ietf / moderators Goto Github PK

I suspect that it would be worth specifically addressing the line between "people" and "their ideas." This is likely to be contentious, but it's already come up more than once.

For example, which of the following statements would be considered unprofessional commentary? (all names are just examples)

• "Bob, your idea sucks, is stupid, and will kill the Internet."
• "Zoe's term as Area Director has been a disaster, and we'll be very happy when she's gone."
• "Bruce, your idea is flawed, because it doesn't consider the following situations..."
• "Joe works for FooCorp, and we know that they're Evil, so his idea should be rejected."
• "Only an idiot would come up with this idea."

Personally, I think the criticism of Bruce's idea is the only one above that's a professional message. That's because even though the rest attack an idea, they do so without actual valid criticism (typically, detailed arguments as to why the idea is bad), so they are effectively using the idea as a proxy to attack the person. Nice hack, if you can get away with it.

This work guides the SAA. At some point we should ask about the proper roles of other players.

To main

Sometimes people who really do add value can also just lose it. I saw this as a working group chair, and in consultation with my AD and with the person involved, we agreed to moderate his mail to the list instead of banning him entirely. Having this sort of tool in the tool box requires some assessment as to whether the person has contributed value, or is just a troll.

Unprofessional behavior means inundating a large number of people with a large number of messages in a short period of time. These sorts of rapid-fire responses are generally not well considered, especially if they are heated.

Advice should be given to the community as a whole that when conversations are heated, people should avoid such rapid-fire responses, pause, and reflect on the conversation, and then respond, if one feels the need.

I think it might be worth it to add insults as a subsection to the Descriptions section of unprofessional commentary. One clear example comes to mind: "The idea you proposed is stupid." It would be nice to send a clear message that this kind of language is unprofessional, and not tolerated on IETF lists.

The current subsection on "personal attacks" mentions "insulting others", but I worry some might interpret that as allowing insulting others' ideas.

One thing that appears to be missing from the SOP is Level
3, at which you recommend, or encourage others to initiate, a PR
action under RFC 3683. It seems clear, at least to me, that one
14 day suspension, a brief pause, and then more of the same
would not serve either the individual or the community very
well.

This text:

A report made by a member of the SAA team should include a suggested action in response, including specific text for email(s) to the original poster and/or the list, as appropriate.

Seems to read better as "A report made to a member of the SAA team..." or is this just jumping to the response from the SAA and that's also called a "report"?

Right now, the first/default activity of the SAA is to quietly intervene by sending the person in question a private e-mail, with no public activity.

I think it's worth examining the thinking behind this.

ISTM it's attempting to be sensitive to the feelings of the person whose behaviour in question, giving them a chance to correct their behaviour without being publicly shamed.

While I don't think we should be trying to shame people, this approach leaves me wondering if we've considered the other effects of this policy -- namely that the target of the behaviour is left having to endure how they've been treated, and the rest of the community is left with the impression that the behaviour may or may not have be acceptable.

In particular, if there isn't any visible reaction to unacceptable behaviour on the list, is there a risk that that effectively moves "the line" towards unacceptable behaviour?

Possible mitigations (not mutually exclusive in every case):

• Don't have a level 0; every incident is mentioned on-list, to give visibility
• Require some kind of restorative justice process to be followed
• Publish a weekly/monthly summary about how many incidents there were at each level

Calling it "friendly suggestion" uses a subjective term that the recipient may not agree with. Some other term (e.g. "initial suggestion") may capture the interaction in more neutral terms.

Suggestion to update the email template to the list to include:
(a) A sentence and pointer to the moderation rules (perhaps both https://github.com/ietf/Moderators/ and https://github.com/ietf/Moderators/blob/main/sop.md).
(b) A sentence that clarifies that this is the second step (level 1) in the moderation procedure, and that individual posters had previously been privately notified that their posting was against the community guidelines and yet they continued to post.

It's not clear whether there are circumstances where the SAA can immediately escalate to level 1 or level 2.

If someone says something so vile / obviously unprofessional that half of the list is wondering if there's going to be a response, it seems like it's a good candidate for immediate escalation.

The SAA should exercise caution in their role. The SAA should avoid engaging in ambiguous circumstances. Mike St. John's initial response to the call for comments on the SoW was a good example. As I wrote on list, to me that was a last call comment to a topic that had been hotly debated in plenary, and Mike was inbounds responding to the IETF list. Without relitigating all of that, what appears as overly aggressive intervention lands us in meta-discussions about what's appropriate.

To avoid unnecessary embarrassment, might it make sense to offer the individual the option of not having that note sent, with the caveat that continued violation can lead to a posting rights suspension?

The aim should be to provide some education in the process -- minimizing threats about punishments of the "or else" variety-- the situation is very different than a policy and set of procedures that seem to focus on escalation and punishment. In that regard, at least some of the templates mentioned in your note may deserve another look to adopt more of a "we are here to help you work better and more effectively within the IETF" tone and rather less of "you have been bad and, if you don't stop it, we are going to punish you". If an extra template or piece of one that includes "you have been participating in the IETF for many years and hence know better but might need a reminder" would help, consider that too.

The filename is the problem (https://github.com/ietf/Moderators/blob/main/unprofessional-commentary.md).
When renaming it, we need to also update links in the email templates that point to it.

The template https://github.com/linuxwolf/ietf-saa/blob/master/email-templates/announcements.txt seems to only indicate an involuntary suspension of posting rights... maybe reword to allow the language to accommodate a voluntary cooling off and then involuntary suspension?

Not sure I am doing this correctly in github, but regarding:

Personal attacks
Personal attacks include attacking people for their opinions, beliefs, or ideas rather than criticizing the opinions, beliefs, and ideas themselves. Personal attacks also include name-calling and insulting, demeaning, or belittling others.

I think it is good this is clear it is about attacking "people". For example, someone saying that Bob is stupid.

I think two things are missing from the IETF-SSA. These are:

Criticism of the actions of the IETF leadership bodies should be given a lot of latitude. Not to the point where it falls into personal attacks on people (that is, individuals) in the leadership body, but it should be fine to criticize the actions of a leadership body. Other wise, the SSA will be perceived as being used to censor the IETF community to protect the leadership.

The other is that with the exception of the IETF Chair, members of the SAA team should not hold IETF leadership positions. I understand the issues of getting people to do this job, but there is too much inherent conflict of interest otherwise.

There should be another (perhaps optional) level between 1 and 2, i.e., a second public warning before the temporary posting rights restriction.

I see this as effectively giving you the option of a slower escalation path in cases where that might be helpful. You would still retain the immediate escalation option when appropriate.