Talk:Wikimedia Foundation Annual Plan/2025-2026/Global Trends/Common global standards for NPOV policies

Research contributions

Please share here research, comparisons and other work around NPOV principles and policies that we can learn from for this work.

Community and other resources

Please share here examples of NPOV policies from different Wikimedia projects so we can map what already exists.

• d:Q4656487#sitelinks-wikipedia has links to all of the policies. WhatamIdoing (talk) 20:07, 4 April 2025 (UTC)

  the link collection in phab:T389445 might also be interesting. --Johannnes89 (talk) 06:50, 5 April 2025 (UTC)

• I think it's important to understand the differing roles and concepts of "neutrality" in different (i.e., non-Wikipedia) projects. For example:
  • The English Wikipedia (w:en:WP:NPOV) says that being neutral means reading a lot of good sources and fairly representing the views of those sources. For example: w:en:Breast cancer awareness has been criticized by a lot of Western (and especially American) academics. It is promoted by charity fundraisers in those same countries, and exploited by businesses, and lauded as life-saving by some patients. A neutral article does not treat academics and money-grubbers and patients the same. A neutral article acknowledges the existence of all of these viewpoints ("POVs") and writes the article to emphasize the academic viewpoint as the most important. Additionally, it promotes the use of a neutral, dispassionate tone when describing facts. (For example, those that use breast cancer awareness as a marketing tool are described at the English Wikipedia as "charities" and "fundraisers" and "businesses", but never using moralistic or judgemental language like "money-grubbers".)
  • Commons (c:COM:NPOV) takes its role as an image host seriously by hosting "bad" images. These should be accurately labeled but not deleted. For example, in the case of a real-world border dispute, a map showing the borderlines claimed by Country A should be described as showing the borders of Country A, including disputed territory claimed by Country B. It should not be described as "the only true and correct official boarders of Country A". In the words of EB White, their goal is closer to "a well-balanced library, not a well-balanced book". The map showing Country A's claimed borders is "balanced" by the existence of a separate map showing Country B's claimed borders.
    Similarly, hoaxes, photomontages, and AI-generated images are acceptable if they are correctly labeled, but unlabeled ones are eligible for deletion. This is because all of these images could be used in a neutral way, e.g., with a caption of "Here is an example of a hoax photo" or "Here is a map showing the Country A's claimed borders". Commons also hosts technically bad photos, because other projects need to be able to write, e.g., "Here is an example of a blurry photograph, showing what happens if you don't focus the camera correctly".
  • Wikivoyage (voy:WV:Be fair) eschews Wikipedia's "neutral tone" in favor of a fair-minded but lively writing style. Beyond that, the project is more concerned with personal experiences and useful descriptions. Where Wikipedia might write "According to The Big Book of Reviews, the building has a strobe light", Wikivoyage might write "Located in a historic building, the bar has live music every weekend, with a disco-style dance floor". Also, where Wikipedia would even-handedly report both praise and criticism, Wikivoyage prefers to make positive recommendations, and to omit attractions that its contributors do not recommend. Where real-world controversies affect travelers, they try to fairly describe the situation from a purely utilitarian viewpoint. For example, rather than saying whether a given patch of dirt "really" belongs to Country A or Country B, they will note that the dispute exists, and that as a purely practical matter, a traveler will need to get a visa from Country B.
  • Wikisource (s:WS:NPOV) hosts accurate copies of authentic old documents, without a central selection process to make sure the documents have the "right" viewpoint or are (e.g.,) evenly distributed by age and location, and without "correcting" (i.e., corrupting) historical documents to present modern viewpoints. An accurate, faithful copy of a historical document is the goal, even if that old document uses offensively outdated language, expresses offensively outdated ideas, etc.
• Similar differences are found across most projects. I'm not sure that it will be easy, or even possible, to write a policy of any significant length or actionable detail that applies equally to all projects. However, I believe that a shorter, more values-focused statement – something similar to foundation:Resolution:Biographies of living people – is possible. WhatamIdoing (talk) 21:40, 4 April 2025 (UTC)

Mandatory sources

On Verifiability, which is derived from NOPV: In July 2013, the Lusophone Wikipedia community came to the conclusion that the existing concepts of NPOV were insufficient and ineffective, and decided that sources should become mandatory, and any unsourced content may be removed from articles by editors without further explanation, after which it may only return to the article if it's followed inline by a reliable source or sources which directly confirms it. This draconian policy of zero tolerance with unsourced content, which I'm not sure exists in any other project, appears to have decisively contributed to and have been followed by a significant increase in the reputation of the project in the Lusophone world - which until then was not very good, to put it mildly -, most notably and noticeably in Portugal and Brazil. It has also caused a dramatic change of attitude among editors, since reinstatement of removed content, even if "true", became strictly forbidden if a reliable source was not immediately presented.

