Wikipedia talk:Verifiability

Self-published sources for statements that ascribe the information to the publisher

In light of the dispute at Talk:Microsoft Windows#Request for comment on reliability of a self-published work as a source of info about themselves, I think it should be added to the policy, explicitly, that any self-published source can be used as a source of information for any statement that ascribe the information to the publisher, and that conditions 1-3 at Wikipedia:Reliable sources#Self-published and questionable sources as sources on themselves need not be met for the ascribed information. (e.g. "According to X, whatever is claimed on X's official website") This is essentially a special case of self-published sources of information about themselves, as it is about the views expressed by themselves, and nothing more. This will prevent confusion and similar disputes. Sovmeeya (talk) 19:15, 15 September 2024 (UTC)

I disagree with this proposal as it violates WP:WEIGHT and opens the door to pushing into articles any claim made by a company about its competitors. Schazjmd (talk) 19:28, 15 September 2024 (UTC)

I don't think there's confusion with that. They should not be able to make claims about third parties, especially when those comments are obviously self serving. -- LCU ActivelyDisinterested «@» °∆t° 00:09, 16 September 2024 (UTC)

Oppose ActivelyDisinterested, that is exactly what the proposal means. The OP does not say that statements cannot be self-serving or concern other entities, and in fact, explicitly says that conditions 1 (not unduly self-serving or exceptional claims) and 2 (not involving claims about third parties) of WP:SELFSOURCE do not have to be followed as long as we ascribe the claim to the to the source. And losing condition 3 (not involve claims about events not directly related to the subject) opens us up to even more problematic claims. Meters (talk) 09:17, 16 September 2024 (UTC)

And I find this whole proposal to be very premature. The OP raised an RFC about specific content on Microsoft Windows, and is attempting to implement a general solution here to get their way (and more) even before the RFC has closed on the specific article issue raised. Note that as yet there has been no support for the OP's position in the RFC. Meters (talk) 09:26, 16 September 2024 (UTC)

I read the discussion at Microsoft Windows that's way I said they shouldn't be able to do that. I also completely agree that trying to change policy to win a content dispute is a bad idea. -- LCU ActivelyDisinterested «@» °∆t° 09:49, 16 September 2024 (UTC)

Oppose Let me tell you about myself, starting with what I think about you, is not ABOUTSELF. It fails at least one of either 2. It does not involve claims about third parties; or 3. It does not involve claims about events not directly related to the source;. Unadulterated sophistry. - Rotary Engine ^talk 09:58, 16 September 2024 (UTC)

The policy permits my proposal as it is now. The purpose of the proposal is merely to say explicitly that this use of a self-published source is permitted. To prevent confusion.

It's not true that the such use necessarily violates WP:WEIGHT. Schazjmd's arguments has nothing to do with the questions of Verifiability and Source Reliability, and does not belong here. They are fine example of exactly why my proposal is necessary.

When we ascribe something to X, the source is used for the statement as a WHOLE - including the "According to X," part, so it's obviously a reliable statement that complies with the policy AS IT IS NOW. I understand your concerns, such statement might not be appropriate for some OTHER reason, but this should be judged on a case by case basis. One thing for sure, though - a self-published source for such statement will ALWAYS BE RELIABLE.

The purpose of Wikipedia:Verifiability is to assure "people using the encyclopedia can check that the information comes from a reliable source." Nothing more. You can't get any more reliable for someone's expressed views than their official website.

The matter has been previously discussed at the Teahouse, where I presented the question and two editors agreed with me.

I've withdrawn the RfC at Talk:Microsoft Windows#Request for comment on reliability of a self-published work as a source of info about themselves, and we'll continue discussing the matter here. Sovmeeya (talk) 19:15, 16 September 2024 (UTC)

@Sovmeeya, the part you're ignoring is that WP:ABOUTSELF means "about themselves". It does not mean what they say about any other entity other than themselves. When Digital Confidence says something about any other company, it's no longer about themselves. Schazjmd (talk) 19:21, 16 September 2024 (UTC)

I'm not ignoring that. Any statement that begins with "According to X," is a 100% statement about X, not about any third parties, regardless of how this statement continues. It's entirely "about themselves". It's not certain that their expressed views are correct, but it's certainly certain that these are their expressed views.

It might be X that makes a statement about a third party, but the WHOLE statement only mentions it second hand, without endorsing it, and in compliance with Wikipedia sourcing policies. Sovmeeya (talk) 19:31, 16 September 2024 (UTC)

It is them stating their opinion about a third party. Adding attribution does not change that. Schazjmd (talk) 19:34, 16 September 2024 (UTC)

Adding attribution does not change that, but it does make the statement verifiable and the source reliable. (for the statement as a WHOLE) Verifiability and Source Reliability is what this policy is all about. Sovmeeya (talk) 19:44, 16 September 2024 (UTC)

It does not make the source reliable. WP:ABOUTSELF only permits statements about themselves, not what they say about anyone else. Digital Confidence's self-published content about Digital Confidence is about them; anything Digital Confidence self-publishes about any other entity is not about Digital Confidence. Schazjmd (talk) 19:53, 16 September 2024 (UTC)

The policy as it's written now is confusing. That's why I've made this proposal.

Forget about the policy for a second, and tell me this:

If it has been established that a certain website is the official website of X, and there is no reasonable doubt to its authenticity, (it doesn't appears to have been hacked). In that website, it's said something.

Consider this statement: "According to X, something".

1. Can you verify that X has expressed something? (by checking if something is said on the website)
2. Do you have any doubt for the truthfulness of the statement that "X has expressed something", if you find that something is indeed said in the website?

If the answer to the first question is yes, and to the second is no, then the statement is verifiable and the source is reliable. Sovmeeya (talk) 20:26, 16 September 2024 (UTC)

No that is not the case, which is why there is additional guidance about self-published sources. -- LCU ActivelyDisinterested «@» °∆t° 20:54, 16 September 2024 (UTC)

The context you gave at the Teahouse and the content you wanted to add with the RFC don't match. At the Teahouse you asked about According to Digital Confidence, the built-in metadata stripper is flawed (although I would argue form limited being more neutral), which the source might be reliable for but could still be undue. While at the RFC was for According to Digital Confidence, the Remove Properties and Personal Information feature has a very limited support of file formats and metadata elements, and has a misleading user interface. These are not the same. Context is critical and how much you can rely on a self published source is important.

Simply adding "According to" to the beginning of a sentence isn't some kind of magic that allows any content to be added from a self-published source. -- LCU ActivelyDisinterested «@» °∆t° 20:52, 16 September 2024 (UTC)

The discussion at the Teahouse was limited to the question of Source Reliability of a self-published source about itself, as it should be here. I kept it short to save space. Adding the other details about the nature of the criticism makes no difference to the question of Source Reliability of a self-published source about itself.

And yes, adding "According to" to the beginning of a statement does make it a COMPLETELY different statement - one that is about the publisher, and is permitted by THIS policy. Sovmeeya (talk) 18:56, 17 September 2024 (UTC)

No it's doesn't, as everyone replying to this thread has made clear. -- LCU ActivelyDisinterested «@» °∆t° 01:46, 18 September 2024 (UTC)

Not everyone. Blueboar agrees with me. See his comment below. And that's on top of the two editors at the Teahouse. Sovmeeya (talk) 09:50, 22 September 2024 (UTC)

At the very most, Blueboar agreed with a far less dogmatic formulation of your premise. They do not agree at all with the conclusion you've drawn from said premise. Remsense ‥ 论 09:55, 22 September 2024 (UTC)

I'm sorry but your answers now seem to be a case of WP:IDHT. You may dislike the answer but multiple editors have expressed the same opinion. -- LCU ActivelyDisinterested «@» °∆t° 11:53, 22 September 2024 (UTC)

I wonder whether we're confusing WP:SPS with WP:ABOUTSELF. SPS does not have any restrictions like "claims about third parties".

In the instant case, we have:

• A market-dominating software system, and
• A possibly non-notable critic of some detail in one product.

When the critic says "I think a sub-sub-feature of this product is flawed in this very specific way", it might be SPS but it is not ABOUTSELF.

The question isn't whether someone can verify that the critic published that criticism; the only question of any importance is why anybody should care, for which see Wikipedia:Neutral point of view and Wikipedia:What Wikipedia is not. WhatamIdoing (talk) 03:09, 18 September 2024 (UTC)

The other critical question if it's self published would the points laid out in WP:SPS. -- LCU ActivelyDisinterested «@» °∆t° 16:28, 18 September 2024 (UTC)

Yes, and to be clear, those points are:

• "produced by an established subject-matter expert"
• "whose work in the relevant field has previously been published by reliable, independent publications" (bold in the original)

The points that sound like "not unduly self-serving" are in ABOUTSELF. WhatamIdoing (talk) 21:31, 18 September 2024 (UTC)

The claim here is that adding "According to" to the beginning of the sentence makes it an ABOUTSELF statement. -- LCU ActivelyDisinterested «@» °∆t° 21:44, 18 September 2024 (UTC)

I can see why someone might come to that conclusion, but that's not what we intend ABOUTSELF to cover. ABOUTSELF is for "Chris Celebrity said he got married today". ABOUTSELF is not for "Alice Expert said something about someone else". WhatamIdoing (talk) 22:07, 18 September 2024 (UTC)

No. You're ignoring the purpose of THIS policy - Wikipedia:Verifiability. As its title suggest, and as it's written in its lead section, its purpose is to assure "people using the encyclopedia can check that the information comes from a reliable source." Nothing more.

Now if this is the purpose of THIS policy, and if we have a different policy for Due and undue weight, why would THIS policy make a distinction between the two statements in your comment?! it would make no sense, as they are both verifiable the same! (provided that Chris Celebrity and Alice Expert published what it's claimed they've said in their respective official websites)

When we ask if a statement that begins with "According to" is verifiable or not, it makes absolutely no difference what follows the "According to", as long as it's indeed written in the self-published source. What follows could be true, false, short, long, about third parties, outragious. ANYTHING. Sovmeeya (talk) 10:18, 22 September 2024 (UTC)

Though this has been articulated to you several different ways so far, you do not seem to appreciate the central point that policies are written to communicate important norms to the community as needed. You have a very particular problem with what the policies explicitly cover, but it is just that: a problem that is particular to you. You were told what ABOUTSELF is intended to communicate, and you continue to reply as if we're discussing the logical completion of policy instead of the particular practical needs that policy is intended address. Remsense ‥ 论 10:39, 22 September 2024 (UTC)

Also, the statement "According to Digital Confidence, the built-in metadata stripper of Windows is flawed" does NOT involve claims about third parties, since the assertions are not presented as objective facts, but as subjective assertions that Digital Confidence claims. The criteria in Wikipedia:Verifiability#Self-published or questionable sources as sources on themselves does not prohibits it. It prohibits bundling information about themselves with information about third parties that are presented as objective facts. For example: "the built-in metadata stripper of Windows is flawed, as was found in an analysis by Digital Confidence, a company founded in 2009". Sovmeeya (talk) 10:53, 22 September 2024 (UTC)

Also, the statement "According to Digital Confidence, the built-in metadata stripper of Windows is flawed" does NOT involve claims about third parties. Respectfully, that statement DOES involve a claim about a third party - that "the built-in metadata stripper of Windows is flawed". Wikipedia:Verifiability#Self-published or questionable sources as sources on themselves DOES prohibit using self-published sources for such a claim; even if attributed. There is NO limitation in that policy to only claims that are presented as objective facts. Rotary Engine ^talk 23:35, 22 September 2024 (UTC)

There is no EXPLICIT limitation in that policy to only claims that are presented as objective facts. This is why the policy is confusing, and this is why I want to clarify it. But it's a trivial deduction that this is the intended policy based on the purpose of this policy - verifiability.

