Talk:Universal Code of Conduct/Enforcement guidelines

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search


This page is for discussions related to Universal Code of Conduct/Enforcement guidelines.

  Please remember to:


  Discussion navigation:


What is the improvement method[edit]

Hello,

Thank you for posting this.

We were notified that we would get sufficient time to see the textual proposals, and highlight where changes need to be made to them. What is the methodology being planned for us to raise issues and get the UCOC#2 committee to make changes to them, prior to the ratification vote? Nosebagbear (talk) 23:40, 24 January 2022 (UTC)[reply]

The above seems like a reasonable request. Why was there no response? 08:28, 7 March 2022 (UTC)2600:1700:D0A0:21B0:60C6:E2D6:7132:DA34

Translations[edit]

Do we need to translate this page or it will be done by staff? I'm a little bit confused about this. --justinianus | talk 19:01, 25 January 2022 (UTC)[reply]

Justinianus: I removed the template, as most of the translations we had prepared are up - so please feel free to proceed. Thank you! Xeno (WMF) (talk) 20:46, 25 January 2022 (UTC)[reply]

Japanese community[1]--YShibata (WMF) (talk) 12:07, 4 February 2022 (UTC)[reply]

Code enforcement definition - "punishments"[edit]

I suggest "and punishments" be removed from the final paragraph of the code enforcement definition. WMF probably not in the business of training some users of the website to punish other users of the website (even if they deserve it). The preceding "imposing technical restrictions" already covers it better. Levivich (talk) 02:20, 26 January 2022 (UTC)[reply]

Indeed this would actually mean we'd have to amend en:WP:PREVENTATIVE, and I'd much rather see not just a removal of the "and punishments", but a specific inclusion of the concept that sanctions should be preventative, not punitive. Nosebagbear (talk) 09:17, 26 January 2022 (UTC)[reply]

@Nosebagbear and Levivich: I made this change only since I think there was a genuine error somewhere that had this language included. Not going to do it twice, though. –MJLTalk 16:20, 26 January 2022 (UTC)[reply]

The issue I see with en:WP:PREVENTATIVE is that it is too broad in scope, ambiguous, and opens the door to en:WP:NOTCRYSTALBALL which could work against targeted editors who have been forced to endure the wrath of prejudiced admins harboring preconceived notions, or they simply don't like a particular editor because of ideological differences. We're not going to please all of the people all of the time. Atsme📞📧 14:09, 15 March 2022 (UTC)[reply]
"Sanctions" would be better than "Punishments" CFynn (talk) 13:22, 18 March 2022 (UTC)[reply]

Systematic issue[edit]

I think you mean "systemic" instead of "systematic". One means "acting according to a system" and the other "inherent in the system". Vexations (talk) 12:02, 27 January 2022 (UTC)[reply]

There's also "system issues" in the text, so there's "system issues", "systemic issues" and "systematic issues" to choose from. Whichever is chosen, the choice should be a conscious one, not an accidental one. ToBeFree (talk) 09:26, 30 January 2022 (UTC)[reply]
You are right. This was changed back and forth and discussed several times in the committee and the result is not optimal. And yes, there is one "system issue" too. Taylor 49 (talk) 00:13, 3 February 2022 (UTC)[reply]
"systemic" seems right. –SJ talk  00:37, 4 February 2022 (UTC)[reply]
Yeah, I always hear "systemic" used in this context. "System" sounds more like a malfunctioning computer not a widespread problem, and "systematic" is when you classify things. Jo-Jo Eumerus (talk, contributions) 10:28, 4 February 2022 (UTC)[reply]

Description lists / bulleted lists[edit]

The HTML created by starting one line with ";" followed by lines starting with "*" (<dl><dt></dt></dl><ul><li></li></ul>) is a bit, hm, semantically suboptimal. Looking at the current revision of the page, "Escalation paths" is correctly marked as a L4 heading while "Violations related to affiliate governance" is a (misused) description list. ToBeFree (talk) 09:15, 30 January 2022 (UTC)[reply]

"See definition on Meta."[edit]

...with a link, please. ToBeFree (talk) 09:23, 30 January 2022 (UTC)[reply]

Some comments on the most recent version[edit]

Reviewing this version, I have some questions:

  • "Code enforcement is a responsibility of designated functionaries and bodies with technical or decision-making power, such as, but not limited to: local sysops" and "advanced rights holders": I take this includes bureaucrats?
  • "Functionary" in wiki-jargon has a specific meaning: Editors with access to the CheckUser or Oversight permissions. You may want to use a different word here if it's meant to be used more broadly.
  • From "Recommendations for the reporting and processingo" I get the impression the idea is of a tool that only the various editors who have to "enforce the UCoC" have read access to.

And concerns:

  • "Individuals required to acknowledge and adhere to the Universal Code of Conduct will be required to attend training to ensure a common understanding of implementation. Other members of the community will be able to attend this training if they wish to do so." I think it would be good to specify what this exactly entails - physical attendance, reading materials, timing and so on.
  • Who decides on the content of the training? If all local permission-holders are expected to attend such a training, it might strongly influence how they apply and use their permissions, and I'd imagine projects would want to have some control on this.
  • I am sure this has been discussed elsewhere, but at least on enwiki I don't think that the administrator policy and appointment processes are designed for a "wiki-cop" role. Or in other words, their purpose isn't to act as one.

That's it from me, for now. Jo-Jo Eumerus (talk, contributions) 11:04, 2 February 2022 (UTC)[reply]

"advanced rights holders": I take this includes bureaucrats? YES. Taylor 49 (talk) 00:13, 3 February 2022 (UTC)[reply]

The Foundation doesn't understand how our projects work[edit]

A Universal Code of Conduct has to be drafted by the whole community and not by a committee. The Wikimedia projects are based upon grassroots democracy principles. Normally rules, guidelines and so on are discussed and formulated by all people who want to participate collaboratively in an open discussion. Aristocratic concepts like committees are very unusual. Because of this the Universal Code of Conduct is absolutely unacceptable in its current form. The only possibility to fix this is to start again from the very beginning. Chaddy (talk) 21:22, 4 February 2022 (UTC)[reply]

I agree with you ... I'm certainly not going to vote for guidelines on how to enforce a Universal Code of Conduct that, as written today, turns everyday activities like logging or discussing edits from governments or PR companies into "harassment" and "doxing", and describes what any normal editor faced with a flat-earther or conspiracy theorist will have to do as "psychological manipulation". Cheers, --Andreas JN466 18:54, 22 February 2022 (UTC)[reply]
I disagree that the whole community should draft it - that would be chaos. I'd be ok with a balanced mix of representatives from the community, but I'm of the mind that the WMF. ArbCom tends to throw the ball back into the admin court via discretionary sanctions, and it simply is not a productive way to handle all situations - that's when systemic bias opens the door. Atsme📞📧 14:16, 15 March 2022 (UTC)[reply]

U4C or UC4?[edit]

U4C, per common usage, would be an abbreviation for a term with 4 U's followed by a C. The Universal Code of Conduct Coordinating Committee would more properly be UC4. Of course, the element "C4" is a designation for an explosive, which I suspect is why I have yet to see an American airport with alphanumeric gates inlcude a gate C4, but I digress. Seriously, it should probably be changed, lest an inordinate number of other nitpicky Wikimedians make similar complaints. Jclemens (talk) 08:02, 5 February 2022 (UTC)[reply]

Hm?
Numeronyms. ToBeFree (talk) 19:49, 5 February 2022 (UTC)[reply]
@Jclemens, If you have never seen an American airport with a gate C4, you must not have ever been to (or through) DFW or Atlanta Hartsfield -- two of the largest airports in the US. Of course, just traversing through an airport, you won't generally see ALL of the gates... 73.127.147.187 07:07, 21 February 2022 (UTC)[reply]

questions about scope[edit]

I'm trying to make sure I understand this.

  • Local advanced permission holders are charged with primary enforcement of the UCoC
  • On projects with arbitration committees, the committee is also expected to enforce it
  • Real or perceived failure to do so can be reported to the U4C
  • The U4C is described as "peers" of ArbComs, but it also says it is a place for "final appeals" which would seem to indicate it could in fact overrule a local arbcom that, as of right now, is already the final avenue of appeal

Is that correct? Beeblebrox (talk) 23:45, 8 February 2022 (UTC)[reply]

@Beeblebrox: (speaking as only one person here) Correct on all parts except the last one. Both ArbComs and the U4C are places of final appeals. For enwiki, enwiki's arbcom would be the final place of appeal, but on simplewiki (with no arbcom at the moment), the U4C would have that role. –MJLTalk 06:35, 10 February 2022 (UTC)[reply]
@MJL: Thanks for the reply. If that's the case I think it would be best if it were explicitly stated, in order to avoid the impression that you can just go over the head of a local arbcom. Beeblebrox (talk) 01:10, 12 February 2022 (UTC)[reply]

TL;DR: Please summarize what you request us (local community and/or individuals)[edit]

The enforcement guideline (and also UCoC itself) is wikt:TL;DR. I don't know what I and our colleagues must/should/may do for UCoC. Please clarify it in compliance with RFC2119 or defined terms by yourself. Long detailed texts can be left as it is now. We need the summarized version of the current long texts.--aokomoriuta (talk) 12:28, 10 February 2022 (UTC)[reply]

Copyedit[edit]

Can you guys copyedit this document? It's a bit minor, relatively speaking, but aesthetically the document looks a bit messy with grammar/formatting inconsistencies etc. I suppose different people wrote different parts, hence the differences, but surely that can be cleaned up by WMF facilitators at the end. ProcrastinatingReader (talk) 14:30, 11 February 2022 (UTC)[reply]

We might want to fix words like "accusee" for example. I know we mean accused, but in translation, the distinction between accuser and accused is easily lost. See for example accusée. We might also want to avoid things like "formfactor limitations", (form factor is usually written as two words, but even then, why not say "if there is enough space" instead? Words that ought to be hyphenated in some cases include "in person", "high level", "third party", and "decision making", but "with-in" should not. We should probably never use "so as to" or "in order to". "Allow reports to be processed privately by whomever " should use the subject pronoun whoever instead of whomever. "Appeals are not possible in following cases" is missing "the". There's more of course, but perhaps this is enough to convince us that some copy-editing is in order. Thanks, Vexations (talk) 20:21, 11 February 2022 (UTC)[reply]
@User:Vexations: Thank you for proofreading. Support some of your suggestions, but not sure who has the right to edit the document now, because the drafting process has finished, and it is the Foundation who "owns" the text. IMHO both "by whomever" and "by whoever" are equally correct. What's wrong with "in order to"? Taylor 49 (talk) 01:09, 20 February 2022 (UTC)[reply]
Whoever and whomever are not equivalent in English. That's not a matter of opinion. You can look it up in a dictionary. I'll note that in some languages the distinction between who and whom does not exist, so some people whose first language is not English may find that confusing. That is a good reason to avoid using those words. "In order to" is just another way of saying "to". So is "so as to". When writing texts that need to be translated we should avoid such unnecessary ornamentation. Our goal should be to produce a text that is clear, unambiguous and easy to translate. If the committee or the WMF can't do this, they should hire a professional editor who (not whom) can. Vexations (talk) 15:59, 20 February 2022 (UTC)[reply]
But "whomever" is the object of the sentence, not the subject (you wouldn't say "processed privately by he", you'd say "processed privately by him"), so it's correct as written. -- Ahecht (TALK
PAGE
) 17:31, 17 March 2022 (UTC)[reply]
@Ahecht and Taylor 49: "Whomever" is the subject of the relative clause, not its object. As such, it should not be in the objective case ("whomever" is the objective case of "whoever").
Merriam gives the following example: "A prize will be given to whoever solves the riddle."
In other words, the object governed by "to" in this example is not the word "whoever", but the entire relative clause, whose subject is "whoever" and whose verb is "solves" ("whoever solves", as in "s/he solves"). For more examples see https://www.grammarbook.com/grammar/whoever.asp Andreas JN466 19:12, 17 March 2022 (UTC)[reply]

