Wikipedia talk:Reliable sources/Archive 62

This is an archive of past discussions about Wikipedia:Reliable sources. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 55

←

Archive 60

Master's theses should also be ruled "reliable" if they are the only form of academic scholarship talking about a subject

Re: "Masters dissertations and theses are considered reliable only if they can be shown to have had significant scholarly influence."

Historically I've also included master's theses if they are the only form of academic scholarship talking about a subject (as in no PhD theses discuss the topic). This is why I cited a master's thesis - one that was cited by an academic book (The Japanese in Latin America) - for Japanese community of Mexico City (the same thesis is critical in Liceo Mexicano Japones) - PhD theses in the topic dont exist (as of now). Also the master's thesis is from 1983, when master's degrees were far less common and therefore carried more weight.

Many Wikipedia topics don't have a whole lot of "reliable sources" to choose from. The sourcing reqs of course should be more stringent for those that do (say Barack Obama) but for ones that don't: I think the guideline should also state master's theses are acceptable if they are the only form of academic scholarship talking about a subject, with preference perhaps for older theses (prior to the last two decades or so).

P.S. I had even included a further reading to an honor's bachelors thesis from the 1960s (higher education degrees were even less common back then) in an article about some Russian novel... I think it may have been one of the few English sources about it? WhisperToMe (talk) 06:36, 18 August 2019 (UTC)

Current guidance is good - Masters theses (which are often awful) are not a good way to source the "accepted knowledge" that Wikipedia is meant to be reflecting. If a topic is such that only such sourcing is available, that would indicate a WP:N problem. Alexbrn (talk) 09:17, 18 August 2019 (UTC)

Re: Masters theses being awful, wouldn't this depend on the discipline and/or the time period? I'm trying to find some Chronicle of Higher Education articles on the matter but I thought in earlier time periods that master's degrees are more esteemed, though recent periods would not. WhisperToMe (talk) 14:22, 18 August 2019 (UTC)

No way. If the content of a thesis is worthy of referencing, it will be published elsewhere. You should never rely on any student thesis as a reliable source. —SmokeyJoe (talk) 09:25, 18 August 2019 (UTC)
- Wikipedia:Reliable_sources#Scholarship does state "Completed dissertations or theses written as part of the requirements for a doctorate, and which are publicly available (most via interlibrary loan or from Proquest), can be used but care should be exercised, as they are often, in part, primary sources." So PhDs can be used at times. WhisperToMe (talk) 14:22, 18 August 2019 (UTC)

If a masters thesis is cited by an academic book, you can cite the academic book for that particular content. However, the fact that an academic has relied on the thesis for one bit of information doesn't mean that we can rely on it for other bits. Consensus can be sought for exceptions but the default rule that masters theses are below the bar is a good one. Zero^talk 10:05, 18 August 2019 (UTC)

I'd be happy to seek consensus regarding a particular article, though the usage of the thesis was never challenged by another user on ENwiki (admittedly it is a low-traffic article). On ESwiki there were questions regarding admissability but the source was admitted. WhisperToMe (talk) 14:22, 18 August 2019 (UTC)

We should be very cautious about them - in practice the temptation to use them (by us, and sometimes by academics) comes because they are accessible online, where better sources are not (or are in a different language). It's not common that one really makes an original contribution in most fields, and if it does it's likely to be rehashed as a published paper. Johnbod (talk) 14:45, 18 August 2019 (UTC)

Yes, that is often a motive. In this case I wanted to get a licentiate thesis from Switzerland, which seems to be in book form anyway, to support an article on the Japanese school of Geneva (the Mexico thesis I mentioned earlier is also only in book form, and I asked for it because Eswiki editors, fully aware it was cited in a published book, wanted to see the parts of the original thesis) WhisperToMe (talk) 16:02, 18 August 2019 (UTC)

I don’t think a Master’s thesis can be considered reliable. They CAN however be used for finding sourced that ARE reliable (ie bibliography). Also... If there is ONLY one source that can be used to support something stated in one of our articles, I would have serious UNDUE questions... especially if that one source is a Masters thesis (I would have problems even if we were talking a PhD dissertation). Blueboar (talk) 20:49, 18 August 2019 (UTC)
- I went back and looked at some of the articles in which I used a master's thesis that was already cited by a scholarly published book. In Japanese community of Mexico City and Liceo Mexicano Japonés I cited Masterson (The Japanese in Latin America), the work citing the master's thesis by Watanabe, alongside from the thesis itself. I felt the thesis simply filled out/complemented the material cited to Masterson rather than being undue for the topic: an analysis of the Masterson section along with the portions of Watanabe's in my posession would show that Masterson's clearly getting information from Watanabe. However I can see a thesis being undue if it advances a new conclusion/point of view not stated in any published source, including any sources citing it. WhisperToMe (talk) 23:25, 18 August 2019 (UTC)
Master and PhD thesis and doctorates do not have the same rigor of peer review in general as journal publications. (However, many thesis and doctorates will include published papers that have gone through peer review). These should not be used as RSes just because they seem "academic". We need that attention that the current language offers. --Masem (t) 04:25, 19 August 2019 (UTC)
Master's theses do not have sufficient oversight to be considered reliable; if a better source cannot be found for a specific claim, then there is a serious concern of WP:REDFLAG or at least WP:UNDUE. Jayjg ^(talk) 16:28, 19 August 2019 (UTC)
- After consideration, if it's something that's particularly unusual/against mainstream scholarship, definitely agree that it's best to not include it from a master's thesis. WhisperToMe (talk) 00:21, 20 August 2019 (UTC)

In the humanities--especially history, MA theses are important, are evaluated by teams of serious scholars, and have university endorsement. Students who do a poor job will be risking their careers--they can lose their fellowships and damage their long-term teaching-job possibilities .--so this is indeed serious work typically much better quality than media reporting--graduate students can and do take months to get it right, while reporters have a matter of hours to file a report. My experience tells me that on the whole MA theses in history & humanities meet & usually exceed Wikipedia standards. Rjensen (talk) 01:41, 20 August 2019 (UTC)

The Yoshida thesis (2001) was from McGill University and was in anthropology, and the Watanabe thesis (1983) was from California State University and also in anthropology; therefore both are under humanities. The Harayama-Kadokura mémoire de licence (1989) from the University of Geneva is in education, which I think is social sciences (someone told me that the French-speaking Swiss consider a Mémoire de licence equivalent to a U.S. bachelors, which is a bit shocking since the document is 118 pages). WhisperToMe (talk) 04:24, 20 August 2019 (UTC)

People might (or might not) be interested that I have recently said I think it is ok to use a master's thesis as an RS at Talk:Our Lady of Vladimir/GA1, where I am the reviewer (Ok, it should be Virgin of Vladimir). This is because, despite the icon being incredibly famous, the only monograph on it I can find was published in Russian, and then in English in Prague in 1928 (which seems to have contributed to the author getting shot back in Mother Russia). There is some interesting stuff on the post-Yeltzin history, but mainly the thesis repeats and updates nitty-gritty stuff on the many restorations and scientific examinations etc, which doesn't appear to be available in English elsewhere, except the 1928 book. Thoughts welcome. Johnbod (talk) 15:28, 20 August 2019 (UTC)

I do feel if the master's thesis, so long as it's credibly reviewed/had credible oversight, is not particularly unusual/against normal reporting and scholarship, and is one of few sources, or one of few sources in English (or even say, French, which is similar to English), then we should be a bit relaxed and allow it. WhisperToMe (talk) 23:34, 20 August 2019 (UTC)

If it is not particularly unusual/against normal reporting and scholarship, then why do we need it? Why not just use the few sources/few sources in English instead? Jayjg ^(talk) 16:43, 21 August 2019 (UTC)

Because it has additional information (often basic facts, demographics, additional history - essentially how I used Yoshida and Watanabe) that may not be recorded in other secondary sources (though it may be recorded in primary sources). I am already using the other "few sources" but I want more information put in, if possible. For example Masterson, in his scholarly academic book, repeated several points of information that Watanabe stated, but there are additional points in Watanabe's thesis, in the same paragraphs which Masterson used, which were not included in Masterson's book. WhisperToMe (talk) 23:39, 21 August 2019 (UTC)

Basic information can be sourced from primary sources, if it is uncontroversial. And you still haven't solved the issue of it being WP:UNDUE and WP:NOTEVERYTHING; if reliable secondary sources don't care to note something, then it's unlikely Wikipedia should take note of it either. Jayjg ^(talk) 16:06, 22 August 2019 (UTC)

