Wikipedia talk:Neutral point of view

Preferring allegedly neutral sources

@Bob K31416 recently added this sentence:

• "Reliable sources that are more neutral are preferred over biased reliable sources if they both support the same given material.  Biased reliable sources may be used if a more neutral reliable source is not found for the given material."

I don't think this is good advice, and I don't think that the community actually has this preference in practice.

You should use the best sources you can for building an article, but once you know what content is needed, it is not actually important whether a given uncontested fact is supported by a WP:BIASED source or a "more neutral" one. The source needs to be strong enough to support the claim, but almost any source is going to be strong enough to support uncontested facts. Rearranging the deck chairs – I mean, rearranging the little blue clicky numbers, so that the one I perceive to be "more neutral" gets used somewhat more times in an article, and the one I declare to be "biased" is used slightly less, is not an important or valuable activity, especially for the 299 out of 300 page views in which zero readers actually try to read the source.

The important thing is that we get the content right. If the content needs to be "Biden is the 46th president of the US", then it does not actually matter whether that's followed by a ^{[more neutral source]} or a ^{[biased source]}. What matters is that the sentence itself is correct, verifiable, and neutral.

To quote the relevant item from the Wikipedia talk:Verifiability/FAQ:

What if the source is biased?

Sources are allowed to be biased or non-neutral; sometimes these are the best possible sources for supporting information about the different viewpoints held on a controversial subject. However, the resulting Wikipedia articles must maintain a neutral point of view.

The article must be neutral, and it does not matter if the source is. WhatamIdoing (talk) 04:04, 28 July 2024 (UTC)

There is also the issue of who decides which source is 'more neutral', something open to both personal and socio-culture bias. And as you say content has to be verifiable, if the source is biased isn't an issue as long as the content isn't biased. -- LCU ActivelyDisinterested «@» °∆t° 12:17, 28 July 2024 (UTC)

No. The the food article on presidential inauguration festivities, may be reliable, but it is not the source to use for that. And it seems foolish to suggest any source will do. -- Alanscottwalker (talk) 13:21, 28 July 2024 (UTC)

"Not the source to use for" what? That sounds like a perfectly fine source to use, assuming that you are writing a paragraph about the cake served at presidential inauguration festivities, which is, in fact, something we sometimes write about. WhatamIdoing (talk) 16:23, 28 July 2024 (UTC)

For your prior approved use, which should have been obvious. Your defense of crap sources is still crap. -- Alanscottwalker (talk) Alanscottwalker (talk) 18:42, 28 July 2024 (UTC)

So you think that a sentence like "Biden is the 46th president of the US" should not be supported by a newspaper article saying "At the inauguration festivities celebrating the moment when Biden became the 46th president of the US..." Instead, you would like to see it supported by a newspaper article that isn't about the festivities associated with him becoming the 46th president. A bias towards "serious" sources, vs "frivolous" ones, is common in our community, and I personally wouldn't start in the style section to find such information.

However, I don't see how that – the question of whether a fact-checked news article on the front page is better than an equally fact-checked news article in the style section of the same newspaper – is relevant to whether the source is "biased" or "more neutral", which is the subject of this thread. The recently added text says you shouldn't cite even front-page, main news articles from The Guardian or The Daily Telegraph or Mother Jones or The New Republic or Politico or Reason or pretty much any other news source for whether Biden is the 46th president, because they're WP:BIASED. Instead, you should preferentially cite "more neutral" sources, of which exactly none are agreed by everyone to be "more neutral" when it comes to political news. WhatamIdoing (talk) 21:06, 28 July 2024 (UTC)

With respect to your first paragraph, it bares no relationship to what I said. Your last paragraph makes little sense, because it is as of you have no idea what you are writing about or what good sources for the subject are. Indeed, the effect of your argument is, 'Wikipedia does not care about sourcing', we will just send you to look at basically anything. Alanscottwalker (talk) 21:58, 28 July 2024 (UTC)

Let me see if I can break this down:

• Bob boldly added a sentence to this policy. That sentence indicated that, when you have two sources, both of which are reliable, and both of which say the same thing, and one of them is "more neutral" and the other is "biased", then it is a plusgood thing to cite the "more neutral" reliable source instead of the "biased" reliable source. This is all about which source to have in the little blue clicky number, and not at all about what the article says.
• WP:RSP (rightly or wrongly) says that The Guardian is a biased source when it comes to politics.
• Imagine that you have two sources in hand:
  • a bona fide hard-news, fact-checked story from the "biased" The Guardian that says Biden is the 46th president, and
  • any reliable source at all that is "more neutral", e.g., an article in Food & Wine that's primarily about the food served at the inauguration balls^[*], that also says Biden is the 46th president.
• If you want to follow Bob's advice, you would cite the second source, even if it's what you'd call a "crap source", because it's "more neutral" and the newspaper article is "biased". In other words, so long as the source is just barely non-crappy enough to be reliable, this proposal claims that it is more important for the source to seem "neutral" than for the source to be timely, relevant, fact-checked, appropriate to the subject matter, etc.
• If, instead, you want to follow my advice – which I think is also your advice – you would cite the best source, even if the best source is a biased source. In this simplified example, the best source is the internationally recognized news organization and not the foodie magazine.
  • And if you really wanted to follow my advice, but which is probably not your advice, you also would not worry too much about the strength of the source behind an uncontested^[**] statement of fact, and instead focus your energy on things that actually matter, like how to balance domestic vs international efforts, or economic vs social policies, or political vs personal content, etc., since there's a much bigger chance that readers will read some of the article. In Joe Biden, the median source can expect to be clicked on approximately once per quarter million page views. The first sentence, on the other hand, will be read pretty much every time.

[*] Note that this is only an example; the text as submitted prefers any and all "more neutral" reliable sources over any and all "biased" reliable sources without consideration for any other factors. It could be that there's still a small town newspaper somewhere that everyone would agree is "more neutral".

[**] By anyone who is not part of the Election denial movement in the United States. WhatamIdoing (talk) 23:07, 28 July 2024 (UTC)

Your policy view is still appalling, as it is not focused on doing good work, it is at best a transactional dodge from doing good work. Alanscottwalker (talk) 14:42, 29 July 2024 (UTC)

I think I am focused on giving accurate reasons for preferring (or rejecting) a given source. If you're working on an individual article yourself, then a gut feeling may be good enough to decide that you'll choose this source over that source. But WP:Policy writing is hard, and one of the reasons it's hard is because you have to get both the rule and the explanation right. That means, for example, that you have to question whether this rule is about independent sources or secondary ones (at one point, Wikipedia's policies used those words interchangeably), and you have to question whether bias is the most important characteristic to care about, and you have to question whether 'bias' is actually the right word to describe the problem being addressed (answer in this case: No. Bob actually wants a rule that says not to cite reliable sources that say something derogatory about the subject if we can avoid doing so).

I think you might be more intuitive about this than I am (i.e., than I feel I can afford to be with my policy-writing hat on). For example, here:

• Bob says we should always prefer unbiased sources (when we have a choice).
• I say that we need to think about this, because we have a long-standing rule saying that biased sources are okay. Bob's addition would create a contradiction between WP:RS and NPOV.
• You say that if you want to say who the 46th POTUS is, you shouldn't cite an article about the food at the inauguration balls.

That latter point, though I can see the connection now, has nothing to do with WP:BIASED sources. "Shall we cite biased [or derogatory] sources, especially given that RSP has a habit of declaring reputable newspapers to be biased?" has no connection to "Shall we cite a fluffy human interest story about the cake served at an after party, when we could be citing a respectable hard news story instead?" WhatamIdoing (talk) 15:51, 29 July 2024 (UTC)

If this was talking about source independence, that would be one thing, as implicitly the more independent the source is from the topic at head, the more likely it will remain more neutral (though as noted, it can still possibly be biased). But we can't start with expectance of more neutral sources for the problems mentioned above. --Masem (t) 13:35, 28 July 2024 (UTC)

Adding more. Per policy, we also avoid published opinion for facts. Alanscottwalker (talk) 13:42, 28 July 2024 (UTC)

This is true, but it's not relevant. An opinion piece can be quite neutral ("We think voters have a difficult choice between two equally qualified candidates...") and a purely factual piece can be heavily biased (e.g., by presenting only facts that support my view and omitting facts that prove my preference wrong). WhatamIdoing (talk) 16:27, 28 July 2024 (UTC)

Don't give me that facetious stuff. It's like you are rewriting policy in your mind to "always use crap sources." And your argument is not improved by your latest, 'its just a matter of semantics.' We don't use opinion pieces because their purpose is to make facts relative in service of opinion, and that is bias. -- Alanscottwalker (talk) 18:44, 28 July 2024 (UTC)

I believe that if you spend some time in the archives for WP:V and WP:RS, you will find that we prohibit citing opinion pieces for facts because opinion pieces are not generally subject to fact checking (at the few institutions that actually do fact checking these days).

