Wikipedia talk:Reliable sources/Archive 47

This is an archive of past discussions. 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.

This guideline says that

What do you want to add to an article, that you can't find a better source for? WhatamIdoing (talk) 03:20, 29 September 2014 (UTC)

IMDb Pro, instead of simply IMDb, is considered WP:Reliable. For more Wikipedia pages on IMDb, see

talk

) 03:22, 29 September 2014 (UTC)

Regarding my initial post above... I was referring to

talk

) 03:36, 29 September 2014 (UTC)

IMDb Pro is a type of paid user account that gives you extra tools. It gives you access to representation. It also allows entry for people looking for work who have not appeared in films or TV yet, and they can load head shots and their resume. It also lists casting calls and has social networking features. The dates of birth and death for those that appear are not affected. We should have a standard paragraph about what information at IMDB is reliable and what is user generated. We should say that the biographies may not be reliable but that the lists of appearances of actors are, and the casts of films are. Current film information is supplied by the production companies. IMDB is under editorial control and has a link to submit a correction or emendation for editorial review. --Richard Arthur Norton (1958- ) (talk) 17:24, 29 September 2014 (UTC)

I've alerted WP:Film to this discussion.

talk

) 18:08, 29 September 2014 (UTC)

I've found it to be particularly unreliable when it comes to future films. Its my understanding that anyone can submit information to IMDb and that information is screened by IMDb but only for blatant/obvious fallacies. See IMDb help.--TriiipleThreat (talk) 18:24, 29 September 2014 (UTC)

• I think that is the nature of pre-production. Actors and directors and crew become contractually attached to a project, but that legal obligation is time dependent and expires if the project does not meet deadlines. Then availability is checked again and a new set of names become attached to the project for the next time period. --Richard Arthur Norton (1958- ) (talk) 00:43, 30 September 2014 (UTC)

Thanks for weighing in,

talk

) 18:31, 29 September 2014 (UTC)

I am in agreement with

indiscriminant as they try to list everyone, which is not wiki policy.AbramTerger (talk

) 18:40, 29 September 2014 (UTC)

@

Flyer22: I believe IDMb Pro as Richard Arthur Norton mentioned just allows for additional tools/services. It does not provide any additional editorial oversight.--TriiipleThreat (talk

) 18:47, 29 September 2014 (UTC)

• Comment An IP added the sentence that IMDB pro is regarded as a reliable source, and I reverted that edit on the basis that it had not been discussed. I do not have access to IMDB pro, but if it contains all of the content of the free IMDB then it follows it can't be reliable. There may be extra components (in a similar vein to the WGA credits) that may be considered reliable, but if this is the case it would be better to state explicitly which parts of IMDB are acceptable for our sourcing requirements. Betty Logan (talk) 19:16, 29 September 2014 (UTC)
• Comment. I will echo what the others have said. There are much better sources out there than the IMDb, and they should be used in preference to it. Although the IMDb is usually correct, there's a huge gulf between "correct" and "reliable". Reliability implies fact-checking and editorial control. The IMDb is a very useful resource, but it is definitely not reliable. Upcoming films are especially rife with speculation, rumors, and wishful thinking from fans. The WGA credits are reliable, but the crowd-sourced data is not. If I thought a crowd-sourced database was a bad idea, I wouldn't be here. But that doesn't mean that I harbor illusions about its reliability. NinjaRobotPirate (talk) 23:30, 29 September 2014 (UTC)

Can you give us a link to an example of wrong data on a film at IMDB so that we are all looking at the same thing. Remember Variety may published that actor-A is attached to a film in pre-production, which is correct. That actor may not appear in the final film. Delays may have made them unavailable by time principle shooting begins. The director may recast the part after filming begins if the performance is not what they are looking for, and actor-b takes over the role. Actor-A may be cut from the film in editing for time constraints. Actor-A may die before or after filming begins. Yet, the fact that they were attached is correct. IMDB has ample warnings about pre-production specifications. --Richard Arthur Norton (1958- ) (talk) 00:56, 30 September 2014 (UTC)

