Wikipedia talk:Citing sources/Archive 42

@Herostratus: The section doesn't really explain what's suggested by the example's citations. Someone might assume this is acceptable:

• Mary likes cake, but John doesn't.<ref>Mary likes cake</ref><ref>John doesn't like cake.</ref>

This is exactly what's being advised against. "If a sentence or paragraph is footnoted with a source, adding new material to that sentence without a new source—or without making sure it is supported by the existing source—is highly misleading. ... Do not add other facts or assertions into a fully cited paragraph or sentence".

This is the only proper way:

• Mary likes cake,<ref>Mary likes cake</ref> but John doesn't.<ref>John doesn't like cake.</ref>

--Ilovetopaint (talk) 02:19, 31 January 2017 (UTC)

No, it certainly is not the only proper way. Depending on the material and the refs, it is OK to put all the refs for a paragraph at the end of the paragraph, and this has the considerable advantage of not cluttering the prose with a bunch of little numbers; the reader can look up the refs quite as well if they're gathered at the end of the paragraph. It depends. Sometimes it is best to put the ref in the middle of a paragraph or even a sentence, it that aids the reader and clarifies which exact fact is tied to which ref, if this is crucial to the reader being able to verify the statements in the paragraph. It's often not crucial. It depends on the material and the refs. It's partly a matter of taste.

"If a sentence or paragraph is footnoted with a source, adding new material to that sentence without a new source—or without making sure it is supported by the existing source—is highly misleading. ... Do not add other facts or assertions into a fully cited paragraph or sentence" is about something completely different. It says if you add a new fact to a sentence or paragraph you have to add a new ref. It doesn't say where the new ref should be placed. At the end of the paragraph may be fine. Herostratus (talk) 02:40, 31 January 2017 (UTC)

Right. This would also be acceptable:

Mary likes cake, but John doesn't.^[1]

My point is that the current phrasing makes it appear as though it's okay for the two sources to be bundled while containing different facts. Note the emphasized text:

... Sometimes the article is more readable if multiple citations are bundled into a single footnote. For example, when there are multiple sources for a given sentence, and each source applies to the entire sentence, the sources can be placed at the end of the sentence, like this.^[4][5][6][7]

Oh, OK. I see what you're saying. The who-likes-cake example contradicts "and each source applies to the entire sentence", if we assume that in the example Mary likes cake, but John doesn't.<ref name=Mary/><ref name=John/> the ref "Mary" talks about Mary only and the ref "John" talks about John only -- which I do assume, and which I'm confident is the intent and how most everyone would understand it (we could make this manifest, but I think it's pretty obvious).

I think that the actual intent of the passage you cited is to specify that bunding should only be used when each source applies to the entire sentence; it is in the "Bundling citations" section. So it wouldn't apply to the John-Mary-cake example since those ref's aren't bundled.

Assuming this is true, it might be better to rewrite the paragraph

Sometimes the article is more readable if multiple citations are bundled into a single footnote. For example, when there are multiple sources for a given sentence, and each source applies to the entire sentence, the sources can be placed at the end of the sentence, like this.[4][5][6][7] Or they can be bundled into one footnote at the end of the sentence or paragraph, like this.[4]

To something like this:

Sometimes the article is more readable if multiple citations are bundled into a single footnote. For example, when there are multiple sources for a given sentence, the sources can be placed at the end of the sentence, like this.[4][5][6][7] Or -- if each source applies to the entire sentence -- they can be bundled into one footnote at the end of the sentence or paragraph, like this.[4]

The problem with this is 1) why have this rule, and 2) the example that immediately follows ("The sun is pretty big...") absolutely violates this, each bundled ref supports only part of the two-sentence statement.

Probably better would be to remove the clause "and each source applies to the entire sentence" altogether. I don't see how it's helpful. But there could be complications and nuances I'm not seeing, so I'd like to hear input from other editors before doing this. Herostratus (talk) 07:13, 31 January 2017 (UTC)

I don't think there is any question over the intent in "Bundling citations", but the intent in "Text-source integrity" was absolutely not "the ref 'Mary' talks about Mary only and the ref 'John' talks about John only". Both sources are supposed to apply to both Mary and John. Again, note this text:

If a sentence or paragraph is footnoted with a source, adding new material to that sentence without a new source—or without making sure it is supported by the existing source—is highly misleading. ... Do not add other facts or assertions into a fully cited paragraph or sentence.

The reason why is to prevent this scenario:

People who like cake include Jack, Matthew, Luke, Sarah, and Jane.^[2]

Perfectly fine. But then another editor comes along and adds this:

People who like cake include John, Mary, Jack, Matthew, Luke, Sarah, and Jane. Peter doesn't like cake, he likes sandwiches.^{[3][4][2][5][6]}

This makes it look like every source is talking about those individuals. But they're not, and this makes each claim unnecessarily tedious to verify. The citations should either be broken up inside one reference (as demonstrated previously) or written thusly

People who like cake include John,^[3] Mary,^[4] Jack, Matthew, Luke, Sarah, and Jane.^[2] Peter doesn't like cake,^[5] he likes sandwiches.^[6]

-Ilovetopaint (talk) 02:07, 2 February 2017 (UTC)

Well, you're making an incorrect interpretation of the example in in "Text–source integrity" IMO. The first example of OK format is

Mary likes cake, but John doesn't.<ref name=Mary/><ref name=John/>

No indication is given of what, exactly, is in the references named "Mary" and "John". It could be that the ref "Mary" says "Mary likes cake but John doesn't" and the ref named "John" says basically the same thing. But that would be nonsensical, for a couple of reasons. For the first reason why it would make no sense to assume the writer thought that, look at the second example:

Mary likes cake,<ref name=Mary/> but John doesn't.<ref name=John/>

It is stone-cold obvious from this second example that for this example the reference named "Mary" clearly talks about, and only talks about, Mary. And the reference about John talks about, and only talks about, John. That is why they are placed as there, each supporting the particular part of the sentence that the ref covers. If both refs said essentially "Mary likes cake but John doesn't" it would be nonsensical to place them as shown in the example. Right?

Is it possible that the person writing the example intended the refs named "John" and "Mary" to be assumed to contain different text in the two example -- in the first example both references talk about both Mary and John, and in the second example each talks only about Mary or John, respectively? No, it's not possible. No one would write the examples like that, using the same names for two refs that are different in the two different examples. No one would be that foolish.

The second reason that it is safe to assume that the writer intended the refs to say, respectively, "Mary likes cake" and "John doesn't like cake" (rather than both saying "Mary likes cake but John doesn't") is the names of the refs -- "Mary" and "John" rather than "Ref_1" and "Ref_2" or whatever. It is reasonable to assume that the writer gave them those names to indicate what the ref covers. This is less compelling than my first point but supports it -- not that it needs support.

But in your edit here which started this, you don't assume that. You indicate that the intent was to have to example refs contain the same facts (in your edit, you envisioned the refs as giving the facts from two different sources -- but still, the same facts). But IMO that was not the intent and completely changes the meaning of the example.

This is a very big deal and would be huge change, to basically override "The distance between material and its source is a matter of editorial judgment" and "Both of the following are acceptable". I would not support making a change of this magnitude without a full-blown WP:RFC and a wide and deep discussion.

Which would be fine. On the merits, your point is reasonable -- I don't think I agree, but I'm willing to be convinced. I myself pepper refs into the middle of my own sentences often enough. The question is should this now become a requirement for all editors to do. Maybe. But not without an RfC. Herostratus (talk) 07:37, 2 February 2017 (UTC)

I agree that editors should be allowed - as they currently are - to include references inside a sentence at whichever place their editorial judgement says is best. When a reader verifies references, we assume the reader will look up all the references in the sentence, not just the closest one to the fact. Sometimes it makes sense to put a reference in the middle of a sentence, but sometimes it makes more sense to simply put the references together at the end. Both are perfectly standard in high-quality professional and academic writing. Referencing is intended for humans, not for computers, and accordingly it has a lot of flexibility. — Carl (CBM · talk) 12:22, 2 February 2017 (UTC)

Exactly the system isn't really designed to provide a 100% unambiguous correspondence between individual facts and individual sources to verify content (in an machine readable manner of sorts), instead it merely replicates various usages (primarily in printed text) out side of Wikipedia. So if you want to verify a sentence or paragraph in doubt you need to check all sources in the immediate neighbourhood and not just an individual one and that isn't considere too much of a burden for verification. A lot of citations through a sentence or paragraph make the verification of individual facts often more efficient, but they are also less pleasant to read and put an "unnecessary" bureaucratic burden on our authors (our most precious and scarce resource). Another thing is that not every single "fact" or better piece of knowledge (as relation between facts) can be expressed in a single sentence.--Kmhkmh (talk) 13:07, 2 February 2017 (UTC)

"This is a very big deal and would be huge change" — For the fourth time:

If a sentence or paragraph is footnoted with a source, adding new material ... without making sure it is supported by the existing source—is highly misleading. ... Sometimes the article is more readable if multiple citations are bundled into a single footnote. For example, when there are multiple sources for a given sentence, and each source applies to the entire sentence, the sources can be placed at the end of the sentence, like this.[4][5][6][7]

From WP:CITEFOOT:

The citation should be added close to the material it supports, offering text–source integrity. If a word or phrase is particularly contentious, an inline citation may be added next to that word or phrase within the sentence, but it is usually sufficient to add the citation to the end of the clause, sentence, or paragraph, so long as it's clear which source supports which part of the text.

