Wikipedia talk:Manual of Style (footnotes)/Archive 11

Archive 5

←

Archive 9

Archive 10

Archive 11

Archive 12

A proposal to simplify and improve footnote markup in Wikipedia

Wikipedia offers a simple and powerful method to create footnotes (a.k.a. references) through the <ref> element in the wiki markup. The resulting HTML markup, however, is more complex than necessary. This study shows how the number of elements required to represent a footnote can be halved, while improving the reusability of the content. Howcome (talk) 09:42, 31 March 2009 (UTC)

You can use a background image for IE6 & IE7 and still use only the 'a'. s/name/id/ would further reduce the number of chars. ¦ Reisio (talk) 16:49, 31 March 2009 (UTC)

Thanks, I've added the s/name/id idea to the paper and credited you. Background images I'm less sure about, they scale and print poorly. --Howcome (talk) 13:04, 3 April 2009 (UTC)

I haven't looked at the details of this. This may be a valuable suggestion but, even if so, this is not the proper venue for it. This belongs at http://www.mediawiki.org/wiki/Extension_talk:Cite/Cite.php and/or at http://www.mediawiki.org/wiki/Bugzilla. -- Boracay Bill (talk) 00:15, 2 April 2009 (UTC)

Thanks for the pointer, I'll put it up there.--Howcome (talk) 13:04, 3 April 2009 (UTC)

Have you proposed the same at Wikipedia:Village pump (proposals)?

Cpiral (talk) 18:53, 1 September 2009 (UTC)

Frequency of footnotes

Are there any specific guidelines for how frequently to refer to the same source? I've seen articles in which every sentence or even phrase references the same source repeatedly, which makes it horrible to read. On the other hand, in the past I've used a single footnote to back up an entire paragraph of three or four sentences, only to have a {{Refimprove}} slapped on it. What's the general consensus here? –Spudtater (talk • contribs) 11:52, 7 April 2009 (UTC)

General consensus is that a single footnote can be used to back up a paragraph and/or multiple sentences within one paragraph. New paragraphs (or individual items in a list) need the ref repeated. If there is a direct quote in a paragraph, that sentence also needs the source. If the paragraph uses multiple sources, it needs to be clearly noted (so can again be repeats). In general, the same reference does not need to be put individually on back to back sentences within one paragraph. If someone is tagging refimprove, ask them why they tagged, and if that is the reason, explain and point here (and to the many FA/FLs which follow this consensus). -- AnmaFinotera (talk · contribs) 13:38, 7 April 2009 (UTC)

Thanks for that –Spudtater (talk • contribs) 21:22, 27 May 2009 (UTC)

Ref names (with quotes)

Recently I put some "Pet Peeves" mini-essays on my user page, and identified this article as one that I have a peeve about. You can read my peeve here. I would like to change the examples in this article, and add some of the things mentioned (not the whole peeve, just the relevant points). Any comments or objections before I proceed? --A Knight Who Says Ni (talk) 11:54, 18 April 2009 (UTC)

I'd prefer users to be in the habit of using quotes so that when they have a ref name like <ref name="Smith 2004"> it will work how they expect, rather than <ref name=Smith 2004>. I think your view on very short reference names is a bit too simplistic; yes, reference names are better kept short, but many articles refer to Smith's work of 2002, 2004, 2005(a), 2005(b) etc. so a slightly longer name is needed to add clarity and I'd hope we're beyond the 8.3 days of DOS where spaces and punctuation weren't allowed. Rjwilmsi 14:14, 18 April 2009 (UTC)

As I have noted at Help:Cite errors: "Footnote names are case sensitive. They may not be a numeric integer. The quotes are optional unless the name or group includes a space, punctuation or other mark. It is recommended that names and groups be kept simple and restricted to the standard English alphabet and numerals." --Gadget850 (talk) 15:01, 18 April 2009 (UTC)

Style Recommendations in WP:FOOT questioned

I have placed the following at [[WT:CITE#Style Recommendations at WP:FOOT#Style_recommendations:

I dithered about whether to place this here or on WT:FOOT. WP:CITE and WP:FOOT both identify themselves as Editing style guidelines. I decided to place this here because of the {{Nutshell}} info at the head of this page which says, "This citing sources guideline (a) discusses when to use citations, (b) shows how to format individual citations, and (c) provides methods for presenting citations within Wikipedia articles."

WP:FOOT#Style_recommendations contains recommendations regarding placement of superscripted footnote links in relation to punctuation which appear to me to conflict with consensus results from discussions of this subject which I have seen here in the past. I want to raise a yellow flag about this. I will leave a note on WT:FOOT mentioning this and pointing here. -- Boracay Bill (talk) 01:43, 23 April 2009 (UTC)

-- Boracay Bill (talk) 01:45, 23 April 2009 (UTC)

Named references

Is it possible to have something like this: Some paragraph.<ref name="smith"> Some other paragraph.<ref name="smith"> End of article. <ref name="smith">John Smith, 2006 Blah Blah</ref> without the last one showing up on the page? This way we could have cleaner markup in the article proper and separate the article's prose from the refs, which especially with {{citation}} template can be quite hard to read and confusing to new editors. I like finding and inserting refs into articles, but I've begun to hesitate since they make the markup intimidating. Shreevatsa (talk) 18:08, 9 May 2009 (UTC)

Not with the current version of cite.php. ---— Gadget850 (Ed) ^talk 18:23, 9 May 2009 (UTC)

Right. But see this. -- Boracay Bill (talk) 04:00, 10 May 2009 (UTC)

Thanks a lot, Boracay Bill. That's exactly the sort of thing I had in mind... do you know if that extension will make its way into Wikipedia eventually? Regards, Shreevatsa (talk) 13:24, 20 May 2009 (UTC)

I don't know when—or if. I have this modified cite-body.php up on a local test wiki and have noticed that in its present form it seems to have a problem with mismatched links and backlinks in hidden named refs where the group- parameter is specified; I haven't spent any time trying to figure out the cause or a solution. The process for getting such things incorporated into Wikipedia is explained in Wikipedia:Bug reports and feature requests. From what I've seen, (1) that process does not appear to work very well and (2) some changes seem to occur without the process being followed (e.g., the addition of the group= parameter to the production version of cite-body.php). -- Boracay Bill (talk) 23:45, 20 May 2009 (UTC)

(update) See Bugzilla:18890. -- Boracay Bill (talk) 12:51, 24 May 2009 (UTC)

Thanks very much for notifying me! I've added myself to the CC list. Shreevatsa (talk) 15:52, 24 May 2009 (UTC)

(update) There's a test wiki for this available here. -- Boracay Bill (talk) 03:21, 20 June 2009 (UTC)

Actually, in the meantime, it is possible to unclutter the article body even without any software changes, if we are willing to have the superscripts show up in the Notes section: see here for an example. Yes, it is a bit ugly to have them show up like that, but I think the benefits are worth it. Can we discuss this? Shreevatsa (talk) 15:59, 24 May 2009 (UTC)

That is much worse than "a bit ugly"; it is hideous. It does not conform to any real-world convention, and is therefore confusing to the reader. Furthermore, it is easier to maintain (edit) articles if the content of the citation is adjacent to the text that the citation supports, especially if an editor is only editing one section at a time, which is the preferred practice of most experienced editors. Finell (Talk) 17:45, 24 May 2009 (UTC)

Agree that is unacceptably bad. I think a combination of specific notes with a list of references maybe a way out. See e.g. Charlemagne. This is the way historians cite in their publications and you will find this style on several history related articles. Arnoutf (talk) 18:03, 24 May 2009 (UTC)

The "hideousness" can be fixed by just enclosing them in a <div style="display:none;"></div>, I realised. As for keeping references in the same section, I agree with you, but note that already we often have prose in one section referring (by name) to citations whose content is only present in other sections. It would not be too great a loss to have all references consistently placed at the end, IMHO. In any case, we can just keep the references at the end of each section, where they don't interfere with the prose. How does this sound? Shreevatsa (talk) 21:33, 24 May 2009 (UTC)

Such a radical change to Wikipedia's layout and Manual of Style would require a very strong concensus, which is very unlikely. Meanwhile, please follow the exisitng guidelines. Finell (Talk) 00:39, 25 May 2009 (UTC)

It is not a radical change (or any change at all) to either of them, as it does not change the appearance of the article. And the purpose of discussion is to build consensus — is there somewhere else you think this discussion should be carried out? Shreevatsa (talk) 01:11, 25 May 2009 (UTC)

There are problems with the cite.php system. People look for solutions to those problems. The idea you found, Shreevatsa, is one that others have come across before. (There are discussions in the archives here about it.) It's generally been thought to be confusing to readers, since here, clicking on "3b" doesn't take you to the text. If you hide that line of numbers in a "display none" diff, then the "b" is still there but goes nowhere. But what you want is something other editors want - a system that supports something like LaTeX-style markup. I think Bill's patch would be a step in that direction. Gimmetrow 01:42, 25 May 2009 (UTC)

Thanks! Glad to hear I'm not the only one concerned about this. :) You are right; I was just realising that the last backward-reference(?) from a footnote would lead nowhere. That kills my suggestion. I agree we should wait for the appropriate modification to cite.php to get into the Mediawiki version that Wikipedia uses, given this. In the meantime, we can only try our best to keep the citation markup clean, I guess... Regards, Shreevatsa (talk) 02:00, 25 May 2009 (UTC)

The request to define all the references in one section often comes up in discussions at {{reflist}}. See template:Reflist/testcases2 for an example— you still get an extra backlink from the reference list into the hidden section which makes it unacceptable for many.

As to Charlemagne— that is something of a mess. It is neither of the three accepted systems: Footnotes, shortened or parenthetical.

There is also the proposed, but not accepted User:CitationTool/Hybrid referencing. ---— Gadget850 (Ed) ^talk 02:05, 25 May 2009 (UTC)

(outdent) See Bugzilla:18890 and http://siteslot.com/testwiki/ -- Boracay Bill (talk) 12:44, 25 May 2009 (UTC)

Reflist: recent edit

Re this edit. I changed it in my previous edit because the 2 "reflist" references appear absolutely identically in my browser. You have to go into edit mode to see how they're different. I was relying on the following text to explain their uses, and the explanation for "Reflist" was followed by the explanation for "Reflist|2", so .... But now that it's back the way it was, I still see 3 options, 2 of which are identical. -- JackofOz (talk) 03:57, 7 June 2009 (UTC)

REFPUNC

The text of WP:REFPUNC says that editors are not required to place the ref tags in any particular place with respect to punctuation, so long as it's consistent in the article. Footnotes [5] and [6] label the examples as "correct" and "incorrect" locations with respect to punctuation.

Why does footnote [6] contradict the text that it's supposed to be illustrating? WhatamIdoing (talk) 04:13, 20 June 2009 (UTC)

It's been more than a month, with no answer, so I've corrected the footnote to match the text of the section. WhatamIdoing (talk) 18:57, 22 July 2009 (UTC)

I think there needs to be a discussion on if it is acceptable or not. To me, it looks like the footnotes were conflicting, and shouldn't just be changed. If there was a prior discussion that stated it is the proper method that would be helpful. In not, there should probably be a new focus on whether it should be acceptable or not. This page probably isn't watchlisted by many people. --Happy editing! Nehrams2020 (talk • contrib) 19:07, 22 July 2009 (UTC)

There is no possible way to accept both "Some editors prefer the in-house style of journals such as Nature, which place references before punctuation." and "Putting the ref before the punctuation, like Nature does, is always wrong." This blatant self-contradiction needed to be resolved.

In case of such conflicts, I think that WP:IAR and WP:CREEP are relevant: there is no good reason to restrict editors' freedom of judgment on this point, so we should not decree the one True™ answer (despite my not-so-secret hope that they'll all "independently and freely" choose the first option, which I personally think is more aesthetically pleasing). WhatamIdoing (talk) 19:37, 22 July 2009 (UTC)

There obviously should be an option for editors to choose between using Harvard referencing or the <ref> format. These are two different methods of providing inline citations. However, for this case, for a reader to go from article to article and see two conflicting methods of where the citation is placed seems counter-productive. Like you, I prefer the first method, but if community consensus prefers the second method then so be it. The many articles across Wikipedia would look better if there was a uniform setup to the placement throughout. A poll/discussion could tailor the guidelines to point out what is the preferred style (based on what consensus favors). --Happy editing! Nehrams2020 (talk • contrib) 19:55, 22 July 2009 (UTC)

[1] Here's the edit that added the conflicting info. Gimmetrow 22:18, 22 July 2009 (UTC)

Given that this edit put this information in the wrong section, I'll bet that the editor was unaware of the contents of WP:REFPUNC. WhatamIdoing (talk) 16:36, 23 July 2009 (UTC)

ref syntax should be improved

At the moment there is no possibility to know, if a ref element is attached to a sentence or a set of sentences, e.g.

This is sentence 1. This is sentence 2.[ref]Smith (2008), p. 7.[/ref]

Instead, the following syntax should be possible:

This is sentence 1. [ref "Smith (2008), p. 7."]This is sentence 2[/ref]

Here, it is obvious, that sentence 1 cannot be found in the book by Smith. That would allow a more precise knowledge about the origin of parts of an article. Additionally, this would allow mouse-over effects that let pop up a small box with the source. I posted this idea already to the programmers, without any reaction that someone is improving this. 92.229.63.236 (talk) 16:12, 4 July 2009 (UTC)

Here is an alternative which can be implemented as a template (note that I have constructed this example so that the scopes of the refs overlap. This is often the case.):

{{refscope name=smith2008p7|{{refscope name=footnoteXYZ|This is a sentence to which both refs apply.<ref name=footnoteXYZ>This is footnote XYZ.</ref>}} This is sentence to which only smith2008p7 applies.<ref name=smith2008p7>Smith (2008), p. 7.</ref>}}

The template {{Refscope}} might generate HTML similar to the following:

<span class="refscope" id="smith2008p7"><span class="refscope" id="footnoteXYZ">This is a sentence to which both refs apply.<ref name=footnoteXYZ>This is footnote XYZ.</ref></span> This is sentence to which only smith2008p7 applies.<ref name=smith2008p7>Smith (2008), p. 7.</ref></span>

(I think I got that straight -- I'm a lousy enough proofreader that I might have screwed it up.) Mouseover effects, etc. could be applied to span class "refscope" via MediaWiki:Monobook.css and/or via [[User:<username>/monobook.css]]. Personally, I think that there is very little chance of such a template being used widely and a pretty good chance that it would be misapplied much of the time where it is used. Wtmitchell (talk) (earlier Boracay Bill) 01:12, 6 July 2009 (UTC)

uh-oh

I was rewriting (clarifying?) the text at "separating references and explanatory notes", and I'm noticing that the references it produces appear the END of this document. Help! Agradman ^talk/_contribs 04:55, 16 July 2009 (UTC)

Looking at it quickly while running through my watchlist items, I'd say that you should have dummied up the rendering with some variation of {{ref}} and {{note}}. Using the <Ref> mechansim mixed the refs which you intended to render in your example in with the refs for the article in which your example appears. I'm bypassing this for now, but I may get back to it later today. Wtmitchell (talk) (earlier Boracay Bill) 23:01, 16 July 2009 (UTC)

Would it be overkill if we were to implement the content on another page and then transclude it here? Agradman ^talk/_contribs 00:12, 17 July 2009 (UTC)

I've redone the rendered example. It seems to be working as intended now. Wtmitchell (talk) (earlier Boracay Bill) 02:48, 17 July 2009 (UTC)

Missing group parameter?

Shouldn't the line

Claim #5{{#tag:ref|Claim #5 explained.<ref>Nested reference for explanation of claim #5.</ref>}}

contain a group parameter somewhere? Rubenescio (talk) 12:32, 21 July 2009 (UTC)

Thanks. That looks like my error. I've fixed it. I'm presently traveling and checking WP from internet cafes. My participation will be spotty for the next several weeks. Wtmitchell (talk) (earlier Boracay Bill) 00:30, 22 July 2009 (UTC)

can anyone help me with a small problem?

I'm having a bit of a problem with a footnote on the Paddy Bradley article and would be grateful if someone could help me out.

The problem is with Note A, which I have used twice in the body of the article. One small problem which I can't seem to fix.... the second link to Note A (i.e. the link in the "Club" section directs you to the correct footnote at the bottom, but when you click on the "^" symbol beside the note it doesn't redirect you back to the Club paragraph...only to the initial link to Note A.

Any suggestions? Derry Boi (talk) 11:54, 26 July 2009 (UTC)

I have applied a fix. It may or may not be exactly what you had in mind, but you should be able to see how to change it if changes are needed. Wtmitchell (talk) (earlier Boracay Bill) 01:24, 27 July 2009 (UTC)

Thanks very much. Derry Boi (talk) 11:40, 27 July 2009 (UTC)

Absolute Astronomy citations

Has anyone else noticed wikipedia articles citing to absoluteastronomy.com and the cited article turns out to be directly copied from the citing article? For instance, the wikipedia article http://en.wikipedia.org/wiki/Goof cited to http://www.absoluteastronomy.com/topics/Goof#encyclopedia which was itself a copy of the wikipedia article. Nightkey (talk) 15:07, 3 September 2009 (UTC)

See Wikipedia:Mirrors and forks, specifically Wikipedia:Mirrors and forks/Abc#Absolute Astronomy. There are quite a few links to that site,[2] but only a few actually used in articles. Perhaps it is time for a bot to start tagging mirror links. ---— Gadget850 (Ed) ^talk 15:33, 3 September 2009 (UTC)

Another footnote offender is wapedia.mobi Nightkey (talk)

cite.php update

The cite software has been updated to define named references within the reference list:

This is a reference.<ref name="refname1"/>

<references>
<ref name="refname1">content</ref>
</references>

This is a reference.

{{Reflist}} has been updated with |refs=

This is reference 1.<ref name="refname1"/>
This is reference 2.<ref name="refname2"/>
This is reference 3.<ref name="refname3"/>

{{reflist|
refs=
<ref name="refname1">content1</ref>
<ref name="refname2">content2</ref>
<ref name="refname3">content3</ref>
}}

This is reference 1. This is reference 2. This is reference 3.