OTOH, the unfortunate acritical import of the deficient English Wikipedia policies on the definitions of original research and reliable sources to our project on the early days of our Wikipedia, most notably a chronic misunderstanding of one of the most common definitions of what primary sources are, and that they are quite often the best quality sources, as opposed to the secondary ones that merely parrot or summarize them, and are generally of a lesser value concerning that information - as correctly stated in the Wikipedia article about them, but not in the policy -, has caused havoc there since ever, and it's something we definitively must work upon. - Darwin ^Ahoy! 16:45, 27 March 2025 (UTC)

How "primary sources are... quite often the best quality sources", @DarwIn? I cannot think of a ready example.

Counterexamples are galore:

1. The Bible (which one, by the way?) is a a very bad source re "what God (which one, by the way?) is or does".
2. I Ching (易經) is a very bad source about itself or Chinese divination.
3. The Myth of the Twentieth Century (Der Mythus des zwanzigsten Jahrhunderts) is a very bad source about myths, 20th century and itself.
4. Nicolas Rosi (this one: https://www.bbc.com/news/uk-scotland-68004574) and his writings or claims was a very bad source for the article that he wrote about himself back then on enwiki ...

-> do elucidate, together with that "chronic misunderstanding (of its definition)".

Zezen (talk) 20:46, 27 March 2025 (UTC)

@Zezen 1) Prime-minister of X tweets something (primary source, then frowned upon), which is badly quoted by 2) The Times (secondary source, then "reliable" on Wikipedia standards. 1) is the reliable source, and 2) just parrots it, and parrots it badly. Yet it's 2) that the Wikipedia policy says it's the best. And there are countless other examples, with laws and other official records being quoted by some secondary source instead of directly using the original (something we never do in History, except in very exceptional occasions, the original source is generally considered the best), and so on. - Darwin ^Ahoy! 21:47, 27 March 2025 (UTC)

Prime-minister of X tweets something does not per (at least en-WP) NPOV necessarily mean it should be on WP. The Times writing about the tweet is an indicator it should. Gråbergs Gråa Sång (talk) 05:51, 28 March 2025 (UTC)

@Gråbergs Gråa Sång I absolutely agree with you. The Times is crucial to evaluate the notability of the information - and it's the primary source of such notability, on that - but it's obviously less reliable than the primary, original source when it comes to the content that was actually posted by the OP. therefore, both of them have good value, but only when they act as primary sources of the information they transmit. Which is actually the opposite of what the Wikipedia policy says. - Darwin ^Ahoy! 12:56, 28 March 2025 (UTC)

@Zezen About your counterexamples, I believe it suffices to say that the Bible is the only reliable source about what is written in the Bible. You have countless works playing with the supposed biblical fact that Mary Magdalene was a prostitute, when such thing is nowhere to be seen in the Bible itself. All the other cases you listed there are like this: They may be unreliable for what they describe, but they are the only reliable source about the way they are describing it. And if you need that specific information in an article, it is them that you should quote, not some secondary source that just parrots it at best without any added value, even if they also provide some valuable analysis on the original source and facts - of which they would then be the primary - not secondary - sources themselves. - Darwin ^Ahoy! 22:24, 27 March 2025 (UTC)

I'd fundamentally disagree with the notion that verifiability is "derived" from NPOV. It's entirely possible to write a completely verifiable and yet utterly non-neutral article. It would be completely possible to write an article on, for instance, the en:Elgin Marbles which was entirely verifiable and sourced to reliable sources and yet did not mention Greece's two-century campaign to have them returned to Athens. Entirely verifiable and yet entirely non-neutral.