The question in this thread is whether we should prefer sources not because they are "more reliable" (or "less crap", if you prefer), but because they appear to be "more neutral" (or "less biased"). An opinion piece can be neutral; a factual piece can be biased. Therefore, when trying to decide whether to prefer more-neutral/less-biased sources, considerations of whether the source in question has some qualities other than being more-neutral or less-biased (e.g., whether it is an opinion piece or a fact-oriented piece) is a red herring. WhatamIdoing (talk) 21:13, 28 July 2024 (UTC)

Since I've been in those discussions, whatever you believe about them is irrelevant. And as I have explained your ideas of bias and neutral make little sense. Alanscottwalker (talk) 21:48, 28 July 2024 (UTC)

Sure, we were both in a discussion about this in 2012, wherein @Blueboar said There is a line we can draw between "News" and "Opinion". The editorial staff fact checks "News" articles... they may not fact check opinion pieces (such as op-ed columns). Thus, opinion pieces have limited reliability on Wikipedia. WhatamIdoing (talk) 23:35, 28 July 2024 (UTC)

Objectivity of a source with respect to the item which is citing it certainly does affect the real world reliability of the source in that context/instance. We don't have much on this in policies. The item mentioned in the OP sort of works in opposite directions on this. One portion is of little consequence...it only applies where there there are multiple sources available for the same material and just says to prefer the less biased one. The latter part seems to bless use of biased sources when unbiased ones are not available. This is too complicated to comment on because nearly every possible example will involve interaction with other policy provisions. North8000 (talk) 14:42, 28 July 2024 (UTC)

• A concern has to do with cases where material is supported by a source and the source is replaced by a more biased source that doesn't improve the support of the material. Or having an edit reverted that would have replaced a biased source with one that is just as good for supporting the material but with less bias.

Is trying to avoid these kind of situations an acceptable goal? Thanks. Bob K31416 (talk) 19:26, 28 July 2024 (UTC)

I think that's a case in which you want to make case-by-case decisions and defer to any consensus that forms on the individual article's talk page. For example:

• WP:RSP says that "Some editors consider CNN biased". Those editors would probably think that replacing anything cited to CNN with something cited to The Wall Street Journal would be a case of replacing a "biased" source with a "more neutral" source. Others would feel quite the opposite, and would say that the biased WSJ needs to be replaced by the more neutral CNN. I think this would be a complete waste of editors' time, regardless of whether it's replacing CNN with WSJ or vice versa. IMO that sort of edit does not improve anything. I would not make such an edit, and I would not revert anyone who made such an edit.
• OTOH, an article like Donald Trump has problems. We are citing too many WP:PRIMARYNEWS sources. We are citing just too many sources. There are 838 refs on that page. That's not good for anyone. If you replaced 100 of them with sources to a reputable biography, then that would be great, and it really would not matter to me whether the biography in question was Donald Trump is the Best Thing Since Sliced Bread or The Evil Genius Who Will Destroy America or Trump: Advantages and Disadvantages. The fact that Trump was the duly elected 45th president of the United States, and that the Wikipedia article should say this, does not depend one whit on whether the cited source is pro-, anti-, or wishy washy.

WhatamIdoing (talk) 21:32, 28 July 2024 (UTC)

Consider the case of a source for the fact that Councilman John Doe lived in Los Angeles in 1992. One source with the information is titled, "John Doe and DUI". It is about a member of John Doe's staff driving drunk and being arrested for DUI and there was an implication that John Doe tried to get the case dismissed, although it was not clear that John Doe ever did that. It also says that some believe John Doe is the worst councilman. Another source with the residence information is titled, "John Doe Elected to City Council", and is a factual news story without opinion or possible misrepresentations. They both are in traditionally reliable sources and have the information that John Doe lived in Los Angeles in 1992. There is a discussion amongst Wikipedia editors with no other details discussed. Those favoring the DUI source prevailed and noted that it is irrelevant whether or not the source is biased. Which source is preferable for Wikipedia? Bob K31416 (talk) 00:40, 29 July 2024 (UTC)

In terms of what you had added, this example would be a case where we prefer an objective, just-the-news article over an op-ed or something more reaching for analysis. I wouldn't call that a case of being more neutral, but being more objective or "absence of commentary by the journalist". Masem (t) 00:54, 29 July 2024 (UTC)

There's nothing in this story that suggests the DUI news article is anything other than an object, just-the-news article without any 'commentary' by the journalist. That some believe all the other council members are better is, after all, "just the facts". However, we prohibit citing opinion pieces (encompassing both op-eds and the actual 'eds') for factual content. WhatamIdoing (talk) 00:56, 29 July 2024 (UTC)

Its more about the "implication" statement there. And that becomes a tough line to learn how to read that, but the equivalent would be how Fox News tends to report on political news on its "news" programs. Its all based on "fact" but in a way to wildly swing the perception and creating implications that are not there. Now, if its other parties that created the implications and the article's just quoting that, I agree that's a "just the facts" news article. Masem (t) 01:00, 29 July 2024 (UTC)

Every fact has implications, and the fact that the complaint is about 'implications' can be evidence that the journalist is sticking rigidly to the demonstrable facts and refusing to go even one character beyond them.

Journalists sometimes have to write articles in which they know that Paul Politician has tried to exert some improper influence, but can't get anyone to say that on the record. In that case, the factually accurate article may "imply" something, but the implication is nothing but the facts. For example: Imagine that the employee was arrested on Saturday night, and the usual process is to keep people in jail until the first hearing in court, which will be on Monday morning. Imagine, too, that the actual facts are that the politician has called the jailhouse several times a day since the arrest, yelling that the ordinary system is terribly inconvenient for him and their employee needs to be released sooner and the politician is thinking that it's time for the city to review the jail's budget.

But: Imagine that the newspaper's lawyers say that the journalist can't publish the actual facts unless and until they can get one person who will go on record saying – in effect – "Yes, I really did willfully violate the jail's confidentiality rules by telling a reporter all about the phone calls made by concerned friends and family members, so please fire me."

If we knew the facts, we wouldn't be sneering at the journalist for "implying" the actual facts. It's only because we don't know the facts ourselves that we wonder whether this is a case of being unfair to the politician or actually a case of being altogether too gentle on an instance of abuse of power. WhatamIdoing (talk) 01:14, 29 July 2024 (UTC)

Bottom line: If you don't have evidence that the 'implication' is factually wrong, then you should look to the reputation of the source. If it's good (e.g., Los Angeles Times, for a politician in LA), then editors should assume that any 'implications' are warranted and fact-based. WhatamIdoing (talk) 01:17, 29 July 2024 (UTC)

Any. Either. Both. None of the above (in which case, you'll have to supply a better one).

We are not here to WP:RIGHTGREATWRONGS, or even to soothe the minor irritations caused by a headline writer. We are adding sources that make it possible for people using the encyclopedia to check whether the information (e.g., that Doe lived in LA in 1992) comes from a reliable source (e.g., a newspaper article that WP:Directly supportsthat particular sentence). If the newspaper with the public-relations-disapproved headline happens to be a reliable source for the material in question, then it is acceptable to cite the reliable source. Whether to do so, as with everything else, is up to editors working on the article.

Were I unhappy with the consensus, my strategy would likely involve finding a much, much better source and re-writing the article from scratch. WhatamIdoing (talk) 00:55, 29 July 2024 (UTC)

• My problem with the instruction is that it assumes we are starting with material and then finding sources to support it… when we should be starting with the sources and then writing material based on what the sources say. Blueboar (talk) 20:38, 28 July 2024 (UTC)

  Every time someone fixes a {{citation needed}}, that is exactly what we're doing. When you are starting or significantly re-writing an article, you should start with a handful of excellent sources and proceed from there. But when you are just trying to demonstrate that some existing text was not made up by a Wikipedia editor, then you start with the material and find a source to support it. WhatamIdoing (talk) 21:15, 28 July 2024 (UTC)

  But that too is the wrong way to go about it. When we see a cn tag, we should read lots and lots of sources that might (or might not) support it. If necessary, we rewrite the material to better reflect what all those sources say. We might add context and opposing viewpoints, or explain nuance. We might even omit the material completely because not enough sources cover it.

  But, ok, let’s say that there are enough sources that cover the material (as written)… at that point it is simply a matter of using editorial judgment to cite the source we think best supports it.

  That is where different editors can disagree. I might think source X best supports it, while another editor thinks source Y best supports it. We could spend hours debating these sources… But that is really a pointless debate since nothing says we can’t cite both.

