Wikipedia talk:Citing sources/Archive 35

Archive 30

←

Archive 33

Archive 34

Archive 35

Archive 36

Archive 37

→

Archive 40

Foreign-language sources

Consistency would benefit from a note stating that non-Latin-alphabet sources should have their author and title duplicated in transliteration, following the Library of Congress system (which is used worldwide, in English-language libraries, catalogues, bibliographies, and academic publications). This is essential for most English readers to be able to use such references at all. —Michael Z. 2012-11-30 01:28 z

Sensible in principle, but recommending the LoC system specifically is perhaps not the best idea - it's particularly idiosyncratic with some systems, eg Cyrillic. Andrew Gray (talk) 13:12, 13 December 2012 (UTC)

"Indirect speech"

The section In-text attribution begins

In-text attribution should be used with direct speech (a source's words between quotation marks); indirect speech (a source's words without quotation marks); and close paraphrasing.

Using a source's words without quotation marks is not indirect speech, but plagiarism! Indirect speech is a specific form of paraphrasing. This needs to be fixed, pronto! --Thnidu (talk) 16:37, 4 December 2012 (UTC)

Not all indirect quotations are plagiarism (or copyvios, for that matter). There just aren't that many ways to say "The five-year survival rate for Scary Disease is 24%". WhatamIdoing (talk) 23:24, 7 December 2012 (UTC)

It's just another way of reporting someone's words. As long as it's relevant to the article, not excessive (so covered by fair use and not plagiarism) and properly sourced then it can be used much as a quotation. Whether a quotation is better is often a matter of style or editorial choice; text can flow more naturally if indirect speech is used. It's very common in news reporting for example.--JohnBlackburne^words_deeds 23:35, 7 December 2012 (UTC)

So, in-text attribution's pretty much being recommended for all occasions now? Quotes or no, I say: <blockquote>that's the opposite of plagiarism.</blockquote> —Machine Elf ¹⁷³⁵ 14:37, 13 December 2012 (UTC)}

Looks like I was misunderstood. Maybe I wasn't clear enough. The article says (and this is an exact quote)

... indirect speech (a source's words without quotation marks) ...

Using a source's very words is not indirect speech. The first paragraph of the article Indirect speech defines the term (underlining added):

Indirect speech, also called reported speech or indirect discourse, is a means of expressing the content of statements, questions or other utterances, without quoting them explicitly as is done in direct speech. For example, He said "I'm coming" is direct speech, whereas He said he was coming is indirect speech.

The source's words are modified to accord with the context of quotation: I'm coming → [He said] he was coming. Of course you should attribute them, whether you use direct quotation or indirect quotation or further paraphrase (e.g., He said he would be right out). But authorship applies not just to the meaning but to the wording as well, and if you use a source's exact words without acknowledging the source of the wording, you are implicitly claiming that the wording is your own, and that's a type of plagiarism. To avoid that, you acknowledge authorship of the wording by using quotation marks,* along with the reference.

Now, WhatamIdoing is right in pointing out that

There just aren't that many ways to say "The five-year survival rate for Scary Disease is 24%".

If you give proper credit for the source, the exact wording of a short text, especially in a routine or formulaic construction like this example, need not be cited in quotation marks. (This is my own opinion. I'm not a Wikipedia authority, just an academic with a career behind me.) But for longer exact quotations, or even short ones with unusual or striking wording ("A rose-red city half as old as Time"), you'd better use quotation marks along with the attribution, or it's... you know.

I've changed the misleading description to read

indirect speech (a source's words, modified [see the article], without quotation marks);

* Or a block quotation like where I quote WhatamIdoing just above. --Thnidu (talk) 06:19, 15 December 2012 (UTC)

The current format of Intext attribution was pushed by SlimVirgin into several guidelines a few years ago (see here for a list of some previous discussions). Because there have been problems with experienced editors stepping beyond the bounds of what is acceptable under plagiarism, I still think that the guidances over intext attribution should be more restrictive. However that is not the only reason for a tightening up of the wording: If someone writes Sir Robert Armstrong said he told the truth it is a summary of Sir Robert Armstrong said he was "economical with the truth", but if there is no rule in place how does one tell if the former is a summary of his words or his actual words? -- PBS (talk) 18:33, 15 December 2012 (UTC)

Number of guns per capita by country - False numbers!

Hello there, Information displayed on the following address is not true. Numbers for some countries have been altered, raised. These sort of mistakes are not supposed to be published and posted for people to see. We are not talking about number of flowers her, it's guns! It's disturbing for the people involved and dangerous if taken seriously from the ones that do not tend good. Please, help solve this. Regards, A

Shortcut: http://en.wikipedia.org/wiki/Number_of_guns_per_capita_by_country — Preceding unsigned comment added by 80.202.132.64 (talk) 12:44, 17 December 2012 (UTC)

Language=English

I've a bad feeling I know what the response to this question will be, but here goes. If an article widely specifies parameter language=English in its cite templates, even though no other language is involved anywhere in the article, and all the refs are in English anyway, would removing that superfluous parameter to bring the usage into line with the recommendations for using the template (Template:Cite_news#Title "language: The language the source is written in, if not English") constitute a style change requiring prior consent, or is it just a minor bit of housekeeping subject to WP:BRD like any other edit? Colonies Chris (talk) 09:51, 5 December 2012 (UTC)

IMO, in the absence of evidence to the contrary, you may assume that it was unintentional and that changing it is therefore a minor bit of housekeeping, unless and until someone complains. If someone complains, then naturally you would let them revert to include that parameter, maintaining at strict WP:0RR approach yourself. If you want to reduce the odds of anyone yelling at you, though, it really doesn't take very long to put a note on the talk page. WhatamIdoing (talk) 23:26, 7 December 2012 (UTC)

I concur that this is housekeeping. --— Gadget850 (Ed) ^talk 13:24, 13 December 2012 (UTC)

"If an article widely specifies parameter language=English in its cite templates" if an article consistently indicates that sources are in English, then changing that for an article would be a style change and require discussion. Fifelfoo (talk) 20:28, 23 December 2012 (UTC)

Nonsense! It is a clearly established rule that there is no need to specify language in en-wp when the language in question is English. Simply and obviously superfluous and silly. -- Alarics (talk) 20:49, 23 December 2012 (UTC)

How do I cite an entire television series?

