On Fri, Jul 19, 2019 at 2:52 AM Niclas Hedhman <[email protected]> wrote:
> Well, this is kind of ComDev stuff, but anyhow...
>
> I am not particularly fond of the "Comment can reasonably be expected to
> provoke an angry response", because it is incredibly subjective, as well as
> impossible to predict future.
> Possibly change that bullet point to "Comment is using too much emotional
> language and can easily be misunderstood". Similar tuning in others, but
> not important at this point.
>
I was a bit skeptical, but I tried it out in the text, and I do agree that
it's an improvement. Thank you for the suggestion Niclas!
> Instead, I think moderation should not have any different standards than
> the reason for being put on moderation in the first place.
I agree, but I don't see that as my call. Within this committee, Gris is
VP, and I'd like her to make this decision.
> Which means that
> we should delegate back to ComDev's guidelines[1][2], and if they are not
> adequate then patch it there.
>
I partly disagree here. I am not trying to create guidelines that work for
everyone, I'm trying to create guidelines that work for us. If another
nonpmc or committee decides they want to try out moderation, we can offer
what we have and then start thinking about finding the commonalities and
bringing them back to comdev. But I don't want to start with trying to
solve this problem for 65,000 contributors, when I still lack confidence
that we have it right here.
Just as an example: I think the standard on board@ should be different than
the standard on *@diversity, and should be different than the standard on
the various project dev lists. There should be commonalities (and those
commonalities *do* belong at comdev), but the topics on these lists are
different, the subscriber counts are different. The degree to which people
chose to be there or are forced to be there is different. These all play a
role in what expectations are reasonable.
However, Yes, I have been intending to mine the ASF CoC for moderation
messages. Especially section 5 contains a list of things that are somewhat
more objective and easier to explain.
One bit that I think is missing;
> * "Try to avoid using emotionally loaded language, avoid assigning fault
> and avoid absolute stipulations"
>
Yes, I believe the "emotionally loaded language" part of that is covered in
your above suggestion which I've adopted. But the latter part ("assigning
fault", "absolute stipulations") is not something I want to try to judge as
a moderator. I'm worried that I'll miss some, and get tangled in picking
sides if I try to take that on. That really will need to go over to the
level of social moderation and even off-list coaching (another solution
that I'd like to see us use more heavily; people react much more
constructively to criticism when it is not in front of an audience)
> Of course, examples would help greatly.
>
> So, I would like to suggest that
> a) In the Netiquette guidelines it may be good to have "If the netiquette
> rules not followed, you might be put on moderation after N warnings. This
> moderation is limited to X days. Each community within ASF defines N and
> X."
>
> b) Moderation guidelines should also be in ComDev, as a resource for
> other projects/lists. The above list is a starting point there.
>
> c) VP D&I should then set the X and N for these lists, but the higher
> those numbers are, the more backlash can be expected. maybe 30 and 2 are
> reasonable...
>
> d) After X days, moderators should evaluate if the person has improved
> the behavior, and if extension is required, need to point to the posts that
> were rejected and not improved upon successfully, and I suggest that
> decision is posted on private@
>
I have no objections to you taking this to ComDev Niclas, but I'm not
convinced yet that making moderation into a ComDev recommendation is the
right approach. Perhaps you can convince me there. : o) My goal with this
draft is not to create a new set of standards, it's to define a response to
violations of a subset of the standards that we have. That means:
* picking the subset of that standard that we want to handle in technical
moderation.
* defining the response that email recipients will receive.
Best Regards,
Myrle
[1] http://community.apache.org/contributors/etiquette
> [2] http://www.apache.org/dev/contrib-email-tips
<troll>By the way: is that recommendation on HTML still up-to-date? (vi!
emacs!) </troll> <ducks/>
