Open main menu

Wikipedia β

Wikipedia talk:Citing sources

Frequently asked questions (FAQ)
Why doesn't Wikipedia require everyone to use exactly the same style for formatting citations on every single article, regardless of the subject?
Different academic disciplines use different styles because they have different needs and interests. Variations include differences in the choice of information to include, the order in which the information is presented, the punctuation, and the name of the section headings under which the information is presented. There is no house style on Wikipedia, and the community does not want to have the holy war that will happen if we tell people that they must use the style preferred by scientists in articles about history or the style preferred by artists when writing about science. Editors should choose a style that they believe is appropriate for the individual article in question and should never edit-war over the style of citations.
What styles are commonly used?
There are many published style manuals. For British English the Oxford Style Manual is the authoritative source. For American English the Chicago Manual of Style is commonly used by historians and in the fine arts. Other US style guides include APA style which is used by sociologists and psychologists, and The MLA Style Manual which is used in humanities. The Council of Science Editors and Vancouver styles are popular with scientists. Editors on Wikipedia may use any style they like, including styles they have made up themselves. It is unusual for Wikipedia articles to strictly adhere to a formally published academic style.
Isn't everyone required to use clickable footnotes like this[1] in every single article?
Footnotes (also called "<ref> tags") are popular but not required. The purpose of an inline citation is to provide information about where that material came from. Any system that allows someone to figure out which source supports which material achieves that goal and is therefore acceptable. Other styles, such as parenthetical citations, are simpler for new users to understand, are commonly taught in schools, and may be the style preferred by the relevant academic discipline.
Why doesn't Wikipedia require everyone to use citation templates in every single article?
Citation templates have advantages and disadvantages. They provide machine-readable meta data and can be used by editors who don't know how to properly order and format a citation. However, they are intimidating and confusing to most new users, and, if more than a few dozen are used, they make the pages noticeably slower to load. Editors should use their best judgment to decide which format best suits each specific article.
Isn't there a rule that every single sentence requires an inline citation?
No. Wikipedia:Verifiability requires citations based on the content rather than the grammar. Sometimes, one sentence will require multiple inline citations. In other instances, a whole paragraph will not require any inline citations.
Aren't general references prohibited?
A general reference is a citation listed at the end of an article, without any system for linking it to a particular bit of material. In an article that contains more than a couple of sentences, it is more difficult to maintain text-source integrity without using inline citations, but general references can be useful and are not banned. However, they are not adequate if the material is one of four types of content requiring an inline citation. The article Early life of Joseph Smith, Jr. is an example of a featured article that uses some general references.
Can I cite a sign?
Yes, signs, including gravestones, that are displayed in public are considered publications. If the article is using citation templates, then use {{cite sign}}. You may also cite works of art, videos, music album liner notes, sheet music, interviews, recorded speeches, podcasts, television episodes, maps, public mailing lists, ship registers, and a wide variety of other things that are published and accessible to the public.


RfC: Remove the bullet point that starts "adding citation templates..."Edit

The consensus is against the proposal to remove the bullet point "* adding citation templates to an article that already uses a consistent system without templates, or removing citation templates from an article that uses them consistently;" from Wikipedia:Citing sources.

Cunard (talk) 01:08, 12 August 2018 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Remove the bullet point "* adding citation templates to an article that already uses a consistent system without templates, or removing citation templates from an article that uses them consistently;" from Wikipedia:Citing sources, and rely on the standard consensus policy processes such as WP:BRD in its stead. -- PBS (talk) 12:56, 4 July 2018 (UTC)

