Policy talk:Universal Code of Conduct: Difference between revisions

From Wikimedia Foundation Governance Wiki
Content deleted Content added
::Typical. Proclamations just for the sake of publicity, with no regard to substance. ~~~~
Line 386: Line 386:
[[User:Bluerasberry|<span style="background:#cedff2;color:#11e">''' Blue Rasberry '''</span>]][[User talk:Bluerasberry|<span style="background:#cedff2;color:#11e">(talk)</span>]] 00:13, 23 May 2020 (UTC)
[[User:Bluerasberry|<span style="background:#cedff2;color:#11e">''' Blue Rasberry '''</span>]][[User talk:Bluerasberry|<span style="background:#cedff2;color:#11e">(talk)</span>]] 00:13, 23 May 2020 (UTC)
:Not a surprise – the decision was made a long time ago and the 'consultations' were a facade. We'll soon find out what the "close consultation with volunteer contributor communities" means. [[User:EddieHugh|EddieHugh]] ([[User talk:EddieHugh|talk]]) 11:42, 23 May 2020 (UTC)
:Not a surprise – the decision was made a long time ago and the 'consultations' were a facade. We'll soon find out what the "close consultation with volunteer contributor communities" means. [[User:EddieHugh|EddieHugh]] ([[User talk:EddieHugh|talk]]) 11:42, 23 May 2020 (UTC)
::Typical. Proclamations just for the sake of publicity, with no regard to substance. [[User:Nemo_bis|Nemo]] 13:16, 23 May 2020 (UTC)

Revision as of 13:16, 23 May 2020

Redundant with Terms of Use?

How'd this not be redundant with Section 4 of the Terms of Use on refraining from certain activities (and behaviours)? —MarcoAurelio (talk) 20:19, 30 September 2019 (UTC)[reply]

@MarcoAurelio:Thank you for your comment. Section 4 of Wikimedia Terms of Use does cover some behavioural guidelines along with content guidelines such as copyright infringement and paid contributions, however it is not a comprehensive list. It is very open to interpretation and not often applied as a guideline on wiki. The Universal Code of Conduct aims to help communities actually apply the section 4 of the Wikimedia Terms of Use by spelling out in more detail, what is covered by them and more focussed conduct guidelines. --NNair (WMF) (talk) 17:26, 1 October 2019 (UTC)[reply]

Would it not be more efficient to rewrite Section 4 to be more of a "Projects should have rules that enforce [x], support [y], etc..."? Vermont (talk) 17:44, 1 October 2019 (UTC)[reply]
@Vermont: Hi Vermont, interesting point. Any changes to the TOU will have to be approved by the legal team, and we’ll be working with them as this process progresses. A difficulty there is that the section might get lengthy, as language could drift towards legalese. Having a readable and concise TOU is important, so it might be best that the TOU references the UCoC rather than having it live there in full. We’ll let them know about this suggestion, and get their opinions on it.--NNair (WMF) (talk) 06:02, 4 October 2019 (UTC)[reply]
Having a readable and concise UCoC is similarly important. I would hope that your vision for a UCoC is one that is more of a basis for how local conduct policies should look, rather than an attempt at removing all local conduct policies and replacing it with a global, English, UCoC.Vermont (talk) 10:12, 4 October 2019 (UTC)[reply]
^ Very much this. I hope WMF realizes that if they try to force a code of conduct that is written by themselves on the entire community, it will not go well. The community should really be taking the initiative on this matter. --Rschen7754 17:57, 4 October 2019 (UTC)[reply]
This is a good direction I think. Similarly to how the Foundation put out a resolution that all projects should adopt a local policy that upholds basic principles with regard to living persons. GMGtalk 12:30, 6 October 2019 (UTC)[reply]
Template:Rto The contents to be expected in the CoC are in big part already existing conduct policies and guidelines throughout the projects. However these rules are scattered, and hard to find, therefore unknown to many editors. I've spent the greater part of ca. two months to discover the policies and guidelines of enwiki, and I still find new essays and explanations to the rules. A CoC wouldn't impose rules much different from the already existing ones, and there was no mention of any local policies being replaced. The point of an UCoC is to collect in one place the basic conduct guidelines, in a "readable and concise" format, thus old and new users alike can read it and be aware of it. The awareness of these rules will possibly help the communities to call out and address conduct issues even before those escalate to sanctions, and also make us more mindful of our conduct. —Aron Man.🍂 hist🌾 18:53, 26 October 2019 (UTC)[reply]
The TOU were written by, and belong to, the community. Even if changes also require Legal's approval, one can't make substantial changes to it without the community's consent. --Yair rand (talk) 17:08, 6 October 2019 (UTC)[reply]
The BLP framework is something we have been looking at. It is definitely an applicable approach. Similarly, the UCoC is aiming to come up with a basic, minimum standard for behavioural guidelines, and the idea emphasizes the fact that projects with strong guidelines must be able to abide by and uphold their respective local behavioural policies. In terms of process, the Foundation aims to facilitate the process but the whole project exceedingly depends on the feedback and ideas from the community. We’re working on a more detailed outline/timeline for conversations that we’ll post here. --NNair (WMF) (talk) 03:01, 14 October 2019 (UTC)[reply]

Already rejected

If nearly 50 editors randomly come across a proposal to ban the WMF from doing something, and every single one of them supports preventing the WMF from doing that thing, the WMF should probably not do that thing. If you try to push for it anyways, we will have a crisis on our hands, and anyone trying to do any productive work will despair as we yet again need hundreds of volunteers to try to stop the WMF from causing yet another catastrophe. --Yair rand (talk) 17:09, 6 October 2019 (UTC)[reply]

Most of the open-source communities have adapted a Code-of-Conduct in recent years, with success. They had their own, difficult debates. Wikipedia is far behind these communities, in terms of civility, openness, constructive debate, and cooperation. The introduction of a CoC is a step forward, by discouraging unwholesome behaviors, that's so common in our online interactions. It is understandable, however, that some people, who haven't experienced the benefits of a CoC in a project, have fears about the application of new rules. I'm sure, after some time most of these people will find, that being mindful of our conduct, and following a CoC creates a more healthy community, where contributing is a more uplifting experience. — Aron M (talk) 17:32, 26 October 2019 (UTC)[reply]
@Aron M: Could you provide links to a few examples that of such CoCs in other online communities? Thanks, Nsk92 (talk) 17:22, 22 November 2019 (UTC)[reply]
@Nsk92: code of conduct Blue Rasberry (talk) 20:37, 7 February 2020 (UTC)[reply]
Nsk92,Aron Manning https://meta.stackexchange.com/conduct for example. Vexations (talk)

The problem many perceive is not that the community might be healthier with a CoC, but that this initiative is coming from the "T&S team", who (by contract) are required to represent the interests of the WMF above all other considerations. Simply 'managing' the input from experienced volunteers who are actually the community that a CoC will be applied to, is not the same thing as seeing the WMF buy in to limiting their own behaviours as part of agreeing a volunteer driven, volunteer agreed, CoC.

The 100% supported vote linked above was for the board should create "terms of use" that the foundation has to comply with in their dealings with the communities. If WMF employees, contractors and board members were to start there, rather than engaging in an expensive programme of managing stakeholders in order to force a WMF view of the world on everyone else, that would be a super starting point that the community could be reassured about. Amongst other advantages it provides the starting principles for community governance of actions such as the undocumented and un-appeal-able global blocks of long standing unpaid project volunteers by anonymous WMF employees and contractors who have access to the WMF Office account. Not that this is an argument against WMF T&S taking "probably" justifiable actions, just that the community is way, way, overdue for a better governance process of WMF T&S that visibly respects our community shared values of transparency and openness.

Thanks -- (talk) 18:51, 26 October 2019 (UTC)[reply]

The process this pre-consultation is about, would amend existing local rules with a globally unified 'code of conduct' (CoC). When in doubt, the CoC would supersede the local rules. None of the communities actively asked for it. Introduction of the CoC clearly interferes with the decision processes of the local projects. Preventing the WMF from this kind of engagements is at the heart of the 100% supported proposal. By extension, it would effectively prevent the WMF to roll out a universal code of conduct, let alone enforce such a set of rules.
The proposal can indeed be interpreted as a rejection of the CoC initiated by the WMF. Given the unambiguity of the vote I feel any engagement of the WMF against the spirit of the proposal would be a recipe for the next major project crisis.---<(kmk)>- (talk) 02:14, 29 December 2019 (UTC)[reply]

Wikimedia LGBT+ meeting minutes October 2019

Team LGBT+ discussed this proposal in its October meeting. The Wiki LGBT+ community has discussed conduct and misconduct in almost all of its meetings over the past few years and considers itself as a major stakeholder in any broad-reaching conduct proposal. Blue Rasberry (talk) 20:06, 8 October 2019 (UTC)[reply]

Hi User:Bluerasberry, thanks for showing up and letting us know about your discussions! We are aware that such discussions are happening in different places and are hoping to gather the input from them in the upcoming consultation about the Universal Code of Conduct. Actually we kickstarted some of those conversations with our session at Wikimania and we are trying start more of them. So we are definitely looking forward to hearing about your ideas and experiences! --CSteigenberger (WMF) (talk) 09:36, 11 October 2019 (UTC)[reply]
Are we having a session about this in Boston next month? GMGtalk 12:12, 13 October 2019 (UTC)[reply]
Hello @GreenMeansGo: @Bluerasberry:I'll be at WikiConference North America next month and will be available to have one on one meetings, and also to participate in an unconference session, too, about the UCoC. SPoore (WMF) Strategist, Community health initiative (talk) 16:31, 21 October 2019 (UTC)[reply]
@SPoore (WMF): I will be at that event and see you then and there. I will take notes and report back to Wiki LGBT+. Blue Rasberry (talk) 17:12, 21 October 2019 (UTC)[reply]

What on earth

has the preamble of Constitution of India got to do with Universal values of Wikimedia communities? Winged Blades of Godric (talk) 14:41, 18 November 2019 (UTC)[reply]

