On Thu, Jan 7, 2016 at 11:50 PM, Anthony Ferrara <ircmax...@gmail.com> wrote:
> Zeev, > > On Thu, Jan 7, 2016 at 3:50 PM, Zeev Suraski <z...@zend.com> wrote: > >> -----Original Message----- > >> From: Anthony Ferrara [mailto:ircmax...@gmail.com] > >> Sent: Thursday, January 07, 2016 8:15 PM > >> To: internals@lists.php.net > >> Subject: [PHP-DEV] Re: [RFC] [Draft] Adopt Code of Conduct > >> > >> All, > >> > >> On Mon, Jan 4, 2016 at 4:06 PM, Anthony Ferrara <ircmax...@gmail.com> > >> wrote: > >> > >> There has been some discussion asking for a split of the RFC into two. > >> I do not believe that this is a good idea, because the CoC is useless > >> without > >> some sort of resolution strategy (without *anything*). And if we do need > >> to > >> do something (which I firmly believe), then why not do it right the > first > >> time. I > >> am more than willing to evolve this proposal significantly (it's no > where > >> near a > >> final form). This discussion should help it evolve. > > > > First, I firmly believe that having a CoC - without anything extra - is > > anything but useless. Values go a long way. Telling people what you > expect > > of them isn't only the first step towards obtaining that behavior - it's > by > > far the most important step. I suspect anybody who has kids (or that > has a > > reasonably fresh memory of being a kid himself) should be able to vouch > for > > that, and again, I'm bringing up the thesis that the vast majority of us > > here follow the law not because we're afraid of what would happen if we > > don't - but because it's the right thing to do. > > We already have that: https://lwn.net/Articles/452278/ > > The point is many people believe that does not constitute a code of > conduct. It is a worth while thing to have, but it doesn't make the > assurances to others that the project takes bad behavior, harassment > and discrimination seriously. > > That is not why it's not a Code of Conduct. A Code of Conduct does not inherently have to include assurances for what happens if you don't follow it. That's almost by definition outside the scope of the Code itself. One of the most famous codes in civilization, the ten commandments, has no penalties in it (although it's perhaps the author went out of writing space :) The reason Rasmus' email is not a Code of Conduct - or at least not a sufficient one - is that it covers just one issue out of many that can occur. Which is precisely why adopting a wider CoC makes sense. > And I agree with you about not doing something because it isn't right. > However, I'm not attempting to codify what's "right" here. Instead, > it's about communicating to others that we take these things seriously > and hence hold each other to a standard. > Having a CoC which is wider in scope and ratified by a voted RFC rather than an email on some mailing list sends a strong message. Having it in our contributor guidelines would also go a long way. I guess here we fundamentally disagree - it seems that sending the message that 'we take this seriously' - by placing strong emphasis on reporting and penalties - is more important to some than agreeing about the values themselves. For me, the values themselves and communicating them properly and prominently are infinitely more important than the policing mechanism, as I believe that stating them clearly would go a very long way and is anything but useless. > And if we don't have any means at all of holding ourselves to said > standard, what use is the standard? I, for one, believe that setting expectations is one of the most important things in life and minimizes friction tremendously. Just by setting expectations, nothing else, humans can work and interact much better with each other. Agreeing on a standard sets expectations, and while it may seem magical - it can absolutely improve the situation, simply because people would know what's expected of them, and what's unacceptable. Secondly, I'm not against having a mediation team - ad-hoc or otherwise - but giving it powers, and codifying what should be an extreme case - is a very slippery slope. > > Secondly, if we do want to add an extra layer, having a resolution > strategy > > does not have to include penalties - neither proposed ones nor the > > jurisdiction to impose ones. If the RFC stopped at structuring how > people > > can bring up issues and have them discussed and mediated, I doubt the RFC > > would be nearly as controversial as it is right now. > > I think that the resolution strategy needs to have some sort of > penalty, up to and including removal from the project. Otherwise > what's the point of the resolution strategy? The worst thing we can do > is put up a resolution path that people just say "so? why should I > care?". Again, I think I see things differently. To me, that's like saying "What use is it telling my daughters they should always be polite and respectful to others, if I'm not threatening that they'll get punished otherwise?". At least the types of mediation I know - mediation is not at all like a pseudo court. It's about mediation, and hence, has no power to force either side to do anything. I would argue that if it did - the chances for successful mediation go down tremendously for