  As for “choose neutral sources over biased sources”… the only way to know which sources are neutral and which are biased is to read lots and lots of sources. And we may find that a source we initially thought was neutral is, in fact, biased (or vise versa). Blueboar (talk) 00:25, 29 July 2024 (UTC)

  {{citation needed}} is a request for one (1) reliable source. It is not a request for a full review of the subject area. If you wish to treat that tag as meaning {{citation-needed-neutrality-dubious-clarify-by-whom-original-research-POV-statement-copyedit-peacock-tone-undue-weight-relevance-contradictory-speculation-inline}}, then I'd never stop you, but that's not how the community actually handles those in practice. WhatamIdoing (talk) 00:45, 29 July 2024 (UTC)

  Which is one of Wikipedia’s flaws. Blueboar (talk) 01:03, 29 July 2024 (UTC)

  Or perhaps it is evidence that editors know how to prioritize our work. WhatamIdoing (talk) 01:15, 29 July 2024 (UTC)

• I agree with WAID that this policy should primarily be focused on getting the content right. I mostly agree with the advice in the proposal in theory, but even if I agreed 100%, I would still oppose adding it to the policy. Firefangledfeathers (talk / contribs) 01:24, 29 July 2024 (UTC)

  Wikipedia informs the reader with the text and by supplying references where more information on the subject can be found. By unnecessarily choosing a biased reference, the reader is directed towards one point of view. Bob K31416 (talk) 16:27, 29 July 2024 (UTC)

  @Bob K31416, did you know that academic research (doi:10.1145/3366423.3380300) says that 99.7% of the time, readers don't read any refs? The reader isn't "directed towards" any point of view in the refs, because the readers almost never read the refs at all. WhatamIdoing (talk) 23:37, 29 July 2024 (UTC)

  I agree this text should not be included. Our article text should be neutral but our sources need not be. Neutrality is but one aspect of a source's characteristics we can judge and other characteristics may be more important. In some areas editors cannot come to agreement about whether sources are neutral or not, or take a very broad approach to dismissing or approving of sources, without consideration of the author of the piece or type of article. I can't see how it is helpful to fill our article talk pages with arguments of neutrality which ultimately aren't important if the article text remains the same. I think Bob K31416 is confusing sources with external links. Our sources are not "where more information on the subject can be found" but are closer to "proof we didn't just make this up".

  However, I disagree with WAID that stats about how many editors read refs is relevant to how we choose which to cite. It is like saying that because there are so few people editing Wikipedia compared with those reading it, we shouldn't waste our time making the editing experience any better. I agree that having excellent article text is more important than having sources that are better than "sufficient to do the job", we can't just dismiss the important of sources because few readers look at them. I think we should regard our refs as important primarily to other editors and secondly to those readers who are diligent enough to question what they just read. That there are few such diligent readers is not important. They exist and we should serve them. And of course we should serve other editors, as this is a community project written collectively. Having a great source makes others editors (and diligent readers) confident that the text is well supported.

  There is another kind of source, though, and it isn't one that appears at the bottom of our articles. It is the sources that editors have read and perhaps discussed with others on the talk pages. A non-neutral source can be bad for the article not because it necessarily gets the fact wrong, but because it has the weight all wrong. We end up including information that misleads the readers simply be being told to them as though it was important or fair. We might call them "hate facts" as they are factoids that people who hate the subject love to tell each other. I don't know the solution for that though, because our better neutral sources simple don't mention these "hate facts" at all. And our weight policy is written with "big book on a famous battle" in mind, not current affairs and news articles by generally biased press. -- Colin°^Talk 18:14, 31 July 2024 (UTC)

  I'd like to strongly support this whole comment of Colin's, especially A non-neutral source can be bad for the article not because it necessarily gets the fact wrong, but because it has the weight all wrong. In my view, the bias of a source isn't often relevant to whether it publishes accurate "facts", but it is very often relevant to whether or not we should include these "facts" in articles and whether to give them WEIGHT alongside other facts reported by less-biased sources.

  If we do shift focus in this way, it will remove some of the wikilawyering around "this source is so biased it should never be used" vs. "this source is biased but its facts are as good as anyone else's" - a fairly sterile antinomy that wastes editor time. There will be teething issues, though, since I don't think The Guardian's choice of facts on one side of UK political issues is as tendentious as The Telegraph's in the other direction, but current entries at perrenial sources would suggest that it is equally or more tendentious. Still, were we to adopt this as a community principle, we could work out which biases really should count against DUE inclusion, over time. Newimpartial (talk) 22:04, 31 July 2024 (UTC)

Comment - I think the expressed topic, "should a fact be cited to a biased source when a less biased source is available?" is not as important to the enwiki community as are several adjacent topics. Some of these more interesting questions include (in rough order of proximity to the current discussion):

• Should the title of the source be a factor in deciding which source to use, when choices are available? (For example, some RS have headlines the content of which might well fail WP:BLP requirements if stated in an article.)
• What principles should govern (whether and how) facts that are only documented in biased RS are included in articles?
• (related but different) How should editors understand the ways the biases of sources interact with other factors in assessing DUE WEIGHT?
• How do the biases of sources interact with other factors in deciding on terminology in articles (e.g., to assess compliance with WP:NPOV)?

While I am not at all suggesting that the current discussion be expanded to include all of these questions, I think each of them raises more fundamental issues that, if the community reached consensus on any related principles, might resolve questioning about choices of biased and less biased sources en passant, as it were. Newimpartial (talk) 16:42, 29 July 2024 (UTC)

IMO the two most common real-wiki examples where biased sources come into play are:

• Wanting to apply a broad vague negative value-laden term to somebody or some group. My answer on those is to provide more specific informative info in that area rather than such a term. And use sources that do that.
• Using source bias evaluation to try to include or exclude material. Sometimes the warrior just invokes something from wp:verifiability, but usually they also invoke wp:weight. And there, depending on which argument serves their side, they can select and quote parts of policies that say that biased sources are OK, or else seek to disparage the source because it's biased.

Sincerely, North8000 (talk) 17:11, 29 July 2024 (UTC)

But it is often a WP:WEIGHT issue (in fact, that's what I came here to say.) We can use WP:BIASED sources, but an article that relies on a non-representative sample of sources that are all biased in the same direction is going to itself be biased, at least if this leads to it giving their views disproportionate representation. The hard part, of course, is figuring out what is representative, but there's a reason why people with complaints of this nature at WP:RSN often get pointed to WP:DUE instead. And this is the answer to the proposal above; NPOV already touches on how to handle biased sources, it just touches on them in DUE. --Aquillion (talk) 02:26, 28 September 2024 (UTC)

WP:RPLA of WP:YESPOV

A 2010 edit removed the bolded text serving as a destination for WP:YESPOV and WP:NOPOV in contravention of WP:RPLA. I was confused why WP:Vandalism said core content policies of neutral point of view (which does not mean no point of view) but not the NPOV policy itself or its FAQ. To fix this, I archived the content to WP:No point of view.

Should we reintroduce the concept that "'neutral point of view' doesn't meant 'no point of view'"? If not here, we could do it at WP:NPOV/FAQ. 174.89.12.36 (talk) 03:50, 9 September 2024 (UTC)

I don't believe the current destination is astonishing, and is a much better description of how neutrality should be achieved than the decade and a half old text. Policies are not statutes set in stone they change with time and practice. The current wording still shows that NPOV doesn't mean 'No POV', but rather the POV found in reliable sources. -- LCU ActivelyDisinterested «@» °∆t° 12:24, 9 September 2024 (UTC)

Can we have a sentence saying NPOV doesn't mean 'No POV', but rather the POV found in reliable sources? Currently this meaning seems present but diffused throughout the first 3 list items, whereas I would like something boldfaced and more explicit.

Per "better ... than ... old text", I self-rv'ed the 3rd page. Either it's just me who's slow, taking days to realize the redirect intended to target the 3 items or it should really be made more obvious. 174.89.12.36 (talk) 15:21, 9 September 2024 (UTC)

I would be all for such simpler language, but unfortunately past discussion show that it would likely be controversial. Let's see if anyone else has an opinion. -- LCU ActivelyDisinterested «@» °∆t° 16:10, 10 September 2024 (UTC)

I've also reverted your changes to YESPOV and NOPOV, whether the current policy should be changed or not they should point to the current policy. -- LCU ActivelyDisinterested «@» °∆t° 12:41, 9 September 2024 (UTC)

Clarification of BESTSOURCES

I'd added a bit to WP:BESTSOURCES, and was asked to gain consensus first.

What we currently have is

In principle, all articles should be based on reliable, independent, published sources with a reputation for fact-checking and accuracy. When writing about a topic, basing content on the best respected and most authoritative reliable sources helps to prevent bias, undue weight, and other NPOV disagreements. Try the library for reputable books and journal articles, and look online for the most reliable resources. If you need help finding high-quality sources, ask other editors on the talk page of the article you are working on, or ask at the reference desk.

