Wikipedia talk:Manual of Style (footnotes)/Archive 9

Tags placed after punctuation: continuing a previous dispute
As the author of one of those main page articles, I'd say that standardisation of footnote style would be an excellent aim, and I would advocate in-line cites using reference templates and cite php, with the tags placed after punctuation, as our house style. Tim Vickers 05:14, 8 November 2007 (UTC)
 * Why after the punctuation as a house style? --Philip Baird Shearer 12:35, 8 November 2007 (UTC)
 * How about because "before the punctuation" is incredibly unprofessional and illogical and goes against every known published style guide? Using Nature as an example is ludicrous; it and other journals are notorious for making writers jump through hoops to get published and their one-off standards are never used in the professional writing community as example of what to do.  One person at that journal made that decision - that should be no basis for our styles here.  An entire community of writers and editors make the decisions that go into the Chicago Manual.  That's why it's a reference work.  Nature is not a style reference work.  Other commonly references style guides that specify the footnote after punctuation are APA and  MLA.  Every style guide I can think of tells people to put them after.  Preferring "after" is not just a subjective standard.  Commas and periods are visual cues telling the reader when the phrase or sentence ends.  The reason the footnote looks awkward before the punctuation is because it becomes the visual cue.. but it's not the end. --Bloodzombie 14:30, 9 November 2007 (UTC)
 * You actually dare to call Nature unprofessional???????? How many publications do you have in impact factor >25 journals to make that claim? Arnoutf (talk) 23:47, 17 November 2007 (UTC)


 * Concur with TimVickers, Gimmetrow and Bloodzombie for all of the reasons explained by them and others in the past. There is no reason Wikipedia should follow the example set by one publication; Wiki is large enough and important enough to have its own house style.  Those seeking to change this guideline should only do so based on consensus for that change from the broader community, which to my knowledge has not been sought and is not in evidence.  Sandy Georgia  (Talk) 15:47, 9 November 2007 (UTC)
 * Thank you, Sandy; this is another straw man. Nature is by no means alone; it was the example some editor had to hand when the subject came up. Before punctuation is quite common in scientific discourse; especially with bracketed inline references, as opposed to superscripts. Septentrionalis PMAnderson 16:05, 9 November 2007 (UTC)
 * Two more: I have just come across the ICRC Commentaries on the Geneva conventions that use a similar style as that used by the European Union style guide  (as mentioned before in earlier sections of this discussion). --Philip Baird Shearer 01:06, 12 November 2007 (UTC)
 * The ICRC link does not have footnotes, but a form of parenthetic citation—and parenthetic citations come before punctuation in most style guides. On the other hand, the EU page suggests a style with "figure in superscript between parentheses with same value as the text, preceded by a fine space and followed by any punctuation" (3); I don't recall ever seeing that style on Wiki. Gimmetrow 17:03, 14 November 2007 (UTC)
 * ICRC see the sentence "This text therefore laid down the principle that the Convention should be applied from the moment of capture; this was, however, a source of considerable difficulties for the Detaining Power (1)." and then look at the bottom of the article for the footnote (1). I am not suggesting that the EU format is used exactly as shown on Wikipedia pages, because to use it would be silly given the technical workings of the current reference tag method. However both these examples put the reference tag before the punctuation and neither of them are for scientific discourses. I have also shown previously that the International Review of the Red Cross does not specify before or after punctuation for footnoting in their style guide and that both styles are used in articles published in that journal. --Philip Baird Shearer 12:51, 15 November 2007 (UTC)
 * Parenthetic citations (generally) go before punctuation. The ICRC link you provided, and the example from it you quote above, use a parenthetic citation system. It's not footnotes. Gimmetrow 23:34, 15 November 2007 (UTC)
 * The EU guide line calls them "References to footnotes". If they are not references to footnotes why are they numbered and directed to text placed at the bottom of the page? --Philip Baird Shearer (talk) 10:08, 17 November 2007 (UTC)
 * The EU thing is strange. But what is the point of your question? I commented on the ICRC link/example. Gimmetrow 21:37, 17 November 2007 (UTC)
 * You wrote of the ICRC usage "It's not footnotes.", yet the EU guidlline describes it as "References to footnotes". The ICRC commentary is using a similar system to that used by the EU that EU describes as "References to footnotes". What would you call the ICRC method if not reference tags to a footnotes, (because it is not a simple parenthetic citation -- like the Harvard syle -- where the information is contained inside the brackets)?--Philip Baird Shearer (talk) 23:45, 17 November 2007 (UTC)
 * The ICRC looks to me like a number system, and the number system is a variation of parenthetic citation according to MLA. I'm not going to suggest a classification for the EU thing. Gimmetrow 23:55, 17 November 2007 (UTC)


 * (outdent) Straw man? How deftly you ignore my actual point.  I acknowledge that you use Nature as an example; I didn't claim it is the sole example.  My point is that Nature is not a style guide and therefore we shouldn't be taking cues from it.  You actually just added additional merit to my point - Nature is a scientific discourse community whereas Wikipedia as a whole is not.  All the more reason its peculiar styles are not appropriate for Wikipedia.  Passive voice and excessive use of graphs and charts are also quite common in scientific discourse.  Is that where you're going next? --Bloodzombie 16:17, 9 November 2007 (UTC)
 * I didn't ignore your point; I replied to another post altogether. As it happens, I agree that the Chicago Manual of Style is a worthwhile work of reference; it would, for example, be nice if WP:MOS acknowledged its position on punctuation and quotation marks. I discourage no editor from following it.


 * But, by the same token, I would require no editor to follow it when there are other methods which are good practice, and may be more comfortable to the editor or suitable for the field concerned. As for the scientific passive, it has its uses, but I doubt they will come up that often: we should not be tempted to write articles in the first person.  Septentrionalis PMAnderson 03:43, 10 November 2007 (UTC)
 * Nobody disagrees Chicago Manual of Style is helpfull. However, it is not the only Wiki citing standard as Harvard and direct to internet citation are also explicitly allowed. While I agree we should limit the no of styles, I do not see why the punctuation is an issue, while Harvard is allowed. Before punctuation is THE standard in most European journals (including indeed Nature), of course you can call all European publishing unprofessional from a personal impression, but in that case I expect you (ie the person making that claim, that Nature style is unprofessional) to be (at least) a multiple Nobel laureate. If your are not, don't talk about professional use of referencing. Arnoutf (talk) 23:54, 17 November 2007 (UTC)

Rejected
The proposal to remove the house style has been rejected by the community. For those who like to wikilawyer when they don't get their way, a rejected proposal is defined as any "for which consensus for acceptance is not present after a reasonable time period, for which consensus is unclear after a reasonable time period for discussion regardless of whether there is active discussion or not" ... "Making small changes will not change this fact, nor will repetitive arguments."