The error messages are being worked out at Help:Cite errors. ---— Gadget850 (Ed) ^talk 12:47, 17 September 2009 (UTC)

I'm calling these "list-defined references" unless someone comes up with a better name. ---— Gadget850 (Ed) ^talk 14:27, 17 September 2009 (UTC)

Hallelujah! I remember the idea behind this being talked about back in June, it's great to see it actually come about. Woohoo! "list defined references" sounds liek a fine name to me, by the way.
— V = I * R (talk to Ω) 14:49, 17 September 2009 (UTC)

Brilliant! You'll want a hyphen between "list" and "defined" though. OrangeDog (talk • edits) 15:11, 17 September 2009 (UTC)

Nag, nag, nag. ---— Gadget850 (Ed) ^talk 15:16, 17 September 2009 (UTC)

Speaking of error messages, can anyone figure out why I'm getting one at Taner Akcam? Rd232 ^talk 19:25, 17 September 2009 (UTC)

Check <ref name="omroep2">— there is another ref stuck on it without a name. I just converted Arthur Rudolph as a sample since it used groups. ---— Gadget850 (Ed) ^talk 19:39, 17 September 2009 (UTC)

OK, thanks. Rd232 ^talk 19:43, 17 September 2009 (UTC)

I converted Anne Dallas Dudley as another example to use. Kaldari (talk) 23:01, 17 September 2009 (UTC)

See Wikipedia:Village pump (technical)/Archive 65#Footnotes update. I think it is possible to stuff all of this in the citation templates. ---— Gadget850 (Ed) ^talk 02:18, 18 September 2009 (UTC)

By the by, I've updated Help:Footnotes#List-defined references as well as WP:Footnotes. Rd232 ^talk 12:36, 18 September 2009 (UTC)

I created a shortcut WP:LDR and added it to the sectionSPhilbrickT

See {{r}}, a template to combine the inline cite names. ---— Gadget850 (Ed) ^talk 18:11, 22 September 2009 (UTC)

Ugh. Anomie⚔ 22:02, 22 September 2009 (UTC)

And {{sfn}}, a similar template for shortened footnotes. ---— Gadget850 (Ed) ^talk 22:38, 22 September 2009 (UTC)

Help:Footnotes

Propose that the non-guideline portions of Wikipedia:Footnotes be moved to Help:Footnotes.

Some background on the Help namespace. Originally, Help pages were written at Meta and transcluded through templates to each version of Wikipedia. Each Wikipedia would then add on bits specific to that implementation. The Meta Help pages were never updated, and each implementation of Wikipedia has diverged due to differences in CSS, JS, MediaWiki messages and templates. A year or so ago, we scrapped the Meta updates and began our own implementation of Help pages.

Meanwhile, we have a lot of duplication and a lot of differences between Wikipedia:Footnotes and Help:Footnotes. It appears to me that much of the Wikipedia page is not about guidelines, but about how-to, and that belongs on the Help page. ---— Gadget850 (Ed) ^talk 19:11, 20 September 2009 (UTC)

The opposite move makes more sense to me (i.e. redirect the help page here) - in what situation would one want to know about the technical usages of footnotes without the guidelines of how and where to use them? Seems to me that at the same time as we are explaining the tags, we should provide the formatting information as well. Christopher Parham (talk) 19:34, 20 September 2009 (UTC)

This page is about one specific method of crafting footnotes that uses cite.php. The few guidelines here such as style and size and display are universal and should be at Wikipedia:Citing sources; for example, they also apply to Wikipedia:Parenthetical referencing. I suspect that part of the confusion is the use of footnotes as a term to refer to a specific technical implementation. ---— Gadget850 (Ed) ^talk 15:39, 22 September 2009 (UTC)

Ref name (again...)

I've been ignoring this, I suppose in the hope that it would just cease to exist, but what the hell is up with "ref name" anyway? Is this one of those things that was created specifically by Wikipedians who didn't know what they were doing, like the whole category system? Does any actual, usable, academic style of referencing use footnotes like that? What are they for? Things that are only one page? Webpages? I suppose it figures that that is the type of stuff that passes for a good reference on Wikipedia. What about actual books? It has been mentioned before, but they are totally useless for books, even though people try to use them anyway. I don't want to see a footnote "2zf". That doesn't tell me anything, about the reference, about where I am in the footnotes, or about where the footnote is on the page. It doesn't make any sense and we should kill this before it spreads any further. Adam Bishop (talk) 02:46, 23 September 2009 (UTC)

I may be misunderstanding what you are asking, but using <ref name="blah"> allows an editor to reuse a reference. Like done in this article. That way you don't have to type out the ref multiple times.--Rockfang (talk) 04:56, 23 September 2009 (UTC)

I know what it does, but the result is nonsensical. Even in that article, you have 7, 8, two 6s in a row, 9, 10, 11, and back to 3. Huh? This is completely missing the point of a footnote. The idea is not that a piece of information is note 2 and we can reference it again later, it's that the information is sequentially note 2 the first time, and if it needs to be referenced 10 notes later it is now in footnote 12, not footnote 2 again. Summit, Wisconsin does not suffer too much because it's very short, but anything longer than that becomes impossible to follow. Does that make sense? If you need to reuse a reference, why not just type it again in a different note? Or copy and paste if you can't handle typing it again...it's on a computer after all. Or use "ibid" or something that is actually a standard reference, not whatever this is. Adam Bishop (talk) 05:16, 23 September 2009 (UTC)

If you're asking whether anyone else uses numbers as a way of referring to references that are collected at the end, then yes they do. This form of endnote is a common style in some academic journals, including the high profile journal Science. Dragons flight (talk) 06:40, 23 September 2009 (UTC)

You mean Science uses endnotes (which is fine; that's really what I mean here, not footnotes), or that it uses this strange system of jumping back and forth between numbers? Adam Bishop (talk) 07:49, 23 September 2009 (UTC)

Both. Science uses endnotes with the possibility of repeats that appear in essentially the same fashion that our content will render. Dragons flight (talk) 09:15, 23 September 2009 (UTC)

Do you have an example of that? (Possibly online? I can see back issues up to 2004) Adam Bishop (talk) 21:19, 23 September 2009 (UTC)

See [3]. The frequency of repetitions is less than I've seen on many wiki pages, but the same principle is there. Dragons flight (talk) 00:14, 24 September 2009 (UTC)

Hmm...very odd. Is this a standard form of referencing for scientific literature, or is peculiar to Science itself? (And why does this necessarily mean it should be adopted on Wikipedia? Maybe it would work for Wikipedia's science articles, but for, say, a history article, why shouldn't we use a standard history referencing system?) Adam Bishop (talk) 02:26, 24 September 2009 (UTC)

It's certainly not universal, but it is a common approach in scientific literature. As to the question of what Wikipedia "should" do, I don't know. I will say however that I have a general preference for moving towards more standardization in referencing rather than less. We already have so many allowed styles that it can be confusing for editors. Dragons flight (talk) 02:45, 24 September 2009 (UTC)

Reply to Adam: I think I understand you better now. With regards to not using ibid, my thinking is that it should not be used. Consider the possibility of one editor cites a source 3 times in a row with 1 normal listing and two ibid uses. What if a different editor comes along and cites a source between the two ibids and doesn't update the 2nd ibid. That could potentially cause a discrepancy with the references. I do understand your logic of having them in order, and I think that this could be useful possibly in shorter articles, but in longer articles like Barack Obama, it would make the ref section even longer and I think this would be a bad thing.--Rockfang (talk) 07:09, 23 September 2009 (UTC)

Well, we don't have to use ibid, we can just repeat the reference, even it does appear twice in a row. (I do this frequently, assuming that, like you, people don't want to see "ibid" on Wikipedia.) Adam Bishop (talk) 07:49, 23 September 2009 (UTC)

I've only just heard of {{Rp}} and {{R}} (in the section above!), but that seems to fit what you're after to some extent, no? Rd232 ^talk 08:24, 23 September 2009 (UTC)

See Wikipedia:Footnote#Style: "Do not use ibid., Id., or similar abbreviations in footnotes." ---— Gadget850 (Ed) ^talk 09:56, 23 September 2009 (UTC)

The problem with "ibid" could be solved if people just paid attention to what they're doing, but that's not really what I'm concerned about; Rp and R don't help, they do the same thing I'm complaining about, put more than one reference into one note. What's the problem with a longer reference section? Adam Bishop (talk) 21:19, 23 September 2009 (UTC)

A number of things, including maintainability, readability, page size, and not giving a misleading impression of how many sources are used. What do you have against merged references? Rd232 ^talk 09:48, 24 September 2009 (UTC)

Well as I've already mentioned, they don't work for books, and they don't give any useful information about the reference. Each reference should be distinct, even if it is from the same source. They are misleading, because (for example) note 2 does not follow note 10. That should be note 11, obviously. Most importantly, I suppose, is that it is impossible to know how many "note 2"s you have read, especially if there are 5 or 6 or more. Do you expect the reader to keep track of that, so they will know to click back from note 2g instead of 2e? If this is supposed to make references easier to use, then it is a massive failure on that account alone. Also, apparently aside from limited use in scientific literature, no real style of referencing does anything like this. Using a real system is easier to maintain and read, doesn't affect page size in any meaningful way (why does that matter, anyway?), and actually gives a better impression of the sources used. Of course, there are many different styles of referencing and they aren't all compatible, but they are all better than this. Anyone who has written even just a high school essay should know that. Adam Bishop (talk) 12:40, 24 September 2009 (UTC)

With respect, this is a "real system" for scientists, and it is the idea of fully repeating every reference that is unreal. From my point of view doing that would be worse and give a misleading impression that more reference works were consulted than would actually be the case. See my comments below. Dragons flight (talk) 14:38, 24 September 2009 (UTC)

I agree with Dragons flight. In addition, a reader does not need to keep track of how many times a reference is used, the a,b,c, ... in the reflist does that. There is no functional reason for the footnote numbers to be in order in the text (the only reason appears to be historical usage), clicking on any given number takes the reader to the reference. This is a "real system". I use references all day both academic and legal, both modern and 19th century, both paper and electronic. This works better than the unlinked (Bower 2007 p. 37) method, better than the op. cit. method, and better than many other methods, when used properly. Repeating the full citation would appear to be almost ludicrous. Note that is a number of different pages are to be pinpoint referenced in a work, the work can be listed in full under /* Sources */ and the individual pinpoint page cites can be in a short form. A number of articles do this. Similarly if only two pages from different sections the same work are used there can be two separate footnotes, clearly showing the section name and the page number. See for example the citations to Nieuw Nederlandsch biografisch woordenboek onder redactie in the Jacob van Heeckeren tot Enghuizen article. One page is used five times and the other only once. --Bejnar (talk) 16:20, 24 September 2009 (UTC)

I tend to agree with Adam that it was a mistake to present references in this way when the refname feature was introduced. It would be bettter, I think, to adhere to the formatting style that predominates in most published matter (ie. footnotes numbered sequentially in order). However, it is easily avoided by simply not using the refname feature and just manually repeating references. The relevant guidelines address the issue by neither encouraging or discouraging the use of the refname feature for repeated references. [Occasionally one must revert an AWB user since refname is included in its general fixes. It would probably be simple enough to get it removed, which would be a good start if one doesn't like the method.] Christopher Parham (talk) 13:02, 24 September 2009 (UTC)

"Most published matter"? That would be a hard point to prove. I could certainly bury you in libraries worth of scientific works that use either this style or the Harvard style (i.e. references appear in text as statements like "(Smith 1967, p. 34)"). In either case the full details of the actual reference work appears only once at the end. You advocate manually repeating references. For me that is ugly, unnecessary, and at some level just viscerally wrong. Don't take that as a personal insult. I find the divergent opinions here somewhat fascinating. It is just that I can't imagine anyone trained in scientific publishing advocating that, because duplicate full copies of every repeated reference just isn't done anywhere in scientific publishing. History, literature, and law have different conventions than science and engineering, and I think it is hard for people on either side of that divide to be accepting of the conventions of people on the other side. I don't know that there is any easy way to bridge this gap, and I'm certain that not everyone can be satisfied by any single system. (Of course my personal preference is Harvard style referencing, so I am rarely satisfied anyway.  ;-) Dragons flight (talk) 14:38, 24 September 2009 (UTC)

I don't advocate manually repeating references but rather having the refname feature present material as if I were, rather than consolidating references. I stand by my contention that sequential numbering is the norm across most published matter; "scientific works" though filling libraries make up a small fraction of all published material. This is especially true among "popular" works, which are probably the best proxy for what Wikipedia's general interest audience would be expecting. Christopher Parham (talk) 17:50, 24 September 2009 (UTC)

Hmm, apparently I have been unclear...I don't mean including the full reference every time; that would, of course, be ridiculous, and no style does that, as far as I know. What I mean is, the reference can be "Smith, 1967, pg. 34", in (for example) note 2, and if you had another reference on the same page in note 12, note 12 would repeat "Smith, 1967, pg. 34". Since ibid is apparently forbidden here, that reference would be repeated even if it was note 3. (The full citation, at least in the style I use, which I think is APA or APA-esque, is used the first time something is cited, and then the shortened version thereafter.) I hope that makes more sense! I think we actually agree. For an example of what I'm talking about, see William of Tyre, which is mostly written and referenced by me (and ignore any irregularities in the references that I haven't fixed yet). Adam Bishop (talk) 16:21, 24 September 2009 (UTC)

Then you need to take a look at shortened footnotes WP:CITESHORT, which is not discussed on the guidelines page as it is a different reference style. This is part of the problem with this page being titled "Footnotes", as it only applies to the system that directly uses cite.php. ---— Gadget850 (Ed) ^talk 16:55, 24 September 2009 (UTC)

• Adam Bishop says they don't work for books, and they don't give any useful information about the reference. I have seen plenty of articles where ref name works for books and provides pinpoint citation information. (See my reference above.) What ref name ought not to be used for is as a substitute for pinpoint citation information. If the same book or long article is cited multiple times in an article, it should be listed as a /* Source */ in the bibliography section. A short form of footnote can then be used for separate notes in the /* Notes */ field. If the same "page", or in the appropriate case "section" or natural group of pages, is used multiple times then ref name may be used advantageously, otherwise, the editor should provide the pinpoint citation information in short form in separate notes. The problem is not with the tool. --Bejnar (talk) 16:20, 24 September 2009 (UTC)

Shortened footnotes. ---— Gadget850 (Ed) ^talk 16:52, 25 September 2009 (UTC)

List-defined references and error messages

Now that we have the list-defined references, some new use cases have possibly arisen which unfortunately don't play nice with how some refernces are picked up and have error messages are displayed. One such use case is have a central page of references for articles within a certain topic where they commonly reuse the same refernces for example TV Characters within a series may use episode based cites, This method can be archived by truncating a page within the references tags. Unfortunately this causes errors to display which is unfortunate because this would be a excellent use case for this new feature since it's something people have suggested before, For a example of what I'm talking about I have created a example: User:Peachey88/Sandbox/022 for the references and User:Peachey88/Sandbox/023 as the mock article. Could we possibly work something out so we can archive this? Peachey88 ^{(Talk Page · Contribs)} 07:46, 23 September 2009 (UTC)

Addressing the specific errors: User:Peachey88/Sandbox/022 needs a {{reflist}}, an easy fix. Sandbox/02 defines reference "abcd", and is then transcluded into User:Peachey88/Sandbox/023 where it is not used, thus the error— you can't define a reference and not use it.

BTW: clicking on the Cite error message takes you directly to the section in Help:Cite errors for troubleshooting the problem. I have more work to do there.

As you can quickly tell, the current cite.php won't let you do what you are really trying to do. You can't create a page with a bunch of cites and use them as needed on other pages— you have to use every cite or you get "Cite error: <ref> tag with name "$1" defined in <references> is not used in prior text." The discussion on bugzilla that lead to the update specifically asked for this to ensure that all defined cites are actually used. ---— Gadget850 (Ed) ^talk 09:53, 23 September 2009 (UTC)

Discussion re whether to permit use of all-numeric YYYY-MM-DD format in footnotes

FYI -- there is a discussion at [4] as to whether or not to allow the use of the all-numeric YYYY-MM-DD format in footnotes/references.

I'm raising the point here in the event that you would like to follow it or join in. Thanks.--Epeefleche (talk) 07:30, 29 September 2009 (UTC)

• Update: The discussion is now a RFC at Wikipedia:Mosnum/proposal_on_YYYY-MM-DD_numerical_dates. -- Alarics (talk) 19:37, 29 September 2009 (UTC)

Problem with the #tag trick

I can't make the #tag trick described on the project page work. Here's what I'm doing:

• The ref following this is named "Foo2005" <ref name="Foo2005">This is the body of a ref namedFoo2005</ref>
• This uses #tag to generate a ref named Ran1912 inside a ref named Foo2005{{#tag:ref|J. Random, 1912. Cited in <ref name="Foo2005" />|name=Ran1912}}

Here's the results:

• The ref following this is named "Foo2005"
• this uses #tag to generate a ref named Ran1912

References

Is this a bug or a misunderstanding on my part? Wtmitchell (talk) (earlier Boracay Bill) 07:35, 6 October 2009 (UTC)

Two remarks: (1) mistyping of "Foo2005" or "Foo25" — (2) the {{#tag:ref|...}} will make the outside <ref>, not the inside. -- Codicorumus  « msg 09:57, 6 October 2009 (UTC)

Thanks for picking up my bad proofreading. I've fixed that mistyping in the above. and if now works as intended. Wtmitchell (talk) (earlier Boracay Bill) 02:57, 7 October 2009 (UTC)

MOS:PAIC?