  • Support as nominator. In the good old days when Wikipedia was young I never came across citation templates. It seems that the dispute about them started in 2005. For a time it could be argued that it was possibly to have a page without templates, let alone citation templates; and later pages with templates, but not citation templates. However with the deprecation of "ISBN 978-3-16-148410-0" and the introduction {{ISBN|978-3-16-148410-0}} (without prior consensus on the talk page of every article that received a bot alteration see the RfC) and the widespread use of templates such as {{webarchive}} both are put into articles without first gaining consensus on an article's talk page (along with other templates such as {{better source}}), it seems to me that the bullet point in WP:CITEVAR "adding citation templates to an article that already uses a consistent system without templates,..." is irrelevant in many case as all such altered pages already have templates in inline-citations or in the "References" section. I think it is time to remove that whole bullet point and rely on WP:BRD instead. -- PBS (talk) 12:56, 4 July 2018 (UTC)
  • Comment For those who will argue ("'Once more unto the breach...") that the wording should be changed to "adding citation templates to an article that already uses a consistent system without citation templates," using the argument that {{ISBN}} is not a "citation" template--well nor is {{sfn}} or {{efn}} yet I have seen people on this talk page who dislike those template argue that they are--given the widespread use of templates within reference sections and inline-citations, what is it that currently justifies singling out citation template as unique in requiring consensus on the talk page before adding one (and so creating an exception to the usual WP:BRD process)? -- PBS (talk) 12:56, 4 July 2018 (UTC)
    Historically, a few things are treated in this way: ENGVAR is another. The reason is that the choice has no objective "right" answer, so people just argue based on their personal preferences ("I like templates", "I don't", "Well I do". etc.). Since there is no objective source or fact to resolve the disputes, they go on interminably. The point of the rule is to sidestep all that wasted time by cutting off the disputes before they begin. There are not many of these issues that both draw a lot of personal opinion, have no objectively "right" answer, and affect numerous articles. — Carl (CBM · talk) 17:07, 4 July 2018 (UTC)
    ENGVAR is completely different because once an article has content in one dialect or another unless there is a good reason to change it then there is no reason to do so. In the case of citation templates that is not so, because if an article was created nearly 15 years ago then it would not have been created with templated citations, the technology and usage has evolved. To date the change in varieties of English spellings and grammar has been less volatile. Further there is no prohibition on someone adding an American spelling to a British article (it get changed usually within BRD if necessary), what there is a prohibition against is changing the spelling (or syntax) without good reason. In the case of this prohibition, it actually says "adding citation template" which is a different type of prohibition which encompass adding a new citation, not just changing ones already present. -- PBS (talk) 18:15, 4 July 2018 (UTC)
  • Oppose - Weakening Citevar will result in mass changes to reference formats of articles with no benefit to the articles. The rule is in place for a reason.Nigel Ish (talk) 14:52, 4 July 2018 (UTC)
    "The rule is in place for a reason." Are you sure about that as your account is not as old as the "rule", because the rule has morphed over the years and been interpreted by some in a way that is not compatible with the consensus policy. What do you think was the reason for its introduction and why is the usual BRD cycle insufficient to prevent mass changes? -- PBS (talk) 16:29, 4 July 2018 (UTC)
    The problem with mass changes is that they often present a fait accompli: if you log back on to find that 20, or 200, articles on your watchlist have all changes, what would you do to move the BRD process forward? The BRD cycle only works for changes that are made to a small number of articles; it doesn't work for things that get automated or semi-automated, in which case a centralized discussion is the only reasonable option. The current rule is compatible with CONSENSUS, though, like the rest of the MOS. — Carl (CBM · talk) 17:03, 4 July 2018 (UTC)
    I would start a central discussion and if there is no consensus revert them all (as I have done with such instances of mass changes in the in the past). The current rule is not compatible with CONSENSUS because it is used by some to stop the usual BRD cycle. -- PBS (talk) 17:50, 4 July 2018 (UTC)
  • I generally support moving to citation templates, but we need to be really careful a change here isn't interpreted as an invitation for someone to go start making large scale changes across many articles. Such efforts almost always generate more disruption than they are worth. Monty845 14:58, 4 July 2018 (UTC)
    This concern seems particularly valid given that this RFC has been advertised on other pages as "RFC to ease introduction of citation templates to articles not presently using them" - although the change proposed would equally allow users to remove templates from articles that do use them. I suspect that if adding templates became viewed as a form of "cleanup", we would indeed see large-scale projects aiming to convert nearly all articles to use templates. — Carl (CBM · talk) 15:02, 4 July 2018 (UTC)
    diff please to where is it advertised that way. -- PBS (talk) 16:39, 4 July 2018 (UTC)
    @PBS: [1], [2], [3], [4]. — Carl (CBM · talk) 16:59, 4 July 2018 (UTC)
    Thank you, I will change them. -- PBS (talk) 17:42, 4 July 2018 (UTC)
    I have edited PBS's changes because the heading PBS used, 'RfC: Remove the bullet point that starts "adding citation templates..."' makes no sense outside of this talk page. Jc3s5h (talk) 18:35, 4 July 2018 (UTC)
  • Comment Templates are good and bring a slew of benefits (bot maintenance, parameter validation, error checking, metadata, etc.) and should be used when possible. It is now easier than ever to use citation templates, with tools like the WP:REFTOOLBAR and other assisting aids we didn't have ages ago. The opposite (Template -> Plaintext) however is bad and should not be encouraged. Being WP:BOLD on this matter is usually fine but you need to follow WP:1RR if you run into issues. I'm not sure if CITEVAR needs modifying or not, but I support whatever option goes along the lines of what I've described. However, I want to emphasize that mass changes aren't desirable here, this is something that requires a scalpel, not a chainsaw. Headbomb {t · c · p · b} 15:24, 4 July 2018 (UTC)
  • Oppose this has no effect: reread the first para of citevar. Fifelfoo (talk) 16:54, 4 July 2018 (UTC)
    Fifelfoo if it has no effect, then why oppose the removal of superfluous text? -- PBS (talk) 08:43, 6 July 2018 (UTC)
    superfluous Rfcs ought to be opposed as a waste of time that could have been better used. Fifelfoo (talk) 08:48, 6 July 2018 (UTC)
    Fifelfoo are you declaring that if the RfC had started "Keep" instead of "Delete" you would have opposed the RfC? -- PBS (talk) 10:23, 6 July 2018 (UTC)
    yes, but for the worse procedural sin of a reversed onus. Would you like to insinuate any other points? When do I get my free badger. Fifelfoo (talk) 10:38, 6 July 2018 (UTC)
  • Opposed - per the old adage: “If it ain’t broke, don’t fix it”. An article that uses non-templated citations “ain’t broke”. Blueboar (talk) 17:08, 4 July 2018 (UTC)
    • But it is broken Blueboar. I have seen editors remove citations newly added because the citation was added template when other existing citations do not, even when the format makes them identical in appearance to the reader. Secondly with the use of templates like {{ISBN}} this bullet point is redundant. -- PBS (talk) 17:42, 4 July 2018 (UTC)
      Removed completely, or converted to the existing style? Blueboar (talk) 18:09, 4 July 2018 (UTC)
      Removed completely (dif and see article talk page) -- PBS (talk) 18:24, 4 July 2018 (UTC)
      I would draw a distinction between reverting addition of a citation to a page with a consistent non-templated citation style when the editor who added the template is experienced, and is judged well-capable of conforming to this guideline, vs. an edit by an inexperienced editor who can barely add any citation, and probably lacks the skills to conform to this guideline. I take no position on how often the latter appears, it isn't something that made an impression on me. Jc3s5h (talk) 18:51, 4 July 2018 (UTC)
