Wikipedia talk:Reliable sources

Reliable sources quiz

I've expanded the reliable sources quiz for new editors a little more, but it still needs a few more examples before it'll be ready to launch, and I keep having an unexpectedly hard time finding good ones. Could anyone help? {{u|Sdkb}} ^talk 20:31, 6 September 2020 (UTC)

Headlines

The RfC at WP:VPP § RfC: Reliability of headlines has been closed with consensus to include a new subsection to state that headlines are unreliable. In the workshop, Awilley's suggestion garnered the most support:

News headlines are not a reliable source if the information in the headline is not explicitly supported in the body of the source. Headlines are written to quickly and briefly grab readers' attention, and may be overstated or lack context. They are often written by copy editors instead of the researchers and journalists who wrote the articles.

Thryduulf suggested "explicitly noting that headlines are a reliable source for themselves when they are the subject", which would allow headlines to be used along the lines of WP:ABOUTSELF. Skdb Sdkb recommended using the phrase "generally reliable" instead of the word "reliable" to afford some flexibility.

What are your thoughts on these suggestions, and do you have any other proposals for the final text? — Newslinger talk 17:52, 18 September 2020 (UTC)

I would recommend unsplitting that split infinitive in this specific version (without supporting/opposing the suggested text). I keep reading "to" as "too" (which makes sense until you hit the word "grab" which is why I'm tripping). --Izno (talk) 18:48, 18 September 2020 (UTC)

If we do qualify with Skdb's "generally", should we also note special care is required in particular contexts (BLP, medical claims) where sensitivity and accuracy is deemed paramount? 03:43, 21 September 2020 (UTC)Humanengr (talk)

I do not really see a need for the qualification. What is the use case? PackMecEng (talk) 03:50, 21 September 2020 (UTC)

Per the close "there is broad consensus here that we should avoid citing information in a headline that is not supported by the body of the article. … There was also some concern that the proposal is over-broad and does not allow for editorial discretion in edge cases. Sdkb suggested the wording generally unreliable instead of simply unreliable, which I don't think anyone directly objected to. Inclusion of 'general' might satisfy 'some' but shouldn't be included without qualification. Did you want particular justification for the med and bio cases I offered as compromise? Humanengr (talk) 04:31, 21 September 2020 (UTC)

Sure, what are some examples in BLP & Med claims where you would cite a headline that is not supported by the body of the source? PackMecEng (talk) 04:42, 21 September 2020 (UTC)

Re med claims, WP:NOABSTRACT says that even re abstracts … when it comes to actually writing a Wikipedia article, it is misleading to give a full citation for a source after reading only its abstract; the abstract necessarily presents a stripped-down version of the conclusions and omits the background that can be crucial for understanding exactly what the source says, and may not represent the article's actual conclusions. … From the article cited there: Even in large-circulation general medical journals, data in abstracts were commonly inconsistency[sic] with full reports. Those confirm — for this high consequence area — both the RfC's view of headlines and WP:NEWSORG's even the most reputable reporting sometimes contains errors.. Re bios, I note WP:BLPSOURCE's caution re Contentious material … that is … poorly sourced; and WP:PUBLICFIGURE: If an allegation or incident is … well-documented … Humanengr (talk) 06:02, 21 September 2020 (UTC)

Those appear to be reasons not to use headlines. I agree with that which is why above I was saying that the qualification "generally" is not needed. If you have generally in there it means sometimes there are cases where a headline would be useable. PackMecEng (talk) 06:08, 21 September 2020 (UTC)

Ohhh … I mistook your "qualification" to refer to my qualification of Sdbk's 'generally' (as your comment was indented under mine). I agree, 'generally' is not needed. Humanengr (talk) 06:21, 21 September 2020 (UTC)

Same team! PackMecEng (talk) 06:27, 21 September 2020 (UTC)

The headline seems undoubtedly reliable for a sentence like, 'X [journalist/author/critic], wrote [headline].' or 'The Chicago Tribune published an article, [headline]. - Alanscottwalker (talk) 12:47, 21 September 2020 (UTC)