Hi both, thanks for the feedback on the collage; we can look at some of the images for improvement.
@Winged Blades of Godric: The constitution of a democratic country represents people’s voices, people’s representation, and togetherness in diversity. It’s one thing that binds every citizen of a country together irrespective of their differences and cultural values. This aptly resonates with the concept of unity and universal values in our movement. Then why not include a picture of the preamble of the constitution of not only the biggest democracy but also the most culturally diverse country in the world to signify those values? With this picture, I was definitely not focussing on the ‘socialist’ part, but perhaps this one can replace the one that’s already there.
@Risker: I hear your concerns about the history of the Chiang Kai-shek square. I chose it because of how this square holds a sentimental significance for Taiwan as it is now the place where democracy is celebrated, rather than Chiang Kai-shek himself. But I can see how it could be read as divisive in terms of Chinese history.
If you have suggestions for new images, let me know. Best,--NNair (WMF) (talk) 04:57, 20 November 2019 (UTC)[reply]
See this map. No, India is not culturally tolerant. In general, I don't really see the point of promotional images for a UCoC, but if you really want one please use uncontroversial and globally relatable images. Vermont (talk) 10:15, 20 November 2019 (UTC)[reply]
I agree with WBG, Risker and Vermont, and I'm concerned with the direction this is taking. —MarcoAurelio (talk) 12:21, 20 November 2019 (UTC)[reply]
I agree with WBG, Risker, Vermont and MarcoAurelio. To avoid repetition, my lengthy thoughts on the crass insensitivity of the particular images chosen, the impracticality of trying to illustrate "universal values" with any kind of photograph, and the impossibility of a meaningful code of conduct without enforcing cultural imperialism of some kind, are on my en-wiki talk page. Iridescent (talk) 13:19, 20 November 2019 (UTC)[reply]
Hi all, thank you for the inputs about the image. Please do note that the picture in question here is not the final representation of UCoC. And as I said earlier, we can look at more images for improvement. For that do share your suggestions about other pictures that in your view, better represent the values of the movement and are least controversial. I am looking forward to formulating something that’s widely accepted. --NNair (WMF) (talk) 07:13, 21 November 2019 (UTC)[reply]
NNair, are you Chickenwings10? Vermont (talk) 10:46, 21 November 2019 (UTC)[reply]
It's in the shape of a cross. There's really no image or set of images that accurately represent world values without offending someone, because there isn't a set of values shared globally that I think would accurately reflect the UCoC. Vermont (talk) 20:15, 22 November 2019 (UTC)[reply]
I mean, it is (or at least has become) the quintessential secular ethical code. That this manuscript is shaped like a cross is historical accident. GMGtalk 21:28, 22 November 2019 (UTC)[reply]
The cross shape may be an accident, our choosing it would not be, nor could we expect it to be perceived as an accident. WereSpielChequers (talk) 21:50, 22 November 2019 (UTC)[reply]
We're getting somewhat off-topic, but the cross shape is definitely not an accident but an intentional attempt to express continuity between Ancient Greek culture and Byzantine Orthodox Christianity. If we (a) need a CoC and (b) need an image to illustrate (both of which I—along with as far as I can tell virtually everyone not on the WMF payroll—strongly dispute), than I agree that it's a total no-brainer that religious imagery, national symbols, anything pertaining to political issues, or anything that a reasonable person could possibly interpret as such, are totally unacceptable. It's by design, not accident, that the logo of every WMF project is utterly bland.Iridescent (talk) 23:33, 22 November 2019 (UTC)[reply]
If you want to allude to the Hippocratic Oath, there are other images available, such as this one. -- Llywrch (talk) 23:19, 22 November 2019 (UTC)[reply]
But again, we're back to subconscious cultural bias. The entire concept of medical oaths, let alone the Hippocratic Oath in particular, is specific to particular cultures; in much of Africa, for instance, the entire concept is unknown, and even in the West there are relatively few jurisdictions where any form of oath is mandatory rather than voluntary. (Quick RS.) Finding common cultural reference points which have the same meaning to all readers on a global project really is harder than it looks.Iridescent (talk) 23:42, 22 November 2019 (UTC)[reply]
I mean, as far as I am aware, the Hippocratic Oath has been endorsed and updated by the World Health Organization. But regardless, all this is probably just a distraction, and having any image at all is probably not remotely worth this level of debate. Something something en:BIKE. GMGtalk 13:03, 23 November 2019 (UTC)[reply]
Getting severely off the point, but I think you're misremembering. The WHO is a UN project and would never try to enforce a US-specific concept like medical oaths, as it would provoke a global scandal. The (unrelated) World Medical Association, an international trade body for medical professionals, regularly suggests proposed codes of ethics (I assume the Declaration of Geneva is what you're thinking of), but none of them have any particularly wide acceptance. Even in the US, only about 75% of doctors take any kind of oath; if you tried to suggest to medical students at (e.g.) Cambridge University that they should be forced to swear an oath, they'd laugh in your face and politely escort you from the medical school.Iridescent (talk) 15:39, 23 November 2019 (UTC)[reply]
Admittedly completely off topic banter at this point, but ... uhh ... [1] ... And how is a two thousand year old Greek oath a "US-specific concept"? And yes, oaths in general are an archaic concept that are rarely used, but I'm pretty sure we got that bit of tradition you you guys being a former colony an all. GMGtalk 17:43, 24 November 2019 (UTC)[reply]
The difficulties of finding a suitable image may be a distraction, or they may be themselves a symbol of how it's not possible to have a "universal" one-size-fits-all approach to all cultures in the world. Nemo 07:49, 29 November 2019 (UTC)[reply]
The point of questioning the image was manifold: First, it revealed certain biases on the part of the person/group posting it. Second, it illustrated that the same image/representation is perceived very differently between individuals and (likely) cultures. Third, it revealed the irony of trying to apply a uniform interpretation of acceptable behaviour across many different cultures. Finally, it illustrated that, despite the fact that English Wikipedians are constantly berated for behaving badly whenever we dare to step outside of our home project, the entire basis of the proposals and all of the research is based on the supposedly ineffective and unhelpful policies of English Wikipedians. This final point is not an exaggeration. Every international meeting I have attended, I have heard at great length how horrible English Wikipedia is or how nasty English Wikipedians are. And yet, the relevant research is comparing *every other project* to English Wikipedia, and finding *those projects* deficient. I'd say there's some bias here, yes. Risker (talk) 01:36, 4 December 2019 (UTC)[reply]

Universal - total universal

Wenn ich das Wort universal richtig verstehe, dann heißt es sowas wie "allgemein" oder "weltweit". ... Wenn das ernst gemeint ist, dann bin ich ja froh, dass es auch in der universal sprache geschrieben wurde und so von jedem auf dieser Welt verstanden wird. Da braucht man es dann auch nicht übersetzen, versteht ja jeder ... Und da Sarkasmus nicht von jedem verstanden wird nochmal deutlich: Es ist arrogant/überheblich/eine Frechheit für eine angeblich "globales" Bewegung eine verbindliches regelwerk einführen zu wollen und das nur in einer einzigen Sprache anzubieten. ... AGF? Das wäre dann wohl: sorry, aber WMF ist einfach zu dumm oder das "Wiki" im Namen heißt nicht, dass das beauftragen von Übersetzern nicht viele Jahr dauern kann ...Sicherlich Post 13:33, 9 December 2019 (UTC)[reply]

Verstehe ich das richtig, dass Du durchaus für einen solchen globalen Code of Conduct wärest, sofern er in allen Projektsprachen vorhanden wäre?--Schreibvieh (talk) 15:18, 9 December 2019 (UTC)[reply]
Vielmehr ist es so: Wenn etwas, was von "oben" in "wrong language" ankommt, so wird das nicht ernst genommen bzw. fürs Erste mal abgelehnt. Man möchte ja erst mal verstehen, worum es eigentlich überhaupt geht. Geht ja in Unternehmen auch so. Wenn der CEO im DACH-Raum in einem DACH-Unternehmen was in Englisch brabbelt bzw. schreibt, so nehmen die Angestellten das Gebrabbel nicht ernst. Und am Schluss wird es so sein: Der CEO jammert über mangelnde Engagement seiner Untergebenen, die folgen ja seinen Anweisungen nicht. Und tut das in Klartext in Deutsch seinen Angestellten kund. Und die Angestellten so: Hä? Das hören wir jetzt zum ersten Mal. Und rebellieren erst recht. Ist menschlich. Und genau das gilt es zu berücksichtigen.
D.h. das gleich direkt "in Klartext in Deutsch" weitergeben. Also nicht voller Geschwurbel, wie beim Kurier-Artikel. Also so: "Wir wollen kein Autorenschwund mehr. Also neue Regeln. Dies sollen für alle Wikipedias gelten. Beispielsweise: 'Mobbingverbot' oder 'kein Missbrauch der Wikipedia'. Du kannst bei den Regeln mitdiskutieren." (das fasst den Kuriertext einigermassen zusammen). --Filzstift (talk) 15:37, 9 December 2019 (UTC)[reply]
@Schreibvieh: nein, ich habe keine Meinung dazu. Ich bin aber der Meinung, dass wenn man etwas global als verbindlich vorschreiben will, dass es auch in den Sprachen vorliegen muss. Und wenn es eine Diskussion dazu geben soll, dass muss mindestens in den wichtigsten Sprachen vorliegen. Hier liegt es, im typischen WMF-Stil, nur in englisch vor. ...Sicherlich Post 16:09, 9 December 2019 (UTC)[reply]
Da hast Du völlig recht.--Schreibvieh (talk) 12:08, 10 December 2019 (UTC)[reply]

+1 Strongly agree. This complete proposal and consultation process should be made available in at least the most common "world languages". The potential impact of global Wikimedia Codes of Conduct is huge; in case of violations of these codes we will be right back to discussing office actions and user bans. I ask the Foundation to provide access to this discussion to all or most communities by making translations available. --Martina Nolte (talk) 19:23, 10 December 2019 (UTC)[reply]

Bitte denkt daran, dass es sich hier um die Ankündigung eine Projektes handelt, das noch in einer frühen Planungsphase ist. Aktuell kontaktieren wir Communities rund um die Welt in ihren jeweiligen Sprachen, um sie überhaupt darauf aufmerksam zu machen. Wir sind darauf vorbereitet, Kommentare und Fragen in den Sprachen anzunehmen, in denen die Menschen sich wohl fühlen hier zu kommunizieren. Als nächsten Schritt haben wir geplant, gezielt etwa 14 Sprach-Communities zu kontaktieren (Communities, die noch kein ausführliches Regelwerk haben, aber eine relevante Anzahl von aktiven Freiwilligen) und mit ihnen durch per Kurzzeitvertrag angestellte muttersprachliche Facilitators eine vorab Befragung zum Thema durchzuführen. In der Folge wird es eine allgemeine Befragung für alle Wikis geben. Wir haben (auch finanziell abgesichert) im Plan, dass es dabei professionelle Übersetzungen geben wird, insbesondere natürlich für das endgültige Ergebnis. Aber all Einzelschritte in alle Sprachen zu übersetzen übersteigt unsere Möglichkeiten. Jede Hilfe vorhandene Inhalte als Freiwillige zu übersetzen ist deshalb hoch willkommen!
English version (rough translation): Please remember that what you find here is an announcement of a project in a pre planning stage. At the moment we are contacting communities around the world in their languages to make them aware of this. We are ready to take comments and questions in the languages people feel comfortable in communicating. As a next step we will do an outreach to around 14 language communities we have identified as target communities (communities who do not have an extensive set of rules, but have a certain number of active contributors) and do a pre-consultation with them through hired facilitators in their languages. After that there will be a general consultation for all the wikis. We have planned (and budgeted for) some professional translations on the way and certainly for the translation of the final result, but do not have the resources to do translations in all languages for all steps. Helping out by translating content you are able to translate as a volunteer is highly appreciated! --CSteigenberger (WMF) (talk) 10:59, 11 December 2019 (UTC)[reply]
Klar; in der frühen Planungsphase: und in der späten wird dann der finale Entwurf präsentiert wo man ggf. noch die Kommasetzung beeinflussen kann ...
aber gut. Ich gehe von einer "hidden agenda" aus bzw. so hidden ist die gar nicht: WMF will seine Machtposition ggü. den Communitys auf eine Regel stellen. Damit dann Superprotect oder Bans wie gegen Fran frei Hand direkt von WMF aus gemacht werden können. Die Behauptung es wäre da es eine Regelungslücke gäbe wodurch es "zahlreiche Vorfälle" gab die Mobbing ermöglichten ist mit extrem viel AGF wohl Wunschdenken, wahrscheinlicher ist es schlicht eine Lüge: ich habe mehrfach, inklusive ping nachgefragt und einen Belege erbeten. Ich bekam keine Antwort also gehe ich davon aus, dass es schlicht nichts gibt. Wenn es aber keine Fakten bzgl. des Bedarfs gibt ist der Rest auch nur Brot und Spiele fürs Volk. ...Sicherlich Post 14:35, 11 December 2019 (UTC) schade, dass es kein "citation needed" auf de gibt. würde ich direkt im Kurier dransetzen [reply]
Mutmaßung: WMF bezieht sich vielleicht auf die Ergebnisse der Befragung zu Harassment (vor gefühlt zwei Jahren oder so)? Ich zum Beispiel habe damals in der Umfrage auch von Belästigungen, Beleidigungen und Bedrohungen berichtet. Allerdings auch, dass ich trotz Gewaltandrohungen auch von der Foundation nicht geschützt worden war. Die Foundation hatte es vorgezogen, auf mein Hilfegesuch GAR NICHT zu antworten. Seitdem habe ich den Eindruck, dass Office Actions, Superprotect und User-Bans einer politischen Agenda unterliegen, die mit unseren Projektzielen und allgemeinen Anstandsregeln nicht unbedingt irgendetwas gemeinsam haben. --Martina Nolte (talk) 19:04, 11 December 2019 (UTC)[reply]
Bisher ist halt Schweigen im Walde, trotz mehrfacher, ausdrücklicher Nachfrage (mit eigener Zwischenüberschrift und ping; mehr geht on-wiki wohl kaum).
Da kann man nur mutmaßen und das führt, wie man im [Kurier sehen kann, nicht nur bei mir zu der Vermutung, dass WMF hier Superprotect und "Frambans" legitimieren will. Denn genau das könnte man dann ja mit einem, zwangsläufig recht allgemein gefassten Code leicht begründen und zwar sowohl um aktiv zu werden als auch um nicht aktiv zu werden. Das bei WMF-Aktionen irgendwelche persönlichen Befindlichkeiten eine Rolle spielen scheint ja im Framban nicht abwegig (um es mal diplomatisch zu formulieren).
Wenn ich das richtig verstanden habe (man möge mich korrigieren), sind ja auch personell Superprotect, Framban und dieser Code hier ganz eng miteinander verwoben --> Nachtigall, ick hör dir trapsen
Auch lehnt WMF es wohl ab dem unter #First a ToU for the WMF verlinkten zuzustimmen. Die Reaktion geht über ein It's good that you share your thoughts nicht hinaus und diese Formulierung ist wohl die US-Variante und "jaja". Womit Superprotect/Frambans weiter wundbar auf dem Tisch liegen.
... zu der Umfrage (die ich nicht kenne/an die ich mich nicht erinnere) ist dann die Frage; kann man aus den Ergebnissen den Schluss ziehen, dass Regeln fehlen, wie behauptet wird oder vielleicht, dass die Umsetzung mangelhaft ist oder gar, dass die Dinge subtil genug sind um von Regeln kaum wirklich erfasst werden zu können? ...
Bei Androhungen von Gewalt und ähnlichem, sprich juristisch relevantem sehe ich die WMF völlig unabhängig von einem Code in der Verantwortung. Das sie gar nicht reagieren spricht Bände ...
...Sicherlich Post 19:51, 11 December 2019 (UTC)[reply]
@ Martina Nolte. Belästigungen und Beleidigungen habe ich durchaus schon erlebt, Bedrohung Gott sei Dank noch nicht. Ich komme nicht umhin festzustellen, dass es gewiss auch Streithammel gibt die das genau auch suchen. Ich habe meine Lehren daraus gezogen, denn Onlinestreitigkeiten können sich durchaus auf die eigene mentale und physische Gesundheit auswirken. Ein „Code of Conduct“ hört sich erst einmal nicht schlecht an und soweit ich weiß wird zumindest im angelsächsischem Raum, in der realen Welt, ein Verstoß durchaus geahndet. Aber da wären wir schon beim Thema, denn bevor man etwas Neues implementiert sollte man doch auch erst einmal nachsehen, ob das existierende Regelwerk nur Zierde ist und warum es stellenweise nicht durchgesetzt wird. Leider wird man oft mit problematischen Editoren, die auch schon ein gewisses Verhaltensmuster aufweisen, allein gelassen. Das Wort das Du verwendest mag altertümlich klingen aber trifft den Punkt – Anstand. Man darf, muss und sollte teilweise kontrovers diskutieren, doch verstehe ich nicht warum online ein Verhalten oftmals stillschweigend geduldet wird welches im realen Leben einfach vollkommen inakzeptabel ist. Insofern hat das nicht nur mit einer politischen Agenda zu tun sondern schlicht weg auch mit Seilschaften. Meine Erfahrung zumindest. --Catflap08 (talk) 19:27, 2 January 2020 (UTC)[reply]

