Wikinews:Water cooler/policy/archives/2019/April

Using Wikipedia as a "source"?

Currently, the usage of Wikipedia as a source is debated in Bulgarian Wikipedia, so I am curious about it happening in English Wikinews. Initially, I thought that all en-wn articles don't use Wikipedia as a source. However, I found pages using "Wikipedia.org", like this article and that article. Of course that was in mid-2000s. I could not find many newer articles using Wikipedia as a main source. I don't think one interview counts; it mentions "Wikipedia" but doesn't use it as a source. Well, there's another article using Wikipedia as a source, but it primarily involves Wikipedia itself. I could not find local rules tackling the matter, though (off-topic) I found Wikinews:Blog as Source and tagged it as an "essay". --George Ho (talk) 01:50, 5 February 2019 (UTC); edited, 01:51, 5 February 2019 (UTC)

Nor will you find anything here in the modern era using Wikipedia that way. Wikipedia is not useable as a source. Totally, 100% not useable as a source; there's nothing remotely ambiguous about the point. Wikipedia openly acknowledges not being appropriate to use as a source. As we find ourselves having to explain to naive newcomers from time to time, if you want to use some information you find in a Wikipedia article, either the article cites an actual source that contains in the information, in which case you should use that actual source, or it doesn't cite an actual source that contains the information, in which case the information is unverified when it occurs in Wikipedia. --Pi zero (talk) 02:06, 5 February 2019 (UTC)

Wikipedia could really only be a source about what Wikipedia claims about itself. Just like how it would be totally appropriate to say that the website of a political party describes itself as "the foremost leader in democratic socialist thought in America" or how a company's website may boast that it "sold more than 100,000 widgets in Q1!" —Justin (koavf)❤T☮C☺M☯ 02:12, 5 February 2019 (UTC)

That sounds strange to me. The credibility Wikipedia needs to be used as a source appears to me the same as the credibility it needs to be used at all: that is what encyclopaedias are used for. (Except in Wikipedia itself: there it is unacceptable due to self-sourcing and circular sourcing concerns.) Consequently, the position that Wikipedia is not usable as a source appears to me to quietly imply that Wikipedia is not usable at all (except maybe for the lists of sources at the ends of its articles), and therefore has no reason to exist, other than to merely list other sources. This appears to me to be a fallacy which I describe here.

There is also the argument by some academics that Wikipedia cannot be trusted, since anyone can edit it. I describe this fallacy there, too.

Could someone please point to mistakes in this reasoning? -- Григор Гачев (talk) 11:29, 5 February 2019 (UTC)

What you describe as pointless is in fact Wikipedia's raison d'etre; to bring information together into a sensible whole. When written to the best Wikipedia can do, the sources are there to be checked; they can be cited directly. See w:Wikipedia:General disclaimer. BRS ^{(Talk) (Contribs)} 13:04, 5 February 2019 (UTC)

The "sensible whole" argument is precisely my point. By systematizing facts from many sources, a well-written Wikipedia article makes more things clear and thus provides more value than any separate source it cites, helping the reader understand things that they might otherwise miss. (In theory, any reader can assemble that from the sources, like the article editors did. In practice however, this will require amount of work and expertise comparable to the one invested into creating that Wikipedia article. Very small percentage of the readers will have the opportunity and the motivation to do that.)

If that is true however, citing a Wikipedia article is beneficial in cases when the extra value it provides enhances the understanding of the news article (or any other publication) that cites it. Therefore, in such cases citing it is clearly justified, exactly like with any other encyclopaedia. Actually, if the publication that cites it relies on what its sensible whole brings, then citing it appears to me as mandatory as would be citing sources for the facts described there. -- Григор Гачев (talk) 22:53, 5 February 2019 (UTC)

We're not discussing citing a Wikipedia article. We're discussing using a Wikipedia article as a source, which is quite different. There may be confusion arising here due to misunderstanding of the term "source" as used on en.wn. Just as there may be confusion over the meaning of the term "trust". --Pi zero (talk) 23:27, 5 February 2019 (UTC)

