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.

Archive 280

←

Archive 283

Archive 284

Archive 285

Archive 286

Archive 287

→

Archive 290

For the interested, Haaretz on WP:s deprecated sources. Gråbergs Gråa Sång (talk) 22:37, 9 January 2020 (UTC)

Yup, Facebook believes in freedom of speech, we don't: WP:NOTFREESPEECH. Also, “anyone can edit” should be changed to “anyone can edit constructively”. Tgeorgescu (talk) 22:47, 9 January 2020 (UTC)

Nah... anyone can still edit... it’s just that nonconsructive edits will be removed or amended by other editors. Blueboar (talk) 23:36, 9 January 2020 (UTC)

Wikipedia, the encyclopedia that anyone can attempt to edit _signed, Rosguill ^talk 23:45, 9 January 2020 (UTC)

Then certain things cannot be said, thus no freedom of speech. (soapbox alert) this at its heart if the problem with certain attitudes towards absolute Free speech. It is regarded as not only the right to be wrong, but the right to outright lie. This may be partially true, but us not allowing it is not different from me not being able to right for a major newspaper or walking into Number 10 to call Boris a liar to his face (or come to that to go into any of your houses and launch a foul mouthed tirade at your kids).Slatersteven (talk) 10:32, 10 January 2020 (UTC)

Haaretz's choice to refer to deprecation as "delisting" halfway through the article is odd, and their phrasing suggests that we have some sort of immutable hierarchy of validity for various types of publications, but otherwise it's a good read. Their decision to counterpose our against Facebook's content moderation is interesting, and in an ideal world would motivate Facebook to try harder, although I'm not sure the comparison is entirely fair because the goals of the two platforms are rather distinct. _signed, Rosguill ^talk 22:50, 9 January 2020 (UTC)

• What bothers me is that this article shows how most people have misunderstood what “deprecation” means... it is viewed by too many as being a “ban” on using certain sources. However, it is SUPPOSED TO BE a more nuanced “limitation” on usage. Almost every RFC that resulted in “deprecation” includes carve outs and exceptions... situations when the source IS acceptable. We need to do a better job of explaining that. Blueboar (talk) 23:46, 9 January 2020 (UTC)

• Based on the way that some editors are removing the two deprecated sites mentioned there from articles, it clearly does not mean limitation on use to them. I've seen editors remove it and add {{cn}} for innocuous facts. Walter Görlitz (talk) 23:51, 9 January 2020 (UTC)

• Yeah it is a defacto ban. If it was meant to be or not really does not seem to matter unfortunately. PackMecEng (talk) 23:52, 9 January 2020 (UTC)
• I've seen the same thing as Walter. It's one of the reasons why I've universally opposed deprecation. Springee (talk) 00:52, 10 January 2020 (UTC)

• All content, even innocuous facts, should be sourced to RS, unless they are clearly "the sky is blue" type facts. If a fact is only mentioned on unreliable sources, then we can't be sure it's true, and even if true, it doesn't have enough due weight. See my comment below. -- BullRangifer (talk) 00:57, 10 January 2020 (UTC)

• May I suggest you raise that point to the film project and TV project? It is their collected opinions that once a film has been released or TV program has aired the need for sources has ceased, going so far as removing sources that have been supplied up to that point. There are other projects that have taken similar stances, but those are the two that irk me most. However, WP:V does not support the ubiquitous need for sourcing you espouse. It only goes so far as to state "all quotations, and any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation that directly supports the material." Walter Görlitz (talk) 02:20, 10 January 2020 (UTC)

• Walter Görlitz, I find that rather disturbing. We don't accept IMDB because some of its content is crowdsourced and thus not reliable enough. Is "no source at all" somehow more reliable? They are on a slippery slope, and inline citations must be provided if any content is challenged.

I'm not sure if there is any discrepancy between what I wrote and the WP:V policy. The "inline citation" must still come from a RS, and that applies to all article content. Otherwise, basic facts that are of the "sky is blue" variety don't need such citations.

BTW, I'll happily modify my statement, so please continue to discuss and critique it. -- BullRangifer (talk) 02:33, 10 January 2020 (UTC)

All the film and TV people are saying is that a film, TV episode or book is its own reference. But this only applies to what is actually in it. If you are acknowledged in the credits, then that is fine; but going beyond that by saying that we heard somewhere that you were a ghost-writer is WP:OR. Hawkeye7 (discuss) 02:39, 10 January 2020 (UTC)

Well, those are primary sources, so one must be careful. We can use primary sources for uncontroversial facts, but not much else. For example, we cannot use them for interpretation. Without secondary RS, we cannot know what weight to give that information, so such content can be challenged. -- BullRangifer (talk) 03:45, 10 January 2020 (UTC)

Deprecation should have only one exception: "can only be used in its own article, and then only if it's not unduly self-serving." All our content, facts and opinions, must be sourced solely using RS, with that being the only exception. Since my research includes reading content from both RS and unreliable ones, I will often find details in unreliable sources which might be factual, so I then search RS for those facts. If I can find them, I'll use the RS. If they are not found in RS, I can't be certain they are factual, and even if they are, they don't have enough weight to be mentioned, since use in RS guides our determination of due weight. -- BullRangifer (talk) 00:20, 10 January 2020 (UTC)