      Ok... I would agree that we should not REMOVE a citation (completely) simply because it doesn’t conform to the previously established style. Instead, the non-conforming citation should be CONFORMED to the previously established style. And I don’t think it matters who does the conforming. Blueboar (talk) 20:32, 4 July 2018 (UTC)
      But Blueboar the current wording is interpreted that way by some editors even when that change does not affect the visual style; and Jc3s5h this issue is about style and appearance, if the changes conform to the same visual style as the rest of the articles what is the problem with adding templated citations? The addition of plain text citations to articles that use templates often happens (and someone usually (eventually) puts such additions into templates), and this is not an issue with this guidance because it says "changes" not "adds" (unlike with the first part of the bullet point). -- PBS (talk) 09:40, 5 July 2018 (UTC)
      I disagree that "this issue is about style and appearance". The bullet up for removal is about switching between templates or lack of templates. If an editor went into an article that used consistently used citation templates and rewrote all the citations so they visually looked like CS1, but there were no templates, that would go against the bullet in question.
      I don't recall ever seeing someone add a citation to an article with citation templates that mimiced CS1 without using templates; such additions usually do not follow any recognizable style. Also, although it's theoretically possible to mimic CS1, it would not emit any metadata, so it would not be the same. Jc3s5h (talk) 13:21, 5 July 2018 (UTC)
      The current wording does not prohibit such an addition (that half of the bullet point is about changing not adding citations). Further I think that your comment shows that you are looking at this as an editor not a reader. A reader does not notice whether or not there is meta data, but they may notice inconsistencies in visual citation style. Whether or not a template is or is not used has nothing to do with the style used other than the limited range of styles that can be used on a page. -- PBS (talk) 16:03, 5 July 2018 (UTC)
      Granted, the bullet only applies to changing. But I don't recall an editor ever changing one or a few citations in an article that consistently used templates to a style that mimicked CS1 but didn't use templates, or the reverse.
      Reading through the archives of this talk page will reveal many discussions about whether CITEVAR only applies to the rendered appearance of a citation, or it also applies to the wikitext that produces the citation. I can't find any clear consensus on this point, so I do not accept your statements written as if such a consensus existed. (To find such discussions, try entering "wikitext" in the archive search box near the top of the talk page.) Jc3s5h (talk) 16:27, 5 July 2018 (UTC)
      The definition of style in this guideline is about visual style (hence the comments about short citations either being harvard/bracketed against the usage of ref...tag pairs; and the sentence that starts "A number of citation styles exist...". Nowhere in the guideline is the use of templates equated with style. It is only this bullet point (and one additional bracketed comment) that prohibits the use of templates in an article that has none. The reasons for the prohibition has nothing to do with style. -- PBS (talk) 08:23, 6 July 2018 (UTC)
      I disagree; the totality of the talk page archives demonstrates there is no clear consensus as to whether this guideline can apply to undiscussed changes to wikitext that produces no visible change (with the probable exception of the bullet under discussion). Further, this is a companion to the MOS, which has a section Keep markup simple, which serves as a precedent that the MOS and related pages, including this one, apply directly to wikitext even if there is no visible change in the rendered article. Jc3s5h (talk) 12:35, 6 July 2018 (UTC)
    Wikipedia should have a new adage then, because this mostly amounts to If it ain't broke, don't improve it. Headbomb {t · c · p · b} 17:16, 4 July 2018 (UTC)
    @Headbomb: Please write that essay!  — SMcCandlish ¢ 😼  08:46, 5 July 2018 (UTC)
  • Oppose as written, because removing citation templates from articles that already have them is a bad idea, for the reasons that Headbomb describes above. Citation templates provide formatting consistency, automatic linking, and error checking that plain text does not. – Jonesey95 (talk) 17:43, 4 July 2018 (UTC)
  • Comment A related discussion from 2010, which proposed to establish a house citation stye, may be found at Wikipedia:Centralized discussion/Citation discussion. Jc3s5h (talk) 18:39, 4 July 2018 (UTC)
  • Oppose. I believe we all agree that it is preferable to template citations, but that is not question presented here. As to adding or removing templates: this is precisely one of the points where "standard consensus policy processes" is inadequate, where additional emphasis is needed. ♦ J. Johnson (JJ) (talk) 18:50, 4 July 2018 (UTC)
    I'm undecided about whether it's better to template citations. Citation templates in the past simply didn't work for some articles, and that was the environment in which CITEVAR was developed. We will see as the discussion develops whether those who disliked citation templates (and are still around) still dislike them. Jc3s5h (talk) 18:57, 4 July 2018 (UTC)
    So perhaps we are not all in agreement re templates, but, as I said, that is not question presented here. Whether we like, dislike, or dgaf, I think we mostly agree that we don't want people feeling emboldened to just dive in and make changes either way. And I think most of us feel that there still needs to be emphasis right up front to caution editors to not cannon-ball into the pool. ♦ J. Johnson (JJ) (talk) 23:03, 4 July 2018 (UTC)
    J. Johnson Your concern is addressed in the first sentence of the section "Editors should not attempt to change an article's established citation style merely on the grounds of personal preference, to make it match other articles, or without first seeking consensus for the change.". The bullet point is meaningless for the reasons I mention above about other citation templates already existing on most article pages and this prohibition totally failed to stop a bot change that added {{ISBN}} to well over 300,000 pages. Has anyone ever seen a person remove all the citation templates on a page that was consistently using citation templates before such an edit? -- PBS (talk) 09:55, 5 July 2018 (UTC)
    Certainly you have noticed that a few editors hold that seeking consenus first does not apply for "clear improvements". I believe we need to be explicit about some of the changes in the scope of "seek consensus first". However, while that could be an interesting discussion, I don't believe it would have any effect in the current discussion. ♦ J. Johnson (JJ) (talk) 20:20, 5 July 2018 (UTC)
    I certainly don't agree with the very unwise assumption that "we all agree that it is preferable to template citations". This might well be true of the assembled gnomes here, but is not for the wider editing group. I have used templates when the project convention required it, on medical articles, but the tools I got used to no longer work, so even my medical FA mostly has to get along without me now. Templates have all sorts of disadvantages, and their added benefits don't compensate imo. Johnbod (talk) 17:46, 6 July 2018 (UTC)
    It's more accurate to say that there's widespread agreement that templated citations are preferable. This is obvious from the fact that they used more – massively more – especially in GAs and FAs, but also site-wide. It's mostly new editors who don't use them (yet). The small percentage of established editors who hate citation templates are more likely to congregate at this talk page than anywhere else on the system, BTW.  — SMcCandlish ¢ 😼  06:55, 8 July 2018 (UTC)
  • I don't like it. (I am trying to be funny. Guess I failed.) Anyway, I never use citation templates because I am perfectly well satisfied to type out the cites myself; the system is deeply embedded in me; it provides the requisite sources (with links, if there are any); it does not have useless info (like when the cite was added). So that is one answer to the unnamed editor just above who forgot to add the tildes (but that's okay; we all get brain fugues from time to time). Thanks for reading this. Sincerely, BeenAroundAWhile (talk) 19:17, 4 July 2018 (UTC)