There is no change being proposed. If there was, why would this "Text-source integrity" section even exist? What makes it any different from WP:VERIFY or WP:STICKTOSOURCE?-Ilovetopaint (talk) 17:39, 5 February 2017 (UTC)

Having a very few sources at the end of a paragraph is not necessarily too tedious to verify. There are pros and cons for bundling sources at the end or for adding them to individual facts, which approach is best needs to be judged on an individual basis and it is to degree also simply a matter of personal taste. Obviously the more sources you have (in number and content volume) the less appropriate it becomes to simply bundle them at the end, i.e. there comes a point when it is too tedious indeed. However you can also bundle and annotate sources, then bundling at the end (of a sentence or paragraph) can still be ok even with a larger number of sources as long as the annotation spells out explicitly where to find what.--Kmhkmh (talk) 03:04, 2 February 2017 (UTC)

" ... as long as the annotation spells out explicitly where to find what. " — Right. In the example I outlined, this is not the case.--Ilovetopaint (talk) 18:13, 5 February 2017 (UTC)

From the talk page of User talk:LlywelynII:

The template to which user:LlywelynII is referring is {{EB1911}} which is a wrapper around {{Cite EB1911}} which is a wrapper around {{cite encyclopedia}}. The EB1911 wrappers simply fill in some fields and allow a limited number of other fields to be passed onto {{cite encyclopedia}}. One of those filled out is the editor-last=Chisholm and editor-first=Hugh.

{{cite encyclopedia}} includes by default a link to the editor-last link in ref=harv if the author is not set. In the case of EB1911 may articles were written anonymously, so it is common practices to use {{harvnb|Chisholm|1911|p=n}} when adding inline citations.

1. Is citing the editor of a book in short inline citations to be discouraged as user:LlywelynII suggests:
   • should one instead add a parameter author=Anonymous instead?
   • or should one use the explicit ref= such as the one user:LlywelynII prefers: ref={{harvid|''EB''|1911}} so that inline citations can contain {{harvnb|''EB''|1911}}?
2. If the parameter editor-last= is not to be used as a default option when no author is present, then presumably templates such as {{cite encyclopedia}} should be altered to discourage such usage by altering editor parameters to behave like the more recent translator parameters and not be automatically included in ref=harv if no author is given?

I think that this would be a detrimental step to removed links to editors what do other think? When no author is available, is it useful to link short citations to long ones via the editor and year, or is using editor in this way creating a "malformed" short citation? -- PBS (talk) 15:59, 8 February 2017 (UTC)

As a point of clarification: The lack-of-authors-fall-back to |editor-last1=–|editor-last4= + |year= when creating a cs1|2 citation's CITEREF identifier is common to all cs1 templates and to {{citation}}; not just to {{cite encyclopedia}}. The creation of the CITEREF identifier is done by Module:Citation/CS1 which is the engine that renders all of the cs1|2 templates.

—Trappist the monk (talk) 16:24, 8 February 2017 (UTC)

Yes, LlywelynII is quite wrong to imply that a citation of the form "Chisholm, 1911" attributes the work to Chisholm. In the Harvard system, name–date combinations are simply identifiers, and where there's no author, the editor(s) can be used or some other part of the citation, such as the name of the work. Peter coxhead (talk) 17:16, 8 February 2017 (UTC)

Nope, it absolutely attributes the work to Chisholm as its editor. You and I both know that is separate from the authors of the individual entries but the formatting doesn't always make that clear and our readers can (and do) misunderstand what is being stated. The name of the work should be used instead. — LlywelynII 10:34, 10 February 2017 (UTC)

The templates {{cite EB1911}} and {{EB1911}} format the text of the citation in exactly the same way as {{cite encyclopedia}} does. All they do is fill out some of the fields automatically, such as encyclopedia=[[Encyclopædia Britannica Eleventh Edition|Encyclopædia Britannica]], removing that repetitious task from the editor wishing to use the {{cite encyclopedia}} for citing information extracted from EB1911. So the temples no more "absolutely attributes the work to Chisholm as its editor" than does the template {{cite encyclopedia}}. You state "but the formatting doesn't always make that clear" what formatting is unclear (ought the formatting of {{cite encyclopedia}} be changed)? Your state "our readers can (and do) misunderstand what is being stated." what is your evidence for this? -- PBS (talk) 12:32, 11 February 2017 (UTC)

It occurs to me to wonder ... {{EB1911}} is an attribution template; {{Cite EB1911}} is a citation template. Why should a short, in-line template like {{sfn}} link to the attribution template {{EB1911}}? It seems to me that attribution templates should not be made part of a formal reference list or bibliography per se, but rather, should be a separate addendum to those lists.

It also occurs to me to wonder how long an attribution template is valid. An article may once have used whole portions of text from the attributed source but this is Wikipedia: editors and bots can and will edit that original text until it no longer resembles the original source. Once that happens, is the attribution template still valid?

—Trappist the monk (talk) 10:12, 9 February 2017 (UTC)

To answer your two questions. One has to first understand the reason for these attribution templates. The editing community (or at least those who have expressed an opinion on the issue) is divided over the issue of whether third party text should be incorporated into Wikipedia articles. No responsible editor thinks that unauthorised copyrighted material should be incorporated so that is not an issue. What remains are two types of text. That with a copyleft licence that is compatible with the one used by Wikiepdia and the second is public domain text (often from copyright expired sources like EB1911). The former includes text from sources like Citizendium and government sources like the US Federal Government and the British Government. There are hundreds of templates for these in Category:Attribution templates.

How prominently these templates should be displayed depends on ones opinions on plagiarism. One of the arguments against using third party text was that it is plagiarism. Something that one is taught is bad, in many schools and also in the first year at university. Indeed some well known cases of plagiarise have in recent years lead to well known academics being disowned by their peers. To address this issue by those who think that Wikipedia should take text from appropriate legal third sources, it was decided to highlight the source any text that was copied in a way as clearly making the text as copied means that it is not plagiarised because it has not been done surreptitiously. This is the answer to your first question. Yes we could, and can just add an attribution template separate from the citation template, but in that case if the "this text is copied" template is placed in an unusual place then can be argued that text is being copied surreptitiously. Making the attribution part of the citation template stops that charge.

Whether the template has to remain indefinitely has to do with copyright law. If a Wikipedia text is copied by a third party and then modified the Wikipedia licence must still be included, because that is part of the terms of the licence, and so all derived works not matter how change must include it. Likewise text on Wikipedia derived from Citizendium text must also continue to display Citizendium licences even when the wording is far removed from the original. In the case of copyright expired text like EB1911 there is no legal requirement to display anything. It is just a moral imperative (anti-plagiarism). As to when it become unnecessary to us an attribution template, is a matter of judgement, but toolks like Earwig Copyvio Detector at least allows for an easy assessment of what proportion of the text is copied. Also if only one or two paragraph are copied in a much larger article then an inline attribution is sufficient and many of these templates come with an inline attribution switch to alter the words to cover such usage. -- PBS (talk) 17:34, 9 February 2017 (UTC)

A couple of other points of detail. Uses of "{{sfn|Chisholm|1911}}" don't necessarily point to an attribution template; they occasionally point to a citation template. I don't see anything wrong with that, as long as the implied distinction ("these are not EB1911 words, but EB1911 supports this claim") is correct. Second, PBS and I use "{{sfn|Chisholm|1911|p=123}}" as a signpost when the article being attributed is several pages long. It's extra work for us but it makes things easier for the citation-hunter.

On the original point - I frankly don't see the problem with attributing an editor. Is someone using Wikipedia as a source of statistics on author productivity? And, since we are explicitly linking to a source that anyone can check, there's full transparency on the meaning. But, if there is doubt about using Chisholm (and, come to that, Herbermann) in citations - can you explain what we should use instead? And if there is a consensus, can someone write a bot to update the many, many (I can come up with an exact number if you want) occurrences? David Brooks (talk) 18:48, 9 February 2017 (UTC)

Worth noting that the quote above was discussing the replacement of a proper citation with an eyesore "attribution" template.

It's not "wrong" to have the attribution templates and it's not "wrong" to attribute a work to its chief editor. It is unhelpful to prefer boilerplate in place of well-formatted citations and it is extremely unhelpful to attribute individual encyclopedia entries to the chief editor of the entire work. It is far better to attribute the work to the entry's own author or, where unknown, to simply use the name of the work either as EB, Enc. Brit., or Encyclopaedia Britannica, owing to the confusion that easy and needlessly arises from the present formatting.

That said, it was never a question of whether some attribution should exist (it should) but whether the lazy boilerplate attribution templates slapped on during the early stages of Wikipedia, when articles were copied wholesale from EB11, should continue to be permanently preferred to well-cited and inline citations to the quoted material. Of course they shouldn't. What was actually occurring here was the self-righteous claim that the boilerplate should never be replaced with Wikipedia's standard citation formatting and that such formatting will always be incorrect w/r/t material taken from the EB11.

That was nonsense. — LlywelynII 10:44, 10 February 2017 (UTC)

Look, are you suggesting we use up our valuable time budget changing every short citation into a {{sfn|EB1911|...}} (which can at least be automated) or a fully formatted reference (which takes human effort), when the existing form is always enough to let an investigator easily establish the true nature of the citation? Especially given that EB1911 is actually in good shape - some other common PD sources have more obscure paths to verification. Or shall we get on with the much more important task of adding inline citations - any inline citations - to the thousands of articles that lack them? Shall we paint the rusty parts of the fence, or fill the mile-long gaping holes? There's a clear opportunity cost of making an essentially useless change.

