Wikipedia talk:Neutral point of view/Archive 65

Archive 60

←

Archive 63

Archive 64

Archive 65

Archive 66

Archive 67

Archive 68

Sanger criticism of the NPOV policy

Today someone posted a link in the thread Talk:Larry Sanger#Recent Sanger criticism of Wikipedia about Sanger (a Wikipedia co-founder) criticizing the NPOV policy and in general the neutrality of Wikipedia, which he indicates "no longer has an effective neutrality policy". Just thought it may be of interest as it directly relates to the NPOV policy. Regards, Thinker78 (talk) 22:30, 16 October 2023 (UTC)

Despite a few overreach implied claims (eg implying that Wikipedia has deliberately abandoned neutrality) I think that it is a good look at ourselves. In certain areas our NOPV policy implements/entrenches POV rather than avoiding it. IMO this is because it has not evolved to keep up with significant changes in sources/journalism. Most notably the removal of money from journalism by the internet (and resultant decline) and the evolution of sources from Walter Cronkite to biased advocates with far more biased coverage and, more importantly, far more biased inclusion/omission. Of course some folks will think "we like it that way" and seek to deprecate Sanger and voices like him but I would appeal to them to rise above that and just do the best that we can do here. Sincerely, North8000 (talk) 23:05, 16 October 2023 (UTC)

Good points. Journalism certainly has declined a lot and now an oligarchy has control over most of the main news media. In regular life, the counterweight is the innumerable sources of raw news in social media. But in Wikipedia, it translates to using sources for certain articles that are highly biased, opinionated, and not independent, while deprecating truly independent commentators. There are other issues overall like soft official censorship, politics and conflicts of interest that may be driving editorial practices in some scientific journals. Regards, Thinker78 (talk) 03:22, 17 October 2023 (UTC)

Do you have a source for this supposed journalistic decline? Horse Eye's Back (talk) 14:07, 18 October 2023 (UTC)

I think that your comment is spot-on, @North8000. I've been thinking a lot about this, and I can tell you have too. I also think Sanger's critiques get blithely dismissed by a lot of Wikipedians because of Sanger's comments about American politics. That is a shame. He's not an expert in politics, and his opinions are a great deal more conservative than most of ours, but one shouldn't "throw out the baby with the bathwater". He is an expert in epistemology, which I think is the most directly relevant field of academic study for crafting a healthy encyclopedic NPOV policy.

I also think this essay by Sanger, entitled "Why Neutrality?" is worth reading and discussing as a community. Philomathes2357 (talk) 03:23, 17 October 2023 (UTC)

I'll point to this Signpost that came out about a year after the linked post. [[1]] (and there's probably more in the archives for Signpost). There is definitely a significant fraction of editors that think Sanger's completely barking up the wrong tree.

While I think he's expecting far too much from what NPOV has generally established, I have to agree with North here (and something I've long said) that numerous combinations of factors has led to what we normally take as unbiased sources to become more biased if only to make sure they keep readership up, fighting against the Fox News effect. And couple that with many editors feeling perhaps too comfortable in writing as negatively as they can when backed by RSes about groups at the extremes of the political and ideological spectrums, and the problem has been magnified. Masem (t) 03:31, 17 October 2023 (UTC)

I'm more concerned about omissions that degrade the mission of the article from providing information than simply biased language/choice of words. Several things interact with each other. WP:RS def + WP:Weight+ Deprecation of sources+the kinder gentler meaning of wikilawyering = major omissions & undercoverage. North8000 (talk) 19:46, 17 October 2023 (UTC)

You are 100% correct. Omission is a bigger problem than biased word choice. This book discusses extensively the "omission" of important material from mainstream western media. Our sourcing policies, especially in regards to modern politics, should be updated to account for the insights from the book.

As for "false balance", I view that as a trivial concern. The current state of affairs (huge amounts of information about the world being thrown down the memory hole for being "unreliable" or "undue") is a much bigger problem than any "false balance" problems that could hypothetically be presented by a modified NPOV policy. Having a balanced NPOV approach to politics that presents multiple points of view does not mean that we'd have to write "according to geographers, France is in Europe". That's a strawman that I see a lot. Philomathes2357 (talk) 22:59, 17 October 2023 (UTC)

That book pre-dates our NPOV policy... So our NPOV policy likely already accounts for that book. But that aside, do you not see the irony of claiming that NPOV needs to be re-written to reflect a single controversial source? Horse Eye's Back (talk) 14:22, 18 October 2023 (UTC)

Obstruction to include material is a fallout of what I have seen as well. FRINGE/UNDUE tends to get abused in this fashion. Basically, it is a lot of riding on the edge of where WP:RGW comes into play, because editors believing the position put forth by RSes is the "right thing" when we should be aware that even our body of RSes can try to position a specific angle and not give the expected objective, neutral coverage they had traditionally been known for. Masem (t) 12:43, 18 October 2023 (UTC)

When were they known for that and when did it change? Horse Eye's Back (talk) 14:27, 18 October 2023 (UTC)

I needed to look up the 1960s New York Times for an article I was editing, and it was such a joy to read. No labels, no emotive language, no concern that the reported facts play into the hands of the wrong side and need to be counterweighted, etc. Respecting the reader and letting them make their own conclusions. Incredible integrity and a truth-seeking approach compared to the pandering to readers' or owners' tastes that the media is forced to engage in nowadays. PaulT2022 (talk) 15:15, 18 October 2023 (UTC)

The 1960s New York Times was not known for its objective coverage of women and minorities nor for letting women and minorities participate in that coverage. Horse Eye's Back (talk) 15:22, 18 October 2023 (UTC)

"I needed to look up the 1960s New York Times" The newspaper notorious for its discriminatory policies? Per the main articles List of The New York Times controversies: "In November 1979, a federal court gave approval to a settlement between The New York Times and a group of female Times employees who sued alleging sex discrimination. The settlement agreement was effective for a four-year period beginning in January 1979; under the agreement, the company amended existing equal opportunity targets and paid $350,000 in compensation and attorneys' fees for the plaintiffs, but was not required to "pay new or retroactive salary increases, make immediate promotions, revoke past employment practices or substantially change its present affirmative-action programs."" Dimadick (talk) 22:44, 18 October 2023 (UTC)

I'm not sure I see how mistreatment of female and minority employees is related to the fact that news coverage in NYT changed from factual to emotive.

In retrospect, I should've linked https://www.rand.org/pubs/research_briefs/RB10059.html instead of providing an example. PaulT2022 (talk) 23:04, 18 October 2023 (UTC)

Maybe what these editors are trying to communicate to you is that an unemotional style is not the same thing as unbiased content. Wikipedia editors should be able to filter out stylistic factors pretty easily (e.g., skip the dramatic descriptions – perhaps you're thinking of things like "In the end, the complexity of the arrangement almost killed the deal to free the hostages. Officials called it the largest private financial transfer in history. It was surely the most intricate. But for technical details, the hostages could have been released sooner."), but they can't turn "facts about upper-class privileged people" into "facts about everyone". WhatamIdoing (talk) 03:31, 19 October 2023 (UTC)

I'd agree, but it's often more nuanced.

To use your Cinnamon Roll Day example from the thread above, a source presents an expert quote about links between buns and nationalism. In the source, it's presented tongue in cheek (It probably doesn’t hurt that the sugary buns taste good, too.) In Wikipedia, it becomes a serious expert statement on Swedish anti-immigration sentiment. The problem with this style of journalist writing is there's often no hint in the source to what extent the quotes are representative of the mainstream thought, and, more importantly, context that would be present in academic sources is often missing in the modern news. (I.e. it's quite possible that the expert opinion is entirely mainstream in application to the modern following of the day, but may not apply to the time the day was instated in the 90s.)

Another editor in this thread complained that Hamas terrorist attacks would often be described as "massacres", while mass civilian deaths by the other side would use formal terms such as "airstrikes". I could easily see how an attempt to filter this out per your suggestion would be called out as WP:RGW. To use an example "for the other side", there's a story in Times today how Hamas statements were attributed to "Palestinian officials", which I think a lot of people may assume to refer to PLO officials rather than Hamas leadership.

I agree that it's all stylistic choices. Formally speaking, they don't make these sources any less reliable for facts, and they don't introduce unworkable bias.

But they do increase workload on the editors, who have to differentiate between reliable statements on facts and reliable statements on opinions (and very often you'd only have the latter!), as well as the meaning behind the terminology used. It makes it immensely hard to write neutrally and don't introduce original research when collecting and arranging the opinions – often entirely unwittingly. PaulT2022 (talk) 11:43, 19 October 2023 (UTC)

«Another editor in this thread complained that Hamas terrorist attacks would often be described as "massacres", while mass civilian deaths by the other side would use formal terms such as "airstrikes".» => I bet that Larry Sanger agree with this wording. Visite fortuitement prolongée (talk) 18:39, 19 October 2023 (UTC)

I would indeed prefer terrorist attacks to be called terrorist attacks and terrorist organisation spokespeople to be called terrorist organisation spokespeople regardless of what Larry Sanger or anybody else thinks about it.

If anything, Larry Sanger is a perfect example of why indiscriminate use of opinions from news sources is a problem. He gets interviewed only because media like to print "Wikipedia co-founder says don't trust Wikipedia" headlines. He's not a recognised expert on Wikipedia, not part of Wikipedia itself for many years, and yet his opinions are regularly featured by major news outlets to this day. PaulT2022 (talk) 19:28, 19 October 2023 (UTC)

PaulT2022, very well said. We need an FAQ, and the first entry should read: "1. Any mention of Larry Sanger will deleted. His views have been exhaustively discussed and only produce more heat than light. He is not an authority on the subject. Do not waste our time by opening threads about his views on NPOV or by citing him." -- Valjean (talk) (PING me) 23:13, 19 October 2023 (UTC)

So the white male perspective is factual and all others are emotional? Horse Eye's Back (talk) 15:14, 19 October 2023 (UTC)

I don't think the tendency that RAND calls "truth decay" is caused by hiring more non-male and non-white journalists. What an absurd idea. PaulT2022 (talk) 20:06, 19 October 2023 (UTC)

RAND does not appear to comment on the 1960s New York Times and is discussing a tangential concept. Do you think its just coincidence that in the same period you feel that the NYT went from factual to emotional was when the news room diversified? Horse Eye's Back (talk) 20:09, 19 October 2023 (UTC)

I said emotive, not emotional. I believe the shift towards emotive writing to be driven by commercial considerations and pressures from change in how news are consumed (Internet/short attention span vs reading a printed broadsheet).

I understand where you're coming from, but I don't have an interest in discussing biases for the simple reason that it isn't our job to correct them.

As for the style, the RAND discussion isn't tangential: evidence of a shift from a journalistic style based on the use of public language, academic register, references to authority, and event-based reporting to one based more heavily on personal perspective, narration, and subjectivity. ...the three newspapers’ reporting before 2000 used language that was more heavily event- and context-based than it was in stories written after 2000; pre-2000 stories also contained more references to time, official titles, and positions and institutions and used more descriptive, elaborative language to provide story details. In contrast, the team found that post-2000 reporting engaged in more storytelling and emphasized interactions, personal perspective, and emotion more heavily than did stories in the pre-2000 period

My belief is that it matters and the modern style is a lot harder to use neutrally (representing POVs proportionally and without editorial bias) for the reasons I explained above. PaulT2022 (talk) 20:48, 19 October 2023 (UTC)

How are sources which more fully describe the various POVs harder to use to represent POVs proportionally and without editorial bias? The RAND study also says that many of those changes are not negative but related to the internet, you don't need to provide a mini-bio of a figure which someone can copy-paste and google for example. You also need less elaborative language when you have more pictures, maps, and graphics. Horse Eye's Back (talk) 21:16, 19 October 2023 (UTC)

The increasing share of what's reported has become more about who said what ("personal perspective and emotion," as RAND put it) rather than the facts verified by journalists and publishers. As a consequence, editors have to make judgement calls that I would prefer to be made by reliable sources.

First, editors have to make a judgement call on whether to use an attributed statement or not, based on the perceived reliability of the quoted speaker rather than the source.

Second, editors have to decide whether a certain stylistic choice (the aforementioned example of calling Hamas "Palestinian officials" is one recent example I find particularly egregious) is a statement of fact made by a reliable secondary source or a subjective label. PaulT2022 (talk) 22:41, 19 October 2023 (UTC)

First... Thats not actually something that we get to do, we can assess weight but the reliability of the quoted speaker has absolutely no bearing on anything. Second... No we don't... We don't second guess the source like that nor do we do original research. Your argument is divorced from the reality of editing competently and would really only be useful if one were engaging in OR. Horse Eye's Back (talk) 16:38, 28 October 2023 (UTC)

I don't think that's quite true. Editors have to make a judgement call about whether to include everything. Therefore, editors have to make a judgement call on whether to use an attributed statement or not. If their judgement says "Sure, I might include that if Apple's Tim Cook said that, but I won't include it if Elon Musk said that, because Musk has a reputation for saying all sorts of silly things, and besides, the subject of the article is Apple, and who cares about what Musk thinks about Apple" – well, we've just made a judgement call about whether to include something based on the perceived (un)reliability of the quoted speaker. WhatamIdoing (talk) 02:17, 31 October 2023 (UTC)

That doesn't appear to answer the question, I asked about NPOV not factuality. Horse Eye's Back (talk) 15:15, 20 October 2023 (UTC)

Sorry, I could've been more explicit. These issues result in the effective NPOV of Wikipedia articles being determined by the quotes reprinted by the media and the specific subjective language news editors choose to use. The quotes and statements in news are widely considered 'reliably sourced,' (although they're much closer to opinion columns in terms of the editing and fact-checking process) and each expression of opinion quoted in reliable sources ends up being considered in determining NPOV weight. This omits the reality that, because they are quotes and stylistic choices, they are not fact-checked in the same way as a peer-reviewed academic article or a journalistic investigation written in the publication's own voice would be, for example. PaulT2022 (talk) 15:36, 20 October 2023 (UTC)

"the effective NPOV of Wikipedia articles being determined by the quotes reprinted by the media and the specific subjective language news editors choose to use." isn't that what happens in either scenario? You're basically describing how NPOV works, note that the "editorial bias" in the wording is our own bias not that of the publishers. Horse Eye's Back (talk) 15:56, 20 October 2023 (UTC)

PaulT2022, such opinions get attributed when we include them. We document facts and opinions, truths and falsehoods. If the opinion is factual and many other RS agree with it, we may not attribute it as it's a "sky is blue" situation. -- Valjean (talk) (PING me) 15:41, 20 October 2023 (UTC)

I think it's a problem, because "many other RS agree with it" is determined in practice by what opinions RS feature and what language they use. Which is a good thing with academic sources, but it doesn't work well with modern news publications.