Training[edit]

Just to make it sure, because I can not understand this point in the version in any on the languages I speak. Do I read it correctly that all administrators on all WMF projects will be sent to a compulsory UCoC training, and if they refuse to undergo the training they lose the flag?--Ymblanter (talk) 20:03, 16 February 2022 (UTC)[reply]

I read "Individuals required to acknowledge and adhere to the Universal Code of Conduct will be required to attend training to ensure a common understanding of implementation." and interpret that as "training is mandatory for anyone who must adhere to the UCoC". The UCoC says "The Universal Code of Conduct applies equally to all Wikimedians without any exceptions." I don't see how this wording allows for "Other members of the community" for who adherence is optional (me, I hope). My good faith reading of the above is that "adhere" is not meant in the sense of "comply with", "bind oneself to observance" or "stick to the rules", but in the sense of "enforce" or perhaps "support" or "maintain loyaly". Anyway, the current wording is far too vague, impossible to translate, and for the Nth time: PLEASE people, write in clear, plain English. Vexations (talk) 21:49, 16 February 2022 (UTC)[reply]
I hope this gets clarified, otherwise I am going to vote against the document.--Ymblanter (talk) 21:57, 16 February 2022 (UTC)[reply]
"Attend" also implies some things about the training that are incompatible with how the community is organized. AntiCompositeNumber (talk) 04:45, 17 February 2022 (UTC)[reply]
I think "Individuals required to acknowledge and adhere to the Universal Code of Conduct" means people listed in the "Affirmation of the UCoC among certain groups" section. So you wouldn't have to attend this training because although you're require to adhere to the code of conduct, you're not required to "acknowledge" it. But yes, forcing people to undergo compulsory unspecified training on pain of losing permissions doesn't sound like it's going to be very popular. Hut 8.5 18:21, 18 February 2022 (UTC)[reply]
It says "All advanced rights holders" I think that includes all administrators and it affects me, as a New page reviewer (patroller) as well. Vexations (talk) 18:54, 18 February 2022 (UTC)[reply]
@Vexations: No, it doesn't affect you. From the glossary, Advanced rights holder: user who holds administrative rights above typical editing permissions, generally elected through community processes or appointed by Arbitration Committees. This includes, as a non-exhaustive list: local sysops / administrators, functionaries, global sysops, stewards.MJLTalk 23:44, 18 February 2022 (UTC)[reply]
The list of these people includes "All advanced rights holders", this is how I have come to this conclusion.--Ymblanter (talk) 19:45, 18 February 2022 (UTC)[reply]

I'm going to comment as an individual member of the committee because I think people are really overthinking this training thing. Here's how I interpret everything...
Important bits:

  • The following individuals should be required to affirm (through signed declaration or other format to be decided) they will acknowledge and adhere to the Universal Code of Conduct...
  • The users listed above should accomplish the affirmation at the occasion of acquiring the right or role, as well as every re-election, renewal or prolongation, the existing ones do so within a short time after the ratification of these guidelines, with exception of current advanced rights holders with rights that are not up for renewal who will not have a set timeframe to accomplish these affirmations. This may be changed on review after a year following the ratification of these guidelines. Once formed, the U4C will create procedures to facilitate these affirmations.
  • Individuals required to acknowledge and adhere to the Universal Code of Conduct will be required to attend training to ensure a common understanding of implementation. (emphasis added)

Despite what people may think, current admins don't have to affirm anything right now. We basically set the deadline for that to basically whenever they get around to it. Also, the U4C won't exist yet after ratification. That's why so many things are written in future and conditional tense. Also, despite the fact the document says training will be required for new admins, there is no training right now. It also doesn't say anything about what happens when someone refuses to take the training. The U4C won't even exist if we ratify these guidelines, and there is nothing giving them the power currently to do that.
Again, this is just how I read the document. I'm just one user, and there's nothing stopping me from being wrong about how the WMF will interpret these guidelines. –MJLTalk 00:10, 19 February 2022 (UTC)[reply]

It also doesn't say anything about what happens when someone refuses to take the training - it does say that the training is "required", which definitely implies that refusing to take it is not an option and not compatible with holding these permissions. There aren't any qualifiers attached to this word. I'm not exactly keen on the idea of having to take unspecified training in order to be judged fit to keep permissions I've had for over a decade, and if lots of people get desysopped for refusing to take it that could be very damaging to projects. Hut 8.5 10:21, 19 February 2022 (UTC)[reply]
Indeed, this is sounding increasingly like the UCoC can and will be used to remove or block advanced permission holders duly elected by their local communities. I can't support that. We've been trying to make the process for gaining adminship easier for years and now there's this thing that is guaranteed to make otherwise qualified candidates unwilling to run. Beeblebrox (talk) 18:38, 19 February 2022 (UTC)[reply]
@Hut 8.5 & Beeblebrox: Okay, assuming I'm wrong about that and the training is required [even though it's written in future tense with the qualifier will be with no statement as to when (and we know it isn't referring to post-ratification because that's when the guidelines say the WMF should start developing the training)]. The line current advanced rights holders with rights that are not up for renewal who will not have a set timeframe to accomplish these affirmations inherently means that (to me) they are not among the individuals required to acknowledge... the Universal Code of Conduct.
As always, speaking for myself and not the whole committee. –MJLTalk 19:28, 19 February 2022 (UTC)[reply]
It only says there's no set time frame, which implies that at some point the U4C or the foundation might decide that enough time has passed and declare it must be done now, but that's not even the point. You seem to think we're only worried about ourselves, and this little loophole should satisfy us. That is not the case. I don't think any volunteer should be forced to agree to this in order to gain a permission a local community has elected them to. I cannot possibly support this document while there is any such requirement in place. Beeblebrox (talk) 19:50, 19 February 2022 (UTC)[reply]
There is no loophole. "current advanced rights holders...will not have a set timeframe to accomplish these affirmations" is referring to the "Affirmation of the UCoC among certain groups" section, which means the signed statement that you agree to abide by the UCoC. The training isn't one of these "affirmations", so that statement doesn't apply to it. But yes I absolutely agree with Beeblebrox that this isn't a reasonable expectation and I don't think other community members will either. Hut 8.5 10:55, 20 February 2022 (UTC)[reply]
Yeah, I'm going to have to agree with Beeblebrox on this one. I can't see any reason at all for mandatory training and mandatory signing of some hypothetical document in order to become an administrator. Many projects already have a hard time getting qualified candidates to assume that role now. I'm not even sure I can make the case for other functionaries and advanced permissions, and a lot of it depends on the wording of the document itself. The UCoC is a pretty subjective document as it is, and we need to see some history of how it is being interpreted, both globally and locally. I've yet to see a good case for making signing of a document mandatory, since the policy itself is explicit that it covers every person participating in a WMF wiki or a Wikimedia movement activity. Speaking as a functionary (checkuser and oversighter), there's a really big difference between signing off on the confidentiality statement (which is role-specific and very limited), and signing off on this. I mean...not only is going to make it harder to get new administrators, but we're probably going to see active administrators pulling back on many projects. Risker (talk) 05:05, 20 February 2022 (UTC)[reply]
Just going to add here that there is zero mandatory training for any other global policy or for any level of advanced permission. Not a drop. It's on-the-job training for every other role; some local groups may provide limited support and training for some roles, but I couldn't find a project that has mandatory training for any other aspect of adminship, or checkuser, or oversight, or steward. Bluntly put, I don't think the UCoC is as important as the Terms of Use (which also applies to every single user on a WMF wiki), and I have yet to see anything that persuades me that it is such an exceptional document that it requires thousands of users to undergo compulsory education. Risker (talk) 05:27, 20 February 2022 (UTC)[reply]
I don't mean to disrespect those who put their hard work into drafting this, but frankly it reads like it was written be persons who have no clue whatsoever how things have always been done on these projects, and what kind of people are doing the work. Beeblebrox (talk) 22:16, 20 February 2022 (UTC)[reply]
Perhaps we can get someone from the WMF to tell us how exactly how they are going to interpret these guidelines. Vexations (talk) 21:23, 20 February 2022 (UTC)[reply]
Ok, so I will vote against the document. Thank you for replies.--Ymblanter (talk) 10:42, 26 February 2022 (UTC)[reply]

Criticism and support[edit]

If anyone wants to support the UCoC, there is a userbox template avaialble. Taylor 49 (talk) 01:09, 20 February 2022 (UTC)[reply]

And if anyone wants to oppose the UCoC, there is a userbox template available for that too. Vexations (talk) 21:30, 20 February 2022 (UTC)[reply]
I think there is a broad difference between supporting the concept of the UCoC and supporting the outrageous aspects of the enforcement guidelines. Maybe we need a userbox for that. Beeblebrox (talk) 22:17, 20 February 2022 (UTC)[reply]

Counter proposal: pony up the dough[edit]

Current wording: "Code enforcement is a responsibility of designated functionaries and bodies with technical or decision-making power, such as, but not limited to: local sysops, stewards, Arbitration Committee (ArbComs) and their members, event safety coordinators, the Universal Code of Conduct Coordinating Committee (U4C), and the Wikimedia Foundation."

I propose this be changed to "Code enforcement is a responsibility of the Wikimedia Foundation."

They wanted this, they can enforce it.

If they don't have enough staff to enforce it, start hiring on admins and functionaries to assist. If you're employing people it is perfectly fair to mandate policies and have them agreed to in writing, and to require employees do training so they can do the job you hired them to do. It's not fair to foist it on a large group of people who already have unpaid positions granted to them not by the foundation or the U4C, but by their local communities. Beeblebrox (talk) 03:21, 21 February 2022 (UTC)[reply]

@Beeblebrox: Um, as an individual I completely disagree with this approach. That'd give a blank check for the foundation to just override community decisions at will in the name of enforcing the UCoC. Why would anyone want that? –MJLTalk 20:38, 26 February 2022 (UTC)[reply]
My actual preference is to just have the UCoC, with a much toned-down version of the enforcement guidelines, as in, the normally understood on-wiki definition of guidelines as "advice that should normally be followed." If, on the other hand, I have to enforce this, and I have to affirm that I will do so, and have to take training, and if I fail to do any of that I'll be fired, the foundation should be prepared to pay for all that. Trust and Safety enforces egregious violations of the terms of use and could hypothetically be fired for failure to do so, because they are employees. If I'm going to be an employee I expect to be paid. Beeblebrox (talk) 20:54, 26 February 2022 (UTC)[reply]

Right to be Heard[edit]

As times are moving on, I realised I should probably raise this aspect on the talk page, along with a summary of actions taken to date on it.