Could you please explain these misunderstandings? That might resolve the differences in this argument. -- Григор Гачев (talk) 14:15, 6 February 2019 (UTC)

There's a disconnect between what you initially said, which is using Wikipedia to source details to an article, and what you seem to actually want, which is giving due respect to a sister project. To the latter end we have {{sisters}} which we use to give prominent links to Wikipedia (and, occasionally, others as appropriate). The basic style used is here. By way of context, prior to this template, templates such as {{wikipediapar}} were often used; while they weren't always placed there, they used to be often placed alongside the other sources. It was pointed out this were inappropriate because this implied details were being taken from there and that an external links section might be more appropriate; this created the further paradox that we would be treating sister projects more low than we treat true outgoing links to our sources. So the solution was to create that template and an associated section in appropriate articles, placing links to sister projects above sources. I've just now, while typing this out, added the updating of those articles to the archive curation tasks. BRS ^{(Talk) (Contribs)} 21:22, 9 February 2019 (UTC)

Seems that I have given the wrong impression. While giving due respect to a sister project is undoubtedly an important thing, I am not seeking it here.

Being by nature topic overviews, Wikipedia articles can and often do make visible topic tendencies and links that aren't visible in any their sources separately. I believe that if a publication - eg. a Wikinews news - is referencing or building upon such tendencies and links, then it is not just using the Wikipedia article - it is using that article as a source, and for a good and justified reason. If that is true, then it appears proper to me to point in this Wikinews article the Wikipedia article as a source, as these tendencies and links aren't evident in the sources of that article.

Alternatively, the Wikinews article would have to list all or most of the sources of the Wikipedia article, and hope that the reader of the news will read them all and will assemble in his/her head an equivalent of the Wikipedia article that cites them. That appears to me a big waste of the reader's time and efforts, with significant chance that they will not be able to find these tendencies and links even after reading all sources. (Not every reader has knowledge and logic on par with the editors who wrote the Wikipedia article.) Or, the Wikinews article will have to put these sources in a separate section, together with a list of the tendencies and links that would be understood after reading them all. That appears to me absurd, since it would be much easier to just either copy the appropriate text from the Wikipedia article, or to link to it somehow from the sources section. So, listing the Wikipedia article as a source appears to me to be the only proper action in this case.

The equivalent of this problem in the scientific articles is resolved through the equivalent of this solution - articles that build on notions, shown in overview publications, list these overview publications as sources. That is another reason that makes me believe that listing in such a case the Wikipedia article as a source is the proper action.

Still, I wonder if I might be wrong, and actively seek arguments that will prove it. -- Григор Гачев (talk) 21:38, 16 February 2019 (UTC)

The sad truth is that some Wikipedia articles are good and accurate and some are not. Timing, the social dynamics of the people working on the article, the accessibility of the subject matter, and other things can all affect the accuracy of the article. There could be 95% of sources that say "most widgets are blue" but if the "widgets are red" fans get to the article first, then they are likely to accumulate like-minded friends and there is reasonably likely to be a Wikipedia consensus that widgets are red. Participants in project Wiki are amateurs or at least functioning in an amateur capacity. I've heard of professionals in obscure subjects who were driven out of Wikipedia articles in their area of expertise when they tried to correct factual problems.

The general rule of thumb is "Read the Wikipedia article. Then click on the tags to see the source that the article cites for the facts you want. Then look at those sources with your own eyes. Then if they're good, use them yourself."

In my early days, an article on grammar said, "Most linguists, especially sociolinguists, think [distinguishing between 'which' and 'that' in a specific way] is a pretty silly idea." I looked at the source and it was really one guy who happened to be a pro sociolinguist who said that. The original Wikiauthor who wrote that line made the kind of generalization that people make out loud all the time when working from memory, but they put it in an article where it was likely to be taken as gospel by anyone who read it.

