2015-02-19 16:51 GMT+02:00 Pierre Joye <pierre....@gmail.com>:

> On Thu, Feb 19, 2015 at 6:33 AM, Zeev Suraski <z...@zend.com> wrote:
> >> -----Original Message-----
> >> From: Pierre Joye [mailto:pierre....@gmail.com]
> >> Sent: Thursday, February 19, 2015 4:09 PM
> >> To: Zeev Suraski
> >> Cc: Anthony Ferrara; PHP internals
> >> Subject: Re: [PHP-DEV] Using Other Channels (was Scalar Type
> Declarations
> >> v0.5)
> >>
> >> We have seen off list discussions or pressures many times in the pasts.
> I
> >> have
> >> (other too but I can only talk for myself) been said an insane amount of
> >> times
> >> to stop private discussions, for anything related to php core. There is
> no
> >> exception to this rule. I repeat:
> >> There is NO exception to this rule.
> >
> > I disagree.  Completely.
>
> I did not expect you to agree. I would be surprised if you do.
>
> > I think 1:1 or group discussions are completely legitimate, and they're
> not
> > only legitimate - they can be very productive.
> >
> > There's a huge gap between making decisions in closed groups, or doing
> 'arm
> > bending' - and private discussions, which I repeat, are completely
> > legitimate.
> > Private discussions happen all the time, in countless mediums.
> Conferences,
> > emails, IRC (public in theory, private in practice), Twitter DMs and
> more.
> >
> > There's absolutely nothing wrong with discussing an idea directly with
> one
> > or more people before bugging thousands of people with it.  They'd all be
> > better served if the idea presented to them already had a slightly more
> than
> > just one person standing behind it (which is an inherent side effect of
> > barring private discussions as you suggest), and had some of the initial
> > issues weeded out.
>
> To discuss at an idea or concept at events and co? Indeed. We all do
> that. The difference is how to move it to a group discussions and grab
> other people thoughts to actually get it done, with consensus. The
> latter part is totally absent using your process.
>
> Discussing,  working, implementing something for weeks or months
> privately? NDA and all that? Sorry, this is not the PHP I want.
>
> Weeding out the initial issues? I wonder which magic you use to be
> able to figure out all issues based on the experiences or thoughts of
> a couple of people. It does not work. What you say is that a very
> small group is able to pin point the perfect proposal better than
> actually discussing it on the open channels. This is wrong in so many
> ways.
>
> Cheers,
> Pierre
>
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: http://www.php.net/unsub.php
>
>
I think this starts to go the route of putting things into absolute. Ideal
things tend not to happen/work in the real world to the letter.

Some things just don't work out by themselves. The Type Hinting RFC's are
an anomaly in the regular PHP Core workflow and need some creative
handling. My personal view is that the RFC requires a mediator - the
person. who is more or less impartial to what type of hints get into PHP
and is able to handle communications between groups and bring them
together. RFC needs a triage. It needs someone to make a game plan and
implement it. Not someone developing one or the other proposal, but a
strictly managing entity. Because the further we go, the more things are
need to be taken care of. I believe the Type Hint project reached the
point, where it can't be handled by a single person. It touches into some
other RFC's in one or the other way, sometimes maybe requiring some
adjustments in the related RFC or the type hints itself.

Also, whatever way it goes at this point, the RFC's are not going to make
into 7.0 if we follow the release process properly. So, there needs some
decision making to happen considering the PHP itself, like reserving
keywords for 7.0, doing typehints for 7.1 and stuff alike.

Maybe it does not look like I have described it from inside, but from the
side it looks like a disaster is brewing and it's gonna blow any second now
leading to rushed decisions that may screw up things. I saw similar
situations a few times in my carrier and I certainly have that feel right
now. I'd give it 50/50 chance that something will go wrong here.

Reply via email to