This all began with Talk:Michelle Thomas. The discussion concerns the wording that says IMDB is unreliable and must be removed as a source for a birth date. The other questions are whether we have to know an absolute truth, or just use the best information known at the time even if it is fuzzy. This made me aware that we need to distinguish user generated content at IMDB, from content under the editorial control of IMDB. --Richard Arthur Norton (1958- ) (talk) 01:01, 30 September 2014 (UTC)

You've missed the entire point. It doesn't matter how often the IMDb is right, it's still not reliable. Reliability and correctness are related but different concepts on Wikipedia. Some dude's blog might be right about a bit of data, but it's still not a reliable source. Find a better source for your data than the IMDb. If you can't, then it's

know it to be true but can't find a reliable source to support it. I routinely remove/revert these additions, and I'm afraid you'll find no support from me. NinjaRobotPirate (talk

) 03:17, 30 September 2014 (UTC)

Where is your evidence that dates of birth and death are user submitted and not under IMDB editorial control? The New York Times is wrong every day, that is why they list their corrections the following day. Here is a cut and paste from my introduction at the top of this debate:

IMDB was sued for displaying the real age of an actress and not what she submitted to the editors. "Hoang also made an issue of how IMDb had allegedly used credit information when she signed up to IMDb's Pro account and how IMDb's employees used a third party verification website to gain information to use in her profile." see this article The submission rules on the webpage for adding yourself as an actor say that "[biographical information] will be sent to the IMDb Data Editors for checking." You can submit a plot summary, an actor's biography, you can rate a movie, you can leave comments, you can report an error for review. You can not go in an change Denzel Washington's birthday like you can in Wikipedia. This is what we call editorial control, just like the New York Times. --Richard Arthur Norton (1958- ) (talk) 13:41, 30 September 2014 (UTC)

Not reliable. Just because they were sued for one piece of information once doesn't make them a reliable source for the rest of their database. The fact that you're trying to use it in a case where it disagrees with the actor's tombstone and NY Times obituary makes this a particularly bad example of how reliable IMDB's supposed to be. Should the family correct the tombstone based on the supposed editorial superiority of IMDB? This is getting absurd. Even if you think IMDB is credible for the material (I don't) then there are still better sources for the information and IMDB shouldn't be treated as equivalent to the NY Times. We're supposed to cite what better sources say. __ E L A Q U E A T E 14:12, 30 September 2014 (UTC)

• If you read the article you would see that there are six references for her death from reliable sources, evenly split over the two dates. This tends to happen when someone dies close to midnight. The New York Times is not witnessing her death directly, it reads: "Thomas died December 22 in Manhattan, according to her publicist." I am sure they are quoting the publicist 100% correctly, but the publicist may be just as wrong as anyone else quoted that uses the other date. --Richard Arthur Norton (1958- ) (talk) 19:07, 1 October 2014 (UTC)

Well, then, I think we've that "example of wrong data" at IMDb, then. Under no circumstances should you use a website with such a mediocre reputation to contradict a major daily newspaper. Instead, why don't you contact IMDb and tell them that they likely have an error? WhatamIdoing (talk) 14:58, 30 September 2014 (UTC)

The important thing to get across to editors and indirectly to readers is that we can never cover advance information about films because it's so likely to be changed. Another point worth remembering is that we don't absolutely have to have the date of birth of every actor, and we should look for other sources alongside IMDB. Itsmejudith (talk) 16:03, 30 September 2014 (UTC)

The New York Times is error prone too, they print corrections every day because the articles are under editorial control. After the

Jason Blair affair the New York Times had to hire an ombudsman to address the systemic problems. We have evidence from the IMDB lawsuit that IMDB is under editorial control. If they are unreliable, we should not be even linking to them in the external links, anymore than we should be linking to someone's fan blog. Finding individual errors does not make the entire data set unreliable. They just need to be under editorial control to make corrections when they are pointed out. --Richard Arthur Norton (1958- ) (talk

) 18:59, 1 October 2014 (UTC)

External links are not required to be reliable; see

WP:ELMAYBE

.

