Wikipedia talk:Citing sources Archive 29

This is an archive of past discussions on Wikipedia:Citing sources. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 25

←

Archive 27

→

Tools

(removed long section that was posted) SlimVirgin ^{TALK contribs} 23:35, 18 January 2010 (UTC)

I think that this information should be restored--194.65.151.101 (talk) 17:03, 18 January 2010 (UTC)

It hasn't been removed. SlimVirgin ^{TALK contribs} 23:35, 18 January 2010 (UTC)

It was - compare this with this. At least, the section "citation tools" disapeared--194.65.151.101 (talk) 10:43, 19 January 2010 (UTC)

Making thesis titles italicized

I've recommended changing the citation style of {{cite thesis}} so that it more closely matches current APA guidelines. Please join the discussion at Template talk:Cite thesis. Thanks. Kaldari (talk) 04:23, 20 January 2010 (UTC)

Citing someone else citing something

I'm not at all certain about these changes. I can see the advantages when you're dealing with weak sources ("Susie's blog says that Mike's book says..."), but there are other times when the fact that Expert Emily cites Original Joe actually makes the claim stronger, not weaker. If a standard medical textbook gives you a couple of footnotes behind a sentence, you don't actually need to say "Sure, Janeway's Immunobiology says it, but we can't trust a team of world-famous immunologists to get anything right; they were relying on these three papers for that claim, so this claim is 'really' from the three papers."

I've had an editor claim that essentially any medical fact could be rejected on the grounds that it came from a primary source: textbooks and reviews are liberally studded with footnotes to original experiments, and reports of original experiments are (by definition) always primary sources -- and therefore every single medical fact can be challenged as "only" being supported by a primary source. (You won't be surprised to hear that he only applied this standard to facts he didn't like.)

I agree that there may be some limited circumstances in which identifying all the sources back to the beginning is helpful to the reader (although I propose that if you're that uncertain of the facts, you're better off not including it at all), but that circumstance is frankly quite rare in most scientific topics and essentially non-existent when using high-quality secondary sources. The text as currently written essentially demands that we re-cite every single secondary scientific source named in Wikipedia to include the footnotes from the sources we're currently naming. This is silly at best, and more probably actually unhelpful to the reader. WhatamIdoing (talk) 00:52, 16 January 2010 (UTC)

The requirement you are complaining about, "the text as currently written essentially demands that we re-cite every single secondary scientific source named in Wikipedia to include the footnotes from the sources we're currently naming", does not exist. --Jc3s5h (talk) 01:45, 16 January 2010 (UTC)

It does, actually. See:

Unless you examine the book yourself, your reference is the webpage, not the book, but you must, in turn, make clear that the webpage cited the book. It's important to be clear about this for two reasons: (a) because the credibility of your edit rests on the webpage, which may have misinterpreted the book, and (b) because you must avoid giving the impression that the webpage author is the original source.

I'm pretty good with my native language, and the word "must" doesn't leave open the option of not telling the reader that a medical text, or a review article, cited something other than itself in making a factual claim. For myself, I'm perfectly comfortable telling readers that a fact came out of Janeway's without bothering to "make clear that the webpage cited" something else. WhatamIdoing (talk) 02:40, 16 January 2010 (UTC)

Here's the scenario that needs to be addressed:

Editor reads Janeway's. Editor adds a fact from the highly regarded medical textbook to Wikipedia. Janeway's paragraph about the fact cites a paper about an experiment. Should the editor:

Cite only Janeway's?
Cite only the experiment directly (after duly looking it up)?
Cite both?

Routine practice across thousands of articles is to cite only the secondary source. When editors have cited only the original experiment, they have been told doing so violates (the previous version of) this rule.

The new rule -- and it is a new rule -- says that editors must always cite both. I see no benefit to copying the refs out of the medical textbooks that we're citing. WhatamIdoing (talk) 02:54, 16 January 2010 (UTC)

It's not a new rule on Wikipedia, and it's a standard rule in academia: that you say where you saw the material, and cite the original source where appropriate. You can't say, "According to Smith," when it's really according to Jones. That's the point of the rule. Regarding your example above, it depends on the context. If it would be misleading to cite only Janeways, then you have to cit their source too. If not, not. SlimVirgin ^{TALK contribs} 04:48, 16 January 2010 (UTC)

WhatamIdoing: in context, the rule is saying you must not cite the paper about the experiment if you have not read the paper. You can cite Janeway's and not the experiment paper. If you have a reason to state that Janeway's cited the experiment paper, you can, but you don't have to. If a new requirement that you must trace all statements to their original source were to be instituted, it wouldn't be in a section named "Say where you found the material". --Jc3s5h (talk) 04:59, 16 January 2010 (UTC)

Also, if editors are being told they shouldn't cite only the experiment paper (given that they have "duly looked it up", which I take to mean they read it), no version of the rule under discussion forbids that. If one were to apply that logic, I couldn't use any source I first discovered in a Wikipedia article, because that's where I first found the source, and I'm not supposed to cite Wikipedia articles. Such an interpretation is absurd. The only valid reason to not to rely on the experiment paper alone is that it might be a primary source, and those can be tricky to use. But that is an issue for WP:Reliable sources, not for this guideline. --Jc3s5h (talk) 05:08, 16 January 2010 (UTC)

No, we've actually had problems with editors citing only the experiment, without looking it up, which isn't allowed by any system. But WP:PSTS and WP:MEDRS strongly encourage editors to rely on the secondary source. WhatamIdoing (talk) 05:57, 16 January 2010 (UTC)

This needn't be a primary/secondary issue, so it's best not confuse that issue with this one. What we usually see are secondary sources citing other secondary sources. SlimVirgin ^{TALK contribs} 06:04, 16 January 2010 (UTC)

Nope: That's what you see, not what "we" see. This part of "we" is telling you that medical texts fairly often name primary sources. WhatamIdoing (talk) 06:07, 16 January 2010 (UTC)

I don't think that you're reading what's actually written. Look at it with fresh eyes, as if you were a newbie who didn't already know what we meant. It says "you must, in turn, make clear that [Janeway's] cited [an experiment]". It doesn't say "if it's doubtful or extraordinary or you think that Janeway's might have gotten wrong, then you must make it clear that Janeway's is relying on the scientific literature instead of pulling this supposed fact out of its collective hat". It doesn't say "use your judgment." It doesn't say, "it depends on the context." It doesn't say, "you can, but you don't have to". It actually says, in plain English, you must name your source and you must name the source that your source relies on. No options or exceptions are mentioned.

If what it actually says, in black and white, isn't what you mean for it to say -- and I doubt it is, from your comments, and from the fact that this unqualified requirement is a major departure from both previous guidance and current practice -- then it needs to be changed to say what you mean. WhatamIdoing (talk) 06:07, 16 January 2010 (UTC)

I've qualified it to take your concerns into account. [1] Does this work? SlimVirgin ^{TALK contribs} 06:15, 16 January 2010 (UTC)

Also, this is not a new rule. It was you who removed it just 11 days ago. [2] SlimVirgin ^{TALK contribs} 06:17, 16 January 2010 (UTC)

must vs should

Nope, "must" is a new rule. The previous text never required that the underlying source be named.

You have once again removed the statement that "It is not necessary to say how you obtained a book or other source. A book's contents are assumed to be the same whether you buy it, borrow it from a library, or read an image of it on a computer screen." We have an actual problem with real editors that we are trying to fix with this sentence: people claiming that if you read the Google books copy of ISBN 0123456789, then you can't cite the book because the electronic version is somehow not sufficiently the same as the original paper copy to meet the requirements to "say where you got it". Please restore the text again. WhatamIdoing (talk) 06:43, 16 January 2010 (UTC)

Before you changed it, it said, "Unless you examine the book yourself, your reference is then the Web page, not the book. You should, in turn, make it clear in the reference that the Web page cited the book." [3] This has been standard practice on WP for as long as we've been careful about sourcing, and standard practice in academia too. Anyway, does the new version work for you? [4]

As for your other point, what does Google have to do with buying a book or borrowing it from a library? And can you give us a few examples of editors making that argument about Google? SlimVirgin ^{TALK contribs} 07:44, 16 January 2010 (UTC)

The point is to make it clear that a book reproduced in Google Books is not an "intermediate source" but is exactly equivalent to a physical copy of the book. You may think, as I did, that this is so obvious that it ought not to need clarifying, but apparently not: please read the section almost at the top of this page entitled "Google books". -- Alarics (talk) 09:32, 16 January 2010 (UTC)

If this is an issue that multiple editors are raising, I agree we could add something about it, so long as we don't talk about borrowing books from libraries. But if it's only one or two editors discussing it, it's not worth adding, in my view. Because when we cite, what we're saying is, "I looked at a page, and this is what it said, and that page was attributed to name, title, publisher, date, page number. To get into the various ways in which we can perceive pages—that way lies madness. SlimVirgin ^{TALK contribs} 14:40, 16 January 2010 (UTC)

I took a look at the Google books section above. It's just one editor who misunderstood what the section said. SlimVirgin ^{TALK contribs} 16:13, 16 January 2010 (UTC)

I will note that there are certain cases where citing the original or official records can be argued. For example when one cites laws, government documents like constitutions, and international treaties. Do you cite the website that contains a reproduction of the United States Declaration of Independence or do you cite the United States Declaration of Independence but append something like "Retrieved 2010-01-01 from www.example.com/declaration.html"? Lambanog (talk) 11:03, 16 January 2010 (UTC)

You can do either. I think we need to be careful not to be too fractal-esque about this. We could cite the Declaration, which we saw on this website, which we viewed with this brower, on computer X, with operating system Y. Common sense has to kick in at some point. SlimVirgin ^{TALK contribs} 14:40, 16 January 2010 (UTC)

I'm a little confused about this entire thread. Is anyone seriously suggesting that it is OK to cite a source from some third party work without mentioning that third party and without direct access to the original? That should be unacceptable by any standard. older ≠ wiser 20:33, 16 January 2010 (UTC)

No, it's the other way round. The issue is whether and to what extent, when I cite Book B (which I have read) which cites Book A (which I have not read), I am obliged to mention Book A in my footnote—as in "A 2009, p. 1, cited in B 2010, p. 2". We agree that I should not say, "A 2009, p. 1" alone. The question is whether it is always okay to say, "B 2010, p. 2" alone. My argument is no, that it is often preferable, and sometimes important, to cite both. SlimVirgin ^{TALK contribs} 20:44, 16 January 2010 (UTC)

I've seen at least two unrelated incidents within the last year: I don't pretend that many editors have a problem figuring out our citation norms. However, I don't think that they are unreasonably misunderstanding: The standard title is "Say where you got it", and there's no antecedent for "it". Does 'it' refer to "the book" or "the information"? If the instruction is to "Say where you got the book", then the correct answer is going to be books.google.com, or from the university library, or from the bookstore (or whatever is accurate).

I've also seen a small number of instances in which an editor has removed a proper bibliographic citation and replaced(!) them with URLs to Google books. This could be motivated by any number of behaviors, but a good-faith effort to "say where you read the book" is the one that springs to my mind.

SV, I'm curious what you have against libraries here. You can buy an eBook, but I believe that "borrowing from the library" has such a strong association with dead-tree publications that it should convey the concept of "original paper == faithful non-paper" much more effectively than a reference to any bookseller. WhatamIdoing (talk) 01:28, 17 January 2010 (UTC)

Chicago Manual of Style rule

What we're discussing here is what the above calls rule 17.274: Citations taken from secondary sources (a sadly ambiguous phrase for our purposes). The rule says:

"A secondary source is a source that quotes or paraphrases another source. An example would be Sontag's On Photography cited in Zelizer's book Remembering to Forget. Use the format below only if you are unable to examine the original source material (e. g. Sontag's On Photography) ...

"Give the citation for the original material (use the appropriate citation format for your source e.g. book or article) followed by the words "Quoted in." Then give the citation information for your secondary source as an annotation.

"Reference list example:

"Sontag, S. 1977. On photography. New York: Anchor Books. Quoted in B. Zelizer, Remembering to forget: Holocaust memory through the camera's eye (Chicago: University of Chicago Press, 2003), 11."

This is standard practice in academia, and it's what WP:CITE has recommended for some time, but at the suggestion of WhatamIdoing, it has been changed to say that doing the above is preferable, not that we must do it. The sub-section below is about something else entirely, which is getting mixed up with the above. SlimVirgin ^{TALK contribs} 20:54, 16 January 2010 (UTC)

I think SlimVirgin has mischaracterized what the guideline says. But rather than quibble about our respective interpretations of the guideline, I would like to suggest that it is inherently impossible to explain what we want when the guideline intermingles the case of an editor who has not read the quoted source with an editor who has read it. I suggest the treatment of these two cases be divided into two paragraphs or sections. --Jc3s5h (talk) 21:23, 16 January 2010 (UTC)

So the historians are now all of academia? Chicago is not used in hard sciences, ever. It's only rarely used in even the "soft" sciences.

Unraveling the chain back to the original source makes sense for historians. When you're talking about old medical texts, you want to be able to say that da Carpi cites Mundinus who cites Galen as his authority -- "who said what when" is the history of the idea, after all.

But for current science, nobody does that. You might, for certain purposes, describe something like that in the text, but never in the footnotes. I can't even find anything in CSE that hints at how you'd format it. (We can't follow URM because it solves the problem by directly prohibiting authors from citing secondary sources at all.) I see no reason to force everyone to use the approach best suited for historians: What's functional for history is not optimal for every single subject. WhatamIdoing (talk) 01:56, 17 January 2010 (UTC)

Also, what Chicago is talking about is not how we actually use this section on a daily basis. We use this section primarily (but not exclusively) for non-quotation situations -- for "Susie's blog says that Mike's book says that [name of political party] is a bunch of mindless jerks", in which what we really have is not Mike's book at all, or even a little tiny snip of it, but Susie's personal reaction to Mike's book. I'm not sure that even Chicago would recommend naming Mike's book in this situation; it seems more sensible to say that you read a blog and leave it at that.

In the articles I most often work on, the claims are usually "Sick Susie's blog (or someone's press release) says that this new study proves that ___ cures cancer." Editors should not trust non-reliable sources to get these things right. On the other hand, if Expert Emily cites the study in a proper paper (especially if combined with other studies in a solid review), then there's no need to name all the underlying literature, and you should simply say where you got the information, and not bother the reader with where Expert Emily got the information. WhatamIdoing (talk) 04:36, 17 January 2010 (UTC)

Secondary Citation of Books

It seems to me that, as discussed above, there are two clearcut cases in citing books, with a number of gray areas in between:

A digital source (web page, etc.) that provides a complete scan of a book can be considered to be the same as the book itself and can legitimately be cited as the book.
A source (whether electronic or print) that provides selections from a book should be cited John Doe, Title 1 (further publication details), as cited in Mary Roe, "title 2," (further publication details).

In between those are various gray areas, for which I would suggest (without being too dogmatic):

Partial scans (e.g. Google Books limited preview). Here it seems to be a judgment call, with the intervening source cited when the citing editor feels it limited their access to the book.
Transcripts that purport to provide a complete and accurate text of a book (e.g. Project Gutenberg). Here the intervening source should probably be cited, just as we would cite a specific edition, to reflect the possible variations in the production of the text being cited.

Somehow the guideline should make this clear, without being too verbose. --SteveMcCluskey (talk) 15:14, 16 January 2010 (UTC); edited 15:25, 16 January 2010 (UTC)

Google books isn't citing anything. It is reproducing. The "say where you got it" section talks about sources who cite other sources. That has nothing to do with Google books or the Gutenberg Project. We're mixing up apples and oranges. SlimVirgin ^{TALK contribs} 16:15, 16 January 2010 (UTC)

I think I see where we're talking at cross purposes. The "Say where you got it" section says: "Unless you examine the book yourself, your reference is the webpage, not the book." The situation I (and others) are considering is where an editor does not have a physical copy of the book, but accesses an electronic copy of it through Google books, the Gutenberg Project, etc. The question, in the light of the passage I just quoted, is how should the editor cite that book. For my suggested answers to that question, read above. --SteveMcCluskey (talk) 20:17, 16 January 2010 (UTC)

For something that is presumed to be an accurate reproduction of the original, the citation would be for the original book (article or whatever), with some indication of the reproduction used. older ≠ wiser 20:27, 16 January 2010 (UTC)

When it comes to Google books, there's no need to state that you read it on Google books, or link to the page in question; you just need to cite the book itself. Reading a scan on Google books is the same thing as "examining the book yourself". Note, however, that this doesn't apply to every website. If someone claims to have scanned a book and uploaded the images to his personal website, it cannot be assumed to be an accurate reflection of the source book. Jayjg ^(talk) 04:49, 17 January 2010 (UTC)

One source that provides abstracts, some full scans, and some full text, of books and journals in their area of interest is the Astrophysics Data System. This is what they have to say about giving them credit:

How should I acknowledge ADS?

If you wish to acknowledge us in a publication, kindly use a phrase such as the following:

"This research has made use of NASA's Astrophysics Data System Bibliographic Services" Thanks! [5]

They don't say anything about how to format the citations to the works they provide access to, nor have I come across any citations in books or journals that mention ADS, but I infer from their statement above that they don't consider it necessary to mention them in the citation to one of the works they provide access to. --Jc3s5h (talk) 16:42, 16 January 2010 (UTC)

In the absence of a reason to worry about discrepancies (e.g., you read an early draft that's since been amended), then you should generally not specify what I'll describe as the medium you used. The point is to say where the information came from, not to say whether you read the book on paper, on an Amazon Kindle, or at Google Books. If it's available online, you can provide a link as a convenience to readers, but you shouldn't do anything that misleads anyone into thinking that the medium changes the contents.

The primary problem that we must address here is using an unreliable source's claim about the contents of a reliable source. There was something the other month at Multiple chemical sensitivity that would be relevant: A website said that the Austrian government had made some pronouncement that declared MCS to have a non-psychological etiology. (The actual letter, which was also available online, says nothing of the sort.) Either version meets the minimum standard, but frankly I'd rather have this kind of thing sourced directly to the unreliable blog, rather than misleading the reader into thinking that the blog is reliable because it claims to have accurately represented a government source.

There are certainly times that citing both is relevant, but those may be best handled in text rather than citations (e.g., "Jane said in her book Stuff that Sir Roger de Coverley's letters indicate..."). WhatamIdoing (talk) 02:11, 17 January 2010 (UTC)

One para for accessible source, one para for inaccessible source

To reduce confusion, I suggest replacing the section with two paragraphs, one for the case where the quoted source is not accessible, the other for the case when it is accessible. Please consider the following:

Say where you found the material

Occasionally you may have access to a source (say, Jones' web page) which quotes another source (say, Smith's book, which says "the sun is hot" on page 99). The quote provides all the information you wish to put in the article. The best thing to do would be to obtain Smith's book, but if this is not possible, you could write "the sun is hot (Smith, 2009, p. 99, as cited in Jones, 2010)" (using APA style with a full reference list near the end of the article). If Jones fully accepted Smith's conclusion, you could also paraphrase "Sol has a high temperature (Jones, 2010)". It would be wrong to write "the sun is hot (Smith 2010, p. 99)" because you never read Smith's book.

If you are fortunate enough to have access to both sources, and the Smith book completely covers the point in the article, it suffices to write "the sun is hot (Smith 2010, p. 99)". However, you could cite both, as above, if there is a reason to. For example Smith's work might be self-published, so the citation in Jones would lend credibility to it.

--Jc3s5h (talk) 02:35, 17 January 2010 (UTC)

I'm more concerned with non-quotations, which aren't addressed here. IMO, if you got information about a medical report out of a patient's blog, you preferentially shouldn't use it at all, but if you're going to use it, you should be sure to name the patient's blog as the sole source -- without putting any lipstick on that pig about the blog claiming to be based on a high-quality source. You just say where you got the information, and stop. WhatamIdoing (talk) 04:42, 17 January 2010 (UTC)

What's the problem here?

I am not aware that the guideline being discussed here has been problematical. What is all the fuss about?

Obviously, "Say where you found it" does not mean, Where did you get the book/journal/newspaper/etc. Looking at featured articles, and observing that citations in them do not mention Amazon.com, Barnes & Noble, or the New York Public Library, confirms this. It is also confirmed by the lack of fields for book store=, library=, news stand=, and friend who lent it to you= in citation templates. Why is someone bringing up this red herring?
From that it should follow that a faithful image of pages from a book, journal, etc.—on a web site, in a course syllabus, or wherever—counts as "you examin[ing] the book yourself".
"For example, a webpage may provide information that the page's author attributes to a book. Unless you examine the book yourself, your reference is the webpage, not the book." (Emphasis added) The second sentence in this quotation of the guideline illuminates the first sentence. The second sentence is not talking about Google Books; it is talking about attribution by an intermediate source (what the Chicago Manual of Style, and most of the world outside Wikipedia, refers to as a secondary source).
Using a quotation from an intermediate source should normally be avoided. If you can't find the original online, go to the library! Or, forget the quotation from the original, paraphrase the intermediate source, and cite only the intermediate source. The "quoted in" alternative should only be used as a last resort.
If you use Bloom's summary of Hamlet in an article, you cite Bloom, not Hamlet.
The content on a newspaper's or magazine's web site may not exactly match the printed article. Often, the web version will have updates and revisions that do not appear in the print edition. If you use the web edition, cite the web edition.
"Susie's blog" most likely is not a reliable source for anything. We don't care what she "says that Mike's book says". Unless Susie is really Harold Bloom.
No guideline can capture all good practices in using and citing sources. Editorial judgment is required. For example, if you are relying on Source A, and Source A has an extended discussion of Source B, you should either (1) find Source B and use it exclusively (especially if Source A does not add anything significant), (2) find Source B and cite both, or (3) cite Source A and indicate that Source A is relying on source B. On the other hand, it is not always, or even usually, necessary to cite a source's sources. The Durants' The Story of Civilization is studded with footnotes. Many other history books are not, even though they make as much use of secondary sources as the Durants do. If I cite The Story of Civilization as the source of an historical fact, I usually won't mention footnoted sourc(e). If The Story of Civilization attributes a statement or opinion to another source, I will credit the other source.
Regardless of who does or doesn't follow the Chicago citation format, the material quoted from Chicago above is the usual practice, the preferred practice in Wikipedia, and makes perfect sense.—Finell 02:33, 19 January 2010 (UTC)

An effort to address apparent confusion is not a red herring: Actual editors (note my deliberate use of the plural) have misunderstood the guideline as requiring editors to name the source of the book/magazine/newspaper rather than to name the source of the information. This might be because someone explained things incorrectly to them once upon a time; it might because they have limited English skills; it might be for any number of reasons. What's important is that it has happened, and that we can prevent it from happening again by being clearer about what we mean here.