    @BeenAroundAWhile: Your claim that access/retrieved date is "useless info [about] when the cite was added" is false; it's useful info about when the citation was last checked as being valid for the material for which it's offered as the source. Any number of intervening editors could have messed with it and inadvertently [we hope!] falsified the verification status. It's important, and you need to include such a date even in manual, untemplated citations.  — SMcCandlish ¢ 😼  06:06, 5 July 2018 (UTC)
    And there's nothing wrong with writing everything manually the first time around. The point is to let gnomes more easily convert those to templated citations for enhanced functionality, consistency, error checking, metadata, etc. Headbomb {t · c · p · b} 20:05, 4 July 2018 (UTC)
    You do realize that changing citation style is still possible on an article by article basis... so the gnomes just need to discuss the change on the talk page first, and gain consensus for it. Of course, they would need to convince others that the ability to easily gnome is worth the change (not everyone cares about gnoming). Blueboar (talk) 00:55, 5 July 2018 (UTC)
    WP:BURO. Wikipedia belongs to the WP:BOLD. Begging for permission before editing is contrary to the spirit of Wikipedia. Headbomb {t · c · p · b} 01:12, 5 July 2018 (UTC)
    Yes. "the gnomes just need to discuss the change on the talk page first, and gain consensus for it" isn't really true. It's what CITE advises to avoid dispute, but we're free to IAR this, and we do it all the time, with nearly zero dispute. For one thing – whether the template-haters like it or not – there is a generalized consensus site-wide (observable in actual practice rather than rule twiddling omphaloskepsis) that templated citations are preferable because they provide consistent formatting, error checking, and useful metadata. Disputes arise when some asshat comes to a well-developed article with highly consistent CS1 citations and changes them to CS2, or kills the {{sfn}} arrangement of the citations, or flips from Harvard to Vancouver, or undoes WP:LDR, or other things in that vein. That kind of change likely does need consensus, and not seeking it, repeatedly and in the face of objections, is likely to get you ANIed. No one is going to get ANIed for converting substandard citations into good ones. Someone who tried to ANI you for it is fairly likely to get boomeranged on a POINT and OWN basis, if your improved cites are in fact improvements. Fortunately, the number of cite-format obsessives has dwindled, as has the number of careless or intentionally disruptive cite changers, so this'll probably never actually play out in kangaroo court at the drama boards.  — SMcCandlish ¢ 😼  06:06, 5 July 2018 (UTC)
  • Oppose. I like citation templates but I can see this leading to mass slow-motion edit wars between proponents of different citation systems. We're better off keeping the current wording. —David Eppstein (talk) 20:43, 4 July 2018 (UTC)
  • Oppose per Nigel and Eppstein.--Kmhkmh (talk) 01:03, 5 July 2018 (UTC)
  • Support removing the first half of it. Templated citations provide benefits, untemplated ones do not.  — SMcCandlish ¢ 😼  06:08, 5 July 2018 (UTC)
  • Also support removing the first half per SMcCandlish. · · · Peter (Southwood) (talk): 09:00, 5 July 2018 (UTC)
  • Oppose as drafted as per McCandlish. Consistently converting to templated citations is beneficial in so many ways. Peter coxhead (talk) 11:08, 5 July 2018 (UTC)
  • Oppose per Blueboar (and I routinely do use templates - but I also recognize that some fights just aren't worth it). Ealdgyth - Talk 12:17, 5 July 2018 (UTC)
  • Comment I find it interesting that no one has commented on the insertion of {{ISBN}} and {{webarchive}} to pages that do not use citation templates. Can such an addition be reverted under this bullet point? If not why not? -- PBS (talk) 16:03, 5 July 2018 (UTC)
    I believe that in the phrase from the bullet, "citation templates to an article that already uses a consistent system without template", the phrase "citation templates" was meant to apply to the CS1 and CS2 citation templates, the similar templates that preceded them and are now gone or out of favor, and any future templates that serve the same purpose. It does not apply to general use templates that are used within a non-templated citation. For example, if the title of an article contained scientific notation, I could render it with {{e}} in a non-templated citation. Templates that are of little use outside of citations are not quite so obvious, but I would put {{ISBN}} in the same category as {{e}} rather than treating it as a "citation template" in the sense of the bullet. Jc3s5h (talk) 16:39, 5 July 2018 (UTC)
    Those would be identifier templates, not citation templates. Headbomb {t · c · p · b} 16:48, 5 July 2018 (UTC)
    What about templates that are not based on CS1 or CS2 like {{cite wikisource}} or wrapper templates {{Cite ODNB}}. I think that this is a false dichotomy to say that {{harvnb}} is a citation template while {{webarchive}} and {{ISBN}} are not, because {{webarchive}} and {{ISBN}} and others are usually used part of a citation, and the point about not using citation templates is not meant to be "I don't like them" but such arguments as "using templates in inline citations makes it hard to read the paragraph" presumably the use of any template in a citation fits that description. -- PBS (talk) 17:33, 5 July 2018 (UTC)
    I disagree. The extra space occupied by citation templates vs. untemplated citations is only one of many reasons some editors prefer untemplated citations. If typing templates that are primarily used in citations, without actually being CS1 templates or something that serves the same purpose, occupies excessive space in the wikitext, or is more trouble than it's worth, that can be addressed on a case-by-case basis. If such a template isn't worthwhile, both templated and untemplated citations alike would benefit from the deprecation of the template in question. Jc3s5h (talk) 17:57, 5 July 2018 (UTC)
    Aye. We have a long history of merging away citation-related templates that aren't much use, and turning their intended utility into a parameter in CS1/CS2 (which are now the same codebase themselves).  — SMcCandlish ¢ 😼  21:37, 5 July 2018 (UTC)
    Jc3s5h, I think that there is no good reason for not mixing templated and non-templated citations in the same article providing they use the same style. What legitimate reasons have you come across for editors insisting that templates are not used -- Other than space and the consistent use of a different style? -- PBS (talk) 08:23, 6 July 2018 (UTC)
    One obvious good reason is that, while those who have learned to write templates in an external style have learned to put the elements in the correct order with the correct typographic treatment, there is probably no one on the planet who has memorized how to write a citation that looks like a CS1 template without the aid of the template markup. We should not expect editors to develop this new skill for no good reason, and there is no reason to expect anyone would want to develop that skill.
    Another good reason, which I believe has been mentioned a few times in the archive (but the archives are too vast to find the mentions) is that wikitext markup language is essentially a programming language, and programmers are passionate about consistent formatting and consistent idioms in source code. A language closely related to wikitext is TEX; for the first search result I found about the desire to properly format the source code in this related language see this StackExchange post. Jc3s5h (talk) 12:47, 6 July 2018 (UTC)
    Last, First. "Title". Website. Publisher: Location. Accessed January 1, 1980. --Izno (talk) 13:15, 6 July 2018 (UTC)
    Damn, "Retrieved". Still, the point you should make isn't "people should not learn the skill of emulating CS1/2" but instead "CS1/2 change and it's just easier to deal with that change by using the template instead of hand-writing those citations". --Izno (talk) 13:18, 6 July 2018 (UTC)
    @Izno: That's because you haven't seen those horror of those who try to hardcode something like Kolmann, Matthew A.; Crofts, Stephanie B.; Dean, Mason N.; Summers, Adam P.; Lovejoy, Nathan R. (2015-12-01). "Morphology does not predict performance: jaw curvature and prey crushing in durophagous stingrays". ''Journal of Experimental Biology''. '''218''' (24): 3941–3949. [[Digital object identifier|doi]]:10.1242/jeb.127340. [[International Standard Serial Number|ISSN]] 0022-0949. [[PubMed Identifier|PMID]] 26567348. directly in articles. Headbomb {t · c · p · b} 16:38, 6 July 2018 (UTC)