Just to give one example of an exception specified in RFC: Daily Mail - old reporting (from historical days when the Mail was under different management) is considered reliable. So, at a minimum, we are SUPPOSED to check when the report was published before we remove it. Context is important. It bothers me that people are ignoring this. Blueboar (talk) 00:34, 10 January 2020 (UTC)

Good point. Is there a cut-off date at the deprecation notice? -- BullRangifer (talk) 00:44, 10 January 2020 (UTC)

No, it's just something someone said in passing in the RFC. Nor is it clear if this was before or after "HURRAH FOR THE BLACKSHIRTS" - David Gerard (talk) 07:19, 10 January 2020 (UTC)

• In Wikipedia terms, this is pretty straightforward: if, like OANN, a site has a history of publishing conspiracy theories or falsehoods without robust fact-checking and retraction, then including any content from that source puts Wikipedia editors in the position of arbiters of truth. We have to decide which stories are true and which are false, rather than allowing the normal process of journalistic checks and balances to do this. We can only do that by seeing if sources we consider reliable, say the same thing. And if they do, well, we should be using the reliable sources instead.

That's what deprecated means. You can't trust it without verifying from another, more reliable source, in which case why would you not use the more reliable source instead? Deprecated is not the same as "generally unreliable", it's a specific outcome we reserve only for sites that are sufficiently deceptive that we should never use them. I personally would never use the Washington Examiner for anything, but it's not deprecated. WorldNetDaily is. That distinction is valid and entirely consistent with the five pillars. Guy (help!) 11:36, 12 January 2020 (UTC)

A problem for Wikipedia

The point is made in Haaretz that the preponderance of right-wing sources among our deprecated list is interpreted by conservatives in much the same way as Twitter bans on hate speech: most of the affected sources are conservative, therefore this is anti-conservative bias. Obviously it's not, but there are more liberal Wikipedians than conservative (for reasons that are inherent to what Wikipedia is) so we need to work doubly hard to be sure we hold all sources to the same standard.

I have given this a lot of thought recently, and the discussion above re. the Haaretz piece highlights an issue I think we need to address more clearly: reliability versus bias (or, as Springee put it above, WP:V versus WP:WEIGHT).

• A biased source is likely to be treated as requiring attribution.
• An unreliable source is likely to be deprecated.

Bias and unreliability correlate (albeit rather asymmetrically, see below). We have deprecated only a few of the hyper-partisan left sources because, with a few notable exceptions, they are still more accurate than hyper-partisan right sources, so we treat them as biased (attribution required) rather than deprecated. Not always: AlterNet is generally more unreliable than the deprecated Breitbart. IMO AlterNet should be deprecated, and you could make an argument that based on current data Breitbart could be "generally unreliable" rather than deprecated (albeit that this would make little practical difference).

Our challenge as Wikipedians is to demonstrate that our decisions about source reliability are based on solid, rational, empirically factual grounds, and not on our politics. Part of that is ensuring that sources are treated similarly according to their position on objective bias / accuracy axes. Some sources only rate on political leaning, and that's dangerous for at least two reasons:

1. Fox News online and CNN cable have similar levels of bias but fact-checking shows CNN cable to be more reliable than Fox News;
2. Wikipedia treats bias and accuracy separately.

Bluntly, I do not think we should ever use any source outside the Ad Fontes "green box of joy" as a source of fact. That would put The Daily Beast, CNN cable, MSNBC, Mother Jones, The Intercept and so on outside the realm of usable sources for facts - at least in articles on current politics. In my view, nothing of value would be lost.

Background
The following discussion has been closed. Please do not modify it.
Right-partisan sources are more biased, more consistent with each other, and less factually accurate The way right-partisan media works has changed significantly since 2015. I have often referenced Network Propaganda, an excellent book that I think every regular here should read. This documents the different incentives that have driven asymmetric bias in American media. What the data shows is that right-leaning media is currently dominated by a positive feedback loop where social shares and hence advertising revenue are driven by the extent to which a source agrees with other sources in the bubble. Fox suffers penalties if it published accurate information that conflicts with the conservative narrative. Mainstream media is dominated by a negative feedback system of fact-checking. Washington Post will suffer if it published factually inaccurate stories however much they may align with liberal talking points. Hyper-partisan left sources are as ideology driven as hyper-partisan right, but they are likely to be contradicted or fact-checked by mainstream sources and, crucially, their audience is almost certainly consuming those sources, whereas consumers of conservative media often (and in many sources not just often but generally) are not. Sean Hannity probably does not care if the Washington Post rates his claims as Pants-On-Fire, but Rachel Maddow probably does. If you look at the Ad Fontes chart, a useful guide which rates sources on a continuum by both accuracy and bias on separate axes, what you see is exactly what that asymmmetry would predict: that the average of right-leaning sources is significantly more biased and significantly less reliable than the average for left-leaning sources. With the exception of the New York Post, IJR and Reason, pretty much all common right-leaning sources are now outside the region of mainstream journalism, and all have significant issues with factual inaccuracy and promotion of conspiracy theories. On the Ad Fontes chart, starting with New York Post and working left, there's a continuum of first increasing and then decreasing accuracy as you move from leans right through neutral and into leans left and on to partisan and hyper-partisan left. But if you go to the right, there's a huge gap before you hit the right-leaning sources. And 40% of American readers live in a world where these sources are True and anything that says otherwise is Fake News. I've read several scholarly sources that agree on this. That feeds into conflict here The result is that conservatives generally see unanimity between the sources they consider reliable, which may be unanimously contradicted by sources we consider reliable. A lay understanding of the psychology of cognitive dissonance is sufficent to understand this, and to know that we cannot fix it. We are doomed to have these debates forever because we cannot change the underlying mismatch between mainstream reality and that of the conservative media bubble. Mainstream has come to be seen as the opposite of conservative, and that's a real problem for us at this page, and for any editor active in current politics.