Zur Erbauung: Universal Tellerwäscher ;)
Zum Inhalt: Dier anglozentristische Laden hat sich noch nie tatsächlich für Diversität interessiert, wenn es ums Eingemachte ging. Obwohl es sogar bezahlte Community Liason Leute gab, die seltsame Sprachen sprachen, haben die nicht etwa als Liason mit den nicht-anglophonen Gemeinschaften so etwas wie das hier übersetzt, was ich für eine Kernaufgabe einer solchen Stelle halten würde, sondern das dann per Massennachricht auf englisch in die ganzen Projekte gespammt. Vor lauter Geld nicht grade gehen können, aber nicht mal die Höflichkeit besitzen, wenigstens die 20-30 häufigsten Sprachen asap zu nehmen. Grüße vom Sänger ♫(Reden) 19:46, 11 December 2019 (UTC)[reply]

" we need to ensure that everyone who would like to share important knowledge on our platforms feels comfortable and welcome to join us", also: "Wir müssen sicherstellen, dass jeder und jede, der oder die wichtiges Wissen auf unseren Plattformen teilen will, sich wohl fühlt und den Eindruck hat, bei uns willlkommen zu sein." How come that I don't feel comfortable and welcome at all on this page? Wie kommts dann, dass ich mich auf dieser Seite nicht wohl und willkommen fühle? Might the language issue play a part in it? Vielleicht spielt das Sprachproblem da eine Rolle? Or else the impression that the decision is already made and the call for actively engaging in the discussion only means that we should help to implement it? Oder vielleicht der Eindruck, dass die Entscheiduhg schon getroffen ist und die Einladung, sich "aktiv in der Diskussion zu engagieren", bloß bedeutet, dass wir helfen sollen, diese Entscheidung umzusetzen?Mautpreller (talk) 11:44, 12 December 2019 (UTC)[reply]

Da hier das Fehlen einer Antwort bemängelt wird, sowie vermutet wird es könne an der Sprache liegen, möchte ich einfach noch einmal auf meine Hinweise etwas weiter oben verweisen, sollten sie in der Menge des Textes untergegangen sein. --CSteigenberger (WMF) (talk) 09:31, 17 December 2019 (UTC)[reply]

Inzwischen ist es immerhin auf deutsch zu finden; dank einer IP. Andere Sprachen: nix. ....
Zitat: "Aktuell kontaktieren wir Communities rund um die Welt in ihren jeweiligen Sprachen, um sie überhaupt darauf aufmerksam zu machen." - toll und dann kommen sie hierher und; ah englisch. Kann ich nicht; nix für mich. ... Das macht natürlich den Prozess einfacher, weil man sich nur mit den üblichen meta-hanseln rumärgern muss und sich auch rumspricht, dass dieses Meta irgendwas englisches ist. ... und wenn dann später wieder ein ankündigung kommt; dann ists auch völlig irrelevant in welchen sprachen es vorliegt. Die Leute sind ausreichen abgeschreckt. .... und das hat IMO auch nichts mit "anfangsphase" o.ä. zu tun. auch bei "fertigen" Sachen werden finden isch übersetzungen nur vereinzeilt und unsystematisch. Gerade mal in meinen beiträgen geguckt: Strategy/Wikimedia movement/2018-20/Working Groups/Diversity/Recommendations/1 - english only. beim Wikimedia Foundation Medium-term plan 2019 siehts besser aus. Aber spanisch: 27%, Französisch & Russisch 4%. Deutsch 100%: dank eines Wikipedianers. Der übergeordnete Wikimedia Foundation Medium-term plan 2019/Annual Plan 2019-2020 mal gar nicht, die anderen Unterpunkte auch so zufällig. ... Zwei Thesen: WMF glaubt nicht, dass jmd. der nicht englsich spricht irgendwas wertvolles beitragen kann oder WMF ist sowieso egal was hier irgendwer versteht oder nicht, denn WMF weiß ja was WMF will. Das reicht. Negativ? ja und bin hier nochmal vorbeigekommen weil cih diesen Beitrag las ...Sicherlich Post 18:42, 25 December 2019 (UTC)[reply]

Falscher Name

Larger communities can use this universal code of conduct to review and polish their existing policies by seeing the bigger picture around the movement. New communities and communities that have only started to develop their own policies can look at the code of conduct for guidance and take it as a baseline until they have developed their own set of rules to build upon it.

So wie sich das anhört müsste das eigentlich nicht "Universeller Verhaltenskodex" sondern "Beispiel eines Verhaltenskodexes" heißen. "Universell" impliziert, dass er universell gelten würde. Habitator terrae (talk) 15:44, 9 December 2019 (UTC)[reply]

Indeed, default or model (as in: A successful example to be copied, with or without modifications) seem much more appropriate terms, if the rationale is indeed filling a regulatory gap in smaller communities. A possible analogy could also be seen in a rather short and basic monastic rule, with possibly quite extensive local consuetudines. Some of the institutions, structures and ways of proceeding that have sprung up on the larger projects could perhaps be replicated here on meta for smaller projects by pooling of their resources. Although I am using religious analogies I do not see a place for an infallible organisation in this context. Those cases potentially requiring coordination with law enforcement and/or those which could end up in court should probably best be handled by a permanent and professional entity though. --HHill (talk) 17:30, 15 December 2019 (UTC)[reply]

Simply no

The Wikimedia movement (does such a movement really exist? in my opinion: no) does not need a binding set of ethical guidelines. On the contrary, an attempt to enforce such a "binding set" will destroy the very essence of the community projects. It is extreme hubris to think that any organization could actually "ensure" (!!) that everyone of good will will feel "comfortable" (!!). I expect the kind of fake participation that is usual in WMF ideas (as in the Strategy Process): WE have decided to implement a universal code of conduct. YOU are permitted to take part in a constructive discussion about the details. However, my constructive argument is simply: forget it. In the best case, this is a useless endeavour. In the worst case, it is a Big Brother idea. Nothing useful can come out of this but much dangerous and hazardous.--Mautpreller (talk) 19:38, 9 December 2019 (UTC)[reply]

+1 We're all adults here and the self-government of my wikicommunity is working. The last thing my community needs is WMF making rules for us, because we can better make them ourselves. This UCoC-idea is totally superflous. Nasiruddin (talk) 08:15, 10 December 2019 (UTC)[reply]

I am really baffled by one section of the "Community Insights" Report, i.e. the devastating answers to the questions "I am consulted sufficiently by the Wikimedia Foundation" and "I feel that my voice is heard in Wikimedia Foundation decisions". T&S seems to completely ignore this (as was the case in the whole Strategy Process). Their therapy seems to be "more of the same".Mautpreller (talk) 10:31, 10 December 2019 (UTC)[reply]

One more remark: This text is in all essential features written by me. However, it is not a code of conduct (although it was integrated in the T&S assessment of such codes). It aims to convince people of a way how to criticize articles in a review. It does not prescribe rules, ethical values or confessions and is definitely not binding. It is a free offer, not a set of rules. The very idea would be destroyed if one tried to transform it into a "code". If you really want to enable more and better participation as you say again and again, you should think about such offers. It is definitely counterproductive to look for ways how one can deny access to users, which seems to be the only idea of the Universal Code of Conduct project. This can only cause harm and should be quickly abandoned.Mautpreller (talk) 12:45, 10 December 2019 (UTC)[reply]

Allowing diversity

Wikimedia truly is a global player. Our communities span around the globe. Diversity does not only exist within these communities but also between communities. Communication styles vary big times between countries and continents. At Commons, one of our global projects, it is helpful and expected that administrators consider a user's nationality when assessing and judging their communication. For example, German and Dutch folks (no offense, I am German with some Dutch roots myself), are known for a very blunt and candid communication style which can be perceived as rude or even as aggressive and offensive by - let's say - an American or Australian reader. Wikimedia Commons seems to have found strategies to work with the diversity of their users. Any global Wikimedia code of conduct, if really necessary at all, should allow wiggle room for such international and intercultural diversity. Please do not try to apply a cookie cutter globally. --Martina Nolte (talk) 19:56, 10 December 2019 (UTC)[reply]

enforcement?