This seems rather beside the point; not every editor has access to a top-quality access academic library. It is not at all uncommon for fairly basic factual information on medieval artworks (full dimensions, restorations etc) only to be set out large works in German from a century ago, and probably printed in under one thousand copies. It is rather unreasonable to expect these can be accessed. The Worldcat record for the book I mentioned above shows only 51 copies worldwide: 4 copies in libraries in the UK (1 public library), 30 in US (1 public library), and others in Europe (mostly Germany, with several public libraries). Of course if someone later can access these later they can be added, or used to replace a thesis. Johnbod (talk) 18:06, 22 August 2019 (UTC)

The citing of the material in that section in an academic book means its not UNDUE. Its instead an expansion of DUE material; when Masterson cited from Watanabe, the relevant section became DUE because Masterson said so, and Masterson's conclusions are not in conflict with other sources.
IMO NOTEVERYTHING is best applied to "dense" articles where there are plenty of sources in English; it's impossible to cram everything into Barack Obama. The Mexico City and Montreal Japanese schools have sources in Japanese, which most of us can't read. Also Watanabe covers material that they may not have. I second Johnbod's point about accessibility of information.
WhisperToMe (talk) 20:59, 22 August 2019 (UTC)

The fact that an academic book decided to source some information, and ignore others, means the ignored information is WP:UNDUE. WP:NOTEVERYTHING applies to every article on Wikipedia. Regarding Johnbod's point, if you can translate the material from Japanese, then use the Japanese source. If you can access the German library, then use the source in that library. And if you can't read or access the information that supposedly exists in foreign languages or inaccessible reliable sources, then you have no way of knowing if the material is there, and what it says, and an unreliable source (such as a Master's thesis) doesn't help in that regard since it is, well, unreliable. Jayjg ^(talk) 18:38, 23 August 2019 (UTC)

To claim that basic information such as the restoration history of an incredibly well-known artwork breaches WP:UNDUE or WP:NOTEVERYTHING is simply absurd. I can't be bothered to continue the argument, but there doesn't seem to be much agreement with you. Johnbod (talk) 22:23, 23 August 2019 (UTC)

OK, so after all this I still feel a bit at sea. I know very little about how colleges and universities work, so let me make up a scenario and ask some basic questions.

Let's say that here I have here a master's thesis, Morphology of Saco Bay Coastal Inlets. It was written by Violet Anne Tower in 1952 and is held by the University of Southern Maine. I want to use it in the article Biddeford Pool. I don't want to to use it to establish the notability of Biddeford Pool; that is already established, I think. However, there's nothing in the article about the entity was formed, and I think that'd be worthwhile to include in an article about a landform. (Right? I'm going to assume that this doesn't violate WP:DONTDOTHIS or WP:DONTDOTHAT or WP:FORGODSSAKESNOTTHAT or any of the other strictures invoked above.)

So anyway then, the question is, how reliable is this instrument (a masters thesis) for statements of fact?

If Morphology of Saco Bay Coastal Inlets says "Biddeford Pool was created about 16,000 years ago during the last retreat of the glaciers...", can I use that or not? That is my question. Let's say that Violet Anne Tower took soundings and studied local rock striations herself to conclude this, so there's no deeper source. How confident can we be that Violet Anne Tower is correct and that Biddeford Pool was not in fact created in the last two thousand years by storm action? Violet Ann Tower herself is nobody to trust. She is 22 and has a bachelor's degree and she could just be guessing. So we are bereft of being able to lean at all the author's expertise and reputation, which is usually a helpful part of vetting sources, and entirely dependent on the structure in which the instrument was created, which I guess means the grownups who vet these things.

I assume that someone reads these things? How many -- one, or more? Do they catch errors? Most errors, or only glaring errors? If they do catch them, do they correct them? If Violet Anne Tower is sloppy and not too bright, is her advisor going to say "Wait, your data doesn't support that particular conclusion with sufficient clarity, take that out"? Or is her advisor just going to be like "OK Violet, you worked hard, nice work, accepted"? How much does it differ between institutions? Does it matter if the the thesis is in the archives of Harvard rather than University of Southern Maine? What if it's Lindsey Wilson College? Does it differ a lot between advisors within an institution? Is one advisor just trying to get to retirement with minimum work, while the one in the next office is extremely thorough? Or do institutions have people seeing to it that there are no sloppy time-servers doing this? This is a lot of questions, I know, but it would interest me to know the answers to a few, at least. Herostratus (talk) 00:06, 24 August 2019 (UTC)

Generally, master's theses don't contain original research like that, or when they do (in the sciences) it is typically published as normal papers, with the supervisor(s) as co-author; where they do it should be treated with great caution. Otherwise, the answers to all your questions have very variable answers, even more than published papers or books. No doubt most theses are fairly good, but some may be pretty bad. The institution matters rather a lot. Hence we try to avoid them. But then you'll have noticed that Wikipedia articles are stuffed full of newspaper sources, written for the most part by journalists who are no sort of expert, and got their information who knows where, or stuff off the internet for which the same is true. Johnbod (talk) 02:02, 24 August 2019 (UTC)

This is why for hard sciences, especially medicine, newspapers are not treated as reliable sources. WP:MEDRS talks about this. However for humanities and social sciences newspapers are often treated as RSes. If the reliability of masters' theses, in a sum, is about the same as the reliablity of newspaper sources, we should encourage or discourage their use in a similar way we do newspapers; allow it or disallow it depending on the situation. To be more precise, we could inquire with experts in each relevant field if a Wikipedian challenges the use of a master's thesis in the article. Each field and each institution has its own conventions, so professors who are experts in their fields can provide guidance. WhisperToMe (talk) 00:38, 25 August 2019 (UTC)

In my view, a Master's thesis available on-line, but not published in a peer-reviewed publication, is never a reliable source on Wikipedia unless it is something that is widely cited in other reliable secondary sources. If someone were to use one, other than for minor attribution, in an FAC, I would give them a hard time (i.e. rigorously question its use by rigorously evaluating its content) As I've written elsewhere, "Writing is a compact between an author and their reader. In an encyclopedia that compact is about faithfully translating a complex subject whose paper trail is sometimes jargon-ridden in language and concepts familiar to an ordinary layperson." If it is getting to the point that you have to rely on Masters theses for content development, then you are violating that compact, that trust the layperson implicitly places in you, for you are not being faithful in your grand precis of the subject. You are essentially trading uncertainty for certainty. Abandon that topic and pick up something vital that has sources aplenty, but goes abegging for contributors. Unfortunately, a lot of editors consider Wikipedia to be some kind of gray zone, a no man's land, or happy medium, between conventional encyclopedia writing (such as they usually do in Britannica) and original research, such as done in an article submitted to a journal. As for an editor's question about the specific 1952 thesis, I would worry about something else. In the last 70 years, the chronological estimates of glaciation in New England in the last ice age, have been revised downward (to my limited knowledge). My views don't just apply to Master's theses. If you are going through the literature with a fine-toothed comb, picking up secondary sources here and there, and making the decision yourself about which is more important and which not, you are again violating that compact. You need to rely on tertiary sources, such as widely used textbooks or literature reviews in journals for that. If something is not in them, then you are again being unfaithful by emphasizing the content they are being cited for. Due weight is talked about everywhere, but implementing it in encyclopedic writing is tricky business. Fowler&fowler «Talk» 14:58, 11 September 2019 (UTC)

Relying on other tertiary sources (introductory university-level textbooks, general and specialty encyclopedias) would be great guidance for some topics. Others sadly don't have any entries in such sources, though... WhisperToMe (talk) 12:55, 8 October 2019 (UTC)

Memoirs as RS

Do they qualify as RS when used in articles about the subject or the subject's works? At Wikipedia:Featured article candidates/Mullum Malarum/archive4, Fowler&fowler disputed this. The FAC failed. Now I want to get this cleared before I attempt another FAC. --Kailash29792 (talk) 17:18, 8 September 2019 (UTC)

When used for information about the subject, or the subject's works, a memoir would be a primary source, which could be a reliable source but must be used with care, as explained on WP:Reliable sources. It could be a secondary source when the author is writing about works written by others. Jc3s5h (talk) 17:25, 8 September 2019 (UTC)

