Wikipedia talk:Citing sources

Referencing individual page numbers by embedding pdf links

I stumbled across Metropolitan Manila Transit Corporation which cites different pages of several sources multiple times and is using {{Rp}} to do so. That's perfectly OK per WP:IBID and is something done by many others. What I've never seen done before is the embedding of links to individual pdf pages into the "Rp" syntax (like ). I've looked a both the documentation for the template and WP:REFPAGE but was unable to find any type of guidance regarding this kind of thing. An example of what I'm talking about can be see at Metropolitan Manila Transit Corporation#cite_ref-JUMSUT-MT-C2-1984_33-0. It seems like there should be better way to do such a thing (e.g., WP:SFN) but perhaps it's not wrong per se. Has anyone else ever come across this type of thing before? In principle, this seems like a potential problem per WP:CS:EMBED since this using embedded citations is a style that was deprecated more than ten years ago (I think). -- Marchjuly (talk) 05:15, 22 January 2026 (UTC)

I have never seen it before with {{rp}}. However, I don't think it is a problem with respect to WP:CS:EMBED because it is not an embedded citation as such; that is, the external link does not appear in the article content body, but as part of the inline citation. Template:Reference page § Intent says {{rp}} is an alternative to the method of using shortened footnotes and then Template:Sfnp § Adding a URL for the page or location explicitly allows this If a specific link to the page or section is available, a URL can be added to the location or page number, so I'd deduce from that that it's OK. Cheers, SunloungerFrog (talk) 06:22, 22 January 2026 (UTC)

Thank you for the response. However, in the alternatives you cite, the embedded link appears in the "References" or "Notes" section pretty much the same way it would for any regularly formatted reference; in this particular case, the embedded link actually does appear in the body of the article, i.e., withing the footnote marker itself. Moreover, the first two examples given in Template:Sfnp#Adding a URL for the page or location look like parenthetical references, a style that was deprecated a few years back; so, it seems possible that part of the template's documentation wasn't updated to reflect parenthetical referencing is no longer considered OK. -- Marchjuly (talk) 06:56, 22 January 2026 (UTC)

I see what you mean, though I would say that it meets the spirit of the rules, which is that bare urls within body text like this are bad in the absence of fuller bibliographic information; in this case there is fuller bibliographic information, and I would not myself consider it to be against WP:CS:EMBED or WP:NOELBODY.

With respect to Template:Sfnp § Adding a URL for the page or location, the fact that {{harv}} (the first two rows in the table) is deprecated (maybe unless wrapped in <ref>...</ref> tags?) does not to my mind take away from the statement that URLs are permitted to be added to the location or page number in other shortened footnote templates, and so, I think, by extension within {{Rp}} too.

More generally, I think that doing something like this is practically helpful for the reader - it makes it quicker and easier to get to the actual source material and helps avoid duplicating bibliographic information - so why not? Cheers, SunloungerFrog (talk) 09:37, 22 January 2026 (UTC)

Just as a heads-up: We'll do the global roll-out of sub-referencing this year. One of the use cases we've observed is adding the page number in combination with a link to individual pages, e.g. de:Doris Stockhausen#Einzelnachweise. Johannes Richter (WMDE) (talk) 10:48, 22 January 2026 (UTC)

That actually looks quite interesting. -- Marchjuly (talk) 11:15, 22 January 2026 (UTC)

Yes, I agree. But it's also a visible change, so expect some editors to object. (After all, we get about three-quarter million registered editors a year, so it'd be weird if everyone agreed on anything.) WhatamIdoing (talk) 01:26, 27 January 2026 (UTC)

The {{rp}} technique is an ugly relic and should be phased out as the opportunity arises. True it is not formally deprecated but it is a typographic disaster, it makes pages look like crap graffiti on a subway train. To make it even worse, as this technique would do, crosses a line that should not be crossed. Embedded citations are explicitly deprecated and this is an embedded citation in flashing neon lights. The right way to do this is to use {{sfn}} or {{sfnp}} – SMcCandlish has done a great essay at User:SMcCandlish/How to use the sfnp family of templates if anybody would like more details. Meanwhile, revert. --𝕁𝕄𝔽 (talk) 11:01, 22 January 2026 (UTC)

While I agree with much of what you posted, this isn't a case of simply reverting a change made by someone else. The creator of the article, who is also the primary contributer, actually did this from the get go; so, it wasn't a case of someone showing up and introducing a new citation style mid-stream. I've notified the creator of this discussion; so, perhaps they will comment. -- Marchjuly (talk) 11:15, 22 January 2026 (UTC)

It's certainly reasonable for that person to have input, but they don't have "more say" than anyone else. There's a frequent misunderstanding about WP:CITEVAR and all the MOS:VAR provisions, that we defer to the preferences of first/earlier editors. That's not the case. All of them are clear that only when consensus cannot be reached on which variation of something is preferable do we defer to the version used in the first [non-stub] version of the article (or of the material in question). This is simply to have some kind of fall-back default, not because the first major contributor has a supervote.  — SMcCandlish ☏ ¢ 😼  17:39, 24 January 2026 (UTC)

@SMcCandlish: I agree with this, and I probably should've made that more clear it my prior post. I was just trying to note that there was no other citation style to "revert" back to per se. Since this was used from the beginning, it would mean a new style would need to be introduced. So, that's one of the reasons I notified the article's creator (who introduced this style) of this discussion. Once everyone is on board that a change is needed, then how to change it should be fairly easy to discuss on the article's talk page. -- Marchjuly (talk) 21:10, 24 January 2026 (UTC)

Thanks for letting me know. I actually didn't check any policy or guidelines before using links with {{Rp}}; I just saw a few articles use it when citing a book source by using it to link directly to the specific page of the book on Google Books. Though given that most of the times I've used it were for mere PDFs on websites and not actual direct links to pages on Google Books, I'll probably consider limiting the use of links with {{Rp}} in the article to only those that directly link to the page. Oh, and I look forward to using the sub-referencing feature instead when it comes out. Ganmatthew (talk • contribs) 16:27, 22 January 2026 (UTC)

Proposed deprecation of {{rp}} and related templates

There are really two considerations here, and they are not actually at cross purposes.

First, the community completely deprecated parenthetical citations (WP:PAREN). This is the shoehorning of citation data (authors, dates, page numbers, etc.) into article text as inline parenthetical asides. "Parenthetical" here has its original and primary meaning of an extraneous annotation or comment; it does not refer in particular to round brackets, which are only called "parentheses" in American English (and only because of their frequent use for parenthetical-in-the-original-sense annotations). So, yes, {{rp}} is, like all other parenthetical citation, deprecated by the community and should be replaced, whether or not anyone's bothered putting {{Deprecated template}} in that particular template's documentation.

Second, we should generally do what is helpful for readers and not do what is unhelpful for readers. Confusing, geeky gibberish like "^[12]:349" is not helpful to readers, but an impediment to their reading. On the other hand, it is certainly useful to link (in the citation, not in the middle of article text) to specific online page scans in a source when we can do so. This can save a reader (including an editor) anywhere from minutes to months (depending on how hard it is to otherwise get ahold of the source and go through it) in an effort to determine whether a cited work actually supports the claim in our article. A simple way to do this in short footnotes: {{sfnp|Smith|Garcia|2003|p=[https://www.archive.org/whatever 349]}}</ref>. This has for {{em|practical}} purposes the same use as <code><nowiki><ref name="Smith & Garcia (2003)" />{{rp|[https://www.archive.org/whatever 349]}}, except better in at least three ways: It keeps the citation data in the refs section instead of stuffing it into an inlined parenthetical, and it links to the full citation of the Smith & Garcia (2003) source in the bibliography, and it associates the page number with the source identifier in an immediately human-readable way (versus something like "^[12]:349", which requires the reader to click around and try to figure out what "[12]" refers to).

While Template:Rp served a purpose when I created it in 2007, it was actually made obsolete within only a couple of years by improvements to the <ref> system and further development of templates to work with those improvements (with more improvements coming soon, like sub-referencing). {{Rp}} is now just an unslightly and pointless cancer. We really should not have drawn out article-by-article discussions about replacing it, when there is already an across-the-board community consensus for replacement of all parenthetic citation methods. Just do it, in a way that doesn't lose functionality (like linked pages). PS: If a link has been put around a page number but does not go to the page but only to an entire PDF (or an Internet Archive "Login and borrow" page), that is reader-hateful behavior, and the link should instead be in |url= in the main citation (with |access=registration if needed). Similarly, any URL that just goes to a Google Books or Amazon or whatever page about the book without providing any access to the book text should just be removed, since it serves no function but advertising/sales.  — SMcCandlish ☏ ¢ 😼  17:39, 24 January 2026 (UTC)

Interesting! But if {{rp}} is de facto deprecated, maybe we should put {{Deprecated template}} into its documentation? That would certainly help to facilitate transitions away from it and discourage its usage in new cases (which still happens quite a lot). Gawaon (talk) 18:24, 24 January 2026 (UTC)

I think that would be a great idea. Cheers, SunloungerFrog (talk) 18:28, 24 January 2026 (UTC)

Yes, this is long overdue.  — SMcCandlish ☏ ¢ 😼  18:08, 25 January 2026 (UTC)

The only reason I've seen suggested for using {{rp}} is that it allows the reuse of a refname while also allowing for different page numbers with each instance. This is exactly what sub-referencing will allow, so rp will become redundant once sub-referencing goes life. -- LCU ActivelyDisinterested «@» °∆t° 20:16, 24 January 2026 (UTC)

As I've indicated above, {{rp}} is already obsolete, and has been for years, even without sub-referencing. You can reuse the same ref with a different page number at each instance in this manner (among others):

Article text making claim no. one.<ref>{{cite book |last=Smith |first=Jane |date=2006 |title=Big Important Book |publisher=Miskatonic University Press |pages=22–23}}</ref> Claim no. two.{{sfnp|Smith|2006|p=47}} Claim no. three.{{sfnp|Smith|2006|pp=139–140}} Claim no. four, which needs some annotation.<ref>{{harvp|Smith|2006|pp=ix, 2}}. Some editorial annotations here.</ref>

It's easy and consistent. See my tutorial for more details: User:SMcCandlish/How to use the sfnp family of templates.  — SMcCandlish ☏ ¢ 😼  18:08, 25 January 2026 (UTC)

My point was in regard to editors who dislike using short form refs. Rather than saying that RP has any real current use. -- LCU ActivelyDisinterested «@» °∆t° 18:56, 25 January 2026 (UTC)

"dislike" doesn't convey the attitude of editors invoking WP:CITEVAR when they resist any improvement in an article's citation mechanisms. I wonder how the forthcoming subreferencing feature will fare. -- Michael Bednarek (talk) 01:02, 26 January 2026 (UTC)

Yes, {{rp}} is slightly ugly^[133:12] but so are sub-references^[133.12]. This is not a significant improvement and rp is more concise. Johnjbarton (talk) 01:52, 26 January 2026 (UTC)

Current policy allows for a wide breadth of different referencing styles, CITEVAR says the style of an article can be changed if their is consensus to do so. Curmudgeons not liking their articles changed is just something that has to be worked through.

RP isn't styled ^[133:12] instead the difference between RP and sub-refs is ^[133]:12 and ^[133.12]. So between leaving page numbers in the body of the article, and having the page numbers in the reference section. -- LCU ActivelyDisinterested «@» °∆t° 08:51, 26 January 2026 (UTC)

Morever, there are often several page numbers or a range and they have two or three digits, while the first nine different subreferences to any given reference (which should be enough in nearly all cases) have just one digit, so the actual difference is rather between ^{[133]:123–137,151} and ^[133.2]. To me, at least, the latter looks indeed more reader-friendly and less cluttered. Gawaon (talk) 08:59, 26 January 2026 (UTC)

"...and rp is more concise". It's not. "^[23:12]" is precisely the same length as "^[23]:12", but is also easier to parse than the latter, as it groups the entire citation inside [...], instead of slpitting it into [one]:two parts with different formatting. Rp has always been problematic in this regard, but was not "fixable" because the <ref> extension is what generated the [...] markup. And that's without considering Gawaon's point about "^{[133]:123–137,151}" and "^[133.2]". In actual practice, a sub-ref of the latter sort will always be shorter than a corresponding rp instance, except in the unusual cases that the same source is cited 10+ times, at page numbers shorter than 10 (i.e. to produce something like "^[133]:7" vs. "^[133:12].  — SMcCandlish ☏ ¢ 😼  09:33, 26 January 2026 (UTC)

Sub-referencing – Reference Previews

We've seen quite a number of dewiki articles using the same source more than 10+ times with different details (e.g. de:Magdalena Spínola – 22 different sub-references), but readers don't mind about footnote numbers like ^[1.22] according to our user research, especially given the benefit of showing both the main bibliographic information and page number (or other details) in Reference Previews, just like with regular references.

However we are thinking about improvements to the reference list, because users pointed out that there's a lot of unused space when an article uses more than just a few sub-references.

We've published our learnings from the first months of sub-referencing on German Wikipedia in m:WMDE Technical Wishes/Sub-referencing/Learnings – including our next steps in the development of sub-references and our plan for the global rollout this year.

If anyone want's to try Sub-referencing: It is deployed on German Wikipedia and testwiki as well as the beta cluster. Johannes Richter (WMDE) (talk) 10:37, 26 January 2026 (UTC)

But this creates the annoying two-layer sources thing, so it does not replace rp. Johnjbarton (talk) 01:53, 26 January 2026 (UTC)

SMcCandlish said: "Parenthetical" here has its original and primary meaning of an extraneous annotation or comment; it does not refer in particular to round brackets [...] {{rp}} is, like all other parenthetical citation, deprecated by the community and should be replaced, whether or not anyone's bothered putting {{Deprecated template}} in that particular template's documentation. This does not seem to be a valid interpretation of the parenthetical citation deprecation. Neither the content guideline at WP:PAREN nor the successful proposal for parenthetical citation deprecation at WP:PARREF mention {{Rp}} or anything like it. Both only mention in-text citations using parentheses as described in Parenthetical referencing. There may be good reasons to deprecate {{Rp}}, but the deprecation of parenthetical citations at WP:PARREF is not one of them. Biogeographist (talk) 02:46, 26 January 2026 (UTC)

That's classic WP:Wikilawyering. If what you believe were correct, then the entire WP:PAREN decision could be instantly circumvented at will (and with no recourse other than a new anti-parenthetical-citations RFC to stop you) simply by tweaking all the old Harvard referencing templates to emit things like "[Garcia 2012]" or "Garcia [2012]" instead of the original "(Garcia 2012)" or "Garcia (2012)" [depending on which brand of the template you chose, as to where it put brackets]. Let's not be silly. There is not a person in this e-room who will believe for a second that you could get away with that. The obvious and only sensible interpretation of the entire discussion that resulted in WP:PAREN is that the community does not want citation detailia injected inline into articles, interrupting the reader in mid-sentence (any more than the minimal "^[2]" already does).  — SMcCandlish ☏ ¢ 😼  09:28, 26 January 2026 (UTC)

You must be joking, because I don't believe that anyone would simply replace parentheses with square brackets to implement parenthetical referencing, nor have I seen anyone try to do that; you're just making up a thought experiment about an implausibility! (And then you're accusing me of being "silly" and "wikilawyering"?) In the huge discussion at WP:PARREF, the proposal said I am merely proposing that we do not use inline, non software based, text parentheticals, and the only mention of {{rp}} in that discussion was an approving statement about it being an acceptable way to cite multiple pages from the same book, a statement which only elicited agreement. So it can't be right that the WP:PARREF discussion applies to {{rp}}; there was ample opportunity to explicitly say in the discussion that the deprecation applied to {{rp}}, but the exact opposite was said instead: {{rp}} was considered an acceptable alternative. Biogeographist (talk) 14:10, 26 January 2026 (UTC)

• "Second, we should generally do what is helpful for readers and not do what is unhelpful for readers."

I disagree with the unsupported claims about what readers want. (The evidence in the world suggests they want AI summaries). But if we did believe that citation marking is unhelpful for readers it would not be very difficult to change the rendering to omit footnotes altogether.

In my opinion "verifiability" is much more important than "readability" in the parts of Wikipedia I work on. Readability is "nice to have"; verifiability is essential. Helping editors add sources and check sources is a priority. Johnjbarton (talk) 17:31, 26 January 2026 (UTC)

@Gawaon, SunloungerFrog, and SMcCandlish: If you are going to put {{Deprecated template}} into the documentation of {{Rp}}, as the three of you said above that you want to do, you should do the same to {{R}}, which produces the same kind of superscript inline page numbers. (There may be other templates that do the same?) A week ago I added a reference using {{Sfn}} (which was the template I instinctively used to reference multiple page numbers) to Against Empathy, then a few days later I noticed that other references already on the page used {{R}}, so I changed my {{Sfn}} references to {{R}} per WP:CITEVAR. I had never heard before that templates like {{Rp}} and {{R}} are deprecated, so you definitely need to do a better job of getting the word out about that, if it's true. If I had known about the deprecation, I would have done the opposite and changed the existing {{R}} templates in that article to {{Sfn}}. Biogeographist (talk) 02:27, 26 January 2026 (UTC)

I had in fact never heard of {{R}}, but it seems to be just a shortcut for <ref name="whatever" />? Re-using named references doesn't led to clutter and certainly won't be deprecated, so I cannot see a good reason to deprecate {{R}}. That said, whenever you need additional information (such as a specific page number), named references alone won't do the trick, and in such cases I'd always prefer and suggest {{sfn}} over ugly and unreadable {{rp}}. Gawaon (talk) 08:14, 26 January 2026 (UTC)

When I use {{r}} I use it only as a shortcut for named refs but it can also add page numbers like {{rp}}. In the unlikely case that we deprecate {{rp}} (which I do not like but do not think should be deprecated) then it is only that feature of {{r}} that would be concerned, not the whole template. But the argument for applying WP:PARREF makes almost as little sense as deprecating all footnotes because after all we format footnotes as numbers in brackets and WP:PARREF was about numbers in parens so it should apply equally well to footnote markers. (Of course it doesn't apply, neither to footnote markers nor to page numbers attached to footnote markers.) —David Eppstein (talk) 08:22, 26 January 2026 (UTC)

{{R}} has a "feature" to do what {{Rp}} does; so just that aspect of it should eventually be deprecated and then (after uses of that are replaced with something else) the "feature" removed.  — SMcCandlish ☏ ¢ 😼  09:28, 26 January 2026 (UTC)

Fortunately, {{R}} can be easily modified to use Sub-referencing once it's deployed, e.g. {{R|Foo|p=30|quote=This is a quote from page 30}} would just output <ref name="Foo" details="Page: 30&#10;Quotation&#58; &quot;This is a quote from page 30&quot;" /> instead of <ref name="Foo" /><sup><span title="Page: 30&#10;Quotation&#58; &quot;This is a quote from page 30&quot;">30</span></sup>. Hardest part would be getting consensus to do so. --Ahecht (TALK
PAGE) 16:13, 16 July 2026 (UTC)

I use {{rp}} because it is the closest to the inline citation style that I prefer (which looks something like [HW: 123]). This is standard citation style although it depends on the field. I believe they should not be deprecated. Rather, the subreferencing feature should provide options to support various well-established citation styles (including inline parenthetical citations, deprecated here but not everywhere) used and prefered by different groups of people. 慈居 (talk) 19:08, 19 February 2026 (UTC)

Support or oppose deprecation

Summary section to permit simple expressions of support or opposition without cluttering the debate above.