A "Code of Conduct" - in addition to Terms of Services and Wikiquette guidelines - will only be as valuable as enforcement processes, and enforcing authorities come with it. Nothing has been said in the introduction about this aspect. Are we talking about office actions and global/partial user bans again, now based on violations of a global code of conduct? Recently, the consultation on user bans (after "Framgate") showed a very clear result: Communities highly appreciate self-management, including "local" conflict resolution and handling of disruptive community members. Wikimedia stated something like "we heard you, loud and clear". But based on the introductionary statements on codes of conduct, I am afraid that Wikimedia is simply trying to give a new name to an old (already dismissed) approach. Please don't. --Martina Nolte (talk) 19:56, 10 December 2019 (UTC)[reply]

IMO: thats what the Code is about. To give WMF the excuse to use Superprotect or ban people as they wish. ...Sicherlich Post 14:38, 11 December 2019 (UTC)[reply]
Hello @Martina Nolte: and @Sicherlich:. The UCoC project is not an effort to undermine community enforcement of policy, or to replace it with Foundation staff. It aims to support community practices by consolidating basic values around behaviour across the whole movement, especially beneficial to smaller language projects that have not written specific conduct policies yet. Local, volunteer enforcement is the model it is working with. Patrick Earley (WMF) (talk) 19:38, 13 December 2019 (UTC)[reply]
Die Botschaft hör' ich wohl, allein mir fehlt der Glaube. (I hear the message well but lack faiths constant trust) ... sorry, but it sounds to me like "Niemand hat die Absicht eine Mauer zu errichten" (Walter Ulbricht, some weeks bevor the berlin wall was build) ...Sicherlich Post 20:05, 13 December 2019 (UTC)[reply]
@User:PEarley (WMF) for your understanding I repeat the reasoning I wrote before (somewhere): I remember two harsh steps against the community by WMF: Superprotect and Framban. It took a lot of effort and kind of "outcry" by the community to get it corrected. In both cases Jan Eissfeldt was involved. Now a Team which is (if I get it right) lead by him is starting a Code of Conduct but that has nothing to do with WMF seeking ways to get a legitimation for using its powers? ... Well, thats hard to believe to me. ...Sicherlich Post 22:41, 13 December 2019 (UTC)[reply]
Trust is always a voluntary thing. You can't demand it, it is there or not there. I, for my part, don't trust Trust & Safety. Not least because they have not proved to be trustworthy in the past. You won't change this if you don't do anything to earn trust. This idea is not suitable to do this.Mautpreller (talk) 11:28, 14 December 2019 (UTC)[reply]
@User:PEarley (WMF): Could you please explain in more detail how the implementation of a global Code of Conduct would look like in a local project? Let's say: German Wikipedia has their set of rules comparable to w:Wikipedia:Wikiquette and w:Wikipedia:No personal attacks and w:Wikipedia:Assume good faith. In the left bottom navigation you see that a LOT of Wikipedias have local guidelines and rules addressing user conduct. The rules are not exactly the same but the general ideas and goals seem to be alike. Local administrators apply those rules and guidelines in cases of misconduct including insulting, threatening, and otherwise disruptive behavior. Local administrators know their community's "local" communication style (I.e. they can well distinguish between candid, rude, and offensive communication); they also usually know the "historical" background of interpersonal conflicts; and they have a pretty good idea if a disruptive user is using offensive language possibly "by accident" in a single incident, or if they are willingly and repeatedly offending others. Accordingly administrators would implement administrative sanctions (i.e. topic bans, interaction bans with specific users, temporary edit bans, permanent/indefinite edit bans). Blocking policies would regulate appeals against sanctions. Additionally, users can call for mediation, and arbitration; or they can initiate a community vote on a users ban. What I am trying to say: This is typically a pretty complex system with which communities address and manage misconduct and interpersonal conflicts. Now, how would a global "Code of Conduct" play into this? What exactly would be the benefit of a global code of conduct in addition to these local rules and processes? According to your user page you are the Manager, Policy - Trust & Safety. As you know: Policy is binding; it comes with enforcement and sanctions in case of violations. Would the Code of Conduct be a part of the Terms of Use; and thus supersede local policies and processes? Would Foundation staff take Office Actions in cases where a local community is not capable (in the view of Trust&Safety) to handle disruptive users appropriately and effectively? If not why would there be a need for a global policy? You talked about small communities. I agree that there might be a support need. But wouldn't it be much more effective to talk to those communities directly and offer them assistance with developing conduct policies and/or with enforcing those rules? Last but not least question: Is the Foundation willing to include an initial sentence to the Code of Conduct that the policy only applies to those communities that do not have rules and guidelines on user conduct, and/or are lacking enforcement processes, and/or enforcing manpower? --Martina Nolte (talk) 04:08, 15 December 2019 (UTC)[reply]

First a ToU for the WMF

Here was a wonderful suggestion for a ToU, that the WMF should give itself in regard of their relationship to the communities. Unless they stop trying to rule from above and start listening to the communities and behave as the facilitator they are, not the leader, nbody will have any trust in them any longer. The WMF ist one of the problems in terms of conduct, they should Change quickly. Grüße vom Sänger ♫(Reden) 15:00, 11 December 2019 (UTC)[reply]

The WMF does not have ToU for staff that are different from those that apply to everyone else, but there is a Code of Conduct that applies especially to staff and board members. You can read it here. All staff members are bound by their contracts to adhere to this Code of Conduct. --CSteigenberger (WMF) (talk) 09:38, 17 December 2019 (UTC)[reply]
? So? It seems to me the Policy of WMF and the suggested ToU have nothing in common!? I doubt that the intention is that "WMF signs something". Its the content that matters, right? ...Sicherlich Post 18:25, 25 December 2019 (UTC)[reply]
That was a non-answer. We want a ToU for the organisation WMF, the service organisation of the Wikiverse, that gets all legitimacy from the authors of the different projects and has absolutely no legitimacy on its own. They have botched quite a lot in last years, they have acted mote then once in the last years against the communities, and they don't seem to have learned that much from their grave mistakes. They should eat a lot of humble pie and have a lot to apologise for to the communities. That ToU as binding guidelines would be a nice start. There is absolutely nothing in it, that cannot be subscribed by the WMF.
Das war jetzt nur eine völlig unzureichende Nichtantwort. Wir wollen ToU für die Organisation WMF, für die Serviceorganisatuion des Wikiversums, die ihre gesamte Legitimation von den AutorInnen der verschiedenen Projekte bekommt und keinerlei selbständige Legitimation besitzt. Die WMF hat in den letzten Jahren viel Porzellan zerschlagen, sie haben mehrfach gegen die Community gearbeitet, und sie scheinen nichts aus ihren groben und bösen Fehlern gelernt zu haben. Sie sollten ordentlich Kreide fressen und sich endlich glaubhaft bei den Communities entschuldigen. Diese ToU als bindende Richtlinie wäre ein netter Anfang. Es gibt absolut nicht da drin, was nicht von der WMF unterschrieben werden könnte. Grüße vom Sänger ♫(Reden) 15:28, 26 December 2019 (UTC)[reply]
Hallo User:Sänger. Anmerkung: ToU, also "Terms of Use", sind eine vertragliche Vereinbarung zwischen Personen ("Customer") und einem Dienstleister ("Service Provider"). Theoretisch könnte man die Community als Anbieter des Dienstes 'Schreiben einer Enzyklopädie' betrachten. Das knirscht aber an allen Ecken und Enden. Es fängt damit an, dass die (globale) Community rein praktisch nicht wie eine kohärente Entität handeln kann. Auch auf der anderen Seite gibt es begriffliche Reibungen. Die WMF ist eben keine Person sondern eine Organisation. Deswegen halte ich auch für den Vorschlag von Tinz die Bezeichnung "Code of Conduct", also "Verhaltenskodex" für passender. Viele Grüße, -<(kmk)>- (talk) 08:56, 29 December 2019 (UTC)[reply]
Du hast im Prinzip recht, ich habe schlicht die Terminologie von Tinz übernommen. Wichtig ist vor allem, dass die endlich aufhören so zu tun, als seien sie der Boss sondern anfangen mit den tatsächlichen Bossen, den Communities zusammenzuarbeiten. Bisher haben sie schon des öfteren aus reiner Machgier riesige Konflikte vom Zaun gebrochen, und irgendwie kommt es einem angesichts von FRAMBAN nicht so vor, als würden sie ihr Verhalten jemals reflektieren. Grüße vom Sänger ♫(Reden) 15:45, 3 January 2020 (UTC)[reply]
Die WMF ist in der realen Welt verhaftet und somit nicht logischerweise Teil einer digitalen „Kumbaya-Wolke“ oder einer „Community“ die jedes Mitglied der „Community“ anders definiert. Eine ToU muss für alle gelten daraus ergibt sich dann ein ToC der auch für alle gilt. Die Frage eines ToC adressiert aber ein Problem welches gelöst werden muss, wenn nicht, ist der Käs gegessen und Wikipedia über kurz oder lang Geschichte und ein Eintrag in einer anderen Enzyklopädie.--Catflap08 (talk) 21:58, 3 January 2020 (UTC)[reply]

An oxymoron: binding guidelines

What's the realtionship between existing local rules and UCoC?

  • a binding set of ethical guidelines
  • the need for a global set of conduct rules
  • These guidelines
  • The universal code of conduct will apply to all of us - staff and volunteers alike, all around the globe
  • it aims to provide a basic level of norms that everyone working on the projects will be asked to follow and build upon.
  • Larger communities can use this universal code of conduct to review and polish their existing policies by seeing the bigger picture around the movement. New communities and communities that have only started to develop their own policies can look at the code of conduct for guidance and take it as a baseline until they have developed their own set of rules to build upon it.
  • A similar question goes for the enforcement of the guidelines once they are drafted. The Wikimedia movement is run by the communities, therefore they remain the prime authority to roll out the guidelines. As per the current plan, the responsibility of enforcement will be handled by the communities and administrators of various wiki projects. However, thoughts and suggestions on alternative means of enforcement are welcome.

Summury: They're binding guidelines you can follow or not, but you're forced to do so because you have the authority not to do it. --Der-Wir-Ing ("DWI") talk 23:11, 12 December 2019 (UTC)[reply]

There is a creative tension between the aim for a global code and the aim for an inclusive global organisation that is open to everyone regardless of their cultural background. At this stage, where we don't have any specific rules being proposed, it is possible to come up with the sort of self contradictory phrasing that English speaking marketing people are likely to understand as keeping options open, but which is anything from disconcerting to offputting and illogical to people who are trying to translate this into other languages, raising alarm bells for people who have seen previous US based attempts to impose US standards on other cultures, and probably discriminatory against our friends on the Autism spectrum. My hope is that the Foundation has people who can shift this dialogue to something like "We can hopefully all agree that vandalism is wrong, though sometimes you need to understand a language and culture to spot subtle vandalism. In some languages CONVERSING IN ALL CAPS is considered to be shouting and deprecated. But not all languages have upper and lower cases. For communities that have a single case script such as Georgian, SHOUTING is a truly alien concept. So we can probably agree that don't vandalise Wikipedia should be a universal rule, but don't write in all caps might go in the English language code and other languages where people consider it relevant."
My fear is that the Foundation will make the sort of mistakes US based "global" organisations are known for. Like many I have negative experiences both of the WMF and of other US based organisations. Over a decade ago in "real life" I had some US colleagues who were rolling out a US English rude word checker to all Emails going in and out of a global organisation. They thought I was being difficult when I pointed out that my German colleagues were having phone calls from clients whose German language emails were being rejected for including German words and names that were profanities in American English. I hope the WMF won't be quite so crass; But given past insensitivities and mistakes by the WMF, it would be safe to assume that this project will be high risk for the WMF as it plays to its weaknesses rather than its strengths. WereSpielChequers (talk) 10:03, 15 December 2019 (UTC)[reply]
Indeed. It's also contradictory to state that the thing will subject both staffers and volunteers, when in practice it's an unilateral proclamation from the WMF staff towards everyone else. Inevitably, the power structure is unequal and contrary to all our principles.
To make sure that every stakeholder is equally considered, to counter the perception that it's something flowing from WMF staff to everyone else, to avoid conflicts of interest of the staff as well as anglo-centric and USA-centric contents, the WMF should step aside from this process. The stakeholders should be able to select a facilitator with proven experience in facilitating professional codes etc., necessarily from a non-USA and non-English country, and then all Wikimedia groups will have the same standing in discussing with the facilitator and contributing to a text. I'm sure there are such facilitators to be found in places like Geneva, Brussels, São Paulo and others where countless international social organisations operate; I know OECD has some comprehensive procedure to develop such intercultural documents.
Once a text is produced, WMF will not have a final say in its contents, but can endorse it, just as WMF can endorse the universal declaration of human rights or the Geneva conventions. Of course the international law of war is "binding" on everyone, but because it's external to Wikimedia nobody expects a subsidiary of the International Criminal Court to ever be set up in the WMF SF office to put a wikimedian on trial. Nemo 10:42, 15 December 2019 (UTC)[reply]
Thank you for highlighting the inconsistent wording. It is apparent that whoever drafted this either has absolutely no idea what those words mean to local communities, and/or they're intentionally trying to differentiate their writing by using what they percieve as synonyms (rule/guideline, for example) and perhaps unknowingly, what local communities percieve as antonyms. Regardless, it's a huge oversight that shouldn't have made it out of whatever drafting process the WMF employee(s) who wrote this went through. If the WMF is so set on having a set of guidelines/rules/laws/divine ordinances to trump local project autonomy, at least have it defined and explained by someone who knows about the projects from the perspective of those whose autonomy they're writing away. Vermont (talk) 11:51, 15 December 2019 (UTC)[reply]

Which communities have been involved in this discussion?