Take something like https://www.thetimes.co.uk/article/larry-sanger-i-wouldnt-trust-wikipedia-and-i-helped-to-invent-it-cflrhmdhx for example.

Someone not familiar with the background would conclude that Larry Sanger is a decorated expert in encyclopedias (the president of the Knowledge Standards Foundation) and Yet now a form of censorship has set in, says Sanger, thanks to a political agenda among a core of its volunteer editors, as well as those working for the Wikimedia Foundation. must be factual because there's no mention in the article that fact-checkers couldn't confirm this or that other views on censorship in Wikipedia exist.

Yet we wouldn't feature this viewpoint because editors choose to ignore the subjective perspective reliable sources chosen to propagate and decide "He is not an authority on the subject". Which is probably a right call in the case of Sanger.

The Cinnamon Roll Day example is another example, where there's not enough context in the source to evaluate whether opinion from a reliable source is being taken out of context.

The result is that Wikipedia's viewpoint on topics is determined by editors' agreement on who to consider an expert, whose quotes in reliable sources are factual, and which ones aren't, and which labels are subjective editorial embellishments and which are factual statements. This is the opposite of how NPOV should work, IMO, as using such sources in practice forces editors to cease being neutral observers and make judgment calls beyond simple source reliability. PaulT2022 (talk) 16:10, 20 October 2023 (UTC)

With regards to the first hypothetical how would the editor arriving at a conclusion not stated by the source not be WP:OR? Horse Eye's Back (talk) 17:12, 20 October 2023 (UTC)

In this specific case, by deciding that the expert isn't authoritative on the matter, despite multiple reliable sources presenting his subjective views without qualifying them as fringe.

Imagine someone searching "Wikipedia bias" in Google News and attempting to rewrite Wikipedia#Coverage_of_topics_and_systemic_bias according to "the prominence of viewpoints in reliable sources". The section would turn into a disastrous hit piece, if written by the letter of NPOV. PaulT2022 (talk) 18:17, 20 October 2023 (UTC)

I'm asking how they arrive at the conclusion that "Larry Sanger is a decorated expert in encyclopedias" if the source doesn't say that. Horse Eye's Back (talk) 18:44, 20 October 2023 (UTC)

Because his viewpoint is featured by multiple major newspapers.

That's my point exactly.

Editors indeed do own research to decide whether he is an authoritative expert or not. They don't follow NPOV's "prominence in published reliable sources" to determine his opinion's weight. PaulT2022 (talk) 19:05, 20 October 2023 (UTC)

We are required to follow NPOV's "prominence in published reliable sources" to determine his opinion's weight. Are you confusing NPOV with WP:RS? You seem to be describing the process by which we would conclude whether Sanger is a reliable source or not... Not the one by which we would determine what amount of weight would satisfy NPOV. Horse Eye's Back (talk) 19:09, 20 October 2023 (UTC)

I like your approach, but Sanger isn't the source. The sources are The Times, The New York Times and thousands of other newspapers who publish his point of view.

If there were a policy or guideline stating that a quote published by a reliable source essentially equals WP:RSOPINION for the purposes of establishing reliability and viewpoint prominence, it would prevent many ad-hoc disagreements. PaulT2022 (talk) 22:56, 20 October 2023 (UTC)

Then why are you treating them like a source? Horse Eye's Back (talk) 00:51, 21 October 2023 (UTC)

Something like Andre's proposal to make explicit that DUE WEIGHT is proportional to RELIABILITY would do it. PaulT2022 (talk) 23:09, 20 October 2023 (UTC)

Its an interesting proposal, but you can't make explicit something which doesn't exist at all... We would have to write new policy because current consensus is that weight is not proportional to reliability. Horse Eye's Back (talk) 00:51, 21 October 2023 (UTC)

This is basically a major issue right now - editors have a personal viewpoint that happens to align with some of the reliable sources out there, and thus they push on that point claiming that the RSes back them. Instead, we should keep our personal feelings about a topic out of writing about it, and instead look to see what type of commentary develops naturally from what the whole of reliable sources say. Cherry picking sources is the most basic version of this, but I have too often seen editors (individually or as a group) trying to push a WP:RGW-type approach because they feel the need to label and criticize these people - the most recent case being trying to deny calling Andy Ngo a journalist because some fraction of sources didn't call him that.

There is a very thin but blurred line between following NPOV and UNDUE as to include well-sourced commentary as well as exclude very minor views, and trying to edit in a RGW approach that puts far too much weight on some aspects that may not fairly follow NPOV/UNDUE but because its coming from reliable sources, its forced into these articles. Sometimes its hard to see the difference between these which is why most of these cases when this happens are highly contentious, since those editing a bit over the RGW line feel they are following the sources.

What we really need to do is try to avoid including any type of commentary or criticism of topics (particularly controversial or contentious ones) until we can naturally see such opinions fall out of the reliable sources. UNDUE needs to reflect that short-term positions are not the type of content we should reflect, but instead put more weight on the long-term. The short-term may end up being the same as the long-term, but its far easier to justify the long-term as being a proper approach to NPOV/UNDUE writing than trying to push things in the short term. Masem (t) 19:16, 20 October 2023 (UTC)

UNDUE is about culling the fringe. Read all the sources, take a reading of the views, and express, proportionately, in a way that distills without any kind of novel expression of synthesis or original research, an accurate essence of topics. An art, not a science. Therefore, we must respect the soft boundaries and blurry lines. Naturally, everyone is going to do this slightly differently, but the bottom line is that Wikipedia is built incremental and piecemeal. We have to be vigilant, and also embrace bold change. We have to protect what's been built, and also keep out the problems and the weakening changes. What that means to me for UNDUE is that it's not that UNDUE needs to be stronger. UNDUE is already used to keep out a lot of things that it doesn't exactly touch on. In my view, we should just make explicit that DUE WEIGHT is proportional to RELIABILITY. Andre🚐 22:21, 20 October 2023 (UTC)

Just to be clear, I'm not against using news sources altogether, especially for lesser details. I'm speaking mainly of the issues that arise when they're used to determine WP:WEIGHT and when articles are written primarily based on news media. PaulT2022 (talk) 16:17, 20 October 2023 (UTC)

Paul, when you write "The result is that Wikipedia's viewpoint on topics is determined by editors' agreement on ...", you are pointing out the single most prominent weakness of Wikipedia and all human endeavors. People are involved. Do you have a solution to that problem? Our safeguard here is that editors of all different persuasions reading all types of sources have an input here. That's why I have a box on my talk page. Look at it. Here's the quote from it: "The best content is developed through civil collaboration between editors who hold opposing points of view." Wikis work on the principle that "No one knows everything, but everyone knows something." Each editor has their own biases, POV, and favorite sources. That is unavoidable. We tend to balance each other, because someone else, who may hold an opposing POV, will point out our errors and blind spots, and may provide sourcing that punctures our preconceived ideas and misunderstandings. That is a good thing. -- Valjean (talk) (PING me) 16:40, 20 October 2023 (UTC)

I agree with everything you say, Valjean.

My qualms are with the NPOV wording being prone to wikilawyering due to the binary nature of the "reliable source" definition in WP:WEIGHT and WP:RS. PaulT2022 (talk) 18:33, 20 October 2023 (UTC)

@PaulT2022's comments are on point. I think you've hit upon some of the most important big picture issues that I've been trying to talk about. I'd like to know what you think of the following:

I think one of the core issues might be that, in NPOV, and even more importantly, in RS, we use the same approach and assign the same labels to scientific sources and political sources. This might be a mistake. The underlying assumptions and methods of the scientific enterprise are not the same assumptions and methods used in modern journalism, or even modern academia in the humanities and social sciences.

One illustrative example among thousands: at an article, this source, entitled "Reorienting Hong Kong's Resistance: Leftism, Decoloniality, and Internationalism", is used, verbatim, in Wikivoice to label a news outlet, The Grayzone as "known for misleading reporting". The specific chapter quoted is entitled "How to Abolish the Hong Kong Police". When I questioned the use of Wikivoice to make a direct quotation from this source, which makes only a passing mention of The Grayzone, I was told that it is a "high quality academic source", because it was published by Palgrave Macmillan.

I accept the premise that this is an academic source, and may fit some definition of "high quality". But my point is that this work is not in any way similar to science. There is no evidence that anything resembling the scientific method was used to reach the conclusions printed in the book. It has a clear point of view (the title should give that away). It is written by people sympathetic to the Hong Kong resistance movement, who wish to promote that movement to a wider academic audience. It may be valuable as a rich, fertile source of academically-framed opinions about the Hong Kong resistance movement. But it is not the same thing as other high-quality academic sources, like a systematic study of butterfly morphology published in the journal Nature.

I think we could use Wikivoice to directly quote a study of butterfly morphology in Nature for an article about a certain species of butterfly. But we should not be using Wikivoice to quote Hong Kong resistance advocates casting vague, passing aspersions at news outlets that have been critical of the Hong Kong resistance. We should not equivocate between these two sources by putting them in the same box of "high quality academic sources". Each one of them may indeed be both high quality and academic, but not in the same way.

For these reasons, I keep coming back to the idea that we should have a separate methodology for assessing reliability and weight, an equivalent of MEDRS, for modern news and politics (NEWSRS or POLRS). That might sound crazy or radical, but it might also be necessary. Philomathes2357 (talk) 02:46, 21 October 2023 (UTC)

Wikipedia has not abandoned neutrality de jure, but community resolution processes often fail to uphold the claim that consensus is based on reasons rather than votes. Usually, numbers carry the day - even when they are based on irrelevant arguments and feelings. Sennalen (talk) 23:26, 11 November 2023 (UTC)

Yeah, a textbook is a high quality source, Philomathes2357. Andre🚐 02:50, 21 October 2023 (UTC)

I agree. Philomathes2357 (talk) 03:06, 21 October 2023 (UTC)

Good. So, since we agree that under present policy, it's a Wikivoice fact that Greyzone is far-left and known for misleading reporting. You can propose POLRS but, I do not think it should or will success if it is on the basis of making the sourcing standards for politics exclude textbooks and news. Andre🚐 03:31, 21 October 2023 (UTC)

No. I agree with your first comment, but I completely disagree with your second comment. Philomathes2357 (talk) 04:56, 21 October 2023 (UTC)

Is this not covered by WP:OR already? "published sources that are directly related to the topic" (emphasis from the original). PaulT2022 (talk) 09:29, 21 October 2023 (UTC)

In theory, perhaps. It comes down to your definition of "directly related". Having too broad of a definition of "directly related" can cause problems, too. If "Reorienting the Hong Kong Resistance" is too directly related to The Grayzone to be usable, wouldn't that mean that we could never use NPR or PBS as a source when writing about US politics? It's a slippery slope, and I'm not convinced that the "OR" policy adequately addresses my concerns.

Note that this is just an example/case study...I could cite dozens and dozens of others.

In practice, sources like this are used all the time, especially when they espouse political views that Wikipedians largely agree with. I guarantee that if "Reorienting the Hong Kong Resistance" praised The Grayzone in passing, rather than criticizing it, the source would have been mocked and torn to shreds on the talk page, or at minimum, thrown down the memory hole for being "undue".

I'm not saying that the source isn't an academic source, or that it isn't high-quality in some ways - all I'm saying is that the enterprise the authors are engaged in is, qualitatively, miles away from "science", such that a blanket description of "high quality academic source" is a suboptimal approach.

This also highlights a topic that was the subject of a very long thread here - the apparent inability of a lot of people to distinguish between "facts" and "opinions". Some folks seem to genuinely believe that a source like "Hong Kong Resistance" is spitting "facts" and not opinions, in the same way that a scientific journal writes facts.

So the argument I see a lot, "oh really, we shouldn't use Wikivoice for this 'high-quality academic source'? What's next, we can't use Wikivoice to say the earth is a sphere?" completely misses the point. So do allusions to "sky is blue", as both arguments draw a false equivalence between scientific academic sources and humanities-based academic sources. That is current policy, but it's untenable. Philomathes2357 (talk) 00:03, 22 October 2023 (UTC)

The assumption apparently made in this comment, that Wikipedia articles draw on high-quality scientific academic sources largely for facts while Wikipedia articles draw on high-quality humanities academic sources largely for opinions, seems implausible on the face of it. I don't think this assumption parses accurately what kind of statements scientific sources make, what kind of statements humanistic sources make, and how either type of source is typically used in Wikipedia articles. Newimpartial (talk) 01:49, 22 October 2023 (UTC)

@Newimpartial, I don't mean to come off as rude, but your summary is actually the opposite of what I'm saying. Let me try again.

What I'm saying is that high-quality academic scientific sources and high-quality academic humanistic sources are both being used for facts, in an equivalent manner. The declarative sentences printed within both types of sources are often summarized simply as "high-quality, academic sources of fact".

Many editors mistakenly take 'opinions from humanities sources to be "facts", because their status as "academic sources" suggests that academic scientific and academic humanistic sources are equivalent in their tone, style, methodologies, and goals.

My objection is: They are both useful, but they are not the same, and should not be treated the same.

Academic humanistic sources typically do not use the scientific method to formulate their declarative sentences. They also tolerate, and in some cases actively promote, the injection of a political and social worldview in a way that academic scientific sources do not. Writers of academic humanities publications are free to express their personal opinions on a variety of social, cultural, and political issues in the author's voice, with declarative sentences, and no explicit indication that opinions, rather than empirical facts, are being expressed.