• Support per SMcCandlish's proposition and good typography. --𝕁𝕄𝔽 (talk) 10:56, 26 January 2026 (UTC)
• Support per SMcCandlish and because {{rp}} produces ugly clutter that reduces readability, especially with several page numbers, page ranges, or when there are two or more references in a row (all of which are fairly common). Gawaon (talk) 11:09, 26 January 2026 (UTC)
• Oppose until sub-referencing, which has a similar inline look as {{rp}}, as others have mentioned above, is implemented in English Wikipedia. Biogeographist (talk) 14:30, 26 January 2026 (UTC)
• Oppose The readability claim is a personal preference I do not share. The alternatives are not better or are worse. — Preceding unsigned comment added by Johnjbarton (talk • contribs) 17:21, 26 January 2026 (UTC)
• Oppose per Biogeographraphist and Johnjbarton. None of the alternatives currently available offer objective improvements.WP:CITEVAR makes it clear that preference for one form of citation style over another are principally subjective and that changes to an article's style may only be done based on consensus of the editors of the article concerned. Editors not supporting your personal preference are not "curmudgeons" as they have been labelled above any more than you are curmudgeonly for not supporting their equally valid preference. Thryduulf (talk) 17:34, 26 January 2026 (UTC)
• Oppose: Not only is sub-referencing not even available on enwiki yet, it doesn't solve one of the major problems with Harvard-style referencing (which rp does solve): Citing a large number of individual pages in one source produces a large stack of items in the references section. Until that problem is solved, rp is important to have. The arguments for depreciating rp seem to be based on personal preference; I'm particularly unconvinced by the readability/clutter concerns because sub-referencing will look almost the same. Pi.1415926535 (talk) 18:32, 26 January 2026 (UTC)

  We are currently investigating different ideas on how to address the issue of reference lists getting cluttered with lots of sub-references. Interestingly our user testing shows that most readers rely on the reference pop-up instead of using the reference list (which is in line with previous research that most readers don't even scroll to the end of the page), but we recognize the importance of for editors and for those readers who do care about working with the reference list. --Johannes Richter (WMDE) (talk) 19:19, 26 January 2026 (UTC)

  Which suggests that the reference number and page need not be shown for any citation format. A small mark for the pop-up would suffice for most readers.^* Johnjbarton (talk) 19:38, 26 January 2026 (UTC)

  Sub-referencing won't look almost the same in many cases, see the discussion above; and of course there are other alternatives with even better readability/less clutter, such as {{sfn}} and friends. Gawaon (talk) 21:53, 26 January 2026 (UTC)

• Oppose At least until the community has had a chance to use sub-referencing on a large scale. -- LCU ActivelyDisinterested «@» °∆t° 18:44, 26 January 2026 (UTC)
• Strong oppose. Sub-referencing does not deprecate the utility of rp as the page number is not shown in-text, and thus is a useless second click. The fact that shortened footnotes can do the same job as rp doesn't mean that one is redundant to the other: depending on the situation, either referencing style may be more useful. In terms of utility, it's easier to edit and maintain an article that reuses references and uses rp for page number than one that has shortened footnotes (besides the fact that sfns require two reference sections and thus needlessly take up article space.) Rp is also somewhat easier to use in the VE than sfns. Cremastra (talk · contribs) 23:21, 26 January 2026 (UTC)

  To add to this: shortened footnotes are little better than parenthetical citations squirrelled away into a footnote. This is needless obfuscation for the reader. Why not give them the whole footnote, as we have a very convenient format to do so with? Cremastra (talk · contribs) 00:23, 27 January 2026 (UTC)

  @Cremastra I hope you don't mind me asking, it might be relevant for future improvements to our feature: What do you mean with "second click"? Perhaps that's a misunderstanding, sub-referencing shows the page number in the same pop-up as the main information (example), making it accessible in-text without any extra click. --Johannes Richter (WMDE) (talk) 07:28, 27 January 2026 (UTC)

  Sorry, that was very badly worded. I meant one still has to hover over the footnote to get to the page number, as opposed to putting it in-text directly. Cremastra (talk · contribs) 12:59, 27 January 2026 (UTC)

• Support deprecation Even if sub-referencing never happens. {{rp}}'s cryptic numbers attached to the in-article superscript rather than being anywhere near the actual reference were a terrible idea from the start. Anomie⚔ 23:36, 26 January 2026 (UTC)
• Oppose per WP:NOTBURO and WP:CREEP. We do not need more and more rules about exactly how citations need to conform to some format that only bots are capable of formatting correctly. I don't happen to like {{rp}}-produced page number referencing but deprecation is going in the wrong direction. The argument that this will preempt the subreferencing format before we even have a chance to test it is also strong. —David Eppstein (talk) 23:52, 26 January 2026 (UTC)
• Support per SMcCandlish. Prefer short references. Ifly6 (talk) 23:55, 26 January 2026 (UTC)
• Oppose per Thryduulf and Cremastra. As a reader, shortened footnotes irritate me, so as an editor, I avoid them. I don't think rp is an ideal solution, but I prefer it to sfn both as an editor and as a reader. I'm looking forward to the implementation of subreferencing and seeing how that plays out. When I started editing, I found it frustrating that there are so many different ways to do things, but that's how it is and I don't find it likely that the project will ever land on One True Way for anything. As it is, those who prefer shortened footnotes use them to write their articles and those who prefer rp use it to write their articles and CITEVAR tells them to adjust when they work on an article using the other method. I don't see a widespread adoption of either method that would justify calling either the only method that should be used. Schazjmd (talk) 00:06, 27 January 2026 (UTC)

  I spent some time last year experimenting with different styles of citations (sfn vs rp, and template vs handwritten). Handwritten is not as laborious as it sounds and greatly improves the readability of the wikitext; the downsides, of course, are obvious. I wrote or expanded several articles with sfns, and while I don't hate them, I find them cumbersome to handle, especially in the VE which I use for article writing. I usually find myself defaulting to the rp system, as it's the most comfortable for me. Different strokes for different folks, I guess. Cremastra (talk · contribs) 00:26, 27 January 2026 (UTC)

• Oppose - Rp does get annoying, especially when working with page ranges, but I disagree that there's any obviously better alternative right now. Would be happy to reconsider when sub-referencing is implemented. For my part, as a reader I've always found sfn more reader unfriendly than rp. 95% of the time when I click or hover over a citation, I want to see what the citation is, and 5% of the time I'm actually following it through to where I'd need a page number. But most of the time, Sfn just takes me to a last name, year, and page number, then I have to scroll around and cross-reference that with a separate list before I can even see what the source is. It's, to me, the worst of all the systems (other than the non hyperlinked ones, I guess), but I digress. I'd have to see the discussion deprecating parenthetical references, but the argument above that Rp is already deprecated just sounds like wikilawyering. I wonder how many people, when they hear "parenthetical references", imagine something other than a reference in parentheses. Indeed, "regarding something contained in parentheses" is a common definition of "parenthetical", so just thoroughly unpersuaded on that end, without having a link to the actual discussion. — Rhododendrites ^talk \\ 00:25, 27 January 2026 (UTC)
• Support, for reasons I've given in great detail above. It also means, necessarily, deprecating essentially identical functionality in {{R}} and any other template.  — SMcCandlish ☏ ¢ 😼  01:08, 27 January 2026 (UTC)
• Oppose per Cremastra and Rhododendrites. Nikkimaria (talk) 01:07, 27 January 2026 (UTC)
• Oppose, Schazjmd and Rhododendrites have explained how the rp system while imperfect is more helpful for readers than some other systems, others have also made good points I won't repeat. Surprised it has to be said that the consensus on parenthetical referencing applies to parenthetical referencing, and is not about things that are not parenthetical referencing. CMD (talk) 01:49, 27 January 2026 (UTC)
• Oppose, at least until sub-referencing becomes available. I think sub-referencing has significant advantages, but there are situations in which citing numerous different pages from a single book would result in an unwieldy number of sub-references. Boghog (talk) 09:50, 27 January 2026 (UTC)

  Indeed, we need to see how subreferencing copes with situations like List of lakes of Yukon where circa 70 different pages of one source are used (and several other sources are used over 20 times each) a significant number of times) before we can see whether it can replace {{rp}}. Short footnotes are terrible in these situations though so it's unlikely to be the worst form of referencing available. Thryduulf (talk) 04:24, 29 January 2026 (UTC)

  Lists are indeed a use case where we frequently observe a higher number of sub-references than in regular articles (e.g. de:Liste der Straßen in Bad Honnef#Einzelnachweise und Anmerkungen). We've received a couple of suggestions how to deal with that, happy to hear your thoughts as well! We are currently looking at different ideas and will ask for community feedback once we've completed our initial investigation. Johannes Richter (WMDE) (talk) 09:19, 29 January 2026 (UTC)

  Maybe such sublists could be automatically rendered in multicolumn format on wider screens? In that way they wouldn't need to much vertical space. Gawaon (talk) 09:24, 29 January 2026 (UTC)

  That's one of the ideas we're investigating, but it probably just works for page numbers and is less suitable if sub-references include more information, e.g. de:Ophiuride#Einzelnachweise. Johannes Richter (WMDE) (talk) 12:14, 29 January 2026 (UTC)

  (edit conflict) I agree with the multicolumn suggestion. Also, if possible, when a reference is clicked on the actual reference details should be at the top of the screen with the specific page continuing to be highlighted. At present if I click e.g. de:Liste_der_Straßen_in_Bad_Honnef#cite_note-236 I can tell that the reference is to page 88 of something, but I have to scroll upwards to find what. Thryduulf (talk) 12:14, 29 January 2026 (UTC)

  Thanks for your feedback! That's another suggestion we're considering. Other ideas included collapsing sub-references in the reference list, only showing a few of them (including the one a reader clicked on) if a single source has been used a lot of times. We're doing UX research on articles with lot's of sub-references and will ask for community feedback once we're closer to proposing a solution. Johannes Richter (WMDE) (talk) 13:13, 29 January 2026 (UTC)

  You may want to review MOS:COLLAPSE before suggesting that here. Anomie⚔ 00:51, 30 January 2026 (UTC)

• Oppose until the sub-referencing feature becomes widely available.Ganmatthew (talk • contribs) 17:50, 27 January 2026 (UTC)
• Oppose on the grounds that it's instruction creep, that it would rule out a working system that is arguably the best fit for some circumstances, and that the "it's already deprecated" argument is wiki-lawyering in the face of the facts. The consensus of the 2020 discussion that deprecated parenthetical referencing says nothing about {{rp}}. The entire discussion mentions the {{rp}} template all of twice, with one editor liking it and another not (and both of those editors supported the deprecation proposal). Nothing in the proposal wording itself implies that "parenthetical" is to be interpreted in a broad way. It means in parentheses, not secondary, auxiliary, ancillary, etc. Stepwise Continuous Dysfunction (talk) 02:43, 28 January 2026 (UTC)
• Support, {{rp}} is an absolute blight on Wikipedia. Headbomb {t · c · p · b} 07:46, 28 January 2026 (UTC)
• Oppose. If anything, {{sfn}} should be deprecated as a travesty upon editing and readers eyes. Katzrockso (talk) 04:02, 29 January 2026 (UTC)
• Oppose as I personally find it a better reader experience than {{sfn}}, as Rhododendrites explains quite nicely. It's also more straightforward to manage as an editor if there are only a few sources that are cited multiple times with different page numbers; in these cases {{sfn}} becomes overkill.

I also agree that the 2020 discussion does not concern {{rp}}, given that it is scantly mentioned in the actual content of the discussion itself; any attempt to construe it as such is a stretch IMO.  novov talk edits 06:29, 29 January 2026 (UTC)

• Oppose, it is a stretch to construe the 2020 discussion as deprecating {{rp}}, and we should wait for sub-referencing to be established before we can take more definitive decisions on this topic. Chaotic Enby (talk · contribs) 15:42, 29 January 2026 (UTC)

  Additionally, given the impact of the change, I wonder if it may be relevant to make this into a RfC and put a notice on WP:CENT. Chaotic Enby (talk · contribs) 15:49, 29 January 2026 (UTC)

  That sounds like a good idea. Stepwise Continuous Dysfunction (talk) 20:31, 30 January 2026 (UTC)

• Support once sub-referencing is available. For this to work, it'll need an example article converted over from superscript page numbers to sub-reference page numbers. The {{rp}} template is widely used, so it'll need a proper RfC to deprecate it.
  Also, that RfC should probably mention the widely used shorthand template, {{r}} (which is used across 34,000 pages to create over 40,000 superscript page numbers).
  Also also, it should probably deal with the concept of superscript page numbers. Thousands more pages use handwritten superscript page numbers following references, like Madrid § Tourism, for example. There are also niche templates creating these, like {{ran}} and {{listref}}, which were both made to solve the issue created on pages where named references would create dozens and dozens of backlinks, making the references unreadable.
  Above, I see some debate about whether {{rp}} was deprecated when parenthetical referencing was deprecated, to avoid confusion this time we should try to be clear with an explicit RfC about this style of citation. Rjjⁱⁱⁱ (talk) 03:12, 30 January 2026 (UTC)

  It is false that {{r}} is merely a shorthand for {{rp}} and false that it's primary use is to create superscript page numbers. Mostly, {{r|something|another}} is a shorthand for <ref name=something/><ref name=another/>. It can also add page numbers but in my experience usually does not. The same statistics page you list shows that maybe only one in seven uses of {{r}} include page numbers. Please do not confidently advocate deprecation of something you obviously are unfamiliar with. —David Eppstein (talk) 06:09, 30 January 2026 (UTC)

  From what I understand of @Rjjiii's comment, they weren't saying that {{r}} was a shorthand of {{rp}} specifically, or that this was its primary use. In fact, as you've pointed out, that same page shows {{r}} being used 284,000 times, including for other purposes. Maybe I misread their comment, but it doesn't seem like they were making a claim that strong. Chaotic Enby (talk · contribs) 06:15, 30 January 2026 (UTC)

  @David Eppstein: It is false that {{r}} is merely a shorthand for {{rp}} and false that it's primary use is to create superscript page numbers." I agree? Click the "over 40,000" link. The |page= parameter is used over 40,000 times. It doesn't really make sense to deprecate {{rp}} and leave the same functionality in other templates. "The same statistics page you list shows that maybe only one in seven uses of include page numbers." 41,183 / 284,578 is one in seven, but it's also thousands of footnotes. Can {{r}} create sub-references instead? If not, what should be done with the |page= parameter? Do you not think those are relevant questions here? Rjjⁱⁱⁱ (talk) 06:28, 30 January 2026 (UTC)

  I don't know whether r can be made to create subreferences because we do not yet have subreferences. Until we have enough experience with them to consider questions like converting r+pages to use them I think this RFC is premature. —David Eppstein (talk) 06:46, 30 January 2026 (UTC)

• Oppose because alternative formatting sacrifices salience—you get a context-free stack of page numbers when perusing the reference list. (It is almost never important to see a list evidencing the fact that pages 1, 2, 3, 5, 7 and 9 are cited. It's more important, though only when reading deeply and critically, to see which page is cited for which proposition, as you read those propositions.) Maybe a plurality of readers don't care about page numbers and only want to see them after a click; that could be a user preference, and even a default. But I certainly dislike a click or mouseover that delivers me to an incomplete citation or takes me away from the sentence I just read. TheFeds 08:42, 31 January 2026 (UTC)
• Oppose. Subreferencing will not provide the exact style that rp does. Both subreferencing and rp are only close to a standard inline citation style that is my favorite ([AB2: 123, §1.2]). I am still going to prefer rp over subreferencing – the more you write at the bottom the more often you will have to go to the bottom (either directly or using reference preview). 慈居 (talk) 18:14, 19 February 2026 (UTC)
• Support. For example, This parameter is very confusing for non-editor readers. What are they supposed to understand from it? A math equation? What does 1/22 even mean? IdanST (talk) 12:27, 23 February 2026 (UTC)

  Superscript numbers in the first place are confusing if you haven't seen them before. Cremastra (talk · contribs) 16:25, 23 February 2026 (UTC)

  And if you hate the colon in superscript numbers, you'll probably hate the period in the forthcoming sub-referencing feature too.^[1.22] Biogeographist (talk) 17:04, 23 February 2026 (UTC)

  Where are you seeing "1/22"? "[1]:22" looks nothing like a math equation. And hovering over it will reveal "Page: 22". --Ahecht (TALK
  PAGE) 16:43, 23 April 2026 (UTC)

  1:22 is a common division notation in non-English languages, e.g. de:Division (Mathematik)#Schreibweisen, it:Divisione (matematica)#Descrizione, fr:Division#Vocabulaire et notations. Historique. Johannnes89 (talk) 09:25, 19 July 2026 (UTC)

• Strong oppose, {{rp}} makes it easier to do WP:V checks and the reference section is tidier. It's also simpler. We should not force everyone to use {{sfn}} or put page numbers in the ref. How about we let people have preferences? Kowal2701 (talk, contribs) 09:59, 26 March 2026 (UTC)
• Oppose until sub-referencing is implemented per Biogeographist. --Ahecht (TALK
  PAGE) 16:44, 23 April 2026 (UTC)
• Very weak support. Only if it's handled similarly to the way WP:PAREN was handled. No active projects, no bots, no unsolicited edits with the sole purpose of removing them, no judgement of people who still use them anyway. Let local editors do what they want. Don't make it a thing. Just looked it up -- {{rp}}, {{r}}, {{rma}} and {{rmn}} are used in 100K+ articles. They'll probably never disappear, so don't get your hopes up. (I personally think they are inscrutable, ugly, idiosyncratic and unnecessary. Use short cites. They do the same thing, but are clear, readable, familiar and fully automated. ---- CharlesTGillingham (talk) 19:37, 27 June 2026 (UTC)

  Short cites (sfn) are horrible, jumpy, confusing and difficult to edit. Deprecate them. Johnjbarton (talk) 19:01, 16 July 2026 (UTC)

Citing a US Trademark?

What's the best way to cite a US Trademark. Right now in Big Duck I've just got U.S. Trademark 71,318,066 but my reviewer asks if there's something more that should be included and I don't know if there is or not. RoySmith (talk) 23:39, 5 May 2026 (UTC)

It's possible that certain external style guides would include more information (Maybe the registrant's name? Or a date?), but we don't have any rules like that. WhatamIdoing (talk) 05:14, 27 May 2026 (UTC)

Conflicting information

Two sets of sources give different information. Much of what is going on is discussed here. I believe the newspaper reporters and those who wrote the headlines sincerely believed what they were saying was true. I may be asking in the wrong place, but what is the proper way to handle this?— Vchimpanzee • talk • contributions • 21:46, 11 May 2026 (UTC)

If there is conflicting information in equally reliable sources, then that should be discussed in the article. If news reporting is shown to be wrong by later reporting or printed works, then only use the latter. -- LCU ActivelyDisinterested «@» °∆t° 12:49, 12 May 2026 (UTC)

ActivelyDisinterested I don't know if there were any corrections later, but is what I did in The Arras all right?— Vchimpanzee • talk • contributions • 22:02, 13 May 2026 (UTC)

I can't very well throw out sources where the error is in the headline. Have I done enough to show what is going on in The Arras?— Vchimpanzee • talk • contributions • 20:51, 12 May 2026 (UTC)

For newspaper articles, headlines often aren't written by the article's author but by someone else, so it's best to disregard any information or errors in them. Look at the body of the article instead. Gawaon (talk) 20:56, 12 May 2026 (UTC)

Anyway, I would appreciate anyone looking to see if I have handled the situation correctly.— Vchimpanzee • talk • contributions • 21:56, 12 May 2026 (UTC)

I guess it's true on Wikipedia that if no one objects, you've done it right.— Vchimpanzee • talk • contributions • 17:43, 15 May 2026 (UTC)

See WP:HEADLINES: "News headlines—including subheadlines—are not a reliable source." If "the error is in the headline", then ignore the headline. WhatamIdoing (talk) 05:15, 27 May 2026 (UTC)

Minor grammatical error in Variation in citation methods

The second paragraph under Variation in citation methods has the sentence:

An article where all or most of the citations fail to provide needed bibliographic data does not have a consistent citation style and can be changed freely to insert such data.

I almost moved the position of the "and" resulting in:

An article where all or most of the citations fail to provide needed bibliographic data and does not have a consistent citation style and can be changed freely to insert such data.

Should I just go ahead with this change or am I missing something? My Gussie (talk) 12:48, 14 May 2026 (UTC)

It is correct as written. Parse it as two sentences: An article where all or most of the citations fail to provide needed bibliographic data does not have a consistent citation style. Such citations can be changed freely to insert such data. DrKay (talk) 12:55, 14 May 2026 (UTC)

Gotcha. I'm therefore thinking that the only improvement would be to just add a comma just before the and. Does this significantly improve readability? My Gussie (talk) 17:31, 14 May 2026 (UTC)

I'm going ahead with adding the comma. My Gussie (talk) 15:00, 15 May 2026 (UTC)

Thank you -- I agree that the comma improves readability. But I think DrKay's two-sentence version is even better. --DavidCary (talk) 18:06, 4 June 2026 (UTC)

Alternative to "page number" for ebooks should be added

Maybe it is somewhere else? But this would be the place to find a short how-to answer to this. Quite urgently needed. Jp1008 (talk) 04:47, 27 May 2026 (UTC)

You can use chapter= or at= for a specific location. PARAKANYAA (talk) 05:07, 27 May 2026 (UTC)

I've added this to the WP:EBOOK section. @Jp1008, did you find the EBOOK section (under "Identifying parts of a source" > "Books and print articles" in this guideline). WhatamIdoing (talk) 05:20, 27 May 2026 (UTC)

If you need inspiration for the help page being more useful (I can experiment), adding some examples would be nice, in particular, what are the suggested best practices to follow the "at=". Should I put in in quotes? Paragraph number? Referencing is important and the easier we make it for newbies, the better. eBooks are now the most common way to read books.
Jp1008 (talk) 21:18, 27 May 2026 (UTC)

Amaze, amaze! Thank you Jp1008 (talk) 21:08, 27 May 2026 (UTC)

Citing newspaper pages without lettered sections

This is an issue when citing old newspapers.

Most newspapers I have seen split their sections up by letter, e.g. page A5, or B10, sometimes they will do A-5 or 10-B but the principle is the same. I have also seen sections numbered separately so it's like 1-5 or 3-2 or something.

However, some newspapers would do it where they would totally restart their numbering in every section, and then not indicate this in the page number. So there would be the main section, it goes say pages 1 to page 30, then it would get to the lifestyle section, it would be pages 1 to 20, but there would be no separate indication in the page number. When citing such papers I have always just written the page number in the page= parameter and not indicated the section at all. But this seems problematic to me, as if I was checking such a reference, I would not know what section it was in. Any ideas?

Sorry if this isn't clear. PARAKANYAA (talk) 21:23, 27 May 2026 (UTC)

I'd suggest writing it as "p. 14 (lifestyle section)" or "lifestyle section, p. 14". Gawaon (talk) 07:08, 28 May 2026 (UTC)

@PARAKANYAA, in Help:CS1 templates, use |department= to indicate the section. Izno (talk) 16:58, 6 July 2026 (UTC)

Making clear that an old source is being cited

In Atlantic slave trade#British abolitionism we have a reference Williams 2021, pp. 105–106, 120–122. If the reader hovers over the reference, they see a date of 2021. But the reference applies to Williams, Eric (2021) [1944]. Capitalism and Slavery (Third ed.). Chapel Hill: University of North Carolina Press. ISBN 978-1-4696-6369-2., the key date of which is "orig-date" which is 1944. Therefore we have a situation that completely misleads the reader by suggesting that a recent source is being quoted, when in fact it is more than 80 years old. Whilst, within the subject, historians still argue about the relevance of Williams' opinions, (and the result of that argument wavers about over time) it still seems wrong that Wikipedia suggests that the origin of this idea is recent.

Is there a better way to inform the reader that these ideas go back to 1944? Is the reference as it exists in the article correct and, if so, is it right for a reference that "follows the rules" to mislead in this way.