An invitation and a link to this pre-consultation page had been posted at the community newspaper page at German Wikipedia. Did WMF invite any other community to contribute to this discussion? Where do other communities provide their input and feedback? --Martina Nolte (talk) 04:29, 15 December 2019 (UTC)[reply]

We are still trying to get more translations and will definitely do a drive to invite more people to the conversation after we have hired facilitators for targeted languages early next year. We expect some of the conversations to happen at local projects or even off-wiki (some reactions are reaching us this way already). We will try to bring the results of those conversations back here to Meta, like we already did for the conversations that happened at conferences (another way we are inviting people) here. --CSteigenberger (WMF) (talk) 09:44, 17 December 2019 (UTC)[reply]
The translation service Deepl exists and provides surprisingly good results -- much better than Google-translate. With this AI driven service reasonable translations are just a few mouse clicks away. Yes, they don't provide each and every language in the wikipedia universe. But German, English, French, Spanish, Italian, Dutch, Polish, Portuguese and Russian would get the translation task a long way. I would have expected the WMF to use this service already. ---<(kmk)>- (talk) 09:08, 29 December 2019 (UTC)[reply]

Wikimedia_movement recommendations

@CSteigenberger (WMF): A month went by and it doesn't look like more communities have been involved in this discussion. But in the meantime, the preparations of a new movement strategy paper made it to the next stage and the recommendation still includes a global "Code of Conduct" - despite all the negative reactions you received here. Can you please explain this? What is the purpose of this consultation? In which way did the comments influence the process and decision making? --Martina Nolte (talk) 22:29, 20 January 2020 (UTC)[reply]

OAuth

I won't take part in any discussion that involves OAuth. There is not the slightest reason why I should disclose my email to the WMF in order to talk about a subject. They can contact me via Wikimail, that's enough. A further reason for suspicion and distrust.--Mautpreller (talk) 12:19, 29 December 2019 (UTC)[reply]

Ein vollständiges Impressum für das Projekt wäre hilfreich incl. Fax-Nummer. Wo ist das? --07:44, 3 January 2020 (UTC)

Paradox

If I'm not wrong, a "code of conduct" works on a voluntary base, so nobody is really obliged to follow the rules. But then demanding a "binding set of ethical guidelines" is contradictory.--Sinuhe20 (talk) 13:13, 2 January 2020 (UTC)[reply]

#An oxymoron: binding guidelines --Der-Wir-Ing ("DWI") talk 14:34, 2 January 2020 (UTC)[reply]
A "code of conduct" explains how to do cooperation on a voluntary base. Note: a scientific review of cooperation does not exsist, so no one has a guideline. The Council of Europe was asked to wright a review for its own sakes but has not answered yet. --84.62.141.109 08:50, 6 January 2020 (UTC)[reply]

Community Insights Report Statistic

Per User:Aron Manning's revert of my edit: The source is the same as the source of the 40% statistic, used in that same line. It's Community Insights report (2018), the SS08 chart, the line on policies. In the same way that the the "Quite a bit" and "A lot" statistics were added together to achieve 40%, it makes sense to add "None" and "A little" together to get 48%. The presentation of the 40% rather than the 48% felt misleading, hence why I added the larger statistic, and I'm reverting the revert. 98.113.245.219 22:30, 12 January 2020 (UTC)[reply]

@98.113.245.219: Thanks for the explanation! —Aron Man.🍂 edits🌾 03:56, 13 January 2020 (UTC)[reply]

Draft?

Where is the draft? If there isn't a draft yet, can we please create a page to hold one? EllenCT (talk) 17:11, 21 January 2020 (UTC)[reply]

