Wikipedia talk:Reliable sources

Propose to be able to cite application source code

Sometimes there are articles which are about an application or piece of software where it can be difficult to find reliable secondary sources describing application features, known issues, etc, as they may not even exist in the currently allowable list. I propose the idea that if actual source code is publicly available, that it should be allowed as a reliable source. The source code doesn't have a point of view, and cannot lie or make exaggerated claims (the comments might, but the actual code cannot). If source code can be pointed to, then it should be acceptable. For example, the GIMP article has a lot of references to gimp.org for release notes (which is technically not even allowed, as it a website promoting a product), but the GIMP source code itself is the source itself, and should be an indisputable source of whether or not a certain feature (or bug) exists. --Thoric (talk) 20:56, 11 December 2020 (UTC)

This would be equivalent to a primary source. As to what a piece of code does, that probably needs interpretation by someone, eg that some piece of code contains a bug. So that is where a secondary source would be needed to avoid original research by the code reader/Wikipedia writer. Graeme Bartlett (talk) 01:08, 1 January 2021 (UTC)

@Thoric, why do you think that citing "a website promoting a product" is disallowed? It wouldn't demonstrate notability (because it's not independent of the subject), but it's just as reliable as citing a politician's campaign website, or a tweet from a celebrity. WhatamIdoing (talk) 23:10, 5 February 2021 (UTC)

Per WhatamIdoing, some primary sourcing is allowed, we just don't want articles that only or primarily rely on sourcing from the primary website as that would be self-promotional/lack of notability.

But if there is software bug or the like, we want 3rd parties to explain that bug, we absolutely do not want editors to go and point that out , as that would be original research. --Masem (t) 23:15, 5 February 2021 (UTC)

If there is a software bug, we want someone who is not a Wikipedia editor to point that out. We don't need Wikipedia:Third-party sources for that; if there's a bug in Microsoft Windows, and they post a note about the bug on their corporate blog, then it's fine to cite that non-independent website.

I strongly agree that the overall article needs to be WP:Based upon sources that are independent and (ideally) secondary. WhatamIdoing (talk) 23:38, 5 February 2021 (UTC)

But that said, the severity of the bug could be an issue. Coming from video games, I have seen situations that have escalated to WP problems of players upset over a small, trivial bug that most sources have not discussed, that the developer has stated exists but do not have time for resolution on. We could source the dev to say this bug exists, but if no 3rd parties are talking about it, is it really significant? We don't want WP editors to be trying to determine what are the most critical bugs in a software program if there's no guidance behind that. --Masem (t) 23:44, 5 February 2021 (UTC)

The source would still be reliable for that claim. A non-independent source doesn't carry much WP:WEIGHT, so it might not be included in the Wikipedia article at all, but the source is still reliable and still able to verify the claim. The problem of being able to fully verify trivial, insignificant details in sources that are reliable for those trivial, insignificant details is exactly why Verifiability does not guarantee inclusion. It's reliable; we still might not use it. WhatamIdoing (talk) 02:38, 6 February 2021 (UTC)

If we're not talking about notability (something which is subjective), but talking about a fact (something which is objective, true, or false, and not an opinion), then why are primary sources not allowed, especially if we are talking about a feature of a software application? Why do we require use of a secondary source to cite such a thing, when anyone can see that it is true or not? --Thoric (talk) 17:46, 8 February 2021 (UTC)

Because primary sources usually require interpretation which is often a violation of Wikipedia policy. If they don't require interpretation, there is still a question of due weight: if something is only found in a primary source, does it really belong in an encyclopedia article? Just because something can be found in a reliable source doesn't mean that we should or must include it in an article. ElKevbo (talk) 19:55, 8 February 2021 (UTC)