(This question also applies to situations where editors cite old books on seamanship which have appeared in reprinted form, when there are modern works that analyse all the old primary sources and therefore are better sources. Unfortunately the primary sources tend to be free on the internet or are relatively cheap, whilst editors would either have to buy a recent book or go and visit a library.) ThoughtIdRetired _TIR 12:48, 4 June 2026 (UTC)

• There is no single “correct” answer to dealing with “old” sources. While many will be outdated, others are will be considered seminal works and are still well respected. Take it on a case by case basis. That said, I agree that our citations should reflect when the source was first published, as well as the publication date of whichever subsequent edition/reprinting we are citing. Blueboar (talk) 13:07, 4 June 2026 (UTC)

In this case, the Third Edition was published by the University of North Carolina Press in 2021. So the citation is correct. I think any explanation about the age of the underlying source should be made in the article text. Jc3s5h (talk) 13:35, 4 June 2026 (UTC)

I suppose the important question here is whether there are significant differences between the 1944 first edition and the 2021 third edition? If so, then the 2021 date is more important than the 1944 date. If not, then 1944 is more important. Blueboar (talk) 16:31, 4 June 2026 (UTC)

You always cite the edition you have, for verifiability reasons. PARAKANYAA (talk) 16:46, 4 June 2026 (UTC)

Old sources are outdated by new sources, not just by the passage of time. An old source isn't immediately an issue, although many have to be handled with care, and sometimes they're the most appropriate source. Absolutely the use of old sources are because they're usually free, but their not necessarily primary sources. It would be great if editors all had access to new works but that's not always possible. Although if you ask nicely the WMF may buy them for you (WP:Resource support pilot). -- LCU ActivelyDisinterested «@» °∆t° 17:05, 4 June 2026 (UTC)

Yes, I get the point that the principle of verifiability dictates citing the edition to hand. What I am introducing here is the incorrect impression that the reader may take away that this is a recent work. Later editions, I understand, continue with the two central tenets of the original work, that slavery produced a significantly large part of Britain's national income and that falling profits from sugar plantations drove emancipation. A quick read suggests that these are new ideas, when they were formulated with much less data than is available to modern historians. So the price of ease of verifiability is that the reader does not easily learn the date of the origin of these ideas. ThoughtIdRetired _TIR 17:15, 4 June 2026 (UTC)

Few readers will bother checking citation details anyway, so even if there were something like an "original-date" field, it would likely have little effect. Also, even newer books can spread older ideas 😉. Generally, I'd consider some in-text attribution, such as "in his 1944 classic, Williams said", more useful in such cases. Gawaon (talk) 17:22, 4 June 2026 (UTC)

If it's any consolation, data shows that almost no readers check the citations. PARAKANYAA (talk) 19:43, 4 June 2026 (UTC)

To be clear, Williams' ideas are highly relevant to the modern debate. What the reader should learn is that they have been around for 82 years. ThoughtIdRetired _TIR 17:18, 4 June 2026 (UTC)

If there are new sources that challenge or clarify Williams ideas those should be put in the article. If his ideas are unchallenged then the age of his work isn't relevant. If someone wrote the definitive answer 75 years ago then modern historians may find no need to create new works. -- LCU ActivelyDisinterested «@» °∆t° 19:08, 4 June 2026 (UTC)

Here is a best-practices hack that will do you what you want:

"Voilà"

• Voltaire (2021) [1752]. Le Micromégas.

Look at the source -- I've expanded {{sfn}} and used {{harvp}} so I could put "[1752]" into the short citation as additional citation text.

I left a note at Module:Footnotes to handle this better in the future, if you're interested. ---- CharlesTGillingham (talk) 21:24, 27 June 2026 (UTC)

Does using any shortened footnotes mean the entire article, even non paginated sources like websites, has to use sfns?

@Sirfurboy Seems to think that because any reference (or, most references) in an article uses an sfn, all references on the article must be sfns, including non-paginated sources like websites, or it violates the "consistency" rule of this guideline. I do not agree, as consistency does not mean every class of reference must be treated identifically, like how we have cite web, cite av media, etc. It is regular practice on hundreds of FAs to do it this way, with sfns on paginated refs or sources with multiple page, and there is no line in this guideline that says this is inappropriate. For examples of articles that use this style from the TFAs scheduled for this month, see Voss (collection) and Early life and education of Donald Trump. PARAKANYAA (talk) 23:59, 4 June 2026 (UTC)

There is no such requirement. Nikkimaria (talk) 00:03, 5 June 2026 (UTC)

I understand that WP:CITEVAR mandates that articles properly tagged with {{Use shortened footnotes}} or {{Use list-defined references}} must not have their citation style changed. -- Michael Bednarek (talk) 03:40, 5 June 2026 (UTC)

Even then, "shortened footnotes" are only relevant in cases where the same source is cited again but with some details (usually the cited page or section) changed. For that, sfn and friends are great. If the same source is cited just once, the issue of shortening never arises; and if it's cited repeatedly but without changed details, we would just use a named reference invoked several times. Gawaon (talk) 06:48, 5 June 2026 (UTC)

I repeat, articles properly tagged with {{Use shortened footnotes}}, which means "exclusively". -- Michael Bednarek (talk) 07:10, 5 June 2026 (UTC)

I guess so, though that seems practical only for articles that mostly cite print publications, and fairly impractical for references to web pages and similar sources. But that would have to be discussed on the article's talk page in case of disagreements. Gawaon (talk) 07:29, 5 June 2026 (UTC)

Yes, those that use it exclusively. However I note that even that template's description lists this as an option: It should not be applied to articles that use a mix of standard and shortened footnotes unless there is a consensus on the article's talk page that the article should be converted to shortened footnotes with a separate list of full citations. PARAKANYAA (talk) 11:49, 5 June 2026 (UTC)

The page in question is Christian Identity. There are four references that PARAKANYAA reverted from my attempts to match the cite style: (Comparet, 2007), (Downey 2013), (Fink, 2020) and (Wexler et al., 2024). Note that each of these has an author and Wexler et al. has page numbers. Now Help:Shortened footnotes is clear that this cite style implements a hybrid with Harvard style referencing with 4 rationales for doing so. The first rationale does not apply for the three references that have no page numbers, but not using Sfn for these subverts the other three reasons for using Sfn. In particular, these references are not now in the alphabetised bibliography. We have a Harvard style bibliography that is now inconsistent in that four of the named references (a vanishing minority of all of them) are not in the Harvard style bibliography but perversely in the footnote section. I am not aware of any guideline that says we should entertain inconsistent citation styles merely over whether a source has page numbers. Is there one? The article in question is looking at the history and development of a movement, largely relying on academic sourcing. The references in question are named references. What is the reason for reverting to an inconsistent cite style? Sirfurboy🏄 (talk) 08:06, 5 June 2026 (UTC)

Sfns do not mandate adhering to every single aspect of Harvard style and there's absolutely no reason to suggest that is the case. Help:Shortened footnotes is not a guideline and it is lists some opinions on why one may or may not use sfns, but some thoughts on one may or may not choose to use them. At no point at that help page does it says every citation in a page with sfns must use all sfns. There is nothing inconsistent about using sfns with paginated sources and long refs with non-paginated sources, if the style is to do that, which it has been for years. PARAKANYAA (talk) 11:48, 5 June 2026 (UTC)

WP:CITEVARYES has:

The following are standard practice:

...

imposing one style on an article with inconsistent citation styles (e.g., some of the citations in footnotes and others as parenthetical references): an improvement because it makes the citations easier to understand and edit;

I still don't understand why you would want some of the unpaginated references to be inconsistent with the other references, and I still don't see reference to any guideline that we should do this. Sirfurboy🏄 (talk) 12:50, 5 June 2026 (UTC)

We are not doing some citations in footnotes and others as parenthetical references, but a consistent style. Multiple other editors here, and the template:Use shortened footnotes, argue that this is an accepted style: It should not be applied to articles that use a mix of standard and shortened footnotes unless there is a consensus on the article's talk page that the article should be converted to shortened footnotes with a separate list of full citations.

Where is any guideline saying this isn't fine? And the answer about why one would want this is that there is no reason to make a footnote short in this context because it makes you require an extra click to require the exact same amount of information for 0 reason. PARAKANYAA (talk) 12:53, 5 June 2026 (UTC)

Who, other than you, is arguing that Christian Identity, as it is right now, in this version, is using a consistent style? Sirfurboy🏄 (talk) 12:56, 5 June 2026 (UTC)

I am saying that it is a consistent style to use a mix of sfns and long references, which you dispute, and that the prohibition on "citations in footnotes and others as parenthetical references" does not equate to "if there is one sfn in an article every single citation must use sfns". The status quo on the page is a mess and the only consistent style it ever used was all long references. PARAKANYAA (talk) 13:02, 5 June 2026 (UTC)

Short form references can be, and regularly are, used for all types of sources including non-paginated sources. If the article style is to use short form references then converting a new references to match that style would be correct per CITEVAR. If the style has always been mixed, with only paginated sources using short form references, then that article style should be followed. CITEVAR says to follow the articles style and discuss any changes on the articles talk page. -- LCU ActivelyDisinterested «@» °∆t° 08:42, 5 June 2026 (UTC)

The last revision of Christian Identity before the most recent spate of edits appear to be 16 April. At that point there were definitely non-paginated sources using short form references, "ADL 2017" as an example, but looking through the articles history it doesn't appear that at any point did all non-paginated sources use short form references. So I don't think anyone can claim a consistent style. -- LCU ActivelyDisinterested «@» °∆t° 08:58, 5 June 2026 (UTC)

That version had 108 citations, all but 4 short. ISTM that the article's established citation style is short, and converting the few outliers is fully conforming with CITEVAR. -- Michael Bednarek (talk) 11:27, 5 June 2026 (UTC)

At no point did the non-paginated sources ever use short references, so it is not the established style to do so. PARAKANYAA (talk) 11:44, 5 June 2026 (UTC)

Incorrect. The short references: ADL 2017, Southern Poverty Law Center, James 2012, The Joplin Globe 2001, middlebury.edu 2024, The Los Angeles Times 1999, The News and Advance 1991, and Lloyd 1995 are all unpaginated. The reversion is inconsistent with these. Sirfurboy🏄 (talk) 12:11, 5 June 2026 (UTC)

Huh, I agree that is inconsistent. The last consistent style the article used was long footnotes entirely . The article has never used short footnotes entirely. Should we change back to that? PARAKANYAA (talk) 13:01, 5 June 2026 (UTC)

Right, so having established the inconsistency, we can make it consistent by converting 4 references to Sfn, or converting 105 references to long footnotes. I would suggest we go with Sfn now, but as I suggested yesterday on the article talk page, we could have a fuller discussion about which style we should be using. Note that the style was only long form in that diff because Butlerblog converted all the paranthetical references to that form in 2021, and the page historically had used parenthetical references. The same editor then converted all references to Sfn a few years later, correctly noting that paranthetical referencing had originally been there. We should not be agreeing a new style here, but on the article talk page, and with Butlerblog taking part. I'll note that my own preference would actually be harvnb, which approximates, in a web friendly manner, the footnotes‑and‑bibliography styles that you would generally find in a history text. Because the shorter reference is wrapped in a reference tag, the footnote can be annotated, whilst allowing for a clear alphabetised bibliography. But that is a discussion for the talk page. Sirfurboy🏄 (talk) 13:23, 5 June 2026 (UTC)

I don't want to derail the above discussion by getting too specific about the Christian Identity article, so I'll limit it to say I'm happy to engage in article specific discussion at that article's TP. But specific to this discussion, I also prefer to see consistency, and obviously I lean to {{sfn}} (and admittedly I could be swayed to {{harvnb}}). I think I understand what @PARAKANYAA is saying, and I don't necessarily disagree with the premise entirely, but from an assessment point of view, it's better to have it one way or the other. There are always exceptions to that, so it just needs to be considered on a case-by-case basis and by consensus. ButlerBlog (talk) 14:33, 5 June 2026 (UTC)

I won't dispute not using it on this article if there is consensus for that (though I feel like if we're going to do that, we need to split the bibliography by type for readability), but the principle that the mix of standard and shortened footnotes depending on the type of reference is inappropriate or inconsistent in any article I fully disagree with. PARAKANYAA (talk) 15:33, 5 June 2026 (UTC)

Or, we can move the handful of short references without pages to be long refs, and have a style that is consistent? If we're going to be quoting documentation pages, It should not be applied to articles that use a mix of standard and shortened footnotes unless there is a consensus on the article's talk page that the article should be converted to shortened footnotes with a separate list of full citations. Adding shortened footnotes to sources with no pages creates a completely useless hassle if you want to access the references, that is why I oppose it. But if we're doing that on this one page per consensus, then ok, that is the style, but as Nikkimaria and ActivelyDisinterested said there is no issue with having mixing them as the style if that is what is chosen. PARAKANYAA (talk) 15:16, 5 June 2026 (UTC)

I agree; consistency can be achieved in many different ways – there is no single "correct" answer. Sfn can be used for newspapers and similar unpaginated sources, but it's often more hassle than it's worth. Gawaon (talk) 20:21, 5 June 2026 (UTC)

Parenthetical claims

I didn't find an example for this issue. Where should the citation be placed for a parenthetical claim? For example:

• Something fascinating is discussed in this sentence. (Now we have an aside.¹) Then the encyclopedia continued.

or

• Something fascinating is discussed in this sentence. (Now we have an aside.)¹ Then the encyclopedia continued.

To me the citation should be inside the parenthesis, grouped with the sentence. Johnjbarton (talk) 16:53, 19 June 2026 (UTC)

Inside the parens I'd say. I think I did it that way on one article but later broke it out for tidyness. — Very Polite Person (talk/contribs) 16:56, 19 June 2026 (UTC)

I'd place it after the parentheses, since citations are generally placed after closing punctuation, and that includes the closing parenthesis. But it's best to avoid full-sentence or multi-sentence parentheses altogether – if it doesn't fit into the main text, either omit it or make it an explanatory note instead. Gawaon (talk) 02:44, 20 June 2026 (UTC)

@Johnjbarton see WP:REFPUNCT. Izno (talk) 17:21, 6 July 2026 (UTC)

Thanks that is what I was looking for. Johnjbarton (talk) 22:02, 6 July 2026 (UTC)

Citing a book available on archive.org

I thought there was some way to distinguish books available via archive.org, such as

• King-Hele, D. (1964). Theory of Satellite Orbits in an Atmosphere. United Kingdom: Butterworths.

However I found no information about this on in the article. Hints? Johnjbarton (talk) 20:46, 25 June 2026 (UTC)

{{cite book
 |last=King-Hele
 |first=D.
 |date=1964
 |title=Theory of Satellite Orbits in an Atmosphere
 |publisher=Butterworths
 |location=London
 |page=6
 |url=https://archive.org/details/theoryofsatellit0000king/page/6/mode/2up
 |url-access=registration
 |via=Archive.org
}}

King-Hele, D. (1964). Theory of Satellite Orbits in an Atmosphere. London: Butterworths. p. 6 – via Archive.org.

—Trappist the monk (talk) 21:17, 25 June 2026 (UTC)

Check out how these are setup between Archive and Newspapers (list defined refs, just edit on References section): Amanda_Jones_(inventor)#References — Very Polite Person (talk/contribs) 01:14, 26 June 2026 (UTC)

WP:PAREN details

Is there a guideline or something that works out the details, e.g. What cases, exactly, has the community deprecated? How exactly should editors fix the parenthetical citations they find? etc. ---- CharlesTGillingham (talk) 11:00, 26 June 2026 (UTC)

• According to the section as written, it is deprecated in running text (i.e., article lead and body) and only when the format is "Statement of fact (Mr. Knowitall 2012: 20)". The recommended action is to use a ref-tag based citation style. I don't think there's anything spelling it out more explicitly than what is already here. — Chris Woodrich (talk) 11:46, 26 June 2026 (UTC)
• Anything that cites the source in text was deprecated. So as well as the example mentioned by Chris Woodrich, another example would be:
  "Smith 2020 says that ..."
  that should become something like;
  "In 2020 Smith said ..."
  I don't think there any exact guidance on any of this, only the understanding that if you cite something it should be inside ref tags. -- LCU ActivelyDisinterested «@» °∆t° 14:27, 26 June 2026 (UTC)

I'm satisfied there is no guide except this one. See my next topic.----CharlesTGillingham (talk) 15:06, 27 June 2026 (UTC)

References

Smith, John (2020). example.

Parenthetical references in explanatory notes

EDIT: Please scroll down to reboot below. ---- CharlesTGillingham (talk) 04:28, 28 June 2026 (UTC)

Consider these cases.

Y The sun is pretty big.. Voltaire points out that size is relative; to the giants of Sirius the sun is quite small.

Y The sun is pretty big.. Voltaire (1752) points out that size is relative; to the giants of Sirius the sun is quite small.

N The sun is pretty big.. Voltaire points out that size is relative; to the giants of Sirius the sun is quite small. (Voltaire 1752)

Y The sun is pretty big.

Y The sun is pretty big.

? The sun is pretty big.

• Sagan, Carl (1986). Size of the sun.
• Voltaire (1752). Le Micromégas.

I would argue that the last case is deprecated under WP:PAREN (or should be). I interpret the rule as this: Parenthetical references are banned everywhere, in-text attributions are allowed everywhere.

There seems to be some confusion about this, as I've seen different assertions made by different editors.

Does you all agree with me? Why or why not? ---- CharlesTGillingham (talk) 16:13, 27 June 2026 (UTC)

Parenthetical references are not banned and not everywhere. They are deprecated (significantly weaker than banned), in article text only. It is completely standard to have parenthetical references in footnotes with longer bibliography items later. There is absolutely no requirement for "explanatory" footnotes to be separated out from other footnotes and treated separately. If we make footnotes with <ref> it doesn't even work to put more footnotes inside them, so a footnote-in-footnote referencing style is impossible; footnotes in footnotes can be done with some templates but that way lies madness . If a footnote has a longer explanation, it can still mix with a parenthetical reference. —David Eppstein (talk) 16:34, 27 June 2026 (UTC)

I'm only talking about deprecation (not "banning"). The example above demonstrates that there is a system in place to put citation footnotes in explanatory footnotes (used 300K+ articles). I am not proposing deprecating WP:short citations (used ~500K articles). The examples above have short citations (that was deliberate). Short cites are not parenthetical references. For one thing, they normally don't have parenthesis (as shown above). ---- CharlesTGillingham (talk) 18:41, 27 June 2026 (UTC)

Save us from gnomes who think that rule-following and rule-enforcing is more important than being able to include explanatory text in reference footnotes without levels on levels on levels of indirection. —David Eppstein (talk) 18:56, 27 June 2026 (UTC)

Hey, I'm just trying to update the document pages for {{harv}} and company and I don't know what to write. They made the rule, I don't what it is. I'm just asking what they think the rule is. I'm looking for an answer: is this deprecated or not? ---- CharlesTGillingham (talk) 21:29, 27 June 2026 (UTC)

Any of the note versions would be acceptable. However your second example isn't, it's exactly what PERAN was all about. -- LCU ActivelyDisinterested «@» °∆t° 21:31, 27 June 2026 (UTC)

<retracted -- I'm working on it>----CharlesTGillingham (talk) 22:05, 27 June 2026 (UTC)

That you for a clear answer on example 6.

Example two is problematic, and I'd be inclined to agree with you. However, the consensus at village pump was to ban "parenthetical style citations". According to Wikipedia, parenthetical referencing looks like this: "(Smith 2008, p. 1) or (Smith 2008:1)" (Harvard) or like this "(Smith Smith's Book 1) (MLA). It's arguable whether that applies to in-text attribution.

(And David, sorry I snipped at you.) ---- CharlesTGillingham (talk) 22:28, 27 June 2026 (UTC)

I think the answer you'll get out of this discussion will depend heavily on who participates; some people who didn't like the deprecation of parenthetical referencing in the first place will seize on any opportunity to try to limit it, including advocating for "Author (year)" somehow being a valid citation and advocating for "explanatory notes aren't article body" despite the deprecation in the RFC close not being limited so strictly.

My takes:

• "Author (year)" might be a valid form of in-text attribution (depending on the context) but isn't a non-deprecated citation format. A <ref> or other non-deprecated format is still needed for the citation. Your second example, to be valid, should look more like Voltaire (1752) points out that size is relative; to the giants of Sirius the sun is quite small., although you might also want to rewrite it more like Voltaire, in 1752, points out... to flow more naturally while preserving the in-text attribution.
• Explanatory notes in an article are subject to the deprecation of parenthetical referencing in the same manner as the body text (and Special:Diff/1288012813 should be reverted). The distinction "directly inlined into articles" was to specify that short-form references like <ref>Author year</ref> and {{sfn}} are not included in the deprecation, not to continue to allow parenthetical referencing in explanatory footnotes, image captions, and whatever else.
• Footnotes-in-footnotes (or, more specifically, "references in explanatory footnotes") work fine if you use templates like {{efn}}. The "madness" linked above is an example of far too many explanatory footnotes containing far too much content, not a problem related to this discussion.

Anomie⚔ 23:09, 27 June 2026 (UTC)

Are you arguing that {{sfnp}} is an invalid way of making shortened citations and that WP:CITESHORT can only be done using {{sfn}} instead? Seriously?? The 23240 articles that transclude sfnp suggest that this dogmatism is misguided. Next I imagine you'll start trying to modify our cite templates to take away the parentheses they add around their years. —David Eppstein (talk) 01:39, 28 June 2026 (UTC)

{{sfnp}} is fine. I think you know that, and are trying to misinterpret things to pick a fight. Please don't. Anomie⚔ 01:59, 28 June 2026 (UTC)

Just to be clear: we are not discussing the merits of short citations, we are not discussing the use of templates vs. handwritten short cites. I used {{sfn}} in the example because they combine automatically in the reflist without refnames, that's all. It was less typing. ---- CharlesTGillingham (talk) 02:55, 28 June 2026 (UTC)

My understanding is that parenthetical references are deprecated in running text. They are still allowed, and indeed very common, in footnotes, and that includes explanatory notes. A note is a note. Gawaon (talk) 02:20, 28 June 2026 (UTC)

I also don't understand the point of this discussion, since WP:PAREN explicitly says so. Gawaon (talk) 02:27, 28 June 2026 (UTC)

The problem is that your understanding isn't documented anywhere, at least not unambiguously. In the previous section I asked if anybody had written a guide that breaks out these borderline things (in-text attributions and explanatory notes) and gives clear guidance. There isn't. So I'm asking. ---- CharlesTGillingham (talk) 02:55, 28 June 2026 (UTC)

I realize now that there is a major issue I had completely overlooked. If you use <ref>...</ref> to hold your explanatory footnotes, then you can't cite anything with out some kind of parenthetical references (or other short citations with parenthesis). So we're forcing the reader to interpret these (apparently abhorrent) parenthetical references only because we refuse to use {{efn}}, which solves the problem? That doesn't see right to me. But, if that's what you all agreed to, then fine. But you have to write it out clearly because it doesn't make sense. ----CharlesTGillingham (talk) 03:03, 28 June 2026 (UTC)

Yes, short citations, which may or may not include parentheses depending on how they are written, are of course not deprecated and never will be – but they have to be placed in notes. If that was your concern, I'm glad it's resolved. Gawaon (talk) 03:14, 28 June 2026 (UTC)

---

Again, we're not talking about short citations.

My actual question is what is the consensus about how WP:PAREN applies to explanatory notes, so I can document it correctly.

Let's start over. Here are the examples again, this time without short citations or {{sfn}} or {{harv}} templates, and with Anomie's interpretation of in-text attributions. I need {{efn}} because that's the only way to use <ref>...</ref> in an explanatory footnote.

Consider these cases.

Y The sun is pretty big. Voltaire points out that size is relative; to the giants of Sirius the sun is quite small.