In a recent edit, someone added "See also: MOS:PAIC" to the section on ref tags and punctuation. Does anyone else find MOS:PAIC to be misleading and inappropriately prescriptive? Not that that doesn't describe much of the MOS in general. Anomie⚔ 20:14, 10 October 2009 (UTC)

It is the job of a style manual to be prescriptive. In this case, though, the prescription was on this page at WP:REFPUNCT rather than at MOS:PAIC, so I've undone the edit. Shreevatsa (talk) 21:35, 10 October 2009 (UTC)

MOSPAIC is wrong; the way the MOS pages aren't synced is eternally frustrating. But there's another issue here: when and how did Nature ref punctuation creep back into this page? The last discussions I participated in did not favor that inclusion. Can anyone point me to a section in talk archives explaining how that worked its way back into this page? Wiki has an in-house style, and the last I knew, the Nature style was rejected. SandyGeorgia (Talk) 16:45, 20 October 2009 (UTC)

I completely share your frustration with MOS - and have been known to wonder aloud why we think having a Manual of Style anyone can edit is anything other than barking mad. How indeed in these circumstances can MOSPAIC, or any other part of MOS be "wrong". There is no requirement for consistency, and no few examples of the reverse. Not sure about the Nature thing. Ben MacDui 19:17, 20 October 2009 (UTC)

I believe the current wording, where the Nature system is described as acceptable but not most common, was the result of just the series of conversations you are talking about. I refer to those taking place between roughly around the turn of 2007-2008. It was less popular but not completely rejected. As far as I know, the current wording has been in place since that discussion. I support it as it stands; moving punctuation from before a period to after it does not improve Wikipedia for our readers, who have no plausible basis for caring about this issue. Christopher Parham (talk) 19:33, 20 October 2009 (UTC)

OK, I'll take your word for it (partly because I don't have time to look it up :) SandyGeorgia (Talk) 19:55, 20 October 2009 (UTC)

Incidentally, MOS:PAIC does not appear to contradict this guideline, unless someone has changed it just now. "Citations are generally place after the punctuation" is true and consistent with this page, which also has information about another acceptable (but less common) style. Christopher Parham (talk) 19:49, 20 October 2009 (UTC)

Ref tags and punctuation

The section Ref tags and punctuation misses the point when it makes the footnote-before-or-after-punctuation a question of style: The actual meaning changes depending with the before/after choice. In the latter case the footnote refers to the entire sentence (or, on the outside, an unusually independent clause); in the former, the footnote refers to the last word, last few words, or the last clause. Consider

Examples of planets include earth, mars, and jupiter[1].

Examples of planets include earth, mars, and jupiter.[1]

The first makes [1] refer to jupiter alone, e.g. as a source for the fact that jupiter is a planet. The second makes [1] refer to the entire sentence, e.g. as a source for the fact that earth, mars, and jupiter are all planets.

I strongly encourage a change of the current guideline to reflect the above. Not only is this the logical way to do it, but this is also the only way I can recall having seen in a styleguide previous to WP. 94.220.249.30 (talk) 19:01, 6 November 2009 (UTC)

In practice it doesn't work that way because when you see a footnote you have no idea whether it adheres to the style guide or not; so even if you change the guide as you suggest, when "blah blah.[1]" appears in Wikipedia there's no way to know exactly what [1] covers without consulting the source yourself. Logical systems work well for publications with a coherent editing process, but not for Wikipedia. Systems where information is conveyed by punctuation placement are not useful in a world where most punctuation is just placed according to some other rule (or randomly) and there's no way to tell what's deliberate and what is accidental - it's a bit like firefighters trying to communicate with smoke signals. Christopher Parham (talk) 20:32, 6 November 2009 (UTC)

Nested references with "name"

Is it possible to reuse a nested reference like a normal reference by giving it a "name" parameter. {{#tag:ref|Claim #5 explained.<ref>Nested reference for explanation of claim #5.</ref>|group="nb" name="some-name"}} did not work here. bamse (talk) 23:05, 6 November 2009 (UTC)

Never mind. I found my mistake. bamse (talk) 16:15, 7 November 2009 (UTC)

Reference VS Ref

The terminology is inconsistent in the article. — CpiralCpiral 00:00, 2 December 2009 (UTC)

"The article"? I assume that you mean that it is inconsistent at WP:FOOTNOTES (a non-article page).

Can you please give me an example of what you mean? I saw no uses of "ref" that didn't refer to things inside angle brackets in my quick scan. WhatamIdoing (talk) 07:26, 2 December 2009 (UTC)

List defined references

I have a prototype tool which can manually apply these in most cases, if there are one or two articles where it is needed drop me a talk page note. Rich Farmbrough, 14:40, 22 December 2009 (UTC).

Most reference text at bottom

I understand there is a technique that allows the bulk of the reference text to be placed in the References section at the bottom, and refer to them in the text using named references. I can't find the technique in the project page. I suspect it has only been discussed on talk pages. Can someone who knows how to do this document it on the project page? --Jc3s5h (talk) 22:43, 4 January 2010 (UTC)

See WP:LDR (subsection of the footnotes page). Christopher Parham (talk) 22:48, 4 January 2010 (UTC)

There is also a technique which allows an editor to place the wikitext for the bulk of references or footnotes in endmatter sections rather than inline with the wikitext for the article prose. One article where this has been done is Timeline of Philippine Sovereignty. Wtmitchell (talk) (earlier Boracay Bill) 02:26, 6 January 2010 (UTC)

Timeline of Philippine Sovereignty is using list defined references and shortened footnotes. ---— Gadget850 (Ed) ^talk 02:21, 12 January 2010 (UTC)

Relevant discussion at WP:VPP

See here. MickMacNee (talk) 00:05, 7 January 2010 (UTC)

Claim #5 - double ref'd

Is there possibly a developer solution to nested ref tagging? Doing with the #tag:ref method leaves the article you've editted hard to edit by less experienced users who expect standard ref tags. 76.66.197.17 (talk) 22:29, 10 January 2010 (UTC)

I've had this thought also; in fact, I've had a #tag:ref which I placed in an article removed -- probably because it looked strange to whoever removed it. One possible approach might be to wrap the #tag:ref generaton in a template. if the template is named "Tagref", an invocation like {{Tagref|ref body|name=refname}} would be a drop-in replacement for <ref name=refname|ref body>, except that Tagref invocations could be nested.

{{Tagref}} (if that's the name used) might be invoked similarly to the following:

• {{Tagref|ab cd|name=one}}
• {{Tagref|ab cd<ref>nested ref</ref>|name=two}}
• {{Tagref|ab cd{{Tagref|nested tagref invocation|name=three}}|name=four}}

The body of the template might look something like the following.

{{#if:{{{1|}}}<!--
 -->|{{#if:{{{name|}}}<!--
  -->|{{#Tag:Ref|{{{1}}}|name={{{name}}}}}<!--
  -->|{{error|Error:''name'' parameter must be defined.}}<!--
 -->}}<!--
 -->|{{Error|Error:empty ref body.}}<!--
-->}}

Wtmitchell (talk) (earlier Boracay Bill) 00:20, 12 January 2010 (UTC)

"Group" parameter in list-defined references

I suggest we remove the "group" parameters from the example in the list-defined references section, since it wouldn't be used in an article, right? We can still have it in the code that shows what it looks like though. --Apoc2400 (talk) 18:10, 18 January 2010 (UTC)

It is not needed. You can ensure that separate examples get grouped separately by including |close=1 in the {{reflist}} for the example, but not the <pre>...</pre>. This closes all references from the last {{reflist}} to the current {{reflist}}. ---— Gadget850 (Ed) ^talk 18:20, 18 January 2010 (UTC)

Relevant discussion

There is a discussion occurring at Wikipedia:Centralized discussion/Wikipedia Citation Style. Your participation would be appreciated.
— V = I * R (talk to Ohms law) 02:06, 19 January 2010 (UTC)

Why group="nb" ?

Is there a specific reason the letters "nb" are used for the notes group attribute? ---kilbad (talk) 23:02, 3 February 2010 (UTC)

I'm guessing that it is intended as an abbreviation for "notes-bibliography". See this. Wtmitchell (talk) (earlier Boracay Bill) 23:55, 4 February 2010 (UTC)

Nota bene. -— Gadget850 (Ed) ^talk 01:13, 5 February 2010 (UTC)

Ref names

I have been arguing this fact with User:Racepacket tonight because for the past week he has been removing the quotation marks that have been used to designate the name parameter in the ref tags. I have asked him to add them when he adds new references, as he tends to not use them. However, tonight he has been removing them wholesale from the article.

This guideline says they are not entirely necessary. But what is to be done if some editors do use them even when they are not necessary and one editor does not use them when they are not necessary? Surely removing the usage wholesale is akin to WP:SPELLING. And I am aware that it is WP:LAME but at this point Racepacket is accusing me of disrupting bringing the article to GA status because I am putting the quotation marks back in after he has removed them all.—Ryūlóng (竜龙) 06:28, 8 February 2010 (UTC)

He has just done it again.—Ryūlóng (竜龙) 06:29, 8 February 2010 (UTC)

space and ref

Link to related discussion Wikipedia:Content noticeboard#User:Stemonitis and space in front of ref tag. --Snek01 (talk) 16:39, 9 February 2010 (UTC)

Apparently, this page contradicts the main MOS in this regard. See the discussion here: WT:MOS#Contradiction regarding inline citations. Ucucha 14:10, 12 February 2010 (UTC)

REFPUNC, again

This page has long contained "permission" for editors to format footnotes in any reasonable, consistent style -- that is, without requiring an editors to use a style that some other editor declares to be "better" (in the sense of being prettier). I've reverted the change, because I don't think that Wikipedia benefits from needless and functionless WP:CREEPiness, but if there are logical arguments that favor making one system mandatory, I'd be happy to hear them. WhatamIdoing (talk) 02:17, 13 February 2010 (UTC)

It's very odd that you consider my halving of the length, and significant simplification of the contents, of this section to be instruction creep. Very easy to fling around your CREEP link, but is it appropriate? Now, I'll repeat what I've said at the MoS, which has a long-standing rule against the ungainly, disruptive pre-positioning of ref tags: The major sources (as copiously cited at WP:FN) say to put ref tags after the punctuation. It was always an absurd practice to put them before, and everyone but Nature, bless its heart, has realised this. The problem is acute with multiple tags, such as ... the cow chewed its cud^[3][5][6], and burped with satisfaction.

Many WPian editors insist on the bunching of many ref tags; there is no rule against this, and occasionally it is justified.

FAC has long objected to the practice of the dangling punctuation mark after superscript numerals—it's easy to see why. Giving editors carte blanche to do what they like in such a basic aspect of formatting goes counter to the practice of every other publishing house. One form is used throughout. Let's not be amateurish for the sake of someone's thirst for editorial laissez faire. I'm reverting your reversion. Please discuss here or at the MoS, which, again, has not allowed inside tags for some time. Tony (talk) 02:38, 13 February 2010 (UTC)

I see a current edit war at WP:MOS over this point (e.g. [5][6][7]). It's absurd to declare that something that's being actively edit warred over is something that has firm consensus.

I think your decisiont to change this text to an instruction to follow your personally preferred aesthetic style easily qualifies as "instruction creep". WhatamIdoing (talk) 02:54, 13 February 2010 (UTC)

It is not relevant that it is my personal preference for a style of ref tags that is numerical, bracketed and superscript (the last is the most relevant). Nature uses inline full-size numbers, which are reasonable (just as Harvard et al. are) before the punctuation. In WP's system, the problem arises from the scrappy visual appearance created by a large amount of white space between the end of the last word and the punctuation mark. I had no hand in creating the existing guidance. Tony (talk) 03:14, 14 February 2010 (UTC)

The issue of the "scrappy visual appearance" is an argument from personal aesthetic style. Not everyone agrees that a non-breaking space causes a "scrappy visual appearance". Some people (apparently) think that the "large amount of white space" looks better than having everything crammed together. You are not wrong; they are not wrong. Humans are not required to have identical aesthetic preferences. WhatamIdoing (talk) 04:10, 14 February 2010 (UTC)

I would tend to agree with WhatamIdoing; I don't see a basis for your proposal other than the desire to impose personal preferences on others. The two methods are equally clear and readable, generally speaking (although certainly particular individuals might find one method or the other more challenging). I would prefer to maintain the current text. Christopher Parham (talk) 13:28, 16 February 2010 (UTC)

There is a clear consensus on the MOS thread supporting the edit made by Tony. You claim, WhatamIdoing, that the proposed changes are an example of WP:CREEPiness. It is your own preferred version that "creeps". It is much longer, more complex and requires unnecessary work for editors establishing consistency and which style should be used on a given page. It is also unjustified in terms of common usage. You also say "if there are logical arguments that favor making one system mandatory, I'd be happy to hear them". You will find the logical arguments by carefully reading the MOS thread. --Epipelagic (talk) 03:48, 18 February 2010 (UTC)

Generally, I think that's a bit of forum-shopping: you can't just move every discussion to MOS talk, where there is a clear historical preference for greater regulation regardless of merit. I don't think the discussion on this page points to any consensus at all to overturn language that was established through a very in-depth discussion involving many users over months. Christopher Parham (talk) 13:35, 18 February 2010 (UTC)

Epipelagic, "You must do it exactly this way" is instruction creep. "Any system you like is okay with us, and here are a couple of examples" is verbose, but not instruction creep. WhatamIdoing (talk) 05:14, 19 February 2010 (UTC)

But the old version on WP:FN prohibits mixing styles. That's worse creep, in terms of adding to the editor load, than anything in the straightforward MOS version. --Epipelagic (talk) 02:01, 20 February 2010 (UTC)

• I don't see clear consensus on this change. I prefer the ref after the comma or the period, but I don't see Wikipedia consensus on that, regardless of what may exist within a particular discussion. --Bejnar (talk) 04:14, 18 February 2010 (UTC)

Proposal: citations after punctuation only

This thread needs to be be struck, so I have archived it. There is already an active thread above, #Proposal: Inline citations, where voting is currently underway. Both threads were initiated by Darkfrog24. The only diffeence is that the second is the inverse of the other, making it difficult to directly merge votes. This seem to be yet another stratagem by the gatekeepers on this page to circumvent and discount the views of the wider Wikipedia community. --Epipelagic (talk) 03:45, 19 February 2010 (UTC)

This isn't about thwarting anyone. The argument on WT:FN has been "This page is not bound by a conversation that took place on a different talk page," so I started another one here, both to satisfy that argument and to allow anyone on WT:FN who was not also on WT: MoS to have a say in the matter.

As for it being the inverse, the truth is that while I have no strong objections to either allowing both styles or requiring only one, I do feel that giving Wikipedia editors one set of instructions on the MoS and another on WP:FN is a problem. If anything, the fact that I'm the one suggesting that we go with a consensus—and WT:MoS did reach consensus—with which I initially disagreed ought to give the second proposal more weight. Someone is going to have to just bite a lip and accept a set of instructions that they don't like. It's happened on far more contentious issues than this one. Darkfrog24 (talk) 13:23, 19 February 2010 (UTC)

I'm not sure why we are speaking about stratagems here; the language initially on this page represented the consensus of dozens of editors on this page over a period of weeks if not months. Asking for some discussion on talk before changing the language here is a perfectly reasonable step. Christopher Parham (talk) 14:33, 19 February 2010 (UTC)

Archive of redundant thread
The following discussion has been closed. Please do not modify it.
We have a problem. I am sure we can all agree that it would be better for Wikipedia if we stopped giving the editors contradictory instructions. Right now, WP:FN allows both after-punctuation and before-punctuation inline citations, while WP:MoS allows only after-punctuation. After a long discussion with many good points on both sides, the MoS talk page reached a consensus that it should not change its guideline to permit both styles. To resolve this, I propose that WP:FN endorse only after-punctuation inline citations. If there is anyone here on WT:FN who didn't get to participate (or fully participate) in the discussion over on WT:MoS, let's hear you now. Every interested party deserves a say. Darkfrog24 (talk) 00:42, 19 February 2010 (UTC) Oppose - the main purpose of such rules is to give MOS editors something to feel good about. Adding regulation in this area won't make our articles clearer or more informative, while certainly adding additional work as good-faith editors bring articles into conformance with the MOS style du jour. That work is better devoted to constructive pursuits. Christopher Parham (talk) 00:53, 19 February 2010 (UTC) Comment- a strange collection of invalid reasons. Saying "the main purpose of such rules is to give MOS editors something to feel good about" is not a reason at all, it is attacking the integrity of those who disagree with you in a fatuous way. If one were to respond in kind, it might be to say that the main purpose of not having proper rules is to allow FN editors to avoid thinking at all. But that would be equally fatuous, so I have struck it. A straightforward after-punctuation rule hugely simplifies things for editors in ways that more than compensate for any time lost bringing good-faith editors into conformance with the MOS style. (those editors usually have to brought in line for many reasons apart from placement). --Epipelagic (talk) 01:36, 19 February 2010 (UTC) Support - for the many reasons outlined in the previous discussion thread. The rule is the simplest one, flows better, is in almost complete conformity with mainstream usage, simplifies tasks for editors and reduces the scope for edit wars. There is no need to endlessly keep establishing which style has precedence in a given article. --Epipelagic (talk) 01:34, 19 February 2010 (UTC)

• As an addendum, I support Epipelagic's edit, which Christopher Parham seems intent on thwarting. There is clear consensus not to mix before-punctuation with small superscript numerals; there has been for some time, actually. Tony (talk) 03:56, 19 February 2010 (UTC)
• I agree with Tony that there's consensus for footnotes after punctuation. Eubulides (talk) 07:12, 19 February 2010 (UTC)
• See my comments above. in #Proposal: Inline citations There are good reasons not to mandate one style. -- PBS (talk) 04:54, 24 March 2010 (UTC)

Footnotes and closing parentheses

I just checked the Chicago Manual of Style's discussion of footnotes and punctuation, and it turns out that we're citing an old version (14th edition); the current (15th) edition differs slightly, in that it says it's occasionally OK to put a footnote number before a closing parenthesis that's in the middle of a sentence. So I propose the following change in order to keep us up to date with the current edition:

• "When a reference tag coincides with punctuation, it is placed immediately after the punctuation, except for dashes and (occasionally) for closing parentheses within a sentence.<ref>The Chicago Manual of Style, 14th ed. 1993, Clause 15.8, p. 494 – "The superior numerals used for note reference numbers in the text should follow any punctuation marks except the dash, which they precede. The numbers should also be placed outside closing parentheses." 15th ed. 2003, Clause 16.30 – "A note number should be placed at the end of a sentence or at the end of a clause. The number follows any punctuation mark except for the dash, which it precedes. It follows a closing parenthesis.... For a parenthetical phrase within a sentence, it may occasionally be appropriate to place the note number before the closing parenthesis." – See also CMoS Online, Style Q&A, Punctuation, cited 19 February 2010.</ref>"

Eubulides (talk) 07:12, 19 February 2010 (UTC)

The Oxford Guide to Style has ". . . but inside the closing parenthesis when referring solely to matter within the parentheses", which seems sensible.--Boson (talk) 07:42, 19 February 2010 (UTC)

What circumstances, I wonder, does CMOS have in mind for before-closing-parenthesis placement? Tony (talk) 08:08, 19 February 2010 (UTC)

As far as I know, the CMoS does not say. It does give two examples, one either way:

• "This," wrote George Templeton Strong, "is what our tailors can do." (In an earlier book he had said quite the opposite.)²
• Men and their unions, as they entered industrial work, negotiated two things: young women would be laid off once they married (the commonly acknowledged "marriage bar"¹), and men would be paid a "family wage."

