Wikipedia talk:Verifiability

The project page associated with this talk page is an official policy on Wikipedia. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing, and don't panic.

Exceptional claims require exceptional sources

I question the first sentence, "Any exceptional claim requires multiple high-quality sources." It is cited to various statements that exceptional evidence or proof is required. To me that means we need a good reliable source from a recent, relevant high-quality secondary publication that explains where it obtained its information. That allows us to investigate whether the source has accurately reported the information and we can always challenge it with other sources that contradict it.

The problem with the current wording is that it is an invitation to source bomb, to provide numerous sources, none of which are conclusive, and require anyone challenging it to read through each of them. In fact when a news source reports news inaccurately, it is often picked up on wire services and reported in hundreds of other publications.

I suggest instead that it read "Any exceptional claim requires a high quality source."

TFD (talk) 22:38, 28 November 2015 (UTC)

Can you give an actual example of your hypothesized scenario actually happening? EEng (talk) 22:59, 28 November 2015 (UTC)

The current wording is fine. Any single source can be in error, regardless of its quality. Therefore, if a single source were to report, for example, that a cold fusion experiment demonstrated that a magic box could produce 1 MW net power output for 24 hours the article would be cautious about how that result was reported, even if the source would normally be regarded as high quality. If it was a recent report, the concern would be that a mistake might have been made (if not fraud—a lot of money is made when magic devices get good reports), and we should wait for confirmation by other sources before reporting it as a known fact. If it was an old report, the concern would be that there should have been other experiments to replicate the result, and the fact they cannot be found suggests the single finding was not significant. Johnuniq (talk) 01:37, 29 November 2015 (UTC)

Concur with Johnuniq that the current wording is best, for the reasons he stated. Regards, TransporterMan (TALK) 17:42, 29 November 2015 (UTC)

Me too, especially since the problem the proposed change claims to solve appears to by hypothetical. EEng (talk) 02:44, 30 November 2015 (UTC)

Not to mention if someone is determined enough they will invent some loophole in whatever rule we come up with. I think this should only be changed if the rule is being exploited on a consistent basis or there is genuine confusion about it.--72.0.200.133 (talk) 18:31, 30 November 2015 (UTC)

It appears that TFD does not appreciate the problem with exceptional claims, nor the proper application of WP:WEIGHT. A claim is exceptional when it is contrary to the majority viewpoint, and if there is only "a" source in support then (to quote Jimbo) "it doesn't belong in Wikipedia, except perhaps in some ancilliary article."

TFD may also be unaware that high-quality sources (i.e., journals) often publish work that makes exceptional claims. Publication in itself does not make a claim credible or grant approval, it only exposes the claim for consideration. "In fact", the wire services, and the popular media generally, are more likely to light-up on a report that "Einstein got it wrong" then on all those boring, hard-to-understand reports that Einstein still has it right. Determination of scientific consensus can be challenging because there is rarely any headline when that consensus comes together, and because WP editors are often unfamiliar with what secondary sources are truly reliable. I believe this is also true in other fields. ~ J. Johnson (JJ) (talk) 00:11, 18 December 2015 (UTC)

Talk about scholarly citations on Wikipedia

The WMF analytics team is presenting a pair of talks on the subject of m:Wikipedia as the front matter to all research this Friday December 4, 2015 (two days from now) at 12:00 noon PST. They will talk about unique identifiers and scholarly citations in Wikipedia. One of the speakers is from CrossRef, and his subject is primarily about how Wikipedia citations bring non-scientists into contact with the scientific literature. The talk will be broadcast on YouTube, so that anyone can watch it. There is more information on Meta. Please join if you're interested. WhatamIdoing (talk) 19:59, 2 December 2015 (UTC)

'Verifiability' is poorly defined

The lead says: verifiability means that anyone using the encyclopedia can check that the information comes from a reliable source. What does verifiability mean here in a practical sense? That

1. a link to a reliable source is provided in the article, or merely that
2. such a source must exist somewhere, or
3. something else.