I'd also fundamentally disagree with your assertion that primary sources are quite often the best quality sources whereas secondary ones merely parrot or summarize them. It's ironic, really, that you think that there is a chronic misunderstanding of what a primary source is: if you think a secondary source merely parrots or summarises primary ones, you have a serious misunderstanding of what a secondary source is. High-quality secondary sources interpret and analyse primary sources. At least in the subject areas that I am interested in, this is absolutely not trivial and leaving the selection, interpretation, and analysis of primary sources up to editors with no guarantee of expertise and unknown viewpoints is a dangerous task indeed. As an example, ancient sources tell us that en:Aspasia was put on trial for impiety; modern scholarship doubts whether this trial ever happened. Privileging the primary sources in this case leads to the omission of the predominant modern scholarly view: if that's considered neutrality, then NPOV is fundamentally broken. This is hardly an isolated example. Caeciliusinhorto (talk) 21:31, 27 March 2025 (UTC)

@Caeciliusinhorto And who decides it is not neutral, despite being sourced to reliable sources? You, the editor? And people is supposed to trust you, and not what was published and peer reviewed? - Darwin ^Ahoy! 21:49, 27 March 2025 (UTC)

And people is supposed to trust you, and not what was published and peer reviewed? I don't understand what your argument is here. You are the one arguing that published and peer reviewed secondary sources merely parrot or summarize primary sources. The idea that our articles should be based on published and ideally peer reviewed reliable secondary sources is the fundamental core of English Wikipedia's sourcing policies with which you apparently disagree. Caeciliusinhorto (talk) 21:57, 27 March 2025 (UTC)

@Caeciliusinhorto Yes, if they are a secondary sources, that's exactly what they do. They either parrot the original source, in variable degrees of reliability, or attempt to summarize according to their own bias. That's the definition of what secondary sources are in History or Science. If they provide their own analysis or interpretation of what was reported in some original source, they are being the primary, original source of that analysis or interpretation as well. There's no point in quoting a journalist quoting a law that was approved by the government, when you can directly quote the document itself, without engaging in "who tells a tale adds a tail" sort of dynamics. - Darwin ^Ahoy! 22:05, 27 March 2025 (UTC)

Yes, if they are a secondary sources, that's exactly what they do. That's the definition. If they provide their own analysis or interpretation of what was reported in some original source, they are being the primary, original source of that analysis or interpretation as well. That's not what the phrases "primary source" and "secondary source" mean either in common English usage or in the English Wikipedia policy you are criticising. You objected in your original post to en:WP:PRIMARY, which defines a primary source as original materials that are close to an event, and are often accounts written by people who are directly involved and a secondary source as one which provides thought and reflection based on primary sources, generally at least one step removed from an event. It contains analysis, evaluation, interpretation, or synthesis of the facts, evidence, concepts, and ideas taken from primary sources.

As for whether directly quoting a law is a good idea: I would agree that usually a journalist is not the best source here, but often nor is the law itself. Legalese is famously opaque to non-specialists. There are many cases where a law is difficult to interpret for a layman and we should absolutely cite a secondary source to interpret what that law means. If e.g. an American law refers to a en:public figure, it is potentially misleading to simply quote the law without providing a secondary source which explains what that means – a public figure in law is not necessarily the same as is meant by the phrase in a non-technical sense. Caeciliusinhorto (talk) 22:18, 27 March 2025 (UTC)

@Caeciliusinhorto "providing a secondary source which explains what that means" - Then what you are quoting is a primary source (for the explanation). It is not a secondary source anymore. And as a good practice you should also include the original text the explanation refers to - using its primary, original source, and not the secondary one, which you have no idea if it's quoting right or not (and there's no point in using it for that, anyway, if you have access to the original one). I'm using the definition at primary source, btw, which is also in line with the way we usually define them in Portuguese, so I believe there's no language barrier issues involved here. - Darwin ^Ahoy! 22:29, 27 March 2025 (UTC)

Your insistence on calling this a primary source because it is primary for the interpretation is deeply confusing and unhelpful. In normal Emglish the fact that it interprets a primary source is precisely what makes it a secondary source.

