Open main menu

Talk:Strategy/Wikimedia movement/2018-20/Working Groups/Community Health/Recommendations/Rules and regulations, decision making processes and leadership

Active discussions
The following discussion is closed. Please do not modify it.
Most likely, new comments will not be taken into account by the new three Working Group members in their work of developing the final Recommendations. You are free however to continue discussing in the spirit of "discussing about Wikipedia is a work in progress". :)

CommentEdit

I don't know where there's an opportunity to comment on these, but oppose any global policies. Each project sets its own rules, and for good reason. We should not have global policies which may work fine for one project and terribly on another; that violates the longstanding principle that each project is independent and self-governing. Seraphimblade (talk) 12:18, 10 August 2019 (UTC)

Pinging author SGrabarczuk (WMF) since I'm not sure if you're watching this page. If there's somewhere else I should comment, please let me know, but that is not apparent. Seraphimblade (talk) 12:33, 10 August 2019 (UTC)
Seraphimblade, this is the right place for your comment :) Each recommendation may be commented on its talk page. This is even advisable. Yes, I watch the page, and I hope that the Working Group members will watch it as well. SGrabarczuk (WMF) (talk) 13:03, 10 August 2019 (UTC)
SGrabarczuk (WMF), thanks, I found where it says that now. So to answer the questions posed:
  • What do you like about this recommendation? What do you dislike? Why?
  • It might work as a set of suggestions. Not a set of mandates. Different approaches work for different projects, and projects are self-governed, not dictated to from the WMF or Meta. They're not all bad ideas, but what is a bad idea is to make them mandatory across the board. (There are also some just bad suggestions; functionaries should not be "term limited". Experience is a good thing. Also, there should not be a universal code of conduct for on- and offline; interaction online is fundamentally different.)
  • What does this recommendation mean for you in contexts that are meaningful to you - your language, your projects, your region, your affiliate group, your identity? How would you be affected?
  • It would mean that the English Wikipedia, Commons, and any other projects in which I participate, would no longer be able to self-govern and set their own rules. For the English Wikipedia, I would think the Foundation would have learned we will fight such encroachment tooth and nail after the Fram incident. Did you not learn the lesson that time? Because it quite hurt our community, and I would really rather not go through it again.
  • What is important to keep in mind when implementing this recommendation? What values and ideas are important to preserve? Are there any red lines we should not cross?
  • As above, you are proposing to cross the biggest, reddest line there is—community self-governance. And again, keep in mind what happened the last time we did that. Many of the users who left due to that incident never returned, and our community is poorer for it. You made the mistake of hamfistedness with Superprotect, with Visual Editor, with Fram, and in many other cases. Every single time, it has exploded spectacularly and ended terribly. How many reminders do you need that you must not ever do that again? The WMF exists to serve the communities, not to rule them.
  • Does this recommendation support our Strategic Direction to become the support system for the whole free knowledge movement?
  • No. Attacking and destroying the volunteer community is directly at odds with that direction, and any attempt to undermine self-governance will destroy it. You should already know that from the last time you did it, which was barely a couple months ago. Is there something at the WMF which keeps these lessons from sinking in?
  • In general, what we have already done has worked, and worked well, for nearly two decades. We neither need nor want massive, sweeping changes dictated from above, and you must refrain from that. WMF IS NOT IN CHARGE, and it better figure that out, right quick. We welcome your suggestions and your advice, but not any edicts from on high. Seraphimblade (talk) 15:24, 10 August 2019 (UTC)
Let us not shoot the messenger. Most of this was probably drawn up before the Fram incident, possibly by people not involved in that particular fiasco. Other than that, I agree with the sentiments expressed by Seraphimblade. WMF need to tread very carefully here. There may be a significant disconnect between the people working on this strategy stuff and the people who are producing and maintaining the content. · · · Peter (Southwood) (talk): 20:32, 11 August 2019 (UTC)
As I understand it, this set of recommendations was drafted by this group of people, most of whom do not appear affiliated with the WMF − so, to reemphasize this point, this was not drafted by « the WMF ».
(In general, and while I understand the sentiment behind the points made (and would agree to some), I find it unhelpful to frame this as “Us vs. Them”.)
Jean-Fred (talk) 23:06, 11 August 2019 (UTC)
Jean-Frédéric, the WMF chose the members and subjects that these reports would be on, and paid a substantial amount of money for them. So, honestly, I think that's at most a distinction without a difference. This entire thing took place at the direction of the WMF, regardless of who actually did it. Seraphimblade (talk) 23:51, 11 August 2019 (UTC)
Not really. The WMF is funding this process yes, but not running it. The topics for discussion by the working groups were decided at the 2018 Wikimedia Conference, not by the WMF. Looking at the strategy core team, which chose the working group members, I don't see anyone who had previously worked for the WMF. Then, of course, a large majority of the members of the working groups are not from the WMF, and include a fair number of people who have had significant disagreements with the WMF in the past. It is really hard for me to imagine a strategy process for the Wikimedia movement in which the WMF played less of a role. Of course you are welcome to disagree with the recommendations, but don't pretend this is all some kind of WMF conspiracy. Chris Keating (The Land) (talk) 08:42, 12 August 2019 (UTC)
Not too much to add to what The Land said, just one point: It’s a bit unclear what you mean by “[WMF] paid a substantial amount of money for them”: while the process was indeed funded, the folks on the working groups (bar a few of WMF staff) were there in vounteer capacities and were not paid for their work on it. (not sure whether that’s what you were implying, but making it very clear). Jean-Fred (talk) 09:10, 12 August 2019 (UTC)
@The Land: - Agree with you that WMF is not much involved. On the other hand, the application process report states:- The 24.4% of non-organized applicants is not proportional to the actual size of the non-organized part of the movement ... Contrary to their next line, that back-pats themselves for having reached out well-enough, no-body from en-wiki (apart from those who are actively involved with affiliates/user-groups) knew damn anything about this survey (and it's alleged vitality) including ArbCom and all that. It was after the Framgate fiasco -- when these formation of working groups has been long over and the surveys were nearing completion -- that the first public notice was posted by a benevolent arbitrator over VP, (who somehow landed up over here by chance). That led to some publicity at the dying phases and its just that. I leave it to your better judgement, as to whether the volunteer community of en/de/nl, who are not involved in any affiliates/user-groups are represented in the working groups proportionately -- AFAIS, there's a hugely disproportionate presence of volunteer folks from affiliated bodies. You need to appreciate that intersectionality plays at different tiers and there exists classes of volunteers. Winged Blades of Godric (talk) 09:15, 12 August 2019 (UTC)
It looks like 12/13 of those individuals are affiliated with the WMF, or at least affiliated with a WMF Affiliate. --Yair rand (talk) 19:09, 12 August 2019 (UTC)
I agree with what Peter wrote up there : there is a disconnect, but I would add there is more than one, and there is also an internal disconnect within communities. Our movement needs to build counter powers to make place for people who are not at the moment the most welcome to contribute, and this involves a way of handling harrassment when it is not dealt with properly within the community. There are problems that are underlined by people who do not belong to white educated male of the northen hemisphere. Seraphimblade has written "In general, what we have already done has worked, and worked well, for nearly two decades". Well no. More and more people are complaining about informal organized lobbies who set out to target people they dont accept on the projects. We need to build systmes to adress cases when the community does not handle harrassement in a fair way. And this is being asked by members of the community, members who keep saying they have been targeted by harrassment, especially contributors active in the gender gap reduction. This is being constantly asked, mediatized over and over again. So room has to be made so that this demand sinks into the community's ears, because I do see the foundation quite active on these subjects and - as is justly remarked above, the WMF should not have to embark on this ship. But in fact WE - the people harrassed - are constantly asking the WMF and chapters to DO SOMETHING, and it is reassuring to see they do try, becuase it means our voices can be heard, and somebody can be made accountable in fine.
Some influent members of the community just do not see the offense happening just under their nose, they sometimes are in a state of pure denial, and very often the person who discloses the offense is in turn accused of witch hunting, which is something utterly unfair and profoundly destabilising to experiment.
No project, absolutely no project should be allowed to let harrassement be pursued because of the solidarity of its influent members. There should be counter powers. Either the community sets them up, either the WMF, chapters and UG will ne pressed to undetake whatever actions it takes for it to stop.
NO to harrasssment. Shluss, Punkt. Nattes à chat (talk) 21:35, 18 August 2019 (UTC)
[I'd like to add something privately, so I'm using my normal (volunteer) account.] I agree with you Nattes à chat, however, I feel I must emphasize two things. First of all, please let's not use the model "white educated male of the Northen hemisphere". Leaving obvious inequalities within the US or Western Europe aside, please take an example: white educated lower-class males from North Asia or Central or Eastern Europe have little in common with wealthy WASPs from the US or old Western European elytes. The meaning of education does differ in various groups and places. Inequalities between genders may differ (I mean, in some countries, females are massively unequal to men, and in some, they're less unequal). What's more, it's possible for a wealthy citizen of India (note: Global South, Southern hemisphere, not necesarily male) to be privileged over an average educated inhabitant of Latvia (white, European Union, Northern hemisphere). Conclusion: some models may be useful and show real tendencies, but there's a risk of oversimplification. Secondly, I doubt if the harrassment is the only argument for a change of responsibilities between the amorphic online communities and organized groups. It may be a good example, but it's not necesarily the main reason. This first point overlaps with the Diversity, and second one - with the R&R, but let's take this into account in this conversation as well. Tar Lócesilion (talk) 19:26, 27 August 2019 (UTC)

Term limits for functionariesEdit

I am not sure about this... Is the suggestion to only allow check users to be a check user for 6 years and than remove that ability? Term limits are different from needing to reconfirm. Doc James (talk · contribs · email) 13:47, 10 August 2019 (UTC)

James, you were in the working group that produced these reports; how can you be uncertain? Winged Blades of Godric (talk) 09:09, 11 August 2019 (UTC)
I was unable to make the last meeting. I encourage everyone to go thru and provide feedback on the strategy statements.Doc James (talk · contribs · email) 10:07, 11 August 2019 (UTC)
  • What purpose would this serve? Is there any evidence that there is a problem and that this may fix it? · · · Peter (Southwood) (talk): 04:38, 12 August 2019 (UTC)
  • Winged Blades of Godric or anyone, where can I find who was on these working groups? SarahSV talk 21:21, 14 August 2019 (UTC)
      • User:SlimVirgin now that we have drafts, which are an excellent start, I think it makes it much easier for the wider movement to engage with the strategy and I encourage people to do so. Doc James (talk · contribs · email) 06:46, 15 August 2019 (UTC)
    • It's okay, I found it. Doc James, a huge problem with these groups is that they focus on people who go to meetings. There's a lot of talk about diversity, but the movement excludes people who can't do the off-wiki activities for various reasons. SarahSV talk 21:25, 14 August 2019 (UTC)
Agree that, in general, term limits are probably less useful a tool than fixed-term positions with requirement for re-appointment to ensure that functionaries are in line with evolving community expectations. T.Shafee(Evo﹠Evo)talk 01:22, 15 August 2019 (UTC)
Should we be expecting any responses to requests for clarification from the workgroups? Do they have spokespersons? · · · Peter (Southwood) (talk): 11:40, 15 August 2019 (UTC)
Based on the "mandatory training" discussion on the talkpage of another recommendation, I am wondering if the recommendation uses "functionary" to refer to something other than Checkuser and Oversight. I take it, what is the definition? Jo-Jo Eumerus (talk, contributions) 16:28, 15 August 2019 (UTC)
Hello, when I read "term limits", I supposed indeed that it was meant that the term ends, but that the "functionary" can be immediately a candidate again for a new term. That is a good thing. If "term limit" is meant to limit a term and then prohibit the person to be a candidate again (for a certain time), I would strongly disagree. In many small wikis this would just not work. Don't forget that small wikis depend on a very limited number of people. Prohibiting someone to renew the term does not automatically create new ("diverse") candidates. I find the proposal much to radical and not considering the immense differences between wikis and positions. For example, the number of the members of an ArbCom is limited, the potential number of admins is unlimited. --Ziko (talk) 07:50, 17 August 2019 (UTC)
Prohibiting immediately repeated candidacy might be problematic on small wikis, however on big wikis it might be beneficial, as there is significant power-concentration, and a change in control structures could relax that a bit. — Aron M (talk) 08:30, 21 August 2019 (UTC)
If reRfA is required there should not be any need for an obligatory gap, as any issues can be brought up at the new RfA, and skilled workers need not be thrown out of the labour pool without due cause. · · · Peter (Southwood) (talk): 07:27, 22 August 2019 (UTC)
  • While we are discussing this point, has anyone proposed an actual term limit yet? The efficacy and difficulties would both be strongly influenced by how long the proposed limit would be.· · · Peter (Southwood) (talk): 07:32, 22 August 2019 (UTC)
  • term limits … to make room for new leadership to come through implies that there is a limited number of "seats at the table". If there were, say, 400 active admins, that doesn't prevent person 401 from being granted admin abilities. Plus, anybody can participate in policy discussions on en-WP, you don't need somebody to stand down from their role for you to get a say. Do the authors of this realise that the wikis aren't governed by elected committees or parliaments? Pelagic (talk) 10:28, 27 August 2019 (UTC)

