Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:106219
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
	s=turbo-smtp; d=php.net;
	h=Received:Received:X-TurboSMTP-Tracking:X-Gm-Message-State:X-Google-Smtp-Source:X-Received:MIME-Version:References:In-Reply-To:From:Date:X-Gmail-Original-Message-Id:Message-ID:Subject:To:Content-Type;
	b=DU+4yY3W10GXCKAzFZS5JNjjPlu0O6tuCt9JirQI7VMxh5df4L+c0jIun4DQ/C
	+xLwN+rx+X/O7+QJsJoqRkgR1lpwN1YAQzUAtmqwWlbdR7ctado/3rpfo/o6ddK9
	ylIG+RZX+6YBuZ+LTPUGv/D3H/d2BqWlAW/BS1tCsa4LI=;
MIME-Version: 1.0
References: <2311901d53767$1c5aa780$550ff680$@gmail.com>
In-Reply-To: <2311901d53767$1c5aa780$550ff680$@gmail.com>
Date: Tue, 16 Jul 2019 07:18:33 -0700
Message-ID: <CAJHtznx7u4wKDEjjARCtNd9o-HxCDzd_qERC0Js480sJ+Z9q4A@mail.gmail.com>
To: PHP internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary="000000000000db9397058dcd0d86"
Subject: Re: hebrevc() and other 'contentious' 7.4 proposed deprecations
From: zeev@php.net (Zeev Suraski)

--000000000000db9397058dcd0d86
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Given apparently nobody has paid any attention to this email (both in terms
of my support of deprecating hebrevc(), and my request to reconsider
supporting proposals with substantial numbers of 'nay' voters) - I'm
resending it one more time for consideration:

On Wed, Jul 10, 2019 at 2:33 PM <zeev@php.net> wrote:

> Two separate topics on this message:
>
>
>
> First =E2=80=93 I wanted to point out that my fierce defense of the hebre=
v()
> function does not in fact extend to hebrevc().  As much as I think the RF=
C
> was really wrong about hebrev(), and we got scarily close to deprecating =
a
> functionality that =E2=80=93 while somewhat esoteric =E2=80=93 can be ext=
remely useful
> and cannot be easily replicated in any way =E2=80=93 I have to say that I=
 think
> the RFC is pretty much correct on hebrevc().  I don't think it's very
> plausible hebrevc() is still in use today =E2=80=93 and even if we're mis=
sing
> something and it is =E2=80=93 it can be implemented in a one liner with 1=
00.00%
> compatibility.  While I don't think it brings much value to deprecate it =
=E2=80=93
> perhaps sending the message that you shouldn't be using it for HTML bears
> *some* level of value.  I voted in favor.
>
>
>
> Now, with that said =E2=80=93 I would *really* encourage everyone who vot=
ed on
> this RFC (as well as ones who haven't) to take a look at what I would cal=
l
> the 'contentious votes' in there.  In a nutshell, votes with a substantia=
l
> amount of people voting against the deprecation.  If you voted 'yes' for
> one of these =E2=80=93 please consider, for a moment, whether your positi=
on on it
> is "It's evil, I really think we're better off without it" or whether it'=
s
> more of a "I don't think it's very useful".   If it's the former =E2=80=
=93 by all
> means, keep your vote.  But if it's the latter =E2=80=93 please consider =
the
> possibility that the fact that a substantial number of people feel strong=
ly
> enough about keeping it to vote against the deprecation (and let's admit =
it
> =E2=80=93 against the odds), may mean it is, in fact, useful =E2=80=93 ev=
en if you don't
> find it useful yourself.
>
>
>
> While we can argue whether consensus-based voting makes sense for votes i=
n
> general, I think it's tenfold more important when dealing with
> deprecations.  If there's a substantial minority that thinks a feature is
> still useful =E2=80=93 we should keep it =E2=80=93 unless there's a real =
tangible cost
> associated with keeping it.  For most of the proposed deprecations =E2=80=
=93 that
> cost is simply not there.
>
>
>
> For reference, this is what consensus looks like:
>
> https://www.dropbox.com/s/asfgt98rss3xyw2/consensus.PNG?dl=3D0
>
> https://www.dropbox.com/s/iia7ua4xh6bihe3/consensus2.PNG?dl=3D0
>
>
>
> And this is what it doesn't look like:
>
> https://www.dropbox.com/s/56jdl2v1lpxba49/no-consensus.PNG?dl=3D0
>
> https://www.dropbox.com/s/hj8jozuun7a4w42/no-consensus2.PNG?dl=3D0
>
>
>
> To connect with the first point =E2=80=93 the hebrevc() vote certainly lo=
oks a
> lot more like the latter than the former, but I do believe it's mostly
> related to confusion with hebrev() and as the author of both =E2=80=93 I =
feel
> comfortable supporting its removal :)
>
>
>
> Thanks for your consideration,
>
>
>
> Zeev
>

--000000000000db9397058dcd0d86--