That's not a criticism, that's a simple observation of the fact that the two enterprises, science and the humanities, are approached differently in the academy - a fact that is not currently reflected in Wikipedia's MOS, but should be. Philomathes2357 (talk) 02:34, 22 October 2023 (UTC)

It's not that consensus academic social science results are facts, epistemologically. It's that Wikipedia treats them as such. It's a convenient fiction to enable the writing of an encyclopedia, a general purpose overview reference work. Andre🚐 02:42, 22 October 2023 (UTC)

Are you confused? They're all sciences... The natural sciences, the social sciences, and the formal sciences. The natural sciences and the social sciences are both based on empirical evidence. Are you perhaps confusing the social sciences with the formal sciences which do not share the same empirical base? Horse Eye's Back (talk) 02:44, 22 October 2023 (UTC)

Philomathes, I am well aware that Wikipedia editors carry a wide range of largely folkloric epistemologies when it comes to facts and opinions, and sciences and humanities. You appear to be claiming that analysis in the humanities reflects political and social worldview in a way that turns the results of such inquiry into opinion, in a sense different from the way that statements of scientific theory are also dictated by their paradigms rather than their supporting evidence. However, I am not aware of any significant support for such a muscular distinction in contemporary epistemology. (Also, I have difficulty imagining how you can continue to participate in this discussion without violating your topic ban concerning politics, broadly construed, if such a topic ban still applies.) Newimpartial (talk) 11:02, 22 October 2023 (UTC)

@Andrevan - thank you for being candid. My question would be, why is Wikipedia in the business of writing "convenient fiction" about important topics? Do you think our readers, who trust us, are aware that we've redefined the term "fact", and that we are relying on "fiction" to write our articles?

Why not treat facts as facts, and opinions as opinions, exactly as NPOV says? I think that creating a novel, unusual definition of "fact" so that we can write political opinions in Wikivoice is a disservice to the reader.

@Horse Eye's Back - that's a semantic trick that misses the point. It's is a good demonstration of the equivocation fallacy that I'm trying to point out.

@Newimpartial - it's not that every declarative sentence in the academic humanities is necessarily an opinion. My point is that they are often opinions, and if we assume that academic scientific and academic humanistic sources are equivalent, as current policy does, most editors overlook these opinions and assume that, since they're "academic", they must be facts.

This has been discussed by other authors, here's just one example. Here's another. Natural sciences, social sciences, and humanities are overlapping, but distinct in their methodologies, including their epistemologies. Philomathes2357 (talk) 19:11, 22 October 2023 (UTC)

If thats a semantic trick then your entire argument amounts to nothing more than a semantic trick. Surely that is not what you mean to be implying. Horse Eye's Back (talk) 16:41, 28 October 2023 (UTC)

@Horse Eye's Back, I think Philomathes2357 was saying that you had engaged in a semantic trick. He was talking about "academic humanistic sources", which you re-labeled as "social sciences" and declared that since (according to you) they're sciences, they're good.

So just so we can talk about the same thing, here are some "social sciences": Religious studies, Comparative law, and Art history. Are these fields that you associate with the Scientific method (i.e., formulating a hypothesis and then testing it in an experiment)? WhatamIdoing (talk) 02:26, 31 October 2023 (UTC)

Yes those are fields which are inherently associated with the Scientific method, thats how almost any research in those three fields is going to be conducted. Horse Eye's Back (talk) 13:56, 31 October 2023 (UTC)

I'm sorry @Horse Eye's Back, that's the wrong answer. They do not conduct their research by making use of the scientific method. Academic journals about art history do not have scientific experiments and hypotheses in them in any way that resembles, say, physics or botany. They're qualitatively different in their methods and goals. Philomathes2357 (talk) 05:21, 3 November 2023 (UTC)

I don't think I've read an Art History paper without a tested hypothesis, they do absolutely use the scientific method. Are you under the impression that they don't have literature reviews and the sort in physics and botany? There is more than one research methodology in academia. Horse Eye's Back (talk) 15:08, 3 November 2023 (UTC)

Really? Here's a link to an art history paper in a mid-range journal from a mid-range academic publisher: https://sciendo.com/article/10.2478/mik-2022-0009 The journal name is Art History & Criticism. Please identify the hypothesis in this article. I can't find one. (I chose this article because it was the shortest [modulo the book review] in the most recent issue of the journal.) WhatamIdoing (talk) 19:54, 4 November 2023 (UTC)

I'm thrown a bit by the workmanlike translation from Persian so I can't say for certain what the author is asserting but doesn't it appear to be art criticism of a historical subject not art history? Horse Eye's Back (talk) 20:08, 4 November 2023 (UTC)

Not every publication in the hard sciences has an easily identifiable and clear hypothesis, either, but that doesn't make it unscientific. Not every paper is trying to prove a new result. Some are review studies, some are replicating others' work, some are simple tables of data and observation. Anyway, as far as this art history article, it's absolutely a rigorous set of observations, with citations. The contention or the evidence being advanced concerns the impact of Western art on Iran. But this is a work of textual analysis, definitely reliable, and definitely academic. Andre🚐 21:09, 4 November 2023 (UTC)

Definitely reliable, definitely academic, and definitely not using the scientific method. Consider:

Does it make an observation? Yes. Does it ask a question? No. Does it offer a testable hypothesis to answer that non-existent question? No. Does it make any predictions at all? No. Does it test those non-existent predictions? No. Ergo, it is not using the scientific method.

There's nothing wrong with that. Non-science is not nonsense. But we should not pretend that every valuable work of scholarship is Science. WhatamIdoing (talk) 02:34, 5 November 2023 (UTC)

I disagree, but we're largely arguing semantics. Science is not defined as anything that makes predictions. It doesn't have to hit every step in the scientific method to be science. The social sciences, as we've established, very often do not make falsifiable hypotheses. That doesn't mean it's not scientific. Andre🚐 03:01, 5 November 2023 (UTC)

Note that the linked page Science includes a section for just such "non-science" as explained in the Social Science section "Due to the limitations of conducting controlled experiments involving large groups of individuals or complex situations, social scientists may adopt other research methods such as the historical method, case studies, and cross-cultural studies. Moreover, if quantitative information is available, social scientists may rely on statistical approaches to better understand social relationships and processes." Horse Eye's Back (talk) 03:06, 5 November 2023 (UTC)

@Horse Eye's Back Exactly. I'm glad you quoted that passage. That is a good explanation why the social sciences, and even more so the humanities, are qualitatively different in their methods and goals.

One can play a semantic trick and say that, since they're all "sciences", they're all the same. But there's a widely recognized distinction in academia between "science" and these other "sciences". Think about it. I've never attended or worked at a college or university where the "science" building housed, say, the sociology or polisci departments. Those are in a different building, often a building shared with the humanities departments. And if someone tells me they have a "STEM" degree, I know they're not talking about art history. In fact, if someone told me in a job interview that they had a "STEM" degree, and I hired them, and later found out that they had studied the "science" of art history, I would feel misled.

They are not the same, and hence my point: I don't think it's appropriate to treat them the same in RS policy by putting them all under a single label, "high quality academic sources". There needs to be more nuance. I'd like to have a conversation about what form that nuance should or could take, but we need to be on the same page about definitions first. Philomathes2357 (talk) 04:00, 5 November 2023 (UTC)

But you haven't given evidence, from this passage or elsewhere, that differences between experimental and observational methods (and many departments housed in the "Science" building use observational methods) result in differences in the kinds of statements made in the resulting published sources, or in how we should treat these sources in Wikipedia articles. When that passage refers to the historical method, case studies, and cross-cultural studies, those are specific observational (or quasi-experimental) methods. The passage does not at all indicate that social sciences are qualitatively different in their methods and goals - your conclusion, unlike the work of humanistic scholars, is not supported by the evidence you offer. Your move from nuances of difference in methodological emphasis to "qualitative differences in goals" looks from here like a semantic trick.

You seem to believe that there is a "qualitative" difference in epistemological status that results in most findings of "sciences" being statements of "fact" and most findings of "humanities" and social science being statements of "opinion". However, you haven't provided any support for this position aside from your own strongly held opinion and a few "semantic tricks". Neither the literature presented in this discussion, nor the overall consensus within the philosophy of science generally, supports treating one set of disciplines as generating facts and the other as generating opinions - such an interptetation seems to caricature and thereby falsify what "both" sets of disciplines actually do and claim. Newimpartial (talk) 11:54, 5 November 2023 (UTC)

@Newimpartial Thanks for your comment and for engaging in a rigorous way. My position is not that one discipline generates facts and others generate opinions in the black-and-white, binary way you describe. That would be a caricature of the disciplines, if it were my position. Of course, I readily concede that opinions (in the form of untested hypotheses or informed speculation about experimental results) exist in scientific journals, and any humanities journal surely abounds with factual statements.

What I am saying is that expressing personal opinions in many non-scientific contexts, often in declarative form, is seen as an important and legitimate way to write academic works. To put it another way: it's not acceptable to include your political opinions in a physics journal. It's totally cool - encouraged, even - for scholars to weave their personal political opinions into a narrative in a humanities journal.

I'm a bit surprised by the objections I'm reading here, so I'm not sure what evidence this audience would find compelling. I could give you a list of polisci/humanities academic literature that expresses a range of social values that are personal to the author, and not derived from any formal method. I'll stick to one example for brevity's sake.

One Example

How to Abolish the Hong Kong Police. This work of academic literature advocates for a certain worldview, a certain set of outcomes in an ongoing political dispute, and presupposes a variety of social values. One presupposition is that the Hong Kong police should be "abolished". In the abstract, it states: "In grappling with this question, we encourage Hongkongers to engage with diverse decolonial and abolitionist struggles across the globe, along with existing and ongoing proto-abolitionist practices at home."

I'm not knocking it, this is academic literature. Within the field, it's considered a perfectly legitimate way to write. But it is not using anything approximating the scientific method, and it presupposes and explicitly advocates for specific political positions, like "engaging in...decolonial and abolitionist struggles". This type of explicit advocacy and exegesis of a values-driven point of view on world affairs is nowhere to be found in the physical sciences. You will not find a declarative sentence in, say, an astrophysics journal, that expresses the political opinions of the authors as empirical fact.

Therefore, from a Wikipedia writing point of view, a declarative statement of "fact" from this sort of humanities-based academic work should not be apprehended in the same way as a statement of "fact" from a research article in "The Astrophysical Journal". This presents a problem, because currently, all declarative sentences from any academic source are treated as "academic facts", with no distinction between the disciplines.

Would, perhaps, several more examples of this type suffice to demonstrate my point? I could give you as many as you want. I already linked to several sources above that attempt to probe the differences between scientific and humanistic literature. Honestly, in over 10 years in academia, I have never encountered the opinion that the physical sciences and the humanities are epistemologically or methodologically the same until I met @Horse Eye's Back, so I guess I'm a bit taken aback and confused as to how to address that novel theory.

I would love to have a deeper discussion about the nuanced differences between the disciplines here, or what theoretical changes to RS policy might constructively reflect those differences, but unfortunately we're still stuck on "astrophysics and art history are both sciences in the same way, utilizing the same epistemology and methodology". Once we can dispel that, I'd love to engage with you more on the details. Philomathes2357 (talk) 19:24, 5 November 2023 (UTC)

Briefly: you haven't shown that the article presents as fact that the Hong Kong police should be abolished, nor have you shown the more relevant thing - that any wikipedia article would present such a normative/proscriptive statement "as fact" even if the source had actually done so. You seem instead to be making a kind of slippery slope argument, that the mere presence of normative or prescriptive statements in a source casts doubt on factual statements presented in that same source. If so, you need to offer a basis for treating sources that way, since it seems like an attempt to override WP:V as a core policy simply to serve your own preferences (or, dare I say, aesthetics).

Also, I am under the impression that in invoking a proposal to abolish the Hong Kong Police as an example, you have violated your topic ban (against discussing politics in general) both re: the letter of the ban and concerning its intended purpose. My advice, then, would be for you to strike the example and to withdraw from the discussion. Newimpartial (talk) 20:03, 5 November 2023 (UTC)

Good points. I agree, the mere presence of normative statements does not cast doubt on factual statements presented in the same source. My contention is more subtle than that. Let me take a couple of steps back. I appreciate you engaging.

The root of my objection goes back to WP:NPOV. NPOV clearly states that we must avoid stating facts as opinions, and avoid stating opinions as facts. However, there's no formal process editors can go through to ascertain whether or not a given passage from an RS is "fact" or "opinion". That would be fine, if editors had the freedom to apply a little common sense to parse fact from opinion. But I've also observed a widespread lack of ability, and an even more widespread lack of interest, in determining on a case by case basis whether a given statement is a "fact" or an "opinion". Instead, the un-written policy seems to be that if an RS writes a declarative sentence, and the RS does not explicitly label that declarative sentence an "opinion", it should be treated as a fact. This un-written rule of thumb has been crystallized in a popular slogan: "verifiability is truth". That slogan has been repeated to me several times, and I've seen it on several userpages.

The problem is, opinions exist outside of "opinion" pieces. One will occasionally find an opinion in a work of scientific literature. One will frequently find opinions in humanistic literature. And one will constantly find opinions in modern political news. Often, these opinions take the form of declarative sentences, and even if it's obvious that it's an opinion, editors will say something like "there's no indication that this was intended as an opinion piece, so we should treat it as fact".

An example from the Hong Kong piece: it says in the piece that a US news outlet, The Grayzone, is "known for misleading reporting in service of authoritarian regimes". No source is cited, no examples are given, and obviously, "known for misleading reporting" is an opinion. Also, note that Grayzone has been sharply critical of the Hong Kong protests. Given that fact, is it any surprise that a text entitled "How to Abolish the Hong Kong Police", written by and for dedicated members of the Hong Kong resistance, would find a harsh critic of the movement to be "misleading"?

However, that excerpt is currently in the Grayzone article, word-for-word, in Wikivoice - even though it's clearly an opinion. When I pointed out that this is problematic, I was told that it is a "high quality academic source" and that academic sources are not opinion pieces. So, even though this is an opinion, and one that was not derived from anything approximating a scientific approach, it is currently presented in mainspace as a fact in the same way current rigorously-derived scientific consensus on the Big Bang might be presented on its article. Does that example clarify what I'm talking about?

The fact that I'm attempting to highlight is that, while scientific sources and non-scientific sources both contain facts and opinions, you are much more likely to encounter proscriptive statements, opinions, and value judgements within non-scientific, humanistic academic literature than in scientific literature. The prevalence of opinions and value judgements is so much more prevalent in the humanities (and, to a lesser degree, the social sciences) that we can not simply assume that declarative sentences in the humanities should be understood as expressions of fact.

I don't know what the solution is - maybe it would entail changes to RS policy, maybe not, I'd like to hear others' thoughts on that in this discussion or a future one - but just saying "welp, it's an academic source, and they didn't say upfront that it was an opinion, so I guess it must be a fact!" is untenable.

