Wikipedia talk:Citing sources

From Wikipedia, the free encyclopedia
  (Redirected from Wikipedia talk:CITE)
Jump to: navigation, search
WikiProject Manual of Style
WikiProject icon This page falls within the scope of WikiProject Manual of Style, a drive to identify and address contradictions and redundancies, improve language, and coordinate the pages that form the MoS guidelines.
the Wikipedia Help Project  
WikiProject icon This page is within the scope of the Wikipedia Help Project, a collaborative effort to improve Wikipedia's help documentation for readers and contributors. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks. To browse help related resources see the help menu or help directory. Or ask for help on your talk page and a volunteer will visit you there.
 ???  This page does not require a rating on the project's quality scale.
 ???  This page has not yet received a rating on the project's importance scale.

Multiple uses for citations re: topic sentence: "Citations are used to ..."[edit]

The lead sentence is incomplete (and/or ambiguous) and inaccurate as stated. It is not completely the case that: "Citations are used to identify the reliable sources on which an article is based."

  1. Citations have multiple uses, not just one.
  2. Citations are identified with specific factual claims, not entire articles.


"Citations are used to identify, locate, assess, and access the reliable source(s) on which a factual claim in an article is based."

I could cite some reliable sources on the multiple uses of citations, if that is not deemed obvious.

That a specific citation is not applicable to an article as a whole seems self evident, but I stand to be corrected.

Opinions? Paulscrawl (talk) 03:44, 7 July 2014 (UTC)

Even if it were strictly speaking true, I don't think that it would be helpful. The purpose on the English Wikipedia is to tell the reader (who may also be an editor) where that piece of information comes from, i.e., "identify". Source assessment is something that editors do, not citations. WhatamIdoing (talk) 05:34, 7 July 2014 (UTC)
The problem is the lead sentence does not make it clear who is using the citation. The editor who inserts a citation does indeed use the citation to identify the reliable sources on which an article is based. The reader may use the citation to do the things described by Paulscrawl. The reader can do whatever he/she wants to with the citations. If the reader has printed the article, the reader can even use the citations to line a birdcage. Perhaps there is a way to clarify this point.
Citations can be inline, in which case they are likely to support a specific claim in the article, or they can be general, in which case the citation supports, to some degree, the entire article. Perhaps the most common general citation occurs in articles about books, plays, or movies. The book, play, or movie is cited by implication (if not explicitly), and it supports all factual claims about the content of the book, play, or movie. Jc3s5h (talk) 20:55, 7 July 2014 (UTC)
  Agree with W. Citations identify and help locate a source, but they are not used (nor should be) to assess a source, nor even to note an editor's assessment of the source. If such an assessment is useful it can be added as a comment following the citation, but the citation itself does not do that. ~ J. Johnson (JJ) (talk) 21:05, 7 July 2014 (UTC)
The editor would not put the editor's assessment of the quality of the source within the citation. However, the reader, upon seeing a statement in Wikipedia with a footnote number next to it, might wonder if the source is a good source. The reader then clicks on the number and discovers the source is published by Yale University Press. The reader might conclude it is a good source. So in that sense the reader uses the citation to assess the quality of the source, although you could say the assessment in the reader's mind is a by-product of identifying the source. Jc3s5h (talk) 21:58, 7 July 2014 (UTC)
I would say that. I'd also say that if I click the ISBN and end up at a website that sells the book or find a library that holds it, and I get it, then my locating, buying, and accessing it, too, is a byproduct of identifying the source. WhatamIdoing (talk) 15:01, 14 July 2014 (UTC)

Thanks for the thoughtful feedback. Still digesting some reliable sources on multiple uses for making citations and multiple uses readers make of citations, which I'll apply to Citation. (I apologize for posting this here, rather than on that main article's talk page.) WP:WHYCITE needs work in any case to tighten coherence with WP:VERIFY.

One quick point, to clear up ambiguity about agency of assessment: to be made clear in Citation#Law, Legal_citation Bluebook, and ALWD Citation Manual: in law, citations are used not merely to identify (a prerequisite for all other uses, granted), but also to explicitly access authority, and are, in fact, arranged in order of legal authority (constitution, statute, cases, ranked by authority of court, and within hierarchy level, by date, regulations, books, articles) both to make and enable assessment. Rather like arranging citations by levels of evidence (meta-analysis, randomized controlled trial, controlled trial, cohort study, etc) in some published systematic reviews of evidence based medicine. Different fields use citations in different ways. Identification is not the only reason these varied citation manuals give as their purpose for writer and reader. More later, after doing some more work. Thanks again for your time and input - you helped me clarify my thoughts for revision. Paulscrawl (talk) 06:21, 16 July 2014 (UTC)