Determining verifiability of "According to X," statements is trivial: if what follows "According to X," is in X's website - it is verifiable, and X's website is a reliable source for such statement. The only relevant criteria is that there is no reasonable doubt as to the authenticity of X's website. (it hasn't been hacked) Sovmeeya (talk) 14:12, 23 September 2024 (UTC)

I don't see that it is particularly confusing. Most respondents in this discussion do not seem confused.

The purpose of the policy is indeed verifiability, but the policy is explicit that this is verifiability in reliable sources; and both verifiable and reliable are terms of art, with meanings distinct from the plain English.

The policy then goes on to define reliable sources; where it is explicit that, with certain limited exceptions, self published sources are not reliable (term of art). The exception for self published sources is: a) subject matter experts, b) information about the publishers themselves; with a defined set of rules for b), including that the information "not involve claims about third parties".

Those rules are intended to prevent exactly the type of end around that is being proposed here.

The source directly supporting the proposed content is a necessary but not sufficient condition for inclusion.

I also concur entirely with the comments by Blueboar below around NPOV & DUE weight. But for mine, they are a second hurdle, when the horse has already fallen at the first: reliability. Rotary Engine ^talk 00:12, 24 September 2024 (UTC)

I'll try it a little differently this time. Direct quotes of self-published work. Reminder - we're talking only about verifiability now, not DUE WEIGHT. Consider the following statements:

1. <START>"Company A states on its website the following: "A was founded in the year 2020""<END> (cite A's website)
2. <START>"Company A states on its website the following: "Our tests shows that product B by our competitor company C, totally sucks, and using it will kill you by electrocution""<END> (cite A's website)

To verify the 1st statement, you can check A's website and see if the quote appears there. A's website is a reliable source for the 1st statement.

To verify the 2nd statement, you can check A's website and see if the quote appears there. A's website is a reliable source for the 2nd statement.

Same thing! Anything else would be an absurd!

By contrast, the statement <START>"Product B by company C totally sucks, and using it will kill you by electrocution"<END> is not verifiable, and A's website is not a reliable source for it.

Don't you agree? Sovmeeya (talk) 10:58, 25 September 2024 (UTC)

The statement of your second company is unreliable for the product, so your or anyone else's observation of it, does not meet the standards of verifiability on the product. There are many unreliable things that can be observed, just not put on Wikipedia. This should not be this hard by now, what you think of the word, when you think of the word "verify" is irrelevant (although you actually seem to have an oddly cabined view: 'to verify; is regularly concerned with quality and proper use of evidence, not just any evidence). But, how a source is being used is relevant, is it being used for unreliable commentary. Alanscottwalker (talk) 11:21, 25 September 2024 (UTC)

I've edited my comment to clarify the scope of the statements.

The quality of evidence of a direct quote is the highest you can get. The statements only mean that what is within the quotes appears in the A's website. Nothing more. It does not mean that what's within the quotes is necessarily true.

These two statements are equivalent:

1. <START>"Company A states on its website the following: "Our tests shows that product B by our competitor company C, totally sucks, and using it will kill you by electrocution""<END>
2. <START>"Company A states on its website the following: "Our tests shows that product B by our competitor company C, totally sucks, and using it will kill you by electrocution", but what company A states might not be true."<END>

In the first statement it is implicitly that what's within the quotes might not be true. In the second statement it's explicit. This covers all possibilities, so the two statements, as a WHOLE, are necessarily true. Sovmeeya (talk) 12:07, 25 September 2024 (UTC)

Come on. You either are not listening, interested in philosophy and so in the wrong place, or simply have no ability to properly use information and sources in context. When you use what you want to use for unreliable commentary on the topic, it cannot be verifiable for the topic. No one cares what you think is true, the policy cares when someone is trying to shove unreliable commentary in a particular article. Alanscottwalker (talk) 14:40, 25 September 2024 (UTC)

No, I don't agree. Because, per WP:V, verifiability (Wikipedia term of art) isn't just you can check the source, it's you can check the source AND the source has certain characteristics. For the most part, as a simplification, those characteristics are a reputation for fact checking & accuracy. For self-published sources, those characteristics are detailed in WP:SELFPUB and WP:ABOUTSELF. Rotary Engine ^talk 14:57, 29 September 2024 (UTC)

Oppose Current policy already allows it under certain circumstances/conditions. There's no need to go beyond that with wording that would be used as categorically greenlighting it, overriding the current policy restrictions. North8000 (talk) 21:15, 16 September 2024 (UTC)

I agree with the contention that when we write: “According to person X, ‘Y an idiot’” (cite X’s website) we are making a statement about X’s opinion, and not a statement about Y. It is verifiable that X has stated this opinion.

However… Verifiability is NOT THE ONLY ISSUE here. We have to ask whether mentioning X’s opinion is appropriate or not. THAT is a function of DUE WEIGHT. It might be DUE to mention it in the article about X, but be UNDUE to mention in the article about Y.

(extreme example: Hitler’s views on Jews are verifiably sourced to Mein Kamph, but there will be a very very limited range of articles where his views would be appropriate to mention - even with attribution. Essentially, they would be DUE to mention in the article about Hitler himself, and definitely NOT in an article on Judaism).

To relate this to the debate under discussion: the question isn’t really about verifiability (whether we can reliably verify that Digital Confidence has an opinion), but how much WEIGHT to give that opinion. Blueboar (talk) 22:41, 16 September 2024 (UTC)

That's what I've been saying early on at Talk:Microsoft Windows. Another editor there didn't agree with me. He and most editors in this discussion, so far, think that the proposed statement violates Wikipedia:Verifiability. Evidently, the policy as it is written now is confusing! That's why I'm asking to EXPLICITLY make it crystal clear in the policy that such use of a self-published source NEVER violates Wikipedia:Verifiability. (although it might be inappropriate for other reasons) When we'll do that, we could move on to the question of DUE WEIGHT at Talk:Microsoft Windows. This will also prevent confusion, unnecessary disputes, and waste of time in the future, for similar statements. Sovmeeya (talk) 18:40, 17 September 2024 (UTC)

Thing is, this actually isn’t a question that comes up all that often, and amending policy to “clarify” how it should be applied in rare situations almost always causes unforeseen headaches.

So we are reluctant to amend policy without seeing a proposal for specific wording - and then giving a lot of thought as to how that proposed wording might be misused by Wikilawyers to cause even more debates than the status quo language causes.

Sometimes it is actually easier to occasionally have to explain “no, that’s not what this passage of policy means” than it is to amend the passage itself. Blueboar (talk) 19:03, 17 September 2024 (UTC)

We are making a statement about X's opinion, but that statement involves a claim about Y; which fails WP:ABOUTSELF #2. Rotary Engine ^talk 23:39, 22 September 2024 (UTC)

A LOT depends on context - is the article focused on X or is it focused on Y? Blueboar (talk) 11:31, 23 September 2024 (UTC)

Uh oh, you're getting confused now. Perhaps forgotten what this discussion all about by now. Context has nothing to do with the question of verifiability of "According to X," statements. Could be relevant to DUE WEIGHT, but not to verifiability. Wikipedia:Reliable sources also have a "Context matters" section, but it's not the kind of "context" relevant to the question of verifiability of "According to X," statements. Better stick to what you've initially written above.

Determining verifiability of "According to X," statements is trivial: if what follows "According to X," is in X's website - it is verifiable, and X's website is a reliable source for such statement. The only relevant criteria is that there is no reasonable doubt as to the authenticity of X's website. (it hasn't been hacked) Sovmeeya (talk) 14:07, 23 September 2024 (UTC)

Our policies and guidelines all work together… and thus need to be discussed together. The issue here is actually Reliability (which is an aspect of Verifiability)… most self published sources (ie those not published by acknowledged experts) are NOT considered reliable for claims about third parties… but ARE considered reliable for claims about themselves and their beliefs - IN articles about the person or people who hold those beliefs. This is why we can cite flat earth proponents in the article about Flat Earth… but NOT in the article on Earth. Blueboar (talk) 14:43, 23 September 2024 (UTC)

That's not what you initially said. Anyway, claims could be reliable or not, regardless of what articles you put them in. In what articles we should mention a particular statement, if any, is a question of DUE WEIGHT. DUE WEIGHT and Verifiability (including Source Reliability) are two different and completely independent questions, covered by two separate and independent policies.

When we consider a statement for inclusion, we should start by asking if it's verifiable, and if the answer is yes, continue to ask if it's DUE WEIGHT for a particular article.

For "According to X," statements, the statement will ALWAYS be verifiable if what follows "According to X," is in X's website. Period. From this point, we need to consider DUE WEIGHT. Sovmeeya (talk) 14:57, 23 September 2024 (UTC)

Yes, if you have a WP:Published source in which X says ____, then "According to X, ____" will always be verifiable. It won't always be verifiable-because-ABOUTSELF-subclause, but it will be ordinary-main-policy-verifiable (which, lest you misunderstand, is better).

But:

• See also Wikipedia:Verifiability#Verifiability does not guarantee inclusion, because being verifiable is Necessary but not sufficient.
• See also Wikipedia:Due weight, which one company's criticism of one detail in one subunit of a huge software system isn't.

Which is to say: The statement is verifiable, and you still don't get to put it in that article. WhatamIdoing (talk) 15:41, 23 September 2024 (UTC)

We're talking about a case where X's website is a self-published source. That is, the author and publisher are the same, as in "Business, charitable, and personal websites". That's as per "Examples of self-published sources" at Wikipedia:Identifying and using self-published works#Identifying self-published sources. So it's not "ordinary-main-policy-verifiable".

I don't need you to tell me that "Verifiability does not guarantee inclusion", I've said that over a month ago at Talk:Microsoft Windows#Privacy features addition reverted, as well as numerous times here, including in the very comment you were replying to.

When we'll finish clarifying here that "According to X," statements are ALWAYS verifiable, we could move on to discuss Due weight at Talk:Microsoft Windows. Sovmeeya (talk) 11:17, 25 September 2024 (UTC)

No. Not All Ways. you should know this by now. We measure all things in the context of articles. Alanscottwalker (talk) 15:44, 25 September 2024 (UTC)

The wording right now is a little ambiguous, but it works in most situations. As I have said during the RFC discussion, there are multiple considerations, including how elaborate a claim is and the scope of the article. Senorangel (talk) 02:58, 18 September 2024 (UTC)

I'll give another example to everyone here: Lets say, hypothetically, Tylor Swift tweeted on her official social media page the following: "I honestly believe that Earth is flat". Now consider the following statement: "Tylor Swift said that Earth is flat". It has the following traits:

1. It's self-published
2. The ascribed info (second handed) involves a third party (Earth)
3. The ascribed info (second handed) involves an exceptional claim (outrageous, contradicts proven solid scientific evidence, contradicts view held by all experts in the field, view held by a negligible minority)

Shouldn't we mention this on the Wikipedia article on Swift as well as the article Modern flat Earth beliefs? Of course we should!

