llvm dev - May 2016 - Resuming the discussion of establishing an LLVM code of conduct

On 5 May 2016 at 23:06, Tanya Lattner <tanyalattner at llvm.org>
wrote:
> The point is that you wouldn’t know that from looking at the alternative
code of conduct. I would need to spend months pouring through mailing lists
posts and watching the community to feel its a good and safe place.
Right, this is a very good point for having something written down.

> One of the many reasons a code of conduct is useful is to show outsiders
what our community is like, what we think is acceptable and what is not. They
read that, then they know how they are expected to behave and how they will be
treated. They are much more likely to be a part of that community.
My argument is that this is much more important in the US than in most
other places (see my response to Hal). But the US is a big enough
chunk that we cannot ignore.

However, we also can't ignore that the US is not the *only* source,
and for a few of us, having a code that is overly powerful while being
overly vague is a reason to *leave* the project. Some even mentioned
forking it.

Then my question is: how many people are we prepared to lose, and how
many are we expecting to gain? More importantly, can we gain without
losing?
>From the few passionate responses against the code in its current
form, it would be naive to say we could. So, can we change the code in
order to not lose those types of people?

Mind you, those that responded are but a few who *can* respond. The
kind of impasse this CoC creates, affects people that rarely
communicate, especially on controversial subjects such as this. And
those people don't blog about their problems, they just go be
productive elsewhere. You won't know they're gone until it's too
late.

> But there are other goals of a CoC that can not be ignored and are not met
by the alternative CoC.
Such as?

--renato

On May 5, 2016, at 3:18 PM, Renato Golin via llvm-dev <llvm-dev at
lists.llvm.org> wrote:
>> One of the many reasons a code of conduct is useful is to show
outsiders what our community is like, what we think is acceptable and what is
not. They read that, then they know how they are expected to behave and how they
will be treated. They are much more likely to be a part of that community.
> 
> My argument is that this is much more important in the US than in most
> other places (see my response to Hal). But the US is a big enough
> chunk that we cannot ignore.
Renato, I’m confused about your approach here.  At this point, you seem
persistently interested in discussing whether having a code of conduct is the
right thing or not.  This is missing the point: we’re committed to it, and want
to make sure that we get reasonable processes and policies in place.

If we find out that they are problematic in practice, we can and will course
correct.

-Chris

On 5 May 2016 at 23:22, Chris Lattner <clattner at apple.com>
wrote:
> Renato, I’m confused about your approach here.  At this point, you seem
persistently interested in discussing whether having a code of conduct is the
right thing or not.
No, I'm asking for arguments in favour. So far, arguments were mostly
against, because all those in favour didn't have to discuss against
the status quo.

This led to those against lacking understanding of the real arguments.
I think Tanya is putting forward good arguments, so that's cool.

>  This is missing the point: we’re committed to it, and want to make sure
that we get reasonable processes and policies in place.
>
> If we find out that they are problematic in practice, we can and will
course correct.
I think some of us are finding it problematic in practice, but we
don't seem to be correcting it.

Chandler did some corrections, most of them welcome, but the most
important thing hasn't been even discussed: who and how are we
choosing the committee. This makes a lot of difference as to how the
CoC will be interpreted.

--renato

> On May 5, 2016, at 3:18 PM, Renato Golin <renato.golin at linaro.org>
wrote:
> 
> On 5 May 2016 at 23:06, Tanya Lattner <tanyalattner at llvm.org>
wrote:
>> The point is that you wouldn’t know that from looking at the
alternative code of conduct. I would need to spend months pouring through
mailing lists posts and watching the community to feel its a good and safe
place.
> 
> Right, this is a very good point for having something written down.
> 
> 
>> One of the many reasons a code of conduct is useful is to show
outsiders what our community is like, what we think is acceptable and what is
not. They read that, then they know how they are expected to behave and how they
will be treated. They are much more likely to be a part of that community.
> 
> My argument is that this is much more important in the US than in most
> other places (see my response to Hal). But the US is a big enough
> chunk that we cannot ignore.
> 
> However, we also can't ignore that the US is not the *only* source,
> and for a few of us, having a code that is overly powerful while being
> overly vague is a reason to *leave* the project. Some even mentioned
> forking it.
> 
> Then my question is: how many people are we prepared to lose, and how
> many are we expecting to gain? More importantly, can we gain without
> losing?
> 
> From the few passionate responses against the code in its current
> form, it would be naive to say we could. So, can we change the code in
> order to not lose those types of people?
> 
> Mind you, those that responded are but a few who *can* respond. The
> kind of impasse this CoC creates, affects people that rarely
> communicate, especially on controversial subjects such as this. And
> those people don't blog about their problems, they just go be
> productive elsewhere. You won't know they're gone until it's
too late.
Quite frankly, I find some of the responses to be very concerning and
disturbing. Comments about how its ok to behave in certain ways that many find
offensive, sexist, or racist, is extremely disappointing and if those people
want to leave the community then I am fine with it.