As for the idea that it is good practice to include the primary (in the sense of the policy we are discussing) source: that isn't always all that helpful. In the case of Aspasia the primary source (e.g. Plutarch's text) is written in ancient Greek. To provide a version which is actually checkable by the kind of person likely to actually read our article, we need to filter it through several layers of interpretation already (we don't have an autograph manuscript, so we have potentially multiple steps of manuscript transmission and then editing before we get the "standard" Greek text, which is of no use to the layperson so it would be better to provide an English translation. Every step in that chain of transmission is more doubtful than that a modern scholar publishing in a peer reviewed journal has misread the source badly enough to significantly impact their interpretation. And this is by far our most straightforward primary source: the fragments of comedy involve several more layers of transmission and interpretation). Even if we provide that, how is our layperson to decide if the conclusions drawn by our scholar are valid? At best this is supplemental to the scholarly sources; at worst it is liable to lead the layman into active error where they misunderstand a difficult-to-interpret primary source. Caeciliusinhorto (talk) 23:01, 27 March 2025 (UTC)

@Caeciliusinhorto In the case you mention the original source is not easily available, so you have to work with what you have, and your solution probably is the best - noting, however, that the so called "secondary source" in that situation is acting like a primary source as well - as it is the closest to the original a common person can get. But in rather common situations such as the tweet I mentioned above, a country legislation, a 19th century chancel, etc., ideally you use 1) the original source for the fact itself 2) sources that provide insight, analysis, etc. over it. Both are primary (original) sources relatively to the information they provide, even if they have different degrees of proximity to the subject they are dealing with. - Darwin ^Ahoy! 23:13, 27 March 2025 (UTC)

You seem to be distinguishing between secondary literature and secondary sources, a distinction which is not adopted in WP:PSTS. Fgnievinski (talk) 07:00, 12 April 2025 (UTC)

@Fgnievinski It seems to be what is followed in Primary source, though. Apparently the Wikipedia community has followed some fringe definition of what secondary sources are, eventually mixing a number of different contexts together in some kind of original research of themselves, and elected them as "most reliable" very early in the process, poisoning everything around since then. - Darwin ^Ahoy! 14:37, 18 April 2025 (UTC)

@Caeciliusinhorto On the case of Aspasia, you should always, at all times, use primary sources for that. The older ones to state what they say, that she was "put on trial for impiety", and the more recent ones to provide the modern interpretation, of which they are the primary sources themselves. Otherwise, without checking and using the primary source, you risk using a "modern interpretation" of a misquote of the original, which while obviously unreliable (having been poisoned from the start) is fairly common on that area of research, as lazy researchers tend to parrot other parrots, or they themselves may have distorted the original reading. Secondary sources have no place there, as well, they are always of inferior quality. - Darwin ^Ahoy! 21:55, 27 March 2025 (UTC)

The real "unfortunate acritical import of the deficient English Wikipedia policies" lies in the import of American values, in particular, the concept of "bright lines". Military historians like myself are trained to weigh the value of different sources critically, but on Wikipedia obstacles are placed in the way of that. The usual ethos is to prefer primary sources over secondary ones. But there are other restrictions on the correct use of sources, such as the absurd blanket prohibition on using master's theses when many of them are excellent. Hawkeye7 (talk) 03:40, 28 March 2025 (UTC)

@Hawkeye7 More than a decade ago we faced the same concern about master and degree dissertations, and even doctorate thesis, specially as in Portugal and Brazil it is common to find such works of very deplorable quality, even if all of them have some degree of peer review. Initially the community tended to refuse them, but since there is some peer review (and sometimes very good peer review) we now accept them, but only as primary sources. Secondary sources, e.g. a 2017 doctorate thesis claiming that a witch was burnt in São Paulo in 1765 without even presenting the primary source where they got that from, are obviously unreliable, as are dissertations sourcing stuff to 16th century documents when that information is nowhere to be found on the mentioned primary sources. So we use them evaluating their reputation on a case by case basis. Sometimes they are of excellent quality, sometimes it's just made up crap that somehow passes the revision. But I totally agree that such sources can't be dismissed just because of their nature, in particular the doctorate thesis which usually face a quality juri before being released. - Darwin ^Ahoy! 12:50, 28 March 2025 (UTC)

My own doctoral thesis was reviewed by three international experts. (The master's thesis was reviewed by two, but is more often cited.) I generally regard secondary sources as unreliable, and so double-check the facts against primary sources whenever possible. I pass over popular books that don't have footnotes for this reason. But the purists decry this practice as "original research". Hawkeye7 (talk) 22:25, 28 March 2025 (UTC)

Darwin, you write: „a significant increase in the reputation of the project in the Lusophone world“ - my odd question in this context: can you provide any citations? Regards Sargoth (talk) 20:34, 29 March 2025 (UTC)