Those numbers I mentioned. Related to your assertion "It is far better to attribute the work to the entry's own author...", there are now only 17 pages that have {{sfn|Chisholm|1911}} but have a named author in the EB1911 template. There were 33, but I handled about half of them, and may as well do the rest; this is a trivial effort. And remember that some of the articles were written by Hugh himself. On the larger point, there are currently 1,915 pages containing 6,806 {{sfn|Chisholm|1911}} tags. As I said, there are more important things to do and I absolutely will not spend any time on them, nor will I switch to introducing an inconsistency, unless there is a wider consensus that the current habit is wrong and exactly what the new standard should be. David Brooks (talk) 06:47, 11 February 2017 (UTC)

@LlywelynII I am curious why you use the format ref={{harvid|''EB''|1911}} and not ref=harv when there is an EB1911 article author as you did here in Anthracite and here in Arad Fort (there are many more examples but these were the first two in the list). When available, why do you not display the author in the long citation, (you suppress the display by using display-authors=0) and why did you not use the "author-last|year" for the short citation? -- PBS (talk) 12:32, 11 February 2017 (UTC)

@LlywelynII I raised this section on this talk page specifically to talk about whether it is acceptable to use the automatically formatted coupling of "editor-last|date" in short citations to an long encyclopaedia citation, such as {{sfn|Chisholm|1911}} no-one has supported your assertions that to do so "well-meaning editors have attributed articles personally to Hugh Chisholm because of EB’s shitty default template)". Therefore I do not think that your edit to August Hermann Francke has a consensus for altering the short citations {{sfn|Chisholm|1911}} because they are confusing. Removing the attribution template when the the earwig tool suggested that 92.5% confidence" that most of the text in the article was a copied from EB1911, on the grounds that the citation style is confusing is baseless, but that is something for a different discussion on Wikipedia talk:Plagiarism, as it is out of scope for this guideline. -- PBS (talk) 12:32, 11 February 2017 (UTC)

In the source editor there is a cite->template->(whatever)->URL generation tool, in which a user can post a url, click a magnifying glass, then get a citation.

In the visual editor there is a similar tool at cite->URL.

Can someone tell me the name of this tool? I am seeking the technical specifications which describe what metadata fields this tool checks on external websites to find title, author, etc. Thanks. Blue Rasberry (talk) 21:23, 10 February 2017 (UTC)

@User:Bluerasberry there are several tools that I use see user:PBS/Notes#Format citation tools -- PBS (talk) 14:07, 11 February 2017 (UTC)

@Bluerasberry: I believe you may looking for Wikipedia:RefToolbar. Headbomb {talk / contribs / physics / books} 15:02, 11 February 2017 (UTC)

Thanks both of you. This is what I wanted. Blue Rasberry (talk) 11:13, 13 February 2017 (UTC)

Can we reference (for example) reliable books, and how would this be done? Nimzo321 (talk) 20:03, 19 February 2017 (UTC)

Yes of course, it is very widely done and indeed is generally preferable IMO. See template:cite book. --John Maynard Friedman (talk) 20:20, 19 February 2017 (UTC)

The parameters of {{Cite book}} and {{Cite web}} can be daunting. If you know a book's ISBN, or a web pages's URL, the yadkard tool does a great job of providing a filled-out template in useful forms. Similarly, if the book is on Google Books, you can use the Google Books citation tool to generate a citation that links to that url. David Brooks (talk) 04:58, 20 February 2017 (UTC)

Thanks a lot, that's good to know Nimzo321 (talk) 23:33, 21 February 2017 (UTC)

Here's a question. I have seen an interesting case of citation, which the citation has a link to a personal blog or a internet forum article which contains the screenshot/picture of the article. Is it okay to use? For instance, Decentralization policy in K League's first citation is ^[1] Kingkenny1967 (talk) 05:24, 9 March 2017 (UTC)

The place where the screenshot/picture is stored should be a reliable source, at least to the extent that it can be relied on to store a true copy of the original, without making any alterations. A personal blog belonging to someone no one has ever heard of, or a internet forum are not acceptable, even for such a low level of reliability. The rare exception would be if a person who is a reliable source in his/her own right, such as a respected scholar or journalist, has publicly acknowledged his/her identity on an internet forum. The acknowledgement would itself have to appear in a reliable source, not in the internet forum. Jc3s5h (talk) 13:54, 9 March 2017 (UTC)

A tricky question. You are supposed to cite where you saw the source, so (strictly speaking) you would be citing a (presumably reliable) source indirectly via an unreliable source. And how do you know that the image you saw wasn't altered in some way? Not entirely different from a library sending you a photocopy, except that we have greater trust in the institutional integrity of a well-known library. It seems to come back to all knowledge being provisional, and you have to consider any given item in the totality of a broad range of "sources", context, experiences, etc., ~ J. Johnson (JJ) (talk) 00:02, 10 March 2017 (UTC)

We need to cite the source we saw, though a source article or books could be online in several places. Pages from the same edition of a book can be visible in both Google Books and on Amazon. Linking to the source is another thing. Often sources under copyright are posted in places that are violating the copyright, so can't be linked to. StarryGrandma (talk) 00:37, 10 March 2017 (UTC)

Addressing the question of reliability: it there is a probably-reliable source but the only access you have to it is a reference or quotation in an unreliable source (which includes the case when an unreliable website refers to a reliable book) then you can't use it at all. In order for you to use a claim, the whole chain of information flow from a reliable source to the Wikipedia page has to be verified by you. There is a body of opinion that we can link to a copy of a reliable source hosted in an unreliable place, see the essay WP:CONVENIENCE, but this is just a convenience for readers and does not excuse us from the need to check that the copy is faithful to the original. Zero^talk 00:54, 10 March 2017 (UTC)

There is no statement in this guideline requiring an in-line citation to be relevant to the sentence to which it is attached. I propose a sentence could be added under "Text source integrity" along the lines of:

"The citation and any specified page number or quotation must have direct relevance to the sentence or clause against which it is placed"

Oncenawhile (talk) 09:34, 12 March 2017 (UTC)

• Oppose There are at least to issues with that. First there is intentionally no requirement that breaks down the location of the footnote down to a sentence but just to the general neighbourhood, which a word, a half sentence, a a single sentence, several sentences or even a paragraph. Which of those is best depends on the concrete case and is to degree up to the discretion of the editor. Secondly that the footnote/citation must be be related to the content in its neighbourhood (usually source it) is so obvious that we don't need the suggested additional line, which feels to me a bit like WP:CREEP.--Kmhkmh (talk) 11:23, 12 March 2017 (UTC)

I basically agree with Kmhkmh... although, I have seen situations where a (minor) bit of content was edited out of an article, while the citation supporting it was retained (thus resulting in a superfluous citation that does not support any content). However, I don't think that is what the OP is talking about... and superfluous citations can be dealt with through talk page discussion (I have found that, once you explain what happened, few editors object to removing the superfluous citation). Blueboar (talk) 13:06, 12 March 2017 (UTC)

I also agree with @Kmhkmh:, in that the flexibility point in his/her first two sentences is an important requirement I hadn't considered.

There are other reasons why editors try to push citations which are not needed, such as either (a) they have conflict-of-interest-related reasons to add links to a particular work, or (b) they don't like the overall slant of a particular RS so would rather use an RS with their preferred overall slant, even if the latter doesn't actually support the text.

@Kmhkmh: @Blueboar: could I try an amended version on you both:

"The citation and any specified page number or quotation must have direct relevance to specific information in the article, however, this may be as little as a word or a half sentence, or as much as a paragraph."

Oncenawhile (talk) 14:32, 12 March 2017 (UTC)

That is better and certainly addresses my first concern, but I still don't quite see why we need that explicit addition. That citations need to relate to the article's content (in the neighbourhood of the citation) is plainly obvious to me and I haven't come across any serious editor disputing that. Hence there should be no problem with removing superfluous citations (leftover) or intentionally deceiving citations based on the current policy, i.e. additional more explicit policy backup imho isn't really needed here. A person blocking such a removal has no leg to stand on even with the current policy and will be overruled by other editors. So as long as it is not becoming a common phenomenon/problem the current policies should be good enough to deal with occasional case. However if more editors are convinced we need it, I wouldn't object, but as a general rule I prefer the policies as concise and short as possible and not every occasional individual dispute needs to be covered explicitly by policy (attempting to do that exactly leads to WP:CREEP).--Kmhkmh (talk) 14:54, 12 March 2017 (UTC)

Absolutely not. It is allowable to collect all citations at the end of a paragraph; that is not an unusual practice in scholarly writing. Furthermore, the source may support the entire paragraph, not just one sentence. Jc3s5h (talk) 15:32, 12 March 2017 (UTC)

Is there a preferred format for citing a historical document, such as a letter held in special collections at a library? Klossoke (talk) 11:38, 5 April 2017 (UTC)klossoke

There aren't any preferred formats at all. Look at the other citations in the article to see what style is being used (e.g. Citation style 1, Chicago Manual of Style, APA style). If you can identify a style, use whatever format is recommended for that style. If there is no discernible style, you would be doing the encyclopedia a service by first discussing the issue on the article's talk page and then putting all the citations into a uniform style. Jc3s5h (talk) 13:52, 5 April 2017 (UTC)

Ah, do we have any expectation of publication? There is a huge difference between information disseminated widely in a publication where there is at least some nominal overview by the editors, and someone's letter buried in basement of some library. With original materials there are often questions of authenticity, even provenance. If (say) Paula Jones writes an article referring to a letter by Bill Brown, she (and her editors) are vouching for the existence of the letter, and that a particular Bill Brown wrote it. (Just as the editors vouch for Paula Jones.) Where a WP editor finds an interesting (and possibly heretofore unknown) original document, that seems to be more of a case of "first report", and its evaluation more the function of a journalist or researcher. So what is the threshold or character of "publication" that has to be met before we can use something like a letter? ~ J. Johnson (JJ) (talk) 21:44, 5 April 2017 (UTC)

Well certainly a ref has to be at least possibly accessible. A document held in a private collection wouldn't qualify. A semi-public library such as the Boston Athenæum... you normally have to buy a membership (if they'll even let you -- they have to like the cut of your jib I assume), which is a pretty high bar (and although "Researchers can make appointments to see items in the special collections", I doubt that "I just want to check a ref I saw on the internet" would qualify), so I dunno about that...

But I believe I can cite a book written in Serbian and published in 1820, and there are only three copies publicly available, in libraries in Serbia. They're not going to inter-library-loan them -- to check the ref you'd have to go Serbia (and engage a translator too, if you don't know Serbian). Is that really that much different than a single copy of an original document held at (say) the Karpeles Manuscript Library Museum in Fort Wayne?

