Draft code of conduct for R project

[hturner]

The possibility of a having code of conduct (CoC) for the R project was discussed at our kick-off meeting. The issue came up again when we applied to Outreachy as they expect participating projects to have a code of conduct. There was some discussion among the R Foundation after this and it seemed that people were open to having a code of conduct to cover R project spaces, e.g. the mailing lists, Bugzilla. This would also be useful to cover R Contribution Working Group events not covered by an existing CoC (e.g. the useR! conference CoC).

This would require some work to consider the CoC used by other projects and how it might work for the R project, including how it can be implemented in practice. It would be good to know if anyone is interested to help with drafting a proposal for the R Foundation/R Core to consider.

[llrs]

I am currently at the code of conduct committee of Bioconductor. There is some information about what and how it operates. See the website for more information: https://bioconductor.github.io/bioc_coc_multilingual/
I could help anyone wanting to take the lead.
There is also the rOpenSci code of conduct: https://ropensci.org/code-of-conduct/

[hturner]

The issue of a code of conduct for the R project is still a long-term goal, but we require a code of conduct for events such as further Collaboration Campfires/Office Hours that are not covered by a partner's CoC. This should be quick to turn around as the scope is much smaller, so we will start on this.

[hturner]

This is a duplicate of #30 (also opened by me!) where there is some helpful discussion and links. I will close that one now we have the link here for reference.

[brainwane]

(Disclaimer: I'm not an R user or R programmer, and am writing based on my expertise in other open source package management systems and improving other open source projects' ability to recruit and grow contributors.)

Based on the conversation in this Fediverse thread, there's a desire among many R people to have a more hospitable atmosphere both on the mailing lists and when submitting/resubmitting packages to CRAN. Thus it might be helpful to have a code of conduct to formalize expectations in digital spaces, and to have a lightweight way to follow up on incident reports without having to assemble a Court of Arbitration (section 10 of the R Foundation statutes).

I noticed that this issue is "In progress" -- could you share what the current specific step in progress is, and how other interested people could help?

I also have a specific suggestion: in your code of conduct, explicitly state that it's okay to forward private correspondence to the Code of Conduct working group/committee, even without the consent of that correspondence's original author, as long as such forwarding does not break a contract of confidentiality. Evidently, behavior in private email correspondence is a concern multiple R people have raised. But, per this mailing list thread, at least some people believe such correspondence ought not be published without the author's consent, even when the correspondence was from a CRAN volunteer in their capacity as a member of the CRAN team. So, as this commenter suggests, perhaps CRAN submission correspondence, such as acceptances and rejections, ought to be "accessible in case of code of conduct violations by maintainers or CRAN people." It would be nice if that correspondence were automatically centrally digitally archived for such access, but that would require some infrastructure work. Instead of or until that, an ombuds or a Code of Conduct body could at least receive forwards of such correspondence to review it.

Thank you for your work.

[dlaehnemann]

Another good starting point for a Code of Conduct is the Contributor Covenant. While the name focuses more on the contribution side (so mailing lists, Bugzilla, things here on GitHub, I guess), I think it could also be used for events. But I have no experience of this in action in a larger community, so it might make sense to reach out to someone in the projects already using it.

[shannonpileggi]

I guess from the opening comment it would be R Core/R Foundation that needs to approve a CoC?

And then who would be agreeing to adhere to the CoC? Those same institutions plus... individuals who submit packages to CRAN?

And if there is a violation, who would that be reported to and what action would be taken as a consequence?

[llrs]

At Bioconductor, the other repository besides CRAN, it is stated that anyone participating in the project (even on social media) should be aware of the Code of Conduct (CoC). Depending on how it is written/approved it could affect anyone submitting a bug report, a package, participating on the mailing list, forums, slack or any official part of the project.

There is a committee (I'm currently serving that committee) that receive those incidents and decide the course of action (see enforcement section).

The committee was initially formed by some senior people and other respected members of the community. They (we) serve the committee for up to 2 years. When someone steps down there is a call for new people to serve the committee/project. In my opinion, in this case the text and who would serve on such a committee, should be agreed from the R core and the R foundation (and perhaps there should be always one or more members of the R Foundation on the committee) and would be enforced at least on CRAN communication, mailing lists, bugzilla and the svn repository.

In case of an incident between the leadership of the project and the CoC committee there is also an ambudsperson from a different organization on the CoC committee to mediate. People from the CoC committee involved in an incident cannot participate in the investigation of said incident.

The consequences of an incident can be from closing it/none (after communicating with the reporte), to the exclusion of people (temporary or permanent) of the project (see procedure and the CoC). Other roles involve checking if people who want to join other governing bodies do not adhere to the values and principles of the CoC, or assisting with requests from the community regarding the CoC.

[shannonpileggi]

Thank you for sharing, @llrs!

Another exmaple here from the python community that I saw on mastodon. This post shares the outcome of a CoC violoation and also links to the corresponding contributing guidelines.

https://discuss.python.org/t/three-month-suspension-for-a-core-developer/60250