Now,

• If the first option is correct, which I think it is, the quoted statement contradicts what follows. Combining the statement with All material in Wikipedia mainspace, [...], must be verifiable. would then give All material in Wikipedia mainspace, [...], must link to a reliable source. But then it says: All quotations, and any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation that directly supports the material. Now it suddenly says that only a subset of all content needs a source (mathematically that is not a contradiction, but in normal language it is). Also: Any material that needs a source but does not have one may be removed implies that not all material needs a source.
• If the second option is correct, we wouldn't even need to include the existing source to the article, unless other guidelines require that, such as BLP. I think that is not the correct interpretation of verifiability.

I tried to work my way through the archives where related issues have been discussed many times, but I can't find a clear answer to my question. It would be really helpful to editors if we can improve the definition of the title word of this guideline as it is one of the core content policies, and if needed we need to solve the contradiction. Gap9551 (talk) 21:32, 10 December 2015 (UTC)

All material must be verifiable, but only certain material requires an inline citation (which need not be a hyperlink; it could be a citation to a paper book or the like). Some material is so uncontentious and well known that the reader who hasn't already verified the claim just through the process of living can easily verify the claim by looking it up in innumerable readily available sources. For example, I don't need a book to tell me Paris is the capital of France, I've been there. If I don't happen to know who the senators are from Idaho, it will be very easy for me to find a reliable source. Jc3s5h (talk) 21:56, 10 December 2015 (UTC)

• It means all three of the following:-

1. All material in the encyclopaedia must be verifiable, in that a reliable source for it must exist somewhere; and
2. All material in the encyclopaedia which has been challenged must be verified, in that a reliable source for it must be provided by means of an inline citation, or else the material can be removed; and
3. All material in the encyclopaedia which is likely to be challenged must be verified, in that a reliable source for it must be provided by means of an inline citation, or else the material can be removed.

Hope this helps—S Marshall T/C

Thanks both, that does help. So my second option is correct after all, not the first (in the general sense, only in specific cases), which wasn't obvious from the current definition verifiability means that anyone using the encyclopedia can check that the information comes from a reliable source. Unless I'm the only person who finds the current phrasing of the lead unclear, would it worth explicitly stating a version of your clarifications in the project page, in particular that non-contentious material need not necessarily have a source listed in the article provided it exists somewhere (but that it may still be good practise to add one)? Gap9551 (talk) 22:31, 10 December 2015 (UTC)

stronger wording?

I've been editing WP since 2006, and I refer to WP:V a lot. I really like this sentence about WP:

Its content is determined by previously published information rather than the beliefs or experiences of its editors.

Plenty of editors think that they can just edit WP in ways that make sense to them. That's what I thought when I started years ago. I have edited a lot of controversial pages, such as Jesus, Historical Jesus, Purgatory, and Evolutionary Psychology. Getting editors to be true to published sources is often a challenge. Based on my experiences, I have a simple suggestion to make this line stronger. Change it to:

Its content is determined by previously published information rather than the opinions or experiences of its editors.

Here's my reasoning.

1. Editors freely share their opinions but not their beliefs. Someone who believes in creationism might oppose a reference to evolution, but they will do so by stating an opinion rather than by referencing their own beliefs. They'll say, "This reference to evolution is outside the scope of this article," rather than "I'm a creationist and want to remove this reference to evolution."
2. The term "opinions" is broader. Sometimes editors have opinions that are not connected to anything that would normally be called a "belief." For example, I often hear that a page should describe all sides of an issue without favoring one side over another. That's an opinion but I wouldn't call it a belief.
3. Editors are not supposed to speculate about the personal beliefs of other editors. If I say, "WP isn't based on your beliefs," that requires me to infer what their beliefs are and how they affect their editing. But if I say, "WP isn't based on your opinions," that's legitimate because their opinions are right there on the Talk page.

WP:V is a touchstone for good WP editing, and I think this change will make this important sentence even stronger than it already is. Thanks. Jonathan Tweet (talk) 20:59, 17 December 2015 (UTC)

• Week Opposed - I think you are overthinking it... opinions are beliefs... no one would state an opinion that they don't believe. Blueboar (talk) 13:35, 18 December 2015 (UTC)

• Oppose. Even though I do sometimes state opinions that I do not entirely believe in (there are some subtle differences), yet I find JT's argument in favor underwhelming, and even too much of a stretch. Overthought, indeed. ~ J. Johnson (JJ) (talk) 21:20, 18 December 2015 (UTC)