Not only that the fact that the ascribed info is false does not makes the statement inappropriate for inclusion in Wikipedia, it's precisely the reason why it should be included! She is very famous, has many fans, and therefore influential. (we shouldn't be discussing here things about DUE WEIGHT, but I wanted to give an extreme case to make my point)

What does that tell you all? that what Wikipedia:Verifiability says is not what you think it says!

In particular, the conditions at Wikipedia:Reliable sources#Self-published and questionable sources as sources on themselves does not prohibits it. It prohibits bundling information about themselves with information about third parties that are presented as objective facts. For example: "the built-in metadata stripper of Windows is flawed, as was found in an analysis by Digital Confidence, a company founded in 2009". There is no EXPLICIT limitation in that policy to only claims that are presented as objective facts. This is why the policy is confusing, and this is why I want to clarify it. But it's a trivial deduction that this is the intended policy based on the purpose of this policy - verifiability.

Determining verifiability of "According to X," statements is trivial: if what follows "According to X," is in X's website - it is verifiable, and X's website is a reliable source for such statement. The only relevant criteria is that there is no reasonable doubt as to the authenticity of X's website. Sovmeeya (talk) 14:21, 23 September 2024 (UTC)

No. Earth is not a third party, it is not a party at all, it is a thing without agency. Alanscottwalker (talk) 14:24, 23 September 2024 (UTC)

However… if Swift says “Trump is a poo-poo head”, we can mention her self published opinion about Trump (properly attributed) in the article about Swift… but NOT in the article about Trump. Again… context matters. Blueboar (talk) 14:50, 23 September 2024 (UTC)

Blueboar, the hypothetical source the new editor has given in the example is a self-published social media post. That would violate WP:BLPSPS.

Compare a musician tweeting an opinion ABOUT music:

• Self-published source: "The rhythm in this song is challenging".
• Article content: Y "Mel Musician once described the rhythm as 'challenging'."

the same musician tweeting a fact ABOUT music:

• Self-published source: "That song is in the key of B♭".
• Article content: Y "Mel Musician said she plays the song in the key of B♭."

with the same musician tweeting ABOUT herSELF:

• Self-published source: "Today is my birthday!".
• Article content: Y "Mel Musician said her birthday is September 23rd."

and the same musician tweeting ABOUT someone else:

• Self-published source: "Today is my friend Chris Celebrity's birthday!".
• Article content: N "Mel Musician said Chris Celebrity's birthday is September 23rd."

Self-published sources cannot be used to support content that are ABOUT another living person, full stop. It does not matter if it's an opinion or fact. Such a source is unusable per WP:BLPSPS. WhatamIdoing (talk) 15:54, 23 September 2024 (UTC)

Yes. Thank you. But I'll add for clarity of the new editor, instead of "Article content" in the above chart, you should read that as "Potential article content", because as WhatamIdoing and others stress elsewhere: just having a source is not enough, it is a piece that goes into the multiple considerations. Alanscottwalker (talk) 10:33, 24 September 2024 (UTC)

An example of self-published source about a third party living person:

The ex-wife of the president of a large country has posted on her verified social network page that she has divorced the president because he used to regularly get drunk and beat her. No other evidence are available to support this claim.

Consider the following statement: "The ex-wife of the president has said that he used to regularly get drunk and beat her."

It has the following traits:

1. It's self-published
2. The ascribed info (second handed) involves a third party (the president)
3. The ascribed info (second handed) is about a living person
4. The ascribed info (second handed) involves an exceptional claim of objectionable nature
5. The ex-wife is non-notable for an article on herself on Wikipedia

Shouldn't we mention this on the Wikipedia article on the president? Sovmeeya (talk) 13:34, 25 September 2024 (UTC)

If the source is saying "she said", then it is not the wife's own statement. And the issue is whether the source (not the wife) is reliable for that and a bunch of other things. If you are saying "she said", just because you want to talk about her tweets, you are definitely not reliable, and her bare tweets are not reliable for anyone else, except herself. Alanscottwalker (talk) 15:19, 25 September 2024 (UTC)

To answer the question Shouldn't we mention this on the Wikipedia article on the president?: The community decided not to use such sources in this way. This is a self-imposed restriction.

In the model that all sources are reliable for something, the ex-wife's social media post is a reliable, non-independent, primary, self-published source for what she posted. You could not use this to make any claim that "it's true" in any form: not that it's true that he got drunk, not that it's true that this was her actual reason for divorce, merely that it's true that she said this. (Even then, we'd probably want some reason to believe that her social media account hadn't been hacked.)

Now: Is this verifiable? Yes. A reader "can check" that this was actually posted in her account, by going to the post and reading it. All sources/documents are considered reliable for claims about the exact words that are in that document. Nobody will click the link to her social media post, read the words "I divorced him because he used to get drunk and beat me", and say "What's going on? This post is just her saying that tomorrow is her birthday". They're all going to say "Yes, I guess she did post that."

But: This is a disallowed source. It is nominally verifiable but still not acceptable under our policies. Because using that post to talk about her ex-husband is banned under our policies, it doesn't particularly matter whether it's verifiable. You still can't use it. WhatamIdoing (talk) 18:06, 25 September 2024 (UTC)

What you've written in paragraphs 2 and 3 are absolutely true. That's exactly what I've been saying all along.

You claim that such statement is still disallowed by this policy, despite being "nominally verifiable". That's an exceptional claim! self-published sources are weak, and so the limitations imposed by this policy are very justified and make a lot of sense for the content of self-published sources, when it's presented as objective facts! But they are pointless and not justified at all for statements that ascribe info to the source.

On top of that, a policy titled "Verifiability" has no place for limitations that has nothing to do with verifiability. Just like we have a separate policy for WP:UNDUE, we can have a separate policy for additional limitations, if in fact the community wishes to impose them. This could be titled Wikipedia:Limitations for statements that ascribe info to the source in self-published sources, or even better - Wikipedia:Pointless limitations for statements that ascribe info to the source in self-published sources. I'm sure Wikipedia servers have enough room for one more policy page.

Consequently, I would have accepted your claim that these limitations are imposed by this policy only if it had been stated explicitly that they apply for statements that ascribe info to the source in self-published sources, which is not. But let's not argue about it any longer! my proposal here is to amend the policy anyway. I'm calling it "clarifying the existing policy". You want to call it "a policy change"? so be it. The result is the same. I've created an RfC for the proposed change below. If anyone have sensible arguments for why any limitations should be imposed for AUTOMATICALLY dismissing statements that ascribe info to the source in self-published sources, they can present them there. Just remember - arguments that concern WP:UNDUE should not be used to reject my proposal, as they are completely irrelevant. (e.g. "A LOT depends on context - is the article focused on X or is it focused on Y") Sovmeeya (talk) 16:54, 6 October 2024 (UTC)

Well, it looks like you've been indeffed, but just in case you're reading this page anyway, mind the gap between not acceptable under our policies (notice my use of the plural word policies) and disallowed by this policy (notice your use of the singular policy). WhatamIdoing (talk) 21:58, 11 October 2024 (UTC)

About this: There is no EXPLICIT limitation in that policy to only claims that are presented as objective facts. This is why the policy is confusing, and this is why I want to clarify it. But it's a trivial deduction that this is the intended policy based on the purpose of this policy - verifiability.

Sovmeeya, I'm sure this "trivial deduction" feels logical to you, but the deduction is wrong. The purpose of this policy is to explain one (1) of the multiple requirements for getting content into Wikipedia. The requirement explained on this page is that the material – whether facts or opinions, which are treated equally in this policy [*] – must have been provably published somewhere else (i.e., off wiki), by a source that we consider "reliable".

There is nothing in this policy or any other that says Digital Confidence's view about Microsoft should be included in Wikipedia. BTW, if you happen to be connected to this company, this would probably be a good time to review Wikipedia:Conflict of interest and to let the team know that their criticism of Microsoft isn't likely to be included in the English Wikipedia.

[*] Facts and opinions are treated the same because we assert facts about opinions, rather than the opinions themselves. "According to Mel Musician, the rhythm as 'challenging'" is a fact – a fact that is about her opinion, but still a fact. WhatamIdoing (talk) 16:15, 23 September 2024 (UTC)

• Oppose: For the layman, the casual reader of Wikipedia, the difference between "2+2=5" and "According to John Doe, 2+2=5" is just a technicality. That's why this type of statements are allowed only on the current special circumstances. Also, the proposal can easily lead to loads of Argument from authority. Cambalachero (talk) 20:52, 6 October 2024 (UTC)
• Question: Why was so much time wasted on this ridiculous proposal? EEng 05:39, 7 October 2024 (UTC)

  I think the WMF should invest in some sort of fund that we can dip into to bribe folks with whatever amount they think they should expect to see from being promoted onwiki—just so we don't have to feel guilty about not arguing with them, or indeed feel guilty when we argue with them. Remsense ‥ 论 12:00, 8 October 2024 (UTC)

  Sometimes, it's helpful to explain things even when it seems hopeless because there exist people – rare, perhaps, but real ones – who have similar questions but don't want to ask themselves. WhatamIdoing (talk) 23:17, 11 October 2024 (UTC)

• Oppose in strongest possible terms. This is obviously not going to happen but it's important enough to weigh in anyway. Attribution is not a magic cure-all; it still puts something before the reader, which means we still have some responsibility to source it properly. WP:RSOPINION establishes that there is still a threshold of fact-checking and accuracy that must be met for opinion (ie. only some sources are usable for opinions), just a lower one than for in-article statements of fact. We cannot put something like "ResearchCorp says that John Doe is a pedophile" to an article sourced solely to ResearchCorp's website; nor can we use it for something like "ResearchCorp says that they have the cure for cancer" or something like that. Statements that are exceptional, or BLP-sensitive, or clearly self-serving require a secondary source exist for them so that we can provide necessary context; if no secondary source exists then we can't be covering them at all. And there's another case that is often overlooked - the practice of so-called "nutpicking", where an editor could look for the most absurd or incriminating statements from a BLP's own site or the like, and use WP:ABOUTSELF to post them in order to make that person look dumb or crazy, even if no secondary sources have covered those statements. All of these things require limits on how we can use non-RS primary sources about themselves. --Aquillion (talk) 04:08, 12 October 2024 (UTC)

Reliability of sources

I've noticed that many right-wing sources on this site are considered unreliable; my question is: why? I'm politically neutral and, considering that the politics of Italy and the United States are different, I have asked myself this question. Thanks in advance. JacktheBrown (talk) 22:48, 28 September 2024 (UTC)

That's not a question with a simple answer, but whether a source is perceived as right wing or left wing shouldn't be a consideration in questions of reliability. If you look into discussions about sources the issue are rarely ideological. Instead the cause is that media organisation have diverged since the days of print media, and that divergence has impacted media on different parts of the political spectrum differently. That change in the real world has then had an impact on Wikipedia. -- LCU ActivelyDisinterested «@» °∆t° 10:47, 29 September 2024 (UTC)

@ActivelyDisinterested: the thing that seems strange to me is that extreme left-wing sources aren't banned (maybe I'm wrong), while extreme right-wing sources are.
I repeat: I don't know the political situation in the United States in detail. In Italy we have a right-wing government; it's a government that many Italians support, because Italy is quite conservative, which for the Italian context is a very good thing; for the US context, however, it seems not, since the American right has, for example, denied climate change, which unfortunately exists.