I've recently seen confusion at an article talk on whether a list of BESTSOURCES (which often gets linked here) excludes other sources from being used, even if they have consensus for doing so, if they aren't on an article's BESTSOURCES list. I'm thinking it might be helpful to make it clear, with something like:

At some articles, editors may develop a BESTSOURCES list of sources that have already gained consensus for use. The creation of such a list does not imply that no other sources may be used in an article but only that content cited to other sources may require consensus for use of that source for that content.

Valereee (talk) 18:23, 27 September 2024 (UTC)

Besides the confusion you saw at that article, are there other pages/disputes where this guidance would help? Firefangledfeathers (talk / contribs) 18:27, 27 September 2024 (UTC)

I can't say I've personally seen the same arguments elsewhere, but I do understand where it's coming from and why there might be argument on both sides, with one arguing "Can't use it, not on the list" and the other arguing "But it's a minor point simply not easily found without a complete reading of all sources". In some CTs, it might be helpful to make sure there's clarity that a BESTSOURCES list doesn't limit editors from gaining consensus to use a source not on the list. Valereee (talk) 18:44, 27 September 2024 (UTC)

Oppose. If there really are articles where editors have made a BESTSOURCES list, I fear that will help whoever comes first against change, like the FAQs on talk pages that say this has already been discussed so go away, but consensus is not required for what's on our list. I fear too that "source" will mean the publisher, as happens on WP:RSN too often. Peter Gulutzan (talk) 18:43, 27 September 2024 (UTC)

@Peter Gulutzan, can you expand? This is actually to clarify that editors can use sources not on a list, not that they cannot use those sources. I feel like your objection is to the opposite happening -- people saying "Not on the list, can't use" -- which is what I was trying to prevent? Maybe I'm misreading your post? Valereee (talk) 18:46, 27 September 2024 (UTC)

Saying as policy that people may make article BESTSOURCES lists is blessing them. Peter Gulutzan (talk) 20:04, 27 September 2024 (UTC)

Ah! I understand, I think. Yeah, I didn't mean 'may' as in 'allowed to' (much less 'encouraged to'). I mean 'may' as in 'sometimes decided to'. Hm, it's a good point. Language, so important. Valereee (talk) 11:23, 28 September 2024 (UTC)

I had the same thought when I saw you addition. The language in the sentence could make it seem that such lists have policy backing, rather than being a page consensus like FAQs. There's similar problems with "but only that content cited to other sources may require consensus for use of that source for that content.", as it could imply that editors need pre-approval to edit (rather than edits could be reverted and need discussion). -- LCU ActivelyDisinterested «@» °∆t° 14:29, 28 September 2024 (UTC)

I agree, and a list of "best sources" can be used as a cudgel to WP:OWN an article. Even if created with the best of intentions, it is easily subject to misinterpretation and abuse. I think the current language is fine and does not warrant an addition. Coretheapple (talk) 13:17, 2 October 2024 (UTC)

The current wording just says that they help prevent bias. This is pretty gentle for somebody to interpret as "can't use other sources". Also, a consensus for use would presumably mean picking bestsources by consensus. Finally, if somebody is just trying to use this clause to knock out certain viewpoints, wp:weight (flawed as it is) and other guidance weighs in towards including them. Plus a list idea sounds awfully prescriptive. Sincerely, North8000 (talk) 18:47, 27 September 2024 (UTC)

@North8000, at contentious topics, it's often helpful to maintain a list of sources that people with various viewpoints have agreed over time are among the best sources available. Such a list typically is created through a long process -- months and longer -- of discussion. It's to prevent every new addition being argued over because of objections to the source itself. Instead such objections are limited to only those that aren't yet on the list. :D So fewer such objections! :D Valereee (talk) 18:53, 27 September 2024 (UTC)

Hmm. At first, I assumed that this was another dispute over Wikipedia:WikiProject Video games/Sources, where the apparently level of discussion for many items was two editors agreeing with each other more than a decade ago, and their word is now holy writ for all articles, especially for their dis-recommendations.

Is the general idea here something like this?

• Editors (using their best judgement, etc.) agree that certain sources are appropriate and useful for a specific article (or a very small number of articles, e.g., COVID-19 and COVID-19 pandemic).
• For convenience, they post a list of what the sources are.
• If someone cites a source that isn't on the list of pre-approved sources, then reversions happen and complaints are lodged.

If so, then... I'm really not sure. On the one hand, if you're going to enforce that kind of agreement, then you probably ought to advertise it. On the other hand, pre-banning nearly every source in the world except the tiny fraction of sources already on the 'approved' list sounds like an extremely bad idea. What if a great source gets published tomorrow? Do you first need to have a discussion to put it on the approved list?

I think that broad categories would be more useful than individual books (e.g., don't recommend "21st edition of Harrison's Principles of Internal Medicine" specifically; instead, recommend "medical school textbooks", or, if you need to be more narrow, "medical books recommended by Doody's Core Titles during the last five years").

I also think that it would be more useful for editors to think about the sources that form the basis for the article vs those that are used for little bits. Using a Christmas tree as an analogy, you normally want to use a small number of high-quality sources to write the "tree" of an article, but you can use dozens of sources, from a wider variety of types and quality, for individual "decorations".

When I wrote Breast cancer awareness, I leaned heavily on one book (33% of inline citations) and strongly on three other sources (11%, 11%, and 9%). These four sources account for almost two thirds of the article, and they provide a pretty strong basis for the article. The addition of an "unapproved" source here or there, even if that source has a wildly different POV, is not going to change the overall article. WhatamIdoing (talk) 23:21, 27 September 2024 (UTC)

As a minor point, I really dislike the phrase "a BESTSOURCES list". Editors might choose to develop "a list of recommended sources" or even "a list of relevant high-quality sources", but I dislike giving it an WP:UPPERCASE name. That implies that this is an institutionalized thing, and possibly even a best practice, rather than just an unusual thing that some editors might voluntarily choose to do. WhatamIdoing (talk) 00:39, 28 September 2024 (UTC)

Agree! Andre🚐 01:03, 28 September 2024 (UTC)

No. It's the way to go. It puts emphasis on what the collective we, should be doing and it is a best practice, finding and conveying the best sources we can. Not best practice is using crap sources for any old garbage, and even worse just stumbling along with crap sources for any old garbage. It is also likely among best process in those cases (which is far from every article, indeed policy/guideline is really not called upon extensively to settle most articles) when process to overcome extensive disagreements is needed. --Alanscottwalker (talk) 09:52, 28 September 2024 (UTC)

Disagree. It can lead to unintentional gatekeeping or cherrypicking, especially if the criteria are arbitrarily defined. Weight is in all WP:RS and we are supposed to reflect all non-FRINGE minority POVs with proportionate weight. I agree with WAID that it's not a best practice and not an institutionalized thing, it's something that some editors are doing on particularly contentious articles, just like sometimes there are special consensus items or special FAQs. In general, it's not a scalable solution, and what people used to call anti-wiki. Andre🚐 10:09, 28 September 2024 (UTC)

That makes no sense. Your complaint then is writing every article is unintentional gate keeping and cherry picking. Alanscottwalker (talk) 10:12, 28 September 2024 (UTC)

I'm not sure what you mean. Don't get me wrong, I think it's a useful exercise. I think the output of a BESTSOURCES source survey could inform a lot of decisions about structure or weight for a lead. I like WAID's Christmas tree idea. But, I think that the policy should not prescribe the practice of going through assembling a BESTSOURCES list, for the reasons I explained. Andre🚐 10:15, 28 September 2024 (UTC)

The reason we have articles is to have discreet bodies of knowledge based on sourcing. Doing our best work in that is the goal. And how we are supposed to arrive at those articles is through discussion and agreement. (At any rate, there are crap filled Christmas trees and very fine Christmas trees, too). Alanscottwalker (talk) 10:23, 28 September 2024 (UTC)

I prefer not to call articles "crap," it's derogatory and it denigrates the good faith contributions of volunteers. I prefer to think of articles as living things that start out as embryos and eventually become fertilized and grow into little chickens that hatch. But a tree growing from a seed is good too. One of the main ways or reasons why Wikipedia works is because the process engenders aggregation. When disputes arise, there is discussion and negotiation, but that's essentially a blocking process. It's like instead of telling an ad hoc group, a task force or a tiger team, to go solve problems, you instead create a meeting with oversight and committees; WP:NOTBURO. Andre🚐 10:31, 28 September 2024 (UTC)

You're still not making sense, no one is told to go there, they are invited to discuss, and they don't have to if they don't want to. And it also makes no sense to say all articles must come into being in that way, good work readily finds agreement -- also, very fine, even great articles from the start (maybe they need a formatting fix here and there) are more than welcome - the more the merrier, in fact. We are literally for people making themselves knowledgeable about a subject and then writing. Finally, crap articles are not the responsibility of any one editor, they are the responsibility of the project. -- Alanscottwalker (talk) 10:46, 28 September 2024 (UTC)