I'm not a fan of the show Dance Moms, but my wife is, so I've had to sit through virtually every episode. At the article, there have been attempts to remove content, claiming BLP issues and that we must avoid primary sources but the content is definitely supported by the episodes, with many common themes running through the series. It's virtually impossible to cite each episode; I could do so using {{cite episode}} but the article would become over-referenced very quickly. In TV articles it's not normally the practice to cite everything if it's supported by the episodes, but I feel I'm going to have to, for the benefit of a couple of editors, but I really can't think of a good way to. Any suggestions? --AussieLegend (✉) 08:15, 23 December 2012 (UTC)

This sounds like horrific original research, primarily in the form of the exegesis of the articles subject: the re-explaining of what "Dance Moms" itself is from the text "Dance Moms". This is bad practice. If it wasn't mentioned in a secondary source, then why is the concept WP:WEIGHTy enough to include in an article? Finally, citations should indicate the source in the work where the object can be found—if you can't cite it to an episode and a timestamp or time range, you shouldn't be making a claim. Fifelfoo (talk) 20:27, 23 December 2012 (UTC)

The content can be cited to specific times in episodes, in several actually. It means cherry picking references because, as I said, the same themes run throughout the series. It's one of those horrible reality series where purely original content could be compressed into one or two episodes, the rest is just the dance moms bitching about Abby or one of the other moms, the dance moms fighting with Abby, Abby bitching about the Candy Apples, the Candy Apples bitching about Abby, .... and on it goes. --AussieLegend (✉) 02:57, 24 December 2012 (UTC)

How to cite webpage subdivisions?

A problem I and others have grappled with in various ways, but without any great satisfaction: How should we cite subdivisions within a webpage or blog?

Within printed works (or digital works that follow printed forms, including some web pages) we have chapters, but my sensibilities rebel at calling (say) a 60 word comment on a blog a "chapter". "Verse" would seem appropriate (in the sense of the biblical "book, chapter, and verse"), but we don't have a |verse= parameter, and anyway that usage would be confusing. "Section" would be good, but there's no |section=. |at= seems useful here, but at what level?

I think it would help to have a clearer conception of the various possible subdivisions of a webpage, and citation parameters best suited for each. E.g.:

1. A web site (e.g.: "Wikipedia"); typically with a fully-qualified domain name (one dot, no slashes).
2. A website that is an independent subsection of the preceeding (e.g., has a slash or extra dot in the url), not otherwise related to the parent or any sibling sites. (E.g.: "Pruned", hosted at pruned.blogspot.com, or "The Discovery of Global Warming" at www.aip.org/history/climate/index.htm). How can we distinguish these from sites and pages?
3. A web page of a site (unique urls), possibly even denominated as chapters [E.g., here (Weart) and here (IPCC AR4).]
4. Textually identified or segregated sections, possibly as distinct pages (e.g., IPCC) or as anchored references within a page (e.g., Weart). This may include blog entries, which may lack specific anchors.

Instances at all of these levels (though not all simultaneously) can be cited using the existing tools, but the results tend to be haphazard, and the means used irregular and confusing. As a start, how might we clarify these subdivisions? ~ J. Johnson (JJ) (talk) 23:10, 26 December 2012 (UTC)

Positioning of Text–source integrity, Bundling citations

I think that these sections

• Text–source integrity
• Bundling citations
• In-text attribution

which are not central to this guideline are better off blow "Inline citations" and "General References". -- PBS (talk) 02:57, 30 December 2012 (UTC)

Why do you think they're not central to this guideline? I know you don't like them, but that's a separate issue from whether they're important parts of the guideline. I see editors cite them a lot, so the attempt to bury them really isn't appropriate. SlimVirgin ^(talk) 03:13, 30 December 2012 (UTC)

I see that you have again altered the ordering without consensus what is the hurry? Please wait until there is a consensus here on the talk page one way or another for such a change.

You wrote in the edit history "the order was changed without consensus, so I'm restoring these sections to be next to inline citations" when was the order changed without consensus?

The central point to this guideline is about inline citations and general references. I do not think that the guideline benefits from the a change in ordering that you made.

Your point about "I see editors cite them a lot" is addressed by links to the sections. I am more concerned with the editor who read this guideline from the top as a fresh document and I think that the ordering is better as it was before you made your change. -- PBS (talk) 10:49, 30 December 2012 (UTC)

How to deal with...

...the first reference at Names of Asian cities in different languages: References ("KNAB, the Place Names Database of EKI") with scores of links? Thanks. —  AjaxSmack  01:44, 2 January 2013 (UTC)

Yes, that is ugly: a single named ref replicated 126 times. The basic problem is that the entire table is derived from a single source. (Which rather makes me wonder if there might be copyvio problem.) My suggestion is to use a single general reference: "All of these entries are taken from ....." ~ J. Johnson (JJ) (talk) 00:10, 3 January 2013 (UTC)

In reality, only a small fraction are from that source, so that wouldn't work. —  AjaxSmack  22:04, 3 January 2013 (UTC)

Huh??? Correct me if I am wrong, but what I see in the reflist of that article are six references. One has two back-links (used twice in the article?), four have only one backlink. But the first one has 128 backlinks. How can you possibly claim that 128 citations out of 134 is "only a small fraction"? ~ J. Johnson (JJ) (talk) 02:00, 4 January 2013 (UTC)

Sorry. I meant that only a small fraction of the total article entries are from that source. Others are interwikilinked and many are unsourced. Therefore, I'm not sure how an "All of these entries are taken from ....." statement might work. —  AjaxSmack  03:00, 4 January 2013 (UTC)

I see. That suggests this article has a major problem with lack of sources. I can't help you with that. ~ J. Johnson (JJ) (talk) 23:03, 4 January 2013 (UTC)

As to the issue with backlinks, I created {{listref}} just for this purpose. --— Gadget850 (Ed) ^talk 00:14, 5 January 2013 (UTC)

Thanks. That's exactly what I was looking for and I have applied it. —  AjaxSmack  04:55, 5 January 2013 (UTC)

Getting citations like Wikipedia

I am trying to setup my wiki to have the exact same styling of citations as Wikipedia. To test this I copied the Wind Power page and got the Cite extension. However, when I looked at my page there was a bunch of <ref'> errors and missing templates. I started trying to copy the templates and now the wiki gives a Fatal Error for time executed. Is there a way I can set up my wiki to have exactly the same Cite as Wikipedia? 2001:558:6045:B8:25C3:F55B:1D25:DE3E (talk) 20:28, 6 January 2013 (UTC)