Of course, a historical document is a primary source, and if it is so obscure that no professional historian has seen fit to reference it in a secondary source... why would we want to? Herostratus (talk) 00:47, 6 April 2017 (UTC)

Exactly.... the problem with using such primary documents is that there is a huge potential for Original Research. Blueboar (talk) 02:29, 6 April 2017 (UTC)

That is why we have the wording in WP:PSTS: "Policy: Unless restricted by another policy, primary sources that have been reputably published may be used in Wikipedia, but only with care, because it is easy to misuse them". (my bolding) There is yards of text in the archives in WT:PSTS about those two words, but the the general consensus is that unpublished stuff in archives is usually not acceptable. That does not mean that this question is unreasonable because a secondary reliable source may reference an item in an archive, but use a referencing style that is not the same as that used in a Wikipedia article. In which case one needs to know how to format the citation. -- PBS (talk) 06:42, 6 April 2017 (UTC)

PBS wrote "That does not mean that this question is unreasonable because a secondary reliable source may reference an item in an archive, but use a referencing style that is not the same as that used in a Wikipedia article. In which case one needs to know how to format the citation." Wikipedia editors cite what they read. If a secondary source cites an item in an archive, the Wikipedia cites the secondary source. If the archive is important, a free-form note could be added to the end of the citation explaining that the secondary source cites an item in an archive, giving as much detail about the archive as seems relevant. Jc3s5h (talk) 12:48, 6 April 2017 (UTC)

A footnote in WP:V about what is published states "This includes material such as documents in publicly accessible archives, inscriptions on monuments, gravestones, etc., that are available for anyone to see." Jc3s5h (talk) 12:52, 6 April 2017 (UTC)

As Blueboar said, there is a huge potential for mischief. E.g., if some anonymous editor from Sarajevo says he's read this 1820 item, how do we know that he's not making it up? Or like Herostratus said: if "no professional historian has seen fit to reference it in a secondary source" (or even just seen it), and if it is significant in some way, then possibly it should be brought to the attention of a professional historian. Otherwise it is not properly within the received knowledge of humankind that encyclopedias summarize.

There is also a bit imprecision as what are "primary documents/sources" in history, and in the sciences. In the sciences a primary source (see WP:PRIMARY) is the vetted and published paper, not someone's lab notes, or a letter to Dad describing how the experiment went. May be some day we'll get a better grasp of just what "publication" is. ~ J. Johnson (JJ) (talk) 20:16, 6 April 2017 (UTC)

Is there any tool to find duplicate citations?--Marc Lacoste (talk) 14:52, 24 March 2017 (UTC)

That's a heavy question. First of all, are you talking about within a given article? Or across all articles (the "main" name space)?

I assume you mean proper citations of sources (i.e., not every use of the <ref>...</ref> tags), and perhaps only "full" citations (with the full details of the source), and not the "short cites" (e.g., "Smith, 2001") that refer to a full citation. Citations encoded in templates (such as {{citation}} and the {{cite xxx}} family) must also be assumed, as the alternative – free-form text – is just too slippery to deal with.

There is also the question of what constitutes a "duplicate". Are two templated citations considered duplicates if one has a spelling error? Or one uses full first names, and the other only first initials? (There's a bot running around that likes to merge "duplicate" refs into named-refs, which I foil with slight differences in punctuation, etc.)

If by "duplicate citations" you mean multiple references to the same source, there is an attractive notion of searching for certain templates with matching values for some parameter (such as |title=). But, again, there is the problem of trivial differences. Various identifiers would seem workable, though citations often don't include them even where they are available. ISBNs show another problem: multiple values for what are essentially the same source.

Another consideration is whether you have a particular task in mind (where a few judicious searches might suffice), or a tool for general use. The latter is at least a two-pipe problem. ~ J. Johnson (JJ) (talk) 21:07, 5 April 2017 (UTC)

Thanks for your reply. I'm not sure I comprehend all your suggestions, so I'll precise my example : I was working to make a WP:good article of Airbus A320 Family and the reviewer noted that it should better comply with WP:DUPCITES : "Please combine precisely duplicated full citations, in keeping with the existing citation style". But there are 145 Airbus_A320_family#References, so if I want to check manually everyone, I have to compare the first one with 144 others, the 2d with 143, etc. : 10440 comparisons. Of course there are shortcuts, buts still, that's a computer job, not a human one. Comparison can be done on the title and/or the url for links. Am I more clear? (what's the bot merging "duplicate" refs into named-refs?) --Marc Lacoste (talk) 14:47, 6 April 2017 (UTC)

It's not enough to compare just the title and URL. Page numbers or other locations within the work (if using citation templates, the page, pages, or loc parameters), edition, and publication date should be compared. For sites that keep changing, the access dates should be compared. If you want to combine citations with different access dates that are otherwise the same, you should view the current version of the source and see if it still supports all the claims it's cited for, and if so, combine them and change the access date to today. Jc3s5h (talk) 15:34, 6 April 2017 (UTC)

reFill will do this! That particular article, however, does not appear to contain any exactly duplicated refs not already combined using ref names: Noyster (talk), 15:45, 6 April 2017 (UTC)

@Jc3s5h : I just want to find duplicates, but to manually deduplicate them.
@Noyster : Thanks! I'll try it on other articles.--Marc Lacoste (talk) 16:48, 6 April 2017 (UTC)

Ah! It is amazing how having a detail or two can collapse a problem from the broad realm of possibilites I described, to a rather mundane one of a particular task in a particular article.

I had forgotten about DUPCITES , but there you have it: "Please combine precisely duplicated full citations." The rationale behind this is that (per standard bibliographic practice) the full citation of a source, with the all of the bibliographical details, is needed only once in an article. Unfortunately, the authors of DUPCITE did not quite grasp this. Coupled with some common WP misapprehensions and mis-practices, such as mixing full citations with in-source locations, so-called "references" such as "Smith, Frank, Book, ... p. 60" and "Smith, Frank, Book, ... p. 70" are not "precisely duplicated", resulting in duplication of the actual full citation.

Allow me to suggest an alternative. First, in a given article each source should have exactly one full citation, no more nor less. This can be located either:

1) in the first note ("reference") where the source is cited,

2) at the end of a chapter or section (not usually done on Wikipedia), or

3) at the end of the article in a special section (typically titled "References", "Sources", "Bibliography", or such).

Second, use short cites ("Smith, 2010, p. 60") as the in-line links to the full citation (where ever it is located).

It's easy! And gets to the result DUPCITES is groping for. ~ J. Johnson (JJ) (talk) 19:49, 6 April 2017 (UTC)

Of course, DUPCITE only applies (as it says) "in keeping with the existing citation style". It does not recommend changing the style to short cites solely to try to avoid duplicated references, if the existing style does not already use short cites. — Carl (CBM · talk) 20:02, 6 April 2017 (UTC)

I stand corrected: that should have been (and now is) WP:DUPCITES (aka MOS:DUPCITES), not WP:DUPCITE. The latter points to what AWB deems a "fix", which I consider quite dubious.

WP:DUPCITES says: "Any method that is consistent with the existing citation style (if any) may be used, or consensus can be sought to change the existing style" [emphasis added]. It then proceeds to describe use of short cites. BTW, I consider multiple but not "precisely duplicated" full citations not a "style", but amateurish. ~ J. Johnson (JJ) (talk) 20:46, 6 April 2017 (UTC)

This is a neutral notice that an RfC on citing regarding use of primary sources has begun at Talk:RuPaul's Drag Race#Request for comment. --Tenebrae (talk) 14:26, 28 April 2017 (UTC)

I'm a bit stumped as to what to do about Excellent Sheep - it seems to me that it is quite likely notable and would survive a AfD, and yet it has all those references, some of which are completely dead and broken and would need time to repair. As far as I can see, the references are serving no useful purpose and I'd be inclined to just prune, but then.. I'm not sure what the best thing to do is. Ideally someone with knowledge of the subject will improve the stub, but meanwhile there are a whole load of refs there which aren't doing anything. JMWt (talk) 18:32, 1 May 2017 (UTC)

IMO, just leave it. I like to do this sort of thing myself just to preclude an AFD on notability grounds. If you feel uncomfortable with a stub (in general) consider merging to a related topic--in the case of a book, its series or its creator. --Izno (talk) 19:23, 1 May 2017 (UTC)

A compromise might be to repair a few import ones and move the rest to th article's talk page, so that it remains available and visible to other editor who might wants to extend the article. Another option is to simply rename the "References" section into "Further reading". Other than I'd also agree with Izno to in doubt just leave it as it is. Also for repair or clarification it might be a good idea to simply check with original editor (first), he is still active. --Kmhkmh (talk) 19:51, 1 May 2017 (UTC)

Not quite sure how to format this one. I have a translation of

• Korotkevich, Ye S., L. M. Savatyugin, and V. A. Morev. "Through drilling a shelf glacier in the region of Novolazarev Station [Skvoznoye burenie shelfovogo lednika v raione stantsii Novolazarevskoy]." Soviet Antarctic Expedition Bulletin 98 (1978): 49-52.

which is the MLA format given at this link. What I have in hand says on the first page "Translated for the National Science Foundation (etc.) by the Al-Ahram Center for Scientific Translations 1979". By WP:SAYWHEREYOUGOTIT I don't think I should be citing the original, but there's no evidence this translation ever appeared as more than a typescript produced by the NSF. Any advice on the right way to cite this? Mike Christie (talk - contribs - library) 00:33, 26 April 2017 (UTC)

There are probably many of these typescripts around. When I was a graduate student, the woman whose official title in the Astronomy department was "Computer" also translated articles from Russian journals for the faculty. Having a reliable translation like this is no different than if you went to a university library, got a copy of the article, and had someone translate it for you. If you want you can put (translated for the National Science Foundation in 1979) after the reference. StarryGrandma (talk) 17:28, 26 April 2017 (UTC)

