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

Renato Golin via llvm-dev llvm-dev at lists.llvm.org
Fri May 6 11:28:27 PDT 2016

On 6 May 2016 at 19:16, Philip Reames via llvm-dev
<llvm-dev at lists.llvm.org> wrote:
> 1) Person A makes a documented serious physical threat against Person B who
> is a member of the LLVM community.  Person A does not then get to come into
> the community and continue harassing Person B.  We can and could say Person
> A is not welcome; at minimum, all of Person A's communications should be
> moderated.

Are we proposing actively singling out people in our community? This
doesn't scale and is just plain offensive. "Serious" is a matter of

> 2) Person A has multiple convictions for sexual assault or other violent
> crime.  Person A does not get to attend LLVM events.

Are we proposing background checks for participating on our community?
This could have so many legal problems in so many countries...

> 3) Person A (an existing LLVM contributor) takes a technical discussion from
> LLVM with Person B into an alternate channel so as to personally attack
> person B without being subject to CoC. Workaround does not work, still a
> violation of CoC.

I wonder how much powerless are we, today, to deal with that.

We already have the power to moderate, ban, and publicly denounce people.

We already have to power to revoke commit access, revert patches,
unlink buildbots.

And we already can do that with no explanations necessary, but we can
always add explanations by email.

I don't think in such a case, many people would complain, either.


More information about the llvm-dev mailing list