Re: [python-committers] PEP 8015: Organization of the Python community

2018-10-15 Thread Victor Stinner 
Le ven. 12 oct. 2018 à 20:33, Brett Cannon  a écrit :
>> Python became too big to work as an unique team anymore, people
>> naturally have grouped themself as teams to work more closely on
>> specific topics, sometimes called "Special Interest Group" (SIG).
>>
>> Team members are Python contributors and Python core developers. The
>> team is responsible to select who can join the team and how.
>
> How is this bootstrapped? Do I get to declare myself the "import team" and 
> then I get to choose who joins after that?

I don't want to formalize the Python community too much. I don't think
that it's needed to have a process to create a group. It seems like in
the past, some people started to talk about a topic a litlte bit
off-topic for a list, then someone proposed to create a SIG. A mailing
list have been created. That's it. Sometimes, the list dies after a
few messages. Sometimes, the list becomes very popular.

Contributors are free to organize and group themselves, but the need
the board to make concrete changes in Python: merge changes, accept
PEPs, etc.


>> Board members must be Python core developers.  It is important that the
>> members of the board reflect the diversity of Python' users and
>> contributors. A small step to ensure that is to enforce that two members
>> cannot work for the same company (or subsidiaries of the same company).
>> In addition, to encourage more people to get involved, a core developer
>> can only be a board member twice (up to 6 years total).
>
> Is the two-term limit forever, or just consecutively?

I propose forever. In your life, you can be a board member for 6
years. It's designed to rotate frequently. So it's very different of
the previous organization using a single BDFL for life :-)


>> To boot
>> Special Case: Board Members And PEPs
>> 
>>
>> A board member cannot be a PEP delegate.
>>
>> A board member can offer a PEP, but cannot decide how their own PEP is
>> approved.
>
> So do the two other board members then make the decision? Or is there some 
> third person who will step in to make up the loss of a vote (e.g. the release 
> manager if they happen to not already be a board member)?

Yes, the two other members have to decide how a PEP is decided. These
two people are free to ask the opinion or support of anyone help :-)
Again, I don't think that it's neeed to formalize too much here.


>> The organization of this workgroup is defined by the
>> `ConductWG Charter `_.
>
> Is this here to mean the expectation that the conduct WG will manage CoC 
> issues for the core development team?

I don't want to put this responsibility on the board. So yes,
conflicts between core developers will be handled by the conduct WG.
By the way, technically, I think that it's fine if a board member is
also part of the conduct WG. But they would have to behave and
communicate differently when having the "board hat" or the "conduct WG
hat".

But I'm open to other propositions how to handle such conflict :-)

Victor
___
python-committers mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-committers
Code of Conduct: https://www.python.org/psf/codeofconduct/

Re: [python-committers] PEP 8015: Organization of the Python community

2018-10-15 Thread Antoine Pitrou 

Le 15/10/2018 à 10:30, Victor Stinner a écrit :
> 
>>> The organization of this workgroup is defined by the
>>> `ConductWG Charter `_.
>>
>> Is this here to mean the expectation that the conduct WG will manage CoC 
>> issues for the core development team?
> 
> I don't want to put this responsibility on the board. So yes,
> conflicts between core developers will be handled by the conduct WG.

How does that work?  The conduct WG doesn't even seem to have published
procedures.  Also we cannot expect them to follow the history of
interpersonal interactions in Python core development.

> But I'm open to other propositions how to handle such conflict :-)

If something is required, I would suggest some kind of ethical committee
that would be comprised of core developers.  I would also suggest their
mission to focus on appeasement rather than punishment.

Regards

Antoine.
___
python-committers mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-committers
Code of Conduct: https://www.python.org/psf/codeofconduct/

Re: [python-committers] PEP 8015: Organization of the Python community

2018-10-15 Thread Victor Stinner 
Le lun. 15 oct. 2018 à 11:35, Antoine Pitrou  a écrit :
> > I don't want to put this responsibility on the board. So yes,
> > conflicts between core developers will be handled by the conduct WG.
>
> How does that work?  The conduct WG doesn't even seem to have published
> procedures.  Also we cannot expect them to follow the history of
> interpersonal interactions in Python core development.

https://wiki.python.org/psf/ConductWG/Charter contains many information.


> > But I'm open to other propositions how to handle such conflict :-)
>
> If something is required, I would suggest some kind of ethical committee
> that would be comprised of core developers.  I would also suggest their
> mission to focus on appeasement rather than punishment.

"punishment" Oh, you seem to have a very bad opinion about the conduct
WG :-( Why do you have a so bad opinion? I don't think that the goal
of the conduct WG is to punish.

I don't know if they already published transparency reports, but I'm
quite sure that removing messages or ban someone is the least common
option and only taken for the worst incidents. I'm quite sure that
many issues are handled in private with people involved in the
incident and that no punishment action has been taken.

... That's just guesses, I don't know much about this conduct group
:-) It seems to be one of your concern, the lack of transparency?

I'm not sure how creating a *new* group would be way better: more fair
and more transparent?