Clarity on that point is fairly cheap. For example, we could re-title the section from "Say where you found it" to "Say where you found the information". WhatamIdoing (talk) 00:50, 22 January 2010 (UTC)

Minor formatting error

On the Economy of Queensland page there are two references numbered 25. Could someone explain/fix this error? I presume it is caused by a format error but I can't identify it. - Shiftchange (talk) 02:10, 20 January 2010 (UTC)

This is working correctly. Both footnotes are apparently intended to cite the same source. See WP:REFNAME for explanation of how this works. ---- CharlesGillingham (talk) 05:20, 20 January 2010 (UTC)

Actually the problem still exists. It appears/disappears depending on my browser's (which is Seamonkey 1.18) font size and sometimes the repeated reference number is 24. It doesn't appear in Firefox on the same laptop. - Shiftchange (talk) 07:42, 20 January 2010 (UTC)

This sounds like a problem specific to your browser. If you don't mind, could you describe it to the editors at cite.php? --- CharlesGillingham (talk) 09:38, 20 January 2010 (UTC)

My revert

Charles, I reverted one of your edits, the one where you remove "references," and the explanation of what "source" means on Wikipedia. SlimVirgin ^{TALK contribs} 07:10, 20 January 2010 (UTC)

Well, I thought that the "Use of Terms" section should describe how these terms are used in this guideline, and why. This guideline has to be careful to use these terms precisely. In normal usage, there is a lot of overlap between "reference", "source", "footnote" and "citation", but here we have to stick to their core meaning to avoid ambiguity. The term "references" is almost always ambiguous, but luckily more precise alternatives exist, i.e. "source", "citation" or "footnote". Sentences that use "references" may appear clear to the writer, but can be ambiguous for the reader. (It gets worse when people refer to "footnotes" as "references", as they often do. You could, for example, say the "reference containing a reference of the reference" to mean "the footnote containing a citation of the source". It's fine to call footnotes "references" on a talk page or in some other guideline, but it becomes problematic here.)

Anyway, that's why I went through the guideline last year and made sure the use of terms was consistent. (Or was it two years ago? I can't remember.) I did the same tonight, and I thought it might be helpful to make the use of terms section a little more specific to this guideline. I won't revert your revert: the section is perfectly accurate (although a little pointless) as it stands. ---- CharlesGillingham (talk) 10:04, 20 January 2010 (UTC)

I agree. It's just that the term "reference" is used a lot on WP, so we may as well make clear that it's used interchangeably. Perhaps we could try to clarify some more. SlimVirgin ^{TALK contribs} 01:34, 22 January 2010 (UTC)

Embedded links

I'd like to remove Wikipedia:Citing_sources#Embedded_links, because we really don't allow embedded links to be used instead of citations nowadays. Any objections? SlimVirgin ^{TALK contribs} 20:11, 17 January 2010 (UTC)

We don't encourage articles to use only embedded links, but we wouldn't delete one, or reject one from an IP editor, because links were in that form. Also, if one were using parenthetical referencing, one might wish to use both an embedded link and a parenthetical cite as one possible mechanism to give the reader quick access to the source. I think the section should remain, but perhaps state that such citations should be converted to a better format if a willing editor can be found. We should also consider whether there is any decent system that ,in each citation, there is both an embedded links and a more conventional cite. --Jc3s5h (talk) 20:27, 17 January 2010 (UTC)

This page seem to say that the section containing <references /> or {{reflist}} should be called "Notes". When did that happen? I see it called "References" far more often, and that is what I use. I can see the case when there are two sections, one with numbered references and one ith general ones, but what is wrong with "References" when there is only one section? --Apoc2400 (talk) 21:25, 18 January 2010 (UTC)

Both are valid, really, and I agree the page should make it clearer as I've had seen some editors try to argue that References is not valid, when of course it is and makes far more sense, IMHO, for a pure reference section with all the references in a single section. -- AnmaFinotera (talk · contribs) 21:36, 18 January 2010 (UTC)

The "correct" title depends on what style guide you're following (if any). Here are the most common academic systems:

Chicago: "Center the title Bibliography about one inch from the top of the page"[6] (used by historians)
APA: "In APA style, the alphabetical list of works cited, which appears at the end of the paper, is titled 'References.'"[7] (used by sociologists and psychologists)
MLA: "Center the title Works Cited about one inch from the top of the page."[8] (used in humanities)
CSE: "Center the title References (or Cited References) and then list the works you have cited in the paper; do not include other works you may have read."[9] (used by scientists)

Wikipedia overwhelmingly uses ==References== (following CSE and APA standards). ==Notes== is second place (and also gets used for non-citation explanatory text). Other options are distinctly uncommon, but editors are allowed to use their best judgment.

Keeping this page lined up with actual practice has been a challenge for years now: for some reason, the people who are most likely to spend their time working on pages like this (a thankless task) are the editors who are least likely to personally approve of the choices made by the majority of editors. You can boldly correct the page if you think it would be helpful to editors that consult this page to have ==References== used more frequently in the examples. WhatamIdoing (talk) 01:03, 22 January 2010 (UTC)

The standard thing here is to use Notes if there's only one section, and Notes and References if you're using short refs followed by full citations in a references section—that's what I most often see at FAC. But people can choose which they prefer. SlimVirgin ^{TALK contribs} 01:30, 22 January 2010 (UTC)

... which is what the page says: "Sections containing citations are usually called 'Notes' or 'References'." SlimVirgin ^{TALK contribs} 01:32, 22 January 2010 (UTC)

If by "the standard thing" you mean "what is actually done in Wikipedia articles", then you're wrong: Wikipedia overwhelmingly uses ==References== when is there is only one section for citations. (This is not my anecdotal impression: we pulled a random sample of a couple thousand articles last year and ran the numbers.) WhatamIdoing (talk) 23:27, 22 January 2010 (UTC)

If only one header is used, it's going to look odd with footnotes containing commentary in a References section. But regardless, why does it matter? The guideline says people can choose. SlimVirgin ^{TALK contribs} 23:40, 22 January 2010 (UTC)

It is unusual for articles to contain any explanatory footnotes.

Why it matters: Using the most common heading in the examples may make the examples more easily understood by people reading the page. We could "choose" to use ==Argelphratz== in our examples (as this is a technically permitted option), but we don't choose it for the examples here because it wouldn't be as immediately identified by our editors as a heading for citations. It is at least possible that ==References== would be more helpful than the distinctly less common ==Notes== for the same reason. WhatamIdoing (talk) 22:31, 25 January 2010 (UTC)

This is a very good point. The more choices we show the reader, the longer the learning curve, the more confusing and useless the guideline is. I think the right call, in every case, is that the examples should display the most common choice, the first few lines of text should describe the most common choice, and then the remainder of the paragraph should describe the alternatives. This is what we should do if our intent is to be useful to the readers of the guide---- CharlesGillingham (talk) 17:41, 26 January 2010 (UTC)

Notes and References are the most common headers. People also use Bibliography and, less often, Sources. An article coming to FAC with any of these headers would be accepted. There's no need always to pin people down, so long as what they use makes sense, and there's no internal inconsistency or confusion. Also, it's not at all unusual for articles to contain explanatory footnotes; quite the opposite, in my experience. SlimVirgin ^{TALK contribs} 17:59, 26 January 2010 (UTC)

Then you're editing a remarkably unusual set of articles. If you want some data, as opposed to anecdotal impressions, you might try the list at User:WhatamIdoing/Header_frequency. You'll want to re-acquire the headers (it's been many months since this list was generated), but we found very, very few articles that contained any explanatory footnotes.

Speaking very broadly (and from memory), half the articles we looked at used ==References==, a fifth of the articles used nothing (stubs, unref'd articles, or articles using a section heading that we didn't grep for), one-twentieth contained ==Notes==, and one-thirtieth contained both ==Refs== and ==Notes== -- and most of those were short vs full citations, not citations vs explanatory notes. (The balance of articles, for those that have added up one-half, one-fifth, one-twentieth, and one-thirtieth and come up thirteen-sixtieths short of a whole) contained ==External links== or ==See also== but not note- or ref-related headers.)

If you can find a set of articles that shows a significantly different pattern, then I'd definitely like to hear about it. WhatamIdoing (talk) 02:09, 27 January 2010 (UTC)

invisible to the reader?

hey, just for my info, when was the "invisible to the reader" bit added to accessdate info? Um this goes against everything I've seen for the past couple years... not saying it's bad, but, stuff just gets changed too d*mn much. • Ling.Nut 07:18, 24 January 2010 (UTC)

It was added quite recently. A couple of people tried to remove it, myself included, but it was restored. First, there's no need for access dates for things that are in print, because when e.g. newspapers change their articles, they do it throughout the day, so if that's the issue, we'd need time of access, not date. Secondly, there's no need at all to make the access dates invisible. That's not part of any citation style I've ever seen. SlimVirgin ^{TALK contribs} 07:33, 24 January 2010 (UTC)

I have come here from an ongoing thread about this at Template talk:Cite web #Hidden access dates. The "invisible to the reader" stuff was added added in June 2008, as a result of Wikipedia talk:Citing sources/Archive 25 #Retrieval dates for online versions of old printed sources, again, a 2008 (through 2009!) discussion. I agree with SV that there's no need for access dates for newspaper articles, and that the practice of commented-out access dates is inappropriate. Tony1's recent edit saying that the access date should be a UTC timestamp is also not appropriate: first, almost nobody does it that way, and second, almost nobody will do it that way even if it's in the guideline (how many editors can be bothered to generate UTC accurately?). To try to reflect all this I installed this edit: it removes the stuff about commented-out access dates and removes the claim that it's typical to need access dates when citing newspaper articles (that may have been true at some point, but it's not true now). This edit also says that an access date for a web page is required if the publication date is unknown; that's good practice and is relatively common (though perhaps not as common as it should be). Eubulides (talk) 09:30, 24 January 2010 (UTC)

I think that's most appropriate. It doesn't have to be mandated to have accessdate's included every time (and it's not always appropriate), but neither should commenting out accessdates be encouraged, which is what the previous wording seemed to imply. — Huntster (t @ c) 10:46, 24 January 2010 (UTC)

Am I hearing that this conversation is scattered across different forums? That sucks. • Ling.Nut 11:47, 24 January 2010 (UTC)

I believe commenting out the access date was a compromise between those who wanted an access date even if there is a publication date, and those who thought that redundant. Please refer to the archived discussion.

I've modified Eubulides' edit slightly to restore the visible status quo ante of showing only the publication date if there is one and requiring a retrieval date otherwise. --EnOreg (talk) 16:45, 24 January 2010 (UTC)

There is no consensus to remove accessdate, so teh edit has been restored. There is no such "status quo" claiming that accessdate should only be used if there is no publication date. They are separate parameters, with separate purposes, and both valid per the community usage and lack of consensus to change it. -- AnmaFinotera (talk · contribs) 17:10, 24 January 2010 (UTC)

The status quo of this guideline before Eubulides' edit today was to hide the access date if there is a publication date. This has been a long standing consensus reached on this very discussion page. The new discussion in this section has only started this morning which is way too little time to claim any new consenus, in particular on such a fundamental guideline.

Eubulides' edit removed hidden access dates which I mostly agree with even though there was no proper consensus. But the new wording suggests that there should be a visible access date for Web sources even if there is a publication date anyway. I disagree with this detail and as there was no time for consensus it needs to be changed back. That's what I tried to do while preserving the rest. My change was reverted by AnmaFinotera.

If we don't find a way to remove only the contentious part from the new version then we need to revert all edits since this morning back to the last proper consensus. I would hope we can find a more constructive alternative. --EnOreg (talk) 18:07, 24 January 2010 (UTC)

As I have said before and elsewhere, I am against commenting out parameters, epsecially ones that are widely used. Having it in a form on the documentation page is not the same as making it mandatory, just makes it easier for user to fill in the most common used parameters. Debresser (talk) 12:50, 24 January 2010 (UTC)
I agree with Debresser. Hiding the access date is not helpful. And I am firmly opposed to removing it. Having the access date has helped me reconstruct/recover references from deadlinks/mangled links. older ≠ wiser 14:37, 24 January 2010 (UTC)

The wording I installed was not intended to imply that there must be a visible access date for sources that also have a publication date. (I was trying to reflect consensus, but now see that I worded things inartfully.) I see no consensus for recommending a visible access date in that case; on the contrary, there's considerable sentiment, both here and in previous discussion, for both generally recommending it and generally recommending against it. To better reflect consensus (as opposed to what I wish the consensus would be) I suggest this further change:

"the date you retrieved it (~~required~~ optional if the publication date is ~~unknown~~ known)"

This is a bit clearer anyway. Eubulides (talk) 18:33, 24 January 2010 (UTC)

The main thing is obtaining consensus before you start making all kinds of edits. Debresser (talk) 18:49, 24 January 2010 (UTC)

Gosh. What a novel idea, Debresser. And yet I kinda like it.• Ling.Nut 19:09, 24 January 2010 (UTC)

Yes, that is precisely what my previous comment was attempting to do. The goal is for the page to accurately reflect consensus. Eubulides (talk) 19:31, 24 January 2010 (UTC)

The way it is now makes sense to me, because we need to keep it simple. No access dates required for anything in print. Access dates for material on the web required if the publication date is unknown. The invisible thing was bizarre and is thankfully gone. People can add access dates to whatever else they fancy (I've even seen editors add the date they read a book, which shows how much we confuse them), but at least this guideline doesn't try to force them. SlimVirgin ^{TALK contribs} 03:17, 25 January 2010 (UTC)

Unfortunately the way it is now can easily be misinterpreted to disagree with what the previous comment advises. WP:CITEHOW currently says "Citations for World Wide web articles typically include:... the date you retrieved it (required if the publication date is unknown)". This makes it appear that access dates are recommended even when the publication date is known, which isn't part of the consensus. The proposed rewording to "the date you retrieved it (optional if the publication date is known)" is less likely to be misinterpreted. Eubulides (talk) 04:00, 25 January 2010 (UTC)

Makes sense. I'd be fine with that. SlimVirgin ^{TALK contribs} 04:29, 25 January 2010 (UTC)

Yes, sounds good. Tony (talk) 08:51, 25 January 2010 (UTC)

(unindent) It is my view that for content retrieved from the web, even when it is simply an online copy of a printed text, an access date is always desireable (although not required) and I always provide one. This is because even such web content may change. For example Project Gutenberg texts were mentioned in one of the discussions on this point as stabel sites that need not suppy an acess date. But PG routinely posts corrected versions at the original URL, moving earlier versions to new, more obscure, URLs. The metadata indicating this is typically on a different, albiet related, web page from the actual content. An access date may allow detection of cases where the citation was to an earlier corrected version. Other sources of digitized versions of print content may make similar changes. In cases of link rot an access date may also be useful in finding useful versions of a page on the internet archive Wayback Machine. (Note that the NY Times has announced plans to require registration and payment for viewing online archives after a small number of free accesses per reader.) For pure web content, that is content that is not simply a copy of a printed original, I think that an access date ought to be required, whether a publication date is provided or not. There is always at least the possibility of silent changes, and the even greater possibility of link rot. If users are in fact usign accesss dates in place of publication dates, that is obviously wrong. Correcting that is a matter of education, including clear instructions on citation templates. Let's look at the four major style guides (as summarized in Research and Documentation in the Electronic Age, Fourth Edition by Diana Hacker):

MLA
- For a "SHORT WORK FROM A WEB SITE" (which is stated to be "articles, poems, and other documents that are not book length or that appear as internal pages" )the MLA says "Include the following elements: author's name; title of the short work, in quotation marks; title of the site, italicized; sponsor of the site; date of publication or last update; medium; and your date of access." Note that both date of publication 'and access date are specified. (see http://www.dianahacker.com/resdoc/p04_c08_s2.html)
- For an "ONLINE BOOK" MLA says "Give the print publication information for the work, if available (see items 6–19), followed by the title of the Web site, the medium, and your date of access." Again note date of access is not considered redundant with date of publication. (see http://www.dianahacker.com/resdoc/p04_c08_s2.html)
- For a "WORK FROM A DATABASE" MLA says "For sources retrieved from a library's subscription database, first list the publication information for the source (see items 20–26). Then give the name of the database, italicized; the medium; and your date of access." Again note date of access is not considered redundant with date of publication. (see http://www.dianahacker.com/resdoc/p04_c08_s2.html)
- For an "ARTICLE IN AN ONLINE JOURNAL" MLA says "give publication information as for a print journal .... Then give the medium and your date of access." Again note date of access is not considered redundant with date of publication. (see http://www.dianahacker.com/resdoc/p04_c08_s2.html)
- For an "ARTICLE IN AN ONLINE MAGAZINE OR NEWSPAPER" MLA says Give the author; the title of the article (in quotation marks); the title of the magazine or newspaper (italicized); the sponsor or publisher of the site (use "N.p." if there is none); the date of publication; the medium; and your date of access." (same note and source)
- For an "ENTIRE WEBLOG (BLOG)" MLA says "Give the author's name; the title of the blog, italicized; the sponsor or publisher of the blog (use "N.p." if there is none); and the date of the most recent update. Then give the medium and your date of access." (same note and source)
- For a "POSTING TO AN ONLINE DISCUSSION LIST" MLA says Begin with the author's name, followed by the title or subject line, in quotation marks (use the label "Online posting" if the posting has no title); the title of the Web site on which the discusson list is found, italicized; the sponsor or publisher of the site (use "N.p." if there is none); the date of publication; the medium; and your date of access." (same note and source)
- For a "WORK OF ART" MLA says "For artworks found online, omit the medium of composition and include the title of the Web site, the medium, and your date of access." (same note and source)
- For a "CARTOON" MLA says "To cite an online cartoon, instead of publication information give the title of the Web site; the sponsor or publisher; the date; the medium; and your date of access." (same note and source)
- For a "MAP OR CHART" MLA says "Add the medium and, for an online source, the sponsor or publisher and the date of access." (same note and source)
- For a "RADIO OR TELEVISION PROGRAM" MLA says "For a program you accessed online, after the information about the program give the network, the date, the title of the Web site, the medium ("Web"), and the date of access." (same note and source)
- "Other sources (including online versions)" MLA says "For sources obtained on the Web, ... end the citation with the medium and your date of access." (same note and source)
APA
- For an "ARTICLE FROM AN ONLINE PERIODICAL" APA does not specifically mention an access date, but it example shows one: "Ashe, D. D., & McCutcheon, L. E. (2001). Shyness, loneliness, and attitude toward celebrities. Current Research in Social Psychology, 6(9), 124–133. Retrieved July 3, 2001, from http://www.uiowa.edu/" (see http://www.dianahacker.com/resdoc/p04_c09_s2.html)
- For a "DOCUMENT FROM A WEB SITE" APA says "Give your date of access only if the source itself has no date." (see http://www.dianahacker.com/resdoc/p04_c09_s2.html)
- For an "ENTRY IN A WIKI" APA says "Unlike most other entries for electronic sources, the entry for a wiki includes the date of retrieval because the content of a wiki is often not stable" (see http://www.dianahacker.com/resdoc/p04_c09_s2.html)
CMS
- "NOTE: The Chicago Manual of Style does not advise including the date you accessed a Web source, but you may provide an access date after the URL if the cited material is time-sensitive" (http://www.dianahacker.com/resdoc/p04_c10_s2.html)
CSE
- For "Electronic sources" CSE says "CSE guidelines for Web sites and subscription services require publication information as for books.... In addition, include an update date if one is available and your date of access." (see http://www.dianahacker.com/resdoc/p04_c11_s2.html)
- In accord with this general directive, Access dates are specified (or shown in examples) for HOME PAGE OF A WEB SITE, SHORT WORK FROM A WEB SITE, ONLINE BOOK, and ARTICLE IN AN ONLINE PERIODICAL. "Other sources (print and electronic)" show "[Cited <date>]" in the examples for online versions, but not for print versions. This is the form used elsewhere for access dates (see http://www.dianahacker.com/resdoc/p04_c11_s2.html)

In short, while not all style guides recommend access dates for online references, several do, and only 1 suggests that the presence of an access dae depends on the absence of a publication date. DES ^(talk) 17:58, 25 January 2010 (UTC)

The previous comment says that two style guides (APA and CMS) recommend against access dates in this situation, and two guides (MLA and CSE) favor them. This is evidence against our recommending access dates in this situation, as Wikipedia:Citing sources purposely does not take sides in the what-style-to-use debate. Saying that access dates are optional lets editors choose the style they prefer, which is the consensus position here. Eubulides (talk) 04:08, 26 January 2010 (UTC)