@Sargoth not a direct citation, but a lot of meaningful hints. By 2018 or so Kondzilla, Sony and other well known record labels acting in Brazil started requesting a bio of the artist in wiki.pt as a prerequisite for signing a contract with them. We know that because a lot of them were caught paying people or even attempting themselves to create their bio in order to get the contracts. Also, criticism online dramatically changed from "Wikipedia is unreliable" to "Wikipedia is biased", which tells a lot. And well known companies and political pundits, specially from Brazil, but also from Portugal, started paying people to make or improve their articles on the platform. It also became much more easy to get partnerships with cultural and educational institutions. And (as a not so good new development, but also related to that) way more people started caring enough about our content to sue us when they disagreed with the information in the article - myself included. And these are just a few, there are a lot more. Nothing of this happened until 2013-2014, AFAIK. Self respect and esteem by that time was so low that you can find media pieces with power Wikipedists saying themselves that the quality of the project was low, that it was not reliable and it was kind of a weird hobby they had. It was indeed a dramatic change, which can be directly attributed to the draconian approach to sourcing information, which was the only meaningful thing which changed since then in our policies. - Darwin ^Ahoy! 21:02, 29 March 2025 (UTC)

Thank you for your answer. Anecdotal evidence is not totally invalid, but hard external sources are generally more convincing. Could you at least link to some of those criticism online please? The last argumentation, change of overall behaviour since 2014, can also been observed in Germany although nothing changed source-wised (sources are mandatory, but old unsourced articles or with sources listed as literature [there is no "further reading" section] are left untouched, more or less) Best regards Sargoth (talk) 21:37, 29 March 2025 (UTC)

@Sargoth Here is one from the Right and another from the Left, including a note from the Brazilian Press Association claiming Wikipedia is biased for not blindly accepting stuff from different POVs (in fact, obviously biased sources, such as that 247 platform). There are more, but those 2 were very famous. As for Portugal, there was this case, among others, where Wikipedia was condemned by Portugal Supreme Court to delete an article of a Luso-American businessman and reveal the identity of its editors (a condemnation that we never respected, btw). From what I recall, the process directly mentions a number of times the reputation and impact of Wikipedia as a source of information and therefore the need to exclude that particular information from there. Furthermore, a cursory search on the Portuguese jurisprudence reveals that Wikipedia is routinely used as a source of information on court (e.g., by the Supreme Court itself).

Like wiki.de, we also didn't remove immediately all the unsourced content that already existed (and even new one added afterwards), but anyone willing to do that can do it without being hindered by editors claiming "I know it is true", "I have seen it in the source and will eventually add it at some point" and whatelse - and a number of editors do that kind of maintenance in a daily basis. Anything unsourced or badly sourced can be immediately discarded and can't be reinstated unless a proper source is there. It ended for good all disputes related to that, which were a lot, a true bliss. And there was this very noticeable increase on reputation. - Darwin ^Ahoy! 22:11, 29 March 2025 (UTC)

Criticism

WMF is being played by the quasi-fascist movement that has taken over the Republican Party and the government of the United States. There is nothing wrong with Wikipedia's established guidelines and policies and our twenty years of actual practice. The Trumpist far right is attempting to subvert and suppress any and all public sources of information that contradict their "alternative facts" counternarrative and they are using allegations of "bias" as a mechanism to subvert and silence Wikipedia. WMF is playing right into their hands by even wading into this so-called "issue" — they are probably pissing down their collective legs over Katharine Maher having to testify in front of Marjorie Taylor Greene & Co., worried that their turn is next. Well, it is coming and we expect a lot more than the abject repentance and total surrender that she offered. So let's just drop this whole matter, shall we? —Tim Davenport, Corvallis, OR ///// Carrite (talk) 00:27, 28 March 2025 (UTC)

The existential threat WMF sees is the division of Wikipedia (and possibly the internet) into American and non-American parts. To most Americans it is a given that they invented the Wikipedia and the internet, and they deserve to control it. This comes up regularly on the English Wikipedia when American admins attempt to shut down discussions before editors in other time zones can respond. Hawkeye7 (talk) 22:54, 28 March 2025 (UTC)