That would fall under the aboutself situation that Thryduulf described wouldn't it? PackMecEng (talk) 13:14, 21 September 2020 (UTC)

• @Newslinger: Given the discussion above, I would be bold and install your proposed text plus Thryduulf's suggestion, leaving out Skdb's suggestion for now. Mz7 (talk) 19:50, 23 September 2020 (UTC)
• Oops, I'm just coming across this now (the ping above failed; I'm Sdkb, not Skdb). The reason I'd prefer "generally" is largely per our fifth pillar norm that we don't make hard rules except when absolutely necessary, but also that the main argument for headlines not being reliable, that they're generally not written by the same author and generally don't receive the same level of scrutiny, is not something that applies universally. It's not hard to come up with a hypothetical: next week, the RSP Greenlit Times announces that they're adopting a new policy in which all writers write their own headlines and all headlines are put through three rounds of fact checking before publication. That may sound a little far-fetched, but the idea of The New Yorker putting out a "how we write our headlines" insider story that makes it clear every headline goes through TNY's famous fact-checking process is not. If we haven't built flexibility into our rule, we won't be able to handle something like that. As I said in the main discussion, we don't want to deprecate New Yorker headlines more strongly than National Enquirer body text, which is what we'd be doing, since even National Enquirer has "generally". {{u|Sdkb}} ^talk 05:52, 24 September 2020 (UTC)

Addition of ‘generally’ here would not be in isolation; it would be in the context of rankings at WP:RSP and would allow a ‘generally reliable’ ranking at RSP to generally (!) trump consensus re headlines — as in But here we’re talking about a generally reliable source with a reputation for fact checking and accuracy. Humanengr (talk) 19:59, 25 September 2020 (UTC)

I've fixed the misspelling, sorry about that. My comment used {{np}} to avoid pinging anyone, since I did not want to invite only those who were mentioned in the comment. I did publish a short notification of this discussion at WP:VPP § Continuation of workshop. — Newslinger talk 16:01, 26 September 2020 (UTC)

• I don’t see a need to hedge things with a “generally”. All we need to say is: “Headlines are not considered reliable except in WP:ABOUTSELF situations”. That makes it clear and unambiguous. The ONLY situation in which it is appropriate to cite ANY headline is in an ABOUTSELF situation. Period, end of statement. Blueboar (talk) 20:27, 25 September 2020 (UTC)

  Doesn't this present a problem for ordinary citation construction? We must always cite the title of the article. If the title itself is not considered reliable, why should it be included prominently in the citation, where it will be seen in the footnotes, and the wikitext being edited? Should article titles come tagged with a little disclaimer: "warning: not actually reliable in and of itself."? Elizium23 (talk) 20:30, 25 September 2020 (UTC)

Ok... (sigh) I suppose I should not have used the word “cite” to explain why I prefer blunt and unambiguous language (I forgot that “cite” can get confused with “citation”).

So let me amend, using slightly different wording... I think we should say: “The text of a news Headline is not considered reliable for verifying information, except in WP:ABOUTSELF situations”.

I say this because we want to make it unambiguously clear that we can not reliably verify information by pointing to the text of a headline - the ONLY thing that a headline verifies is the text of the headline itself.

This is distinct from using the headline as a “title” when formulating a citation. Obviously, we need to do that. Blueboar (talk) 21:06, 25 September 2020 (UTC)

This is still inadequately rigid. It does nothing to address the RSP Greenlit Times issue I posed above. {{u|Sdkb}} ^talk 00:40, 3 October 2020 (UTC)

Sdkb, if a periodical announced a new policy in which all writers write their own headlines and all headlines are put through three rounds of fact checking before publication, start an RFC carving out a headline exception for that agency. Schazjmd (talk) 01:08, 3 October 2020 (UTC)

Schazjmd, due to WP:CONLEVEL, it wouldn't be possible to carve out an exception unless the discussion was at least as broad as the very broad VPP discussion that's brought us here. The point is that we should build flexibility into the rule now, while we're creating it, so that we do not later have to amend it to address flaws as they reveal themselves. {{u|Sdkb}} ^talk 01:18, 3 October 2020 (UTC)