Development of expected normsEdit

The strategy page makes it sound like WMF will issue commands from on-high about what is acceptable or unacceptable conduct. However, for those projects where there are already significant community-approved processes for policy setting and dispute resolution, it is essential that there be a meaningful dialog between WMF and the community about how conduct will be classified. The recent debacle at the English Wikipedia over the Fram controversy, as well as the clear directive from the WMF Board of Trustees resulting from that controversy, make it very clear that the larger communities have already had extensive experience in what does and does not work, and may very well understand the nuances of these things better than WMF staff do. This needs to be a genuinely collaborative project and not a set of divine commandments. --Tryptofish (talk) 16:32, 10 August 2019 (UTC)

It would be difficult for the larger communities not to understand the nuances of these things better than the WMF as they have already demonstrated their poor judgement more than once. We are dealing with communities who operate in a way that experts find surprising - read the research reports - so any claims that they have hired or will hire professionals should be received with skepticism. We do not need "professionals" who think they know what they are doing, with professional arrogance. We are breaking new ground here. Jimbo is not going to be happy with the people who break his serendipitous creation. The optimal norms for a functional wiki may significantly vary over its development and with the composition of its members. Who knows what effects sudden and sweeping changes may have on an ecosystem that functions significantly differently from the well studied and plausibly explained examples? The smaller wikis need an environment which encourages growth. The larger wikis need an environment that allows quality assurance. We do not need social experiments without a control group and no ethics committee oversight, pushing questionable political agendas composed largely by outsiders. · · · Peter (Southwood) (talk): 05:14, 12 August 2019 (UTC)
I totally agree with you Peter on a theoretical level, but I also believe that harrassement is not dealt with properly, and this comes from what I have experienced establishing les sans pagEs on the francophone wikipedia. Lobbies with political agendas and conflict of interests clearly are able to do a lot of harm out there. I think we need some systems establishing counterpowers to adress these cases. The Signpost is clearly one such an instance of counter power. The WMF is another, T&S is one, but we need more because the most efficient harrassers are influent people who tend to cumulate responsibilities : being involved / employed in a chapter, while contributing as an sysop and also making decision at WMF level is something very current, and it is always difficult to "disagree" with such people. If you do, you expose yourself to many many problems.
We do not need professional arrogance ++, we do not need influent contributor arrogance ++, we do need efficient counter powers that can be set in motion to check from time to time if WMF, or the Community has not been dealing with cases of harrassment fairly. Let's not be naive and talk only of harrassment within the "community" as opposed to chapters and WMF. Harrassment does not limit itself to the online projects ot the "community". It is everywhere. It is a culture that we can choose to nurture by refusing to adress it, or not. We have to set up counter powers to each set of identified powers that have become over arrogant with the years.Nattes à chat (talk) 21:58, 18 August 2019 (UTC)
We are stuck between a rock and a hard place, and need to get out of it without losing too much, either of our productive members, or quality assurance. The projects started off with the anyone can edit philosphy, and we have lots of evidence that while in practice, anyone who can jump through a few minor technical hoops can physically edit, there are a lot of people who edit badly, either through lack of competence in the subject matter, indifference or ignorance of how to edit constructively, intentional disruptive behaviour, a desire to push an unbalanced PoV, to promote disinformation, push other people around, or other problematic behaviour. Sometimes these behaviours are from people who genuinely believe they are working in the best interests of the project, and I think we have some people who believe that they are right and because they are right, they have the right to use tactics outside of behavioural expectations, and to stretch the rules to breaking point in pursuit of their right understanding of the good of the project. Some may even believe that it is necessary to destroy the projects for the greater good. Meanwhile we are all expected to assume good faith. Sometimes abominations are perpetrated in good faith.
I do not think that a group of enthusiastic amateurs with little personal involvement will come up with a system that works first try, and there is unlikely to be anyone who has much personal involvement in more than a small handful of projects, and the skills to come up with a workable solution for all the projects. I am not optimistic that the damage that could be done will be repairable. By getting each project to develop their own system, with external feedback, rather than the other way round, there is a better chance that at least some may come up with workable systems, and the risk of irreversible havoc should be reduced. We need expert advice, not professional advice. Sometimes they are the same thing, but not necessarily. Arbitrators should be acceptable to both sides in a conflict.
Harassment is a particularly difficult problem. There is sometimes genuine harassment, where someone persecutes someone else over a period when the victim has done no objective harm, or the persecution is disproportionate to the harm done, but that is not the only reason for a perception of harassment. Often people feel harassed when what is happening is objective criticism, sometimes even when it is politely stated, and some people may confuse a perception of harassment with hurt feelings. A persom who in good faith is producing unacceptable content, or who themselves are behaving disruptively, in a way that they do not understand is disruptive, may easily feel harassed when criticised for their actions. We cannot afford to counter-harass people who are providing justifiable objective criticism within the civility constraints. They are our defence against inappropriate and low quality content. We have no way of knowing the real motivations of an editor, and can only reasonably react based on what they actually do. Both the innocent and the guilty claim innocence. I see value in thightening up the civility/politeness constraints, though we must take into account the wide range of social and cultural backrounds our editors come from. There are still a lot of ad hominem arguments and attempts to coerce by passive-aggressive behaviour, and I am sure that at least some people are unaware that they are doing this. Others may be gaming the system by pushing the apparent limits of acceptable behaviour deliberately, and contentious behaviour can be entirely polite, yet remain disruptive.
Maybe a group of behavioural advisors could be put together for each project, who are sufficiently skilled to identify when someone is being disruptive, and can be called on for an opinion, without necessarily having authority or tools to take action. They could be called on to advise when someone feels that a comment is ad hominem, unreasonable, unjustified by policy, or is logically incorrect. Such feedback may educate unintentionally disruptive editors, who might otherwise get blocked for reasons they do not understand. Calling in a previously uninvolved third party may diffuse developing conflict. Cheers, · · · Peter (Southwood) (talk): 07:50, 19 August 2019 (UTC)
Nattes isn't talking about editors "producing unacceptable content", that's the simple case. The difficulties lie in conflicts between good editors - experienced, or not -. These tend to turn into ad hominem arguments / incivilities / bullying / system gaming / benefiting from a friendly admin. What's your suggestion to handle such cases? — Aron M (talk) 00:41, 20 August 2019 (UTC)
Education might help. Part of the problem is with people thinking that because something seems reasonable to them, that it must be permissible, when it actually transgresses the rules. Competence is required, not only in the topic one wishes to edit and the style of editing required, but also in fitting in with the community and managing newcomers fairly. · · · Peter (Southwood) (talk): 04:58, 20 August 2019 (UTC)

SupportEdit

I strongly support the development of a base set of conduct rules covering the entire Wikimedia ecosystem, along with standard effective and efficient enforcement processes. I view this as probably the single most important recommendation to the long-term health of our wikiverse.

For example, I believe it is appropriate to globally say that posting an address or phone number on-wiki of another editor or reader (and therefore, really, any living human) is unacceptable, without express on-wiki permission.

It is crucial that this code of conduct be developed with input from wiki editors, potential editors, readers, and WMF staff, and approved by the Board.

As the UN succeeded in devising its Universal Declaration of Human Rights, surely we can come to consensus on a minimal set of behavioral rules that apply to all participants in all our projects.

Libcub (talk) 05:08, 11 August 2019 (UTC)

Perhaps, but who is the "we" in "we can come to consensus"? In the UN, no member voted against the human rights proposal. Is that the standard that will be applied here (all projects will get a vote), or will it be another dictat from on high? If it's the latter, there is likely to be considerable opposition and major negative effects. EddieHugh (talk) 11:27, 11 August 2019 (UTC)
There may be some difficulty in clearly defining what is acceptable and what is not. It is not possible to reach a real consensus unless everyone in the discussion understands what the others mean. This will require some really good, clear choice of terminology and expert translation, and no woolly terminology that is later claimed to mean something else. The details really will be easier to sort out for each language separately, otherwise there is a very real risk of riding roughshod over the cultural diversity that currently exists and imposing the norms of the majority group. · · · Peter (Southwood) (talk): 20:46, 11 August 2019 (UTC)

Hello, I see a lot of positive things in these recommendations: limits of terms, secret voting on people, better training etc. What I would be not able to agree to would be quotas, e.g. that a certain percentage of admins or ArbCom members have to belong to certain subgroups. That would limit the active voting right. Ziko (talk) 14:09, 12 August 2019 (UTC)

