Welcome! Log In Create A New Profile

Advanced

[PHP-DEV] [RFC][Discuss] Increase non-syntax runtime-impacting RFC voting threshold to 60%

Posted by Sara Golemon 
https://wiki.php.net/rfc/rfc.voting-threshold

This topic has come up on the mailing list a few times, so I'd like to
formally open the topic for discussion.

I'm generally pretty liberal when it comes to allowing the PHP
language to evolve and explore its identity, but the truth is a
feature that has 30 people vote against it and 31 people vote in favor
of it is not a mandate by any stretch of the imagination. It's an
opportunity to examine why a divide exists and if we're all being
honest with each other, improve the original idea before it becomes a
maintenance burden.

Please note the "Open Question". I'm not all that sure 60% is enough
of a mandate either, but I wanted to be conservative in my
conservatism. If folks think 2/3 is more appropriate (and consistent
with syntax changes), I'm happy to change this number before we move
to voting phase.

-Sara

Or, as Ze'ev once famously said, "Give the language a rest".

--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
On 13.09.2017 at 22:42, Sara Golemon wrote:

> https://wiki.php.net/rfc/rfc.voting-threshold
>
> This topic has come up on the mailing list a few times, so I'd like to
> formally open the topic for discussion.
>
> I'm generally pretty liberal when it comes to allowing the PHP
> language to evolve and explore its identity, but the truth is a
> feature that has 30 people vote against it and 31 people vote in favor
> of it is not a mandate by any stretch of the imagination. It's an
> opportunity to examine why a divide exists and if we're all being
> honest with each other, improve the original idea before it becomes a
> maintenance burden.

Thanks for taking the initiative on this!

FTR: Zeev has also started preparing an RFC which includes this voting
threshold, besides further issues: https://wiki.php.net/rfc/voting2017.

> Please note the "Open Question". I'm not all that sure 60% is enough
> of a mandate either, but I wanted to be conservative in my
> conservatism. If folks think 2/3 is more appropriate (and consistent
> with syntax changes), I'm happy to change this number before we move
> to voting phase.

IMHO, a 2/3 majority would be most suitable for any changes to php-src.
Most votes have even been clearer, and I believe most (if not all) which
would have failed a 2/3 threshold would have failed 60% as well. (Zeev
presented more detailed stats on this list a while ago.) Having a 2/3
threshold for all php-src change related votes would at least avoid the
discussion into which category the vote falls, though.

--
Christoph M. Becker

--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
> -----Original Message-----
> From: Christoph M. Becker [mailto:[email protected]]
> Sent: Thursday, September 14, 2017 12:25 AM
> To: Sara Golemon <[email protected]>; PHP internals <[email protected]>
> Subject: [PHP-DEV] Re: [RFC][Discuss] Increase non-syntax runtime-impacting
> RFC votingthreshold to 60%
>
> On 13.09.2017 at 22:42, Sara Golemon wrote:
>
> > https://wiki.php.net/rfc/rfc.voting-threshold
> >
> > This topic has come up on the mailing list a few times, so I'd like to
> > formally open the topic for discussion.
> >
> > I'm generally pretty liberal when it comes to allowing the PHP
> > language to evolve and explore its identity, but the truth is a
> > feature that has 30 people vote against it and 31 people vote in favor
> > of it is not a mandate by any stretch of the imagination. It's an
> > opportunity to examine why a divide exists and if we're all being
> > honest with each other, improve the original idea before it becomes a
> > maintenance burden.
>
> Thanks for taking the initiative on this!
>
> FTR: Zeev has also started preparing an RFC which includes this voting
> threshold, besides further issues: https://wiki.php.net/rfc/voting2017.


Oh wow, I was hoping to bake it for a while longer before public scrutiny :) Still a lot of work to go on it.

Realistically I'm only going to bring it up for discussion sometime around late October because I'm actually going to be off the grid for most of the 2nd half of September and most of October.

> IMHO, a 2/3 majority would be most suitable for any changes to php-src.
> Most votes have even been clearer, and I believe most (if not all) which would
> have failed a 2/3 threshold would have failed 60% as well. (Zeev presented
> more detailed stats on this list a while ago.) Having a 2/3 threshold for all php-
> src change related votes would at least avoid the discussion into which category
> the vote falls, though.