Returning to the main topic: the thing that seems strange to me is that extreme left-wing sources aren't banned (maybe I'm wrong), while extreme right-wing sources are. JacktheBrown (talk) 14:11, 29 September 2024 (UTC)

extreme left-wing sources aren't banned (maybe I'm wrong) yes you age +are+ wrong. Take for example in British news media The Canary and Skwarkbox are both considered unreliable, and are on the far left of politics. Also to be clear sources aren't banned so much as actively discouraged if the consensus is that they are unreliable.

Whether governments or voters are of a particular part of the political spectrum is also not a consideration in judging the reliability of sources. -- LCU ActivelyDisinterested «@» °∆t° 14:21, 29 September 2024 (UTC)

@ActivelyDisinterested: thank you very much for this detailed and very useful explanation. JacktheBrown (talk) 14:23, 29 September 2024 (UTC)

The primary reason media sources get Deprecated (which is not quite a “ban”), is that they have repeatedly been shown to not fact check (or, in some cases, shown to completely invent “facts” which they report).

As to why right leaning sources seem to be more likely to be deprecated than left leaning ones… this is due to the fact that right wing sources tend to get more scrutiny. It’s no secret that Wikipedia attracts academics, who tend to be a somewhat left leaning group. They notice (and complain) when right-wing sources don’t fact check… and they tend to be a bit more forgiving when left-wing sources do the same.

That being said, we have deprecated a few left-wing sources when enough evidence has been presented to show they are not properly fact checking. It’s difficult, but possible. Blueboar (talk) 15:04, 29 September 2024 (UTC)

@Blueboar: Recognizing and admitting this problem is certainly a big step forward, but actions are more important than words; I would like neutrality not to be compromised in the encyclopedia. JacktheBrown (talk) 20:12, 12 October 2024 (UTC)

User:JacktheBrown - Outside of politics the banning also happens. I think it is just the mechanics or effect of WP:RSP. For the UK it seems more an elitism thing than a political leaning -- for example it seems most British press by volume is excluded as low class. RSP seems to interpret 'reliable' to mean 'respected' or 'truth' to that WP editor, excluding consideration of 'available' or 'accurate' from 'verifiability'. (You can elsewhere see discussions on a paywalled source or remote paper being preferred despite readers generally not having access to such or WP:VNT) The terms RSP uses are "Blacklisted" meaning mechanically edits including that site are blocked, or "Deprecated" meaning the guidance includes "generally prohibited" by automatic warning of such an editor and removal of such edits by third parties, or "Generally unreliable" meaning outside "exceptional circumstances" not to be used with removal by third parties and pings in TALK. If you want to see what discussions on what to ban are like, they are mixed in at WP:RSN. Cheers Markbassett (talk) 04:22, 13 October 2024 (UTC)

RfC on amending the policy to explicitly allow any statement that ascribes information to the source

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Should we amend this policy to explicitly allow any statement that inline ascribes information to the source, including any self-published source? Sovmeeya (talk) 16:45, 6 October 2024 (UTC)

Should we add to this policy under the section "Sources that are usually not reliable" a sub-section named "Statements that ascribe information to the source", and under it write the following:

"Any statement that inline ascribe information to the source is verifiable, and the source is a reliable for it, as long as there is no reasonable doubt as to the ascribed information authenticity. This include self-published sources. Consequently, any such statement is allowed under this policy.

There are no limits to the source and to the ascribed information, which may include unduly self-serving and exceptional claims; claims about third parties, including living people and deceased people; and claims about events not directly related to the source. The claims may be suspected as true, suspected as false, verified as true, or verified as false.

The ascribed information may be direct quotes or a summery of the information in the source in other words.

Examples:

• "According to X, Earth is flat"
• "X says "Y an idiot, a thief, and a fornicator""
• "X says he's the smartest man in the world"
• "X claims 2+2=5"
• "According to X, aliens from Mars are going to destroy the Earth in the year 2025"

"

The matter has been extensively discussed in the following: (the counts are of UNIQUE editors across all discussions)

• Wikipedia talk:Verifiability#Self-published sources for statements that ascribe the information to the publisher
  • 7 editors have disagreed with me that the Wikipedia:Verifiability policy currently allows such statements. Out of these, 3 believes Wikipedia:Verifiability depends on context, ("is the article focused on X or is it focused on Y", and 1 believes that such statements are "nominally verifiable", but if self-published, still disallowed by the Wikipedia:Verifiability policy.
• Wikipedia:Teahouse/Questions/Archive_1233#Is self-published work a reliable and usable source of info about themselves?
  • 2 editors have agreed with me that such statements are allowed by the current Wikipedia:Verifiability policy, although may be inappropriate due to WP:UNDUE.
• Talk:Microsoft Windows#Request for comment on reliability of a self-published work as a source of info about themselves
  • 3 editors have disagreed with me that the Wikipedia:Verifiability policy currently allows such statements
• Talk:Microsoft Windows#Privacy features addition reverted
  • 1 editor have disagreed with me that the Wikipedia:Verifiability policy currently allows such statements

Statements that ascribe info to the source in self-published sources, such as personal or cooperate websites, blogs, and social media, are a special and trivial case of self-published or questionable sources as sources on themselves. In such statement, only the info that the source has published something is presented as a fact, whereas the content of what was published is presented as subjective assertions.

The purpose of this policy should be only the verifiability of info by reliable sources. Self-published sources are weak, and so the policy currently have the following limitations on the type of info that can be based on such sources:

"Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer."

"

1. The material is neither unduly self-serving nor an exceptional claim;
2. It does not involve claims about third parties;
3. It does not involve claims about events not directly related to the source;
4. There is no reasonable doubt as to its authenticity; and
5. The article is not based primarily on such sources.

"

These limitations are very justified and make a lot of sense for the content of self-published sources, when it's presented as objective facts! But they are pointless and not justified at all for statements that ascribe info to the source. The purpose of these limitations is to assure that info in Wikipedia is reliable. The reliability for "X said Y" (cite X's website) is the highest possible. It's trivially verifiable. In such statement, Wikipedia vouches only for the simple fact that X have said Y. Wikipedia does not vouches for Y. Any reader understands that X could be mistaken or lying.

In this era of social networks, many famous/important/influential people have accounts they use to publish info about themselves, or their views/testimonies of other people/things/events. These include politicians, head of countries, head of large corporations, celebs, and all sorts of VIP's. What they publish can be important info for Wikipedia articles, usually about themselves, and occasionally about other matters. This is governed by WP:UNDUE, and should be judged on a case by case basis. Wikipedia should never AUTOMATICALLY dismiss statements that ascribe info to the source in self-published sources. My proposal, that have no exceptions, will prevent future disputes and save a lot of time.

Determining if a statement is appropriate for a certain article is a two stage process:

1. Is the statement verifiable? (this is independent of the article)
2. Does the statement pass the WP:UNDUE test? (this in article dependent)

These are two different and completely independent questions, covered by two separate and independent policies.

Wikipedia:Verifiability already explicitly says that "Verifiability does not guarantee inclusion" and refers to WP:UNDUE. So my proposal will not categorically greenlight any such statements. Inappropriate statements will fail WP:UNDUE. So there is really no reason for a concern.

Verifiability is only a function of the statement and the source. It's not a function of the context article. "Water is a compound with the chemical formula H2O." is verifiable statement for the article Water and for the article Tylor Swift. It just not relevant to the latter. Do not mix Verifiability and WP:UNDUE! It's very wrong! Wikipedia:Verifiability should only impose limits that are relevant to the verifiability by reliable sources of info that is presented as objective facts. Nothing more.

Pay attention: Arguments that concern WP:UNDUE should not be used to reject my proposal, as they are completely irrelevant. (e.g. "A LOT depends on context - is the article focused on X or is it focused on Y")

Hypothetical examples of how my proposal could apply, where mentioning statements with ascribed info is clearly essential:

• The ex-wife of the president of a large country has posted on her verified social network page that she has divorced the president because he used to regularly get drunk and beat her. The ex-wife is not notable for an article, and no other evidence are available to support this claim.
  • "The ex-wife of the president has said that he used to regularly get drunk and beat her." is a verifiable statement with a reliable source, and is due weight for the article on the president.
• An influential celebrity X tweeted on her official social media page the following: "I honestly believe that Earth is flat".
  • "X said that Earth is flat". is a verifiable statement with a reliable source, and is due weight for the article on X as well as for the article Modern flat Earth beliefs. Sovmeeya (talk) 16:50, 6 October 2024 (UTC)

Oppose Nothing has changed since last month's discussion above. Sovmeeya's attempts to get their own way is becoming tendentious and a time sink. Schazjmd (talk) 16:53, 6 October 2024 (UTC)

• Oppose. Adding 'According to X' in front of a statement does not somehow immunize it from verifiability concerns. If adopted, this loophole is so broad that it would amount to throwing away WP:V entirely. - MrOllie (talk) 17:02, 6 October 2024 (UTC)
• Malformed. Per WP:RFC, opening statements should be brief. I don't think this qualifies. DonIago (talk) 17:13, 6 October 2024 (UTC)
• Oppose and close as Malformed. That is the least brief or neutral RFC question I've yet to see. It also completely ignores the preceding discussion. -- LCU ActivelyDisinterested «@» °∆t° 17:37, 6 October 2024 (UTC)

  It isn't Malformed! The RfC statement is: "Should we amend this policy to explicitly allow any statement that inline ascribes information to the source, including any self-published source?" That's brief and neutral. As per Wikipedia:Requests for comment#Statement should be neutral and brief: "If you have lots to say on the issue, give and sign a brief statement in the initial description and publish the page, then edit the page again and place additional comments below your first statement and timestamp." You fail to understand a policy. Again. Doesn't that tell you something? Sovmeeya (talk) 17:51, 6 October 2024 (UTC)

  You continue to make uncivil remarks because other editors don't agree with you. -- LCU ActivelyDisinterested «@» °∆t° 17:53, 6 October 2024 (UTC)

  Even if the RFC isn't malformed, your own summary indicates that in the prior discussions a significant number of editors disagreed with you relative to the number of editors who agreed with you, so I have to wonder what you hope to achieve by escalating this. DonIago (talk) 18:05, 6 October 2024 (UTC)