Use Wikipedia's sources, not Wikipedia directly. This way, if Wikipedia got it wrong, there's a reasonable chance that you'll catch it. Who knows? You might even go back and fix the Wikipedia article! ...I'd recommend wearing Kevlar, though, because messengers can and do get shot. Darkfrog24 (talk) 05:07, 27 February 2019 (UTC)

I can't agree more with your thinking - but not completely with your conclusions. There still can be Wikipedia articles that are correct about the facts - and my guess is that most of them are. My logic is that "if that is true, the rejecting the usage of these as sources is probably an overreaction to the fewer articles that aren't correct".

I am trying to test that logic by coming from the opposite way. Wikipedia in general shuns primary sources and strongly prefer secondary, even tertiary sources when possible. However, every level of sourcing introduces additional possibility for mistakes. Should then we reverse this policy and emphasize on primary sources, because the secondary / tertiary ones may contain more mistakes? In fact, should be shun the primary sources too, in favor of directly analyzing the data ourselves - any analysis the primary source does may also contain mistakes?... This idea appears wrong to me. If higher-level sources are acceptable, then Wikipedia articles should logically be acceptable too.

It is true that Wikipedia articles can contain mistakes. However, the other sources can contain mistakes too: nobody is infallible. It appears to me that the typical mistakes in collaborative non-professional projects like Wikipedia and different from those in single-person or collective-produced highly professional articles. The first kind of mistakes are those you describe. The second kind are the professional opinions that must be diverse in order to develop any science (or any other kind of fact establishing and describing) properly, despite that the truth can be only one. A lot of these will not match the truth, sometimes even when it is already known: even then, it usually takes a lot of time and discussion to push all other opinions into the fringes.

As a result, we tend to overestimate the mistakes in the Wikipedia articles, and to vastly underestimate the mistakes in the other sources. In reality, the percentage and the graveness of both kinds might be of the same magnitude. If it turns out to be so (I haven't measured it), then maybe it is time to re-evaluate the "Wikipedia is not a source" rule, at least for the other WMF projects. (Within Wikipedia it might be preserved even then, to avoid the possibility for circular sourcing.)

Is anyone able to logically or factually refute these lines of thinking? I would love to test them. -- Григор Гачев (talk) 18:41, 4 April 2019 (UTC)

You say "perform the analysis ourselves." But we are not experts, or if we are, cannot prove it.

The people analyzing data for traditionally published sources have all had the opportunity to disclose their credentials, which may be as official as certifications and professional CVs, degrees and training programs, but at the very least their real names. We do none of that here.

We're all real people who work real jobs to earn our livings, so presumably we've all got some level of expertise in something, but we cannot confirm that to our fellow Wikinewsies without giving up our anonymity. Darkfrog24 (talk) 18:47, 4 April 2019 (UTC)

Note that it doesn't matter whether or not things stated in a Wikipedia article happen to be correct. That's irrelevant. The nature of the vetting/sourcing process is paramount. (Btw, doing analysis is forbidden under our neutrality policy.) --Pi zero (talk) 19:29, 4 April 2019 (UTC)

In wikivnews Religious related usernames allowed? Or Prohibited?

b:Using Wikibooks/Setting Up A User Account#Choosing a Username

"Offensive usernames:"

Usernames that invoke the name of a religious figure or religion in a distasteful, disrespectful, or provocative way,

Please answer me (Taribos (talk) 15:23, 30 April 2019 (UTC))

Hi, Taribos. The determining factor isn't whether the username is related to religion, but whether it's offensive. Our local policy here, Wikinews:Username, says to avoid offensive usernames, and the recommendations you link to, at our sister Wikibooks, look pretty reasonable. It's probably best to avoid religion-related usernames, but whether a particular username is a problem must be decided on a case-by-case basis. --Pi zero (talk) 16:58, 30 April 2019 (UTC)