I agree, and the only exception that may make sense is the addition of functionality under an extension's namespace/pseudo namespace. E.g., I'm not sure we need a 2/3 vote for a new oci8_*() function, or a new method added to ext/mysqli. There are no downwards compatibility considerations, and it seems reasonable enough that the subject matter experts (the extension maintainers) will be given jurisdiction here. To be honest, I'm not sure we need a vote at all for those, 2/3 or otherwise. But this is pretty much the only example I can think of.

That said, I do think that if we finally have the mental strength and stamina to tackle the laconic 2011 Voting RFC, we should tackle it more thoroughly and try to solve as many of the issues that came up over the years. This is what I'm attempting to do in the RFC I started drafting a couple of days ago - and that I'm NOT YET PUBLICLY DISCUSSING :)

Zeev
On 13.09.2017 at 23:38, Zeev Suraski wrote:

>> -----Original Message-----
>> From: Christoph M. Becker [mailto:[email protected]]
>> Sent: Thursday, September 14, 2017 12:25 AM
>> To: Sara Golemon <[email protected]>; PHP internals <[email protected]>
>> Subject: [PHP-DEV] Re: [RFC][Discuss] Increase non-syntax runtime-impacting
>> RFC votingthreshold to 60%
>>
>> FTR: Zeev has also started preparing an RFC which includes this voting
>> threshold, besides further issues: https://wiki.php.net/rfc/voting2017.
>
> Oh wow, I was hoping to bake it for a while longer before public scrutiny :)

If you want to hide changes from https://wiki.php.net/feed.php, you
have to mark them as "Minor Changes" (right below the "Edit summary"). :)

--
Christoph M. Becker

--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
On 14 Sep 2017, at 1:03, Christoph M. Becker <[email protected]<mailto:[email protected]>> wrote:

On 13.09.2017 at 23:38, Zeev Suraski wrote:

-----Original Message-----
From: Christoph M. Becker [mailto:[email protected]]
Sent: Thursday, September 14, 2017 12:25 AM
To: Sara Golemon <[email protected]<mailto:[email protected]>>; PHP internals <[email protected]<mailto:[email protected]>>
Subject: [PHP-DEV] Re: [RFC][Discuss] Increase non-syntax runtime-impacting
RFC votingthreshold to 60%

FTR: Zeev has also started preparing an RFC which includes this voting
threshold, besides further issues: https://wiki.php.net/rfc/voting2017.

Oh wow, I was hoping to bake it for a while longer before public scrutiny :)

If you want to hide changes from https://wiki.php.net/feed.php, you
have to mark them as "Minor Changes" (right below the "Edit summary").

I'm not into hiding, it's not confidential, just wasn't quite ready for the spotlight just yet, especially as I won't be around to discuss it for over a month :)

Zeev
On 14.09.2017 at 00:08m, Zeev Suraski wrote:

> On 14 Sep 2017, at 1:03, Christoph M. Becker <[email protected]<mailto:[email protected]>> wrote:
>
> On 13.09.2017 at 23:38, Zeev Suraski wrote:
>
> -----Original Message-----
> From: Christoph M. Becker [mailto:[email protected]]
> Sent: Thursday, September 14, 2017 12:25 AM
> To: Sara Golemon <[email protected]<mailto:[email protected]>>; PHP internals <[email protected]<mailto:[email protected]>>
> Subject: [PHP-DEV] Re: [RFC][Discuss] Increase non-syntax runtime-impacting
> RFC votingthreshold to 60%
>
> FTR: Zeev has also started preparing an RFC which includes this voting
> threshold, besides further issues: https://wiki.php.net/rfc/voting2017.
>
> Oh wow, I was hoping to bake it for a while longer before public scrutiny :)
>
> If you want to hide changes from https://wiki.php.net/feed.php, you
> have to mark them as "Minor Changes" (right below the "Edit summary").
>
> I'm not into hiding, it's not confidential, just wasn't quite ready for the spotlight just yet, especially as I won't be around to discuss it for over a month :)

I didn't want to bring it into the spotlight – I just wanted to avoid
having two partially overlapping RFCs going unnoticed by the respective
authors. That's also why I've snipped most parts of your previous
reply, even though I'd have to say something about it. :)

