Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:97800
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: error (pb1.pair.com: domain pthreads.org from 74.125.82.45 cause and error)
MIME-Version: 1.0
In-Reply-To: <4d79d06b-671e-02c8-0e8d-d019ff57b7e0@gmail.com>
References: <CAL=_i_nuF0y8TC_GrWsX5e77sP-jKq+ed374SW0e=VxSPniVjg@mail.gmail.com>
 <CAF+90c935-XMWDeyLEuw-Pr7Njh+Q=ZR7HT+isD9Rry-R-g2uw@mail.gmail.com>
 <f432ec0d-1836-1c5c-43d5-455c0257459e@heigl.org> <47370a17-d428-b30d-183b-c1ec96d47872@gmail.com>
 <CAL=_i_m4zegT41zOKc7+zXYN-487umJ9NKDM0KMKqMZhY81vVw@mail.gmail.com>
 <8307a455-1f57-b289-74cb-71c3a0410c92@gmail.com> <CAL=_i_k3q3-gpj2B9mS_NjsfxYjjRi0a9Wq6GHD3GEYexh=wQw@mail.gmail.com>
 <4d79d06b-671e-02c8-0e8d-d019ff57b7e0@gmail.com>
Date: Tue, 17 Jan 2017 08:19:46 +0000
Message-ID: <CAL=_i_neXs5qdsyKKRoT9aXr5xH=5H1BTunjk8O5=ZZzd-X7Pw@mail.gmail.com>
To: Stanislav Malyshev <smalyshev@gmail.com>
Cc: Andreas Heigl <andreas@heigl.org>, Nikita Popov <nikita.ppv@gmail.com>, 
	PHP internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary=f403045e9c969b8fa7054645f6ed
Subject: Re: [PHP-DEV] bugsnet cleanup
From: pthreads@pthreads.org (Joe Watkins)

--f403045e9c969b8fa7054645f6ed
Content-Type: text/plain; charset=UTF-8

Morning Stas :)

> True, but some people can not write an RFC, or not ready to do it,
> however they still can request and discuss features.

They can do that, but they can also start a discussion on internals if they
want to encourage someone else to do their work for them.

The RFC document already states that things that come without patches are
highly unlikely to happen, things that come without RFC (where they are
required) are even less likely to happen.

> I just think dropping all of them is going too far

Well, I'm not really suggesting that, I am looking for criteria to drop
those that are realistically never going to get attention, because they are
not applicable, or for other reasons. But I'm not suggesting we just drop a
bunch because we can't be bothered with them.

> I also disagree that bugs
> is not a useful tool, the problem is not the tool but that people are
> not using it.

Maybe it's not entirely useless, but it's not as useful as other tools we
have at our disposal. As already mentioned, the noise created on github
when pull requests are made is much more effective at generating
conversation than bugs on bugsnet.

We can either spend our time developing tools - note that the last commit
to bugsnet was by Rasmus to fix a "feature" that has *never* worked, a
feature that we need in order to do a good job, that is still incomplete -
useless is not too strong a word here - or we can spend our time developing
PHP, using tools developed by industry leaders in open source collaboration.

Sure, part of the problem is that people are not using bugsnet, but I do
see github as a partial solution to that - people are using github; you
open issues and pull requests and you make noise, you interrupt everyone's
day that is using github.

> Why not? It's a working system which we own and we can suit to our needs.

When things have been broken forever, and when there is no interest in
actually developing these tools, then the observation that we own it, and
that we can mould it to suit are needs are moot - these things are not, as
a matter of fact, relevant, precisely because it does not suit our needs,
and nobody is developing it to meet our needs.

> Source is not hosted on github, it's mirrored to github.

Aware :) But to the outsider, that doesn't matter, they can open a pull
request on github as they can for anything else hosted on github.

>  if anything it is missing a bunch of features we need and do have on
bugs.php.net

The only thing that is legitimately missing is a way to handle security
bugs, as far as I can see ... maybe I'm missing something obvious ?

