Hi Den tir. 16. jun. 2020 kl. 00.41 skrev Deleu <deleu...@gmail.com>: > People arguing BC breaks without even knowing the scope of the change > clearly show biased.
I am sorry but I do not think you understand the scale of which the PHP project is at. Any change we make to the language has consequences for hundreds of millions of websites running PHP, potentially millions of developers who work with PHP and so on. Therefore any change that breaks backwards compatibility in any way has to be justified. We have a rather strict BC policy, something that allowed you and millions of others to easily upgrade from PHP5 to PHP7 with next to no changes for the most part. Do I personally believe that a change of name for some directives, potentially more, are justified? No I do not. That is my personal bias here. Let's assume that 10% of the current user base of PHP upgrades to whatever version a change like this is implemented. The number of work hours spent on investigating, updating, testing and patching these BC breaks which are changes for the sake of change is a crazy amount of hours invested into it. Opcache is a very popular extension, changing an ini directive means change of build systems, you can certainly argue that these changes could potentially just be changed by a tool, but even doing so will have cost a substantial amount of hours to implement and test. It is easily in the thousands of hours, a normal work year for me is about 1900 hours in terms of hours for just one person. Demanding that our users should invest so many hours besides the usual amount for already upgrading to a PHP version is lunacy, especially if the change is to try censor something that has no correlation to any racial slurs. So to say that the arguments about BC breaks (which I believe I was the only one to post about in this thread) without knowing the scope of the change is void. Yes, any policy for backwards compatibility breaks can easily be classified as biased, because they are an opinion of the project as a whole, or rather, a policy. > As white men, we're being dismissive, insensitive and strongly suggesting > we don't want change. While people may not feel offended by any of these > terms being discussed, this thread alone already serves as reason for > people to feel like there's no room for diversity in the internal community > of php. The "we" in this is extremely biased, it attempts to force me to feel as an inferior human (to steal the term from Larry above), because I do not agree with your request for a change. The classification you just did there is something I personally would feel offended by, because you attempt to use my ethnicity as an argument for why I feel the way I feel. > I believe that if we cannot come together to take the small (potentially > insignificant) step towards making changes that signal a welcoming > environment, how are we going to actually take the big steps? We could start by taking steps that matters for once, censoring words that have no correlation to any racial issue because it might offend someone because it has the word black in it. What about whitespace? Am I a nothing, an empty space just because I am caucassian? There are other issues we should tackle to make PHP better, after all, we have a major version in the works, set to release later this year. Something (excuse my bias here) is way more important than trying to justify backwards compatibility breaks for no reason. -- regards, Kalle Sommer Nielsen ka...@php.net -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
