Talk:Etiquette/Process for Moderation

From OpenStreetMap Wiki
Jump to navigation Jump to search

The draft document was open for comments and suggestions from July 23 to August 18, 2021. Thank you for your comments and suggestions! The Subcommittee did our best to incorporate feedback and respond to each thread.

decision making process

they will make a decision by consensus

I interpret this as a “100% of the moderation team must approve & there must be 100% quorum”. The other meaning of “rough agreement” cannot apply to something important like “banning people from the mailing lists”. (cf. similar DWG ban policy) ____Amᵃᵖanda (talk) 09:06, 25 July 2021 (UTC)

The moderation team will make decisions as a group based on conversation and experience, but it’s not a legislative body with a formal voting process. Its decision-making would be similar to the OSMF board’s day-to-day business outside its monthly meetings. [subcommittee] SJFriedl (talk) 16:28, 18 August 2021 (UTC)

a public building will be held??!?!

A public town hall will be held the week of August 9 - exact date and time TBD. All participation welcome and encouraged!

What does this mean? How is a building held? Amᵃᵖanda (talk) 09:31, 25 July 2021 (UTC)

"Town hall" also means "meeting". See the 3rd definition here: --seav (talk) 11:18, 28 July 2021 (UTC)
Should probably just use the word "meeting" to avoid confusion then ;) GinaroZ (talk) 14:00, 30 July 2021 (UTC)

Language will be changed to say 'public meeting' [subcommittee] SJFriedl (talk) 16:28, 18 August 2021 (UTC)

Request for a public or private apology

A potential response from the moderation team could be to request a public or private apology from someone who has violated the OSMF Etiquette Guidelines. Question: could there be an instance where the moderators determine that a violator can be reinstated to the mailing lists only if the violator offers a public or private apology? If this is the case, then I do not agree with this potential response, as the apology would likely not be sincere. - Courtney Clark

A person is as good as their word. As a simple example, someone insult the intelligence of another person. They get asked to make an apology, and they say that their previous message was unfounded. In this example, you're right that they may be saying so just to get unbanned. This may work a couple times, but then both the readers and moderators will not accept further messages.
A second point is : what other avenue do you suggest for "redemption" ? For more serious violations, sure a temp/perma ban seems more useful, but for non-repeated, "lighter" (as in impact on the community) violations an apology seems wholly sufficient to me. --Gileri (talk) 09:10, 1 August 2021 (UTC)

I would expect requesting a public apology might come from discussion with the offender. E.g. Yeah, I didn't really mean that or on consideration I shouldn't have said it. Okay, can you apologise publicly and then we'll be in a good place to close this up.

Also, it's not excluding private apologies and no public action being taken (maybe wise if the thread went dead) LastGrape/Gregory 17:52, 6 August 2021 (UTC)

I do not think it is reasonable or fair to demand an apology, or worse, an admission of guilt, as a precondition for avoiding or ending a ban. I have been on both sides of this, and my experience tells that this concept just does not work. Either the offender agrees that their behavior was wrong, in which case they are likely to sincerely apologize on their own anyway, or they do not, in which case you are effectively demanding that they "plead guilty", which at best leads to a fake apology, but often will just be perceived as extremely unfair and unacceptable, deadlocking the dispute resolution. --Kevin Kofler (talk) 12:34, 18 August 2021 (UTC)

We agree this does not seem to be an effective response. We will remove this bullet. [subcommittee] SJFriedl (talk) 16:29, 18 August 2021 (UTC)

A polite email?

A formal warning email to individuals from a designated OSMF moderation account could be clearer than "A polite email" as that could be anything polite

Aslo 'Appeal Requests' could be more widely understood than Reconsideration Requests, especially by those who do not have english as a first language

This should indeed be clearer. Just a polite mail will frequently not be taken seriously, e.g. a moderator request formulated nicely as "Please refrain from doing X" has been interpreted literally as optional and ignored several times in the past. Also what I am missing from the list of measures is such a request to cease a certain offensive behaviour. "Polite mail outlining the issue" definitely sounds much weaker than that. --Nop (talk) 18:16, 17 August 2021 (UTC)

Agreed. Will change the language to read "a formal warning email to individuals". To the second point, that email could include a request to cease a certain behaviour. Reminder that this is not an exhaustive list of actions available to the moderation team. We intend to promote translations of the entire document to enable comprehension in as many languages as possible. [subcommittee] SJFriedl (talk) 16:29, 18 August 2021 (UTC)

Ukrainian community suggestions regarding moderation limits

As we are currently finding our way out of a conflict with DWG which, as we see it, abuses its position and bans users depending on their speech, we suggest the following:

1. Explicitly state in the Etiquette Policy that CWG, not DWG, holds exclusive power regarding moderation of communications. Currently DWG appeals to Etiquette Policy as the grounds for applying their bans. That's something we consider inappropriate, as DWG's domain is, as its name suggests, data, not communication.

2. Explicitly state which channels undergo such moderation. If it covers exclusively the mailing lists, we suggest phrasing the policy accordingly. We would also like to remind that OSM communication includes OSM forum, and OSM forum moderation is being performed by moderators voted for and elected by the respective communities (e.g. national communities). We feel that this situation is adequate and needs no change - and setting up explicit lists of places being moderated according to the policy eliminates the possibility of it being applied as broadly as possible (exactly what DWG currently tries to do).