As so often, CMOS sucks itself. I don't go along with that "maybe if it's time-sensitive" bit. Here, we need to keep track of things. Tony (talk) 07:23, 26 January 2010 (UTC)

Having read through the archived discussions I don't see a clear consensus against access dates, I see a number of people on each side, with some of the "against" people being perhaps more persistent on the matter. But loud voice and persistence are not supposed to set consensus. I find the arguments for using access dates significantly stronger than the arguments against them, and that the style guides are split means that using them is not (as some have claimed in the past) contrary to common practice. More over, the vast majority of actual cites that provide any metadata at all on Wikipedia provide access dates, in my experience. If we are supposed to describe what editors actually do, then we should recommend them, or at least not recommend against them. Myself, i am more willing to be prescriptive than some, and i would have no problem with a pretty strong recommendation in favor of access dates. DES ^(talk) 15:03, 26 January 2010 (UTC)

There is no claim here that there is a consensus against access dates, and there's no suggestion that they be prohibited. However, clearly there is no consensus for requiring them either, at least not in the case under discussion. The matter is controversial, and so the guideline should neither require them nor prohibit them. Many citations provide access dates, but more lack them than provide them. The "loud voice and persistence" argument cuts both ways. Eubulides (talk) 04:59, 27 January 2010 (UTC)

Template:Harvard reference

Template:Harvard reference has been nominated for deletion. 70.29.210.242 (talk) 11:17, 29 January 2010 (UTC)

As the template page nor talk page is tagged, the discussion is at Wikipedia:Templates for discussion/Log/2010 January 28. -- AnmaFinotera (talk · contribs) 14:35, 29 January 2010 (UTC)

Endnotes X2 (Citation Manager)

Is anybody aware of a plugin that will allow me to use my endnotes library(it is a citation manager) to quickly add citations to the articles I am writing. It is a longshot but I have become spoiled by the ease with which I can cite, and would LOVE to find a way to use it with wikipedia. Chrislk02 ^{Chris Kreider} 23:39, 29 January 2010 (UTC)

Wikipedia:Embedded citations

I've MfD'd that page because it has a totally unclear status. Posting notice here because it's linked from Wikipedia:Citing sources#Embedded links. Pcap ping 04:05, 30 January 2010 (UTC)

I tagged it as {{supplement}} based on the discussion there. Pcap ping 13:58, 30 January 2010 (UTC)

Duplicating references on an article

Recently I have been confronted with a reference requirement that has gotten me somewhat puzzled. An editor has recently pointed out to me that when references are used, even if they are already listed in a citations section, they must again be listed in a seperate references or bibliography section, in alphabetical order. This to me is an unecessary duplication of effort and clutters up the article unnecessarily. I agree whole heartedly that the article must have references please don't get me wrong, but to list them twice, is just uneeded in my opinion. An example of this is Jared C. Monti. Monti has a "Notes" section with all the references used and know has a duplicate "References" section with the exact same references, only in alphabetical order. My question here is, is this really necessary? --Kumioko (talk) 01:37, 1 February 2010 (UTC)

There is no requirement that sources used inside footnotes must also be listed in a separate "bibliography" section. Some pages do list them that way, particularly if the footnotes have only abbreviated reference information. But if the footnotes have full reference information then an additional section is not necessary nor required. — Carl (CBM · talk) 01:45, 1 February 2010 (UTC)

Thanks, and I agree with needing a full reference if using abbeviated refs. --Kumioko (talk) 01:50, 1 February 2010 (UTC)

It is not required in general, but it is required to conform to the established citation style in a given article. If all the sources in an article are listed in alphabetical order in a References section, except the one you add, a reader is justified in thinking that the only references that were used were the ones listed in alphabetical order, and the reader would think there is no need to search through the notes section for any additional sources. --Jc3s5h (talk) 01:58, 1 February 2010 (UTC)

Sorry Jc3s5h your comment is somewhat confusing. If I create an inline citation they are listed in the order they are added as sources from top to bottom so they would almost never appear in alphabetical order. If the requirement is that they must be visible to the reader in alphabetical order then in essence there must be a separate section. Is that true? Hegvald, if you look at the example I gave above, all that is happening is that I am duplicating references and in this example I think its unnecessary. For another example please look at Smedley Butler. This article uses A LOT of different references, abbreviated citations and a lot of other things and in this case I can see where it is is useful to have a seperate section due to the volume of references. Even in this case though I would argue that the only references that "HAVE" to be in the separate references section are the full references used as abbreviations in the article. Schmidt, 1998, p. 210 fro example would need to have a complete reference in the references section. Just to clarify I am not trying to negate what the other editor has told me or to be argumentative. I am just trying to understand an item that has become a conusing aspect of wiki-editing for me. Especially as of late and as I intend to become much more involved in GA and better class article preperation an reviews I need to get clear on the use of references.--Kumioko (talk) 02:49, 1 February 2010 (UTC)

There are two advantages to using a bibliography or list of references (or "Works cited" or whatever you want to call it).

The first is that footnotes can be made shorter, as noted above, as all the bibliographic information can be relegated to that section. Particularly if you want to use those cumbersome citation templates. (I can see how these may be useful from a database point-of-view, but putting them in the middle of the text makes the text very difficult to work with.)
The second is that it allows the reader to get a quick overview of which references have been used in an article. The latter is more important with large articles with many footnotes. It is very useful when you are familiar with a topic and its sources and wonder if the author(s) of the Wikipedia article has/have utilized some particular source or author that you would expect to see in such an article.

Personally I prefer to use a list of references and like to see it in articles I read. --Hegvald (talk) 02:39, 1 February 2010 (UTC)

(EC)There is no such requirement. Many editors like it, many don't. I personally find it ugly and useless and confusing and would never use it on my own articles. I prefer a single reference section, without the whole redundancy of "footnotes" or "harvard" referencing. It is, however, considered polite to follow whatever style is in place on an article, however, so if the article already uses that style, its best to use it (or a primary editor or the article should be willing to convert your reference for you since it is far less user friendly). If your reference was the first properly formatted one on the article, you have no obligation to change it and the other editor is in no position to say you must. -- AnmaFinotera (talk · contribs) 02:50, 1 February 2010 (UTC)

Using the article Jared C. Monti as an example, it has 13 sources in the Notes section and in the References section. Anyone taking a casual look at it would suppose that every source is listed in both sections. If Kumioko were to add a new fact, and provide a citation to a new source in only the Notes section, a reader who expected the References section to be a complete list of all the sources would be deceived.

This particular article could probably benefit from shortening the citations in some way, but the change should be discussed on the article's talk page.

The type of article that benefits most from separate Notes and References sections are ones where there are many citations to particular pages in a few books. Then a short citation can be given for each page in the correct place, with one entry in the References section to give the full details about the source. --Jc3s5h (talk) 03:14, 1 February 2010 (UTC)

There is no requirement that every source cited in Notes must also be listed in the separate References section, and the contrary practice is quite common. See, for example, Jack the Ripper, a recently-promoted featured article, which contains many citations in the Notes section that are not repeated in its References section. Eubulides (talk) 06:21, 1 February 2010 (UTC)

The practice when I started editing in 2004 was always to list your sources in alphabetical order in a References section at the end. This allowed the reader quickly to scan which sources you'd used. The reason that existed at the time was we used URLs as inline citations. Once we had access to a good footnote system it became unnecessary to list everything in a References section, but still desirable, because using reflist to list the citations leaves them unclear and confusing for the reader. Listing citations in a References section also means you can use short refs inline, which reduces in-text clutter and makes things easier for future editors in edit mode. I agree that, if this isn't currently mandatory, there's no need to try to force it on people, but I still think we should promote it as the ideal. I also find, as an editor, than it helps to clarify in my own mind whether my research has been thorough enough. SlimVirgin ^{TALK contribs} 07:31, 1 February 2010 (UTC)

I disagree. While it may be ideal for you, it is not ideal for everyone and its better to promote the actual guidelines which allows editors to use what is best for them an appropriate for the type of article, and future editors coming after the first to use a valid style needs to just respect their style and keep it unless they get consensus on that article to change it. Trying to force people to use a cumbersome and rather ugly system (to some of us) is not improving the encyclopedia. Promoting referencing period is. -- AnmaFinotera (talk · contribs) 14:18, 1 February 2010 (UTC)

I disagree with your disagree, and raise you an "accept" the idea that the easiest systems should win out, regardless of the present jumble of reference styles in use. What is especially perturbing is the proliferation of completely made-up bibliographical notations. FWiW (with suitable and appropriate tongue in cheek) Bzuk (talk) 14:23, 1 February 2010 (UTC).

Not sure what you mean Bzuk, could you clarify? --Kumioko (talk) 14:28, 1 February 2010 (UTC)

In the publishing world, in which I occasionally dabble, the use of a Notes (footnotes, endnotes, we call them citations) section is an established means of identifying individual sources to verify/attribute reference sources. In scholarly works, the Harvard Citation is a ready means of accomplishing the same effect. The citation links to the bibliographical record (bibliography, in Wiki world called the References) that is typically a list of sources of information that were read/used in research and provide the reader with a further source of resources. The usual style guides employed by a publishing house include the Modern Language Association (MLA) style guide (used in Social Sciences predominately), the American Psychological Association (APA) (Sciences, scholarly works) and the vast number of offshoots, Chicago/Turabian and ad infinitum. The selection of a "house style" is what authors and editors in the "real world" will work with; typically with a general (overall) and line editor following the style guidelines in order to establish a notes and bibliography addendum. With all the availability of various style guidelines, the choice for an editor on a Wiki article should be: choose a style that works, stick with it and not necessarily make up a style that the average reader would probably not understand. FWiW Bzuk (talk) 15:04, 1 February 2010 (UTC).

Thank you, that completely makes sense know. --Kumioko (talk) 15:24, 1 February 2010 (UTC)

correction/addition to 55 days in Peking citation

The ambassador's comments (cited) missed the significance of his remarks: Sir Arthur Robinson - "If China is a cow, Your Majesty, she is indeed a marvelous animal. She gives meat as well as milk...." 68.101.235.54 (talk) 17:32, 1 February 2010 (UTC) please consider adding to his statement <ref> from the screen dialog /<ref> ... 'and she rises to continue to spread her influence ... ' in order to post the importance of the compliment to Chinese influence in the world at this (1900) history mimicked lesson. thank you.. A Heston / Niven lover —Preceding unsigned comment added by 68.101.235.54 (talk) 17:32, 1 February 2010

RFC on introduction of {{r}} for citations

You will may wish to comment at this RFC, which I raised because {{r}} has been imposed without prior discussion on some articles, is not currently document in well-used citation guidelines, and because the implications for some well-used tools are not clear. --Philcha (talk) 10:55, 2 February 2010 (UTC)

List-defined references

This nice method is imho simpler and better than what's currently tought under How to present citations#Footnote system, because it avoids long refs within the main text. Under Help:Footnotes#List-defined references it says: "As of September 2009, the cite software allows named references to be defined within the reference list rather than in the article text. This can make editing articles much easier, particularly on heavily cited sections." See a Norwegian article that uses it. One can also use groups: See List-defined references-section under Wikipedia:Footnotes#Advanced. Dugnad (talk) 07:09, 7 February 2010 (UTC)

That's your personal opinion, of course, and not shared by all editors, or even the majority. Most do not find it "simpler" nor "better", as it requires editing in two windows at once and makes it far more difficult to see what's referencing what while editing. -- AnmaFinotera (talk · contribs) 23:40, 7 February 2010 (UTC)

(ec) Without prejudice to other styles, I do think that WP:LDR shouldn't be ignored on this page. Right now there is no mention of it at all. I'd suggest adding at least a link to it from the section titled "Footnote system" and probably a short description. Dragons flight (talk) 23:54, 7 February 2010 (UTC)

I think a new subsection between "shortened footnotes" and "parenthetical referencing" makes sense. — Carl (CBM · talk) 23:56, 7 February 2010 (UTC)

We should probably mention that system here. On the other hand, once an article has an established system, that should be respected. Getting to choose which citation system to use is a nice incentive to create and new, well-referenced articles <smile> — Carl (CBM · talk) 23:51, 7 February 2010 (UTC)

General references

Okay. I'm wondering if Wikipedia:Citing_sources#General references should be removed or reworded too, because we don't allow this anymore either. SlimVirgin ^{TALK contribs} 21:41, 17 January 2010 (UTC)

We absolutely do allow general references. Cases when general references would be appropriate would be articles about well-known uncontentious topics, which could be supported by general references to textbooks. It would usually be easy to find the supporting text by using the index or table of contents of the textbooks.

Some might claim that it is better to use secondary references rather than text books, which might be thought of as tertiary references. The problem with that is that no self-respecting journal will publish articles on elementary topics, so one would have to go to the archives building at a university library to find journal articles from the 19th century to source some articles. --Jc3s5h (talk) 22:03, 17 January 2010 (UTC)

Can you think of a decent article that relies on general references? I haven't seen one for a long time. SlimVirgin ^{TALK contribs} 01:02, 18 January 2010 (UTC)

Many lists use general references, because it would be quite redundant to have an inline citation to the same reference for each listed item. Dabomb87 (talk) 13:37, 19 January 2010 (UTC)

See for example List of birds of Maryland, which uses a couple inline citations to verify a few sentences in the lead, and a general reference that covers all of the species (imagine using an inline citation of that same web link for all 435 species). Dabomb87 (talk) 13:52, 19 January 2010 (UTC)

I think both should either be removed or reworded to better discourage their use, and encourage changing any instances to more appropriate options. In particular, I think the general references section needs overhauling or removing. While Jc3s5h makes an interesting point, I think that example shows why the section should be removed. Putting some textbooks (or any other reference) and claiming it supports the whole article is rarely accurate nor helpful. Far too often people will put a general reference at the end fo an article, when all it actually supports is one section of text. Or, an article will have a general reference, but then its expanded by others who either add no references or add specific references - what does the general reference then actually cite? Inlines and other referencing systems are far clearer and more explicit. In the case of "uncontentious topics", the references still should be there, with a further reading section, if desired, to note even more supporting text. -- AnmaFinotera (talk · contribs) 22:26, 17 January 2010 (UTC)

AnmaFinotera's comments have some merit. I would add that there are some _______* who go around making subtle changes to numbers just for fun; inline citations would make it easier for those who are not experts to correct this kind of vandalism. However, I think this is the wrong page to make this suggestion. Since this page is mostly about the mechanics of citing sources, not the type or quantity of sources needed, I think this needs to be raised on WP:V.

Please fill in the blank with the nastiest word you can think of.

--Jc3s5h (talk) 22:43, 17 January 2010 (UTC)

WP:V deals only with material that needs a reference -- anything challenged or likely to be challenged, and quotations -- and it requires inline citations. Nothing else needs a reference, so what we're saying here is that material that doesn't need a reference may be supported by a general reference -- a pointless thing to say. I think we should remove this section. It's a throwback to the days when references weren't demanded as much as they are now, and didn't have to be as specific, and retaining it is misleading. SlimVirgin ^{TALK contribs} 01:06, 18 January 2010 (UTC)

I disagree. The verifiability policy underlies the notability guideline, so general references can serve to establish the notability of a topic even in an article that contains no quotations and no statements that have been or are likely to be challenged. Furthermore, verifiability is not the only reason to include references; allowing readers to explore the subject further is another reason, and general references serve that purpose. --Jc3s5h (talk) 01:42, 18 January 2010 (UTC)

Looking again, I see that both WP:V and this guideline have leads that emphasize that we provide sources to prove we aren't liars. The benefit of allowing the reader to learn more by reading the references seems to be treated either as an afterthought, or not at all. --Jc3s5h (talk) 01:52, 18 January 2010 (UTC)

I personally dislike general references. However, there are some featured articles that include general references. There are many "perfect stubs" that use solely general references. It's done, and therefore it should be included in this page. If, on the other hand, the section were to be copyedited in a way that includes a convincing plea for inline refs and explicit permission to allow conversion from general refs to inline refs (of any style of inline refs, not just <ref> tags), then you won't hear a word in complaint from me.
I don't exactly mind embedded links, although I consider them inferior and undesirable. I think that they are an excellent method of getting newbies to provide refs. I think that the last paragraph, which discourages its use, is just about right. I strongly support the continued inclusion of the second paragraph, as that improper non-reference external link is a bit of a pet peeve of mine. WhatamIdoing (talk) 02:40, 18 January 2010 (UTC)

There are several valid uses of "general references" (here this means entries in the "references" section that are not pointed to by inline citations). As pointed out above, they are useful in stubs and other short articles, particularly for establishing notability. But they are also useful as a way for editors to note references that should be cited in the article. Often I have seen a random editor add a good, but forgotten, reference to the references section.

Most importantly, as Jc3s5h alludes, general references are useful for articles that are written primarily from textbooks. This is because the best way for a reader to learn about the topic (and, incidentally, verify the content of our article) would be to simply read about the topic in more detail in a good textbook. So providing references to a collection of good textbooks is an important service of such articles. For example, the references section itself can be divided into subsections for "textbooks" and "other references".

Sometimes, when people feel bad about not having an inline citation for a "general reference", they simply attach an inline citation to the first sentence of the article, or explain in a footnote which general references can be consulted for a general overview. Examples: Mathematical logic and Aldol reaction. — Carl (CBM · talk) 03:41, 18 January 2010 (UTC)

Should that kind of resource not be in Further reading, though? The point of a References section is to list the material used as a source, not the material that ought to be used. SlimVirgin ^{TALK contribs} 05:39, 18 January 2010 (UTC)

It's difficult to tell what material has been used as a source at some time in the past, so moving things to "further reading" is always somewhat tricky. The current text says general references are "more likely to be appropriate for relatively undeveloped articles or those covering a very simple or narrow topic." That seems accurate to me. — Carl (CBM · talk) 13:35, 18 January 2010 (UTC)

I would support increasing the strength of the language against embedded links at this time. This style is being phased out. At the same time, the guideline should continue to highlight that citations in any format, recommended or otherwise, are far preferable to no citations at all. As for general references, I disagree that we don't allow this. General references are useful in shorter articles and as a way to support material that does not require an inline citation. Christopher Parham (talk) 15:36, 19 January 2010 (UTC)

What if we described embedded links as forms of temporary references (similar to the discussion in WP:Verification methods). This would kill three birds with one stone -- it would make it clear that these are not preferred, emphasize that they should not be removed and encourage people to use some reference even if they can't figure out how to format it. ---- CharlesGillingham (talk) 05:15, 20 January 2010 (UTC)

Yes, this would be a good solution. Christopher Parham (talk) 15:40, 22 January 2010 (UTC)

FYI: I have reverted a change that required all biographical material to have an inline reference. This was likely added by SlimVirgin during her massive rewrite; it did not exist in the Jan 15 version for instance. [10] The change contradicts the section right below it to boot, and it wasn't discussed here. The lack of inline references has been used by activist admins to delete or prod sourced BLP articles on the grounds that they are "unsourced"; mostly athletes articles that clearly qualify per WP:ATHLETE, but where the entire wiki article is based on a profile on an site that certainly qualifies as WP:RS, e.g. ATP's web site for tennis players. Pcap ping 20:45, 8 February 2010 (UTC)

Clicking on the ref tag when editing

It used to be that when I clicked on the references icon box when editing an article the 2 sets of ref brackets came up with an 'insert text here' comment, the latter being already highlighted, but in recent edits it is no longer being highlighted. Has something changed on the template set-up or have I perhaps inadvertently changed something? Thanks. Eldumpo (talk) 22:12, 9 February 2010 (UTC)

Like this?

<ref>Insert footnote text here</ref> It seems to work for me. I'm using Monobook skin, Mozilla Firefox 3.0.17, Windows XP. --Redrose64 (talk) 22:37, 9 February 2010 (UTC)

It also works for me with Firefox 3.5.7. Log out, and try it as an IP user. If the problem goes away, it has something to do with your preferences. If not, it's probably something to do with your windowing environment (perhaps your background color is white rather than Redrose64's dark blue?). Eubulides (talk) 00:06, 10 February 2010 (UTC)

Thanks for responses. I tried logging out and it still didn't work. I'm not aware of any changes that have been made to my Windows environment since it stopped showing up blue. How do I check that it is still showing up (but as white)? I'm using IE8. Eldumpo (talk) 08:36, 10 February 2010 (UTC)

References on talk pages

I've started a thread at Wikipedia talk:Talk page guidelines#References on talk pages to get input on handling references in draft text on talk pages. If anybody has input or ideas please contribute there. Will Beback talk 21:14, 12 February 2010 (UTC)

I need some guidance

I'm just training myself up to be an article reviewer and I am running afoul where I point out an unsourced claim which someone then informs me is from the source at the end of the following sentence, or possibly at the end of the paragraph.

The problem for me is that there is absolutely no visible difference between a sentence that has been inserted by a diligent editor who has gone on to reference the sentence a little way after and a sentence that has been inserted unverified and unverifiable in the middle of an otherwise sourced piece of text.

I can't see how one is to know unless one questions it and asks for clarification and this leads me to think that - although it sounds extreme - we should begin to consider a guideline that says "where there is a full stop there should be a citation number". Now, it would have to be a guideline rather than a policy; I would hate for us to lose lots of content because a change sent people on a rampage to "clean up" articles of unreferenced sentences. But I can't see any other way to greater protect us from mischievous edits entering into otherwise sound text and to help someone reviewing an article.