As for The New York Times vs IMDb, the reputation of the daily newspaper for fact-checking and editorial control dramatically exceeds the website. The newspaper wins hands down, every time, and if there really was a problem with what they published, then you should be able to find a published correction about it. If you're convinced that they're wrong, then they even publish directions on how to let them know about an alleged error. WhatamIdoing (talk) 20:49, 1 October 2014 (UTC)

So does IMDB: http://www.imdb.com/help/?adding/ I am sorry you do not like IMDB, but to convince me you are going to have to give me examples of incorrect information at IMDB, and not just data that conflicts with other sources, but 100% wrong. At Talk:Michelle Thomas we have six reliable sources split over two death dates. --Richard Arthur Norton (1958- ) (talk) 00:10, 12 October 2014 (UTC)

Well, off the top of my head, IMDB has this insanely dumb trivia item that Jennifer Lee, the director-screenwriter of Frozen, is the sister of Walt Disney Animation Studios visual development artist Brittney Lee (who did work on Frozen and also did illustrations for one of the children's books that was released for the film). This is like, face-palm, jaw-dropping, blows-my-mind stupid, the kind of idiotic trivia submission that implies IMDb is not even running a basic Google quality control/fact-checking test on submissions (or whomever they hired to do it is not doing their job). First, the two Lees look nothing alike (see here and here). Second, Lee is one of the most common English surnames. Third, Brittney Lee grew up in western Pennsylvania while Jennifer Lee grew up in Rhode Island with her sister, Amy Leigh Kaiser, who is now an English teacher at a high school. Fourth, it's clear from this interview (as well as this op-ed) that she has just one sister. I've seen several other dumb errors over the years that implies IMDB has poor quality control but this is the best example that comes to mind right now. --Coolcaesar (talk) 08:59, 12 October 2014 (UTC)

Speaking as someone who's edited his own actor profile, and that of a friend, I can attest that IMDB is authoritative and fact-checked for such matters as castings, which episodes, personal information et al., and that its fact-checking requirements are rigorous....I know because I've had to verify my own data, and been through their procedures. Same applies to "locations" - their editors take a while to verify and fact-check things, but they docheck with the production companies to verify whether you were in such and so an episode, or as for which locations etc. For things like trivia listings and such, and forum/discussion boards, it's not useful as citations, of course not. The claim that it is not fact-checked is complete rubbish, and while my own attestation of this might be construed as "original research", the claim for anyone who has not experienced their procedures first-hand to claim that fact-checking by them does not exist is..... a claim only, and pretentious.Skookum1 (talk) 06:15, 13 October 2014 (UTC)

But the point remains whether an alternate source could still be found for those things. A credit from a film is better than a reference from IMDB. If an actor is uncredited in the film, I would want a reference beside IMDB to ensure notability since IMDB credits tend to be

indiscriminant.AbramTerger (talk

) 15:52, 13 October 2014 (UTC)

I do apologise but I am very new to Wiki as such I am checking my Q relates to identifying reliable sources but I am directed to the talk page, now I am on the talk page and highlighted above 'this is the talk page for discussing improvements to the Identifying reliable sources page' just who do I need to talk to when I keep getting redirected again to ask a Q in the first instance if someone can help --Ab747000 (talk) 00:39, 7 November 2014 (UTC)

No problem Ab747000. If you're want to talk to other editors about a specific source, that you want to use on a specific page, go here: Wikipedia:Reliable sources/Noticeboard. It's a noticeboard where you'll get more help sorting out specific questions from more editors. Read some of the other questions on that page and you'll get the hang of it. (This talk page is for discussing changes or improvements to the overall guidance we give generally, on the guidance page.)__ E L A Q U E A T E 01:16, 7 November 2014 (UTC)

A description (South Korea remains a major destination for travelling Japanese men who exploit children through prostitution.) is written in a source (ECPAT report) with a citation (a blog which is a translation of Munhwa.com). However the citation discusses nothing about the child prostitution at all.