3. Explicitly state that OSM supports freedom of speech as a general rule of thumb. We've seen troubling signs in the mailing lists, with some people claiming that freedom of speech is somehow "ideological" oe "political" and goes against the values of their countries and/or communities. We, however, strongly believe that Universal Declaration of Human Rights, articles 18-19, is just that: universal. We believe that these rights should apply to anyone and everyone by default - unless these people actually seek to harm other peoples' rights. For us, that's what moderation is, in a nutshell. We would also like to remind that some of the OSM practices are implicitly based on the concepts of freedom of thought and freedom of speech - such as Any tags you like.

Kilkenni (talk) 11:53, 27 July 2021 (UTC)

For more context for the above, see here - both the diary entry and my reply to it. SomeoneElse (talk) 08:23, 28 July 2021 (UTC)

We have received a Scope of Work from the OSMF board to cover only talk@ and osmf-talk@ email lists with this process for moderation. For now, this moderation process will not cover communications with DWG or other working groups. If the proposed process proves to be successful thanks to community support and participation, we would love to see it applied in areas beyond the initial scope. [subcommittee] SJFriedl (talk) 16:32, 18 August 2021 (UTC)

Incident Documentation should be public by default

Just like bans are documented, I believe restrictive measures such as temporary or permanent blocks should be documented. Based on the guidelines, blocks are far from the first action for most violations, so justifying them to the community should be easy. That provides some balance between users and moderators, and allow spotting, discussing and acting on possible over-zealous moderation as a community. A last point, I think that by being transparent on who agreed to block someone would protect the CWG reputation and trust uf there are a few ill-intentionned moderators. --Gileri (talk) 09:22, 1 August 2021 (UTC)

The purpose of moderation process is to smooth communications on the lists, rather than to focus on punishments or violations. The Moderation Team will provide updates to the OSMF board for transparency rather than the overall community so the board can have oversight on the moderation team. [subcommittee] SJFriedl (talk) 16:30, 18 August 2021 (UTC)

It sounds like you are avoiding the issue. We all agree that the purpose is to smooth communication, but as outlined here violations and punishments are unavoidable. Forming a team with a lot of discretion regarding what punishment to apply, little oversight and no transparency to the public is ripe for intentional, or unintentional abuse. We have a lot of examples all around the world about that, no need to bring the same issues in OSM communities. We also have examples of silencing tactics being used to target well-meaning and really productive members of our community.
If the Moderation Team can stand behind the punishments it delivers, why the need to keep them secret ? --Gileri (talk) 18:18, 18 August 2021 (UTC)
I agree that transparency can be an important part of accountability, but in this case keeping things a bit more private protects the person that has been subject to the offense. If results of a complaint are made public, this creates a risk of even more judgement from the broader community and could result in further abuse of the offended person. --MaggieMaps
>If results of a complaint are made public, this creates a risk of even more judgement from the broader community and could result in further abuse of the offended person.
I believe in most jurisdictions final decisions are made public, and the retaliation may be way harsher than mean comments online. Also, if someone is banned, they will be allowed to control the discourse more easily on a medium outside the Moderation Team purview anyway. I don't see how keeping bans under wraps address that. Targeted and untargetted campaign will happen, and in high-value cases the Moderation Team will surely need to communicate on their actions anyway. --Gileri (talk) 17:26, 19 August 2021 (UTC)

Requirements for moderators' messages

The OSM community consists of a large diversity of people from various cultures and countries. We do have cases where people struggle to follow rules from a different culture (e.g. due to limited knowledge of the English language). Moderators' actions should not only protect the individuals from insults, harrassment etc. but we inclusive and welcoming as well. They should try to support onboarding of community members from foreign cultures. This means that someone whom the moderators act against needs to be told what they did wrong. The "violator" should be able to understand the message, i.e. it should include a translation if necessary.

Therefore, I propose to add the following section:

Requirements for Communication

Messages by the moderators to people violating the rules must meet the following requirements:

  • They must include a brief description of the violation (e.g. "on 15 December 2021, you insulted Joe Average on the Talk mailing list (see quote of your email below)").
  • The violated rule must be named.
  • The message must inform about the right to appeal if the user is banned or put under moderation.
  • If the moderators can assume that the recipient might be unable to understand the English message, a translation should be included.

--Nakaner (talk) 11:30, 18 August 2021 (UTC)

We will add to the Incident Response section: "When responding to a report, the moderator should include a brief description of the violation, the rule that was violated, and acknowledgement of the right to appeal." For right now, the talk lists are where this will be applied and they are currently in English. If the process is successful here, we definitely encourage the expansion beyond English. [subcommittee] SJFriedl (talk) 16:30, 18 August 2021 (UTC)

Incident Documentation legal status

The incident documentation will necessarily include personal information about individuals, some of which is likely to be potentially damaging to them (eg. x acted as a bully, or y has such and such sexual status, or y and z were once partners).

  • What are the legal responsibilities of OSM if this documentation is hacked and broadcast?
  • Where will OSM list members give their consent to keeping this personal information under the European GDPR? Will existing list members be required to give retrospective consent?
  • How will data subject access rights be implemented without allowing access to other subjects' data (eg. when x attacks y in some way and x wants to see their own record)?

Marinheiro (talk) 17:39, 27 August 2021 (UTC)

research: "educate people about why .. "


"Does Transparency in Moderation Really Matter?: User Behavior After Content Removal Explanations on Reddit"

"Adopting a topic modeling approach, we show that removal explanations often provide information that educate users about the social norms of the community, thereby (theoretically) preparing them to become a productive member. We build regression models that show evidence of removal explanations playing a role in future user activity. Most importantly, we show that offering explanations for content moderation reduces the odds of future post removals. Additionally, explanations provided by human moderators did not have a significant advantage over explanations provided by bots for reducing future post removals."

IMHO: the other research is also importants :

--ImreSamu (talk) 08:32, 3 September 2021 (UTC)