Re: [gentoo-dev] [RFC pre-GLEP] Gentoo Git Workflow

[Daniel Campbell]

On 09/10/2017 02:34 AM, Michał Górny wrote:
> W dniu nie, 10.09.2017 o godzinie 00∶39 -0700, użytkownik Daniel
> Campbell napisał:
>> On 09/09/2017 12:47 AM, Michał Górny wrote:
>>> W dniu pią, 08.09.2017 o godzinie 17∶19 -0400, użytkownik Rich Freeman
>>> napisał:
>>>> On Tue, Jul 25, 2017 at 4:05 AM, Michał Górny <mgorny@g.o> wrote:
>>>>>
>>>>> What do you think about it? Is there anything else that needs being
>>>>> covered?
>>>>>
>>>>
>>>> FYI - if anybody does want to make any comments on the proposed
>>>> devmanual changes to implement the new tags please comment at:
>>>>
>>>> https://github.com/gentoo/devmanual.gentoo.org/pull/72
>>>>
>>>> For that matter, if you want to even know what the proposed changes
>>>> are you should also visit the link.
>>>>
>>>> List replies seem to be discouraged.
>>>>
>>>> I realize that some prefer to limit comments to media that are purely
>>>> open source.  I guess the FOSS Linux kernel provided /dev/null still
>>>> exists as an alternative.  Honestly, I'm not sure which of the options
>>>> are more likely to get read.
>>>>
>>>
>>> TL;DR: Rich, I would really appreciate if you stopped the flamebaits.
>>> I understand that you think you're doing Gentoo a favor but you're not.
>>>
>>> The footers were discussed to death in this very thread. I've heard your
>>> opinions. However, as far as I'm concerned (and as I've pointed out) you
>>> did literally *nothing* to push your ideas forward for 2+ years.
>>>
>>> Since I've done all the work, I've did it my way and for the reasons
>>> I've explained multiple times. If you disagree, them I'm sorry but
>>> in life you don't get to have everything your way. Especially if all you
>>> do is complain and expect others to do the work for you.
>>>
>>> I understand that you're unhappy and since you have no valid arguments,
>>> all you can do is try to get more people to support you and shout.
>>> Revive the bikeshed as many times as possible, try to make a lot of
>>> noise and block changes. Worst case, you've blocked something you didn't
>>> like. Best case, you're finally get others so discouraged that they'll
>>> do things your way just so that you stop.
>>>
>>> Rich, this is not a kindergarten. It's time you learn to lose,
>>> and accept the consequences. If you can't do that, the door out is open,
>>> and you're free to leave anytime you want.
>>>
>>
>> This behavior is not befitting Gentoo leadership. Limiting commentary to
>> a walled garden outside Gentoo control violates one of our goals
>> (independence), and the incendiary retorts are no more mature than the
>> behavior you're criticizing. Nothing will change in the way people
>> respond to you until you learn how to treat others.
>>
>> By all means, I'm glad we're seeing some action on which tags to settle
>> with. It's been a mess of confusion ("which tags do I use? Will this
>> tick someone off?", etc), and will be easier to build on once we figure
>> out the tags that'll work best. It'll be awesome to get automatic bug
>> closing from a commit, and I suspect it'll bring a lot of good. Settling
>> on tags allows us to automate more. However, as a Council member, the
>> Gentoo community looks to you and your behavior as an example. Is this
>> the example you want to set for our community?
>>
>> On the GitHub Issue, you called this mailing list "completely useless"
>> [1], and then you sent your message above a little later. Would you want
>> to work with someone who talks to you the way you treated Rich?
>
> Yes, I did call it useless *multiple times*, and I've pointed out
> the problems. Considering they were ignored and the quality of
> the mailing list hasn't improved, this statement still stands.
>
> Rich should have talked to me if he had problems with understanding my
> comment or missed the context to it. What he did instead is beginning
> a public stone throwing session. This is not a kind of behavior I am
> going to accept, or respond kindly to.
>
> It's elementary. If you have a problem with me, talk to me first. Not go
> pointing fingers and shouting 'this person is bad'.

That's a good policy, one most of us can agree with I think. Sarcasm
isn't often understood (or appreciated) by all, but Rich's comment
really didn't do anything except highlight how discussion was being
moved. This creates a divide in the information and discussion available
to other developers. You and others are free to do this, but others are
right to criticize it also. Calling an entire community "useless" is not
going to magically make it better.

>
>> None of this bickering is motivating or inspiring to those who read it,
>> and leadership should know better than to stoop to this level publicly.
>> You will not get more developer activity, agreement, cooperation, or
>> contribution by berating your fellow developers. In fact, Gentoo is
>> known for its bickering developer community. You are in a position to
>> change that. You asserted in #gentoo-trustees that the Council *is*
>> Gentoo's leadership.
>>
>> Is this your brand of leadership?
>>
>> ~zlg
>>
>> [1] https://dev.gentoo.org/~zlg/useless.png
>>
>> (screenshotted since GitHub conversations can be curated.)
>
> I'm not going to answer to your political propaganda. We don't need more
> politicians like you in Gentoo. We need actual people who do stuff
> rather than talk about theory to the point when everybody is so tired
> nobody wants to do anything anymore.
>
>
I'm not looking for political propaganda. What would I gain from it? I
don't really have any strong connections here. In fact I'm probably
ticking a few people off. I'm calling your practices out for what they
are, since nobody else appears willing or able to. It's flat-out
dishonest to complain about a mailing list on one site, then
deliberately add flames to the fire less than a day later on the very
list you complained about. You are contributing to the very thing you
claim to detest. I don't see that as political, because it affects the
work that you, I, and other Gentoo devs do. It damages relationships
between developers, and if something isn't done about it, Gentoo will
continue falling into irrelevance because we can't get our ship to sail
smoothly. We all have the ability to improve this, and I gain nothing by
writing this e-mail.