Everything requires some modicum of interpretation -- including the comprehension and rephrasing of a secondary source. We accept that skilled wikipedians comprehend highly technical secondary sources on a regular basis, even when the subject matter may be too technical for someone without knowledge in that area to adequately comprehend. Plenty of wikipedians are skilled enough in different programming languages to at least be able to point out something like, "There is source code here indicating support for feature X (citing line of code). The documentation describes how to use feature X (citing application documentation)." I understand that saying something like, "When running the application, I can find and use feature X", is considered to be original research, but citing a reference to the source code is little different than citing a research paper. --Thoric (talk) 19:30, 10 February 2021 (UTC)

You are allowed to cite primary sources. However, you can only cite them for the content that a normal, non-expert, generally educated person can see in the source. This means that you can cite computer code (if it has been published). However, what you could actually write on the basis of that would sound more like 'Line 2238 contains a comment that says "You are Not Expected to Understand This"' than "The source code indicates support for feature X". This is because a generally educated non-expert won't be able to tell whether the source code indicates support for feature X by looking at the lines of code, and honestly, neither can you, unless you read the whole thing and determined that the code actually worked and wasn't broken or nullified by some other piece of code. (Think about the critical difference between laying the ground work for a feature that might appear in the future vs the feature actually being present.) WhatamIdoing (talk) 05:54, 14 March 2021 (UTC)

Keep context in mind when arguing claims

I'd like to add this to the project page with the above heading. The section WP:CONTEXTMATTERS or WP:RSCONTEXT deals with 'reliability of source' whereas this section would deal with the issue of biased generalisation of claims.

Theories dealing with subjective interpretations and implementations need to be contextualised in concrete conditions. This contextualisation needs to account for specific historic development and overlapping characteristics without trying to extrapolate to other contexts with different historic backgrounds and conditions. There are a set of topics that are contentious due to the nature of their implementation and corresponding effects. The contention could arise from the lack of contextualisation, the fallacy of mixing theory with practice, giving WP:UNDUE weightage to certain examples or because the point of view is not supported by WP:RS. Due to this, editors will almost inevitably confuse opposing views, based on different contexts, as violating neutral point of view. If you think you've spotted one, please spend a moment before trying to report them straight to the incident noticeboard and consider dispute resolution. Vikram Vincent 09:03, 5 March 2021 (UTC)

Based on the discussion below the one line I'd like to propose, while dealing with contentious topics or arguments, is: Clearly state the context of the claims that you are making. Vikram Vincent 10:48, 14 March 2021 (UTC)

I'm having some difficulty understanding the proposed text. Could you provide an example of how this would be applied to article content? — Newslinger talk 10:20, 5 March 2021 (UTC)

Newslinger I would prefer to keep the examples on my user subpage due to the contentious nature of the topics. User:Vincentvikram/Always_keep_context_in_mind_when_arguing_claims Vikram Vincent 10:46, 5 March 2021 (UTC)

I think this is covered by Wikipedia:No original research § Synthesis of published material (WP:SYNTH) in a more general way. Is there something missing from that policy (or other existing policies and guidelines) that would be covered by the proposed text? Since this guideline is focused on evaluating the reliability of a source, guidance on matters outside of this scope would probably fit better elsewhere. — Newslinger talk 11:06, 5 March 2021 (UTC)

Newslinger Noted. Which venue would be appropriate? Vikram Vincent 11:52, 5 March 2021 (UTC)

Personally, I think it would be best to complete the essay at User:Vincentvikram/Always keep context in mind when arguing claims before proposing any changes to policies or guidelines. Since the proposed text discusses a range of topics (content disputes, conduct disputes, and dispute resolution procedures), it would be challenging to include the text into a policy or guideline page in its current form. You would most likely have more success by breaking the proposed text into smaller portions that can be implemented as amendments to existing policy/guideline sections, if you can identify any policies/guidelines that would benefit from clarification.