I sense this has probably been proposed before, so I would be grateful if anyone can point me towards previous discussion of similar ideas. --bodnotbod (talk) 10:30, 16 February 2010 (UTC)

I'm interested, too. Paradoctor (talk) 11:08, 16 February 2010 (UTC)

WP:When to cite, although I don't entirely agree with it (think it sets the minimum rather than absolute standard; many people treat it as an absolute limit of some sort). The idea of putting a cite after every full stop will meet with scathing resistance, BTW. • Ling.Nut 13:11, 16 February 2010 (UTC)

Not everywhere, it would seem. Paradoctor (talk) 13:31, 16 February 2010 (UTC)

Well, that's a BLP, so the more the merrier! --Redrose64 (talk) 13:42, 16 February 2010 (UTC)

What about Juice_Plus#Nutrients_and_phytochemicals? Paradoctor (talk) 14:10, 16 February 2010 (UTC)

or more specifically, WP:When to cite#Text-source relationship. --Redrose64 (talk) 13:21, 16 February 2010 (UTC)

The essay's talkpage doesn't contain much about the question of citation density. Anything else? Paradoctor (talk) 13:31, 16 February 2010 (UTC)

Isn't the obvious solution just to check the source at the end of the paragraph to see if it backs up the claim you are questioning? Christopher Parham (talk) 14:37, 16 February 2010 (UTC)

There are those who have a problem with that. See also the discussion linked at the end of the section. Paradoctor (talk) 14:51, 16 February 2010 (UTC)

I'm not sure what those discussions have to do with my statement. I am simply saying that if Bodnotbod is encountering this problem, a simple change to his own methods might suffice to resolve the problem, without requiring sweeping guideline changes that would have significant implementation issues. Christopher Parham (talk) 14:57, 16 February 2010 (UTC)

When people follow your advice, this will lead to exactly to this kind of discussion. Paradoctor (talk) 15:11, 16 February 2010 (UTC)

it will not work without a basic understanding of the subject.Wdl1961 (talk) 15:24, 16 February 2010 (UTC)

Christopher makes an entirely valid point. I will endeavour to do so. Of course often sources are printed and one would go to considerable lengths which aren't very practical to someone whose interest in a subject may not extend beyond trying to do a good article review. Thanks for the links to other en:wp pages. I'll have a look at those. --bodnotbod (talk) 15:32, 16 February 2010 (UTC)

The general rule of thumb is that a citation is expected to support the otherwise-uncited text that precedes the citation in its paragraph. There's no need to put a citation after every sentence. Eubulides (talk) 17:12, 16 February 2010 (UTC)

by definition:if you can not detect the error there is not a problem for you to solve.Wdl1961 (talk) 17:44, 16 February 2010 (UTC)

Hmmm.... How are editors supposed to interpret Eubulides "rule of thumb" when adding to unsourced articles? The rule seems to say that after I append a sourced statement δ to a paragraph of unsourced statements α. β. γ., readers will assume earlier statements by other contributors are covered by my source because the paragraph reads "α. β. γ. δ.^source". What can I do to prevent that assumption, assuming I can't just re-cast the sentence to put my statement first? Should I insert a {{fact}} template before my statement (e.g. "α. β. γ.^{[citation needed]} δ.^source"), or after every unsourced statement in that paragraph (e.g. "α.^{[citation needed]} β.^{[citation needed]} γ.^{[citation needed]} δ.^source")? Maybe this isn't as simple as it might appear. - Pointillist (talk) 23:34, 16 February 2010 (UTC)

The rule of thumb I gave is for well-constructed articles. Obviously if one's starting with a poorly-constructed article there will be intermediate states where the article is still poorly-constructed. If it's really bad, adding a {{fact}} tag is appropriate (and this is true even if you're not adding sourced statements). Eubulides (talk) 00:22, 17 February 2010 (UTC)

Bodnot, if you're reviewing an article, why not simply read the source after the paragraph in question? Adding a ref after every sentence or part thereof is untidy looking and should be avoided where it's reasonable to do so, which is almost always. SlimVirgin ^{TALK contribs} 00:32, 17 February 2010 (UTC)

Sorry, I see you've already agreed to do that. :) SlimVirgin ^{TALK contribs} 00:33, 17 February 2010 (UTC)

Perhaps there's too much concern about the visual appearance of inline citations. I agree they're pretty ugly in most browsers, but the recent Over-sourcing? and Can too many references be a bad thing? discussions seem to be almost entirely about "untidy looking" "clutter". There are two other factors we should bear in mind:

Citations are necessary when what you are publishing is "the encyclopedia anyone can vandalize": not even the most ardent editor wants to watchlist each of his/her contributions in perpetuity. If we weren't concerned about clutter surely we'd want as many citations as possible, including use of hard-to-verify references.
Anyway, clutter is optional when rendering pages: it would be easy to exclude citations and references from the default page view, with the option to show them on demand or via a preferences setting. Doing that would sideline most of the angst about {{Citation}} template performance, too.

I know there are other issues, but surely finding solutions to visual clutter is a better way to satisfy Postdlf's-PhD-wife-emetic-test than avoiding citations and reducing verifiability. - Pointillist (talk) 01:34, 17 February 2010 (UTC)

She'll get a kick out of the fact that her academically sensitive stomach could be the basis for an editing standard. postdlf (talk) 14:46, 17 February 2010 (UTC)

The clutter makes articles hard to read and edit in edit mode, and that leads to bad writing. It's just as easy to click on a reference at the end of a paragraph than to click on one at the end of the sentence. I think we get mesmerized by multiple references. If you look at new articles about current events that are being edited by lots of people, it's always a feature of them that there are lots of unnecessary references. Amy Bishop is charged with murder and that sentence will have five refs after it, as though each ref is some kind of magic spell that makes the sentence more accurate. Experienced editors working on well-developed articles don't do that. We shouldn't be encouraging Wikipedians to do what inexperienced editors do. SlimVirgin ^{TALK contribs} 01:41, 17 February 2010 (UTC)

Yes, adding multiple citations for the same statement is one of the other issues—new articles about current events are a problem area generally, arguably outside the remit of an encyclopedia. But ignoring those issues, the fact is that clutter in reading mode – and the difficulty of working with citations and references in editing mode – are both just technical limitations of the current platform. Those issues could be solved, and should be brainstormed properly without assuming Wikipedia's infrastructure is set in concrete. The alternative proposal (that every reader should check the cited source at the end of a paragraph to see which statements in the preceding paragraph are supported) is pretty sloppy thinking, IMO. That wouldn't really work even if they all had Bodleian or Cambridge University Library cards, would it? - Pointillist (talk) 02:10, 17 February 2010 (UTC)

It's fine to say those issues can be resolved technically, but until that happens we have to be able to read what we're editing. What do you see as the difference between clicking on a footnote at the end of a sentence and one at the end of the paragraph? (I'm struggling to see how a Bodleian library card might fit into this. :)) SlimVirgin ^{TALK contribs} 02:57, 17 February 2010 (UTC)

Oh sorry, SlimVirgin, I just meant that – even if every reader had a superb library on hand – looking up each cited-at-end-of-paragraph reference would be such a pain that no-one would do it, i.e. it's better to keep citations closer to the statements they support. You know all this stuff far better than I do, viz. the mind-blowing quality of the referencing in 1948 Palestinian exodus from Lydda and Ramla. I was just disappointed to see so much discussion about visual clutter but no mention of the benefits of fixing the platform. - Pointillist (talk) 09:58, 17 February 2010 (UTC)

Mind-blowing is right, though; that's part of the problem. :) Whenever possible, I try to keep my refs at the end of a paragraph. If there are multiple points that need to be supported, I might write something like,<ref>For the statement about the death, see Smith 1994, p. 1; for the quote from Jones, see Edwards 2010, p. 3.</ref> with links if available. That way it gets squished between one set of ref tags, which is tidier, but people can easily see which source supports which point. SlimVirgin ^{TALK contribs} 10:09, 17 February 2010 (UTC)

Re clutter in edit mode - I try to avoid this by the two-stage reference method, ie using either {{sfn}} or <ref>{{harvnb}}</ref> in the inline text, and get all the {{cite book}} out of the way down at the bottom. --Redrose64 (talk) 10:03, 17 February 2010 (UTC)

Both SlimVirgin and Redrose64 describe excellent ways to handle citations that are the best ways to deal with and avoid the problems noted above. I wish we could make those the standards because there really is no advantage to doing it another way. postdlf (talk) 14:46, 17 February 2010 (UTC)

That was interesting. The more I learn about Wikipedia, the more I'm convinced that we don't need more editors, we need more developers. Paradoctor (talk) 17:58, 17 February 2010 (UTC)

Ideal citation method?

Greetings; can anyone please advise me of what is the preferred method for making citations in wikipedia? I have seen quite a number across different pages. Just wanting to find out what is the ideal. No need for a long explanation, a link will be fine.--TheSoundAndTheFury (talk) 14:11, 17 February 2010 (UTC)

There is no one preferred method. Different editors like different styles, and there are several acceptable ones, including using footnotes, citation templates, harvard ref, the newer list-defined thing, etc. If you read this page (the page itself, not this talk page), you'll see the ones that are considered acceptable. Pick the one you like for use in articles that have no established style, and use the established style in ones that are using a valid one. -- AnmaFinotera (talk · contribs) 14:21, 17 February 2010 (UTC)

Easier said than done, but the intent is a good one. FWiW Bzuk (talk) 14:43, 17 February 2010 (UTC)

I see. Is there a goal that any given page should stick to the same citation method? I am just getting started, and don't know why I didn't join Wikipedia earlier, but I am pretty anal about things like referencing. When a page is somewhat idle, is it okay to overhaul a spotty referencing system, and standardise it? Also, sometimes books referenced in the article and bibliography section also appear in the "Further reading" section. This seems redundant, especially if the "bibliography" method is used. I would like to prevent full book titles from appearing needlessly multiple times, is that okay?--TheSoundAndTheFury (talk) 03:54, 18 February 2010 (UTC)

Yes, the style should be consistent within an article. Its okay to overhaul if it does not have an acceptable standard style (i.e. all bare URLs, half-written refs, etc). Its always acceptable to standardize ref formatting, however if a valid style has been established, it is considered polite to stick with that style unless you achieve talk page consensus to change it (or if a reasonable amount of time, say 2-4 weeks, passes with no objection to a proposal on the talk page). I'm also very anal about referencing, and have my preferred style. Generally, I'll just not work on articles that use another style that I can't stand to use, unless I can get consensus to change the style. For the second question, if something is being used as a ref, it should not also be in further reading. See WP:FURTHER. -- AnmaFinotera (talk · contribs) 04:04, 18 February 2010 (UTC)

Excellent, AnmaFinotera. Thank you. --TheSoundAndTheFury (talk) 08:58, 18 February 2010 (UTC)

Citation of a book

I'm trying to cite the book Big & Rich: All Access using Harvard cites. The book is written by the duo's members, Big Kenny and John Rich, along with Allen Rucker. Since "Kenny" is his first name, how would I cite it? "Big Kenny, Rich and Rucker, p. x"? Ten Pound Hammer, his otters and a clue-bat • ^{(Many otters • One bat • One hammer)} 21:42, 25 February 2010 (UTC)

Big is not the official last name, but is possibly used as such, so this is not an easy issue. Do you have the book? The Library of Congres or other bibliographic information (usually on one of the first pages even before TOC usually provide how the authors are listed in the official catalogues. Alternatively you can try to find the book in a library catalogue or ask a librarian how to do this. They are experts in these issues. Arnoutf (talk) 21:57, 25 February 2010 (UTC)

His real name is William Kenneth Alphin, and most biographical sources list him under "B" for Big. He's also listed as "Big Kenny" in the jacket, not "Kenny, Big." Ten Pound Hammer, his otters and a clue-bat • ^{(Many otters • One bat • One hammer)} 22:01, 25 February 2010 (UTC)

Check the sources through Special:BookSources/1599957175. For example Britannica.[11] ---— Gadget850 (Ed) ^talk 22:05, 25 February 2010 (UTC)

No results on Britannica. BookSources says it's "Big Kenny." Ten Pound Hammer, his otters and a clue-bat • ^{(Many otters • One bat • One hammer)} 22:13, 25 February 2010 (UTC)

One way to cite it is as follows:

John Rich; Big Kenny; Allen Rucker (2007). Big & Rich: All Access. Center Street. ISBN 9781599957173. Wtmitchell (talk) (earlier Boracay Bill) 00:39, 27 February 2010 (UTC)

To that I would add

|ref={{harvid|Rich|Big Kenny|Rucker|2007}}

so that your Harvard referencing will link, assuming that those are of the form

{{harvnb|Rich|Big Kenny|Rucker|2007|p=123}}

--Redrose64 (talk) 09:50, 27 February 2010 (UTC)

Why is the citing of references so complicated?

Wikipedia has a Utopian vision of everyone on the planet contributing to a knowledge database. However, the ability of an average person to cite a source is effectively impossible. I have read and re-read the complicated "language" of citation and cannot figure them out without referring them to an IT major. Simply stating a Web source at the end of a line in an article becomes a convoluted adventure, it makes you wonder how many people are turned off from contributing to Wikipedia because of its esoteric nature. —Preceding unsigned comment added by 216.28.8.149 (talk) 19:26, 23 February 2010 (UTC)

In general, if you don't know how to do something, just ask for help or request it in the appropriate place, normally the article's talkpage. If you have a specific problem, please feel free to contact me on my talkpage. Happy editing, Paradoctor (talk) 20:43, 23 February 2010 (UTC)

Another suggestion: just do your best to give the source, using common sense. Other editors will be along soon enough to add the fancy formatting details or, if the information you give really is incomplete, to request missing data (page numbers, year of publication, whatever). Please try to remember that such requests are not personal criticisms, but rather are meant to be helpful.—Jerome Kohl (talk) 20:51, 23 February 2010 (UTC)

I agree, 216, and I think it's one of our biggest problems. For Wikipedia to continue into the future, we need to keep attracting new editors who will stick with us, but the complexity of the rules is putting people off. SlimVirgin ^{TALK contribs} 10:12, 24 February 2010 (UTC)

What do you complicated about citing sources? Perhaps the first thing to notice is that the system of citing sources, is really a few alternate systems. If you are starting an article, or the first to add sources to an article, you may choose from the system you find easiest. If you are not the first to add sources to an article, just "blindly" follow what the previous editors have done (ie, if they used some sort of template and plugged their website into a certain blank, copy and paste the whole template and plug your website into that blank). Hyacinth (talk) 11:15, 24 February 2010 (UTC)

(ec x 2) I agree that the complexity of the rules is putting people off - in many areas, not just citing. I started an "essay" at User:Philcha/Sandbox/Producing_a_Good_Article "simple, common-sense summaries of these rules that will help editors to get articles to GA status the vast majority of the time" - plus some hints and tools. After a quick read through, I feel it should be simplified further, with additional mid-level essays between the simplified top-level essay and the details of WP's policies and guidelines. I'd also want to use a less formal type than that of WP's policies and guidelines, as I agree with 216.28.8.149's concern over its complicated language.

I still think GA is a reasonable objective, as GA's rules are quite simple.

Would others be interested in comment and/or construction to a "simple guide". --Philcha (talk) 11:25, 24 February 2010 (UTC)

(ec) After another a quick read of User:Philcha/Sandbox/Producing_a_Good_Article, I suggest the top-level should almost a cookbook, in which each item would cover: the simple but common case; how to detect more complex cases and get advice.

Hi, SlimVirgin, thanks for the adjustments to User:Philcha/Sandbox/Producing_a_Good_Article. Sorry for suggesting the objective should change to a cookbook while you were doing this. --Philcha (talk) 12:37, 24 February 2010 (UTC)

To citing sources? Hyacinth (talk) 12:21, 24 February 2010 (UTC)

Re specifying citing sources, I think we should aim for new editors to produce more than bare URLs. --Philcha (talk) 12:40, 24 February 2010 (UTC)

See Wikipedia:Referencing for beginners. Ty 12:48, 24 February 2010 (UTC)

I've had the refTools gadget enabled so long I sometimes forget that some people don't have it. Adding a cite without it would be a royal pain. Adding one with it is as simple as filling in the boxes. What is the downside of changing the default so that it is available for all editors, except for those who might want to turn it off (although I can't imagine why?). I realize this isn't the place to make that happen, but before I propose at Village Pump, I'd like to see some feedback.--SPhilbrick T 13:42, 24 February 2010 (UTC)

Some people don't like filling in boxes and find it easier and more pleasant to type out the data. Ty 12:32, 26 February 2010 (UTC)

Those editors who really prefer to type out the data can, even if the Cite option is available. I still haven't heard a reason why the default assumption is to not show the box. --SPhilbrick T 15:06, 26 February 2010 (UTC)

The RefToolbar is under new development; see Wikipedia talk:RefToolbar#Please try the new version!. ---— Gadget850 (Ed) ^talk 16:43, 24 February 2010 (UTC)

An important thing to remember here is that if an article already has an established referencing style that does not use templates, editors should continue to add references without using templates. So this tool needs to make sure that, if there are not already reference templates, they are not introduced inadvertently by editors who are "just using the citation tool". — Carl (CBM · talk) 16:50, 24 February 2010 (UTC)

IMO the only good reasons for not using citation templates and tools like are RefToolbar: WP is not good at tell new editors what tools are available; the article is old and gets negligible maintenance. At way, the article is likely to one or more of have: refs with inconsistent formatting already; negligible number of refs. In either case, I'd formatting a few with citation templates and wait for a few days. If there are objections, I'd just tell the objectors that I'll go and work on some other article. I would not try to hard-coding formatting and would not spend time to work out any existing formatting rules. —Preceding unsigned comment added by Philcha (talk • contribs) 23:15, 25 February 2010

So, it seems that the citing of sources it complicated because:

there are multiple academic citation styles
there are multiple Wikipedia citations options, such as templates, inline, <ref>...
these are not compatible with each other
the academic citation style and Wikipedia citation option you find in an article is what you are supposed to stick with (meaning everybody needs to know every style & option)

Hyacinth (talk) 12:15, 26 February 2010 (UTC)

"Inline" means that a part of each citation is a marker that says which part of the text is supported by which citation. The rest of the citation is in 1 or 2 parts:

For citations to different pages of a book use, the inline marker links to a part that says e.g. "Author (date), page N". A further part is shared by all citations that use the same book and gives (fairly full) details.
For smaller works, e.g. journal articles or web pages, where you generally don't specify pages, the inline marker can link to the details directly.

Either way, you eventually have to give the details of the source (book, journal article, web page, etc.). The formatting of the details must be consistent within a WP article - the details for books, journal articles, web pages, etc. must have the same elements (title, date, etc.) in the same order and typography.

Citation templates are a way of formatting of the details in a consistent manner. --Philcha (talk) 15:59, 26 February 2010 (UTC)

Agree with the last sentiment completely. The templates are just a useful way of ensure the data is uniformly presented. A new user should *not* be told to use templates, or to cite is a particular way. If they are citing a website, just tell them to give the URL! Someone else will eventually come by and clean it up, and even if they don't, we should only really be concerned that the information is sourced, not how it is visually presented. Yes, having everything nice and pretty is the ideal end result, but isn't that big of a deal for our purposes. — Huntster (t @ c) 15:40, 27 February 2010 (UTC)

Huntster's "just tell them to give the URL" assume's the source is a web site - probably, but not certain. For other types of source, you need most of the info up front, you can't reconstruct from 1 or 2 elements. In such cases the templates tell the original user what info is needed - and even better if tools like refTools highlight the most important items.

This is not just about neatness. If a source is not properly identified, it will not comply with WP:V, and an edit that relies on that source can be removed. --Philcha (talk) 22:07, 27 February 2010 (UTC)

I don't think a journalist or an academic should or would be allowed to go, "Oh, yeah, here's that source or the rest of that source, or here's the part the makes it verifiable and stuff," some indeterminate amount of time after printing something, and still be considered credible. While we should never bite newbies, if you're going to drive a car there are some rules you're expected to follow. Newbies already know how to use a computer, they're already writing about a topic. They cut and paste a little more information.

Can we make it easier? Probably. But without being specific about what could be simpler at some point our complaints began to sound like an infommercial showing someone helpless flailing about with citations: "There has to be a better way!" Hyacinth (talk) 03:09, 28 February 2010 (UTC)

It's simple - citation templates plus a form-based tool such as refTools is the easier way to produce citations. WP makes it look difficult - it buries the info under N levels of guides, instead of providing it up front; and the guides then focus on the internal details of citation templates rather than on the form-based tools. --Philcha (talk) 11:45, 28 February 2010 (UTC)

Amen. Note that I produced the citation I offered in the section below by searching Google Books, then plugging the URL http://books.google.com/books?id=DLJuPQAACAAJ into this tool, and then tweaking the generated cite. Wtmitchell (talk) (earlier Boracay Bill) 02:05, 1 March 2010 (UTC)

Styles and templates

As the project page notes (here), a number of different citation styles exist. Some articles use style X, some use style Y, but articles are supposed to be internally consistent. Differences between styles involve such things as

order of presentation of items
rules for italicization and/or bolding of individual items
inter-item punctuation, number of authors to name individually before saying "et. al."
how many of those named authors to name in associated shortened footnotes or parenthetical references
how to present volume/issue/chapter/work information
etc.

(that's my impression based on discussions I've seen here and based on a quick look at [12], [13], [14], [15], and a few other sources)

The {{citation}} and {{cite xxx}} templates use one particular style which may or may not conform well to any particular style guide outside of Wikipedia. This situation implies that articles using citation styles other than the style supported by those templates might necessarily need to use use hand-crafted citations. I've been wondering if it makes sense to look at adding a style= parameter to the {{citation}}, {{cite xxx}} and {{citation/core}} templates, enhancing them to support that.