Unpaginated Electronic Law Journals

Some electronic law journal services (such as Quicklaw) provide some law journal articles in PDF format with pages different from those which would have been found in the print copy of the journal. The pages are simply sequential within the article, rather than representing the actual print copy pages. So an article that in fact begins on p. 173 of the journal and runs to p. 190 will in Quicklaw PDF appear only with the page numbers 1 through 24, which often range longer even than the printed article, since the margins, typeface, etc. differ from the print version. Now that libraries are cancelling printed journal subscriptions, there is also no way to access the print versions. One authoritative guide, the McGill Guide to Uniform Legal CItation, cryptically says: "If there is no system for pinpointing (i.e., page citation), use a PDF copy of the source that is provided by the Publishers, and then cite to the page number." But they say nothing about proper formats for this process, and the result would seem to be ridiculous, e. g., Author, Article Title, date, volume Journal Name, starting page (say 173) then cited page, (say 3). But the page cited obviously cannot be earlier than the starting page of the article! (Originally or previously posted at Wikipedia talk:Citing sources/Archive 29 by 184.151.127.205 (talk), January 5, 2013, then shortly removed from there (as not a proper addition to an archive) by me: Nick Levinson (talk) 20:10, 5 January 2013 (UTC))

My solution to discrepant page numbers is to cite a page with something like this, using hypothetical numbers:

p. 178 (p. 6 per PDF viewer)

I don't name the PDF viewer program because there are now several and probably all use the same pagination since that would depend on the file. Although a viewer may reveal the file publisher's original pagination as well as the viewer's generated pagination, I don't count on all PDF viewers doing so for the file.

Nick Levinson (talk) 20:10, 5 January 2013 (UTC)

Similar problem with the pdfs of some scientific articles, where the visible ("printed") pagination is not that of the printed issue, but starts at 1. I also have several cases of a large document that contains other documents, which are valid sources in their own right, with their own pagination. In one case I simply put the pdf pagination (which is the most convenient level) in brackets, with a note explaining that.

I have seen pdfs where the pdf pagination was adjusted to match the printed pagination (such as the chapters in the IPCC's AR4). I believe that is set in the pdf file, and nothing to do with the viewer. Though that might depend on the pdf version. ~ J. Johnson (JJ) (talk) 20:27, 6 January 2013 (UTC)

Sometimes people describe the location, with labels like "On third page of Chapter 2" or "Section on 'Methodology'" or "Search for the phrase foo". WhatamIdoing (talk) 22:26, 9 January 2013 (UTC)

In-text attribution

The wording of WP:INTEXT needs tweaking, I think. It says

... In-text attribution should be used with direct speech (a source's words between quotation marks); indirect speech ...

MOS:QUOTE says

... Format a long quote ... as a block quotation ... Do not enclose block quotations in quotation marks ...

One alternative would be to just drop the words in parentheses. Another would be to expand them:

... In-text attribution should be used with direct speech; indirect speech ...

... In-text attribution should be used with direct speech (a source's words between quotation marks or a block quotation); indirect speech ...

Comments? Aymatth2 (talk) 16:55, 25 January 2013 (UTC)

Good idea, and done: "a source's words between quotation marks or as a block quotation." [1] SlimVirgin ^(talk) 23:49, 26 January 2013 (UTC)

I have reverted the moves to the two sections "Text-source integrity" and "In-text attribution" (but not the change in text). "Text-source integrity" is not describing an Inline citation method (so it shoudl not be a subsection of "Inline citations"). Also I think it should be kept next to text bundling and TB section still needs editing along the Lines I suggested previously. I think that "Text-source integrity" should remain near the bottom of this guideline because it is not describing citation method, and probably needs a home in a different guideline, but as no one has suggested a better home it may as well remain near the bottom of this one. I do not think it should be moved up because it will lead some to mistakenly infer that in-text-attribution is a citation method. -- PBS (talk) 10:12, 27 January 2013 (UTC)

PBS, please stop reverting every edit I make to this page. You've placed these sections out on a limb away from their context. If you want to do that, please gain consensus for it. I know you don't like them, but that doesn't mean they're not an integral part of how to write and place citations. SlimVirgin ^(talk) 17:49, 27 January 2013 (UTC)

SV you comment implies that I am reverting you edits through bad faith. That is not true. I have only reverted those changes I do not think are an improvement to the guideline, and I have explained above in detail why I made those reverts.

SV it is not I who have placed them anywhere -- I have restored them to where they were before you moved them without discussing the changes and gaining consensus to move them. It is you who made a bold move, but instead of then discussing it and gaining a consensus for the moves, you have decided not to bother to do that, but have reverted my revert instead. Why are you not following the advise at WP:BRD explaining your edit and waiting to see if there is a consensus for the move? I think it would have been more constructive if you would have addressed the reasons I gave for my reversals. But instead of doing that you say "please gain consensus for it" as if it were I who was making changes! Why have you brought this conversation down to the level of personal accusations instead of addressing the substantive points I raised? -- PBS (talk) 19:56, 27 January 2013 (UTC)

• To heck with this... I'm unwatching this page and don't care how it ends up. I shall continue to reference articles in this manner - it got through WP:GAN like that, so I don't see that there's anything wrong with the way I do things. The constant bickering on this page is no longer of any interest to me. --Redrose64 (talk) 20:50, 27 January 2013 (UTC)

Essay on citing sources with Zotero

I started an essay on citing sources using the drag-and-drop method with Zotero at User:A13ean/Zotero, and would appreciate outside input, refinements and suggestions. If enough people think it would be helpful, I will make it a public essay and link to it under the citation templates and tools section. Thanks, a13ean (talk) 21:53, 5 February 2013 (UTC)

Citing Social Networks

I was wondering if it is possible to cite social networks' content as a resource. Many twitter accounts are verified now, and their owners use twitter as the main means of communication. In addition, there are public figures and organizations that maintain Facebook pages as means of communication with the public Asaifm (talk) 09:31, 1 February 2013 (UTC)