Translations are not the original, and the differences can be profound. (E.g., canali → canals.) If you didn't read the original (and can't even read the original language, or perhaps just don't trust your grasp of it), you cite the translation. If the translation is a manuscript that was was not itself published, then you might have some RS issues, and might even be verging on OR. If you think there is good enough reason to rely on the translation nonetheless, I would suggest citing it in the usual way (e.g., "Krotokevich, et al., <giving the publication details of the original> translated by ...."), followed by details of regarding the translation, and especially where it is to be found. ~ J. Johnson (JJ) (talk) 22:25, 2 May 2017 (UTC)

I have am impression that footnotes themselves (that is, their text – whatever is between the <ref></ref> tags) must be terminated with a period (unless there is other punctuation), and it seems that most citation templates do that. All examples here (WP:REF) also apparently follow this trend, but I cannot find whether this rule actually exists in WP and where it is codified. Moreover, examples in WP:FN seem to use the final punctuation quite randomly. So the question is: do/should we have any rules regarding this? — Mikhail Ryazanov (talk) 01:22, 25 April 2017 (UTC)

See WP:CITEVAR. Since no single style has been adopted by Wikipedia, how to end a citation will vary from article to article. Jc3s5h (talk) 01:38, 25 April 2017 (UTC)

Which styles (using footnotes/endnotes) do not have terminal punctuation? — Mikhail Ryazanov (talk) 06:14, 26 April 2017 (UTC)

I am not aware of any printed style manuals that suggest citation footnotes with no terminal punctuation. But we do not require that our citations follow any outside model; editors are free to invent a style that does not use terminal punctuation. Jc3s5h (talk) 12:20, 26 April 2017 (UTC)

The CS2 style produced by {{citation}} (rather than the CS1 style produced by the "cite family" of templates) produces commas between items and no terminal full stop. The commas are similar to, for example, Elsevier's numbered style (see Elsevier Standard Reference Styles), but all of Elsevier's styles have a terminal full stop. Personally I would prefer to see the terminal punctuation; it's what I have always recommended to my students (see this summary). Presumably there was some consensus early on not to style CS2 in this way. Peter coxhead (talk) 16:28, 26 April 2017 (UTC)

The postscript parameter for {{citation}} allows the editor to provide terminal punctuation if desired. StarryGrandma (talk) 16:51, 26 April 2017 (UTC)

One way the citation template are designed to be used allows editors to put additional information after the template, separated by commas, knowing that the template generates no final punctuation of its own. So adding final punctuation would break those articles. For example, this is one way to add information about reprints or translations of papers. — Carl (CBM · talk) 16:55, 26 April 2017 (UTC)

But the idea was that in these cases, after the additional information, the period should be put manually? — Mikhail Ryazanov (talk) 03:08, 27 April 2017 (UTC)

I think that many articles did do that, but I don't know that it was considered mandatory. — Carl (CBM · talk) 13:44, 27 April 2017 (UTC)

Ah, CBM makes a good point. Consider a CS2 citation like

• {{citation |last=Smith |first=A. |date=2016 |title=Some book |publication-place=London |publisher=Williams}} → Smith, A. (2016), Some book, London: Williams

As CS2 uses commas between items, you can add extra information manually with the same separator:

• {{citation |last=Smith |first=A. |date=2016 |title=Some book |publication-place=London |publisher=Williams |postscript=, extra1, extra2}} → Smith, A. (2016), Some book, London: Williams, extra1, extra2{{citation}}: CS1 maint: postscript (link)

whereas if the template automatically inserted a stop at the end you'd get

• Smith, A. (2016), Some book, London: Williams. extra1, extra2{{citation}}: CS1 maint: postscript (link)

So I now see why CS2 works as it does.

@Mikhail Ryazanov: so it's optional to insert the "." at the end of a CS2 citation using |postscript=. My experience is that this isn't generally done, but it would be ok if used consistently in an article without changing any existing citation styles. Peter coxhead (talk) 11:17, 27 April 2017 (UTC)

The style I am more used to is {{citation |last=Smith |first=A. |date=2016 |title=Some book |publication-place=London |publisher=Williams}}, extra1, extra2. Even more stuff. which relies on the lack of ending punctuation in the template. — Carl (CBM · talk) 13:44, 27 April 2017 (UTC)

Standard practice (check any style book) is that footnotes (endnotes) have terminal punctuation (e.g., periods), notes being text. However, citations (whether full or short) do not necessarily have terminal periods, because several of them might be strung together in a note.

The default inclusion of a final period in the {cite} templates was poor design, probably arising from a confusion that a citation is a footnote. That alone I consider a good reason for using {{citation}}. ~ J. Johnson (JJ) (talk) 21:55, 27 April 2017 (UTC)

Inclusion of multiple citations in one reference (footnote/endnote) is a rare case, so, in my opinion, it would be better if templates had periods by default, with an option to suppress it.

But anyway, returning to the original question, is the idea that "whatever is between the <ref></ref> tags must be terminated with a period" correct? — Mikhail Ryazanov (talk) 02:05, 29 April 2017 (UTC)

No. It is optional. DrKay (talk) 07:02, 29 April 2017 (UTC)

Notes (as produced by <ref>...</ref> tags) are basically text, just like the main text, which is to say, composed of sentences (citations being a kind of special sentence). So more precisely, they should have terminal punctuation. But not necessarily a period.

Having multiple citations (usually, short cites, but can also be full citations) in a note is uncommon only because most editors don't know how to merge them, and instead string them out as multiple notes. (All of those ^[1][2][3]... sequences are candidates for merging into a single note.) But there are also cases not of multiple citations, but where one wants to append information that is not well handled in a template, such as multiple ISBNs, or a section or paragraph number that one wants along with a page number.

This notion that having to type a single "." is so burdensome it must be done automatically, while not having it requires typing "|postscript=none", seems bass-ackwards. ~ J. Johnson (JJ) (talk) 20:58, 29 April 2017 (UTC)

@DrKay: Any proofs? — Mikhail Ryazanov (talk) 05:24, 2 May 2017 (UTC)

Proof of what? Existence? Easy enough to demonstrate: add a note somewhere that says (more or less) "This source addresses the question of 'What is <blank>?'". QED, terminal punctuation that is not a period.

If you mean "proof" (demonstration?) of some WP policy or rule mandating terminal periods (or such): can you find one that says sentences must end in periods? ~ J. Johnson (JJ) (talk) 22:04, 2 May 2017 (UTC)

I was asking DrKay, who replied: "No. It is optional." to my question about terminal punctuation. So I wanted to see a reference to any WP policy (or at least any style guide) that say so. (J. Johnson, please read this whole section carefully from the very beginning before answering. I have an impression that you do not understand what this discussion is about.) — Mikhail Ryazanov (talk) 09:03, 3 May 2017 (UTC)

You're asking for a rule, and I'm asking "why?". Please read carefully before responding. ~ J. Johnson (JJ) (talk) 21:42, 3 May 2017 (UTC)

Opinions are needed on the following: Wikipedia talk:Citation overkill#Citations. A permalink for it is here. Flyer22 Reborn (talk) 06:19, 9 May 2017 (UTC)

Please see/join this discussion about a bot conversion of some existing reference lists making use of unordered lists (*) instead of definition lists (:). —TheDJ (talk • contribs) 14:07, 12 May 2017 (UTC)

To keep the discussion in one place if interested please comment at Wikipedia talk:Embedded citations#Historical -- PBS (talk) 11:35, 26 May 2017 (UTC)

SV I don't object to your changes, but it seems a random thing to do without a reason. Please explain why you thought the change necessary and if you intended a subtle change in meaning. -- PBS (talk) 11:05, 26 May 2017 (UTC)

Hi PBS, I've been intending to make that edit for ages, because I regularly see editors misunderstand it. There was no change in meaning. SarahSV ^(talk) 20:31, 26 May 2017 (UTC)

Likewise, I have no problem with the re-write and I agree you haven't altered the meaning but I am intrigued as to how editors "misunderstand" it? Betty Logan (talk) 20:46, 26 May 2017 (UTC)

People continue to think it means they must say where they saw a source. But that's not what it's about; it's about when they didn't see the original source material. I think the misunderstanding stems from not realizing that we're referring to a standard academic practice, so mentioning that makes it clearer. SarahSV ^(talk) 03:49, 28 May 2017 (UTC)

I removed the part between the dashes in "If you have read Smith's book yourself—or an academic article or any other kind of source—that's all you have to cite." because it can be read as that some academic article will do instead of Smith. Zero^talk 06:10, 28 May 2017 (UTC)

A side issue regarding this rule, which should be somewhere but not necessarily here, is that the both the original and intermediate sources have to be reliable. Actually this is the main reason for having the rule at all: to prevent reliable sources being cited via undisclosed unreliable sources. But how should this be presented? Zero^talk 06:12, 28 May 2017 (UTC)

I had a disagreement with another editor over the use of URLs in citations that follow a major style guide (the Bluebook for U.S. legal works):

According to this page (see WP:CITESTYLE), Wikipedia does not have a single house style, though citations within any given article should follow a consistent style. A number of citation styles exist including those described in the Wikipedia articles for Citation, APA style, ASA style, MLA style, The Chicago Manual of Style, Author-date referencing, the Vancouver system and Bluebook.

Because the style guides are designed for use in print, they all include the URL spelled out rather than link an element of the citation like the Wikipedia Citation Style 1 templates, which link the title of the web page or article:

• Using cite web: Gee, Mark (November 11, 2016). "How-To: Photograph The Moon". Popular Photography. Bonnier Corporation. Retrieved 14 May 2017.
• Using cite news: Choe, Sang-Hun (May 13, 2017). "North Korea Launches a Missile, Its First Test After an Election in the South". The New York Times. Retrieved 14 May 2017.