In the phase 2 consultations, overwhelmingly the most requested change (about 40 individuals) to the original draft was the formal inclusion of the Right To Be Heard (RTBH). This includes two primary facets: that of literally being heard before being sanctioned and that of hearing all the evidence against you. Especially the latter sense conflicts with an absolute right to anonymity.

There's a few different variants used, so I'll propose the following for the purposes of discussion: "Individuals should have the right to tell their side of any conduct matter, prior to a decision being made on any sanctions. This also necessitates hearing the evidence against them in order to be able to make specific rebuttals, rather than merely the general charges."

Response to "include it in phase 3"[edit]

This falls down on two primary grounds: balancing of rights and out of scope

  1. The right to be heard is a countervailing right to the right to anonymity. The two cannot co-exist in complete form (an absolute right to accuser anonymity prohibits much of a right to be heard, and vice-versa). The current draft UCOC#2 documentation includes numerous mentions to a right to accuser anonymity in the various levels of conduct proceeding. However, it does not include any clear-cut, definitive and understandable, statements of right to be heard. This leads to a clear statement to communities which one must be given priority. Were we leaving the discussion for phase 3, the balancing test should be fully created at that stage, with both being given a level playing field.
  2. Phase 3 has a specific scope "the Wikimedia Foundation will facilitate a process to draft, in the form of a constitution, the remainder of the U4C process, handle any other logistics necessary to establish the U4C, and help facilitate the initial election procedures.". This scope only allows changes pertaining to the U4C. Any additions regarding a right to be heard would therefore not apply to 95+% of conduct decisions covered by the UCOC.

Update[edit]

Talking to @SNg (WMF): as a follow-up to a UCOC panel conversation (and thus public), it's now been clarified that despite significant amounts of the UCOC#2 policy text only having its first form at Christmas, it couldn't be changed prior to the vote. I don't know if that was always the intention, and if so, why it wasn't prominently noted back in Dec.

Currently the UCOC doesn't have a equitable balancing test, and indeed the only aspect that could be RTBH oriented pertains strictly to the U4C, while anonymity safeguards are global.

RTBH isn't practical/wise in all cases[edit]

Here, I've included a more detailed proposal - please expand and have a read, and review

Example of a proposed RTBH amendment with some exceptions

But the right to be heard is not practical, wise, or safe in all cases[edit]

Right to prior reply: exceptions[edit]

Absolutely! While there will always be edge cases, the following categories would be logical exclusions from a guaranteed right to a reasonable opportunity to reply prior to sanctions (communities could of course still choose to offer the rights for some of these):

  1. Vandalism and equivalent levels of disruptive editing
  2. Threats of harm (including aspects such as wilful doxxing) and egregious personal attacks
  3. Legal threats
  4. Edit warring generating sanctions less than 72 hours
  5. Socking

Right to hear evidence: exceptions[edit]

Then there is the much narrower category of cases where the accused should not have the right to hear all the evidence against them. Inclusions:

  1. Socking - where the accused is a repeat offender and evidence is likely to aid their future socking
  2. Cases where a clear indication that the accused would (and could) retaliate off-wiki

Right to later reply: ongoing misconduct[edit]

Finally, there would definitely be cases where providing a right to be heard prior to sanctions would be non-viable, outside of the exceptions listed. In short, this could be summarised as "cases with likelihood of ongoing/recurring misconduct". Communities would be left the right to define where this line would be drawn. But in cases where someone was sanctioned prior to a right to be heard, then any appeal should require a specific decision for a block to remain (that is a "no consensus" would revert any sanction).

Nosebagbear (talk) 09:34, 21 February 2022 (UTC)[reply]

Discussion[edit]