Hi Ziko, Better training is fairly obviously a positive, but what do you consider positive about limits on terms and secret voting? · · · Peter (Southwood) (talk): 16:59, 16 August 2019 (UTC)
Hello Pbsouthwood, I don't feel that strong about limited terms, but it is normal to limit the time that is followed by a positive vote. If an admin is inactive, this is a good way to end the office. Secret voting: people should feel free to vote without being afraid of repercussions, but also: candidates experience the present day process as quite toxic, and I think that the open voting often "enriched" with comments about the candidate is contributing to the problem.
Training: I expect that this point can becoming quite controversial, as it would require a way to train the candidates which might not work totally without knowing the candidate's identity. So, for example, if someone wants to be an admin candidate, he/she would have to follow a videotelephone course of some kind. Ziko (talk) 21:06, 16 August 2019 (UTC)
Hi Ziko, I don't follow your statement limit the time that is followed by a positive vote. could you try to clarify? I appreciate that English is not your home language, so maybe you could try Dutch (which I can read a bit), or German, (where I largely rely on Google translate, but can often work out the meaning eventually).
About training: There are online training options which could be linked to the username and remain entirely anonymous. (we did specify "better" training) A well-designed system would allow a user to sign up for a module, be assessed and have the assessment result (including date) publicly linked to their username. If it is considered better, only successful assessments need be linked, so you could keep trying till you get it right without embarrasssment. Many of the oppose arguments relate to competence. These can be reduced if the candidate has demonstrated competence to objective criteria and this is recorded.
About secret voting. I am mainly with Nosebagbear (see below) on this one. Discussion is very important at RfA. Voting without discussion comes down to whether the voter likes the candidate, and has no clear bearing on whether other people should trust them. When the community is large and diverse, one does not know the people who work in areas that do not interest oneself. Comments by people you trust are important. Comments by people you don't know are less useful, because you don't know where they are coming from, but they can open up useful lines of enquiry so due diligence can be done. Anonymised votes and comments are dangerous as no-one knows who the voter is, and cannot judge whether they are just shy, or trolls, sockpuppets, meatpuppets, people with no obvious experience etc. There would be a danger of rigged elections, which is not easy with the current ENWP system, where enough eyes are on the participants that it is unlikely that inappropriate participants will not be called out. With a discussion system the closers can assess the validity of support and oppose arguments, and it is the oppose arguments that are most important as they show where there may be problems, so they should be thoroughly examined and discussed if there is any validity to them. · · · Peter (Southwood) (talk): 07:46, 17 August 2019 (UTC)
Hello, see above. It is a good thing if an admin term has to be renewed, but I hope the reccomendation does not mean that after a term you are not allowed to be an admin anymore (or for a certain period of time).
Secret voting: You can still discuss about candidates on the talk page. But the voting itself should be secret, so that you cannot see who voted for whom. Your account needs eligibility to vote, and We can detect double voting. --Ziko (talk) 07:54, 17 August 2019 (UTC)
OK, that is clear enough. I personally would not be strongly opposed to a limited term for admins, but there might be some people who would not consider re-requesting or re-election worth the hassle, and it is a hassle. we might lose more than we can afford. We might end up with an even more unbalanced collection of admins. I would strongly oppose any restriction to multiple terms, or even a gap requirement for admins completing a term in good standing. They are not policymakers, they are the maintenance staff. If they make policy it is with the same standing as other trusted editors.
I don't see the point in secret voting if there is non-secret discussion on the talk page. Disconnecting the discussion from support and opposition would make it more difficult to assess consensus.
I would be more inclined to support a recall system, to the extent that I am personally voluntarily comnitted to one. That too would (or should) be an in-community policy.
Some RfAs are more "toxic" (a word I hesitate to use, as it is so often misused) than others. This is often due to abuse of process because the rules of engagement are pretty vague, and we have a lot of wikilawyers. Also some candidates are more controversial than others, so more debate is likely.
I would also be happy to see more specific requirements for support and opposition statements at RfAs. For example, support and oppose statements should indicate why, clearly enough that other users can understand why the person is supporting or opposing. This helps consensus assessment, and avoids muck of the bickering. It would be a major break in tradition, but I think it is a thing that is broken and needs to be fixed. · · · Peter (Southwood) (talk): 08:52, 17 August 2019 (UTC)
  • I mean, it's not an issue if voters were anonymised, but things are discovered on RfA all the time, and they lead to questions. If it's like ARBCOM, with separated questions and votes, then quite a few voters will just not read the Q&A aspect. So I'm concerned about a reduced accuracy in the RfA. Nosebagbear (talk) 22:39, 16 August 2019 (UTC)

Be clear about the change in behaviour that you wantEdit

If you are going to keep someone in the community with a warning, or allow them to return after a fixed term block or unblock after an unblock appeal, you need to be clear to them what behaviour they should avoid in future. This is important for the previously sanctioned editor, for people who might learn from them, and for administrators and others who need to know in future whether this person merits a block for repeating what they were blocked for in the past or warned for doing some new sort of problem. I suggest we list "fixed term blocks for unclear reasons" as a form of toxic behaviour to be avoided. There are occasions when we need to permanently ban someone for reasons that can't be listed online for outing or other reasons, and frankly if we are permanently banning or blocking someone the process can work efficiently even if the banned person isn't sure why they were banned. But a fixed term block for reasons that are undisclosed to the blocked editor is setting a member of the community up to fail, and fail in ways that can be toxic to them and others who interact with them. After recent events on EN Wikipedia it would be helpful to have this listed movement wide as an example of toxic behaviour and poor leadership to be avoided. WereSpielChequers (talk) 10:56, 11 August 2019 (UTC)

Hear hear. MER-C (talk) 14:15, 11 August 2019 (UTC)
It seems the chickens are coming home to roost. The Fram incident is having repercussions here too. The up side, if there is one, is that there will probably be more intense scrutiny of any suggestions made here. The down side is that some bridges are now ash. Some care will be needed to avoid burning any more. · · · Peter (Southwood) (talk): 20:18, 11 August 2019 (UTC)
The WMF is very close to exhausting the patience of the community - after WP:FRAM and IP Editing: Privacy Enhancement and Abuse Mitigation, the latter I consider to be a stab in the back. There are editors from en.wp who feel this rubbish is not worth their time debunking and are content to just point and laugh/sigh (see w:WP:FRAM, w:User talk:Sitush#And). The WMF can start to repair community relations by explicitly junking the worst of the proposals put forward. MER-C (talk) 07:45, 12 August 2019 (UTC)
and vice versa. do you really want to use the phrase "stab in the back."? you could start by trying some collaboration. Slowking4 (talk) 19:20, 12 August 2019 (UTC)
I did word this as deprecating the specific behaviour, not the individuals responsible for the specific incident. I didn't check, and can't think of another example, but among our thousands of admins and the millions of blocks that have been issued I'm sure this isn't the first time such a mistake has been made. WereSpielChequers (talk) 14:31, 15 August 2019 (UTC)

Potential hazards to editors from repressive countriesEdit

If we are going to set demographic distribution requirements for users with advanced permissions, we will end up forcing those users to make public some amount of those demographics. And that could be a very bad thing. We obviously want to enable and welcome participation from people who come from repressive countries or cultures, and we want them to be able to participate here freely. Forcing users to divulge information that could lead to them being persecuted would accomplish the opposite of what we want, and preclude their participation. --Tryptofish (talk) 18:37, 12 August 2019 (UTC)

Furthermore, this contradicts the idea behind IP Editing: Privacy Enhancement and Abuse Mitigation directly. MER-C (talk) 18:59, 12 August 2019 (UTC)

Term limitsEdit

Given that there are no limits on the number of "functionaries" on any project (unless that limit is imposed by the community on that project), there is no genuine barrier to adding more functionaries, regardless of whether the role has set terms. That is, there is no set number of bureaucrats, checkusers, oversighters, unless that particular project has decided to set that limit. Why would term limits be something that would be globally imposed, rather than decentralized to the projects themselves? Almost all projects that have functionaries have locally established some sort of rules, which may include standing for periodic election, minimal activity standards, ensuring consistent performance standards, and so on; the rules are determined by the individual community and reflect the needs and expectations of those communities. What evidence is there that the lack of functionary term limits across all projects has led to specific issues? Are these issues confined to a specific project, or are they generalized issues? Risker (talk) 19:12, 12 August 2019 (UTC)

to the extent hoarding power is a systemic problem across communities, then a global solution is indicated. and we aspire to democratic values rather than aristocratic. given it is a perpetual proposal, perhaps training in use of tools in a harm minimization way would be better. Slowking4 (talk) 13:49, 20 August 2019 (UTC)
to the extent there is implicit bias in voting for leadership positions, term limits will not make room, rather there will have to be culture change. Slowking4 (talk) 08:47, 21 August 2019 (UTC)
@Slowking4: - Is English your first language? If not, you can feel free to use a language of your choice (since this is Meta). Winged Blades of Godric (talk) 19:12, 7 September 2019 (UTC)
"He who wishes to exert a useful influence must be careful to insult nothing. Let him not be troubled by what seems absurd, but concentrate his energies to the creation of what is good. He must not demolish, but build" ... --Slowking4 (talk) 21:32, 7 September 2019 (UTC)
It seems to me the Foundation is starting a culture change with these recommendations. Culture change takes time, and a lot of effort, however. — Aron M (talk) 09:08, 21 August 2019 (UTC)
I've been giving a lot of thought lately about this idea given some posts Risker has made on en around a dearth of leaders. Today I learned that there are concerns about the number of stewards as well. Aron Manning is suggesting this is because a culture change is underway. I don't understand how a culture shift is responsible for these shortages. It might be in response to a "generation" shift in leaders (e.g. the people who got us through the start-up and growth phases are fading away). But if so then we, as a movement, haven't done a good job of filling that leadership gap. And I think, a document like a strategic plan for the next 10 years, is a great way to identify this as an area for time thought and attention by the foundation. Best, Barkeep49 (talk) 19:42, 5 September 2019 (UTC)
hoarding power is a systemic problem across communities What is this hoarding of power that you write of, Slowking4, and is it truly a systemic problem? —Pelagic (talk) 10:58, 27 August 2019 (UTC)
In the case of ENWP there is no hoarding of power by any single person, there is growth of influence, as people are recognised to be expert or specialist in their particular policy interests and some groups develop a lot of influence. The way consensus is used to make policy means that a large influential group can drive policy to some extent, because consensus is usually nuanced and generally led by people with a lot of experience in the specific aspect in question. Policy is also generally descriptive, so a large group of people with a common goal can have a strong influence on some things, for example WikiProject Medicine can enforce the use of medical reliable sources at a far more rigorous level of evidence than normally required on the principle that readers will be using Wikipedia to make personal decisions which could affect their health, and that it would be irresponsible to allow bad medical information on Wikipedia. Sometimes this is good, sometimes not so good. There are camp followers who use the influence of these groups to overzealously apply their personal interpretations of policy, sometimes to the extent that they overstep the mark too often and end up indeffed. If Wikipedia had a large and influental group of people who wanted to push a particular point of view for politicqal or commercial purposes, some drift would occur. This can be done from inside a project, or if that fails, from outside, by, for example, attempting to remove the incumbent influential groups to make way for new influential groups with different agendas. A sudden large enough influece gap could lead to a shift in influence in unpredictable directions, with unpredictable consequences. · · · Peter (Southwood) (talk): 15:13, 27 August 2019 (UTC)
bingo [1] --Slowking4 (talk) 21:37, 7 September 2019 (UTC)
Me, I am thinking that a lot of these Influential People get their influence not from their user rights. They are informal power positions. So term limits would not help. Jo-Jo Eumerus (talk, contributions) 08:22, 8 September 2019 (UTC)

Strong supportEdit

The findings of the draft portray very well the experience of less vocal groups on English Wikipedia (I presume other projects as well).
From own and others' experience: issues like lack of civility and respect in communication, bullying, harassment, threats, combative behaviour happen every day.
The established community processes efficiently handle trivial issues (vandalism, blatant disruption), but non-trivial issues listed above, are either ignored, or result in fruitless drama usually.