With this particular issue (git commit tags), I understand bikeshed will
happen. Seeking to divide the community between your hand-picked "these
people do things (I like), so they're better than the others" and
everyone else is destructive to our distribution. It's akin to building
a second bikeshed because you didn't like the color of the first one.
Why should anyone work with you if you treat them like trash? Because
your hands are in many cookie jars? Because you're a Council member?
Because you're competent? Libre software projects cannot exist purely in
a technical environment as long as there is a human element. You need
social merit, and social skill, to get anywhere with collaboration, or
you will end up with too much work and not enough time (because you've
driven everyone away). That is the beginning of a distro's downward
curve -- when collaboration is no longer enjoyable to those who are part
of the group. You and I don't know each other personally, but based on
where I've seen your name, it's clear you have plenty of technical
merit. And some people are your friends (I assume), so you're not
bankrupt in social merit, either. Why do you feel entitled to treat
others in this fashion? You are already an established member of our
distro, and there's no need to lash out at others.

Note that I didn't claim that I was good at the social part. Most of us
could do better, myself included. Many programmers are gruff and
insolent, so it's difficult to learn from them. I have trouble with
people who are quick to belittle one's intelligence for making mistakes,
so it affects my ability to get good feedback. I don't fancy being
abused, having lived through enough of it in my personal life, so I
don't seek out those who will abuse me. As a result, I contribute less
to Gentoo than I'd like to. I suspect I'm not alone in that.

I'm not saying we should flatter each other and put on kid gloves. I'm
saying we should treat each other as reasonable adults working on
software to the betterment of everyone involved. A big part of Gentoo's
problems are in how it deals with developers making mistakes, or the
poor accessibility of decisions or processes that are often discarded as
"this was already covered", often without links to rationale. QA, for
example, means jack if its information is not widely accessible,
discussed, and enforced on a consistent basis. I can count on one hand
the number of useful QA docs I've managed to find.

Does that mean QA as a whole is crap? No. It means that there is a
disconnect between those who write the standards and those who need to
write *to* those standards. Fix that, and your hatred of "do nothing"s
will fix itself because your fellow developers will have better
resources to educate themselves with. It has the knock-on effect of
convenience: if someone makes a dumb mistake, someone can tell them what
they screwed up and link to something that will educate them. IRC bots
do this all the time, especially in #vim, so it's not like this is some
insurmountable hurdle. Until we get to such a point, it's the
responsibility of the more experienced developers to assist those they
perceive as "bad" to improve, or their criticism has no weight or
credibility. Again, that doesn't mean hand-holding, coddling, or
anything of the sort. It means getting people up to speed and
understanding that mistakes happen due to incomplete or conflicting
information, not malice. If you take a look at other areas of life,
you'll see that it's how humanity retains its collective knowledge:
passing it down to the less-experienced. Computer science wouldn't have
gotten anywhere if the findings and techniques weren't shared. The same
is true of Gentoo.

A fine example is a recent review I had with floppym in #gentoo-python.
To be honest, I was expecting to deal with berating and belittlement,
like what happens when I deal with you. But I reserved my judgment and
went ahead anyway. floppym was a big surprise. I learned more from him
in 10-20 minutes than I have in most other reviews, and now I know more
so I can avoid making those mistakes in the future. (I kept notes) I
thanked him for his time and willingness to teach me a few things, and
Gentoo improved a little bit that day, both technically and socially. I
hope the experience was as pleasant for him.

(To clarify, I recall floppym and I having a minor tiff months ago. We
both overcame that history to improve Gentoo. If you're reading this,
Mike, thanks again.)

Situations like the above are probably not that special; I assume it
happens a lot. But typically, nobody in Gentoo really points that stuff
out and congratulates it. I never read about how developers X, Y, and Z
pulled together and made a useful feature in Gentoo possible. I never
read about the work arch testers and CI and infra put into making our
lives smoother. I don't see any stories told about how things came to be
for Gentoo. It's shrouded in documents spread across all of infra --
assuming it exists in the first place -- and as a result, the hard work
people put into this distro doesn't get a lot of attention. That could
be a large source of the general discontentment among some of us.

Maybe we should give our developers more credit when they do great work
instead of hating on them when they screw up. This method focuses on the
good instead of the bad, providing positive reinforcement, which is
proven to work in psychology. People like to know they're contributing
to something and making something better. If they're not, they want to
know *why* their solution is broken and *how* to fix it, so they can do
better.

If you aren't interested in any of that, I have to wonder why you chose
to run for a role that deals just as much with people as it does code.
If you have the inclination, I'd be willing to discuss that in another
thread, private or public. I've gone far enough off-topic already.

(Apologies for verbosity. It's necessary to ensure I don't get
misunderstood.)

~zlg
-- 
Daniel Campbell - Gentoo Developer, Trustee, Treasurer
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net
fpr: AE03 9064 AE00 053C 270C  1DE4 6F7A 9091 1EA0 55D6