Instead of saying I don't make sense, you can just say you don't agree with my theory of collaboration or my opinion on source selection process. Wikipedia works because of decentralization, good faith, a friendly environment for volunteers, and the fact that any task can be easily decomposed into many smaller tasks. The authoritative source list makes sense as a bulwark for disruption in some special cases, similar to ECP protection, but let's not advocate for it in the majority of cases or claim it is a desirable state. Such centralized, gated processes are a necessary compromise to keep the order and the trains running on time, but they are as I said what we used to call anti-wiki, meaning antithetical to the principles of collaboration that people are supposed to know and internalize. They lead to a system whereby a central process usurps the distributed nature of wikicollaboration. You may not agree but it's completely coherent as an ideology, continuing to say that I'm not making sense is uncharitable at best. Andre🚐 20:31, 28 September 2024 (UTC)

Re: I really dislike the phrase "a BESTSOURCES list". Hm, fair point, and really, it may be the basis of the confusion I'm seeing. Maybe this is something that should go into an essay? Maybe WP:Developing a list of sources that have gained consensus at an article that would cover both the value of creating such a list and the fact that the creation of such a list doesn't mean other sources can't be used, but that they'd need consensus for whatever content is being cited to them? And also that consensus for what items are included on the list can also change. Valereee (talk) 11:34, 28 September 2024 (UTC)

I think an essay would give you a lot more room to explain what it is, why it is, and how it works. I would recommend writing that page no matter what does/doesn't happen here. WhatamIdoing (talk) 21:39, 28 September 2024 (UTC)

I've created a crappy draft at WP:CONSOURCE. Please all feel free to edit. Valereee (talk) 22:42, 28 September 2024 (UTC)

It is an institutional thing. It is a best practice. Levivich (talk) 12:49, 28 September 2024 (UTC)

Is it really an institutional thing? Can you name 10 articles that have taken this approach, off the top of your head? I certainly can't. How many times have you seen people productively working on an article say "Hey, let's take a time out from writing and come up with a short list of pre-approved sources"? My answer is "never".

I suspect this exists only in a handful of articles about heavily disputed subjects, solely because it was voluntarily adopted by the regular editors of that article. By "heavily disputed", I mean way more disputed than Donald Trump, which presumably wouldn't be citing 800+ sources if restricting articles to a handful of pre-approved sources were truly an institutionalized thing. WhatamIdoing (talk) 21:38, 28 September 2024 (UTC)

I view the main problem with this kind of list (whatever you call it) as creating a potential WP:OWN issue in articles about subjects that have evoked strong feelings, such as popular culture figures with fan bases. One in particular comes to mind. A book came out on the subject of the article, arguably the most thoroughly researched book ever published on that person, which delved into the darker aspects of that person's persona. The book was immediately slammed by the editors on the page, in my view without sufficient basis. I happened upon the book some years after it was trashed on the talk page, and was surprised to see that it had been used as a source in only a very limited basis. While a "best sources" list was not created, because thankfully the proposed paragraph was not in the NPOV policy, such a list no doubt would have excluded that new book, which made the popular culture figure seem less heroic than was portrayed by other sources. Local consesnsus can at times be skewed by prejudice and POV issues. Consensus can change. Editors come and go. So what a consensus of editors (sometimes just three or four people) view as "best" at any given time should not be stated in semi-formal language on a talk page. Coretheapple (talk) 13:35, 2 October 2024 (UTC)

Anytime editors don't agree with a local consensus, an RFC can be bought to challenge it so that is not a valid objection imo. Selfstudier (talk) 13:54, 2 October 2024 (UTC)

Re: If someone cites a source that isn't on the list of pre-approved sources, then reversions happen and complaints are lodged. No, the general idea is to prevent this from happening by establishing that the simple fact there is a list of sources that have consensus does not mean other sources cannot be used, just that they may need to gain consensus for the content they're being used for. Valereee (talk) 13:19, 28 September 2024 (UTC)

I am involved in active dispute on this. I agree with North8000 and Peter. The policy doesn't prescribe the creation of lists of BESTSOURCES. Editors can still do that, but the policy does not prescribe this activity, and I have serious reservations about making that a formal part of NPOV, because it could accidentally and unintentionally encourage a blind spot in source selection or an exclusivity of some sources to the expense of others. Andre🚐 22:35, 27 September 2024 (UTC)

Well, of course, it makes sense to agree on a list through discussion. To actually take "read widely" seriously, then a reading list is just the thing. Evaluation of "new" sources will also have a template to later follow. You are going to have to have some agreement, and many interim agreements, regardless, because the reading list will also in general be in the article's sources. (maybe just drop the last clause). Alanscottwalker (talk) 23:41, 27 September 2024 (UTC)

I have to say that I really do not understand why anyone would object to a consensual effort to produce a list of the best sources, particularly for complex and controversial articles. The ideal would be to deal with issues of weight so that not so great but acceptable sources do not trump great sources by virtue of number, edit warring or anything else. Not saying it should be required as policy but why not encouraged as policy? Selfstudier (talk) 13:27, 1 October 2024 (UTC)

It's astounding. How do you collaborate with people who think that gathering the best sources is somehow a problem? It's step one, and if anyone thinks it's not step one, well, I got nothing. Levivich (talk) 14:51, 2 October 2024 (UTC)

Because the policy doesn't say "gather a list of the best sources." It says that you should use the best sources, which are the most reputable books and journal articles. It does not say that you should create a prescriptive list. Again, I think that can be a useful exercise, but the policy should not and does not say that inclusion and exclusion criteria should be so exacting. Andre🚐 20:26, 2 October 2024 (UTC)

Not only does this seem like hair splitting but the "prescriptive" and "exacting" bits are strawmen.Selfstudier (talk) 22:39, 2 October 2024 (UTC)

Wikipedia policies are intentionally vague and open-ended to account for the general nature of the project and the fact that many different things have different treatments. Andre🚐 22:55, 2 October 2024 (UTC)

I find Wikipedia policies to be fairly clear; the interpretation of those policies, however, is incredibly dynamic. For example, "use the best sources" means "you do not need to use the best sources, other sources are OK to use, too." Levivich (talk) 23:46, 2 October 2024 (UTC)

That's kind of what I mean. the wiggle room is there for interpretation. Wikipedia:The rules are principles. So talking about a specific activity or action should be less used unless it's very clear that this activity or action is prescribed, which making the list isn't, currently. The point is that there is more than one path through a process to a principle. Andre🚐 00:01, 3 October 2024 (UTC)

I am very against such WP:LOCALCONSENSUS on sourcing and would frankly add a line to WP:BESTSOURCES reading Which sources are the best sources is determined solely by the state of the field at a given time. Settling on a specified list of sources necessarily means excluding some of the best sources and is therefore discouraged. Loki (talk) 03:34, 3 October 2024 (UTC)

While I don't want to add this to the policy either necessarily, I do agree with you that this is kind of my opinion or the reason for my opposition to the opposite policy change. Andre🚐 03:38, 3 October 2024 (UTC)

Examples

I stumbled across this today and thought it might be helpful to provide examples of where the idea of a list of sources that editors at a topic or article have gained consensus are good has been used. Valereee (talk) 17:52, 28 September 2024 (UTC)

• Wikipedia:WikiProject UK Railways/Sources
• Wikipedia:WikiProject Video games/Sources (Masem (t) 18:19, 28 September 2024 (UTC))
• Wikipedia:WikiProject Football/Booklist and Wikipedia:WikiProject Football/Links
• WP:LDS/RS Levivich (talk) 18:44, 28 September 2024 (UTC)

  Aren't these project RS lists, I thought this was about approved source lists on article talk pages? -- LCU ActivelyDisinterested «@» °∆t° 19:51, 28 September 2024 (UTC)

  I wouldn't use the word "approved" but, yeah. This is a little self-centered, but I'm only really aware of the articles where BESTSOURCES lists have been compiled where I've been directly involved, so here are some examples I remember:

  • Talk:Zionism#Best sources
  • Talk:Nakba#Core sources
  • Talk:Alexandros Schinas/Archive 2#Sources
  • Talk:Murder of George Floyd/Archive 4#Source analysis table
  • Talk:The Holocaust/Archive 36#TOC Analysis

  The last two are basically BESTSOURCES lists although I didn't call them that. Levivich (talk) 23:22, 28 September 2024 (UTC)

  So, WAID requested a bar of 10 articles (Is it really an institutional thing? Can you name 10 articles that have taken this approach, off the top of your head?), which, if there were an institutional best practice, I would assume should be attainable, after all, 6.5m articles in enwiki, and 6500 featured articles, so if my numbers and math are at all correct, that's 0.10%, or one tenth of 1 percent featured articles, and you've given 5, which is .07% of that. One of them is not a good example as it's a current evolving situation, so hardly counts, the one prompting this attempt to change policy. If I do the numbers with 40,000+ good articles it's even more stark. At 50% of the incredibly skimpy bar, with that caveat that all of the examples involve the direct involvement of you. 10 articles is an exceedingly low number, I would say, if there are over 6000 featured articles, and over 40,000 good articles, which is a really impressive number, don't get me wrong, community, but also a staggeringly tiny number considering Wikipedia has existed for over 20 years.