  • Pinging both those in favour of a RTBH, against, or in-between on the relevant phase 2 discussion, apologies if already aware @Gnom, Mautpreller, DBarthel (WMF), Deryck Chan, Frank Schulenburg, MBq, Mirji, Ghilt, Nasiruddin, Aschmidt, Gardini, Schlesinger, Andim, Voyager, Chaddy, Tkarcher, Lyzzy, Der-Wir-Ing, Marcus Cyron, Lesless, Atsirlin, Holder, Don-kun, Túrelio, Bubo bubo, Ca$e, Millbart, Bernhard Wallisch, Sänger, A.Savin, Uranus95, Fano, Habitator terrae, DaizY, Luke081515, Neozoon, Hogü-456, Jo-Jo Eumerus, Niki.L, CennoxX, Vermont, Tgr, and Martinvl:. If I've missed anyone, please, please ping them. Nosebagbear (talk) 12:57, 25 February 2022 (UTC)[reply]
  • The current absence was highlighted by @Gnom: in one of the roundtables, and I'm currently working to arrange a dedicated call (rather than looped into one of the others) on it through Stella Ng. Nosebagbear (talk) 09:34, 21 February 2022 (UTC)[reply]
    From my perspective, the main reason for why especially the Wikimedia Foundation and the newly created bodies should, at least in principle, recognise the right of the accused to be heard (while there should be exceptions, without question): We need the volunteer community to have trust in the enforcement decisions of the WMF and the newly created bodies. If they, however, reserve the right to decide without involving the accused at all, we are jeopardising that trust. Gnom (talk) 22:20, 21 February 2022 (UTC)[reply]
  • This is going to need a lot of fine tuning; all too often "right to be heard" becomes/is interpreted as "right to hit back" and while en:Audi alteram partem is a fundamental principle of justice, the latter isn't and instead tends to become a problem in its own right. Jo-Jo Eumerus (talk, contributions) 10:31, 22 February 2022 (UTC)[reply]
    @Jo-Jo Eumerus: What do you mean by "right to hit back"? As far as I know the right to "Widerklage" is accepted. Habitator terrae (talk) 14:24, 25 February 2022 (UTC)[reply]
    "Right to hit back" means when rather than actually presenting a defence of your behaviour, you raise (often non-germane or non-excusing) counter-accusations. These tend to derail the discussion into a back-and-forth or pot-vs-kettle games that make it impossible to resolve the problem. I see no evidence that this right is actually widely accepted. Jo-Jo Eumerus (talk, contributions) 16:00, 25 February 2022 (UTC)[reply]
    @Jo-Jo Eumerus: At least it is accepted (and a right) in the German legal system (to ensure the Grundsatz der Waffengleichheit, in Englisch words, a fair proceeding), I don't see anything other for other legal systems, because interwikilinks exists. Habitator terrae (talk) 17:01, 25 February 2022 (UTC)[reply]
  • RTBH is a conditio sine qua non, en:star chambers are something for dictators, not for an open, transparent and inclusive community. I was (and still am) on the receiving end of a star chamber procedure in MediaWiki, so I know what I'm talking about.
    Yes, there are certain circumstances, where more privacy may be necessary, especially in regard of doxxing and such, but that should always be very well reasoned and be an exception. But even there the RTBH should only be restricted for clear cut trolls, others should have some kind of anonymised kind of RTBH, so not to jeopardise the possible victim of doxxing or mobbing. Grüße vom Sänger ♫(Reden) 13:07, 25 February 2022 (UTC)[reply]
    I fully agree with Sänger and Gnom. Don-kun (talk) 16:18, 25 February 2022 (UTC)[reply]
  • There is a paradox in demanding that there be exceptions to the right to be heard. Who determines whether there has been sock-puppeting or whether there is a legal threat? In the latter case, unless the administrator concerned has legal experience are they competent to make the call? In the case of sock-puppeting, why is the blocked editor not allowed to put forward a defence? The alleged sock puppet might have been launched by a malicious editor who lied in order to get the blocked editor blocked and who is now launching a sock-puppet to further discredit the blocked editor.
As a real-life example, what happens if the police arrest somebody in relation to a crime. In most civilised societies the police are obliged to lay a format charge against the person that they arrested within a specified time or bring them in front of a lower court in order to hold them while an investigation is on-going. If sock-puppeting and legal threats are sufficiently serious, then some process must exist whereby the accused editor is prevented from editing other than answering the case against him. The process must be such that if nothing is done, then the block will automatically lapse.
Finally, if the administrators fail to act against a malicious editor and the victim resorts to going to law, is this not a failure on the part of the Wikipedia project concerned? Rather than place sanctions on the litigant for going to law, any administrators who were involved should hang their heads in shame. Martinvl (talk) 21:34, 25 February 2022 (UTC)[reply]
Yes, that is a paradox, and that is a real problem. That's why trust is paramount.
That's why people, who were involved in extreme anti-wikimedian activity, like implementation of SuperProtect or such, must not have any say in such procedures.
That's why those, who are involved in in enforcing the UCoC must have wide backing in the community (backing with WMF staff is nice but not necessary), so elections of some kind have to be implemented.
I can't normally be, that the accused is not heard at all (nazis and such ilk the clear exception, they gave up voluntarily on such rights), but it can't as well be (imho even less), that victims of doxxing and mobbing can be abused by procedural stuff.
It's sometimes a tightrope, I remember some case from an off-line meeting, where two people with some special and in the concrete situation antagonistic mentalities clashed, and only one was heard, the other was reprimanded without any chance to be heard. I probably could search for this, it was made public here somewhere and afaik an wikimedia-I around that time, but will not do so now. Grüße vom Sänger ♫(Reden) 09:33, 26 February 2022 (UTC)[reply]
As you can see from my proposal, @Martinvl: the exceptions of being heard at all from me are absolutely miniscule - the broader grouping is those who should be heard prior to being sanctioned (or, more accurately, given an opportunity for thus). In lots of those cases the ongoing threat is significant, and if we waited until they'd been heard we'd invariably get far more issues. I'm not sure I have ever seen someone go to the courts for failure to act on a malicious editor - the legal threats I've seen are almost invariably because someone didn't like some content and decided to threaten lawsuits over them. I'm a strong supporter of DOLT, and in rare cases I've seen a legitimate libel claim, dealt with it, and as soon as the editor confirms that they're satisfied and retract the legal threat, have unblocked. Nosebagbear (talk) 12:54, 27 February 2022 (UTC)[reply]
@Nosebagbear: I appreciate your comments about DOLT. However there is more to the situation of being denied the right to be heard that the threat of legal action. I would like your view to the question "Is it appropriate for an editor who is blocked to ignore Wikipedia rules to counter what he honestly believes to be a potential and imminent risk to human life?" I would also like comments from @JBW: and @Nick-D:. Once I have your reaction, I will explain the full details of the situation to which I am referring and I think that you will see why the right to be heard should never be over-ridden. Martinvl (talk) 22:12, 7 March 2022 (UTC)[reply]
I would probably be inclined to say that such an inherently flawed hypothetical should act if, and only if, there was not another way to resolve such things, and such a belief was reasonable.
Putting aside more nuanced off-wiki options that it would not be reasonable to assume awareness of (e.g. asking on the Discord), or slower options like asking on a user talk page, if someone can make a genuine case for imminent risk to human life than the emergency contact is readily findable by the briefest of search engine checks and from my own contacts to them, is indeed a quick process. Nosebagbear (talk) 22:35, 7 March 2022 (UTC)[reply]
Martinvl, It is not reasonable to try to give an answer to a question of that kind posed without context. It is a loaded question, phrased in such a way as to try to make only one answer seem reasonable. It is open to reading in such a way that "no" means "no, it is never appropriate", so as to make it seem that anyone not answering "yes" is being unreasonable, but it is also open to reading as meaning that "yes" means "yes, that is always appropriate", so that anyone answering "yes" is committing themselves to supporting an absurdly over-broad statement. Attempting to produce some all-embracing answer covering every possibility would be futile. Why not "explain the full details of the situation to which [you are] referring" before I give you an answer, rather than this absurd attempt to get me to commit myself to a position without knowing what that position is? I can't conceive of any good reason why you should wish to hide from me what it is you are asking me about, rather than fully disclosing what it is, so that I can make a reasonable response, appropriate to the case concerned, rather than making a stab in the dark. Of all the countless editors and administrators on all the Wikimedia projects who have so far taken no part in this discussion, you haven't chosen me at random to ping; you have asked the question with a view to relating it to some incident or other concerning me in some way. What constructive purpose do you intend to achieve by getting me to make some generic statement without knowledge of what incident that is, so that I am at risk of giving an answer which doesn't apply to that incident, rather than giving me the full facts, so that I can make an informed and relevant response? In view of what this whole discussion is about, there is considerable irony there. JBW (talk) 14:48, 28 June 2022 (UTC)[reply]
@JBW, Nick-D, and Nosebagbear: During the early stages of the Lindt Cafe siege (Sydney 2014), somebody posted information on Wikipedia that they had gleaned from Facebook. A debate ensued between two administrators as to whether or not that posting should be kept in place. I recalled that the storming of the Iranian Embassy in 1980 was broadcast live on television. I understood that this was due to an oversight on the part of the SAS planning procedures. Had those holding the Iranian Embassy been watching the TV broadcasts, they would have had a few vital seconds to prepare for the SAS raid and possibly murder all the hostages.
I believed that there was a possibility that something like this could have happened in Sydney. In the Sydney siege, police efforts were hampered by intense media coverage, especially from social media. My analysis of the real-life situation situation suggested that Wikipedia should not be giving up-to-the-minute coverage as an accidental posting of details of a rescue operation could jeopardise the operation. This is implicit in Wikipedia's policy of en:WP:NOTNEWS. Since I believed that human life could be at risk, and that removing any risk to human life supercedes Wikipedia rules I intervened (See en:User talk:Martinvl#Australian terrorist incident).
When @Wee Curry Monster: raised a complaint against me, he had not seen the e-mail that I wrote. In that e-mail, I explicitly told the recipient that I was blocked and that the recipient should use his own judgement on how to procede. I don't know what WCM told User:Nick-D, but Nick-D's response had many points that could be debated. In the above paragraphs I believe that I have fully justifed my actions in respect of the Australian terrorist incident. I therefore take grave exception to your comment 'There are no "unfounded allegations". Everything stated in the edit you linked to is unambiguously true.'
Does this explain where I was coming from?
For the record, User:JBW (aka User:JamesBWatson) removed my access to e-mail less than 24 hours after WCM made a complaint even though the English Wikipedia requires at least 24 hours. Moreover, I had made clear arrangement for my case to be heard, but JBW ran rough-shod over them. Martinvl (talk) 21:34, 28 June 2022 (UTC)[reply]
I am reluctantly commenting as I was pinged by Martin.
For the record, it would appear that the above explanation is at odds with the time stamps. [2] is the original post made by Martin. The timestamp being 08:45, 16 December 2014 (UTC), in Sydney Australia this would be 16:45, 16 December 2014 (AEST). The Lindt Cafe siege ended at 02:13 local time, so at the time of Martin making his posting the siege had been over for nearly 15 hours. Given the blanket coverage he referred to, it is unlikely that he would not be aware of this. Other editors were already dealing with a possible link between a wikipedia editor and the perpetrator, an admin was already involved and there was no need for Martin's intervention whilst blocked. The material is now redacted so I cannot link to it but review of the talk page archive by a member of the WMF can confirm this narrative. I ask you to draw your own conclusions but it would appear to me that this over dramatic scenario was nothing more that an editor known for extreme wikilawyering conjuring up a scenario where he was portrayed as the victim of an over-zealous admin.[reply]
Further, this was not an isolated example of Martin using his talk page to influence other editors. Eg [3] where he had used the email functionality to contact other editors. He had been repeatedly warned not to do so, one of the occasions he chose to do so was to contact myself via email making a number of gratuitous and offensive references to my mental health (I am a PTSD sufferer).
Martin was originally topic banned and then blocked on the English Wikipedia because he is an extreme wikilawyer, making him a disruptive presence on the platform. To quote one editor he displayed "tendentious editing, lack of collegiality and collaborative spirit, battleground attitude, IDHT and extreme Wikilawyering".
Relevant discussions Removal of talk page & email access, 2nd Topic Ban appeal,First Topic Ban appeal,Original ANI discussion. There are multiple further examples at ANI, AN.
There are several topic ban appeals and unblock requests on his talk page [4]. He has been blocked from access to UTRS for spamming the system. He is also blocked permanently on simple.wiki for exactly the same disruptive behaviour [5]. The continuing theme in all of these is a complete lack of any appreciation for the disruptive nature of his editing that lead to his block, instead he is persisting with the theme that he was wronged.
I would not think it an exaggeration to suggest that an editor nursing a grievance for nearly a decade and one who exploits any chance to air it, is precisely the sort of editor the proposal under discussion here was designed to protect the community from. Wee Curry Monster (talk) 08:56, 29 June 2022 (UTC)[reply]
@Wee Curry Monster: When I contacted User:Nick-D regarding your PTSD, you had made some wild comments about me (I do not recall what they were). A day or two earlier, you had wanted to shut your account down (or something of that order). Whatever it was, it suggested to me that you were [mentally] in a "bad place". I suspected that the comments you made were the result of your mental state at the time. I wanted to ask you quietly to reconsider the statement, but to do so without passing on my e-mail address to you or to publicise my suspicions about your mental state. The most discreet way that I could think of was to do so via your mentor Nick-D. Unfortunately Nick-D failed to assume good faith and also to understand that people suffering from PTSD can be suicidal and I did not want to tip you over the edge. (I don't know how serious your condition is, but I saw how it affected my father who spent several years in a Nazi labour camp). Will you please for a moment assume good faith and accept that whatever our differences might, I did not want to exploit your condition.
There were gross irregularities in all of the incidents that you listed, many of which involved you. However this is not the forum to discuss the details of these irregularities – I have however discussed many of them at a higher level in this forum to try to ensure that they are not repeated. Martinvl (talk) 20:55, 29 June 2022 (UTC)[reply]
I can further demonstrate that the narrative that Martin has tried to portray here is entirely false. Removal of his email access did not follow his comments on the Lindt Cafe siege. It was for an entirely different incident on 4 January 2015. I noticed a message posted by an editor [6] responding to an email request from Martin to change the article en:International Bank Account Number, Martin immediately removed it [7]. I discussed this with my mentor Nick-D [8]. Nick suggested this was prima facie evidence of an attempt to recruit meat puppets with the email functionality and as this was not the first occasion advised I post at en:WP:AN asking for review by an uninvolved admin. I did so [9], it was reviewed and JBW as an uninvolved admin chose to revoke his talk page and email access [10]. His comment There are no "unfounded allegations". Everything stated in the edit you linked to is unambiguously true. relates to that discussion and have nothing to do with the Lindt Cafe siege the previous month. @JBW: acted entirely appropriately, there were no irregularities in his treatment of Martin.
[11] I also went back to look at why I was discussing a wikibreak with my mentor Nick. One of the principal reasons I was taking a wikibreak was to get away from the stress generated by Martin's repeated accusations against me. This link also refers to the content of the email and rather than expressing concern for me, they made a series of unpleasant accusations; so much so they were immediately deleted. I also deleted the rather patronising and unpleasant email he sent me and the cumulative effect of this was I withdrew my support for a proposal to unblock Martin with conditions [12]. You will also note from this link that even his supporters gave up on him frustrated by his conduct.
I just wish to repeat my support for @Nosebagbear:'s commentary that the right to be heard is not practical, wise, or safe in all cases. Martin is one of those cases, as the result of his nearly decade long obsession with me I have wasted many hours of editing time that I would rather have spent on creating content. The community does need protection from editors who would exploit a right to be heard to do damage to the project. Wee Curry Monster (talk) 18:39, 1 July 2022 (UTC)[reply]
Bluntly, this entire subthread is little more than Martinvl, an editor who is indefinitely blocked on en.wiki, demonstrating why he is indefinitely blocked on en.wiki. In December 2019, User:JBW posted a message on Martinvl's en.wiki talk page which expands on this in detail.
Since we're on this, Martinvl appears to believe (or at least to argue) that he was not given the right to be heard in the proceedings that led to his being blocked on en.wiki. This is despite the fact that there was a nine-day long ANI discussion that he engaged in, and despite the fact that he engaged with every other proceeding (and indeed started most of them - including the one where he got blocked). Based on this, if we adopt a right to be heard, it is worth being clear that:
  1. It is sufficient to be given the opportunity to be heard. If the editor chooses not to take that opportunity, or chooses to disrupt the process by which they would be heard, that's not the fault of the community or any other editor.
  2. A right to be heard is not a right to be agreed with. In that ANI, Martinvl refused to engage with the substantive complaints about his behaviour, instead focussing on procedural questions that were quickly dismissed by the community. The fact that they were dismissed does not mean that he was not heard, it means that his position was not accepted.
Kahastok (talk) 22:15, 1 July 2022 (UTC)[reply]
@Kahastok:
  • Firstly, in my view JBW's posting in December 2019 was in extremely bad taste. The full background is that I spotted a note written five years previously on a Wikipedian's talk page to the effect he had died (presumably written by a relative). There was no mention of his death on en:WP:RIP. I felt it appropriate that this person be commemorated in the normal way and that his account be tidied up, so I offered to provide the approriate verifiable information from Ancestry.com to enable this to be done. The deceased Wikipedian had supplied enough personal information on his home page that it was possible to confirm that the document in question was in fact about him.
  • Secondly, may I draw to attention that Wikipedia adheres to the law of the State of California and on reading the case Pinsker v. Pacific Coast Society of Orthodontists , 12 Cal.3d 541 (https://scocal.stanford.edu/opinion/pinsker-v-pacific-coast-society-orthodontists-27828), I believe the process followed at ANI does not follow the requirements of due process. Briefly, in this case, the judge said that the US constitution guarantees "Freedom of association" which means that golf clubs, which exist soley for the benefit of its members, due process need not be followed. However, organisations that rpovide a public service such as proffesional bodies are required to follow due process when reslving disputes between its members. I believe that Wikipedia falls into the latter category. If this is the case, the right to due process is a legal requirement which negates everythign that you heard.
Martinvl (talk) 23:08, 1 July 2022 (UTC)[reply]
Thank you for another fine demonstration of why you were blocked on en.wiki.
In case anyone's interested, Martinvl lost the last person on en.wiki willing to give him the benefit of the doubt on 16 November 2014, largely because of this kind of BS. Kahastok (talk) 09:47, 2 July 2022 (UTC)[reply]
  • Thank you for pinging, Nosebagbear. I appreciate very much your efforts towards a consensus. I agree that 'right to be heard' does not mean 'right to harm' or 'right to keep up the conflict'. But i do not concur with a list of codified exceptions to something which I would consider a fundamental part of online cooperations. I'd like to put it this way: A User prone to enforcements should be informed by the enforcing body, and given the opportunity to depict their views. They should be allowed to react without having to use a formal appeal process, and without a-priori exceptions. If they decide to write a statement, someone should read it, and take it into account (or reject it). No more, no less. Our job is to provide a friendly space for everyone, and thes means to expel a small fraction of users. But we are not law enforcement, and we are not entitled to judge or punish people. Keeping up communication is pivotal. --MBq (talk) 09:45, 28 February 2022 (UTC)[reply]
    @MBq:, I'm not sure what "prone" means in this context - logically it would suggest more those recurring troublemakers (likely the low-grade type, as we don't normally permit recurring high-grade problem editors), rather than specifically those being considered for sanctions. To me, the evidentiary aspect of RTBH is even more crucial than the "heard prior to a sanction" aspect. That needs a couple of minor write-outs to avoid people opposing it on very narrow grounds. I'm open to a more general phrasing for the "talk before action" aspects. Since the WMF have decided that no changes will be occurring prior to the vote, I will be staunchly opposing in order to get the committee both more time and to ensure they commit more time to this aspect. Though if we're reworking, there are other bits that could be improved Nosebagbear (talk) 10:47, 28 February 2022 (UTC)[reply]
    "prone" as in having a tendency or disposition towards being the subject of enforcement; liable. For example, I am prone to UCoC enforcement because I grew up in a culture where we frequently use casual expressions of endearment that are considered extremely rude by most North Americans. My casual exchanges with a member of the same group could easily be misconstrued as hostile even though they are the exact opposite in the right context. Vexations (talk) 17:42, 28 February 2022 (UTC)[reply]
    So that is roughly what I'd mooted - and I'm not sure what fraction of instances of conduct cases complex enough that en-wiki would use ANI rather than AIV (etc) are committed by those "prone". I'd have suspected that a majority would be by non-prone individuals, and we'd still want them talked to. Nosebagbear (talk) 19:38, 28 February 2022 (UTC)[reply]
  • Thanks Nosebagbear. I understand, and frankly disagree with, the current bias towards the right to anonymity and against the right to respond, but it isn't a deal-breaker in terms of my support for enacting some kind of UCOC/U4C. As others have said in this section, thank you for your ongoing effort to try to find a consensus, particularly by trying to enshrine the right to respond in the enforcement procedures. Deryck C. 23:08, 18 April 2022 (UTC)[reply]