Proposals

I guess what I am arguing is that we should do a few things to formalise the way we conduct discussions that may result in widespread changes to the project and external commentary.

1. Segregate WP:RSP into two or three blocks: one for general websites (VHChartz, TV Tropes and the like), one for media, and one for actual fake news (News Front etc).
2. Introduce a template for RfC discussion of sources (that is, the entire source rather than one or a few uses of it for a specific fact).
   1. Include an assessment of impact. Deprecating WorldnetDaily, Occupy Democrats or InfoWars has no real effect on Wikipedia because no competent editor would use them anyway, but deprecating Breitbart and the Daily Mail had a substantial impact.
   2. Be explicit about the difference between bias and accuracy, and include this in debates.
   3. An RfC for any source with more than $THRESHOLD number of uses (perhaps 1,000?) should be advertised at WP:CENT.
3. Agree a consistent set of tools for assessing reliability, that minimises the possibility of political bias - so, look for right-leaning fact check sites, and exclude all fact-checkers that do not assess media from both sides of the spectrum.

What do others think? Guy (help!) 14:14, 12 January 2020 (UTC)

I'm not seeing the argument for any sweeping changes (and I'd even disagree with your "a problem for Wikipedia" framing.) The Haaretz piece largely praises our system; the way we do things now is working, on the whole. Our articles - the high-profile ones, the ones that get enough attention and focus to avoid getting hijacked by a single committed editor, anyway, which is the real risk but only happens on peripheral topics - are largely accurate, reliable, and evenhanded. Yes, if you zoom in close on the talk pages there's a ton of ink being spilled in the AP2 topic area, but if you zoom out most of that is actually pretty minor - it's being spilled over minor wording tweaks or a sentence or two that wouldn't be utterly unacceptable either way. Compared to the vast gulf in the world as a whole, we've actually done pretty well at finding consensus. A lot of other coverage says similar things - that Wikipedia has remained largely reliable in an era where many other online information channels are fracturing or showing their flaws. Our system is, mostly, working, so I don't see how that supports an argument for changing direction. Your argument mostly seems to be that people who live in a bubble of unreliable sources are going to look at our articles and see them as biased, but that's always going to be the case - it has nothing to do with the arcane intricacies of how we handle reliable sources, and everything to with eg. our article on Global Warming not saying what the facts-optional talking head on TV tells them it should. That is not something we can fix; the problem is not on our end. --Aquillion (talk) 08:54, 13 January 2020 (UTC)

Aquillion, don't get me wrong, I am really not overblowing this, but there is no doubt that conservatives perceive a serious bias against them as a result of the majority of deprecated sources being conservative - there's a perception that we deprecate due to bias, and do this asymmetrically, whereas in fact we deprecate due to accuracy and it is merely coincidental that the most biased and most conservative sources are also the least accurate.

That's why I'd like to formalise the RfC process: to show that the assessment is fair. I agree that the issue of conspiracy theorising and outright falsehood in the right wing media bubble is very much not our problem to fix, but it is our problem to manage and I think we could do a better job of it. I am encouraged that you don't see this as a major issue, though - I would be happy to discover that I am being oversensitive to this form of criticism. Guy (help!) 09:25, 13 January 2020 (UTC)

If you followed that path, here is what you would see: First, to the extent that anyone pays any attention to the change at all, it would be covered, in the sources that the people you're trying to convince trust, as "Wikipedia admits it has a problem with bias against us, takes piddling efforts to fix it." Second, perceiving the change as a sign that sufficient pressure can get Wikipedia to reflect their views, the complaints about bias would increase sharply in volume. Third, nobody who thought the system was unfair before would think it was fair now, because what they actually want is, again, our articles on the hot-button topics they care about to reflect their view of the world (and because, again, by changing policy in response to complaints that it is unfair, you've given people with strong political preferences about our content and policy an added incentive to continue to be intransigent in hopes of extracting more concessions.) And, finally, all those practical issues aside, it would be a bad rationale to change our policies - our goal is not to appear fair, our goal is to be fair. One of the reasons we've been so successful at maintaining factual accuracy when so many other sites have failed is because we categorically reject the false balance of prioritizing the appearance of accuracy. Our job is to reflect the best available sources, not to sell Wikipedia to people (though I think that the site's success shows that maintaining that laser-tight focus on getting things right will, ultimately, be successful at winning people over.) --Aquillion (talk) 12:45, 13 January 2020 (UTC)

Clarity and thoughtfulness surrounding the deprecation of sources seems like an excellent idea, but maybe not approaching from this direction? Conservatives may often be correct to "perceive a serious bias against them", but are so very often wrong in identifying the causes and horrid at arguing solutions. Sourcing policies on Wikipedia may allow bias to flourish in some areas, but as Aquillion points out working towards the goal of neutrality is "mostly working" on high profile articles. I'd disagree with him tho that this is the "system" working, i would say it is the result of competent and thoughtful editors expending a great deal of effort to make the system work.—eric 13:26, 13 January 2020 (UTC)