  • When two editors - both a benefit to the project - have a conflict, seeking a cooperative solution with compromises on both sides would be the good-faith effort to make. This seldom happens, and combative "discussions" are more usual on the noticeboards. To resolve these non-trivial issues to the benefit of all parties, trained mediators are necessary., however this concept might sound alien to many editors initially.
  • The noticeboards cannot provide equal treatment to editors: occasionally even harsh behaviour is accepted, on other occasions minor disagreements are blown out of proportions. Everybody has an individual understanding of the policies, thus the outcome of disputes is strongly influenced by the number of supporters of an editor, and their interpretation of policies.
  • Preferential treatment is a recurrent issue, also exhibited in the severity of sanctions. Mistakes by new editors, who unknowingly break one of the infinite number of rules (policies, guidelines), are often punished with swift blocks, without first explaining the rules. They have no supporters to question these actions, and there is no place for complaints without getting indeffed and talk-page-access removed. These editors experience this as a harsh, exclusivist action, thus never return. We mostly hear of their experiences from off-wiki sources and the media.
On the other hand long-time editors' breaking of rules is often overlooked, and there are so-called "unblockables". We have seen in the fram drama, that despite multiple reports through years, such cases are ignored, even by the ArbCom. There are no other "community processes" on enwiki to resolve these issues. — Aron M (talk) 01:31, 13 August 2019 (UTC)
Generally it takes a lot to get talk page access removed. Yes new editors can end up blocked for legal threats or copyright violations or spamming fairly quickly. What the community is looking for is assurances that this will not happen again. If instead the "newbie" simple continues the legal threats against the person who blocked them yes talk page access will be removed. Doc James (talk · contribs · email) 03:14, 13 August 2019 (UTC)
actually, no it does not, all it takes is an admin who wants to shut someone up. and as you see below, more common than you might think. Slowking4 (talk) 13:51, 20 August 2019 (UTC)
Although TPA removal is generally hard, even taken to noticeboards, then rejected (AN, ANI), there are less-visible cases when it's revoked without reason, no legal threats involved. It's another "inconsistent application of rules". — Aron M (talk) 18:08, 13 August 2019 (UTC)
I would love some examples, please. Winged Blades of Godric (talk) 18:56, 13 August 2019 (UTC)
Yes, I agree with WBG and would also like some examples. If we have admins who are issuing routine blocks with TPA revoked absent abuse of the talk page, that's a problem that needs to be addressed. Seraphimblade (talk) 09:43, 14 August 2019 (UTC)
Indeed. However, I believe this is not the place to share and resolve specific issues, but to discuss the proposal, which has generalized findings such as "inconsistent application of rules". Inconsistent TPA handling is just an example for that. — Aron M (talk) 18:29, 14 August 2019 (UTC)
Aron Manning, since it was you who brought up the subject, it would be reasonable to expect you to be able to support your point with dertails. Otherwise it is not persuasive that the rules are actually being applied inconsistently to a significant extent, or that this is a problem that needs to be fixed. · · · Peter (Southwood) (talk): 11:51, 15 August 2019 (UTC)
Examples of revoked TPA without reason given: [main], [public places], [old private] account. For the curious: socking allegations are rebuted on my user page, with the appropriate policy referred, that allows this usage. Sadly those rules are ignored in this case, and the right to appeal was not given to a fairly new user (see TPA). This is just one example to confirm the findings of the research of "unclear rules", "inconsistent application of rules". — Aron M (talk) 06:39, 17 August 2019 (UTC)
What happened to your ArbCom appeal? Also, did T&S respond? Winged Blades of Godric (talk) 09:59, 17 August 2019 (UTC)
This is not the forum to discuss this, but if you ask (further discussion on user TP, please, comments will be moved there):
The appeal was redirected to TP, where I was accused of abusive use of my unused alt account, then TPA was revoked, then the UTRS appeal was redirected to my alt account's TP, subsequently that TPA was revoked too, then UTRS access was revoked, all while ArbCom failed to reply since 30 June.
"Once incoming mail has cleared list moderation, each message shall be acknowledged with a standard message and processed by the coordinating arbitrator or their deputy within 24 hours of receipt".
Procedural error?
You need to explain how you know about my ArbCom appeal, T&S report, and my email list subscriptions, you mentioned above. Neither of this is public information. — Aron M (talk) 18:08, 17 August 2019 (UTC)

┌─────────────────────────────────┘
Oh, please... you guys know the problem. I've seen it happen a lot. It has happened to me, twice. Once you're blocked you're only allowed to appeal, so if you try to say the block is unjustified (as it often is if you run up against one of the House edit-warriors or one of their tolerated sockpuppets) your talk page will be zipped up in a heartbeat. For many more examples do some research on off-wiki fora. You know where they are WiBloG (you are a member of one), Doc, Seraph, ... I don't want to do all your digging for you... suffice it to say, it's a common occurrence for new members -- flabbergasted that they've just been railroaded out because they had inappropriate ideas and/or didn't know all the rules. Be aware, too, when people have their reputation unfairly smeared they're very likely to think about saying "that's libelous" (because it is), without knowing that they can lose TP access just for saying that (though in point of fact, policy says that complaining of libel is not a legal threat). For a very simple example, please explain the so-called super-mario problem without showing you understand the double standard that new users face... how many new users could tell, say, Bishonen, to buzz off (but with one of the stronger words the cool admins love to use in their RfAs) and edit the next day? SashiRolls (talk) 23:56, 16 August 2019 (UTC)

I don't know whatever club-membership you are alleging of me but I claim to not have any connection to anybody over any external site, who claims to have some kind of connection to me. (Impersonation, might-be but I am not bothered, anyways.) We are discussing this over a Wikimedia platform and certainly, don't see any reason to do research on off-wiki fora.
This was a damn good revocation of TPA despite your feelings of not deserving a mile-long block log. People with some kind of inappropriate ideas certainly needs to be railroaded out.
I have told some admin to fuck off, in my earliest days and yet maintain a pristine block log. Unless and until, you manage to exhibit extreme dumbassery coupled with strong doses of deception and lack of self-awareness, you manage to hold on to your TPA over en.Winged Blades of Godric (talk) 09:59, 17 August 2019 (UTC)
As a public service, I have popped up the comments of your "WBG"/"maybe WBG" impersonator in the Governance thread over at Wikipediocracy so that you can complain to the admins there. However, I agree with Aron that here we need to remain focused on the problem of dealing with problematic insiders monopolising governance discussions / RFA discussions, etc.SashiRolls (talk) 21:39, 19 August 2019 (UTC)
1) I have to disprove this opinion. I was always polite and respectful, never exhibiting the uncivilities, that you admit, yet my TPA was revoked after a complaint, and for no reason on my 3 accounts. The "inconsistent application of rules" is apparent.
2) Please consider, that this is a public forum to discuss the proposals, not your opinion about Sashi. If you need to do that, the user talk page would be the appropriate place to post these walls of text. — Aron M (talk) 18:16, 17 August 2019 (UTC)

SuggestionsEdit

  • The ineffectiveness of current dispute resolution processes is also a result of the high stakes: blocking might mean the end of one's wikipedia presence.
The blocking policy states "Blocks are used to prevent damage or disruption to Wikipedia, not to punish users". Such disruption can be stopped with focused (article/category/topic/interaction) blocks for one day, that escalate to 3 days/1 week/4 weeks if the issue is recurrent. Category blocks have been already proposed as a development, to my best knowledge.
Universal (project-wide) and indef blocks should only be used when an account's sole purpose is disruption, or accumulated multiple focused blocks. With further development, the mediawiki software can ensure blocks are applied incrementally, and only to the articles/categories the editor participated in.
  • Reporting misconduct is difficult, and has the risk of backfiring (ex. being treated as personal attack). For this reason, many issues are not reported. A "safe" space needs to be introduced for reporting, where the reporter's behaviour only influences the outcome of the report, and there's no threat of punishment. This space can be the planned user reporting system.
  • Reports should focus on facts. A tool can be developed that helps editors collect the edits for evidence, by selecting the relevant text on pages, then selecting the relevant policy or guideline, and entering a subjective evaluation of the edit, then repeating it for other edits.
  • This tool can be used by administrators too, to generate the notice for sanctioned editors, thus making it clear to the editor what caused the sanction, while guaranteeing the transparency of such actions. — Aron M (talk) 01:31, 13 August 2019 (UTC)
Aron Manning, I agree in principle with much of what you have expressed above, but would like some clarification if possible.
  1. Is there any reliable evidence that the two processes specified in the first bullet point actually achieve the claimed effect, that it is likely to be effective in other communities, and will scale well?
  2. Is there a more explicit specification for the tool mentioned for collecting evidence?
· · · Peter (Southwood) (talk): 16:04, 13 August 2019 (UTC)
2. This is a suggestion, not a specification, for the implementation of the user reporting system, which is in planning phase to my best knowledge. Do you have any specific questions regarding the details? We can discuss it in a dedicated thread.
1. These processes work reliably on German Wikipedia (stats) since 2004. There's a good English language introduction in this comment on WP:Fram, and DeWiki editors should be asked for their experience. I'll ping the commenter. — Aron M (talk) 18:34, 14 August 2019 (UTC)
2. I would want to see evidence that such software is reasonably practicable before pinning any hope to it.
1. Is there a study showing that the processes work reliably on German Wikipedia? The links you provided do not appear to be helpful.· · · Peter (Southwood) (talk): 11:58, 15 August 2019 (UTC)
1. As I said, you should ask German WP editors.
2. How would be there such evidence in the lack of the software, or even a mature vision of it? To fill a small part of the gap I detailed and drafted my suggestion at the User reporting system consultation. I'm looking forward to seeing your suggestions. — Aron M (talk) 21:56, 15 August 2019 (UTC)
I misunderstood. I thought you were referring to software for which a feasibuility study had been done. It seems that it is hypothetical, which is a bit of a drawback. In principle it could be useful, and I would support further study and development of the tool. · · · Peter (Southwood) (talk): 18:19, 16 August 2019 (UTC)

German Wikipedia questionsEdit

Hello @PerfektesChaos: would you, and others from DeWiki like to comment on the usefulness of these 2 processes? Thank you in advance. de:WP:Administratoren/Probleme, and de:WP:Adminwiederwahl/Intro (translated: admin re-election) — Aron M (talk) 18:34, 14 August 2019 (UTC)

I was invited to comment on two of the processes at de:wp, i.e. WP:AP (conflicts with administrators using their tools) and WP:AWW (recall of administrators).

  • WP:AP is the older process, initiated on 18 March 2004 by Eloquence. The principal idea was to open a process for complaints against administrators who violated policy when using administrative actions. The complainant can open a case, describe his or her view of the conflict, and invite the admin to comment. Afterwards we have an open discussion (much like an RFC) which is closed by another admin. If there is indeed a consensus that policy was violated, there are three possible consequences, i.e. to warn the administrator, to temporary de-sysop him or her, or a regular de-sysop. This process more or less remained unchanged since 2004 with the exception of a minor adaptation that such a case must not be closed before 24 hours have passed since opening it. This process is used frequently (see the archives) but many see it as unsatisfactory as sanctions against admins are extremely rare. In fact, we have right now a policy proposal which intends to change this process such that not admins close these cases but members of an independent elected group of non-admins. The proposal provides some statistics for all the cases from 2017 up to 15 June 2019 according to which 56 cases were closed where no violation was seen, 11 cases were closed with a warning but no sanction, and three cases were retracted. In none of the cases we had a de-sysop, be it temporary or not.
  • WP:AWW (recall of admins) is a process which was introduced in 2009. For every admin we have a separate page (see WP:AWW for an index, this one is mine) where every eligible user can vote for a recall. Votes expire automatically after half a year. If there are 50 active votes for a recall or if 25 votes for recall were submitted within the last 30 days, an admin has to rerun for adminship within 30 days or will be de-sysoped. Freshly elected admins are excempt for a year from this process. Likewise admins with extra privileges ('crats, oversighters, checkusers) as they have to rerun every two years. Statistics of this process are available at de:Wikipedia:Adminwiederwahl/Kurzstatistik. Since this process was introduced, 227 admins were recalled, of these 94 did not rerun, 55 lost the new candidacy, and 78 were re-elected successfully. As sanctions at WP:AP or sanctions of admins through the arbcom are extremely rare, this is the main process for de-sysops at de:wp.