Double "Staff"[edit]

The text currently says: "All Wikimedia Foundation staff, Board members, Wikimedia affiliate board members, staff and contractors".

I suspect that the word "staff" is not supposed to be written twice :) Amir E. Aharoni (talk) 11:39, 21 February 2022 (UTC)[reply]

Affiliates are a separate thing from the foundation and have their own staff, pretty sure that's what that is meant to indicate. Beeblebrox (talk) 18:52, 21 February 2022 (UTC)[reply]
Oh, OK. Facepalm :)
I missed the "affiliates" in the long comma list.
Perhaps, to avoid such misreadings, it can say "All Wikimedia Foundation staff and Board members, as well as Wikimedia affiliate board members, staff and contractors". --Amir E. Aharoni (talk) 20:50, 21 February 2022 (UTC)[reply]

Translation paragraph[edit]

There's this paragraph now:

The original version of UCoC Enforcement Guidelines is in English. It will be translated into other languages used on Wikimedia projects. In the event of any differences in meaning between the original English version and a translation, the original English version should take precedence and be the one decisions are based on.

Some issues with it:

  • "It will be translated into other languages used on Wikimedia projects." - this is ambiguous. Translated by whom? It may be interpreted as if it's going to be translated by WMF staff to all the languages used on Wikimedia projects.
  • "the original English version should take precedence" - "should take" or "will take"? Or maybe simple "takes"? When writing rules, there should be less ambiguity.
  • "be the one decisions are based on" - It's grammatically correct, but slightly hard to parse. Maybe it should be changed to: "be the one, on which decisions are based". But maybe it's just me :) Amir E. Aharoni (talk) 11:43, 21 February 2022 (UTC)[reply]
Just say "in the event of any conflict of interpretation between the English version of the UCoC and any translation, the most recent English version shall prevail" instead. Vexations (talk) 12:20, 21 February 2022 (UTC)[reply]
  • I agree with the Vexations' version: it is more consistent with usual legal conventions for differences. In addition, the "should take precedence" is fairly problematic. TE(æ)A,ea. (talk) 17:34, 21 February 2022 (UTC)[reply]
I disagree with the clause's content. It means that for example a Spanish-language project would have to understand the English wording to fully enforce the clause, rather than just interpret the Spanish-language clause. I would restrict this close to multi-language projects. --NaBUru38 (talk) 21:31, 21 February 2022 (UTC)[reply]
  • NaBUru38: I think a more acceptable alternative would be to have verified translations into different languages; although, if there were not translations into all languages, that could lead to problems in case of conflicts. However, my and Amir's problems were more with the structure of the clause, rather than with the content. TE(æ)A,ea. (talk) 21:35, 21 February 2022 (UTC)[reply]
    I did most of the translation into Dutch. I did my best to provide a faithful translation. However, I cannot guarantee that the translation accurately represents what the original was meant to say, because the original was so poorly worded that I can't say with any degree of certainty that I know how to interpret it. Neither does anyone else, as far as I can tell. To a degree, it is possible to faithfully translate a problematic text into an equally problematic one, which is what I did. Nobody should rely on a volunteer's translation of the UCoC, or agree to abide by it unless they have also read and understood the original. Vexations (talk) 12:04, 22 February 2022 (UTC)[reply]

"Such as, but not limited to"[edit]

I've been making a few changes to the page, and have come across the phrase "such as" (in "such as, but not limited to"). From experience (with support from Google Search), "including, but not limited to" is a much more common expression. I would have changed the phrasing, but its consistency makes me wonder if this was intentional (or somewhat stylistic), or whether I should just keep the phrasing. TE(æ)A,ea. (talk) 17:38, 21 February 2022 (UTC)[reply]

Remarks from Wikimedia Deutschland[edit]

(link to the German version of these remarks)

Assessment[edit]

Wikimedia Deutschland thanks the volunteers for their work and commitment to the drafting and consultation processes. The new proposal for the implementation guidelines has become much clearer, many misunderstandings, redundancies and inaccuracies have been cleared up and some points from our original feedback have been taken into account. Wikimedia Deutschland still has four main points of criticism:

  1. UCoC Coordinating Committee (U4C) is better explained, but the overloading of this important committee has not been removed.
  2. Protection of personal data vs. transparency: Public reporting is quite common in the Movement but has serious consequences for the protection of those affected and in the case of false accusations.
  3. Rehabilitation in case of false allegations and resocialization of sanctioned persons are insufficiently taken into account.
  4. The right to be heard should apply to cases that are not open to public scrutiny, unless there are good reasons not to do so.

Wikimedia Deutschland recommends that all interested parties participate in the vote in March. A high level of participation in the vote is crucial.

The four points of criticism[edit]

In cross-team consultations as well as coordination with our Board and the Executive Director, we have identified the following four points:

1. U4C - UCoC Coordinating Committee[edit]

The new UCoC Coordinating Committee (U4C) to be established to enforce the UCoC at global level is described more thoroughly and clearly in the new proposal, as are its roles, responsibilities and composition. In addition, it is stated more precisely that in the future, too, responsibility will initially lie with the project community and affiliate structures, and will only be escalated to the U4C if they are unable to handle a case or if there are no such structures. A previously feared "reigning in" of the community as well as WMDE structures is thus left out.

However, the overloading of this important committee has not been eliminated. In WMDE's view, the committee as designed is neither sufficiently well set up nor supported and thus runs the risk of failing as a central body. The committee is described as a volunteer body. However, the preventive and responsive tasks of the U4C require qualified, experienced and paid staff or extensively assisted volunteer members. It must be ensured that the U4C is supported by a well-functioning secretariat and thus becomes workable and functional. This secretariat should then take over the concrete case processing in order to ensure speed and process reliability and to counteract the exhaustion of volunteers. Furthermore, an annual change of members is counterproductive, as functioning cooperation (trust, work structure, division of tasks/delegation to full-time staff, familiarisation with ongoing cases, etc.) is essential for professional work, which is sensitive and critical for those affected and, if necessary, the public and will therefore be under special observation.

WMDE welcomes the decision for a transitional/introduction process and the U4C Building Committee to lead to the establishment of the U4C, and particularly support the pragmatics of having Maggie Dennis (Vice President of Community Resilience and Sustainability at WMF) select the members of this Building Committee through a criteria-led process rather than having the Building Committee itself elected through a lengthy process.

2. Protection of personal data vs. transparency[edit]

There is still a default assumption in the current proposed enforcement guidelines that most cases will be handled and archived publicly. This is currently standard practice in de-WP for on-wiki cases, which are public anyway, the username often does not allow any conclusions to be drawn about the identity of the persons concerned and they are therefore not GDPR-relevant (cf. vandalism report). In rare exceptional cases, however, there have also been non-public requests to the arbitration committee, which then led to criticism in the community.

(Wiki)public wiki-hounding, doxing, defamation and false accusations and thus the problematic handling of personal data are also a problem in de-WP, as is known from the problem analyses in the WMDE project Online Communication Culture, among others. It can be argued that corresponding (wiki)public marking as UCoC-violation/deletion/archiving of attacks and harassment is necessary in order not to let them stand unchallenged for weeks.

Nevertheless, case handling (fact-finding through hearings or research, discussion of interventions and sanctions if necessary) should be confidential and not public, especially to protect those involved (affected as well as people under suspicion). It is often advisable that cases are known only to a small group, all of whom have given a written confidentiality agreement and are trained for this work. In addition, victims must be able to remain anonymous and, for example, report attacks non-publicly for their protection and safety.

It is important to raise awareness and educate communities and staff on the protection of victims and build trust by providing information on reporting processes and sanction decisions, as planned within the framework of the care and safety concept.

3. Rehabilitation and resocialization[edit]

There is still a lack of recognition that the processing of a case does not end with the sanctions. In addition to the further care of those affected, further employment with offenders should be described if necessary and/or consequences for risk assessment and the procedures to be adapted in prevention/responsive work. Ways to rehabilitation, resocialization or the right to be forgotten of wrongly suspected persons are not listed. In WMDE's work, rehabilitation of the wrongly suspected person will be an explicit part of good intervention in the care concept, as this contributes to the acceptance of hotlines/case management structures.

4. Right to be heard[edit]

The first draft did not address the issue of the right of the suspect to be heard. The rights of the person under suspicion were not mentioned at all. The new proposal does not now include this as a standard clause, but does establish a possibility to be heard in a global Wikimedia policy.