The FAC did not fail because of only this issue. There were likely others. The problem here is that we have a book of recollections of J. Mahendran, the director of a movie, Mullum Malarum: Mahendran, J. (2013) [2004]. Cinemavum Naanum [Cinema and Me] (in Tamil). Karpagam Publications., published in the Tamil language, unavailable under the author's name on WorldCat (see here and here), or in the Library of Congress Catalog, or on Google books (all three list Tamil language publications); never translated; very likely not peer-reviewed before publication; and to my knowledge, not reviewed in an English language journal or newspaper after publication (see here) These recollections were being employed for citing text added to the Development section of the movie . Its assessment proved very difficult for this reviewer. Fowler&fowler «Talk» 18:24, 8 September 2019 (UTC)

Fowler&fowler, now I've added this link to the Bibliography section. Is it good? Or is this a better link? At least now do you accept the book is usable? --Kailash29792 (talk) 08:40, 9 September 2019 (UTC)

Obviously, none are. Of course not, that uncataloged memoir is not appropriate except by way of one or two vignettes that support statements in reliable secondary sources. Please take out those links. They are spam. Fowler&fowler «Talk» 11:00, 9 September 2019 (UTC)

Since udumalai.com showed the book on sale, I did not believe it was spam. Now can any other admin share their view about the book Cinemavum Naanum? --Kailash29792 (talk) 11:35, 9 September 2019 (UTC)

Jc3s5h, after assessing Fowler's comments, do you have anything to say about this book? But I personally believe it should not be dismissed as a non-RS. --Kailash29792 (talk) 06:08, 11 September 2019 (UTC)

I am not familiar with Indian book publishing so I won't venture any opinion about this book. Jc3s5h (talk) 16:05, 11 September 2019 (UTC)

For Metropolitan Correctional Center, Chicago I used Orange is the New Black as a source. I attributed some statements so that they credit author Piper Kerman rather than stating it entirely as fact but others were stated as fact. WhisperToMe (talk) 13:08, 8 October 2019 (UTC)

Thanks WhisperToMe. The same should work for Cinemavum Naanum, where I can do something like "According to Mahendran's book CN..." instead of replacing the source as Fowler suggests, because I can't find many alternative sources for claims in the book. --Kailash29792 (talk) 13:17, 8 October 2019 (UTC)

Comment by Ldjhdata

I am still bewildered about why Wiki seems to be asking the impossible in the case where I edited in a factual paragraph but it was removed (censored) as soon as spotted. There are always going to be cases where someone (me in this instance) knows from first hand information that the detail edited in is correct, yet how on earth can one provide verifiable proof when the origin of the story was hushed up by the people involved. Or, to put it another way, is Wiki not the place to post factual material within an article that is known to be severely limited by the original writer who was clearly not aware of the truth? Why does Wiki appear to be prepared to publish incomplete material whilst censoring evidence from a source that offered irrefutable evidence?

L D J Harris Ldjhdata (talk) 12:58, 10 October 2019 (UTC)

All we know about you is you have access to the internet. There is no way to distinguish a sincere Wikipedia editor with accurate first-hand knowledge from a bald-faced liar, or an agent of a totalitarian regime who's assigned mission is to spread disinformation. Jc3s5h (talk) 13:12, 10 October 2019 (UTC)

Crime in Cape Town was so bad,the ruling community moved to Stellenbosch temporarily / Purchase of Cape by Britain from the Dutch

Daily Mail bulk removals (again)

Once again, despite WP:DAILYMAIL, we're seeing bulk runs to remove or tag DM sources. This is being done on the basis of "match the domain name", nothing more. Some of these aren't even being read (some of them have been dead archive links), they're just beng removed.

We do not have consensus for such a bulk action. WP:DAILYMAIL does not give that.
There is an edit box warning in place against restoring these (I don't know any RfC conclusion for such a box, if anyone has it handy it would be helpful to note it here). That's a strong disincentive to many editors against restoring (that's obviously its intention), yet there is no justification for removing them.

Examples:

https://en.wikipedia.org/w/index.php?title=Winterval&diff=923882178&oldid=923881149 (simply removed "rm non-RS" - but there is no such policy)
https://en.wikipedia.org/w/index.php?title=Walker,_Newcastle_upon_Tyne&curid=2446938&diff=923891660&oldid=915659675 ({{deprecated inline}} tag)
https://en.wikipedia.org/w/index.php?title=1958_Mars_Bluff_B-47_nuclear_weapon_loss_incident&diff=prev&oldid=923747289 Using AWB "removing deprecated sources (Wikipedia:Reliable sources/Perennial sources)", despite then creating orphaned citations to a now-missing ref.
https://en.wikipedia.org/w/index.php?title=Web_blocking_in_the_United_Kingdom&type=revision&diff=923814395&oldid=921938665 "rm. Daily Mail" removing multiple sources from an article on UK web censorship, where the DM's fulminating reactions can be seen as an essential part of adequate coverage of the topic.

We do not have consensus for such a bulk removal. Such unconsidered removals are damaging - that has been the agreement of past RfCs, and so we don't do it, and we don't do it automatically in bulk. The removals are also being done in an automated and unchecked fashion, such that they leave technical breakages such as orphaned cites. These automated bulk removals without per-article consideration should stop. Andy Dingley (talk) 13:36, 31 October 2019 (UTC)

Leaving {{deprecated inline}} is fine: while the DM ban is not against wholesale remove, we don't want DM refs ultimately, until DM is essential to the topic itself. Removing DM cites, even if that left behind a {{cn}} is not helpful - one can use the DM cite to find a non-DM cite that should be replaceable, and the DM ban wasn't "must get rid of immediately'. --Masem (t) 14:33, 31 October 2019 (UTC)

Should DeviantArt be Inserted Into the User-Generated Listing?

RfC on mainstream newspapers as RS

Deprecated sources

Tommy Roberts

Interview marked as US

Self-published and questionable sources as sources on themselves

This section causes repeated confusion as people think it is carte blanche to source a substantial proportion of an article from the subject's own websites, with just the occasional independent source thrown in to establish notability. I think we should add something along the lines of:

Use of self-sourced material should be de minimis, the great majority of any article must be drawn from independent sources.

For an example, as of today, Knights of Columbus has around 80 citations to the organization itself, which account for between a third and half of the content of the article. In general material that is not included in reliable secondary sources but only on a subject's own website, is not encyclopaedic, by common consent. ABOUTSELF appears to be used on occasion as an end-run around that. Guy (help!) 22:02, 23 November 2019 (UTC)

Imho the 5th condition ("The article is not based primarily on such sources") is sufficient, which comes down to: at least 50% of the article should be drawn from independent reliable sources (for clarity, press reports promulgating hardly more than the subject's press releases, or comparable sources, are not independent enough in this type of assessment – I've seen attempts to calculate them in as "independent" go down at AfD). 50% is really enough: the alternative could be unsourced content, which is worse. An example of a recently created article balancing around 50% is Bach Digital: here all independent sources are in German (or something that looks like a machine translation), in which case having the same content in English from the WP:SELFSOURCE is definitely an advantage. --Francis Schonken (talk) 22:38, 23 November 2019 (UTC)

I think we should strengthen that condition along the lines of Guy's suggestion. I've gotten similar thrown at me when I've gone to deal with at least one problematic editor on video games articles. That said, I think WP:V is the place this conversation should be since the section title in question is there; maybe JzG meant this topic to be on WT:V? -Izno (talk) 01:23, 6 December 2019 (UTC)

Izno, not really, as it's in this guideline not V, but I have no objection to posting a link there for more eyes. Guy (help!) 11:12, 6 December 2019 (UTC)

@JzG: The content is copy-pasted verbatim from WP:ABOUTSELF (with a little policy/guideline drift; possibly the c+p happened the other way). --Izno (talk) 15:56, 6 December 2019 (UTC)

I've frequently referred to sources used under WP:ABOUTSELF as primary source equivalents. If this were written into policy, WP:PSTS would automatically limit their use:

Wikipedia articles should be based on reliable, published secondary sources and, to a lesser extent, on tertiary sources and primary sources. Secondary or tertiary sources are needed to establish the topic's notability and to avoid novel interpretations of primary sources. All analyses and interpretive or synthetic claims about primary sources must be referenced to a secondary or tertiary source, and must not be an original analysis of the primary-source material by Wikipedia editors.

WP:SELFSOURCE is a copy of WP:ABOUTSELF, so any changes here could also be proposed at WT:V. — Newslinger talk 09:55, 8 December 2019 (UTC)

Guy, I wonder whether we should take a more holistic view of this question.

On the one hand, more than 50% of the article should come from independent sources.

In a technically separate rule, more than 50% of the article should come from secondary sources.

And per the rationale given at WP:WHYN, you still need to have enough content to write an article.