Before initiating a general Policy/Guideline issue here, I would like you to comment as a case study at Talk:Prostitution in South Korea#Reverted removal of the Japanese sex tourists in South Korea section. All suggestions are welcomed.―― Phoenix7777 (talk) 01:48, 30 October 2014 (UTC)

It doesn't matter what the (now-current) version of the blog is, because (a) we're not using that blog ourselves and (b) we don't require our sources to tell us how they came by their information in the first place (see the last item in the /FAQ).

One could, I suppose, make the claim that by screwing up their citations (or by getting their citations right, but the blog has been edited since then?), that they might not be an example of ideal fact-checking.

One would also want to consider whether or not this is the only reliable source in the world that says this, too. WhatamIdoing (talk) 03:33, 30 October 2014 (UTC)

That last comment is the key. We may have grounds to question the reliability of the ECPAT report (for this specific sentence)... but the solution is to simply find another reliable source for the statement. There are lots of sources that note how that SK is a destination in the Japanese sex tourism industry, and for child-sex tourism in particular. Blueboar (talk) 12:17, 30 October 2014 (UTC)

Where, though, is the requirement that this solution be applied?

WP:DUE requires that reliable sources disagreeing with or differing from the statement be given appropriate weight, but I don't see a requirement anywhere that another source confirming the statement be added if a WP editor disagrees with what ECPAT (acknowledged as an RS in relation to this topic) said. I don't know whether ECPAT got this right or not, but it would be improper for me to question it based on my own knowledge/belief/opinion/POV. Wtmitchell (talk)

(earlier Boracay Bill) 23:41, 9 November 2014 (UTC)

WP editors are not reliable sources. WP editors are probably among the most unreliable of unreliable sources. Opinions/POVs/judgements by WP editors about individual assertions made by sources considered reliable for a topic do not in themselves hold any weight whatever in impugning the reliability of those individual assertions by sources considered reliable for a topic. Contrary opinions/POVs/judgements expressed by other sources also considered reliable for the topic do have weight, and should be given due weight. Wtmitchell (talk) (earlier Boracay Bill) 22:05, 10 November 2014 (UTC)

Of course any editor can question whether any source is reliable for a particular statement, regardless of whether it's generally found reliable for other things. There's nothing wrong with Blueboar expressing doubt about whether a source is reliable for a certain statement; it's a requirement for all editors to make assessments like these, constantly, in discussion or alone. Characterizing editors weighing reliability as OR is wrong-headed. It's required to edit in the first place.

Now if you're saying editors (within consensus) can't determine who we judge to be a reliable source for a particular claim, then who would? Who then decides what reliable sources are reliable enough to decide what other sources are reliable? Should we add nested levels of citations for the sources of the citations themselves (as The New York Times found to be reliable by Professor Wimperschnootz who is vouched for by Professor McDeedee who we heard is good from...)? This is rabbit-hole silliness. Editors (within consensus) are always ultimately going to be the ones who decide if any single particular source should be considered reliable enough for any particular claim. There's no way around it. As we aren't infallible, we add citations so that readers have the ability to check what we're basing our editorial choices on. Editors aren't reliable for article material, but we are the only ones available to judge what sources we consider reliable enough to repeat in a context (in negotiated consensus with each other, of course). __ E L A Q U E A T E 23:02, 10 November 2014 (UTC)

Shorter: No source is blanket considered reliable for every claim it could possibly make. Assessing whether a particular source could be considered reliable for specific individual claims is routine editing, regardless of whether the source was considered reliable for other things.__ E L A Q U E A T E 23:19, 10 November 2014 (UTC)

Here is a report of a peer-reviewed (and presumably copy-edited) journal article where the phrase "(Should we cite the crappy Gabor paper here?)" was included in the finished work. The journal was Ethology.