--AFBorchert (talk) 05:52, 16 August 2019 (UTC)

I will respond in detail in September.
Basically both instruments are valuable to limit power misuse when elected people are networking over a decade, but both instruments themselves may be misused by some pressure groups against the purpose of the instruments.
While both tools are important in case of real power misuse, their daily usage is more like a revenge of those who have been blocked for good reason etc. Unfortunately it became a regular sequence that someone violated project rules, being blocked a few days, and vice versa opening an Admin Problem. Naturally, the outcome is rather poor. Claiming to be sentenced but innocent does not wash all sins away.
The descriptions of the procedures by AFBorchert are correct.
Geetings --PerfektesChaos (talk) 19:05, 19 August 2019 (UTC)
This is already useful information, but nothing surprising. I would welcome any further details.· · · Peter (Southwood) (talk): 16:33, 21 August 2019 (UTC)
I don't speak German, and this is merely hearsay, but I hear that the admin-recall process results in controversial topics becoming open warfare zones. Admins become unwilling to deal with ugly situations. As experienced editors generally know, a good and impartial admin stepping in to clean up a major conflict is likely to make enemies from both sides of the dispute. Those enemies pile up and threaten to torpedo any admin dedicated enough (or stupid enough) to deal with such problems.
There are pros and cons to both sides of the admin-recall issue. Alsee (talk) 22:34, 11 September 2019 (UTC)
I also vaguely recall that it was as an ineffective payback during the Superprotect affair a few years back, whereby the volunteer account of a WMF staffer who had flipped the switch was deadmined. Jo-Jo Eumerus (talk, contributions) 07:18, 12 September 2019 (UTC)

Model Code of Conduct rather than Universal Code of ConductEdit

As I previously noted under the Diversity Working Group's discussion of the same recommendation: A Code of Conduct will only be effective if it has "buy in" from the community that it governs. Also, behavioral expectation are different on different communities. The same situation occurs in the United States where each state adopts its own laws. Because states want to develop best practices and be consistent when practical, a system of "Model Laws" have developed. A set of legal scholars draft Model Laws and then recommend them to all of the state legislatures for adoption. The Wikimedia movement could do the same thing with a Code of Conduct. A process could draft a Model Code of Conduct that would be sent to all of the projects to consider, amend and adopt. This has the advantage of allowing local adaptations to reflect unique community expectations as well as to engage a much wider section of each community than participates in this strategic consultation. Many members of the community have already said that they would view a "top down" imposition of a universal code of conduct to be an unwarranted intrusion on the internal management of their communities. A Model Code avoid this problem. Hlevy2 (talk) 13:57, 12 August 2019 (UTC)

This would be a better way to go. To minimise the amount of bickering over the details, which can drive a discussion quite rapidly to no consensus, it would be useful if each recommendation was accompnied by a reasoned explanation of the problem it was intended to solve, and why it was decided that the specific item would be likely to solve that problem and not cause other problems, or at least why it should be better than the status quo. Claims of fact should be well referenced fron reliable sources. Cases where the policy has worked in practice should be listed where available. It is far easier to get buy-in for a proposal which cannot be shot down in flames on sight, and it will be scrutinised by many eyes looking for problems. Consider how the proposals will help further the goals of the project, and how they may increase the burden on volunteers. Avoid the latter, it is a deal breaker. Cheers, · · · Peter (Southwood) (talk): 14:29, 13 August 2019 (UTC)
Per-project (non-universal) CoC would have serious problems:
  • It needs to be read and accepted individually on each project, that a user visits. This would be irritating for the user and a nightmare for the developers.
  • The individual CoCs would not differ much. Unlike our policies, a CoC is focused, compact, and general. There's very little to win with the differences, but it comes at a high price.
  • The individual projects could possibly customize the CoC in a way that ignores specific issues, that's difficult to face for the community.
— Aron M (talk) 04:52, 15 August 2019 (UTC)
The user would accept the Term of Service which would incorporate the Code of Conduct as adopted by each project from time to time. So there would be no need to display the CoC as the user visits each project.Hlevy2 (talk) 18:47, 4 September 2019 (UTC)
A universal CoC is likely to have even worse problems if it is a mismatch to any community.
A universal CoC would have to be acceptable to all communities. That would make it a very drawn-out process to develop and approve, and may not be possible. Top down or outside imposition is likely to be disastrous. This combination is impractical. · · · Peter (Southwood) (talk): 10:38, 15 August 2019 (UTC)
Not everybody imagines a disaster. Introducing CoC is a hurdle in every community. We deal with it. — Aron M (talk) 01:46, 16 August 2019 (UTC)
Failing to consider management of consequences for a known significant risk is not just irresponsible, it is negligence and failure of due diligence. · · · Peter (Southwood) (talk): 07:48, 16 August 2019 (UTC)
 ? If this was an answer, what do you mean by it? — Aron M (talk) 08:23, 16 August 2019 (UTC)
An answer is generally a response to a question. My previous statement was not a response to a question, just a comment on the preceding commentary, so technically not intended as an answer. It is mainly a response to the comment that "Not everybody imagines a disaster", which may be true, but is not an adequate rebuttal of the perception by some that disastrous consequences may occur. I will expand on the meaning I had hoped to communicate, as an answer to the preceding question (what do I mean by it):
  • Several commentators have indicated that they percieve significant risks of disruption to project communities as a probable consequence of following this recommendation. Even if they are not technically a disaster, they may be serious setbacks.
  • The proposers of the recommendation, and the groups that they represent, have an ethical responsibility to avoid harm to the project communities.
  • To demonstrate that they have carried out their responsibilities with due diligence, they need to show that they have taken these predictions into account, have considered the risk, and have taken reasonably practicable steps to mitigate the predicted consequences. There are several possible routes.
    • The predicted consequences can be shown to be acceptable, either absolutely, or as the best option available.
    • The probability of the predicted damaging consequences occurring can be shown to be sufficiently remote.
    • Procedures can be set up before implementation to deal satisfactorily with the consequences if they do occur.
At this point I see no evidence that any of these things have been attempted. · · · Peter (Southwood) (talk): 16:26, 16 August 2019 (UTC)
  • Strongly agree that a model code of conduct would be much better, for the reasons stated above. You can however split the model into different levels of importance along the lines of "this is recommended to include in your model" to "this is more or less mandatory, if you do not include this in your model, please explain your reasoning". And it would be good to somehow "force" communities to have the discussion to come to a CoC for their project, rather than just make it a thing which might come up. So make having a CoC mandatory. --Jan-Bart (talk) 06:01, 16 August 2019 (UTC)
    • I agree that making a CoC mandatory would be reasonable per Jan-Bart's rationale above. Not trying to make the same CoC mandatory for all projects. · · · Peter (Southwood) (talk): 07:48, 16 August 2019 (UTC)
Conduct on a small project is much different than on a large project. For example, if only two or three editors visit each day, doing Recent Change Patrol should never be considered "wikihounding." If anything, WP:OWN violations are a more serious matter than WP:HARASSMENT on small projects. We need to provide incentives for collaboration, not threaten users with a T&S investigation if they try to interact with the other editors. Hlevy2 (talk) 18:47, 4 September 2019 (UTC)
  • @Aron Manning: - you say there's "very little to win with the differences", that seems a tough argument to make without a specific draft CoC to be discussing. More importantly "The individual projects could possibly customize the CoC in a way that ignores specific issues, that's difficult to face for the community" - what about individual communities finding it difficult to face because the WMF (far more than the other local communities) have decided something that ignores a specific state of affairs on a local community. That ignore could mean either "fails to resolve an issue" or "causes an issue" Nosebagbear (talk) 14:13, 15 August 2019 (UTC)
  • Besides that I find Jan-Bart argument very good, somebody please tell, what exact differences do you anticipate between individual projects' CoCs? Specifically, what civility rules do you think won't be applicable to one project, but valid for another, eg. enwiki vs. enwikt, or commons? — Aron M (talk) 08:30, 16 August 2019 (UTC)
@Aron Manning: So that's tricky to lay out examples without someone actually giving us a draft CoC to review. Instead I have to work with what I don't think would be wanted for en-wiki but is mooted in the various suggestions seen in the drafts elsewhere and on working group discussions etc (I assume they must be wanted in some project(s) or they'd be terrible ideas). But the rigidity of what falls under it (some might say that saying "Fuck this" to edit summaries would be against it, but while I supported stopping that, the en-wiki community was firmly against a rigid prohibition - but others might be in favour). A couple of drafts want to prohibit any satire or joke based on a difference of political views - no exceptions for when you're joking on the userpage with an editor who is perfectly game for it. The jurisdictional issue is a big one - who is going to resolve complaints, we've seen that en-wiki and de-wiki are ardently against the WMF resolving any complaints outside their traditional 4 areas (Harm, self-harm, legal and CSE). Even within en-wiki at this moment there's a firm discussion on how much fairness and transparency can be traded away to handle private complaints. I can well imagine a disagreement on where that line should be drawn. Nosebagbear (talk) 08:41, 16 August 2019 (UTC)
There's a lot going on in this answer, many of it worth a discussion, that you could start new threads with. There's no example for the question above, tho.
For what to expect from a CoC, take a look at a common template CoC used in many free software projects: https://www.contributor-covenant.org/version/1/4/code-of-conduct
I expect something along those lines. It's pretty general. Imho a CoC is more of a mission statement, than exactly outlined rules, with minute details, like our policies. As it's so general, there's not much that can differ between projects. On the other hand each project has its own "sensitivities": how incidents are weighed. Those differences are untouched by a CoC, thus each project can keep its own nuances.
I guess a CoC is less threatening, if one has already experienced it. — Aron M (talk) 09:01, 16 August 2019 (UTC)
While there are a couple of unique points, probably not worth creating a thread until my "nope" thread gets some discussion. I do suspect that the trust issue I laid in that thread is particularly important here - the burning of bridges makes it hard to believe that it would never (whether now or in 3 years) be used in an undesired way. And a couple of those facets are distinctly not minute details - jurisdictional decisions in particular. The other sense is that if it isn't such a major change, than the positives stated in the rationales would seem likely to be muted at best. Nosebagbear (talk) 09:17, 16 August 2019 (UTC)

Question for writersEdit

In regard to: "Prevent accumulation of power by introducing term limits, a maximum to the number of offices one person can hold..." What is your definition of "offices" in this sentence? Vermont (talk) 13:36, 13 August 2019 (UTC)

I too would like an answer to this reasonable and very relevant question. · · · Peter (Southwood) (talk): 10:41, 15 August 2019 (UTC)

Clarification neededEdit

Establish new structures which allow equitable distribution of power in all structural levels of communities and invest in developing trust in these structures within the communities

  1. How is it proposed to establish these structures? This would have a strong critical influence on developing trust in these structures
  2. What "power" is referred to here?
  3. What distribution of such "power" would be considered equitable? · · · Peter (Southwood) (talk): 14:47, 13 August 2019 (UTC)
Would distribution be measured by individual or households? For example, if the President of Wikimedia Australia marries the President of Wikimedia Indonesia, should the rule require one of them to step down? What about siblings? Is disclosure during the election process sufficient or should there be a flat prohibition? In many projects, members of the ArbCom (or leadership committee) are also Admins, would the rule cover that?

More clarification neededEdit

Empower communities for responsible and inclusive self governance, so they can grow and adapt to meet the future challenges posed by the Strategy 2030,

  • This sounds nice, but what does it mean? · · · Peter (Southwood) (talk): 14:51, 13 August 2019 (UTC)
  • How does this tie in with imposing a movement-wide single Code of Conduct? · · · Peter (Southwood) (talk): 16:32, 16 August 2019 (UTC)

Yet more clarity neededEdit

Facilitate and support community leadership by capacity building and support for (potential) leaders, and introduce (mandatory) training for certain community roles.