psychological reasons - both of the mediators and the subjects. Of course, our challenge is that unlike mediation, where you have the option of going to court if mediation fails - we don't have a very good conflict resolution mechanism, short of a public vote. But should an extreme case of an extreme case (gross violation followed by complete failure of mediation) dictate our mechanism? I don't think so. Here, the fact that even if PHP isn't free of harassment - it's certainly not an epidemic - should dictate which direction is more sensible. If it was an epidemic - I might have thought differently. > > The problems begin as soon as we try to create some sort of a > > mini-judicial-body, that has substantial powers, governs based on > loosely > > written rules, has zero tools and experience in getting to the bottom of > > things or determining the truth between two or more quarrelling parties. > > Thinking we can do that when we failed agreeing on infinitely simpler > things > > is remarkably optimistic. > > I'm not saying the current team I have proposed is good. I'm not > saying we need to be firm with it That's usually the problem. I very much respect the fact that you're very open to feedback and have modified your original RFC substantially and realize how difficult it is. But the problem is that what you're trying to solve is simply too complex. Sorry for sounding like a broken record, but we're not legislators nor lawyers, and no matter how much we work on that - whatever structure we come up with to enforce conflict resolution is going to be riddled with holes and fail or be abused in unpredictable ways sooner or later. > > However, I think time and time again it's been proven that the court > of public opinion is a poor judge of these types of situations. The > recent edits that I have been making to the RFC reflect the reduction > in power of the team significantly. What I do want to keep is a safe > and private place for these resolutions to occur in. > And like I said, I think the newer drafts are way better than the original one. But I still think that attempting to codify the response - beyond having a mediation team - whose job is exclusively to mediate - would bring a lot more bad than good > In extremely significant cases decisions will need to be public, but > with a private team like this at least the information gathering step > can be done in a non-biased manner with a team. That will also happen with a mediation team. If mediation fails - and again, I see no reason to believe this is going to be anything but an extremely extreme case - we don't have good options beyond the court of public opinion, as much as I agree with you it can sometimes be a poor judge (heck, it voted in favor of STH... JOKE!) > > I disagree we NEED to do something. PHP is not in a situation where > it's in > > an absolute need of a CoC, and the fact it's thriving without one and > that > > nobody appears to be coming up with examples as to why we must have one > > beyond future-proofing attests to that. Yes, it's not perfect - but as > Stas > > said, that RFC isn't a magic wand that would make it perfect. That > said, I > > think adopting a CoC is a good idea, much like I teach my daughters > what's > > right and what's wrong without telling them what would happen if they > don't > > follow my guidance. Whenever I have to resort to penalties (which I'm > happy > > to say rarely happens) - I've failed, and I virtually always regret it. > > I don't believe we literally need to do something in the sense that > the project will die if we don't. With that said, I do believe that > adopting the right one will do a lot of good for the project and > community. So it's not a life or death need, I would say it's > something we should definitely try to do. Another way to look at it is that if we adopt a CoC that stops at mediation, we can use it for a couple of years and see how it goes. We wouldn't be standing out as the first or second or 1000th project to go down that route. It's very common. If it fails, we can always vote to beef it up. This doesn't work in the opposite direction - once we establish a body with bylaws and structure and code, it'll be almost impossible to undo it - unless it fails spectacularly and with very clear evidence - while it's more likely to fail silently with little evidence. > > I'm still interested in hearing more about the four explicit threats of > > violence you mentioned. > > As I said before, I do not wish to discuss my personal matters in > public. I only said that because there was implication on list that > nothing has ever happened before, and I was showing that just my > experience should act as a counterpoint to that. > I missed that, and I fully respect your right to privacy. The reason I wanted you to share this is that one of the key issues for opponents of the 'toothful' RFC is that the same dry facts can be perceived by one side as X and the other as Y. What one calls harassment - another may call argument. What one may call bullying - another may call discussion. Threats of violence too can range from mild ("you should be banned from the project") to the extreme ("I know where you live and I'm going to kill you"). Thanks, Zeev