So I wonder whether there is some minimum system here, by which we say that if you can't write (say) 300 hundred words from WP:INDY sources, then you can't have an article on that subject at all, and that if your independent sources only allow you to write n words, then the total article length can't exceed 2n. Does that sound about right to you? WhatamIdoing (talk) 03:52, 30 December 2019 (UTC)

Non-independent and primary sources should be used mostly for providing details on topics identified from independent, reliable, secondary sources. --Ronz (talk) 04:43, 30 December 2019 (UTC)

Too many citations of the same source

Hey, all! I recently encountered a situation involving a Wikipedia article that cites multiple pieces of writing from the same generally reliable source (i.e., The Hollywood Reporter or THR), two of which are reviews that support statements regarding the article's subject's film performances. The article originally contained three reviews from THR, but one was removed on the basis of there being other reliable sources that exist. Regardless of whether said basis is true, I have not found a guideline that states whether it is more appropriate to have two citations of the same reliable source and one from another, or if three that lead to the same source is perfectly acceptable. What I'd like to ask is: Is there such a thing as too many citations of different pieces of writing from one reliable source in an article? Thank you very much, all! KyleJoan^talk 16:25, 10 December 2019 (UTC)

WP:ONESOURCE might be interesting, but it's not part of the WP:PAGs. The best solution is to improve the article with a greater variety of sourcing. Alexbrn (talk) 16:27, 10 December 2019 (UTC)

The article I referenced already contains many citations of many different generally reliable sources, so in this situation, given that WP:ONESOURCE is met, do you believe there could be too many citations of the same source? Thank you for referring me to that essay as well! A very helpful read. KyleJoan^talk 16:34, 10 December 2019 (UTC)

What article? This sounds more like a WP:NPOV question – we should reflect the balance of reliable sources proportionately, though there are many factors that go into assessing that. Alexbrn (talk) 16:39, 10 December 2019 (UTC)

It's Timothée Chalamet. Sorry for not mentioning it earlier! I was avoiding giving any tells regarding my position. The core of the situation was that there were two reviews that supported the same positive statement regarding one of his film performances–one from The Guardian and one from THR. The THR citation was removed per my earlier explanation, and I was trying to gauge how justified the removal was. As it stands, said positive statement is only supported by the Guardian citation. KyleJoan^talk 16:48, 10 December 2019 (UTC)

Note: The positive statement itself was never disputed; it was the repeated use of THR, even though other reliable sources are repeatedly utilized across the article as well. KyleJoan^talk 16:55, 10 December 2019 (UTC)

I'm not seeing the problem here. THR is reliable for this kind of content, and it's multiple articles, not excessive use of the same article. It doesn't overwhelm the article and there's no obvious evidence that THR is biased, which could in turn bias the article. Is this just a stylistic preference? What's your issue with the source please? Guy (help!) 10:32, 12 December 2019 (UTC)
In all honesty, I do not see the problem with the citation either. Another editor with a different view felt that since there are already two other THR reviews of the subject's other films in the article, then–per WP:COMMONSENSE–find a different one. It could be a stylistic preference, but they never explained further. KyleJoan^talk 14:41, 12 December 2019 (UTC)
Are these three articles by three reporters in the same publication, or three articles by the same writer in the same publication? That can affect the POV, especially on more subjective matters. WhatamIdoing (talk) 03:45, 30 December 2019 (UTC)
It's the former (i.e., three articles by three reporters in the same publication). KyleJoan^talk 08:19, 30 December 2019 (UTC)
That is less concerning than the single-journalist situation. It might be slightly preferable to cite a wider diversity of publications, but it does not sound important to me. WhatamIdoing (talk) 04:36, 2 January 2020 (UTC)

RfC at TERF

Fiction as a reference

Consider Adding a section recognizing Regulated Public Report Filing as "RS"

Hi Wise Wikipedians who care about Reliable Source guideline,

I'd like to propose adding a reliable source of "Public Report Filing" that's being regulated. Examples are "Amazon's SEC Annual Report", "Wikimedia Foundation's Form 1023". I suspect currently it falls under self-publish and can't be used as Reliable Source. They are rich of information and can add to Wikipedia a lot of new facts and help for validation. Certainly part of these reports are subjective, such as executives' discussion of the market of the organization, and those shall be considered biased, but other parts are objective, such as who are the board of members, what are their annual revenue, etc.

I'd like to argument since they are public and under regulation, they can be treated RS unless proven not. I'd like to add such suggestions in the WP:RS in new paragraph but before go down that effort, I'd like to ask for early feedback:

1. Has this been discussed/decided before, I might be missing out?
2. What's your opinion on this direction?

Thank you!

xinbenlv ^{Talk, Remember to "ping" me} 19:25, 19 January 2020 (UTC)

I think this falls in the "reliable primary source" category. Our guideline probably deserves some mention of this type of sources under "Reliability in specific contexts". Deryck C. 11:40, 20 January 2020 (UTC)

Xinbenlv, are you aware (a) of ABOUTSELF and (b) that this guideline is not a policy, so doesn't really set the "rules" about this kind of thing? If there is a rule or guideline which prohibits the use that you want to make of such reports, you'll need to figure that out and propose the change there, not here. Finally, just in case: If you're wanting to use such a public record as a source about a living person, that's prohibited, see BLPPRIMARY (along with the two following sections of that policy, which may also be of interest). Regards, TransporterMan (TALK) 18:58, 20 January 2020 (UTC)

I want to support the point raised by Xinbenlv here. This is a distinct class of document that meets the "published sources with a reputation for fact-checking and accuracy" segment of reliable sources and occurs in a regulatory context in which factual information is legally required both by the regulatory agency and (for public corporations) the fiduciary expectations of the marketplace. I would suggest these be deal with through clarification of how these sources differ from others listed in WP:ABOUTSELF, and how they can be reliably used to provide factual information such as: quantity and location of corporate facilities, membership of corporate boards, reported revenue and profits, personnel, etc. This is an area where there is a vast reporting apparatus of news sources that are not more reliable than the underlying primary source documents, don't engage in additional fact-checking, and aren't subject to strong legal liability for mis-statements in the way that the regulated filings are. It is in fact better (from a reliability standpoint) to use Exxon's corporate income and profits filing than to use Oil and Gas Weekly’s reporting of the same number; they're not checking Exxon's balance sheet, just reporting it.

With respect to BLPPRIMARY, this echoes a problem that has come up at WP:N/ACADEMICS. (There, we often have notable people whose primary career details are only listed by their own webpages or those of their institution. The latter—"XY is an distinguished professor of history"—are extremely reliable, but non-"third party" sources.) It is almost never the case that, for example, someone's corporate board membership receives independent, third-party coverage from a reliable source. Such facts are not akin to other, personally invasive "public records searches" (for reasons that I think should be obvious, but could be explained in a guideline) and a regulated filing should be acceptable source to demonstrate this. I acknowledge that such a shift would requiring vetting at BLP.--Carwil (talk) 12:59, 21 January 2020 (UTC)

Firstly, I have only skimmed through this as I've got to fly....but wanted to link into the content directly the above. I agree that official publications should be used (content available) as RS; to my mind, not being allowed implies some sort of morally-driven WP censorship; if it's there in the public domain, anyone can access but due to the high search-engine returns for WP, it could be more widely disseminated? I've recently written this quote at a bio Talk (to show I knew of it): "Exercise extreme caution in using primary sources. Do not use trial transcripts and other court records, or other public documents, to support assertions about a living person. Do not use public records that include personal details, such as date of birth, home value, traffic citations, vehicle registrations, and home or business addresses." This has been worded as a 'blanket' covering all and every aspects; there's nothing stopping a supposedly-reliable third-party from directly-accessing the same content and (re)publishing, making it cite-able.

Picking up on the example above of non-press coverage, the same applies to journeymen journalists - they might be rarely or never written about except in, perhaps, scandal, accident or eulogy, the latter if they're lucky. As journalists, a bio could link to other articles. I have written such an article, but I was lucky to find (and bought-in) a magazine obituary (about 1988). If there's no internet equivalent, it is indeed difficult. I have another drafted another several years ago, recently died, but couldn't find any obituary. Another is unwritten, with only one obituary I know of, at basically a glorified WP:SPS, the owner of which having been heavily criticised for wanting to contribute to WP based on his own experience and website. I later saw the same site being referenced (Sump) by one of the same critics.