N The sun is pretty big. Voltaire (1752) points out that size is relative; to the giants of Sirius the sun is quite small.

Y The sun is pretty big. Voltaire (1752) points out that size is relative; to the giants of Sirius the sun is quite small.

N The sun is pretty big. Voltaire points out that size is relative; to the giants of Sirius the sun is quite small. (Voltaire 1752)

Y The sun is pretty big.

Y? The sun is pretty big.

Y The sun is pretty big.

Y? The sun is pretty big.

Is this the correct interpretation of WP:PAREN? What is the argument in favor of the cases with question marks? Why are these okay, exactly? ---- CharlesTGillingham (talk) 04:14, 28 June 2026 (UTC)

Again, there is no requirement that explanatory footnotes be in their own separate section with efn. It is entirely acceptable to mix them with reference-only footnotes using ref. When you do that, the only formats that the explanatory footnote can take are the ones that you have question-marked. Forcing editors to use efn because of an unjustified fear of parentheses is putting the cart before the horse. —David Eppstein (talk) 05:09, 28 June 2026 (UTC)

{{refn}} can be used instead of {{efn}} if an article is mixing explanatory footnotes with references in one list. Anomie⚔ 12:40, 28 June 2026 (UTC)

Yes, all of those you marked with a Y are okay, question mark or not. WP:PAREN says so explicitly: "This also does not affect explanatory footnotes." So which variant to use is largely a matter of personal style and, more importantly, the style already dominant in the article in question. Personally, I'd tend to use the style of your first example, with ref-footnotes added even within efn notes. I also think that style is most common, but it's not something the MOS strictly requires. Gawaon (talk) 05:28, 28 June 2026 (UTC)

@CharlesTGillingham, we added the sentence "This also does not affect explanatory footnotes" to this guideline over a year ago. And it sounds to me like you're asking us, "Does this affect explanatory footnotes?" I know you're smart, so it's weird to see you asking this. What's really going on? Did someone claim that the rule is different from what the guideline says? Does it need to be made even plainer or explained in great detail? WhatamIdoing (talk) 05:58, 28 June 2026 (UTC)

I haven't looked at cite stuff for about 15 years. I came across WP:PAREN by accident and I thought it might be an opportunity to simplify the usage of the {{harv}} family, so I tried to dive back in. But then I saw the sentence about explanatory footnotes. I just couldn't fathom the reason --- it just doesn't make any sense from a reader's point of view. If they were bad for readers in the main body, then they are just as bad for readers in explanatory footnotes as well. If this part of the decision is just because of a technical limitation (i.e. no nested refs), then say that. Otherwise it's completely mysterious. ---- CharlesTGillingham (talk) 10:44, 28 June 2026 (UTC)

Ah, so you understand the current content of the guideline but want to challenge/change it. Perhaps you could have said so instead of asking odd questions; that would have been clearer. Gawaon (talk) 12:43, 28 June 2026 (UTC)

As above, I'd say the first question-mark may be considered as an in-text attribution without a (non-deprecated) citation. An actual citation should probably be added. The second is a deprecated parenthetical reference. Special:Diff/1288012813 was a mistake following a discussion mostly among people opposed to the deprecation in the first place. Anomie⚔ 12:38, 28 June 2026 (UTC)

I hear you, but it seems you are in the minority. ---- CharlesTGillingham (talk) 14:26, 28 June 2026 (UTC)

Like I said, I think the answer you'll get here depends heavily on who chooses to participate. This page doesn't seem very heavily attended, and seems to skew towards people who opposed the original deprecation. You might get a wider input by taking this to Wikipedia:Village pump (policy) or the like. Anomie⚔ 00:38, 29 June 2026 (UTC)

@Gawaon: I know what it says. I don't understand it. I've backed off having an opinion. I'm asking why the consensus thought this was the right choice. I'm looking for an explanation, an argument that this was the right choice. Is because of the technical issue? (No nested <ref>...</ref>) Is it a Three-fifths compromise? (That is, everyone got sick of arguing and just chose something with no rationale). Does it actually make sense to any of you? This used to be a place for rational debate. ---- CharlesTGillingham (talk) 14:26, 28 June 2026 (UTC)

I wasn't involved at that time, but I'd still say "a note is a note", so it seems plausible enough to me. Short references are parenthetical references, but placed in footnotes, so we can't and wouldn't want to deprecate them there. Hence allowing their placement in other kinds of notes is logical enough – forbidding it in one type but allowing it in another would be harder to justify, and it's doubtful whether whatever benefits that policy might bring would be worth it.

Also, I think the main goal of deprecation was to avoid interruptions in the reading flow. Most readers skip all notes, explanatory or not, and if they do read them, the interruption happens anyway, so that issue doesn't arise if you're already in a note. Gawaon (talk) 15:20, 28 June 2026 (UTC)

Thank you. (That was an actual, civil answer to the question "why?") If I understanding you correctly, you're saying: citations and footnotes sort of look the same in Wikipedia (in edit mode), so you can't base a rule on a difference that editors can't see. ---- CharlesTGillingham (talk) 16:18, 28 June 2026 (UTC)

I doubt that's what Gawaon was saying, since editors should be able to very easily see the difference between {{sfn|Voltaire|1752}} (or <ref>{{harv|Voltaire|1752}}</ref> or whatever) and <ref>Voltaire points out that size is relative; to the giants of Sirius the sun is quite small. (Voltaire 1752)</ref> (or similar with {{efn}} or {{refn}}). I think, instead, he's claiming that the deprecation was only because of "interruptions in the reading flow" in prose, and non-prose content such as explanatory notes should be allowed to continue using parenthetical referencing; I don't see that distinction in looking at the close of the deprecation RFC, however. Anomie⚔ 01:00, 29 June 2026 (UTC)

Well, maybe I was a bit unfair. He says "A note is a note" that they are the same. He doesn't think it's helpful to make a distinction between prose-like EFN's and code-like cites.

I agree with you that the "flow is interrupted" in an EFN; that's what my examples prove, and that's what confused me in the first place. But I acknowledge these editors have other considerations. They feel are more important.

Like: avoiding "multiple layers of indirection" (i.e. footnotes-in-footnotes); avoiding templates for EFNs ({{efn}}/{{refn}}); not having a separate sections for EFNs and cites; having a clear easily-stated rule to settle arguments.

That's what I'm hearing. ---- CharlesTGillingham (talk) 12:07, 29 June 2026 (UTC)

I note that there's no actual requirement to have separate sections for explanatory footnotes and cites, and a clearly stated rule could as clearly state "this includes explanatory footnotes" as the reverse. As for indirection, I suppose that depends on whether you consider "(Smith 2026)" as not being an indirection. Anomie⚔ 13:00, 29 June 2026 (UTC)

Just to close this out, here's what I've learned.

Parenthetical referencing is allowed in explanatory footnotes because:

1. The only alternative is footnotes-in-footnotes and, for some editors, footnotes-in-footnotes is an unreasonable degree of complexity.
2. Footnotes-in-footnotes requires using a template such as {{efn}} or {{refn}} (due to limitations of cite.php) and:
   1. Some editors are reluctant to use templates.
   2. Using {{efn}} requires EFNs be indexed by letters or other symbols rather than numbers
   3. Using {{efn}} requires EFNs to be grouped together in one list and citations another (which can be concatenated). Some editors prefer that footnotes appear in the order they first appeared in the text.
3. The rule is simpler to understand and enforce if it is based solely <ref>...</ref> without making a distinction between citations and footnotes.
4. "Interruptions in reading flow" doesn't really matter in footnotes.

(EDIT: Anomie explains that {{refn}} eliminates 2.3 and 2.4 as problems.)

My own opinion is this:

1. I can see how this is a matter of opinion. Anomie points out that (Smith 2008, p. 1), or (Smith 2000, p. 1) is also "directed" at the full citation, so they have the same "level of misdirections" as a footnote without the automation.
2. The second is due to limits in cite.php and would evaporate if it ever supported nested footnotes (this was first asked for 20 years ago). It took less time to go from V2s to the moon.)
3. I disagree with the third, because, as Anomie pointed out, editors aren't idiots; they can tell the difference between a text-like explanatory footnote and a code-like citation.
4. Anomie and I strongly disagree with the last one. I think my examples prove that the "flow" is interrupted in explanatory footnotes and I can't think of any reasonable reason that this doesn't matter.

One last comment. I had to really read between the lines in order to wrestle these 4 to 6 objections out of you all. At the top, I showed that I had doubts on this and I asked who agrees with it and why. I expected answers of the form. "I agree with it because X, Y and Z." Instead, you offered indignant assertions without any reasoning whatsoever, many of them off topic. It's like you all have given up on winning with sound arguments and valid reasoning. I'm confused and disappointed. ---- CharlesTGillingham (talk) 22:26, 29 June 2026 (UTC)

While 2.1 is strange (and disconnected from the reality of current editing) but not totally invalid, 2.2 and 2.3 are easily countered by the existence of {{refn}}. Anomie⚔ 23:17, 29 June 2026 (UTC)

I tested {{refn}} and couldn't figure out how to get it to work. Never mind, I got it to work. That's how {{efn}} should have worked! Too late. Sorry. Changed my mind after seeing examples below. ---- CharlesTGillingham (talk) 02:27, 30 June 2026 (UTC)

That feels a bit very unfair. You seemed to be asking for an explanation of what exactly WP:PAREN entails, writing: "I would argue that the last case is deprecated under WP:PAREN". We provided that explanation, pointing out that the wording in WP:PAREN is clear and unambiguous. Now you say you had to "wrestle ... objections" out of us and that was harder than it should have been. But since you had never asked for objections in the first place (to what, anyway? – that too became only clear over time), it's odd to blame us for that. Next time, if you make your point explicit in your first post rather than expecting us to read between the lines and blaming our failure to immediately do so, that would make things easier for everyone, yourself included. Gawaon (talk) 03:32, 30 June 2026 (UTC)

Okay, fair enough. Here's my excuse: I thought (stupidly) that WP:PAREN was basically this: Special:PermanentLink/976895063#Deprecate parenthetical citations, which was the first thing I read. I didn't realize that you all had made some more decisions. So, yes, I could have said that differently, I see how I was asking the wrong question. I meant to ask Who agrees with this policy and why?. But you have to admit there was a lot of bad attitude in people's responses. I think somebody called me a "gnome". ---- CharlesTGillingham (talk) 03:49, 30 June 2026 (UTC)

Well, the link you give already has, in the closure section: "This discussion supports the deprecation only of parenthetical style citations directly inlined into articles. It does not deprecate the use of the entire citation format when it is used within <ref></ref> tags." (emphasis added) In my understanding, an {{efn}} is essentially a different way to write a ref that exists for technical reasons and to allow better footnote grouping. It's not something very different and certainly not "directly inline" in an article, as that clearly is supposed to refer to the main text flow. So I'd still maintain that edits made to WP:PAREN since then logically follow from the closure decision you cite rather than modifying it in any meaningful way.

As for "bad attitude", keep in mind that we're to AGF. And I don't think a gnome is as bad as you make them out to be. Gawaon (talk) 05:59, 30 June 2026 (UTC)

So more specifically, if you want to get your preferred change while adhering to the spirit and letter of that closure decision, you would have to prove that {{efn}}'s are "directly inlined into articles", while ref's are not. I'd be sincerely surprised if you managed to pull that off. Gawaon (talk) 06:19, 30 June 2026 (UTC)

Oh and I thought of one more reason.

5. We shouldn't be deprecating anything that uniquely identifies a source.

5. I actually agree with this one, but it's too late for that now. Now I just wish it was consistent. (And, full disclosure, my proximate problem is that it's a lot more work to document inconsistent rules.)

---

CharlesTGillingham (talk) 03:15, 30 June 2026 (UTC)

_____

Here are three examples of the things we're talking about.

1. Explanatory footnotes in the same section with citation footnotes. Footnote uses normal citations.

Markup	Renders as
The Sun is very big.<ref>{{cite book|last=Miller|year=2000|title=Size of the Sun}}</ref>{{refn|The moon is also very big.<ref>{{cite book|last=Hsu|year=2001|title=Size of the Moon}}</ref>}} == References and Notes == {{reflist}}	The Sun is very big. References and Notes

2. Explanatory footnotes in their own section. Footnote uses normal citation.

Markup	Renders as
The Sun is very big.<ref>{{cite book|last=Miller|year=2002|title=Size of the Sun}}</ref>{{efn|The moon is also very big.<ref>{{cite book|last=Hsu|year=2003|title=Size of the Moon}}</ref>}} == Notes == {{notelist}} == References == {{reflist}}	The Sun is very big. Notes References

3. Explanatory footnotes in the same section with citation footnotes. Footnotes use parenthetical references.

Markup	Renders as
The Sun is very big.<ref>{{cite book|last=Miller|year=2004|title=Size of the Sun}}</ref>{{refn|The moon is also very big. {{harv|Hsu|2005}}</ref>}} == References == {{reflist}} == Bibliography == * {{cite book|last=Hsu|year=2003|title=Size of the Moon}}	The Sun is very big. References Bibliography Hsu (2005). Size of the Moon.

Parenthetical citations are deprecated everywhere except in explanatory footnotes. The question at issue is this: Why? Why is it so important to permit editors to use case #3, when we've deprecated so much else?

(Oh, and if you think I've misrepresented this peculiar case, please copy & edit my examples and show us exactly what you think "parenthetical references are used in explanatory notes, but not in the text body" looks like in an actual article. ---- CharlesTGillingham (talk) 03:15, 30 June 2026 (UTC)

There's tons of other stuff that's still permitted too, so your examples feel very strange and arbitrary to me. Gawaon (talk) 03:42, 30 June 2026 (UTC)

Also, in your case #3, you have "References" (as is normal) on the left, but it accidentally (and unusually) became "References and Notes" on the right. Gawaon (talk) 03:46, 30 June 2026 (UTC)

Fixed the example. ---- CharlesTGillingham (talk) 03:58, 30 June 2026 (UTC)

I tried to do everything using the most popular, the most "normal" way. I did the references with citation templates in <ref>...</ref>, the most common way on Wikipedia, the way the visual editor does it. I did the parenthetical references the most common way as well, with {{harv}} linking into a full citation in the references section. I submit that this is the most likely way an editor would interpret "parenthetical references used in explanatory notes, but not in the text body". Is there a way you think is more likely? Copy my examples, edit them and show me. ---- CharlesTGillingham (talk) 03:58, 30 June 2026 (UTC)

Your case #3 doesn't feel normal to me at all. More typically I'd expect in such a case both books, Miller (2004) and Hsu (2005), to be listed in the Bibliography, and so ref 1 to contain just the short reference "Miller (2004).". Then it seems logical enough that both ref's are written in the same style, with short author–year citations given within the footnote and full bibliographic details following in the Bibliography. Gawaon (talk) 06:13, 30 June 2026 (UTC)

"Why is it so important to permit editors to ..." is the wrong question. The better form is "Why is it so important to forbid editors to ...". Our rules allow editors to do lots of things; see for instance WP:CITEVAR. In most cases, when what they do is problematic, it can be handled in the normal course of editing by other editors without forcing everyone to comply with strict rules. This is better, when possible, because rules can often backfire in circumstances unexpected by the rule-makers that cause everyone to do something in a worse way. They are only needed when we have a problem that the normal process of editing and consensus is inadequate to handle. The discussion that caused parenthetical citations to become deprecated presented examples of articles that had actual problems. Above, I have presented an example of an article with nested efns that became a problem. Have you presented any examples of actual articles where parentheses in explanatory footnotes became so overloaded that they became difficult to understand, and where editors were so insistant on preserving that format that the normal process of article editing and local consensus needed a global rule-making process to override it? No. You appear to be so focused on creating more rules that you have stopped considering consequences for actual editing of content. —David Eppstein (talk) 05:20, 30 June 2026 (UTC)

I agree with you about rules. I am not arguing for adding a rule, I am arguing for removing an exception. All the concerns you have about rules apply equally well to exceptions. Exceptions have to be documented. Exceptions have to be learned by new editors. Exceptions should only made when the exception actually solves a problem. Exceptions have unintended consequences. ---- CharlesTGillingham (talk) 09:59, 30 June 2026 (UTC)

No. You are arguing for making a rule more strict. You have not provided any justification for doing so other than an apparent love for more strict rules. More strict rules are, in general, not a good thing and are generally unhelpful for what should be the goal of every Wikipedia editor: making better content. —David Eppstein (talk) 07:13, 2 July 2026 (UTC)

I believe the answer to your "Why?" is related to your item #5 above: some people disagree with the community consensus for deprecation, and so want to bring back parenthetical referencing wherever they can. Anomie⚔ 12:04, 30 June 2026 (UTC)

4. Here's an example like Gawaon suggested. That looks a lot better.

Markup	Renders as
The Sun is very big.<ref>{{harvnb|Miller|2008}}.</ref><ref>The moon is also very big {{harv|Hsu|2007}}.</ref> == Citations == {{reflist}} == Sources == * {{cite book|last=Hsu|year=2007|title=Size of the Moon}} * {{cite book|last=Miller|year=2008|title=Size of the Sun}}	The Sun is very big. Citations Sources Hsu (2007). Size of the Moon. Miller (2008). Size of the Sun.

Example 3 looked it had mixed citation formats (i.e., it didn't respect WP:CITESTYLE). But this example looks okay. I see the footnotes as a kind of "short footnote with text" rather "article text in an explanatory footnote with a parenthetical citation." The style feels consistent. Readers of an article like won't be "confused" (that's the word from the original 2020 decision), because they will be accustomed to seeing author-date in the short footnotes.

I don't think I have to document "parenthetical referencing" at {{harv/doc}}. The documentation of {{harv}} will say "{{harv}} is for short footnotes" and this example won't be documented, but the fact that there are variations on short footnotes will be. ---- CharlesTGillingham (talk) 10:39, 30 June 2026 (UTC)

I've very slightly edited that example to use {{harvnb}} for the first case ({{sfn}} without surrounding ref would be the obvious alternative) since I've never seen a whole footnote enclosed in parentheses and don't think that formatting detail would make any sense. Hope that's okay; revert me otherwise. Gawaon (talk) 12:35, 30 June 2026 (UTC)

And for the record, since you write "like Gawaon suggested": now (with my edit) it looks as I had intended it. Gawaon (talk) 12:40, 30 June 2026 (UTC)

Good. It's a convincing example. ---- CharlesTGillingham (talk) 19:19, 30 June 2026 (UTC)

Good. A minor point I noticed is that its "Citations" subheading is actively misleading, since ref 2 is more than a mere citation. So a heading like "References" (most usual), "Notes", "Notes and references", "Footnotes", or whatever would make more sense. Gawaon (talk) 02:19, 1 July 2026 (UTC)

MOS:REFERENCES names both ==Citations== and ==Sources== as "questionable in some contexts". Most editors avoid those two. WhatamIdoing (talk) 02:46, 1 July 2026 (UTC)

Yes, the names "References" and "Bibliography" from example 3 are better and work just fine here too. Gawaon (talk) 05:21, 1 July 2026 (UTC)

There's another reason: When we make a change that causes an inconvenient number (or complexity) of articles to be 'in violation' of a rule, then we normally try to make that change as small or narrowly as possible. Instead of saying "anything vaguely or approximately like ___ is now bad", we say "anything exactly like this, in exactly this circumstance, is bad". We'd rather have people prioritize re-formatting things such as "The sun is really big (Miller 2008, p. 1)" than to have them worrying about how to do a more complicated thing with a nested footnote. Most editors could re-format 10 or more simple parentheticals in the amount of time that it takes them to figure out how to fix one nested footnote.

And, of course, the most important reason: Because that's what the consensus actually was. A consensus doesn't have to make sense to any individual to still be the community's consensus. WhatamIdoing (talk) 02:45, 1 July 2026 (UTC)

Well said. Gawaon (talk) 05:21, 1 July 2026 (UTC)

then we normally try to make that change as small or narrowly as possible I'd say we generally try to be clear about the change, but not "as small or narrowly as possible". Usually we try to avoid confusing loopholes like the one being claimed here.

the most important reason: Because that's what the consensus actually was. I don't see that at Wikipedia:Village pump (proposals)/Archive 171#Parenthetical citation closure, although the close isn't entirely unambiguous if you wikilawyer hard enough. Because, yes, you could claim a very narrow interpretation of "directly inlined into articles" and a very wide one of "used within <ref></ref> tags", as deprecation-opposers have done here, even though that's nor supported by the discussion. Anomie⚔ 11:14, 1 July 2026 (UTC)

See my "sincerely surprised if you managed to pull that off" remark above. Interestingly, none of those opposed to the status quo has replied to that challenge yet. Personally, the closure seems fairly clear and unambiguous to me. The explicit remarks that the deprecation only applies to "directly inlined" content and not within ref tags leave little room for plausible alternative interpretations. Gawaon (talk) 12:36, 1 July 2026 (UTC)

I doubt I'll convince you, but to me it seems very clear that "directly inline" is to contrast with the superficially similar WP:CITESHORT style, and the "in <ref></ref> tags" is similarly in direct reference to that style (to avoid a claim that {{sfn}} is required for WP:CITESHORT). These clarifications were specifically aimed at preserving WP:CITESHORT, not at carving loopholes that no one in the discussion argued for to continue to allow parenthetical referencing in explanatory footnotes. Anomie⚔ 13:00, 1 July 2026 (UTC)

I think that we probably need to get more input than just the handful of people participating so far. But I'm also wary of trying to get people to read through the convoluted discussion above. Does everyone agree that the section below accurately states and illustrates the dispute? Feel free to directly edit it to make minor corrections, but if you think I've misstated (or stated non-neutrally) let's discuss it. If we can agree on the statement below, I intend to post an invite on WP:Village pump (policy) pointing at that. Anomie⚔ 13:48, 1 July 2026 (UTC)

Seeing no objections and people starting to comment below, I've posted the invitation. Let's see if we get enough fresh input to have a reliable read on consensus. Anomie⚔ 11:42, 2 July 2026 (UTC)

An attempt to clearly state the question

The dispute is over whether explanatory footnotes are or are not covered by the 2020 deprecation of parenthetical referencing. See in the following example.

Markup	Renders as
{{#parsoid	This is clearly deprecated (Seraphimblade 2020). As well as similar wikitext with the same result (Seraphimblade 2020). This is clearly not deprecated. As well as similar wikitext with the same result. This line has explanatory notes. References Bibliography Seraphimblade (2020). "Deprecate parenthetical citations § Parenthetical citation closure".