Comment. There is a change in the view toward citation-related software by The Chicago Manual of Style since this bullet was put in place. In the 14th and 16th edition I can't find any mention of citation management software. In the 17th edtition (2017) on pages 744–5 there is a subsection on citation management tools. The tone of the subsection seems to presume scholars now use citation management tools; it specifically mentions Zotero, which is capable of creating Wikipedia citation templates. But as Chicago says, the results of such tools need to be manually inspected to check that the tool worked properly. Jc3s5h (talk) 14:02, 6 July 2018 (UTC)

  • Oppose per Blueboar and Ealdgyth. Hchc2009 (talk) 17:56, 6 July 2018 (UTC)
  • Oppose with no prejudice against someone trying to change the citation style though the dispute resolution/consensus building process. I recently encountered a very uncooperative editor who decided to change the citations on a well-developed article from CS1 to Harvnb (among other things). It was very disruptive. Without a clear guideline, addressing such disruption would be very difficult.- MrX 🖋 18:50, 6 July 2018 (UTC)

Consensus? As of 18:50 I see one "Support" (from PBS), and eleven "Oppose". (Also two "support removing the first half" [SMcCandlish and Peter], but that is not the question presented.) Though not unanimous, consensus is clearly against the proposal. Are we done with this? ♦ J. Johnson (JJ) (talk) 21:14, 6 July 2018 (UTC)