You probably can cite verified identities in the social network domain, such like a person or company facebook or twitter account, but if it is relevant information, then it usually can be found in other better suited sources outside the social network domain as (the official webseite, a press release, interview/statement in "regular" media). Facebook in particular (though popular with companies and inviduals) I would avoid at all cost, since it is not accessible without registration.--Kmhkmh (talk) 13:02, 1 February 2013 (UTC)

Thank you for your response and pieces of advice. I understand now how to proceed. Keep social networks as a last resort. Asaifm (talk) 14:18, 1 February 2013 (UTC)

I would rank Twitter comments (Twits?) below blogs as hardly useful for an encyclopedia. ~ J. Johnson (JJ) (talk) 23:13, 1 February 2013 (UTC)

You might cite Twitter if you are quoting a message. WP:RS recommends citing original sources for direct quotations, even if they are available in what we would normally consider to be better sources. Beyond that, I'm not sure why you would do so, but maybe there is another appropriate use. WhatamIdoing (talk) 16:23, 2 February 2013 (UTC)

I think the "original source" for a quote is often where it was quoted, not the original utterance itself. The difference is that in the process of "quoting" somebody else essentially vouches that "so-and-so" said "this". For an editor to state that he heard Joe Blow say "quote" while waiting for the bus smacks of original research. And it seems to me that Twitter is really not significantly different from oral speech. If something was significant I expect it would be picked up at other sources. Perhaps even a blog! ~ J. Johnson (JJ) (talk) 00:29, 6 February 2013 (UTC)

That might be reasonable, but it's not our advice. I was around for some of those conversations, and we definitely meant the original-original utterance (assuming it wasn't unrecorded and therefore ephemeral oral speech), not the first copy after the original. WhatamIdoing (talk) 16:38, 9 February 2013 (UTC)

Any way to cite info received in an email?

I have been working on National Football League Rookie of the Year Award and in an email conversation with an NFL PR manager I learned that there were more than 3 million votes cast for this year's Pepsi award, but this information wasn't listed in the press release. I think I already know the answer, but is there any way for me to include this information in the article? SGMD1 ^Talk/_Contribs 15:58, 7 February 2013 (UTC)

Only if it was published in a reliable source. --— Gadget850 (Ed) ^talk 16:33, 7 February 2013 (UTC)

Emails are private communications, and so fail WP:V. --Redrose64 (talk) 16:57, 7 February 2013 (UTC)

I have had problems with "Peerage .Com" the site is not a reliable source, but usually Daryl Lundy (the site's author) cites reliable sources, so his site can be cited via WP:SAYWHEREYOUREADIT. Unfortunately sometimes his source is a email he has received, which of course (unless it was from a published well known academic in the field) is not a reliable source. -- PBS (talk) 10:53, 9 February 2013 (UTC)

Discussion on Meta regarding WebCite service

Please come participate in this discussion: meta:WebCite. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 18:59, 9 February 2013 (UTC)

Removing superfluous citation parameters; cutting down bureaucracy

Over the past few months, following the procedure insisted on by contributors to this page, I have repeatedly gone through the process of notifying in advance on an article's talk page, my intention to remove superfluous and unhelpful cite template parameters such as 'location=New York' for The New York Times, or 'publisher=BBC' for BBC Sport or 'publisher=MLB Advanced Media' for references to MLB.com. (These changes bring the citations closer to the style-independent guidelines for using the citation template, e.g. WP:CITE#Newspaper articles "city of publication, if not included in name of newspaper"). After a week or so without objections, I go ahead and make the change. I have made these improvements - in tandem with many other gnoming fixes - to more than 50 articles, and in not a single case has anyone objected (or even commented) on the talk page during that waiting period; in only two cases out of those fifty-odd has any editor objected at all, and those objections came after I had made the changes following the waiting period (those two ended up reverted). It seems to me that the process of asking permission beforehand is completely pointless. In the unusual event of anyone objecting, they only object - or even notice - when the change is actually made. It would make much more sense for this sort of change to be subject to the usual WP:BRD rule applicable to changes in general. An objection rate of only 4% suggests that they are scarcely controversial. Colonies Chris (talk) 20:21, 1 February 2013 (UTC)

I remove these parameters from articles without discussion when I see them. People fill them in just because they're there; in the case of new editors they may feel they're expected to. SlimVirgin ^(talk) 20:27, 1 February 2013 (UTC)

If asking ahead of time is too irritating, the solution is to leave the articles as they are. Editors looking at an article they haven't edited before should strive to keep the existing citation style and adapt any new citations accordingly; the last thing they should do is try to change the style based on their own preference about which material is "superfluous". Editors who like to do gnoming need to be particularly careful to restrain any urge to standardize citations as they go. — Carl (CBM · talk) 21:29, 1 February 2013 (UTC)

Removing that the New York Times is published in New York, or that the BBC publishes BBC Sport doesn't count as changing the citation style. I've seen a lot of complaints about this from new editors, who see these citations in an article and think they have to copy them, then spend time trying to find out who the publisher of some obscure newspaper is. Removing superfluous parameters is a service. I just wish it could be made clearer at the template level that they're almost never needed in the first place. SlimVirgin ^(talk) 21:55, 1 February 2013 (UTC)

Yes, it certainly does count as changing the citation style, if every reference in the article has a location specified and the locations are then removed. The decision whether a location needs to be specified for every reference, for just some references, or whether it should be specified for no references in an article is the epitome of a style decision. — Carl (CBM · talk) 01:28, 2 February 2013 (UTC)

This whole discussion hinges on what is considered a 'style' change; for the more hardline contributors to this page, virtually any change to citations constitutes a style change. First of all Carl's statement that gnoming editors 'should not change the style to suit their own preferences' is inappropriate here. The changes I make are bringing citations into line with the style-independent guidelines for the use of the cite template. Not 'my personal preference', but agreed-by-consensus guidelines that apply to any and all citation styles. Second, the unsupported assumption that every element of the citations must be the result of a conscious 'style decision'; all experienced editors know that articles and their citations tend to grow up in a haphazard fashion, with - as SV notes - some editors supplying 'location' and 'publisher' (or even 'language=English') just because the template parameters exist, or because they saw someone else do it somewhere. Some simply misunderstand the parameters and code 'publisher=Newsweek' or 'author=Associated Press', or use 'cite web' when they should have used 'cite news'; are these also to be considered the inviolable results of a style decision? The template documentation is already quite clear on the subject of publisher and location (WP:CITE#What_information_to_include); the only time 'publisher' is even mentioned is for books, and for 'location', the documentation specifically states 'city of publication, if not included in name of newspaper'. No objector has been able to explain how this superfluous information might be valuable to someone checking a reference. But none of this cuts any ice with those who are determined to see any tiny fix as a forbidden style change. Colonies Chris (talk) 15:12, 2 February 2013 (UTC)

It is perfectly reasonable for an editor to say, "the citation style I establish for this article includes the location for every print reference", and equally reasonable for them to say, "the citation style I establish for this article never includes the location of any print reference". That's why it is called "style" after all, because it is a more or less arbitrary decision. In particular: information may be included in references for purely stylistic reasons even if the information is not directly valuable for someone checking the reference.