  • Looks nice in principle, but a little vague.
  1. Which community roles?
  2. Who will probide the training?
  3. Who will set the training standards and assessment criteria?
  4. Who will assess the competence of candidates?
  5. How will assessment be done? Will records be open?
  6. How will this process be made transparent and apopropriate to the specific community?

· · · Peter (Southwood) (talk): 15:01, 13 August 2019 (UTC)

Questionable claimEdit

By having a balanced distribution of power within communities, we can achieve knowledge equity.

  • This is something of a sweeping statement. I would like to know what evidence supports this claim. Maybe I don't understand what is meant here, in which case clarification would help. For example,
  1. What is a balanced distribution of power in this context and how would it be measured to demonstrate whether or not it exists?
  2. How is knowledge equity defined in this context (please link to definition), and how is it to be measured to show that it has been achieved?

· · · Peter (Southwood) (talk): 15:12, 13 August 2019 (UTC)

In some countries, there is a lower literacy rate than in highly industrialized countries. "Knowledge equity" could mean providing access to knowledge to both literate and illiterate people. Is balancing the distribution of decision making roles between literate and illiterate people a necessary precondition to attempting to provide access to knowledge to iliterate people? Hlevy2 (talk) 20:07, 13 August 2019 (UTC)

ConflictEdit

While I recognise that the level and type of conflict is often excessive, conflict at the dialectic level is an integral part of our quality assurance process. We come to consensus on difficult topics as a result of the discussions and arguments that arise out of conflict. If everyone was on the same side of the fence, a seriously unbalanced point of view could emerge.(Wikitruth Through Wikiorder David A. Hoffman and Salil Mehra Emory Law Journal, Vol. 59, 2010, Temple University Legal Studies Research Paper No. 2009-17.)
We cannot - and should not even try to - eliminate all conflict. What we do need to reduce is ad hominem conflict which is counterproductive, and is used by many as the weapon of choice, often as a means of baiting the opposition into a flagrant incivility, in the hope of getting them blocked and "winning" by default. · · · Peter (Southwood) (talk): 15:30, 13 August 2019 (UTC)

"conflict at the dialectic level is an integral part of our quality assurance process." except conflict on wiki is typically battleground and ad hominem. the intellectual dialectical critique is rare, and when you ask people to engage on talk, they typically continue to delete and edit war, and block to enforce their edits.
it is not integral, there are other methods of producing consensus, by incorporating mediators in a process. and quality is typically improved by meetups, and individual effort. the group breaking news effort is a small subset. this community has chosen to privilege battleground.Slowking4 (talk) 14:04, 20 August 2019 (UTC)
Yes, the battlefield and ad hominem part should go, but we should keep the part that works, even if it only works part of the time in the current environment. It is a useful tool when used appropriately. · · · Peter (Southwood) (talk): 16:14, 20 August 2019 (UTC)
I would call debate that conflict, which is beneficial, and an integral part of creating the encyclopedia. It's important to recognize, when debate escalates to disruptive conflict (battleground and ad hominem). Although the barrier between the two is subjective, we have to draw a line, when the ad hominem arguments over-weigh the constructive arguments. This happens more often than not, and contrary to what is beneficial to encyclopedic content, often the ad hominem arguments win by force and threat of sanctions. This dynamic is closely related to bullying and/or harassment, that we would like to be less common. — Aron M (talk) 08:21, 21 August 2019 (UTC)
See dialectic, but debate is close enough most of the time.
As I said, the battlefield and ad hominem should go. Preferably completely, and that will require people to be better at recognising them when they occur. I think part of the problem is that a lot of people do not recognise ad hominen amd passive aggressive behaviour when they are using it. The problem with calling them out when they occur, is that is usually percieved to be ad hominem and battlefield behaviour in itself. Maybe a system by which a neutral third party who is skilled in recognising logical fallacy and other undesirable tactics could be pinged to comment on a discussion, and point out all cases of undesirable behaviour in the discussion, without prejudice. As sort of behavioural user check. All participants in the discussion would get checked, including the one who requested the check. (Just brainstorming, I don't know how practicable this could be, it could be a very busy and rather unpopular post, so there should be several available, a sort of "Arbcom lite".). · · · Peter (Southwood) (talk): 15:44, 21 August 2019 (UTC)

Consensus buildingEdit

It would be very interesting to see what the authors consider an appropriate process of consensus building, how thay propose assessing consensus, what thay consider illegitimate process, and how they propose to deal with it. If this cannot be articulated effectively, there would be little point in advocating for it. · · · Peter (Southwood) (talk): 15:48, 13 August 2019 (UTC)

Another request for clarificationEdit

Actively encouraging and supporting individuals from previously underrepresented groups to move to positions of leadership

  • What positions of leadership would these be, and how would the individuals be recognised as leaders? I comment from the point of view of Wikipedian on ENWP, where to the best of my understanding, leaders are recognised in specific contexts by virtue of them being seen to be competent in that field, and by them standing up for our principles and policies. We do not consider our functionaries to be leaders because of the permissions they hold, which are theoretically awarded because they need them and are trusted to use them appropriately. We do not generally take into account what groups these people may belong to, and I would not even ask - it is a private matter. Arbcom members are elected because people think they are likely to judge behaviour impartially, have a reasonable history of doing so, and are unlikely to overstep their authority, which is very limited. This may differ on other projects, so some clarification is necessary. · · · Peter (Southwood) (talk): 16:28, 13 August 2019 (UTC)
  • I agree. The best way to avoid discrimination is to avoid asking or telling about your self-identified characteristics. It is simply not relevant, and there is no guarantee that the self-identification accurately reflects the real world characteristics. Because the movement generally does not know the characteristics of its sign-on users, it has never discriminated against them, and there is no past history of discrimination to remedy. The Working Group needs to explain this more fully and tie it to your desired goal of Knowledge Equity. Thanks, Hlevy2 (talk) 20:41, 13 August 2019 (UTC)

Election processes versus consensus selectionEdit

... and ensuring that all community election processes are designed to encourage both candidacies and voting (e.g secret ballots when voting on people)

  • Is this an intention to impose secret ballot elections to replace consensus targeted discussion systems where these are the norm?
  • If so, what majority will be considered necessary to carry a vote? Will there be a minimum voter turnout to legitimise a result? How will voters be encouraged to engage when they have had no personal interaction with the candidate? · · · Peter (Southwood) (talk): 17:18, 13 August 2019 (UTC)
  • As someone who became an admin on my second attempt, can I point out the advantages of a structured discussion over a secret ballot? When people oppose and give their reasons it is possible in that discussion to clear up misconceptions or respond to mistakes being pointed out. It is also possible to come back a few months later, look at the arguments for oppose and run when enough of them have been resolved. If my first request for adminship had been rejected by secret ballot I'm not sure I'd have run again, how could I? Elections work when you are electing a fixed number of people such as an arbitration committee, they don't work when you are deciding whether an individual meets a standard. WereSpielChequers (talk) 14:54, 15 August 2019 (UTC)
  • Another question, do we have evidence that using secret ballots does actually encourage candidacies and voting? Jo-Jo Eumerus (talk, contributions) 13:28, 3 September 2019 (UTC)
    [Placeholder]. Winged Blades of Godric (talk) 13:54, 3 September 2019 (UTC)

Sanctions Against CommunitiesEdit

The recommendation states, "Sanctions can apply to individuals causing justified concerns, as well as to communities that consistently fail to act against such individuals." I understand the concept of individual action, accountability and responsibility, but more thought should be given to the concept of collective responsibility and sanctions. For example, suppose English Wikipedia adopts a standard of using pronouns in based on long term historic usage. However, a group of transgendered individuals become very vocal in demanding a different pronoun policy, claiming that they "feel endangered" by the requirement of traditional pronoun choices even in articles discussing non-transgendered individuals. If the English Wikipedia decides not to bend to their demands, can it be "sanctioned"? Conversely, if the English Wikipedia decides not to take action against the more emphatic advocacy of the group, could supporters of traditional pronoun choices escalate the matter and demand "sanctions" against a user group that represents the interests of transgendered people? The notion of group or corporate culpability is very tricky and complex, and the Working Group should clarify what it has in mind. Thank you. Hlevy2 (talk) 20:19, 13 August 2019 (UTC)

Guilt by association? Dangerous ground here. Who gets to decide which concerns are justified? Who decides which communities are failing to act? Is the community to be sanctioned for failing to have rules which forbid certain behaviour, or are members tasked with behavioural enforcement to be sanctioned for following the rules that they are appointed to enforce because they do not include rules that someone else thinks should exist? · · · Peter (Southwood) (talk): 06:14, 14 August 2019 (UTC)
What kind of sanctions could be applied against communities of volunteers that could produce a positive result? · · · Peter (Southwood) (talk): 16:43, 16 August 2019 (UTC)
I wonder whether their willingness for community sanctions are because we are a "Quasi-Anarchy" Nosebagbear (talk) 19:55, 18 August 2019 (UTC)
Possibly because there is no other easy way to put pressure on a herd of cats diverse group of volunteers with extremely varied levels of commitment and attention. I guess that comes to much the same thing. "Caedite eos. Novit enim Dominus qui sunt eius". · · · Peter (Southwood) (talk): 04:40, 20 August 2019 (UTC)
i like the concept of collective responsibility. we all share responsibility in the toxic community english has become, and the actions we take and support will be our epitaph. Slowking4 (talk) 14:11, 20 August 2019 (UTC)
What is easy is irrelevant - difficile iudiciis malum iura faciunt or perhaps, if it isn't too gauche to give a damn about fairness in this day and age - Fiat justitia ruat caelum Nosebagbear (talk) 14:23, 20 August 2019 (UTC)
I would go for the fairness and justice but without having the sky fall on our heads, and since I still edit there, would prefer to delay the need for an epitaph. The "toxic community" is also partly the responsibility of some of those who are no longer there. · · · Peter (Southwood) (talk): 15:39, 20 August 2019 (UTC)

Partial support to the global Code of ConductEdit

I think a global CoC would be a good idea for global spaces: international mailing lists, meta-wiki, maybe IRC channels and Telegram groups too? But when it comes to specific communities, this CoC should not be a rule but a guideline or an example so that the local communities can develop and enforce their own behavioural rules.