If the Code of Conduct was really not representing what is already happening in
our community then it needs to be reworded. There are many many people who
aren’t even commenting at all because they fear being attacked in this thread.
There are many who aren’t responding because they agree we need one and just
want this thing done and don’t want to talk about it anymore.

I don’t think we will agree about the need for a code of conduct. I just don’t
see it happening. Its not a US versus everywhere else thing. All of the issues
the code of conduct touches on are universal across everywhere. I feel strongly
that we need a code of conduct for the strength, health, and future of our
community and I don’t see anything in Chandler’s draft that makes me feel
concerned. Is it wordy and long? Maybe a little.  But, I think it represents the
people of this community and the respect we have for each other. If we find out
we were wrong, then we can make changes to it as we go. Nothing is ever set in
stone and you learn through experience.
> 
> 
>> But there are other goals of a CoC that can not be ignored and are not
met by the alternative CoC.
> 
> Such as?
I was referring to the goal of showing outsiders what our community is like and
showing it is an indeed a safe and inviting place for everyone.

-Tanya
> 
> --renato

Am 06.05.2016 um 01:26 schrieb Tanya Lattner via
llvm-dev:
>
> There are
> many many people who aren’t even commenting at all because they fear
> being attacked in this thread.
Now you're contradicting LLVM's own description that it's a friendly
community.

Also:
How many is "many, many", actually? How many of these are really in 
fear, how many are just trying to impose their mindset without actually 
planning to contribute in earnest, how many are so fearful that they 
should really seek professional help?

 > There are many who aren’t responding
> because they agree we need one and just want this thing done and
> don’t want to talk about it anymore.
Well, I can imagine that those who want to push in that direction don't 
want the opposition.
> I don’t think we will agree about the need for a code of conduct. I
> just don’t see it happening. Its not a US versus everywhere else
> thing. All of the issues the code of conduct touches on are universal
> across everywhere.
Harassment is indeed universal.
The ideas about how to best deal that with are not.

 > I feel strongly that we need a code of conduct for
> the strength, health, and future of our community and I don’t see
> anything in Chandler’s draft that makes me feel concerned.
The essence of that argument is that since you don't feel concerns, 
nobody else should.
I do not think I'll follow that.

 > Is it
> wordy and long? Maybe a little.
Actually it seems to be improving, but from my perspective, it's still 
far too overspecific and too easily abused.
Note that my concerns are different from Renatos: Renato fears a power 
grab by the enforcement committee, I fear that regardless of how the 
committee is elected the whole thing is open to degeneration.

 > But, I think it represents the
> people of this community and the respect we have for each other.
In that case, we wouldn't need a code of conduct at all.

There's an inconsistency here.
This "the respect we have for each other" bit is in line with the 
argument that the CoC's purpose is to keep things the way they are. 
Above, you said that people have fear of speaking up, which is a pretty 
strong contradiction.
It's one of those things that make alarms go off.

 > If
> we find out we were wrong, then we can make changes to it as we go.
> Nothing is ever set in stone and you learn through experience.
*shrug* I have had experience with that kind of approach. With codes of 
conduct, set up with the best of intentions, and in the end, people just 
chose to ignore them.
> I was referring to the goal of showing outsiders what our community
> is like and showing it is an indeed a safe and inviting place for
> everyone.
A self-description is as conclusive as any code of conduct.
It's not the text that counts, whether it's enforceable or not; what 
counts is what's actually happening, and no text will change much of that.

BTW the community shouldn't describe itself.

Aww dammit. I'm writing far too much on that topic. Again.
Ah well.

