Hi! > It is broad for a reason. If harassment that's obviously connected > with the project (it would need to be obviously connected) happens > off-list, that's still problematic. I think limiting the scope to just > the project territories is dangerous as it provides too easy of a way > for members to cause problems with no resolution possible.
I think approaching it with the idea that it can resolve any disagreement is very dangerous, as it implies scope creep and trying to police other's speech and actions in venues having little or nothing to do with the project. I understand it is not *your* intent *now*, but once it's in place, you have no control over it, and it would develop by its own laws. And documented experience shows that there are a number of people willing to abuse CoCs (including in projects that they have little or no prior involvement with) to play politics, bully, troll or raise their own profile, and inviting this conduct by having undefined expandable scope is asking for trouble. There are documented cases for that. Yes, there will be real bad incidents which this CoC will not be able to handle. It is completely fine (not the incidents, but this property of CoC). You can not control the whole world with one CoC. Let's limit ourselves to building better community, not to try and control everybody everywhere. > Unless we lift that ban just for the appeal. Meaning that the banned > individual requests an appeal, so they are unbanned for that single > thread until it is resolved... Requests from whom? What about pre-appeal process? I think if we talking about actions like long-term bans, process should be automatic, not something that is optional and should be requested by undefined means. > The other side is far more serious though. Many MANY people avoid > coming forward about incidents because they are afraid they will be > labeled. Simply look at *EVERY* woman who has come forward to talk I know, but if we want to take community-wide action, I see no other way but letting the community know what happened. Due to the nature of the incident, it may be possible to conceal people involved, or may not be possible (i.e. if you know X and Y were feuding and then X is sanctioned, it's pretty clear what is the cause) - but if you want community consensus, I do not see how you plan to get one without telling the whole story as much as possible. Telling one side is not good and would be wildly misleading - I have seen enough cases where one side tells one story, other says another, and the impartial evidence (logs, emails, etc.) tells the third. Not having the full story is a guarantee for unjust actions. > other side is innocent. No matter the strength of the evidence. Simply > because no matter how strong and clear the evidence is, some people > simply won't believe it. That's fine, we do not need *everybody* to believe it. That's why we have votes, etc. But if only 5 people ever know what it is about, I do not see how they can be given power to decide for the whole community. > Well, what's the alternative? To let them continue to cause trouble? I If they continue, there can be emergency measures. But then it would be pretty clear-cut. If there is any reasonable doubt, there should be no bans, at least not without consulting the community. > would never vote for a ban (temporary or permanent) unless there was a > strong pattern of significant abuse, and I think many here would agree I believe you. But you are not the only person voting - in fact, I'm not even sure if you'll be the person voting at all (are you going to be member of the committee?). > As in the appeals process, perhaps we can figure out wording to lift > the ban on the mailing list during the RFC process, under the explicit > understanding that the accused offender not post to other threads > until the RFC process is done. Would that at least partially resolve > that concern? I would rather replace ban with moderation (assuming we can have one person moderated in our list software), with moderators be a wide set of established contributors. While we can disagree on the details, I assume most of us can distinguish egregious abuse from something else. > So I think the argument that "we're too big for that" is a bit of a stretch... That's not the argument. The argument is "we're too different and it's too important to just take preformulated one". Especially when it has issues with both wording (too punitive and negative) and scope (described above). We have seen much better examples, from Drupal to Python to Django. > I have received no less than 4 direct threats of violence that were > directly due to my involvement with the Scalar Type Declarations RFC. That's bad and obviously unacceptable. Were those list members? Do you think a threat of being banned from the list would prevent them from doing it? > I believe that both Zeev and myself crossed significant lines during > that discussion as well, to which there should have been some level of > recourse or moderator that could have stepped in to cool us down and > help. Was it harassment? Do you think you and Zeev should have been banned from the list for your conduct? > Since posting this RFC, there have been people openly speculating > about my gender, sexual orientation and other personal matters. In > contexts that are purely obvious that it is connected to this RFC, and > hence the project. Are you implying that mentioning one's gender is harassment? That would be very worrying in its broad scope. I hope we do not intend to police the internet for any discussion that may involve anyone from the community and try to intervene? Because I certainly don't remember anyone's gender being discussed on any community resource. Otherwise, there's a lot of crap going over the internet, and I don't see how our CoC is going to change that. > And that's just me. I know for a fact that several other people have > had incidents. I know that several people avoid internals and the > project because of fear of incidents. I won't speak for them, that's > their prerogative. What kind of "incidents"? I know there are heated discussions sometimes, and it can be too much for people (I myself had to take time off from the list several time because it is too exhausting and nerve-wrecking) but I don't see how CoC would change any of that - heated argument is certainly not harassment, and having CoC team intervene in the course of even a heated discussion would only be hurtful, in my opinion, by further poisoning the atmosphere with power plays. > But please stop pretending nothing's ever happened. My experience > alone should be enough to justify. I never said "nothing's ever happened". The claim was specifically about people scared away from contributing to PHP because there's no CoC, not about anything happening ever. To substantiate that, we need example of: 1. Harassment that is against CoC and warrants counteraction 2. One that CoC would have actually stopped (if somebody is being harassed on Twitter, our CoC can't do anything about it) 3. One that would scare a person away Luckily, your experience seems to not be of such kind - in fact, except for the threats of violence, it seems other examples do not fall under either of three. > Because the argument stands on its own. We shouldn't have to talk If it stands on its own, the let it do that and drop the "crowds of imaginary scared contributors" argument. There's a lot of things here that could (and I am sure does) scare people away, but not having formal CoC is nowhere near the top. > arguments stand on their own. And the vast majority of people agree > that even if nobody was hurt, it's worth having (the smoke detector > argument). Smoke detector is fine. Smoke enforcer that beats you up when you are having a barbecue on a backyard or enjoying a cigar on the beach is not. I'm fine with having a smoke detector. Even fine with smoke cleaner. It's smoke enforcer with indefinitely broad scope that I have problem with. -- Stas Malyshev smalys...@gmail.com -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