Another example of WP inconsistency is a cult film starring Adrienne Barbeau and Keir Dullea; without a hard-published (1980s) reference, there's never going to be a stand-alone article, just an inclusion into the filmography (again, I bought a copy). If it was on WP, then afterwards it would likely be copied into wider webform.--Rocknrollmancer (talk) 14:05, 21 January 2020 (UTC)

I would agree that the documents and filings being discussed are reliable. However, reliability is only one part of the usage puzzle... the real problem is that such documents and filings are also PRIMARY sources - and there are strong limits as to HOW we are allowed to use primary sources (See: WP:No original research for more on that). Blueboar (talk) 14:27, 21 January 2020 (UTC)

Marathi.tv

Adding guidance to define reliable sources (news organization)

Daily Mail TV reviews

Are Daily Mail TV reviews really included in the moratorium on it as a source (as per this diff)? I can understand news reporting being subject to it, but a review is just one columnist's opinion of a television programme, play, album etc. I'm puzzled why they are mixed in with the moratorium on the Daily Mail's news operation. - X201 (talk) 16:58, 16 February 2020 (UTC)

You might want to look at the November 2019 WP:RSN thread Opinions in the Daily Mail. Peter Gulutzan (talk) 17:18, 16 February 2020 (UTC)

The Daily Mail was proscribed for its seeming inability to report facts accurately, and in some cases making up quotes and falsely attributing them. IMO this does not extend to columns, reviews and Op-eds, where the DM is primarily functioning in a primary source capacity for the author. The scope of the discussion was very narrow and these aspects were not discussed in any great detail from what I recall. What was discussed was the Daily Mail's factual reporting capabilities. Unfortunately the closing statement is very open-ended and is open to interpretation. Ultimately though I would say the ban only applies when it is being used as a secondary source (these were the only examples that were put forward) and not as a primary source. Proscribing it as a primary source would create difficulties: for example, when the DM accused Ralph Miliband of being unpatriotic, his son Ed Miliband wrote an Op-ed in the DM to rebut their accusations. The only source for the piece is the DM itself, even if other newspapers reported on it. In such a case I would say that the DM story is not a reliable secondary source for facts and claims about Ralph Milibad, but the op-ed by Ed Miliband is a reliable primary source for what Ed wrote. Betty Logan (talk) 22:09, 16 February 2020 (UTC)

There is a global ban on most British national newspapers, including the Daily Mail. This includes their summary removal from all articles, including any content sourced to them, and any adjacent sources from other places. So far it's the Daily Mail that's most affected, but the others (Express downwards) have been removed as well. The ban is implemented by David Gerard who is pursuing a personal crusade to do so. There is no attempt at meeting any reasonable editing standards in doing so - sentences are regularly torn in half or odd fragments left behind, or the layout of whole pages trashed and then left that way: .

There is no policy to support this. There is no RFC to support this. WP:DAILYMAIL is specific that it does not support an absolute ban like this, nor does it support a campaign of immediate and total removal. Nor does WP:DEPS support this. Yet that's what's being done.

A look at Special:Contributions/David Gerard will show how out of control this has got. Despite a past agreement that these sources would be tagged rather than removed, or that the sources would be removed rather than the content, or even that articles simply wouldn't be broken by careless edits which fall below WP:CIR, none of this is happening. We just have a handful of editors (and at this scale it's just one) ripping articles to pieces. Andy Dingley (talk) 03:52, 17 February 2020 (UTC)

Request for a restatement in the positive

@User:Jc3s5h, regarding [this reversion].

"there is nothing reliable in [Wikipedia] that is not citable with something else."

I find the preceding statement most unclear: a double negative pendant on the vague terms "citable" and "something else". Since you did not like my effort to clarify, would you please do me the favor of offering your own positive restatement? Dayirmiter (talk) 05:11, 16 February 2020 (UTC)

The passage was added in substantially its present form by Schierbecker in 2011.

I think it means that the parts of Wikipedia that are reliable either have a citation to a reliable source, or finding a citation to support the claim would be straightforward. Parts that have no citation and for which finding a citation would be challenging are not reliable. The part could be traced to the editor who added it, but that editor, like all Wikipedia editors, is considered unreliable. Jc3s5h (talk) 06:19, 16 February 2020 (UTC)

User:Jc3s5h Thank you for your reply. Since the fact of who authored the passage in question is irrelevant, I will ignore your first and last sentences. Also, since our concern is what is on the page, the conjecture that "finding a citation . . . would be" this or that has no bearing. Excluding those parts and "I think it means that", and supplying the word in brackets, your reply reads as follows.

The parts of Wikipedia [that] are reliable have a citation to a reliable source. Parts that have no citation are not reliable.

I think this will work. Is this agreeable to you? I also suggest omitting the introductory clause about original research as not necessary to, and possibly muddying, the point being made.

Further comment? If not, I'll make the edit in a day or two. Dayirmiter (talk) 05:28, 17 February 2020 (UTC)

Your edit means a statement like "Paris is the capital of France" should be considered unreliable if it does not have a citation. Your version would encourage editors to provide citations to provide citations for claims that could be looked up in an elementary school library in five minutes. There is no need to clutter articles with such citations. If you make the change you will be reverted immediately.

The sentence you choose to ignore, "the part could be traced to the editor who added it, but that editor, like all Wikipedia editors, is considered unreliable", has nothing to do with who added a passage to this guideline. It means that whatever knowledge and ability the editor who added a claim to Wikipedia doesn't count; what counts is that either the claim has a citation to a reliable source, or the claim is common knowledge that is already known by many readers, and can easily be looked up in a wide assortment of reference works for those who don't already know it. Jc3s5h (talk) 13:57, 17 February 2020 (UTC)

W3Schools

Muflihun.com

Sood (as in surname and caste)

Adding clarifty to RSOPINION

This is related to the section about the Daily Mail and TV Reviews, which later I had separately asked over at WP:RSN about (see in "Daily Mail and RSOPINION"), but I want to discuss generically (not in context of Daily Mail) the RSOPINION section, as that's a point of contention at the base of this, and feels like it needs clarity. Whether this needs a larger RFC or not depends.

In the above linked discussion, there are some editors that read RSOPINION that come to believe this means this only allows for opinion that come from reliable sources to be used, as, when combined with WP:WEIGHT, we are not supposed to use non-reliable sources for anything (fact or opinion). But this seem to get into a circular argument on how a reliable source is defined, and, strictly read, can significantly limit what sources can be available to review.

RSOPINION as given appears to define another type of reliable source, that being when the source is giving the opinion of the author of the piece. It should not matter where that piece is published, though I think part of the current language in RSOPINION creates the stumbling block is the line "A prime example of this is opinion pieces in sources recognized as reliable." which is not wrong, but gives the impression this is the extent of how far RSOPINION can go.

Mind you, the other side of the problem is that one could see this could create an endless supply of sources that would challenge a proper evaluation of WEIGHT. You'd have people arguing that we'd need to include all the opinions of the various far-right conspiracy theorists on various political topics, for example, of which there are an endless supply of blogs. Or to extend the TV show concept, the numerous armchair critics that go on about their favorite shows that go "woke". Absolutely fair point and we need to avoid that.

I personally believe there is an implicit understanding in WEIGHT that we give more weight to experts in a specific field or from journalists and analysts that write for RS-for-facts over "nobodies", even if those nobodies outweigh the experts 100 to 1. (eg we don't give in to review bombing). This is already the basis of WP:FRINGE, but when we get to less fringe-y topics, like, say abortion rights, this concept basically says that we have more than enough RSOPINION on both sides of the debate (parties fighting pro-life or pro-choice in the numerous legal cases) that we don't need to touch random RSOPINION blogs about that. When we talk a television episode, we can look to whom is reviewing it rather than where they necessary published and determine if a noted critic that they probably have a good deal of WEIGHT to bring compared to random YouTube reviewer #3235. (The lack of mention of "experts" or similar authority in WEIGHT is a surprise as I read it in this light now, but would make sense to consider).

As an added point, from a recent discussion at WT:VG, we'd also probably like to make sure that when using RSOPINION to draw opinion from a source that is normally not an RS for some reason, we'd still like that source to have some type of editorial control that we know that opinion likely has gone through so it likely isn't some type of uncontrolled rant or bs like that. For example, we know that Forbes Contributors articles received only minimal editorial control so that we do not consider them RS-for-fact, but that minimal editorial is sufficient to this end to say that they can be used for RSOPINION.

To that end, I'm tossing up two points of discussion:

Should RSOPINION be read to include opinions published in sources not normally reliable for fact? If so, do we need to add language to be clear about that?
Assuming "yes" to the first question, should we add further language to RSOPINION to stress the that expert/authorative opinions used as RSOPINION be given much more WEIGHT compared to others?

