RE: [PHP-DEV] A call for help (urgent)
| From: | Zeev Suraski | Date: | Sat, 17 May 2014 18:43:29 +0000 |
| Subject: | RE: [PHP-DEV] A call for help (urgent) | ||
| References: | 1 2 3 4 5 6 | Groups: | php.internals |
| Request: | Send a blank email to internals+get-74286@lists.php.net to get a copy of this message | ||
> -----Original Message-----
> From: bruno@chalopin.fr [mailto:bruno@chalopin.fr]
> Sent: Saturday, May 17, 2014 9:31 PM
> To: PHP internals
> Subject: Re: [PHP-DEV] A call for help (urgent)
>
>
> Le 17 mai 2014 à 18:03, Zeev Suraski <zeev@zend.com> a écrit :
>
> > But much
> > like I wouldn't dream of telling the docs team how they should do
> > their job, I don't expect the opposite to happen either.
>
> So if I understand you suggest that only core developers should decide
for core
> modifications ? Isn’t it against the voting RFC which you’re one of the
authors ?
It may be against what’s written there (which is why I need to beg as
opposed to just point people to it) - but not its spirit.
The voting RFC was written with language functions, features and processes
in mind. Not implementation. Things like Traits - which most certainly
involves Core, or return type hinting - these are the things the Voting
RFC was created for. As you pointed out, I'm one who those who wrote much
of the text of it, I should know.
I didn't even dream that we'll have RFCs that deal with low-level
implementation up for vote. If I did, there'd be a separate dedicated
section for it.
> That’s a really weird vision of an open source community for me.
It actually isn't. Open Source is typically some form of meritocracy, and
those who have merit in a given component are those who own it. The RFC
process was designed to give a wider range of 'stakeholders' a say in the
language direction - and I think it greatly helped PHP in the last few
years. However it was never designed to give those stakeholders a say
about its implementation. Unfortunately, I wasn't sufficiently
forward-looking to make that clear in the RFC. I still think this can
work without changes to the Voting RFC, again, assuming people do not vote
on implementation RFCs that don't directly concern the parts of the code
they own.
Zeev