Please don't get sidetracked over whether the explanatory notes should use {{efn}} to separate them from the references. That's a matter outside of the current dispute. Anomie⚔ 13:48, 1 July 2026 (UTC)

• I find ² acceptable. -- Michael Bednarek (talk) 15:05, 1 July 2026 (UTC)
  • That's nice, but doesn't really speak to the question. Back in 2020 you also found parenthetical references acceptable. Anomie⚔ 11:39, 2 July 2026 (UTC)
• I frankly don't see a possible dispute over the interpretation of the 2020 closure here. Anomie, you're attempting to overturn parts of the 2020 closure, especially the part where it says: "It does not deprecate the use of the entire citation format [= parenthetical style citations] when it is used within <ref></ref> tags". That allows your note 2 without even the shadow of a doubt. Seeking a new consensus that modifies the old one is of course your right, but please be honest about what you're doing here. Gawaon (talk) 03:23, 2 July 2026 (UTC)

  So my interpretation of the closure, as well as the current wording of WP:PAREN, is that author-year-style citations within footnotes were not deprecated, and that includes cases where they share the same footnote with explanatory text. The closure states so explicitly for their occurrence "within <ref></ref> tags", which includes note 2 from the example. More reasonable doubts could be made for explanatory notes written using {{efn}} or similar templates and rendered in a separate section, but to keep things simple and straightforward to explain, my preferred interpretation is that they should remain allowed in all kinds of notes. Which does not mean one has to use them there, of course. It's a permission, not an obligation. Gawaon (talk) 13:59, 2 July 2026 (UTC)

• Covered by deprecation When I look at the deprecation discussion, I see no one arguing for differing treatment of explanatory footnotes versus the main article text; the one !vote even mentioning them was advocating for keeping parenthetical referencing because they considered and insufficiently distinct. Instead I see people talking about whether parenthetical referencing should be deprecated with no limitations on location. Meanwhile, I see a lot of discussion, starting with the initial statement itself, that wanted to be sure to preserve WP:SHORTCITE. The references to keeping {{harv}} "when wrapped in <ref></ref>" are typically paired with reference to keeping {{sfn}}, and a few even explicitly give examples like <ref>{{harv}}</ref> that make it particularly clear they're referring specifically to WP:SHORTCITE-style.
  Therefore, when I read the close and see text like "directly inlined into articles" and "when it is used within <ref></ref> tags", I find the most consistent interpretation is that these qualifiers are attempting to clearly exclude WP:SHORTCITE from being covered by the deprecation, not that they're introducing a loophole for keeping parenthetical citations if they're inside explanatory footnotes. Anomie⚔ 11:39, 2 July 2026 (UTC)
  • Above Gawaon accuses me of being "dishonest" in this position. IMO I'm being entirely honest, while I think he is wikilawyering to try to keep some vestige of the deprecated format he wishes was not deprecated. Anomie⚔ 11:39, 2 July 2026 (UTC)

    My apologies if I came across as "accusing"; that was not my intention! However, please don't attribute any undocumented "wishes" to me. I certainly have no wish to use the author–year style in the running text, and I've already mentioned that I personally avoid it in {{efn}} notes. But my personal style and preferences don't dictate our guidelines, nor my interpretation of them. Gawaon (talk) 13:46, 2 July 2026 (UTC)

• From the close: "Inline parenthetical referencing should be deprecated." In my view, parenthetical referencing in a footnote isn't parenthetical referencing inline, so in the black letter of the rules, strictly speaking, I'm going to say no. I don't like this because the outcome doesn't make much sense to me, but it's what it says.—S Marshall T/C 13:32, 2 July 2026 (UTC)
  • Turns out I was wrong, the close really was intended to only cover the article body, as strange and unsupported by the discussion as that still seems to me. If people want to start a new RFC to make the guidance make sense, I'd be in favor of that, but I don't have the energy to pursue it myself right now. Anomie⚔ 11:00, 3 July 2026 (UTC)

  I'm curious as to why you choose the interpretation that doesn't make sense over the interpretation I suggested just above, where "inline" is there to mean "in prose" as a contrast with WP:SHORTCITE rather than meaning "only in the body text". Anomie⚔ 19:00, 2 July 2026 (UTC)

  Because the closer wrote: Please see the section #Parenthetical citation closure for details and rationale. I followed the hyperlink and read that section, and the words I saw were: This discussion supports the deprecation only of parenthetical style citations directly inlined into articles. It does not deprecate the use of the entire citation format when it is used within <ref></ref> tags, nor the use of the {{sfn}} and {{harv}} templates. The closer's intent is crystal clear to me, even though I think it's a bit silly.—S Marshall T/C 21:52, 2 July 2026 (UTC)

  Seems extremely silly to me to choose the less sensible reading that would be kind of a WP:SUPERVOTE if it were the case, but 🤷. Anomie⚔ 23:24, 2 July 2026 (UTC)

  I just realized I had Seraphimblade confused with someone else, and he's still around. So I went with the most straightforward method to try to clarify the intent of the close. Maybe we'll get a real answer as to what the intended meaning was? 😀 Anomie⚔ 23:51, 2 July 2026 (UTC)

  He has replied ("I intended it to be a very narrow close"), so we guess we can consider this issue settled now. Gawaon (talk) 02:09, 3 July 2026 (UTC)

• I would never use Harvard on-wiki myself, but I have to agree with Gawaon and S Marshall. I'm just not seeing the ambiguity here, as explanatory footnotes are fundamentally not inline text. Speaking subjectively, I do think this carve out does make sense as it allows consistency with a short citation system of the form <ref>{{harv}}</ref>, which was explicitly noted as not deprecated in the 2020 RfC's closure. UpTheOctave! • 8^va? 15:28, 2 July 2026 (UTC)
• No, per UpTheOctave. Nikkimaria (talk) 23:56, 2 July 2026 (UTC)
• Not covered and they don't need to be. I proposed the deprecation in 2020, and I didn't intend parentheticals in footnotes to be covered. Having parentheticals in a footnote is fine and even helpful when explanatory footnotes are needed. CaptainEek ^{Edits Ho Cap'n!}⚓ 04:53, 5 July 2026 (UTC)

Edits to the guideline

(First let me apologize that my ignorance and haste caused such a hoopla in the previous section. Hopefully, the rest of what I have to contribute here will be useful. Thank you to the editors who stuck with me until I got my head around it.)

I've boldly added two "good" examples developed by User:Gawaon and myself to WP:CITE. See WP:CITE#Citations in explanatory footnotes. The examples should be uncontroversial and they are clearly marked as suggestions. I tried to use only the most normal/popular/preferred methods in every case. If there is a different example which is less controversial, please let me know (or add it yourself).

I think these examples are necessary, my experience proves it. It wasn’t clear to me what you all were thinking and I needed working example to really see it. I think we all need that, actually.

Please discuss these examples my recent edits (and only my recent edits) under this header. Let's not talk about policy in this section, if we can help it. ----CharlesTGillingham (talk) 15:09, 1 July 2026 (UTC)

The edits are presumably Special:Diff/1361846497/1362030205? I object to the second example, that's exactly what's under dispute above. I'd say either delete it, or make it very clear that it's disputed. Also, BTW, WP:EXPLNOTE goes to a different page, and WP:EXPLNOTESECT is broken due to the changed section title. Anomie⚔ 15:32, 1 July 2026 (UTC)

I fixed the short cut. Not sure why the short cut's not working. Short cut's fixed. (Thanks to User:RedRose).

I'd rather not take it out. It is currently acceptable, after all. But I can put some kind of note on it, I guess. Is there a template tag of some kind I could use? (EDITED) ---- CharlesTGillingham (talk) 15:51, 1 July 2026 (UTC)

Actually, I think it's better for your purposes if it's there. The question at issue is this: is this example okay? Your goal is: consensus to remove this example or give it a N. It focusses this debate on a salient issue rather than on abstract principles. ----CharlesTGillingham (talk) 16:26, 1 July 2026 (UTC)

I have to wonder why someone who hasn't made a substantive content edit since April is putting so much effort into telling other editors how to edit content. —David Eppstein (talk) 01:48, 3 July 2026 (UTC)

Argument from authority. Irrelevant. ---- CharlesTGillingham (talk) 16:21, 3 July 2026 (UTC)

Shortened footnotes using "sfn" are wonky

I have again encountered an article where "sfn" is used. These are a nightmare to edit. Is there any support for prohibiting this "sfn" style, and just insisting on inline citations? Magnolia677 (talk) 19:18, 1 July 2026 (UTC)

Sfns are easy, inline citations to the same book 60+ times on 60 different pages are impossible to verify otherwise. PARAKANYAA (talk) 19:29, 1 July 2026 (UTC)

No per CITEVAR, although also per CITEVAR people shouldn't be changing inline cites to short foot notes either (at least unless it's discussed on the talk page like any change to an articles established style). -- LCU ActivelyDisinterested «@» °∆t° 19:40, 1 July 2026 (UTC)

WP:CITEVAR Not what citevar says. It clarifies that you should not do it "merely on the grounds of personal preference or to make it match other articles", not ever. If someone contests that, of course, you advance to discussion. PARAKANYAA (talk) 19:44, 1 July 2026 (UTC)

I ran into this yesterday and tried to persuade the editor to use the same style as all the other citations on the article. It was no use. My experience with these sfn's is they're wonky and finicky. And they add books to the reference list that mix in with the inline citations, so you confuse them as being "further reading". Why not just use <ref name=XYZ>{{rep|20}} or 21 or 28...or whatever page you're citing? Magnolia677 (talk) 19:46, 1 July 2026 (UTC)

How is it finicky or wonky? It does exactly the same thing every time. RP is the worst template, it clogs up the text of the article, isn't even clear on what it is trying to convey. If any should be deprecated it is RP. PARAKANYAA (talk) 19:49, 1 July 2026 (UTC)

Sfn is like a flip-phone; fun back in the day but editors have moved on to a simpler style. Not sure I'm going to get any traction with this proposal though. Magnolia677 (talk) 19:54, 1 July 2026 (UTC)

Perhaps you can give concrete examples of what you mean by wonky and finicky? Without those, I suspect that your proposal will be interpreted as WP:IDONTLIKEIT...

—Trappist the monk (talk) 20:02, 1 July 2026 (UTC)

What style have they moved on to? There was a large discussion about deprecating RP, which saw quite a bit of support, so that's clearly not it. PARAKANYAA (talk) 20:45, 1 July 2026 (UTC)

Very few edits actually include a page number. Most are usually citing websites or news articles. Magnolia677 (talk) 20:51, 1 July 2026 (UTC)

Yeah, but reliable book sources are always better. Most people cite trashy sources that should be replaced. Are you suggesting we no longer cite books?? PARAKANYAA (talk) 20:53, 1 July 2026 (UTC)

I'm suggesting we cite whatever reliable source is appropriate...just not in an insufferably hard-to-use style most editors stopped using 15 years ago. Magnolia677 (talk) 20:57, 1 July 2026 (UTC)

Most editors did not stop using it for books. Most editors have stopped using RP tags. PARAKANYAA (talk) 21:38, 1 July 2026 (UTC)

I find sfn very easy to use. It follows a short citation style that is very common in academic writing and requires only three pieces of information: author last name, year of publication, and page number(s). What is wonky and finicky about that? voorts (talk/contributions) 21:50, 2 July 2026 (UTC)

I don't get your point and I certainly don't agree. {{sfn}} is extremely useful when referring to different parts of the same work (whether pages, chapters, or whatever). Gawaon (talk) 03:10, 2 July 2026 (UTC)

Look at the "Toponymy" section at Scarsdale, New York...{{Sfn|Hansen|1954|p=9}}{{Sfn|Reische|Brownell|1976|p=17}} Where's the rest of it? Where's the URL and the name of the source? Oh...way down there! Magnolia677 (talk) 20:15, 1 July 2026 (UTC)

Nothing wonky and finicky about those; they are working as they are designed to work.

—Trappist the monk (talk) 21:44, 1 July 2026 (UTC)

I'm restating things here, but {{sfn}} is absolutely necessary if you are citing many pages from a book, which often happens with topics that are interesting enough to have whole books written about them. So you can't get rid of it if it's needed.

There are articles that need {{sfn}} for some sources, but use ordinary citations-in-footnotes for sources used only once. This "mixed" style is a valid style that you all could use, if no one objects. This style is actually pretty common.

The most important thing is that the text is verifiable. The rest someone can fix later. ---- CharlesTGillingham (talk) 22:13, 1 July 2026 (UTC)

I do think that using sfns for sources without pages (as I've seen done) is stupid and useless and adds nothing but an extra click for all involved, though. PARAKANYAA (talk) 22:15, 1 July 2026 (UTC)

@Magnolia677: You said "Where's the rest of it? Where's the URL and the name of the source? Oh...way down there!" There's no way to do multiple citations to a single source without this problem. Either you copy the citation into every footnote where's it's used, or you create pointers (i.e. sfn or refname) to main citation. If you have a pointer, then the main citation is going to be "way down there" somewhere. Think about it. This isn't a problem with sfn (or refnames), this is a consequence of combining identical/matching footnotes. ---- CharlesTGillingham (talk) 22:47, 1 July 2026 (UTC)

<ref name=XYZ> allows you to do multiple citations using a single source...and much simpler. Magnolia677 (talk) 23:03, 1 July 2026 (UTC)

Not to different pages. Do you think it's acceptable to cite the full range of a book? PARAKANYAA (talk) 23:08, 1 July 2026 (UTC)

<<

@Magnolia677: How is that easier? They're exactly the same amount of work! <refname = Smith2009p3 /> and {{sfn|Smith|2009|p3}} make the same tradeoff:

1. - You have to go somewhere else to find the full citation
2. + No clutter in the edit window
3. + No need to type the full citation twice

The differences:

1. + {{sfn}} handles multiple pages of the same source
2. - {{sfn}} takes the reader two clicks to get to the source
3. - Using {{sfn}} for sources that aren't repeated is a pain for both editors and readers, but, articles with mixed short footnotes and citations-in-footnotes are ugly. Refnames, on the other hand, mix well with citations in footnotes.----CharlesTGillingham (talk)

And another thing or two:

1. + {{sfn}}'s "refname" (that is, "year" and "date", the thing you have to remember when you're adding another citation to the source) is transparent, standardized, human readable, machine readable, familiar, easy to remember. Refnames, on the other hand, are often inscrutable computer code like "DGH_3mC" or ":3".
2. + {{sfn}}'s full citations are in an alphabetized list, which can be organized with sub-headers. It's useful to readers. refnames full citation is buried in a random section of the article and is vulnerable to being accidentally deleted if the section is deleted. ---- CharlesTGillingham (talk) 12:44, 2 July 2026 (UTC)

   Where would refname = Smith2009p3 come from, though? {{sfn}} is typically used for page references that are unique, so there won't be any named reference one could re-use. (And in the case of two {{sfn}} templates happening to go to the same page or range, the templating system is smart enough to merge them into one note, without any explicit naming required.) Gawaon (talk) 13:38, 2 July 2026 (UTC)

   That's the alternative if you want to specify individual pages and not use sfns. You just duplicate the long ref and use refname. PARAKANYAA (talk) 19:13, 2 July 2026 (UTC)

I was trying to show that adding a named ref and adding {{sfn}} take the same amount of work. Then I kind of went off. Forgive me. ---- CharlesTGillingham (talk) 18:22, 2 July 2026 (UTC)

In-text attribution

This table appears in Template:sfn/doc#In-text attribution. I don't think it belongs there, because it is only tangentially connected {{sfn}}. I think it belongs here. If there's no objection, I'm adding it.

I didn't build this table, so don't shoot the messenger.

---

Since the September 2020 request for comments, in-text attributions require an additional inline citation at the end of sentence. This table summarizes several examples discussed and the consensus reached there:

Example	Consensus
The Sun is pretty big. (Miller 2005, p. 1)	Formally deprecated; replace with "The Sun is pretty big."
The paper by Eek (2020) showed...	Wordy, but may be valid as in-text attribution; use "Eek showed..." instead
Eek (p. 35) showed...	Formally deprecated; use "Eek showed..." instead
In 2020, Pelagic et al. found that...	Wordy, but may be valid as in-text attribution.
Smith (2008) argued...	Formally deprecated; see the following for a non-deprecated alternative
Smith (2008) argued...	May be valid as in-text attribution in contexts where both year and author are encyclopedic information; consider rewriting as "Smith, in 2008, argued..."
Smith argued...	Standard non-deprecated format

---

---- CharlesTGillingham (talk) 18:15, 2 July 2026 (UTC)

The table makes sense at Template:Harvard citation documentation, where it's currently located, as those templates are used to generate the sorts of things shown as deprecated in the table. Putting it in this page would probably give undue weight to the deprecated styles it's warning against; at most we might adjust the text in the N box at WP:PAREN to include some of the different deprecated formats. OTOH, if you want to somehow hide it from Template:Sfn/doc where that transcludes Template:Harvard citation documentation, that could make sense since it's probably not really relevant to {{sfn}} either. Anomie⚔ 19:08, 2 July 2026 (UTC)

This decision doesn't apply exclusively to the templates -- it applies to hand written harvard references as well (e.g., see #1 in the table). The issues that it represents are not issues that the template document discusses --- the template documentation is all about "how use the the template to do things with the template that are difficult or tricky". There's no do this/don't do that. It's a how-to page not a policy page. It doesn't belong there.

That doesn't necessarily mean that it belongs here, however, but this is the only well-traveled place that I know it that talks about in-text attributions. It could go in WP:Inline citations, in the section about deprecated and badly formatted citations, I suppose. But that guideline isn't particularly well traveled. We're not trying to bury it -- the 2020 decision is hard to get your head around and editors shouldn't have to search for the information. That's my goal here. ---- CharlesTGillingham (talk) 19:30, 2 July 2026 (UTC)

Note the table isn't just about in-text attributions, it's specifically about the intersection of in-text attribution and the deprecation of parenthetical referencing. The only places other than Template:Harvard citation documentation (i.e. the templates used for parenthetical referencing) that have that same overlap are WP:PAREN (but it'd be better incorporated into the existing N example, maybe with a Y example added) and Wikipedia:Inline citation#Deprecated citation styles. Personally I think the 2020 decision is fairly easy to get your head around, if you can identify and ignore the specious arguments people make about it. 😀 Anomie⚔ 19:54, 2 July 2026 (UTC)

You said "Template:Harvard citation documentation (i.e. the templates used for parenthetical referencing)" This isn't a fair characterization (I think you were misled by the title, which is a misnomer. I'm currently in the process of fixing that.) This document only talks about short footnotes. That's what it's about. It describes how {{harv}} family templates are used in short footnotes (that is, as a variation on {{sfn}} with different formatting). It has nothing to do with in-text parenthetical referencing. ---- CharlesTGillingham (talk) 23:23, 2 July 2026 (UTC)

{{harv}} itself, which includes parentheses, was the primary template to do parenthetical referencing before that was deprecated. It may still be used like <ref>{{harv|...}}</ref> to do short footnotes, although {{harvnb}} is more usual for that use case. Anomie⚔ 23:29, 2 July 2026 (UTC)

You're not following, and this isn't important, but: the template documentation for {{sfn}} and {{harv}} only describes their use in shortened footnotes. I don't see any reason to add material about parenthetical referencing into it now. Let's just agree it belongs here. ---- CharlesTGillingham (talk) 23:48, 2 July 2026 (UTC)

Always? Going back to a 2019 version of that page, I find The Harv family of templates are used for creating short-cites (or "short citations"), typically used as in-line citations, ... right at the top. That was changed following the deprecation of parenthetical referencing. Anomie⚔ 23:58, 2 July 2026 (UTC) For clarity, the sentence this was replying to was removed in Special:Diff/1362287300. Anomie⚔ 00:02, 3 July 2026 (UTC)

As for whether it should be added back if you've removed it, 🤷. I think it might be useful to note that the original use case is now deprecated, to make that easier to find for people looking at the documentation. Anomie⚔ 00:02, 3 July 2026 (UTC)

Yes, I checked that too. Should have said "for a long time". In 2021, I refocussed the documentation on shortened footnotes.

There is a paragraph in the led (that only shows for {{harv}} templates) that explains WP:PAREN. ---- CharlesTGillingham (talk) 00:17, 3 July 2026 (UTC)

This table mixes up two different things.

The real question should be: Is it ok to use "Smith (2008) argued" as a style of in-text attribution, with a separate footnote at the end of the sentence as the reference (possibly to a different source that discusses Smith rather than a direct reference to Smith)? Yes, although if you want to say this then other equivalent phrasings like "In 2008, Smith argued..." can often be less awkward. Is it ok to treat the "Smith (2008)" part of the text as being a reference itself, and omit the final footnote? No.

However, the two variations are shown with a wikilink to the Smith 2008 reference in one case and without the wikilink in the other case. That is totally irrelevant. The wikilink does not change whether it is an in-text attribution or a parenthetical reference; it is the footnote that matters. The unlinked version, without its footnote, would still be a problematic parenthetical reference; it still needs a footnote to make clear that it is not intended as a reference itself. By the same token, the linked version, with a footnote, is still an in-text attribution; the link is a courtesy to readers who might want to more directly find the source discussed in the attribution (which, again, might be different than the source in the footnote reference) and the link does not make it into a reference. —David Eppstein (talk) 20:39, 2 July 2026 (UTC)

I agree that presence/absence of the wikilink is completely irrelevant to what the table is trying to explain. Here's the same table without wikilinks, and I amend my proposal to adding this table to the article. ---- CharlesTGillingham (talk) 23:23, 2 July 2026 (UTC)

Also added the Cross/Check form that is currently used at WP:PAREN (Anomie's suggestion) ---- CharlesTGillingham (talk) 00:17, 3 July 2026 (UTC)

---

Since the September 2020 request for comments, in-text attributions require an additional inline citation at the end of sentence. This table summarizes several examples discussed and the consensus reached there:

Example	Consensus
N The Sun is pretty big. (Miller 2005, p. 1)	Formally deprecated; replace with "The Sun is pretty big."
Y The paper by Eek (2020) showed...	Wordy, but may be valid as in-text attribution; use "Eek showed..." instead
N Eek (p. 35) showed...	Formally deprecated; use "Eek showed..." instead
Y In 2020, Pelagic et al. found that...	Wordy, but may be valid as in-text attribution.
N Smith (2008) argued...	Formally deprecated; see the following for a non-deprecated alternative
Y Smith (2008) argued...	May be valid as in-text attribution in contexts where both year and author are encyclopedic information; consider rewriting as "Smith, in 2008, argued..."
Y Smith argued...	Standard non-deprecated format

---

The question is where to put it:

1. WP:PAREN after the example.
2. WP:INTEXT after the other advice.
3. WP:Inline citation#In-text attribution, and add a sentence linking to it in WP:PAREN and WP:INTEXT.