"But there was never a previous consensus!"

Yes there was. The house style was established as early as October 2005, and has been in consistent, widespread use since then. The fact that it was followed and not disputed demonstrates consensus. "Silence implies consent if there is adequate exposure to the community."

"But consensus can change!"

Yes! But please read the actual policy on the matter. "Wikipedia's decisions are not based on the number of people who showed up and voted a particular way on a particular day. It is based on a system of good reasons." The proposal to remove the house style is not based on good reasons, but on personal preference. "Attempts to change consensus must be based on a clear engagement with the reasons behind the current consensus". No one has addressed any problems with the previous consensus except their own personal dislike of it.


 * False:
 * There never was a previous consensus.
 * Several reasons have been expressed against both having a "house style", and against this one. As it happens, I myself put footnotes after punctuation, so only half of these sway me; but they are clear. Septentrionalis PMAnderson 22:23, 18 November 2007 (UTC)

Editors aren't required to follow the house style, and most won't even be aware of it. It simply serves as a guide for those who are aware of it and want to fix up an article they are working on. Removing the house style would make editing Wikipedia needlessly complex, especially if editors are to keep articles internally consistent (which everyone agrees is desirable). Instead of learning the house style once and using it everywhere, they would be required to examine the punctuation usage of every reference in every article they edit, or slog through the article's history searching for the first usage, to make sure they are conforming to the article's preferred style. It's ridiculous. The "Varieties of English" rule was added for a good reason, and even the "defer to the first contributor" part of the rule is a last resort. It's not a catch-all principle to be used any time three editors disagree with a guideline.
 * This paragraph is self-contradictory. If most editors won't even be aware of the existence of a house style, then its absence makes no difference to their editing. For those editors who care about this misbegotten issue, making sure that a given article is consistent on placement will be easier than moving all the footnotes to the other side of punctuation. Septentrionalis PMAnderson 22:22, 18 November 2007 (UTC)

Two or three editors disliking a policy and disruptively revert warring to their preferred version does not invalidate it. If these editors refuse to accept this, please begin dispute resolution.

(As an aside, this is an incredibly minor minor trivial tiny detail. I can't believe how fanatical you are about it.  Who cares?  You've been dragging this on for months and months.  Are you not embarrassed by your own tenacity?  Is this sort of continuing behavior a side effect of our rigid adherence to civility?) — Omegatron 00:12, 18 November 2007 (UTC)


 * The Wrong Version indeed. Omegatron, the text you used has lost quite a bit from the last version before this all began. To describe the house style in the original terms, while taking advantage of some simplifications, I would recommend:

Place a ref tag at the end of the term, phrase, sentence, or paragraph to which the note refers.

When a ref tag coincides with punctuation, it should be placed after punctuation without an intervening space, and likewise for successive ref tags. The exception is a dash —which should follow the ref tag. This is the format recommended by the Chicago Manual of Style and the MLA Handbook.
 * This includes a new ref at the end. Gimmetrow 02:34, 18 November 2007 (UTC)
 * MLA handbook is an irrelevant addition, we choose the Chicage style, fair enough, we do not need a (North American) book that is most likely based on Chicage Style in its advice first of all, because it adds nothing, secondly because it creates the image that this is indeed the most accepted style in science writing (look there are two seperate sources for it). For that reason I say leave MLA out. Arnoutf (talk) 11:56, 18 November 2007 (UTC)
 * Yes it is a small thing, being blown out of proportion by editors accusing editors who do not like the North American (this was established in the long texts above) of being bad editors. The choice where to place the ref is a choice, only that. No convincing arguments have been given why one choice is better than another. I can live with the current guideline if only all editors involved in this debate acknowledge that it is a choice, a subjective choice, a historic choice and nothing else (ie if they drop all reference to beauty, quality of writing etc etc). Arnoutf (talk) 11:45, 18 November 2007 (UTC)
 * The fact that it's the recommended style for footnotes by CMoS and MLA does not, in any way, say that footnotes need to or should be used—MLA seems to discourage them. Nor does it imply footnotes are more common in any particular discipline than some other citation system; many journals use parenthetics. Gimmetrow 18:54, 18 November 2007 (UTC)
 * Of course, one might also wonder at the fanatacism for rules-making displayed by those who believe Wikipedia should be uniform in "incredibly minor minor trivial tiny detail[s]" -- in my experience, this sort of guideline serves less as a guide for those wanting to fix up an article, and more as a tool for those who want to create more arcane ways in which perfectly acceptable articles need to be "fixed." It's absurd to talk about consensus being founded on good reasons, and then argue that there is consensus for the current version, which is founded on nothing but pedantry. Christopher Parham (talk) 19:17, 18 November 2007 (UTC)
 * This also omits the real flaw here; Consensus for the "house style" is not now present, and has never been present. It is therefore equally rejected. Septentrionalis PMAnderson 22:16, 18 November 2007 (UTC)

editprotected This should, of course, have the protected tag, since it has been protected to one side in an ongoing and unsettled dispute. Septentrionalis PMAnderson 22:26, 18 November 2007 (UTC)
 * Go ahead and add it then. And it's the "wrong version" for me, too. Gimmetrow 23:28, 18 November 2007 (UTC)
 * Even if I were an admin, I would probably let someone else do it, as a COI. As a contribution to the discussion, why is it Wrong for you? You have stated the fuller text you would prefer; but what differences there do you see as material? Septentrionalis PMAnderson 23:52, 18 November 2007 (UTC)
 * Sorry, I must have confused you with PBS, who is an admin. This text doesn't say anything about spacing, nor does it have anything related to the additional text PBS and I were working on to attempt to solve this. It's really gone on far too long, unless the ultimate intent is to make an ArbCom case. Gimmetrow 00:07, 19 November 2007 (UTC)

This page is not protected. &mdash; Carl (CBM · talk) 13:10, 19 November 2007 (UTC)

Reference tags and spaces
See previous discussions
 * Wikipedia_talk:Footnotes/archive4
 * Wikipedia talk:Footnotes/archive5
 * Wikipedia talk:Footnotes/archive5
 * Wikipedia talk:Footnotes/archive6
 * - When placed at the end of a clause or sentence the ref tag should be directly after the punctuation mark without an intervening space, in order to prevent the reference number wrapping to the next line. 6-4
 * and Gimmetrow's proposal in the previous section.