Meh... Let the process play out... while I don’t expect a late flurry of supporters to suddenly show up, I suppose it could happen... and people wikilawyer about less. Blueboar (talk) 21:32, 6 July 2018 (UTC)
  • half support We should encourage conversion to cite templates and not put hurdles in the way. But we should discourage their conversion to non templated references. Graeme Bartlett (talk) 04:39, 7 July 2018 (UTC)
    I agree that we should encourage conversion to templates (but not necessarily {cite}), and even discourage removal of templates. But NOT by granting license to freely trample over the sensibilities of others. Even if some cretin with an inadequate education fails to appreciate the rightness of "{cite xxx} for all", the requirement to seek consensus first is no "hurdle" to such conversions; it's barely a speed bump. Encouraging the simple courtesy of a preliminary explanation shows respect and basic civility, while granting a right to "shoot first, explanations optional" only erodes it. ♦ J. Johnson (JJ) (talk) 20:01, 7 July 2018 (UTC)
    To the contrary, it's a massive time-drain and a recipe for endless drama.  — SMcCandlish ¢ 😼  06:55, 8 July 2018 (UTC)
  • Oppose removal. That sentence has prevented a lot of disputes. SarahSV (talk) 05:33, 8 July 2018 (UTC)
  • Oppose I too oppose removing the sentence. If I find an article with an idiosyncratic citation style, I am perfectly capable of posting on the talk page any concerns I have that the citation style in that particular case is hampering development of the article. We can then move towards consensus, or not. The more usual situation I encounter is the mixed citation article, or even worse, the use of short form citation without providing the base bibliographic data. In those cases, I feel free to beef-up the article and provide full and uniform citation, usually with the use of the citation templates. --Bejnar (talk) 01:28, 9 July 2018 (UTC)
  • Oppose as drafted: There are times when I take a crusty old page and do a complete rewrite. I do a lot of work in the areas of computers and engineering, and a lot of those pages were among the first added to Wikipedia, followed by years of neglect. In such cases I certainly don't need some misguided policy telling me that I can't completely redo the citation style. --Guy Macon (talk) 13:50, 10 July 2018 (UTC)
    Yep. This is what those who want to misinterpret CITEVAR as something akin to an inviolable law just refuse to accept. And we'll continue to edit around them while they talk to themselves in their echo-chamber bubble here. At some point the community will catch on that part of the guideline wording here is trying to force rather than reflect actual community best practice. I'm patient.  — SMcCandlish ¢ 😼  22:09, 11 July 2018 (UTC)
  • All you have to do, per CITEVAR is consult the talk page & wait a reasonable interval. But that's JUST TOO HARD for some people, it seems! Johnbod (talk) 19:28, 21 July 2018 (UTC)
  • I am confused Guy Macon the proposal here is to remove a bullet point that is part of WP:CITEVAR tells you that you can not "completely redo the citation style", so why do you oppose the removal of this bullet point? -- PBS (talk) 18:36, 21 July 2018 (UTC)
    You are misinterpreting CITEVAR. It does not mean that you cannot do a complete rewrite of a page without retaining the citation style (or anything else) from the old page. As the guideline says: "Editors should not attempt to change an article's established citation style merely on the grounds of personal preference, to make it match other articles, or without first seeking consensus for the change." Any time you interpret a guideline in such a way that it prevents completely rewriting a page, you are interpreting it incorrectly. Some pages are so bad that it is best to just nuke the page and start over. --Guy Macon (talk) 19:05, 21 July 2018 (UTC)
  • Strong Oppose although I think the guideline should distinguish more between external templates like isbn and bibleref, and actual templated citation styles like sfn etc. Johnbod (talk) 19:28, 21 July 2018 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Guideline on wikilinks within citationsEdit

Should we have a guideline on when to use wikilinks within citations? E.g., if using {{cite news}}, is it better to use "publisher=New York Times" or "publisher=New York Times" "work=The New York Times" or "work=The New York Times"? Does it depend on whether the publisher or author has a Wikipedia article? And should the same publisher/author for multiple refs be linked only the first time?

According to previous discussions here, it's currently up to the editor, which can sometimes lead to confusion. Λυδαcιτγ 01:29, 18 July 2018 (UTC)

@Audacity: Aside: |publisher=New York Times is off-target, twice over (incorrect parameter, incorrect name). If you've been putting the publication name in the publisher field, please stop; it's misleading citation metadata and it's creating a lot of frustrating cleanup work for other editors. It's like confusing Game of Thrones with HBO, or The Magical Mystery Tour with Apple Records. This should be done as |work=[[The New York Times]] (|newspaper=, |journal= are equivalent). The The is part of the title in this case. If it had a publisher with a distinct name, it would be added as, e.g., |publisher=[[News Corporation]]. If it was an AP or Reuters or other newswire story, we also use, e.g., |agency=[[Associated Press]], and if we got it via some intermediary, that'd be something like | In reality in this case, the publisher of NYT is The New York Times Company, so including it is redundant and should just be skipped. I'm pretty sure {{cite news}}'s doc is clear about all this; the problem is people don't read it and just copy-cat badly formatted citations from 2006 or whenever.  — SMcCandlish ¢ 😼  01:59, 18 July 2018 (UTC)
Cheers - updated above. I do use the Work field, though I find the names confusing -- but that's a separate issue. Λυδαcιτγ 03:23, 18 July 2018 (UTC)
We should probably address this general linking-in-cites matter in the guideline itself. Our general MOS:LINK guideline doesn't want links repeated over and over because it causes a "sea of blue" effect. That's arguably even more important here because citation text is dense and small already. We shouldn't be redlinking anything in citations, nor linking stuff more than once. It's reasonable to link stuff on first appearance in a citation even if already linked in the body, because he cites render in a different section, and we often re-link something the first time it re-appears in a new section if separated from the previous link by one or more intervening sections.

A countervailing viewpoint I've encountered is that every citation is a "stand-alone thing" and should be fully linked for everything in it which can be bluelinked. I disagree with that idea, for MOS:LINK reasons and because it makes it harder to tell which link actually is the one to click on to get to the actual source. A complication of linking only in the first cite to the same stuff is that citations can move around with the text and end up in a different order, but this is a fairly minor cleanup matter, and most of our citations are under-linked anyway (at first occurrence), without any measurable negative effect on the encyclopedia.
 — SMcCandlish ¢ 😼  01:59, 18 July 2018 (UTC)