WP:CITEHOW is not a comprehensive list of what may or may not be included in citations; WP:CITEHOW directly says that it is only giving examples of "typical" citations, not that it applies to all citations in all articles. As the guideline says, "While citations should aim to provide the information listed above, Wikipedia does not have a single house style, though citations within any given article should follow a consistent style." Note in particular that it only says "citations should aim to provide" and does not say anything about omitting information. — Carl (CBM · talk) 15:37, 2 February 2013 (UTC)

Perhaps it should say something about omitting information. WhatamIdoing (talk) 16:36, 2 February 2013 (UTC)

I have a view somewhere between the two expressed so far: yes, you can establish a citation style that includes (or omits) these items. But given the very low response rate, I believe it fair to assume that these items are only rarely (about 4% of the time, apparently) intended to be included. So I think that bold improvements with zero edit warring (and even an apology for not knowing in advance that this was one of the rare articles with that item as part of a defined style) would be acceptable for these changes.

I'm thinking about this in practical terms. If 100% of talk-page messages left in advance drew no response, then there is very little point in posting those messages in advance. (There might be a point in leaving a post-edit note about the change, like "Hi, I made this change because I thought you probably didn't want this redundant information, but if I guessed wrong, then please feel free to revert".) If 96% of the changes were accepted (so far), then it's likely that the use of these parameters commonly is not part of the articles' established citation styles. With such lopsided odds, I think it's fair to assume that edits like these should be "boldly correcting the citations to match the article's established citation style" rather than "changing the citation style without consensus". WhatamIdoing (talk) 16:36, 2 February 2013 (UTC)

The point of requiring the talk page posts in advance is that we don't want editors going article-to-article changing citation styles, and requiring editors to post a note first helps us emphasize that changing citation styles is not the same as changing spelling errors. Changing citation styles should be a rare event, based on a specific need of the article, and hence it is reasonable to require editors who plan to make the change to discuss it first. In other words, the frustration mentioned in the original post is intentional, because we want to discourage the sort of editing that is being described. — Carl (CBM · talk) 16:43, 2 February 2013 (UTC)

In case it seems like I am being bureaucratic solely about locations, let me point out that although the specific question here is about locations, there are many other "optional" data that we treat similarly. A good example is with ISBNs, which are also "optional". Some editors go out of their way to include ISBNs for every book reference; other editors prefer never to include them [2]. We would not want someone going around removing all ISBNs from articles because they feel the ISBNs are superfluous; at the same time we don't want editors to go through adding ISBNs to articles that don't include any. The same is true for a lot of other indexing data, such as PubMed or JStor links - there is no general consensus on whether to include the additional information. The primary goal of CITEVAR is to avoid enormous amounts of time-sucking discussion by telling everyone to sit on their hands when they see a different citation style than the one they prefer. — Carl (CBM · talk) 16:56, 2 February 2013 (UTC)

In the general case, I fully agree with you. But where "these items" are concerned (I've bolded the relevant words above, so you won't overlook them), I think that we can safely assume that these specific items are not actually intended to be part of the citation style in approximately 96% of the articles. "These items" are the ones he names above: "BBC" as the publisher when "BBC" appears in the title of the program, "New York" as the location when "New York" appears in the title of the newspaper, and "MLB Advanced Media" for MLB.com, when MLB Advanced Media is the only publisher of that website. WhatamIdoing (talk) 17:34, 2 February 2013 (UTC)

• The New York Times has different editions in different locations and the same applies to other periodicals such as The Economist. BBC Worldwide Learning aka BBC Active is published by Pearson not the BBC now. From such examples, it seems that the parameters in question might be significant in some cases and so the information should not be removed in a rote way. Warden (talk) 17:09, 2 February 2013 (UTC)

  True, but he's not talking about removing those. (In your last example, he should remove "BBC" from the |publisher= field, because it's factually wrong.) WhatamIdoing (talk) 17:34, 2 February 2013 (UTC)

SlimVirgin and Colonies Chris are right. Carl is entirely wrong. It is certainly not a significant style change to remove "location=New York" from a ref citing The New York Times; on the contrary, it is simply following the rules. The rules are clear that location is not required when the location is included in the title of the publication. It is completely wrong and ridiculous to suggest that such changes should be discussed on the talk page of an article with a time limit before doing it. The Economist is a different case altogether because that title does not include the location. -- Alarics (talk) 21:27, 2 February 2013 (UTC)

Even if there were rules to say that the location is not required, there are no rules that say it cannot be included. In order to use the MOS to justify the removal, we would need rules that say it cannot be included. The general MOS principle is that optional styles that have been implemented in an article should be preserved; that is in the lede of WP:MOS, and WP:CITEVAR repeats it here for the specific case of citations. At the moment, including the location of every reference is an optional style. — Carl (CBM · talk) 13:02, 3 February 2013 (UTC)

This is not a MOS guideline. As for the wording in WP:CITEVAR: "'When I use a word,' Humpty Dumpty said, in rather a scornful tone, 'it means just what I choose it to mean — neither more nor less.'". I do not think that a non biased interpretation of changes in "citation style" covers the sort of changes that are being discussed here. Far to many editors abuse the wording of WP:CITEVAR to prevent improvements to citations which have nothing to do with changing the style. The original reason for the wording of CITEVAR was to stop the mass moves from harvard style to footnote3 style citations. The sooner CITEVAR is put back in its box and restricted on what is covered the better it will be for Wikipedia.-- PBS (talk) 16:04, 4 February 2013 (UTC)