Looking forward to your thoughts whenever you have the time to respond. Philomathes2357 (talk) 02:20, 6 November 2023 (UTC)

Philomathes2357, you write: "However, there's no formal process editors can go through to ascertain whether or not a given passage from an RS is "fact" or "opinion"." Unless I'm missing something (entirely possible), that's not true. There are these people called "fact-checkers" and many ways to determine whether a statement is true or false.

Around here, we often talk in shorthand, assuming we have the same knowledge base when that is often not the case. What is an obvious, and therefore assumed, truth/fact to one group of editors (usually "mainstream") who specialize in a topic and know all the RS and what fact-checkers have ruled on the topic, is not at all obvious or even understood by another group of editors (usually "fringe") who get their info from extreme, usually far-right, sources of disinformation and have likely been shielded from the facts. They often lack knowledge of what the rest of the population considers common knowledge. Fox News does not report such things to them, and since they refuse to read or watch what Trump has told them is "fake news" (literally all national and international mainstream sources), they are clueless. They only get what's shared in their bubble, and fact-checking is not allowed. Within the AP2 topic area, this is one of the most frequent causes of friction and wasted time.

So, why do you say that "there's no formal process editors can go through to ascertain whether or not a given passage from an RS is "fact" or "opinion"."? We work with RS which tell us what are facts and opinions all the time. We have know there are fact-checkers who parse this stuff for us. This should not be difficult. -- Valjean (talk) (PING me) 19:07, 6 November 2023 (UTC)

We – Wikipedia editors – have fact checkers? Where's the noticeboard for that? WhatamIdoing (talk) 21:55, 6 November 2023 (UTC)

I assume he means reliable sources such as PolitiFact which we trust for fact-checking other publications. Andre🚐 23:00, 6 November 2023 (UTC)

Correct (now tweaked). We know there are fact-checkers available to us to check, and we use them. Fringe MAGA people refuse to use them and consider them fake news. Philomathes2357, please respond. -- Valjean (talk) (PING me) 01:59, 7 November 2023 (UTC)

Certainly, I agree that fact-checking websites are useful. I look at them all the time. But my concern is not necessarily with false statements, or even with statements that could be within the purview of a fact-checker.

What I have in mind instead are statements where an author's analysis, value judgements, and opinions are issued in a news article, but issued in the form of a declarative sentence.

My current understanding is that, if a piece is not labeled "opinion", and the author does not specifically say something like "in my opinion...", but simply presents their analysis in declarative form, the un-written rule is that we should treat it is a statement of fact. That seems to be the practical implementation of the slogan "verifiability is truth". It's a shortcut, so that editors don't have to discuss whether a given passage is of a factual or opinionated nature.

I've also heard the suggestion that we must treat such statements as statements of fact - to judge for oneself whether a statement is a "fact" or an "opinion" is original research. I don't think that's formal policy, but it does seem to be a common view.

Following from that logic, it's also been implied to me that the deciding factor on whether a given text is "fact" or "opinion" is the intent of the author (i.e. - "this article does not appear to have been intended as an opinion piece, so we should treat declarative sentences therein as statements of fact")

I think there are big problems with that model. Please correct me to the degree you think I've misunderstood. Thanks. Philomathes2357 (talk) 02:23, 7 November 2023 (UTC)

The question is where the statement was made and by whom. If the statement was a normative claim, like "Pie is great," or, "Roosevelt is the best course forward for America," then yeah, I agree, those are just subjective statements, not expert opinions. An expert opinion is different. An expert opinion would be something like, "anthropogenic climate change is likely to lead to sea level rise and an increase in terrestrial and ocean temperatures." That is a fact. Andre🚐 02:26, 7 November 2023 (UTC)

OTOH, maybe we are "talking past each other". I'm talking about fact vs non-facts (easy to distinguish) and you're talking about facts sources vs opinion pieces, and how some have difficulty distinguishing between them. I'm not sure if I'm close to the issue. -- Valjean (talk) (PING me) 19:16, 6 November 2023 (UTC)

It would appear that you're arguing that the differences between the "hard sciences" and the "soft sciences" is real and not just sexism. The research suggests otherwise.[2] Also note that you're making one of the least nuanced arguments here, you actually appear opposed to treating this with nuance and are advocating got broad strokes policy changes. Horse Eye's Back (talk) 16:35, 5 November 2023 (UTC)

There may be cultural biases at play here. In the US, science is natural sciences, and if you do something remarkably similar to natural sciences, only about individual humans (e.g., experimental psychology) or human societies (e.g., behavioral economics), then we will stretch a point and let you claim to be a social science, though we'll yank that status-conferring label right back away from you if you step away from the scientific methods (e.g., Psychoanalysis).

In other parts of the world, notably Germany, science is a very broad term involving any and all sorts of organized knowledge, including, say, Theology and Fine Arts. A retired German professor once told tell me that he had undertaken a "strictly scientific" reading of an ancient document. He firmly believed (and told me) that textual criticism of ancient documents was exactly as scientific as anything involving test tubes in a chemistry lab. I don't agree. That's not what we mean, in English, when we use the word science. But apparently German uses the same word for scholarly and scientific. If you want to talk about "real" science in German, you have to specify knowledge-making-of-nature or the equivalent. That language/culture elevates all scholarly fields equally, whereas in English we value (natural/formal/social/applied) science over humanities. WhatamIdoing (talk) 22:17, 6 November 2023 (UTC)

That's why I say it's semantic. But the point is that social sciences do in fact make hypotheses and rigorous arguments. And many hard science papers do not. The question is about peer review, the process of applying a literature review, and scientific rigor. I agree that there's a substantive difference between reading technical instrument data that means something about bacteria, or whatnot, and reading papers and coming up with interpretation and analyses. But fields like linguistics, for example, make hypotheses, and try to prove them experimentally, all the time. Andre🚐 22:59, 6 November 2023 (UTC)

Linguistics is called scientific because it's "comprehensive, systematic, objective, and precise", and you could be "comprehensive, systematic, objective, and precise" about evaluating a ballet performance, too. Some sub-fields of linguistics make depend significantly on hypotheses, but others don't. Some sub-fields try to prove them experimentally, and others don't.

("Rigorous arguments" aren't actually part of the scientific method.) WhatamIdoing (talk) 04:34, 7 November 2023 (UTC)

In such fields as computer science, decision science and formal logic, which are part of pure and applied mathematics (though, many universities have given them their own buildings... not that it really matters, of course), there are often different amounts of rigor, and the rigor of a proof or an argument is a bit more important than whether your lab book observations were detailed enough. Physics is more rigorous than psychology usually, because it involves models and equations and math more than surveying and statistical methods with human subjects. Still, experimental psychology is absolutely a rigorous social science that follows the scientific method and has very structured experiments. And there are many papers published in computer science, which is definitely a STEM field, that never make a hypothesis or use any element in the grade-school version of the scientific method. Computational linguistics is part of the social sciences but there are definitely tons of papers in that field with very rigorous scientific experiments that check every box in your 8th grade science text's description of what an experiment is. In the real world, it's true that "the pursuit of structured knowledge" can also be considered scientific. It's all immaterial, though, to NPOV and RS policy, where social science papers absolutely are valuable and scientific and no less valuable to writing articles on such topics. Pure math rarely has a real-world experiment in a paper, just a bunch of equations and contentions about abstract quantities. Still, we use the same RS and NPOV approach. If a Fields medalist is writing some assertions about the field of math, and we also have someone else who is just a pop science writer, the former will probably get their assertion written in Wikivoice, and we won't bother to include the latter except as attributed where relevant. Andre🚐 04:44, 7 November 2023 (UTC)

Whether logic (and philosophy and pure math) is properly science or "only" a necessary precondition that makes science possible is apparently up for debate, and I have certainly heard scientists dismiss engineering as "not science".

"The pursuit of structured knowledge" may be (IMO is) a highly desirable thing, but that doesn't mean that the average English speaker/reader will consider art history to be a science, and if we tell them that they can read an old document "scientifically", they'll be confused. WhatamIdoing (talk) 04:54, 7 November 2023 (UTC)

The general rule of thumb, by the way, is that if you have to name your field "___ science", then it's not science. WhatamIdoing (talk) 04:55, 7 November 2023 (UTC)

Tell that to computer scientists. --Masem (t) 05:08, 7 November 2023 (UTC)

I would, but they told me in the first place. [3] [4] :-) WhatamIdoing (talk) 20:18, 13 November 2023 (UTC)

Amusing, but an obvious contradiction if you are arguing we should use the most common dictionary definitions of words. Andre🚐 05:46, 7 November 2023 (UTC)

Yeah, I mean, cool, if you think it's up for debate, that's your prerogative, but the "T" in "STEM field" is technology and the E is there for engineering (and the M is of course math), and I assure you that chemical engineers, biomedical engineers, electrical engineers, and mechanical engineers are "doing science" all day long even though they never record observations from an experiment, since most of those fields are physics and biology and (bio)chemistry. Andre🚐 05:19, 7 November 2023 (UTC)

STEM = Science, Technology, Engineering, and Math, right?

Or, by its own admission, science, not-science, not-science, and not-science, because if they were all obviously science, then "STEM" would just be "S". WhatamIdoing (talk) 22:41, 10 November 2023 (UTC)

I knew you would say that! But again, the real world doesn't come in neat little boxes. If computer science is part of engineering or math, it's not science you say a misnomer, but what about computer engineering? The reality is that Hard and soft science is a line that changes over time -- see the demarcation problem or the exact sciences. These are problems in philosophy and existentialism and epistemology and whatnot. Not on Wikipedia though. For an encyclopedia, it's obvious that the academic fields of math, engineering, etc, follow the exact same rules as physics or astronomy. I am a software engineer and a member of both the ACM and the IEEE (though, I don't get much value out of my membership). Their proceedings are no different from the ACS or any other academic society. The point is that social sciences are also for our purposes, high-quality academic sources whose scientificness, at least as pertains to wiki, shouldn't really be besmirched or questions, certainly not along the lines of allowing political fringe (or, excluding all political material because it's all "subjective") Andre🚐 22:47, 10 November 2023 (UTC)

I'm delighted to have lived up to your expectations. :-)

I agree that the real world is messy. Medicine, for example, is traditionally described as "the art and science of healing" – a phrase old enough that its original meaning would be more aptly rendered in English as "the practice and knowledge of healing". And yet med school students, even if they've never formulated a hypothesis outside of a school assignment, routinely think of themselves as scientists merely because they use the results of scientific research. They have to be reminded that almost none of what they do is, narrowly speaking, doing science. They, like engineers and technicians, are instead using knowledge produced by science. (Well, about half the time. The other half is mostly based on things like "when my Attending was in training, his Attending told him..." or "Nurse said the hospital procedure is...") WhatamIdoing (talk) 23:06, 10 November 2023 (UTC)

Hmm, but, unless you're a medical research doctor. I agree that your family doctor or the ear, nose, and throat guy you see for a sinus checkup isn't a scientist in any meaningful sense of the word, nor really an engineer or a technician in my view, they're just a doctor. There are many doctors in academic fields like epidemiology who are both scientists and doctors. For example Harry Ostrer, who was the topic of controversy in a recent article that I was editing. Thankfully, there's no current dispute. But at one point, there was a fairly active dispute as to whether Dr. Ostrer, who by all account is an academic and teaching physician and is not the kind of guy you can go see unless your disorder is really interesting, was a WP:MEDRS-related topic or source (in a part of an article mentioning Tay-Sachs disease) since he's written popular books on history and genetics. Or a more relevant and timely example is Anthony Fauci, arguably a doctor, scientist, public health expert, and a politician. Andre🚐 23:20, 10 November 2023 (UTC)

Their proceedings are no different from the ACS or any other academic society. – probably better than most. How many academic societies would argue that their science isn't science enough?

https://cacm.acm.org/magazines/2012/10/155530-where-is-the-science-in-computer-science/fulltext --PaulT2022 (talk) 23:33, 10 November 2023 (UTC)

I think it would be more accurate to say that some English-speaking speech communities value (natural/formal/social/applied) science over humanities. And I do not believe that the (mainstream, consensus) perspective of editors on enwiki is one of those speech communities. Certainly WP:V, a core policy, does not uphold the preference WhatamIdoing has presented as what is done in English. There could be more than one speech community operating "in English", I might venture. Newimpartial (talk) 00:04, 7 November 2023 (UTC)

@WhatamIdoing Great comment, great point. The German word for science/scholarship, Wissenschaft, literally translates to "knowledge-ship" or "knowledge-craft". A much broader notion than the English word "science".

@Newimpartial I wouldn't say "some English-speaking speech communities value...science over humanities". I would say that standard American English distinguishes between science and the humanities in a way that say, German, does not.

Put it this way: current RS policy seems to implicitly adopt the German notion of Wissenschaft that WhatamIdoing described above, which treats anything published by an academic publisher as an "academic source of fact". My basic premise is that our RS policy should reflect the common English-language distinctions that WhatamIdoing also described. Philomathes2357 (talk) 03:23, 7 November 2023 (UTC)

Wikipedia isn't for American English. Wikipedia is for all English-language and in fact all world languages. There might be some forms of systemic bias that editors or articles or projects have since most editors either hail from the UK or America. However, I know we have quite a few enwiki editors and readers from Australia, Sweden, Malaysia, Singapore, or, yeah, just about anywhere. And of course you're all complaining about a simple connotation, not a definition, since one of the definitions of science is: the study of some aspect of human behavior, for example, sociology or anthropology. (Collins) a system for organizing the knowledge about a particular subject, especially one that deals with aspects of human behaviour or society (OED) see: Christian Science, creation science. Andre🚐 03:32, 7 November 2023 (UTC)

Actually, I shouldn't have said "American". I'm not aware of any form of English that uses the term "science" like German uses the term Wissenschaft. All forms of English make much finer distinctions, not just American English. Philomathes2357 (talk) 03:36, 7 November 2023 (UTC)

The Danish word for "science" is cool. It's "videnskab". Viden=knowledge and Skab=create, so to a Dane, science is the creation of knowledge. We do not really "know" something if it hasn't been proven to exist. All else is speculation. Karl Popper's concept of falsifiability comes to mind. If it's falsifiable, it can be the subject of scientific inquiry and examination. All else is religion, metaphysics, and general mumbo jumbo. At Wikipedia, where we are supposed to document the "sum of all human knowledge", we define knowledge as that which is described in RS. That includes facts, opinions, and other things mentioned in RS, such as the awareness of the existence of lots of nonsense like pseudoscience, alternative medicine, conspiracy theories, lies, etc. -- Valjean (talk) (PING me) 03:39, 7 November 2023 (UTC)

Andrevan, I notice that you quote the least common/lowest down definitions in your comment. That is not going to give a fair impression of the most common English meanings for the word.