@Sdkb... even in your RSP Greenlit Times scenario, a headline would not sufficiently verify non-aboutself information. We need to point readers to the full text of the news article so they can see the context in which the information is presented. Blueboar (talk) 12:54, 4 October 2020 (UTC)

References

• It has been a while since new input. I have implemented the suggested wording by Newslinger here. PackMecEng (talk) 22:06, 31 October 2020 (UTC)

  PackMecEng, I do not consider my concerns from above to have been adequately addressed, and given the close of the Village pump discussion and the fact that this is a more limited forum, I don't think there's sufficient consensus that no caveat is needed to implement that wording. I'm therefore reverting for now. Perhaps we can issue invitations to some relevant pages to garner additional input here. {{u|Sdkb}} ^talk 22:49, 31 October 2020 (UTC)

  From what I can tell you are the only one disagreeing. Whereas myself, Newslinger, Blueboar, Elizium23, Izno, Schazjmd, and Humanengr seem to be going in the same direction. It is of course open to tweaks, but no need to revert. PackMecEng (talk) 22:55, 31 October 2020 (UTC)

  I'd count myself as going in the same direction as well, just with the tweak of adding generally. You are putting words in the mouth of several of the users you list, as e.g. Izno did not weigh in on whether or not to use that word, and Elizium23 weighed in only with a skeptical query to Blueboar that seems to indicate if anything more a leaning toward my view. But this is a discussion, not a vote, so the pertinent question is arguments, not numbers. My argument about generally was sufficiently strong that Mz7 mentioned it in their close, and in this follow-up discussion here, I do not see anyone convincingly addressing the RSP Greenlit Times point I raised above. {{u|Sdkb}} ^talk 23:56, 31 October 2020 (UTC)

  Sdkb, I did mention your suggestion in the close as a way to potentially get the ball rolling on implementation, but it should not be considered an endorsement. It seems clear from reading this discussion that there isn't consensus in support of generally, so to implement the RfC result after more than a month of delay, I think we should go with PackMecEng's edit. For what it's worth, I do not think this is a big deal; if the hypothetical situation you pose ever does occur, then we can revisit this issue then. At the end of the day, this is merely a guideline and WP:IAR is policy. Mz7 (talk) 00:05, 1 November 2020 (UTC)

  Mz7, fair enough. I don't want to clog the gears of the inevitable so I'll stand aside if no one else here is concerned by the stronger wording. I tend to find that IAR is embraced far more in name than in practice, but should we ever come across a situation like the one I laid out above, it will at least help to have a reference to IAR on record in the discussion that led to the guideline. {{u|Sdkb}} ^talk 06:11, 1 November 2020 (UTC)

  I support PackMecEng's edit implementing Newslinger's suggested wording. Multiple editors agree that it reflects the findings of WP:VPP § RfC: Reliability of headlines, and this discussion has been lingering for six weeks. Schazjmd (talk) 23:03, 31 October 2020 (UTC)

  I support PackMecEng's edit implementing Newslinger's suggested wording; and agree that it reflects the the result of WP:VPP § RfC: Reliability of headlines. - Ryk72 ^talk 23:20, 31 October 2020 (UTC)

  The proposed guidance has already been the standard practice on the reliable sources noticeboard for some time, as shown in the past noticeboard discussions listed in Wikipedia:Village pump (policy)/Archive 160 § Discussion (Headlines). The text above received the strongest support in the village pump RfC, and I agree that it is time to add it into the guideline. If any issues arise, this text can always be amended later. — Newslinger talk 00:01, 1 November 2020 (UTC)

  I would like to reiterate my non-opposition request for reordering the second sentence so that I don't trip over the words. That's all I have to say. --Izno (talk) 02:38, 1 November 2020 (UTC)

  Izno, I support that suggestion for clarity. {{u|Sdkb}} ^talk 06:14, 1 November 2020 (UTC)

   Done [1]. I've gone ahead and readded the text with Izno's suggestion to unsplit that infinitive. Mz7 (talk) 06:46, 1 November 2020 (UTC)

  Does this reordering work:

  Headlines are written to grab readers' attention quickly and briefly, and may be overstated or lack context.