I know a bit about the internals of the affected templates, and believe that this is technically feasible without creating a maintenance nightmare. I haven't asked about this on the template talk pages yet, though, and also haven't explored performance impact problems which I half-suspect this might lead to — I thought it better to first ask here whether or not this might be considered a useful thing to explore. Comments? Wtmitchell (talk) (earlier Boracay Bill) 05:41, 28 February 2010 (UTC)

Please see Wikipedia:Centralized discussion/Citation discussion, particularly its Demo of specific proposal section. Eubulides (talk) 06:22, 28 February 2010 (UTC)

It would be technically feasible. However, it would, in practice, defeat one of the advantages of using citation templates: bringing some measure of uniformity to citations within an article (unlike all other publications, uniformity throughout Wikipedia is probably a lost cause). It is hard enough to get editors to avoid mixing {{citation}} and {{cite xxx}} templates in one article, despite clear instructions. Adding a style= parameter would mean that even citations that use one template (or template family) would wind up being non-uniform. No thanks.—Finell 06:35, 28 February 2010 (UTC)

Eubulides, the implementation I have in mind would be via changes to {{Citation/core}}.

Finell, where you see disadvantages, I see advantages --

WP doesn't require uniformity between articles, only within articles
cites now necessarily hand-crafted could be done with templates
all cites done with templates for a particular supported style would be done without style variations between them
consensus-agreed changes to WP's rendering of a particular citation style could be quickly done (for those cites done with templates)
articles could easily be switched from one supported citation style to another

also, {{Citation}} is now able to mimic the {{cite xxx}} styles pretty well — perhaps well enough that some of the {{cite xxx}} templates could be redone to use {{Citation}} with appropriate style-modifying parameters as an intermediary (pursuing that here would lead to a larger discussion which I'd like to avoid, as it's off-topic here). Wtmitchell (talk) (earlier Boracay Bill) 07:19, 28 February 2010 (UTC)

As I said above, the major problem is not in the proposed coding and the additional parameter itself, but in the expectation (contrary to experience) that Wikipedians will use the parameter at all, let alone correctly. That is also the problem with mixing the two types of template: {{Citation}} can mimic the {{cite xxx}} family if a parameter is used correctly; how often do you see that happen in real life? The practical consequence of adding flexibility to template usage is that the relatively small percentage of us who care about consistent citation formatting will have more clean-up to do than we have now. Also, switching articles "from one supported citation style to another" of contrary to the MOS, and just opens up another opportunity for conflict. So, I repeat: No thanks.—Finell 02:52, 2 March 2010 (UTC)

I'm now traveling and will be for perhaps 2-3 weeks. My WP activity has been reduced by this. You probably have a point about the immediate effect of adding flexibility, but I was thinking that the default effect for those editors not using the added parameter(s) would be to continue producing whatever the consensus default format might be for those citation and cite xxx templates (that is, don't change the current default behavior without consensus to do so). Also, without searching for the precise guideline and section, the MOS allows switching cite styles in an article where there is consensus to do so, and I had in mind the mechanics of switching where such a consensus exists; I did not have in mind cite style switches done without consensus. Wtmitchell (talk) (earlier Boracay Bill) 02:28, 3 March 2010 (UTC)

First Hand Source

This question has bugged me about Wikipedia, and I see someone has question a section of an article I added to using first-hand knowledge. How do you cite first hand knowledge? Let's say for example, you were at the fall of the Berlin Wall. I know many sources, photos and even videos exists of this event, but for the sake of the argument let's say everyone went home, wrote it down in their diaries or whatever and no published or photographic evidence exists of this event. Does this mean for all intent and purposes these things never happened because no one bothered to record them in a published source? How do you address when you KNOW something, but you can't find a source online to back it up? -Eaglescout1984 19:19 17 Feb 2010 (GMT) —Preceding unsigned comment added by 69.68.125.121 (talk)

Get it published by a reliable publisher. It's perfectly reasonable that each publication decide what it wants to publish. You don't send knitting patterns to an astronomy magazine. You don't send an article on hot-rodding a Studebaker to a home decorating magazine. You don't send unpublished stuff to Wikipedia. Jc3s5h (talk) 20:29, 17 February 2010 (UTC)

First hand knowledge isn't consistent with the core policy WP:V. First sentence: "The threshold for inclusion in Wikipedia is verifiability, not truth—what counts is whether readers can verify that material added to Wikipedia has already been published by a reliable source, not whether editors think it is true." Christopher Parham (talk) 21:21, 17 February 2010 (UTC)

If no one has bothered to publish it, then it probably isn't important enough for Wikipedia to include it. WhatamIdoing (talk) 03:42, 19 February 2010 (UTC)

Yeah, I guess when people who aren't famous die building something famous, it isn't important to the elite rich who decide what's important enough to get published. -Eaglescout1984 13:04, 5 March 2010 (GMT) —Preceding unsigned comment added by Eaglescout1984 (talk • contribs)

Say where you got it

Editors are advised that there is an ongoing discussion involving WP:SAYWHEREYOUGOTIT at the RS talk page, Wikipedia_talk:Identifying_reliable_sources#.22Say_where_you_got_it.22_..._what_if_I_got_it_from_a_site_that_violates_copyright.3F. --JN 466 13:42, 8 March 2010 (UTC)

BOT that tidies up references

Hi, I don't know if this is the right place to ask... I am looking for a bot that tidys up references; I would like to know if there is a bot that changes the appearance of references, example here instead of loking this this this.--intraining ^{Jack In} 22:28, 11 March 2010 (UTC)

There was a 'bot that did that, it was User:DumZiBoT but that hasn't been active since June 2009. A typical action of that 'bot was like this. I don't know if other 'bots are doing anything similar. --Redrose64 (talk) 22:51, 11 March 2010 (UTC)

I decided to give that bot a try; the results are excellent, it took about 30 mins to run. This is the result. thanks alot for your help:).--intraining ^{Jack In} 23:52, 11 March 2010 (UTC)

Citing images on Wikimedia Commons

I'm writing an article on a historical figure from here in Brisbane. The only reference that I've been able to find on this person's date of death is from his tombstone, and the only online image that I can find of that is on Commons, here:

File:NHS-JCrase1.jpg

The date is clear enough, and I don't think that this is a case of unacceptable circular citations, but I'm a bit perplexed about the best way to actually include this in the article. Any advice? Lankiveil ^{(speak to me)} 05:29, 12 March 2010 (UTC)

Make a footnote explaining where the information comes from. The cite looks solid, that we have no template for it shouldn't stop you. Paradoctor (talk) 11:16, 12 March 2010 (UTC)

Indeed, here is an example of where this has been done before (if using Firefox or Google Chrome, it'll highlight in blue; but IE, it won't, so see ref 123). If you were to include the gravestone image in your article, then even better. --Redrose64 (talk) 11:39, 12 March 2010 (UTC)

I dont think you will have a lot of problem in this case, as the fact that you want to support is unlikely to be challenged anyway. Arnoutf (talk) 11:51, 12 March 2010 (UTC)

Agree with the above. Please put a wikilink to the file in the footnote, though, like this: [[:File:NHS-JCrase1.jpg|Gravestone of So-and-So]]. Eubulides (talk) 19:02, 12 March 2010 (UTC)

Great, thanks all for the advice :-). Lankiveil ^{(speak to me)} 00:40, 13 March 2010 (UTC).

Sources that aren't really sources...

I'm putting together an article on an event that has become subject of various conspiracy theories (from bona fide academic suggestions to outright crackpot). The facts are referenced to reliable sources. And then there's a review of the alternative suggestions. I suspect that putting them all in one basket under ==Sources== header is not appropriate: all those "alternatives" aren't really my sources. It will look like:

Good, Historian (1899). A respectable account of the ****. London: John Murray. - this is the source for facts and figures, cited at least twenty times.
Modern, Revisionist (2003). Another unconventional explanation of the events of ****. Journal of ***, vol. **, pp.***-***. - this is cited three times as an example of modern critical review of historic sources.
Danbrown, Wannabe (2009). Look no further it was ZOG. ISBN ****. - this is mentioned once as just another modern fantasy.

If I use ==Sources== section strictly for my sources I might end up with numerous calls to the same "non-source" book within ==Notes== section. Where should I place full detail on these "other" books and articles? Is there a guideline that will say "keep them all together" or "no, sheeps and goats separately" ? NVO (talk) 13:17, 7 March 2010 (UTC)

If they're not sources, then they shouldn't be cited as sources. There's no need to give bibliographic detail for the Dan Brown wannabe, just as there's no need for the Holy Grail article to give bibliographic details for The Da Vinci Code even though it discusses that book. Instead, cite the review of the unconventional theories. Eubulides (talk) 19:26, 7 March 2010 (UTC)

If there are any works that would be desirable to mention, but aren't used as sources, you can create a "Further reading" section for those. But I agree with Eubulides that fringe works typically needn't be listed at all. If a fringe work is so notable that it has its own article, you might link that in "See also". --RL0919 (talk) 19:54, 7 March 2010 (UTC)

You might like to look at the usual options, which are listed at WP:FOOTERS. WP:FURTHERREADING in particular sounds like a useful option. WhatamIdoing (talk) 02:22, 13 March 2010 (UTC)

"Further reading" sounds like a recommendation to the reader. Advertizing fringe fiction is something that I would rather avoid. NVO (talk) 07:46, 13 March 2010 (UTC)

Removing a citation (link to a source)

It would be helpful if a section would be added to this page about the Removing of Citations. Citations are often removed by editors, sometimes referencing a reason such as WP:RS, mostly without reason. After the removal, the content is left uncited. I think this is a detrimental practice. It immediately puts the content up for removal also. Proper procedure should be: never delete a citation if you can not add a better one instead of it. If you do not have a better one, leave the citation, and merely discuss it on the discussion page. Would anyone like to comment on this? --BalderV (talk) 13:43, 8 March 2010 (UTC)

If a source is unreliable, it shouldn't be cited, and citations to it should be removed. The corresponding article text could also be removed as well, if it appears dubious; or it could be kept with a {{fact}} tag if it seems OK but challengeable; or it could be kept without citation if it's so obvious that it can't reasonably be challenged. So, the practice is not always detrimental; it depends on the circumstances. Eubulides (talk) 17:37, 8 March 2010 (UTC)

Check the diff - the details of the source in the left column will often give a very indication of whether the source is from a good publisher or Anyone's blog. Then:

Eubulides's comments cover unreliable sources such as Anyone's blog.
If the source looks OK:
- Revert so that the citations and associated text is restored. Make the edit summary e.g. "restored source as seems WP:RS - discuss first at Talk if you disagree".
- If the other editor enters a dialogue, keep WP:CIVIL and WP:AGF:
  - The other editor may have found a source that formally is WP:RS but has content problems, e.g. discredited theory, or taken out of context.
  - OTOH there are less benign situations, e.g. a POV-pusher. If so, you'll have to adapt your approach to the details of the situations - but keep WP:CIVIL and WP:AGF, as some POV-pusher try to provoke their opponents. --Philcha (talk) 09:44, 13 March 2010 (UTC)

Behind a paywall or free registration required?

Is there a standard, guideline, or convention for indicating that a citation is behind a paywall or free registration is required to access the page (typically a verified email address). patsw (talk) 01:58, 18 March 2010 (UTC)

No, generally this information is not regarded as critical for inclusion in citations at all. If you feel it's useful for the reader to be aware of this, putting "requires registration" or some such beside the link would seem fine. Christopher Parham (talk) 03:41, 18 March 2010 (UTC)

There is {{rr}}, also accessible via {{PAYWALL}}: {{rr}} Paradoctor (talk) 08:06, 18 March 2010 (UTC)

I know of two templates available for these purposes: {{Registration required}} and {{Subscription required}}. They should be placed inside the reference, but outside the citation template, for example:

<ref>{{cite web |url=http://www.example.com/ |title=Example |accessdate=18 March 2010 }}{{Registration required}}</ref>

<ref>{{cite web |url=http://www.example.com/ |title=Example |accessdate=18 March 2010 }}{{Subscription required}}</ref>

which produce:

"Example". Retrieved 18 March 2010.(registration required)
"Example". Retrieved 18 March 2010.(subscription required)

There may be others that I'm not aware of. --Redrose64 (talk) 14:44, 18 March 2010 (UTC)

{{Registration required}} and {{Subscription required}}; usually placed in the <ref>...</ref> or other tags and after the citation template. ---— Gadget850 (Ed) ^talk 15:37, 18 March 2010 (UTC)

I didn't expect to have to make the case for such templates, but I am happy to. Faced with an article for which there are dozens of citations, and little time to read then, I believe, readers would be aided with the foreknowledge of their encumbrance rather than wait for the cited page to load only to see that they have reached a paywall or "free registration required". patsw (talk) 00:38, 19 March 2010 (UTC)

I'm less certain that {{Registration required}} and {{Subscription required}} are so helpful:

They only appear after the reader has clicking at least the initial linking the main text. How readers (as opposed as editors) do that?
Do {{Registration required}} and {{Subscription required}} c?
I also note that {{Registration required}} and {{Subscription required}} have no docs. If their developers can't be bothered with production docs with examples of all the referring schemes, why should editors be bothered with the templates? ---Philcha (talk) 06:02, 19 March 2010 (UTC)

We could write "registration required" without a template, perish the thought. :) SlimVirgin ^{TALK contribs} 07:52, 19 March 2010 (UTC)

With my typing, ... :-) And I suspect it may still won't work for all referring schemes, e.g. sfn. --Philcha (talk) 08:48, 19 March 2010 (UTC)

The idea is that the user gets the warning when they're about to click the link for which registration or subscription is required. If the page uses two-stage referencing, then whether the first stage is achieved by means of <ref></ref> or {{sfn}} is immaterial, you still put the {{Registration required}} or {{Subscription required}} in such a place that the message will show at the end of the line which contains the clickable URL in question. That is, the primary consideration is to put it put it immediately after whichever of these is being used:

[http://www.example.com/ Example]{{registration required}}
{{cite web |url=http://www.example.com/ |title=Example |accessdate=19 March 2010 }}{{registration required}}

These templates do have docs. Small docs yes, but they do exist. I suspect that they're small because neither takes any parameters. --Redrose64 (talk) 14:39, 19 March 2010 (UTC)

I've expanded the documentation of Template:Registration required along the lines of the doc for {{dead link}}. --Redrose64 (talk) 15:12, 19 March 2010 (UTC)

Cite Style Error ?

Here's my point about the citation style. If I use

It shows in the Citation section, reference to the very first entry from that book. Say page 24. Evenif my citation content says otherwise.

^[1]

Only if I use

That I get desired results. In many cases where the Citation section pointed to Page 185, my cite was actually pointing to somewhere else!. ^[2]

^ Weal, (1999) p. 51–53
^ Weal (1999), page 53.

Can someone shed some light on this ? Thanks Perseus71 (talk) 04:13, 20 March 2010 (UTC)

Without knowing which article this is in it's hard to be sure, but at a guess I'd say that you've used the same value for the name= attribute on more than one <ref></ref> tag. That is to say, your article probably contains both of the following:

<ref name="Johnweal99">Weal, (1999) page 24</ref>

<ref name="Johnweal99">Weal, (1999) page 53</ref>

The two things to remember are (a) ref names must be unique; and (b) it is only necessary to name a <ref></ref> if exactly the same source was used for two or more different items in the article - if the source was only used once, the name is unnecessary.

Let's consider a hypothetical example. Assume that on page 24 of said book, there are two statements "Elephants are big" and "Giraffes are tall", and that on page 53 we have the statement "Elephants, giraffes and lions are all native to Africa". Now suppose that the article text reads:

There are many large animals in the world, among which are elephants. Elephants come from Africa, as do giraffes. Giraffes are noted for being tall.

We would reference this as follows:

There are many large animals in the world, among which are elephants.<ref name="Johnweal99">Weal, (1999) page 24</ref> Elephants come from Africa, as do giraffes.<ref>Weal, (1999) page 53</ref> Giraffes are noted for being tall.<ref name="Johnweal99" />

which produces:

There are many large animals in the world, among which are elephants.^[1] Elephants come from Africa, as do giraffes.^[2] Giraffes are noted for being tall.^[1]

^ ^a ^b Weal, (1999) page 24
^ Weal, (1999) page 53

Note that the page 53 ref is unnamed, because it's only needed once. However, if the page 53 ref is needed twice, it does need to be named; but the name must be unique. One possibility is to include the page number in the name, as follows:

There are many large animals in the world, among which are elephants.<ref name="Johnweal99">Weal, (1999) page 24</ref> Elephants come from Africa, as do giraffes.<ref name="Johnweal99p53">Weal, (1999) page 53</ref> Giraffes are noted for being tall.<ref name="Johnweal99" /> Another African animal is the lion.<ref name="Johnweal99p53" />

which produces:

There are many large animals in the world, among which are elephants.^[1] Elephants come from Africa, as do giraffes.^[2] Giraffes are noted for being tall.^[1] Another African animal is the lion.^[2]

^ ^a ^b Weal, (1999) page 24
^ ^a ^b Weal, (1999) page 53

If you would like me to fix the article directly, please post the article name here. --Redrose64 (talk) 13:42, 20 March 2010 (UTC)

Presuming you are referring to Jagdgeschwader 11: multiple references have the names "Johnweal06" and "Dannyparker98". I used the error check feature of RefToolbar. I will let you fix these, as the page numbers differ. ---— Gadget850 (Ed) ^talk 15:32, 20 March 2010 (UTC)

Yes. I see that the article is actively being edited by others (including Perseus71), and haven't jumped in to fix these problems. I tend to use names like weal2006 for a general ref to a 2006 book by Weal, like weal2006p5 or weal2006pp27-28 for refs to specific pages in that book. Wtmitchell (talk) (earlier Boracay Bill) 04:40, 21 March 2010 (UTC)

Thank you all for the clarification as well as the offer of help. It is immensely appreciated. Yes, I was referring to JG 11. But that's too an example to the point. This article failed first round of GA for this reason. Yes the pages do differ. I did manage to go through every single cite to change pages. You are welcome to take a look and provide your feedback. Thanks once again. Perseus71 (talk) 05:01, 21 March 2010 (UTC)

Afterthought -

If this is the reason for naming the Cites, then why do you have a "REF=" parameter in the Cite Book ? Isn't that supposed to tie somehow to a cite from that book ?

{{cite book}}

Please let me know. Perseus71 (talk) 13:28, 21 March 2010 (UTC)

Entirely different purpose. The |ref= parameter of {{cite book}} (and most of the other cite templates) is indeed an anchor for an internal link, but the <ref name=></ref> cannot provide that link - its purpose is to avoid duplication of identical refs, and permit their re-use. Let's take our earlier example, but this time enhance the shortened footnotes so that they will link:

There are many large animals in the world, among which are elephants.<ref name="Johnweal99">[[#refWeal1999|Weal, (1999)]] page 24</ref> Elephants come from Africa, as do giraffes.<ref>[[#refWeal1999|Weal, (1999)]] page 53</ref> Giraffes are noted for being tall.<ref name="Johnweal99" />

{{reflist}}

*{{cite book |last=Weal |first=John |year=1999 |title=Example book of animals |publisher=Daily Planet Publishers |location=Metropolis |ref=refWeal1999 }}

which produces:

There are many large animals in the world, among which are elephants.^[1] Elephants come from Africa, as do giraffes.^[2] Giraffes are noted for being tall.^[1]

^ ^a ^b Weal, (1999) page 24
^ Weal, (1999) page 53

Weal, John (1999). Example book of animals. Metropolis: Daily Planet Publishers.

If you click either of the little ^[1]s, these take you to "^{^} ^a ^b Weal, (1999) page 24"; similarly, ^[2] takes you to "^{^} Weal, (1999) page 53". If you click either of those, it takes you to the full citation. The movement is performed in at least Firefox, Chome and IE7, but Firefox and Chrome will also highlight the link target in pale blue; IE7 won't. --Redrose64 (talk) 14:43, 21 March 2010 (UTC)

Ah Ha! That is the most important lesson for me! Immense appreciation. ' Perseus 71 ^talk 17:35, 21 March 2010 (UTC)

See Wikipedia:Citing sources/Example edits for different methods for more info. There have been some recent changes in this area and there are a number articles still using now-deprecated techniques which will no longer highlight the accessed citation. See the last few sections of the talk page of that project page for more information. Wtmitchell (talk) (earlier Boracay Bill) 22:33, 22 March 2010 (UTC)

I should mention {{sfn}} (shortened footnote template), which was designed to make this process dumb proof. Multiple uses of the same {{sfn}} call are automatically combined into a single footnote, with one footnote for each page number. The link in the footnote leads you to the citation at the bottom of the page. The only thing you see in edit mode is the author name, date, and page number. It saves you from having to keep track of "computerese" id names, like #refWeal1999p35. ---- CharlesGillingham (talk) 12:38, 26 March 2010 (UTC)

Image source

The lead section now says "A source is also required when uploading an image." I think this is too broad and incorrect as it stands. If I go to, say, the Tower of London, and take a picture with my own camera, and upload that for use in Tower of London, I don't need to cite a source beyond my own statement (possibly confirmed by email to OTRS) that it is a picture of the Tower of London, nor that I took it. If i upload a picture which was published elsewhere, i must state where I got it, and enough information to determine its copyright status, either public domain, an acceptable license, or available via fair use, as the case may be. But I don't generally need to cite a reliable source that identifies the content of the picture, unless that is challenged in good faith as beien inaccurate.

I therefore propose to change

"A source is also required when uploading an image."

to

"When an image is uploaded, the uploader must state where the image came from, and its copyright status. If the image comes from an online source, a link to the source is normally provided."

