Hi! > Even without that, though, it's clear we *do* have more serious issues > than just "rudeness". When a major contributor is getting death-threats > over an RFC, *there is a problem*. That they're happening off-list > doesn't change the fact that *that is a problem*.
OK, so to evaluate solution to a problem we need to see: 1. There is a problem 2. Solution is possible to implement 3. Solution solves the problem. 4. Solution does not produce the effects worse than the original problem Now, do we have the problem that internals is not the nicest place in the world? Definitely. Does CoC as solution solve it? Possibly, if we apply it really extensively and ban all people that cause anybody to feel any discomfort. That would kill any substantial discussion on the list. Do we have a problem with harassment outside internals (taken broadly)? We do. Can we make CoC that would prevent it? Nope. So, we have a situation where we have a mismatch between a problem and a solution, and that is what the misunderstanding is based on. You and several other people try to prove something we already agree about - that certain problems exist - and forget to prove something that needs to be proven - that what you propose would solve *these* problems in any acceptable way. Instead, the solution (at least part of it) is designed to solve *different* problems, which nobody showed we even had. This mismatch is an issue. > with the risk of those tools being abused. It's not just "it's too > dangerous", but "it's so dangerous that we'd rather have the current > problem." That is, that current problems are tolerable. They are "tolerable" by definition, since we are tolerating them right now :) That, of course, does not mean improvement can't be made. But for that, we need to actually see the path to improvement, not just "do something because something has to be done". > The other "contra" position is to make a CoC toothless. The argument CoC can not have any tooth per se. It's just a promise, as I said. Promise does not enforce itself. People can enforce promise, in different ways. These ways are completely separate from the promise, and I think there's a lot of value in the promise itself. In fact, I think it is a much more significant step than figuring out how to punish people that break the promise. > I'll take that a step further: Having a CoC with no teeth has a higher > risk of abuse than it having teeth, because those who would abuse it can > use that lack of teeth to their advantage. If you talking about insulting people on twitter and reddit, I do not see how any tooth to CoC would change anything. We can't ban people from twitter and reddit (thankfully). > At the same time, though, if someone is being maliciously hostile what > great cover! A private email is not a PHP-Group managed resource, so no > rules! Twitter, ha, no rules! Reddit? LOL like they enforce That's not true. The fact that PHP community does not enforce these rules, does not mean there are no rules at all. It just means we do not have responsibility to enforce them. We are not Team PHP World Police. I could be OK with looking into matters directly related to RFCs and alike discussions - but phrasing like "PHP business" open to obvious trolling like "what do you mean heinous acts of $Person to support position $whatever is not your business? Are you $whatever-ist? Obviously you are, and I just finished an article for $MajorNewspaper declaring PHP Group is a nest of $whatever-ists and I'll click "Send" unless you agree to make it your business right this second". So we need to be clear we never promised to get into this. > infrastructure". It's trivial to circumvent otherwise. Now, how do we It is trivial to circumvent anyway. Twitter and reddit as both pseudonymous. > Let's all focus on maximizing the benefit and minimizing the risk. > Pretending that the status quo is oh-so-wonderful accomplishes neither > goal. I don't think anybody pretends oh-so-wonderful. But, see the four points above. Having *some* solution is not enough. It needs to be *good* solution. Going back to the pill analogy, if you're sick, raiding the medicine cabinet and trying random pills may not be a good idea. And saying that does not mean denying that somebody feels unwell - it just means finding the right pill is a good idea before swallowing it. -- Stas Malyshev smalys...@gmail.com -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php