I like the idea of having a group which is external to the core
developer group, since I expect them to be more "fair". For example,
if an "important"/"popular" core dev misbehaves, it may be more
difficult for other core developers to take any action. If the group
handling CoC incidents is external, I expect them to handle incidents
the same way for everyone.

For example, when Guido was the BDFL, if you imagine that BDFL
misbehaved, who would stand up against Guido? Maybe this example is
too hypothetical. Better examples can be found in the Linux community,
their BDFL Linus Torvalds is known to have strong words on the mailing
list:
https://www.newyorker.com/science/elements/after-years-of-abusive-e-mails-the-creator-of-linux-steps-aside

I'm not interested here to discuss if Linus behaviour was appropriate
or not. I'm just asking about the process: who is supposed to discuss
with Linus about his behaviour? Replace "Linus" with any other
famous/popular Linux developer. And why no action has been taken to
reduce the toxicity of the Linux mailing list when Sage Sharp wrote a
blog post explaining clearly that the Linux community has issues?
https://sage.thesharps.us/2015/10/05/closing-a-door/

I'm using the example of Linux because multiple events are well
documented, the situation is getting better, and in the past, there
were many blocker issues to handle "Code of Conflict" incidents.
Moreover, it's easier to criticize a community different than ours
community :-) Our community is perfect, right? :-)

Victor
___
python-committers mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-committers
Code of Conduct: https://www.python.org/psf/codeofconduct/

[python-committers] Note: I published the first draft of PEP 8001

2018-10-15 Thread Łukasz Langa 
See: https://discuss.python.org/t/pep-8001-python-governance-voting-process/233 


Cheers,
Ł


signature.asc
Description: Message signed with OpenPGP
___
python-committers mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-committers
Code of Conduct: https://www.python.org/psf/codeofconduct/

Re: [python-committers] PEP 8015: Organization of the Python community

2018-10-15 Thread Antoine Pitrou 

Le 15/10/2018 à 12:00, Victor Stinner a écrit :
> Le lun. 15 oct. 2018 à 11:35, Antoine Pitrou  a écrit :
>>> I don't want to put this responsibility on the board. So yes,
>>> conflicts between core developers will be handled by the conduct WG.
>>
>> How does that work?  The conduct WG doesn't even seem to have published
>> procedures.  Also we cannot expect them to follow the history of
>> interpersonal interactions in Python core development.
> 
> https://wiki.python.org/psf/ConductWG/Charter contains many information.

That information doesn't seem to include resolving conflicts.  It
advises on policies, creates a set of standard documents, and develops
training materials.

My understanding is that the WG is not meant to handle conflicts by
itself.  Rather, it gives guidance to other bodies on how to handle
conflicts.  So we would have to have our own body for handling
conflicts, such as an ethical committee.

> ... That's just guesses, I don't know much about this conduct group
> :-) It seems to be one of your concern, the lack of transparency?

That, and the fact they're outside the loop of core development, and
don't know anything about the usual social / power dynamics here.  They
also don't have the time to devote that a professional magistrate would
have.

> Replace "Linus" with any other
> famous/popular Linux developer. And why no action has been taken to
> reduce the toxicity of the Linux mailing list when Sage Sharp wrote a
> blog post explaining clearly that the Linux community has issues?

But do you think an external WG would have had the authority to deal
with the problem?  People did criticize Linus' behaviour, even amongst
kernel developers.  But there wasn't any solution exercisable from the
outside.  You couldn't just displace Linus and put someone else in
power, because that wouldn't have been accepted by the broader community.

Only a recognized body formed by the community itself could be
legitimate enough to take action.  Which suggests to me that the CoC WG
can be a source of advice and suggestions, not take decisions by itself.

Regards

Antoine.
___
python-committers mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-committers
Code of Conduct: https://www.python.org/psf/codeofconduct/

Re: [python-committers] discuss.python.org participation

2018-10-15 Thread Benjamin Peterson 


On Fri, Oct 12, 2018, at 11:55, Brett Cannon wrote:
> On Thu, 11 Oct 2018 at 01:30, Antoine Pitrou  wrote:
> 
> >
> > What concerns me is that there are several long-time and/or prominent
> > developers who are not even registered (*) on discuss.python.org.  For
> > example Benjamin Peterson, Larry Hastings, Raymond Hettinger, Stefan
> > Krah, Terry Reedy.
> >
> 
> I believe Larry is currently busy so he might simply have not taken the
> time (and will be occupied into I believe November).
> 
> I will also note that Benjamin, Larry, and Raymond can be a bit quiet at
> times and so they may not have signed up yet because they have not had
> anything to say to compel them to create accounts (Stefan has already
> stated he doesn't like this idea so I'm assuming that might be why he has
> not signed up yet).

Brett is exactly right about me. I'm following along on Discourse and this 
mailing list when I have the time. I plan to read the governance PEPs and vote 
when the time comes. But, I don't have anything useful to say.

At the sprint, I verbally supported Łukasz's plans to try out Discourse. 
Discourse isn't perfect and may feel uncomfortable, but we're also losing 
potential contributors because they find mailing lists uncomfortable.
___
python-committers mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-committers
Code of Conduct: https://www.python.org/psf/codeofconduct/