I mostly agree with Aquillion's comments here, although I would consider perception to still be an important secondary consideration. To the extent that there's an issue here, I would say the easiest way to respond is probably to emphasize the counterexamples that people can point to. The articles tend to mention Occupy Democrats but I've never seen anyone cite MintPress News in this context yet; likewise, Telesur doesn't seem to get mentioned despite being very left-wing by the standards of the American "left-right" axis. If Grayzone (currently still under discussion above) is deprecated then it would probably also be a good example. If that approach is insufficient, or if those examples are challenged for some reason, I suppose there is also the option of making it clearer to outside observers by specifically deprecating a couple more of the thousands of other sources that unambiguously qualify for it. Fake news websites aren't limited by political alignment, after all. Sunrise (talk) 23:22, 13 January 2020 (UTC)

I do not think we need to separate out fake news. At the end of the day any actual fake news site should be depreciated if not out right banned. I think (part from perennial "but I like it" arguments) RSN works fine.Slatersteven (talk) 19:05, 14 January 2020 (UTC)

• Proposals 2.0, 2.2, and 2.3 are great ideas, in my opinion. Currently, editors are copying The Daily Caller RfC because it's linked in the instructions at the top of the page, as determined in the 2019 header text RfC. An RfC template could be edited as improvements are identified. For instance, it would be helpful to remind the RfC starter to provide links to past discussions and prominent examples of usage, and it would be helpful to remind RfC participants to consider context. I agree that bias and reliability should be evaluated independently, and that RfCs on frequently cited sources should be advertised on WP:CENT. — Newslinger talk 09:15, 16 January 2020 (UTC)

It was a good read, but paragraph 4 had a falsehood: "Like all decisions at Wikipedia, the addition of a news outlet to the list is determined by a community vote." Everybody makes mistakes. So, I see the site-by-site approach as almost doomed to failure. I agree Wikipedia has enormous problems, and I would not trust articles for anything. The only way to dig out the full story on most topics with any controversial aspects at all is to carefully review edit histories and talk page histories, and this defeats the whole purpose of having articles. I know equal time is not a thing here, but maybe every article deemed to involve controversy should have a certain, limited amount of space available for the proponents of different sides (possibly more than just two) to make their best cases, like debates. As it stands, the more persistent editors suppress almost everything they disagree with, almost including the existence of disagreement at all. Sorry to be so negative. -- Yae4 (talk) 19:52, 20 January 2020 (UTC)

Your suggestion is incompatible with the undue weight policy, which states: "Giving due weight and avoiding giving undue weight means articles should not give minority views or aspects as much of or as detailed a description as more widely held views or widely supported aspects. Generally, the views of tiny minorities should not be included at all, except perhaps in a 'see also' to an article about those specific views." There are other wikis that are more sympathetic to fringe views, but Wikipedia is not one of them. — Newslinger talk 20:05, 20 January 2020 (UTC)

When discussing possibly making major changes to solve major issues, referring to WP:IAR seems better than referring to current norms. I acknowleged "equal time is not a thing here," but that doesn't preclude allowing some amount of space to be reserved for presenting "minority" views, to lessen accusations of suppression. Anyway, the issue isn't just "theories" or "science" or even "ideas". Here is a simple factoid example. Let's say someone wants to "follow the money," and answer the question, how much is spent on mainstream climate science research and politics, compared with how much is spent on non-mainstream?

Climate_change_denial#Funding: roughly $900 million

Global_warming_controversy#Funding: TL;DR but has a couple numbers, 1.2 and 16 million

Versus:

Global_warming a couple mentions of funding of scientists who disagreed with "scientific consensus," but no number, and no link, so had to do a search to find:

Climate_finance#Flows_of_climate_finance "estimated that climate finance reached $437 billion"

Shouldn't it be easier to find both numbers in an encyclopedia, maybe in one place? -- Yae4 (talk) 15:08, 21 January 2020 (UTC)

"Both numbers"? You've pointed to four types of numbers, and seem to be conflating different things: CC denial funding goes to misinformation and advertising, Climate finance is “finance that aims at reducing emissions, and enhancing sinks of greenhouse gases ...." etc. so is financing action to stabilise the climate, not sums funding scientists of whatever views on CC. Of course, if a reliable secondary source puts together the comparison you seem to be suggesting, then you can cite that. . . dave souza, talk 17:48, 21 January 2020 (UTC)

Dave souza, Previous editors took the trouble to include an easily found number for "denial" funding, but in Wikipedia I haven't been able to find a comparable number for "mainstream" science and politics funding. I did misunderstand the purpose of "Climate finance," so thanks for that. -- Yae4 (talk) 18:33, 21 January 2020 (UTC)

Yae4, perhaps you misunderstand the purpose of this page, which is about reliable sources – "original research isn't accepted here, and you can't ask others to do it for you. If you're interested in the complexities of funding, you may find Willie Soon's arrangements informative, but here the discussion is about assessing good sources. . . dave souza, talk 22:02, 21 January 2020 (UTC)

Dave souza, I understand the purpose, but the process and measures for RS determination look completely arbitrary and biased in application, so I don't understand that, for sure. I'd already seen that article, but thanks again. -- Yae4 (talk) 22:44, 21 January 2020 (UTC)

• My only contribution to this is a famous quote by Stephen Colbert: "Reality has a well-known liberal bias". This is not necessarily true in the world in general, but it certainly is true in the context of these sources. So if Bob ThreeHats is on his laptop, raging that we're not letting him says that Mexicans are planning to invade the USA to be part of George Soros' army based on something he saw on InfoWars/Rush Limbaugh/Breitbart, I really don't see why we should concern ourselves with the appearance of being fair to a person who's idea of being fair means endorsing his prejudice against facts.