"There is nothing wrong with Wikipedia's established guidelines and policies and our twenty years of actual practice" - Yes., there are many issues, actually. And not of lesser importance. I've seen first hand editors there so afraid of talking about certain (usually culture war related) themes - which are actually very much consensual out of that bubble - that they would beg me to not answer them back after contacting me. And obvious evidence of canvassing on WP Discord to forcefully pass fringe theories. Among other things, like the perpetual confusion between primary and secondary sources mentioned above. it doesn't mean in anyway that Taylor Greene or Musk are correct, but yes, there are a number of issues which obviously affect neutrality and should be openly discussed, indeed. - Darwin ^Ahoy! 15:58, 29 March 2025 (UTC)

@Carrite: I suggest it is at least possible the Foundation has more than just English Wikipedia in mind. The word 'global' in the title is one hint in support of this suggestion. In my work I have had occasion to observe quite a few Wikipedias other than English Wikipedia, and I can assure you that the mainstream view and practice of NPOV on English Wikipedia is far from universal across the projects.

Even if you believe status quo on English Wikipedia is perfect, I encourage you to consider the possibility there may be projects with a more tenuous grasp of the NPOV policy, often related to group bias (e.g. in Wikipedias in languages spoken in only one country), which is harder to combat in an insular language community. Asaf (WMF) (talk) 15:10, 1 April 2025 (UTC)

Limit the number (and/or content) of contentious articles

Svwp is like the 15th biggest version and while we have enough admins and patrollers, we do not have enough number of editors for contentious subjects. So our challenge is not how the NPOV policy is written (it is about right) or conflict resolution process but on how to preserve energy for discussion of contentious subject. Our trend is now to limit the day-today contentious content, which makes up less then 0,01% of or total. We did not accept anything of Musks raised right arm for the first six weeks, and after that only with three words in connection to another thing. I am proud of our 99,99% content, not being contentious, and am worried these can be less read because of the 0,01% Yger (talk) 07:34, 28 March 2025 (UTC)

That's very interesting, @Yger! Can you share a little more about how this restriction works in practice? Are edits on contentious topics reverted and users censured? By whom? Who decides what is contentious? When does it become permissible to address it? (is six weeks the standard interval?) How was this policy/habit arrived at? Ijon (talk) 15:13, 1 April 2025 (UTC)

Svwp is not so big, We are only 15-20 ("overseers") who are really active in the managing of this types of discussions, and in this case it was one of these supported by me who stated "it should not be mentioned, before context is secured" supported by one or two others who stated Wp should not be a news media. And then the 3-4 active patrollers deleted any inclusion of this (until six weeks later). We had our worst breakdown in a hard discussion for about eight years in December, and after that all want our work space to be friendly and supportive to each other, and therefor agree to limit/avoid discussions on contentious subjects meaning we avoid write or accept entries on this type of subjects (and we have a log tradition to avoid write articles of news). Yger (talk) 15:39, 1 April 2025 (UTC)

Thank you very much! Ijon (talk) 15:46, 1 April 2025 (UTC)

A working group of who to do what?

Per the Diff post, the plan here is to "convene a working group of active editors, Trustees, researchers, and advisors to explore recommendations for common standards for NPOV policies".

Can there be some clarity on what in tarnation this is supposed to mean? It is written vaguely enough to allow a wide variety of interpretations, and some of them are rather concerning, although I am not sure what was intended. JPxG (talk) 05:31, 1 April 2025 (UTC)

WP:BEANS + Carrite's thread above, be it right or wrong... Zezen (talk) 07:28, 1 April 2025 (UTC)

I am looking for an actual answer. JPxG (talk) 00:14, 2 April 2025 (UTC)

@Mayur-WMF: I'm writing up a summary of this announcement for the Signpost in a couple days -- could I get some clarification on what this section of the post means? Is there any information on who the "advisors" are to be, and what role is envisioned for this working group (and its common standards) with respect to local project governance? Is this meant to provide resources for projects who request assistance, or is it compulsory? JPxG (talk) 00:20, 2 April 2025 (UTC)

Hi @JPxG:, the goal of this working group will be to review neutral point of view (NPOV) policies on the Wikimedia projects and propose recommendations for common standards. The first step will be an analysis on the state of neutrality policies and principles across Wikimedia projects. This will include opportunities for volunteers to share links and suggestions on what resources might be useful. Based on this analysis of existing policies and working closely with communities, the working group will develop specific recommendations for common standards. In terms of advisors, we will invite researchers - who share Wikipedia's goal of encyclopedic content written from a neutral point of view - to contribute to this work. We are planning to share more information in the next two weeks by mid-April. NSzafran-WMF (talk) 21:10, 3 April 2025 (UTC)