Valjean, I think there are several qualities that are commonly used to identify science. One is hypotheses that can be falsified (e.g., whether species can live without oxygen, yes; name of species, no). Another is progressivism, meaning that each generation knows more about the world than the previous one (e.g., physics, yes; history, no). AIUI all the natural sciences have all of the main qualities, but the broader definition seems to want only most of the main qualities, most of the time. WhatamIdoing (talk) 04:49, 7 November 2023 (UTC)

"Lower down" definitions aren't somehow invalid. And the point is it's arguing semantics which is a fool's errand anyway. Every word has many definitions, context, and subtext. The bounds of knowledge aren't determined by common usage. Nor are Wiki policies and guidelines (except for what the article text should say a la COMMONNAME etc). The question of "are things that aren't using the scientific method still scientific" is what we're discussing. Yes, they are, in American and in UK English. Andre🚐 05:25, 7 November 2023 (UTC)

To me, the key point that has not been as prominent in this discussion as it is in WP policy, is that WP:V does not place "science" above other methods of systematic study. On enwiki, "science" is not a privileged source of facts or of findings, conpared to say historical or other qualitative research.

Every once in a while, I see an editor (like Philomathes) try to map the fact/opinion distinction that WP employs onto a science/humanities distinction, which WP does not. These attempts always strike me as falling in the broad category of "misunderstsndings of WP policy". Newimpartial (talk) 11:36, 7 November 2023 (UTC)

I would tend to disagree with that, given that we have policies MEDRS and the slightly weaker SCIRS. When discussing anything in the scientific or medical field, even though these topics may be covered by traditional high-quality media to a degree appropriate for WP, we nearly always defer to what the peer-reviewed publications say when we go into the actual research for the topic. Masem (t) 13:08, 7 November 2023 (UTC)

I should perhaps have been more precise in distinguishing between WP:V and enwiki as a whole - but, as I said, it is WP:V that does not place science above other methods of systematic study. To me, the rules set out that distinguish science topics (and, to a lesser extent, natural science topics) from other topics are about those domains; they establish sourcing requirements specific to those domains, but they do not relate to the distinction between "fact" and "opinion" under discussuon here. Facts are still facts in humanities topics, and opinions are still opinions within science and medicine.

If we had (as perhaps we should) a set of sourcing requirements specific to politics topics on enwiki, or to religious topics, those requirements would presumably aim to establish a high standard of sourcing relevant to those domains; I can't imagine the enwiki community imposing experimental methods on those subjects or saying, in essence, "this isn't science so it it is all just a matter of opinion" - those being the options that might flow from some comments made elsewhere in this discussuon.

On the other hand, privileging review articles and books over ideographic research findings might go some way to improve the treatment of politics and religion on enwiki, but this would be pretty much the direct opposite of the way some editors have wanted to treat hard sciences and humanities as opposing methodologies producing different kinds of findings. Newimpartial (talk) 03:25, 8 November 2023 (UTC)

I fully agree with @Newimpartial: WP:V does not place "science" above other methods of systematic study. None of our polices or guidelines do – not even MEDRS. The WP:PROPOSAL to privilege the Wikipedia:Scientific point of view was rejected eons ago.

It is, however, true that some individual editors personally privilege "science" over other subjects. WhatamIdoing (talk) 22:49, 10 November 2023 (UTC)

Side point, I think one of the issues at play (among many) is the idea that we're meant to document the sum of human knowledge, when in fact the mission is to summarize it, which is significantly different. That's why UNDUE is important because it is meant for us to determine and write as a summary of what are the most oft-repeated opinions or views on a topic, and downplay those that come from a few or even one source. But we've (as a whole) gotten so used to documenting every detail as it gets reported in an RS and lose sight of using our position as editors to make discretionary choices as to what to include. This leads to, among other problems, making articles on controversial topics read like scarlet letters or laundry lists of every negative that an RS has said (leading to excessive cherry picking of opinions), rather than trying to write, as a summary, why the topic is controversial and hitting the points that fall out of an RS source review. There's more than just that that leads to larger NPOV problem, but that's stuff I've already mentioned or beyond the scope here. Short answer: summarizing sources takes more work that rote documentation of sources, but leads to better quality articles in all content areas, including neutrality. Masem (t) 05:03, 7 November 2023 (UTC)

breaking

I think you misunderstood me, Philomathes. A convenient fiction is like money or a corporation or philosophy. It's an abstraction. It's not literally real like an object. It's a mental model. We suspend the fact that yeah, epistemologically, some scientific results especially in social sciences, aren't 100% or even 99%. Maybe they're 80%. But that is good enough to write them as 100% in Wiki text. Andre🚐 19:14, 22 October 2023 (UTC)

Ah, thank you for clarifying. I did slightly misunderstand, but I think I get what you are saying now.

We do largely agree. I too, am comfortable with the current policy of making Wikivoice statements of "fact" to summarize academic consensus "opinions". But - only if those opinions/conclusions were derived through the scientific method.

My objection is that quite a few academic sources in the social sciences/humanities do not use a scientific methodology, and so their conclusions, in many but not all cases, should be best understood as notable opinions from high quality sources, rather than facts that can be written in encyclopedic voice. I think this abstract that I linked above does a decent job of laying out some of the differences between scientific and humanistic approaches to academic work. Philomathes2357 (talk) 19:36, 22 October 2023 (UTC)

I would suggest that if your issue is with the nature of social sciences and humanities then your issue is outside of Wikipedia. It's not up to editors of Wikipedia to determine if the basis of these areas are valid or not. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:52, 22 October 2023 (UTC)

Yeah, unfortunately, we don't have the leverage or leeway that you'd like Philomathes. We need to trust the sources based on other sources. All that policy calls for is reputation based analysis. If a source has a reputation for fact-checking and accuracy, and if an academic source is widely cited absent a deep controversy or a lot of people reviewing it negatively, that's a good sign it should be considered authoritative. Not limited to, but that's to give you an idea. Andre🚐 19:55, 22 October 2023 (UTC)

Where do theories fall on you facts vs opinions spread? Most of the natural sciences stuff you're describing as facts would be more accurately described as theories. Horse Eye's Back (talk) 19:19, 22 October 2023 (UTC)

Well, there are at least two definitions of theory that I'm aware of. The technical, scientific sense, and the colloquial sense. People equivocate between the two all the time. ("my theory is that God created the world 6,000 years ago, and evolution's just a theory too, bro").

The two are similar: both colloquial theories and scientific theories are opinions, not facts in and of themselves. Theories, at least good ones, contain facts, and they are narrative syntheses of a bunch of independent facts and observations. I think the important distinction is that scientific theories are the product of the scientific method, whereas colloquial theories are not.

My contention is that many (but not all) of the declarative statements made in the social sciences and humanities literature fall more into the category of colloquial theories than scientific theories. Philomathes2357 (talk) 19:50, 22 October 2023 (UTC)

This is too philosophical of a discussion. Wikipedia has its own jargon that it uses. So it doesn't really matter that some "theories" like gravity or evolution consist of best case scenarios and descriptions that are more like ideas than facts. They are presented the way that leading scientific authorities believe them. They will change with the times but will lag the cutting edge. Andre🚐 19:56, 22 October 2023 (UTC)

Philomathes: what matters is not what kind of statements the relevant academic sources include, but rather which statements are cited by Wikipedia, and how. The idea that any Wikipedia editor believes that all statements made in academic sources in the humanities are facts and should be cited as such reads like a straw goat argument, and is at the very least unproven. But the idea you seem to be carrying, that statements embodying human knowledge can be grouped into "facts", "scientific theories" and "colloquial theories/opinions", doesn't reflect any well-established contemporary epistemology and certainly lacks the widespread consensus that would allow us to determine Wikipedia content on its basis. This typology seems to be essentially WP:OR/"opinion" on your part.

And by the way, neither of the sources you cited supports the epistemologiclal claims you have been making here; as a survey of undergraduates, the second source may illustrate the concept of "folkloric epistemology", which I introduced above, rather than the characterizations you intended to support. Newimpartial (talk) 11:32, 23 October 2023 (UTC)

break

@PaulT2022, I have a lot of sympathy for this, but that quotation I posted above about hostages ("In the end, the complexity of the arrangement almost killed the deal to free the hostages. Officials called it the largest private financial transfer in history. It was surely the most intricate. But for technical details, the hostages could have been released sooner.") It's almost word for word out of the New York Times in January of 1981 (42 years ago). It's about the Iran hostage crisis. Where I wrote "Officials", the original says "Officials of the former Carter Administration".

While I'm not really surprised to hear that newspapers carry more magazine-style content than they used to, there is no past in which newspapers never used "emotive" language. WhatamIdoing (talk) 21:11, 21 October 2023 (UTC)

@WhatamIdoing, I'm sorry, I didn't realise you had an example involving "officials". I should've been clearer I wasn't commenting on it.

I was referring to another example discussed in an editorial in The Times on 18 October this year. Multiple unquestionably reliable news sources, such as CNN, BBC and WSJ, referred to a Hamas spokesperson (or an official controlled by Hamas) as a "Palestinian official". This phrasing is highly ambiguous in the best case scenario as it may be interpreted to refer to the broadly internationally recognised PLO, which is/was effectively at war with Hamas. The author of the editorial argues that it is a conscious stylistic choice to underplay the fact that Hamas was the sole source for the news.

Currently, by interpreting NPOV/WP:WEIGHT formally, this can be used to argue (wikilawyer) that the prevailing POV is that Hamas should be described as "Palestinian officials". In this case, Wikipedia editors didn't follow the slippery slope of the news sources and al-Ahli Arab Hospital explosion attributes the figures to the "Hamas-run Gaza Health Ministry", which is IMO a more neutral way to describe the source.

These kinds of language choices and the extensive use of attributed quotes are what I referred to as the "emotive" language of the news. I hope this makes sense, and once again, I apologise for any confusion. PaulT2022 (talk) 22:15, 21 October 2023 (UTC)

The Hamas/Palestinian official sounds like a fine distinction that could make for an emotive opinion piece, but that doesn't have much basis in fact. You can be an official in practice without your office, or your right to hold it, being recognized by other governments. The head of a regional health agency would be "an official" regardless of whether you're talking about the Gaza Health Ministry in the Gaza Strip or the (non-governmental) Kaiser Permanente health system in California. I'm glad that we've been more precise, but I don't think that it's "emotive" for a newspaper to recognize that a person is Palestinian and holds an office. WhatamIdoing (talk) 04:16, 22 October 2023 (UTC)

It might be in the context – it's not that either is wrong; it's whether it's neutral or has connotations that hint at which side is good and which is bad. I agree that this example is very mild, but I've seen more egregious ones, such as the BBC's Putinesque delight: 'Hamas staged a lightning operation'.

To stay on the subject, is there a policy that allows editors to write 'Hamas-run Gaza Health Ministry' in place of 'Palestinian officials', when the latter is used by the plurality of sources about this strike? If not, what's the benefit in keeping the letter of WP:WEIGHT as mechanistic as it is and not reflecting best editing practices? PaulT2022 (talk) 22:03, 23 October 2023 (UTC)

I've noticed, especially with newer editors, that there's a desire for a policy that says "Write an encyclopedia article". Editors are expected to Wikipedia:Use our own words. Editors are expected to use Wikipedia:Editorial discretion. Editors are required to be "carefully and critically analyzing a variety of reliable sources and then attempting to convey to the reader the information contained in them fairly, proportionately, and as far as possible without editorial bias". Editors are reminded that If a rule prevents you from improving or maintaining Wikipedia, ignore it. But if we really need a page, with an official policy banner at the top, that actually says "You are officially permitted to choose a different word if the one that is most commonly used in the sources doesn't seem to 'fairly, proportionately, and without editorial bias' communicate the information that is in the sources", then maybe we should give up in despair and close the project. WhatamIdoing (talk) 01:24, 24 October 2023 (UTC)

The problem with being a newer editor is that you may encounter an older editor who insists that the exact same word most newspapers use must be used because of WP:WEIGHT. That happened to me, and I didn't know what you just said. I also didn't know that you could go to this talk page to ask and receive brilliant advice.

Before someone assumes I was trying to add something fringe, it's worth noting that, at least in my experience, the only two editors, who I remember to misuse WP:WEIGHT to wikilawyer their preferred wording in the discussions I participated in, were eventually sanctioned by ArbCom for multiple kinds of disruptive behavior, albeit in a different topic area than I was editing. As a result, I associate the misuse of WEIGHT with disruptive and toxic editing, rather than with newer editors who don't understand its intent. PaulT2022 (talk) 03:02, 24 October 2023 (UTC)

I usually summarize this belief as "If it's not a copyvio, then it's original research". It's wrong, but since our main method of teaching people how to edit is a telephone game in which the rules get twisted slightly with each repetition, then it's not surprising that it happens. Variations on this include insisting that the word used must be the word in the source cited at that end of the sentence ("the source says 'lorry' so you can't say 'truck' in this sentence!") or that even the punctuation must be retained ("the source says 'Alzheimers disease', so you can't say 'Alzheimer's disease' in this sentence!").

@PaulT2022, it's possible that you would find the sentiment in Wikipedia:Bring me a rock to be familiar. WhatamIdoing (talk) 00:22, 27 October 2023 (UTC)

Around 2010 with the decline or change to accountability journalism, with the traditional media having to change tactics to economically survive and compete against Fox News and online independent journalism.[5] Masem (t) 15:15, 18 October 2023 (UTC)

That source linked does not appear to say that. Also note that its accountability journalism which is being decried by the Cronkite worshipping crowd. Horse Eye's Back (talk) 15:22, 18 October 2023 (UTC)

The key there is the economic drivers that traditional media has had to adapt to stay afloat. Its become targeting their coverage to the audience (eg what Fox News has always done) with more aggressive accountability to keep in line with what their audience whats, and that immediately shifts these papers away from objectivity. Its the shift away from investigative reporting (which is more costly but also more objective) to accountability that creates the problem. That's in that article as well as others [6] [7] Masem (t) 16:00, 18 October 2023 (UTC)

You've misunderstood the sources. Investigative journalism and accountability journalism are the same thing. "The declining supply of high-quality accountability journalism, also referred to as investigative or watchdog journalism, can be viewed from an economic perspective as a pricing problem." Horse Eye's Back (talk) 17:31, 18 October 2023 (UTC)

I probably got it wrong yes, but my point is that these articles all point to how, the AP and other sources have engaged in this new form of accountability journalism that leaves objectivity at the door as to be able to retain and grow their pair reader base. It doesn't mean they have gone so biased as to be unusable, but we absolutely should be aware that the media is continuing to fight for themselves to stay relevant in today's online world, and the old adage of being objective reporting of the news is long gone.