WMDE considers this to be a good first step at this point and sufficient for matters on-wiki. In the case of matters that are not open to public scrutiny, the accused should have a right to be heard, unless there are good reasons not to do so.

In Germany, an accused person would have to be informed about the content of the accusation and the name of the reporting person after one month at the latest, if only because the person's data is processed (GDPR). But there are exceptions to this, too, if the reporting person or the threatened person might be endangered by the information to the accused person. If there is sufficient "evidence", sanctions can be imposed in exceptional cases without the threatened person having been heard.

Current studies show that digital violence is becoming an ever-growing problem. Opinion-forming in digital media is becoming more and more polarised and moralised. In the process, those affected often fall by the wayside. The problem is not that false accusations are made, but that there are not enough protections, reporting points and processes for violations of codes of conduct. In the case of digital violence, where sexist, racist or otherwise discriminatory attacks are obviously taking place, it is important to avoid a situation where the guidelines do not also ensure that perpetrators are given a forum for the expected rejection of the accusation and the accused person has to justify this and, in the worst case, the hatred spreads even further.

According to the restorative justice approach, a conscious and explicit hearing of the accused can in certain cases enable a transformation of the conflict through a reparation process. The result of such a pilot project on the platform Reddit shows that a well-moderated hearing can lead to reducing online harassment more efficiently in the long run, by removing the underlying motivations of hate online.

Additional comment on the new proposal[edit]

Alignment of community and affiliate policies with the UCoC.

The current proposal makes it even clearer that the UCoC is a minimum standard that projects and organisations must (and many already do) comply with, but that they can go beyond it. What is not yet so clear is how to get there. In any case, there should be different support services and mechanisms, adapted to the respective needs.

One challenge is that law enforcement on hate on the net is often not yet defined and the legal situation is still thin. Authorities themselves are also not yet fully able to deal with it and are not confident in applying it in the context of the laws that already exist. However, shifting the responsibility to the project and affiliate structures is currently the best option.

The implementation of regional guidelines remains vague at best, but WMDE is optimistic about the situation in German-speaking countries, because a) the communities already have very far-reaching guidelines, responsibilities and implementation methods for dealing with reports and sanctions on-wiki, and b) the German-speaking chapters can provide good support here or bring in external support.--Nicole Ebber (WMDE) (talk) 19:03, 23 February 2022 (UTC)[reply]

My feedback[edit]

After long consideration and local discussion on how it would be implemented on my home Wiki, svwp, I have come to the conclusion I will not support this enforcement Guideline

  • UCoC contains three type of unacceptable behavior, and I see a major difference between 1+2 and 3, and I see these needing different type of enforcement. Content vandalism (3) could be fully referred to to be handled by existing enforcement bodies within the projects. But the first two needs something more, and here something like a "UCOC interpretation specialist" per project would be appropriate. I find that this proposal does not address this as would be needed
  • U4c. This is a strange all new organization. What is needed is a body overseeing the existing project in general with focus on if they are working as should be expected, and where the UCoC enforcement is only part of its responsibility. And I would like to see this reporting to the Global council, be a subcommittee to.

For these reasons I will vote no in the upcoming ratification vote.Yger (talk) 07:45, 24 February 2022 (UTC)[reply]

If the UCoC is so difficult to interpret that it requires a a "UCOC interpretation specialist", then how can we expect a non-specialist, like a a regular editor, to comply with it? We're knowingly imposing rules on people that don't understand these rules. That's unconscionable. Vexations (talk) 12:07, 24 February 2022 (UTC)[reply]
I agree: the UCoC seems subjective and open to interpretation. For example, 2.1 Expected behaviour - Mutual respect calls on all Wikimedians to assume good faith. Article 3, special cases notes "Users engaged in bad faith reporting and persistent unjustified reports risk facing sanctions". How do you report a violation without assuming bad faith? Where is the limit at which you can reasonably stop to assume good faith and be justified in making a report? What can be reported? Would reports be open and public or anonymous? Too many problems. 174.4.229.78 13:13, 9 March 2022 (UTC)[reply]

Clarification requested[edit]

There were several points that I had trouble interpreting. Could someone please answer these questions and change the guideline to eliminate the ambiguities if needed?

  • 3.2 Affirmation of the UCoC among certain groups the existing ones do so within a short time after the ratification of these guidelines, with exception of current advanced rights holders with rights that are not up for renewal who will not have a set timeframe to accomplish these affirmations.: What does "not up for renewal" mean; a. those who'll have the rights for life or b. those whose term will end before "a short time" passes?
  • 3.3 Recommendations for UCoC training for community members Individuals required to acknowledge and adhere to the Universal Code of Conduct will be required to attend training to ensure a common understanding of implementation.: This should probably be "Individuals required to affirm they will acknowledge and adhere", but that aside, will those "who will not have a set timeframe to accomplish these affirmations" also be required to take the training?
  • 3.4 Promoting UCoC awareness Footers on Wikimedia projects and edit confirmation pages for logged-out users: Does this mean footers for all users and edit confirmation pages for logged-out users only?
  • 3.4 Promoting UCoC awareness Prominently displayed at in person events: (It's not clarification that's needed here but) Perhaps it is the UCoC itself and not a link that should be displayed? Also, this is the only item on this list that is a verb phrase rather than a noun phrase.

Thank you.--YTRK (talk) 14:33, 24 February 2022 (UTC)[reply]

@YTRK: Thanks for this ROFL: "Prominently displayed at in person events". Habitator terrae (talk) 19:34, 26 February 2022 (UTC)[reply]
@YTRK: My personal thoughts and not representative of the whole committee:
  • 3.2 The first thing, I think.
  • 3.3 It's ambiguous, and folks have come to conflicting interpretations regarding it.
  • 3.4 It means footers for all users; and separately only for logged-out users, edit confirmation pages
  • 3.4 I think a QR-Code or something similar is the idea? I don't know for sure though. You are right about the grammar thing, but that can't be fixed right now. –MJLTalk 20:27, 26 February 2022 (UTC)[reply]
@MJL Thank you. And apologies for having edited the draft itself.
Regarding 3.2, it seems rather odd that those with a term will have to affirm mid-term while those without will not (not that I think option b is correct). As for 3.4-1, perhaps the footer and the edit confirmation page should be listed separately.
Although I appreciate the work that the draft committee has been doing, I can't help feeling that it is too premature to put to a vote: there's still quite a lot of cosmetic refinement to be done. YTRK (talk) 11:49, 27 February 2022 (UTC)[reply]

Please don't treat sysop the same as staff[edit]

This isn't the understanding of the in Wikipedia (see e.g. de:WP:A#Stellung in der Wikipedia). Habitator terrae (talk) 16:40, 26 February 2022 (UTC)[reply]

This is exactly my point in the above section "Counter proposal: pony up the dough". Beeblebrox (talk) 21:03, 26 February 2022 (UTC)[reply]
Staff must not have any rights over the community, they are just our paid staffers, the community is the boss. Syops/Admins on the other hand are elected by the community to do this, they have proper validation and are vetted by the highest entity in the Wikiverse, the communities. Grüße vom Sänger ♫(Reden) 21:41, 26 February 2022 (UTC)[reply]

Unmarked translations[edit]

This part of the text was not marked for translation. For your information, best regards. --justinianus | talk 08:54, 28 February 2022 (UTC)[reply]

"Enforcement guidelines summary"[edit]

A few things I noticed about the summary.

  • There are four references to a "Enforcement Outline", which I believe are typos of the "Enforcement Guideline".
  • The "Enforcement Outline" mentioned above and "The guidelines" on HOW-points 4&5 probably refers to the same thing, and ought to be written in the same way.
  • HOW-point 4 has a failed link: "[recommendations]".

Please address these.--YTRK (talk) 11:40, 1 March 2022 (UTC)[reply]

Voting for the enforcement guideline now reduces the chances of the UCoC ever getting fixed[edit]

Folks, the UCoC is badly written.

Take the definition of harassment.

This includes: "Disclosure of personal data (Doxing): sharing other contributors' private information, such as name, place of employment, physical or email address without their explicit consent either on the Wikimedia projects or elsewhere, or sharing information concerning their Wikimedia activity outside the projects."

As written, this literally means that Wikimedians will not be allowed to share "information concerning [other contributors'] Wikimedia activity outside the projects". This may not be the intended meaning, but it is the literal meaning – like Fight Club: "The first rule of Fight Club is you do not talk about Fight Club."

What about "place of employment"? There are pages on Wikipedia, in project space and article space, that discuss contributors' place of employment without their consent. Examples:

Wikipedia editors and arbitrators have in the past commented on such cases to the media ... take this seminal article for example, which was directly responsible for the changes to the WMF Terms of Use outlawing undisclosed paid editing:

Almost everything User:Doctree and others told that publication about Wiki-PR falls foul of the letter of UCoC as presently written, doesn't it?

Then there are cases like the ones listed below. From the perspective of the UCoC, as written, all the protagonists in these cases ("David r from Meth Productions", "Qworty", "Wifione") were victims rather than perpetrators:

What about the UCoC's definition of "psychological manipulation"? "Psychological manipulation: Maliciously causing someone to doubt their own perceptions, senses, or understanding with the objective to win an argument or force someone to behave the way you want."

What if someone genuinely and honestly subscribes to fringe beliefs, or is just not competent (think Scots Wikipedia)? They will encounter plenty of volunteers who will try to "cause them to doubt their own perceptions, sense, or understanding with the objective to win an argument" ... and "force" them to stop inserting said fringe beliefs into articles!

As written, the UCoC passage about psychological manipulation can be read to criminalise ordinary debate ... but that is how the Wikipedia sausage is made. It will multiply accusations of "gate-keeping" lobbed against volunteers. There are enough such accusations already, often unjustified; there's no need to provide extra encouragement. It can also be used arbitrarily against anyone who has ever advocated a point of view, or tried to change another editor's mind.

That's before we get to other issues mentioned on this page like the right to be heard ...

So, it's a no from me. Just no. The UCoC is not fit for purpose. Why should anyone vote for enforcing it?