Neutrality isn't painting a picture in dull grey when one side says black, and the other says white. See WP:LUNATICS, but apply it to politics instead of science. Headbomb {t · c · p · b} 23:21, 21 January 2020 (UTC)

Newslinger, Another comment Re: "Fringe theories" as applied to climate change. Wikipedia:Fringe_theories has existed since January 2006,1 i.e. 4 years before Wikipedia:Arbitration/Requests/Case/Climate_change, but there is only one use of the word "fringe" in the Arbitration. By end of February 2006 (new article somewhat stabilized), among 3 examples, climate change was not yet discussed in "fringe theories."2 Today, there is still no mention of climate change, aside from linking to the 10 year old Arbitration case. 3 Given that "fringe theories" is now being frequently applied to skepticism over climate change predictions etc., shouldn't the Wikipedia position be fully explained in some detail at Wikipedia:Fringe_theories? (Aside: I would suggest including a study that documents how many actively publishing, "climate scientists" have reduced their "carbon footprints" to zero or similar, demonstrating true belief, but I can't find one.) I would appreciate if someone who can explain the Wikipedia position would add a section. -- Yae4 (talk) 16:06, 23 January 2020 (UTC)

The fringe guideline isn't particularly at issue in ARBCC, and while the guideline includes examples, it's not a prescriptive list and it applies equally to topics not mentioned as examples. What is at issue is weight policy, as covered by WP:ARBCC#Undue weight. As for your aside suggesting "including a study that documents how many actively publishing, 'climate scientists' have reduced their 'carbon footprints' to zero or similar, demonstrating true belief" merely suggest that you don't understand science. So not a good idea for a section. None of which has any obvious relevance to discussion on quality of sources. . . dave souza, talk 17:51, 24 January 2020 (UTC)

Ouch, add one more not-so-subtle personal attack by an admin to the count. Is there a list of designated admins or other editors who are certified as "qualified" to guide science articles? Saying the wiki-bias problem is about sources is a side step. The real issue is wikipedia has fixed positions on some topics, and "fringe theories" and banning sources with "voting" are just two of the wiki-lawyering methods used to pretend otherwise. -- Yae4 (talk) 14:30, 26 January 2020 (UTC)

Why would you think that randomly introducing an unsourced non-sequitur that is quite obviously designed to imply some sort of hypocrisy or wrongdoing among climate scientists would be relevant to, or well-received in, this discussion? NorthBySouthBaranof (talk) 14:53, 26 January 2020 (UTC)

Why would you ask a fallacious Complex_question? -- Yae4 (talk) 19:27, 28 January 2020 (UTC)

• I don't really see that the Haaretz piece actually indicates anything on our part that really needs solving. Disinformation and conspiracy mongering are currently a larger problem on the political right than the political left, though not to say that it is non-existent on the political left because it isn't. Much of that is probably historical accident, and you can certainly find plenty of instances where the situation was reversed (the French Revolution and the Soviet Bloc spring to mind). Neutrality looks biased to people who are biased. That's not really an "us problem". GMG^talk 16:47, 23 January 2020 (UTC)
  • Something that I would argue does need solving is the fact that, over the course of 2019, the number of deprecated sources went from 6 to 26. I don't think 2019 was a worse year for fake news than 2018 or 2017. It looks more like editors are starting to ignore the intermediate levels of classification. Connor Behan (talk) 18:50, 27 January 2020 (UTC)
    • What I think has happened is that as some sources like the Daily Mail became deprecated, some editors went to other sources to pull similar stories as the Daily Mail; in those types of cases, other newspapers that are going to share the same type of information as the Daily Mail are going to be just as poor as the Daily Mail for the most part. So as more of these new sources get used, we see the need to also deprecate those. It's a small snowball effect. --Masem (t) 19:10, 27 January 2020 (UTC)

• Also, deprecation is still a fairly new practice, and the process takes time to go through. I would argue that the deprecations in 2018 and early 2019 mostly acted as test cases that defined how the practice was developed, building on the original precedent of the Daily Mail, and so there was a backlog of sources that we're still going through today. Besides, it's not as if 26 is more than a small fraction of the fake news websites out there (and of course, fake news is not the only cause of deprecation), so we aren't likely to run out of valid candidates any time soon. Sunrise (talk) 09:26, 31 January 2020 (UTC)

An example

David Gerard (talk · contribs) writes "The source is deprecated - see WP:THESUN - and should not be used at all - even as a placeholder. So I would like to discuss this clearly here. I have seen Gerard remove the source, even for uncontroversial sports reporting as well. Walter Görlitz (talk) 17:33, 14 January 2020 (UTC)

See the sun RFC below.Slatersteven (talk) 17:35, 14 January 2020 (UTC)

No. This is the correct location. As you can see, @Blueboar:, @PackMecEng:, @Springee:, and @BullRangifer: have voiced opinion here and so the discussion belongs here, not in a new location that David Gerard wants to discuss it at. Walter Görlitz (talk) 18:59, 14 January 2020 (UTC)