I know what I used to read 20+ years ago and what I read now, and it is a completely different world in terms of the type of language and approach to describing otherwise similar events or news stories. Masem (t) 23:44, 18 October 2023 (UTC)

Are you confusing objectivity and viewpoint dominance? The news was never reported objectively, it was at one point all reported from the same POV though. Horse Eye's Back (talk) 15:20, 19 October 2023 (UTC)

I think the question is: Does he still read an actual newspaper?

What I read now is a completely different world, but what I read now is mostly online, from a mixed variety of sources, with magazines and newspapers jumbled together. What I read then was almost exclusively a daily newspaper, printed on paper, delivered to my door each morning. They could (and seem to) write exactly the same thing, and yet my own news world would still be very different. WhatamIdoing (talk) 21:14, 21 October 2023 (UTC)

I do wonder what you read Masem, but the good news is that you can easily read NYT articles or WSJ articles from 1972 or 1992. They do change a little bit, but not that much. Andre🚐 03:32, 21 October 2023 (UTC)

And what do you say to all of the experts who say that far from in decline we're currently living in a golden age of news? Horse Eye's Back (talk) 14:07, 18 October 2023 (UTC)

This is an important point. I can’t think of a good remedy, but imo it’s an inevitable potential consequence of the deterrence to journalistic sources. But there is no easy fix to that. Yr Enw (talk) 07:07, 23 October 2023 (UTC)

That comment adding the link is over 3 years old as a note, it isn't from today. Cursed Peace (talk) 19:46, 17 October 2023 (UTC)

I don't really think the Sanger piece from 2020 contains anything that might be termed "actionable intelligence". The position it elaborates is essentially a form of Nietzechean relativism, in which the observable fact that people on different sides of a controversy believe their views to be grounded in reality, with comparable intensity, is used as a basis to present "both sides" of the controversy as though both sets of views in fact have equal epistolological standing. In observable social reality, some views accompany factually-grounded observations while others are on weaker evidential ground, and one of the inconvenient responsibilities of an encyclopaedia is to sort through the sources to distinguish the scenario where perspectives havw equal plausibility, according to the best avaibale sources, from the other scenario. The current NPOV policy meets this responsibility with a good deal more maturity and acument than the "false balance" that Sanger would have us employ, IMO. Newimpartial (talk) 20:10, 17 October 2023 (UTC)

The good news is that it is fixable. North8000 (talk) 19:48, 17 October 2023 (UTC)

There's definitely a huge problem in the way in which bias in English editorial language spills over into Wikipedia language usage and subject naming discussions. The Israeli-Palestinian conflict is pretty prime fodder for this - for example in how words like "massacre" get handed out extremely partially to certain types of event, but not others. Take this list for example, which in this page diff has been expunged of any mass killing events in the recent conflict on the part of the Israelis in favour of only mass killing events on the part of Palestinian militants - an imbalance that has become more ironic as the conflict has gone on, but again, is a function of imbalanced editorial language and probably more than just a little bit of systemic bias. So NPOV here is probably not actually helping us to achieve anything approximating neutrality; on the contrary, it's helping enforce an artificial divide by means of certain linguistic choices, i.e. that of "massacre", which is essentially just one word for a mass killing, but one not used routinely for all mass killings (and only very rarely for airstrikes, for example, as a pretty prime example in the context right here, even if events might contain comparable degrees of mass killing). This might be the sort of thing that is actually better controlled by a strict editorial guideline, much as the MOS:TERRORIST guideline is something that is very effective at stamping out needless debates over biased language in its own little area through its appeal to the fact that one source's terrorist is another source's freedom fighter. There's not necessarily a direct parallel there with what's going on with the word massacre, but some sort of style guide might likewise be a solution here and elsewhere. Iskandar323 (talk) 02:25, 18 October 2023 (UTC)

That's a more specialized area but illustrates that wp:NPOV as written often creates POV tilts rather than preventing them. North8000 (talk) 12:35, 18 October 2023 (UTC)

@Thinker78: Is this the comment you're referencing? "Yeah well they are asking for money now…but why? Why would I give to any media type organization who is so blatantly a left wing communist organization who hides behind the world “neutral”? What a" [8] thats the only addition to that thread from 16 October 2023‎ but it doesn't appear to contain that link. Can you provide the diff of the comment you're referencing? Horse Eye's Back (talk) 14:18, 18 October 2023 (UTC)

I think I didn't notice that the start of the thread was from 3 years ago. I noticed the recent (reverted) comment in my watchlist and assumed the whole thread was this month. Thinker78 (talk) 01:03, 19 October 2023 (UTC)

Today someone posted a link in the thread No, today someone posted a short rant in that thread and got reverted. The link itself has been there for three years. And it is not "criticism", it is disagreement. Sanger is for giving the opinion of experts who studied a subject for forty years the same weight as the opinion of some idiot who googled it for forty seconds and who has a megaphone. Wikipedia is for quality control instead. That was nothing new when someone cited it three years ago, and that stale venting by Sanger, someone who has started several failed Wikipedia-like projects based on his "neutrality" pabulum, is not a reason to start this thread, which has only been good for allowing three people to agree with other and pat each other's backs. --Hob Gadling (talk) 15:44, 18 October 2023 (UTC)

It's also important to point out that while people trying to push this argument will say that the entire news media landscape is biased, we don't just use the news (in fact it's not the best available sourcing) - the best available sourcing is usually academic in nature. The argument people effectively make is that high-quality reporting can't be trusted, academia can't be trusted, nothing can be trusted and therefore instead we should weight things to reflect their own personal beliefs, elevating whatever sources they personally agree with. We rely on the best available sources partially because that's what it means to be an encyclopedia, but partially because the alternatives inevitably result in editors introducing their own biases; in practice, the arguments above amount to people saying that we should disregard the best sources in favor of what a few editors' gut feelings tell them is the "natural" weight based on their Twitter or Reddit or Facebook feeds. Part of the reason why WP:FALSEBALANCE is so poisonous is because when we disregard the quality of sourcing, it allows editors to balance things based on their personal bias and what they perceive as the "normal" balance according to their own bubble and experience, all of which is an open sewer of bias. This isn't something unique to them - everyone lives in their own bubbles and has their own biases - but it's only by relying on source quality that we can move away from it; it's why encyclopedias work the way they do. We obtain neutrality by a rigid focus on covering and weighing things according to high-quality sourcing. Arguments that try to pull us away from that are WP:RIGHTGREATWRONGS (in that most of them implicitly seek to "correct" some vast sweeping problem with all of coverage by starting at Wikipedia) and intrinsically WP:TENDENTIOUS (in that they inevitably push a specific view, outlook, or weighting that isn't backed by encyclopedic sources.) There is a point where insisting that the entire landscape of high-quality sourcing is unreliable and that we ought to be trusting some talking head with a blog instead becomes WP:NOTHERE. --Aquillion (talk) 18:14, 18 October 2023 (UTC)

we don't just use the news (in fact it's not the best available sourcing) - the best available sourcing is usually academic in nature – I don't think this discussion would've been needed if this was universally true. There are plenty of articles written using news sources exclusively or almost exclusively, as well as editors seeking to restrict weight of (reliable) non-news sources based on the amount of news coverage. PaulT2022 (talk) 19:11, 18 October 2023 (UTC)

Bad articles that aren't inline with this policy don't make this policy the issue -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:47, 19 October 2023 (UTC)

WP:WEIGHT is far too susceptible to abuse by wikilawyers. I'm not opposed to your view that WP:RS might be a better place for a discussion how to weight the sources. I believe that having an explicit statement either in a policy or a guideline that recognises that sources aren't equal and, for example, academic can be preferrable to journalistic, including in determining weight, would help. PaulT2022 (talk) 19:57, 19 October 2023 (UTC)

• Sanger makes several claims about issues were the Wikipedia articles are correct, and his personal believes don't match. NPOV isn't that you personal believes are as valid as other people's personal believes. I don't see anything here that would improve Wikipedia, or that needs to be taken into account. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 16:37, 18 October 2023 (UTC)

Yes, if this were something that he posted on WP:ANI or the like as a regular user, it would get dismissed as a content dispute. He's upset that editors have determined that the sourcing doesn't support the weight, focus, and article text that he would prefer in several different articles; that's not a NPOV issue, it's a "Sanger has out-there views that aren't backed by a sober analysis of the sources" issue. He also has major WP:FALSEBALANCE issues in that he presumes that any views he holds must be important and significant and well-supported enough to have a major presence in any relevant article, without regard for the weight and focus of the best available sources. --Aquillion (talk) 17:51, 18 October 2023 (UTC)

That's what always gets me about discussion about Sanger, when you actually go and look at what he said it's full of conspiracy theories and debunked nonsense, I don't understand why anytime is wasted on discussing him. I mean The global warming and MMR vaccine articles are examples; I hardly need to dive into these pages, since it is quite enough to say that they endorse definite positions that scientific minorities reject, really just because the link with autism has been thoroughly rejected, the original study shown to be fraudulent, and it's author struck off the medical register, we should still include the "opposite" view for balance? He'd have the article on Earth start with the possibility that it's flat. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:54, 18 October 2023 (UTC)

Just a gentle reminder, this isn't a discussion of Sanger's personal beliefs, it's a discussion of NPOV policy in light of several critiques, not only from Sanger, but from established editors here. I commented earlier that these conversations often go off the rails because people want to make it about Sanger's personal beliefs - this is an unfortunate demonstration. Philomathes2357 (talk) 20:22, 18 October 2023 (UTC)

If that is the case then why does the section title say "Sanger criticism of the NPOV policy"? This is clearly a discussion of Sanger's personal beliefs as they pertain to the NPOV policy. Horse Eye's Back (talk) 20:39, 18 October 2023 (UTC)

But it was started by Sanger being brought up, otherwise nothing has been shown as an issue. Yes there are editors who hold differing opinions, there always are on all issues. We're not as many who criticise Wikipedia seem to believe one united cabal. There are good articles and bad, but there is nothing fundamentally wrong with the idea of FALSEBALANCE. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 20:40, 18 October 2023 (UTC)

• No, this discussion is solely and entirely about Sanger's personal beliefs. The crux of his position is that he feels that Wikipedia is biased because his personal beliefs are insufficiently represented here and because the way be balance sources doesn't reflect his personal perspective; obviously, in order to evaluate whether he's correct, we have to dig into those beliefs to determine whether what he wants in our articles actually reflects the balance of the best available sources. --Aquillion (talk) 21:39, 18 October 2023 (UTC)

IMO some of the arguments here are against straw man mis-characterizations of what Sanger wrote. IMO the most prevalent POV problems come from combining wp:weight with wikiklawyering which the skews what (undisputed factual material) is and isn't covered in the articles.) Sincerely, North8000 (talk) 21:25, 18 October 2023 (UTC)

The problem is that no one is completely unbiased; to reduce the effect of that bias, we determine the weight and focus of articles by looking at the best available sources, not by ask editors for a "gut check" of what they think balance should look like. This is the crux of WP:FALSEBALANCE - editors' gut feelings are always going to be biased by what they consume; Sanger personally believes in eg. a serious controversy over MMR vaccines that high-quality sources don't support. His complaint is that, fundamentally, he wants articles to reflect what he believes the world looks like, and feels that the way we balance due weight should reflect his gut feelings. This is an extremely common perspective among editors who have strong personal beliefs that they believe to be neutral, because they look at articles and get distressed when the balance of sources doesn't reflect their own experiences. I think that Sanger does genuinely believe that eg. MMR vaccines are actually medically controversial! But that shows why the argument he makes can't be taken seriously. "Sanger's gut feelings" aren't something we can use to balance articles. Obviously he wouldn't frame it that way because to him, his gut feelings and WP:FRINGE perspectives are accurate and true and any failure to give them the respect that he personally believes they deserve them means we're not neutral, but that's the gist of his examples and his actual arguments. Obviously when someone holds to fringe beliefs as fiercely and devotedly as Sanger does, and refuses to accept that they are fringe, any article he doesn't agree with is going to be the result of "wikilawyering", but the ultimate reason why the articles don't say what he wants them to say is because he holds beliefs that are wildly out of line with high-quality coverage. Hence why he hates WP:FALSEBALANCE, because it prevents him from just asserting that his personal beliefs need to be the barometer for "balance", and forces him to back his position up with high-quality sources - which he cannot do. --Aquillion (talk) 21:39, 18 October 2023 (UTC)

What undisputed facts are not covered by the MMR vaccine article, the Global warming article, the abortion article, the alternative medicine article? These are not straw man mis-characterisations of his arguments, these are his arguments.

As I said above there are good articles and bad, but a fundamental issue with NPOV? Not based on Sanger's arguments. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:50, 18 October 2023 (UTC)

I think a major flaw in reasoning when having these NPOV discussions is equating political reporting with scientific research. Person A brings up an NPOV issue, wherein factual information about a political event is excluded from an article because of Wikilawyers with a blatant POV of their own citing "weight" and "false balance", and person B says "well, if we include that information, next thing you know, we won't be able to say that the earth is a sphere".