The question is how should the URLs be formatted when using one of the major style guides outside Wikipedia, eg. APA, ASA, Bluebook, the Chicago Manual of Style, MLA, or Vancouver system? Should the URLs be formatted as prescribed in the style guides, with the URL displayed in full? Eg. (in Bluebook style):

Ronald Mann, Opinion analysis: Court uses cheerleader uniform case to validate broad copyright in industrial designs, SCOTUSblog (March 22, 2017), http://www.scotusblog.com/2017/03/opinion-analysis-court-uses-cheerleader-uniform-case-validate-broad-copyright-industrial-designs/

And is this type of URL really a "bare URL"? AHeneen (talk) 12:31, 14 May 2017 (UTC)

I wonder about accessibility. For readers using screen reader tools, how are those citations spoken? "aych tee tee pee ess colon slash slash double-yew double-yew double-yew period ..."? I can imagine that such readings would become unintelligible when they include percent encoding and query strings. I would think that all readers, regardless of visual abilities, would prefer url labels that are contextually meaningful rather than 'technically correct' per the style guide.

I suspect that most style guides are focused primarily on printed paper. While articles from Wikipedia can be printed, the encyclopedia is first and foremost an on-line resource. Rules that are necessary for printed paper may not be necessary here; the url rules being an exemplar. If a reader needs the url in its raw form, it is available with a mouse click.

The style used in Star Athletica, L. L. C. v. Varsity Brands, Inc. includes Id. and Ibid. which are dependent on the relative positioning of other citations in the article text. These are very fragile because anyone can move or insert or delete anything at any time and so break the tenuous connections between these dependent citations. Discouraged by WP:CITE; see WP:IBID.

—Trappist the monk (talk) 13:34, 14 May 2017 (UTC)

I replaced all uses of Id. and Ibid. in the article. Regarding readability, I am neutral: one the one hand, I can understand how the plain URLs can be annoying with screen readers, but, on the other hand, there's also the need to have the URL when printing (Citation style 1 templates print the URL). A relevant style guideline is MOS:LINKSTYLE: The text needs to make sense to readers who cannot follow links. Users may print articles or read offline, and Wikipedia content may be encountered in republished form, often without links. AHeneen (talk) 15:20, 14 May 2017 (UTC)

Actually since 26 September 2015 cs1|2 templates did not include urls in print versions (discussion here). Apparently no one noticed so thanks for the prod to get it fixed.

—Trappist the monk (talk) 16:51, 16 May 2017 (UTC)

Also linkrot affects these types of citation. Despite claims to the contrary, some of the URLs in the cited article would not be easy to find on archive.org as insufficient metadata is presented alongside to facilitate a suitable search should the URL itself just go dead. As for typing in URLs from printed material, does that really happen? In almost all such URLs in the cited article, there are over 100 characters to type in, many of which aren't English words. That's unlikely to be hand-typed by anyone ever. The Rambling Man (talk) 15:46, 14 May 2017 (UTC)

Could you also please explain how " Slip op. at 2-3 (quoting 799 F. 3d 468, 471, 491-492 (2015), available at http://www.leagle.com/decision/In%20FCO%2020150819101/VARSITY%20BRANDS,%20INC.%20v.%20STAR%20ATHLETICA,%20LLC" is sufficient enough for (when the link goes dead) anyone to go looking for the archived link? What does "Slip op. at 2-3 (quoting 799 F. 3d 468, 471, 491-492 (2015)" mean to anyone outside the author of the article please? The Rambling Man (talk) 19:53, 15 May 2017 (UTC)

Finally, there's at least one other legal case nom at DYK right now, i.e. R v Jordan (2016). This appears to not use bare URLs. Does that make it wrong, right or something else? It certainly makes for a better reader experience, avoiding those barely parseable URLs. If this article can use this approach, why not Star Athletica, L. L. C. v. Varsity Brands, Inc.? The Rambling Man (talk) 20:49, 15 May 2017 (UTC)

• This guideline already explicitly recommends against the sufficiency of bare URLs: "improving existing citations by adding missing information, such as by replacing bare URLs with full bibliographic citations: an improvement because it aids verifiability, and fights linkrot;" (bold mine). I'm not sure why that is unclear for the OP, they are already being told, in plain English and in simple terms, to convert bare URLs as a best practice. Using MLA or Chicago Manual of Style for guidance where Wikipedia has none is fine, but we have explicit, in-house rules that couldn't be clearer. --Jayron32 16:31, 16 May 2017 (UTC)

• My 2c, as this has popped up in a number of places now. There are multiple competing issues here.

1. Bluebook is a valid citation style

2. As described, Bluebook is a bare citation contains bare URL's, as *Wikipedia* defines a bare citation URL.

3. Wikipedia encourages bare citations citations with bare URL's to be changed to a more informative citation. (WP:CS in general)

4. Wikipedia *heavily discourages* (to the level of having an arbitration case) changing an already established citation style to something else. (WP:CITEVAR specifically)

5. DYK doesnt allow bare citations.

So lets get the easy one out of the way first, DYK is its own project with its own rules. If you want your articles to be eligible for DYK, you have to satisfy the DYK requirements. This means not using a bare citation style URL. You can still use Bluebook or other 'bare' styles, but it wont be eligible for DYK. The ways around this (in order of effectiveness) 1. Change the citation style. 2. Convince reviewers at DYK to IAR this particular rule as in some areas 'Bare' styles are heavily used. 3. Convince DYK to change its rules on a more permanent basis to allow 'bare' citations in some areas. If you dont want to attempt any of those 3 actions, then keeping Bluebook (and other 'bare' styles) means an article is extremely unlikely to be accepted at DYK.

WP:CS in general encourages 'bare' citations to be improved. However WP:CITEVAR is explicit that no citation style can be forced on an article, and has the arbcom case to back it up. As WP:CS is a best practice guideline, it should be followed in most cases as following it results in a better article for the broad audience who are reading wikipedia. It does not mean it has to be followed in all cases, and where it conflicts with CITEVAR, something that has already been in front of ARBCOM is going to come out on top (absent another arbcom ruling).

So to sum up: You can use Bluebook (or other 'bare' styles) if you want, and anyone attempting to enforce a style against WP:CITEVAR risks coming up before ARBCOM, but you have to accept that while using a 'bare' citation style, the article will be locked out of a number of community-led processes like DYK. I would also say that the point of a wikipedia article is to appeal to a broad audience, and be understandable by a broad audience. If you use a citation style that is predominantly used by the legal profession, its not really tailoring the article to a broad audience is it, even if the article itself is a legal-based one. Only in death does duty end (talk) 11:11, 17 May 2017 (UTC)

Bluebook citations do not use bare citations because the URL is accompanied by other information that allows those who understand that citation style to locate the source even if the URL goes dead. Whether a citation is "bare" isn't about whether the URL is directly visible in the article, it's about whether the source can be found even if the URL goes dead. Jc3s5h (talk) 11:35, 17 May 2017 (UTC)

It uses a bare URL. Bare URL's for most wikipedia editors constitute a bare citation where the URL is the primary part of the citation. This isnt a view that I alone hold. You can go argue at DYK that using Bluebook isnt a bare citation style, but as far as I recall that argument was already held and rejected because DYK does not allow bare URL's. You are correct that it isnt entirely a bare-citation style, so I will amend the above. It doesnt change that ultimately if people want to use any style they can, but have to accept where that style conflicts with individual projects, they can either change it or not participate. Only in death does duty end (talk) 11:40, 17 May 2017 (UTC)

If you look at WP:CITE, for which this is the talk page, you find

• improving existing citations by adding missing information, such as by replacing bare URLs with full bibliographic citations: an improvement because it aids verifiability, and fights linkrot;

There are a few other statements in WP:CITE that also indicate that whether a citation is bare or not has to do with the presence or absence of the necessary bibliographic information; there is no mention of any encouragement or discouragement of making the URL directly visible in the rendered article. Jc3s5h (talk) 12:00, 17 May 2017 (UTC)

However Bluebook citestyle routinely contains bare URL's which are disallowed in places like DYK. It still doesnt change the fact that you cant make someone replace bare URL's if they dont want to per CITEVAR. Regardless of what the wider WP:CITE says. Only in death does duty end (talk) 12:10, 17 May 2017 (UTC)

You are correct, no one can make you improve the article (the correct link, by the way, is WP:NOTCOMPULSORY), however, you can also not make anyone at DYK post your article for whatever criteria they want. People have reached a consensus that bare URLs are insufficient. You can't make them post your article at DYK in contravention of that consensus. --Jayron32 14:05, 17 May 2017 (UTC)

I don't believe Bluebook allows bare URLs (as "bare" is understood in WP:CITE). Please provide a link to a law journal article where the journal requires Bluebook and the article contains a bare URL.

The bullet point I mentioned above,

• improving existing citations by adding missing information, such as by replacing bare URLs with full bibliographic citations: an improvement because it aids verifiability, and fights linkrot;

is part of the Generally considered helpful section. Actions in that section may be taken without asking for consensus on the talk page, and in spite of CITEVAR, unless there is something special about the article that would render the word "generally" inapplicable. So if an article contained, throughout its history, only bare URLs (as understood in WP:CITE), any editor could come along and rewrite all the citations to follow the citation style of the editor's choice without seeking consensus on the article's talk page. Jc3s5h (talk) 12:28, 17 May 2017 (UTC)

The way I read it, WP:CS is satisfied by a full Bluebook citation given the text such as by replacing bare URLs with full bibliographic citations, which does not require that a bare URL have 'text'-make sure to read the full clause. So I agree, from this point, this is a thing specific to WP:DYK (per OID). However, WP:Accessibility#Links says to avoid unintuitive link text--I wonder if that suggestion also extends to no link text. @Graham87: Thoughts on no-text URLs as an accessibility issue? What is the difference between a link, no text; a link, bad text, and a link, good text? --Izno (talk) 12:53, 17 May 2017 (UTC)

A link with no text generates exactly the same output for all readers (the URL), which can usually be skipped over fairly easily by arrowing past it. Graham87 13:55, 17 May 2017 (UTC)

Maybe OID is correct to the extent that maybe the editors who determine whether an article appears in DYK won't accept "exposed" URLs. But DYK's Supplementary guidelines states "References in the article must not be bare URLs (e.g., http://example.com or )....

