Add a brief "code of conduct" section to the Governance Model

[dscho]

It is better to state clearly expectations and intentions than to assume quietly that everybody agrees.

So let's add a concise section to the governance model, maybe modeled after http://software-carpentry.org/conduct.html.

[dscho]

Here is an excellent start: http://contributor-covenant.org/version/1/1/0/code_of_conduct.md
Probably should start a discussion whether contributors are okay with this version as-is, or suggest improvements.

[frogonwheels]

Yes! I like the contributor-covenant.org one. On first reading I really can't see anything I'd change.

[PhilipOakley]

[re-sending, hopefully using the right return email address]
It is better to state clearly expectations and intentions than to assume quietly that everybody agrees.

So let's add a concise section to the governance model, maybe modeled after http://software-carpentry.org/conduct.html.

I'd be cautious about simply adding such a model. While it is well intentioned, and tries to summaries the ideals and concepts that are being attempted, they can become a rod for one's back.

There is potentially a great difficulty in deciding how 'we' [0] will discriminate (that is, make decisions about difficult issues) that are somehow non-discriminatory in the eyes of all.

I see from the issues [dscho's link; 2] that there has been some discussions on the opal project (view all issues at that project, including closed).

A quick google search for "code of conduct for open source" offered a few others such as Twitter's [2], with it's 'strive to' section, which acknowledges that it can't be perfect.

Codes of conduct (especially if espressed too tightly) are a potential can of worms, perhaps becoming a 'wicked issue' [3].

Civility and tolerance are important aspects here.

Philip

[0] who would be the 'we' who actually wants to / has to / can make such decisions, and make them stick??
[1] EthicalSource/contributor_covenant@48dc590
[2] https://engineering.twitter.com/opensource/code-of-conduct
[3] https://en.wikipedia.org/wiki/Wicked_problem

[dscho]

who would be the 'we' who actually wants to / has to / can make such decisions, and make them stick??

As described in the governance model, it would be my (i.e. the project lead's) burden to arbitrate, and if failing to reach a consensus, to decide.

Civility and tolerance are important aspects here.

Absolutely. So far, we have managed to be quite civil and tolerant most of the time, but I feel that we can do better.

Having said that, I think it is necessary to describe explicitly the code of conduct that we, the developers, contributors and project lead of the Git for Windows project, are willing to abide by.

As you said, nobody is perfect, so we can only strive to live by the code of conduct, and sometimes we fail. The idea is that we fail gracefully and do better next time 'round.

Now, I do not expect anybody to disagree with the text in https://github.com/CoralineAda/contributor_covenant/blob/e3ae4dcbf749bbbe06ca67e532899ef6845ee198/CODE_OF_CONDUCT.md, but I see that there is still quite a bit of wiggle room.

For example, I personally find it offensive to half of the population of this wonderful planet to refer to users exclusively in the masculine form. I would much rather have an equal number of female and male contributors and users in our project, and I would rather encourage that by referring to a user as "her", because let's face it, so far roughly 100% of our contributors have been male, and it does not help the tone on the mailing list nor in the tickets. Our track record, therefore, is abysmal.

I do not want to dictate to anybody how to think and feel, of course, but I believe it is important to have this discussion so we can be on the same page about how we want to conduct development in this project, and then document that by publishing the code of conduct on which we all can agree.

[PhilipOakley]

Civility and tolerance are important aspects here.

Absolutely. So far, we have managed to be quite civil and tolerant most of the time, but I feel that we can do better.

Agreed.

...

For example, I personally find it offensive to half of the population of this wonderful planet to refer to users exclusively in the masculine form. I would much rather have an equal number of female and male contributors and users in our project, and I would rather encourage that by referring to a user as "her", because let's face it, so far roughly 100% of our contributors have been male, and it does not help the tone on the mailing list nor in the tickets. Our track record, therefore, is abysmal.

One thing I notice about the he/she problem is the tendency to afford all the sensible stuff to 'she' and stupid stuff to 'he' (or similar stereotypes) (e.g. in the various git threads).