Whatever you all think. ---- CharlesTGillingham (talk) 00:32, 3 July 2026 (UTC)

As others have already mentioned, none of these is valid as in-text attribution, because in-text attribution is deprecated. Hence, the actual attribution must, in all cases, appear in a footnote. Authors and years may be mentioned in the text, but not as attribution. Gawaon (talk) 02:01, 3 July 2026 (UTC)

In-text referencing is deprecated, when in certain parenthetical formats. (In-text referencing by means of footnote markers is not deprecated.) In-text attribution (that is, having article text that happens to mention that some scholar said or published something) is also not deprecated. How are we going to write an article like Wiles's proof of Fermat's Last Theorem without mentioning that Andrew Wiles published a proof of Fermat's Last Theorem? That is an in-text attribution. The question is merely how to write and format our articles in such a way that we distinguish their references (the materials readers can go to to check their veracity) from their attributions (the encyclopedic description of who said what when, in articles for which this sort of description is relevant as content), especially in cases where both of those two different things involve the same publications. —David Eppstein (talk) 06:13, 3 July 2026 (UTC)

Where do you think the table should go? ---- CharlesTGillingham (talk) 08:36, 3 July 2026 (UTC)

Fair enough and I can accept that. Even so, I don't think the table should go anywhere. What point would it serve? Gawaon (talk) 09:19, 3 July 2026 (UTC)

Because it's completely undocumented right now. There is nothing in the guide that says explicitly that in-text attributions require an additional inline citation at the end of sentence. EDIT: That's not correct. ---- CharlesTGillingham (talk) 16:04, 3 July 2026 (UTC)

All claims that could reasonably be challenged require citations. We do not need to separately call out specific types of claims. Doing so would be a mistake because it would give excuses to editors to avoid providing citations for other claims that are not specifically called out. —David Eppstein (talk) 19:29, 3 July 2026 (UTC)

Not the question at issue. ---- ~2026-38138-45 (talk) 19:48, 3 July 2026 (UTC)

Of course it is. CTG wants a rule that claims of the form "Scholar So-and-So published this" need an inline citation. But we already have a rule that all claims that could reasonably be challenged need an inline citation. A claim of this form without a citation could reasonably be challenged on the basis that we don't know how to verify the claim. So the general rule applies and we do not need a more specific rule. —David Eppstein (talk) 20:14, 3 July 2026 (UTC)

The question at issue is what to do with the table --- CharlesTGillingham (talk) 22:18, 3 July 2026 (UTC)

I take it your answer is "toss it" and I'm fine with that. (I brought it here because doesn't belong where it was {{sfn/doc}}, which I just rebuilt, and I didn't want to just toss it.) ---- CharlesTGillingham (talk) 22:39, 3 July 2026 (UTC)

Did you overlook WP:PAREN? Considering that we recently discussed it at length, you should know it well. Gawaon (talk) 03:29, 4 July 2026 (UTC)

[EDIT: let me rephrase]

This question is about the table. Where does it go? What do you think we're talking about? ---- CharlesTGillingham (talk) 10:25, 4 July 2026 (UTC)

CLOSE Toss the table, per Gawaon. Don't restore it to {{sfn}}, per me. WP:INTEXT has all the guidance we need and it doesn't belong in {{sfn}}. The table is still here if any one disagrees. ---- CharlesTGillingham (talk) 10:45, 4 July 2026 (UTC)

Well I wrote the table, it seems it still answers a question that has not been answered elsewhere. This guideline seems as good a place as any to keep it. Also Gawaon's comment is wrong - only 2 of the 6 variants discussed were formally deprecated, the others had no consensus. You can read the RFC again if you don't believe me. Mathnerd314159 (talk) 04:26, 5 July 2026 (UTC)

You can use some of them for attribution, but the actual citation follows in all cases in a footnote. Which is the important part. Gawaon (talk) 07:16, 5 July 2026 (UTC)

And that's not what the RFC decided. To quote from the discussion: 'I don't agree that every article with "occasional in-text {{harvtxt}} references" is necessarily badly written or unreadable enough to justify universally deprecating the style.' And ultimately the RFC was retargeted to focus only on replacing parenthetical citations at the end of sentences with footnotes, and to be silent on the general concept of inline citation style. But I nonetheless see many editors citing WP:PAREN to justify wholesale replacement of in-text citations with footnotes, in violation of WP:CITEVAR. Mathnerd314159 (talk) 13:44, 5 July 2026 (UTC)

Well, if the closure decision did indeed carve out an exception for "occasional in-text {{harvtxt}} references" (whatever that's supposed to mean), I'm sure you can cite the exact statement where it says so. Incidentally, the {{harvtxt}} documentation itself seems to be better informed than you, pointing out that "harvtxt should not appear within the body text of an article outside of <ref>...</ref> tags." Gawaon (talk) 15:22, 5 July 2026 (UTC)

That statement was added by CharlesTGillingham during the recent kerfuffle; it did not appear 1 month ago. If you want to quote the closure decision, it says, "This discussion supports the deprecation only of parenthetical style citations directly inlined into articles. It does not deprecate[...] the use of the {{sfn}} and {{harv}} templates." Of which the harvtxt and harvnb templates are considered harv templates, and were the ones specifically brought up during the discussion. Mathnerd314159 (talk) 01:26, 6 July 2026 (UTC)

@Mathnerd314159: The harvard citation templates' documentation has said basically the same thing since 2021: "Also note that inline use of these templates, i.e. use of {{harv}} without <ref>...</ref> tags around it, was deprecated in September 2020. See also WP:PAREN."history I didn't change this in the recent rebuild. ---- CharlesTGillingham (talk) 10:19, 7 July 2026 (UTC)

Yes, the harv family is still allowed, but only within footnotes. This was pointed out in an initial note by the proposer, CaptainEek: "This is NOT a proposal to ban Template:sfn, or Template:Harv (as long as it is properly nested in a ref tag)" (emphasis added), and later confirmed by the closer, who explicitly noted that parenthetical references (whether produced by {{harv}} or in some other way) are still allowed "within <ref></ref> tags", but not outside. CaptainEek also explicitly stated that the deprecation was meant to include cases such as "The paper by Eek (2020) showed" and "Eek (p. 35)", and the closing statement did not reject or modify this. Case closed, and I won't participate in this discussion any further, as I need my time for other things. Gawaon (talk) 03:32, 6 July 2026 (UTC)

I'm just trying to understand your reasoning here:

• CaptainEek writes a broad, vague proposal
• The discussion critiques the proposal heavily and CaptainEek modifies and clarifies it several times, in somewhat contradictory ways
• The closer says "In sum, the supporters outweigh the opponents, so some form of statement should be added to WP:CITEVAR, but I won't specify the wording"
• The final statement added (and still present in WP:CITEVAR) is very narrow, only showing parenthetical citations at the end of sentences
• Your conclusion from this sequence of events is that "The closing statement did not reject or modify CaptainEek's proposal", and that her (heavily critiqued) proposal is what is binding.

I'm not a very good wikilawyer, but that doesn't seem right. Mathnerd314159 (talk) 17:41, 6 July 2026 (UTC)

No, CaptainEek's notes were not contradictory, and they were added on 6 August, just one day after her initial proposal. The actual close was one month later. So it's clear that what was discussed during that whole time included her two notes. If parts of them had failed to reach consensus, the closer would have said so. But he didn't.

Also, the core sentence from the closing statement is: "This discussion supports the deprecation only of parenthetical style citations directly inlined into articles." That was meant to still allow them within footnotes, as the recent discussion on his talk page resolved, but not elsewhere.

Are you seriously going to tell me that sentence-final placement is "directly inlined into articles", but sentence-middle is not? In case you try that, I'll preemptively reply with one word: No. Gawaon (talk) 03:44, 7 July 2026 (UTC)

Here's the thing: people disagree. To keep people from fighting, you need clear, simple rules. It's better if they have adequate study and valid logic behind them, but that's not strictly necessary. We have to have something to point to solve disagreements. The only thing we have is the language of the 2020 decision. It's the letter of the law. ---- CharlesTGillingham (talk)

In-text attributions have required an additional inline citation since Mar 16, 2011, when this edit added the clear language "An inline citation should follow the attribution, usually at the end of the sentence or paragraph in question."

So, if we dig a bit deeper into the logic, we have that in-text attributions: (1) may "uniquely identify the source"; (2) may have text-source integrity (i.e. are "inline"); (3) may fulfill WP:BURDEN by allowing readers to verify the text; but (4) they're not an approved "citation style" in Wikipedia; so (5) you (or later editors) should add another inline citation in an acceptable citation style.

That guidance explains all the examples in your table, so strictly speaking, Gawaon is right: the table isn't necessary. (Except maybe to clarify and emphasize the guidance from 2011, so no one gets confused and unintentionally starts an argument.) ---- CharlesTGillingham (talk) 11:55, 7 July 2026 (UTC)

I'll make this concrete. So there is the article Gödel's incompleteness theorems. There are a lot of uses of the Harvard templates. I'm not saying it's a good article overall - there are numerous actual parenthetical citations using {{harv}}, that clearly have to be replaced with footnotes. But I want to focus in particular on the section Relationship with computability and how it uses {{harvtxt}} to reference the papers directly, with statements such as "Kleene (1943) presented". And so what I'm hearing from you is this isn't an acceptable form of citation - but at least to me it seems like a clear and simple way to cite the material. Mathnerd314159 (talk) 13:55, 7 July 2026 (UTC)

It's acceptable as attribution, deprecated as citation. And the article's authors know that too, since, as you might have noticed, they have an {{sfn|Kleene|1943}} at the end of the relevant statement – so that's as it should be.

Personally, I think the section suffers from reading like it's written for scientists, not for a lay audience. Changing "Kleene (1943) presented a proof" to "In 1943, Stephen Cole Kleene presented a proof" would be nicer to read and present additional useful information (the author's full name and a link to his article). As written, it's already acceptable, for sure, but there seems to be room for improvement. Gawaon (talk) 15:29, 7 July 2026 (UTC)

OK, I think you have made your position somewhat less clear, because in CaptainEek's proposal, there is the August 6 "note to the note" that "use of The paper by Eek (2020) showed [should be] phased out, in the interest of consistency.", which I would interpret to mean that "Kleene (1943) presented" is also deprecated. But then I would point to:

1. Biogeographist's opposition that I quoted before, and then that
2. the proposal was further clarified to a specific modification of WP:CITEVAR in September which didn't mention this kind of in-text attribution+reference combination

as reasons I don't think CaptainEek's note to the the note is actually what was approved as the consensus. Mathnerd314159 (talk) 20:56, 7 July 2026 (UTC)

You can think what you want, but don't be surprised that few others will follow you in that idiosyncratic interpretation. A single commenter's opinion in a discussion obviously does not override the closer's decision, and if it wasn't explicitly mentioned in a post-close edit, that means that no particular exception was carved out for it. Gawaon (talk) 02:27, 8 July 2026 (UTC)

Also, it's revealing that you refuse to engage with the core sentence from the closer's statement, cited above, which directly contradicts your claim that some parenthetical-style citations directly inlined into articles are still permitted. But pretending it's not there won't make it go away. Gawaon (talk) 02:34, 8 July 2026 (UTC)

Well I have engaged it - I quoted it before you did. But the problem is it's ambiguous. "Parenthetical style citations" is nowhere defined in the closure. CaptainEek's second note says she explicitly meant to include attributions like "Smith (2008) said", but the suggested wording added after that and the final wording and examples in WP:CITEVAR do not list these. And it has been argued earlier in this thread whether it is a citation at all or simply an attribution, so the plain language does not weigh in either. Mathnerd314159 (talk) 18:15, 9 July 2026 (UTC)

Fortunately we have an article about it: Parenthetical referencing, sparing the closer the need to repeat everything said there. The "Jones (2001)" case is explicitly mentioned there, as second item in the "How to cite" section, and called "narrative style citations". So the ambiguity you wish for does not actually exist. Gawaon (talk) 02:47, 10 July 2026 (UTC)

Right, so in APA, per the ref in that bullet, the case we have been discussing is called a "narrative citation" and is distinct from a parenthetical citation. The remaining question is whether the closer was familiar with this APA definition when they wrote the closure and specifically intended to only exclude parenthetical citations as opposed to narrative citations, or if "inline parenthetical citations" was a more vague term referring to APA style generally. Mathnerd314159 (talk) 04:12, 10 July 2026 (UTC)

You may or may not have noticed that there are parentheses in "Jones (2001)" and that it's given as an example in the Parenthetical referencing article, as well as in one of CaptainEek's original notes. It's a (very common) variant of the parenthetical referencing style, not a distinct style of its own. Considering all this evidence, I believe you can easily answer this "question" for yourself. If you seriously can't, Seraphimblade is still around and you could just ask him. But I'd consider this a waste of his time, as the answer is so obvious. Gawaon (talk) 05:28, 10 July 2026 (UTC)

Well if you read the APA guide, "In 2016, Koehler noted..." is also considered a narrative citation, and this doesn't have any parentheses. And then in the discussion, the term "narrative citation" was never mentioned, even though narrative citations were mentioned in the mainspace Parenthetical referencing article. In fact though the historical WP:PAREN article does not discuss narrative citations. So the conclusion I would draw is that CaptainEek tried to add narrative citations in for potential discussion and deprecation as well, but this additional category of citation was not recognized by either Seraphimblade nor anybody else in the 2020 RFC besides the brief mention in early opposition comments by Biogeographist. So considering all the evidence, I would conclude that narrative citations were not deprecated in the RFC, not explicitly but simply because they were not discussed beyond some passing mentions. Mathnerd314159 (talk) 18:36, 10 July 2026 (UTC)

You can write "In 2016, Koehler noted" all you want, but that's not a clearly recognizable citation, so don't be surprised if somebody adds a {{citation needed}} tag to your sentence or simply adds a proper footnote citation at the end. Wikipedia is self-repairing to some degree. Gawaon (talk) 03:39, 11 July 2026 (UTC)

That's where you would do something like "In 2016, Koehler noted", and the year would be a link to the paper. This would be verifiable, clearly recognizable, and follow APA style for citations (valid narrative citation). If we check WP:INLINE we see that "The most common method [of associating text with sources] is numbered footnotes within the text, but other forms are also used on occasion.", so it is valid to use citation formats besides footnotes. And as we have discussed, the RFC did not deprecate narrative citations. So it seems like this would be a perfectly acceptable way to write a citation under current policy. And then by extension if that is allowed, then using {{harvnb}} or {{harvtxt}} to write "Koehler 2016 noted" or "Koehler (2016) noted" would also be allowed as they are minor stylistic variations on narrative citations (contrary to the statements added by CharlesTGillingham to the documentation for those templates). Mathnerd314159 (talk) 18:00, 11 July 2026 (UTC)

I'll stick to the COMMONSENSE view that "narrative citations", especially since they nearly always contain parentheses, are a variant of parenthetical referencing, as per our article on it, which reflects the typical view. So they were deprecated by the 2020 RfC, in agreement with CaptainEek's original notes. Prove me wrong by citing a statement from the closer where he explicitly describes them as still permitted. If you can't, I'll consider this discussion over. Gawaon (talk) 02:50, 12 July 2026 (UTC)

If your only justification is your own common sense, then right below that there's WP:NOCOMMON. You haven't explained why narrative citations are bad in articles. Narrative citations do not have the same issues as parenthetical citations; they do not clutter the text. They actually clutter it less than the typical numbered footnote, if used well. I've explained my interpretation of the discussion at length - Captain Eek's note, Biogeographist's opposition, Francis Schonken's reframing to sidestep the issue, the close that is ambiguous, the final wording in WP:PAREN that illustrates parenthetical but not narrative citations. I could equally ask you to prove me wrong by citing a statement from the closer where he explicitly describes narrative citations as part of the style that is being deprecated. Mathnerd314159 (talk) 16:17, 12 July 2026 (UTC)

Not so: "Alice Jones (2008, p. 33) argues that the sun is, in fact, bigger than you think" (our own example of deprecated narrative-style citations) is just as interruptive to the reading flow as "Some authors argue that the sun is, in fact, bigger than you think (Jones 2008, p. 33)" would be. And both are equally "parenthetical"; a superficial look at them is enough to show that. Gawaon (talk) 03:55, 13 July 2026 (UTC)

In-text attributions and inline citations (ANSWERED)

True or False? The consensus is that an in-text attribution <EDIT> of a fact that is likely to be challenged </EDIT> requires an additional inline citation. at the end of the sentence, <EDIT> when the in-text attribution already uniquely identifies a source. </EDIT>

The answer, per WP:INTEXT is True (see final line of first paragraph of WP:INTEXT)

(and Thus the table in the previous section is correct and uncontroversial.) ---- CharlesTGillingham (talk) 10:21, 4 July 2026 (UTC)

• No idea. But whoever made the table in previous section seemed very confident that there was a consensus at the 2020 request for comments that the answer was True. ---- CharlesTGillingham (talk) 16:16, 3 July 2026 (UTC)
• Unsure about the basic idea, but we certainly shouldn't say "at the end of the sentence" since the endnote numeral may be placed at some other point in the text. Also, any consensus that might exist does not apply to attribution within endnotes, no matter whether they are citations or explanatory endnotes. Jc3s5h (talk) 16:49, 3 July 2026 (UTC)

  Of course. Fixed --- ~2026-38138-45 (talk) 19:54, 3 July 2026 (UTC)

• This is basic WP:V. An inline citation is needed if and only if the fact or claim that the in-text attribution supports is "challenged or likely to be challenged". Otherwise, it's not necessary.—S Marshall T/C 20:26, 3 July 2026 (UTC)

  The question at issue is whether an in-text attribution counts as a form of inline citation. If it does, WP:V is satisfied, regardless of a second inline citation. Note that "in-text attribution" is listed as type of citation in the first section of the guide.

  I edited the question to avoid this confusion going forward. Fixed ---- CharlesTGillingham (talk) 22:16, 3 July 2026 (UTC)

  In-text citations are deprecated. Remember? So if a citation is needed, the in-text attribution by itself can't provide it. Gawaon (talk) 03:34, 4 July 2026 (UTC)

  We're talking about in-text attribution, not in-text citations. As described in the second paragraph of the first section of the guide, and the section WP:INTEXT. ----CharlesTGillingham (talk) 10:02, 4 July 2026 (UTC)

The OP asked the same thing over at Wikipedia talk:Inline citation, and the only accurate answer is it's more complicated than that. We don't need a little blue clicky number saying "Act III of Hamlet" if the sentence says "In Act III of Hamlet, Hamlet says 'To be, or not to be, that is the question'."

• Does this require an inline citation? Obviously yes, because it's a direct quotation.
• Does it require an inline citation in the form of a little blue clicky number? No, that's not the community's practice.

In this particular case (but not in the more common cases), the WP:INTEXT attribution is the only inline citation we need. It's silly to have "Act III of Hamlet" at both the start of the sentence and at the end of it, and it would be a WP:NOTBURO violation to insist upon it.

I need to make some time to get back to the Wikipedia:Inline citation page and see how much still needs to be cleaned up. Last I checked, almost everything except little blue clicky numbers had been lumped in with WP:PAREN as his idea of "deprecated". (Any interested editor should feel free to beat me to it.) WhatamIdoing (talk) 04:27, 18 July 2026 (UTC)

I had a discussion with @Gawaon above in #In-text attribution, it seems there is a lot of disagreement about this issue as we now have 2 editors supporting each side of yes/no. Maybe we need another RFC. :-) Mathnerd314159 (talk) 01:38, 19 July 2026 (UTC)

Wikipedia:Village pump (proposals)/Archive 171#Deprecate parenthetical citations was about parenthetical citations, which would look like this:

“ Hamlet says 'To be, or not to be, that is the question' (Shakespeare, Hamlet, Act III). ”

Parenthetical citations don't apply to sentences that contain no parentheses ("brackets" in British English). This:

“ In Act III of Hamlet, Hamlet says 'To be, or not to be, that is the question'. ”

is (a) not deprecated by the 2020 RFC linked above and (b) not in need of a little blue clicky number at the end, because anyone who reads the sentence already knows what the source is. Adding a little blue clicky number that redundantly repeats exactly the same content, like this:

“ In Act III of Hamlet, Hamlet says 'To be, or not to be, that is the question'."

”

would basically be silly. Educated people know how to find a copy of Hamlet all by themselves; they don't need someone to give them the name of a publisher.

The key thing to know, however, is that most of the time, you're not quoting Shakespeare (or Plato or the Bible or any other extremely famous work that educated people should recognize). Most of the time, in-text attribution sounds more like this:

“ Historian I.M. Portant from Big University said "views have changed over the centuries" ”

– and that does require a full citation at the end. WhatamIdoing (talk) 02:14, 19 July 2026 (UTC)

In your example, would you suggest that "Historian I.M. Portant in her 2019 article 'Big Significant Findings' said 'views have changed over the centuries'" does not require further citation? Nikkimaria (talk) 02:18, 19 July 2026 (UTC)

It's not enough information to find the article. A full description of an article would require writing the name of the author, date, article title, and magazine/journal into the sentence. A sentence such as:

• "Historian I.M. Portant in her 2019 article 'Big Significant Findings' in the Journal of Important Things said 'views have changed over the centuries'."

would be sufficient – it meets the WP:V goal of identifying the source and the CITE statement that "what matters most is that you provide enough information to identify the source. Others will improve the formatting if needed" – but it also would be bad style, so it's against the MOS (e.g., in WP:MEDSAY) and someone would need to "improve the formatting", preferably by reformatting nearly all of that description into {{cite journal}} inside a little blue clicky number. WhatamIdoing (talk) 02:30, 19 July 2026 (UTC)

I'd suggest that in the majority of cases an author-year-title combo would be sufficient to uniquely identify the source, but thanks for clarifying. Nikkimaria (talk) 02:36, 19 July 2026 (UTC)

If the source is easily available online, it probably would be enough. But if the article were a bit older, especially if it was published in an offline or now-defunct periodical, then author-year-title is probably not enough. WhatamIdoing (talk) 02:51, 19 July 2026 (UTC)

Thanks for the explanation, WhatamIdoing! I was just going to write that I doubt there's a real disagreement here. I, too, would agree that no citation footnote is needed for the Hamlet case – here the inline attribution is entirely sufficient, since it unambiguously identifies both work and location within the work, and it's a natural way to express this. But that's no longer the case once author, year and/or title, and possibly page number are required, as is the case for nearly all newer and less well‑known works. So, for the question "How do I cite this 2021 paper from two largely unknown academic authors?" the Hamlet example is simply irrelevant. Different problem, different solution, and almost certainly one that'll involve a footnote. Gawaon (talk) 02:35, 19 July 2026 (UTC)

IMO the relevance of the Hamlet example is about the importance of not screwing up the written rules by writing them as if the non-famous examples were the only ones that ever happened in any of our 7,220,331 articles. Whatever we write on this subject needs to work for the common situation, but it also needs to work for WP:PLOTSUM content and the occasional Shakespearean quotation, too. WhatamIdoing (talk) 02:56, 19 July 2026 (UTC)

Citing multiple pages of the same source with subreferencing

the confusion mentioned in § Citing multiple pages of the same source does not exist when using details= ((subreferencing). The text should clarify the use of ibid in conjunction with subreferencing.