--
Christoph M. Becker

--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
On Thu, Sep 14, 2017 at 4:42 AM, Sara Golemon <[email protected]> wrote:

> https://wiki.php.net/rfc/rfc.voting-threshold
>
> This topic has come up on the mailing list a few times, so I'd like to
> formally open the topic for discussion.
>
> I'm generally pretty liberal when it comes to allowing the PHP
> language to evolve and explore its identity, but the truth is a
> feature that has 30 people vote against it and 31 people vote in favor
> of it is not a mandate by any stretch of the imagination. It's an
> opportunity to examine why a divide exists and if we're all being
> honest with each other, improve the original idea before it becomes a
> maintenance burden.
>
> Please note the "Open Question". I'm not all that sure 60% is enough
> of a mandate either, but I wanted to be conservative in my
> conservatism. If folks think 2/3 is more appropriate (and consistent
> with syntax changes), I'm happy to change this number before we move
> to voting phase.
>
> -Sara
>
> Or, as Ze'ev once famously said, "Give the language a rest".
>

+1 on this, though I would strongly recommend to use a 2/3 threshold for
all RFCs, be they language, library or procedural. We're having this
discussion on nearly every single RFC (seriously, even the UUID RFC which
is as non-language as these things get had arguments about this) and this
would be a good chance to simplify the rules.

I would also explicitly note that the voting threshold applies to the
primary RFC vote only, while secondary votes are simple majority votes.

Nikita
Le 13/09/2017 à 22:42, Sara Golemon a écrit :
> https://wiki.php.net/rfc/rfc.voting-threshold
>
> This topic has come up on the mailing list a few times, so I'd like to
> formally open the topic for discussion.
>
> I'm generally pretty liberal when it comes to allowing the PHP
> language to evolve and explore its identity, but the truth is a
> feature that has 30 people vote against it and 31 people vote in favor
> of it is not a mandate by any stretch of the imagination.

+1, such result show we fail to have a good consensus,
and that ~50% will be unhappy.

So indeed, having a higher threshold seems a good idea.

Remi

> It's an
> opportunity to examine why a divide exists and if we're all being
> honest with each other, improve the original idea before it becomes a
> maintenance burden.
>
> Please note the "Open Question". I'm not all that sure 60% is enough
> of a mandate either, but I wanted to be conservative in my
> conservatism. If folks think 2/3 is more appropriate (and consistent
> with syntax changes), I'm happy to change this number before we move
> to voting phase.
>
> -Sara
>
> Or, as Ze'ev once famously said, "Give the language a rest".
>
Whoops, just noticed I sent this to Zeev directly last night rather than
the list.

---------- Forwarded message ----------
From: Sara Golemon <[email protected]>
Date: Wed, Sep 13, 2017 at 10:28 PM
Subject: Re: [PHP-DEV] [RFC][Discuss] Increase non-syntax runtime-impacting
RFC voting threshold to 60%
To: Zeev Suraski <[email protected]>


On Wed, Sep 13, 2017 at 5:00 PM, Zeev Suraski <[email protected]> wrote:

> Remarkable timing. In the last few days I’ve been working on a completely
> rewritten RFC workflow & voting RFC. It’s a bit premature, but given you
> just brought up on of the main points - I want to share it with you.
>
> Ah! Great minds think alike, and apparently so do we. Looks like you've
put quite a lot more thought and time in, and a replacement standard would
leave less ambiguity than a mere update, so I'll withdraw the one I put up
and wait for you to formally open discussion on yours.


> Note - I’m not going to bring it up for discussion in the immediate future
> - because I’m going on a 3.5wk vacation starting next week - but if all
> goes well I think I’ll bring it up for discussion towards the end of
> October.
>
> Yeaaaahhhh, don't you have a conference coming up as well?



> Can you take a look?
>
> https://wiki.php.net/rfc/voting2017
>
> Again, won’t be publishing it in the immediate future so even if you
> absolutely hate it, be sure to tell that just to me
Hi!

> I would also explicitly note that the voting threshold applies to the
> primary RFC vote only, while secondary votes are simple majority votes.

This sounds like a good compromise.

--
Stas Malyshev
smalyshev@gmail.com

--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
Sorry, only registered users may post in this forum.

Click here to login