Hello User:EllenCT At the moment, the discussion is focussed upon collecting feedback on how the community feels about a code of conduct for the movement. Before drafting the code, it is essential to have structured dialogues with diverse communities and take their inputs. We don't want to produce something hastily without conferring with stakeholders in many languages. Therefore, the next step is to launch mini-consultations for a range of language communities on the subject. Contractors to support this work will be onboarded soon. Thus, the question of drafting the CoC will not surface until those consultations with community are completed and their input has been documented, and reviewed.--NNair (WMF) (talk) 06:33, 22 January 2020 (UTC)[reply]
Hello Nnair, thank you for the explanation. Could this be stated on the content page itself? Ziko (talk) 10:22, 22 January 2020 (UTC)[reply]
@NNair (WMF): do you think it is possible to ask the community how they feel about the abstract idea of a movement code of conduct without specifying a concrete example? Isn't anyone likely to be undecided going to say, "it depends on what's in it?" Isn't it better to advance a specific proposal and ask people what they like and don't like about it, than ask them to endorse the idea of a Code without any specific text? Can we start with the mw:Code of Conduct and the wmf:Friendly space policy? EllenCT (talk) 14:16, 22 January 2020 (UTC)[reply]
@EllenCT: Personal opinion: You make a good point that the "abstract idea" is hard to sell to those, who have no experience of a Code of Conduct. As I have that experience from real-life (meatspace) volunteering and open-source projects, the answer to the question of a CoC for me is an obvious "yes and let's start discussing it's contents".
I was baffled by all the drama in reaction to this proposal. I believe the good-faith response would be to support discussing a CoC, not to reject it without investing as little as to search the internet to see what's a CoC, how wide-spread it's use is (100K+ projects for the most common Contributor Covenant) and what its purpose is (more resources.) These should help editors understand what to expect. —Aron Man.🍂 edits🌾 00:10, 23 January 2020 (UTC)[reply]
Sorry, have no good faith left. My experience with any kind of codes of conduct is pretty bad. They are invariably used to get rid of people that are unpopular, that are deemed to be disturbers, destructive, and so on. They do literally nothing to give voice to the underprivileged and oppressed. On the contrary, they constitute a kind of "shadow law" that is not subjected to the normatives of transparency and rule of law. The underprivileged and oppressed are not enabled to take part in a discourse and to voice their own interests and ideas (which obviously would be a very good thing). On the contrary, they are used as a legitimation resource in a power struggle. It's always them who are the first victims of a "code of conduct" because there is no reason why they should be more friendly, positive and nice than other human beings, but they more often lack the resources to hide behind a civil surface. "Protective spaces" for people who are afraid to go out in the open (and may have very good reasons to be afraid) are a very good idea that I support by all means. But you cannot turn the "movement" as a whole into a protective space. This means silencing the voices we need, rather than giving them their own personal voice. We need "politics in the first person", not politics by advocates "for" others. We need an atmosphere where "it is safe to be unpopular" (en:Adlai Stevenson II). What happens here is the exact opposite, and that is very sad in my opinion.Mautpreller (talk) 09:14, 23 January 2020 (UTC)[reply]
In addition: That is a mistake that has been so often made in so many movements. Invoking the victims of capitalism, of sexism, of racism, of harrassment, "for" the ideas of the movement. There is no doubt that these power structures exist and that they have victimized many people. They also exist in the Wikimedia projects, that's clear. It is a good thing to fight against them. But that must not mean that these groups are used as an "invocation object". They will have to stand in for themselves (and get all support they need). It's a bad idea to replace their voices by the voices of advocates who "know" what is best for them.Mautpreller (talk) 09:36, 23 January 2020 (UTC)[reply]
@Mautpreller: Thanks for this deep response. You've described perfectly what I experience on the English Wikipedia, without the use of any CoC. I'm not sure a CoC would address those issues directly, but I have some suggestions for that too. —Aron Man.🍂 edits🌾 15:39, 24 January 2020 (UTC)[reply]
I don't like the good-faith-argument, when it is about a powerful organisation at one side and pretty powerless users at the other side, and the consequences of new implemented rules are potentially the latter being excluded by the first. This kind of "we all want the same good, we are all in the same boat, so have good faith, we will do the best" is used here far too often to cover the different interests and different point of views, and exactly this can be a problem with a CoC, that is made in the "best interest of us all" but only from the side, that has the power to define these best interests. Like Mautpreller often demands, a user policy should implement rights for the users. Only then it is a good foundation for a communication between organisation and users at eye level. And only when this communication is at eye level, really good things will come out of it. --Magiers (talk) 12:35, 23 January 2020 (UTC)[reply]
@Magiers: This powerless users versus the monster WMF is a made-up illusion that creates discord in the community. I see zero credibility in it: the biggest communities have more power than the WMF to the extent, that the WMF cannot do anything about violations of global policies and the Terms of Use by contributors in high positions. On the other part about user rights being declared I agree: rules and rights should be in balance. The lack of due processes and right to appeal in all governance processes is very unhealthy, causing injustice and abuse. —Aron Man.🍂 edits🌾 15:53, 24 January 2020 (UTC)[reply]
I did not speak of "monster" and I object to the accusation, that I "create discord" just because I argue a different opinion. I spoke about power, that is not bad in intself, but is unequally distributed, when one side is legislative and potentially judiciary and executive in union. Then to demand, the single users shall have good faith, is not convincing. I expect proofs that things are aiming in the right direction. So let's start with a draft, what the COC aims in detail. Otherwhise I cannot give a carte blanche to nebulous or undisclosed plans. --Magiers (talk) 08:35, 25 January 2020 (UTC)[reply]
@Magiers: I'm certainly not accusing you with anything, don't worry. This is not a forum to make accusations, but to engage in civil discourse. I was highlighting the fact that the WMF is treated as a "monster" in certain vocal circles. I do not imply that it's something to be "accused" of. Everybody has their right to their opinion.
What I'd like to point out is that "WMF" is an umbrella term for many smaller organizations and hundreds of people, including many volunteers, not just employees. Treating all of them negatively because of grievances - both valid and over-exaggerated -, towards some members is a gross generalization and hurtful towards those, who invest months of work into projects. It creates a counterproductive, unhealthy environment, that I called "discord". This effect is not a conscious intention, but an effect nonetheless, which I wanted to raise awareness of. Blaming any particular person for this would be as wrong and misguided as blaming a vaguely-defined umbrella organization for all the troubles, when some members of the communities directly contribute to the uncivil behaviors, that a CoC would denounce.
To see what a draft might look like please see The Contributor Covenant CoC and understand that this consultation is not about the exact wording of a CoC, but – in my opinion – to gather feedback from the community of what's expected from a CoC, so that a future draft can be written with the community's feedback taken into account. I hope I helped to clarify the purpose. Thank you for this discussion. —Aron Man.🍂 edits🌾 10:03, 25 January 2020 (UTC)[reply]
Without a draft, it's impossible to know if a CoC has the potential to be used as a way of tone policing so that anyone who seems to be an irritating dissenter will be silenced, or whether it will be ineffective at handing people who casually throw in "f**k you" just to be hostile because they want to demonstrate how great they are at "free speech", or whether it just means that there will be an effective secret police force with non-transparent reporting and assessment processes that perma-bans people without any chance of appeal.
All these things already happen on our projects, but formalizing them in a CoC for all projects and events is unlikely to be a crowd pleaser. -- (talk) 09:56, 23 January 2020 (UTC)[reply]
Thanks Mautpreller for pinging me on this. I can see you have had bad experiences with CoCs in the past. That doesn’t mean our CoC needs to be one of those. Aron Man made some good points. Many other communities have CoCs and it’s rather odd for such a large community to not have one. I have seen them work both well and poorly. It largely depended on how the CoC was implemented by leadership (For poor examples, see any subreddit without distributed and diverse leadership). Usually situations of silencing and other bad implementations of CoCs had directly to do with the leadership. Right now, this is just proposing having a universal CoC. Could you maybe look at some examples of where CoCs are going well? There is even the CoC for technical spaces that happened a few years back. There was loads of discussion and it was a slow process, but it was successful. Let me know if you want me to share some other examples and I’ll gladly do so. Best, Jackiekoerner (talk) 15:42, 23 January 2020 (UTC)[reply]
There is no "leadership" in or above a Wikipedia project and this is good and not bad. "Leadership" by any "Movement body" or WMF organization over autonomous projects is a) not legitimate, b) not to be trusted. There is already some experience with the "Trust & Safety" team attempts which is very bad. There are no clear procedures, minimal requirements of a fair process are not fulfilled. No, I don't accept that.--Mautpreller (talk) 16:01, 23 January 2020 (UTC)[reply]
+1. Also, FWIW, the Technical spaces CoC was edit-warred into place by the WMF without consensus, and people let it sit because it was just too exhausting to keep fighting off the WMF. (It set in place a WMF-appointed dictatorship with no accountability, able to do more-or-less whatever they wanted.)
In general, setting conduct policy on the projects is none of the WMF's business, and setting French Wikipedia conduct policy is none of English Wikipedia's business. --Yair rand (talk) 17:58, 23 January 2020 (UTC)[reply]
I can imagine situations, where laissez-faire towards some projects might not be warranted, iirc one of the Balkan language WP was once taken over by nationalistic racist admins who harassed NPOV authors, and was it Azebaijan, where the dictator took over the WP? In such extreme circumstances I could imagine some intervention by legal, but usually you are right, of course. Grüße vom Sänger ♫(Reden) 18:47, 23 January 2020 (UTC)[reply]
@Mautpreller: Every project has its leadership and established power structures, just like the WMF has it's governance structures. The only difference is that it's undeclared, that it's only visible if you follow the events. In this form this is less transparent than declared statuses and responsibilities, leading to the lack of accountability in the projects, which leads to injustice and abuse. Fortunately, that's not an everyday event, but happens too often and creates an unhealthy editing environment, where fear of abuse negatively affect the way editors edit and interact. —Aron Man.🍂 edits🌾 16:05, 24 January 2020 (UTC)[reply]
Certainly, every project has its power structure (but not leadership, that's not correct). But these powers are limited. The "Code of conduct" attempt, however, introduces unlimited and, more important, uncontrolled power. I could see this very well in several actions of the Trust&Safety team. It's easy to see: there is no way to defend yourself if a "leader" says you are guilty of any transgression. For two reasons: because it is not clearly defined what exactly is a transgression and because there is no fair trial with formalized procedures. The main problem is that all Coc's you are using as examples are highlighting lofty values and "pledges". Almost each and every action might be defined as a transgression against such values and "pledges" because they are so lofty. A well-proven maxim for any kind of diciplinary measures is to exactly define negative behaviour that may entail sanctions, but not positive values because they are open to different interpretations and the danger is enormous that these vague global confessions are made use of for highly problematic interests. Even more important is the "fair trial" issue. You absolutely and always need clear procedures for notification, defence, and appeal. There are examples for this in admin actions in the Wikipedias (sometimes they fall short of this requirement but on the whole it is at least acknowledged as important). In the Contributor Covenant CoC the opposite is the case: There is no concern of procedures, fair trials, appeals, and so on. The T&S "trials" did not even notify the persons they sanctioned, did not state the transgressions sanctioned, did not permit any defence and no appeal. I could say they failed to do so, but this is not correct, they did it on purpose. - These forms of sanctions and proecesses are fundamentally unacceptable. They lead to kafka-like situations. They fail to comply with fundamental requirements of fair process. If you try to push through with them, you will provoke a degree of "discord" you didn't dream of before. And rightly so. No emancipatory project should be allowed to fall back behind these requirements that were the achievements of hard and long-standing fights of the civil rights movements. Mautpreller (talk) 22:25, 26 January 2020 (UTC)[reply]
@Mautpreller: Many replies, sorry.
Certainly, every project has its power structure [...] But these powers are limited.
Indefinite unilateral blocks are within that power. An admin has the right to de-platform an editor with less edits than the admin, if they choose to do so. This does happen (not giving names). That's about as much damage as can be done. A very high limit, I would say.
[CoC] introduces unlimited and, more important, uncontrolled power
The CoC is not the proposal of the feared / distrusted T&S. 1) It's an open-source industry standard in the last 5 years and 2) it's proposed by the working groups to address conduct issues with an industry-proven practice.
If the power to sanction based on the CoC would be given only to the T&S (in its current form), then I would agree with you. I've strongly advocated transparency and accountability for T&S, with no effect. There are 2 reasons why I don't worry about this, though:
  1. The T&S does not have the capacity to handle all reports about user conduct. It's practically impossible to concentrate that power in a small team. The obvious solution to handle reports is to create a team of volunteers and - hopefully - some employed professionals trained in dispute resolution (mediation / arbitration) to handle reports. This means the evaluation and decision-making will be in the hand of volunteers or the communities, if the report is addressed to one of the dramaboards, just like now. It would be unfeasible to give this "power" solely to the T&S. I find such ideas unrealistic and lacking the effort to think about how this can be implemented. These thoughts are important to be addressed, but if repeated in overwhelming quantity then this dramatic presumptions of bad faith distract from the true purpose and benefits of a CoC. I also note that there are editors who maybe don't want to adjust their conduct according to a CoC. I assume they also voted to oppose.
  2. The Principle of Accountability is introduced with the recommendations. At the present and in the past there were no checks and balances to ensure accountability. en:WP:ADMINACCT is ca. 10 lines compared to hundreds of pages that govern editor actions and conduct. Even that 10 lines is generally ignored and unactionable. This is a movement-wide issue, T&S is not the only group that needs to improve in this regard, but all the communities.
It is within the purview of the recommendations to develop processes to ensure accountability. The best and most important process to start implementing accountability is the user reporting process, that would handle complaints about CoC violations. I've explored how this can be done and drafted a design proposal for the planned tool to transparently and accountably handle reports and the discussions in the evaluation process (with "notification, defense, and appeal"). Preventing the abuse that many editors fear depends on properly designed processes with fairness guarantees, not the CoC.
It's easy to see: there is no way to defend yourself if a "leader" says you are guilty of any transgression.
As I've said above, this does happen in the communities. You can call them "admins in high regard", if you wish. They word the policies and also apply those, therefore I've used the shorter "leaders".
For two reasons: because it is not clearly defined what exactly is a transgression and because there is no fair trial with formalized procedures.
The two most vague and widely interpreted policies are en:wp:nothere (not even a policy, but used as such) and en:wp:disrupt. An admin opining that any of these apply is enough justification for an indefinite block, which is practically not appealable, thus usually becomes a "de facto ban". There is no procedure required for such decision, no warning, no community review necessary. It can happen out of the blue and it does occasionally. The current situation in the communities is not an ounce better than what you assume. This power is there and used just like you describe it: unlimited and uncontrolled.
Even more important is the "fair trial" issue.
I strongly agree on that as you can see in the contributions I've linked. Trials are not the concern of the CoC. The linked resources have propositions for handling of cases mostly in the context of events (in meatspace). Online the requirements are very different. In person we are more truthful, more accountable and more empathic. Online it's much easier to hide or misinterpret facts or ignore the consequences of our actions and the pain caused to others. Online tools and procedures have to address these with transparency and procedural constraints ensured by the tools (implemented by the software). The current policies and guidelines lack such constraints, giving a very wide range of actions to handle specific incidents. This results in preferential treatment and undue application of rules. All these procedures need to be developed further to ensure fair trials in any case, including cases regarding the CoC.
No emancipatory project should be allowed to fall back behind these requirements that were the achievements of hard and long-standing fights of the civil rights movements.
I agree. So why did we allow Wikipedia's reality to be the opposite of civil rights? Why are bullied new users told they have no rights, the exit is thataway (literally)? I can't empathize with these presumptions of a power grab because the power grab is already a reality, just in different hands. —Aron Man.🍂 edits🌾 09:34, 27 January 2020 (UTC)[reply]
I agree that there are examples of blatant power abuse in the communities. However, there are at least procedures to control and limit this. In de.wp, there is the de:Wikipedia:Sperrprüfung ("block review") and the de:Wikipedia:Administratoren/Probleme ("review of admin action problems"), as well as the ArbCom (de:Wikipedia:Schiedsgericht. I certainly don't want to idealize these mechanisms and criticized them often and sometimes violently because they are not good enough. But they are at least there and they have some consequences (e.g. that it is common understanding that any blocked user has the right to open a "block review" and has to be unblocked to do so). How should we decide whether another kind to deal with "conduct problems" would be better? For my part, I only see one possibility: does this kind of "problem dealing" offer better ways to limit and control power? But the suggestion of a "Universal Code of Conduct" does, as far as I can see, nothing in this direction. The Contributor Covenant CoC, highlighted as an "industrial standard", certainly does not do it. In this situation, I think mistrust is fully justified. The danger of a change to the worse is overwhelming and I don't see any guarantees that this will not be so. I appreciate your attempts to ensure some fundamental rights in these processes and to give more weight to mediation and professional help rather than sanction and exclusion, but this is not part of the suggestion, only your approach to it. As long as the suggestion itself does not contain any thoughts about "checks and balances", limitation and control, I shall remain a staunch opponent to a Universal Code of Conduct.Mautpreller (talk) 10:46, 27 January 2020 (UTC)[reply]
In addition: I wrote myself a kind of recommendation on the German Wikipedia which is by now more than ten years old. This is de:Wikipedia:Kritik-Knigge, might be translated as "recommendations for criticism in article reviews". However, this is not a "code of conduct" and it does not specify any rules to be enforced. It rather intends to set an example for a change of perspectives. How can criticism, even fundamental criticism, be voiced so that it is acceptable? It relies on experience from professional real-life editing work, and it tries to show that it is important to consider the author's perspective rather than purely insisting on rules. But this kind of recommendation can only work if it refrains from any kind of enforcement. It is certainly not the only way to a better discussion (sometimes you will need enforcement) but it is one way among others. It relies on argument, persuasion and conviction instead of hard and fast rules. In my evaluation, even the existence of such a recommendation helps sometimes. But I don't see anything that could be compared with this idea in the present UCoC discussion.Mautpreller (talk) 11:30, 27 January 2020 (UTC)[reply]
@Mautpreller:: Thank you for linking the German Wikipedia's processes. In the first round of the consultation we have discussed those in detail and I was impressed by those processes. Later that year the community desysop RfC on enwiki listed more big communities that have similar processes. Of course each process is only as effective as those who operate it, but the German WP at least has processes. The English WP has nothing: no block review, no admin review, no "admin problems" and no community desysop. As in game theory, the rules/systems determine how the 'game' is played. Without rules it's free-for-all. That's enwiki at the present.
A CoC, however, has nothing to do with abuse, preferential treatment and other governance issues. That's the matter of transparency (presenting the evidence for decisions) and accountability (the enforcement of rules for the enforcers). Ensuring these is a target on its own, parallel to the CoC. —Aron Man.🍂 edits🌾 13:48, 29 January 2020 (UTC)[reply]
The fundamental design flaw of Wiki*edia is: merging judicial and executive power. This image (off-wiki) demonstrates the issue with that. That's the reason why the governance structures fail and toxicity is rampant. To resolve this issue these roles (rights) have to be separated and both kept accountable to the other and the community. Resolving that issue has nothing to do with the CoC. —Aron Man.🍂 edits🌾 14:06, 29 January 2020 (UTC)[reply]
I am sorry, but I don't quite believe it. Why is it impossible to get a clear response to the question whether the CoC is something to be enforced? I'm afraid it will be so. If so, the questions of accountability, transparency and fair trial are vital in the CoC. And the question is also crucial who will enforce it. Above, the CoC for technical spaces and for RL gatherings were highlighted as examples of successful implementation. Well, I don't agree. The Romaine case and the MZMcBride are very telling examples.--Mautpreller (talk) 19:06, 31 January 2020 (UTC)[reply]
  • @Jackiekoerner: - two things with one of your earlier statements. One is that the larger communities do have COCs, though there's differing levels of summarisation into a single document "code" - as very few users are cross-wiki (and those that are bluntly not really in need of a standardised UCOC), there isn't much to be gained by a standard document. If the WMF would like each community to come up with a single document summation of key rules, which then linked out to the full documents, that would be a completely different discussion which I'd happily support. The leaders aspect is something that makes experience in successful implementation not especially transferable - projects don't have leaders, whether that be a local community or the entire Wikimedia set of projects. Almost all decisions would be implemented by a general consensus. The WMF is specifically refuted as holding any leadership position over the communities (perhaps a good analogy might be how the UN Sec-Gen is the Chief administrative officer of the UN, not their CEO). I've only talked to a handful of active mediawiki editors, so it could be sample bias, but 3/4 them say they were unhappy with the COC that they, even now, both feel was de facto forced upon the community. Nosebagbear (talk) 18:08, 24 January 2020 (UTC)[reply]
@Nosebagbear: I'm unaware of any kind of CoC on enwiki - our wiki -, despite my extensive knowledge of policies. If you mean that those policies already include some points usually found in CoCs, then consider how many editors (especially casual or new editors) are aware of those policies, have read them (all the hundreds of pages) and know how to reference them. One of the benefits of a CoC is concision and focus: collecting the most fundamental expectations regarding conduct. The next benefit is to reinforce the importance of these policies, like en:WP:CIV. As you might be aware, the civility policy - despite being one of the 5 pillars - is mostly ignored and not actionable. How do you propose communities should counter the everyday bullying and the quite common harassment, if those values were not reinforced? This is the question the Movement is trying to answer with the introduction of a CoC, that benefited most (100K+) and biggest of the open-source projects in this regard. What would be your solution? —Aron Man.🍂 edits🌾 18:42, 24 January 2020 (UTC)[reply]
I'd want to know what a UCOC could include and how it would solve current issue(s) on en-wiki that the current set-up (or a summarisation of the current setup) couldn't, and do so without causing any significant new problem. An example doesn't have to be the one ultimately opted for, but it has to be a viable possibility to consider supporting the concept. Nosebagbear (talk) 18:46, 24 January 2020 (UTC)[reply]
@Nosebagbear: what a UCOC could include: See the most common Contributor Covenant CoC. how it would solve current issue(s): Doing some research will help you with that. I wouldn't assume that this will be explained individually to every editor. In every big project participants need to be prepared on their own volition, otherwise the project falls apart. Did you want to answer my previous question "What would be your solution?" —Aron Man.🍂 edits🌾 10:16, 25 January 2020 (UTC)[reply]
For informational purposes: The above user, User:Aron_Manning, was banned from the English Wikipedia project in June, for violating the project's conduct policies. I wouldn't ordinarily bring this up, but the repeated use of misleading wording like "our wiki" and the arguing specifically against the guidelines under which the user was banned, indicate to me that this user is not here to participate productively. --Yair rand (talk) 03:40, 11 March 2020 (UTC)[reply]

Is this teaching tolerance toward cultures or just mere dos and don'ts?