This is more exploratory, and if this is a controversial direction, then maybe an RFC will be needed. --Masem (t) 15:00, 3 April 2020 (UTC)

Masem, thanks so much for broaching this important subject. Here are a few thoughts.

I have studied and written about this and related subjects for many years, and this essay deals with it: NPOV means neutral editing, not neutral content. There I deal with how to deal with biased sources and content. We should use all RS, including biased sources, and do it properly when creating content. Biased sources just require more care, including attribution.

To me, WEIGHT is determined exclusively by coverage in independent RS, never by coverage in self-published sources or unreliable sources. They get ZERO weight in determining "probable cause" for use. Otherwise, in principle, all unreliable sources, even blacklisted ones, can be used under certain controlled conditions.

Your points: (1) would need added language; (2) definitely. -- Valjean (talk) 17:51, 3 April 2020 (UTC)

Add "Medium" to user generated content section

Since a bunch of other social media, etc. websites are listed, and since I noticed that the website Medium (website) isn't listed (it's a "blog host", where people can self-publish), I was wondering if we could put it in here. I had to revert an edit on the YIMBY page (made by an admin nonetheless) that used this obviously poor source (the underlying data may have been fine, but it was clearly original research).CrenshawLine (talk) 18:15, 12 April 2020 (UTC)

While I agree that Medium (RSP entry) primarily hosts user-generated personal blogs, the list of examples in WP:UGC is already very long, and there is no possible way for us to make it exhaustive. It might be better to stick to generic categories (i.e. "personal websites, personal blogs, group blogs, content farms, Internet forums, social media sites, most wikis, and other collaboratively generated websites"), instead. Medium clearly falls under "personal blogs" and "group blogs". Everyone makes mistakes (including administrators), and a polite reminder is usually enough to make things right. — Newslinger talk 04:37, 14 April 2020 (UTC)

Will do. I've worked together with said admin to improve that page. Maybe in the spirit of your comment we could condense that section then, and remove some of the redundant social media sites?CrenshawLine (talk) 06:31, 14 April 2020 (UTC)

Sure, here's my draft:

Content from websites whose content is largely user-generated is also generally unacceptable. Sites with user-generated content include personal websites, personal and group blogs, content farms, Internet forums, social media sites, online video platforms, most wikis (including Wikipedia), and other collaboratively created websites.

I've added "online video platforms", and Wikipedia is the only example left in the sentence. Do you have any suggestions for other categories that could be listed? — Newslinger talk 08:21, 14 April 2020 (UTC)

Looks good. Maybe Image hosting service? CrenshawLine (talk) 22:08, 14 April 2020 (UTC)

Second draft:

Content from websites whose content is largely user-generated is also generally unacceptable. Sites with user-generated content include personal websites, personal and group blogs (excluding newspaper and magazine blogs), content farms, Internet forums, social media sites, video and image hosting services, most wikis (including Wikipedia), and other collaboratively created websites.

I've also added an exception for WP:NEWSBLOG to be consistent with WP:SPS. How is this? — Newslinger talk 02:54, 15 April 2020 (UTC)

I am but one of many wikipdia users, but i think it's fantastic.CrenshawLine (talk) 05:24, 15 April 2020 (UTC)

Thanks for your input! As no one else has commented in the last couple of days, I've implemented the new wording in Special:Diff/951043117. If anyone objects to this, please say so. — Newslinger talk 05:30, 15 April 2020 (UTC)

Newslinger, thank you for your attention on this. Don't want to complicate things, but as always there are minor points that get overlooked in any general statement. When it comes to "personal websites", and the music-related articles I work on here, I can't help wondering about the likes of robertchristgau.com and pauldunoyer.com. These are sites that serve as archives of articles, reviews and essays by Robert Christgau and Paul Du Noyer, respectively; similar sites exist for Greil Marcus and a good few other well-known writers, I'm sure. Just wonder whether mention of "personal websites" needs to allow for this type of scenario. Thanks, JG66 (talk) 10:00, 15 April 2020 (UTC)

Hi JG66, the phrase "personal websites" was actually already in the guideline before I made the recent changes. Fortunately, subject-matter experts such as Christgau and Du Noyer are exempt from WP:RSSELF (and WP:UGC, if the content is verified in some way to be written by the author) according to this line in WP:RSSELF: "Self-published expert sources may be considered reliable when produced by an established expert on the subject matter, whose work in the relevant field has previously been published by reliable, independent publications." However, the exemption does not apply to third-party claims about living persons, as this line in the WP:SPS policy takes precedence: "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." I'm going to add the living persons exception to WP:RSSELF to be consistent with WP:SPS and WP:BLPSPS. — Newslinger talk 10:08, 15 April 2020 (UTC)

Hi Newslinger. Sure I realise(d) the phrase was there beforehand – and fair enough if you think it's okay as is. My query concerned the bald statement, without qualification, at WP:UGC, but I'm sure you've got a better handle across the whole guideline. No probs. JG66 (talk) 13:06, 15 April 2020 (UTC)

Wikimedia Project Grant Proposal on Disinformation

Social media post by an organization

At first blush, and without any in depth study, I would say that The Ninth Judicial Circuit Historical Society would likely be a RS. If I found information on their website about a third party, I wouldn't hesitate to use it. What about on their Facebook page, however? Can we consider that to be an extension of their webpage? I am currently working on an article about the first female lawyer in Alaska and not sure if this brief biography is acceptable or not. --Slugger O'Toole (talk) 12:53, 3 May 2020 (UTC)

Reliabilty is tied to the publisher, not the medium. An exception would be if there is doubt that this is, in fact, their FB account. It's not a verified account, but that's not reason in itself to contest it, unless others have reason to doubt its authenticity.—Bagumba (talk) 13:57, 3 May 2020 (UTC)

Bagumba, Thank you. That's what I was thinking as well. Perhaps the language could be updated to make this clearer. -- Slugger O'Toole (talk) 14:10, 3 May 2020 (UTC)

Do you have an example where you think it was unclearly written?—Bagumba (talk) 15:50, 3 May 2020 (UTC)

Bagumba, This sentence reads as if it applies to individuals, but not to reliable organizations like a historical society: "Self-published expert sources may be considered reliable when produced by an established expert on the subject matter, whose work in the relevant field has previously been published by reliable, independent publications" (emphasis mine). -- Slugger O'Toole (talk) 16:12, 3 May 2020 (UTC)

But that text doesn't distinguish the reliability of a website vs. social media. It's a seperate issue about whether an org is technically an "expert".—Bagumba (talk) 16:24, 3 May 2020 (UTC)

I recommend not using it because it is on Facebook. Here is another source about the subject , but it seems to be slightly at variance with the Facebook post. I think this demonstrates why it is best not to use user generated content such as Facebook. Anyway, maybe try posting your question at the reliable sources noticeboard. You might get more input there. ---Steve Quinn (talk) 19:46, 3 May 2020 (UTC)

On second thought, the two sources might be similar in this instance. I'll leave it to more discerning minds than mine to reach a conclusion on that :>) ---Steve Quinn (talk) 19:55, 3 May 2020 (UTC)

I think this is a better link to the above Google Books source. See pages 66-69. It should open on page 66 by my reckoning. ---Steve Quinn (talk) 20:09, 3 May 2020 (UTC)

Also, the Facebook content looks like unattributed copyrighted material. This also is a no-no on Wikipedia. ---Steve Quinn (talk) 20:40, 3 May 2020 (UTC)

According to a new PDF...

