citations - Are there instances where citing Wikipedia is allowed?

For example, suppose you're writing an article about triangles, and you want to include a short proof of Pythagoras' theorem, but you can't quite remember how to derive it, so you pop over to: https://en.wikipedia.org/wiki/Pythagorean_theorem#Algebraic_proofs and check the proof there. It works mathematically, makes sense and is as far as you can tell, correct, so you include it. Now the proof on Wikipedia doesn't have a citation because probably someone has been taught it and copied it over to Wikipedia, but it's still correct. The only thing you could cite for the proof would be Wikipedia, would a citation be ok or should it just be copied down without a citation?

Edit: Ok, there have been many responses. Most with a firm "no", but some giving the reason: "Just get the proof from a book" since it's so common, but suppose then it wasn't a common proof of something. Maybe it's not very well known or something but suppose there is a small Wikipedia page with the correct mathematical proof. How much effort are people going to go to to find a proof in a book when there's a (correct (which we know, becuase it's maths and we can check it)) proof on Wikipedia? Why is it so bad to say: "I found this proof on Wikipedia, and the maths checks out so it doesn't matter who wrote it, it is correct, but that's where I found it." Or if you do find it somewhere else why not say: "Proof taken from Wikipedia and verified by the proof shown in "Triangles and their properties, Nature, 2014, p113 etc..."

Answer

Are there instances where citing Wikipedia is allowed?

I'm not going to try to answer the question of "is allowed" or "isn't allowed" but instead I'm going to try and answer if citing Wikipedia is, in general, a good idea or not for research papers.

Here's five points I can think of for and five points against.

Good idea:

• If Wikipedia is your source, and you learn something from what Wikipedia's editors have provided, then citing Wikipedia is the Nice Thing To Do™.



• Wikipedia is a convenient, well-known, easily and freely accessible source (vs. final-copy published papers that are pay-per-view if not out of your wallet, then indirectly out of your university's).


• Content aims to be written in an accessible manner targeted at a wide audience (vs. arguably the bulk of technical articles published in journals and conferences). Plus there's clickable links for jargon! Sweet!


• Notable or Featured Articles will probably have undergone more scrutiny from experts (w.r.t. bias, technical correctness, sources cited, etc.) than your average peer-reviewed article on a similar topic.


• At least with Wikipedia, readers should know where they stand with such a citation. Compare this with the artifice of citing papers for claims made in poor-quality journals or conferences that can be considered authoritative solely on the basis of being "peer-reviewed".

Bad idea:

• A Wikipedia article is subject to frequent editing and the content you cite it for may not be there when the reader goes looking for it (adding a date-accessed only partly addresses the issue since it will be difficult for the reader to start going through versions edit but as per the comments below, you can use a citing service within Wikipedia to cite a specific version of a page).


• Relatedly, the content may have been correct at the time of citing but errors may be edited into the content at a later point, making you look bad by association (edit, again, citing a specific version could work around that)


• There is no minimal guarantee of the quality for most Wikipedia articles (which peer-review, at least in theory, should provide, depending perhaps on the venue).



• Wikipedia is not a first-party source: in theory, no new knowledge is "created" in Wikipedia. Where feasible, you should try to attribute knowledge to those who create it (and optionally thereafter add a reference to those who report it or better describe it in a book or survey)


• Attribution for the content of a Wikipedia article is difficult: multiple editors may be involved in a certain piece of content, many editors are anonymous, etc. This means that there is no culpability for information (unlike peer-reviewed papers where authors have something to lose by publishing crap). This opens up further possibilities, such as authors anonymously adding content to Wikipedia that they can cite to support the claims in their paper.

Okay, back to academic convention and "isn't alloweds" and "is alloweds" ...

---

Maybe it's not very well known or something but suppose there is a small Wikipedia page with the correct mathematical proof. How much effort are people going to go to to find a proof in a book when there's a (correct (which we know, because it's maths and we can check it)) proof on Wikipedia?

According to Wikipedia policies, it should have a reference to an original source. Cite that.

If it doesn't have a reference, or if you copy how the proof is expressed directly, I would recommend to add a footnote (rather than a formal citation) stating the Wikipedia article you found the proof on and how you verified it (and the CC-BY-SA licence if you copy the expression).

Generally avoid citing web-pages (unless they are normative online standards or something) if only to avoid getting your hand bitten off by a reviewer. Using footnotes is much safer in this regards and serves a similar-ish purpose (though it won't be counted as a formal citation).

In research, if enough reviewers believe something isn't allowed, then de facto it isn't allowed.

Or if you do find it somewhere else why not say: "Proof taken from Wikipedia and verified by the proof shown in "Triangles and their properties, Nature, 2014, p113 etc..."

Why not say "Proof sketched for me by Fred, the guy who sits next to me, who had previously read it in 'Triangles and their properties, Nature, 2014, p113 etc...'"?

In other words, what's important is the reference, not how you found.