Its an example of what? what has the sun got to do with the above?Slatersteven (talk) 19:06, 14 January 2020 (UTC)

Did you read the article? "Another British tabloid, The Sun (published by Rupert Murdoch’s News Group), was added this past year, as were similar outlets in Iran and Venezuela. The same goes for the left-wing Facebook group Occupy Democrats, which was delisted as a valid source of information." There was a brief discussion about editors removing deprecated sources and tagging instead of fixing. Essentially equating deprecated with blacklisted. This is an example of an editor removing a reference to a site that we have deprecated rather than 1) tagging is as unreliable or 2) finding a source that could be used in its place. He goes on to complain about me below. Walter Görlitz (talk) 08:37, 15 January 2020 (UTC)

https://www.google.com/search?q=alignment+dungeons+and+dragons&sxsrf=ACYBGNR6TA-J2BnVmNqv2VRTclm_kwPoHg:1580449827674&source=lnms&tbm=nws&sa=X&ved=2ahUKEwiCgcG0kq3nAhUUFzQIHbEjAw8Q_AUoA3oECDcQBQ&biw=1094&bih=474&dpr=1.25

Lots of articles. They OK? How about CBR? Peregrine Fisher (talk) 05:54, 31 January 2020 (UTC)

This seems like trivia that shouldn't be included in an encyclopedia article, except maybe in the Legacy section of Alignment (Dungeons & Dragons) as examples of how widespread this meme is. _signed, Rosguill ^talk 06:49, 31 January 2020 (UTC)

This is completely inappropriate. Guy (help!) 09:29, 31 January 2020 (UTC)

List of Latinised names contains a long list taken from the book The record interpreter : a collection of abbreviations, Latin words and names used in English historical manuscripts and records by Charles Trice Martin, published in 1910. The list in our article purports to be a list of names "Coined by Anglo-Norman scribes", but in the book it's only stated to be a list of "Latin Forms of English Surnames" - see the Internet Archive copy here.

So apart from the apparently inaccurate claim about what the list represents, my question here is whether such a list of names published in 1910 can be a reliable source for this article in view of the amount of research that must have taken place since then. I've tried raising the issue on Talk:List of Latinised names - noting an earlier concern about the reliability of the list that was raised by an apparently knowledgeable editor in 2012 - but had no response other than being reverted twice when I tried removing it.  —SMALLJIM  20:38, 29 January 2020 (UTC)

An old book is least of the issues I see here. As you pointed out, that section title is a pure original research. Then, using as a main source some webpage with unclear reliability (coxresearcher.com), which copied data from an old book, instead of the book itself, is not the way one should cite sources in an encyclopedia. Finally, ignoring discussion on the article talkpage and simply pushing preferred revision via reverts is not helping to improve the article. Note revert to the previous stable revision may be justified (per BRD), but lack of communication thereafter certainly is not. Inviting Lobsterthermidor to this discussion. Pavlor (talk) 07:44, 30 January 2020 (UTC)

Thanks @Pavlor: other problems aside, I'm pleased you agree that there's no evidence that the list actually represents Latinised names that were coined by Anglo-Norman scribes, as the article claims it is. Regarding my second point, a quick search in Google Scholar for '"personal names" latinization' throws up many recent hits (not all relevant, of course), showing that this is a field still under active research. So, getting back to RS, I suggest that this 1910 source is not reliable enough to base anything in this subject area on (other than what was the state of knowledge in the early 20th century). Does anyone disagree?  —SMALLJIM  20:15, 30 January 2020 (UTC)

I will certainly not dismiss a source only because it is over 100 years old. The core issue here is not the source itself, but its use in the article. Nowhere in the book it states these names are for anglo-norman period only, quite the opposite. So, this book "may be" a reliable source for examples of Latin names used in medieval reords and manuscripts, but NOT (really big NOT) as a source for "Latinised names coined by Anglo-Norman scribes". In this very case, the information presented in the article is not sourced at all and the use of this source is grossly misleading. Pavlor (talk) 06:50, 31 January 2020 (UTC)

Thanks again. It obviously needs to come out of the article. I'll try again, referring to this discussion.  —SMALLJIM  19:26, 31 January 2020 (UTC)

Would this source be allowed to be used in a page about Saudi expansion? https://omniatlas.com/maps/southern-asia/19220505/ 79.67.65.129 (talk) 00:35, 1 February 2020 (UTC)

Would the daily beast be RS for the following material in People's Mujahedin of Iran?

• A few months after the executions, relatives were handed plastic bags with their children's belongings. By October that year many thousands of prisoners had been executed without trial or appeal.

I have to note that wp:RSP says about this source that "Most editors consider The Daily Beast a biased or opinionated source. Some editors advise caution when using this source for controversial statements of fact related to living persons."Saff V. (talk) 13:11, 28 January 2020 (UTC)

You, ah, skipped the most important part "The Daily Beast is considered generally reliable for news. Most editors consider The Daily Beast a biased or opinionated source. Some editors advise caution when using this source for controversial statements of fact related to living persons.” Horse Eye Jack (talk) 04:11, 29 January 2020 (UTC)