While we can't assign bugs on github to anyone, we can ping people and get
their attention ... and assigning bugs on bugsnet doesn't seem to be all
that effective anyway. Some bugs have been assigned for years.

> Exactly. Including using existing system we have, not just dropping everything
and moving to github which doesn't have features we need.

The existing system we have is somewhere between broken and ineffective, in
my opinion. It's all fine to say that we should develop the tools we need,
but *nobody is doing, or has been doing that to an acceptable level*. We
have extremely limited resources, if anyone does come along that wants to
work on systems there are far more important things to fix than working on
ancient dilapidated software, things such as our mailing lists, lxr,
jenkins, and really important stuff are more deserving of attention, in my
opinion.

> I'm sorry I don't understand this. How bugs.php.net is not open?

I said it is open, but it's not being used, it's not in your face like
github is for a lot of us, it's in a "dark corner of the internet".

Maybe it was too early to suggest we drop it completely, it really doesn't
cover the security issue case, but putting your fingers in your ears and
saying "everything is fine, we just need people to come visit" is not going
to solve the problem that there are several thousand open bugs on bugsnet,
and several thousand active php watchers - would be contributors - on
github, and several miles of nothing in between them.

It would be interesting to compare the analytics of php-src on github, and
the analytics (or traffic) on bugsnet ... I have a feeling that php gets
MUCH more attention on github than it does on bugsnet ... Why not go where
the people are ?

Maybe we can develop ways to use *everything* more effectively, but with
such limited resources, I don't know why we would ignore industry leading
tools that we already have, in favour of tools that nobody is, or has an
interest in, developing or maintaining to an acceptable level.

Cheers
Joe


On Mon, Jan 16, 2017 at 8:37 PM, Stanislav Malyshev <smalyshev@gmail.com>
wrote:

> Hi!
>
> > A lot of the time a feature request is just not enough; it requires at
> > least a good discussion, if not an RFC.
>
> True, but some people can not write an RFC, or not ready to do it,
> however they still can request and discuss features.
>
> > That there were feature requests open on bugsnet for more than a decade,
> > some without comments, some open as long as 15 years, should be a hint
> > that it is not useful as a collaboration tool anymore, and we have at
> > our disposal some of the best collaboration tools on offer for free.
>
> I do not disagree that we need to screen them and sort them, I just
> think dropping all of them is going too far. I also disagree that bugs
> is not a useful tool, the problem is not the tool but that people are
> not using it. If we don't have somebody to go over bugs, no matter what
> the tool they will be left behind. If we have somebody, I don't see what
> is the problem with bugs.php.net to use it.
>
> > All I really want to do at the moment is cleanup, but to be perfectly
> > honest I am not sure why we are using bugsnet for anything, given we
>
> Why not? It's a working system which we own and we can suit to our needs.
>
> > took the effort to switch over to git, source is hosted on github, the
>
> Source is not hosted on github, it's mirrored to github.
>
> > vast majority of PECL extensions are hosted on github (or bitbucket or
> > some other collab+vcs solution), and they come with far superior
> > collaboration tools to the ones that nobody bothers to maintain for
> php-src.
>
> Again, the problem is not the tools (which I don't see btw how github is
> that superior - if anything it is missing a bunch of features we need
> and do have on bugs.php.net) but that almost nobody has been using them.
> If people start to use them - and start fixing things that may be
> missing too - we can just bugs.php.net just fine.
>
> > I think this deserves consideration, we should be making the most out of
> > what is on offer.
>
> Exactly. Including using existing system we have, not just dropping
> everything and moving to github which doesn't have features we need.
>
> > I also think that doing things in the wide open has unseen benefits,
> > while bugsnet is open, it's in a dark corner of the internet that not
> > enough people bother to visit.
>
> I'm sorry I don't understand this. How bugs.php.net is not open?
>
> --
> Stas Malyshev
> smalyshev@gmail.com
>

--f403045e9c969b8fa7054645f6ed--