Globalisation is difficult

On the English language Wikipedia we have a reliable sources noticeboard and our policies are neutral between free and paywalled sites let alone offline sources and between English and non English sources. But in practice we are much more likely to cite a free site that is written in English than a paywalled or non English source. I think the reliable sources noticeboard works well and I'd recommend that model to language versions that don't yet have such a resource. Perhaps one outcome from this initiative would be to encourage or enable closer cooperation between different language versions, and in some cases even rely on the relevant language Wikipedia to make decisions as to whether or not a particular source is reliable. Though as someone who speaks neither Arabic or Hebrew I rather expect that those two language versions of Wikipedia will have some topics where the concept of a neutral point of view and a reliable source may be at odds. On the one language I speak sources will vary widely on some topics when you compare say The Irish Independent and Deseret News not only do you see different emphasis, but like the famous Dundee headline for a report of the sinking of the Titanic "Local man lost at sea", many sources have a geographic bias to their coverage. The WMF has made some significant steps towards breaking the paywall barrier with the Wikipedia Library. But last I checked its priority seemed not to be on our most controversial topics. I think the WMF could usefully use a small amount of its income to buy subscriptions for editors in controversial topics, and especially subscriptions to factual sources with contrasting biases. This could even be a condition for people active and problematic in a particular topic that they are given a subscription, and for a period of time are limited to only editing in the controversial topic where they can cite their edits to articles from a subscription to a source with a different POV to theirs. Going back to the "local man lost at sea" trope, it is understandable that an Irish Newspaper will pay more attention to Irish peacekeepers being visited or Irish citizens being deported, a British Newspaper once said "fog in channel, continent cut off". We can be explicitly non parochial. Neutrality for a global project should require us to treat all lives lost as equally "local", equally important, or better, the death of civilians, children, health workers and journalists as more tragic than the death of combatants. But we also need to be able to spot jokes, satire and hyperbole and such things aren't always as clearly presented as a cartoon in the way that "fog in channel, continent cut off" was. WereSpielChequers (talk) 23:26, 1 April 2025 (UTC)

At enwiki, Wikipedia articles are required to be neutral, but the reliable sources themselves are not. For example, a neutral article can be created from "This political group says X" and "The other political group says Y", as long as you fairly, proportionately, and accurately represent all of the viewpoints in significant sources.

In terms of non-English content, I believe that English has become the lingua franca of science and scholarly sources in general, so the days when (e.g.,) chemists should learn German to keep up with their field, but poets need to learn French and philosophers need to learn Latin are probably gone. Non-English sources are probably more useful at the English Wikipedia for recent events and for "niche" articles. We can probably write about an article about an African country, like Nigeria from only English sources, but we probably can't write about a niche subject, like an important traditional Igbo religious leader or a locally popular Hausa author, from only English sources.

The Wikipedia Library provides a lot of non-English sources. Here's some examples:

• Sabinet [1] – includes Afrikaans and other African languages
• Al Manhal [2] – Arabic scholarly papers
• Taaghche [3] – Persian books and newspapers
• Magiran [4] – Persian newspapers and scientific journals
• Les Jours [5] – French media
• Haaretz Hebrew [6] – Hebrew newspaper
• TheMarker [7] – Hebrew business news
• Economic & Political Weekly [8] – variously in Hindi, Tamil, Kannada, Telugu, and Marathi

What else would you want to see? WhatamIdoing (talk) 04:58, 6 April 2025 (UTC)

• “In terms of non-English content, I believe that English has become the lingua franca of science and scholarly sources in general, so the days when (e.g.,) chemists should learn German to keep up with their field, but poets need to learn French and philosophers need to learn Latin are probably gone.”
  • This is only correct in some situations, probably not all.
• “Non-English sources are probably more useful at the English Wikipedia for recent events and for "niche" articles.”
  • Not really, though we may have different definitions for the word “niche”.
• “We can probably write about an article about an African country, like Nigeria from only English sources”
  • But that won’t be a good and comprehensive article.
• “but we probably can't write about a niche subject, like an important traditional Igbo religious leader or a locally popular Hausa author, from only English sources.”
  • Is a locally popular scientist a “niche” subject? (E.g., [9] .) Even if it is, if someone can create an FA for s/he, the person probably won’t be that “niche” a few years later.