Many people<ref name=aardvarks>{{cite book
 | title   = My favorite aardvark
 }}
</ref> have written about aardvarks, but my favorites are foo<ref name=aardvarks details={{cite book
 | title   = ibid
 | section = How to cook an aardvark
 | quote   = A meat thermometer is essential.
 | author  = foo
 }}
" /> and bar.<ref name=aardvarks details={{cite book
 | title   = ibid
 | section = How to train an aardvark
 | author  = bar
 }}
" />

Since sections show directly under the book citation, the referent of ibid is clear.-- Shmuel (Seymour J.) Metz Username:Chatul (talk) 16:35, 9 July 2026 (UTC)

<EDIT>

@Chatul: The code you offered above does not seem to work, and I'm not sure what you meant to write. --- CharlesTGillingham (talk) 23:08, 10 July 2026 (UTC)

I don't think subreferencing has been implemented yet. -- LCU ActivelyDisinterested «@» °∆t° 13:09, 11 July 2026 (UTC)

Works at test.wikipedia.org if the value assigned to the details= attribute is properly quoted. But, using a cs1|2 template in details= is not good practice. In this example, {{cite book}} creates COinS metadata for two books both titled ibid each written by a different author but both missing crucial bibliographic details that give book metadata meaning. For those readers who consume wikipedia references using reference management software there is no connection between My favorite aardvark and either of the two ibid 'books'.

Because subreferencing renders sub-indexed 'details' ([1.1] is a subindex of [1]) below and indented from the main reference, the use of ibid., ib., loc. cit., op. cit. is unnecessary extraneous stuff.

I suspect that for chapter subref details of a cited book, we might want to create a specialized template to serve that purpose {{cite book subref}} or some-such; much like {{harvc}} for example.

—Trappist the monk (talk) 15:07, 11 July 2026 (UTC)

Better yet, a |subref=yes parameter making |title= optional. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 11:53, 13 July 2026 (UTC)

We've seen community members experimenting with templates and sub-referencing during testing on betawiki and mentioned one example on our project page: m:WMDE Technical Wishes/Sub-referencing#Templates and sub-references. We've also shared some learnings in our pilot wiki report . Personally I agree that creating a specialized template might be the better option compared to making changes to {{cite book}}. Johannes Richter (WMDE) (talk) 07:22, 14 July 2026 (UTC)

I will only address one sentence in the original post, "the text should clarify the use of ibid in conjunction with subreferencing." Since subreferencing is not implemented as of 11 July 2026 on the real English Wikipedia, the "Citing sources" should not say a single word about it. Jc3s5h (talk) 16:15, 11 July 2026 (UTC)

Info: We haven't set a date yet (see phab:T414094 for our deployment plan), but we'll probably deploy sub-referencing to English Wikipedia by end of August / beginning of September, unless enwiki community members would like us to speed up the process for this project. Johannes Richter (WMDE) (talk) 07:28, 14 July 2026 (UTC)

Thanks for the update. Please don't speed it up unnecessarily, the best outcome is that it is implemented well and with helpful documentation. CMD (talk) 10:27, 14 July 2026 (UTC)

Markup	Renders as
{{#parsoid	Many people have written about aardvarks, but my favorites are foo and bar. References

Hatnote bonanza

Hello User:Gawaon. I'm trying to simplify the mess of hatnotes at the top of this page, as it's intimidating and we want newer editors to not be overwhelmed at each step of their editing journey. You reverted my edit removing the hatnote on citing Wikipedia. By what route do you think people come to this page accidently? If you Google "citing Wikipedia" the correct page is on top. Many of the other hatnotes are also not necessary imo.

• For information on referencing citations in Wikipedia articles, see Help:Footnotes and Wikipedia:Inline citation.
  • It doesn't make clear how these related pages differ from citing sources. Do we need both? We already link well to help pages for citations.
• For information about citing Wikipedia articles for use in work outside of Wikipedia, see Wikipedia:Citing Wikipedia.
  • See above
• "WP:CITE" and "WP:REF" redirect here. For the Citation Needed information page, see WP:CITENEED. For the reference desk, see WP:REFD.
  • I don't think people would expect the citation needed page here, so I would just keep the ref
• "WP:CS" redirects here. For the policy on clean start, see Wikipedia:Clean start. For the computer science WikiProject, see Wikipedia:WikiProject Computer science.
  • Fine
• "WP:UNCITED" redirects here. For the policy on whether or how quickly to remove uncited material, see WP:BURDENWAIT.
  • Fine.

Can we cut the hatnotes by half? Will still be intimidating, but not as much. In solidarity, —Femke (talk) 🐦 21:32, 9 July 2026 (UTC)

This page has been the example I use of what not to do. Very frustrating for anyone new to have to read through a paragraph of notes before they even know what this page is all about....... Perhaps do what we do for noticeboards and collapse the nav template underneath all the intro templates. Moxy🍁 23:27, 9 July 2026 (UTC)

Hi Femke, fair point! I've reverted my revert. I also would not be opposed to removing some of the redirect notes, but I can't judge how needed they are. Gawaon (talk) 02:37, 10 July 2026 (UTC)

Embedded links

Is Wikipedia:Citing_sources#Avoid_embedded_links simply trying to say avoid external links in the body of the article? There is a lot of waffle, and I'm not sure I'm missing anything, with that quite weird link to the guardian. Might be a section that can be halved. In solidarity, —Femke (talk) 🐦 06:50, 10 July 2026 (UTC)

It seems to be conflating two different things: don't use inline-to-article-text links as a citation style, and don't use bare urls as citations. The discussion of the ancient history of Wikipedia is also not particularly useful. —David Eppstein (talk) 06:58, 10 July 2026 (UTC)

It's trying to say don't do this ([https://example.com]) or this ([https://example.com this]) in article prose, and that external links should either be in ref tags or the external links section. I'm not sure the history of Wikipedia is needed to explain that. -- LCU ActivelyDisinterested «@» °∆t° 12:28, 10 July 2026 (UTC)

Bookmarklets

There is a reference to bookmarklets in Wikipedia:Citing_sources#Preventing_and_repairing_dead_links. I had never heard of these things and wonder if people actually use them? Can it be removed (I'm just trying to make this page a bit less intimidating). In solidarity, —Femke (talk) 🐦 20:56, 10 July 2026 (UTC)

I still have a bookmarklet in my active browser bookmarks (not one of these ones). They still do exist and work. In case you're unfamiliar: they're a way of making things that look like bookmarks that run Javascripts. I think for Wikipedia purposes most people instead use custom scripts that put actions into the Wikipedia toolbar. —David Eppstein (talk) 21:32, 10 July 2026 (UTC)

FYI: Bookmarklet. Apparently they've died and been resurrected at least once, so I'm not surprised you're not familar with the concept. —ClaudineChionh (she/her · talk · email) 00:18, 17 July 2026 (UTC)

Template:R - quote related feature enhancement discussion

May be of interest here: Template talk:R#Easy ready enhancement - Display quotation when no locator is supplied — Very Polite Person (talk/contribs) 17:22, 15 July 2026 (UTC)

Is this an OK way to cite several patents in a bundle

1. https://en.wikipedia.org/w/index.php?title=Amanda_Jones_(inventor)&oldid=1364631313#cite_ref-Jones_patents_1873_26-0
2. https://en.wikipedia.org/w/index.php?title=Amanda_Jones_(inventor)&oldid=1364631313#cite_note-Jones_patents_1873-26

Here is the reference in-line and down in Refs. — Very Polite Person (talk/contribs) 23:33, 17 July 2026 (UTC)

{{multiref}}

—Trappist the monk (talk) 23:43, 17 July 2026 (UTC)

Yes, {{multiref}} is your friend. Gawaon (talk) 02:25, 18 July 2026 (UTC)

What you've done is acceptable, but feel free to use other formats if you'd like. Also, I congratulate you on the first ref to WP:PATENTS I've seen in a couple of years that didn't violate either WP:PRIMARY or WP:RS. WhatamIdoing (talk) 04:33, 18 July 2026 (UTC)

CITEVAR and sfn templates

Wikipedia:Dispute resolution noticeboard#Tetragrammaton is a dispute over adding {{sfn}} templates to an article over the objections of the other editor. The comments suggest that the editors are finding WP:CITEVAR to be unclear on this point. I suggest therefore that we add, to the list of "to be avoided", an explicit statement, such as:

• adding {{sfn}} templates to an article that already uses a consistent system without those templates, or removing the {{sfn}} templates from an article that uses them consistently;

Any objections? WhatamIdoing (talk) 20:53, 19 July 2026 (UTC)

Sounds like it's needed, the issue seems to stem from CITESTYLE. It says "A number of citation styles exist, including APA style, ASA style, MLA style, The Chicago Manual of Style, the Vancouver system and Bluebook", which is an incomplete picture of what editors mean when the say "cite style". -- LCU ActivelyDisinterested «@» °∆t° 21:01, 19 July 2026 (UTC)

In my opinion the sentence A number of citation styles exist, including APA style, ASA style, MLA style, The Chicago Manual of Style, the Vancouver system and Bluebook. does not belong in this article. It's not relevant to Wikipedia citation style. The sentence should list the major specific citation styles actually used in Wikipedia. Johnjbarton (talk) 02:22, 20 July 2026 (UTC)

I agree it would be better if it described Wikipedia practices. -- LCU ActivelyDisinterested «@» °∆t° 16:21, 20 July 2026 (UTC)

However, Wikipedia's own practices are based on and closely informed by these and similar styles, and they are clearly meant to guide the "Nearly any consistent style can be used" that follows a bit later. (These and similar styles can be used, but there's no blanket permission to invent new ones.) Gawaon (talk) 03:18, 21 July 2026 (UTC)

@Gawaon, as the person most responsible for CITEVAR's creation 15 years ago, let me assure you that I actually did not mean that external style guides were supposed to guide anything. I'm sorry if this surprises you or makes you think that we were somehow less than perfect back in the day, when the guidelines were handed down on stone tablets, but I actually meant something closer to "If you can somehow manage to get consensus on that article's talk page for turning all the citations purple in that article, then congratulations: CITEVAR is going to protect your nonsense, so y'all quit edit warring over it already."

The "nearly any consistent style" language was introduced a few months later by @Jc3s5h, who reasonably wanted to point out that some elements of a citation style might be prohibited for other reasons. Specifically, he wanted CITEVAR to clearly state that parts of Wikipedia:Manual of Style/Dates and numbers#Formats apply to citation styles, because 9/8/2011 doesn't work for an international audience (is that August 9th or September 8th?). MOS:DATENUM is not the only policy or guideline that could cramp someone's style (e.g., an editor can't have a "citation style" of violating copyrights through excessively long quotations), but I promise that "it needs to be similar to one of the listed external styles" is not a rule, goal, or even fond wish for CITEVAR. CITEVAR exists to prevent and stop stupid edit wars. If you can stop a stupid edit war by getting consensus to have the citation style be Page 2 of a 2018 Place University Press publication, "Important Book" by John Doe. format, then CITEVAR will support that. WhatamIdoing (talk) 04:54, 21 July 2026 (UTC)

My worry is rather the other way around: if you start with something consistent but unreasonable (like that invented example or "no short citations ever!"), how do you prevent that unreasonableness from persisting? You might say something like "Talk page consensus will quickly override it", and that's probably true if, say, four or more editors get involved, but in my experience, that's often not the case. Often there are just two, one in favour of the status quo, one in favour of changing it, and in that case the status quo usually wins, since there's "no consensus for a change". So if CITEVAR is indeed as open to purely invented and unprecedented styles as you seem to suggest, how would you prevent such a scenario? Gawaon (talk) 05:51, 21 July 2026 (UTC)

CITEVAR does not care if someone thinks it's "unreasonable". CITEVAR cares about whether there are edit wars.

Consensus for citation styles is formed the same way as any other consensus. If consensus cannot be formed either way, then CITEVAR imposes a default (see the line about "defer to the style used by the first major contributor"; see also WP:ENGVAR and WP:STYLEVAR for similar processes, and note that it's not a case of "status quo wins". Also, you may wish to read WP:STATUSQUO, whose contents do not match what some editors claim it says). If you like the specified default, then I guess you could sit down and do your best to stonewall the decision. If you don't, then start at the top of Wikipedia:Dispute resolution and work your way through all the relevant options. None of this should really be surprising or overly daunting to an editor with your level of experience. WhatamIdoing (talk) 06:14, 21 July 2026 (UTC)

I have no major problems with CITEVAR, but I might continue to object to proposed additions to CITEVARNO that I consider unnecessary or unhelpful (such as the one you'd originally suggested above). Gawaon (talk) 10:38, 21 July 2026 (UTC)

Maybe a nutshell could be added to CITEVAR along the lines of "Don't change how references are done in an article without first getting consensus". -- LCU ActivelyDisinterested «@» °∆t° 11:44, 21 July 2026 (UTC)

How would that differ from what CITEVAR already says? And WP:CITEVARYES explicitly allows some changes, such as adding details, merging repeated references, and removing inconsistencies and errors. Gawaon (talk) 12:54, 21 July 2026 (UTC)

Because if we put it in a big box with red, blinking text, maybe people would understand the point. WhatamIdoing (talk) 17:13, 21 July 2026 (UTC)

I agree that's what it already says, but others are actively disagreeing. -- LCU ActivelyDisinterested «@» °∆t° 11:48, 22 July 2026 (UTC)

I'm very sceptical about the "adding" part, since sfn or similar means may simply be needed to refer back to other citations (say a different page number in an already cited book). How to resolve that if they are not allowed? If that situation hasn't arisen in an article so far, I would rather interpret it as "sfn hasn't been needed yet", not "sfn is not allowed". If there's a way to consistently resolve this situation other than "just repeat all the details that stay the same", then I could maybe be convinced that that's a consistent, sfn-free citation style – but how would it work? Gawaon (talk) 21:57, 19 July 2026 (UTC)

"Converting"? WhatamIdoing (talk) 22:46, 19 July 2026 (UTC)

What do you mean by "Converting"? I don't get it. Gawaon (talk) 02:56, 20 July 2026 (UTC)

Instead of talking about "adding sfn" to an article, maybe we should talk about "converting an article to use sfn", i.e., instead of another style for presenting sfn's content. WhatamIdoing (talk) 03:01, 20 July 2026 (UTC)

No, like I wrote in reply to Nikkimaria below: CITEVAR is about what the reader sees, not about what the editor sees. So converting to sfn or some other template format requires discussion if it changes the rendered page (placement of parentheses or similar details), but not if it doesn't. Gawaon (talk) 03:16, 20 July 2026 (UTC)

Your assertion, as Nikkimaria told you below, is not true. In particular, CITEVAR has settled the fights over whether to use citation templates vs manual formatting, and those differences are not visible to the reader. WhatamIdoing (talk) 03:21, 20 July 2026 (UTC)

Sfn isn't the only possibility to resolve that situation - you can use other templates, or even create short cites with no templates at all. Soon subreferencing will also be an option. So calling for a discussion about how best to approach that under CITEVAR seems very reasonable. I'd suggest amending "adding citation templates to an article that already uses a consistent system without templates" to include changing what templates are being used, more broadly, or alternatively changing "switching between major citation styles" to "switching between major citation styles/templates". Nikkimaria (talk) 02:10, 20 July 2026 (UTC)

But the objection in this case is any templates, no? Edit: Any template used to specify pages, anyway. The status quo is just repeating the same information over and over. I don't think that even constitutes a "consistent system". PARAKANYAA (talk) 02:13, 20 July 2026 (UTC)

If the objection is any templates, the existing wording covers that. But it looks like the dispute in question isn't that.

The status quo is just repeating the same information over and over. I don't think that even constitutes a "consistent system". You might not feel it's the best system, but it could be consistent to repeat information. (Speaking generally; I'm not sure that's the case in the dispute in question.) (Edited). Nikkimaria (talk) 02:18, 20 July 2026 (UTC)

Does this reading of CITEVAR not prohibit adding page numbers? If an article consistently cites the full book range without pages, that is a "consistent system" and we cannot change it. Adding that would be prohibited, then? PARAKANYAA (talk) 02:16, 20 July 2026 (UTC)

If there are no pages, no; that would fall under CITEVARYES. If there are page ranges of some kind, I don't know that we've ever arrived at consensus about how narrow they need to be. Nikkimaria (talk) 02:26, 20 July 2026 (UTC)

Why would it fall under CITEVARYES? Pages are not required. PARAKANYAA (talk) 02:36, 20 July 2026 (UTC)

Per WP:PAGENUM. Nikkimaria (talk) 02:41, 20 July 2026 (UTC)

Which is strangely bereft of details, especially since WP:V says "Cite the source clearly, ideally giving page number(s)—though sometimes a section, chapter, or other division is appropriate instead; see Wikipedia:Citing sources for details of how to do this". WhatamIdoing (talk) 02:44, 20 July 2026 (UTC)

"which part" is not pages per se. In fact it says page numbers are "not required"; "Page numbers are not required for a reference to the book or article as a whole". Could be chapters in books with different authors. Many books don't have pages. What I am getting from this is in principle adding pages can and in fact usually is a citevar violation. PARAKANYAA (talk) 02:48, 20 July 2026 (UTC)

When page numbers or equivalent means of identifying a portion of a book are appropriate to include, then they should be included. "Page numbers are not required for a reference to the book or article as a whole", but that's a tolerably rare occurrence. We do sometimes refer to a book as a whole ("Rachel Carson argued in her book Silent Spring that the pesticide DDT was causing environmental damage"), but usually we're citing a book for a much smaller statement ("DDT was first synthesized in 1874"), and that portion should be identified properly, with page numbers or their equivalent. This is not optional, and the few editors who have claimed that they can cite a whole book for something that's on a single page, because anyone who really wants to know if they're putting lies in an article can just spend a couple of hours reading the whole book – times however many books they cited – have not been successful in arguing that adding page numbers is an unreasonable imposition on their rights as an author, or their right to choose a citation style unencumbered by page numbers. WhatamIdoing (talk) 03:17, 20 July 2026 (UTC)

Maybe it should be that way but I am not sure that is actually the case in practice. PARAKANYAA (talk) 03:19, 20 July 2026 (UTC)

If you can find even one FAs that has an accepted style of citing books without giving relevant page numbers anywhere in the article, then I'll propose a change to the WP:PAGENUM guideline that says "Page numbers are optional if you call it a 'citation style' " myself. WhatamIdoing (talk) 03:23, 20 July 2026 (UTC)

No, generally forbidding page numbers is indeed not a consistent citation style, and CITEVAR already says so: "An article where all or most of the citations fail to provide needed bibliographic data does not have a consistent citation style, and can be changed freely to insert such data. Such data includes ... page numbers (where relevant)." Gawaon (talk) 03:35, 20 July 2026 (UTC)

The consensus we arrived at is that page ranges for books should generally be as narrow as possible, keeping in mind three important facts:

1. Sometimes "as narrow as possible" is the entire book.
2. If you're asking for a page range shorter than a couple of pages in a book, then there are more important things you could do with your life.
3. A scholarly journal article is not a book. The correct and normal way to cite a journal article is to give the full range of pages in the bound volume, even if a smaller page range could be given.

WhatamIdoing (talk) 02:43, 20 July 2026 (UTC)

Four. Four important facts:

4. Some books are in formats that do not have page numbers, and when that happens some other way of specifying as narrow as possible a position within the book should be found.

(I do agree with everything else you say here though.) —David Eppstein (talk) 06:41, 20 July 2026 (UTC)

Yes, I agree, "just repeating the same information over and over" is very much NOT a consistent and complete citation style – it simply has a gap, namely in not yet covering how short citations are handled. Filling such a gap is allowed without prior discussion. (Which is not an argument against seeking such a discussion, which may still be a good idea – but not a requirement per CITEVAR.) Gawaon (talk) 03:10, 20 July 2026 (UTC)

1. John Doe (2018). Important Book. Place University Press. ISBN 9781111111111. p. 1.
2. John Doe (2018). Important Book. Place University Press. ISBN 9781111111111. p. 2.
3. John Doe (2018). Important Book. Place University Press. ISBN 9781111111111. p. 3.

Those citations are consistent and complete. You could argue that short cites would be of benefit, but it wouldn't be for reasons of consistency. Nikkimaria (talk) 03:38, 20 July 2026 (UTC)

You can write it like that if you really want (though why would you?), but you can't expect other editors to do the same. Also, the consistency in that style is very much threatened – an editor might notice that the ISBN is actually 9781111171121, correct it in one or two places, but overlook the other(s). Gawaon (talk) 03:54, 20 July 2026 (UTC)

Consistency in any style is fragile, and often not followed by other editors. Nikkimaria (talk) 03:58, 20 July 2026 (UTC)

You can't expect other editors, especially new or inexperienced editors, to do the same, but "what matters most is that you provide enough information to identify the source. Others will improve the formatting if needed". CITEVAR won't let you unilaterally, in the face of active opposition, convert the style of an article to what you believe is the One True™ Style, but it does allow one-off mistakes to be corrected. WhatamIdoing (talk) 04:00, 20 July 2026 (UTC)

But note that consistency is not enough. I don't think anything in WP:Citing sources was meant to read as suggesting that editors could simply invent a new citation style and then force everyone else to follow it, by rejecting all suggestions to change it. See especially WP:CITESTYLE which refers to existing citation styles, listing several of them. It also says "Wikipedia merged several formats into Citation Style 1 and Citation Style 2", which suggests that merged or hybrid forms are allowed – it's not necessary to strictly follow any published style guide in every detail. But that's not a blanket permission for invention without precedent. Consider if it were otherwise. Then an article's style could be that, instead of "John Doe (2018). Important Book. Place University Press. p. 2.", all citations must have the form:

Page 2 of a 2018 Place University Press publication, "Important Book" by John Doe.