I've been thinking especially after reading feedback, most of them negative. Is the Code intended to teach users how to tolerate diverse cultures, give them a lot of respect, promote cultural relativism, or just a mere set of dos and don'ts without sufficiently explaining why the Code should exist in the first place? If neither, then what else does the Code intend? George Ho (talk) 20:47, 31 January 2020 (UTC)[reply]

@George Ho: This is a very good point. There's an example in the stackexchange CoC of such dos and don'ts: [2]. Imo such guidance should be an explanatory supplement to the CoC: my assumption (and real-life experience) is that the primary purpose of a CoC is to reinforce community values about civility and give behavioral guidelines, which would benefit from such explanations. —AronM🍂 edits🌾 01:48, 8 February 2020 (UTC)[reply]
@George Ho: from what I've read, it seems that all the objection stem from that fact that no matter what anyone proposes, it is unenforceable. Mostly that's because such a code of conduct is prescriptive, rather than descriptive. If we found a way to articulate existing consensus (express what we already all agree to) it would be much less charged. And of course there is one thing that we all agree on (on Wikipedia at least): Anything that disrupts the creation and maintenance of the encyclopedia is forbidden. Don't harass the people who write and maintain it. Maybe that's all it ought to say. Adopt a clear definition of harassment, like the Ontario Human Rights Code does: engaging in a course of vexatious comment or conduct that is known or ought reasonably to be known to be unwelcome. Vexations (talk) 18:06, 8 February 2020 (UTC)[reply]

Discuss Space

I thought the plan was to freeze Discuss Space by March 31 2020. Does is still make sense to list it in Universal_Code_of_Conduct#An evolving process? Vexations (talk)

Had anyone advanced a draft over there? I'm not on Spaces. EllenCT (talk) 18:08, 10 March 2020 (UTC)[reply]
I haven't seen a draft for a Code of Conduct there. The use of Discourse itself may be relevant though, as it has some features that facilitate civil conversation. They sort of have it built-in. See https://blog.discourse.org/2013/03/the-universal-rules-of-civilized-discourse/ for more info.

What is it good for? What is insufficient presently?

I write from experience in Swahili wikipedia (and as an observer of African language wikipedias). This debate looks not really clear to me. What is it supposed to change? I see the 2015 report on harassment with some examples of insults. Could those wikipedia communities not handle it? From my experiences of 13 years as admin I do not remember anything of this type in swwiki. Of course it has to do with small or very small communities.

I can actually not see what a global code is to change. Who outside a given community is to decide if a certain expression in Swahili, Malegassy, Somali or Tigrinya is beyond the tolerable? Looks like a huge bureaucratic exercise with unclear parameters. Kipala (talk) 17:17, 10 March 2020 (UTC)[reply]

Proposal: this Code should be developed independent of the Foundation

I propose that the Foundation be given no control, influence, sway, or any other ability to determine the outcome of Code of Conduct determinations, and all such determinations be placed under the control of neutral third parties, because of the Foundation and Foundation staff's inability to demonstrate freedom from conflicts of interest. EllenCT (talk) 09:59, 12 March 2020 (UTC)[reply]

The main page discusses mainly about wikiquette, safe space and personal conflicts. But a code of conduct must include other aspects, such as conflict of interests as EllenCT says. Cntributing to the Wikimedia project while having conflicts of interest should be considered inappropriate conduct.
However, the Foundation's past and current behavior is not the only reason to support EllenCT's proposal. We are a movement, and major issues such as creating a code of conduct should be decided by the community as a whole, not by a few Foundation people. --NaBUru38 (talk) 22:51, 15 March 2020 (UTC)[reply]
Catch 22: who appoints 'neutral third party' and pays for their time and effort? isn't it the Foundation? Retired electrician (talk) 22:44, 23 March 2020 (UTC)[reply]
@Retired electrician: No, of course not. The Wikimedia community is capable of organizing these things for themselves. The WMF are paid administrators who are outsiders to the Wikimedia community and its culture. The Wikimedia community can represent itself. The normal process is
  1. The Wikimedia Foundation publicly announces the budget for Code of Conduct Development, including investment of staff time
  2. some reasonable portion of that budget goes to empowering the Wikimedia community to better represent itself
  3. The Wikimedia community, who are stakeholders in the Code of Conduct, participates in the discussion with power at minimum equal to the WMF, and ideally with power and resources beyond the WMF
A code of conduct comes from an empowered and resourced community. To have it come from outsiders or consultants is an error. Blue Rasberry (talk) 00:31, 19 April 2020 (UTC)[reply]

Enforcement, not code

We already have the WMF terms of use. A universal code of conduct is not really needed.

Neither the WMF nor stewards currently have the capacity to handle escalations of debates that local communities could resolve. That sentiment may have arisen from the various times WMF attempted to intervene in English Wikipedia matters; every time the community rebuffs.

But many small wikis don't have a robust community that takes due account of the vocal opinions editors who don't have advanced privileges. Azerbaijani Wikipedia was a recent example of local dispute resolution failure. Now we have the Russian Wikipedia wanting to escalate to WMF level. More broadly, we also have interwiki disputes, such as those relating to tools deployment, Wikidata integration, and Commons integration. At the moment, it seems that whichever of English Wikipedia or German Wikipedia wants to exert themselves more will win by default, without any real guiding principle of dispute resolution.

What we really need is a WMF appeals court for disputes that the local community cannot resolve. It ought to hear disputes for which local community discussions have got to a dead end, whether the dispute concerns only one wiki or is an interwiki disagreement. Most of the time this appeals court will feel like an RfC or Arbitration case on the English Wikipedia, because small wikis have fewer layers of dispute resolution. But other times it'll function as a super-appeals court that has authority. Deryck C. 12:48, 20 March 2020 (UTC)[reply]

When the new strategy would look for more inclusivity, this discussion is exclusive

Sorry but I figured out about this discussion because I follow international discussions and I am active in several communities but I can assure that some communities did not know about it. Surely the Italian community is ignoring that there is a discussion about an Universal Code of Conduct and we are speaking about one of the biggest community. I assume that this is included in the new reccomendations of the new strategy but these reccomandations also say that the movement must be more inclusive. My feeling is that this discussion is quite exclusive and it's a pity that a discussion about a Code of Conduct starts with a limited group of communities. I suggest to extend the translations and open it to more communities. --Ilario (talk) 10:37, 30 March 2020 (UTC)[reply]

@Ilario: Barely anyone knows about this, but it doesn't really matter because it's already clear that the proposal is practically unanimously opposed. The WMF is starting discussions in a bunch of languages, and they'll get some more walls of opposes, and either the WMF will keep pushing or they won't. There's not much point in trying to pull more attention to this, at least for now. --Yair rand (talk) 18:21, 31 March 2020 (UTC)[reply]

Surveys in progress March 2020

User surveys are in progress. I wish that the Wikimedia community could get notice of when the WMF does surveys and how the WMF will report results. How many communities are invited into this survey?

I am anxious that the WMF has originated some odd ideas in the Code of Conduct which did not come from the Wikimedia community and would be controversial in the wiki community. Why do surveys without disclosing the fact of the survey?

There are so many points of universal agreement. I hope no one risks the likely consensus and delays the advancement of discussion by failing to include the Wikimedia community in the development of any code of conduct. Blue Rasberry (talk) 00:36, 19 April 2020 (UTC)[reply]

WMF please fund Wikimedia communities to organize local discussions on this

The path to a universal code of conduct includes the WMF giving money to Wikimedia community groups who organize themselves to discuss Wikimedia community conduct without any pressure, guidance, or direction from Wikimedia Foundation staff and consultants. There are lots of Wikimedia community groups who have discussed this already. With money to support the administration of grassroots research, notetaking, and documentation, the Wikimedia community is capable of building some consensus among itself.

I am going to guess and estimate that the money that the Wikimedia Foundation has spent in staff time, consultants, researchers, and administrative support on developing a Universal Code of Conduct is about US$400,000. The WMF can report for itself what budget has gone into this and also the specific amount of money which it has granted to the community for the same purpose.

There is so much expertise in this space in the Wikimedia community from so many different perspectives. It is completely impossible for the WMF to go this alone without the community.

Please WMF include the Wikimedia community in the normal conventional routine and sane way. Please do not plan any code of conduct discussion in the repeatedly attempted rollout model of superprotect / image viewer / FRAM / rebranding / fundraising weirdness / Knowledge Engine. Please make sure that multiple Wikimedia community volunteers are at the forefront of this proposal. Any investment that the WMF made in proposing and documenting a model for growing Wikimedia community consensus would be a great investment aspect of the Code of Conduct project. Any passable code of conduct is going to include a rule that says "the only person fit to speak for a community group is a volunteer member who has the support of that group to speak", and no WMF staffer will ever fit that description any more than any corporation or empowered entity can represent consumers or the common person. The Wikimedia community does not need anyone to speak or advocate for it, but does appreciate Wikimedia Foundation support - money - to sponsor the administrative base from which common people can empower themselves and organize their own discussions.