The policy village pump is the most visible venue for policy/guideline suggestions, and my advice is to start a discussion there (after completing the essay) to determine if any portion of the essay content can be incorporated into a relevant policy/guideline. There is a possibility that there would be no consensus for any changes, and if that is the case, there is nothing wrong with adding the essay to the essay directory. ​— Newslinger talk 21:42, 5 March 2021 (UTC) Sorry, I missed a word. — Newslinger talk 08:04, 6 March 2021 (UTC)

Thanks! Vikram Vincent 04:36, 6 March 2021 (UTC)

• Oppose. Anything legitimate it has to say is redundant and not needed per Newslinger, the userspace essay is a WP:COATRACK for pro-Marxism-Leninism material, and the whole origin of this "contextualization" stuff is to aid and abet POV pushing. Per the history related at Wikipedia:Miscellany for deletion/User:Vincentvikram/Yes Marxism-Leninism: Originally, there was an ANI thread about a user who was pushing a Marxist-Leninist POV and WP:NONAZIS was invoked as a point of comparison, and Vincentvikram was defending the user; the reported user (not Vincentvikram) has since been topic banned after another ANI thread. [1] Vincentvikram created a draft essay after the first ANI thread called "Yes Marxism-Leninism" which was clearly a WP:POINTy counter to NONAZIS. I nominated it for deletion at MfD. Vincentvikram moved it to his userspace and later changed the framing to be about "contextualization", but the tide was turning against it; finally before it could be deleted by the discussion, he deleted the nominated page as CSD-U1 and recreated it (or something very similar) under the new name User:Vincentvikram/Always keep context in mind when arguing claims. I consider this a clear attempt to evade the MfD process, and if anyone wishes to nominate the new version, I'll support this; but I gave up at that point. Crossroads ^-talk- 04:18, 6 March 2021 (UTC)

Crossroads while your entire commentary is fine, your sentence the whole origin of this "contextualization" stuff is to aid and abet POV pushing is toxic and far from the truth. My PhD work has been around historical thinking and my interest stems from there. You attributing motive where none exists is against WP:AGF. I think you need to redact that sentence and apologise cause this is not how an editor treats others irrespective of the disagreement. Vikram Vincent 09:39, 6 March 2021 (UTC)

I don't think the March 5th proposal is helpful. It sounds confusing: All theories are required to have implementations in concrete contexts? Abstractness is banned? But then I'm not allowed to mix this theory with any practical applications, so I can't give you the concrete stuff? There's probably something in there, but this isn't ready. You might find it more effective to be specific, like "Don't claim that since one person one a specific lawsuit, then everyone else will win all the other lawsuits, too." WhatamIdoing (talk) 06:04, 14 March 2021 (UTC)

Hi WhatamIdoing! My thoughts were based on the use of causal claims from non-(strict)experimental studies or observations. aquillion gives a good explanation in the following section, What happens if a reliable source makes a false claim, while dealing with claims in statistics. The proposal is simple and can be boiled down to one line:Clearly state the context of the claims that you are making. This would be very useful to contentious topics. Vikram Vincent 06:39, 14 March 2021 (UTC)

I don't think that most editors will understand what you mean by "the context of the claims". I suspect that you mean something like "Don't write This drug costs US$0.14 but instead write According to Named Survey, the government of Costa Rica paid about US$0.14 per 250 mg pill in 2015". Other people might think that "the context" means that they need to say something about the political leanings of the survey organization, or that the whole thing assumes a capitalist system, or that this drug is cheaper than another drug, or any of hundreds of other things. WhatamIdoing (talk) 17:33, 14 March 2021 (UTC)

hmm yes, or any of the hundred other things seems to actually be the possible application... Vikram Vincent 14:44, 22 March 2021 (UTC)

What happens when a reliable source become geoblocked?

Given that edit, https://en.wikipedia.org/w/index.php?title=Istanbul_Airport&oldid=1012106079 , what happens if a source cannot be read and thus no longer gives proof? --Bouzinac (talk) 18:22, 14 March 2021 (UTC)

A geoblocked source is approximately equivalent to a source in dead paper. It is still a reasonable source to access, even if you can't. --Izno (talk) 20:49, 14 March 2021 (UTC)

I get the idea. I am all the more confused as https://www.dhmi.gov.tr/ : normal looks to tell it is not geoblocked...