Provided all citations follow that pattern, that would be a consistent style. Or an article could have a Bibliography section, but sorted by the last author's first name instead of the first author's last name; or grouped first by publisher, then by year, and only then by author. That too would be a consistent style.

But are such styles covered by CITEVAR, forcing other editors to honour them? No, and not for lack of consistency (which is indeed there), but for lack of precedent outside of Wikipedia. They are invented styles, while CITEVAR is meant to decide between established styles (including minor variants and hybrids of them).

Nikkimaria's hypothetical "just repeat every detail in every footnote" style is equally invented – it's not a style any major style guide or publisher would recommend or follow. As such, it fails the "established style" test and editors are free to improve on it, reducing the redundancy and bringing it more in line with actual, established styles. Gawaon (talk) 05:38, 20 July 2026 (UTC)

CITEVAR actually is "meant to read as suggesting that editors could simply invent a new citation style". They cannot force other editors to follow that style merely "by rejecting all suggestions to change it". However, changing away from that style – any style – requires those "other editors" to actually have a chat on the talk page that demonstrates a consensus to change it over the objections of the inventor. WhatamIdoing (talk) 06:17, 20 July 2026 (UTC)

Note: "Established" means that it's been used in that Wikipedia article in all/almost all of the sources cited on the page. We don't need "precedent outside of Wikipedia". If we did, then we'd have to stop using all of the popular citation templates, because none of them have "precedent outside of Wikipedia" or conform to any external style. WhatamIdoing (talk) 06:19, 20 July 2026 (UTC)

Like I said, minor or merged variants of established styles are unproblematic, and CITESTYLE itself describes CS1 and CS2 as such merged variants. Their own outputs differ only in minor details, such as which punctuation is used. And both are variants of a style that's used in millions of books – whether used in a References section or in footnotes, you'll easily find books whose footnotes or References/Bibliography sections will look so similar that you'll be hard-pressed to find major differences. Rather than being new inventions for Wikipedia, they are both standardized variants of a citation style that has been common in the English-speaking world (and outside of it) for many decades. Gawaon (talk) 09:26, 20 July 2026 (UTC)

No, CITEVAR is about styles, i.e. about what the reader sees. It's not about the use or non-use of templates as such, since templates are invisible to the reader. A requirement that says "(a) short citations must look such-and-such and (b) they must be handwritten, using a template to produce the same effect is not allowed" is not a citation style. The (a) part is the style and it's fine as such, but (b) is unrelated to styling (since readers won't see the difference) and just an attempt to make other editors' lives unnecessarily hard. Gawaon (talk) 03:05, 20 July 2026 (UTC)

Two of the three points under CITEVARNO are about things that are (mostly) invisible to the reader. Nikkimaria (talk) 03:09, 20 July 2026 (UTC)

Yeah, in my view the "adding citation templates to an article that already uses a consistent system without templates" part is a mistake and should simply be removed. It doesn't follow from CITEVAR itself (not a style issue) and it's totally unclear what detrimental effect templates could have. The "removing citation templates from an article that uses them consistently" part makes sense despite not being a style issue as such, since it would almost certainly make the style less consistent over time. Editors would struggle to correctly format details that were previously handled by templates, so it would pose an unnecessary burden to editors and lead to a loss of style consistency over time. Gawaon (talk) 03:26, 20 July 2026 (UTC)

You're welcome to that view, but at this point the inclusion of template/no template in CITEVAR is very well established. (Plus templates aren't a magic bullet). Nikkimaria (talk) 03:38, 20 July 2026 (UTC)

Be that as it may, as I understand it, one side wasn't arguing against the sfn template in particular, but against the use of short references at all. That's not covered by CITEVAR and related sections, and every serious citation style ever used in academia or elsewhere has had some means of achieving them. Gawaon (talk) 03:48, 20 July 2026 (UTC)

I don't see any reason why it wouldn't be covered - it's even reader-visible. And given that we have multiple ways of achieving multiple citations with different pagination, discussion seems prudent. Nikkimaria (talk) 03:58, 20 July 2026 (UTC)

I don't think that the Vancouver system uses short refs. https://www.nature.com/nature/for-authors/formatting-guide also doesn't say anything about short refs. From a quick poke around a few sites, it doesn't seem to be a thing in the hard sciences. WhatamIdoing (talk) 05:12, 20 July 2026 (UTC)

It rather seems that in these styles, all references are short – just a number. And indeed there doesn't seem to be an explicit mechanism to refer to a different page number/range in the same publication, but my understanding is that you would list the publication just once, referring to it multiple times without specifying exact page(s). I guess that's considered acceptable in these publications, since citations are mostly to other journal articles which will usually be short. But it's not a precedent for "just repeat the details multiple times" (each cited publication is listed just once) and we have already established that "just omit the specific page ranges, and good luck to readers trying to locate the exact source" is not a style accepted by CITEVAR (it's explicit about forbidding that). Gawaon (talk) 06:13, 20 July 2026 (UTC)

If you were citing a book (uncommon, but it does happen) in these styles, you'd probably only be citing a single chapter or a single part. If you need to cite multiple parts of the same book, AIUI you would repeat the full citation and just change the chapter/page numbers.

The question of repeating full citations separately every single time a source is cited came up on this page maybe ~15 years ago. This was literally a little blue clicky number to a short news article, and then if he cited the same short news article later in the article, he wanted another/separate little blue clicky number with exactly identical contents. He turned up here because a bot (or AWB script?) kept merging the identical refs into unified named ref. I don't think we were very sympathetic, but technically that is a citation style, and iff he managed to get consensus to do his weird thing, then it would be allowable under WP:CITEVAR. WhatamIdoing (talk) 06:27, 20 July 2026 (UTC)

I just noticed a detail all of us might have overlooked so far: WP:CITEVARYES explicitly refers to WP:DUPREF, which says: "Citations to different pages or parts of the same source can also be combined (preserving the distinct parts of the citations), as described in Help:References and page numbers." So partial merges (remove the duplicated part, keep just the differences, likely different page ranges) are already explicitly allowed by CITEVARYES, and if the question was "can a page-local style forbid them?", the answer is a solid no.

If there's already a page-local style on how such partial merges are handled (say via {{sfn}} or, in the future, via subreferences), then it has to be respected per CITEVAR, or consensus sought for a change. If not, any suitable style should be acceptable, since none will be a violation of an already established style for handling such situations. Gawaon (talk) 13:10, 21 July 2026 (UTC)

Of course, Wikipedia:Citing sources is a guideline, and guidelines can't be overwritten by local consensus per CONLEVEL. So DUPREF ("Citations to different pages or parts of the same source can also be combined") would be valid in all articles even if CITEVARYES didn't refer to it – it's not a question of style, it's a guideline. But the additional mention in CITEVARYES can't hurt, and it sure helped me find it. Gawaon (talk) 16:48, 21 July 2026 (UTC)

If you want to combine these:

• Expert, Alice (2001). The Sun is Really Big. Academic Press. p. 15.
• Expert, Alice (2001). The Sun is Really Big. Academic Press. p. 50.

into these:

• Expert, 2001. p. 15.
• Expert, 2001. p. 50.
• Expert, Alice (2001). The Sun is Really Big. Academic Press.

then you have to get consensus on the talk page first. DUPREF is not a get-out-of-CITEVAR-and-annoy-people-free card. WhatamIdoing (talk) 17:17, 21 July 2026 (UTC)

I think you could reasonably combine them into

• Expert, Alice (2001). The Sun is Really Big. Academic Press. pp. 15, 50.

without CITEVAR coming into play. But if the same book is being cited for many different pages, this solution would lose specificity for individual citations. —David Eppstein (talk) 18:24, 21 July 2026 (UTC)

I have used that style, and it really upset an editor once. The Google Books link went to the first page, and he reverted because the information was on the second cited page. Given how mad he was, I assume he was feeling very stupid. WhatamIdoing (talk) 02:51, 22 July 2026 (UTC)

In such cases I sometimes put the gbooks link on the page numbers, not on the book title. —David Eppstein (talk) 05:27, 22 July 2026 (UTC)

I considered removing the links entirely. Readers aren't likely to use them (it's not exciting content). WhatamIdoing (talk) 16:36, 22 July 2026 (UTC)

I could also combine them into

• Expert, Alice (2001). The Sun is Really Big. Academic Press. p. 15.
• Expert 2001, p. 50.

without needing to involve the talk page first. If somebody objects and prefers a different manner of combining them, they might revert me and/or open a talk page discussion, and then we'll discuss. But CITEVAR doesn't magically revoke everything else our guidelines say; it's not a super-guideline that beats all the others. In my understanding (and in actual practice), the page-local citation style described by CITEVAR is about filling the gaps: citation details not already regulated elsewhere in our own guidelines. DUPREF's permission to combine partially redundant citations is not revoked by CITEVAR – but if there's already a locally established style on how to do so, you have to respect it, or seek consensus for a change first. Seen like that, there's no conflict between these different parts of our guideline; they complement each other harmoniously. Gawaon (talk) 02:05, 22 July 2026 (UTC)

No, that change is one that we expect to be discussed. WhatamIdoing (talk) 02:48, 22 July 2026 (UTC)

You do, but why? CITEVARYES explicitly says the opposite, describing "combining duplicate citations (see § Duplicate citations, above)" as "standard practice".

Also, why do you assume bad faith? An editor starting to combine partially or wholly duplicate citations is very likely not trying to "annoy" other editors, but simply to make the article better. At least you should assume so unless you've clear, specific proof that they aren't. Gawaon (talk) 03:26, 22 July 2026 (UTC)

But it's not a case of "combining duplicate citations", because:

(a) They aren't actually "duplicates", and

(b) The result isn't two citations being turned into one citation. Instead, it's two citations being turned into two citations, and one of those is in a different style. WhatamIdoing (talk) 03:47, 22 July 2026 (UTC)

However, such combining of partially duplicate citations is described in the "Duplicate citations" section; CITEVARYES refers to that section, and nothing in either section suggests that the permission applies only to precisely duplicated citations. As for the "different style", I don't buy that, since it would mean that many of our articles actually use two citation styles, one for long and another for short citations. They might look different, granted, but I'll maintain that they actually aren't two styles, but just parts of the same style. Just as journal and newspaper articles will likely be handled somewhat differently from book references, both are likewise elements of the same style, not different styles.

Hence "any method that is consistent with the existing citation style (if any) may be used" when combining partially duplicated references, just as DUPREF says. As long as that specific detail of the overall style hasn't been established yet, that gives you considerable freedom. Though, of course, other editors may dispute your specific choice, and then the discussion will start, as usual. It's the same thing as when you add the first book reference: you'll subtly extend the style, filling a gap that had existed before (how are book references handled?), but such gap-filling is normal and doesn't require a prior announcement. Gawaon (talk) 05:32, 22 July 2026 (UTC)

According to the second sentence of the second paragraph of DUPREF – which you quote selectively, leaving out the half of the sentence about seeking consensus to change the citation style – here's how it works:

• If there is no established citation style, then someone should establish one. (There is nothing special about DUPREF for this; this is ordinary CITE rules.)
• If there is an established citation style and that established citation style is adequate for handling partially duplicated references, then you should follow the established citation style. This includes, BTW, a citation style that says not to merge the refs, but instead to leave each one as a separate full citation. That approach is often preferred in high-traffic/high-dispute articles, so that removing a sentence and its source doesn't break the rest of the sources. (There is nothing special about DUPREF for this; this is ordinary CITEVAR rules.)
• If there is an established citation style and short citations aren't consistent with the existing citations style and you'd prefer to use short citations, then you have the option of trying to obtain consensus for a change. Note that it says "consensus can be sought", and not "consensus must be sought"; this is because there is no requirement to use short citations in any article, ever. (There is nothing special about DUPREF for this; this is ordinary CITEVAR rules.)

WhatamIdoing (talk) 16:54, 22 July 2026 (UTC)

So it's a CONLEVEL issue. The guideline (DUPCITES) says:

• "Citations to different pages or parts of the same source can also be combined (preserving the distinct parts of the citations)"

So let's assume the page-local style says:

• Citations to different pages or parts of the same source cannot be combined. Just duplicate everything except for the part that's different.

So who wins, the guideline or the page-local style?

My understanding is that the guideline wins, per CONLEVEL: "Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale." DUPCITES has wiki-wide community consensus; the page-local style has only local consensus.

Now there's the second DUPCITES sentence you've quoted: "Any method that is consistent with the existing citation style (if any) may be used, or consensus can be sought to change the existing style."

In my understanding that's about the how, not about the whether. The permission already granted is unconditional. Were it otherwise, I would expect the first sentence to say so explicitly, stating something like: "Citations to different pages or parts of the same source can also be combined, provided the page-local citation style offers a method to do so." But it doesn't make such a restriction. Hence the consequences of "there's no such established citation style or method yet" simply mean that a suitable method needs to be found and used (which may or may not involve a talk page discussion), but not that the general "right to combine" already granted in the first sentence is now locally revoked.

WhatamIdoing, could you explain why you seem to think that the usual CONLEVEL order should in such cases be reversed, with DUPCITES's general permission being revoked per merely local consensus? Gawaon (talk) 05:00, 23 July 2026 (UTC)

I think that you don't understand CONLEVEL. CONLEVEL is not when the regular editors of an article form a consensus to use a citation style that you don't think is the best one. CONLEVEL is when WikiProject Composers said that other editors aren't allowed to put infoboxes in "their" articles, and eventually we told them to knock it off. (That multi-year dispute was well before your time, but it was the main impetus for developing CONLEVEL into its present form.)

Did you happen to grow up with an adult who, if you asked if you "can" go outside, would reply that "you can, but you may not" or "I don't know if you can, but you may not"? DUPREF says you "can" do this. The local citation style that you describe says that even if you "can", you still "may not" – unless you get consensus to change the style for that article.

I don't know what you want at this point. Do you want me to re-write the guideline to be unambiguously clear that this guideline does not endorse imposing any form of short citations on an article against the consensus of editors working on that article?

Or we could just add a line that says "No matter what any other policy, guideline, or editor says, if someone tells you that your edits amount to changing the citation style, then they're right (even if you don't think it really ought to count as a different style, and even if you think there's a rule somewhere that authorizes you to impose your style over their objections), and you need obtain a clear consensus for your proposed style before you revert or edit, on pain of you getting blocked for edit warring." We could even put a shortcut of WP:GAWAON next to it, so it'd be easy for you to find it. WhatamIdoing (talk) 03:07, 24 July 2026 (UTC)

Right now I think DUPCITES is fine and in no need of changes. And I think we all know that edit warring, especially if continued or repeated, is unacceptable, so there's no need to repeat that.

I would also appreciate an absence of personal attacks. Gawaon (talk) 03:20, 24 July 2026 (UTC)

WAID, one situation I've run into repeatedly (maybe three or four times) is an article I've worked on getting more viewers because it is TFA, and an editor then changing it to sfn (which I don't use). I think in every case it ended amicably, but at least a couple of times I've been told that as regular editor of the article I have no more say in the citation format than any other editor who happens to stop by to contribute, and have been out!voted. To be honest, though I don't like the outcome, I don't think there's any way to privilege regular editors over drive-by editors without violating OWN. So I think your phrasing -- "when the regular editors of an article form a consensus" -- is not usable in a policy or guideline, at least not without explaining when it is insufficient. Mike Christie (talk - contribs - library) 14:50, 27 July 2026 (UTC)

"when editors come to a consensus at an article to use a citation style ..." is probably a better way of putting it. Although I don't think that changes what WAID meant. -- LCU ActivelyDisinterested «@» °∆t° 15:06, 27 July 2026 (UTC)

I agree that that's better phrasing. Personally I don't like it when brief visitors to an article change one valid format to another valid format, but I don't think there's a way of putting the brakes on that sort of edit without violating our policies. Mike Christie (talk - contribs - library) 16:35, 27 July 2026 (UTC)

I also don't like it when an editor dedicates their time to getting the One True™ Citation Style installed at as many articles as possible. It's a WP:FAIT problem. If someone wants to install {{sfn}}, which is not used in ~97.5% of articles, then they need to discuss it first, and if it's an FA, a consensus to change needs to be pretty strong.

The problem with drive-by editors out!voting the editors who are maintaining the article is that there are three possible outcomes: the regular contributors accept the change they dislike; the regular contributors quit, and the article quality declines; the regular contributors bide their time and then have a second discussion in a few months to reverse the prior 'consensus'. All of these are 'legal' responses under our policies. None of them result in happy editors + the new citation style + article quality being maintained. Like the good, fast, and cheap triangle, when the regular contributors dislike the proposed change, we can usually only get two of the desirable ends. WhatamIdoing (talk) 22:58, 27 July 2026 (UTC)

Following up on this discussion, it seems that these are the three points of inclarity around CITEVAR:

• A 'citation style' encompasses both what is visible to readers and what is visible to editors (e.g., the use or non-use of a template).
• A 'citation style' encompasses the decision about the use or non-use of {{sfn}} and other forms of short citations in an article.
• A 'citation style' can be completely invented by a Wikipedia editor and is not limited to either popular citation templates or to styles used in various academic disciplines.

Have I missed anything? WhatamIdoing (talk) 17:23, 25 July 2026 (UTC)

The parenthesis of the first clause could include white spacing in templates as well as their usage. Personally I think crammed and roomy are crimes against the very nature of existence, but they shouldn't be changed if they are consistent. -- LCU ActivelyDisinterested «@» °∆t° 19:13, 25 July 2026 (UTC)

I agree in principle, but I'm not sure that I'd mention that in the guideline. In practice, whitespacing is defined per-template in Wikipedia:TemplateData on the template's /doc page, so most new additions won't follow any per-article standard. Consequently, most editors simply accept whatever spacing is added.

Drama over whitespacing happens every now and again, but it's mostly about someone's patience expiring rather than a serious dispute. In principle, editors shouldn't intentionally do a mass conversion from one amount/placement of whitespace to another without first discussing it, and any editor who wants to maintain a consistent level of pretty-printing need only adjust the whitespace to the locally chosen style. The drama comes from "How dare he add a citation without doing the whitespace my way! I'm so tired of cleaning up after lazy editors, especially the ones using the visual editor who can't even see that they're getting the whitespace wrong!" vs "If you want the white space arranged a particular way, then WP:SOFIXIT yourself". The latter of the two positions is the one more clearly supported by this guideline, as well as by the Wikipedia:Editing policy. WhatamIdoing (talk) 19:29, 25 July 2026 (UTC)

My experience is that the various citation repair bots do not respect preferences for whitespace in citation templates: they will leave the spacing alone in parts of the citation that they do not change but when they add or remove material in a citation they will not use exactly the same whitespace convention that was already in place. In part it is a difficult problem: how do you recognize what the convention is? And in part it is extra effort in programming the bot that is not reflected in improved outcomes for readers. The same goes for parameter ordering. So if we are going to be more strict about CITEVAR for this kind of issue, we have a problem with bots flouting CITEVAR. On the other hand, I wouldn't want to see human gnomes going around enforcing their own preference that all citations must be in crammed spaceless format (or whatever else) when it is not part of a more general reference consistency or metadata cleanup. —David Eppstein (talk) 19:30, 25 July 2026 (UTC)

It's less about minor differences and more about editors edit warring over changing the white spacing in the whole article. Template data controls parameters not white spacing. -- LCU ActivelyDisinterested «@» °∆t° 20:10, 25 July 2026 (UTC)

Nope. See mw:Help:TemplateData#Custom formats. In {{cite web}}, we've been using the format of "{{_ |_=_}}" for the last few years. WhatamIdoing (talk) 04:25, 26 July 2026 (UTC)

I missed that, it doesn't appear in the displayed text. I wonder who put it there, I doubt it has any consensus behind it. -- LCU ActivelyDisinterested «@» °∆t° 12:54, 26 July 2026 (UTC)

It was discussed a few years ago on multiple pages that I know of (one round started at WT:MED), and far as I know, nobody has complained about it since then. In fact, I recently saw someone complaining about the opposite the other day (that the template they inserted didn't add spaces between parameters, and they wanted it to). WhatamIdoing (talk) 03:19, 27 July 2026 (UTC)

I've seen repeated complaints about white spacing, and how white spacing should be handled, as well as edit warring over it. Like arguing over the referencing style it's a waste of editors time. -- LCU ActivelyDisinterested «@» °∆t° 10:19, 27 July 2026 (UTC)

Feel free to ping me to the next two or three of those discussions. I'm willing to help. WhatamIdoing (talk) 22:49, 27 July 2026 (UTC)

I'd call them "open issues" rather than "points of inclarity", since apparently they are interpreted differently by different editors. Conclusively resolving them one way or the other may require RfCs or similar means of finding consensus. Gawaon (talk) 03:25, 26 July 2026 (UTC)

I ran across an oddball citation style that may interest some of you; instead of ordinary little blue clicky numbers, it says or . I don't like it myself, but it's a consistently applied citation style that's been in this article since 2012, so WP:CITEVAR says that any change to the style would have to be preceded by obtaining consensus on the talk page. WhatamIdoing (talk) 17:07, 25 July 2026 (UTC)

citing DVD commentary tracks

Is there any guidance for referencing DVD commentary tracks in a Wikipedia article? My first inclination is to use the "Cite AV media" template, but there doesn't seem to be explicit support for referring to commentary tracks or alternate audio tracks. Maybe it's better to forgo template usage. Any ideas, or samples I can look at?

Bsammon (talk) 00:32, 3 August 2026 (UTC)

You could use Space Pilot 3000 for some examples of citations to commentary tracks. DonIago (talk) 02:09, 3 August 2026 (UTC)

Thanks! That's a good start. That page uses the "cite video" template, which appears to have recently been deprecated and replaced by "Cite AV media". I'll attempt to use "Cite AV" following the example of the use of "cite video" in that page, and see if it works. Bsammon (talk) 06:28, 3 August 2026 (UTC)

Good luck! I've used it myself, but infrequently, and not in a long time now. DonIago (talk) 19:25, 3 August 2026 (UTC)