These examples are consistent with the Oxford style guide's advice that Boson quotes. Eubulides (talk) 09:32, 19 February 2010 (UTC)

Do we need to provide this level of detail? Has anyone ever encountered a real dispute on this point? (Dispute, meaning "something we actually had to discuss on an article's talk page because someone objected when it was corrected," not "editors are sometimes sloppy," which I freely grant as true, but not as proof of a dispute.) WhatamIdoing (talk) 22:12, 19 February 2010 (UTC)

We need not go into the same level of detail as the CMoS. All we need to do is to mention that citations can go just before ")", as well as just after. That's what the proposed wording change attempts to do. If you prefer, we could make it even terser, as in "When a reference tag coincides with punctuation, it is placed immediately after the punctuation, except for dashes and (occasionally) parentheses." Eubulides (talk) 22:47, 19 February 2010 (UTC)

Merging refs when different quotes used from same source

Question: in the first sentence of this page [8] 3 footnotes (1/3/4) all go to the same book, same page. But because they provide different contextual quotes within the footnote, I can't see how to merge them. This is partly untidy, but partly an NPOV problem in suggesting wider sourcing than is actually the case. Is there a way to merge these refs and keep the quotes? (Other than the obvious solution of lumping everything into one footnote.) I've seen somewhere something clever about citing books, and I don't remember it. Rd232 ^talk 11:01, 19 February 2010 (UTC)

I suggest using short citations in the second and third refs. You might like the system used at Nitrogen narcosis, which is discussed here. WhatamIdoing (talk) 22:07, 19 February 2010 (UTC)

Nitrogen narcosis uses explanatory notes as documented at WP:REFNOTE. User_talk:WhatamIdoing/Archive_3#Nitrogen_narcosis describes shortened footnotes as documented at WP:CITESHORT. ---— Gadget850 (Ed) ^talk 22:43, 19 February 2010 (UTC)

Square brackets

I'm curious as to why the inline tags for references employ square brackets around the number. I've noted some mild but persistent consternation over the size of inline tags and how they create irregular line spacing in articles. However, it seems a large contributor to the vertical size of tags are the seemingly unnecessary brackets; why do we not simply have numbers, as most publications use, as do Spanish and French Wikipedias? (Polish Wikipedia uses the right size of inline tag.) --Ħ MIESIANIACAL 18:32, 21 February 2010 (UTC)

I don't know the reason. However, in some math and science articles, there are equations with exponents which also have citations. The brackets help distinguish a number that is a citation from a number that is an exponent. Jc3s5h (talk) 20:09, 21 February 2010 (UTC)

You can remove the brackets through your CSS— see Help:Cite messages#Customizing. ---— Gadget850 (Ed) ^talk 21:25, 21 February 2010 (UTC)

I just spent an hour looking through all that and I feel like an idiot; it's all gibberish to me! Regardless, shouldn't we be making Wikipedia appear graphically professional for all users, not just those with unique CSS (especially those who are too stupid to figure out how to make their own CSS, like me!)? --Ħ MIESIANIACAL 06:11, 22 February 2010 (UTC)

French Wikipedia really does seem to have it right, inline cites there neither using redundant brackets nor pushing lines of text apart. Why does English Wikipedia not adopt the same? --Ħ MIESIANIACAL 06:16, 22 February 2010 (UTC)

Where English Wikipedia uses "^[3][4]", French Wikipedia uses "³⁴". Is that really an improvement? It looks ambiguous to me, since it can easily be confused with footnote number 34. I'm not saying the English system is perfect (I'd rather that it lost the brackets), but the French system isn't better. Eubulides (talk) 07:25, 22 February 2010 (UTC)

I agree that, on the face of it, doing away with brackets and not pushing lines apart is much more elegant. Is there a problem with this^1,2 ? There is an issue with an equation, perhaps requiring the editor to add a leading sentence fragment and citing that. Or maybe that issue could be overcome by following Nature, and underscoring, like this ^1,2 ? --Epipelagic (talk) 09:51, 22 February 2010 (UTC)

I can make the inline cite link underlined. Adding a comma as a separator can be done, but will need a modification to a system message page. ---— Gadget850 (Ed) ^talk 00:41, 23 February 2010 (UTC)

I suspect there are too many cited equations to change the behavior in a way that will make equations confusing. Rewriting all the equations is not an option. Jc3s5h (talk) 16:59, 22 February 2010 (UTC)

I'm not sure that's a good reason. I've just checked some mathematical articles, and found very few directly citing equations. In any case, the parser that formats raw text for display could easily add square brackets or otherwise delineate citations, perhaps by adding leading spaces followed by an mdash, if they are placed directly after equations. --Epipelagic (talk) 23:43, 22 February 2010 (UTC)

(unindent) There is no way for a parser to reliably recognize an equation. Jc3s5h (talk) 00:08, 23 February 2010 (UTC)

Why do you say that? Math markup goes inside <math> ... </math>. The parser has to recognize the math equation before passing it to TeX. --Epipelagic (talk) 00:34, 23 February 2010 (UTC)

According to Wikipedia:Manual of Style (mathematics)#Typesetting of mathematical formulae "One may set formulae using LaTeX or, for simple formulae, using HTML; both are acceptable and widely used...". And that just covers the mathematics articles. Many other articles in the area of science and technology would also be affected. Jc3s5h (talk) 01:31, 23 February 2010 (UTC)

So let's see if we can accommodate this issue, which affects a tiny proportion of Wikipedia articles, instead of letting it abort an improved formatting that works for the vast majority of articles. Maybe just underscoring the citation, like Nature does, would do it? --Epipelagic (talk) 02:04, 23 February 2010 (UTC)

Gadget, are there any coding issues with using citations without brackets, and maybe adjusting their height in some compromise way so the space between lines is even, as is done on the French Wikipedia? And would there be any difficulty in special casing math equations if they are directly cited, apart perhaps from those coded in HTML? --Epipelagic (talk) 01:17, 23 February 2010 (UTC)

• From a technical viewpoint, the brackets can be removed sitewide. You will have to gain consensus on this, and will have to make a really good case for it. This change can be done on an individual basis, and I can certainly help anyone who has a problem there.
• What is the problem with line spacing? As noted at Help:Cite messages#Classes and CSS IE7 has issues with the markup we use to keep the line height even, but I am not aware of any other problems.
• Can someone give me an example of an equation with a cite?
• Multiple inline cites have a defined separator— this is currently a space. It can be changed, but will change for everyone. To make it customizable, it can be wrapped in a CSS class. Again, this will require consensus.

---— Gadget850 (Ed) ^talk 14:52, 23 February 2010 (UTC)

One potential positive of the brackets is that they increase the target size of the links... ¹ is quite a small target. All in all I think this is an aesthetic issue on which there will be disagreement and there much urgency to make changes. Christopher Parham (talk) 15:14, 23 February 2010 (UTC)

I can't see how there'd be disagreement: the inline tags, as they are now, push lines of text apart. I'm not sure who'd think irregular line spacing is aesthetically acceptable. --Ħ MIESIANIACAL 19:11, 7 March 2010 (UTC)

The article Conversion of units has many expressions (or equations) with cites that could be mistaken for exponents, were it not for the brackets. These don't use the math markup for the whole expression, but sometimes use math markup for individual characters. I don't recall the names of articles where I've seen math markup combined with cites. Jc3s5h (talk) 15:18, 23 February 2010 (UTC)

These issues are rare, and shouldn't be a significant problem. Still there seems no interest, so there is no point in pursuing the matter. That will keep you happy Christopher. Change is such an unseemly thing. --Epipelagic (talk) 22:24, 23 February 2010 (UTC)

personally I don't have an opinion on the matter. Christopher Parham (talk) 22:34, 23 February 2010 (UTC)

General style question involving LDR and more

I do a lot of writing about lemurs and plan to eventually re-write and standardize all of that broad topic's 100+ articles. I just discovered LDR and would like to implement it in my upcoming, massive re-write of the Lemur article. However, before I do it, I would like to get some guidance. I must admit that I'm not an expert on the various citation styles, and I'm only used to what I've used in college or see in the academic literature (biology). But from my experiences, the organizational approach that I've seen on Wiki has confused me a little. Let me try to explain...

Some of the larger FAs out there (including some that I have written) have separate "Notes" and "References" sections. "Notes" is where the {{reflist}} goes, and "References" is where the {{refbegin}} and {{refend}} go. I've been putting all my large books under "References" and then using short footnotes (like those generated by {{sfn}}) to cite individual pages or chapters to those books up in the text body (and listed under "Notes"). Now I can keep doing this, but given that {{r}} can specify page numbers, then why bother separating the full citation and the short footnote into separate sections? Why not just use the long citation only (through LDR) and use the page number when you call {{r}}?

There is another case that partly answers my question but raises its own. Some of my books are collections of articles written by various researchers. Depending on which article I cite in the book, the authors (and "contribution") will be different, but the editors and book info will not change. This is where the Notes/References division appears to help. However, where do you draw the line? Journal, news, web, and other under "Notes", and all books under "References"? Or should only books with different contributors be placed under "References"? And should you only use the |chapter= parameter when your Wiki article only uses a particular chapter from the book, and use page numbers through {{r}} otherwise?

In short, here are the five situations I deal with on a daily basis:

1. Journal articles (some with page numbers, others without), news articles, web sites, etc.
2. Books where I cite just a page or two here or there... usually for just to support a brief factual statement
3. Books with regular chapters (no contributors) where I cite the one or more chapters regularly
4. Books with chapters written by various contributors
5. Books with chapters of collected articles by various contributors

I just need a way to organize all of this cleanly and consistently. Unfortunately, dividing things into "Notes" and "References" makes this a little confusing.

Lastly, I'm not happy with the heading names and how things are being organized. Journal articles, news articles, etc. are not "Notes"—they're references. But if we're pulling out books to list separately, and then point to individual pages, chapters, contributions through ref tags, the journal references (etc.) have to get mixed in with the notes because there is only one {{reflist}} on the page. Am I missing something? Is there another way to do all of this, or an excellent article that I can look at for an example?

Feel free to take this discussion to my talk page if you don't want to clutter up this discussion page for the sake of helping me. – VisionHolder « talk » 17:24, 28 February 2010 (UTC)

And then as I take a break and look around at articles that interest me, I run upon Bat, which subdivides its "References" section with a "General references" area and a level 3 heading "Further reading". This approach might solve the latter issue with section naming, but still leaves me to deal with the five types of sources and how to organize them. – VisionHolder « talk » 18:30, 28 February 2010 (UTC)

Try something like <ref group=nb>a note</ref>, <ref name=test group=ref1>cite a journal article</ref><ref group=ref2cite one type of book></ref>, <ref group=ref3>cite another type of book</ref> and {{reflist |group=nb}}, {{reflist |group=ref1}}, {{reflist |group=ref2}}, {{reflist |group=ref3}}. Wtmitchell (talk) (earlier Boracay Bill) 02:55, 1 March 2010 (UTC)

I'm really sorry... I completely missed your reply until now. I think I see what you are getting at. So should I put several subheadings under a References section for things like Notes, Journal articles, Books, etc.? Either way, I almost need to see a specific example. – VisionHolder « talk » 16:39, 4 March 2010 (UTC)

• MOS:APPENDIX discourages the use of level 3 headers in these sections.
• Is there some reason that you can't just put all of this information in a single section? AFAIK, there's no rule against combining short and full citations in the same list. WhatamIdoing (talk) 22:17, 15 March 2010 (UTC)

I already have moved over to using definition list headings as MOS:APPENDIX suggests. As for keeping the lists separate, I prefer to do that to make things more readable. Especially when I'm using short references to cite an article in a chapter of a book, with different authors and editors for the articles and book (in the same volume). I have an example I'm working on, but it will probably take another week or two before it gets published. If you want, keep an eye on Lemur and when you notice a re-write get posted, you'll see what I've come up with. All-in-all, I'm pretty content with what I've come up with at this point. If I get to FAC and changes are demanded, I will consider it. – VisionHolder « talk » 22:08, 18 March 2010 (UTC)

Style recommendations revisited

Currently reads:

---

• Do not use ibid., Id., or similar abbreviations in footnotes. Other editors who add new references to the article may not take the time to correct Ibid references broken by their addition (op. cit. is less problematic in that it should refer explicitly to a citation contained in the article). However, not all readers are familiar with the meaning of the terms. If a reference is reused in more than one footnote, it is preferable to use the format "Smith, Short Title, 182" rather than "Ibid, 182", so as to avoid these problems, or use named references if appropriate.

---

I'd like to expand it to something like:

---

• Do not use ibid., Id., or similar abbreviations in footnotes. Other editors who add new references to the article may not take the time to correct Ibid references broken by their addition (op. cit. is less problematic in that it should refer explicitly to a citation contained in the article). However, not all readers are familiar with the meaning of the terms. If a reference is reused in more than one footnote, it is preferable to use the format "Smith, Short Title, 182" rather than "Ibid, 182", so as to avoid these problems, or use named references if appropriate. Repeated citations that differ only in the specific page referred to may use unambiguous op. cit. references such as "Smith and Jones, op. cit., p. 123" (where "Smith and Jones" is expanded with a year and subscript if necessary, e.g. "Smith and Jones (1892b), op. cit., p. 123").

---

Comments? Urhixidur (talk) 20:27, 5 March 2010 (UTC)

Question about LDR

The LDR section doesn't appear to explain something. If I insert, say, a {{reflist}} template and only pass some of the article's references via the "ref=" parameter, do the remainder of the inline citations (in the same group) also get listed? Or does the resulting list consist exclusively of those references passed via the "ref=" parameter? It would help if the section could clarify this. Thank you.—RJH (talk) 15:58, 18 March 2010 (UTC)

You can mix inline defined and list defined references and both will be parsed by the same {{reflist}}. ---— Gadget850 (Ed) ^talk 17:09, 18 March 2010 (UTC)

Could the section text be modified to explain this? It might also be good to mention that the order of the references in LDR is arbitrary; they will be listed in the resulting reflist in the order they are used in the article. Thank you.—RJH (talk) 20:41, 18 March 2010 (UTC)

Neat formatting trick

I just came up with this trick for formatting refs, and I'm curious if anyone has come up with it before.

It makes editing much clearer if refs are started on their own line; this doesn't cause a line break in the text, but it does for some reason create extra whitespace before the reference. For example:

This is a line of text.
<ref>Ref 1</ref>
<ref>Ref 2</ref>

This is a line of text.

In order to remove the extra whitespace, the trick is to comment out the line break itself:

This is a line of text.<!--
--><ref>Ref 1</ref><!--
--><ref>Ref 2</ref>

This is a line of text.

...which looks much better. Has anyone come up with this trick before?

Antony-22 (talk) 23:53, 25 March 2010 (UTC)

There are a number of ideas for dealing with formatting. Similarly you can just change the location of the hard return, like this:

This is a line of text.<ref>
Ref 1</ref><ref>
Ref 2</ref>

This results in the standard formatting, e.g.: This is a line of text.

You might also like to look into list-defined references, which allow you to move all the refs to the end of the page (leaving only "<ref name=Foo />" in the article text). WhatamIdoing (talk) 04:11, 26 March 2010 (UTC)

Yes, everybody has their own preferences about how to format inline citations, so as a result they tend to flip-flop around a lot as each editor imposes their own standard. WP:LDR gets the citations out of the way of the article text; it's an elegant solution, although it's also a nuisance to migrate the citations. Eventually, I hope that there will be a bot to implement it, at least for all the named citations. That way you just need to give each cite a name, then run the bot.—RJH (talk) 15:49, 26 March 2010 (UTC)

Contradiction regarding inline citations

This discussion was started on Wikipedia talk:Manual of Style. It has been transferred so it can continue here, since it appears that the issues involved can only be resolved on this page

WP:MOS#Punctuation and inline citations says:

"Place inline citations after any punctuation such as a comma or period, with no intervening space"

WP:REFPUNCT says:

"When the reference tag coincides with punctuation, it is most commonly placed immediately after the punctuation, except for dashes, as recommended by the Chicago Manual of Style and other style guides.[5][6] Some editors prefer the in-house style of journals such as Nature, which place references before punctuation. If an article has evolved using predominantly one style of ref tag placement, the whole article should conform to that style unless there is a consensus to change it."

That is, only one style is acceptable in the main MOS, but both styles are acceptable in REFPUNCT. Ucucha 23:58, 11 February 2010 (UTC)

Wow. Nice catch. I prefer the look of ref-after-punct, but I don't see any inherent superiority either way. We should probably adopt a choose-either-but-maintain-intra-article-consistency policy. Darkfrog24 (talk) 00:16, 12 February 2010 (UTC)

Me too, but primarily I want the contradiction to be resolved. Ucucha 00:19, 12 February 2010 (UTC)

I've always seen it after the dots in Wikipedia, and people copy-editing from before dots to after dots, never the other way around. So here's a vote for always after the dots from me. Headbomb {^ταλκ_{κοντριβς} – WP Physics} 00:21, 12 February 2010 (UTC)

For a long time, the agreement was ref after punc. Then a small group of editors managed to push through on CITE, as I recall, that we could choose. That's where the contradiction comes from. SlimVirgin ^{TALK contribs} 00:22, 12 February 2010 (UTC)

i'm firmly in the after-punctuation school as well, but what are the arguments that this "small group of editors" had in favour of the pre-punctuation style? Sssoul (talk) 10:49, 12 February 2010 (UTC)

It's used by some respected publications, is the main reason. The change was made by Darkfrog less than a month ago: [9], so far as I can tell without discussion. It doesn't appear he understood the change to be material, or intended it to have an impact on the meaning. I have reverted to the standing version for the moment. Christopher Parham (talk) 14:31, 12 February 2010 (UTC)

I changed it to the imperative. How the heck does that constitute a change in policy? (Other than expressing it as a policy rather than as a description.) I really don't see how the phrase "place inline citations after other punctuation" tells people that they may place them either before or after as they see fit. Darkfrog24 (talk) 14:53, 12 February 2010 (UTC)

Don't you see the difference between "this is what is generally done" and "do this"? Ucucha 14:56, 12 February 2010 (UTC)

Yes. One of them is a general statement about what happens. In this case, the statement is inaccurate because it's generally done one way in some publications and the other way in others. The other one is an instruction. Because the purpose of the MoS is to tell people what to do, it should be phrased in the imperative. The MoS is not a writeup of a linguistic study. It's not even a Wikipedia article. It's a set of instructions and it should not pretend to be anything else.