As a number of editors have expressed concerns over the issue of spaces between ref tags and punctuation, which is relevant whatever style is used, I have added the following to the text:
 * To avoid problems with auto-formatting, there should be no spaces between references tags and punctuation marks unless the spaces used are non-breaking spaces &amp;nbsp;.

--Philip Baird Shearer (talk) 14:10, 19 November 2007 (UTC)
 * There should be no spaces at all, nbsp or otherwise. When did that change? Lara  ❤ Love  14:26, 19 November 2007 (UTC)
 * It didn't, as far as I know. I removed it.  Sandy Georgia  (Talk) 14:54, 19 November 2007 (UTC)
 * I have no idea what's going on here. The optional use of non-breaking spaces has been discussed before, and pretty soundly rejected. I remember this because I once proposed text similar to that, only to be shot down. Gimmetrow 22:37, 19 November 2007 (UTC)
 * Then I misunderstood your edit in the section above; let's have reasons to proscribe them, if any. Septentrionalis PMAnderson 22:42, 19 November 2007 (UTC)
 * Glad we agree on that, at least. Gimmetrow 23:16, 19 November 2007 (UTC)

Change of first sentence
The first sentence currently reads:
 * Some words, phrases or facts must be referenced mid-sentence, while others are referenced at the end.

I changed it to
 * Place a ref tag at the end of the term, phrase, sentence, or paragraph to which the note refers.

It seems to me to say the same thing, and PBS seems to like it too. I don't think I've ever seen a wikipedia text put ref marks at the beginning of the term, phrase, sentence or paragraph to which notes refer. So why is this "disputable"? Gimmetrow 23:21, 19 November 2007 (UTC)

Sebastiani
As for what follows, it does seem perfectly uncontroversial. Named tags can be used to produce the annotation used in Horace François Bastien Sébastiani de La Porta, and this was sufficiently unpopular to deny the article FA.
 * Extensive use of named references can result in multiple footnotes at the same point, each citing a different page or on-line source, as here. Some editors dislike this, preferring that the different references at each point be combined into a single note; the article cited was denied the status of a featured article, largely on this ground.\

Septentrionalis PMAnderson 22:42, 19 November 2007 (UTC)
 * Two main issues. 1) that's a debatable interpretation of what happened on that FAC, and 2) why in the world is it necessary or useful to link a specific FAC page? [To the best of my knowledge I wasn't involved in that FAC.] Gimmetrow 22:47, 19 November 2007 (UTC)
 * Feel free to debate it (please observe that although I have a preference on the matter, I did not oppose); there were other issues, and this claims only a large part.
 * Why mention it? To establish that there is such a sentiment on the part of some editors; both to avoid this discussion and to warn the next FA aspirant who finds his article shot down. I don't suppose the FAC needs to be in here, but the article should be, to demonstrate what level of multiple noting is deprecated. Septentrionalis PMAnderson 22:58, 19 November 2007 (UTC)
 * We don't need to mention every sentiment, do we? There have been discussions before to discourage multiple tags together, and there wasn't strong enough support to put it in the guideline here. Gimmetrow 23:16, 19 November 2007 (UTC)
 * Let's consider Joe Editor, who has carefully followed this page, taken the results to FA, and had it rejected; Joe would have a right to feel that he should have been warned. Remember, this is not don't multiply link, but don't do it to excess. Septentrionalis PMAnderson 00:09, 20 November 2007 (UTC)
 * This isn't really an issue with named refs though, it's an issue with citing the same sentence to multiple sources "to excess". That's arguably WP:CITE's domain, though I wouldn't object to something noting that some editors don't like successive refs, and that successive refs can be combined in various ways, including bullets. I also see the section on named refs has been renamed, breaking a link in the "Style" section. Also, the "ibid" text has been duplicated under named refs too. I would like to remove the "ibid" paragraph from the named refs section, unless someone presents a good reason why it needs to be there. As it stands it disrupts the flow of the section and confuses the issue. Gimmetrow 00:33, 20 November 2007 (UTC)
 * I think that all those who contribute to this talk page are so familiar with the Wikipedia footnote/reference methods and we have difficulty lookinh at the issue from the point of view of an editor who, although familiar with footnotes, is not familiar with Wikipedia footnotes. For those who are familiar with footnotes in printed material but are new to footnotes in an electronic media an explanation of why ibid should not be used is probably of great value. I know I have come across several articles were ibid has been used and checking through the history to work out if the placements are still correct before fixing them is no fun. --Philip Baird Shearer (talk) 01:08, 20 November 2007 (UTC)
 * Does it need to be there twice? I only want to remove one discussion of ibid, so the discussion of named refs keeps on focus. There's another discussion of the same point at Footnotes. Gimmetrow 01:31, 20 November 2007 (UTC)
 * In the first section is probably better (and remove it from the second section). How about putting it or something similar as the first paragraph, that way it would not break up the rest of the text as I agree it does at the moment. --Philip Baird Shearer (talk) 01:59, 20 November 2007 (UTC)
 * Conceptually, it doesn't seem to me to fit in that section. Named refs are used for citing the same reference anywhere in the document. Ibid is used for citing the same reference in immediate succession, and where ibids occur in wikitext, they are not part of a named ref. That's why it's a style issue, not a named ref issue. Gimmetrow 02:11, 20 November 2007 (UTC)
 * I agree that it does not seem to conceptually fit into the Named refs section as it is currently written. Named refs, though, are conceptually a hyperlink-mechanism replacement for unadorned paper-media ibids. Also, the danger of using ibids in separate cite.php footnotes (high risk that subsequent inserted footnotes will separate ibids from the citation to they should refer, and difficulty of fixing this problem when it occurs and is not quickly recognized). I think that it is useful to mention both these points in the Named refs section, though perhaps the explanation of the details should be removed to the Style recommendations section. -- Boracay Bill (talk) 03:52, 20 November 2007 (UTC)