  In the prior discussions there was a dispute about what the policy CURRENTLY says. I believe that the limitations in the current policy are only for info that's presented as OBJECTIVE FACTS. Most of the other editors believe that the limitations also applies to ascribed info in self-published sources. This RfC is for changing the policy regardless of what the policy currently says. No limitations should be to automatically dismiss "According to X" statements. Hopefully, the RfC will attract more editors. Sovmeeya (talk) 18:18, 6 October 2024 (UTC)

  Given how this discussion is progressing, I would encourage you to consider withdrawing this RfC as a gesture of good faith. DonIago (talk) 18:36, 6 October 2024 (UTC)

  Can I suggest that sometimes, even if you're right and everyone else is a fool, you have to accept that others don't agree with you and move on. -- LCU ActivelyDisinterested «@» °∆t° 18:51, 6 October 2024 (UTC)

• Oppose. Contrary to multiple core Wikipedia policies. AndyTheGrump (talk) 17:56, 6 October 2024 (UTC)

  What policies? WP:UNDUE? Have you read what I've written? WP:UNDUE is the second stage for determining if a statement is appropriate for inclusion. My proposal only says that "According to X" statements should not be automatically dismissed on the grounds that it violates Wikipedia:Verifiability. Any potential statement would still have to pass WP:UNDUE to be included. No problem! Sovmeeya (talk) 18:32, 6 October 2024 (UTC)

• Oppose Allowing "any" such statements is way too broad, easy to imagine all kinds of no-nos. Selfstudier (talk) 18:21, 6 October 2024 (UTC)

  If it so easy, why don't you share with us a "no-no", that also passes WP:UNDUE? Sovmeeya (talk) 18:27, 6 October 2024 (UTC)

• Oppose WP:Core content policies wisely says Because they complement each other, they should not be interpreted in isolation from one another. When Sovmeeya insists that These are two different and completely independent questions, covered by two separate and independent policies, the editor is wrong and policies must always be interpreted in relationship with other policies. This proposal is mediocre reductionism. Cullen328 (talk) 18:54, 6 October 2024 (UTC)

  Wikipedia:Core content policies is merely an "explanatory essay", not a policy or guideline. Wikipedia:Verifiability and WP:UNDUE complement each other indeed, but there is no relationship between them. The two questions are completely independent of each other. If a potential statement fails either, it will not be included. Sovmeeya (talk) 19:32, 6 October 2024 (UTC)

• The core issue if this RFC is: Are attributed statements of opinion verifiable by citing the opinionator directly?

If we quote someone (or closely paraphrase what they say) we do need to verify that we are quoting the person accurately, and cite where the quote originated.

That said… there are lots of OTHER policies that limit when it is appropriate to include a quote (such as UNDUE) in the first place… and I do think we need to point readers to those other policies. Our policies should not be read in isolation, but as an interwoven set. Blueboar (talk) 19:17, 6 October 2024 (UTC)

Yes. That's exactly what I'm saying. As I've said above, this policy (Wikipedia:Verifiability) already explicitly says that "Verifiability does not guarantee inclusion" and points to WP:UNDUE. So my proposal will not categorically greenlight any such statements. Inappropriate statements will fail WP:UNDUE. So there is really no reason for a concern. Sovmeeya (talk) 19:21, 6 October 2024 (UTC)

Thing is, WP:SPS most definitely does apply when editors neglect to add in-text attribution… when they state opinions as if they were fact (“This proves that the earth is flat” as opposed to “According to Ima Nutcase, ‘This proves that the earth is flat’”).

An unattributed opinion stated as if it were fact is NOT verifiable by an SPS. That distinction is important to mention, and is directly related to this policy.

When, how and whether to include SPS opinions is an area where there is an overlap between UNDUE and WP:V. So both policies need to mention it and support each other. Blueboar (talk) 20:22, 6 October 2024 (UTC)

• Oppose. The current policy sets the bar at about the right height. We should not be using self-published sources for content that involves claims about third parties or about events not directly related to the source. Rotary Engine ^talk 20:10, 6 October 2024 (UTC)
• Oppose I can't support as is though I see it has some points. My main problem is a due type, the information would be a primary source and we really do need a secondary source or for it in some way up front like if the topic is a company then information that is perfectly obvious on its home page. We can't go around trawling primary sources as reporters and make our own news. If noone else has picked up on the president's wifes's blog then we shouldn't have what it says in Wikipedia. NadVolum (talk) 21:38, 6 October 2024 (UTC)
• Oppose. One of the standard considerations in favor of the current policy is that, if something is really worth including, it will be picked up by some reliable source, and we won't have to rely on the SPS. To take your example of the unhappy ex-wife: If the president's ex-wife said that he used to regularly get drunk and beat her, responsible media outlets in that country would make sure that the post really came from her, would ask the ex-president for his side of the story, and would then publish an article that we could cite. If the ex-wife's post attracts no such attention, per your hypothetical that "no other evidence are available to support this claim"? The most likely explanation would be that the media outlets in that large country, familiar with the people involved, know that the poor woman has more or less cracked up, and that she's routinely vending lies about her ex-husband. Maybe last month she posted that he was dealing drugs, etc. They've looked into her latest allegation and decided that it's garbage not worth reporting on. The alternative explanation is that, in today's clickbait-hungry media world, a bombshell allegation against the country's president sank without a bubble. That's not plausible. So I think your example illustrates the merit of the current policy. We could verify that she made the allegation but including it, if there's no other support for it, would still be UNDUE. JamesMLane t c 02:15, 7 October 2024 (UTC)
• Oppose – eh, why? If you're running into friction or other issues here—if you feel you can only clearly justify something via a policy with this enshrined—that's probably an indication your idea of what should be included in a tertiary source is wrong. Remsense ‥ 论 02:20, 7 October 2024 (UTC)

• Oppose This is just another attempt (the third that I know of) by the OP to get some version of their desired result on this. We have been over this. Bludgeoning every response and repeatedly opening this is not helping. Meters (talk) 03:10, 7 October 2024 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Closing the RfC above

Perpetuating closed discussion

The above RfC on amending the policy to explicitly allow any statement that ascribes information to the source has unjustifiably been closed by User:Levivich less than 24 hours after start, saying that "No chance that consensus will form to repeal WP:V#Verifiability does not guarantee inclusion." But it's not an attempt to "repeal WP:V#Verifiability does not guarantee inclusion."! Most participants so far are the same as in prior discussions. I'm trying to attract more. We should leave this RfC open for at least a week. Sovmeeya (talk) 10:40, 8 October 2024 (UTC) any self-published source can be used as a source of information for any statement that ascribe the information to the publisher We already can and do use self-published sources to verify statements when we feel the inclusion of said statements is justified in context. if a statement did not rely on self-published sources in order to be verified, this change wouldn't make any difference regarding its inclusion. if inclusion of a statement was seen as justified per WP:NPOV and WP:ONUS, this change wouldn't make any difference either. So this change has no function as policy unless it impinges on WP:ONUS. Remsense ‥ 论 11:10, 8 October 2024 (UTC) The proposed change impacts poorly with WP:DUE@WP:NPOV, which requires viewpoints to be represented in proportion to the prominence in reliable sources. As currently written, WP:V excludes self-published sources from that body of reliable sources, unless those sources meet the conditions in WP:EXPERTSPS or in WP:ABOUTSELF. This is a good thing. We should not need to trawl through the quagmire of self-published sources to determine whether content meets NPOV. Rotary Engine talk 11:41, 8 October 2024 (UTC) You don't say? That's what I've been saying all along! OVER AND OVER!! In all discussions! I've said that over a month ago at Talk:Microsoft Windows#Privacy features addition reverted. ("There are two different and independent questions here") There are editors that think Wikipedia:Verifiability strictly prohibits this in some cases. (when the ascribed statements involve third-parties, etc.) Hence the RfC, which was prematurely and unjustifiably been closed for an invalid and false reason! It aims to explicitly state this to prevent disputes. Obviously, it does not impinges on WP:ONUS! (as I've clearly written in the RfC: "Inappropriate statements will fail WP:UNDUE. So there is really no reason for a concern.") Reopen it now!. Sovmeeya (talk) 12:30, 8 October 2024 (UTC) Nah. Remsense ‥ 论 12:34, 8 October 2024 (UTC) The people closing this RFC (which has no chance of going your way, see WP:SNOW) are doing you a favor. You've already been blocked for bludgeoning once, it is time to disengage and find another way to contribute to the encyclopedia, not get yourself blocked for edit warring to keep the RFC open. MrOllie (talk) 12:45, 8 October 2024 (UTC)

Note: I blocked Sovmeeya for a week for disruption related to this dispute. They are unlikely to be able to respond here during that time. Firefangledfeathers (talk / contribs) 12:53, 8 October 2024 (UTC)

The RFC should be re-re-closed, and this discussion hatted. The level of IDHT has now just become disruptive. -- LCU ActivelyDisinterested «@» °∆t° 13:22, 8 October 2024 (UTC)

 Done NebY (talk) 13:38, 8 October 2024 (UTC)

Source display

We have achieved a consensus at the List of common misconceptions to split one of our Special:LongPages. Once we decide exactly how to split it, that page will be converted to a very short list (e.g., links to * List of common misconceptions (A–G), or * List of common misconceptions about history, or whatever is decided).

Some editors really appreciate the one-stop-shopping aspect of the long page, but there are so many refs that it's run into the Help:Template limits problem. If there were no refs on the page, of course, we wouldn't see that problem. An editor has very kindly mocked up an option at User:S Marshall/Sandbox/List of common misconceptions demo that would transclude the real lists (e.g., the subject-specific lists) into a single "List of common misconceptions (one page)" that doesn't display the refs. To see the refs, you would click through to the real lists, where you would find identical wording, but this time with the refs shown. To be clear, this display style is meant to be in addition to, rather than instead of, the real lists.

Our question is: Would this be acceptable in the main namespace in terms of the WP:V policy? WhatamIdoing (talk) 19:13, 15 October 2024 (UTC)

I wouldn't think it's a problem in general, but in specific instances it could be. If the information is a direct quote, material that has been challenged, material that is likely to be challenged, or, probably the most important one, contentious material about a BLP, then it would appear to conflict with WP:V/WP:BLP. Not including that information, or only including those references in the transclusion could be a solution. But that may become a mess of include and noinclude tags.