However, changing "inline citations are placed after" to "place inline citations after" does not change the intended meaning of either. The MoS meant for people to place the citations after the punctuation before my change. After my change it expressed that more clearly.

If it's the removal of "generally" that troubles you, then the way to phrase the MoS is "Generally, place citations after punctuation." Darkfrog24 (talk) 15:01, 12 February 2010 (UTC)

I'm happy with any language that reflects the consensus developed at WP:FN: that before & after are both acceptable, that articles should be internally consistent, and that articles consistently in either style should not be changed to the other without talk page consensus. I'm not comfortable with the imperative language in this case given that the exsiting consensus on the subpage doesn't encourage or discourage either method. Christopher Parham (talk) 15:16, 12 February 2010 (UTC)

We're on the same page there. If the consensus changes, we should change the MoS to reflect it. However, what confuses me is why you seem to think that changing "they're placed after" to "place them after" did anything but rephrase what was already there. If "place them after" bothers you because you feel it doesn't reflect consensus, then why doesn't "they're placed after" also bother you? Darkfrog24 (talk) 15:55, 12 February 2010 (UTC)

Becuase one tells people to do something a specific way, and the other does not. Imperative langauge associated with this method in particular indicates that we are telling people to do it a certain way, indicating that we prefer it to be done that way vs. another way. We don't have such a preference. We merely observe that one style is more common, to educate readers about what to expect when editing the encyclopedia. I would also be fine with imperative langauge, if the section was expanded to also present the alternative method and explain that there is no preference. Christopher Parham (talk) 16:02, 12 February 2010 (UTC)

Observations that one style is more common than another belong in the articles, where they can be sourced. The previous text, "are placed after" means to instruct people but instead makes a factually inaccurate statement. "They're placed after" doesn't mean "this is more common than that" either literally or in practice. Everything on the MoS is interpreted as a rule by the editors who read it. We should acknowledge that and use the most honest and direct language available to us.

I'm still a bit confused as to why you changed the imperative to the indicative rather than discussing consensus, but I feel like I have a better grasp of where you're coming from now. Darkfrog24 (talk) 16:18, 12 February 2010 (UTC)

The statement that a particular style is more common is intended to refer to conditions on Wikipedia, not elsewhere in society. If you would like to rewrite the section to reflect the "rules" directly, I would write this:

1. Citations may go before or after punctuation.
2. Articles should be consistent in using one method or the other.
3. Articles that consistently use one method should not be converted to the other. (This point is redundant to one of the MOS's general principles)

As to consensus, I am assuming that the existing consensus, reflected in the language at WP:FN, is operative. If you disagree with it, the best place to discuss it is that talk page; otherwise, the goal here should be to reflect that langauge accurately. Christopher Parham (talk) 16:48, 12 February 2010 (UTC)

BTW, I reverted your replacement of your text. Your edit implies that we generally would like people to add citations after punctuation. There is no consensus to support this claim, indeed the most recently developed consensus says the opposite, that we are indifferent. Christopher Parham (talk) 18:39, 12 February 2010 (UTC)

I think I see the difference: From your perspective the consensus has changed, and from mine it has yet to be established that consensus has changed.

According to Slim, the consensus over on FN is sketchy. Usually, this is not an issue, but in this case, we need to discuss whether "a small group of editors managed to push through" constitutes a proper Wikipedia consensus that the MoS should reflect or whether FN is in error. It's probably fine, but it's worth asking around first. Headbomb, at least, seems to have something to say about it, and half the time Wavelength jumps into these conversations with some excellent reasoning behind a rule that the rest of us had thought was fluff.

I expect the proposed change to be approved relatively quickly. It might have been a small group of editors, but the proposal itself seems good. Darkfrog24 (talk) 20:54, 12 February 2010 (UTC)

This is the strange part, though, Chris. If you believe that the consensus is "either after or before," then why are you changing the MoS from one phrasing that does not reflect that consensus to another phrasing that also does not reflect that consensus? That really doesn't make sense. Imperative vs. indicative has nothing to do with before or after. The change from indicative to imperative had nothing whatsoever to do with anything that was or wasn't being discussed on WP:FN. It was solely about clarity. Darkfrog24 (talk) 21:00, 12 February 2010 (UTC)

I don't know why Slim believes that the consensus is sketchy; WP:FN has been in its current state for a year or more, and WP:MOS has had the language I reverted to in place for a long time, prior to your change in January. I don't see why you can't understand the difference between stating that something is generally done a certain way (makes no value judgment between methods) and commanding that something be done a particular way (a clear value judgment in favor of the method so ordered). What about my wording indicates that inline citations should be placed after punctuation? That said, if you have a wording that would make the MOS more reflective of the existing guideline's recommendation, feel free to add it, but I don't think changing "most people do" to "people should" better reflects the notion that people have an option and neither choice is recommended or discouraged. But we should also keep the language simple, for people who just want a way to format their page, and not be presented with every acceptable style (that's why at the moment, discussion of the second acceptable method is kept to the FN guideline). Christopher Parham (talk) 21:18, 12 February 2010 (UTC)

In reviewing this discussion it appears that we had consensus for the previous (i.e., pre-January-25) wording "Inline citations are generally placed after any punctuation" and the revisions since then haven't resulted in any material improvements or substantive changes (but have aroused controversy) so for now I've reverted to that wording. I suggest that we discuss further wording improvements here. Eubulides (talk) 21:16, 12 February 2010 (UTC)

Before and after January 1, the MoS instructed readers to use only punct-before, even if WP:FN did not. The proposal below suggests changing it to specifically allow both punct-before and punct-after, which would bring it into agreement with WP:FN.

The January 25 change did not change the rule. It simply rephrased what was already there, an instruction to place citations after the closing punctuation with no mention of any other style. The MoS should not make general statements about what happens or what "is done" for two reasons. 1. Anything stated on the MoS, even in the indicative, will be interpreted as a hard-and-fast rule, whether we mean it that way or not. Even gentle statements have been the subjects of edit wars. 2. The purpose of the MoS is not to describe what happens in English in general (the literal meaning of "this is done") but rather to tell people what to do on Wikipedia. It is supposed to make value judgments. The LQ section is a good example. It originally said "punctuation is placed inside the quotation marks if it is part of the quoted material and outside if it is not." That statement isn't true, not even on Wikipedia! Phrasing the MoS in the imperative makes its instructions more direct and less likely to be misinterpreted.

The previous wording did not actually say, "this system is more common than another" or "pick out what you want." It didn't even say "this happens but so does that." It said "this happens," which in practice means "do this and nothing else." We can tell because Wikipedia users have been changing punct-after to punct-before but not the other way around. I changed the wording, not the effective meaning. Darkfrog24 (talk) 23:10, 12 February 2010 (UTC)

Whether or not the January 25 change changed the official "rule", it is perceived to have changed the rule. The wording before January 25 indeed made a value judgment (basically, it was "we advise you to put punctuation before refs, like almost everybody does, but it's not absolutely required"), and in effect the pre-January-25 wording establishes a bias against punctuation-before-refs even if it doesn't absolutely prohibit it. Many of the changes since January 25 have either raised the bar further, or have lowered the bar, and it would be helpful to discuss this wording here rather than moving the bar around without consensus. Eubulides (talk) 01:50, 13 February 2010 (UTC)

This issue needs to be cleared up, and I would like to see a proper discussion. An editor was recently blocked for edit warring over precisely this issue. He was trying to revert formats entered by one of the original instigators of the idea that Nature-style citations were an alternative. But this instigator was also adding a non-breaking space before the citation. You can see what this approach looks like in this article. In my view, this approach offends KISS and look untidy. It means there is a notable break at the end of the sentence, followed by the citation index number in square brackets, and then finally we reach the period. This is not what the eye is use to seeing. The flow that marks the end of the sentence is lost, and instead the sentence jerks to an end.

Further, the Wikipedia format for citation index numbers is to place them inside square brackets. This is not the way it is done in Nature. As you can see from here, Nature underlines the index numbers, and definitely does not add a leading space. Because of these two factors, the sentences in Nature do not jerk at the end in such an ungainly way.

The advantage of the before punctuation style of Nature is that they have a precise scope. Because the citation is inside the punctuation, the scope is unambiguous. For example, if a sentence has a subclause bracketed with commas, then placing a citation just before the second comma clearly confines the scope of the citation to that particular subclause, and does not include the preceding part of the sentence. This gives a high level of unambiguous precision about the scope of the citation. It also means that at most you can cite a complete sentence. If you want to cite a paragraph, then each sentence needs to be separately cited.

By contrast, after punctuation style citations have an ambiguous scope. When placed at the end of a sentence they can reasonably be construed to refer to some or all of the the text from the citation back till another citation occurs or the start of the paragraph is reached.

What would be most logical would be to use both styles within the same article, depending on the scope of the citation. Thus, you would use before-style to cite one sentence, and after-style to cite multiple sentences. However, this is not something the general reader would be attuned to.

What I would not like to see is a pernickety rule that a Wikipedia article must use one style or the other. This will make for more unnecessary edit wars, like the one above (it's bad enough with American verses British English, having to trawl through the history to see which rules)

Currently I favour either the use of one style only (and after-style is the simplest and smoothest in practice) or of both styles. In the case of both styles, I favour allowing them to be mixed in the same article, with editors attuned to scope tailoring them as they need. This gives precision to those who want it, and those who don't won't really be bothered anyway.

In all cases, no spaces. I would also like to see a review of the square bracket around the citation index number. The Nature-style, which underlines the index numbers, flows more cleanly, particularly if citations are to be placed before punctuation. Square brackets could kick in when multiple citations are present. --Epipelagic (talk) 02:04, 13 February 2010 (UTC)

Now that is an impressive and well-thought-out proposal. However, I don't think that allowing both in and out in the same article is best. The readers would think "Huh, this is a bit sloppy" long before they thought that any difference in meaning was meant. We'd also have to teach the editors how to use it. Darkfrog24 (talk) 02:49, 13 February 2010 (UTC)

Please see What Nature does below. Eubulides (talk) 04:48, 13 February 2010 (UTC)

The purpose of the whole Manual of Style is unification, standardization, consistency, uniformity. Standardization is good for readers, it allows using of Bots and it also helps to avoid misunderstanding between wikipedians. Manual of Style clearly provides exact way what to do on examples. If somebody thinks, that s/he can write in his/her "new own style", then it is against spirit of Manual of Style. --Snek01 (talk) 22:59, 13 February 2010 (UTC)

What Nature does

• "The advantage of the before punctuation style of Nature is that they have a precise scope." No, Nature's style has the same scope problem that the after-punctuation style does. Nature always puts the citation before the punctuation, even if it "logically" would follow the punctuation, so one cannot tell in a Nature article if the citation is merely about the preceding parenthetical clause or about the entire earlier part of the sentence, or (indeed) whether the scope extends all the way to the previous citation. What Epipelagic's comment is proposing is a new style, which we could call "logical footnoting" by analogy with logical quotation. Does any high-quality publication use this style?
• "the Wikipedia format for citation index numbers is to place them inside square brackets. This is not the way it is done in Nature." Quite right, and a further point should be made: in HTML Nature underlines and comma-separates normal-size numbers like this^1,2; but in print they omit the underlines and use a smaller font, like this^1,2. Does any high-quality journal use Wikipedia-style superscript bracketed footnotes before punctuation, like this?

If the answer to either question is "no", that suggests Wikipedia articles shouldn't do it either. Wikipedia is not a good place for typographic innovation. Eubulides (talk) 04:48, 13 February 2010 (UTC)

I would prefer something like the Nature HTML style to be introduced on Wikipedia (presumably through a change in Cite.php or Common.css); the brackets are ugly, unnecessary, and (as far as I know) not in line with usage in high-quality reliable sources. Ucucha 04:59, 13 February 2010 (UTC)

• I agree that the brackets are distracting, unnecessary, and not in line with other high-quality publications. Can someone with some expertise in Cite.php etc. please comment? I assume that part of the problem is the hassle of dealing with multiple citations at the same location. They way Wikipedia currently does it, every citation is treated the same; but the way Nature (and many other publications) do it, the first citation in a string is treated differently, because it's not preceded by a comma
• This sounds like a problem that skins could solve. If you use some skins, the simple superscripts-with-commas style^1,2 is indeed underlined; you can see this by revisiting this thread in the Cologne Blue skin, for example. So the best way to implement your suggestion may be to use a simple style like that, and let skins insert brackets and commas as desired. The skin should also decide whether to put the punctuation before or after the superscripts: the default should be before, but an editor could express a preference for after by using the appropriate skin. Surely all this is implementable to someone with sufficient expertise.

Eubulides (talk) 19:23, 13 February 2010 (UTC)

So what Epipelagic has proposed, while interesting, would be a brand new style that none of our readers will have heard of. Darkfrog24 (talk) 13:47, 13 February 2010 (UTC)

The critical point is whether the ref tags are superscript and small-font. Nature has neither; nor do many of the parenthetical systems such as Harvard and Vancouver. WP uses superscript small-font, which leaves an ugly whit-space distance between the end of the last word and an isolated punctuation mark. Tony (talk) 00:47, 19 February 2010 (UTC)

Proposal: Inline citations

Inline citations may be placed either before or after closing punctuation, such as periods and commas, so long as each article is consistent.

Darkfrog24 (talk) 16:45, 12 February 2010 (UTC)

Mild support: I like the look of punct-before, but I see nothing inherently superior about it. Darkfrog24 (talk) 16:45, 12 February 2010 (UTC)

I would say that discussion of what to do in this area should be taken to WP:FN, if you want to change guidance on the matter. This would consolidate it with previous discussions on the matter and ensure consistency between the MOS and the footnotes guideline. Christopher Parham (talk) 16:49, 12 February 2010 (UTC)

This proposal suggests that we should change the MoS's existing instructions, which amount to "only use punct-before," to match those already on WP:FN, which allow both punct-before and punct-after. I doubt anyone on WP:FN would object to that. If we reach a consensus here that punct-after should not be allowed on Wikipedia, then we can take it to WP:FN. Darkfrog24 (talk) 23:16, 12 February 2010 (UTC)

The relevant guidelines on WP:FN are changing as we write this, so it's premature to propose to change wording here based on consensus there. Also, if you read the old discussions there, you'll see that there will probably be objections to any change: it's a contentious issue. Eubulides (talk) 01:50, 13 February 2010 (UTC)

Oppose No, the major sources say to put ref tags after the punctuation. It was always an absurd practice to put them before, and everyone but Nature, bless its heart, has realised this. The problem is acute with multiple tags, such as ... the cow chewed its cud^[3][5][6], and burped with satisfaction. Tony (talk) 02:09, 13 February 2010 (UTC)

• Support I don't think that there is any good reason to demand that editors use only one approach or another: the issue here is fundamentally personal preferences/aesthetic opinion. I would also support a statement that says that punct-before is more common because it is true, and because I think it would encourage people to use that style (which is, as it happens, my own personal preference -- just not one that I'm prepared to force onto other editors). WhatamIdoing (talk) 02:56, 13 February 2010 (UTC)

Oppose: I am not aware of any professionally published work that places a footnote call before a punction mark: it is typographically ugly, and even worse with multiple footnotes, as Tony points out. By the way, the proposal should be clarified: the discussion is about footnotes specifically, not inline citations generally (which would include Harvard-style references in parentheses).—Finell 04:10, 13 February 2010 (UTC)

As noted repeatedly in these discussions, Nature (journal) does exactly that. As one of the most widely respected scientific journals, I suspect that it meets your standard of a "professional published work". WhatamIdoing (talk) 18:02, 13 February 2010 (UTC)

Oppose: It seems that Nature is the only major journal using "before" style, possibly because this style has a more precise scope, as discussed above. But I oppose allowing this style as an alternative. Firstly, because its positioning within the structure of a sentence implies that its scope is restricted to that sentence. I do not believe most editors will intend that. Secondly, allowing two styles adds an unnecessary complexity and burden on editors, with no compensating gain. It will no doubt lead to a further rule that only one style should be used in a given article. This opens the way to unnecessary edit wars, and time consuming searches of the edit history to see which style takes precedence. --Epipelagic (talk) 05:33, 14 February 2010 (UTC)

Oppose. The question is whether both styles should be allowed, so: no—we have to have consistency at WP. "Aesthetic" freedom can't be article-dependent—surely? Which do we go with? The preference of the first editor to apply a reference? Nah. (I will admit that I personally prefer the ref-before-the-punctuation look; but that's another issue, and I'm aware that I'm in the minority.)  HWV258.  09:03, 14 February 2010 (UTC)

Strong Oppose - I am very much against this proposal for mechanical consistency, achieved by abandoning all existing rationale for one or other placement, yet not achieving consistency across the encyclopedia. Sorry, but it sounds like the worst of both worlds! I would consider supporting allowing both placements, and stating guidance pertaining to the meaning of each, given the greater degree of precision afforded. I see no reason why doing so would introduce any greater burden or risk than any other guideline. PL290 (talk) 11:55, 14 February 2010 (UTC)

I'm sorry, I'm confused. The goal of this proposal ("Inline citations may be placed either before or after closing punctuation, such as periods and commas, so long as each article is consistent") is to remove the existing demand for "consistency across the encyclopedia" and to explicitly allow either placement style. Your opposition says that you "would consider supporting allowing both" styles -- which, as far as I can tell, is exactly what the proposal does. But you've labeled your comment "strong oppose". What exactly are you opposing? The existing, disputed one-size-fits-all requirement, or the proposed permission to do what seems best in any given article? WhatamIdoing (talk) 22:42, 14 February 2010 (UTC)

A proposal to, as you put it, "do what seems best in any given article" regarding citation placement, is exactly what I'm saying I'd consider supporting. What I'm opposing is, "so long as each article is consistent." I think the only point in losing the one-size-fits-all consistency across the encyclopedia would be if greater precision were allowed each time a citation is used within an article. I recognize that some dispute whether such precision is really possible or meaningful, but if the general conclusion is that it's not, then let's at least have consistency across the encyclopedia. PL290 (talk) 15:54, 15 February 2010 (UTC)

Oppose Place them all after (see the multi-ref example from Tony above for why). Headbomb {^ταλκ_{κοντριβς} – WP Physics} 14:34, 15 February 2010 (UTC)

