[llvm-dev] Resuming the discussion of establishing an LLVM code of conduct

Jon Roelofs via llvm-dev llvm-dev at lists.llvm.org
Fri May 6 13:57:51 PDT 2016

On Fri, May 6, 2016 at 2:46 PM, Chandler Carruth <chandlerc at gmail.com>

> On Fri, May 6, 2016 at 1:38 PM Jon Roelofs <jroelofs at gmail.com> wrote:
>> On Fri, May 6, 2016 at 2:16 PM, Chandler Carruth <chandlerc at gmail.com>
>> wrote:
>>> On Fri, May 6, 2016 at 1:04 PM Jon Roelofs via llvm-dev <
>>> llvm-dev at lists.llvm.org> wrote:
>>>> On Friday, May 6, 2016, Renato Golin <renato.golin at linaro.org> wrote:
>>>>> On 6 May 2016 at 19:34, David Blaikie <dblaikie at gmail.com> wrote:
>>>>> > This isn't just about what we can do today, but about explaining it
>>>>> to
>>>>> > people who haven't seen us do it/don't know what the community norms
>>>>> are. So
>>>>> > that when evaluating which communities they might want to be
>>>>> involved in,
>>>>> > they have some confidence that this one might be compatible with
>>>>> their
>>>>> > comfort/needs/etc.
>>>>> The CoC can do that on its own. We were talking specifically about the
>>>>> "external media" clause.
>>>> I don't know if this has already been answered in the current thread or
>>>> the previous discussion of an llvm CoC, but: What is the intended
>>>> resolution of an issue like: https://github.com/opal/opal/issues/941 where,
>>>> IIUC, someone from the community makes politically incorrect/unpopular statements
>>>> outside of the community on so-called "external media" (without attacking
>>>> or harassing anyone in particular), but keeps his/her direct interactions
>>>> with the community on topic, engineering related, and non-discriminatory?
>>> I can't speak for how a future committee would look at this, but I can
>>> certainly tell you what I would personally look for and I would have
>>> serious concerns if the response differed significantly:
>>> Until there is some particular reason to believe that the communication
>>> elsewhere is having a serious negative effect on members of the community
>>> and their ability to effectively participate in the community, its none of
>>> our business. But if there are serious negative effects, then trying to
>>> find some way to deal with those seems reasonable. I suspect what that
>>> looks like will depend almost entirely on the circumstances that arise.
>>> As a hypothetical, if someone widely promotes positions that are
>>> sufficiently hostile and unwelcoming to contributors that their mere
>>> presence has a stifling effect and makes those contributors feel incapable
>>> of interacting with the community, no matter how separate the external
>>> behavior is kept or how much we do to try to keep things separate, I would
>>> personally hope the committee would step in because I think that would have
>>> a really negative effect on members of the community.
>>> But it is also nearly impossible to predict what kinds of behavior would
>>> or wouldn't have negative effects (especially these kinds of effects) in
>>> the abstract or based on what happens in a different community with
>>> different people. So I don't really know where the issue you cite falls. I
>>> can imagine context where it could go either way, and I would hope that the
>>> committee pays a *lot* of attention to that kind of context.
>>> Anyways, my perspective on how this would work.
>>> -Chandler
>> Thanks, that's very helpful.
>> If your points here could be clarified in the document itself, that would
>> address my main concern with it: that the CoC might be used to censor those
>> that the community disagrees with, in cases where the speech itself is as
>> you put it: "none of our business".
> I'm not sure what clarifications would help here as this is already how I
> interpret the words there. (Beyond the suggestion earlier in the thread to
> qualify with "rarely" one sentence which I'm going to apply now...)

Adding "rarely" works for me.

> Are there concrete changes that would help clarify this in your mind?

Sorry, I didn't have anything specific in mind. Wordsmithing is not my

Lang's modus tollens "proof" in the other part of the thread really helped
clarify this for me, so I think that along with the change you mention is
probably enough.

Thanks again,


> -Chandler
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20160506/b2263107/attachment-0001.html>

More information about the llvm-dev mailing list