  ? Humanengr (talk) 11:40, 1 November 2020 (UTC)

Book publisher reliability list

Do we have an WP:RSP equivalent? When I'm faced with print books, I'm having hard time determining if the publisher is good for fact checking (and if then for what fields) and what publishers are useless for notability establishment purposes. Graywalls (talk) 20:19, 20 September 2020 (UTC)

How are the better sort of fact checkers getting on with the Bible and the Quran?PatrickGuinness (talk) 13:38, 19 October 2020 (UTC)

Is a gravestone a reliable source?

I'm trying to find a RS for Edward G. Faile being buried at Saint Paul's Church. Surprisingly, I can't find this mentioned anywhere, but I do know he was buried there because I stood in front of his gravestone and took a photo of it. How can I cite this? -- RoySmith (talk) 15:21, 22 September 2020 (UTC)

Well, duh, it turns out I found a RS, but I'm still curious about the basic question; can a photo of a gravestone be used as a RS? -- RoySmith (talk) 15:28, 22 September 2020 (UTC)

No. A reliable source requires a publisher with a reputation for fact-checking and accuracy. Gravestones are typically put up by families of the deceased and are, therefore, self-published sources (not about the persons who published it, the survivors, but about a third person, the deceased). But perhaps a bigger problem than sourcing is that this is prohibited original research to assert that this Edward G. Faile is the same Edward G. Faile as Edward G. Faile. You might, for example, argue that the dates of birth and death match those established for the article's Faile by reliable sources, but that is prohibited synthesis, taking information from one source and combining it with another source to come to a conclusion stated by neither source. Regards, TransporterMan (TALK) 15:37, 22 September 2020 (UTC) PS: Let me amend that slightly. It's not a RS for establishing that he's buried there. But the question is always, "Reliable for what?" Let's say that you had, through other RS established this to be his grave and had also established through a RS that he, himself, had designed the headstone and wrote the inscription (thus fitting into the self-published source exceptions). This headstone could, in those circumstances, be a RS for his dates of birth and death date of birth. — TransporterMan (TALK) 15:49, 22 September 2020 (UTC)

He can be a reliable self published source for the date of his own death? Hyperbolick (talk) 17:00, 22 September 2020 (UTC)

<Self-trout.> Fixed, thanks. Best regards, TransporterMan (TALK) 19:15, 23 September 2020 (UTC)

TransporterMan, lol, self-trout indeed! {{u|Sdkb}} ^talk 05:58, 24 September 2020 (UTC)

There are quite a few archive search results on WP:RSN regarding graves and gravestones (especially WP:RSP#Find a Grave). --Izno (talk) 16:50, 22 September 2020 (UTC)

A grave stone is a reliable PRIMARY source for saying what the text that appears carved on the stone is. As with all primary sources, I would be cautious about using it for anything else. Blueboar (talk) 11:03, 24 September 2020 (UTC)

I agree with only being reliable for showing what it says on the face. You can take a look at this and it can get pretty confusing for typical people and possibly connect it to the wrong person. https://schmidtgen.com/wordpress/2013/10/20/how-to-use-jr-sr-ii-iii-etc-with-cartoons/ Since published items often don't show people's entire name with the prefix and suffixes, it's quite plausible to link it to the wrong person in this naming situation or people who have extremely common names. Graywalls (talk) 06:03, 3 October 2020 (UTC)

To add to what others have mentioned, gravestones are primary sources that may not always agree with published sources. Dates and spellings of names may be different from the consensus view in the published record (and a headstone by itself doesn't prove a body was buried beneath it, merely that the headstone exists). Similarly, dates and spellings in census forms and marriage certificates may not always agree with published information (many people obscure their true age, and a slip of a census taker's pen can mislead). When there is a discrepancy, deciding which records are "true" is not for Wikipedians to settle in articles, but for scholars to analyze and publish elsewhere with their opinions on likelihood. --Animalparty! (talk) 20:32, 11 October 2020 (UTC)

