Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:109093
Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 12.4 \(3445.104.11\))
In-Reply-To: <CA+kxMuRkCZjWbuR1yrw8KxcM20ZxZ-7Zu8s=RAxbspLkGsjO=Q@mail.gmail.com>
Date: Mon, 16 Mar 2020 23:01:31 -0400
Cc: PHP Internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID: <2AE639B7-7775-48CC-8435-2F455D429E6A@newclarity.net>
References: <14383D05-EA33-4CD2-9648-40AA29A837A5@newclarity.net>
 <CA+kxMuRkCZjWbuR1yrw8KxcM20ZxZ-7Zu8s=RAxbspLkGsjO=Q@mail.gmail.com>
To: Dan Ackroyd <Danack@basereality.com>
Subject: Re: [PHP-DEV] Capturing reasons for votes for historical sake?
From: mike@newclarity.net (Mike Schinkel)

> On Mar 16, 2020, at 9:20 PM, Dan Ackroyd <Danack@basereality.com> =
wrote:
>=20
> On Mon, 16 Mar 2020 at 18:29, Mike Schinkel <mike@newclarity.net> =
wrote:
>>=20
>> Would it be possible to add a feature when voting were people either =
need to type in
>> a one to two sentence reason why they voted "no" on a proposal
>=20
> Pretty strong no for me.

Thank you for your reply.  Let me start with the end of your reply, what =
you state was your biggest issue:

> And the big issue, where possibly I disagree with your reasons for =
wanting this:
>=20
>> and their reasons are really important when it comes
>> to future consideration of the same issue
>=20
> This kind of makes an assumption that all RFCs should have some path
> to being passed.
>=20
> For some RFCs, there just isn't a good path forward for them.

Actually no.  You are assuming wrong.

If we had history we could better see the RFCs that had valid reasons =
for not passing and that should not be revisited vs. the ones that =
didn't pass for some technicality and didn't pass because the timing was =
not right that should be revisited. =20

If we knew WHY RFCs were turned down then the next time someone brings =
the issue up again =E2=80=94 which we they will =E2=80=94 we could say =
"Here, read this. Chances are this idea will never pass for the reasons =
explained, so don't waste all of our bandwidth (unless you can counter =
all those objections, which you probably cannot.)"

So actually part of what I am asking for is a way to put the bad ones =
down, and for good.

> btw I think there's a bigger problem on some RFCs being accepted,
> particularly where people who aren't core maintainers are voting on
> things that really need an informed understanding of internals.

I can certainly understand that perspective. But what's the proposal to =
improve that situation? =20

Actually, I think you are under-appreciating what I proposed. Consider =
if instead of a "Yes/No" prompting people to vote, there was a list of =
reasons to vote "no" from those who already voted no. Isn't it likely =
that people who got to review that list would be more likely to vote =
"no" in the case of something that should not actually be accepted?

(To be fair, if we did that we'd probably need a list of reason why =
people votes "yes" too so that voters could see why others voted yes and =
no.)

(BTW, I do think having a good history of why things failed might keep =
bad ones from being approved in the future "accidentally.")

>> However in PHP we have no way of knowing why people voted against a =
proposal
>=20
> I think this type of analysis belongs off-list. Which is why:
> https://github.com/Danack/RfcCodex/blob/master/rfc_codex.md

That's nice, but requires superhuman efforts on one individual's part, =
is contained on the GitHub account controlled by one individual, and is =
inherently slanted by one individual's bias, even if only unconscious =
bias. =20

So it is really not a solution for the PHP community in general, =
notwithstanding that it is a nice resource that you have provided of =
your own industriousness.

> * Explaining exactly why you're voting no can be hard as there can be
> multiple overlapping reasons for voting no.

1. Nothing I proposed would stop someone from listing multiple reasons.  =
Free-form text would allow someone to write a dissertation, if they =
desire.=20

2. If someone votes "no" but cannot put it into words, it is really fair =
that are driving the destiny of PHP used by ~5 million programmers?

3. There is nothing that In I am proposing that would stop you from just =
typing in "I really don't like this" and leave it at that.

> * It sets up arguments about what is and isn't a valid reason for =
voting no.

Not sure how this is relevant.  Once it is voted down it doesn't matter =
if people felt the reasons were valid or not, it would still be voted =
down.

BTW, I am not proposing that the reasons be tied to individuals. Those =
would be separated from the vote and displayed anonymously.  So who =
specifically would they argue with?  Unless of course you are earnestly =
concerned about this syndrome: https://imgur.com/gallery/91sn32Q

> * It drives people who might want to vote no away from the project, if
> they have to take the extra time to justify their position.

That is why I proposed that they could select from the list of all other =
things people have provided. If someone does not want to type a reason =
then just pick from one of the reasons other people already submitted.

> * Shifts the part of the burden for coming up with a plan for how to
> get the RFC implemented onto people who are not in favour of the RFC.

Currently it takes herculean effort to get almost anything approved, but =
it takes effectively zero effort to stifle the hard work someone invests =
in trying to improve PHP. Is it really just that all their work can be =
nullified by a simple thumbs down like an emperor deciding the death of =
a gladiator? (sorry, couldn't resist using that analogy. ;-)

BTW, I am thinking of the outrageous amount of work Paul M Jones is =
putting into Server-Side Request and Response Objects (v2) and fear for =
him that all his effort will be for naught, and he won't even have a =
concise list of reasons why it was voted down. The best he will be able =
to do is infer from the comments in thousands of the messages why people =
voted down. But he still won't know.=20

Frankly it also feels to me that people seeing that will be a lot more =
likely to be be driven away from the project than the fact that people =
who have already gotten the right to vote on the project actually have =
to write a sentence or two to go along with their votes. I know that if =
people vote down his work after all his effort I will feel less like =
contributing for fear of the same outcome. But being told that I am =
expected to help future people by recording my reason to vote? Hell, I'd =
feel like I was being granted an honor and I would want to rise to the =
occassion.

It is almost like you are arguing the with great power comes no =
responsibility (sorry, couldn't resist using allusion, either. ;-)

> * The phrase "this is a terrible idea and I don't have enough crayons
> to explain why to you" would be more likely to be used, and wouldn't
> add much to the discussion.

In which case we would be no worse off than we are now.

However, if even only 1/3 the people who voted actually did provide an =
intelligible reason then we'd be much farther along to know how and why =
we should reconsider something in the future, OR NOT. =20

And this might stop a *lot* of wasted effort. Otherwise we continue to =
be "Those who are ignorant of the past are doomed to repeat it." =20

-Mike

P.S. Some modifications to the proposal might include:

1. As Jakob Givoni proposed: a list of predetermined reasons even the =
first voter course pick from a list , OR they would write their own if =
they want.

2. Create a two (2) phase process. =20

Phase 1: Solicit people enter a short concise reason to vote "no" or =
"yes", or to select to join one of the existing reasons. This is =
basically everyones way to lobby for yes or no.

Phase 2: Do the actually vote where people can select a reason from the =
list to go with their vote, or enter a new one.