As a point of fact, I'm the person that originally wrote and proposed CITEVAR two years ago, and that limited application, although important, was not my only goal. WhatamIdoing (talk) 16:51, 9 February 2013 (UTC)

• It might be helpful to clarify that in neither case where an editor objected did they argue that they were making a style decision. Their objections can be summed up as 'someone put these in a long while back and I don't know what good they are but don't touch them'. Colonies Chris (talk) 10:09, 3 February 2013 (UTC)

When "location" is not required (because it is already part of the name of the publication), but somebody has put it in anyway, it can be removed by any editor without discussion because it is obviously completely superfluous. It doesn't need the rules to say so explicitly. It's not a style change, it's just common sense. Why on earth would anyone want to keep it? I cannot see why we even need to discuss this. -- Alarics (talk) 20:19, 3 February 2013 (UTC)

The best way to proceed is simply to remove those parameters without discussion, but if someone reverts and says they intended it that way – that is, they intended it as a style choice, rather than having made an error – either discuss it with them or move on, but don't revert them. SlimVirgin ^(talk) 20:45, 3 February 2013 (UTC)

Should do the same for omitted ISBNs and DOIs as well? Rather than assuming that their omission was intentional, users could just add them to all articles that don't include them, and see whether anyone reverts. — Carl (CBM · talk) 12:52, 4 February 2013 (UTC)

Yes, that's how it's normally handled. Then if someone has left them out deliberately, they can say so and revert. SlimVirgin ^(talk) 00:20, 5 February 2013 (UTC)

No, I am sure that if someone started going through and adding these to hundreds of articles, including FAs, someone would refer to CITEVAR to ask them to stop. — Carl (CBM · talk) 00:56, 5 February 2013 (UTC)

Granted, if they were doing it en masse, they'd be asked to stop. SlimVirgin ^(talk) 01:24, 5 February 2013 (UTC)

• Carl - earlier in this dscussion you wrote "The decision whether a location needs to be specified for every reference, for just some references, or whether it should be specified for no references in an article is the epitome of a style decision." (emphasis added). So even where the practice within an article is inconsistent, we must assume that the totality of editors have collectively - but implicitly - decided that inconsistency is the citation style of the article? Look at this version of Ryan Zimmerman, before I did any work on it. This has 29 citations. Of those, seven are citing the Washington Post. Two of the citations specify a location and publisher, the others do not. One of them has 'location=Washington, DC', the other has 'location=Virginia Beach, Virginia'. Was User:Nikkimaria right to revert my attempt to bring some consistency to this mess because prior consent is required even for this sort of tidy up? Colonies Chris (talk) 10:25, 4 February 2013 (UTC)

It is important to this conversation to note that CBM is apposed to any and all attempts to standardize anything. He has attempted to block almost all attempts at standardization within Wikipedia and has a reputation for blocking editors who don't wish to do things his way. Be careful Chris that you don't end up getting blocked by CBM. He is an admin and is well known for his anti standardization sentiments. He also a well established critic of AWB so if you use AWB be careful there too. He is one of a number of Programmer/editors that are looking for any reason to get AWB completely outlawed on wiki. 108.28.162.125 (talk) 12:34, 4 February 2013 (UTC)

• • The inconsistency there could use cleanup, but a note on the talk page first would have been reasonable. There is nothing urgent about the edits that couldn't wait for a little while, after all, and it is a sign of respect for fellow editors to wait to see what they think about such things. — Carl (CBM · talk) 12:52, 4 February 2013 (UTC)
    • And so we complete the circle. Read my very first contribution to this discussion. Experience has shown that no-one pays any attention to such talk page notices. They are completely pointless. Any rare editor who objects to the changes once they've been made has the revert facility easily available. Colonies Chris (talk) 14:13, 4 February 2013 (UTC)

But the point of the talk page notices is not that lots of people will respond to them. The point is to give people the opportunity to respond to them, and to emphasize that a sound reason needs to be given before the change is made. There is a second point, as well. Experience has shown that editors sometimes go overboard with standardizing references, so the talk page notices are also intended to keep the such edits to a moderate pace, which encourages editors to limit the style changes to articles where the changes are truly needed.

There is a sound reason that we have CITEVAR in the first place, rather than just telling editors to revert changes they don't like. We tried the other way back in the old days and it didn't work! There's no reason to think it would work better now than it did then. ENGVAR and CITEVAR have been extremely effective at reducing the amount of time wasted with people arguing about style changes. — Carl (CBM · talk) 15:25, 4 February 2013 (UTC)

You are the only person arguing that the sort of minor fixes I'm describing constitute a 'style' change. Everyone else sees them as just a bit of housekeeping within whatever style is in use. WP:CITE#Variation_in_citation_methods gives examples of the sort of things that are to be considered style changes not to be undertaken without prior agreement; they are substantial changes, not minor fixes of this sort. Colonies Chris (talk) 15:50, 4 February 2013 (UTC)

No, he isn't; I agree that some of your edits are contrary to CITEVAR. (You must have forgotten to notify me that I was being discussed here). In fact, you yourself argued here that the presence of publisher for newspaper constituted a change in style. The whole point of CITEVAR is to avoid situations like this one, in which you got into an edit war over the inclusion of publishers. See also previous discussion here. Nikkimaria (talk) 16:11, 4 February 2013 (UTC)

Since I'm fully aware that you obsessively stalk my every move, there was no need to notify you. Strangely, you choose not to mention, in the so-called 'edit war' you refer to, that the editor had reverted my good faith changes with a hysterical accuation of 'vandalism' and even reported me to WP:ANV and was of course turned away from there. Colonies Chris (talk) 17:41, 4 February 2013 (UTC)

Strangely, you choose to miss the point entirely: these courtesies that you dismiss as "pointless" are important to a collaborative project. Nikkimaria (talk) 22:11, 4 February 2013 (UTC)

Carl is now trying to claim that the statement in MOS about "location" not being needed when the location is already in the title of the publication means it is optional, on a par with including the ISBN for a book. It is not. The sentence in the MOS was not intended to mean that. It was meant to deter people from using that parameter in those circumstances, but to urge them to use it otherwise, as in "city of publication, if not included in name of newspaper". It is quite different from the ISBN, which is an optional extra - always useful but never compulsory. The location of a newspaper IS necessary, unless it is already obvious, in which case its inclusion is not a "nice to have" but a "completely pointless". CITEVAR has no bearing on any of this. -- Alarics (talk) 16:28, 4 February 2013 (UTC)