(unindent) What if the paragraph about ibid in the named refs section were deleted, but a sentence were added to the end saying "Named refs can be used much as loc. cit. or ibid would be used in a paper document (but see also Style)"? It would fit fairly well with the note about deleting a named ref with content. And that note should probably refer to named refs with content rather than simply named refs. Gimmetrow 04:51, 20 November 2007 (UTC)
 * But see cautions in Style would be better. Septentrionalis PMAnderson 06:31, 20 November 2007 (UTC)

The Sebastiani article is made possible by named references to this extent: there are 23 separate citations of Sebastiani's obit. each of which links to its own iteration of a footnote containing the same information. Without the named ref mechanic, this would be obviously cumbersome, and there would be one footnote of the form 'Gentleman's Magazine, "Seastiani"; Trumbull, Napoleon p. 34...' instead of two, three or five footnotes, each containing full-form citations. (I don't think the article is overcited; it's just overfootnoted.)

The warning should be here, because this is where editors come to find out about footnotes; it can be in WP:CITE as well.Septentrionalis PMAnderson 06:30, 20 November 2007 (UTC)

Proposed extension for page numbers
As a programmer and a technical writer I see the current &lt;ref&gt; implementation as poor design. It's so close to working really well, but that little bit it's missing would make all the difference.

Many articles will cite the same source (say, a book) multiple times, but with each citation pointing to a different page in the book. Conceptually this is still the same source — it would appear only once in a bibliography — and the page number is a separate piece of information.

However the ref tags as they stand don't make use of this important fact, and actively discourage people providing detailed references. Just think, if you're citing a single book five times throughout an article, would you write those five references like this:

Or would you write

I and the majority of other Wikipedians do the latter, until we are hit with a   warning, by which time the book has generally gone back to the library. The ref tag offers this handy mechanism for re-using references by setting the name attribute, which 9 out of ten times means that we leave out page numbers.

A slight alteration to the ; ; ; ;

which is still a pretty compact way of typing references. This approach has a host of benefits:
 * It is completely backward compatible with existing wiki markup.
 * It encourages people to enter page numbers, rather than actively discouraging them from doing so.
 * It allows the reflist to be intelligently formatted so as to only repeat the minimal information and thus be more readable. The people who write the extension can take this as far as they want, for instance abbreviating to "Ibid. p. 18." instead of re-listing the entire citation for consecutive listings, or providing only "Monster (1999) p. 18." if there are other intervening listings.
 * (This would be automatic; currently editors trying to cite different pages of the same document will try to make similar abbreviations, but they have to keep aware of the ordering of references, which is a pain.)


 * The markup is relatively simple and brief and won't overwhelm the article text.
 * It is conceptually more elegant, and will bring pleasure to whoever programs the extension, whoever writes the citation tags, and whoever reads the generated output.

I don't know who I need to convince with this, or where to make my case, so I'm doing it here. Fuzzypeg ☻ 23:10, 25 November 2007 (UTC)


 * An approach to this which works with existing tools would be to use:

; ; ;
 * and, in a Bibliography section, use:


 * This labels the citation in the bibliography section with an ID of CITEREFMonster1999, and links to that from each of harvnbs in the &lt;ref>s expanded by &lt;references/>.


 * Using this approach, there's no backlink from the bibliography section citation to the page-numbered references &mdash; the browser's Back button is used instead. With multiple backlinks labeled a b c d e f g, nobody ever knows which backlink to click anyhow. One thing which could be done to improve the user interface here would be to provide an editor-insertable clickable icon which would emulate the browser's Back button by attempting with javascript to navigate one page back in the browser's history. -- Boracay Bill (talk) 01:07, 26 November 2007 (UTC)

Something is wrong with footnotes!
I just edited Ivorian Civil War and didn't touch the footnotes, although I added another wikilink. Here's the revision in question: http://en.wikipedia.org/w/index.php?title=Ivorian_Civil_War&oldid=174660026

My edit had the very odd side effect of replacing many of the "ref" tags with strings that look like: UNIQ22615145405235f4-ref-00000006-QINU

Does anyone know what the heck is going on? ǝɹʎℲxoɯ ( contrib ) 19:21, 29 November 2007 (UTC)


 * Just ran across the same thing at Sirtaki. --Fang Aili talk 19:27, 29 November 2007 (UTC)
 * Reports of Wiki-wide strange issues, not only footnotes, due to a software update, perhaps follow WP:VPT to see if it's straightened out? Sandy Georgia  (Talk) 19:48, 29 November 2007 (UTC)
 * Follow up at the Village Pump Technical. Sandy Georgia  (Talk) 19:58, 29 November 2007 (UTC)

Seems to be working OK now. Gimmetrow 20:01, 29 November 2007 (UTC)

"list first, citation second" way of generating footnotes?
Hey, I'm new to Wikipedia, so this may be a dumb question. In putting together an article where single references are used multiple times, it occured to me that it would be easier (for me, at least) to first create a named list of references, THEN place them into the text via name (). Of course, all you have to do is type in one instance of the whole reference and name it, then use the name tag for subsequent uses, but (for me at least) it's a whole lot easier to reference a pre-created list as I work through the article vs. having to scroll back to recall what the heck I named it (or refer to a written cheat sheet or Wordpad window) & remember to paste the whole thing the first time I use it. So, in short: has anyone thought of a "list first, citation second" way of generating footnotes? It may seem like a backward way of approaching the creation of footnotes, but then not all of us eat with a fork in our right hand. . .--Pgagnon999 (talk) 01:49, 6 December 2007 (UTC) To be a little clearer, it would look something like this:

First, create a list of references with a name for each. It would be invisible, so it could exist anywhere in the article.

Second, plug the references into the text by name (or, if they are single references, optionally just type them in here once).

Third, put the tag references/ under the "Reference" section at the end of the article. This also would simplify copying reference names or blocks of reference names for multiple articles.--Pgagnon999 (talk) 02:02, 6 December 2007 (UTC)


 * No, many people have thought of this before, but wiki doesn't seem to support a good way to implement it in an article without the article having some other quirk or deficiency. At least, nobody has figured one yet. See also Archive 6 and  Archive 7 for earlier discussions. Gimmetrow 02:03, 6 December 2007 (UTC)

Referenced at the end
Wording before change
 * (1) "Some words, phrases or facts must be referenced mid-sentence, while others are referenced at the end."

