[python-committers] CoC violation (was: Retire or reword the "Beautiful is better than ugly" Zen clause)

[Donald Stufft]

> On Sep 21, 2018, at 8:59 AM, Paul Moore <p.f.moore at gmail.com> wrote:
> 
> On Fri, 21 Sep 2018 at 13:30, Donald Stufft <donald at stufft.io> wrote:
>> So part of being and open and welcoming community, is knowing and understanding that words, images, etc like that can make people feel like we’re either a group that will directly engage in those attacks that have been associated with them in the past, or at least won’t come to their aid if someone does initiate those kinds of attack.
> 
> I'm going to take this one comment and respond to it out of context.
> But generally, I agree with everything you said.
> 
> My biggest concern is that we're starting to build a community where
> people feel exposed to attack for "CoC violation" accusations over
> simple misunderstandings, or careless wordings. Or, for that matter,
> using terminology that they weren't aware was unacceptable. Not "being
> called out (by the offended party), apologising and moving on", but
> going straight to policy complaints by people (maybe even people not
> directly upset) assuming offense could be claimed. That's clearly
> nothing like the sort of problems people with real reason for
> sensitivity have to live under, but nevertheless it's not a
> comfortable place for people to learn how to interact.
> 
> Balance, forgiveness, and a mature level of empathy are what's
> *really* needed ("among the things that are needed...":-)). Not
> policies. Policies should be weapons of last resort.
> 
> Paul

So I don’t think that being called out by the aggrieved party is the right response generally for these sorts of things. I mean, ultimately it depends on the specific instance, but often times having the person who is feeling attacked call out the other person, what’s going to happen is that person is going to feel compelled to respond back in kind and “defend” themselves. Having a neutral third party there to mediate and calm the situation down is immensely helpful.

I mean, if you personally did something that made me feel uncomfortable, I’d probably personally handle it, because we have a  rapport already, but if someone else did there’s a chance I wouldn’t (either because there might be history there where a specific instance finally spilled over, or because I’m angry/hurt/whatever and I don’t trust myself to respond).

This also falls into the feeling exposed to attack bit. Generally what the CoC does should be private, though it’s tough to balance that out with being transparent too. For instance, we don’t really want to turn CoC enforcement into it’s own sort of shame. If you were to report me, ideally the way it would play out is some member of the moderation team / CoC team / whatever would privately contact me, and tell me to knock it off or whatever. Generally other people shouldn’t know (unless one of the two sides of the issues chooses to divulge it) that it happened (although it’s good to publish anonymized reports too). There should not be some sort of record that the dastardly Paul said something bad once and had to get reprimanded.

Where it gets harder is when more drastic measures are to be taken. If someone gets banned for a day in a sort of timeout, should that be public? Probably not since we want them to come back and ideally be positive contributors from that point out, and feeling like they’ve been put up on display is probably not conducive towards that, and being gone for a day is not likely to be something where other people notice the absence and start to question it. What about a week? A month? Permanent?

Personally I think that publicizing that a particular person had some action taken against them is probably the wrong path to take in all severity levels, and that the CoC team should probably publish some sort of anonymized reports. These reports basically serve to show people who are worried about feeling safe/welcome in the community, that if they have a problem they’re likely to be heard and helped, without putting particular people “on blast”.

Unfortunately our tooling and process isn’t really “here” yet, for instance in the specific case we’re talking about, if that person was jsut silently banned than it can feel a bit kafkaesque and since the record of his statement is permanent and can’t be hidden or something, people looking in from the outside don’t know that it wasn’t acceptable since they don’t see any action to have taken place. The ideal situation is probably that the original post ends up edited, marked,  or hidden in some fashion (but doesn’t just disappear) to say that it was inappropriate in some way (think what GitHub does here with hidden posts) but that doesn’t otherwise create some sort of notification. 

I however, think policies are great! Particularly in a diverse community where the cultural norms may vary widely amongst all of the participants. It helps document what the community expects from people, tells you what the process to take is for remediation of a bad situation, and for the people taking those steps, provides a framework to determine what the best possible outcome is.

Or to put another way, the choice isn’t between not having a policy, and having a policy, the choice is between having an implicit policy where what is allowed and what isn’t is left up to the whims of whoever happens to be around at the moment, where enforcement is likely to be very uneven based on who the offender is etc, versus an explicit policy that spells out the expectations for everyone (and explicit is better than implicit after all ;)).

The PSF CoC falls short in this area, since it’s mostly a vague, aspirational document that leaves a lot to be implicit versus one that is more specific.