First you fix WHAT you want to enforce, then you vote to enforce it. --Andreas JN466 00:55, 5 March 2022 (UTC)[reply]

@Andreas: Hey there! While I understand you may not agree with the UCoC, a vote against the enforcement guidelines because you have issues with the UCoC is unlikely to produce an outcome where the UCoC can be immediately re-written or changed. For the record, I'm not speaking officially here, but I personally think I'm probably right.
If you want to bring up concerns with the UCoC itself, the best place to do so would be here.
That being said, it's your vote. If your only reservation about these guidelines is the thing its enforcing, then don't let me stop you from expressing that. –MJLTalk 20:32, 6 March 2022 (UTC)[reply]
Hey, MJL, you are quite right in that there will be no immediate change to the UCoC wording. I don't expect that. And of course I and others already did bring up these concerns on the UCoC talk page [13][14] (where they remained unanswered), as well as on the mailing list [15][16], where I was told last year that no wording changes to the UCoC would be entertained until sometime in 2023. So voting for enforcement now would mean that a defective UCoC will be in force for at least a year and realistically much longer, which doesn't make sense. I'd rather we spend that year getting the UCoC right. --Andreas JN466 21:21, 6 March 2022 (UTC)[reply]
We're made plenty of efforts to make improvements to the UCoC, and in my experience that has been especially difficult, tedious and requiring an unreasonable amount of effort to get clarity on the meaning of the text, and unwillingness to make the necessary changes. The committee that produced it has been unresponsive to inquiries, and they appear to be unaccountable for their work. There is plenty wrong with the proposed enforcement as well, but I cannot agree to how we're going to enforce something, when it is so fatally flawed that it can't be enforced at all. Vexations (talk) 22:17, 6 March 2022 (UTC)[reply]
Quite. It's also worth noting that hundreds of Wikimedia staff (WMF and affiliates) are being encouraged to participate in the vote (even if they have never edited any of the wiki projects), and that the threshold to approve the enforcement guidelines is just 50% rather than the customary supermajority. To my mind, this means the vote is stacked in favour of approval from the beginning. I wouldn't be "campaigning" in this way if it were otherwise. --Andreas JN466 01:40, 7 March 2022 (UTC)[reply]
I would be very surprised to see the votes from staff members be larger than the margin of success/opposition of the vote's outcome. I know personally I am going to be encouraging every editor I know to weigh in on the enforcement guidelines. –MJLTalk 04:16, 7 March 2022 (UTC)[reply]
How would you find out? There is no commitment here to report the breakdown of (1) ordinary, non-staff project volunteers (2) staff who do volunteer in the projects (3) staff who do not volunteer in the projects. Anyway, even if the UCoC were perfect the enforcement guidelines are not. The concerns raised by others above in #Training alone would lead me to vote no. Given that neither the document which admins would be required to sign nor the training they'd be required to undergo exist today, you are in effect asking people to write the WMF a blank check – which, given its track record, seems like a very, very bad idea. --Andreas JN466 13:19, 7 March 2022 (UTC)[reply]
https://vote.wikimedia.org/wiki/Special:SecurePoll/list/1341 includes the names of everyone who has voted, at least. Of course, WMF staff accounts can be identified by the (WMF) suffix, but you'd have to look for staff voting with volunteer accounts manually. AntiCompositeNumber (talk) 15:45, 7 March 2022 (UTC)[reply]
@AntiCompositeNumber: This list was visible earlier but has now been hidden. Andreas JN466 06:38, 6 April 2022 (UTC)[reply]
@Andreas: I was more-so referring to your "psychological manipulation" points which I have not seen addressed on the talk page. If you brought it up there, I would respond because I have many thoughts on your points relating to Scots Wikipedia. –MJLTalk 04:16, 7 March 2022 (UTC) [reply]
@MJL: [17] --Andreas JN466 13:55, 7 March 2022 (UTC) [reply]

My feedback about this guidelines[edit]

I'm voting Oppose Oppose this guidelines, by next reasons:

  • Decisions made by the Legal Department or the Trust&Safety team must also should be subject to mandatory appeal.
  • Moreover, BOTH the Legal Department and the Trust&Safety team must, in this respect, should be accountable not to the Wikimedia Foundation, but to the community, represented by an community-elected body (directly, without intermediaries in the form of appointees/stuff).
  • A different position is unacceptable, as it leads to lack of control and actual irresponsibility (as well as to insubordination), making the process unclear and inaccessible for evaluation.
  • The community must have clear feedback on such cases, and this feedback must be of a controlling kind of.

Kaganer (talk) 19:45, 16 March 2022 (UTC)[reply]

Updating discussion links to point to this page[edit]

Most of the links that seem to be to the latest UCoC discussions, including in navboxes and overviews of the process, point to Talk:Universal Code of Conduct/Enforcement draft guidelines review when since last month they should probably point here (or at least point to both).

Can someone who's been working on those pages confirm + help update? Since these are all translated pages, I don't want to make partially-synched pages based on a confusion that might create unnecessary work. –SJ talk  04:21, 17 March 2022 (UTC)[reply]

Support Support redirecting all links to the this talkpage. Kaganer (talk) 08:27, 17 March 2022 (UTC)[reply]
@Xeno (WMF): Can you make that happen? I don't have translation administrator rights. –MJLTalk 15:55, 17 March 2022 (UTC)[reply]
Sj/Kaganer/MJL: I've updated the nav box and added a see also to point here to the top of the former review talk page. Could you let me know where else links are pointing to that talk page that should be updated? Xeno (WMF) (talk) 16:31, 17 March 2022 (UTC)[reply]
@Xeno (WMF): I just updated {{Universal Code of Conduct/Talk}} to also point here as well (while removing the link to Talk:Universal Code of Conduct/Policy text which doesn't seem to be as watched anymore). That just needs to be marked for translation. Face-smile.svgMJLTalk 16:58, 17 March 2022 (UTC)[reply]
Thanks MJL, I've done that. Xeno (WMF) (talk) 17:07, 17 March 2022 (UTC)[reply]

Legitimacy[edit]

I see that only 1750 votes have been cast. For such a big decision, which fundamentally alters the structure of the Wikimedia projects, one would expect a participation at least equal to that of the image filter referendum. At this rate, the outcome of this process will never be legitimate.

I encourage the Wikimedia Foundation board of trustee not to ratify a decision approved by an absolute minority of the Wikimedia projects' active users. Nemo 17:19, 19 March 2022 (UTC)[reply]

Yes, absolutely correct. As the owner, WMF can of course do whatever it wants...
But with a participation of just under 2400 users, this vote is not democratically legitimate at all. I assume that there are significantly more active users in all Wikimedia projects. Even very clearly more...
You can compare this with a referendum in a democratic country. There, a certain quorum must be reached in order to prevent a small minority from deciding what a much larger majority does not want.
Perhaps it would be better if the WMF would simply refrain from creating a bureaucratic monster and simply trust the communities to regulate such things.
Best regards --Udo T. (talk) 09:29, 30 March 2022 (UTC)[reply]
EN.WP, in recent years, has a turnout of at least 1,500 users for it's yearly arbitration committee elections, and that's just one project, so yeah, turnout was quite low. I strongly suspect that a primary reason for that is that the UCoC was initially sold as really only being about the smaller projects that don't have things like arbitration committees,large functionary teams, and hundreds of admins, like EN.WP, DE.WP, and others have, and that it wouldn't really effect the big projects. With these enforcement guidelines that is no longer the case, but it has not been easy to get people to care about it. Beeblebrox (talk) 22:29, 31 March 2022 (UTC)[reply]

Where are you seeing the count of votes? Bluerasberry (talk) 00:15, 6 April 2022 (UTC)[reply]

@Bluerasberry: Universal Code of Conduct/Enforcement guidelines/Voting/Results AntiCompositeNumber (talk) 03:06, 6 April 2022 (UTC)[reply]

Résultats provisoires ?[edit]

Est t'il possible d'avoir au moins un résultat provisoire, même si derrière des vérifications ont lieu ?

Il est quelque peu totalement incompréhensible que la fondation prenne plus de 3 semaines pour annoncer le résultat de la consultation alors qu'il est connu au moins dans les grandes lignes... --Fanchb29 (talk) 06:44, 30 March 2022 (UTC)[reply]

Bonjour @Fanchb29,
SecurePoll ne permet pas d'avoir les résultats provisoires tant que les vérifications ne sont pas terminées. La période de vérification est de 2 semaines et non 3 et cela commençait dès le lendemain de la clôture du vote. MPossoupe (WMF) (talk) 10:18, 30 March 2022 (UTC)[reply]

Copy of ballot?[edit]

Is a copy of the ballot available on wiki anywhere? Bluerasberry (talk) 00:17, 6 April 2022 (UTC)[reply]

Sample ballot
The ballot in English looked like File:SecurePoll interface - UCoC Enforcement Guidelines Ratification vote.png. Later, a language selector was added above. Xeno (WMF) (talk) 00:28, 6 April 2022 (UTC)[reply]
Thanks, I put this image in the report for The Signpost. Bluerasberry (talk) 15:40, 6 April 2022 (UTC)[reply]

Report of results[edit]

On 5 April a representative from the Wikimedia Foundation's Trust and Safety team reported results to the Wikimedia-l mailing list.

Is this the place for on-wiki discussion? Thanks. Bluerasberry (talk) 00:18, 6 April 2022 (UTC)[reply]

Thanks, I've cross-posted the wiki-text version of announcement immediately below. Xeno (WMF) (talk) 00:42, 6 April 2022 (UTC)[reply]
This is what I wanted, thanks. Bluerasberry (talk) 15:41, 6 April 2022 (UTC)[reply]

Results from the Universal Code of Conduct Enforcement guidelines ratification vote[edit]

You can find this message translated into additional languages on Meta-wiki.

Hello all,

We would like to thank the over 2,300 Wikimedians who participated in the recently concluded community vote on the Enforcement Guidelines for the Universal Code of Conduct (UCoC). At this time, the volunteer scrutinizing group has completed the review of the accuracy of the vote and the final results are available on Meta-wiki. A quick summary can be found below:

  • 58.6% Yes, 41.4% No
  • Contributors from 128 home wikis participated in the vote
  • Over thirty languages were supported in the ballot

What this outcome means is that there is enough support for the Board to review the document. It does not mean that the Enforcement Guidelines are automatically complete.

From here, the project team will collate and summarize the comments provided in the voting process, and publish them on Meta-wiki. The Enforcement Guidelines will be submitted to the Board of Trustees for their consideration. The Board will review input given during the vote, and examine whether there are aspects of the Guidelines that need further refinement. If so, these comments, and the input provided through Meta-wiki and other community conversations, will provide a good starting point for revising the Guidelines to meet the needs expressed by communities in the voter's responses.

In the event the Board moves forward with ratification, the UCoC project team will begin supporting specific proposals in the Guidelines. Some of these proposals include working with community members to form the U4C Building Committee, starting consultations on training, and supporting conversations on improving our reporting systems. There is still a lot to be done, but we will be able to move into the next phase of this work.

Many people took part in making sure the policy and the enforcement guidelines work for our communities. We will continue to collaboratively work on the details of the strong proposals outlined in the Guidelines as presented by the Wikimedians who engaged with the project in different ways over the last year.

Once again, we thank everyone who participated in the ratification of the Enforcement Guidelines.

For more information regarding the results, please refer to the Results page.