Discussed wording change (see Archive 7)
 * (2) "Place a ref tag at the end of the term, phrase, sentence, or paragraph to which the note refers."

I've been busy on other things but I think this needs further discussion before the change is implemented. The problem I have with this wording is the word "paragraph". Using it implies that the method to be used is not the Nature (journal) like one. I think it is better to leave it as "at the end" and it does not re-open this argument. Wording (1) works for both methods. --Philip Baird Shearer (talk) 12:22, 6 December 2007 (UTC)


 * I don't really see how "or paragraph" in any way rules out a style which cites by sentence, and unless I've misunderstood something, this is a text *you* suggested a few months ago. What argument does this re-open? Gimmetrow 17:19, 6 December 2007 (UTC)


 * Nature style notation can not have a note at the end of a paragraph as the notes are within a sentence. --Philip Baird Shearer (talk) 19:05, 6 December 2007 (UTC)


 * If it's the last sentence of the paragraph, it's the end of the paragraph. What am I missing here? And why did you want this changed to begin with, if it's so wrong? Gimmetrow 23:52, 7 December 2007 (UTC)


 * I reserve the right to change my mind on something :-) and on reflection I see the wording awkward when we have one example style that places ref tags inside punctuation. One of the problems with CMS citations is that one can not tell if a reference tag at the end of the last sentence in a paragraph is a reference for the whole paragraph or just the last sentence. This ambiguity is removed with the Nature style. Personally I would like to combine the two, as it is convenient to use inside the punctuation for a sentence and out outside as a citation for the preceding text from the start of the paragraph or from the previous citation (which ever comes first). This would be particularly useful in Wikipedia articles given the collective editing that goes on, but I appreciate that this option is not on the table at the moment. So an end of paragraph is not something one can have if one is using reference tags inside punctuation. --Philip Baird Shearer (talk) 22:58, 13 December 2007 (UTC)


 * What evidence do you have that this "ambiguity is removed with the Nature style"? Gimmetrow 03:32, 14 December 2007 (UTC)


 * Why did you change this over at WP:CITE? We are by no means at a consensus. Gimmetrow 04:16, 14 December 2007 (UTC)


 * Why are we still talking about Nature style notation when there is no consensus to change Wiki's in-house style to that of one external journal? Sandy Georgia  (Talk) 12:06, 19 December 2007 (UTC)
 * No-one has proposed changing "Wiki's in-house style to that of one external journal" while I have been watching this page; I doubt anyone has ever done so. Please try discussing the proposals which have actually been made.  It is this sort of profound carelessness which makes this discussion so tiring and pointless. Septentrionalis PMAnderson 18:50, 20 December 2007 (UTC)


 * Well precisely. Listen folks, shoving superscript square-brackets numerals before punctuation is just not appropriate. Nature, am I right, doesn't superscript their numerals. Even so, I think it sucks, because (1) when more than one numeral is cited, it looks ever so clumsy; (2) it interferes with the natural integrity of the sentence. I don't know why this is even being discussed, let alone why people are somehow pushing for the clumsy version. Wikipedia is no longer a child, and needs to establish its own formatting to suit its particular mode and readership. Tony   (talk)  14:23, 19 December 2007 (UTC)
 * In shorter form, WP:IDONTLIKEIT. Tony can place footnotes as it pleases him; no-one has ever attempted to prohibit him from doing so. As it happens, I will also place them after punctuation, but I see no reason, and no consensus, for a "house style."
 * If Wikipedia is no longer a child, neither are its editors children; they should be free to follow their own judgment. This, as with WP:ENGVAR, is a point on which there is no consensus, and no likelihood of consensus. We have suggested a simple statement of the positions here, and been continually reverted.  Removing the disputed section, as unsupported. Septentrionalis PMAnderson 20:57, 19 December 2007 (UTC)
 * I have to disagree. Without a consistent style, as any professional publication has, Wikipedia reads like it was written by children. It's ironic that Wikipedians refer to each other as editors, because if there's anything Wikipedia is in desperate need of, it's an editor. --JHP (talk) 04:18, 27 February 2008 (UTC)
 * I fear I have to support the generality of JHP's views (if not the specific phraseology). It is not just the inconsistency involved in using a variety of styles, it is the difficulty experienced by newer users in understanding what is and what is not appropriate and that changes are repeatedly made to MOS without there being any system in place to alert long-suffering editors. WP:ENGVAR works because it is usually fairly obvious from the article what an editor might expect to have to do. Our citations and references 'guidelines' produce a (to me) bewildering variety of 'correct' procedures - that of course only work if a particular article is internally consistent. There is no requirement at all that articles on similar subjects be consistent.
 * Haswell-Smith, Hamish (2004) The Scottish Islands. Edinburgh. Canongate. Pages 314–26.
 * Barrow, G.W.S., “The Reign of William the Lion”, in Scotland and Its Neighbours In the Middle Ages, (London, 1992), pp. 67–89.
 * McDonald, Hugh (1910). Rambles Round Glasgow. John Smith & Son, 197.
 * (1 November, 1973) "Oral Questions to the Minister of Foreign Affairs". Dáil Éireann. Retrieved on 17 January 2007.