The linked WP:Bare URLs essay states "Note that some citation styles, such as the MLA style, use full bibliographic citation that happen to display the text of the URL in addition to proper identifying information, like the author, date, and title of the publication. These are not considered bare URLs." So if OID is correct, then it appears DYK isn't following its own guidelines. Jc3s5h (talk) 14:49, 17 May 2017 (UTC)

Very good point, here is then full lead from WP:Bare URLs with my emphasis underlined:

A bare URL is a URL cited as a reference for some information in an article without any accompanying information about the linked page. In other words, it is just the text out of the URL bar of your browser copied and pasted into the Wiki text, inserted between the <ref> tags or simply provided as an external link, without title, author, date, or any of the usual information necessary for a bibliographic citation or useful for addressing link rot. Note that some citation styles, such as the MLA style, use full bibliographic citation that happen to display the text of the URL in addition to proper identifying information, like the author, date, and title of the publication. These are not considered bare URLs.

Here is what a true "bare URL" is on lines one and two (when not accompanied by any other bibliographic information) followed by Bluebook-style citations on lines three (an online-only source) and four (a print source, which a third party has made available online):

2. http://www.ipwatchdog.com/2017/03/22/copyrights-supreme-court-star-athletica-v-varsity-brands/id=79767/
3. Gene Quinn & Steve Brachmann, Copyrights at the Supreme Court: Star Athletica v. Varsity Brands, IPWatchdog (March 22, 2017), http://www.ipwatchdog.com/2017/03/22/copyrights-supreme-court-star-athletica-v-varsity-brands/id=79767/ ("The Court did not decide whether the chevron stripes were themselves original and thus subject to copyright protection once removed from the cheerleading uniform.").
4. 17 U.S.C. § 113(a), available at https://www.law.cornell.edu/uscode/text/17/113.

Example 4 is a properly-formatted legal citation to a statute. The citation style of legal citations varies by country. Here, "17 U.S.C. § 113(a)" means Title 17, Section 113, Subsection (a) of the United States Code. The "official" US Code is only published in print and the link is to a third-party source that has posted it online and is merely included in the citation for the benefit of the reader, so according to the Bluebook style, the link is preceded by "available at". That is basically akin to including a Google Books URL in a citation for a page of a print book and including "via Google Books". However, the Bluebook doesn't have a way of denoting the website name of the third-party URL, which is why it's just "available at https://www.law.cornell.edu/uscode/text/17/113". In example 3 above, the source is an internet article not published in print, so the URL is necessary to locate the original source (although the website name is included as part of the citation) and isn't preceded by "available at" because that URL is the original source. The most recent edition of the Bluebook (in 2015) now allows the addition of archival URLs in brackets after the URL, so when others added archival URLs to the article, I reformatted them to match the Bluebook citation style.

Most Wikipedia citations to particular cases (see case citation) are to cases that have been published in print (at least most case citations on Wikipedia), so like the US Code example above, the URL is a third-party source that merely benefits the reader and is not necessary to find the original source. For example, "Brown v. Board of Education, 347 U.S. 483 (1954)." was published beginning on page 483 of volume 347 of the United States Reports law reporter and was a US Supreme Court decision (because only year is provided; lower courts will be specified in the parentheses, eg. "Fla. 1954" indicates a decision of the Florida Supreme Court and "9th Cir. 2017" is a decision of the United States Court of Appeals for the Ninth Circuit). Most published US cases can be quickly found online by entering the by entering the page, volume, and reporter (example for "410 U.S. 113"). However, the article that started this discussion was just decided in March, so it has not been published in a reporter yet. The "slip opinion" (abbreviated "slip op.") was released on the Supreme Court's website, which is why the article repeatedly uses the phrase "slip op." While it should be quite intuitive that the slip opinion being cited is that of the case that is the subject of the article, to avoid any ambiguity, I changed all cases to include the short citation "Star Athletica" before "slip op."

So the bottom line is that Bluebook-style citations and the citations in the article that caused this issue have sufficient information to locate the original source, which may be only available in print, and are not "bare URLs" as defined at WP:Bare URLs. (Note: I will be busy and may not reply until the weekend, May 20-21). AHeneen (talk) 08:21, 18 May 2017 (UTC)

I note you haven't answered a single one of my questions yet have made some edits to the article, I guess to cover some of those issues. You think a normal reader understands what "slip opinion" means? I think you're mistaken. All the jargon, the lingo, the in-universe stuff that is needed to relocate these possible dead links is sufficient to request that we have more human-readable content and less raw http. The bottom line is that raw URLs are not necessary, and should be avoided, as many legal articles seem to achieve without this huge debate to try to keep something which is hideous to look at, hard to maintain and impossible to follow. Let's stick to using regular Wikipedia style with citations and then these particular legal articles might be visible on the main page. Right now, with their plethora of raw URLs, they are not going anywhere near the main page. The Rambling Man (talk) 18:43, 18 May 2017 (UTC)

Also, I checked every single legal case at featured article level, just Regents of the Univ. of Cal. v. Bakke had raw URLs, all of which would be subject to linkrot because the associated information would be probably inadequate to use the internet archive to find the details. Why would we use a style guide in this particular article which our featured articles about legal cases most definitely does not? It seems absurd. The Rambling Man (talk) 19:49, 18 May 2017 (UTC)

What questions haven't been answered? As mentioned in the very first two comments from Cwmhiraeth's talk page and several times here, WP:CITESTYLE allows any citation style and specifically includes the Bluebook, its usage is proper per Wikipedia:Manual of Style/Legal#Referencing style, and that WP:CITEVAR states that the citation style of an article shouldn't be changed because of personal preference or "to make it match other articles" (so your comments about FA law cases is a red herring). While not mentioned previously, note that the Bluebook has been included on this page (WP:CS) since at least as far back as the end of 2008 as a hatnote () and in prose worded similar to the current wording since at least the end of 2011 ().

I explained in my previous comment how the citations are basically structured so that the original source can be found. For sources originally/officially online (example 3 in my previous comment), the citation includes the author, title, website name, and accessdate so like Template:Cite web there is enough bibliographic info to locate the source (if it's still online). For sources originally/officially in print, the links to third-party postings of the content aren't subject to linkrot because there is enough information to locate the original source (even if you may not be familiar with how the citation is structured, because WP:CITEVAR allows Bluebook citations). Although not mentioned in any of my previous comments, WP:Linkrot says (emphasis added): The first way to prevent link rot is to avoid bare URLs by recording as much of the exact title, author, publisher and date of the source as possible. Optionally, also add the accessdate. If the link goes bad, this added information can help a future Wikipedian, either editor or reader, locate a new source for the original text, either online or a print copy. What matters is that someone can find the original source in the case of print sources, not the third party URL online. Plus, the text can be highlighted, right-clicked, and most internet browsers have an option to search the internet (using the browser's default search engine). In the case of laws, most are available from many third parties just from the first page of search results.

While covered by the remark on WP:CITEVAR (no need "to make it match other articles") in the first paragraph of this comment, R v Jordan is a Canadian case and uses a different style guide: the McGill Guide, which the MoS allows ("While any citation style may be used in an article (see WP:CITEVAR), for articles on cases, case law, or subjects which use a large amount of case law, it is recommended that editors use the referencing style for the jurisdiction that heard that case or for which that legal subject applies. ... Canada, consider using the McGill Guide.") While I am not familiar with that style guide, according to page 4 of this overview of McGill Guide citations, the internet references in R v Jordan are not properly formatted as the URL should appear in angle brackets (eg. "<https://www.thestar.com/news/crime/2016/10/16/courts-scramble-under-new-rules-on-trial-delays.html>"). The two raw URLs in Regents of the Univ. of Cal. v. Bakke were not properly formatted: one was a bare URL (presumably added after passing FA) and the other should have been a piped link. I fixed both of them. AHeneen (talk) 09:47, 21 May 2017 (UTC)

• Comment: For the record, I concur with AHeneen's analysis here. Bluebook style is perfectly acceptable, and there is nothing wrong with including URLs in the citation when there is additional identifying information about the source. I should note that the Bluebook's URL requirement (Rule 18.2.1) is designed for print sources (i.e. when it is not possible to create hyperlinks), so that readers can easily access sources on the internet without entering the title into a search engine. When using Bluebook citations on other websites (e.g. on Wikipedia), I think it is perfectly within the spirit of Bluebook Rule 18.2.1 to simply use a hyperlink (in fact, I wouldn't be surprised if future editions of the Bluebook address this issue). For that reason, I use hyperlinks when writing Bluebook-style citations on Wikipedia, though I don't think there's anything wrong with adding the bare URL to the end of the citation. Best, -- Notecardforfree (talk) 05:13, 4 June 2017 (UTC)

Opinions are needed on the following matter: Wikipedia talk:Citation overkill#Should this essay be changed to encourage more citations?. Flyer22 Reborn (talk) 00:51, 9 June 2017 (UTC)

A WP:Permalink for it is here. Flyer22 Reborn (talk) 00:52, 9 June 2017 (UTC)

In the first example for WP:INTEXT which is written in a white box above, the in-text attribution mentions the source for the example statement as "John Rawls", and John Rawls has a Wikipedia page about him, but what if the source lacks credentials and lacks a Wikipedia page about them? Would it still make sense to mention the name of the person in an in-text attribution? I ask this question, because I ran across this issue. I made this 15:25, 12 June 2017 edit where I removed mentioning the author's name and the title of the book in favor of mentioning that the source was a book about forensic anthropology for the in-text attribution. In my edit summary for that edit, I said, "... Mentioning the author's name in the Wikipedia page does not add to the article unless the author had their own Wikipedia page." Due to the size limitations of the edit summary, I was not able to also add in the edit summary that I thought that adding the author's credentials would be another case where mentioning the author as an in-text attribution would be sensible. In a later 05:00, 13 June 2017 edit, I added the author's credentials, but what if the author's credentials were not available? What would be the best in-text attribution then? Would the best in-text attribution consist of mentioning the book's title, Forensic Anthropology, and the author's name, Bradley J. Adams? Would the best in-text attribution consist of just mentioning the author's name, Bradley J. Adams? Would the best in-text attribution consist of just mentioning the book's title, Forensic Anthropology? Would the best in-text attribution consist of mentioning a description of the source as "a book about forensic anthropology"? Most importantly, I want to know about what people think about just mentioning the author's name without mentioning the book's name, without mentioning the book's description and without mentioning the author's credentials, because I think that an in-text attribution that just said that "Bradley J. Adams said" would not help the reader know about the perspective and reliability of the source. The reader would be wondering, "who is a Bradley J. Adams, and is he a reliable source for the statement attributed to him?" I think that this Wikipedia article should have an example like this where an in-text attribution with just the author's name for an author who lacks a Wikipedia page about them is an example, so Wikipedia editors can see if such an example is deemed good, indicated by the green check mark (Y), or deemed bad, indicated by the red cross (N).--Ephert (talk) 18:29, 13 June 2017 (UTC)

I previously discussed at WT:V whether Wikipedia is using too many sources (read more there), though I nearly botched the discussion that I started. This time, I figured out what else to discuss related to this and where to discuss my curiosity about style of referencing: maybe here.