  Maybe not all of those articles are at all disputed or controversial, but even if you say that there are only 1000 controversial articles on the level of a major political figure, ideology or a major crisis or atrocity, I would still think that if this were an institutional best practice and something employed at any reasonable percentage of article writing, we should be able to find well over 500 examples. The fact that we can't even get 5, or 1% of that, isn't a good sign that this is actually a common practice or something that the policy should specifically prescribe doing or encourage, or that there's a project-wide consensus that this is how we must do things. Now again, I do think sometimes it is worth doing. In the examples you gave, at least for 2 of them I can completely understand and endorse keeping out journal articles, even if published in a reputable journal with academic credentials, because not all material published by reliable academics isn't FRINGE when it comes to the Holocaust. And I can imagine a similar scenario about reliable current events and news stories for the George Floyd thing. But for a 20th century political ideology or event, where there are multiple reputable journals in the topic area, defining a source list as only books is arbitrary and capricious. And you can say I shouldn't bring that up here because this is WT:NPOV not Talk:Zionism, but you just listed that as one of your examples, of 5. Andre🚐 05:53, 29 September 2024 (UTC)

  I don't think talk pages discussing the reliability of sources is a problem, but they shouldn't be used to stop other editors from editing. Edits might be reverted because they need discussion, but should never be reverted because they don't come from a pre-approved list of sources. That is just pre-approval of edits with an additional step and goes against being bold.

  I think part of the problem here is language, using 'BESTSOURCES list' implies that there can be no others, but that's not what WP:BESTSOURCES is about. -- LCU ActivelyDisinterested «@» °∆t° 10:20, 29 September 2024 (UTC)

  I don't think using 'BESTSOURCES list' implies that there can be no others, and have seen no evidence that it's ever taken that way. For example, at Talk:Zionism, literally not one editor thinks that or has suggested it. At none of the examples I've posted has even one editor suggested it. I've never seen anyone suggest that or interpret it that way. Levivich (talk) 13:17, 29 September 2024 (UTC)

  Thanks for posting the list of five articles that have made lists of sources. There are non-list approaches (e.g., the one-off message on the talk page), and there's also what you might call the 'WORSTSOURCES lists' (e.g., WT:COVID-19 still has a note about Worldometer being disrecommended, and WP:VG disrecommends a long list of websites).

  This makes me wonder whether the point here has little to do with "making a list of the best sources" and instead is more like "sometimes we already have a consensus about whether _____ is an acceptable source for this article, and sometimes we even write that down for your convenience, so you can quit asking already thankyouverymuch". WhatamIdoing (talk) 20:56, 30 September 2024 (UTC)

  +1. That's how I was reading the policy. Valereee (talk) 13:33, 1 October 2024 (UTC)

These are not at all the same thing. It's long been the case that separate topic areas and wikiprojects developed their own mini-RSP, that's not the same as developing a list on a specific article. The reason why video games developed their own source list for reliability is because they use things like gaming magazines or websites like IGN which probably wouldn't be reliable for anything else, so they're developing a list of sources that is typically less reliable by normal standards, but still usable for that topic area. That hasn't been the case for history or politics, which have more stringent standards than video games. If the WikiProject Jewish history or Judaism developed a recommended source list for Zionism, one would assume that it would include a number of Zionist sources, wouldn't that be a fair assumption? Probably a bunch of Israeli Zionist sources in Hebrew. It doesn't make sense to say that selecting a group of authoritative history books is the same thing as having a list of gaming industry publications that are or aren't reliable, it's a huge stretch. Andre🚐 19:50, 28 September 2024 (UTC)

I've seen Paul Siebert taking a similar approach, tho not always creating an explicit "list". As i recall there was even a paper written about one of the articles involved? It is nice to see this section get some attention and hopefully the useless "try the library", "look online", "ask at the reference desk" can be replaced with some real guidance on what "best sources" means and how to find them. fiveby(zero) 13:16, 29 September 2024 (UTC)

Drafting

Here's the current contents of this section, numbered for convenience:

1. In principle, all articles should be based on reliable, independent, published sources with a reputation for fact-checking and accuracy.
2. When writing about a topic, basing content on the best respected and most authoritative reliable sources helps to prevent bias, undue weight, and other NPOV disagreements.
3. Try the library for reputable books and journal articles, and look online for the most reliable resources.
4. If you need help finding high-quality sources, ask other editors on the talk page of the article you are working on, or ask at the reference desk.

It appears that the goal is to add something like "If editors have formed a consensus about the suitability of certain sources (pro or con), it would be collegial if they shared that somehow instead of just yelling at editors who are trying to help for not magically knowing about prior discussions." Also, maybe be careful about making lists, because they create risks for bias and ownership.

I think this new goal could fit between sentences 2 and 3.

Looking at it more generally, sentences 3 and 4 are maybe not providing a lot of value. Would it be better to point at a page like Wikipedia:Find your source (or something better than that)?

Sentence 2 might benefit from being broadened to explicitly mention sources from varying POVs (e.g., World War II shouldn't be written entirely from sources from just one of the belligerents).

One of the missing things is a statement about what The Best™ Sources are. I don't know how to write a sentence that says "The best sources are...", because it varies so significantly by subject. The best type of source for a current event is not the best type of source for a historical event.

What else could we improve in this paragraph? WhatamIdoing (talk) 04:39, 3 October 2024 (UTC)

I find all of the above insightful and largely agree. I think the spirit of sentence 2 and 3 is basically - use better sources if they are available, reliable books and journals are better than less reliable books and journals and websites and news links. I don't like "best" in general as a superlative as in best practice, we just mean that better sources are better. I agree that something about WP:CONTEXTMATTERS is part of it. I also definitely agree that balance and weight are part of this. This reminds of a point you once made - WP:DUE really only talks about viewpoints. Maybe this is a place to remind people about WEIGHT and NPOV being something that theoretically produces an WP:IMPARTIAL tone. Andre🚐 04:47, 3 October 2024 (UTC)

I think that if we are going to add a line like this somewhere, it doesn't go in WP:BESTSOURCES. WP:BESTSOURCES is about preferring the best sources (and not merely treating any reliable source as equal). Saying "if you make a list of sources you should publish it" doesn't fit here. (And I'm not just saying that because I think that such list making is categorically against the policy to begin with. :P ) Loki (talk) 17:37, 3 October 2024 (UTC)

I think WhatamIdoing's observation re Sentence 2 is especially important and valuable. Coretheapple (talk) 18:16, 3 October 2024 (UTC)

I appreciate this discussion and WAID's point about Sentence 2. NPOV is about representing the range of viewpoints found in reliable sources and helping the reader understand why issues are controversial. Our experience in many topic areas is that using higher-quality sources helps us meet these goals because the x best sources represent the body of reliable sources on the topic. However in some topic areas this might not be the case, and restricting source usage could have the effect of excluding some points of view. Clayoquot (talk | contribs) 14:14, 6 October 2024 (UTC)

For a subject like WWII, we could find high-quality scholarly sources for all major POVs (e.g., reputable books from academic publishers in each country). For a subject like Anti-psychiatry (in which the people with institutionally recognized power have a different POV than people without that power) that may be less true. In that case, even an exactly matching rule could have unequal effects. WhatamIdoing (talk) 01:38, 11 October 2024 (UTC)

Clarification for known issues or criticism sections

It seems like known issues or criticism sections are not allowed in Wikipedia acording to @InfiniteNexus see Talk:Pixel_9#Known issues section and neutrality. I am deply worried on this agresion on the neutrality of Wikipedia. Many articles have similar sections and discussions in the past settled these sections as valid. See Talk:Pixel 5#Known issues section blanking Could someone clarify? 90.167.218.96 (talk) 14:57, 29 September 2024 (UTC)

This user is beating a WP:DEADHORSE. It has already been explained to them that a section for "known issues" is (1) unencyclopedic and contravenes WP:NOTCHANGELOG, being an indiscriminate list of trivial matters that belong on a help center or issue tracking system, not Wikipedia; (2) non-neutral, as it directly goes against WP:CSECTION and WP:TRIVIA by having a section dedicated to non-notable software bugs, which also has issues with WP:UNDUE; and (3) unnecessary, as any major controversies can and will be integrated in existing sections, as it is currently being done. I'll once again remind the user that (1) two wrongs don't make a right; (2) consensus is not determined by the number of raw votes but by the strength of the arguments presented; and (3) they should stop going around in circles by simply repeating their arguments. InfiniteNexus (talk) 16:54, 29 September 2024 (UTC)

A more neutral framing is "Reception", which makes room for the full range of opinions. Calling the section "criticism" is non-neutral because the heading naturally excludes positive comments. Yes, you can find articles from the past that fail WP:NPOV. But they should be tagged and fixed, not used as examples of why we ignore WP:NPOV. Shooterwalker (talk) 18:54, 29 September 2024 (UTC)

OK, thanks. I will move the issues under the Reception section. 84.78.243.26 (talk) 06:25, 30 September 2024 (UTC)

84.78, you need to do three things here:

1. Use neutral language, like "Reception". "Reception", by the way, implies that it's about what relevant experts think.
2. Don't put any criticism in a separate section if it could be reasonably placed in an existing one. It's better to have something like "It has 12345 GB, which has been criticized as too little storage" than to have "It has 12345 GB" in one section and "It has been criticized for having too little storage" in another.
3. Don't include every single known or suspected problem. Only include the problems that multiple independent reliable sources believe are significant. For example, if multiple computer magazines say "This device only has a foo, and it really ought to have a baz", then that's probably fine, but you probably shouldn't include anything that can only be sourced to a single source. If it's a significant problem, then you should be able to find multiple reports. If you can only find one website (especially if it's largely driven by sales commissions, customer reviews, or social media), then you shouldn't include it.