Any objections? DES ^(talk) 02:17, 27 March 2010 (UTC)

Yes, I think this is an improvement because it will be less likely to confuse readers. A legitimate "source" for an image is "own work". We should probably limit our use of source to the definitions used in WP:V and WP:RS. WhatamIdoing (talk) 01:44, 28 March 2010 (UTC)

I support that change, David. SlimVirgin ^{TALK contribs} 04:25, 28 March 2010 (UTC)

Multiple cites to the same source?

Is it possible/advisable to cite the same source multiple times in an article without having to duplicate the citation? If so, how? Thanks. --Eamonnca1 (talk) 20:16, 29 March 2010 (UTC)

Use <ref name=xxx> (replacing xxx with a suitable "name" for each new citatation and avoiding duplicating those names). First time, include the whole citation between that and the closing </ref>. All the other times use <ref name=xxx />. The reflist will include a^, b^, c^ etc to show the mutliple entries (Addendum - it is there, in Wikipedia:Citing_sources#Footnote_system, but it could be much clearer, IMO.) --Jubilee♫clipman 20:39, 29 March 2010 (UTC)

Thanks. --Eamonnca1 (talk) 20:53, 29 March 2010 (UTC)

No problem! --Jubilee♫clipman 20:55, 29 March 2010 (UTC)

See WP:REFNAME. The problem is that we try to give the same information in different places. ---— Gadget850 (Ed) ^talk 20:48, 29 March 2010 (UTC)

You edit conficted me just as I was about to link to WP:REFNAME myself! I agree that these guides do try to give the same info multiple time amd often with subtly different (even contraditory) advice. Could we not just use {{main|WP:REFNAME}}, {{see|WP:REFNAME}} or some such at that point in the text and be done with it? --Jubilee♫clipman 20:55, 29 March 2010 (UTC)

List-defined references

I propose that the "LazyDog" examples of List-defined references be changed to make them more consistent with the other existing footnote examples. In particular, removing the group="Ref" parameter this way makes things clearer as group="Ref" is not part of LDR functionality.

It should go something like this...

The example below shows what list-defined references look like in the edit box:

The Sun is pretty big,<ref name=Miller2005p23/> but the Moon is not so big.<ref name=Brown2006/> The Sun is also quite hot.<ref name=Miller2005p34/> ==Notes== {{reflist|refs= <ref name=Miller2005p23>Miller, E: ''The Sun'', page 23. Academic Press, 2005.</ref> <ref name=Miller2005p34>Miller, E: ''The Sun'', page 34. Academic Press, 2005.</ref> <ref name=Brown2006>Brown, R: "Size of the Moon", ''Scientific American'', 51(78):46</ref> }}

Below is how this would look in the article, once you had previewed or saved your edited section:

The Sun is pretty big,^[1] but the Moon is not so big.^[2] The Sun is also quite hot.^[3]

Notes

^ Miller, E: The Sun, page 23. Academic Press, 2005.

^ Brown, R: "Size of the Moon", Scientific American, 51(78):46.

^ Miller, E: The Sun, page 34. Academic Press, 2005.

It should also be pointed out that not all references within the article body text need to be list-defined; ordinary footnote references (fully enclosed with <ref> and </ref> tags) will display as normal along with any list-defined ones. - I do not believe that this is widely understood.

The part which says, "As with other citation templates, these should not be added to articles that already have a stable referencing system..." also needs to be reconsidered. It's based on similar misunderstanding. List-defined references don't involve additional templates and are not in conflict with regular footnote references; they can easily coexist.

--SallyScot (talk) 10:55, 28 March 2010 (UTC)

Good points, with which I generally agree. I'd be happier seeing examples and renderings of the referenced cites, but I understand that it's a tradeoff between pedantic clarity and page length

I am of two minds re the last point made. I wouldn't consider it a conflict with existing style, but I'm sure that some would.

As a sidebar re mention of digressive content in the first point, I am responsible for adding digressive information unrelated to specific topic points in the WP:Citing sources/Example edits for different methods sub-page on the basis of consensus presumed from lack of objection (see discussion, or lack of, here). I still think that that is useful on balance, but understand that this can defocus a tightly-focused topic, I'm happy to let other editors decide the merits of that.

Seems OK to me. You will have to use a group, as there are other <ref>...</ref> tags on the page. Just leave it out of the pre list and add a comment to the other markup so editors know why it is there.

I do agree that LDR and traditional references are not a conflict.

We are presenting the same info at Help:Footnotes#List-defined_references, Wikipedia:Citing_sources#List-defined_references and Wikipedia:Footnotes#List-defined_references. Would it make more sense to fork this to its own help page?

---— Gadget850 (Ed) ^talk 13:13, 28 March 2010 (UTC)

I agree with Gadget850, a group parameter is needed in the implementation if the examples, because otherwise their example reference section will be mixed up with other example reference sections on the page. However, the displayed example wiki-code need not use the group parameter, because it is not an essential part of the LDR feature. DES ^(talk) 14:45, 28 March 2010 (UTC)

As to SallyScot's argument that LDRs can co-exist with in-text references, this is technically possible. However i think it leads to confusion. The major advantage of using LDRs, IMO, is that all references can be found grouped together -- having some references use LDRs and some not is IMO highly undesirable. If an article uses LDRs exclusively and an in-text ref is added, this seems to me unwelcome in much the same way that if all existing references use citation templates and a non-templated ref is added -- in either case I would edit to change the newly added ref to the existing style. Contrawise, if a stable article does not use LDRs, introducing a single one would seem to me to add confusion, and changing all the existing refs to LDR format would seem to require consensus, or at least talk-page suggestion and lack of objection. DES ^(talk) 14:45, 28 March 2010 (UTC)

Yes— technically possible and cause no issues, but not desirable from the standpoint of consistency. An editnotice would probably be appropriate for articles using LDR. ---— Gadget850 (Ed) ^talk 15:25, 28 March 2010 (UTC)

I refer you to User:DESiegel/Ref fmt which I have substed onto the talk pages of several articles where I have added significant numbers of references (generally previously unreferenced or largely unreferenced articles) in LDR format. I assume, perhaps unwisely, that people editing and adding references will look at the talk page, and if they don't, it is still there to point to. A recent example is Talk:GameCrush. DES ^(talk) 16:23, 28 March 2010 (UTC)

I had forgotten that I had already done this: Template:Editnotices/Page/Arthur Rudolph. ---— Gadget850 (Ed) ^talk 16:53, 28 March 2010 (UTC)

I think the "no edit warring" notice is entirely about behavior, not functionality: Editors shouldn't insist that a given article be converted to list-defined references over the objections of other editors. Edit warring to introduce (or remove) LDR is exactly as silly as edit warring over whether there should be a blank line after section headings, or whether citation templates should be typed in a vertical/list format or in a single line.

In particular, because this is new, I expect some editors to be confused by the unexpected appearance of LDR. We don't want endless reversions from people whose real problem is that they can't figure out how LDR works. WhatamIdoing (talk) 19:29, 28 March 2010 (UTC)

(after ec) Obviously edit warring is never good, and particularly not over this sort of thing. But the notices are more than "No edit warring" IMO, they are about consistency. If all references in a given article are in LDR form and a new ref is introduced in non-LDR form, i would convert automatically to LDR format. Contrawise, if a single LDR-format ref is introduced into an article that already has several non-LDR refs, i would convert that to match the existing refs. And in either case I would probably start a talk-page discussion seeking consensus over reference format. DES ^(talk) 19:44, 28 March 2010 (UTC)

IMO, internal consistency is unimportant. LDR and regular refs produce exactly the same thing from the perspective of the reader -- just like including, or skipping, a blank line after a ==section heading== is invisible to the reader. As a result, having "mixed styles" doesn't bother me -- because the style isn't mixed where it counts. I therefore don't think that we want to issue, or even imply, an all-or-nothing command.

But I think that we do want to have a reminder that edit warring over the internal formatting is stupid. You know -- if WP:Wikipedia is not that important, then the invisible-to-the-reader, purely internal formatting of refs is really, really, really, really not important enough to insist that it be "my" way or "your" way. WhatamIdoing (talk) 00:12, 29 March 2010 (UTC)-

I disagree. Much the same argument could be applied to citation templates: manual formatting could produce the exact same result. But we still recommend that an article be all one way or the other. If LDR has a value, it is, IMO, largely in collecting all citation details into a single section. mixing styles sacrifices this. Therefore I would and will routinely change new references to the existing style, whether LDR or in-body, if there is enough of an existing style to be worth adhering to, and start talk page discussion on the matter. Others, I suspect, will do the same. DES ^(talk) 00:51, 29 March 2010 (UTC)

Do we? I haven't ever found a recommendation that an article either use citation templates for either 0% or 100% of refs. Can you provide a direct quotation from a relevant guideline that supports your belief? WhatamIdoing (talk) 01:22, 29 March 2010 (UTC)

From WP:CT: "Because templates can be contentious, editors should not add citation templates, or change an article with a consistent citation format to another, without gaining consensus". From WP:FOOT: "Converting citation styles should not be done without first gaining consensus for the change on the article's talk page." And in any case it has been my experience that in practice editors are strongly encouraged to be uniform in citation style for a given article. DES ^(talk) 03:05, 29 March 2010 (UTC)

WP:CT doesn't say that if you use a citation template here, you must use it in every single citation, and if you don't use the templates for all of them, then you must never use a template for any of them. It's not an all-or-nothing rule. The rules only say that you may not forcibly impose your own preference on an article.

Certainly editors are encouraged to follow a uniform style (it is required for Featured Articles) -- but even at FAC, Wikipedia measures that uniformity according to what the reader sees, not what the editor typed. WhatamIdoing (talk) 04:26, 29 March 2010 (UTC)

--

A group parameter is not needed in the example rendering. The example rendering is achieved here using a construct of some raw html in order to avoid clashing with other references and notes, in exactly the same way as in the Footnotes and Shortened footnotes examples already do to avoid clashing with each other.

List-defined references and in-text references in the same article aren't really much less consistent than re-use of references from another section (e.g. <ref name="multiple"/>), which nobody seems to have an issue with.

I think you have to trade-off any perceived, and I would argue fairly minor, inconsistencies against the real advantages of list-defined references, which also need to be taken in to consideration. Full in-text footnote references can clutter the the article text, making it more difficult to read and edit. Judicious use of list-defined references, which may be focused where this clutter is worst, will help remove this clutter. This could otherwise become too onerous a task if other editors are generally encouraged to believe and insist that internal consistency is everything and it has to be all or nothing when converting to list-defined references.

--SallyScot (talk) 19:42, 28 March 2010 (UTC)

I disagree here, I think it is better to seek per-article (or possibly per-section) consensus on reference format, and then to edit for consistency one way or the other. I agree about clutter, although LDR currently has its own issues -- it generally requires double edits if section editing is being used, and shows a cite error briefly between the two edits. DES ^(talk) 19:48, 28 March 2010 (UTC)

TIP: If your browser supports tabs, edit the reference section in one tab and the article in another. ---— Gadget850 (Ed) ^talk 00:12, 29 March 2010 (UTC)

Granted, but there still must be two edits, and a brief period with a cite error. There is also an increased chance of edit conflicts, and if you aren't very careful with timing, you can get an edit conflict with yourself. Also, when doing section editing, the ref toolbar doesn't recognize existing named refs. Not huge, and I use LDRs routinely, but there are downsides to the current implementation. DES ^(talk) 00:51, 29 March 2010 (UTC)

As there have been no objections simply to the changing of the examples of List-defined references to make them more consistent with the other existing footnote examples I've implemented that part for the time-being. I'll come back to the wording about consensus and stable referencing systems and consistent style in due course.

--SallyScot (talk) 22:37, 28 March 2010 (UTC)

All the embedded markup makes it harder for future editors to edit, but it isn't worth arguing. BTW: <tt>...</tt> will not be supported by HTML 5, which we are slowly moving towards. ---— Gadget850 (Ed) ^talk 00:01, 30 March 2010 (UTC)

Kudos to Sally for fixing the format. It is pain, I know, having edited a few of those myself. (Shame on whoever reverted her edit without first trying to figure out what she had done.) I guess we need to fix these examples without <tt> shortly. But please, let's keep the nice color highlighting in these examples. I think it helps the reader and readers come first. ---- CharlesGillingham (talk) 18:11, 30 March 2010 (UTC)

Just a comment re DES' comment on the inevitability of a brief citation error when using LDR. It's only inevitable if one adds the references in two saved edits. There is however, a way to avoid this, by selecting "edit this page" (rather than editing a section), adding the inline citation and the reference in the same single edit, and then using preview to check that all is well. In my own practice, if I am adding a reference in any style at all (LDR or not), I never edit via only the section. The reason for this is that I always want to preview how the reference reads and correct any errors, and editing only the section doesn't allow me to do this, no matter what style is used. I adopted this practice before I encountered LDR, and have simply continued it with LDR, without any problems.

Indeed I am often intrigued when I see work by more experienced editors (i.e. those who have been at it longer than me, and/or are more prolific than me) doing multiple saves simply because they edit only per section and can't see citation errors until they've saved. I acknowledge BTW that DES' comment specifies "double edits if section editing is being used", so I am not implying unawareness of the alternative approach which I use, only noting that it is my way of dealing with references regardless of the referencing style. Indeed, the only time I edit per section is when I am adding text that does not require a reference, either in an article, or per discussion pages etc. Also BTW, I am impressed with the constructive efforts I see going on here. Regards Wotnow (talk) 01:16, 5 April 2010 (UTC)

In the wrong section

I also think the List-defined references subsection is currently sitting in the wrong place, under the wrong section. It doesn't belong under section 6 - Citation templates and tools, but rather under subsection 4.2 - Inline citations (under How to present citations). - And, as list-defined references display the same as the Footnote system, it should probably be a subsection of that.

--SallyScot (talk) 20:55, 30 March 2010 (UTC)

You are probably right -- I suspect it was added lower down because it was a new facility. DES ^(talk) 22:59, 30 March 2010 (UTC)

Carl suggested it here. Dugnad (talk) 12:10, 2 April 2010 (UTC)

Citation section titles

As a continuation of my comments here, I'd like to propose that WP:CITE be changed to give the editor a choice of the section title to use for a bibliographical list, most likely between Bibliography, References and Sources. Currently WP:CITE recommends "References", but I believe this is wrong because real reference works don't use such section titles. While most of my nonfiction literature at home is in the Hebrew language, of 8 English-language books that cite sources, 5 use "Bibliography", 1 uses "Bibliography/Sources" and 2 have no such section. I have found that shorter works usually either have a Bibliography section or no such section. So, what is the justification for "References" and can be broaden the choice? —Ynhockey ^(Talk) 09:43, 2 April 2010 (UTC)

The references section is not normally a bibliography. When using one of the shortened footnotes or parenthetical references styles there will be a "sources" section which is to some extent a Bibliography, or more accurately a "Works cited" section. Properly a bibliography is a list of relevant works, which may not all be cited, and in any case a bibliography does not normally list a work more than once, which our "References" sections often do. "References" is probably the most common section name for this purpose. My personal preference is "Notes" because that is what best describes what this section holds (at least in a ref/reflist citation style), except when the article uses List-defined references. Then the reference detail is all in the section, so I call it "References". DES ^(talk) 09:56, 2 April 2010 (UTC)

I am only referring to bibliographical lists, which do not have repeating references. The section you're talking about is called "Notes". —Ynhockey ^(Talk) 11:15, 2 April 2010 (UTC)

In practice I think you will find that the section which holds a reflist template in an article that does not use any of the shortened footnote systems is more often called "References" than anything else, and i think that is the section whose name the page is discussing. DES ^(talk) 17:08, 2 April 2010 (UTC)

Again, we are talking about different things here. This page (WP:CITE) also talks about a bibliographical list that can be used in Harvard-style citations in a section below the short footnote section (the one that has {{reflist}}). It does not make sense to call the section "References". I have yet to see one real-life reference work where such a format is used. —Ynhockey ^(Talk) 00:41, 3 April 2010 (UTC)

It's best not to use the word bibliography in this discussion because it has two meanings. If you mean sources, what is wrong with calling that section a References section? It's not the only option, of course, but why wouldn't it be okay? SlimVirgin ^{talk contribs} 01:40, 3 April 2010 (UTC)

Just to clarify, lists of sources can be called Notes, References, Works cited, and Sources. Often editors will call the shortened list "Notes," and the more detailed refs, "References," but that varies.

Bibliography, strictly speaking, refers to any relevant reading, not necessarily sources, though I've seen editors use it as the header for the sources section. SlimVirgin ^{talk contribs} 01:43, 3 April 2010 (UTC)

There is nothing seriously wrong with calling it references, but it's not right either. I believe it does not accurately describe what the section is actually for, and indeed, has several meanings. I believe the most accurate word would be "Sources" anyway, but this is not used in any Wikipedia article. In any case, I am not asking WP:CITE to recommend "Bibliography", just to present it as one of several options (which, IMO, "Sources" should also be one of). —Ynhockey ^(Talk) 01:45, 3 April 2010 (UTC)

Bibliography means something else. What would be wrong with references? It's used quite commonly outside WP. I've seen it used to mean bibliography but it usually means sources. And yes, sources is used as a header. Not often, but I've seen it. SlimVirgin ^{talk contribs} 01:56, 3 April 2010 (UTC)

Let's get our fact straight.

the section "How to present citations" says: "Sections containing citations are usually called "Notes" or "References." Many editors prefer to reserve the section heading "Bibliography" for complete lists of published works in authors' biographies."
- Seems to be using "References" for a section containing actual citations.
The section "Inline citations" says: "In most cases, an inline citation is required, either in addition to, or instead of, a full citation in the References section, depending on which citation method is being used..."
- Unclear to me. May be using "references" for a "list of works cited"
The section "Shortened footnotes" says: "As before, the list of footnotes is automatically generated in a "Notes" or "Footnotes" section. A full citation is then added in a "References" section."
- Unclear to me. May be using "references" for a "list of works cited"
The "Parenthetical referencing" section says: "The full citation is then added at the end of the article to a "References" or "Works cited" section."
- Gives an option on name.
The section "Embedded links" says "A full citation is also required in a References section at the end of the article."
- Seems to be using "References" for a section containing actual detailed citations.
The section "Say where you found it" says:"...you could write: "Smith 2005, p. 100, cited in Jones 2010," between ref tags, with full citations in the References section."
- Seems to be using "References" for a section containing actual detailed citations, but is note quite clear.

Now, Ynhockey, to which of these suggestions/guideliens on section names are you objecting, precisely? what rewording would you suggest? DES ^(talk) 03:59, 3 April 2010 (UTC)

Ynhockey, I suggest that you read this. After reading it, come back and let us know if you still believe that "real reference works don't use such section titles". WhatamIdoing (talk) 05:19, 3 April 2010 (UTC)

Consistent style

Since SV removed basically everything I did while I was typing this message, you'll need to click here to see my proposal. WhatamIdoing (talk) 22:25, 31 March 2010 (UTC)

I have moved the repetitive statements about a consistent style into a single section.

I have also updated the page to explain what we mean by 'consistent style', which is (importantly) consistency in what the reader sees, not what the editor sees. I am credibly assured that an article with consistent citations, but inconsistent internal formatting, would pass FAC (although nobody can imagine why a person would deliberately choose to do that).

My goals for what editors will learn from this section are something like this:

Edit warring is evil.
Readers matter more than editors.
Mixing internal formatting (e.g., template-based citations and non-template citations) is technically permitted, as long as the reader can't tell the difference.
This means that if you're working on (e.g.,) a 'templates' article, and somebody adds a hand-formatted ref (done perfectly, so the reader couldn't tell which citation used a template and which was hand-formatted), then you are permitted to convert it to a citation template, but nobody's going to force you convert it. (Substitute "LDR", "regular footnote", "hand-formatted", etc. for 'templates'.)
If you're working on (e.g.,) an author-date/parenthetical article, and somebody adds a ref-style footnote (a difference that is highly visible to the reader), then you are required to convert it to whichever is the dominant or consensus-approved style.

If you disagree with these goals, please help me refine them. If you think that you can improve the text to better achieve these goals, please WP:BOLDly do so. Thanks, WhatamIdoing (talk) 22:17, 31 March 2010 (UTC)

We already had a section that covered this, so I've moved its wording into your new subsection, and moved your subsection to the top of the citation style section. You introduced some new ideas, and you added that internal consistency doesn't matter. That would need further consensus. Personally I'd be opposed to saying articles don't have to be internally consistent, from the perspective of the reader or editor. SlimVirgin ^{talk contribs} 22:20, 31 March 2010 (UTC)

And I oppose anything that says or implies that readers matter more than editors. It's a false dichtomy, in part because our readers often are our editors and vice versa, and in part because badly written articles aren't good for readers, so what helps editors often helps readers too. SlimVirgin ^{talk contribs} 22:23, 31 March 2010 (UTC)

So given an article that is mostly list-defined references, and an editor adds a non-list-defined reference, you would absolutely require, without exception, that the new citation to be converted to a list-defined reference?