I could see the issue of CIRCULAR being brought up, but it's not being verified by another Wikipedia article it is content transcluded from another Wikipedia article. So I don't think that's a concern. -- LCU ActivelyDisinterested «@» °∆t° 19:25, 15 October 2024 (UTC)

I tend to lean towards the “information needs to be cited in every article in which it appears” mode of thinking. If you are transcluding cited information from one article to another, why not also transclude the citations? It doesn’t add extra work. Blueboar (talk) 00:20, 16 October 2024 (UTC)

There are so many of them that all the citations won't display. There are server-side limits to how much template content you can put on a single page. After a certain point, it just stops rendering all subsequent templates, and everything else is an error message. This is one of the two driving forces behind the decision to split the page. WhatamIdoing (talk) 00:27, 16 October 2024 (UTC)

Washington Post & LA Times

Are these two newspapers still usable sources after the recent interference by the billionaire owners showed that a fact-based reporting can possibly be surpressed by them when it may bring trouble to the billionaires and their businesses by one of the 2024 candidates for US-President? This question is brought to you by the series 'Questions at the Dawn of Fascism'. --Jensbest (talk) 12:14, 28 October 2024 (UTC)

Any decision by the owner(s) of a newspaper on editorial page policy does not necessarily say anything about the reliability of the newspaper. Editorials, including election endorsements, and opinion pieces, are not generally generally used as reliable sources, anyway. Donald Albury 12:41, 28 October 2024 (UTC)

That's not the point. The point is that an owner has interfered because he fears reactions by a possibly autoritarian US-President. It is the obvious attempt to appease Trump. Trump often and clearly expressed his desire to harm businesses which he dislikes. Considering this history and that the tone of Trump has become fully fascistic, it is not safe to consider the reporting of these two newspapers based on facts and untouched by their billionaire owners. --Jensbest (talk) 13:02, 28 October 2024 (UTC)

You are in the wrong place. The right place is WP:RSN. Zero^talk 13:16, 28 October 2024 (UTC)

The editor earned a WP:TBAN about Trump eight years ago, and has now been blocked for violating it. WhatamIdoing (talk) 18:05, 28 October 2024 (UTC)

ONUS a blank check?

We've discussed this in the past and I'm wondering what the current thinking is on ONUS as a reason to remove something without any other reason, even long-standing, multi-years long content that once had a consensus to remain presumably. Is it enough for a few editors to remove something then claim ONUS? What about if an RFC is started? Usually, an RFC means that the status quo remains for the duration, and WP:NOCON means the status quo remains. Does ONUS still have the ability though to just be a blank check to remove anything at all that a few editors don't like? Andre🚐 19:33, 28 October 2024 (UTC)

I've always supported ONUS but I don't believe, or have ever believed, that the wording of ONUS can be used as a reason for removing content. If content is removed for a valid reason and is restored, then ONUS is a reason it shouldn't have been restored until there is consensus for it. It's shouldn't be the reason that the content was removed in the first place. -- LCU ActivelyDisinterested «@» °∆t° 19:55, 28 October 2024 (UTC)

Right, but what about a situation where content has a consensus at a past time, or was added and stood for years at a reasonbly well-watched article, then a few editors come along and create a no-consensus situation due to either inactivity of the editors who originally added/supported that or simply attrition/changing perspectives and userbases, and use ONUS as a justification for "nocon -> remove or change" as opposed to "nocon -> status quo"? If you catch my drift. Andre🚐 19:57, 28 October 2024 (UTC)

I guess my feeling is that if someone is going to claim ONUS as a reason for removing content, they'd better have an underlying explanation ready, and it would probably be best to provide some insight into that reason in the edit summary rather than simply claiming ONUS. I also think unless it's linked, newcomers may very well have no idea what "ONUS" as an edit summary is even referencing. "Better before" isn't much better than "ONUS", but at least it gives a bit more of an inkling as to the reverting editor's mindset.

I'm not readily aware of any situations like the one you've described, though I can't rule out the possibility...however, consensus can change. If editors who came to Consensus A have all gone inactive, and a new number of editors come along six months later with no knowledge of the prior consensus and achieve Consensus B without the prior set of users engaging with them, I'd argue that that's an indication that consensus may indeed have changed. I'm not quite sure I'm addressing the scenario you're describing, though. DonIago (talk) 20:03, 28 October 2024 (UTC)

Yes, but we're talking about a scenario where there's no consensus whether consensus has changed. Therefore, usually, WP:NOCON means status quo. Andre🚐 20:05, 28 October 2024 (UTC)

Isn't that why dispute resolution processes exist? DonIago (talk) 20:07, 28 October 2024 (UTC)

I mean, sure. Although that's a very different discussion. I'd say my recent experience with dispute resolution has left me with some ideas for improvement, as a former mediator with the defunct MEDCOM. Which wasn't always the best either. But you still haven't really answered my question. Which is OK, it's OK to answer my question with another question. I will then answer that with another scenario: What if we went through dispute resolution and the outcome was to have an RFC (which is an outcome), and we're getting NOCON again? Equally strong camps, equally strong arguments (as in the WAID style thought experiment) Andre🚐 20:10, 28 October 2024 (UTC)

There was an RFC on that exact thought experiment, it was never closed as there was no consensus. -- LCU ActivelyDisinterested «@» °∆t° 20:16, 28 October 2024 (UTC)

Yep. Exactly. Which means further discussion and here we are. Andre🚐 20:19, 28 October 2024 (UTC)

Yes and all this has been said before. "It depends on the specifics" is the best there is. Editors have been blocked for running around quoting ONUS removing content and being disruptive, and of course the same is true on NOCON and stokewalling. Being open to others ideas and viewpoints while working to find a solution is the best you can do. If editors are displaying behavioural issues and being disruptive there are other means to resolve the issue. -- LCU ActivelyDisinterested «@» °∆t° 20:24, 28 October 2024 (UTC)

At some point if editors who prefer a difference consensus aren't getting their desired result even after going through an RfC, maybe they need to accept that there simply isn't a consensus for their preferred version, and perhaps try again in six months or such? Anyone who's spent significant time on Wikipedia likely has experience in losing an argument and hopefully trying to be graceful in defeat. DonIago (talk) 20:26, 28 October 2024 (UTC)

I have always thought that being long standing on its own barely counts as a consensus, it's certainly very weak. Consensus shouldn't stand forever, the project will become moribund.

The real issue is that both ways have issues, some editors misuse ONUS to remove content just because they dislike it and at the same time other editors misuse NOCON as a way to stonewall. -- LCU ActivelyDisinterested «@» °∆t° 20:05, 28 October 2024 (UTC)

I agree that NOCON is a form of stonewalling if abused, but what about a legitimately stipulated situation where something was added on a well-watched article, had a consensus at one point, and then a NOCON exists today? Does ONUS therefore allow removing anything that can be fought to a NOCON? As opposed to the usual situation as I understand it that a NOCON means status quo, except perhaps in extreme situations such as a BLP UNSOURCED where policy is clearly on one side. I'm talking about an even content dispute NOCON which leads to removal through ONUS. Is that legit? Andre🚐 20:07, 28 October 2024 (UTC)

Maybe? That's the best answer there is. It's not something that can be given a satisfactory answer in a general sense, only in a specific case by case basis. Discuss it, discuss it some more, discuss it at a venue with more with a bigger audience, have a formal discussion, all the normal dispute resolution processes. -- LCU ActivelyDisinterested «@» °∆t° 20:11, 28 October 2024 (UTC)

“Consensus can change”… so a prior consensus can be overturned by a newer consensus. If there is currently a consensus to omit or remove material from an article, then that material should be omitted or removed… even if there used to be a consensus to include that material.

Of course, whether consensus has in fact changed (or not) can sometimes be a matter of dispute. If so, use the normal dispute resolution steps (seek 3rd party opinions, file an RFC, etc) to resolve the issue. Blueboar (talk) 20:11, 28 October 2024 (UTC)

and when/if the RFC ends with no consensus being clear to omit or remove the material, does ONUS default to removing it? Or does NOCON default to the status quo? Andre🚐 20:15, 28 October 2024 (UTC)

It's starting to feel as though you're asking for a global remedy to a problem that should be handled locally if and when it arises. DonIago (talk) 20:58, 28 October 2024 (UTC)

I'm not asking for any specific remedy. If the answer remains unclear then it's unclear. Andre🚐 21:41, 28 October 2024 (UTC)

Technically, NOCON doesn't "default" to anything, as it's not a rule. It is a claim that "the common result is" to retain the STATUSQUO. Perhaps we should make that clearer, e.g., When discussions of proposals to add, modify, or remove material in articles end without consensus, the common result – but not a result that is required by any policy, including this one – is to retain the version of the article as it was prior to the proposal or bold edit. WhatamIdoing (talk) 23:07, 29 October 2024 (UTC)

That would resolve the conflict by making ONUS supercede NOCON. Not my preference but a valid resolution of the tension. However, I would argue that this is an ambiguity in the current wording. I would read the current wording as defining the "common result" as carrying the force of policy/guideline (like all guidelines, not always observed strictly) and then there are 3 bullet points defining the bounded, but not exhaustive, exceptions to the "common" result. Certainly, while this may not be an accurate interpretation of the policy, it's a common one. Andre🚐 23:20, 29 October 2024 (UTC)

Well, I didn't write that sentence originally, so I can't tell you what the original intention was for the wording choice, but based on what I know of the now-blocked editor, he would be very happy to have you interpret it as requiring inclusion, or at least STATUSQUO.

When I write that something is the "common" choice, I never mean that it is "carrying the force of policy/guideline"; I mean what it says, i.e., it is a statement of fact about the frequency of an event. For example, when I wrote the paragraph in MOS:FNNR about which section heading to use for ref lists, it was based on a survey of a random set of articles. It strictly says which section headings are most common, as a point of objective fact, without telling editors that they should prefer the most common. We did expect that editors would use this information to voluntarily choose the most common section heading (and they have), but it is not "a rule" and it is not a matter of "force". WhatamIdoing (talk) 23:36, 29 October 2024 (UTC)

My guess is that given how old this part is, it was a case of someone writing something down that was observed before its incantation was transcribed. In keeping with the section in policy wherever it is - I know you know the one - that says that all of the policies and guidelines are merely description of a consensus and not the thing in itself. AFAIK, it's always been the case in RFCs and VFDs/AFDs that nocon meant keep. Whereas the ONUS/BURDEN supporting remove thing seems much more recent. Without having exhumed the remains myself to confirm that. There also might be an alternate older wording of this concept. Maybe in the deletion or RFC policy if not here. Andre🚐 23:43, 29 October 2024 (UTC)

The NOCON expansion to prefer QUO was discussed repeatedly before the addition was made, and the editor knew that there were concerns (from me and others) about it.