I guess my point is there is peer review and then there is peer review, and there's probably a bigger drop in reliability from The New England Journal of Medicine and The University of Southern North Dakota at Hoople Review of Multidisciplinary Ethnic Studies then there is from The University of Southern North Dakota at Hoople Review of Multidisciplinary Ethnic Studies and your Uncle Dwight's grocery list.

I'm not sure if we ought to change

When available, academic and peer-reviewed publications, scholarly monographs, and textbooks are usually the most reliable sources. However, some scholarly material may be outdated, in competition with alternative theories, or controversial within the relevant field.

to

When available, respected academic and peer-reviewed publications, scholarly monographs, and textbooks are usually the most reliable sources. However, some scholarly material may be outdated, in competition with alternative theories, controversial within the relevant field, or of indifferent quality.

but maybe. Herostratus (talk) 01:19, 12 November 2014 (UTC)

I wonder what someone like User:MastCell or User:Colin would make of that. With their work on MEDRS over the years, I suspect that they've seen just about everything.

My initial thought is that "respected" would be loved by a certain class of POV pushers, and that "of indifferent quality" is nothing more than the truth (but is it a truth we want to acknowledge here?). WhatamIdoing (talk) 02:11, 12 November 2014 (UTC)

All journals are going to have occasional editorial collapses. If they're rare they don't usually affect overall reputation and level of "respect". The journal in your example is going to have about the same amount of academic respectability next month as it did last month, despite their blooper. It is a reminder that RS aren't somehow infallible, but that's true whether it's New England or North Dakota. Changing the guidance would change nothing of substance here, and might cloud the original points. __ E L A Q U E A T E 02:41, 12 November 2014 (UTC)

Hah. Since I was pinged, here are my thoughts, in no particular order. I think it's fairly common to scatter commentary like this throughout a manuscript when one is drafting it. I certainly do it, as a way of communicating informally with coauthors who aren't physically on-site. Of course, I also go through and remove those comments before submitting a manuscript, in order to avoid embarrassment. At the same time, as the Slate article correctly notes, most quality journals employ editors and proofreaders who work on a manuscript after it has been accepted for publication, and tidy it up (in particular, references often need to be converted to the "house" format; grammatical errors need to be fixed, especially when the primary author is not a native English speaker; and tables and figures often need to be touched up). I suppose smaller journals, like Ethology, don't splurge on proofreading staff. I guess I would hope that, as a matter of simple common sense, we don't treat Etholgy as equivalent to Science or NEJM. If we do, then all the policy-tweaking on Earth won't save us. And finally, my favorite example is from the Slate article: "RESULTS: In this study, we have used (insert statistical method here) to compile unique DNA methylation signatures..." That's the most honest description of statistical methods I've read in an omics study in quite some time. MastCell ^Talk 04:55, 12 November 2014 (UTC)

It's an isolated gaffe. It doesn't discredit Ethology in a fundamental way. This particular one has gotten attention because it's funny, but the NEJM also issues corrections for its typos and omissions. "Reliable source" does not mean "has never made a public error", it means they have a reputation for only rarely making mistakes, and admitting and correcting when they make them. __ E L A Q U E A T E 17:51, 12 November 2014 (UTC)

There is no such thing as a completely reliable source and, unless the source is really trying, there is no such thing as a completely unreliable source.

The text of the project page currently begins, "Wikipedia articles should be based on reliable, published sources"

I think more accuracy would be achieved if it began with something like, "Wikipedia articles should be based on published sources that are considered to be relatively reliable".

Suggestions of wording would be appreciates but top sources may get things and there is always the potential that other sources, perhaps backed with video or other evidence, may provide good usable information. Sources should not necessarily be polarised between good and bad and the accuracy and reliability of materials may depend on the individual researcher/reporter/writer/editing staff involved.