Ah, I could wish you had mentioned Bluebook at the start, as perhaps it is the model you have in mind. It is also the prime example of a "citation" system that (as you mention) explicitly ranks different sources by their putative authority. There are editors who claim to be using a Bluebook style here, but I have never checked if they have incorporated the ranking aspects. It would be rather cumbersome to use here: most readers would not understand such a system, and I doubt if there are any four editors who would agree as to how it might be adapted for Wikipedia. ~ J. Johnson (JJ) (talk) 23:07, 16 July 2014 (UTC)

Non-English language citations[edit]

Where I live, in Germany, most of the articles are stubs. I have tried to find English language links, to help expand these stubs, but have only found German ones thus far. Is it okay to use German language citations, if no English language links are available? — Preceding unsigned comment added by Torspedia (talkcontribs) 20:09, 12 July 2014

@Torspedia: Please see WP:NONENG. --Redrose64 (talk) 00:17, 13 July 2014 (UTC)
@Redrose64: Ta for that. I do hope the sources that I have won't cause too many problems, for I really have not found anything in English. I can now start expanding those stubs, especially for the place where I live and surrounding settlements. — Preceding unsigned comment added by Torspedia (talkcontribs)
Good luck, and happy editing! WhatamIdoing (talk) 15:03, 14 July 2014 (UTC)

Reference required after direct quotations spanning multiple sentences?[edit]

(Copied from WT:MOS): I've seen some users say that if direct quotations span multiple sentences, they are to be followed by a reference for every sentence. As in:

Commenting on (something), Joe Shmoe of The Times said, "blah blah blah."[1] Although he thought (something), Shmoe ultimately felt that, "yada yada yada."[1] Shmoe concluded his piece stating, "yap yap yap."[1]

Instead of:

Commenting on (something), Joe Shmoe of The Times said, "blah blah blah." Although he thought (something), Shmoe ultimately felt that, "yada yada yada." Shmoe concluded his piece stating, "yap yap yap."[1]

Is this a requirement? I don't see it at MOS:QUOTE. 23W 03:21, 14 July 2014 (UTC)

I don't believe that this fairly rare situation has been explicitly covered in any guidance. Putting on your "best editorial judgment" hat, what do you think is clearest? WhatamIdoing (talk) 15:05, 14 July 2014 (UTC)
At least one other person at WT:MOS thinks the second example is clear enough, given that it's "competently written". Some people would still assert that it needs references after every sentence per WP:MINREF, even though neither WP:V nor MINREF itself makes note of consecutive sentences. I just wish it were more clear, as it's a fairly common construct for things like critical reception sections. 23W 21:02, 14 July 2014 (UTC)
I'd rather know your opinion, though. Do you think it's clear enough? Do you think that typical readers will find it clear enough? WhatamIdoing (talk) 00:41, 15 July 2014 (UTC)
To the original poster: In the hypothetical example you provide, as a reader I would: a) appreciate the reduction in unnecessarily duplicated citations; b) see the coherence in the several sentences (clearly indicating one source); and thus, c) expect only one citation at the end of the sequence. As an editor, I would: 1) anticipate citation needed flags on individual sentences, especially if one was particularly controversial or quote worthy, or if a currently newsworthy article is likely to be substantially abridged as events unfold; 2) act defensively by making sure my transitional clauses are strong and stylistically unified to make it clear that, however reformatted or edited by other editors long after this edit drops off recent article history, the single citation applies to the whole sequence; and 3) consider posting first as a separate paragraph to make that fact transparent to diligent article history checkers, and only then editing to incorporate into existing paragraph, if desired for continuity. Don't forget: a good edit summary or two goes a long way in preventing or reverting undesired changes or challenges! Just my personal opinion, of course. Hope that helps. Paulscrawl (talk) 04:13, 16 July 2014 (UTC)

Free open access versions of journal articles behind a paywall[edit]

I believe the article should address access to journal articles that exist behind a paywall. The cite journal template currently states "url:URL of an online location where the text of the publication can be found". You can "find" journal articles at sites like Sage and Sciencedirect, but can you read them? Oftentimes, authors will deposit a copy of their article at their institutional repository like DSpace@MIT So for example, the article "Human-automated path planning optimization and decision support" is available behind a paywall at Science Direct, but the MIT authors have also deposited an open access copy that is free to read at Dspace@MIT. When citing a journal article, should wikipedia editors try to locate a free open access copy of an article and use that in the URL field? If the publisher URL should still be preferred, should the free open access URL be included in another field?Lugevas (talk) 14:27, 25 July 2014 (UTC)