Put another way: is |refs= a "citation style" in your mind? WhatamIdoing (talk) 22:29, 31 March 2010 (UTC)

This is my change to the section, by the way. This is What's version. SlimVirgin ^{talk contribs} 22:31, 31 March 2010 (UTC)

This is a guideline, W, so nothing is absolutely required, but I think if you were to go to FAC they'd wonder why you were using different citation styles. One problem with mixing templates and manual, for instance, is that the templates are controlled centrally, and some of them use citation styles that don't exist outside WP. So if that were ever to be changed—and I hope that does happen one day—one set of citations would change in an article, but not another set. This wouldn't show up in the watchlists for the article, so it might go unnoticed. SlimVirgin ^{talk contribs} 22:34, 31 March 2010 (UTC)

I would consider a change to or from list-defined refs to be a change in citation style. Similarly, a change that converts an article to using citation templates, or removes the templates in favor of plain text, is a change in citation style. So not every change in citation style is visible to the reader. Of course anyone can add a new reference, and eventually it will be converted to match the rest of the article. But converting all the refs fro one style to another, even if there is no visible change in the rendered output, is one of the things this guideline discourages. — Carl (CBM · talk) 22:39, 31 March 2010 (UTC)

I think what W means is that it's not a problem if the article mixes up styles, and probably most of the time it isn't so long as the reader can't see it, but I don't think the guideline should say it's not a problem. Internal consistency is preferable, in part for the reason I outlined above. SlimVirgin ^{talk contribs} 22:47, 31 March 2010 (UTC)

Yes, I agree. — Carl (CBM · talk) 22:54, 31 March 2010 (UTC)

I don't agree that readers matter more than editors: both are on equal footing. However, I agree that the contents of the edit page are less important than those of the actual article page (uncivil hidden comments etc, aside). If two styles of text formatting produce the exact same result, what's the problem: — vs — anyone? I do think that one style of citing should be used though: either paretheticals or footnotes but not both. Whether the sources themselves are placed in-line within the <ref name=name></ref> or under a {{Reflist}} |refs= section is not important unless it causes confusion for editors. {{Cite}} is also not mandatory, so we can't force people to use that either, even if it is used predominantly, IMO. Basically, the more important thing to consider is the actual display; the internal formatting techniques used are entirely secondary. Confusing Mr/Ms Potential Newbie Editor—or even Prof Experienced Sysop—isn't helpful, though, so consistancy even in formatting techniques is probably no bad thing --Jubilee ♫clipman 23:41, 31 March 2010 (UTC)

As I said above, J, the templates can be changed centrally. If we mix and match templates and manual, it would mean some of the citation styles would change, but not all. Then the difference would be visible without our realizing it. It's therefore not a good thing for a guideline to recommend. I also can't see the point of recommending it—editors do fix up articles over time to make them consistent, and they'll keep on doing that no matter what this guideline says. SlimVirgin ^{talk contribs} 00:12, 1 April 2010 (UTC)

Perhaps this is a simpler way of expressing the important concept: Chicago is a citation style. MLA is a citation style. Vancouver is a citation style.

HTML is not a citation style. {{Reflist}} is not a citation style. MS Word is not a citation style. Wikimedia parser functions are not a citation style. {{Citation}} is not a citation style. These are tools for formatting citations in a given style, but they are not, themselves, actual styles.

Put another way, these produce the same results and therefore use the same style:

{{cite book |title=The Sun is Really Big |author=Smith, John |year=2009}}
Smith, John (2009). ''The Sun is Really Big''

These do not use the same citation style, even though they use exactly the same formatting tools, because they produce different results:

Smith, John (2009). ''The Sun is Really Big''
John Smith. ''The Sun is Really Big.'' 2009.

Nor do these, even though they use the same template:

{{cite book |title=The Sun is Really Big |author=Smith, John |year=2009}}
{{cite book |title=The Sun is Really Big |author=John Smith |year=2009}}

It is my opinion that a "citation style" and "the tools used to create that citation style" are different things. Wikipedia might (or might not) choose to require consistency in both, but can we agree that it would be a "both" situation -- that is, that the italics button in the edit box is not part of a citation style? WhatamIdoing (talk) 23:48, 31 March 2010 (UTC)

Yes, please let's separate out style from tool. Talking about both at the same time is confusing. Consistency in citation style is required; consistency in citation tool usage or lack thereof doesn't appear to be required. That's what I was attempting to say, above, in a round about manner --Jubilee ♫clipman 23:59, 31 March 2010 (UTC)

You're recommending mixing up formats, or at least saying it's okay, which boils down to the same thing for our purposes. SlimVirgin ^{talk contribs} 00:14, 1 April 2010 (UTC)

The motivation behind the language in the guideline is both:

Don't change the reader-visible part of the style (e.g. from Harvard refs to footnotes or back). By the way, {{reflist}} is reader-visible, because it changes the font size.
Also, don't change the reader-invisible method that is used to format the citations. For example, don't add or remove citation templates in bulk.

The point of this language is analogous to WP:ENGVAR. Because there is no general agreement on which citation style might be best, or how to format the references in wiki text, we simply stick with what is already established in each article. So for the purposes of this guideline, "citation style" has always encompassed both of the bullet points above. — Carl (CBM · talk) 00:15, 1 April 2010 (UTC)

Editors don't like it if you arrive at an article they've worked hard on and add a ref in a different format. I often have that problem when editing articles with templates, because I don't know how to use them, so I copy the style but then have to add an apology in the edit summary (sorry, don't know how to add this using a template). Sure enough, if it's a page someone cares about, a template appears nanoseconds later. SlimVirgin ^{talk contribs} 00:46, 1 April 2010 (UTC)

I think you've just answered your own question: use the method you are most comfortable with; if others feel compelled to change your addition to follow the established style, great, if not, ownm... move on. That's my attidute, anyway. BTW, I was neither recommending nor even allowing anything, I was merely observing, though perhaps I could have been clearer on that --Jubilee ♫clipman 01:09, 1 April 2010 (UTC)

Addendum: I have installed the Cite Tool and use that almost every time I place a fn ref, BTW, whatever the prevelent system happens to be. I don't use it if I am merely adding refs to an under-developed start class or a stub, though, as I add those simply to help other editors develop the article --Jubilee ♫clipman 01:17, 1 April 2010 (UTC)

It's true that you don't have to follow the established style when adding a new reference – someone else can fix it later. It's changing existing references to another style that's more problematical, and that's what this guideline is supposed to be addressing. If you're just adding a single new reference, sure, you could add it in some reasonable way and someone else can handle the reformatting later. The same goes for most other other stylistic things. For example, if I type the word "color" into the article on England, someone else will change it to "colour" later. But if I change every existing instance of "colour" to "color" in that article, that's a problem. The same holds for reference formatting.

However, we don't want to write a guideline that tells people to ignore these things altogether. It's better to take a second to see what style to use before adding a new reference or using a particular spelling. We can recognize that an occasional lapse is not the end of the world without encouraging more frequent lapses. — Carl (CBM · talk) 01:39, 1 April 2010 (UTC)

What editors "have to" do matters here: Featured articles are apparently required to comply with this page (under criteria "2c").

In conflating citation style with the software mechanism used to format the text, we're not being clear. Editors have, in good faith, reached entirely opposite conclusions as a result of our previous text (see, e.g., DES' comments on 29 March). I think that we should fix this confusion.

We can fix this confusion and require both "visible" and "invisible" consistency, if that's what we think the community wants.

I'm not convinced that the community actually requires or wants this, but I'm open to arguments on this point. I'm not, however, really open to arguments that we should continue pretending that a text formatting technique is a citation style. The fact that this guideline sloppily conflated the two for years is IMO not an argument in favor of retaining the confusion any longer.

Does anyone actually object to treating the two things as separate issues?

Separating them would probably entail adding definitions to WP:CITE#Use of terms along these lines:

"A citation style is a consistent method of presenting bibliographic descriptions of referenced sources. Citation styles usually prescribe the contents of the bibliographic record, the order in which elements are presented, and a method of connecting the citation to relevant part of the article.

"A citation formatting technique is what the editor types to make the software display the bibliographic information in the citation style that the editors have chosen. HTML and citation templates are examples of formatting techniques."

Does anyone object to this? WhatamIdoing (talk) 02:56, 1 April 2010 (UTC)

I don't know what "HTML" means here. Do you mean wikitext? — Carl (CBM · talk) 11:20, 1 April 2010 (UTC)

I don't mind if we say how we're using the terms "style" and "format," so long as we don't say anything that implies inconsistent styles or formats or systems, or whatever we call them, are okay. The relationship between this page and FAs is not only that FAs follow this page, but that this page tries to follow FAs too, i.e. best practice, and I've not seen a recent FA that uses different formats. And in large measure that's because different formats can produce different styles. The distinction isn't as clear in practice on WP. SlimVirgin ^{talk contribs} 04:52, 1 April 2010 (UTC)

I certainly agree that the two factors should be separated out in the MoS, even that is achieved only by defining the terms unequivocally and using the exact same terms for the exact same meaning throughout. Basically, we take the lead of legal documents which define their terms at the start and proceed to use those terms exactly as defined in the actual text. Not that we need to take that precise approach, just that we need to follow the example of unequivocal consistancy --Jubilee ♫clipman 07:30, 1 April 2010 (UTC)

Style vs format

I don't think a long expansion is necessary. We could simply add a sentence to the "consistent style" section saying something like "Consistency is desirable both in the visible appearance of references and in the method used to format references in an article's source code." — Carl (CBM · talk) 11:33, 1 April 2010 (UTC)

I disagree with WhatamIdoing and agree with everyone else. If you say that the source wikitext doesn't matter as long as the appearance is the same, then you are opening the door for fans of a particular citation "format" (like templates, or list defined references, or {{rp}}) to start converting articles, perhaps even using bots. This is disruptive at best and destructive at worst.

I like Carl's version, in the previous post. It just carefully separates the issues, without redefining words in common usage (i.e. "style" and "format"). ---- CharlesGillingham (talk) 20:03, 1 April 2010 (UTC)

I agree that the separation between "style" and "formatting" that WhatamIdoing proposes has some validity, although not as much as he seems to think. In particular, two citations, one using a template and one using manual formatting may produce the same display today, but the template could be changed centrally tomorrow and they two would no longer be consistent. That is one reason to use templates -- to allow central decisions to adjust the displayed formatting style. In any case, i think this guideline should strongly advise, although not mandate, consistency in both displayed style and internal formatting. We should be explict about that. Whether we define separate terms is IMO not highly important as we should give the same advice about both. DES ^(talk) 20:45, 1 April 2010 (UTC)

I must take issue here. The citation templates do not always follow the prescribed style. There have been occasions I've had to be creative with the templates to get the desired output that conforms to the prescribed style. Between the prescribed style and the template output there should be guidance on which takes precedence. Template output can pretty much be considered the de facto Wikipedia style—but it is not recognized. If it is logical that the template is adjusted to conform to prescribed style, then it logically follows prescribed style should take precedence over the template output. Format in the usually invisible editing sphere should not be a big deal. Whether it is template or manual entry should not be an issue most of the time. Templates are an aid and should be considered as nothing more than that. Requiring their use is instruction creep, including for "consistency". Lambanog (talk) 11:50, 2 April 2010 (UTC)

If by "the prescribed style" you mean a style following some standard format such as MLA, or Chicago, or APA, then yes, the citation templates must be "used creatively" to get precisely the format specified by the outside style guide. Note that no such style guide is "prescribed" for wikipedia articles, and so there is no "prescribed style" to "take precedence over the template output".

The "guidance" is that each article may use whatever style the editors determine by consensus, and by default whatever style the first major contributor has used. Such as style may be implemented with or without the use of templates.

But it has been the case that changes, thought of as improvements, are made in the output of some or all of the citation templates. When such changes are made, articles that use the templates will change their display styles, and if an articel uses a mix of templates and manually formatted citations to achieve a consistent result, that consistency may be broken at any time. Therefore it is better either to consistently use templates or to consistently use manual formatting, so that future template changes will not make the citations withing a single articel inconsistent. DES ^(talk) 17:33, 2 April 2010 (UTC)

Your arguments about the possible mess caused by updating the templates are sound, but they're limited to the single issue of citation templates.

<i> and '' are both equally effective means of turning on italic text formatting. If we absolutely require consistency in formatting techniques, you will be required to pick one or the other, even though there is no possibility of any template change ever affecting the output (because no templates are involved).

There is far more to formatting than the "I hate citation templates" debate. WhatamIdoing (talk) 03:02, 3 April 2010 (UTC)

Just a reminder about this issue: Is there any reason to demand non-template internal consistency? Should FACs really care whether italic font is invoked with HTML tags or with wiki code in each and every citation? WhatamIdoing (talk) 03:13, 9 April 2010 (UTC)

I just added 2 short paragraphs to the 'Consistent style" section which i think reflect the consensus in this thread. this pair of edits are my change. DES ^(talk) 21:39, 1 April 2010 (UTC)

Parenthetical referencing links

In the section "Parenthetical referencing" there were two hatnote links. One of them also appeared in the first sentence of the section. Another editor has removed that one from the hatnote. This is at least technically correct, but in this case I think the double hatnote may have had some value. Does anyone agree? DES ^(talk) 10:40, 2 April 2010 (UTC)

Sorry, I just saw this, and yes, I agree. :) SlimVirgin ^{talk contribs} 08:33, 6 April 2010 (UTC)

LOC format not acceptable for cites/footnotes?

A drive-by editor changed the format of all the cites in California Gold Rush, claiming that "LOC format not acceptable style," and citing this page. Perhaps I'm missing the reference, but I'm not seeing support for this change on this page. Is the Library of Congress format (i.e., only capitalizing the first word and proper names) acceptable in Wikipedia? NorCalHistory (talk) 14:35, 11 April 2010 (UTC)

Any consistent style is acceptable, but it would be interesting to know where this style is documented. Jc3s5h (talk) 15:48, 11 April 2010 (UTC)

I never saw the LOC title format before. If it is going to be used in articles, then it should be documented so that other editors understand it. Here and Wikipedia:Manual of Style (capital letters). An editnotice for the article would help alert editors, since this appears to be a style that is not common. ---— Gadget850 (Ed) ^talk 16:06, 11 April 2010 (UTC)

"LOC" isn't even a citation style; it's a set of rules for creating library catalog entries. Nowhere is it listed here or here or here. The most widely used styles– CMOS, MLA, and APA– all require that major words (basically all words except prepositions) in a title be capitalized. (And, by the way, use of the term "drive-by editor" is both a personal attack and reflects a lack of good faith.) 75.2.209.226 (talk) 19:34, 11 April 2010 (UTC)

It think LOC format is the Anglo-American Cataloguing Rules (AACR2). As such, this is not a citaiton style, but a catalog style. ---— Gadget850 (Ed) ^talk 19:48, 11 April 2010 (UTC)

Absolutely correct. (Spoken as a former librarian.) 75.2.209.226 (talk) 20:08, 11 April 2010 (UTC)

Correct, but irrelevant. This guideline actually allows editors to invent their own citation styles if they choose. Sentence case is not only not prohibited, but it is also widely used in medicine- and science-related articles (Query: Why don't we ever get any science librarians in these discussions? Why is it always a language or history specialist?) -- as well as practically every citation that is automatically generated from the ISBN, because the ISBN database is (IMO) badly formatted. WhatamIdoing (talk) 20:35, 11 April 2010 (UTC)

And what does science or medicine have to do with the California Gold Rush? Talk about irrelevant... Sure, editors can create their own citation styles. That's just what WP needs- more chaos. 75.2.209.226 (talk) 21:14, 11 April 2010 (UTC)

A science librarian would be familiar with CSE (ISBN 097796650X), which is the required style guide at nearly every major university for all hard sciences, and which specifically requires sentence case. The hard science folks rarely seem to mistake "the conventions used in my field" for "the conventions used in every single academic field".

This guideline does not permit you to impose the True™ style on any article over the objections of other editors. If you want to use a different style, then you need to go win friends and influence people over at the article, instead of insulting them here. WhatamIdoing (talk) 21:44, 11 April 2010 (UTC)

Wow! I have no idea what generated that unintelligible, off-topic tirade, that completely mischaracterized my posts, nor why you have such a bug about discussing scientific citation style, but please try to stay on topic. The facts are these: 1. The OP asked about using "LOC style," not about using CSE style; 2. There is no such thing as "LOC" citation style; 3. The article in question is about the California Gold Rush, which is a historic topic, not a scientific one; 4. Historians generally use CMOS; 5. No one made any claims about "the conventions used in every single academic field." In fact, I cited several different commonly used styles; 6. I did not insult you; 7. I am not a historian; 8. I am not a librarian. 75.2.209.226 (talk) 22:30, 11 April 2010 (UTC)

Any consistent citation style is acceptable, and styles should not be changed based only on personal preference. In other words: any time you are tempted to go through and change all the refs in an article from one consistent style to another, simply because you prefer a different style, you should reconsider. — Carl (CBM · talk) 23:52, 11 April 2010 (UTC)

Hear, hear—unless you're removing citation templates, in which case you're on the side of God. :p SlimVirgin ^{talk contribs} 00:01, 12 April 2010 (UTC)

Wikipedia:CITE#Consistent_style#Consistent style says, "You should follow the style already established in an article, if it has one; where there is disagreement, the style used by the first editor to use one should be respected." It looks to me as if this edit initially established the citation style for the article. It also looks to me as if any disagreement on this should be discussed on the article's talk page. Wtmitchell (talk) (earlier Boracay Bill) 00:19, 12 April 2010 (UTC)

What is there now is absolutely consistent with this edit, which established the citation style in the article. So no changes are necessary, right? 75.2.209.226 (talk) 01:09, 12 April 2010 (UTC)

It looks to me like the initial bibliography was added in September 2006 and, as early as October 2006, the references had already changed [16] as part of a larger set of edits responding to a peer review [Wikipedia:Peer review/California Gold Rush/archive1] that asked for footnotes. Given that history, and the fact that the style then sat basically unchanged from October 2006 until this year, I would use the October 2006 style (the one that sat for 3.5 years) as the "default style" for this article. Certainly no good reason has been given to change it in 2010. — Carl (CBM · talk) 01:33, 12 April 2010 (UTC)

No good reason was given for changing it in 2006. It was purely NorCalHistory's personal preference for "LOC style." He began to add references using his personal style, instead of the one already established, and then gradually changed what was already there. See: 1, 2, 3, 4, 5, etc. This is really starting to sound like an ownership issue. NorCalHistory took over the article, changing previously established style to suit himself, and now, four years later, still won't let go. And apparently you agree that if an editor can successfully change a consistent, recognizable style to his own personal preferences, and fend off all other editors for several years, that the editor then owns the article. 75.2.209.226 (talk) 02:02, 12 April 2010 (UTC)

NorCalHistory also added inline citations, which were not there before. The editor who added the references originally (rjensen) edited the bibliography again in October 2006 after the style were changed, using the changed style [17], so at least implicitly that person agreed with the change at the time. And this was all during one month back in 2006, after which time the style was stable for 3.5 years. — Carl (CBM · talk) 02:28, 12 April 2010 (UTC)

Is it acceptable to cite a web page that requires login?

While editing Abhinav Kumar Mogha, I tried to access the athlete biography link and wasn't able to because it requires login. Is there documentation about whether this is acceptable for a reference in a Wikipedia article? --Auntof6 (talk) 02:35, 14 April 2010 (UTC)

See Wikipedia:Reliable sources/Cost. If the problem were having to pay in order to be granted a userid and password, the link would be acceptable. On the other hand, if the link is only accessible to a select few, as opposed to anyone who was willing to pay for a subscription, then it is not a publication and is not acceptable as a citation.Jc3s5h (talk) 03:19, 14 April 2010 (UTC)

The Times paywall

See Wikipedia:Village_pump_(miscellaneous)#The_Times_paywall. Ty 08:01, 23 April 2010 (UTC)

The ref name tag doesn't work with certain letters

I tried to create a reference starting with "Ö", but it wasn't accepted, so I had to change it into "O" instead. Why isn't the code working with umlauts? --MoRsE (talk) 07:49, 24 April 2010 (UTC)

Works for me. What do you mean by "not accepted"? ---— Gadget850 (Ed) ^talk 10:18, 24 April 2010 (UTC)

<ref name=Öscar>Öscar test</ref><ref name=Öscar/>
<ref name="Öscar">"Öscar" test</ref><ref name="Öscar"/>

^[1]^[1] ^[1]^[1]

References

^ ^a ^b ^c ^d Öscar test Cite error: The named reference "Öscar" was defined multiple times with different content (see the help page).

Hmm weird...well, it's ok if it was just a temporary hickup. --MoRsE (talk) 13:55, 24 April 2010 (UTC)

Wait. It doesn't work when you call it the second time unless you enclose it in quotes. It works when it is defined because the Cite parser doesn't actually use the refname if a named ref is only used once; when it is invoked, the parser must choke when the name contains a non-ASCII character. When you enclose it in quotes, either the Cite.php or the MediaWiki parser (probably the latter) encodes the "Ö" as ".C3.96". This is documented at WP:REFNAME, but you have to read the footnote.---— Gadget850 (Ed) ^talk 14:31, 24 April 2010 (UTC)

To obviate such problems, perhaps we just need to make it clear that the refname must always be enclosed in quote marks? --Jubilee ♫clipman 14:41, 24 April 2010 (UTC)

Or use standard ASCII or {{r}}. I keep it simple: last name or journal+date all in lower case. ---— Gadget850 (Ed) ^talk 14:55, 24 April 2010 (UTC)

But Öscar and Oscar are not the same word. Never seen {{r}} before: that might work also. Thanks for that --Jubilee ♫clipman 15:27, 24 April 2010 (UTC)

Ah— the sample above should have generated the message Cite error: Invalid <ref> tag; refs with no content must have a name., but error messages are disabled on talk pages. The Cite error part links to the appropriate help section. ---— Gadget850 (Ed) ^talk 15:00, 24 April 2010 (UTC)

I'm sure I have seen error messages on talkpages before, in particular those relating to <ref></ref>. Perhaps they have been deprecated very recently? --Jubilee ♫clipman 15:27, 24 April 2010 (UTC)

The messages are controlled by {{broken ref}}— messages show only on main, user, template, category, help and file pages, but not talk. Updates to Cite a year or so ago caused a lot of confusion and discussion when error messages started appearing on talk pages, so I disabled them after much discussion. Been a while since I had my head in this, so I clean forgot that this should have thrown an error. ---— Gadget850 (Ed) ^talk 15:55, 24 April 2010 (UTC)

Ah, that explains it. Thanks (both for explaining and for disabling the messages) --Jubilee ♫clipman 16:25, 24 April 2010 (UTC)

Going back to the original issue, do we need to explain how to use non-ascii characters in citations (quotes, {{r}} etc), perhaps discouraging their use at the same time? --Jubilee ♫clipman 16:29, 24 April 2010 (UTC)

{{R}} is not an issue, since it uses #tag:ref which results in quotes being rendered. It is documented at WP:REFNAME— I think I added it, but it looks like it got moved to a footnote and the recommendation to not use non-ASCII was removed. ---— Gadget850 (Ed) ^talk 16:47, 24 April 2010 (UTC)

Great work guys. This is something I'd never would have figured out myself. :) --MoRsE (talk) 18:48, 24 April 2010 (UTC)

Two column reflist?