Completely unreliable....can pay to add whatever you want to a grave marker....famous example...Scott Wilson (2016). Resting Places: The Burial Sites of More Than 14,000 Famous Persons, 3d ed. McFarland. p. 165. ISBN 978-1-4766-2599-7. Crawford, Joan (Lucille LeSueur, March 23, 1904 – May 10, 1977) San Antonio born film star.... Her ashes were placed in the vault beside the coffin of her husband, with the crypt listing her birth year as 1908..--Moxy 🍁 20:46, 11 October 2020 (UTC)

Douban

Moved to Wikipedia:Reliable sources/Noticeboard

 – Bsherr (talk) 23:29, 25 September 2020 (UTC)

source related content dispute. RSN or NPOV/N?

In the article squatting, I'm in a disagreement about contents that I believe is to be undue using sources which I believe to be comparable to Indymedia.org (which is a source that is crossed out red on WP:RSP as well as adding these sources alongside already present reliable source. I have little doubt that these sites are disreputable. I think introducing contents from such articles into article is inappropriate. I'm not really uncertain about the reliability of the source, but the other editor do not agree on it. Is this a source related matter for RSN or is it a due weight issue for NPOV/N instead? My contention is over the presence of sources, as well as contents based on sources 1,3,4 and 5 in the example below

Example

The building which once housed Neary's Hotel on Parnell St in Dublin's north inner city, was occupied in 2015 and renamed The Barricade Inn by squatters.

In protest, squatters in Amsterdam had occupied a former fire department the week before the law began (returning it to the owners control on 30 September) and a riot occurred on 1 October when the police blocked a protest and led a horse charge upon it. In Nijmegen (on 2 October), there was also a riot.

Graywalls (talk) 04:33, 3 October 2020 (UTC)

References

​The article from The Irish Times does not link to (or even mention) any of the questionable sources. Since these websites are all used as secondary sources, this is purely a question of reliability and would be best suited for the reliable sources noticeboard. — Newslinger talk 13:04, 4 October 2020 (UTC)

Showbiz Cheat Sheet

I was looking up Clark Middleton, and saw something from Showbiz Cheat Sheet.

https://www.cheatsheet.com/entertainment/the-blacklist-star-clark-middleton-on-the-condition-hes-lived-with-since-the-age-of-4.html/

Is this a reliable source? MikaelaArsenault (talk) 00:50, 6 October 2020 (UTC)

Article that blend commentary and reporting of facts

Are editors aware of any discussions how best to handle articles that blend factual reporting and commentary? I found one relevant prior discussion [[2]] but I was curious if there was a larger one. I think this was one of the big criticisms of Fox News in that it would report the same objective facts as other sources but their subjective commentary was considered questionable. Outside of the many Fox News discussions is anyone aware of a more general discussion? If not is here or RSN a better place to start one? Springee (talk) 13:52, 12 October 2020 (UTC)

This is probably something that would need to be decided on a case-by-case basis. -- Calidum 14:48, 12 October 2020 (UTC)

Official website of writers-critics

Can we cite an official website of writers-critics. I want to cite two websites: https://www.deepagahlot.com/ of Deepa Gahlot & http://bhawanasomaaya.com/ of Bhawana Somaaya. Thanks. --Gazal world (talk) 18:04, 12 October 2020 (UTC)

It depends on the context. For basic facts like birthday, family, where they hail from, you can use self published source especially for within the article about the subject themselves but be sure to read the criteria at WP:ABOUTSELF. Most certainly not if it's for promotional puff. Graywalls (talk) 01:25, 16 October 2020 (UTC)

@Graywalls: These both person are well known theatre and film critics. I want to cite the critical reviews of some plays which they have posted on their website. Can I cite them ? --Gazal world (talk) 15:43, 19 October 2020 (UTC)

Well known critics can probably be cited under the “Expert exemption” clause of WP:SELF, with in-text attribution (and phrased as opinion, not presented as fact). Blueboar (talk) 16:07, 19 October 2020 (UTC)

@Graywalls & @Blueboar: Thanks for answering my question. --Gazal world (talk) 20:41, 20 October 2020 (UTC)