This shouldn't be necessary but I'll add that I only would agree with a "partially global CoC" as long as it has a wide consensus in the community. Which would be tricky, but possible. --Unapersona (talk) 10:37, 14 August 2019 (UTC)

  • This is a reasonable suggestion. I can support in principle. · · · Peter (Southwood) (talk): 12:32, 14 August 2019 (UTC)
  • If there are items in a CoC that prove particularly controversial, perhaps there can be a core policy for all with optional add-ons that each community agrees to? There are examples in draft (also posted here). I'm strongly in favour of a 3-part structure of 1) positive behaviours to strive for, 1) red-line behaviours that are unacceptable, and 3) a clear process for handling issues (in public or private as appropriate). That's also the structure that was aimed for in this CoC draft which includes a couple of items that are specifically only relevant to that community (e.g. maintaining financial records is relevant to Chapters and Thematic organisations, but not enwiki as a whole, whereas release of confidential information also applies to enwiki arbcom). T.Shafee(Evo﹠Evo)talk 01:49, 15 August 2019 (UTC)

Side note on representationEdit

Another thing that might be worth considering when including representatives in bodies such as the one that would oversee a CoC, or even for the WMF trustees board:

  • A representative of the reader/user community (i.e. specifically not an editor)
  • A representative of new editors (e.g. themselves not more than 1 year as an editor, or involved in outreach/editathons/WikiEdu)
  • A representative of people who are not yet a contributor but might later be (e.g. other open knowledge communities to bring an external perspecitve)

T.Shafee(Evo﹠Evo)talk 01:39, 15 August 2019 (UTC)

How would one ensure that representatives of these groups have any useful understanding of the communities whose CoC they would be overseeing (or is that not considered necessary)? How would they be selected (other than being a member of the relevant group, obviously) · · · Peter (Southwood) (talk): 10:30, 15 August 2019 (UTC)
Also, I can't imagine there's that many readers who could be organised well enough to stand for an election (elections being pushed elsewhere so appointment seems unlikely), answer questions etc, and still not make any edits. Nosebagbear (talk) 08:45, 16 August 2019 (UTC)
There can certainly be focus groups of new editors and/or users who do not edit, but there is no reason to entrust such people with the political and moral leadership of the organization. For one thing, such individuals have less of an investment in the success of the project. or may have an undisclosed COI. Why trust a stranger over someone with a demonstrated commitment to the movement and a know set of skills? Hlevy2 (talk) 18:59, 4 September 2019 (UTC)

Feasibility issueEdit

Before any realistic decision could be made about a universal Code of Conduct, a draft should be composed to demonstrate feasibility. If something close enough to workable is presented, we can see if it looks reasonable to continue the effort. Until there is a generally useful example to study, there is no strong reason to assume that such a thing that would be acceptable to the projects is possible. · · · Peter (Southwood) (talk): 12:13, 15 August 2019 (UTC)

Office LimitsEdit

It suggests "a maximum to the number of offices one person can hold," - but offices have a degree of needing to cluster in order to enable individuals to carry out their roles. Look at en-wiki Arbitrators - they are functionally always admins, and would need to be in order to implement blocks and see deleted content. They get Checkuser rights to be able to investigate and Oversight in order to be able to see oversighted information in sensitive cases.

Those with the latter rights are often OTRS agents (whether an Arb or not), because they already have access to some queues for their specific roles, and it's helpful to have full access.

Putting aside the benefits of having multiple rights to carry out certain capabilities - there are relatively few editors who the community has very high trust in. Senior functionaries can only be drawn from this pool - limiting roles can mean insufficient coverage Nosebagbear (talk) 14:22, 15 August 2019 (UTC)

Notwithstanding the theme of your comment (which I agree with); non-admin oversighter(s) can delete content, view any deleted content and un-delete them:-) Winged Blades of Godric (talk) 14:57, 16 August 2019 (UTC)
Besides, one can simply remedy the issue by creating an Arbitrator usergroup that holds all the necessary permissions if need be. Yes, I know that CU and Oversight are perhaps too sensitive to be added to such a right. If memory serves, dewiki Checkusers by purpose also generally only apply the "read" parts of the admin toolset, so adding these rights to the CheckUser usergroup would help there too. Jo-Jo Eumerus (talk, contributions) 16:15, 16 August 2019 (UTC)
Agree and creating such an usergroup with the relevant rights, is a matter of seconds.
I don't get the sensitive aspects. Stewards flip the CU & OS bits for all newly elected arbitrators (unless someone does actively reject them) near-automatically upon a request by ArbCom; if our hypothetical arbitrator usergroup were to be created, Stewards would just flip it instead. Obviously, Stewards will be the sole people with the ability to flip Arbitrator usergroup and upon discovery of any gross abuse, can insta-eject folks (as happens currently).
This is a social problem in entirety rather than a technical one. We actively refuse to elect anyone as arbitrator, unless he's proved his worth in the roles of lesser mortals. Winged Blades of Godric (talk) 13:13, 17 August 2019 (UTC)
What permissions would likely be necessary for an arbitrator usergroup? I would guess view deleted and oversighted content, what else? · · · Peter (Southwood) (talk): 16:18, 21 August 2019 (UTC)
Checkuser, checkuser-log, ancillary view deleted rights (there is more than one such user right) and edit filter viewing rights. Jo-Jo Eumerus (talk, contributions) 08:20, 8 September 2019 (UTC)

Again, this is something that absolutely should be project- or affiliate-determined. There is no evidence provided, nor logical reason given, for insisting on a global standard that would apply to 100% of all projects and affiliates. There already exist minimum standards and requirements for tasks like checkuser, oversight, and steward. Different projects and affiliates have different needs. The fact that this section is so nebulous that it doesn't explain what roles it applies to is problematic. Does it include administrators? Does it include executive directors? This really significantly conflicts with many of the other recommendations from several other working groups that are focused on decentralization of power from a monolithic central body, rather than increasing centralization and control. It's symptomatic of a bigger conflict between the recommendations of various groups. Risker (talk) 19:10, 21 August 2019 (UTC)

An interesting point - this supposedly decentralising recommendation is actually centralising control Nosebagbear (talk) 19:55, 21 August 2019 (UTC)
I believe office limits are a solution to the partially diagnosed problem. We should consider the dissemintation of permissions itself. See Special:ListGroupRights. Why sysops (or any other groups with advanced permissions) are able to do what they are able to do? Maybe some permissions should be splitted to different groups? Maybe there should be a proliferation of groups with advanced permissions? In my opinion, inequality might be due to notions of big deal, hierarchy, and power related with the permissions, regardless of whether these notions are philosophically legitimate or not. This of course requires research, as it's difficult for involved volunteers to assess the situation globally and objectively. (This is my own POV, so I use this account.) Tar Lócesilion (talk) 17:09, 28 August 2019 (UTC)
@Tar Lócesilion: - but this doesn't handle the issue that various rights have to stack in order to operate properly. For example, a sysop doing copyright violation work, needs to be able to see deleted content, delete content in violation, block recurring violators and in certain cases, protect pages. The same could be said in libel instances, and quite a few others. Splitting rights into lots of advances permissions would cause lots of admins to have to go acquire them to effectively perform their job. A fair chunk of short-term disruption and then a heavy group overlap. There's been quite a lot of active unbundling over the last decade, most of it positive. But "the Big Three" are fairly core. Outside "above" sysop, ('Crats, CU/OS, really niche rights) there's only a couple of other rights given. Stewards obviously need them all because they're the rights-handlers, and the sole advanced permissions users for many wikis. Nosebagbear (talk) 17:23, 28 August 2019 (UTC)
Maybe there exist such different roles that require different sets of permissions (X, Y, and Z) and instead of "sysop doing X", "sysop doing Y", "sysop doing Z" (not necesarily "sysop"!) we could have "group X doing X", "group Y doing Y", and "group Z doing Z"? What's more, plesae note that on various wikis, there are various cultures related to groups of permissions. Again, this clearly does require research. Tar Lócesilion (talk) 18:13, 28 August 2019 (UTC)
Obviously I can only speak to my personal knowledge and experience of en-wiki, but only a small fraction (it's there, but it's small) of editors devote themselves to 1 core activity (or groups of activities, AfD & Deletion Review being pretty similar, for example). Most do a few different areas. I'd actually be concerned, and sad, if there were "cultural" (rights-enforced or not) splits like that. There is a single big example of that in enwiki - Arbitration Enforcement. Nosebagbear (talk) 18:31, 28 August 2019 (UTC)

Nope to Unified CoCEdit

Part 1 of this working paper doesn't even stop to consider why a unified code of conduct won't (or even might not) work.

No consideration that communities are different to each other. That different communities view the nature of a wikipedia work space differently (and certainly different to the WMF). That the best processes for handling issues also varies (and varies both project to project and over time)

No consideration that stricter behavioural rules might handicap non-harassing competence. It doesn't think how to handle the loss of non-problematic editors who don't like the new rules. It states it's essential to "collaborative working environments" - office-style is not necessarily the most collaborative.

It also could do with considering the fact that the WMF (specifically T&S in this area), who presumably would be heavily involved in in forcing a minimum CoC, as well as processes for implementing such, is now heavily distrusted by both en-wiki and de-wiki, the largest communities (indications of issues with both belgian wiki and zh-wiki arose I know, but I don't know the comparative depth of feeling there).

The idea of sanctions against communities as a whole is downright terrifying - individuals sanctioned for the (in)actions of others, decided by the WMF. Someone above said it sounded Orwellian, and I think they're right.

I am not against (in fact, I'm strongly in favour of) significant improvements in clarity on civility etc, with some additional bright lines. But so many cases will always be edges cases, wherever you put the base line, and so I dispute that clarity can always be provided.

Do we have some solid data on how many users behave poorly on their 2nd (or 3rd etc) project because they weren't aware of that community's expectations (but didn't cause issues on their home project)? I can understand being confused in many policy areas changing from one to the next, but I wouldn't have thought any project's behavioural rules were so different that it was a frequent cause of issues.

@Doc James: - I know you weren't there for the final meeting, but are you in favour of all 15 recommendations? Is there one or more people from the working group we can pose questions to about them and the rationales? Nosebagbear (talk) 14:45, 15 August 2019 (UTC)

What exact differences do you anticipate between individual project CoCs?Edit

Specifically, what civility rules do you think would be different (not applicable to one project, but valid for another), 1) between same-language wikis (enwiki vs. enwikt, or commons, etc.) and 2) between different language wikis? — Aron M (talk) 16:41, 17 August 2019 (UTC)

  • To whom is this question addressed? Without specification one would expect it to be the Community Health Working Group, who published the content page, but as they seem to have recommended a single CoC for all Wikimedia projects, that would be implausible. If it is addressing the commentators, which one(s)?
  • My own opinion is that it is impracticable to predict with any useful level of confidence, maybe impossible, and in the absence of a credible draft, even more so, which is exactly the reason why a single CoC is unlikely to be workable. From personal experience on only a few projects, in only two languages, I would expect somne differences, which would emerge during development of the CoCs. If perchance the differences are reconcilable, there is nothing to stop two or more projects agreeing on a unified CoC for those projects, but this should come after they have had the opportunity to come to a workable CoC of their own. Translation is possibly an even bigger problem for a single CoC. If there were to be one, would the English language version be considered authoritative? Is this likely to be acceptable to the other languages or will it be seen as yet another example of neo-colonialism? · · · Peter (Southwood) (talk): 18:40, 17 August 2019 (UTC)
For example, the Simple English Wikipedia would have a code of conduct written in Simple English (which may be a useful lesson for the WMF lawyers.) Hlevy2 (talk) 11:52, 18 August 2019 (UTC)
That will be an interesting thing to see. It might have to be somewhat longer than the standard English version. · · · Peter (Southwood) (talk): 16:23, 18 August 2019 (UTC)

Equality PrincipleEdit

A key to the laws of the United States is an equality principle. That is, laws are administered without regard to race, religion, age, nationality or sex. Would the Model Code of Conduct contain a similar equal treatment rule? Will there be "equal protection of the laws?" Equal protection is essential for maintaining a NPOV encyclopedia and a healthy community. Hlevy2 (talk) 12:06, 18 August 2019 (UTC)