Support. Consistent with the existing consensus at the relevant guideline, as the MOS guidance on this matter ought to be. Christopher Parham (talk) 22:24, 18 February 2010 (UTC)

Comment - this appears to be gibberish. Can you explain what this means? --Epipelagic (talk) 03:56, 19 February 2010 (UTC)

I believe that ChrisP means that the proposal, allowing both styles so long as each article is consistent, matches WP:FN, the "relevant guideline." He also seems to believe that MoS should follow WP:FN et al and not the other way around. Darkfrog24 (talk) 13:27, 19 February 2010 (UTC)

Perhaps that is what he meant to say. It would helpful if he offered at least one reason, in light of the above discussion, why the old guideline at WP:FN should prevail. --Epipelagic (talk) 01:51, 20 February 2010 (UTC)

Oppose Nobody is forcing a newbie to write in the only excellent style. Both styles may be good, but this is a guideline and it should tend into an unification. Then robots could be able to change it from the one good style into one wikipedia unified style. It is step forward from "not bad" to "not also bad, but better". Unified style can not harm wikipedia, it can improve it. --Snek01 (talk) 02:39, 20 February 2010 (UTC)

Oppose. There are tons of minor typographic idiosyncrasies that we don't allow. This should be one of them. Strad (talk) 21:16, 20 February 2010 (UTC)

• Support. notes either side of punctuation. This has been the position for several years and after very long arguments, because people started to use bots to "correct" nature style footnote placement and understandably other editors who regularly edited those articles objected as for them footnotes after punctuation looked odd objected. Just as those who think it looks odd before would if we were to go totally for the nature style. BTW if an editor chooses to look in the archives it is clear that it is a lot more than just nature that uses this format. So the compromise agreed that both styles should be used and that there should be no change from one style to another. This is well documented in the archives of this guideline and WP:CITE. I will provide links if that is required, but people commenting here ought to be aware of the previous discussions.-- PBS (talk) 03:06, 24 March 2010 (UTC)
• Comment The wording "When the reference tag coincides with punctuation, it is most commonly placed immediately after the punctuation, except for dashes, as recommended by the Chicago Manual of Style and other style guides.[5][6] Some editors prefer the in-house style of journals such as Nature, which place references before punctuation. If an article has evolved using predominantly one style of ref tag placement, the whole article should conform to that style unless there is a consensus to change it." is the compromise wording that was on this page and on WP:CITE last time I took an interest in this subject and I think should be restored, to prevent bot alterations and edit wars. -- PBS (talk) 03:09, 24 March 2010 (UTC)

• Comment SlimVirgin writes above "For a long time, the agreement was ref after punc. Then a small group of editors managed to push through on CITE, as I recall, that we could choose." I don't think her recollections are correct here is a link to the archives "Wikipedia talk:Footnotes/Archive 8#Reference tags". It lists most of the sections to that date where this was discussed. In a section called Wikipedia talk:Footnotes/Archive 7#What do we agree on? (straw poll on ref tag placement) look for bolding "his has been Wikipedia's house style since 2006-04-04. 4-4" SV and some others agreed with the statement but a number of other editors disagreed with the satement. But read the archives and make up your own mind as to which view is correct -- PBS (talk) 21:36, 13 April 2010 (UTC)

Spaces before citations

See also Wikipedia talk:Footnotes/Archive 9#Reference tags and spaces -- PBS (talk) 21:36, 13 April 2010 (UTC)

In the guideline texts, there are actually two contentious issues: whether to place refs before or after punctuation and whether or not to place spaces before refs. The second issue appears to get missed, but we do need some discussion on it, considering that the Stemonitis-Snek01 incident that led indirectly to this discussion was in large part about placing spaces before refs. I am creating this subheading for discussion about this issue alone. Ucucha 03:01, 13 February 2010 (UTC)

• If spaces are allowed, they need to be non-breaking spaces, because you don't want the ref to get separated from the text by line wrap (especially if the ref is the very last thing in a paragraph). I don't like spaces, but I don't see a good reason to actively prohibit (non-breaking) spaces. WhatamIdoing (talk) 03:15, 13 February 2010 (UTC)

• I do not think that spaces are best in this case. (Thank you so much for making a new section. You're probably headed off a good deal of muddle.) Darkfrog24 (talk) 03:31, 13 February 2010 (UTC)

I am not aware of any professionally published work that inserts a space before a footnote call. Why are even non-contentious issues contentious here?—Finell 04:15, 13 February 2010 (UTC)

This issue caused two blocks, so it's contentious. Assuming you are correct about professionally published works, thank you for contributing that fact. I didn't know that, and such facts should contribute toward settlement. Art LaPella (talk) 05:09, 13 February 2010 (UTC)

I've seen it in 19th century books, along with spaces before semi-colons and colons. It is unfashionable in modern typesetting. I don't, however, believe that "uncommon" or "unfashionable" is the same as "we should prohibit this and block editors that use it". (I wonder whether having it all crammed together causes WP:ACCESS problems... I doubt it, but I'll go ask.) WhatamIdoing (talk) 18:43, 13 February 2010 (UTC)

I oppose allowing non-breaking spaces before citations. There seems to be no general acceptance of this practice. It is also against common sense. The end of a sentence is marked by a period followed by a space before the start of the next sentence. Adding the citation without a leading space clearly attaches the citation to the preceding sentence. Adding a leading space suspends the citation between the two sentences, pointing to neither. Allowing non-breaking spaces will add unnecessary complexity and impose an additional burden on editors with no compensating gain. Checking that non-breaking spaces have been used, and not ordinary spaces, will be (since it can only be checked in the source text) boring and time consuming. --Epipelagic (talk) 06:22, 14 February 2010 (UTC)

• Oppose spaces as making ref tags more cluttery. They need to be as unobtrusive as possible, and superscript numerals in square brackets—the way WP does it—causes awful clutter when either spaced or before the punctuation. Tony (talk) 07:21, 14 February 2010 (UTC)
• Oppose. Per Tony. Too ugly. Headbomb {^ταλκ_{κοντριβς} – WP Physics} 14:37, 15 February 2010 (UTC)
• Oppose. There is no needed a new style. --Snek01 (talk) 00:43, 19 February 2010 (UTC)

Comment. For those who desire a space before the inline cite, you can add it through your CSS. See Help:Cite messages#Customizing. ---— Gadget850 (Ed) ^talk 21:34, 20 February 2010 (UTC)

Fixing the text

Now that the proposal to bring the MoS into compliance with WP:FN has not reached an affirmative consensus, we should address the problem with the existing text.

Inline citations are placed after any punctuation such as a comma or period, without intervening space:

Because this phrasing has been interpreted both as "place citations after" and as "citations may be placed either after or before," I suggest that we replace it with something less ambiguous, such as the following:

Place inline citations after any punctuation, such as a period or comma. Do not put a space between the punctuation and the citation.

Darkfrog24 (talk) 17:20, 14 February 2010 (UTC)

• This appears to be the emerging consensus in the two above sections: we should not allow spaces or before-punctuation refs. I see arguments either way: on the one hand, I don't like prescriptivism where different editors prefer different styles; and on the other hand, this will make copyediting new articles easier, which often use all kinds of styles together, because the copyeditor doesn't have to worry which style is appropriate. Ucucha 16:12, 15 February 2010 (UTC)

Truth be told, I'm fine with either so long as the MoS states it clearly. Darkfrog24 (talk) 16:18, 15 February 2010 (UTC)

Does this text prohibit mid-sentence inline citations? Consider the case of two unrelated sources, supporting sentence fragments:

• Alternate names are One^[1] and Two.^[2]
• Alternate names are One^[1] and Two^[2].

Does this text demand that these be re-drafted as:

• Alternate names are One and Two.^[1][2]

PL290's is right: What's good enough in general may not work everywhere. My first example might reasonably suggest that the first source supports only the first word, and the second source supports both names. The second (pre-punct) format is clearer about what the source is actually supposed to be supporting, but it appears to be prohibited by this language. WhatamIdoing (talk) 00:10, 16 February 2010 (UTC)

You are right, of course, about the fine details of scope. But the reality is that almost no editors, who have not read the earlier discussion, will bother themselves with such niceties, and even fewer readers will have any awareness of them. The same ambiguities exist across the mainstream academic literature. As discussed above, we have no brief to innovate a rigour here which does not exist elsewhere. There is sometimes the idea that if MOS gets definitive or prescriptive then it is lumbering editors with needless bureaucracy. But in this case, if there are not straightforward rules, conforming to mainstream usage, the result will be needless grief for editors, trying to maintain consistency between alternate options, and leaving open the possibility of unnecessary edit wars of the type we have seen recently. The most straightforward compromise rule is simply "place citations after punctuation with no leading space". Normally, a citation within a sentence would refer to a sentence fragment which can be bracketed with commas, and there is no issue. The problematic sentence fragment could be recast as

• Alternate names are One,^[1] and Two.^[2]

Like most rules, there are occasional exceptions, hard to codify, maybe best left to common sense. I suppose dashes are an exception.

• The sailors set out on a raft—the start of an epic ordeal—^[1]but finally hit land.

should be

• The sailors set out on a raft—the start of an epic ordeal^[1]—but finally hit land. --Epipelagic (talk) 07:15, 17 February 2010 (UTC)

Ditto with parentheses. Didn't the MoS already say something like that (at least for parentheses) somewhere? ― A._di_M. (formerly Army1987) 10:57, 17 February 2010 (UTC)

This problem could be trivially addressed by using a descriptive statement rather than the imperious command apparently preferred by some: "Ref tags are normally placed after any punctuation such as a comma or period, without intervening space." It does not have the effect of actually prohibiting punct-after, but it is accurate in both its description of Wikipedia's community-wide consensus and in its description of the actual rules for that system.

Also, Epipelagic, your effort at recasting my example is wrong: It is improper to add an extraneous comma in that phrase. Serial commas require a minimum of three nouns ("One, Two, and Three" instead of "One, and Two"). WhatamIdoing (talk) 05:27, 19 February 2010 (UTC)

I appreciate that, I was suggesting a compromise. Maybe an exception to the three nouns rule, since when citations are involved it looks okay. But there is no way you will get logical positioning of citations unless both styles, before- and after-, are allowed in the same article. Personally, that's the style I'd like to see, but for reasons discussed above, it is not practical. Adopting the before-style has problems similar to the after-style, because the before-style implicitly confines its scope to one sentence. If you want to cite a paragraph with one source, then strictly you would need to cite every sentence separately with the same source. Would look horrible. --Epipelagic (talk) 06:50, 19 February 2010 (UTC)

Named reference definitions should be separate from named reference instances or there should be an option to hide certain references

I'm a programmer, so I'm using programming terminology here. If there is a clearer way to state this, please let me know.

The problem with named references is that one of the reference instances (i.e. a footnote) must contain the reference definition - the body of the reference between the <ref name="foo"> and </ref> tags. This makes it difficult to perform major alterations without accidentally removing the reference definition and thus making dependent reference instances (<ref name="foo"/>) produce an error. It creates a hidden dependence between certain footnotes that define other footnotes and when an article has many footnotes it can become very confusing.

For now I am working around this problem by keeping a list of reference definitions at the end of the article. This has made it far easier for me to remove and change entire sections without breaking all the references in the article. This is unacceptably messy for finished articles, though, as they appear at the end of the text in the article as [2][3][4], etc. It seems a simple change would be a flag attribute in the <ref> tag to make certain references hidden so the reference definitions at the end of the article could be hidden.

Of course there might be some existing option I am overlooking. I hope so. Factomancer (talk) 02:34, 13 April 2010 (UTC)

I think you want list defined references. ---— Gadget850 (Ed) ^talk 03:04, 13 April 2010 (UTC)

Thanks! That's exactly what I was looking for. Factomancer (talk) 18:54, 13 April 2010 (UTC)

Towards standardizing footnote names

I have been thinking for some time that we should, when possible, standardize the names of footnotes. The vast majority of footnotes are given author-date names, such as <ref name=Smith2009>, and I think this is the defacto standard.

I don't see this as a rule that we should be enforced. There have to be exceptions to this rule: there are many thousands of footnotes to sources where there is no author. In these cases we should use names based on a corporate author if possible (for example we might use <ref name="Rolling Stone2009">). I'm not sure if there are many other exceptions, however. A few hundred perhaps.

Again, I don't see this as a prescriptive rule. I see it as a gentle recommendation that we make in all our examples, a standard we strive for when we are editing articles ourselves and a vision of how future templates and citation tools should work. For example, it may be possible to combine {{sfn}}, {{citation}} and list defined references into a system that handles all the anchor names under the covers, using footnote names of the form AuthorYearPagenumber and shortened footnote links of the form given by {{harvid}}.

I think there is nothing uglier that footnote name like "GGFRB2". Editing an article with these notes is more like computer programming than writing, and there are many talented writers and researchers who are no good at reading computer code. ---- CharlesGillingham (talk) 21:20, 15 April 2010 (UTC)

You mean not a policy, but a style guideline? If you hope for it to gain any traction, probably best to propose alternatives, get consensus, and make it part of the Wikipedia:Footnotes style guidelines. Also, what happens when there's no author or no date? If we're going to do this, the guidelines should cover the different possibilities. -Frazzydee|✍ 21:49, 15 April 2010 (UTC)

Sorry but I'm not seeing a significant benefit to standardizing footnote names. It's completely transparent to the reader and only matters to the editor. For the latter it would just be another obstacle to producing a decent article. I think we have too many poorly cited articles to address without worrying about this level of detail.—RJH (talk) 21:52, 15 April 2010 (UTC)

I could see this included as a section on practices, but not as a guideline. I have been using just the last name for the most part or a journal name + ISO date, all in lower case with no spaces so I don't have to use quote marks. When using list defined references, I alphabetize the list by reference name, but I know some others who try to keep it in the same order as the inline cites, to the extent of adding commented numbers to match the cite numbers. If you want to see crazy refnames, there are a whole series of articles like Plutarch of Eretria. ---— Gadget850 (Ed) ^talk 23:08, 15 April 2010 (UTC)

We're not trying to prescribe (like medicine) a standard for the community to follow. We're trying to describe what the community is already doing. If you can demonstrate that a large section of the community is already doing this, then I'd be happy to document that consensus. If not, then the WP:ESSAYspace is thataway, and you're welcome to add your opinions to it. WhatamIdoing (talk) 00:47, 16 April 2010 (UTC)

(1) I would seriously doubt that "name=AuthorYear" is very common (what is the sample size?) I personally use "Author-page" or "Author-firstpage-lastpage" and would only put in AuthorYear-page if I needed to use it for disambiguation purposes. (2) I totally disagree with your argument WhatamIdoing, we do not describe what the community is doing, because guidelines are used within a positive loopback. If something is placed in a guideline, it is assumed to be the preferred option. Take for example the very long argument about the placing of footnotes ref-tags at punctuation. When it was described as being after punctuation, because if was an easy rule to follow, people launched bots to place all footnote ref-tags after punctuation. Take another example, in the MOS the majority of articles had all their dates linked primarily because it was the guidance given in the MOS. After a very bitter debate in the MOS the majority wanted the links removed (but there was not a general consensus that it should be done) -- it was the practice to link them. The result was that bots have been used to go through articles removing them. So it is just not true to say that guidelines only describe what the community is already doing, particularly given that WP:CONLIMITED states that guidelines are the consensus, not that they describe what editors do. -- PBS (talk) 07:28, 16 April 2010 (UTC)

I also tend to use Author-page, as I frequently have cites to two distinct places in the same book/article. --Bejnar (talk) 07:19, 18 April 2010 (UTC)

I guess I should make it clear where I'm coming from. I could sum up my position like this: "Wikitext should not look like computer code. It should resemble prose as much as possible." The reason, of course, is that there are talented potential contributors who have difficulty editing wikitext because they have no background in computer science. This is especially true when adding citations, where there are multiple choices, all of which resemble computer code more than normal citations. I think this is the underlying issue of several citation format debates, and the most important motivation for improving our citation implementation.

I'm aware that this is not an issue that can be addressed in the style guidelines at this time. My hope is that future tools and templates will one day make it make it possible for contributors to add citations without consulting a style guide at all. I mentioned this one issue here to draw it to the attention of people who are working on list-defined reference, templates like {{r}} or {{sfn}} and tools like WP:REFTOOLS, many of whom watch this page. ---- CharlesGillingham (talk) 20:14, 16 April 2010 (UTC)

I know what you mean, the learning curve to write an article has gone from not much more than writing a simple note on a newsgroup or an internet forum to something much more complicated, it must put off a lot of none computer literate people. The use of templates in citations is a very big hurdle and must put some people off bothering to cite their text, particularly if the article they are considering adding to has only templated citations to use as an example (because as beginners they will not have looked at WP:CITE and realise there are other ways of citing sources. -- PBS (talk) 01:30, 17 April 2010 (UTC)

I notice a lot of misunderstanding of the template terms. --Bejnar (talk) 07:19, 18 April 2010 (UTC)

The task of inserting templates could perhaps be simplified through the use of an addition of a template editor to the insertion tool near the bottom of edit window. I.e. select 'template' from the 'insert' menu, type in a valid template name, then a template form appears. Fill in the form and then the template would be inserted at the cursor point. Of course the template editors would need the ability to customize the form.—RJH (talk) 15:48, 17 April 2010 (UTC)

Yes, this is precisely the direction that WP:REFTOOLS is headed. ---- CharlesGillingham (talk) 21:41, 17 April 2010 (UTC)

Very nice. Hopefully at some point it will include options for entering identifier lookup codes such as doi, as well as cite conference/encyclopedia templates.—RJH (talk) 19:35, 18 April 2010 (UTC)

refToolPlus supports ISBN and DOI lookup, encylopedia, press release, map and reflist. ---— Gadget850 (Ed) ^talk 19:52, 18 April 2010 (UTC)

Looks good. The one concern I might have is that the list of coauthors and editors will often exceed the size of those small boxes.—RJH (talk) 20:08, 19 April 2010 (UTC)

LDR bot?