All the above are gleaned (with a minimum of effort) from featured articles or lists on Scottish subjects. The alternative to the currently confusing and rather amateur-looking approach is an editorial group of some kind to create consistency and a genuine house style. I am not holding my breath. Ben MacDuiTalk /  Walk  13:55, 16 March 2008 (UTC)

Error in web site footnotes
You need the http:// in order for web link footnotes to work correctly. This is missing here and in the main citation section. (so you need http://www.example.org not just www.example.org). Should this be changed? MDuchek (talk) 23:46, 7 December 2007 (UTC)
 * Please add them. Gimmetrow 23:52, 7 December 2007 (UTC)

Template based implementation examples
Comments on template based implementation examples to overcome difficulties with editing due to our non-see-what-you-get editing interface led me to add:


 * See Category:Specific source templates for some template based implementation examples.

to the page. WAS 4.250 (talk) 18:16, 13 December 2007 (UTC)

copy-edits
FYI: I did some copy-edits in a couple of places. I don't believe I changed the substance of anything. You can review the total of my edits with this diff. --Lquilter (talk) 01:45, 18 December 2007 (UTC)


 * "To give a footnote an unique identifier" -> "To give a footnote a unique identifier" Why? --Philip Baird Shearer (talk) 09:50, 18 December 2007 (UTC)
 * Probably related to this. (It is "A unique" by American standards, but with British it is "An Unique.") -- Boracay Bill (talk) 12:27, 18 December 2007 (UTC)
 * Nothing to do with BrE or AmE, as far as I can tell (there's only one answer in that link about AmE/BrE distinctions, and I don't think it correct). Per Fowler: "A is now usual also before vowels preceded in fact though not in appearance by the sound of y or w".  As the link Boracay Bill offers, "unique" is "yoo-neek", so begins with "y", so requires suggests "a". Carre (talk) 15:35, 19 December 2007 (UTC) ( edited by Carre (talk) 19:55, 19 December 2007 (UTC) )

From Talk:Morayshire Railway

 * also posted at Template talk:Reflist

This section contains footnotes. I beleive for consistency that should be followed in this case. I will be reverting the page back to using "reflist" as the guideline suggests. --Rockfang (talk) 18:42, 22 December 2007 (UTC)


 * This contradicts Template:reflist which states there is no concensus to use the small font version. There is no reason in this article to use the small font version. In addition FN indicates that small font has some disadvantages and conversely that some editors prefer the smaller font. I do not like the small font. In this article there are only ten referenced items, and I see not advantage for the smaller font. --Stewart (talk) 18:54, 22 December 2007 (UTC)
 * In this case, Template:reflist is incorrect. If you check the history of FN a lot of stuff has been removed because people edited it without consensus.  The "how to use" section is still on there because it has reached consensus. With regards to your small font point: sometimes small font does have disadvantages.  Some people might have poor eyesight even with glasses and the smaller font could be harder for them to read.  I would prefer to be consistent though.  Even though list isn't extremely long at only 10 items, that number was reached as consensus at the most applicable guideline to follow: FN --Rockfang (talk) 19:19, 22 December 2007 (UTC)


 * The difficulty I have with this is you have used a template which has its guidance notes and then informed me that the guidance notes are incorrect and superceded by a MoS elsewhere in Wikipedia. I will reluctantly bow to you edit, however it is important that the guidance notes with the   are corrected to reflect WP:FN otherwise others will follow the notes to the template and revert like I did. --Stewart (talk) 19:45, 22 December 2007 (UTC)
 * Rockfang has forgotten the general rule WP:CONSENSUS. As this page says about such issues: However, as with all citation styles, consensus should be achieved on the talk page before implementing such a change. Talk about reasonably, and reach agreement, or leave it as it was.   Septentrionalis PMAnderson 22:28, 22 December 2007 (UTC)
 * Actually I have not. I just re-read WP:CONSENSUS. In it, is the line:
 * If we find that a particular consensus happens often, we write it down as a guideline, to save people the time having to discuss the same principles over and over.


 * In this case, consensus was reached with FN. I'm not going to every single article's talk page to ask for consensus if the article doesn't follow FN.  In fact, most articles I've viewed so far already meet this guideline. I do have a suggestion though.  If you don't like the guideline, feel free to debate it on it's talk page.  Maybe consensus there will change.  Until it does, I will continue to apply the guideline to articles I come across. --Rockfang (talk) 22:47, 22 December 2007 (UTC)
 * Please see Bold, Revert, Discuss. If you are the only user who cares, fine: you are consensus. But if you are reverted, as you have been, stop and discuss. (And it really is no more trouble, in the long run, to begin by posting "Does anybody care if we switch?" on the talkpage, and wait 24 hours. You can even make a template with your argument, although it should go in your userspace.)  The size of footnotes is not the Most Important Thing; this page does not excuse you from the burden of common civility, especially since it says to discuss these matters.Septentrionalis PMAnderson 23:02, 22 December 2007 (UTC)

I think, as a result of this conversation, that the recommendation should be made fuzzier, precisely to avoid this sort of rule-mongering. The article that brought this up has, at the moment, exactly 10 footnotes; it may have 9 (or 1, or 11) the next time it's rewritten. They aren't citations, but note the 2005 value of 19th-century currency, so it may be more important than usual to have them legible. How about "generally appropriate with less than a dozen" and "more than half a dozen" to allow some slack, and a comment that articles will differ on which looks best?

I've used reflist with a single note, myself, to mark that it differed from the rest of the stub, so we may simply want to say it depends. Septentrionalis PMAnderson 23:13, 22 December 2007 (UTC)

I have been using reflist as a rule, even with a single reference. The argument of issues with text size would have been a great reason to forbid use of the reflist template, but it seems farfetched at best to say that those with weak vision will be better able to read the smaller text because there are ten or more references. Both for consistency and for distinguishing references from the rest of the text, there seems to be little reason to have a hard and fast restriction on use of the template when there are fewer than 10 references. There are few better examples of arbitrary rule creep than the 10-reference cutoff. Furthermore, it astounds me that anyone would spend their time on this encyclopedia going from page to page, counting the references and removing the template where the count is under ten. Isn't there anything else that needs work here? Alansohn (talk) 03:56, 23 December 2007 (UTC)
 * Just to update this page as well: I've decided to give up trying to show my side on this matter. From now on, if I change articles to fit WP:FN and someone reverts it, I'm just gonna leave it. But I do appreciate how you beleive there is a certain way I should spend my time editing Wikipedia. --Rockfang (talk) 06:28, 23 December 2007 (UTC)

"Furthermore, it astounds me that anyone would spend their time on this encyclopedia going from page to page, counting the references and removing the template where the count is under ten."
 * Likewise, why would anyone change a perfectly functional

There's an example of what it looks like at right. You can see that it collects and presents backlinks to particular pages in the same way that it does backlinks to named references currently. It also sorts references to pages, using a natural sorting algorithm so that everything sorts nicely (as can be seen in the picture).