That would have the interesting effect in making a number of the working groups' suggestions (quotas etc) and the WMF's current policy (e.g. minimum age for certain positions) be in breach of it. Nosebagbear (talk) 13:12, 18 August 2019 (UTC)
Not necessarily. There are minimum ages for buying alcohol and driving a car in the US. Quotas might be more of a problem, but I am not a lawyer, particularly not in US law. · · · Peter (Southwood) (talk): 16:27, 18 August 2019 (UTC)
But contributors are anonymous...? --Yair rand (talk) 01:17, 20 August 2019 (UTC)
Staying anonymous is a valid way to prevent discrimination. The problem is motivation for a pattern of actions. For example, if an admin takes a number of actions against a user for a series of policy violations, the user, if applicable, could claim "I am being harassed because I am a woman." Under the equality principle, the user, if applicable, could claim "I am being harassed because I am a man." The admin's conduct should be measured without regard for the identity characterists of the user. WMF projects, as healthy communities, should treat everyone with fairness and respect. Nobody is "lesser" or marginalized in a healthy community and everyone is treated based upon merit and individual conduct. Hlevy2 (talk) 03:53, 20 August 2019 (UTC)
Anonymous does not always mean not identifiable as a member of a group. Discrimination is not usually against a specific person, it is more usually against the group which the person identifies is identified with, which is often recognisable from editing habits. Harassment is often more personal, and does not necessarily relate to group membership or identity. · · · Peter (Southwood) (talk): 04:27, 20 August 2019 (UTC
Yes, admin duties are very complicated and context dependent. The question is will the training that admins receive under this proposal teach them that WMF projects do not discriminate on the basis of race, religion, age, nationality or sex? If the WMF spends money on developing training materials, how will we ensure that the materials reflect our values including the equality principle? Will WMF work from pre-existing training materials of other organizations that train a large number of customer service (e.g., call center) staff? Hlevy2 (talk) 08:23, 20 August 2019 (UTC)
Any training would have to align with the code of conduct, the terms of use, and the policies of the specific project. I would expect input from WMF, members of the relevant project, and anyone else with an interest. I do not see much chance of WMF managing to set up an acceptable system on their own. Besides issues of competence and disconnection, it would be a lot of work. Crowdsourcing is the way to go for many reasons. · · · Peter (Southwood) (talk): 08:53, 20 August 2019 (UTC)
The Working Group recommendations are so vague that it is difficult to discuss them. We need to articulate what values will form the basis of the training, who will develop the training materials, and how the communities can guarantee that the training materials reflect our values. By expressly adopting the equality principle first, then all training can be measured against that. Hlevy2 (talk) 10:42, 20 August 2019 (UTC)
Except we can't do that because we'd have to get guarantees that other recommendations that might clash with it had been canned first. Nosebagbear (talk) 10:49, 20 August 2019 (UTC)
To set up traiming we need to decide what we are going to assess, and how we are going to assess it. This is occupational skills level training, not a PhD. Most of the knowledge requirements can be assessed using multiple choice and other questions which can be marked by an automated system on a platform like Moodle, so the operating labour overhead can be low. Much of the learning can be directed by linking to policy pages, which will always be up to date. Some could be directed by linking to help pages, which might get them inproved too.
There will be an administrative overhead for the software and general admin, which could be managed by a combination of WMF staff and project volunteers with the requisite skills. We probably have dozens of users with these skills already.
Keeping the assessment questions up to date with policy changes will be a bit of work, but not very much as policies tend to change slowly.
There should be little need for personal intervention in training - a formal course could be arranged a few times a year for those who need or want it. Most learning for qualifying for functionary appointments should be possible without intervention, just someone to ask questions of in cases of ambuguity, change or doubt.
Training in use of tools will require a bit of work, but we are used to working on wikis, so everone can help. Learning how to learn on the training system could be considered a filter of a kind. Most of the work would be setting up the assessment question database, and most of that work will be working out what we want people to know.
Projects can share compmemts that are of common value, and modify others for theit own needs. Same with translations.
Many of the modules would also be appropriate for educating people who have difficulty comimg to terms with project policies. If someone is causing a problem, they can be required to pass a relevant module befor continuing to edit. After they pass either they will avoid the problem because they understand what they should do, of if they continue prolematic behaviour it will be an indication that the training needs to be inproved, or the user in not here to build the project. This could be a far more proactive and preventative system than blocking for a fixed period. Where competence is required, we can help provide that competence. Maybe it will work. I think it is worth a try. · · · Peter (Southwood) (talk): 12:27, 20 August 2019 (UTC)
Your points seem to be talking about general wikipedia training. We are discussing (and raising concerns with) the WG's desire for functionary (or Admin & Functionary) training. They don't need to be trained on policy, the training is more (supposedly) designed to be tailored for things like conflict resolution. Proper conflict resolution training could never be done by multiple choice. Hell it couldn't even be assessed by case studies because there's multiple routes and can be tailored to the individuals (both participants and reviewer). Nosebagbear (talk) 12:35, 20 August 2019 (UTC)
You are mostly right, but there is value to assessment of the knowledge and skills of candidate admins etc in that there are often disputes during RfA as to whether the candidate is sufficiently skilled and knowledgeable in the relevant technical aspects which could be demonstrated to some extent by having passed some tests on the topics. This could possibly make RfA a bit smoother. Not all admins are involved in conflict resolution, and there is no need to be an admin to resolve a conflict. I would go so far as to say that some people could probably never learn to resolve conflicts because of their underlying philosophy, and others can with little or no training because they have the necessary empathy. If the WGs think that no-one without conflict resolution skills should be admins then they are going to eliminate a large portion of people who have other essential skills. These people are useful in their own way. I certainly have no confidence in WMF being competent to train or assess anyone in conflict resolution - we probably do a better job than them already, and we are not in any way doing it well. In commercial diving it is generally accepted that it is easier to make a good welder into an adequate diver than to train a diver to weld well. The same concept holds here. My opinion is that many conflicts can be avoided by education, so they would not need to be resolved. Also sometimes the people in conflicts want to "win", rather than come to a resolution where the problem can be solved. Training people to prevent and avoid conflict includes ensuring that they know the rules and can apply them correctly, and there are some of our admins who do not seem to understand or apply some of those rules correctly. I see a significant amount of passive-aggressive behaviour that goes by unchallenged, often apparently as an attempt to bait an opponent into more blatant misbehaviour. Ad hominem in the more extreme forms is usually challenged, but a lot of more subtle examples slide by, and people get away with logical fallacies far too often. Just having more people rcognise these problems should help. We should do what is possible in the hope that some improvement follows.
Anyway, your opinions are often insightful and I would like to know your take on this point. · · · Peter (Southwood) (talk): 13:38, 20 August 2019 (UTC)

We can agree on broad principles, such as "admin training is a good thing", but are we missing a few overarching principles such as "training should teach that equality should be fostered"? Don't adopt a rule to be applied to some other identity group if it can't be applied to everyone? Perhaps the "movement charter" will incorporate such important principles, but they should also be baked into all other programs and recommendations. Hlevy2 (talk) 00:03, 3 September 2019 (UTC)

I think you're getting pretty far into operational topics, and not strategy here. "Equality should be fostered" may be a strategic objective, and "there should be training offered for administrators" may be as well, but detailing the nature of either would be more operational than strategic; there is even a good question as to whether or not conflating these two objectives would be desirable; I'm assuming the "equality" principle is one that should apply regardless of the number of administrators on a project - especially since there are dozens of projects with fewer than three administrators, and many that have no administrators at all. Risker (talk) 02:42, 3 September 2019 (UTC)
Risker, I respectfully disagree. In today's parlance, there is a big difference between equality and equity. The movement should guarantee equality (e.g., equal treatment and fair application of the rules) to all users. The general goal of achieving "knowledge equity" should never be an excuse for mistreating users in an unequal fashion. So without regard to operational issues, we should make clear that adopting a goal for 2030 of "knowledge equity" should never be a justification for violating the equality principle. Hlevy2 (talk) 19:14, 4 September 2019 (UTC)

Prevent accumulation of power seems to be a bad ideaEdit

To be functioning, Wikicommunities need people who have power because they build up trust over time. If term limits would be placed on being an admin that would lead to loss of organizational capital. The same goes for Bureaucrats. Many projects would likely substantially damaged if they would need to introduce term limits for such functions. ChristianKl❫ 09:25, 20 August 2019 (UTC)

It depends on what one means by "power". Mostly we need trusted people with advanced technical "permissions". Whether they thay have power, and how much power they have is debatable and varies between projects. I have permissions on a couple of projects to do things like delete and undelete pages, block vandals etc, which is very useful, and someone has to do it, but to persuade people that a policy change is needed, or that a specific page should be deleted or undeleted, or specific content should be allowed in a particular article, I have no more power than anyone else who can argue logically and persuasively, and follows the rules. Term limits on the ability to do the necessary maintenance would be counterproductive. · · · Peter (Southwood) (talk): 12:48, 20 August 2019 (UTC)

Engagement by the Working GroupEdit

So far I have seen none. Does this mean they do not consider it appropriate to clarify the many problems raised, or it this a policy thing that they have to stick with? Something else? I would ask why not, but would there be any point? They have not answered any other questions. Either way it is not doing their credibility any favours. · · · Peter (Southwood) (talk): 07:56, 22 August 2019 (UTC)

Absolutely, they should all be back from Wikimania by now, and there wasn't any engagement before they left either (notwithstanding Doc James who raised a query on a WG's recommendations he was in, due to being unable to make the last meeting) Nosebagbear (talk) 08:27, 22 August 2019 (UTC)
It's clearly not a rule that they're not allowed or not supposed to engage. Members of the Advocacy group have been quite engaged in answering questions and addressing concerns raised. It certainly would be nice to see engagement in a similar way from some other groups, especially when their proposals were widely found to be unclear or objectionable. Seraphimblade (talk) 13:29, 22 August 2019 (UTC)
Not following this for more than a week; did the CommHealth_WorkGroup engage? Page Nosebagbear ... Winged Blades of Godric (talk) 13:56, 3 September 2019 (UTC)
@Winged Blades of Godric: - nope. Literally while you were typing dropped another message here Talk:Strategy/Wikimedia movement/2018-20/Working Groups, and I've done so a few other places. At least in the 3 groups I'm editing in, I've not seen any activity. Not helpful Nosebagbear (talk) 14:00, 3 September 2019 (UTC)

From Catalan SalonEdit

Link here.

Conduct codeEdit

Sounds good. Fears: If somebody who made a mess with FRAM is going to give us lessons about how to deal with conflicts, reticence could be found.

Support to health support and Trust and safety. Doubts about implementation. Community members with health issues are like sick parents. We can't abandon them.

Talks about 'Wikipedia addicts'. When servers fall, some users have abstinence syndrome. Jokes but sincere worries.

LeadershipEdit

Limits for wikipedia community functionaries but the document doesn’t talks about limits for functionaries of the WMF structures.

Seems like nobody knows Meta. The knowledge wide platform could be Meta. Invest in product, improve flow, but don't abandon products.

That a longtime admin could leave is good (and worked in ca.wiki) when they assume new goals and improve other projects. Because they become users with an admin point of view. (jokes about maoism). It is a good idea also for diversity. Longtime admins and patrollers make good tutors.

About leadership: How can we assure the renewal of leaders? We need an environment that assures that. CH should work on what happens when long time leaders feel that the project is theirs and are afrraid that if they leave it would sunk.

Redefining formal and informal powerEdit

Limitation of mandates: why only in positions on wiki? There are problems with minor wikis of few volunteers to be admin, but except for Arb-com, in the WMF charges, there are enough candidates, but we find that there are lengthy mandates and they are extended, creating the opposite situation to the requested one in the recommendation.

In the case of the affiliates, there may also be cases where those who direct them are always the same people, creating problems of lack of representativeness due to non-renewal. It is feared that the protection of those in a situation of risk leads to accountabilty problems. Fear of an increase in the bureaucracy of the movement.

Conduct codeEdit

Squeaks that a code should be accepted in every local communities (projects). A solution would be that there was a global one but that there would be the option to make a particular code adapted to context, and comptible in values with the general. If there is no local code, the global one must be applies.

We see a Dangers if a code of global behavior evolves into a path to punish criticism. Doubts that WMF is still able to make deep changes in its structure.

Return to "Strategy/Wikimedia movement/2018-20/Working Groups/Community Health/Recommendations/Rules and regulations, decision making processes and leadership" page.