• Yes, Daily Beast should be treated with caution, but in this case I think the reporting does not align with any of their known biases, so it is probably fine; that said, if it is contradicted by any more reliable source, then don't use. --MaximumIdeas (talk) 14:09, 28 January 2020 (UTC)
• I always prefer a better source. In general, if the Daily Beast is the only source, it's likely either hyperbole or WP:UNDUE. Guy (help!) 22:54, 28 January 2020 (UTC)
  • I wouldn't go that far. They're not liars. But sure, nothing wrong with attribution - David Gerard (talk) 23:13, 28 January 2020 (UTC)
• Agree with the above. I don't believe TDB is in any way falsifying their stories, but they often a bit empathatic to drive a point to the reader and may use some exaggerated wording. Use with attribution for things that may seem contestable (as the above). --Masem (t) 01:37, 29 January 2020 (UTC)
• This is one of those ones were there is a large disconnect between the news pieces and the opinion pieces. The opinion pieces are to be avoided at *all* costs but the news pieces are generally high quality and factual. Bias does not disqualify if other conditions are met per WP:RS and WP:VERIFY. When in doubt in cases like this I attribute. Horse Eye Jack (talk) 04:08, 29 January 2020 (UTC)
• For a website that is "always skeptical" and "irreverent" and having a "sharp opinion in the arena of politics, pop-culture and power, The Daily Beast does not strike me as having as high journalistic standards as The Washington Post or The Wall Street Journal, but beyond its frequent opinionated editorials and its rather conversational and sometimes exaggerated language (which can be induced from that About page), it seems that it is considered to be generally reliable. GaɱingFørFuɲ³⁶⁵ 01:16, 1 February 2020 (UTC)

Is City A.M. a reliable source for financial news?GDX420 (talk) 15:53, 1 February 2020 (UTC)

What inspires you to ask? - David Gerard (talk) 17:00, 1 February 2020 (UTC)

https://www.cityam.com/revealed-starling-bank-chief-edged-out-as-billionaire-backer-boosts-stake/ At Anne Boden, presumably. The source has been used in a number of articles, See 4 Vexations (talk) 17:10, 1 February 2020 (UTC)

The above website has been added as ref to a leading-person's mention (in this change) within an article about a UK free-to-air television programme, currently very heavily aired. Rather than just revert as non-RS I thought I'd run it by you - actually states "Wiki" - in case you have/have not encountered it. Thx.--Rocknrollmancer (talk) 18:56, 1 February 2020 (UTC)

Hi, according to our article The Washington Times, "has drawn controversy for publishing racist content, including commentary and conspiracy theories about United States president Barack Obama" and "It has published material promoting Islamophobia", The Columbia Journalism said this about this newspaper, "The Washington Times is like no major city daily in America in the way that it wears its political heart on its sleeve. No major paper in America would dare be so partisan." In 1998 the Egyptian newspaper Al-Ahram wrote that The Washington Times editorial policy was "rabidly anti-Arab, anti-Muslim and pro-Israel." In a 2016 report, the Muslim advocacy group Council on American-Islamic Relations listed The Washington Times among media outlets it said "regularly demonstrates or supports Islamophobic themes." This is all from the wikipedia article of The Washington Times see history for attribution. You can also see other things like their support for Trump etc. The question is, is this a reliable source for the following paragraph.

However, a State Department spokesman reported that "Washington did not claim the exile group was involved in the assassination of scientists in Iran."

Here are some problems with this paragraph:

• The official is not named.
• The text in the quotes is not found in Google except in the Washington Times.
• We have a report from the NBC, a highly reliable source definitely more than this one, that says

  Two senior U.S. officials confirmed for NBC News the MEK’s role in the assassinations, with one senior official saying, 'All your inclinations are correct.'"

  5
• The paragraph starts with "However" which somehow makes all of the previous content claims weak although it is much reliable and much clear.--SharʿabSalam▼ (talk) 23:53, 24 January 2020 (UTC)
• I think the Washington Times is an unreliable source due to its history of bias and inaccuracy. There is no consensus for general exclusion, but if there's good reason to doubt a story in the Washington Times then it should be excluded. I never use it at all. Guy (help!) 10:03, 25 January 2020 (UTC)

I agree with Guy, I wouldn't call The Washington Times a reliable source either, and the part in the article you're bringing up does seem fishy, if thats what youre asking. SageSolomon (talk) 23:10, 27 January 2020 (UTC)

• Per https://en.wikipedia.org?pojem=Wikipedia:Reliable_sources/Perennial_sources , Washington Times is "marginally reliable." I want to note that the NBC News quote above and the Washington Times quote are not actually contradictory (though they seem so at first) -- they are consistent with each other if the US government does not publicly accuse the MEK of this, but anonymous senior officials in Washington also told NBC News that they were involved. The appropriate discussion for such details in on the talk page there, but you may want to consider the idea of including both things. --MaximumIdeas (talk) 14:06, 28 January 2020 (UTC)
• It is one of those newspapers that "follow along" the news, but it is also rife with right-wing opinion, never mind the troubling history of the newspaper's bias for whites (like me, unfortunately). It also has failed numerous fact-checks, and apparently is a purveyor of some falsehoods and hoaxes. It is a shame because there are a few decent folks that write or have written for the newspaper. Unfortunately, it has the same mediocre journalistic integrity as Fox News, but hey, at least it is not the Daily Mail! GaɱingFørFuɲ³⁶⁵ 03:44, 1 February 2020 (UTC)

Is there a consensus then that The Washington Times is unreliable? An editor has intepreted it that way and removed this source from the article. Stefka Bulgaria (talk) 09:25, 2 February 2020 (UTC)

No. This is what The Washington Times - Wednesday, July 24, 2019 (also available on wayback) actually said:

A State Department spokesman at the time said Washington did not claim the exile group was involved in the assassination of scientists in Iran.