It's moderately well know that [we humans] are very bad at producing random selections for things we want to look like random (such as generic gender attribution, or picking random passwords), and the consequent inverse problem that we attribute [unusual] random chances to deliberate choice [123456 is random, in the lottery?] - not really sure how these are to be solved though - unless we re-engineer humans;-) .

...
I'll see if I have sufficient time to think through some of the governance & code of conduct issues.

Philip

[dscho]

Well, I would like to not overthink this. If nobody has any valid objections, we can just go with the Contributors' Covenant and take it from there.

[mhagger]

During a Git community meltdown a couple of years ago, I wrote what I thought would be good community guidelines for the Git project:

https://github.com/mhagger/git/commit/c6e6196be8fab3d48b12c4e42eceae6937538dee

It is heavily weighted towards requiring civility as opposed to fostering diversity, simply because that was the problem we were having at the time. On the plus side, it is oriented towards the way the Git project is organized and run.

Feel free to look at it and, if you like, steal parts.

There was some good discussion about my proposal on the mailing list at the time, but because the discussion took place in the midst an exceedingly nasty flame war I hesitate to provide a link.

[dscho]

Thank you @mhagger!

[dscho]

Maybe http://todogroup.org/opencodeofconduct/#Git%20for%20Windows/git-for-windows@googlegroups.com is even better; it builds on the Contributor's Covenant, and adds more of the points mentioned by @mhagger.

[nalla]

That model just got my final vote.

[dscho]

@nalla thanks! I would like to give @mhagger a chance to weigh in before making it a fact (even if I imagine that he is more than happy with the Open Code of Conduct), okay?

[nalla]

Of course!

[dscho]

Thanks! To make it easier to discuss, I turned this issue into a Pull Request.

[mhagger]

I'm not involved enough in this community to have earned a vote. But the code of conduct you linked to seems fine, assuming you really plan to go to the effort of setting up a response team like it describes.

[PhilipOakley]

I liked it. It has a nice neutral, rather than preaching, tone setting out clear aspirations.

Michael's point about a response team would need addressing as it suggests a large team is available, and the "will contact you personally" is perhaps open to misinterpretation.

also (a petty) s/Git%20for%20Windows/Git-for-Windows/

[dscho]

Okay, I just adjusted it to reflect the fact that there is no "response team". Here is the interdiff:

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index c62628d..df1eee4 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -28,14 +28,13 @@ characteristics above, including participants with disabilities.

 ### Reporting Issues

-If you experience or witness unacceptable behavior—or have any other concerns—please report it by contacting us via **git-for-windows@googlegroups.com**. All reports will be handled with discretion. In your report please include:
+If you experience or witness unacceptable behavior—or have any other concerns—please report it by contacting us via **johannes.schindelin@gmx.de**. All reports will be handled with discretion. In your report please include:

 - Your contact information.
-- Names (real, nicknames, or pseudonyms) of any individuals involved. If there are additional witnesses, please
-include them as well. Your account of what occurred, and if you believe the incident is ongoing. If there is a publicly available record (e.g. a mailing list archive or a public IRC logger), please include a link.
+- Names (real, nicknames, or pseudonyms) of any individuals involved. If there are additional witnesses, please include them as well. Your account of what occurred, and if you believe the incident is ongoing. If there is a publicly available record (e.g. a mailing list archive or a public IRC logger), please include a link.
 - Any additional information that may be helpful.

-After filing a report, a representative will contact you personally. If the person who is harassing you is part of the response team, they will recuse themselves from handling your incident. A representative will then review the incident, follow up with any additional questions, and make a decision as to how to respond. We will respect confidentiality requests for the purpose of protecting victims of abuse.
+After filing a report, the Git for Windows maintainer will contact you personally. If this happens to be the person who is harassing you, they will recuse themselves from handling your incident and ask another trusted Git for Windows contributor to handle it instead, follow up with any additional questions, and make a decision as to how to respond. We will respect confidentiality requests for the purpose of protecting victims of abuse.

