4478