That entire section in the MOS is just a description of "typical" citations and "examples" (it uses both words). It is not in any way intended to describe what must or must not be included, because we do not have any uniform rule about that. Indeed, the page says "While citations should aim to provide the information listed above, Wikipedia does not have a single house style, though citations within any given article should follow a consistent style." The "information listed above" is the list of examples in the section "what to include". — Carl (CBM · talk) 18:18, 4 February 2013 (UTC)

The point of CITEVAR is to use common sense and not have revert wars over style issues. It makes no sense to add the location or publisher of the New York Times or BBC Sport. What matters is whether the parameter helps the readers. It helps them to know the name of the newspaper or broadcaster, the date, the title of the report. Those issues let them judge the relevance of the topic and the quality of the source. Adding location usually tells them nothing, but sometimes it will help. If you've sourced an article about Birmingham to the Evening News, they can't judge whether it's the Evening News (Birmingham), or the Evening News (Tokyo); the former is a good source for Birmingham, the latter probably not. So judge each instance on its merits, rather than according to some rule, or just because the parameter is there. SlimVirgin ^(talk) 00:20, 5 February 2013 (UTC)

If "use common sense" worked, we wouldn't need CITEVAR in the first place. But too often different people have different "common sense", and in the case of citation styles none is really any better than another, so CITEVAR instead says: "because it doesn't really matter what style is used, just sit on your hands and keep the style that is already there". The choice of style is also not very related to whether the style "helps readers", because everyone thinks their preferred style helps readers. Some people think that including ISBNs for every reference helps readers; some think that including locations does. And CITEVAR also covers things such as templates that are invisible to readers. We should all remember that without CITEVAR, there are plenty of gnoming editors who would start adding ISBNs and citation templates to every article. — Carl (CBM · talk) 00:56, 5 February 2013 (UTC)

I do agree with you about the importance of CITEVAR, and it's true that it has dramatically cut down on style wars. SlimVirgin ^(talk) 01:22, 5 February 2013 (UTC)

"Common sense" is embodied in the agreed guidelines for the use of the cite template at WP:CITE, which are independent of any specific citation style. I suggest that the wording of CITEVAR be changed to make clear that any change which solely brings citations into closer alignment with the guidelines at WP:CITE is acceptable (subject to WP:BRD, as ever), and any change which goes beyond that requires prior consent. Colonies Chris (talk) 10:32, 5 February 2013 (UTC)

The guidelines here are simply examples; they are not intended to constitute a house style. I undid a change to the policy because it would give people open license to make all sorts of changes (e.g. adding ISBNs and DOIs) en masse. Remember that for anything this guideline says is unilaterally allowed, someone will decide to go through large numbers of articles to do it. Thus we need to be extremely cautious in granting this permission. — Carl (CBM · talk) 14:43, 6 February 2013 (UTC)

I agree with CBM, this guideline provides only a brief summary of how citation templates might be used, and is ill-suited for going into detail. If the community wants to go into detail, it should do so at Help:Citation Style 1 or some comparable page for the the Citation template. Jc3s5h (talk) 14:51, 6 February 2013 (UTC)

@CBM: Those guidelines describe ISBNs and DOIs as 'optional'. That means that adding them would not be 'bringing citations closer to the guidelines', and would therefore not be acceptable under my proposed wording. @Jc3s5h: the point of referring to the guidelines at WP:CITE is that they are best-practice recommendations not tied to any specific citation style such as CS1, and they apply whether or not citation templates are used. Colonies Chris (talk) 14:59, 6 February 2013 (UTC)

Including locations is also optional; the guideline only says that citations "typically include" "city of publication, if not included in name of newspaper". That does not say that they may not include the location when it is included in the name of the newspaper. Sometimes they have to include it: the Manchester Journal is from Manchester Vermont, the Manchester Evening News is from Manchester, England. But the language you are proposing would allow someone to blindly remove the location from both of these because it is included in the name of the newspaper. — Carl (CBM · talk) 15:18, 6 February 2013 (UTC)

Do you really think that such bizarre behaviour is so much of a danger that we need to make sure we ban it? These are guidelines, not legislation: we don't have write the rules so tightly that we could prosecute someone in a Wikipedia court. People don't randomly remove useful information just because there's a tiny loophole in the rules that doesn't explicitly ban it. (And besides, in a citation I would expect the Manchester Journal's location to be given as Manchester, Vermont, not just Manchester, or for the paper to be referred to as the Manchester (Vermont) Journal.) Colonies Chris (talk) 15:41, 6 February 2013 (UTC)

Yes, we do need to consider experience, which shows there are quite a few editors who take any specific guidance here or in the MOS and view it as a rule that should be implemented in many articles. Your proposed wording would say that someone could entirely remove the location from every reference to the Manchester Journal, just as they could for the Los Angeles Times or the New York Post (neither of which lists a state or country in the title).

The fact that the location is included in the name of the paper is relevant to the way the reference is presented, but it is not the only factor; the prominence of the paper and chance for confusion are also relevant. So for example we might omit the location from the Manchester Journal in an article about Manchester Vermont but include it in an article about a national topic with no close ties to Vermont. But the proposed wording would say that an editor could remove the location blindly without regard to context.

The wording in the examples in this guideline is not very precise because it is just an example; but if people are going to refer to it as "moving closer to the guideline" then the guideline language would have to be much more precise about when to use the location – but at that point we are beginning to develop a house citation style, and we would have to do something that could handle situations like the previous paragraph. — Carl (CBM · talk) 16:02, 6 February 2013 (UTC)

One cannot legislate for every conceivable case. Some common sense is required (I realise that this is an attribute that some people sadly lack). Obviously in the case of Manchester you need to be more specific when it is not "the" Manchester -- Manchester VT or Manchester NH, etc. On the point about ISBNs, this is the first time I have come across a suggestion that it could ever be regarded as unacceptable, on grounds of CITEVAR, to include an ISBN in a book reference. Surely that cannot be right. -- Alarics (talk) 22:28, 6 February 2013 (UTC)