WP:AFFILIATE wording

UGC: Public opinion aggregators are adequately reliable sources of information about public opinion

In discussing the reception of any movie or TV show, I think it would be appropriate to include IMDB ratings, metacritic user scores, and rotten tomatoes user scores. In this case we are not trying to source FACTS from UGC; we essentially just reporting on the results of a public opinion poll. You wouldn't ban reporting on public opinion polls just because the ultimate source of the data is UGC. If anything, the critic scores on these sites are less reliable than the user scores, because critics can be individually bribed, or be a tiny out-of-touch minority. Studies have shown a slightly negative correlation between critical opinion of movies and public opinion of movies. 2600:8801:B04:2000:200F:3CF2:D738:E9EB (talk) 22:25, 12 May 2020 (UTC)

Specifically I disagree with this revert: https://en.wikipedia.org/w/index.php?title=Star_Trek%3A_Discovery&type=revision&diff=955954215&oldid=955952726 2600:8801:B04:2000:200F:3CF2:D738:E9EB (talk) 22:29, 12 May 2020 (UTC)

Isn't that already done in "Reception" sections of media? But you have to be specific, not generalize, based on a single source. "was rated highly on IMDB - cite: IMDB rating" rather than "People liked it - cite: IMDB rating" - Keith D. Tyler ¶ 22:43, 12 May 2020 (UTC)

More to the point, I would say you should be able to say "is highly rated" by citing multiple common sources of ratings, such as RT, IMDB, Metacritic, et. al. But not just one by itself. - Keith D. Tyler ¶ 22:46, 12 May 2020 (UTC)

I looked at the revert and I looked up the cited reason and dear God that policy section is a goddamn nightmare. I'd like to see the consensus discussion that led to that. Sigh. - Keith D. Tyler ¶ 22:50, 12 May 2020 (UTC)

I disagree with the suggestion. It is not very difficult to create or have access to a large number of accounts on IMDb (RSP entry), Metacritic (RSP entry), and Rotten Tomatoes (RSP entry) to manipulate the user ratings. These user scores are frequently review bombed by groups of fans/detractors (e.g. Animal Crossing: New Horizons) or even individuals (e.g. Kunai) for reasons unrelated to the overall quality of the product, and are poor indicators of the general public's opinion. See Review bomb § Notable examples for more examples. Metacritic and Rotten Tomatoes have strong reputations for selecting credible reviewers in their panels of critics. — Newslinger talk 04:00, 13 May 2020 (UTC)

Reliability of private research foundations

Are there any sort of restrictions on sourcing information from private research foundations? I've found lot of research that is published by private foundations that are largely funded and administrated by people with very high levels of wealth, and the research that these foundations put out is pretty clearly showing a bias toward the interests of very wealthy people. Does Wikipedia consider such sources to be reliable? Personally I don't at all but I'm curious as to what others think. And I'm referring to research and articles that the foundation self-publishes, not studies that they perform which are then published in peer-reviewed academic journals. --Señorsnazzypants talk 07:59, 14 May 2020 (UTC)

Can you provide an example of these foundations? Many think tanks are treated similarly to advocacy organizations, and can be used with in-text attribution under WP:BIASED in some situations. — Newslinger talk 08:41, 14 May 2020 (UTC)

I'll give you one off the top of my head since I came across it recently, that being the Cato Institute. Here for example is a recent article about how illegal immigrants in the US allegedly commit crimes at a disproportionately low rate; however, the article repeatedly uses data on all immigrants as if it represents illegal immigrants specifically, and of course this entire concept is discredited by ICE (a taxpayer funded, publicly run agency) annuals reports showing that over 90% of ICE's arrestees have a criminal record and average 4 crimes or pending charges per person.

All that being said, the Cato Institute is run largely by finance professionals and wealthy business owners, which relates to the original question of biased research organizations. I think you would be right to call Cato a "Think Tank" but I wasn't sure if that was the appropriate term for it. --Señorsnazzypants talk 10:32, 14 May 2020 (UTC)

The Cato Institute (RSP entry) was discussed twice on the reliable sources noticeboard. In these discussions, editors acknowledged that the organization is biased or opinionated, and believed that its research should be presented as attributed opinions when it is used in articles. The last discussion was in 2015, so if you'd like to start a new discussion on the noticeboard to gauge the current consensus, please do so. The only other think tank on the perennial sources list is the Center for Economic and Policy Research (RSP entry), and the consensus on that one is more or less the same. Most advocacy organizations and think tanks are considered situational sources, although there are a few exceptions. — Newslinger talk 13:45, 14 May 2020 (UTC)

Why raise a double standard? In wp:rs there is no criteria for actual reliability, expertise on the topic, independence, minimal bias, objectivity, not being under the control of people with an agenda (self-interest or otherwise) etc., The same for treatment of sources in wp:ver and wp:npov. It just requires certain attributes (published, layer of editorial review etc.). If want all of those things, or to assign greater weight to sources which have them (which I think we should do), we should put it into the policy & guidelines. Until then, IMO we should not apply a double standard of selecting particular sources and requiring them to meet a higher unwritten standard. North8000 (talk) 14:18, 14 May 2020 (UTC)

I don't think I was, I'm asking a question to clarify how we regard and indicate the reliability of certain types of sources, that's all. --Señorsnazzypants talk 15:12, 14 May 2020 (UTC)

@Señorsnazzypants: I'm sorry, I didn't intend to sound like a criticism of your question, and I should have worded it differently to avoid that. I was was speaking about this occurring in general. Or, more to the main point, the lack of requiring actual reliability & objectivity except for certain situations. Sincerely, North8000 (talk) 15:33, 14 May 2020 (UTC)

At least historically, before there were all these RFCs on RSN, it was quite normal for awkward sources like the Cato Institute to come multiple times on RSN. All sourcing discussions were then seen as context dependent. So there could be a new question for every specific controversy.--Andrew Lancaster (talk) 18:04, 14 May 2020 (UTC)

These discussions are still a popular option for situational sources. However, nobody has brought up the Cato Institute on the noticeboard in almost five years. — Newslinger talk 18:08, 14 May 2020 (UTC)

Business Insider

Reliable sources for medicine, nutrition and toxicology should state dose

Propaganda pieces are not WP:RS

RS for examples

It's been my understanding for some time that RS has exceptions when *the act of the source is the fact* rather than the content of the act of the source. For example, if you want to say "The National Enquirer said X," it is appropriate to link to the National Enquirer when they say X. Is this no longer the case? I'm seeing a lot of reverts using RS as justification of those cases. - Keith D. Tyler ¶ 23:31, 9 May 2020 (UTC)

Keith, yes, this is still the rule – see the FAQ at the top of this page, not to mention the reason that WP:ABOUTSELF is in the WP:V policy – but since the creation of Wikipedia:Reliable sources/Perennial sources, there are some editors who seem to have made it their mission in life to remove all such sources, even when they are used appropriately and are reliable for the content in question. I believe in some cases, there's a sort of "These are not the droids you are looking for" code that can be added to a citation template, but it is a bit like arguing with a brick wall. (Yes, several editors did predict exactly this problem when RSP was proposed.) WhatamIdoing (talk) 00:36, 10 May 2020 (UTC)

Is the claim "The National Enquirer said X" also cited to a reliable source that fully supports the claim? If not, the claim is undue weight for the article (although WP:ABOUTSELF may apply for the article on the National Enquirer itself). — Newslinger talk 01:49, 10 May 2020 (UTC)

User:Newslinger, the National Enquirer already is a 100% reliable source that fully supports claims about whatever it said. If the claim is "The National Enquirer said <whatever's in an article in that publication>", then you don't need another source to verify the claim. Every publication is 100% reliable for claims about its own contents.

As a completely separate point, WP:DUE is policy and must be observed. But when that's the problem, then the editor removing the content needs should not mislead other editors by telling them that the problem is unreliability. WhatamIdoing (talk) 06:26, 10 May 2020 (UTC)

I agree that editors should ideally mention the due weight policy when removing content that is undue. Although due weight is separate from reliability, the two are related: the first sentence of WP:DUE states, "Neutrality requires that each article or other page in the mainspace fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in the published, reliable sources" (emphasis added). Because the National Enquirer is generally unreliable, what the publication said about a third-party topic is generally undue on the article of the third-party topic, if the Enquirer's coverage is not mentioned by a different reliable source. When a source is undue because of its unreliability, removals of that source are based on both the reliable sources guideline and the due weight policy. Neglecting to mention the due weight policy is less than optimal in terms of semantics, but doesn't reduce the validity of the removals. — Newslinger talk 11:06, 10 May 2020 (UTC)

So then you would have "According to Newsweek, the National Enquirer said X" with a cite to Newsweek saying NE said something, instead of "The NE said X" with a cite to NE actually saying X. That's.... kind of silly imo. I dare say that would be tertiary sourcing.
NOR says "A primary source may be used on Wikipedia only to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source." It fits.- Keith D. Tyler ¶ 22:39, 12 May 2020 (UTC)

No, one would just use "The National Enquirer said X" with a citation to Newsweek. Here, a citation to the National Enquirer (as a primary source) is optional and can be added alongside the citation to Newsweek if there is consensus to do so. However, to establish due weight for this claim on the article of a third-party subject, a citation to a reliable source (not the National Enquirer) is required. — Newslinger talk 04:11, 13 May 2020 (UTC)

The latter form with full chains is used in literary history texts (which Wikipedia RS guideline is generally based on). However, we need to agree on how to add those chains. Alongside or inside the citation to non-primary source? Erkin Alp Güney 11:18, 3 June 2020 (UTC)

I'm a bit late to this party, but I generally endorse what Newslinger has already written. (I will reiterate WAID and Newslinger's frustration with some editors' unfortunate habit of citing WP:RS when they're really applying WP:FRINGE or WP:DUE. In many cases, it's the right action, but citing the wrong policy.)