You can see the listing in the bugtracker here. It would be great if people could review the patch - and fix my code, should it be crappy in spots :) --bainer (talk) 13:03, 23 February 2008 (UTC)
 * Rather than "page=", please call it "note=" or similar to allow flexibility. This could be used to add annotations or comments to specific named refs. If you really wanted to get fancy, you could automatically add "p." for page only when the note is a number, and sort the numbers separate from the annotation comments. Gimmetrow 19:32, 23 February 2008 (UTC)


 * Strong support for use of "note=" over "page=" if just a single extra option is to be given, particularly for say using cite-web template where one can only indicate a particular sub-section given that there is no such thing as a page number. Alternatively allow both parameters, with "page=" sorting into order as described in teh proposal and "notes=" being an unsorted sequential list (somewhat akin to where we choose bullet-pointed lists for References and numbered-pointed items in Footnotes). David Ruben Talk 00:07, 29 February 2008 (UTC)


 * Quite a few other features for cite.php have been discussed over the years. If this seems likely to be implemented, now would be a good time to either add the other features, or make sure the code is structured to add them. Most desired features could be options in the references tag. Do you feel like trying this? Gimmetrow 06:15, 24 February 2008 (UTC)


 * Which suggested features were you thinking of? Now that I'm familiar with the code I could certainly look at implementing some more of them. I chose this to implement because it's useful to me, and because it fits in well with existing referencing practice here at Wikipedia, so it has a reasonable chance of getting turned on on the Wikimedia sites, and looking at the archives here I don't know if all the suggestions for new features would meet those criteria. --bainer (talk) 23:37, 28 February 2008 (UTC)


 * OK, you asked for it. A lot of people would like the ability to define the references without producing a ref mark. Then the references could be defined in a references section and used by name only in the text. You could have in the text, but Smith, Some Title, Alpha Press, 2001 somewhere in a section obviously set aside for references.
 * I used "refdef" in the example. You could implement this by having any "ref"s appearing after references/ behave as an invisible definition. There is a fairly annoying potential problem with this, although the current code doesn't handle refs after references/ "correctly" anyway. A new refdef tag (or similar) is probably the safest implementation.
 * With some additional modification, this would allow the references section to be ordered by the order of the refdefs, rather than the order of appearance of notes in the text. But that's probably too big a change to get turned on at once.
 * Finally, some would like to have two sets of ref tags, one which produces numbered ref marks, and one which produces letters, so the annotations and citations could be split without using the ref/note templates. The alternate code might be called, so you could have This is a text annotation in the text, and have produce a list of them in order with letters. But perhaps this could be handled with the note= field in unnamed refs? Gimmetrow 02:51, 29 February 2008 (UTC)
 * Bugzilla bug 12706 proposes functionality similar to this, and provides code to implement it. -- Boracay Bill (talk) 07:00, 29 February 2008 (UTC)
 * Yes, the decl= in place of name= is a good way to handle invisible definitions. Can this be combined easily with a note= capability, and the ability to do lettered notes in parallel? Gimmetrow 07:13, 29 February 2008 (UTC)
 * I am presently traveling, writing this from an internet cafe, and I'm not inclined to download and examine the code to check this. As I recall, however, Bugzilla bug 12706 as implemented in the code supplied with that bug, would allow footnotes without references (what you're calling ?) and footnotes with references to be grouped separately in any editor-chosen order desired in the expanded References list and would also allow arbitrary headers to be inserted introducing each group. The wikitext syntax is not as you have proposed, and the expanded References list is similar in format to the current References list except to allow editor-chosen grouping and ordering and to allow the insertion of un-numbered headers into the list. All sorts of alternative enhancements are possible, but finding support for enhancements is difficult. Also, note that the code which does the citations (cite.php) is maintained by Wikimedia, and is used elsewhere than in Wikipedia. —Preceding unsigned comment added by Wtmitchell (talk • contribs) 07:54, 29 February 2008 (UTC)
 * OK, I understand. I'll look at it in more detail. Gimmetrow 08:28, 29 February 2008 (UTC)


 * I came here looking for just such a feature, which will make our reference system much more useful. Is it ready to go? - Nunh-huh 19:10, 28 February 2008 (UTC)


 * The patch is still waiting to be reviewed. Although I've tested it myself, people are often blind to errors in their own code, so it needs further testing by other people before it will get implemented. --bainer (talk) 23:37, 28 February 2008 (UTC)


 * I also support changing the parameter from "page=" to "note=", since the first implies to me that the correct usage would be page=24, while the second is more faithful to what you've done: note="Page 24".


 * As far as testing, is it possible to move this to test.wikipedia.org, so people can actually play with it?


 * Finally, I'm personally not at all a proponent of "since you're doing X, why not work on Y and Z too?". I'd rather see smaller changes constantly rolled out than have a big, gnarly one that gets entangled in multiple arguments. -- John Broughton (♫♫) 14:36, 16 March 2008 (UTC)


 * Upon reflection, I think "detail=" would be a better attribute name. We already have templates for notes, and footnotes are notes of sort, and if we ever improve the current cite.php system, I suspect it might be with tags, so having a "note" attribute could well be confusing.-- John Broughton (♫♫) 16:36, 16 March 2008 (UTC)
 * I agree that "detail=" is probably better than "note=", and "note=" better than "page=". In some cases, one might want to include a quote or something rather than just a page number; although I suppose even if it's called "page=", one could still meaningfully put a quote there.  In some cases it would be a set of pages, though, not just one page.
 * The refdef idea would probably be a good thing. Under the current system it can be annoying to try to find which footnote contains the actual text of the reference; or to have the reference deleted when one of the footnotes is deleted. I suppose references that have no footnotes referring to them could still display in the list, so that it would be obvious that they need to be deleted ... or possibly so that they could be used as general references/bibliography for the whole article.  If they don't display, a potential problem might be the accumulation of references that are never actually used.
 * Thank you, bainer, for doing this! Let us know when it's on the test wiki. --Coppertwig (talk) 18:07, 16 March 2008 (UTC)

Straight quotes only supported
Have added a note that only straight quotes are supported for ref naming. This can be a problem where editors prepare text in an external word processor, such as Word, which is set to automatically convert straight quotes into curly (typographers') quotes. Ref names bracketed by curly quotes are ignored, and the ref behaves as though it were unnamed; any later citations of the same ref then cause a citation error. Jayen 466 22:41, 7 March 2008 (UTC)