On Thu, May 5, 2016 at 4:26 PM, Tanya Lattner via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
>
> > On May 5, 2016, at 3:18 PM, Renato Golin <renato.golin at
linaro.org>
> wrote:
> >
> > On 5 May 2016 at 23:06, Tanya Lattner <tanyalattner at llvm.org>
wrote:
> >> The point is that you wouldn’t know that from looking at the
> alternative code of conduct. I would need to spend months pouring through
> mailing lists posts and watching the community to feel its a good and safe
> place.
> >
> > Right, this is a very good point for having something written down.
> >
> >
> >> One of the many reasons a code of conduct is useful is to show
> outsiders what our community is like, what we think is acceptable and what
> is not. They read that, then they know how they are expected to behave and
> how they will be treated. They are much more likely to be a part of that
> community.
> >
> > My argument is that this is much more important in the US than in most
> > other places (see my response to Hal). But the US is a big enough
> > chunk that we cannot ignore.
> >
> > However, we also can't ignore that the US is not the *only*
source,
> > and for a few of us, having a code that is overly powerful while being
> > overly vague is a reason to *leave* the project. Some even mentioned
> > forking it.
> >
> > Then my question is: how many people are we prepared to lose, and how
> > many are we expecting to gain? More importantly, can we gain without
> > losing?
> >
> > From the few passionate responses against the code in its current
> > form, it would be naive to say we could. So, can we change the code in
> > order to not lose those types of people?
> >
> > Mind you, those that responded are but a few who *can* respond. The
> > kind of impasse this CoC creates, affects people that rarely
> > communicate, especially on controversial subjects such as this. And
> > those people don't blog about their problems, they just go be
> > productive elsewhere. You won't know they're gone until
it's too late.
>
> Quite frankly, I find some of the responses to be very concerning and
> disturbing. Comments about how its ok to behave in certain ways that many
> find offensive, sexist, or racist, is extremely disappointing and if those
> people want to leave the community then I am fine with it.
>
> If the Code of Conduct was really not representing what is already
> happening in our community then it needs to be reworded. There are many
> many people who aren’t even commenting at all because they fear being
> attacked in this thread. There are many who aren’t responding because they
> agree we need one and just want this thing done and don’t want to talk
> about it anymore.
>
> I don’t think we will agree about the need for a code of conduct. I just
> don’t see it happening. Its not a US versus everywhere else thing. All of
> the issues the code of conduct touches on are universal across everywhere.
> I feel strongly that we need a code of conduct for the strength, health,
> and future of our community and I don’t see anything in Chandler’s draft
> that makes me feel concerned. Is it wordy and long? Maybe a little.  But, I
> think it represents the people of this community and the respect we have
> for each other. If we find out we were wrong, then we can make changes to
> it as we go. Nothing is ever set in stone and you learn through experience.
>
> >
> >
> >> But there are other goals of a CoC that can not be ignored and are
not
> met by the alternative CoC.
> >
> > Such as?
>
> I was referring to the goal of showing outsiders what our community is
> like and showing it is an indeed a safe and inviting place for everyone.
>
A lot of the discussion in this and other threads revolves around the
"enforcement" part of the CoC. If the purpose is only to show to
others
what we're like, then "enforcement" isn't needed per se, just
explanation.

But IIRC in another thread it was brought up that the "enforcement"
part is
actually needed for "showing it is an indeed a safe and inviting place for
everyone".

So there are actually two separate things:
- we want to codify our existing practices and expectations (this is a
"descriptive" component)
- we want to codify new enforcement practices (this is a
"prescriptive"
component)

I have to admit that adding codified enforcement practices is a major step.
If this is necessary for the CoC to serve its purpose of "showing it is an
indeed a safe and inviting place for everyone" then I think we can add
reasonable enforcement practices (+1 from me). But any change to existing
community practices like this is going to involve discussion, and so far
there hasn't been a focused discussion about this. Can we have one?


Personally, I think that the "enforcement" proposed CoC in its current
form
is a pretty darn good (even split out into a separate "Reporting
Guide"
document). There's some details of choosing the "Code of Conduct
Advisory
Committee" left to do, but overall it seems reasonable.


-- Sean Silva

>
> -Tanya
>
> >
> > --renato
>
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160505/0975774a/attachment.html>

Tanya Lattner via llvm-dev wrote:
> concerning and disturbing
There's an inconsistency here.


<3

(I do seem to be having difficulty articulating my thoughts on this
subject, but I'll try again.)


Tanya Lattner via llvm-dev wrote:
>
> There are many many people who aren’t even commenting at all because
> they fear being attacked in this thread.
I don't know what to do with this.  People interested in LLVM are at
the apex of technical sophistication.  Anyone who wished to comment
anonymously could surely make use of a VPN and a temporary email
account?

Alternately, one imagines a person could email you, requesting you
relay a post on their behalf?

"My ideas aren't sacred. Please challenge them."

This is my mantra, but ... again, I'm at a loss as to how to respond
to someone if they're not operating on this principle.

We have secular bloggers in Bangladesh who are literally being hacked
to death for posting their beliefs.

I don't think that's probably within the scope of concerns for those
expressing opinions in the LLVM community?  (I don't mean this flippantly.
My brain likes to find limit cases and work inward.)

Please help me understand what is literally meant by "they fear being
attacked" in an [llvm-dev] thread.  Is it code for "one might feel
uncomfortable if one's beliefs were challenged"?  Or is there something
more sinister at work?

Genuinely perplexed,

Regards,

Bill