I for one think that requiring a specific filesystem structure or notice in a git repo is quite far afield from the sorts of things that CoC is designed to deal with.
I agree. I do think that as a community of practice we have many unwritten rules and numerous expectations of how we work together. We don't explicitly define the expectation of a README.MD file in repos either.[0] It's a best practice and cultural expectation in our spaces to include one. The code works the same with or with out it.
Yeah, sure a coc.md isn’t “the same”, but both are expected as something we do as a community. If we need to write that down somewhere so there's no repeat confusion on if it's expected or not, that seems like a good compromise. However, I'd like to think we don't have to define everything, but ¯_(ツ)_/¯ .
[0] I'm waiting for someone to contradict me on this risky comparison. :) I could not find anything explicit in https://www.mediawiki.org/wiki/Gerrit/New_repositories or https://www.mediawiki.org/wiki/Manual:Coding_conventions
Yours, Chris K.
On Friday, June 8, 2018, Chris Koerner nobelx@gmail.com wrote:
I for one think that requiring a specific filesystem structure or notice
in
a git repo is quite far afield from the sorts of things that CoC is designed to deal with.
I agree. I do think that as a community of practice we have many unwritten rules and numerous expectations of how we work together. We don't explicitly define the expectation of a README.MD file in repos either.[0] It's a best practice and cultural expectation in our spaces to include one. The code works the same with or with out it.
Yeah, sure a coc.md isn’t “the same”, but both are expected as something we do as a community. If we need to write that down somewhere so there's no repeat confusion on if it's expected or not, that seems like a good compromise. However, I'd like to think we don't have to define everything, but ¯_(ツ)_/¯ .
[0] I'm waiting for someone to contradict me on this risky comparison. :) I could not find anything explicit in https://www.mediawiki.org/wiki/Gerrit/New_repositories or https://www.mediawiki.org/wiki/Manual:Coding_conventions
Yours, Chris K.
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
The issue is, it seems like this is not something we "do" as a community:
* There was a previous discussion about requiring coc.md. there was a lot of arguing and no clear "winner", but a very significant portion of the opinions was that CoC.md was highly recommended but not required if the extension maintainer didnt want it. Thus supporting Yaron's position. https://phabricator.wikimedia.org/T165540#3358929 * There is a general community norm that overriding a -2 by a maintainer of a component is an extraordinary action, even more so when the person doing it is not a maintainer of the extension. This situation is no where near clear cut enough to justify that without discusion * Generally speaking, its usually considered in poorform to have an argument about something, lose the argument (or at least not win it), wait a year until people forget about it, and then try and do the exact same thing.
-- brian
Hi!
I agree. I do think that as a community of practice we have many unwritten rules and numerous expectations of how we work together. We don't explicitly define the expectation of a README.MD file in repos either.[0] It's a best practice and cultural expectation in our spaces to include one. The code works the same with or with out it.
I think there's several aspects here to consider:
a) In WMF technical spaces, and more widely, in the Mediawiki/Wikimedia universe, I think there's universally acknowledged expectation of certain standards of behavior, which in the Wikimedia space have been codified in the CoC. The purpose of these expectations, as I understand them, is to build and maintain an open, welcoming, productive and inspiring community that would support development of Mediawiki and Wikimedia projects. And the CoC is the instrument that we chose to codify and implement those expectations in the Wikimedia spaces, which applies to all of them regardless of technical means chosen to publish or document it. I do not think there is much disagreement about that.
b) How exactly the spaces are managed within this wide framework has a lot of complex and tricky details. Some of which may seem trivial to some people and highly sensitive to other people. Including which files are placed in which repositories, who is allowed to change which repository and for which reasons and procedures, and so on. I think having more clear expectations on that would certainly help.
But beyond that, I think when designing and enforcing the rules for these minute details, we should not lose the sight of why it is done, and not make the process of CoC enforcement go against the goal of having CoC - namely, the welcoming community. If that means sometimes being more flexible, or having a bit more patient discussion and resisting the urge to force your point through, even if I am completely sure I am correct, I think it is still worth it in the long run.
c) Specifically about CoC.md file, I personally think having redundant pointers to the documentation (both technical and about societal norms) is highly welcome, as locating proper docs is notoriously hard and largely unsolved problem with most code. Having the docs is half of the problem (which we also sometimes fail at ;), having it where people would find them is the other half. So adding of the CoC file from this point of view is a smart move.
On the other hand, maintaining a rigid "one size fits all, no exceptions, no discussions, shut up and comply" approach to it feels a bit counter-productive to me. Yes, I foresee the question "if it is a good thing, why not make everybody do it?" - and I could probably easily write a 20-page essay on this topic, but I would limit myself here to this - people have different points of view, and I think being more accommodating in this case is better than having a nice set of checkboxes checked.
What it specifically means for the specific file? I admit I don't have a better proposal than "let's have a community discussion on it". But I think making an open and welcoming community including sometimes being patient in figuring out how exactly to do it. Enforcing having the file in every single repo does not seem to be a pressing concern that would do any harm if not brought into compliance right now. So let's see if we can reach some consensus here.
wikitech-l@lists.wikimedia.org