The specific sentence in ONUS was added two years later, but the edit summary when he updated the shortcut (originally created in 2008 and pointing to the same section as BURDEN) suggests that this was not intended to be "new" content. WhatamIdoing (talk) 00:04, 30 October 2024 (UTC)

That version from 2012 is clearer and more decisive than the current language: In discussions of textual additions or editorial alterations, a lack of consensus results in no change in the article. No "common result" there. It may have been watered down later. But I'm sure that nocon meant keep/status quo even in probably 2005. Here's an essay from 2007. Wikipedia:What "no consensus" means Andre🚐 00:10, 30 October 2024 (UTC)

That essay doesn't address article changes, and on the day it was written, the main point appears to have been: Often, people feel that no consensus should mean that the current status quo prevails. That is not, however, always the case. WhatamIdoing (talk) 00:48, 30 October 2024 (UTC)

I agree, but, that supports the idea that this was the generally practiced procedure at the time. Andre🚐 00:59, 30 October 2024 (UTC)

Maybe. Or maybe that back then we had problems with oversimplification of rules. Or maybe that back then we had problems with wikilawyers claiming that whatever asserted rule supports my preferred outcome is the One True™ Rule. WhatamIdoing (talk) 04:29, 30 October 2024 (UTC)

I see. Well, that's not how I remember it. Andre🚐 05:11, 30 October 2024 (UTC)

I found you, WAID, in 2016 removing some status quo from an old version of BRD. Andre🚐 01:17, 30 October 2024 (UTC)

I have made many edits to BRD over the years. In general, my earlier ones took BRD away from its original concept, which was less about WP:EPTALK and more about what to do when achieving consensus has already proven difficult. WhatamIdoing (talk) 04:27, 30 October 2024 (UTC)

(Hopefully useful links) Wikipedia talk:Verifiability/Archive 74 and Wikipedia talk:Verifiability/Archive 75 host the 2022 debates about ONUS. Wikipedia talk:Verifiability/Archive 77 is the long discussion on BURDEN. ONUS comes up again on Wikipedia talk:Verifiability/Archive 80. All very lengthy discussions, probably helpful to read before reviving the arguments. Schazjmd (talk) 20:35, 28 October 2024 (UTC)

yes, and my name appears in Archive 77 and Archive 80 so I participated. Those archives still don't provide a resolution to the question. Nor does Wikipedia:Requests for comment/When there is no consensus either way Andre🚐 20:43, 28 October 2024 (UTC)

The links are for everyone. Other people are participating here and might not have read those discussions. Schazjmd (talk) 20:45, 28 October 2024 (UTC)

That's true. Thanks. Andre🚐 20:48, 28 October 2024 (UTC)

ONUS has many problems including conflicting with other policies and processes, being out of place (what the heck is a ham-handed arbitrary finger on the scale towards exclusion doing in wp:verifiability?). Also the commonly quoted word "ONUS" is not even in the policy. The original intent was to prevent somebody from using meeting wp:verifiability as an argument for or a way to coerce inclusion. We should fix the whole mess and turn it into a big plus by removing it and substituting "WP:Verifiability is a requirement for inclusion, not a reason for inclusion"' which was the original intent of the wording linked by ONUS. That change would also be neutral regarding inclusion/exclusion because it makes slight balancing changes in either direction. Sincerely,North8000 (talk) 21:41, 28 October 2024 (UTC)

How many times we going to have this discussion ) Selfstudier (talk) 21:47, 28 October 2024 (UTC)

Apparently we need to have it since editors still cite ONUS to remove things. I agree with North8000 that this would solve the problem in a fair way. Andre🚐 21:50, 28 October 2024 (UTC)

We have problems with NOCON, let's remove them both and let the talk page decide through discussion. -- LCU ActivelyDisinterested «@» °∆t° 22:24, 28 October 2024 (UTC)

Something to consider… if there WAS a consensus to include, but is now at NO CONSENSUS, then the consensus has changed… the pendulum is swinging towards not including. So, my “tie breaker” is to omit… and ask again in a year to see whether the pendulum has continued to shift towards omit… or has swung back towards include. Blueboar (talk) 21:53, 28 October 2024 (UTC)

I don't think I agree with that, Blueboar. If there is no consensus to remove material, WP:NOCON would retain the status quo. At least, that's how it's normally worked that I've observed Andre🚐 21:58, 28 October 2024 (UTC)

In principle but that doesn't necessarily apply in all cases and I notice you have studiously avoided being specific. Selfstudier (talk) 22:57, 28 October 2024 (UTC)

Andrevan… Is there “no consensus” to remove the material, or is there “no consensus” to retain the material? Answer: it depends. Ultimately you have to look at how the RFC question was worded, read the comments people made in reply, and figure that out on a case by case basis. Yet another good example of why RFCs are not simply !votes. Blueboar (talk) 12:45, 29 October 2024 (UTC)

That's one of the problematic conflicts of wp:ONUS with the wp:consensus policy. If there is no consensus to remove or reaffirm some long standing material, wp:consensus says that it stays in and wp:Onus says that it comes out. North8000 (talk) 18:54, 29 October 2024 (UTC)

Again… it depends. If the question asked was “Should Xxx be retained?” and there is no consensus to retain, then I would take that lack of consensus as a sign we should remove. Alternatively, if the question asked was “Should Xxx be removed?” and there is no consensus to remove, then I would take that lack of consensus as an indication that we should retain. Blueboar (talk) 19:17, 29 October 2024 (UTC)

Those are the same question, though. Should it be retained or removed - the only differing factor is whether it's there now and is stable and long-standing, or whether it was recently added and its addition is contested. Andre🚐 19:31, 29 October 2024 (UTC)

It's not only that, there is the question of WP:CONLEVEL to consider. Selfstudier (talk) 19:41, 29 October 2024 (UTC)

Yup… which is yet another reason why we can’t have a simple one-size-fits-all “default”. We have to look at the actual discussion. How many editors commented? Did they focus on reasons to include (which failed to gain consensus), or did they focus on reasons to omit (which failed to gain consensus)? How we proceed will depend on all of these nuances. There is no single correct answer. Blueboar (talk) 19:59, 29 October 2024 (UTC)

I don't see the relevance of CONLEVEL. Whether to include or exclude a given bit of material in a specific article has nothing to do with a small group of editors trying to "override community consensus on a wider scale". CONLEVEL does not refer to small discussions. CONLEVEL is about small groups of editors declaring that "their" articles are exempt from site-wide policies. WhatamIdoing (talk) 20:40, 29 October 2024 (UTC)

In my post that started this subthread, (That's one of the problematic conflicts of wp:ONUS with the wp:consensus policy. If there is no consensus to remove or reaffirm some long standing material, wp:consensus says that it stays in and wp:Onus says that it comes out.) I was intending to imply that both questions (remove and keep) got asked and neither attained a consensus. Then ONUS dictates the opposite of wp:nocon. North8000 (talk) 21:25, 29 October 2024 (UTC)

Four years ago, you and I talked about moving ONUS to NOCON, so the contradiction could be side-by-side in the same page.

Recently, there's been talk about moving NOCON out of CON. That seems to have stalled, and not just because a couple of editors were worried about my suggestion that it be moved to a page that is more relevant but doesn't currently say "policy" at the top. WhatamIdoing (talk) 21:33, 29 October 2024 (UTC)

It's been many months since I did this but I went back and analyzed the roots. IMO there was a tacit local agreement with the inclusion (including an edit summary the gist of which is what I said above....to prevent using using verifiably as a way to coerce inclusion) but no big discussion. And certainly NO agreement with the way that it is being used today as I described above. IMO it's a change that would fix many problems and conflicts, aligns with the intent of why it was included in the first place, does not go against any previous large discussion decision, and would be pretty neutral in the inclusion/exclusion equation. Sincerely, North8000 (talk) 23:21, 28 October 2024 (UTC)

While I lean toward North8000, what I'm gathering so far from the discussion is that there still isn't a consensus and it depends. Which is fine. That's how things were last time I checked in. NOCON/PRESERVE is a policy and so is BURDEN/ONUS. However, I do think that last time we discussed it, I think it was ActivelyDisinterested who proposed adding something saying that ONUS can't stand alone as a rationale or when paired with other otherwise invalid rationales, it doesn't add anything. Or clarifying that ONUS specifically doesn't mean that you can remove some long-standing content, claim ONUS if there isn't immediately produced a new consensus to retain it, and have policy justifying that this content has now lost its consensus even if it's been previously considered to have one. I don't think it's theoretical that this is or can be disruptive. ONUS was never intended to be a blank check, that much is clear from the discussion. If we're not going to resolve the conflict with NOCON, at the very least, could we consider an option to limit the scope of potential misuse of ONUS? I just don't think it's a good idea or accurate to the intent of ONUS to allow a situation where long-standing content, with no consensus to change, can be removed by default. This effectively means that 2 editors can have policy endorse removing material that previously, let's say 10 editors supported including, if only 2 of those editors show up for the re-discussion at such time as it is called. Maybe they show up but not until 60 days later, too late. Now you need a new consensus to include that content. I don't see that as a good thing for the integrity of the material on Wikipedia. Andre🚐 20:54, 29 October 2024 (UTC)

Huh… I thought one of the selling points of Wikipedia was that its material is dynamic… stuff gets added, stuff gets changed, stuff gets removed. Blueboar (talk) 21:06, 29 October 2024 (UTC)

Dynamic, yes, but there's a point where it's not a net benefit to keep changing something, particularly a stable and controversial high-quality article. Surely you can come up with an example of this? There's a real example that caused me to come here, but I think better to keep it more abstract for a variety of reasons. There's a problem of defending article quality. I wrote that essay in 2005 and the original title was defending the status quo. I think now that the original title was more along the lines of the point needed. There are other solutions to the problem or maybe you don't agree it's a problem, but I think we're oversimplifying things if you think that Wiki just hums along with people adding or removing stuff, especially in controversial areas, and that every time some long-standing content is challenged that it's a legitimate and helpful challenge. Andre🚐 21:13, 29 October 2024 (UTC)

I'm not so much worried about the 2 editors now but 10 editors give years ago stuff, things change and even if 2 editors removed it 2 can add it back 6 months later. Content changes and I fear the idea it must stand because a group of editors said so at some point in the past just tends towards stonewalling. Content can be removed for many reasons, even just to improve the article. Bloated articles are not better by default.

I don't think ONUS, or VNOT if editors dislike the word, should be used as a initial reason to remove content, as I've said, but my ideas to reformulate the statement didn't get much of any support and I'm in no hurry to try and revive it. -- LCU ActivelyDisinterested «@» °∆t° 22:52, 29 October 2024 (UTC)

• My impression for previous discussions is that the dispute between WP:ONUS and WP:NOCON mostly amounts to editors disagreeing about when text gains a degree of implicit consensus by being longstanding and having people edit it. This is a complex question that we've avoided nailing down (for good reasons; having it too rigid would lead to people arguing over it rather than focusing on actual content.) But as a general rule-of-thumb I feel policy could be a bit more clear on the main points - content that has been seen by many eyes eventually gains a degree of consensus, since each of those people is presumed to have approved of it to some degree, and after that at least some consensus is needed to remove it. While people often use time as a rough handwave (often approximately three months of stability), and the "status quo" or "longstanding" are sometimes used to refer to it, it's really about the eyes; something on an obscure article that few people see might never accrue that sort of consensus, while higher-traffic articles can get it more quickly if it's clear a diverse group of people are editing it and seeing it. And any sort of indication that it's controversial (any objections on talk, say) prevent it from happening. --Aquillion (talk) 14:37, 31 October 2024 (UTC)

If there is no longer consensus to include something, it can be removed. Also a "finger on the scale towards exclusion" is the right way to have it. Almost all content policy is about restricting what can be put into articles and hardly any of it is about restricting what can be taken out. There's no reason for this example to be different. Zero^talk 03:10, 29 October 2024 (UTC)