Regards,

Stella Ng on behalf of the UCoC Project Team

Senior Manager, Trust and Safety Policy

User:SNg (WMF) 00:42, 6 April 2022 (UTC)[reply]

Bonjour User:SNg (WMF),
Il devient compliqué pour des contributeurs comme moi de comprendre ce que la fondation veut faire avec ce code de conduite et les directives.
Déjà le code de conduite n'est pas mis au vote au niveau des plateformes wikimédiennes, empêchant de fait les communautés d'en discuter et de se l'approprier localement.
Mais en plus, les directives, qui sont sensées être l'émanation des souhaits communautaires, qui en temps normal sont validées par les communautés pour être mises en place, ne le sont plus maintenant...
Avec un tel résultat (moins de 60 % de réponses positives), la communauté wikipédienne francophone considère généralement que le résultat est négatif par exemple, et montre qu'il n'y a pas de consensus quand à la rédaction des directives en question.
Là, en faisant intervenir à nouveau le conseil d'administration, vous semblez tenter de donner une légitimité "virtuelle" à ses directives, et cela risque bien de déjà mécontenter les plateformes locales, mais aussi de s’interroger sur la manière des les appliquer.
Pour ma part, je dirais qu'il est peut-être intéressant idéologiquement de s'appuyer sur les pratiques sociales américaines (plus particulièrement autour du siège de la fondation), mais cela ne veut pas pour autant dire que toutes les sociétés "non-américaines" ont les mêmes pratiques. Et il est regrettable qu'il n'en soit pas tenu compte ni dans le code de conduite ni dans ses directives.
Ce code de conduite et les directives rattachées donnent de plus en plus l'impression d'être des documents édités par la fondation pour se donner bonne conscience, qui vont au final créer à terme plus de problèmes qu'être des solutions, et personnellement je le regrette.
Autant la communauté wikipédienne anglophone américaine est peut-être arrivée à un point tel que ces documents représentent bien leur politique interne, autant j'ai pour ma part de plus en plus de difficultés à penser qu'ils représentent les politiques des autres communautés.
Et pour ma part je regrette que la fondation aille sur le terrain de la "morale", car chaque communauté n'a pas les mêmes codes moraux, mais a les mêmes principes concernant l'information et sa diffusion... --Fanchb29 (talk) 06:50, 7 April 2022 (UTC)[reply]
It becomes complicated for contributors like me to understand what the Foundation wants to do with this code of conduct and guidelines. The CoC is already not put to a vote at the level of the Wikimedia platforms, preventing communities from discussing it and making it their own locally. Also these guidelines, which are supposed to be the result of community wishes, which in normal times are approved by the communities in order to be put into practice, are now no longer so... With such a result (less than 60% positive), the French-speaking Wikipedian community generally considers this a lack of consensus. By involving the board of directors again, you seem to be trying to give "virtual" legitimacy to the guidelines; this risks displeasing the local communities, and leads to questions of how to apply them. For my part, I would say it is perhaps interesting ideologically to rely on American social practices (especially around Foundation headquarters), but that does not mean that all "non-Americans" have the same practices. It is regrettable that this is not taken into account either in the CoC or in its guidelines. This code of conduct and the related guidelines give the impression of being documents published by the Foundation to ease their conscience, which will ultimately create more problems than solutions, and personally I regret that... I regret that the foundation is entering the domain of "morality", as each community does not have the same moral codes, but rather the same principles concerning information and its dissemination...
Wenn ich mal der Google-Übersetzung trauen kann, und bei Französisch->Deutsch sollte das schon passen, dann kann ich hier nur zustimmen. Die für solche Sachen erforderliche qualifizierte Mehrheit ist bei weitem nicht erreicht, von Konsens ganz zu schweigen. Der gesamte Text atmet Anglozentrismus, mit viel Bullshit-Bingo und vagem Gelaber, was angesichts solcher Sachen wie FRAMBAN und anderer Totalversagen der "Zentrale" allerdings eher Angst macht denn Trust and Safety suggeriert. Grüße vom Sänger ♫(Reden) 12:25, 7 April 2022 (UTC)[reply]
Ich meine daß, mit der Google-Übersetzung, Sie richtig verstanden haben, was in Französisch gechrieben war. O.Taris (discuter) 09:44, 8 April 2022 (UTC)[reply]
Bonjour@Fanchb29, vous pensez peut-être répondre à @SNg (WMF), mais la modification a en fait été faite par @User:Xeno_(WMF) ici, et non par Stella Ng. Vexations (talk) 14:27, 7 April 2022 (UTC)[reply]
@Vexations : je ne fait que répondre ici au message diffusé à priori au nom de ce compte.
Pour le coup, il est quand même dommage à priori de se retrouver avec un "senior manager Trust en Safety" qui n'a au final pas l'air d'avoir de connaissances particulières dans le domaine du wiki et de sa manière de fonctionner chargé de mettre en place une politique concernant ce domaine...
Mais bon, nous ne sommes pas du tout à la première contradiction près sur wikimédia, en particulier par rapport au fait que le mouvement se veut international mais que la fondation n'arrive que très peu à parler à la communauté autrement qu'en anglais. C'est totalement ridicule niveau inclusion... Et quelque peu idiot quand par la suite elle tente de forcer les communautés à être "encore plus inclusives" elles-même... Fanchb29 (talk) 14:59, 7 April 2022 (UTC)[reply]
PS : au passage, il semble que le fait que ce serait à priori ratifier par la fondation aurait été donné sur la page Universal_Code_of_Conduct/Enforcement_guidelines/Voter_information/fr. Alors comment dire : le 7 mars, à l'ouverture du vote, il n'était pas du tout question de cette manière de ratifier les directives.
Je ferais cela localement sur le wiki que je fréquente, tout le vote serait invalidé immédiatement.
Je ne sais pas si sur wikimédia cela est permis (de changer les modalités de votes et ses conséquences en cours de vote), mais cela me surprends pour ma part grandement... --Fanchb29 (talk) 15:05, 7 April 2022 (UTC)[reply]
Bonjour @Fanchb29,
comme répondu sur le Bistro, je ne vois pas quel changement est intervenu pendant le vote. Dès le 7, il était déjà mentionné que la fondation ferait une ratification en cas de majorité de Oui. Je peux me tromper, mais je ne vois pas dans les diffs un changement de contenu.
Bien à vous,
MPossoupe (WMF) (talk) 16:52, 7 April 2022 (UTC)[reply]
Toute la partie "Un seuil de soutien supérieur à 50% des participants sera nécessaire pour passer à la ratification du Conseil d'administration. À l'heure actuelle, le mouvement n'a pas de pratique unique concernant les processus de vote à suivre (certains processus utilisent quelque chose de plus proche d'une super majorité (⅔), tandis que d'autres utilisent une majorité simple (50% +1), tandis que d'autres évitent complètement un décompte numérique des voix). Pour ce processus, afin de le garder aligné sur la plupart des référendums dans les juridictions du monde réel, un vote à la majorité simple a été choisi." est le 7 marquée comme obsolète et ne représentant pas la version actuellement en vigueur.
Donc déjà aller changer le contenu d'une page d'information en cours de vote, c'est à mon sens très moyen. Sur wikipédia FR, changer le contenu des conditions de vote en cours de vote donne un seul résultat : toute la procédure est annulée.
Mais en plus dans le cas présent, parler d'informations laisse à dire qu'il s'agit de propos facultatifs, pas du tout contraignants. Contrairement à ce qui est indiqué maintenant...
Comme je l'indique sur le bistrot francophone pour le coup, et que je répète ici : si la fondation a l'intention de produire un code et des directives pour pouvoir dire qu'elle l'a fait. Je dirais "tant mieux pour elle", et qu'elle aille en faire la promo si elle le souhaite...
Mais si la fondation a l'intention de provoquer un engouement communautaire de la part des projets (excepté le projet anglophone en ce qui concerne les contributeurs américains et anglais peut-être), il serait quand même un peu temps et même très urgent qu'elle tienne un peu compte dans ses propos; sa manière de communiquer et sa manière de fonctionner des autres communautés et particularités sociales qui ne reflètent pas ce qui peut éventuellement se pratiquer aux USA.
Parce que par exemple, la plupart des communautés ne sont pas là pour se préoccuper de la sexualité (ou non sexualité) des contributeurs, de leur affiliation politique ou religieuse, ou je ne sais quoi d'autres... Même dans certains pays, ce type de considération peut être illégal... Les contributeurs sont juste là pour contribuer... Ce que semble quelque peu oublier la fondation actuellement qui donne l'impression de se préoccuper plus de l'aspect réseau social des projets... Fanchb29 (talk) 17:07, 7 April 2022 (UTC)[reply]
Fanchb29, je ne suis pas si sûr que la Fondation soit tant que cela en phase avec le projet Wikipédia en anglais, en atteste l'histoire du Framban. Il est bien possible que la Fondation soit en décalage avec la plupart des projets qu'elle héberge. O.Taris (discuter) 09:36, 8 April 2022 (UTC)[reply]
Bonjour @Fanchb29,
Le texte n'a pas été changé en coure de vote. C'est la traduction qui n'était sûrement pas à jour d'où la mention obsolète. Mais la version en anglais n'a pas changé. Plusieurs consultations ont eu lieu dans le cadre de ce projet et les commentaires ont été collectés et ont été pris en compte à chaque étape. La ratification a été soumise au vote de la communauté pour décider. Certes, tout n'est pas parfait.
Je serais curieux de savoir les points que vous reprochez au processus
Bien à vous
MPossoupe (WMF) (talk) 15:55, 8 April 2022 (UTC)[reply]
@MPossoupe (WMF) : déjà avoir des modifications de texte en cours de vote.
Tous les contributeurs ne lisent pas l'anglais forcément, surtout de manière fluide...
Changer des mots, qui peuvent inclurent au final un changement de sens au texte, posent pour moi question sur la sincérité du vote dans sa globalité.
Tout ce fait par consensus sur wikimédia par ailleurs. 50 % + 1 vote, ce n'est pas un consensus mais un vote à la majorité simplifiée.
Ce texte représente t'il une communauté "locale" ou toutes les communautés ? Pour ma part, je considère qu'en l'état la question se pose.
Et au final, moi au vu des souvenirs que j'ai et de l'expérience que j'ai des consultations qui ont eu lieu au niveau francophone, je le dit clairement : je suis désolé mais je n'ai pas du tout l'impression que ce qui a été fait représente les consultations et les commentaires qui ont été fait par diverses communautés, bien au contraire même.
Il a par exemple été indiqué dès le départ que plusieurs parties des textes pouvaient posé souci à plusieurs communautés. Ben au final, rien n'a évolué depuis la première présentation du texte du code.
Je ne sais pas ce qui a été discuté au niveau des comités de rédaction, mais pour ma part je ne considère pas pour le coup que leur travail représentent grand chose, vu que même ceux qui ont pu participer à de telles consultations n'ont pas eu ni de la part du comité ni de la part même de la fondation de retours de ces rencontres...
Vous (j'entends par là la fondation et les comités en question) avaient à priori eu des rencontres avec des groupes de contributeurs, mais laissaient franchement l'impression que les groupes ont parlé dans le vide... Fanchb29 (talk) 16:13, 8 April 2022 (UTC)[reply]
...in the end, given the memories I have and the experience I have of the consultations that took place in French, I do not feel that what has been done represents the consultations and comments made by various communities. For example, it was indicated from the start that several parts of the texts could cause concern to several communities. Well, in the end, nothing has changed since the first presentation of the text of the code. I don't know what was discussed at the level of the editorial committees, but even those who were able to participate to such consultations have not gotten any feedback from these meetings from either the committee or from the foundation itself. The foundation and the committees in question had meetings with groups of contributors, but frankly left the impression that those groups spoke in a vacuum.

Minor changes of the text during the voting process[edit]

[18] Taylor 49 (talk) 17:00, 6 April 2022 (UTC)[reply]

Open issues, with proposed changes[edit]

moving to Talk:Universal_Code_of_Conduct#Open_issues,_with_proposed_changes to keep discussions on a single page 20:38, 13 April 2022 (UTC)