 Anyone asked to stop unacceptable behavior is expected to comply immediately. If an individual engages in unacceptable behavior, the representative may take any action they deem appropriate, up to and including a permanent ban from our community without warning.

[PhilipOakley]

Was the middle change (introduce a linefeed) deliberate, or was something else intended.

Could the first change be improved by indicating that the email address is that of the maintainer, e.g. "contacting the maintainer via johannes.schindelin@gmx.de."? (i.e. bring forward the mention of the maintainer to here).

Looks good [https://github.com/dscho/git/blob/code-of-conduct/CONTRIBUTING.md].

[dscho]

Was the middle change (introduce a linefeed) deliberate, or was something else intended.

The linefeed was actually removed (because it was unintended in the first place).

Could the first change be improved by indicating that the email address is that of the maintainer, e.g. "contacting the maintainer via johannes.schindelin@gmx.de."? (i.e. bring forward the mention of the maintainer to here).

Changed.

The interdiff is:

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index df1eee4..af4c7fb 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -28,7 +28,7 @@ characteristics above, including participants with disabilities.

 ### Reporting Issues

-If you experience or witness unacceptable behavior—or have any other concerns—please report it by contacting us via **johannes.schindelin@gmx.de**. All reports will be handled with discretion. In your report please include:
+If you experience or witness unacceptable behavior—or have any other concerns—please report it by contacting Git for Windows' maintainer via **johannes.schindelin@gmx.de**. All reports will be handled with discretion. In your report please include:

 - Your contact information.
 - Names (real, nicknames, or pseudonyms) of any individuals involved. If there are additional witnesses, please include them as well. Your account of what occurred, and if you believe the incident is ongoing. If there is a publicly available record (e.g. a mailing list archive or a public IRC logger), please include a link.

[PhilipOakley]

The sentiment about the need for the C-of-C is something that in some ways is an issue, it's also a reflection on the wider social issues and the development of our civil society. These C-of-C's hopefully help address the institutionalised nature of some of the biases and tendencies we humans commonly have.

In the UK there was a report about 'Institutional Racism' in the London Metropolitan police force, where no individual was personally aggressively racist, but as a whole the combined actions of many small choices made the overall system biased, and demonstrably so.

The point of the C-of-C is to assist in the highlighting and removal of ungrounded biases.

I'd agree that 'equality' (which underlies the C-of-C) is a tricky social term with a contrary meaning for the technically minded. It's like comparing 4 and 7, and making the decision on greatness based on even numbers being somehow better than odd numbers (which would make 4 greater than 7). It's about making the discrimination/decision based on a true assessment of the relevant facts, rather than some stand in false parameter.

[dscho]

I never understood why these codes of conduct have to exist. They seem like silly forced guidelines authored by people that have nothing actually useful to contribute.

The point is that they are not forced: I want it. Nobody forces me to adhere to any code of conduct, but I choose to publicize the rules by which I intend to conduct this project.

And also if women want to contribute, by all means they always could.

Sorry, I find this attitude not welcoming. "If you want to visit me, you know where I live" is not something I would like any of my friends to tell me. Something like "I'd love you to visit me soon!" is more the message I want to convey. It is not so much the what, it is more the how we say it. It matters.

By the way I want encourage diversity out of purely selfish reasons: a diverse, empathy-rich environment is exactly what I like, therefore I want to do everything I can to make this project such a diverse environment.

And one of the things I do to that end is to work on this Pull Request.

[dscho]

The point of the C-of-C is to assist in the highlighting and removal of ungrounded biases.

I could not agree more.

Nobody thinks of themselves biased, nobody thinks that they treat equally competent male and female, or equally competent Caucasian and African-American, differently. Yet there is a growing mountain of evidence that we all fall prey to it. And the only way to counter it is to define the rules by which we want to judge (and be judged) whether we fail to overcome those biases.

And yes, that also entails that we display more empathy and go out of our way to invite people of different sex, of different race, of different background.

[dscho]

Okay, I think this PR has exhausted the discussion ;-) Will merge it in a moment. Thanks, all!