If anyone with power wants to make this code of conduct have community support in the future then they ought to financially sponsor the necessary global community-based discussions for this sooner. The Wikimedia Foundation has enough money at this point that a frank and forward discussion about money should come right at the front of any intersection of WMF proposals and community consensus seeking. Blue Rasberry (talk) 00:53, 19 April 2020 (UTC)[reply]

If they spent anywhere near $400,000 it's a dramatic waste of money considering the only passable version of a UCoC would be one which does not currently interfere with local conduct policies on any projects, which anyone knowledgeable of Wikimedia communities could write up in under two hours. In regard to paying contributors for local discussions, that seems a bit overboard. It'd be a lot easier, and cheaper, to place messages on local noticeboards with MMS and request that the various projects create organized project-space pages for local discussion and connect it to a Wikidata item for WMF staffers to later review. Vermont (talk) 01:33, 19 April 2020 (UTC)[reply]
@Vermont: There are some things which grassroots community volunteers do and some things which come from the administration of a nonprofit organization. It is a fantasy and an error to imagine that volunteers can do for free the things which come from staffed organizations. For any given task, we need to be able to determine when we need volunteer labor and when we need staff labor.
You say when the WMF spends money "it's a dramatic waste" and also that giving wiki community groups money "seems a bit overboard". I get it, I hear lots of wiki people say exactly this, but these wishes are incompatible with each other. The Wikimedia Movement pulls in US$100 million a year and every part gets spent every year. The finance options for addressing any challenge are (1) money directed by WMF staff (2) money directed by community stakeholders (3) money for neither and instead to be spent for some other challenge. The default solution to every challenge in the wiki movement is always either 1 or 3, and never 2 until and unless the Wiki community exerts itself greatly.
You say anyone can synthesize the text in two hours - here are the 1000 pages of text to condense:
I disagree that a knowledgeable Wikipedian can write a passable code of conduct. Instead, I believe that any knowledgeable Wikipedian can easily identify major problems with any text already in existence, including all the codes of conduct for other online communities. Writing a widely acceptable code of conduct might be impossible. I still am not sure.
I also find you very quick to accept that giving all data and community feedback to "WMF staffers to later review" is the solution to this or any arbitrary challenge. Sometimes that works, or sometimes we could go with community partnerships. I would love to be able to reroute the money from WMF staffers to fund research analysis and partnerships to university groups in underserved countries, for example. We could have options. I think it is unfortunate that when money goes to volunteer groups people complain, when it goes to grants people critique the process, when money goes to consultants it is a scandal, but when the WMF hires yet more people then that is silent and unobservable so that is the least negative and most acceptable option.
I advocate for an empowered wiki community in the development of community policy. I see no reason to believe we will increase community empowerment by continuing the status quo. After code of conduct we will have new issues every year. What do you think should happen perpetually when comparable challenges repeatedly arise?
I very much appreciate your responding to me and welcome your wildest proposals of what to do. Blue Rasberry (talk) 19:41, 20 April 2020 (UTC)[reply]
Thank you for your response. My main issue regarding the cost would be that, when implemented, regardless of how good or bad the final product is, there is going to be community dissent. The more the UCoC removes responsibilities from local volunteer communities, the more dissent there would be, and I'd hate to see a lot of money spent on developing a program that is eventually scrapped due to push-back. There are many possibilities less likely to cause problems, such as creating a group of people, staff or not, to enforce a set code in mediating institutional problems on Wikimedia projects, filling the gap that Stewards have been working on of late, such as with azwiki, hrwiki, and a few others. In regard to the "2 hours" comment I made, I meant that many multiple-project community members could formulate a decent proposal for a UCoC in a short amount of time; I was not referring to the extensive research done of currently existing policies. The difference between my comment and your interpretation is, where I used "anyone knowledgeable of Wikimedia communities", you used "Wikipedian". I'd think most global sysops, admins on multiple projects, and stewards are capable of coming up with passable proposals, but I agree that your average editor on a single Wikipedia would not. However, we both seem to agree that most active editors, regardless of project, would be able to identify problems in proposals and voice legitimate opinions on their impact. In reading that collated list of projects and policies over again, I'm a bit intrigued that they left out the Simple English Wikipedia, but that's just my own curiosity and nothing to do with this process. For my "WMF staffers to later review" bit, that's because in the hypothetical situation of sending a mass message to communities, WMF staffers would collate the community responses into one consensus-developed UCoC. In nearly any course of action, there will be multiple points at which WMF staffers would be tasked with judging community consensus and response, and the goal with such a MMS would be to gather community responses for staffers to look through. I hope that is cleared up, and I'm not sure what you refer to by "community partnerships". With regard to community empowerment, it's seemingly become evident that the future holds a continually growing Wikimedia Foundation that, over time, siphons power and responsibilities from local communities while feeling a lessening need to permit community processes to function without WMF involvement. In my view, this is not necessarily bad, but it very well could be. In regard to my proposals, I don't have much so far, but I have some thoughts. Given the niche that I edit in and the experiences I've had, I often don't share the same views of most single-project editors, and any UCoC ideas I conceive would be centered around mediating institutional problems on projects that experience continually permitted admin abuse, copyright violations, racism/intolerance, etc. Multiple projects currently experience or have experienced problems in that regard, and T&S has helped out to some degree, as have stewards and global RfCs, but it's all a bit ambiguous. If there's to be a global code of conduct to hold every project to, I would think it should interfere with community processes as little as possible except when those processes are harmed by institutional issues. As such, the issue isn't what projects do or don't have certain policies, it's a question of what projects enforce them fairly and without extensive bias, which would need some sort of outside force (UCoC committee, like recently proposed?) to manage it. The pages of research don't seem useful to me in that regard; all that is necessary, in my view, is a set of requirements that a project would need to uphold listing basic conduct policies, involved administrator policies (when there's over a set number of administrators), and anti-discrimination policies. For example, a hypothetical UCoC would have come in handy with Til Eulenspiegel rather than a global ban discussion. Take a look at the list of recent global RfCs, and see how many of them focus on issues in a certain wiki where the local establishment is not able to handle whatever issues there are or is the alleged source of the problem. Seems to me about half. Many of these RfC's go weeks or months without coming to a conclusion, some have no discussion and are closed, but for those who do have lengthy discourse at the end there's always questions of jurisdiction, precedent, the ability to enforce a result, etc., and the closer is bound to get hate from whichever side isn't given a positive result. If there is to be a UCoC, let it fill that gap. I know I've answered your last question a bit more broadly than expected, but I hope it gives insight into the lenses through which I view this issue. Thank you, Vermont (talk) 22:25, 20 April 2020 (UTC)[reply]
Bluerasberry, forgot to ping, see above. Also, oy vey, sorry for the wall of text. I just kept typing... Vermont (talk) 22:26, 20 April 2020 (UTC)[reply]
@Vermont: Thanks for your reply. I appreciate it all. I am unable to reply fully for a while and will reflect, thanks. Blue Rasberry (talk) 11:45, 21 April 2020 (UTC)[reply]
WMF employees do not determine consensus. They need to consider it, when it comes to a decision WMF needs to make, and report accurately to the decision maker (for instance the board). If the relevant community has not explicitly determined and summarised the consensus, WMF employees can and should summarise at the best of their abilities. Nemo 05:03, 21 April 2020 (UTC)[reply]
I second the sentiment, but there's also the issue that paying someone to do this kind of work is in effect a bribe to local communities, sometimes very small, to promote the WMF's desired outcome. (But the price usually WMF pays to compliant people is not only financial; it can also come in the form of access to future benefits.) In the worst case, you end up having an army of editors on WMF's payroll capturing millions of users who just don't have the energies to fight for what would work best for their projects.
The WMF grants have sometimes explored ways to avoid this conflict of interest, for instance by focusing on the reimbursement of costs incurred for community work rather than wages. The typical way is to throw some thousands of dollars to an in-person event on a subject WMF likes, but that's not an option now. I agree that alternatives can be found, if there is the will. For instance at some point I was thinking of running in-person reading groups after sending a copy of a book on online abuse that has been recently published in Italian. Such an initiative could happen online too, would have some costs for physical goods and may benefit from a professional facilitator from outside the community without an issue of consensus determination.
I don't necessarily agree that a volunteer cannot write proper policy for our very diverse communities. In fact, if it's possible at all, only volunteers can; while the WMF employees are certain to be unable to do it, not because of how they are but because of how WMF works. It's true however that it takes a lot of work, and some incentives, for the projects to discuss extensively on something that they usually don't feel like discussing, and WMF needs to think how to avoid overloading the community with extra work. Nemo 05:03, 21 April 2020 (UTC)[reply]

Draft proposals

Hello, I added a section for proposals at Universal Code of Conduct#Proposals. I put a draft proposal from me there.

Although I think that my proposal would be good in many ways, a difficult issue is how to support diversity of expressions and opinions, even when those expressions or opinions may offend others, while also supporting civility. At this point, I think that civility policies are best left to local communities.

I welcome others' opinions, including alternate proposals. Thank you, ↠Pine () 03:41, 20 April 2020 (UTC)[reply]

  • I have not marked my proposal for translation because I do not want to request the valuable time of volunteer translators if there are early indications that my proposal will not be approved. However, if others feel that the proposal is good, please feel free to mark it for translation.
  • I tried to avoid breaking any of the existing translation markers. I would appreciate any corrections to errors that I made. Thanks, ↠Pine () 03:49, 20 April 2020 (UTC)[reply]
  1. This seems like a reasonable first approximation. I have not seen anything that would be an obvious problem, but I am fairly sure that there will have to be some changes, as there are likely to be things that are not sufficiently clear to some people.
  2. Why 7 members? Will this be enough? there will be cases where one or more members will want to or have to recuse. How many members would be considered a quorum?
  3. How will they handle language problems, and lack of familiarity with the customs and policies of the various projects? In what language will cases be investigated? How will transparency of process be ensured? · · · Peter (Southwood) (talk): 08:38, 20 April 2020 (UTC)[reply]

We don't need this

Can we just stop it all together before it (the process of developing UCoC) begins for real? It's just a pure waste of time. tufor (talk) 23:11, 21 April 2020 (UTC)[reply]

Instead, PLEASE help us deal with rogue wikis

This initiative clearly has good intentions, but it is also clear it isn't going well. The biggest problem we have are routine squabbles over content. There is little chance of the Foundation doing anything constructive about those squabbles, and even less chance that Foundation efforts would be welcome. When it comes to things like threats, I believe we already know to escalate such issues to be handled by the Foundation. Again, not much chance for the Foundation to do anything new and constructive.

One area that really is a problem is dealing with small wikis that go off the rails. AzWiki and HrWiki come to mind. In some countries we have leaders of government, and academics, and major newspapers all publicly condemning Wikipedia for Holocaust denial and radical propaganda, because that's what's in Wikipedia in that language. That's what Wikipedia contains because the wingnut admins write it that way. Those admins threaten, abuse, run-off, or ban anyone who tries to fix it. Small wikis are get launched with very little vetting of the initial admins. A wiki can turn into a hell-hole if its dominated by abusive admins. Holocaust deniers, rabid-nationalists, historical revisionists, racicists, homophobes, theocrats, whatever. Our process sucks for dealing with that sort of thing. We need a better defined process, but more importantly, that is where we need Foundation help. Right now basically the only option we have is to revoke the bad-admins and dump the wiki into the lap of stewards to manage. Stewards don't want to take the responsibility, the language barrier is a huge problem, and that language is left to fester.

If the Foundation wants "equity" for under-served languages, if the Foundation wants to help users who are being harassed and abused, if the Foundation wants to protect the reputation of Wikipedia as a trusted source for neutral and reliable information, then THIS is where you need to put your efforts. Cleaning up small dysfunctional wikis.

Leave it to the global community to evaluate when it is necessary and appropriate to intervene in a wiki. Once the broader community has made that decision, please oh PLEASE step in and help clean up the mess. Launch an initiative /consultation to write global policies for evaluating and re-booting a wiki. Help set up communication channels for community review of a wiki. If a wiki does need to be rebooted, help us find experienced and responsible editors who can speak the language. If necessary pay them to admin for a year. Maybe hire translators to translate a set of model-policies or other educational-materials into the language. In particular, one of the few things that I find actually effective is when editors are firmly informed: "Wikipedia does not contain Truth. Arguing Truth doesn't work here. Arguing Truth is disruptive. Wikipedia is an accurate summary of what Reliable Sources say." Anyone who is unwilling or unable to put that ahead of their personal beliefs can't be an admin. Alsee (talk) 09:25, 7 May 2020 (UTC)[reply]

Ping NNair (WMF) to (hopefully) take this issue to the team for consideration. Alsee (talk) 09:31, 7 May 2020 (UTC)[reply]

I'd assume that any Universal Code of Conduct would include ways to deal with systematic issues on projects, like those on azwiki, hrwiki, and others; in fact, it's probably one of the only parts of a prospective Universal Code of Conduct that I'd support, and depending on the implementation, strongly so. At the moment, the global RfC system is severely inefficient and there's been very little recourse for people wronged by systematic problems with local administrators. Vermont (talk) 14:03, 7 May 2020 (UTC)[reply]

This will be enacted 30 August 2020

"Develop and introduce a universal code of conduct (UCoC) that will be a binding minimum set of standards across all Wikimedia projects. The first phase, covering policies for in-person and virtual events, technical spaces, and all Wikimedia projects and wikis, and developed in collaboration with the international Wikimedia communities, will be presented to the Board for ratification by August 30, 2020."

Blue Rasberry (talk) 00:13, 23 May 2020 (UTC)[reply]

Not a surprise – the decision was made a long time ago and the 'consultations' were a facade. We'll soon find out what the "close consultation with volunteer contributor communities" means. EddieHugh (talk) 11:42, 23 May 2020 (UTC)[reply]
Typical. Proclamations just for the sake of publicity, with no regard to substance. Nemo 13:16, 23 May 2020 (UTC)[reply]