When I mouseover a citation number, I get a tooltip containing the citation. If I want more info about, say, the publication, should I be expected to go find another citation that links it? It would be easier to just scroll to the top and type the publication name in the search box. Linking only in selected cites, first or otherwise, is a ridiculous idea that shouldn't even be on the table. Not to mention that it would be unmaintainable for a given article in the long term.
Agree that nothing should be redlinked or linked twice in the same cite. If there is to be any written guidance, it should be a gentle suggestion to link anything that can be bluelinked, no more than once in a cite. ―Mandruss  04:06, 18 July 2018 (UTC)
What Mandruss said. Overlinking doesn't apply to citations in the reflist, although they might certainly apply in lists of works in general. Quark#References is a good example. If you click on Ref 7, you really don't care that Ref 6 also has a link to Physical Review Letters. Red links to notable journals/publishers are fine too, since it encourages the creation of those articles.Headbomb {t · c · p · b} 06:23, 18 July 2018 (UTC)
Strongly disagree on the last point. Citations are not for encouraging editing, they're for providing references. If we're going to deny that that a main-prose rationale can be applied to citations, just so we can favor re-re-re-linking the same stuff over and over again, then the same "citations and main text are different" rationale can also be applied against redlinking. It doesn't serve a purpose in the citations, and it makes it look like there's something wrong with the cites, i.e., it casts doubt on their and thus our reliability.  — SMcCandlish ¢ 😼  09:33, 18 July 2018 (UTC)
I have to disagree with We shouldn't be redlinking anything in citations--I think the more-reasonable position is somewhere in the realm of We shouldn't be redlinking anything in citations that we wouldn't expect to be notable, which I think hews a bit more closely to the intent of WP:REDLINK. (But I haven't looked that way recently and am on mobile ATM.) --Izno (talk) 20:10, 18 July 2018 (UTC)
See above.  — SMcCandlish ¢ 😼  03:51, 19 July 2018 (UTC)
What you present therein is a false dichotomy. We don't need to treat citations as exactly like normal text in all cases nor do we need to create exceptions for citations to the way we treat normal text in all the ways in which we might. I don't like seeing seas of blue in my references section, but I also appreciate why someone might decide that such is useful to the reader (namely, ref pop ups, which I believe are live for all readers on both desktop and MobileFrontend). As for "citations aren't for encouraging editing", I don't think I agree with that statement whatsoever. While I agree their main purpose is to provide provenance information, it doesn't strike me as untoward to allow them to serve as potentially adding to our red link backlog. I do also disagree with "it makes us look bad"--it's a wiki. All red links, with that logic in mind, make us look bad, not just in references, which is quite an extreme opinion and unsupported by current consensus (enshrined in REDLINK). Maybe someone needs a refresher on Wiki 101. ;) --Izno (talk) 04:56, 19 July 2018 (UTC)
You're either mistaking every element of my argument or twisting it on purpose. The "false dichotomy" you project is a straw man. I never suggested that because arguments are presented to ignore one rule in citations that none of them can apply. I'm pointing out that "because MoS says so" is a hypocritical argument in favor of redlinking when it's coming from the same people who reject "because MoS says so" as a rationale for something that doesn't suit their preferences (excessive bluelinking). I also never suggested that it's not possible for a citation to serve as a means of noticing a redlink and deciding to do something about it. I said that it's not the purpose of citations. (Analogy: If I really want to, I can use my boot as a hammer, but I will not go out selecting my footwear on the basis of how well it might drive a nail into softwood, and using it that way is apt to have negative effects on its utility for what it was actually designed for.) We have cites to provide verifiability for our article content, more specifically to provide sufficient source information to identify (not treat in full like some kind of bibliographic database) the sources we are using. That segues into a third point: My logic does not even slightly suggest that all redlinks make us look bad. Red links in article prose are part of our "build the Web" article content network of cross-references. Doing them in citations, which are not part of the content but are meta-data, makes it look like our metadata is incomplete or broken. If a citation is sufficient to correctly identify the source, then it is not in fact broken or incomplete for the purposes for which it exists. All that said, I'm more sympathetic to average to being inclusive of source information (e.g. full author names when known, subtitles along with main titles, publisher name and location when it's not redundant, more than one URL-generating identifier, etc.). I'm also not entirely opposed to the idea below of including a redlink when we actually expect that we should have an article for that publisher (or whatever).  — SMcCandlish ¢ 😼  05:23, 19 July 2018 (UTC)

In the past I would also add the wikilinks to the publisher (and sometimes the author) but I was later asked by various authors to stop it. I recall the locic was that it would create excessive incoming links for each article. Thus I see two issues, please feel free to comment on the veracity.

  • SEO google and other search engines will see these wikilinks and place undue importance on some articles such as fortune, etc. This might be less important to us editors, but maybe still to take into account.
  • Inbound link counts. This will make it difficult for us editors to look at Help:What links here.

I think a policy on including or exlcuding the wikilinks would be useful, as audacity (talk · contribs) has suggested. Thanks! Jtbobwaysf (talk) 07:08, 18 July 2018 (UTC)