 * If the external editor can't handle that for that case, it might as well mix up other things in advanced ("ugly") parts of the wiki markup, so there would be hundreds of issues like this. Editors should be purposefully built for wikimedia or text-only. I'm a text-only person myself and I won't tell you what I think of Word. -- Sverdrup (talk) 01:38, 19 March 2008 (UTC)
 * Word has it's uses when preparing long new texts, especially spell-check. Sadly it sometimes takes action that are not wanted. It is the responsibility of the editor using external text-editors to make sure the text inserted in Wikipedia confirms to Wiki formatting. Arnoutf (talk) 07:05, 19 March 2008 (UTC)
 * Yes, of course, but even Firefox has spell-checking for normal browser editing. I would urge everyone to turn off anything in Word that begins with auto- or smart-. -- Sverdrup (talk) 15:25, 20 March 2008 (UTC)

Re: Resizing references
Under WP:FN it states: Although smaller text has some disadvantages, it is common when there is a long list of references (as a rule of thumb, at least ten)....My question is that who decided this "rule of thumb"? Was this based on consensus, or was it just a single editor? And how many agree with this anyway? I feel that having smaller footnotes makes them more distinguishable from external links (in the case of URL references), and makes the footnotes section look more professional (since, from what I've seen, normally footnotes listed in books are smaller than the rest of the text).--  十  八  19:10, 19 March 2008 (UTC)


 * I agree with this rule of thumb but would say that short lists can use normal-sized footnotes but don't have to. It depends what footnotes are used for. If it is for references, see e.g. Barack Obama where even small references already bloat the article by one third. If however footnotes are intended to be read like ordinary text, I can see why normal text size is advisable (e.g. List_of_Carniv%C3%A0le_episodes). I generally like small footnotes for the same reasons as you, but a short list doesn't call me into action. It's a common-sense decision. – sgeureka t•c 19:34, 19 March 2008 (UTC)
 * Side note, but if the references section is that huge, can't we put in a feature to collapse the refs to save space?--  十  八  19:42, 19 March 2008 (UTC)
 * I seem to remember it being discussed in multiple places and the consensus has always been no because we don't want to hide references. Part of showing verifiability is having the references right there, easy to find and see. Or something like that. AnmaFinotera (talk) 19:47, 19 March 2008 (UTC)
 * Yeah, I kind of expected that after I typed out the question. Just wondering.--  十  八  19:51, 19 March 2008 (UTC)
 * I personally like the "rule of thumb." While yes, printed books do have footnotes smaller but that is primarily to save space. Along the same lines, we don't really need to compact references unless there are more than 10, which is when it starts looking unwieldy. If reflist's smaller format is considered better from the get go, then why have the references tag at all? Just dump and replace with reflist. But that has not been done and I suspect it won't be done because the point of reflist is not to make distinguishable from the ELs (which is only a problem if the references are badly formatted in the first place), but reflist is used primarily to compact a large references section to save visual space. My view, anyway, and one I think works well for everything I've edited. Regular size refs look fine until around 10, then it starts getting big so you go to reflist. AnmaFinotera (talk) 19:35, 19 March 2008 (UTC)

Footnotes in sections headings
I somehow imagined that there's a consensus that footnotes shouldn't be used in section headings, but i can't find it anywhere in pages about headings, footnotes or MOS. Is it documented anywhere? --Amir E. Aharoni (talk) 15:17, 20 March 2008 (UTC)
 * There is WP:LEAD -- Boracay Bill (talk) 00:34, 22 March 2008 (UTC)

Footnotes interfering with wikitext readability -- a major concern
First, footnotes is one of the best things ever to happen to wikipedia. Without it there wouldn't be really professional and serious articles like Italian War of 1542–1546. It has also taken Wikipedia articles relating to Web-accessible phenomena to the next level by allowing instant access to deeper sources related to any part of the article.

But unlike infoboxes and tables, which are confined to special places in the article, references and footnotes are coded inline in the wikitext. I think this is a major concern for the Wiki-ness of it all. Suddenly the wikitext... is not simple markup like this. Easy to read and understand. The wikitext has clearly passed into markup where you easily get lost if you don't know the basics of xml-like markup languages, and this is kind of sad.

References are written in place since it is a really sensible idea. References should move with the sourced statement and the note should be updated if the body text is, and vice versa.

It's a hard problem to solve. If a reference is just a simple it would be a pretty acceptable burden on the wikitext. But then we lose the in-text presence that is so important per above.

Has this been discussed before. I try to mediate this by using more whitespace between sourced sentences; the wiki allows one line break without making a section break. However, that is often not enough. The section Wikipedia:Footnotes#Naming.. strikes me with an idea: in-browser syntax highlighting to shade less content-centric parts like footnotes would help this issue (take the example in the linked section and invert it). -- Sverdrup (talk) 14:51, 21 March 2008 (UTC)


 * 12796, if implemented, would address this. As submitted, it includes modified cite.php code to implement it. It is awaiting review. It currently has one supporting vote.-- Boracay Bill (talk) 22:32, 21 March 2008 (UTC)

Multiple footnotes
This is just as aesthetic choice, but, as for me, I don't care for long strings of footnotes like this. It looks ugly and breaks up the text. (I use Apple's Safari web browser. Perhaps footnotes are less obtrusive in other browsers.) So I usually consolidate the references for a single logical point into a single footnote that lists all the sources. This may mean that there is some overlap between footnotes and there are occasionally several footnotes that refer to the same page of the same source. I don't see this is as a problem, since, even with the redundancy it still performs the essential functions of citations, i.e. it verifies the text and provides access to further reading. Any one else have an opinion? (I'm writing this in light of a recent edit to the article Chinese room, and this question is also posted on the talk page there.)  CharlesGillingham (talk) 17:30, 24 March 2008 (UTC)
 * I agree with you. I don't think anyone would have a big problem with doing that. Christopher Parham (talk) 17:52, 24 March 2008 (UTC)

(ec)The string of footnotes is more informative, as it's possible to look in the refs section and see very easily how extensively a particular source has been used.  Ty  17:53, 24 March 2008 (UTC)
 * True, but I'm not sure this bit of information serves any purpose. (Except perhaps to help editors to notice that an article relies to heavily on a single page of a single source, I suppose?) Am I missing something? CharlesGillingham (talk) 06:33, 25 March 2008 (UTC)