Bouzinac, if you want to fix it, the instructions are at WP:DEADREF. This applies to any website that seems to be offline for any reason. WhatamIdoing (talk) 01:09, 15 March 2021 (UTC)

Washington Post and CNN

Suggest the reliability of these outlets be reduced given recent retraction of Georgia Sec of State story. Pkeets (talk) 22:21, 16 March 2021 (UTC)

Redactions are expected behavior for reputable outlets with journalism ethics. No need to change them because they redacted a story, just emphasizes why WP should be cautions of breaking news per WP:NOT#NEWS. --Masem (t) 22:30, 16 March 2021 (UTC)

Of course they're not necessarily reliable, and retractions don't make them good when they've based their stories on anonymous unreliable sources. Like every other news outlet they should have been treated with some caution, and the context should have been examined, before Wikipedia repeated their stories (as happened with Trump–Raffensperger phone call). But why bring this up here rather than on WP:RSN? Peter Gulutzan (talk) 22:49, 16 March 2021 (UTC)

On the contrary, we expect reliable sources to actively and quickly correct their mistakes. ElKevbo (talk) 00:16, 17 March 2021 (UTC)

I agree with @Masem and @ElKevbo: media outlets that issue corrections and retract stories are the reliable ones. The ones that never admit to mistakes are the ones to worry about. Retractions and corrections are common in scientific journals, too. (Obviously, we should stop citing any individual retracted article.) WhatamIdoing (talk) 01:34, 17 March 2021 (UTC)

@Pkeets: Wikipedia is full of activists that will NOT allow any wrongthing to seep through. Fox News is downgraded but nobody thinks fieryfalse but mostly peacefultrue story like this is a problem. Only sources on the "right side of history" will be allowed as top sources on wikipedia, and WaPo and CNN have constantly agreed with the groupthink of the remaining wikiactivists here. They never publish fake news, misinformation, disinformation, because they are on the right side of history, unlike Fox News. Therefore they will never be downgraded, as multiple people here have shown you. 2601:602:9200:1310:1596:19CF:A497:D49F (talk) 18:06, 20 March 2021 (UTC)

Note the current defamation suit brought by Project Veritas against NYTimes: A judge ruled this week for PV in the Motion to Dismiss, calling out malice in the process of injecting opinion into news articles and representing it as fact. To avoid potentially libelous material, that suggests source articles should be evaluated in the future for this failing. Pkeets (talk) 19:33, 20 March 2021 (UTC)

Well, if our own article on Project Veritas is anything to go by... ProcrastinatingReader (talk) 20:12, 20 March 2021 (UTC)

Also, source? ProcrastinatingReader (talk) 20:15, 20 March 2021 (UTC)

Rulings in preliminary motions are made with specific assumptions that may not be substantiated later in trial and should not be used to make significant editorial decisions in Wikipedia (or anywhere else).

And why are we continuing this thread of discussion anyway? ElKevbo (talk) 20:22, 20 March 2021 (UTC)

The ruling though the markup is from a site "friendly" to PV. But I've confirmed that's the right case and that the NYCourt system is reporting that they did deny the Times' motion to dismiss. But as a order to refuse dismissal, that's not case law, only prelim ruling that may not hold in a final review. That said, if that ruling holds true and through subsequent appeals (which I feel it won't), it does point to the problem of this "accountability journalism" that AP has taken the lede in, the mix of factual and op-en without the clarity of being op-ed. --Masem (t) 20:49, 20 March 2021 (UTC)

The newspaper doesn't do "op-ed". The "op" in op-ed stands for "opposite", as in "the newspaper's editorial board prefers the Demican this election, but because we think that being fair is good for business, we're going to let someone with the 'opposite' view explain on the next page why you might want to vote for the Republocrat in a separate opinion piece".