  • We should definitely include wikilinks for authors when we have them. On all citations, not just the first one, so that they can be found easily from the individual reference. This goes in both directions: If we want to know what biases or expertise a source might be expected to have, an article on its author is the first place to look. And if we want to know what topics an author is known for writing about, the list of incoming links to that author's page is again a good first place to look. I generally also add links to newspapers or journals, for a different reason: when we see that a source's publication is bluelinked, we can get an idea how reliable the source is. But I feel more strongly about linking authors than publications. I have less interest in linking publishers, but I don't think that's particularly harmful either. SEAOFBLUE applies to article text, not references. —David Eppstein (talk) 07:25, 18 July 2018 (UTC)
    On linking publishers, I think we should only do it when it pertains closely to reputability. E.g., do link Oxford University Press and SAGE Publications; do not link Salon Media Group or Bantam Books.  — SMcCandlish ¢ 😼  09:37, 18 July 2018 (UTC)
  • Comment: I prefer to link in citations if I feel that the link is beneficial, such as linking to a publisher so that readers and editors can quickly check that the publisher is reliable. Linking the author's name can also be beneficial, and the "authorlink" field helps with that. I don't link authors in references often, though. I might start doing so. Flyer22 Reborn (talk) 05:11, 19 July 2018 (UTC)
  • I could live with the idea of optionally blue-linking author, publisher, and periodical information (not location), when it's likely to be useful to the reader, even in multiple citations – if we're certain this really useful in the cite pop-ups – and even red-linking in cases where we're certain the article should exist. What concerns me are a) I see redlinks in cites again and again for random journalists and book publishers and small-town newspapers that are obviously non-notable, and b) I do not want to be groused at by some citation formatting obsessive for not linking something they think I should have linked. I still don't think it's useful to blue-link either 1) general book publishers like Ballantine, Penguin, Houghton-Mifflin, etc., versus publishers that specialize in academic material (or those that specialize in fringe – reputability cuts both ways), nor 2) publishers that are pretty much synonymous with the publication.  — SMcCandlish ¢ 😼  05:33, 19 July 2018 (UTC)
    In one comment, you (1) state personal opinions about what should and should not be linked (not by parameter but by parameter value), and (2) state as part of your reasoning that you do not want to be groused at by some citation formatting obsessive for not linking something they think I should have linked. I guess you can somehow reconcile the two, but then you're smarter than I am. The simplest and most conflict-free way to do this is not to try to guess what will be of interest to the reader, which guess will vary widely between editors, but to give the reader link access to everything we can. Useful guidance tends to point most editors to the same answers. ―Mandruss  05:51, 19 July 2018 (UTC)
    It's a Wikipedia_talk page. We're apt to express opinions here. There's no conflict of any kind between us offering opinions about what our linking guidelines should be (mostly a matter of why / why not to link something), and – quite separately, in a different domain, not not wanting to be badgered by people in and about mainspace for not doing optional things they prefer were done. They're completely unrelated; the latter is a behavioral matter (another variant of "do not go hassle people on their talk pages or be rude to them in edit summaries because you don't like their writing style" – no one actually has to comply with, or even read, MoS to contribute new material; we're free to clean it up later, and it's more important that people add content and source it than that they get style nitpicks correct. It really only matters if new or careless editors start replacing MoS-correct style with random idiosyncrasies, which becomes disruptive if it doesn't cease). It's also not particularly connected to assessing link value to readers (not a behavioral matter, nor a policy-writing matter). The "[we should] not to try to guess what will be of interest to the reader" idea would completely invalidate most of MOS:LINK, including the part of relevance here. We'd simply link everything that could possibly be linked, on the basis that someone somewhere might appreciate any particular link. Consensus obviously does not favor that approach, or we would not have a long-term guideline against over-linking. We do in fact do this assessment all day every day when we're writing Wikipedia. The argument here is to link more in citations that we normally would. I don't find the argument for it compelling, because people are arguing to link things that are not necessary for identifying and finding the sources; they just want to link it all because they like linking stuff, as far as I can tell. There's not a valid rationale for it that I can see, compared to the more sensible one of linking certain key information pieces (notable authors, publisher when relevant, periodical name, plus including more than one linked identifier when available) repeatedly in cites because of how they're used.  — SMcCandlish ¢ 😼  18:40, 19 July 2018 (UTC)

I agree with Headbomb and Mandruss that the guideline which was primarily designed with having the article main body/text in mind, may not be that helpful if applied (strictly) to other special sections of the article such as infoboxes and the references/notes sections. First of all citations are usually not read by the average reader and they are usually not read like text paragraph or block either, hence that sea of blue argument about impairing reading experience in article's main body doesn't quite apply here. So rather than being concerned how blue links or a sea of blue might impair reading of all citations as whole (or like a text block), we should focused on improving the ease of navigation for individual (or "standalone") citations. Because the primary use case are imho readers looking at a selected few citations individually and hence we should not force those readers to browse all citations just to figure out whether a blue link and/or an online copy of the source exists.

In addition ease of use for editors is more important here than for the article's main body as well, because while only a fraction of the readers will actually deal with citations and sources almost all editors have to.--Kmhkmh (talk) 09:43, 19 July 2018 (UTC)

  • In my view only links to help find the source should be linked. The refs are there to support and lead readers to more info.........dump link spam!!--Moxy (talk) 18:09, 19 July 2018 (UTC) cite format?Edit

What format should I be using to cite from google books? Govvy (talk) 14:39, 25 July 2018 (UTC)

Follow the citation format that already exists in the article. Different articles can have different citation styles, see WP:CITEVAR. Keep in mind that whatever style is followed, it applies to all the sources in the article (but it may vary according to the kind of source, a book vs. a journal article, for example). The citation style recommended by the source is irrelevant; if an article contained a cite to a journal published by the American Psychological Association and The Astronomical Journal, each of which has a different style used within those journals, they would be formatted just the same as all the other journal cites in the Wikipedia article. Jc3s5h (talk) 15:26, 25 July 2018 (UTC)
I s'pect the question is angling for a different fish. @Govvy: the key point is that Google Books is not the publisher; they're just the WP:SAYWHEREYOUGOTIT conduit. In a templated citation do something like {{Cite book |first=Jane |last=Smith |title=Nocturnal Underwater Basketweaving Basics |date=2018 |location=Los Angeles |publisher=Crafty Publishing |access-date=27 July 2018 |via=Google Books}}. In an article not using templated citations, then something like: Smith, Jane (2018); ''Nocturnal Underwater Basketweaving Basics''; Los Angeles: Crafty Publishing; accessed: 27 July 2018, via Google Books., with punctuation and order that matches the rest of the article.  — SMcCandlish ¢ 😼  19:12, 27 July 2018 (UTC)
that makes more sense than cite web format to me, cheers. Govvy (talk) 19:19, 27 July 2018 (UTC)

RfC on access-dates and verifiabilityEdit

  FYI: Pointer to relevant discussion elsewhere.

Please see: Help talk:Citation Style 1#Permit access-date in absence of a URL

Gist: The |access-date= parameter in citation templates indicates the last date at which someone checked whether the cited source actually verifies the claim(s) to which it is attached as a reference. Presently, if the citation does not have a |url= parameter, not only is the display of the date suppressed, it is categorized as an error that people should remove. The RfC asks whether this is the appropriate course of action.  — SMcCandlish ¢ 😼  04:50, 28 July 2018 (UTC)

Proposal to end conflicting date formats within the same citationEdit

  FYI: Pointer to relevant discussion elsewhere.

Please see WT:Manual of Style/Dates and numbers#End "date-forking" into different styles for publication and access/archive in same cite
 — SMcCandlish ¢ 😼  15:24, 29 July 2018 (UTC)

Return to the project page "Citing sources".