That sort of conversation is a symptom of the problem. They (scientific research and modern political journalism) are not even close to being equivalent, but they are given the same blanket description: "reliable". It makes me cringe, and although I'm relieved to see that there are some editors who have noticed this, it troubles me that this isn't self-evident and obvious to everyone. Something's going to have to be done about that sooner or later. Philomathes2357 (talk) 23:06, 18 October 2023 (UTC)

What is to be done? you ask? Despite this hand-wringing, this is polemical. Nothing need be done, as the "problem" you wish to solve is a political one, and not with NPOV. Andre🚐 05:07, 19 October 2023 (UTC)

I'd be all for giving scientific journals more weight than journalists sources when it comes to reliable sourcing, but that would be better handled as part of WP:Reliable Sources. And as to equating scientific research to politics, this thread states with a post from Sanger I which he uses his personal opinions to disagree with scientific literature. If the problem being highlighted is anything like anything he says, Wikipedia is doing fine. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:46, 19 October 2023 (UTC)

"This is an extremely common perspective among editors who have strong personal beliefs that they believe to be neutral" I don't really get that concept. I hold strong personal beliefs in several political, religious, and social topics. I mostly avoid reflecting these in my writings, because we are supposed to be summarizing sources instead of contacting research. And I don't recall arguments in the style of "I am more neutral than you" in talk pages. Dimadick (talk) 23:00, 18 October 2023 (UTC)

When people are certain that their views are correct but uncommon (e.g., society should get rid of money entirely), they tend not to have many illusions about their views being common. If you explain to them that "neutral" is basically Wikipedia's jargon for "mainstream views", then they usually understand that the Wikipedia is going to be just as wrong as the rest of the world.

The problem appears when someone believes their view is not only correct but relatively common (e.g., all this weird weather has nothing to do with decades of pollution; ghosts are real; children should be taught to believe that Santa Claus is real), and they want to have the article reflect the popular views (where "popular" may actually be popular overall, or may be popular only within their Filter bubble) WhatamIdoing (talk) 04:04, 19 October 2023 (UTC)

+1 I've tried explaining the difference between "both-sides" and "a neutral reflection of sources" to a few new editors, and they're always in that latter camp. A general mistrust of mainstream media/science/ideas also tends to be involved. But Wikipedia is built on such sources, and that point goes further than just the NPOV policy. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 11:02, 19 October 2023 (UTC)

Relevant pages:

• Wikipedia:Neutral point of view
• Wikipedia:Yes. We are biased

Visite fortuitement prolongée (talk) 19:50, 18 October 2023 (UTC)

The only thing I'd change about the NPOV policy is the name. Calling it "neutral point-of-view" makes it sound like articles need to be written from a single neutral point-of-view when in fact that's not only not the case, it's almost the opposite of what the policy actually says. Rather, we write articles from the point of view of the sources without inserting any outside point-of-view, including the active "neutral POV" that Sanger appears to want.

I think that the text of the policy is basically perfect, and indeed the only practical way to maintain neutrality that's compatible with WP:V (or even a much broader commitment to reporting what's true). But the name causes many misunderstandings of the sort that tend to lead to WP:FALSEBALANCE. If I was dictator of Wikipedia, I'd rename it to "no outside POV" or "sourced POV" or something like that. Loki (talk) 00:22, 19 October 2023 (UTC)

"No Outside POV" will be misunderstood as rejecting the POV of outside sources, and only accepting the POV of Wikipedia editors. "WP:Reflect the points of view held by high-quality sources" is long but might be difficult to misunderstand. WhatamIdoing (talk) 04:46, 19 October 2023 (UTC)

The current NPOV policy could be more accurately referred to as "MPOV" - mainstream point of view. Philomathes2357 (talk) 07:01, 19 October 2023 (UTC)

I don't think it's necessarily that either, although it's closer. There's definitely cases where Wikipedia's point of view is outside of common definitions of the mainstream. E.g. we're pretty consistently anti- any kind of pseudoscience or woo, which is mainstream among scientists but not among all people. Loki (talk) 08:31, 19 October 2023 (UTC)

One could also look at our treatment of trans issues as one where we follow the science and not the mainstream. Pretty much anytime the science and the mainstream are out of synch we choose the science. Horse Eye's Back (talk) 15:26, 19 October 2023 (UTC)

Sanger is perennially wrong, sour grapes, and not qualified to opine on any topic. This is not a thread worth discussing at all. Andre🚐 05:05, 19 October 2023 (UTC)

Philomathes2357 wrote: "The current NPOV policy could be more accurately referred to as "MPOV" - mainstream point of view." I beg to differ, although there is one sense in which that is true.

NPOV is not a true "point of view" and has little to do with content. Our content and sourcing can be very biased and far from neutral. It describes how editors should remain agnostic and not take sides, censor content, or otherwise let their own opinions and beliefs color what they write. They should document what RS say, even when they disagree with them. In this sense, editors "edit neutrally", and because Wikipedia is its editors, Wikipedia does not take sides and thus remains neutral. (The caveat comes below. )

Wikipedia's only POV is that it gives more due weight to RS over unreliable sources. As a mainstream encyclopedia, it favors sources that fact-check and use the rules of logic and scientific reasoning to determine what is true and what is false or fringe. I have written an essay about this: NPOV means neutral editing, not neutral content. The nutshell:

"NPOV means neutral editing, not neutral content. We do not document exclusively neutral facts or opinions; we write about facts and opinions neutrally. The "Neutral" in NPOV refers to an editorial attitude and mindset; it is not a true "point of view".

"Editors must edit neutrally when they deal with biased content. Since Wikipedia does not take sides, and because it documents all types of biased points of view, often using biased sources, article content cannot be neutral. Source bias must remain evident and unaffected by editorial revisionism, censorship, whitewashing, or political correctness. We document all aspects of reality, whether we like it or not." -- Valjean (talk) (PING me) 15:51, 19 October 2023 (UTC)

I hear ya. I can quibble with some of what you wrote, but that wouldn't be productive. I've definitely perceived a problem, as have a lot of others, but I'm still trying to locate precisely where the root of the problem is. I'm not sure that NPOV is the main problem, although I think that historic iterations of the NPOV policy were vastly superior to today's, at least for some topics.

I'm intrigued by the community's recognition that business as usual was insufficient to deal with the topic of medicine in an encyclopedic manner (hence MEDRS). I've been thinking a lot about that lately. I think it's a bad thing to treat science and modern politics equivalently. If that problem were solved, the issues I perceive with NPOV may dissipate. One thing we can agree on is that NPOV isn't a "point of view"...I'm familiar with Thomas Nagel's writings about the "view from nowhere". Food for thought. Take care. Philomathes2357 (talk) 01:23, 20 October 2023 (UTC)

I tend to look at POV inclusion/omissions of factual useful information as being a bigger wiki-problem. To combine my comments above into a hypothetical example. A US conservative politician Larry gives an economic policy speech and farts while giving the speech. And then a liberal politician Joe disagrees by giving a spun negative description of the speech. MSNBC coverts the fart (because it is interesting) and then says "a speech described as" and gives Joe's description of the speech. The largest news channel (Fox) covers the contents of the speech. A Wiki-lawyer gets the latter excluded because it's Fox. And so the Wiki article covers the fart and description by Joe but doesn't have the actual contents of the speech.North8000 (talk)

Why would this "Wiki-lawyer" need to exclude Fox? No competent editor would have used it in that context in the first place. Horse Eye's Back (talk) 15:19, 20 October 2023 (UTC)

Well my simple answer would be so that the article covers that actual contents of the speech. But in my example I neglected to specify the likely method of exclusion. It would likely be under wp:weight (rather than wp:ver), saying that what is covered by the US's #1 news channel can't count when evaluating weight. Again, to reemphasize, my #1 concern isn't bias per se, it is the resultant un-informativeness of the article North8000 (talk) 18:40, 20 October 2023 (UTC)

I'm not asking for the method of exclusion, I'm asking for the method of inclusion. In order for something to be excluded someone must be trying to include it. On what grounds would a competent good faith editor have used that source in this context? Horse Eye's Back (talk) 18:48, 20 October 2023 (UTC)

This is a sincere, not smartass answer: The grounds is to include that enclyclopedic information on that speech in the article, which is what we are here to do. If you are talking about some other implied question (e.g. grounds for complying with a specific policy or guideline), you'll need to be more specific. Again, this is a sincere, structural post, not a snappy comeback. Sincerely, North8000 (talk) 19:26, 20 October 2023 (UTC)

And where can I find those grounds in policy or guideline? WP:IAR? Horse Eye's Back (talk) 00:56, 21 October 2023 (UTC)

I disagree with the presumption that no "competent good faith editor" should use Fox News in reference to uncontested factual information, in order to make an encyclopedia article more informative to the reader. Why would a competent editor make the decision to leave an article less informative than it otherwise would have been, just to uphold an abstract notion of source purity? That way of thinking about things is a part of the problem North and I are trying to point out.

Please stop and think for a moment about how ludicrous this conversation would sound to the outside world: the #1 most popular source of information about modern politics in the United States, Fox, is expressly forbidden from Wikipedia articles about politics. Our antipathy for Fox is so ingrained that we will purposely deny our readers information that would be useful in an encyclopedic context, just so that we can avoid citing the #1 most popular source about US politics. I think that would sound positively batshit to an awful lot of regular folks.

I understand the impulse to protect Wikipedia's political articles from lies and bullshit. I share that desire. But I think there's been an overcorrection in our sourcing standards, which is what has caused this widespread "omission" that @North8000 refers to. Philomathes2357 (talk) 02:08, 21 October 2023 (UTC)

Propaganda ain't useful. Fox falsified information, on purpose, on many occasions. It should be deprecated. Andre🚐 02:10, 21 October 2023 (UTC)

I think that's a fine way to think in colloquial conversation with friends, or if you're choosing what news you want to read in your free time, but I don't think it's an appropriate approach to writing a serious encyclopedia that is consulted by billions of people around the world for reasonable summaries of politics and news events. I see no encyclopedic or practical utility in deprecating a source about news and politics, other than literal AI-generated fake news spam sites. Even if I accepted your premise that Fox can be essentialized as "propaganda"...wouldn't it be encyclopedically relevant and noteworthy when a (presumably propagandistic) political opinion is expressed by the most popular source for US news and politics?

If our goal is to write a serious encyclopedia, the idea that we should pretend Fox simply doesn't exist is as "fringe" as the ideas expressed on Fox. Philomathes2357 (talk) 03:00, 21 October 2023 (UTC)

Let's talk again after you read the RFCs for Fox News on RSP, the last 2 or 5. The last one that finally got us to GUNREL was because of the revelations of very specific stuff. Or maybe just go watch Succession. Also, I thought you were quoting Chomsky 5 minutes ago. Do you know nothing of Roger Ailes' relationship with Nixon? Or Rupert Murdoch? Or the Koch brothers? How about dark money billionaires? Or -- wait, was your point about NPOV, we need to let unreliable sources in like the Greyzone and Fox News? The Greyzone is unreliable. Sorry, that's the way it is, philosophy aside, pragmatically speaking. You won't be able to change that easily, if at all. Maybe if the Greyzone really extensively cleaned up their act, you could try an RFC in a few years. Continuing to discuss and grousing about policy that won't change isn't productive. You seem like a really smart person who has done a lot of reading. Most people on Wikipedia are. That's kind of just table stakes. NPOV, you observed earlier, is like a "mainstream POV." Now we're getting somewhere. It is, by design. Wikipedia is bell-curve-land. If you fall outside a normal distribution as many of us Wikipedians actually do, cool. But the encyclopedia may be written for you, but it's not from your perspective. It's from the perspective of the extrapolated normality. Andre🚐 03:05, 21 October 2023 (UTC)

Presumably a competent editor is aware of the existence of WP:BLP. Horse Eye's Back (talk) 03:20, 21 October 2023 (UTC)

I think North's thought experiment is legit. I challenge the conservative news media to have some sources that report it that don't suck. Last I checked National Review, WSJ, could still come in. If nobody covered it but Fox, it really probably wasn't that significant. Are you sure it was a politician or just a wannabe like Mike Lindell? Andre🚐 01:01, 21 October 2023 (UTC)

• This is an essay that dresses up "I hate verifiability" under the guise of "bring back real neutrality". It's the cry of every POV pusher I've ever encountered: someone who insists with seething rage that something is true, but fails to produce reliable sources, or produces sources that themselves invalidate their own point.
• Neutrality can't happen in a vaccuum where none of our other policies matter. If one side has sources, and the other doesn't, then we don't report both sides. If something is demonstrably false and verified as false in multiple reliable sources, we would at most mention that the falsehood exists. Sometimes we would save our breath and avoid mentioning the falsehood at all.
• Are there instances where people have wikilawyered certain falsehoods in, or kept certain truths out? Yes. But that's a failure of editing, and not a failure of policy. Shooterwalker (talk) 16:44, 21 October 2023 (UTC)
  • Bingo! -- Valjean (talk) (PING me) 17:19, 21 October 2023 (UTC)

  +1 Andre🚐 17:22, 21 October 2023 (UTC)

My hypothetical was illustrating several things at once. Aside from the obvious ones , in many areas the boring/wonky enclyclopedic stuff is less reported, making it more vulnerable to being knocked out by POV wikilawyering. North8000 (talk) 19:16, 21 October 2023 (UTC)

Also please note my later emphasis that the more common POV wikilawyer tactic to exclude is to use wp:weight rather then WP:ver. It's not about excluding fringe stuff, it's about excluding stuff that nobody doubts is correct. This is very common. North8000 (talk) 15:29, 22 October 2023 (UTC)

Thats generally because of the BLP policy, if you want to change that you need to get "Wikipedia must get the article right. Be very firm about the use of high-quality sources. All quotations and any material challenged or likely to be challenged must be supported by an inline citation to a reliable, published source. Contentious material about living persons (or, in some cases, recently deceased) that is unsourced or poorly sourced—whether the material is negative, positive, neutral, or just questionable—must be removed immediately and without waiting for discussion." stricken as well as get WP:ONUS and WP:BURDEN deprecated. Changing NPOV does nothing for you in that context.

Horse Eye's Back (talk) 20:00, 22 October 2023 (UTC)

????? I don't where you got that from. You are describing something that is rarely a POV issue. My last post didn't repeat it, but per my previous posts, the most common way that POV wikilawyers use to knock out material which nobody is doubting is wp:weight. I think wp:weight got diverted by such folks from it's original intent which was dealing with issues that had opposing views. North8000 (talk) 17:15, 23 October 2023 (UTC)

Just because it's not doubt doesn't mean it should be included. As a counter example no one could doubt a verified quote isn't from the subject making the quote, whether it should be included is a separate discussion. If the subject making the quote is unrelated to the article subject and has no relevant expertise, then whether they said the quote or not is irrelevant. It's entirely appropriate to edit out material that is not in doubt but is not relevant. If the majority of editors thought that it was relevant when it wouldn't be editted out by WEIGHT. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:16, 23 October 2023 (UTC)

When WP:WEIGHT was written, I think editors still had an idea of what belongs in an encyclopedia article. As a result, we apparently didn't think to include statements like "Editors must provide a basic summary of the subject, including contextual information (e.g., when and where the person lived, if the article is about a person)." Now we get questions like "If only one source cited in the article mentions the person's birthdate, should I remove it because it's UNDUE?" WhatamIdoing (talk) 01:37, 24 October 2023 (UTC)

WP:DOB says that you shouldn't include a date of birth if only one source published it, no idea how long that's been in place but I would have thought it's not new. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 10:29, 24 October 2023 (UTC)

WP:DOB does not say that. DOB says that a single self-published social media post from the subject may be usable to verify a birth date.