Is there a bot or script that can re-format the references in an article to use the list-defined reference method? Please? WhatamIdoing (talk) 22:19, 15 March 2010 (UTC)

What criteria would it use to generate the reference names?—RJH (talk) 15:59, 18 March 2010 (UTC)

If we could show consensus for bot conversions to LDR, I'm certain that there are several bot operators who could convert probably 99% of Wikipedia within a month or so. Unfortunately, acceptance of a change to LDR style references has been rather painfully slow to date. It's starting to gain acceptance now, which is terrific, but my sense is that there isn't that level of widespread support for them to where it would be responsible to release a bot (a user script is a slightly different story in my book, as that would be a bot more acceptable, but not by much). I'm perfectly willing to be convinced otherwise, though.
— V = IR (Talk • Contribs) 19:29, 19 March 2010 (UTC)

I'm not really looking for a mass-conversion of all articles: I'm looking for something that will do one at a time, on editor request. This could be a script that I need to install, or a category to list articles in for conversion, or anything like that.

A lot of the articles I edit format refs through Diberri's tool, so they're frequently already named. Some default scheme ("TempName1") could be used for those that weren't named; it would be easy to search and replace. WhatamIdoing (talk) 23:21, 23 March 2010 (UTC)

I ran across a ref named "autogenerated1" the other day, so presumably someone's got ref-naming code. WhatamIdoing (talk) 20:26, 26 March 2010 (UTC)

I know of two tools to give names to references: WP:REFLINKS uses "autogenerated1", WP:AWB can often guess a name based on author & year, else fall back to "ReferenceA". However, neither will convert to list-defined references. Rjwilmsi 20:58, 26 March 2010 (UTC)

The latter would probably work for books. For journals I like to use an abbreviation of the journal name plus the numerical volume and issue. (E.g. mnras535_2, for Monthly Notices of the RAS, vol. 535, issue 2.) It's usually fairly unique and I can tell the source type at a glance. I guess you could use a doi or one of the other identifiers, if they exist.—RJH (talk) 21:10, 26 March 2010 (UTC)

Yes, AWB already has various schemes to derive the reference name, author and year was just an example. Rjwilmsi 21:24, 26 March 2010 (UTC)

Likewise. Thanks.—RJH (talk) 17:08, 27 March 2010 (UTC)

This might do what we want... although for myself, I'm not going to try it until it's reached something more like a foolproof stage. WhatamIdoing (talk) 03:46, 22 April 2010 (UTC)

MoS naming style

There is currently an ongoing discussion about the future of this and others MoS naming style. Please consider the issues raised in the discussion and vote if you wish GnevinAWB (talk) 20:52, 25 April 2010 (UTC)

Headings

I think it would be desirable to make a somewhat stronger recommendation about separation of citation notes and explanatory notes. Recently, I've come articles that keep two sets of notes under the headings "Footnotes" and "Notes" in various combinations. Two examples are Blackbeard and Dutch 1913 battleship proposal. The problem is that the two terms are really synonymous. Considering how extremely rare it is to separate commentary notes from reference notes outside of Wikipedia we should really make more of an effort to keep clearer headings.

I believe "Footnotes" should be avoided as a heading altogether since it is never used in print literature. It is, after all, a typographical term referring to notes on the bottom of each page, not the end of the text, and I don't think I've ever seen it used as a heading in print. What individual editors choose to call citation notes is less of a concern. For refs it could be "Citations", "References" with just "Notes", "Commentary", "Annotations" for the commentary, or a slew of other combinations. What we should consciously avoid, though, are headings which amount to nothing clearer than "Notes A" and "Notes B".

Peter ^Isotalo 07:17, 20 April 2010 (UTC)

If there are no objections, I will go ahead and make a slight amendment along the lines of my suggestion in the next few days.

Peter ^Isotalo 17:38, 27 April 2010 (UTC)

Wikipedia articles are presented on web pages. The convention has arisen to use the term "web page" for what we see in a browser, and to consider it to be one page, even though it often is necessary to scroll to see it all. Since the notes under discussion appear at the foot of the web page, it is acceptable to call them footnotes. Jc3s5h (talk) 18:10, 27 April 2010 (UTC)

To the best of my knowledge, this defintion of the term "footnote" simply doesn't exist outside of Wikipedia. If you print out an article, they will appear as endnotes, pretty much the same as in printed articles or essays. And the whole point of calling something a footnote is if it doesn't require a separate heading.

Peter ^Isotalo 14:13, 28 April 2010 (UTC)

I've referred to the thing that comes at the end of a page as a footnote for 30 years or more, much before Wikipedia existed. Truthkeeper88 (talk) 19:36, 1 May 2010 (UTC)

Yes, that's exactly what I'm saying. At the bottom of a page, right below the text it comments, it's called a footnote. I'm pretty sure it's also common to call endnotes "footnotes" in everyday speech or non-specific descriptions. "Footnotes" as a heading, though, is something I believe I've never seen outside Wikipedia, and especially not right along a heading with the title "Notes".

Peter ^Isotalo 09:32, 2 May 2010 (UTC)

I've had a look at what MLA and the Chicago Manual of Style have to say on the issue, and they seem pretty clear about the separation.

• Chicago: "Footnotes are numbered citations listed at the bottom of each page in the research paper. Endnotes are numbered citations listed on a single page at the end of the research paper."[10]
• MLA: "MLA recommends that all notes be listed on a separate page entitled Notes (centered, no formatting). (Use Note if there is only one note.) The Notes page should appear before the Works Cited page." and describes footnotes as placed "below the text body".[11]
• Cite Right: A Quick Guide to Citation Styles (p. 49) points out that MLA's citations for the Humanities, notes can be used for commentary, but that this requires all citations to be in-text, ei parenthetical.[12]

As far as I can tell, our own article on footnotes (which includes endnotes) describes real life standards and the distinction between the two terms rather well. It's our own application of the citation standards in articles that has generated some pretty novel combinations of styles in certain articles, like the use of several sets of notes and the even stranger habit of placing superscript notes after notes.

Peter ^Isotalo 10:12, 2 May 2010 (UTC)

That makes a lot of sense..given the choice between "endnote" and "footnote", "endnote" would seem to be more appropriate. --Rsl12 (talk) 15:19, 4 May 2010 (UTC)

The use of Footnotes in articles like the ones used as examples such as Blackbeard mirrors precisely the Chicago definition quoted above, when one remembers that each article is a single web page, not a series of pages. The page is often longer than can be viewed on a display without scrolling, but it is nevertheless a single page.Malleus Fatuorum 16:06, 4 May 2010 (UTC)

I haven't had the opportunity to respond here, because I'm extremely busy in the real world, but the response I would have added is the one Malleus posted: Wikipedia articles are not separate sheets of paper. Therefore, the parameters allow more flexibility. Moreover, MLA (which is applied to humanities only) updated recommendations in 2009. In-text citations are no longer required. Keep in mind, also, that different disciplines have different citation methods. Finally, when a paper lands on the desk of a professor with the distinctive superscripts from Wikipedia, it's very clear where the paper came from. When papers were being written on typewriters the superscript was required and easy to format; not until the newest version of MS Word are superscripts easily supported - except here. In my view, this is an issue not worth changing. Truthkeeper88 (talk) 16:54, 4 May 2010 (UTC)

Anyone who's using MLA really ought to be asking at WP:PAREN, not WP:FOOT. MLA has never supported superscripted numbers. WhatamIdoing (talk) 18:37, 4 May 2010 (UTC)

MLA moved from superscripted numbers to parenthetical citations. But, the issue here isn't MLA per se, as much as how to format notes and footnotes on the page, and what constitutes a page. Still don't see a need for change. Truthkeeper88 (talk) 18:43, 4 May 2010 (UTC)

Good points. The more I think about it, the more ambivalent I am to the whole issue. Earlier I said I hoped not to get involved--I should have kept my resolve. :) --Rsl12 (talk) 20:45, 4 May 2010 (UTC)

Okay, sorry for taking so long to reply, but I've been quite busy. I'd just like to stress that citing MLA was really just a way to illustrate how separation of references and commentary notes is handled IRL. The point was that a double set of endnotes is something that is very rare and doesn't appear to occur in any manuals of style.

I don't believe there's much point in trying to extend the definition of "footnote" of our own accord to cover all instances of notes on Wikipedia. The Chicago Manual of Style's default term is after all "notes", which covers both foot- and endnotes alike, and this seems to be the same case with others manuals as well. In fact, every single description of footnotes in Chicago clearly refers to notes that are supposed to appear at the foot of multiple-page documents without a separate heading. Chicago says nothing about web pages at all. Anything about it would obviously be a personal interpretation.

What's worth noting is that the arguments that Wikipedia notes should be called footnotes are quite inconsistent, even without comparing it with IRL manuals of style. Our notes don't even appear at the foot of pages, by rather after the main text and "See also" and before the bibliography, just like normal endnotes. Even if in the very few instances taht they're placed after the bibliography, they are without any exceptions placed before external links and categories. And on top of that, they appear under a separate heading, which footnotes simply never do.

Peter ^Isotalo 11:21, 24 May 2010 (UTC)

Not sure why this is discussed here, as this is about the Cite footnotes system. Wikipedia:Manual of Style (layout) would be more appropriate. ---— Gadget850 (Ed) ^talk 20:41, 26 May 2010 (UTC)

another example for list defined references

I think this example shows a bit more why these were created in the first place and why people might want to use them. I was thinking of adding it to the main article but... im not sure if its 'consensus' .. anyways.. here it is. Thanks Decora (talk) 23:11, 11 May 2010 (UTC)

Separating references from text

(adapted from mediawiki Cite documentation, which was originally by User Tgr)

Sometimes reftags within an article can 'overwhelm' the text. Long cite tags within ref tags can render the original paragraph almost unreadable and uneditable. As of late 2009, mediawiki's Cite extension was modified so that you can move your reftags to the reference section at the end of the page, making it much easier to read and edit the original text. For example, consider this paragraph (color added for instructional purposes):

According to scientists, the Sun is pretty big.<ref name="miller"> {{ cite book | title = The Sun | author = E. Miller | date = 2005 | publisher = New York: Academic Press | date = 2005 | pages = 23-5 | isbn = 12345895 }} </ref> The Moon, however, is not so big<ref name="smith"> {{ cite journal | author = R. Smith | title = Size of the Moon | journal = Scientific American | volume = 46 | date = April 1978 | pages = 44-6. }} </ref>. The Earth is in-between<ref name="miller"/>. But nothing is as big as a supernova <ref name="mrx2007super">{{ cite web | url = http://www.mr-x.sagan.bsa.gov.elbo/stellar_obs_lab/documents/2007/supernova.pdf | format = pdf | title = Supernova: Really, really big | author = Dr Eibert Alnstein | date = 2007 11 2 | accessdate = 2010 5 8 | publisher = Big Space Agency, Republic of Elbonia }} </ref>.

==Notes==

<references/>

This may be considered 'cluttered' by some, as it has become difficult to read and edit the original paragraph. It can be rewritten as follows into a 'less cluttered' version:

According to scientists, the Sun is pretty big.<ref name="miller"/>The Moon, however, is not so big<ref name="smith"/>. The Earth is in-between<ref name="miller"/>. But nothing is as big as a supernova<ref name="mrx2007super"/>.

==Notes==

<references>

<ref name="miller"> {{ cite book | title = The Sun | author = E. Miller | date = 2005 | publisher = New York: Academic Press | date = 2005 | pages = 23-5 | isbn = 12345895 }} </ref>

<ref name="smith"> {{ cite journal | author = R. Smith | title = Size of the Moon | journal = Scientific American | volume = 46 | date = April 1978 | pages = 44-6. }} </ref>

<ref name="mrx2007super">
{{ cite web
| url = http://www.mr-x.sagan.bsa.gov.elbo/stellar_obs_lab/documents/2007/supernova.pdf
| format = pdf
| title = Supernova: Really, really big
| author = Dr Eibert Alnstein
| date = 2007 11 2
| accessdate = 2010 5 8
| publisher = Big Space Agency, Republic of Elbonia }}
</ref>

<references/>

(end of example)

Seems like LDR is worth a separate page. Still need to document practices: how to park unused references, document issues with #tag:ref, using templates such as {{r}} and so forth. ---— Gadget850 (Ed) ^talk 18:00, 12 May 2010 (UTC)

I've been using WP:LDR on several articles of late. The primary drawback of LDR (as I see it) is that when you have an article section open for editing, you then have to perform a second edit to move a citation to the references section. A variant of the split screen editor might be handy; allowing you to edit a section and the references at the same time. One possibility is a link that opens the references section for editing under a new tab on the browser.—RJH (talk) 15:02, 26 May 2010 (UTC)

Another drawback: you need aditional <ref name="..." /> than without LDR. --Philcha (talk) 19:41, 26 May 2010 (UTC)

This is out of date, needs update

It assumes that the <ref> is the most recent. It should also List-defined references, {{Harvard citation}} and {{sfn}} and their strengths and weaknesses. --Philcha (talk) 00:50, 25 May 2010 (UTC)

Your confusions stems from the use of the simple term footnotes to refer to the referencing system provided by the Cite software extension and implemented by <ref>...</ref> tags. Thus, this page is exactly about <ref>...</ref> tags and not the other systems; for an overview, see Wikipedia:Citing sources. ---— Gadget850 (Ed) ^talk 20:29, 26 May 2010 (UTC)

Thanks, I see Wikipedia:Citing sources is in the lead. --Philcha (talk) 21:24, 26 May 2010 (UTC)

You are welcome. We need a better name, but this has been in use for over four years. ---— Gadget850 (Ed) ^talk 22:18, 26 May 2010 (UTC)

Softcopy vs hardcopy

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

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

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

I intend to continue to supply softcopy with hardcopy when available. I will have to check relevant material and dumb one or the other down to match, I suppose. Annoying. Any thoughts or experience from others? Student7 (talk) 00:05, 25 May 2010 (UTC)

How about both, distinguished by the format=... attribute? --Philcha (talk) 00:37, 25 May 2010 (UTC)

Not sure why you are discussing this here, as this is about the Cite footnotes system. Wikipedia talk:Citing sources would be more appropriate. ---— Gadget850 (Ed) ^talk 20:33, 26 May 2010 (UTC)

Thank you. I will move my question there. Student7 (talk) 10:31, 27 May 2010 (UTC)

Footnote3

From the lead:

Editors may also use the older system of template-based footnotes, such as {{ref label}} and {{note label}}.

The use of {{ref}} and variants was the prevalent system before the introduction of {{cite.php}} in December 2005. As documented at Wikipedia:Footnote3, that system has been deprecated for referencing. Thus, the statement in the lead is at odds with the Footnote3 documentation. ---— Gadget850 (Ed) ^talk 02:50, 28 May 2010 (UTC)

REFPUNC doesn't link

WP:REFPUNC links to the top of the page. It should link to the same place as WP:REFPUN and WP:REFPUNCT. (Note this is unrelated to previous threads about the content of that guideline.) Art LaPella (talk) 05:22, 10 June 2010 (UTC)

 Done Fixed. Wtmitchell (talk) (earlier Boracay Bill) 01:19, 11 June 2010 (UTC)

Discontinuiing {{rp}}?

I just ran into {{rp}} for the first time, and I quite frankly don't understand why we ever allowed it. It introduces a highly oblique and original form of referencing that is unintuitive and difficult even for experienced editors and academics to understand without explanation. The only reason that it's around appears to be that some editors feel that certain references can be cited "too many" times. The argument at the bottom of this page is even more cryptic in claiming that there is such a thing as an "exceedingly long entry that cite[s] too many pages to possibly be useful to readers". Where is the logic? If a work is cited several times, no matter how many, it has to be considered necessary for the verifiability of the article. If not, it's obviously the frequency of referencing that must be the problem, not using common reference standards. Considering that Wikipedia is not paper, it seems even stranger to complain about over-long reference sections. Imagine if we argued the same way about the length of articles. Would anyone accept that we invented our own system of abbreviations for common words and implemented in article space? Or new standards of spelling for that matter?

Peter ^Isotalo 13:05, 22 June 2010 (UTC)

Peter, you may be right, but you'll give an example (wikilink). I know some reference techniques that can about 200 for a long article, and that's just making reading slow and hard. --Philcha (talk) 13:59, 22 June 2010 (UTC)

Here are some examples:

• Glossary of cue sports terms has previously been cited as the reason that the template was made in the first place (see template talk:rp). The reasoning here seems somewhat odd to me, since the encyclopedic nature of sources and the list itself precludes the need for most of the page citations (it would be a matter of merely looking up the word in the specified source whether you had a page citation or not). Inventing an entirely new citation layout seems like a way of fixing a perceived problem of "messy" layout by replacing it with a highly ambiguous, experimental and previously unknown type of page reference.
• William Hanna is the article where I saw this template for the first time and reacted to it. In that case, the number of individual citations to a single refs doens't even go beyond 20, which makes the original argument of purportedly excessive number more or less irrelevant. Here it seems to have been reduced entirely to a matter of the personal preference of the primary contributor.
• The most common standard to present page citations is still the method used in articles like today's TFA Ernest Hemingway. There are several variants on this theme and how to actually format notes, but not a single one of them presents page citations separated from the individual work. I don't know of any academic standard of reference or manual of style that even implies that it's a good idea to write page numbers in superscript directly after the note number. The number itself is after all only intended as a reference to a longer text, not to function as some sort of shorthand for that text.

A question, btw: how exactly do you figure a lot of notes would make reading slow? Each note is supposed to be read in relation to the text it supports, and more or less isolated from other notes. They have little or no meaning on their own. I can't imagine that any reader would ever want or even need to go through an entire set of notes from beginning to end. Since the reduction of notes seems rather unnecessary per WP:NOTPAPER, what exactly is the point of the reduction?

Peter ^Isotalo 09:24, 23 June 2010 (UTC)

Sad to see that William Hanna died.

That's a small- to medium-length for FA. Lemur, also FA, is really long and anything that increases reading diffifulty should be avoided. --Philcha (talk) 12:20, 23 June 2010 (UTC)

In general, the spirit of WP:CITE is that essentially any system for citations is acceptable. We should not, in general, change them simply because they use a different style than we would use ourselves.

I think I understand the reasoning behind these. On one hand, many references should have page numbers. On the other hand, with the "named reference" technique, it isn't possible to put that page number into the footnote. You can put a long list of page numbers into the footnote, but that's not very specific.

That argument seems reasonable enough to me. Combined with our general policy not to change reference formatting for stylistic reasons, I think the articles are OK as they stand. — Carl (CBM · talk) 11:26, 23 June 2010 (UTC)

The in-text page number feature is in {{R}}, {{Sfn}}, {{Harvard citation no brackets}}, {{harvnb}}, {{tl}}, {{Harvard citation text}}, {{Harvcoltxt}}, {{Harvcol}}, {{Harvcolnb}} and {{Harvard citations}}.