That said, I don’t have any opinion on the coverage of non-English sources of TWL as I don’t know much about it. Dustfreeworld (talk) 16:59, 6 April 2025 (UTC)

There is no one Wikipedia, and that's good

English Wikipedia has its NPOV policy - German Wikipedia has another. Each of Wikipedia's hundreds of communities has developed a standard that works for them via decades of discussion and consensus. There are established procedures for changing these policies. How could any centrally-developed recommendation for NPOV ever hope to be adopted?

Wikipedia's strength lies in its autonomous communities of editors. I could potentially see some value for this project for very small Wikipedias, perhaps providing a "standard template" NPOV policy for a new Wikipedia to adopt. But once a community reaches a reasonable size, it belongs to its editors, in a very real way. WMF cannot write Wikipedia, in any language. Perhaps it's time to think about a true 2-tier system, where the global movement, supported by the WMF, provides more direct guidance for the many tiny wikis, while the bigger Wikis (those with over 500 active editors, say) take care of themselves, as they currently do.

But frankly, I'm seeing a solution in search of a problem here. We've had one serious issue of "capture" of a Wiki which led to a departure from NPOV - on Croat Wikipedia. That was dealt with, and years ago. What is leading to this discussion now? Ganesha811 (talk) 19:34, 9 April 2025 (UTC)

I will make a general comment about this new NPOV policy idea and responding here seems reasonable. Is the point of the idea to simply write a nice sounding NPOV policy page for each project, or improve/enact/enforce NPOV globally? It would be interesting to evaluate the Croatian Wikipedia NPOV policy before, during and after "capture" (policy page history).

The problem is probably not total destruction of NPOV at a Wikipedia, but partial degradation which is harder to detect and insidious.

I would take a guess that a current problem is around the Israeli-Palestinian conflict: eg the Severe Problems in hewiki RfC and the "Editing for Hate" post at the Wikimedia Forum.

One way to evaluate NPOV would be to look at article interwiki links, titles and redirects. A bot could look at the Israeli-Palestinian topic across hewiki, arwiki and enwiki. Israeli settler violence (Q6087660) has articles in seven language Wikipedias, but not hewiki for example (my discovery) Commander Keane (talk) 23:40, 9 April 2025 (UTC)

@Ganesha811 Even for small/starting projects I don't think that is either needed or even helpful anymore than the general principle of neutrality already enunciated in the pillars.

I 'll share briefly my experience while providing support to the Emakua (Macua) Wikipedia community in Incubator, a community of native speakers consisting on students and academics from the Nampula and Cabo Delgado regions, in North Mozambique, and what was transmitted to me by the academics in my working group. It's a very common situation, and it's essentially not different from many other so called "minority languages", such as the one we have in Portugal, Mirandese:

• Macua is a spoken language mostly used in domestic and rural contexts, without use in a written form outside academic circles - so there are not written sources in Macua. As it's not used for stuff apart from rural & domestic activity, it had never incorporated concepts from contexts different from their own, as there was never such a necessity. This means you don't have any written sources in Macua, and written sources in other languages about the Macua culture are difficult to use, not only because they are often biased or written by outsiders that only looked at teh surface of things, but because many terms used there don't have translation to Macua. So much for our traditional approach to neutrality, it had to be discarded.
• What we actually used as a "reliable source" were the teachers/academics that were with us, which were proposing on the fly new Macua words which could eventually work as translations, and providing structured narratives about the Macua culture that could be written into the Wikipedia articles by the students. This process was being done for the first time in history, by their own information.

Now someone tell me how something more elaborate than what is already written in the 2nd pillar ("We strive for articles with an impartial tone that document and explain major points of view, giving due weight for their prominence. We avoid advocacy, and we characterize information and issues rather than debate them. In some areas there may be just one well-recognized point of view; in others we describe multiple points of view, presenting each accurately and in context rather than as "the truth" or "the best view"") can be used in such contexts - and even there what comes after these initial two sentences, referring verifiable citations and no original research had to be discarded, as it didn't worked there.

Frankly, I also fail to see how something meaningful can be achieved in this debate apart from reinforcing the concept that each project has it's own way to reach neutrality, and even different notions of what is neutral based in their own context and the sources more easily available to them (and the reliability of those is yet another debate). And that we must strive to "be neutral", whatever that means. And that's all. - Darwin ^Ahoy! 09:37, 10 April 2025 (UTC)