But: DOB is only about BLPs, and we write articles about dead people, too. It is always important in an encyclopedia to indicate the "when and where the person lived", even if that information is as vague as "during the 9th century" and "somewhere in the area that is now France". There should never be a question about whether a biography should contain that information. WhatamIdoing (talk) 15:36, 24 October 2023 (UTC)

Oh agree absolutely, especially would agree with statements such as "somewhere in the area that is now France". A lot of Wikipedia's articles contain anachronisms about countries. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 17:53, 24 October 2023 (UTC)

• The issue with your hypothetical is that you framed it in a way that made one way of weighing it "obviously right" (haha, farts.) But we can't rely on gut feelings for how we weigh things, because our gut feelings are biased and unreliable. Consider a much, much more likely alternative: Donald Trump gives a speech about how his opponents are going to raise taxes, and therefore they need to be thrown in jail or murdered in the streets. Or he says that vaccines don't work and that climate change is fake. The vast majority of coverage for the former emphasizes the fact that he called for his opponents to be murdered in the streets and places this in the context of January 6th; the vast majority of coverage for the latter it notes that what he said was false. Fox News' coverage of the former places comparatively far less weight on the calls to kill and murder his opponents, makes no connection to January 6th, and places far more on his discussion of taxes; its coverage of the latter doesn't comment on the accuracy of his comments, or even outright implies there are active scientific controversies involved. How much do we weigh each aspect? Any set of weights is going to make some readers howl bias. The only way we can determine that is by looking at the overarching field of high-quality coverage. And editors who think that Fox News is the highest-quality source available are going to be unhappy with the way we likely end up weighting it. To them, it will feel Fox News is "just stating the facts"; to them, it will feel like they're demanding neutrality when they insist that we weigh things closer to the narrative and weighing that they saw on Fox. Their "gut" says that all that stuff about January 6th and the (obviously, to them, hyperbolic) death threats is no more substantial than the fart from your example. But in practice that's them trying to introduce their gut biases to our article - that is why we demand high-quality sourcing, and why we weigh things according to what the sources say rather than someone going "this aspect is really no more substantial than a fart, amirite?" --Aquillion (talk) 04:11, 31 October 2023 (UTC)

You are tackling a tougher area (bias in coverage) than I was thinking about. I was speaking about excluding factual enclyclopedic information. But I was also reflecting on the reasons that we can no longer depend on the media (or more specifically media that haven't lost a popularity contest at Wikipedia) to be the sole arbiter (on a political topic) on what factual enclyclopedic information gets excluded in an enclyclopedia article. Again, I'm more concenred about excluding factual, useful, enclopedic content than simple bias. North8000 (talk) 14:42, 31 October 2023 (UTC)

No, my point is, they're the same thing. Unless we can back them up by pointing to the overarching weight and focus of the sources, our opinions about what is useful and encyclopedic are merely our biases. Do we include or exclude the fact that, for instance, a politician farted in a speech, or the fact that he threatened to kill his enemies, or the fact that he said something plainly false, or the fact that he was involved in some other thing that some sources consider a world-shattering scandal and others do not cover at all? Which facts you consider relevant and important and WP:DUE are matters of opinion; there are many, many things, for any given topic, which are factually true but which would be absolutely inappropriate and undue for an article on that topic; obviously we do not write articles indiscriminately consisting of every possible fact related to the topic, in every possible context and permutation (and even if we did, of course, we would still have to weight them based on how the article is structured.) Editors understand this intrinsically until something they personally consider vital gets excluded. For example, people exist, somewhere, who doubt the effectiveness of MMR vaccines - the bare existence of such people is, indeed, a fact. This is one of the specific things Sanger is upset about; he thinks we should mention that fact more often, on more articles, and more prominently. But his opinion doesn't carry any weight; we determine whether to mention or not mention it based on coverage in high-quality sources. Sanger doesn't like that, because his gut feelings tell him that it is intrinsically important that those people exist. Similarly, when a politican says something false - it is often a demonstrable fact that it is false. But whether we mention that it is false depends on the weight, focus, and quality of coverage. You can have an article that contains nothing but definite facts and still have it be unacceptably biased, simply based on which facts it chooses to include and how much weight it gives them; and there is no ideal perfect "just the facts" version, since every individual editor is going to have different feelings about which facts are important, how much focus to give them, and so on. But these reflect their biases, not some sort of objective reality. The only way we can produce an encyclopedic article, therefore, is to look at the weighting in the best-quality sources, with more emphasis given to mainstream ones and progressively less emphasis the further on the fringes they are. An editor who comes here with a head full of facts tilted towards what they saw in low-quality or fringe sources, which instilled them with a deep abiding belief that these facts are important, is always going to feel that an unbiased, balanced encyclopedia article on that topic is biased and lopsided. That's their problem, not ours. --Aquillion (talk) 05:01, 1 November 2023 (UTC)

And that argument goes back to my point above that we have the current problem that editors themselves come with their own biases towards topics (not that this group is alone, but there's a fair large number of editors with clear distaste for those on the far right), and think they know what the balance is per UNDUE and other PAG that is based on using the coverage in sources for inclusion. EG: I've talked elsewhere at length about the need to do source surveys before throwing in contentious terms in Wikivoice into articles, as appropriate per UNDUE, but editors still wave their hands, list a small number of sources out of the total available, and claim that that justifies the narrative they would like to see because of their own personal biases. That also goes for removal of info: the whole recent issue of Andy Ngo and being a "journalist" is along the same lines as this. (And to be fair, this is very very common even in trying to write non-contentious stuff. I have seen editors try to apply this to how to label films and other works they feel are "cult classics" or similar with only a few statements from weak sources to back it.) It is why I constantly stress that we should avoid trying to include commentary and opinions that are from the media, uninvolved actors for a topic, from the short term, while waiting for the long-term larger way that such topics should be approached, as you typically will remove the editors' bias in these, your sources are going to be more objective with passage of time, and it will become easier to see what DUE/WEIGHT aspects naturally fall out from all coverage of the sources. The more we focus in the short term on statements that are simple object facts, and worry about the opinion of such aspects later, the more encyclopedic our articles on contemporary topics will be. And in that approach, its very easy to dismiss the need to use any low-quality, fringe sources, because what will be the factual coverage will be what is readily coverage across multiple RSes without any guesswork. Masem (t) 12:36, 1 November 2023 (UTC)

I understand that that is the the current system and the intent of the current system. I'm also saying that sometimes (e.g. on political topics) it doesn't work. Trying to build an enclyclopedia with enclyclopedic content and basing what gets excluded by what our current pathetic fluffy political media does not choose to cover. We should have a framework that guides towards enclyclopedic information and acknowledge some editor judgement guided by that framework. North8000 (talk) 17:07, 3 November 2023 (UTC)

Part of the problem is that many folks do not recognize that there is a problem with what you call "our current pathetic fluffy political media". It appears to me that folks who edit political topics have a lot more faith in political media than the average person I encounter IRL - certainly more than the average political scholar that I know. Some folks around here treat political media reports from "RS" as functionally equivalent in rigor and trustworthiness to academic scientific publications.

Something needs to be done to raise more awareness of the unique characteristics of modern political media. Ideally, the reality of political media would be reflected in the "framework that guides towards encyclopedic information" that you refer to, @North8000, but if a plurality of editors don't even accept the premise that there's a problem here in need of a solution, instituting such a framework is impossible.

How would you suggest that awareness of this issue be raised to the broader community? Philomathes2357 (talk) 19:58, 3 November 2023 (UTC)

@Philomathes2357: Thanks for the question. I think that folks are somewhat aware, but that policies and guidelines (particularly this policy) do not reflect awareness of it. I think that discussions like this raise awareness of the specific issues and also build a little steam towards challenging what needs to be challenged leading to fixing what needs to be fixed. North8000 (talk) 20:23, 4 November 2023 (UTC)

and some people reject the premise. Have you guys realized that the hard sciences have errors, retractions, and for-profit "science" paid-for pharmaceutical companies? Just because it's a "hard science" doesn't make it reliable - and don't get me started on economics! Social sciences do try to do a good job - and it's up to us (and their reputation in other reliable sources) to ensure we focus on the highest quality sources. But experts in political science, or columnists who have been covering a political beat for years, are indeed rightfully considered experts, and Wikipedia reflects their opinions on what is factual and what should be considered de facto factual (i.e,. 80% likely to be factual). But everything in human publications has tons of errors no matter the field. That's why our policies encourage focusing on up-to-date, reliable, and comprehensive reviews - which exist in all social sciences. Andre🚐 21:06, 4 November 2023 (UTC)

@North8000 I agree, these discussions are useful and important. You've said a lot of things in this thread that I agree with, some of which I had never considered before. I've been assembling a list of academic & popular sources that critically analyze modern political media. A few of them are on my user page. If you know of any books, articles, or journal publications that address this topic, I'd like to know about them.

@Andrevan I hear you. You're totally right that scientific publications can be affected by good-faith human error & by corruption. The classic examples being fossil fuel companies paying researchers to "debunk" climate science and pharmaceutical companies paying researchers to speak favorably of their drugs. I agree completely that this problem exists, and that the problem's existence doesn't negate the value of science.

However, I think you are accidentally drawing a false equivalence. There are fundamental differences between humanistic & scientific pursuits in academia, and there are fundamental differences between academic research and the news, and I think these differences should be reflected in RS policy. I also think that our treatment of politics is sub-par, bordering on horrible, and that there are potential policy solutions that will vastly increase the scope & quality of our coverage. It will take a lot of conversations for such policy solutions to take shape, but I think the issue is worth our sustained attention as a community. Philomathes2357 (talk) 22:46, 4 November 2023 (UTC)

And we can agree to disagree, as i oppose such change, and I think Wikipedia's coverage is very good. You're welcome to your opinion, but you also might be trying to WP:RGW. The field of politics in 2023 is going to be what it is. Chaotic and a hot mess. Dubious propaganda left and right. Fascists denying truth, leftists denying pragmatism, and moderates ignoring their peripheral vision. That's life. Unfortunately, science is also subject to politics, squabbles, and holy wars. Just look at stem cell research, genetics, the ethics of abortion, etc. Science and politics are hopelessly intertwined. Yes, there are differences. One of the differences is that we have a much higher sourcing standard for topics of medical or hard science import, whereas we recognize that we cannot hold all sources equal, see, for example, WP:PARITY. I suggest you take a gander at that. Wikipedia will never be welcome to WP:FRINGE in any field. Each content area has different sourcing standards. And RS policy already reflects that academic research and reliable books and journal articles, are given precedence and weight over breaking news or any kind of news analysis in magazines and lay-person periodicals. Andre🚐 22:54, 4 November 2023 (UTC)

What confuses me in this logic, is why would different aspects of a politician's speech need to be weighted using news in the first place. If the speech had lasting impact akin to We shall fight on the beaches, it's highly unlikely that its aspects can be adequately weighted by the contemporaneous news coverage and, more importantly, better quality sources would exist to do so. If the news reports are being used merely as a reliable source for politician's views, surely, WP:PRIMARY would apply. Is the idea here that editors should go beyond facts and perform frequency analysis of the news sources to determine relative importance of politician's views? How is that not original research? If it's neither of these two cases, then why Wikipedia needs to include reactions to the speech that had no lasting significance?

Yes, that core part of WP:weight was unworkable on day one, long before its newer problems. The the putative frequency analysis which it is dependent on impossible to do and doubly impossible to to objectively, and, as you note, would be wp:or if it was done.North8000 (talk) 23:25, 4 November 2023 (UTC)

To use a (hopefully) less controversial example, I looked through London gallery and artists' articles recently and they're extensively sprinkled with mentions of events related to the Stuckists, each supported by multiple national newspapers. 2 out of 7 controversies and 1 out of 7 photographs used in Saatchi Gallery are about the Stuckists. The depth of the content supported by news coverage is non-existent: the Stuckists reported Saatchi to the Office of Fair Trading alleging unfair competition. The complaint was not upheld. It almost looks like COI editing at times, yet, most likely was done in good faith – 'every newspaper wrote about it, we should too'.

I fully support use of amount of coverage to tell fringe views apart from non-fringe, but I'm struggling to understand how using comparative amount of news coverage of different topics to determine what's included in the article is helpful. Content that gets included like this sticks out like a sore thumb and is always obvious to any reader not emotionally involved with the article. PaulT2022 (talk) 19:27, 2 November 2023 (UTC)

• This is sort of a side note… but Masem’s comment reminded me of it… one thing that we (as a community) are very bad about is going back and rewriting articles once the subjects/events are out of the news cycle (and we can better evaluate the historical impact of what was reported at the time). We need to do more of that. Blueboar (talk) 12:50, 1 November 2023 (UTC)

  There are a couple of completely separate but concurrent discussions (I think at VPP and NOT) dealing with how excessively detailed our coverage of current events and those involved has become and the to clear up (including deletion where appropriate). This discussion is certainly related because of the tendency for editors to overload current events that are contentious with every ounce if criticism that can be found, rather than focusing on the bigger, 10+-year picture. Masem (t) 13:41, 1 November 2023 (UTC)

  You may be interested in WT:NOT#WP:NOTNEWS/unfolding news stories if you haven't seen it already. There's a discussion about a new template, Template:Old_news to highlight articles with that issue. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:09, 1 November 2023 (UTC)

  Wikipedia_talk:What_Wikipedia_is_not#WP:NOTNEWS/unfolding_news_stories --Hipal (talk) 20:00, 3 November 2023 (UTC)

  d'oh, my link now works. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:49, 4 November 2023 (UTC)

Sanger criticism of the NPOV policy - break #3

References