That is: there are two basic kinds of opinion pieces: the "editorial", which is written by "the editors", and the "op-ed", which is written by basically anyone else. WhatamIdoing (talk) 02:14, 22 March 2021 (UTC)

@Pkeets: can you clarify what retractions you are referring to? This seems to keep coming up but I cannot find any recent retraction, or even a correction, from the Washington Post let alone CNN related to the Georgia Secretary of State. [2] There was a correction by CNN of a different story, the one about Trump and Georgia's election investigator. This involved the claim Trump said to "find the fraud" and also that the investigator would be a "national hero". I believe the investigator was working for the Secretary of State's office, but the secretary of state himself was not otherwise involved in the story. (Well I think I think he may have confirmed the phone call happened too, but that wasn't the disputed part. AFAWK, he wasn't the source for the erroneous quotes.) Other sources who relied on the Washington Post also corrected their stories [3] Some crazy sources incorrectly claimed that the find the votes part of the Trump–Raffensperger phone call was retracted which was dumb, since AFAICT, that phone call has had the audio and transcriptions available since basically the beginning. But anyway given the high profile correction combined with this craziness, I'm having trouble finding any recent retract which deals with a story about the Georgia Secretary of State. Nil Einne (talk) 11:50, 21 March 2021 (UTC)

I believe it is this story: [4]. --Masem (t) 12:55, 21 March 2021 (UTC)

Has it now become acceptable at Wikipedia to alter someone else's posts and then make raucous comments about it? I'm posting in good faith about issues I feel are important, and I don't appreciate harassment. Pkeets (talk) 01:40, 17 April 2021 (UTC)

Nintendo Everything development interviews

I've asked this question at WT:VG but wasn't really satisfied with the answer.

Nintendo Everything is considered an unreliable source per WP:VG/RS, namely because it's a fan blog site in broad terms. However, they often interview video game developers, and more often than not I see these interviews trying to rewrite an article (this time around Mario & Luigi: Paper Jam). So, would these interviews be considered WP:PRIMARY and allowed, or would they still be unusable? And follow-up question, would it meet WP:GA standards? Panini^🥪 17:00, 19 March 2021 (UTC)

@Panini!, please move this question to Wikipedia:Reliable sources/Noticeboard. WhatamIdoing (talk) 17:14, 19 March 2021 (UTC)

Yikes, was this not the correct place to ask this? Oh, looking at the talk page header now. Sorry about that! Panini^🥪 17:20, 19 March 2021 (UTC)

Dietz Press?

Doing a GA review of Siwanoy, there's a source I'm dubious about:

Barr, Lockwood (1946). Ancient Town of Pelham, Westchester County, New York. Richmond, Va.: Dietz Press. p. 13.

I can't find much on Dietz Press. Their website (https://www.dietzpress.com/) gives precious little information. The google book entry also mentions Higginson Book Company (https://higginsonbooks.com/) which also doesn't say much. I'm guessing this is a vanity press. Anybody know anything about Dietz? -- RoySmith (talk) 15:19, 24 March 2021 (UTC)

@RoySmith, I don't think your instincts have led you in the right direction here. I think it is a small press, but a real one. These websites refer to books printed in the 1940s,[5][6] including a highly regarded work for stamp collectors.[7] While vanity presses did exist in previous centuries, it was not common. This website describes it as a publisher of children's and young adult books.[8] And if they really were a vanity press, they have made it impossible to figure out how to let people pay them to print books for them. WhatamIdoing (talk) 02:06, 27 March 2021 (UTC)

Maybe you were looking for WP:RSN? -- Emir of Wikipedia (talk) 16:58, 27 March 2021 (UTC)

Nature article on "the systemic production of falsified research"

I thought that folks would want to be aware of this.

• Else, Holly; Van Noorden, Richard (2021-03-23). "The fight against fake-paper factories that churn out sham science". Nature. Retrieved 2021-03-29.