That's not how policy or practice works; if we have an RFC over whether to remove an established section of an article (one that previously had consensus of some form), and it reaches "no consensus", the result will be that it is retained. I disagree with your proposal to add a finger on the scale towards exclusion, too; current policy places a priority on stability instead. Adding a finger on the scale towards exclusion encourages WP:STONEWALLing and discourages people from engaging. Generally speaking I don't think that proposals for these sorts of "strong" policies with default outcomes that could drastically change articles are ever workable - heavy-handed policies seem good in people's heads when they picture it giving the "right" answer, but in practice they reduce incentives to engage and compromise, encouraging people to use policy as a bludgeon instead. It's better to have a grey area with. (As an aside, this is, to me, the most frustrating part of discussions about ONUS - people who want to push for an expansive interpretation of it, which would allow for the removal of longstanding text without a consensus to do so, frequently make baffling assertions that this is already how things work, which anyone who edits Wikipedia in controversial areas knows is not the case. Outside of a few situations, like WP:BLP, no-consensus outcomes result in the status quo being retained, not in removal; the standard for disputes is WP:BRD, putting the article back in the status quo until a consensus is reached, not Bold-Revert-Remove.) --Aquillion (talk) 14:37, 31 October 2024 (UTC)

I agree with this and your previous post. WP:ONUS is a finger on the scale towards exclusion and I opined that such a ham-handed arbitrary thing is a bad idea. Zero0000 opined that it is a good idea. So it wasn't a proposal to add something. Sincerely, North8000 (talk) 18:29, 31 October 2024 (UTC)

Agree with North and Aquillion in my humble opinion. While it may not gain consensus to be changed, I encourage a constructive proposal to change text on these lines to make it clearer. Andre🚐 19:12, 31 October 2024 (UTC)

There are differing levels of consensus though. If something has an RFC support for, then maybe it makes sense to maintain inclusion until there is a consensus against. But if something just has an implied consensus through dint of being in the article over time, then once that implied consensus is gone by having been challenged then it should not be retained absent a consensus for it. By making it so default to status quo you are empowering filibustering to retain challenged material that never had any real firm consensus to begin with. Something that say 4 users discussed but is now challenged by 20 shouldn't be retained either. The circumstances matter, and filibustering shouldn't be rewarded. When there was a larger discussion among 20 users, then yes I agree that this consensus holds until a new one forms, but the other way around? Not so much. nableezy - 19:27, 31 October 2024 (UTC)

Well...

If I write an article and include some bit of content, and it 'sticks', then that's evidence of a presumption of consensus.

If we have a discussion that does not prove that presumed consensus to actually exist, should we keep assuming that there is a consensus? WhatamIdoing (talk) 21:05, 31 October 2024 (UTC)

At best, that's silent consensus, conlevel 0, WP:SILENT. Selfstudier (talk) 21:37, 31 October 2024 (UTC)

Do you mean the pre-discussion situation is a silent consensus? Once a discussion has happened, I wouldn't say that we're still silent. WhatamIdoing (talk) 02:07, 1 November 2024 (UTC)

In my opinion, it seems best to conceptualize WP:IMPLICITCONSENSUS only in terms of identifiable individual editors signalling identifiable (if implicit) assents. If an obscure article remains in a given state for seven years and only one editor has ever contributed to it, there's no evidence that another editor is aware of its contents, never mind that they've consented to them. The same would generally hold regarding editors that previously engaged with an article but are no longer active—their consensus can no longer change, so it can't really be considered. Remsense ‥ 论 21:51, 31 October 2024 (UTC)

The above are discussions about the complexities of wp:consensus. This overall discussion is about what has operatively become arbitrary interference/conflict with the wp:consensus process....wp:onus. Which has jumped the track from it's original intent of keeping meeting wp:verifiability from being used to coerce inclusion. Again suggest fixing the whole mess and going back to the original intent by substituting onus with "wp:verifiability is a requirement for inclusion not a reason for inclusion". Sincerely, North8000 (talk) 02:23, 1 November 2024 (UTC)

Apologies for getting sidetracked. If I've read the room correctly: would it be fair to characterize WP:ONUS as a blunt tool to initiate the consensus-making process, rather than as a blunt tool to circumvent it? While I don't think you're wrong in observing it can function as a finger on the scale towards exclusion, I think in most cases it is better described as tending towards stability, because in a majority of applicable situations stability and exclusion happen to be the same outcome. I don't really think it is a natural policy point to apply if one is trying to remove material, unless I'm missing soemthing? Remsense ‥ 论 03:51, 1 November 2024 (UTC)

Not in my experience. ONUS can be cited when a few editors want to remove something, or prevent the addition of something. Verifiability doesn't guarantee inclusion, so this should be removed unless there is a consensus to retain it. Andre🚐 03:58, 1 November 2024 (UTC)

Someone yank me back off-stage if I'm veering off-topic again, but I do think this is directly relevant? It's tough, because I will diverge with opinions stated above (and likely majority opinion in the community, which is fine) in that I don't think presence should be privileged over absence when it comes to discussions on the fringes or outside of WP:V and WP:NPOV. Very often, if we want to entertain cohesion or parsimony as valid editorial goals, we have to fall back on a hamfisted-if-not-false argument that a given content presentation (or any given aspect that doesn't boil down to the claims made by prose) "doesn't reflect sources". WP:PRESERVE says what it says, but it remains indefensible to me that additions should categorically require lower editorial scrutiny than removals do: it is equally important to the quality of articles to discern what they shouldn't say or present, compared to what they should. Remsense ‥ 论 04:15, 1 November 2024 (UTC)

I think I agree with the general sentiment. What I think shouldn't be a permissible practice is to remove something that is long-standing or had consensus in the past citing ONUS. If that BOLD removal is reverted, it should remain unless there is consensus to remove it, not just no consensus to keep it. The BURDEN should be on the BOLD change, not on keeping the status quo. Andre🚐 04:18, 1 November 2024 (UTC)

SPS definition

A discussion on RSN has broken out which revolves around what exactly a self-published source is. From my perspective, WP:USESPS provides the only comprehensive guidance for what is or is not self-published, and it is pretty unambiguous.

However, many of the responses have come down to claiming that USESPS is wrong and should be disregarded or deleted.

Leaving aside the surrounding argument about bias and reliability for this specific source, does anyone uninvolved have any opinions on this? If editors are supposed to disregard the guidance on USESPS, I would like to know. Void if removed (talk) 14:01, 31 October 2024 (UTC)

There was something way back that I had brought up the idea that self published means there is no editorial oversight from a person publishing their work (ignoring copy editing or the act of publishing it on a website).
So someone putting info up on Medium clearly is SPS, while aember of an advocacy org like GLAAD or SPLC will need editorial approve from their in house editors before sonething goes online, and thus is not an SPS. — Masem (t) 14:15, 31 October 2024 (UTC)

Ok, but that isn't at all what USESPS says, and by that measure absolutely every corporate website isn't self-published. So - is USESPS just completely wrong?

My understanding is there is supposed to be some level of separation between the publisher and author. Void if removed (talk) 14:23, 31 October 2024 (UTC)

No, that bears no relation to how WP:SPS has been interpreted until now. Most sources employ their own writers, after all; if the fact that the writers work for the publishers made something a SPS, virtually everything would be a SPS. What the part of USESPS that you're misinterpreting means is that something is a SPS if the writer and publisher are literally the same person, ie. there's no editorial oversight at all; or if the entire thing is presented as a single monolith with no indication of who writes what and therefore no way of determining that there's a distinction between publisher and author. GLAAD lists individual writers for the things it publishes in the acknowledgement section, and those writers are not themselves the publishers, so they don't fall under the part of USESPS that you're citing. Otherwise, by your argument, I could say "well, the New York Times' reporters and writers work for the New York Times, therefore the editorial controls are not independent and it's a SPS." Now, whether the editorial controls are good enough is another story, but you can't just say "the writers work for the publisher, therefore this is a SPS." --Aquillion (talk) 14:50, 31 October 2024 (UTC)

I'm hoping for uninvolved input here, but again: that's not what USESPS says, which is:

If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same.

What I'm trying to get at is why USESPS says things that are completely different to how you and some other editors are interpreting SPS. If Business, charitable, and personal websites are given as a specific example of SPSs, why is this charity's website not self-published? If USESPS says a SPS can have a professional structure in place for deciding whether to publish something, why are people arguing that having some sort of notional approval process means it isn't self published?

I want to know why what the guidance says, and what half a dozen editors are currently saying over on RSN, are so at odds, and which is correct? Void if removed (talk) 16:02, 31 October 2024 (UTC)

Masem, you give an example of Medium.com vs SPLC, but the two collide at https://medium.com/@splcenter I'm sure their PR department does in-house review for both the content on their official Medium.com posts and the content on their https://www.splcenter.org/ website. The level of editorial oversight is the same in both cases, and I suggest that both of these are self-published, because in both cases, the org is posting whatever it wants, whenever it wants. WhatamIdoing (talk) 23:03, 31 October 2024 (UTC)

Counterexample, apparently NYTimes authors are also writing on medium with the NYTimes logo.

https://medium.com/@timesopen

And the WaPo folks just have their own account on there:

https://medium.com/@washingtonpost

Similar counterexamples, all the news media sites out there operate Twitter accounts, and WP:twitter is usually always considered WP:SPS.

Operating on a SPS platform to promote itself does not imply SPS of the organization. Bluethricecreamman (talk) 23:21, 31 October 2024 (UTC)

The definition of self-publishing exempts traditional publishing houses, so news articles are non-self-published no matter where they post.

Of course, any organization can self-publish something. The pages on their websites that tell you how much a subscription costs, or that try to talk you into running ads in the newspaper are all self-published. WhatamIdoing (talk) 02:02, 1 November 2024 (UTC)

I'm not sure that USESPS is the issue so much as this material from WP:V, Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content. Further examples of self-published sources include press releases, the material contained within company websites, advertising campaigns, material published in media by the owner(s)/publisher(s) of the media group, self-released music albums, and electoral manifestos: Springee (talk) 16:27, 31 October 2024 (UTC)

I agree that there are some difficulties in the line "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of the content."

I think that speaks to our hopes and aspirations for reliable sources, but this sentence tacks on some irrelevant things. Self-published material is characterized by the lack of editorial control. Hopefully, usually, that editorial control will involve independent reviewers and even professional standards; hopefully it will involve something validating whether the content is accurate. But a source can be validated for reliability by independent reviewers and still be self-published. WhatamIdoing (talk) 22:36, 31 October 2024 (UTC)

USESPS seems a mess, and doesn't seem to match commonity pratice. If we're relying on an essay that says a self-published source have certain features, but then has to specifically exclude news organisation because they have all the features mentioned, then it's obviously going to cause confusion.

Sources such as SLPC and SBM are accepted as not being self-published, and there are even RFC with that result. Interpreting such sources as self-published doesn't seem to match up with the wording of WP:SPS, and it doesn't appear to line up with the ideas expressed in Self-publishing. Particularly that self-publishing involves the author of the work publishing the work themselves at their own cost. -- LCU ActivelyDisinterested «@» °∆t° 23:49, 31 October 2024 (UTC)

So when a dictionary says that self-publishing means to "publish (a book etc.) oneself rather than through a publishing house", do you find that their exclusion of publishing houses (a group that includes newspapers) to be confusing? WhatamIdoing (talk) 02:05, 1 November 2024 (UTC)