Include open access version in |url=; the paywall version is still available through |doi= or other such parameter.
Cummings, M.L.; Marquez, J.J.; Roy, N. (February 2012). "Human-automated path planning optimization and decision support" (pdf). International Journal of Human-Computer Studies 70 (2): 116–128. doi:10.1016/j.ijhcs.2011.10.001. 
Trappist the monk (talk) 14:51, 25 July 2014 (UTC)
If the only online copy is through a paywall, set |subscription=yes - more at WP:PAYWALL. --Redrose64 (talk) 15:42, 25 July 2014 (UTC)
Thanks for the link to the paywall page. I've been editing Wikipedia on occasion for a decade but sometimes I find it so difficult to find Wikipedia policy and guidelines pages.Lugevas (talk) 17:46, 25 July 2014 (UTC)
Example above, with authors' self-archived version in |url=, links article title to pre-publication version, unsuitable for WP:verification when citing specific page number(s) of journal article, as is common in many fields (and articles up for GA status review, as I understand). In the example case, and many others, the archived version simply does not match publication version in some important aspect (esp. pagination, as here; but also often including: absence of copyright photos, figures, illustrations, maps, or data; absence of cited references; etc.).
I prefer explicit reference to a page, or page range, of a specifically-cited version, as pagination is big in history, law, and related humanities and social sciences, including those citing natural, computer, and mathematical science journals. But my point applies even to citations within those disciplines, which generally cite entire papers, not page(s). You just can't assume the versions are identical without running any one of a number of open source diffpdf programs.
I would put canonical, paginated as published, version in |url= followed by |deadurl=no to force title to hyperlink to that published version when including both |archiveurl= and |archivedate= to helpfully add open access version. Details on usage at template:Cite journal, but in a nutshell, all 4 parameters required for desired behavior.
WP:SAYWHEREYOUGOTIT, indicates you may cite any "true and accurate copy" without specification "(except to the extent that this affects page numbering)" [parenthetically amending perhaps only "ebooks", but generally applicable]. Citing the specific version consulted when using page numbers is also supported by WP:Page numbers. Including |subscription=yes and |doi=, together with non-identically paginated archived version url in |archiveurl= (with it's co-requisite |archivedate=) would be very helpful to both casually curious readers, happy for any accessible version, as well as scholarly verifiers with access to canonical version.
Finally, I prefer to link to PDF container pages rather than direct link to PDF (thus I omit |format=PDF), for both "officially-maintained" archives (university, institutional, WebCite,, etc.) and paywall repositories, as these container pages are very likely more stable URL targets and usually contain valuable bibliographic metadata (CoiNS) easily read by humans and parsed by automated tools such as Zotero. MIT's DSpace calls this the "citable URI" - the metadalink on that page rendered the human- and machine-readable archive date. I see no need to duplicate professionally-maintained archive copies, such as dSpace, on WebCite or (On the other hand, for most authors' "amateur-maintained" self-archive pages - academic or personal home page/blog, etc.- no such standard CoiNS metadata generally exists and there is very little guarantee of a long-lived author &/or URL; I first try archiving such via or WebCite, which will work if container page's robots.txt does not disallow.)
{{cite journal |title=Human-automated path planning optimization and decision support |url= |deadurl=no |subscription=yes |archiveurl= |archivedate=2014-05-14 |first=M.L. |last=Cummings |first2=J.J. |last2=Marquez |first3=N. |last3=Roy |journal=International Journal of Human-Computer Studies |volume=70 |issue=2 |date=February 2012 |pages=116–128 |doi=10.1016/j.ijhcs.2011.10.001}}
Cummings, M.L.; Marquez, J.J.; Roy, N. (February 2012). "Human-automated path planning optimization and decision support". International Journal of Human-Computer Studies 70 (2): 116–128. doi:10.1016/j.ijhcs.2011.10.001. Archived from the original on 2014-05-14. (subscription required (help)). 
The "Archived from the original" tag is an unfortunate hard-coded output of the |archiveurl= &/or |archivedate= parameter(s), perhaps dating to days it assumed use only with copies. It needs to be fixed, especially now that we are in the age of wide-spread and very professionally-maintained institutional repositories for academic archived papers, often pre-publication, not identical to original. But for today, this is what the journal template parameters allow - full transparency of precise cited source; full provision of open access version. Trying to locate a free open access copy of an article is a best practice, in any case. How I stretch the existing parameters to fit my other competing requirements is a personal preference, but I think it has merit.
Paulscrawl (talk) 19:47, 26 July 2014 (UTC)
If I see this "mess", I'd rather recommend don't use the template at all or use the template just for the official citation and add a convenience link to a free online copy outside the template.--Kmhkmh (talk) 21:03, 26 July 2014 (UTC)
This "mess"? Paulscrawl (talk) 23:53, 26 July 2014 (UTC)

Conflicting Sources[edit]

Are there guidelines regarding what to do in the event of conflicting sources? For example, if there is a source in an article that states one thing and significantly more reliable sources can be found that contradict that source? -AnonWikiEditor (talk) 04:08, 1 August 2014 (UTC)