I skimmed through print encyclopedias in past weeks in libraries. I re-realized that most print encyclopedias have seldom or never practiced the style of inline citations. Such practice has been meant usually for academic journals. Over years, inline referencing, by either footnoting or parenthetical referencing, has become a common practice throughout Wikipedia. In print encyclopedias, especially old ones like one of Walt Disney encyclopedias and World Book Encyclopedia, a section or toward the end of an encyclopedia, references, and/or bibliographies has merely been used to merely list references used for sourcing. Those print encyclopedias usually lacked inline citations (footnoting or parenthetical).

However, Wikipedia articles prevalently display inline citations all over articles. Maybe the "Citing sources" guideline is needed to probably combat inaccuracies and vandalism. However, if that's not it, I'm not sure then why else inline referencing has become the growing prevalent style used by most editors. Why not do what most experts have done in the past: merely put sources in "References" or "Bibliography" section, or put them toward the end of the book?

One Citizendium article uses inline referencing, done by experts of the topic. So does one of them, which heavily uses inline citing. In contrast, one Scholarpedia article (didn't know about Scholarpedia until one or two weeks ago) uses the "References" section to list sources, but it doesn't use footnotes or parenthetical referencing. Well, there's WP:NOTPAPER saying that Wikipedia is not a paper encyclopedia. Still, why not do what past print encyclopedias did: just list sources and not make at least one footnote? Since Wikipedia is not paper, why not reformat references to mere list format and no footnotes? --George Ho (talk) 08:50, 15 June 2017 (UTC)

• The difference is that print encyclopedias are written by experts... while Wikipedia is written by amateurs. We do not trust ourselves to always get things right or to know what we are talking about... in line citations give us a way to fact check each other. Blueboar (talk)|
• Blueboar states the reason clearly. Paper encyclopaedias commission experts; we rely on "the public". Inline citations are essential tools to check that opinion, personal observation and original research aren't creeping in. Peter coxhead (talk) 10:26, 15 June 2017 (UTC)

Even articles using just offline (or inaccessible) sources? Even ones using just non-English sources, like Ålgård Line (which I learned via "Random article" tool)? --George Ho (talk) 18:07, 15 June 2017 (UTC)

Yes. Encyclopedias try to summarize human knowledge, but not all human knowledge is recorded in English. And certainly not human knowledge only as found on-line. ~ J. Johnson (JJ) (talk) 18:48, 15 June 2017 (UTC)

It disturbs me how often editors drift into implying that online English sources are somehow superior to any others, as George Ho seems to be doing. We must use all available reliable sources. Peter coxhead (talk) 20:48, 15 June 2017 (UTC)

Um... I also used offline sources for some articles, like Cheers (season 1), where I used microfilms of old newspapers to verify Nielsen ratings. Even I used some Chinese sources to verify some information about dandan youqing. Still, I've occasionally gone to libraries for college work and (mostly) Internet, also used for... sourcing. --George Ho (talk) 22:23, 15 June 2017 (UTC)

Just want to double check something... George, are you perhaps thinking that an "inline" citation must cite an "on line" source? That is definitely not the case. Blueboar (talk) 23:37, 15 June 2017 (UTC)

My original intent was the formatting of references in Wikipedia compared to paper encyclopedias. I'll be specific about an inline citation: an inline citation, either footnote or parenthetical, citing an information. (e.g. Example info.^[1], Example info (Doe, p. 1) ) However, my original intent wasn't discussing online or offline sources. True that online sources are more efficient and faster to check than offline and inaccessible ones. However, online and offline sources are not the main issue, which we can discuss later at either WT:RS or WP:VPP. Back to the original intent, re-reading what the guideline says, a reference usually comes in two forms in one Wikipedia article: an inline citation and a section listing references in full text. In contrast, in paper encyclopedias, a reference is explained in just full detail at the end of either an article or an encyclopedia. Sorry about befuddling the issue a little bit. I hope I'm clear about the main issue as intended in the OP this time. As for the answers, Blueboar's first answer, Peter's first answer, and J. Johnson's answers help a lot and help me answer the questions I may need. Thanks all. --George Ho (talk) 01:43, 16 June 2017 (UTC)

Another note: MOS:REF redirects to this guideline, explaining another reason for coming here to discuss the formatting issue. --George Ho (talk) 01:44, 16 June 2017 (UTC)

• I think Blueboar and Peter missed George's point: on the necessity of inline citations. (Note that he suggests having no footnotes.) Having citations in-line means attribution of specific content (quotations, assertions, etc.) to a specific source. Verifiability is greatly diminished if it requires substantial searching. ~ J. Johnson (JJ) (talk) 18:52, 15 June 2017 (UTC)

I took it for granted that only inline citations allow proper checking; we need to know how to verify every part of an article. So I entirely agree with this point. Peter coxhead (talk) 20:48, 15 June 2017 (UTC)

Does the same apply to featured lists, like list of awards and nominations received by Bradley Cooper? --George Ho (talk) 19:27, 15 June 2017 (UTC)

pls see WP:LISTVERIFY--Moxy (talk) 19:33, 15 June 2017 (UTC)

Another reason is that our articles are mostly edited piecemeal, so closely associating references with the content by citing them inline improves the ‘modularity’, ensuring that the listed sources are actually relevant (or, when using Harvard & similar styles with separate footnotes & refs, at least make this more likely and easier to check).—Odysseus1479 22:14, 15 June 2017 (UTC)

We need some opinions on the following matter: Template talk:Refimprove#Or better. A permalink for it is here. Flyer22 Reborn (talk) 01:47, 17 June 2017 (UTC)

You may be interested in this discussion, which relates to an interpretation of this guideline. Wikipedia:Miscellany for deletion/Wikipedia:Citation underkill. Scribolt (talk) 18:58, 17 June 2017 (UTC)

Not sure we should be spamming new editors with Template:Welcome-citation that links to Wikipedia:Citation underkill over our guidelines or policies on the matter. Was going to ask for deletion as misleading....but perhaps we should just fix the links?--Moxy (talk) 00:14, 24 June 2017 (UTC)

This is not the right venue. I fixed it. There is a link to five pillars and other sources. No worries. QuackGuru (talk) 00:29, 24 June 2017 (UTC)

So your removing the links to a contentious essay with our P/Gs on the matter?--Moxy (talk) 00:34, 24 June 2017 (UTC)

I included different links. Balance is the key. QuackGuru (talk) 00:35, 24 June 2017 (UTC)

So to be clear your still spamming Wikipedia:Citation underkill on new editors pages.--Moxy (talk) 00:37, 24 June 2017 (UTC)

This is the wrong venue. Linking to an essay along with other links is not spamming. What does this have to do with improving this page? QuackGuru (talk) 00:39, 24 June 2017 (UTC)

Your going out of your way to bypass this guideline with an essay you wrote that as you know is contentious.....you dont think its related?--Moxy (talk) 00:44, 24 June 2017 (UTC)

If you think there is a link missing then fix it. QuackGuru (talk) 01:14, 24 June 2017 (UTC)

I have fixed the template to point to this guideline, rather than the essay. Blueboar (talk) 01:33, 24 June 2017 (UTC)

• Template:Welcome-citationunderkill has now been created. WP:ICANTHEARYOU comes to mind.--Moxy (talk) 02:57, 24 June 2017 (UTC)

Moxy and QuackGuru, the "Welcome-citationunderkill" is nominated for discussion. QuackGuru, may I revert Template:Welcome-citation back to its origins, so I can take this to TFD then? Thanks. --George Ho (talk) 03:58, 24 June 2017 (UTC)

I restored it at your request. QuackGuru (talk) 04:44, 24 June 2017 (UTC)

I gave you thanks, QuackGuru. Actually, may you please take it to TFD instead? I voted "delete" on "welcome-citationunderkill". --George Ho (talk) 05:29, 24 June 2017 (UTC)

Why? QuackGuru (talk) 05:58, 24 June 2017 (UTC)

If taken to TFD, and you or someone else nominates it, I would have voted "Keep" because it helps more. If I took it to TFD, I would say that I would request comments on it or something. I don't favor deleting it. Also, see WP:TFD#REASONS. --George Ho (talk) 06:03, 24 June 2017 (UTC)

Actually, I wonder whether you still want to delete it. If not, then let's leave the template alone and use it more often. Would that do? --George Ho (talk) 06:04, 24 June 2017 (UTC)

Greetings and felicitations. I occasionally see in an article's references section an entry along the lines of "Attribution: [Source]", but I've forgotten where in the MoS or elsewhere this specific type of reference is covered. (As opposed to the standard citing of sources.) Would someone please be so kind as to point me in the right direction? —DocWatson42 (talk) 11:57, 26 June 2017 (UTC)