Notice two of the differences: no "However," and no quote marks. And that statement is true and verifiable. From the U.S. Department of State, transcript of "Background Briefing on an Announcement Regarding the Mujahedin-e Khalq": a senior state department official (whom the US Department of State chose not to name) said:

And I should add that the United States Government has not claimed that the MEK was involved in the assassination of scientists in Iran. And that’s really all we’re going to have to say on that.

Peter Gulutzan (talk)

On a BLP an editor wrote The Daily Wire calls him "a prominent liberal and YouTube host". Another editor removed it as "not rs". Checking that on RS/P I found no entry, here I found two archived sections, 241 and 279. From my PoV the sourced statement is good (=true) enough, but "no RS" doesn't help much in a 4th AFD. Please add some summary, e.g., "caution", to RS/P. –84.46.52.200 (talk) 15:30, 30 January 2020 (UTC)

The site has no reputation for fact-checking and accuracy. In fact, fact-checkers say, "DailyWire.com has a tendency to share stories that are taken out of context or not verified".6 The site promotes falsehoods about climate change.7 Snooganssnoogans (talk) 15:37, 30 January 2020 (UTC)

The source is reliable for its own editorial content, so this isn't a WP:RS issue. Saying "The Daily Wire calls someone yada yada yada" is reliably sourced to the original source of the quote, per WP:PRIMARY. Since there is no doubt TDW called the person that, it is not strictly a reliability issue. It would be a reliability issue if we were using the source to speak in Wikipedia's voice, i.e. saying "Person X is a prominent liberal and YouTube host". But in this case, it is merely being used as an WP:ABOUTSELF source, for confirming that TDW did, indeed, make that statement. The question you should be asking is "Is TDW a sufficiently well-respected source in this specific field that its statements of this type represent mainstream or well covered thoughts on this matter" That is, it is a WP:UNDUE issue rather than a reliability issue. TDW did make that statement, but we really want to ask "Why does TDW's writing about this topic matter?" --Jayron32 15:40, 30 January 2020 (UTC)

Worth doing the whole RFC thing? I suspect it deserves deprecation for making stuff up, carelessness with sourcing and running conspiracy theories. Certainly generally unreliable, even if it accidentally runs a true fact - David Gerard (talk) 15:55, 30 January 2020 (UTC)

Not really, so far it's unanimously worse than caution. However, alleged "conspiracy theories" trigger my alarm bells, while some folks in this thread might think that this is a brilliant description for, e.g., Jimmy Dore, Abby Martin, or Grayzone, others including me could think that Anya Parampil is a bad case of "woman in red", and that "prioritize authoritative voices, including news sources like CNN, Fox" on YouTube is utter dubious, with CNN as anti-progressive "conspiracy sans theory". Drawing the line at "pizzagate": –84.46.52.200 (talk) 18:44, 30 January 2020 (UTC)

• This fails WP:UNDUE unless it's been covered in reliable independent secondary sources. I can find you a partisan hack website that says pretty much anything about pretty much anyone. Guy (help!) 09:30, 31 January 2020 (UTC)

  If "no rs" turns out to be "only" UNDUE it's no big deal, mainly I hope that somebody is BOLD and adds this source to RS/P. –01:22, 1 February 2020 (UTC)

• The Daily Wire is not much in the way of news, but rather a platform of commentary forwarded by right-wing pundits such as Ben Shapiro. It is akin to Rush Limbaugh's show, another right-wing platform of commentary. While certainly unaffiliated with the alt-right, sources like these are best avoided in favor of non-opinionated journalism such as The Wall Street Journal. GaɱingFørFuɲ³⁶⁵ 02:27, 1 February 2020 (UTC)
• The Daily Wire isn't in the list at Wikipedia:Reliable sources/Perennial sources, which ranges from the best to the worst of sources. It might be worth opening a discussion, in the hope of reaching consensus on a brief appraisal for general guidance. Narky Blert (talk) 09:33, 1 February 2020 (UTC)

  Narky Blert, agree Guy (help!) 16:20, 2 February 2020 (UTC)

Is the Swedish history magazine Världens Historia reliable? For reference, here's its website, although there's no "About" page and most articles are only available in print. Glades12 (talk) 15:51, 2 February 2020 (UTC)

Also, the ISSN is 0806-4709. Glades12 (talk) 15:56, 2 February 2020 (UTC)

No opinion... but the fact that articles are only available in print is not a factor. We judge reliability based on the reputation of author and publisher, not medium. Blueboar (talk) 16:16, 2 February 2020 (UTC)

Oh, I was aware of that. I was just noting that the website only contains a fraction of all the magazine's articles, so it doesn't give a complete picture. Glades12 (talk) 17:15, 2 February 2020 (UTC)

I'd say generally reliable but not as good as actual history-books. Gråbergs Gråa Sång (talk) 21:41, 2 February 2020 (UTC)

Zdroj:https://en.wikipedia.org?pojem=Wikipedia:Reliable_sources/Noticeboard/Archive_285
Text je dostupný za podmienok Creative Commons Attribution/Share-Alike License 3.0 Unported; prípadne za ďalších podmienok. Podrobnejšie informácie nájdete na stránke Podmienky použitia.

---