WhatamIdoing (talk) 21:07, 30 September 2024 (UTC)

What the others said. Plus such section headings tend to towards including things that would otherwise not merit inclusion.North8000 (talk) 21:28, 30 September 2024 (UTC)

(edit conflict)#3 is the key point. Not every software bug is notable — in fact, most of them aren't. Only if they have attracted widespread coverage from reliable sources, of if they have special significance/relevance, do they warrant a mention. InfiniteNexus (talk) 21:28, 30 September 2024 (UTC)

The later part of this thread focused on neutrality issues but another might be the primary one here. The guidelines / policies that others are referring to (with WP:Not at the core of them) are in essence emphasizing that we are an enclyclopedia covering topics in enclyclopedia-type articles. So this is not "all information" and so you might be seeking to include information that is not enclyclopedia article type information. North8000 (talk) 21:36, 30 September 2024 (UTC)

Do you mean that notable issues with a product should not be allowed in wikipedia and thus be removed? Please, calrify and I will start removing any notable issue from any product page in Wikipedia. 80.103.136.237 (talk) 13:17, 1 October 2024 (UTC)

No, that is not at all what North meant… notable (or perhaps noteworthy) issues with a product (ie those widely reported on) should be mentioned. However, we don’t mention EVERY issue with a product. Trivial issues can be (and usually are) omitted. An examination of the sources, discussion and consensus determines whether a specific issue is noteworthy or trivial. Blueboar (talk) 13:41, 1 October 2024 (UTC)

OK, thanks. All the issues that @InfiniteNexus is trying to hide are notable (accoding to Wikipedia notability definition). 90.167.219.84 (talk) 15:55, 1 October 2024 (UTC)

... No they are not. Please substantiate your arguments rather than simply continue to make claims like this. What "definition" are you referring to, and how so? I also don't appreciate your continual bad-faith assertions that I am making Wikipedia non-neutral, trying to hide information, or seem like a Google employee. InfiniteNexus (talk) 16:04, 1 October 2024 (UTC)

I assume that 90.167 means WP:Due weight instead of Wikipedia:Notability. Notability is Wikipedia's jargon for whether a subject deserves at WP:Separate, stand-alone article, and I don't think they mean to say that each of these consumer complaints deserves a completely separate article. WhatamIdoing (talk) 16:48, 1 October 2024 (UTC)

Good point… DUE vs UNDUE (which is determined by coverage in sources) is the criteria here, not Notability. Blueboar (talk) 16:58, 1 October 2024 (UTC)

The page Wikipedia:Notability does not apply to content within articles, but "notable" is often used generically as a synonym of "noteworthy", i.e. worthy of inclusion on a Wikipedia article. It's important for the IP user to recognize that verifiability does not guarantee inclusion, so just because they can find a source for something doesn't mean it is not trivial and interests a general audience. This is discussed at WP:INDISCRIMINATE. InfiniteNexus (talk) 17:40, 1 October 2024 (UTC)

These issues are supported by multiple sources. So, they are considered WP:Due weight. I think Wikipedia needs to clarify if these are allowed or not as there are many similar articles. 80.103.137.123 (talk) 20:00, 1 October 2024 (UTC)

The first issue was added like this:

• Some owners are reporting a camera tilt issue. Google has not made any comments yet.

The source added for this issue was:

• Some Pixel 9 Pro XL owners are reporting a camera tilt issue

That's one (1) issue with one (1) source. One source ≠ multiple sources. Each complaint needs coverage in multiple sources. How else are we to know whether that's a widespread problem, or just something that a couple of people complained about? WhatamIdoing (talk) 21:54, 1 October 2024 (UTC)

Sure, I will add more sources. 90.167.218.158 (talk) 07:00, 2 October 2024 (UTC)

Either the IP user is intentionally being disruptive or unable to understand the relevant policies and guidelines, as they have once again restored the list of known issues despite being told multiple times that it violates WP:NOT and WP:CSECTION. They have also ignored repeated requests not to edit-war and wait for consensus to fully develop before altering the status quo, so I can no longer assume good faith.

The "sources" that the IP has just added are not reliable (WP:RS). Virtually all of them are newsblogs, and at most two can be considered marginally reliable. The absence of significant coverage from reputable sources is an indicator that these software bugs are not noteworthy for inclusion, and the IP user should WP:DROPTHESTICK at this point. InfiniteNexus (talk) 07:14, 2 October 2024 (UTC)

All of the sources in the page are either from other weblogs or the primary source. Sould we remove the whole article altogether for lack of reliable sources? 90.167.203.206 (talk) 07:37, 2 October 2024 (UTC)

That's not true. The article cites the Associated Press, The Verge, Wired, The New York Times, Bloomberg News, and more. If you're not sure what constitutes a reliable source, please consult WP:RS and see WP:RSPS for a list of common sources. Your tone here is combative, and you seem to avoid addressing the issues at hand, namely, WP:NOT, WP:UNDUE, and WP:CSECTION. Editors have explained to you that an indiscriminate list of software bugs is unencyclopedic and does not belong on Wikipedia, and only if a particular issue has contextual significance and has received significant coverage from reliable sources can it be integrated into other sections in the article. If you are unable to understand this, or simply refuse to work with Wikipedia's policies or guidelines, then let's close this discussion and move on, because competence is required on Wikipedia, and you must be willing to adhere to Wikipedia's policies and guidelines and work with other editors. InfiniteNexus (talk) 16:03, 2 October 2024 (UTC)

@InfiniteNexus, I think this is what the Wikipedia:Dispute resolution processes are for. I doubt the IP has enough experience to know how it works, but they do seem to be trying to comply with all the rules. Would you be willing to show the IP how it's done, by starting a discussion on the talk page, and marching through the list, from Wikipedia:Third opinion all the way to RFC if necessary, addressing each individual item at a time? WhatamIdoing (talk) 16:33, 2 October 2024 (UTC)

Although I have tried to assume good faith, it no longer seems to me that the IP user is trying to follow our PAGs. They have been asked several times, "Please don't edit-war, please don't restore your edit until consensus is reached", yet they continue to do so. They have also been told (by at least four editors), "A standalone list section dedicated to software bugs is unencyclopedic, please don't add it", yet they continue to do so. This discussion has been a de facto 3O request, so I am not optimistic additional discussion will be any more productive. I'm also not sure what an RfC would look like — a question along the lines of "should a list of software bugs be included" would likely yield the response I summarized above: "no, but if a particular issue is noteworthy they can be discussed in other sections" ... which is what I've told the IP since the beginning! I also recognize there may be a language barrier (the IPs geolocate to Spain), but competence is still required and this is not an excuse for disruptive behavior. InfiniteNexus (talk) 17:17, 2 October 2024 (UTC)

We could stop the edit warring at Wikipedia:Requests for page protection, but I think the open question (i.e., the point upon which you and the IP differ) is whether the sources provided indicate that the material is DUE. An RFC question could look like "Shall we include <this sentence>, using <these sources>?" WhatamIdoing (talk) 17:26, 2 October 2024 (UTC)

I would say it seems pretty clear that newsblogs are generally unreliable, especially if they lack meaningful editorial oversight, have a poor track record, tend to publish anything for clicks, and are rarely cited by other reputable sources. I think to justify an exemption to WP:NOTCHANGELOG, there would need to be stronger sources like the ones listed at WP:RSPS — at the very least, it should be reputable enough to have its own Wikipedia article. But what are your thoughts? As I said, this discussion is essentially an informal 3O/RfC/DRN. InfiniteNexus (talk) 17:45, 2 October 2024 (UTC)

I don't know enough about the subject area to know what the usual sourcing standards are. For all I know, these could be highly reputable websites. Alternatively, they could all be AI-generated fakes. I don't have the background information to be able to tell them apart. I find it much easier to evaluate academic sources than news-ish websites. WhatamIdoing (talk) 20:57, 2 October 2024 (UTC)

@InfiniteNexus These news outlets are being used for references in many Wikipedia articles if you want to challenge them then you should challenge them Wikipedia wide. 85.48.187.242 (talk) 09:24, 3 October 2024 (UTC)

I'm starting up a new observation, I have an interest in a particular article, see my log. I'm seeking an analogy to the exception about hearsay evidence - that you can in fact establish that somebody said something - a speech act - while refusing to say that in any sense the truth of what they say is established. So, rather than focusing on the truth of a claim, whether that claim has been stated by a reputable source, that it could be established that a given claim is "out there", which to my way of seeking would involve a lower threshold to establish that - as compared to the truth of what is being claimed. If the current set of wikipedia ideas do not allow for this, I suggest that be a modification. If that is not possible, well, *groan* JohnAugust (talk) 20:42, 15 October 2024 (UTC)

So, there are two policy issues related to this:

1- verifying that the person said it, and that Wikipedia is quoting them accurately. This is covered in WP:V and WP:RS.