That said, I will mention that there are cases where sources aren't robustly reliable for their own contents. Particularly in dealing with online sources, there are many platforms where it is possible for a publisher to modify, replace, or delete content. Access to previous versions, withdrawn/retracted articles, or deleted content - or even acknowledgement of changes that might have taken place - may be left to the honesty and diligence of an individual publisher. Some publishers are extremely transparent about revisions, corrections, and retractions, and some publishers fail badly.

Citing another reliable source in addition to - or even in lieu of - the original publication can be worthwhile in this sort of situation. The most clear-cut example I can think of would be a case where a notable individual tweets something noteworthy on Twitter and then subsequently deletes their tweet. Citation of an independent third-party news story provides a more reliable source for the tweet's content than Twitter itself. TenOfAllTrades(talk) 14:24, 14 May 2020 (UTC)

Do translations need a reliable source?

Summarising non-fiction books

User:DevilTrombone, who I'll invite here, is trying to build a summary of a book at Talk:¡Adios, America! (a book by Ann Coulter. I've tried to explain that we really need reliable sources discussing the book, but he is quoting from Wikipedia:WikiProject Books which says "Summary/Content — report on the content of the book and how it is organized. This can include any thesis and major illustrative examples. Do not try to re-organize the content, just summarize and report it." He says "I am trying to ascertain AC's "thesis and major illustrative examples."

My immediate concern is the Wikiproject guideline which I think is confusing and in fact has confused this new editor. I've got concerns about the sources they are adding to the article but that's a separate issue. Doug Weller talk 17:35, 7 June 2020 (UTC)

I have abandoned the summary approach, but I'm still interested in this topic. --DevilTrombone (talk) 18:14, 7 June 2020 (UTC)

Ideally, for a good summary, there should be sources other than the book itself (such as a review of the book in a scholarly journal where relevant, or, in cases like this, lacking that a newspaper). Of course, a summary of what is covered in the book is not a bad idea, but then again the basic idea of not rewriting it or changing it or adding one's own opinion is covered by WP:OR. RandomCanadian (talk / contribs) 18:44, 7 June 2020 (UTC)

It's very similar to writing about fiction. The work is the source for the summary, which can be given briefly in plain language without detail, summamrizing, without extensive quotes or detailed paraphrase. Nobody should do it without a copy of the book in hand, any more than in fiction. The contents of the book should not be taken from a third party source any more than the plot should be in fiction--if it is you're giving what the reviewer wanted to say , not the author, and it can be quite different and not at all reliable. It's ok to --it's even necessary to say what the argument is, but making the case fior it or giving the casestudies or data used is the job of the book itself, not the WP article In contrast, just as with fiction, interpretation of whether the content is good or sensible of course must be taken from genuine third party sources--just as for fiction. , They way I think of it is, that if someone in conversation mentions the book, the WP article will give you just enough to know what they;re talking about and not sound like an jackass who never heard of it. DGG ( talk ) 05:43, 9 June 2020 (UTC)

Semi-protected edit request on 20 June 2020

Help create a reliable sources quiz for new users!

AfterEllen as a reliable source

PinkNews as a reliable source

Pay sites

Is it valid and/or appropriate to post information that is only available at a pay site? I'm speaking specifically about death certificates during the last 10 years or so. ←Baseball Bugs ^{What's up, Doc?} carrots→ 04:21, 10 July 2020 (UTC)

WP:PAYWALL says yes of using in beyond paid sources, but we would actually never use death certs directly as a reliable source. They are like court documents that we should not be chasing down. --Masem (t) 04:24, 10 July 2020 (UTC)

The specific case is Walter Breuning, whose cause of death was not disclosed in the newspapers but is stated on his death certificate, which is visible on the pay site Ancestry.com. So, is the issue that it's a primary source not corroborated or commented upon by other sources? ←Baseball Bugs ^{What's up, Doc?} carrots→ 04:27, 10 July 2020 (UTC)

If we have to turn to the death cert., and no other source gives it, we don't report it, period. It's not a required piece of data. You ahve the day of death sourced to the newspaper and it implies it was likely of some age/natural cause which is all that is needed, nothing else. Going any farther would be inappropriate reliances on the privacy of court records. --Masem (t) 05:00, 10 July 2020 (UTC)

Very good. Thank you! ←Baseball Bugs ^{What's up, Doc?} carrots→ 05:15, 10 July 2020 (UTC)

I don't see that in the policy. Moreover, I don't see how court documents nor death certificates can't be considered RS. The Rambling Man (Stay indoors, stay safe!!!!) 06:54, 10 July 2020 (UTC)

Ancestry.com appears on Wikipedia:Reliable sources/Perennial sources, and even has it's own anchor on that page (WP:ANCESTRY.COM). The consensus from previous discussions is "Ancestry.com is a genealogy site that hosts a database of primary source documents including marriage and census records. Some of these sources may be usable under WP:BLPPRIMARY, but secondary sources, where available, are usually preferred. Ancestry.com also hosts user-generated content, which is unreliable." So, essentially, as long as you follow the rules set out in WP:PRIMARY, the use of a death certificate sourced to Ancestry.com is acceptable. Harrias ^talk 07:14, 10 July 2020 (UTC)

A death record is an official, legal document, correct? Are official, legal documents generally primarily sources per Wikipedia policy? Futurist110 (talk) 03:19, 11 July 2020 (UTC)

In this specific case, the birth and death dates are available in free sources, while cause of death is not. ←Baseball Bugs ^{What's up, Doc?} carrots→ 08:12, 10 July 2020 (UTC)

So, if Walter Breuning's death record will ever end up on FamilySearch.org and FamilySearch.org will remain free by then, it would then become acceptable to use it as a source for Wikipedia? Futurist110 (talk) 03:17, 11 July 2020 (UTC)

"Puff pieces" in usually reliable sources

The FAQ for Reliable Sources includes the following:

Are there sources that are "always reliable" or sources that are "always unreliable"?

No. The reliability of a source is entirely dependent on the context of the situation, and the statement it is being used to support. Some sources are generally better than others, but reliability is always contextual.

I believe that the policy really needs to underline this somewhat with regard to "puff-pieces" that are commonly found in otherwise credible publications. This impinges particularly on biographical articles: in an interview, the subject, or their family, friends or devotees, will make statements about otherwise undocumented aspects of their life and/or work, and this is repeated unchallenged by the reporter and printed as perfectly in keeping with the journalistic purpose of the publication. Whether the newspaper received any challenge to such a claim we will not usually know. And so a claim made to heighten the story of the newspaper article's subject, printed uncritically, can be reproduced in a Wikipedia article and is essentially immune from any challenge: it might seem vague, onpen to misunderstanding, unlikely or exaggerated, but if an editor tries to remove it, then WP:RS is invoked, and any attempt to qualify the claim is denounced under words to watch. The result is that instead of being able to apply a standard of encyclopaedic truth, we are open to the credulity of a journalist tasked with writing a glowing tribute to somebody with a PR agent for the subject standing at his/her shoulder.

I believe we need something in the 'Reliability in specific contexts' section to address this, perhaps subheaded 'Uncritical appreciations' . As a first draft, this might read something like:

Even usually reliable sources often carry articles (known as puff-pieces) that are largely uncritical, sometimes promotional, in tone, that do not challenge assertions made by or about the subject. Claims that are based solely on such sources may be challenged and removed from Wikipedia articles.

Unfortunately, the tag I would like to apply to such a policy ([[WP:PUFF]]) is already in use in relation to a notability essay.

Thoughts? Kevin McE (talk) 14:44, 17 July 2020 (UTC)

No. Statements made by an interviewee in an interview would fall under WP:PRIMARY. Opinions by the interviewer offered in response to statements of the interviewee would fall under WP:PRIMARY too. That has no relation to the reliability of the source. Also, the use that can be made in Wikipedia of such material is already limited by the WP:NOR policy: no need to add other somewhat conflicting guidance about the same somewhere else. Available tags include, e.g., {{Primary sources}} and {{Attribution needed}} – no need to invent a new one. --Francis Schonken (talk) 15:13, 17 July 2020 (UTC)

Proposal to rewrite WP:NEWSORG

Related Articles