If we discontinue {{Rp}} then we need to the follow up with one of these options:

• Don't cite page numbers
• Don't use named references where the page numbers are different
• Don't use standard footnotes, use only shortened or Harvard

---— Gadget850 (Ed) ^talk 12:23, 23 June 2010 (UTC)

Why would we have to do any of the above? The complaint is strictly about superscript page numbers, not the citation of specific pages.

Peter ^Isotalo 13:06, 23 June 2010 (UTC)

If you remove the in-text page number, where are you going to put it? ---— Gadget850 (Ed) ^talk 13:11, 23 June 2010 (UTC)

In a separate, unique note, of course. That's the recognizable standard used for virtually all reference formats that use notes and in the vast majority of our own articles, including FAs and GAs. Is this in any way problematic?

Peter ^Isotalo 13:36, 23 June 2010 (UTC)

So, for article that use standard footnotes, each citation from the same source but with a separate page number must have a separate and unique entry in the reference list. ---— Gadget850 (Ed) ^talk 14:45, 23 June 2010 (UTC)

That's the near-universal reference standard used for works with reference notes, print and web, academic or otherwise. It's highly legible, easy to read and understand, even for those not used to footnotes. The connection between the page number and the cited work is not obscured by additional clicks (or the considerable distance in printed-out documents) and it's instantly recongizable to just about anyone who's ever taken a university course. In essence, KISS.

The superscript page numbers as they're used in William Hanna lacks most, if not all, of these benefits. I can't recall that I've seen any other reference layout that has received this many independent complaints concerning clarity, despite being fairly uncommon.

Peter ^Isotalo 15:47, 23 June 2010 (UTC)

Of course, Harvard, APA, MLA and other styles include the page number in-text. But this guideline only covers Cite footnotes, so we will ignore that for the moment.

How should we handle list articles that reference the same source hundreds of times and the editor desire to cite the page number?

You will probably be interested in commenting on T15127, a proposal to add in-text page numbers to the Cite system, essentially replacing Rp.

---— Gadget850 (Ed) ^talk 16:42, 23 June 2010 (UTC)

In-text references is not the issue, nor are parenthetical in-text references (a widely-used and recognized standard). The complaint is about page numbers that come in superscript format, separated from the actual specification of the reference, according to a standard that is practically unheard of in most disciplines. The citation of a single source literally hundreds of times is (and should be) an extremely rare occurrence, limited to a few highly specialized articles. And even in those limited few cases, it seems very contrived and technocratic to apply an exotic mix of standards for notes just because some editors feel that a large-ish notes section looks "messy".

There's virtually zero gain for readers in "tidying" an online document of the equivalent of a printed page or two of notes. Nevertheless, the move throws valuable legibility and standardization out the window. And as soon as you come up with a new format, someone sooner or later comes along and starts applying the format to articles that has absolutely no need for it, like William Hanna. And there seems to be plenty more even worse examples of excessive usage in articles like agricultural science (0 uses of the same reference), Albert Einstein (5) and antimony (7). And I didn't even have to search for those; they were the first three random links I clicked on at Special:WhatLinksHere/Template:Rp.

Peter ^Isotalo 17:43, 23 June 2010 (UTC)

I think this argument needs to be made at the individual articles; our policy has always been that any consistent style is acceptable, even if it is not common elsewhere. I did a quick scan, and the Rp template is used on one featured article, Robert Baden-Powell, 1st Baron Baden-Powell, and 51 good articles including your example Albert Einstein. It isn't our place here to force those articles to change. — Carl (CBM · talk) 21:07, 1 July 2010 (UTC)

I agree with Peter, on all his points: for multiple citations of the same source, shortened notes are more standard, more versatile and more readable than super-scripted page numbers. However, I also believe in Wikipedia's principle that editors have a right to experiment with unpopular citation formats, first of all to avoid edit warring and on the off chance that someone invents something that is clearly an improvement. So I don't think there is any reason to take any action against {{Rp}}. However, it is important that supporters of unpopular citation styles avoid changing large numbers of articles. Let newer editors decide which system they like best. ---- CharlesGillingham (talk) 21:29, 2 July 2010 (UTC)

<references/> tag

This page has always permitted a choice on each article between the <references/> tag and the {{reflist}} template. These do generate different output: the {{reflist}} template formats the references with a smaller font size. For articles with many references, the smaller font size of the template might be appropriate. But for articles with only a few references, there's no reason for the smaller font.

I would argue that the smaller font serves only to make the references look ancillary to the article – using a smaller font when there is no limit to the vertical size of the article says that we don't expect readers to actually read the references, that they are only there for show. However, some disagree, and we can work together without agreeing on such stylistic matters.

But we should maintain the current language that allows editors to choose for each article which one they want, and discourages editors from making large-scale changes from one to the other. — Carl (CBM · talk) 21:13, 1 July 2010 (UTC)

Agree. ---— Gadget850 (Ed) ^talk 22:22, 1 July 2010 (UTC)

I disagree, we should maintain a preference for reflist except when references produces better output (per common sense). This also follows practice and what Carl described above. Verbal chat 22:25, 1 July 2010 (UTC)

Current practice is that each article's editors may choose, and that's the right balance. If the editors agree that one version is better, they can choose it. Otherwise, if there are not compelling arguments either way, it's better to keep the established style than to have people argue endlessly and revert back and forth. — Carl (CBM · talk) 23:01, 1 July 2010 (UTC)

If your only problem with {{Reflist}} is the small font size, why don't you address this at Template talk:Reflist or Wikipedia talk:WikiProject Usability? There is no reason to restrict the use of {{Reflist}}, given the additional features of this template (columns, colwidth). --bender235 (talk) 09:12, 2 July 2010 (UTC)

The font size is the only default change. Obviously, changing the number of columns also changes the formatting of the article. For an article with only a few references, those things are also unnecessary. However, I believe many people use {{reflist}} explicitly to acieve the smaller font size, so changing that now would be impossible. — Carl (CBM · talk) 10:27, 2 July 2010 (UTC)

And how about adding a parameter to {{Reflist}} that allows to display the references in normal font size? That could easily done, and leaves you the option to implement {{Reflist}} while keeping the references style, which seems to be utterly important to you. --bender235 (talk) 11:14, 2 July 2010 (UTC)

The difficulty with that is that it would only lead to additional disputes about whether to use the parameter. This is particularly true if someone replaces <references/> with just {{reflist}} (small size) and then someone else adds the large parameter. Now neither person can claim to be restoring the original style, which was just to use <references/>. The rule "use the original style" is useful exactly because it gives an easy and fast resolution to such arguments, which otherwise would go on endlessly. Imagine if we replaced every instance of the word "colour" with a template {{color}} and then said "you can add the ukenglish parameter if you want the spelling with a u". — Carl (CBM · talk) 11:25, 2 July 2010 (UTC)

Policy has always been that you don't have to use citation templates, as long as use is consistent in the article; this should hold for referencing templates as well. BTW: The small font for {{reflist}} can be overridden in personal CSS; it is also broken for a large number of readers, but there has been no consensus to fix it. ---— Gadget850 (Ed) ^talk 11:58, 2 July 2010 (UTC)

Clarity

This article is very unclear about how to accurately add references to an article. Nantucketnoon (talk) 08:57, 25 July 2010 (UTC) I figured it out, but still think it could be more concise. Nantucketnoon (talk) 09:24, 25 July 2010 (UTC)

Endnotes?

Every single reference to footnotes on this MoS page is what in any manual of style would be considered an endnote. As a collection of web documents, Wikipedia does not have anything resembling the page structure of a book, and is unlikely to ever have one. None of our "footnotes" have ever behaved like actual footnotes. I don't personally really care much what terms we use in discussion and everyday talk, but it seems obviously misleading that we use a term that actually means something else.

But rather than debating over whether this should be "(endnotes)" or "(footnotes)", there's the perfectly neutral alternative "(notes)". Any objections to a move and more precise wording?

Peter ^Isotalo 14:59, 28 July 2010 (UTC)

This Wikipedia page is about the numerical in-text citation system that is enabled by the Cite software extension using <ref>...</ref> tags. Contrary to the title, it is not about footnotes or endnotes in general. ---— Gadget850 (Ed) ^talk 13:23, 2 August 2010 (UTC)

That's the technical part of it all, the way the notes are enabled. I'm talking about the description of their actual appearance, which is not unique to Wikipedia. In that sense, it's merely information on how to apply a variant on endnotes to a specific type of text.

Peter ^Isotalo 06:45, 3 August 2010 (UTC)

Previewing footnotes

When I am adding or editing a footnote to a section, I add a temporary <references> tag to check for typos. Sometimes, my human nature lets me forget to remove it before I save. Then I have to publicly admit that I goofed. Worse, I don't notice! Hopefully, this has never happened :-).

Is there a better way to preview a footnote when editing a section? John Harvey (talk) 13:02, 8 September 2010 (UTC)

there was a discussion here after someone asked if this could be added to Edittools with information about some solutions, though I've no experience of them myself.--JohnBlackburne^words_deeds 13:16, 8 September 2010 (UTC)

I use User:Anomie/ajaxpreview.js. ---— Gadget850 (Ed) ^talk 03:04, 25 September 2010 (UTC)

Wikipedia-wide uniformity

I'm sure this has been said before, but having 50 million different standards and letting the contributors to an article decide which should apply is counterproductive and unprofessional. We've got a well established template system that can be applied, and a large number of tools built around that template. That, or another system of similar scope needs to be the one true system for Wikipedia articles. It's stable enough to be machine parseable, it's maintained, and most importantly, it's already there, and we are already deeply invested into it. This isn't something that has to happen overnight - in fact I'd suggest that articles be left alone unless someone is willling to convert them in their entirety to a unified format - but if someone is willing to do that work, I don't think the process should be out there for debate. I do realize that not everyone knows how to use cite templates, but they are well documented, and there's plenty of help out there - and citations added without the appropriate template can just be fixed. Let's quit arguing about the merits of different systems, and find one, and only one that works, and stick to it. Triona (talk) 08:15, 10 August 2010 (UTC)

I'm definitely in favor of more uniform references, but I'm against a uniform adoption citation templates; they are a major liability in terms of user-friendlieness for newcomers. Virtually the typographical appearance can usually be achieved without them. I think a compromise should be met where we at least decide on a handful of standards to choose from rather than accepting literally any standard of any academic discipline out there.

Peter ^Isotalo 10:42, 10 August 2010 (UTC)

Triona you wrote "I don't think the process should be out there for debate." how do you tie that statement into WP:CONSENSUS and in this guideline "Use of such templates is neither encouraged nor discouraged; see WP:CITE." and in WP:CITE "Once a style is selected for an article it is inappropriate to change to another, unless there is a reason that goes beyond mere choice of style." and Wikipedia:Manual of Style#Stability of articles. PBS (talk) 11:35, 10 August 2010 (UTC)

Peter, in re the relative noob-friendliness of templated vs. hand crafted cites, I came to WP as a noob, and found templated cites much more friendly; e.g. something like

• {{cite journal |author=S.E. Jones |title=Muon-Catalysed Fusion Revisited |journal=Nature |volume=321 |pages=127–133 |year=1986 |doi=10.1038/321127a0}}

vs. something like

• S.E. Jones (1986). "Muon-Catalysed Fusion Revisited". ''Nature'' '''321''': 127–133. {{plainlinks|http://en.wikipedia.org/wiki/Digital_object_identifier|doi}}:[http://dx.doi.org/10.1038%2F321127a0 10.1038/321127a0].

Being a noob, I had (for that matter, still have) no idea in the world what order the various parts of a full citation go in, what parts to place in quotes, what to italicize, what to place in parens, what to embolden, etc. — I gathered that there were a bunch of different ways to present citations but didn't know the details of any of them and didn't much care about those details (and still don't); I was (and remain) happy to leave those details to the template designers to work out. Also, should WP ever settle on a WP-wide standard, that standard would be easily implemented for legacy cites which are templated, and a godawful nightmare to implement for legacy cites which are handcrafted. Wtmitchell (talk) (earlier Boracay Bill) 02:02, 11 August 2010 (UTC)

I am impressed that you learn to do that straight off. For most people just put in whatever they think best or within their ability (usually looking at the other citations on the page). Although yes templates are easier to use, it is a steep learning curve over just plonkinag in

• [1]
• John Smith "A book on widget manufacturing" page 123

Which is of course better than so put off by complicated templates or processes that no reference is attempted by new-comer. -- PBS (talk) 07:02, 11 August 2010 (UTC)

I agree with PBS here. Getting used to templates, even for someone who's just editing the text in general, is on the whole rather difficult for a lot of users. That a lot of experienced users testify that they were never bothered by it is hardly surprising, given our general demographics. Learning the order in which (full) refs are written is unavoidable for any experience user, though. For one thing, it will be included in most references of any major usefulness so it's mostly just a matter of straight copying. And there are very few logical deviations from the correct order. Achieving uniformity in articles should not burden newbies, only experienced users who already know how it's done.

And concerning Wikipedia-wide uniformity in references, I'm confident it will never happen. One single standard will not prevail since there are a number of variants that all have something going for them. The types of sources used for any single article and the discipline it's written in will usually determine which to choose. What we should work for, though, is to at least minimize the number of exotic and difficult-to-decipher standards, such as having page numbers after the superscript not (as in William Hanna and a few other articles) or creative mixing of standards.

Peter ^Isotalo 09:12, 12 August 2010 (UTC)

While you are probably right on this Peter, there are significant technical and stylistic advantages towards using the cite templates. I really do think that the way to go forward is to put in enabling language within the MOS that simplifies things for editors trying to move an article to the cite templates, provided that it does not leave the article with inconsistent referencing style (by only partly converting), or cause undue disruption. I'd actually encourage making the full conversion with a single edit, just to minimize disruption. Theoretically, there's an additional advantage to be had, it should become possible for individual users to customize how they view references once they are in a standard format. Triona (talk) 08:24, 17 August 2010 (UTC)

We have tools for using templates to make citations - Magnus' citation builder and refTools - both driven by forms. If I'd know these when I registered, I'd have been much more productive. Within a couple of days newcomers can make citations as easily as more experienced editors. --Philcha (talk) 11:08, 17 August 2010 (UTC)

If we're to judge this based on personal opinion, I know I would have found citation templates a clunky, complicated and unnecessary nuisance if I started editing as a newbie today. Like you, I'm assuming I would share this opinion of templates with plenty of other newcomers. I get the feeling that you're not taking into account that love of software gadgetry is not universal nor equal among all individuals, including Wikipedia editors.

Peter ^Isotalo 22:12, 19 August 2010 (UTC)

I'm confused. I edit a fair amount here, but remain bewildered by the "behind the curtain" formulae to get stuff to look the way I want it to on the page. I was utterly unable to footnote until I stumbled upon the templates, because cutting-and-pasting the format from the article I happened to be editing didn't tell me what info to insert where -- only templates unambiguously tell me, "put page number here", "put title of the book chapter here", "the ISBN number goes here without hyphens -- we will insert them". Could you explain how templates are more confusing than trying to figure out what info goes into a footnote if you don't already know? BTW, I came here to find out what the latest format is for footnotes and found the same bewilderment and lack of consideration I encountered as a newbie: Wikipedia's refusal to tell an editor what format to use is tantamount to encourging him/her to skip citing sources. It's like dialing the Information operator to have a taxi sent to your home toute de suite and, when asked which taxi company I want to use, being told, "I am not allowed to recommend a taxi company to you, so you must tell me which one you want or I can't help you -- even though you just told me you don't know anything about local taxi companies. Sorry!" This is infuriating. When a footnote's called for, why can't we hit a button which brings up a footnote form (either the one most recently used in that article or the one most often used on WP as of that day), then it shows me 1. the format's wiki mark-up with instructions to "substitute your source's info in each slot for the word 'blank'", and 2. the appropriate footnote template, but which tells me what item of info to insert into each slot? Guaranteed to increase citations! FactStraight (talk) 23:02, 5 September 2010 (UTC)

None of that actually requires a template. What you're asking for are instructions or at best subst versions of templates. It's certainly a matter of implementing new technical solutions but it does not require us to add any type of code. As for confusion about citation standards, it is only incidentally associated with templates, since it's a matter of style. Agreeing on using templates in general won't suddenly make everyone agree on which template to use.

And since I'm someone has always used raw templates without code, I'm a bit confused as to why it would be so difficult to add author, year, title and page number without a template. Because that's basically all you need to cite something. And with an online source you need nothing but a link. As far as I'm concerned, templates tend to complicate things immensely by asking for nigh-to-perfectly-formed citations. They can ask for up to a dozen or so superfluous details which newbies don't really need to be bothered with on their first few attempts at adding verifiable facts.

Peter ^Isotalo 07:05, 6 September 2010 (UTC)

We will never achieve side-wide uniformity, but we should be trying. The usual cite templates are a defacto standard and we should be focusing on them and the various tools that support them. The stylistic nits such as just what gets rendered in italics and where commas and periods go are mere details. This is about structuring the elements of references (name, title... moar;) into standard forms that amount to a data structure; we're not paper, we're a website driven by a database, this is fundamental to our future. It's all about being machine readable and enabling ever more tools to help editors working on the site. See Talk:Halle Berry#citation style and Talk:Uma Thurman#References for current article-specific discussions. Cheers, Jack Merridew 22:27, 19 August 2010 (UTC)

A "de facto" standard they are not. That's your opinion.

Peter ^Isotalo 06:45, 20 August 2010 (UTC)

Let me give an example. It is not unusual for a site to suggest an format if you are citing their articles. The Oxford Dictionary of National Biography as an example has a today article which currently is Davidson, William (1786–1820), conspirator at the bottom of their article they give a citation method

Marika Sherwood, ‘Davidson, William (1786–1820)’, Oxford Dictionary of National Biography, Oxford University Press, 2004 accessed 2 Oct 2010

and/or

William Davidson (1786–1820): doi:10.1093/ref:odnb/57029

Now it is not the same as we use in our templates but it contains all the information. If it is added by a new editor who is not familiar with our templates it is more than adequate to meet the standards laid out in WP:PROVEIT, which is all we demand. Whether it is later wrapped up in a template is a matter for guidance and conensus, but I don't think it should be mandated. -- PBS (talk) 01:54, 2 October 2010 (UTC)