2- establishing that Wikipedia should mention the quote in the first place. This is covered at WP:NPOV, and specifically by WP:UNDUE.

Essentially, we want to cite the original (primary) source for Verifiability purposes, but want to cite Secondary sources for DUE Weight purposes. Thus, best practice is to cite both. Blueboar (talk) 21:41, 15 October 2024 (UTC)

The challenge is that saying "So-and-so said ____" implies that ____ is not only true but also relevant/important.

In some areas, we can move to a higher class of sources (e.g., replace social media posts about politician's hairstyles with scholarly sources that look at the same politician's policy stances). In other areas, that's not possible.

Editors will always have to use their judgment to determine whether a given point actually belongs in an encyclopedia article. WhatamIdoing (talk) 00:23, 16 October 2024 (UTC)

OK, thanks for that. I'll see how I go applying this in the next few days.JohnAugust (talk) 00:32, 17 October 2024 (UTC)

Well sourced, journalist written, mainstream corporate media based critical contents is nowhere near as problematic as Awards & Accolades section citing the award granting group, other .org, trade groups. Graywalls (talk) 19:52, 1 November 2024 (UTC)

I agree in theory, and I think this particular example hinges on whether you recognize these websites as "journalist written, mainstream corporate media" vs just some websites. WhatamIdoing (talk) 22:29, 1 November 2024 (UTC)

Rearrange again

This paragraph:

Neutrality assigns weight to viewpoints in proportion to their prominence in reliable sources. However, when reputable sources contradict one another and are relatively equal in prominence, describe both points of view and work for balance. This involves describing the opposing views clearly, drawing on secondary or tertiary sources that describe the disagreement from a disinterested viewpoint.

is currently under ===Balance===. Since it is about viewpoints rather than facts/content more generally, and since it actually links to WP:DUE, I think it should be moved up into the DUE section. Any objections? WhatamIdoing (talk) 23:32, 23 October 2024 (UTC)

"Balance" and "due" are covering a lot of similar ideas. I can see why we may need to shuffle them together. Shooterwalker (talk) 17:35, 24 October 2024 (UTC)

I've been thinking about separating the ideas: DUE for 'viewpoints' ("Deontologists say that the means don't justify the ends, but consequentialists say that they can") and BALANCE for 'facts' (e.g., biographies indicate when and where the person lived, even if sources don't go into great detail about this). However, there is quite a lot that applies to both. WhatamIdoing (talk) 20:12, 24 October 2024 (UTC)

the -est, "one of the first", in xxx superlatives in articles

How do we feel about the liberal use of such sensationalistic superlatives in a lot of articles? I am seeing them everywhere. "the biggest, the first, was featured in best/top 4,523 list of xxx" in the township/Bay Area/region/state/country/time zone. and so on. Even if it's mentioned in reliable sources, I'm seeing this used excessively. Graywalls (talk)

• This is done to establish notability. Blueboar (talk) 22:37, 1 November 2024 (UTC)
• Is it accurate? If so, then I don't think it's a problem, especially if it's just one or two claims per article. I'd rather read "First pizza joint in the Bay Area" than "It's a pizza joint in the Bay Area". It's not sensationalism to report 'favorable' facts.
  Also, Blueboar is correct. Our notion of a Wikipedia:Credible claim of significance pretty much demands that editors add some sort of information along these lines, so we can't complain too much when they do what we insist they do. WhatamIdoing (talk) 22:38, 1 November 2024 (UTC)

I think the issue is the overly liberal use of such superlatives. I do not think that any reasonable editor would object to describing Yuri Gagarin as the first human to journey into outer space. The quality of the sourcing is also important. If the preponderance of reliable sources describe the topic as "the first", then I have no problem including it. But sometimes these claims are sourced to low quality, lazy listicle journalism. As in so many other areas of editing Wikipedia, a healthy dose of common sense and good editorial judgment is required. Cullen328 (talk) 22:51, 1 November 2024 (UTC)

And also, "Awards and Accolades" section being considerably larger than criticism even if there's just as much critical information based on news reports. Graywalls (talk) 22:56, 1 November 2024 (UTC)

Well, listicle formatting may give a false impression there (a list of three awards takes up more screen space than a prose sentence naming the same three awards), but the problem with unbalanced attention isn't really about the superlatives. Someone can be widely recognized as the best guitarist and a person with serious personal problems; a business can produce the finest widgets and the biggest polluter of the town's water supply. And because of the effects of Outrage journalism on WP:RECENTISM, it is sometimes the case that what looks big at the moment turns out to be a temporary blip. It can be difficult to get the balance right, even when you're trying hard. WhatamIdoing (talk) 23:38, 1 November 2024 (UTC)

It's certainly noteworthy to say that they're the worst town polluter though, along with being the finest maker of widget Y. Graywalls (talk) 00:13, 2 November 2024 (UTC)

"was the first restaurant in township to phase out single use flatware, and has the largest solar generation among all sit-down restaurants in the county" sort of thing is what I was referring to. Graywalls (talk) 00:17, 2 November 2024 (UTC)

Assuming that they got some attention from (e.g.) the local newspaper about this, what's wrong with that? That might be what they're actually notable for. Maybe the next paragraph (or next year's version of the article) is going to say that the owners campaigned for a local plastic straw ban, or that their solar array caught fire and burned down the entire historic district. If that's what the sources give attention to that restaurant for, then deciding that this is unimportant is what this policy calls "editorial bias".

In some cases, what's important is that it happens at all. In such a case, the article might say less about "first" and "largest", and instead say something like "The owners are interested in environmental issues and have consequently stopped providing single-use flatware and installed a solar power system". But you really would have to consider everything the source says before deciding what to write. "They're eco-conscious" isn't a good explanation if the facts are that the solar power company was having a contest, and the super-competitive owner was determined to beat his arch-rival, Other Restaurant. WhatamIdoing (talk) 00:52, 2 November 2024 (UTC)

But in grand scheme of thing, those insignificant claims like the biggest snail ever seen (on my property) usually don't belong in an encyclopedia. It can be in editorial gray zone, but those "first in township" like claims are clearly comparable to that watermelon example.

Although I'm not exactly known as being an inclusionist while some others are extreme inclusionist. Graywalls (talk) 04:27, 2 November 2024 (UTC)

If independent sources make a big deal out of _____, then it doesn't matter if _____ seems insignificant to a Wikipedia editor. Instead of imposing our personal biases on the sources (why did they waste space on that unimportant cruft?!), we should reflect the sources as best we can.

Sometimes there are ways to represent the facts in a less-enthusiastic tone. "The biggest, oldest, and best restaurant in Smallville" may become "the only restaurant in Smallville". But ultimately, if the main reason the sources are writing about the restaurant is because they have solar panels on the roof, or a record-setting ball of twine in the garden, or because George Washington Slept Here, the Wikipedia articles really do need to reflect that fact.

Sometimes it's not actually obvious why some "trivia" might be more important than it looks at first blush. For example, "24-bed hospital with the only emergency room within 25 miles of town" probably sounds like pure promotionalism to some editors – but they wouldn't think that if they knew what effect that particular combination of size and distance has on unlocking US federal funding. There could be a hidden importance behind other kinds of "unimportant" facts, too. WhatamIdoing (talk) 07:33, 2 November 2024 (UTC)

I tend to agree if the attention is brought upon by the independent source in an intellectually independent way.

I'll use real example.. " LA Weekly listed it as one of the ten best "Online Resources for Metal Knowledge" in 2013." in Metal Injection, I find questionable. It's just in a list, but it was an editor decision to bring up that it's in a list. Graywalls (talk) 14:51, 2 November 2024 (UTC)