Peaceray (talk) 17:07, 29 March 2021 (UTC)

The "paper mill" problem looks to me like part of a general problem of the volume of publishing being too great for the gatekeepers to inspect properly. If pre-publication peer reviewers could not check the papers for being from bad sources, they surely could not perform a more difficult check for quality or even authenticity of research. Yes, Wikipedia, though much smaller than the sum of scientific research papers accepted annually by even one major scholastic society, suffers similarly as it grows, yet organized efforts to attract new editors mostly concentrate on a hope for adding to the article count. Jim.henderson (talk) 14:55, 1 April 2021 (UTC)

Thanks for the link. Also related: predatory publishing, many journals aren't indexed at relevant places, articles on certain topics may strangely be found in journals that appear to be on other topics (usually an indication it had difficulty to pass peer review in its field), some journals are on warning lists (notable was Beall's, WP has some efforts like WP:RSP and WP:CRAPWATCH), on Wikipedia a particular problem to avoid is citogenesis, astroturfing is increasingly common (disguised fronts for propaganda), then of course, it's unnecessary to mention the importance of secondary reliable sources to evaluate primary sources... —PaleoNeonate – 10:54, 4 April 2021 (UTC)

Question about reliable sources and liberal bias

See also: Wikipedia talk:Reliable sources/Perennial sources § Question about reliable sources and liberal bias

I have read Wikipedia articles on a regular basis (usually daily) for several years now. What I've noticed is that the large majority of reliable sources are liberal/left-leaning news media outlets. This is a fact based on simple analysis (not opinion). Why are so few right-leaning/conservative news outlets considered reliable? I could be wrong on this but I believe WSJ is literally the only right-leaning outlet considered "generally reliable" (i.e. the green check) according to perennial sources page, while there are at least 15 or 20 liberal/left-leaning sources. How can Wikipedia be considered a neutral and balanced encyclopedia when this is the case? K.Q.1997 (talk) 22:37, 7 April 2021 (UTC)

I answered this at Wikipedia talk:Reliable sources/Perennial sources#‎Question about reliable sources and liberal bias, didn't see before posting that that the user had posted the question in two places. – Muboshgu (talk) 23:04, 7 April 2021 (UTC)

Checking back and noticed none of the above editors answered my original question about the bias found throughout Wikipedia on American political articles. Instead people brought up stuff about how Trump is a liar (which I agree with) and how conservative media are all in a cult and embrace alternative facts. This makes me thinks the issue could be the editors themselves who have a partisan bias and are able to skew Wikipedia in one direction. What are the credentials of people who edit Wikipedia articles? Also to be fair I don't follow politics of other countries but as a person who identifies as a centrist on most issues/liberal on social issues like abortion & LGBT rights, the bias in favor of one side is pretty obvious. K.Q.1997 (talk) 22:40, 14 April 2021 (UTC)

There are fewer center-right publications in the American media landscape. To increase this number, the general public would need to support high-quality center-right publications outside of Wikipedia and ensure that they stay in business. On Wikipedia, "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", keeping in mind that "The relative prominence of each viewpoint among Wikipedia editors or the general public is not relevant and should not be considered" (WP:NPOV). — Newslinger talk 04:14, 15 April 2021 (UTC)

Something about NYT, WaPo and CNN

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

---

These two media outlets seem to be consistently failing to verify what they hear from unnamed sources. See the articles from multiple media outlets this week about the unreliability of reports (broken by the NY Times) about the Russian bounty program. Pkeets (talk) 14:30, 16 April 2021 (UTC)

--Orange Mike | Talk 14:51, 16 April 2021 (UTC)

See articles on the Project Veritas videos here where a CNN staffer admitted the outlet's bias: [9] [10] [11] [12] — Preceding unsigned comment added by Pkeets (talk • contribs)

Oh, this is hilarious: we are supposed to deprecate a reliable source, based on an "expose" from an unreliable fraudster and conman???? It is to laugh! --Orange Mike | Talk 14:51, 16 April 2021 (UTC)

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