What you say is true, and I've said so in many comments here over the years. Unfortunately, it is not easy finding a fully satisfactory way to specify this. "relatively" reliable is not really an unambiguous concept-- it can mean either As compared to The Truth, or As compared to other sources. Further, sources are typically of different degrees of reliability for different types of information, even in the same reference. A bio on an official source connected with the subject is very likely to be accurate about the birthplace and date, and considerably less so about the person's importance. The standards for reliability also vary, particularly with respect to BLPs, and even more particularly with respect to negative information in BLPs. The same researcher or writer will be more accurate in some places than others; for example, academic historians will be more authoritatively reliable in their specialty than in making broad generalizations in general works. There are many other qualifying factors, and how to say this right will take some discussion. The present wording "Contextual" is in my opinion a reasonable way to say it, but we shouldindeed go into more detail. DGG ( talk ) 00:23, 14 November 2014 (UTC)

You could try something like, "Wikipedia articles should be based on published sources that are reliable for the content they're being cited for." WhatamIdoing (talk) 06:04, 22 November 2014 (UTC)

The problem with that is that it opens up a whole can of worms - "that source isn't reliable, because it is wrong". Now sometimes that is a legitimate argument (e.g. when a source makes an obvious error), but often it is used when a source doesn't concur with the TRUTH^TM being pushed by a POV-warrior. I suspect that what we really want in sources is that they are expected to be reliable, unless there is strong evidence to the contrary. So for example we expect an academic historian specialising in the English Civil War to know when the Battle of Naseby was fought, and accordingly we cite him or her on trust for the date of the battle, lacking other evidence. AndyTheGrump (talk) 06:36, 22 November 2014 (UTC)

FYI

 – Pointer to relevant discussion elsewhere

Please see Talk:Aspromonte goat#RFC on Italian dairy & farming industry sources for a discussion of whether a website produced by an industry consortium and managed by an Internet publishing/marketing firm, is a presumptively reliable or unreliable primary source for (non-promotional) information. This could have broad implications, because outsourcing of online publication is increasing, not decreasing, as more and more companies see no point in having an on-staff webmaster when firms provide these services on a contractual basis with entire teams of people, with many collective years of experience, and the cost of whom are shared across numerous clients.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  19:56, 21 November 2014 (UTC)

A mining industry consultant who claimed no direct ties to

Mount Polley mine disaster but he's a member of the mining association and works as a consultant in the field, and has a history of POV edits...so the line between contracted and so-called "volunteer" editing on such articles seems very iffy. Similarly, a tide of industry-generated "oil sands" RS were used to mandate the Athabasca "tar sands" in Wikipedia as "oil sands"....when industry funds research and publication using its chosen/misleading jargon, and the deluge of resulting "sources" is used to mandate such POV language, it's an ongoing problem....and there will be those here who defend those repetitive and self-referential sources as "legitimate" instead of being industry-generated.Skookum1 (talk

) 04:45, 22 November 2014 (UTC)

I believe that SMcCandlish is asking a different question: If an industry group hires a firm to create a website for them (rather than hiring an employee to create a website in-house), then is that website reliable or unreliable? For example, if it's okay to cite the "Ruritanian Widget Manufacturers Association" website, written by Joe Employee and Jane Webmaster, for a factual claim (e.g., that manufacturers in Ruritania make a million widgets a year), then is it okay to cite the same information from the "Ruritanian Widget Manufacturers Association" website, only this time the website was written by the employees of Professional Websites R Us instead of by Joe Employee and Jane the in-house webmaster? WhatamIdoing (talk) 06:40, 22 November 2014 (UTC)

Suffice to say I had quite the time correcting a false claim on certain wine-region articles, where the defender of the mistake (a claim that the Sonoran Desert extends into British Columbia, which is counter-factual as the Sonoran Desert ends at the Colorado River...in the real world) fielded article after article from wine and travel magazines as "reliable sources", all clearly playing from the same industry-based press releases. The editor combating me claimed no affiliation to the wine industry......my point is that it's not just overt consultants and companies, who havedisclosed, but those who do not, or as with the Mt Polley item make excuses and evasions and disavowels while acting quite red-handedly as COI/POV.....this is also rife on political bio articles of all kinds, anywhere, where it's cler that "p.r. operatives" are at play, whether as SPAs or "lurkers".Skookum1 (talk) 07:33, 22 November 2014 (UTC)