Recently I have noticed several articles that used the {{reflist|2}} to set the references as two columns ... yet the articles only had one reference. This can cause the single reference to be displayed with a "column gap" in the middle, which looks odd.

Is there some guideline somewhere telling people to always use {{reflist|2}}? Because if there are only a few references, {{reflist}} would be perfectly acceptable instead. --Metropolitan90 (talk) 17:18, 25 April 2010 (UTC)

It's entirely up to the editors at the page. If we really wanted that feature used on every article, then we'd make the template do it automatically. Perhaps at the time that it was set, there were more refs in the article? Also, the multi-column view doesn't work on all web browsers, so the editors might not have noticed how strange it looks. WhatamIdoing (talk) 20:40, 25 April 2010 (UTC)

MoS naming style

There is currently an ongoing discussion about the future of this and others MoS naming style. Please consider the issues raised in the discussion and vote if you wish GnevinAWB (talk) 20:51, 25 April 2010 (UTC)

RfC on merging Words to avoid into Words to watch

Wikipedia talk:Words to watch#RFC. There's a proposal to merge several pages as part of a project to streamline the MoS. One part of the proposal is to merge Words to avoid, Avoid peacock terms, Avoid weasel words, and Avoid neologisms into a new page, Words to watch (W2W). Fresh input would be appreciated at the RfC. SlimVirgin ^{talk contribs} 00:43, 26 April 2010 (UTC)

Nutshell

As written, the nutshell says that this guideline on how to cite sources applies only to inline citations. In fact, the guideline addresses how to cite sources anywhere in a Wikipedia article, including References section (however named; in the real world, Bibliography is the usual term) that does not consist of footnotes. I believed that this should be clarified.—Finell 01:23, 27 April 2010 (UTC)

I don't think a nutshell can, or should be expected to, mention every aspect of the guideline. So, although the guideline mentions general references as well as inline citations, it is not essential to mention general references in the guideline. Also, "inline citations" covers not only footnotes, but parenthetical referencing as well.

The version of the nutshell that Finell attempted to put in place was plainly not in accord with consensus because it stated "The policy on sourcing is Wikipedia:Verifiability, which requires that all content in Wikipedia be supported by citation to reliable sources". That is not the requirement; unchallenged material that is not a direct quote does not need any citation. It is sufficient if a source exists, even if that source is not cited. Jc3s5h (talk) 01:36, 27 April 2010 (UTC)

I am not arguing here for the language that I proposed in my edit. While a nutshell obviously does not restate the entire guideline, it should identify the range of the topic that the guideline covers. To suggest that citation style is only for inline citations is inaccurate and misleading.—Finell 03:05, 27 April 2010 (UTC)

I would prefer the nutshell to read simply "This guideline discusses how to format and present citations." or better still "This guideline discusses how to write citations and where to put them in articles." The other text isn't necessary. ---- CharlesGillingham (talk) 05:46, 27 April 2010 (UTC)

Citing encyclopedia articles

I've seen an increasing number of citations that cite specialized scholarly encyclopedias, without citing the specific article in the encyclopedia from which the information is taken. This leaves the reader with no way to check the reference. I suggest that we add a section on Encyclopedias (making it clear that it refers only to specialized encyclopedias like the Encyclopedia of American Foreign Policy or the Dictionary of National Biography, and continuing to discourage use of general ones like the Brittanica, etc.). The section would recommend inclusion of Article Title, Author (when known), and perhaps volume number and pages, as well as the normal Editor, Book title, publisher and date information. --SteveMcCluskey (talk) 16:01, 1 May 2010 (UTC)

Article linked to book (or periodical) is the standard reference notation. FWiW Bzuk (talk) 17:10, 1 May 2010 (UTC)

Citation proposal - MediaWiki extension

mw:Extension:HarvardReferences - MediaWiki extension that supports "Harvard" references in simple notation is discussed on Wikipedia:Village pump (proposals). X-romix (talk) 12:36, 14 May 2010 (UTC)

stacking sources

I cannot seem to find any info on whether this is allowed or not. There is currently serious questions called into some GANs and FACs at times because of this issue and without clear guidance can make or break and promotion. For the latter, it coming back for a renom can have an editor demand the oppoiste of the last time and again be a make or break issue. We have discussed it (and other issues) at WT:Anime#A-Class, about the particular article School Rumble. The basic rationale behind stacking is to make the articles look cleaner for non-contriversial statements. The argument against it that it would violate the idea that citations need to be clear. Because a lot of the statements in the above article require multiple sources (stuff like citing language translations, media releases, etc.) there would often be a lot of sources after many paragraphs. There have been some other examples (finding them is hard due to term usage, but one example is Wikipedia:Good article reassessment/Preity Zinta/1. It would be extremely helpful if this is cleared up. 陣内 _Jinnai 02:56, 24 May 2010 (UTC)

I think it's fine and pretty consistent with how other publications do it. The only reason not to, really, is that it makes it harder to use the same ref multiple times with named refs. Whether stacking or not, anytime multiple refs are used to support one statement it's important to minimize redundancy and omit the less reliable alternatives. Christopher Parham (talk) 05:44, 24 May 2010 (UTC)

Cite stacking may be appropriate when there are multiple assertions in a single sentence. "Smith was born in Chicago in 1913 to wealthy parents." If those three assertions were found in separate sources then it would make sense to have each item cited separately, in order. For some reason, FAC folks think numbered cites should always be arranged in numerical order, which makes it harder for any future editor (or reader) to verify the material. Will Beback talk 06:08, 24 May 2010 (UTC)

Not sure what's meant by stacking sources, but if Jinnai means combining sources between one set of ref tags, I think it's a good idea and I do it all the time. Otherwise we end up with sentences ending with multiple refs.[1][2][3][4][5]

Far better to end with one.[1] And if we need to make clear who said what, the footnote can say: "For his date of birth, see Smith 2010. For where he met his wife, see Jones 2009, etc". SlimVirgin ^{talk contribs} 09:48, 24 May 2010 (UTC)

Looks like "cite stacking" is placing multiple citations into one reference tag. I see no problem with multiple inline cites, nor with inline cites being out of numerical order. More than a few inline cites may indicate overciting, and the sources should be checked. If the consensus is to stack cites in this manner, then an enhancement to Cite should be requested to automatically do this. It does mean that you can't use named references. ---— Gadget850 (Ed) ^talk 12:01, 24 May 2010 (UTC)

You can use them, but not as often or differently. Sometimes it's appropriate to use the same stacked references more than once, for example. SlimVirgin ^{talk contribs} 13:48, 24 May 2010 (UTC)

I like "cite stacking" (multiple citations into one reference tag), but strongly dislike footnotes out of numerical order. To evaluate the quality of an article I often start with the notes and work backwards, which is &&^%* in long articles with jumbled numbers out of sequence.Rjensen (talk) 12:24, 24 May 2010 (UTC)

I'm not familiar with the issue of footnotes in or out of order. How can the numerical order be controlled? SlimVirgin ^{talk contribs} 13:48, 24 May 2010 (UTC)

If you use named references, then all uses have the same inline cite number, but only the first use will be in numerical order. The only solutions are to not use named references, use shortened references or parenthetical references. Not sure why it is an issue with some folks. ---— Gadget850 (Ed) ^talk 14:20, 24 May 2010 (UTC)

here's an example:^[1]text stuff^[2] more text ^[3] silly stuff; ^[1] great material^[1]; dubious ^[2]; nonsense ^[2]; don;t miss this ^[3]; what is this?^[3]; now add 100 more cites and find me ^[2]; or get lost in the maze ^[2]; elsewhere ^[4]; ^[1]; somewhere^[4]; nowhere^[1]

^ ^a ^b ^c ^d ^e Smith, Something Old (2011)

^ ^a ^b ^c ^d ^e Jones, Something New (1888)

^ ^a ^b ^c Brown, Useful trivia for Wikinuts (2009)

^ ^a ^b Black, Something Lost (date lost)

Rjensen (talk) 14:27, 24 May 2010 (UTC)

I've never seen anyone object to that. If someone at FAC is objecting to it, they shouldn't, because it's unavoidable if we use ref name=. SlimVirgin ^{talk contribs} 14:30, 24 May 2010 (UTC)

User:Will Beback got my initial concern was with what I brought up. There is inconsistant application of whether to allow that type of stacking when the information isn't contriversial and the sources used are all needed (such as in the examples given like noting various media releases of works) which often require their own indivisual citation. It would be good to have some clarity on this as it can become a make-or-break issue depending on who is at the reigns.陣内 _Jinnai 17:53, 24 May 2010 (UTC)

Citation stacking is always allowed, and indeed is often a good idea. Citing sources separately is also allowed. What looks odd at FA is a sentence with lots of refs after it. If that's the only thing that's wrong with an article, I can't imagine anyone opposing because of that alone, so in that sense it would never be make or break. But it can be a sign of careless writing: of sentences and sourcing out of control. Or of a failure to look at the article from the readers' perspective. SlimVirgin ^{talk contribs} 18:05, 24 May 2010 (UTC)

If citation stacking is allowed, then it should be documented as a style. ---— Gadget850 (Ed) ^talk 14:47, 25 May 2010 (UTC)

That's basically what I think should happen as well so there is no confusion in the future.陣内 _Jinnai 18:20, 26 May 2010 (UTC)

Softcopy vs hardcopy

I have had some mild problems with online softcopies of periodicals (newspapers, weekly magazines) not quite matching hard copies. I like to give my fellow editors "a chance" by allowing them to check my work, not being forced to accept what could be a misquote or even a bogus reference. Softcopies may come in multiple editions as it turns out. Apparently they do not share copy editors with their hard copy counterparts.

Most often, the local paper has "box" score type material which isn't picked up in soft copy. I place a comment next to the footnote but include the softcopy for credibility. There was an article on the topic, even if the figures aren't online!

In a recent case, the online number was changed, and a superlative dropped. Another editor caught me to my annoyance since I "knew" what I had read and copied (numbers not the words! :). But he was right. The softcopy was as he had read it, not a whole lot different but different. I have to go with the least common denominator here, which was the soft copy with a 1% lower number and without the superlative. I suspect a "stronger" copy editor. No big deal unless someone checks the hardcopy someday after the softcopy link has died!

I tried to discuss this on another page, and an editor suggested supplying both. 1) Okay, but they don't agree. 2) It "looks funny" to other editors, like I supplied TWO refs (which are really the same), only it is just one. What am I trying to pull? Trying to make it seem "irrefutable?" :) 3) It creates more work for me, which is important since I am lazy!

I intend to continue to supply softcopy with hardcopy when available. I will have to check relevant material and dumb one or the other down to match, I suppose. Annoying. Any thoughts or experience from others? Student7 (talk) 10:30, 27 May 2010 (UTC)

If you check both and see discrepancies, pointing them out is ideal. If you're lazy (if I ever change my name, I think I'll use that as my new middle name), I would only cite the one you actually relied on. Jc3s5h (talk) 14:25, 27 May 2010 (UTC)

Citations following quotations

From WP:Cite#When quoting someone:

You should always add a citation when quoting published material, including the page number if there is one. The citation should be placed either directly after the quotation

"This is a quotation."^[1]

or after a sentence or phrase that introduces the quotation.

Bertrand Russell explained the fallacy as follows:^[2]

Fine, as far as it goes. But it doesn't explain how that should be applied to quotations in the middle of blocks of text. For example, if something read:

Roger Ebert described the film as a "powerful debut" and said Schneider's performance was "a real class act". He went on to praise the rest of the cast, especially those playing Martians.^[3]

Some editors demand that an inline citation follow the quoted part directly (... as "a real class act".^[3] He went on to ...) with a repeat of the same inline at the end. Others, me included it has to be said, are content with using just one at the end of the text to refer to everything that comes before; it can get a little messy otherwise. The guideline is a little fuzzy; hopefully there's an already-formed consensus opinion on this. Cheers, Steve ^T • C 13:26, 29 April 2010 (UTC)

I agree, the standard in publishing is to cite the quotation as near as possible to the statement, given that there could be an excepted quotation. FWiW Bzuk (talk) 13:30, 29 April 2010 (UTC).

I think you're saying you'd go for the "... as 'a real class act'.^[3] He went on to ...^[3]" format, rather than the one I prefer that hasn't got that ... redundancy (IMO); is that right? Sorry, it's the "I agree" that's thrown me. Steve ^T • C 13:56, 29 April 2010 (UTC)

My two cents here: I think a citations should directly follow each sentence that contains a quote. In quotes of living people, I think this is especially needed. I think if you have a run of quoted sentences together, blockquote could be used; however interspersed quotes among non-quotes are best cited with individual citations. —Charles Edward ^{(Talk | Contribs)} 15:20, 29 April 2010 (UTC)

I think the citations in this example follow the consensus above. Would any of you like to check them just to be sure? Thanks - Pointillist (talk) 20:58, 29 April 2010 (UTC)

That would fit my understanding. —Charles Edward ^{(Talk | Contribs)} 21:13, 29 April 2010 (UTC)

I didn't quite follow this discussion... what, exactly, was the consensus that was reached? I'm having the same problem. Like Steve (talk · contribs) above said, I prefer to use a single citation at the end of the material cited, even if that material is two or three paragraphs in length, in order to avoid this: "The band member plays a Stratovarius^[14] and does cocaine all day long^[14] and was last seen alive before he was found dead.^[14]" However, I also will have quotes from the subject... should I then have a citation after the quote, even if that citation is the same as the one at the end of the paragraph(s)? I feel like the above answers were just opinions. What's the actual rule or consensus on this matter? – Kerαunoςcopia^◁_galaxies 03:38, 28 May 2010 (UTC)

I can appreciate what the previous editor is saying about using one cite for three paragraphs. The problem is for later editors who weren't looking when the change was made, or possible "tweaked" by successive authors, or worse (and typically) another paragraph entered in between the others, with or without it's own citation. I would prefer to see an in-line citation on every paragraph, at least. For dense information, I don't mind a citation on each sentence. When there are many facts (as in place articles), sentence cites are often preferable.

But if the guidelines don't say "for each paragraph", it should IMO. Student7 (talk) 16:05, 30 May 2010 (UTC)

But, forgetting about different facts for a moment, does there need to be a citation after every quotation? I'm also finding it difficult to follow this, as people seem to be answering different questions. Must there be a citation every time a quote ends? To repeat Steve's example above (slightly adapting it two include two quotes), which of the following is preferred?

Roger Ebert described the film as a "powerful debut" and said Schneider's performance was "a real class act". He went on to praise the rest of the cast, and said "those playing the Martians were brilliant!"^[3]

or

Roger Ebert described the film as a "powerful debut" and said Schneider's performance was "a real class act".^[3] He went on to praise the rest of the cast, and said "those playing the Martians were brilliant!"^[3]

Sorry if this is repetitive, but I'm not seeing a definitive answer above.--Beloved Freak 11:03, 31 May 2010 (UTC)

I've requested input from WT:MOS and WT:FACR as I'm not sure how much traffic this page gets.--Beloved Freak 11:10, 31 May 2010 (UTC)

I've come here after a link was posted at WT:MOS. Now ref-tag density is not a cut-and-dried matter. Editors need to balance the clarity of which piece of text comes from where, with the ungainly clutter that can result when editors "play safe" and cite on a micro-level. Nothing irritates me more as a reader than seeing [33] at the end of successive sentences. Generally, the reader will assume that a ref tag applies retrospectively if there's no reason to doubt this. Where statements are contentious, of course, a higher density of ref tags may be appropriate. Tony (talk) 11:24, 31 May 2010 (UTC)

Ok, thanks Tony, that makes sense. I take that to mean that you wouldn't necessarily need a citation after every closing quotation mark or after every sentence containing a quotation? That, as long as there is at least one citation for the quoted material, there's no hard-and-fast rule? --Beloved Freak 11:36, 31 May 2010 (UTC)

Yes, but the golden rule is based on the question "Is it clear to the reader?" I've sometimes asked, errr .... how far back does that ref tag cover? That's when the language could be tweaked to make it clearer (there are a number of techniques), or, yes, another ref tag added, earlier in the para. My advice is to try to minimise the density while still being clear, especially when not contentious.

Here's a para I'm looking at just now: "Singapore is the world's fourth leading financial centre[8] and its economy is often ranked amongst the world's top ten most open,[9] competitive[10] and innovative.[11][12]" You'd love to have just one ref tag at the end, after "innovative", but heck, these last three epithets are grand claims, and quite different concepts, so I guess we have to put up with the clutter. This was in the lead, BTW, where some editors feel it's fine to have few or no refs, as long as verification is provided in the body of the article. So ... this bit requires discussion on the article talk page, I guess.

Same article, whole subsection: "The first records of settlement in Singapore are from the 2nd century AD.[31] The island was an outpost of the Sumatran Srivijaya empire and originally had the Javanese name Temasek ('sea town'). Between the 16th and early 19th centuries, Singapore was part of the Sultanate of Johor." Um ... If ref 31 includes the last two sentences, it should go at the end. As it is, I'm not happy to be left dangling without verification of these claims.

Next subsection (British colonial rule (1910–1940): The whole para has just one ref tag, right at the end. This is ideal (no clutter!), and we can presume it is all endorsed in ref 32.

And IMO, the quotation is fine with a single ref tag at the end: Roger Ebert described the film as a "powerful debut" and said Schneider's performance was "a real class act". He went on to praise the rest of the cast, especially those playing Martians.^[3] provided ref 3 contains all of this. That is what the readers will presume, and it's lovely and neat to encompass retrospectively like that.

Para breaks are raised as an issue above; a para break does, indeed, present more reason to re-affirm the source, I have to say. But is it too inflexible to specify a rule based on this observation? I'm unsure. Tony (talk) 11:58, 31 May 2010 (UTC)

Thanks for clarifying. It's a complex situation, but I for one feel that I have a better understanding of it.--Beloved Freak 16:31, 31 May 2010 (UTC)

Makes sense to me. Essentially, two questions seem to have been answered. In Tony's example, cluttered citations are required to keep track of the specifically-cited descriptors. But more importantly to the subject of the discussion, a single citation covering several quotes retroactively can be used once. I think the per-paragraph situation is the editor's call, but it should be used if it helps future editors keep track of what's citing what. – Kerαunoςcopia^◁_galaxies 04:03, 1 June 2010 (UTC)

[Johnweal99-1] Weal, (1999) p. 51–53

[2] Weal (1999), page 53.

[Johnweal99-3] Weal, (1999) page 24

[4] Weal, (1999) page 53

[Johnweal99-5] Weal, (1999) page 24

[Johnweal99p53-6] Weal, (1999) page 53

[Johnweal99-7] Weal, (1999) page 24

[8] Weal, (1999) page 53

[noteLDR01] Miller, E: The Sun, page 23. Academic Press, 2005.

[noteLDR02] Brown, R: "Size of the Moon", Scientific American, 51(78):46.

[noteLDR03] Miller, E: The Sun, page 34. Academic Press, 2005.

[Öscar-9] Öscar test Cite error: The named reference "Öscar" was defined multiple times with different content (see the help page).

[1]

[2]

[1]

[2]

[1]

[2]

[1]

[2]

[1]

[2]

[3]

[1]

[1]

[2]

[3]

[4]

Wikipedia talk:Citing sources/Archive 29

Tools

Making thesis titles italicized

Citing someone else citing something

must vs should

Chicago Manual of Style rule

Secondary Citation of Books

One para for accessible source, one para for inaccessible source

What's the problem here?

Minor formatting error

My revert

Embedded links

Header: Notes or References

invisible to the reader?

Template:Harvard reference

Endnotes X2 (Citation Manager)

Wikipedia:Embedded citations

Duplicating references on an article

correction/addition to 55 days in Peking citation

RFC on introduction of {{r}} for citations

List-defined references

General references

Clicking on the ref tag when editing

References on talk pages

I need some guidance

Ideal citation method?

Citation of a book

Why is the citing of references so complicated?

Styles and templates

First Hand Source

Say where you got it

BOT that tidies up references

Citing images on Wikimedia Commons

Sources that aren't really sources...

Removing a citation (link to a source)

Behind a paywall or free registration required?

Cite Style Error ?

Image source

Multiple cites to the same source?

List-defined references

In the wrong section

Citation section titles

Consistent style

Style vs format

Parenthetical referencing links

LOC format not acceptable for cites/footnotes?

Is it acceptable to cite a web page that requires login?

The Times paywall

The ref name tag doesn't work with certain letters

Two column reflist?

MoS naming style

RfC on merging Words to avoid into Words to watch

Nutshell

Citing encyclopedia articles

Citation proposal - MediaWiki extension

stacking sources

Softcopy vs hardcopy

Citations following quotations