Yes, we don't need to legislate for every specific case because we already have a general statement in WP:CITE: "This information is included in order to identify the source, assist readers in finding it, and (in the case of inline citations) indicate the place in the source where the information is to be found." If a change contributes to this, it's good; if it doesn't, it's not. If, in context, the location is clear, it can safely be omitted because it does not help anyone trying to check a reference. If there is reasonable doubt about location, then the location should be explicitly included. That's what I mean when I say that WP:CITE embodies 'common sense'. Colonies Chris (talk) 10:16, 7 February 2013 (UTC)

There are numerous discrepancies between the examples in this guideline and the citation methods specified by the various printed style guides (and disagreements from one printed guide to another). So the claim that the guidelines in this guideline are best-practice is not true; for many issues, there is no single best practice. Jc3s5h (talk) 23:31, 6 February 2013 (UTC)

The basic elements of an effective citation are the same however you choose to format them. Other style guides may prefer different formatting but the guidelines at WP:CITE are carefully written to be agnostic about formatting; they describe what content is required in a citation; in particular, they omit (publisher) or deprecate (location) when these are not helpful. Colonies Chris (talk) 09:24, 7 February 2013 (UTC)

The change made by Colonies Chris indicated every bit of guidance in WP:CITE had priority over all printed style guides. That's a complete departure from prior understanding of this guideline. Jc3s5h (talk) 14:33, 7 February 2013 (UTC)

I also oppose that change. CITE supports the right of editors to choose any style they want, even if it is not what's considered "best practice" for the average article. See the /FAQ: "Editors on Wikipedia may use any style they like, including styles they have made up themselves."

If someone were to define their citation style as "always includes locations for periodicals", then CITE would support that, even if we think it's silly. (Maybe they wanted to provide it for machine-readability purposes.) This particular one has proven so uncommon that I think its inclusion can be assumed to be accidental until the reverse is proven, but editors are permitted to make that unusual choice. WhatamIdoing (talk) 17:06, 9 February 2013 (UTC)

In the unlikely event that an editor has consciously chosen to cite in a manner different from what our guidelines recommend, they would be entitled to say so and revert. But it shouldn't be necesasary to ask permission before making changes that bring citations into line with style-independent best practice. Colonies Chris (talk) 13:28, 12 February 2013 (UTC)

@Alarics: for an argument that ISBNs should not be blindly inserted, see [3] (personally, I am fine with ISBNs). In general, the fact that people lack common sense is exactly the problem - any specific rule here needs to be relatively safe for people to apply blindly, because some people will apply it blindly. The "What information to include" section is just meant to help people who have no idea how to write a citation, by giving them a general sense of what to include. It's not meant to describe every situation - so telling people they can blindly use it to change the styles in numerous articles is not going to have good results. — Carl (CBM · talk) 00:50, 7 February 2013 (UTC)

Two points:

Actually using BBC for the publisher is not always as dumb as it seems as first. There is no one single editorial team for the website and whether the editor field is filled in or the publisher field is used instead, as the link given for contacting the editor of the page, it is useful information to add. I first came across this when contacting the BBC over one of their pages regarding the River Teme (See Talk:River Teme/Archive 1#BBC source) so the publisher for that page was "BBC Shropshire". -- PBS (talk) 11:28, 9 February 2013 (UTC)

Yes. but in this example the 'publisher' is providing useful information over and above what's in the 'work', whereas 'work=BBC News|publisher=BBC' is merely redundant. Colonies Chris (talk) 17:14, 9 February 2013 (UTC)

"Common sense" is probably not the best phrase to use because common sense is linked to culture and/or education levels. A better turn of phrase is to use the legal definition "Reasonable person" or for our purposes the article The man on the Clapham omnibus which is not so detailed as the "Reasonable person" article, but includes the sentence "The man on the Clapham omnibus is a reasonably educated and intelligent but non-specialist person, against whom the conduct of [a Wikipedia editor] can be measured". -- PBS (talk) 11:28, 9 February 2013 (UTC)

Fair enough - that's a good phrasing of the concept. Colonies Chris (talk) 17:14, 9 February 2013 (UTC)

Lists : Do we need to provide sources for each item ?

This is such a basic issue that I'm sure it has already been addressed but I cannot locate the relevant rule. So here goes: When a list contains nothing but wikipedia entries, with the definitions for each listed item provided in the item's respective wiki entry, do we need to include sources for each item in the list's page too ? Thanks in advance for any guidance. -The Gnome (talk) 11:21, 7 February 2013 (UTC)

You have not provided enough information to give a yes or no answere. It depends on what the list is and whether a specific entry in that list could be seen as outside general knowledge. For example a bullet pointed list of US presidents, would probably not need a citation for Abraham Lincoln (although it might be necessary for less famous presidents included in such a list), but a numbered list that states Lincoln was the 16th president probably does, as that is probably beyond the general knowledge of an English speaking person born outside the United States. However this is not the place to discuss this, as the guidance/policy for such entries relies on WP:V "Attribute all quotations and any material challenged or likely to be challenged to a reliable, published source using an inline citation". Also if the list is in anyway controversial and includes living people then of course the list needs citations (WP:BLP).-- PBS (talk) 10:43, 9 February 2013 (UTC)

It is not WP:LIKELY that list items will be challenged if the information in the list is a basic fact that can be trivially verified by clicking through, even if most people don't know anything about it. For example, if you decided to create a "List of hospitals in Quebec", you probably won't need to provide inline citations to prove that the entries really are hospitals or really are in Quebec (unless someone does actually challenge them, of course). WhatamIdoing (talk) 19:01, 9 February 2013 (UTC)

I will present again the precise context in which I seek advice: A list that contains nothing but items having their own separate wiki articles. The items are described in the list by quoting directly from the relevant wiki articles. The relevant, supporting articles are all referenced. I offer as an example of such a list the List of acronyms: European sovereign-debt crisis. -The Gnome (talk) 09:13, 12 February 2013 (UTC)

See Wikipedia:Manual of Style/Stand-alone lists for guidelines. Remember that Wikipedia articles may be reused, so each article needs to stand on it's own, especially annotated lists. --— Gadget850 (Ed) ^talk 12:28, 12 February 2013 (UTC)

But the lists to which I'm referring are composed exclusively by wiki articles! The references mandated by the relevant style manual are the Wikipedia articles in themselves. I gave an example above. Another example are lists of gay persons, whereby only persons already denoted in their Wikipedia biography as gays are included. Why should such lists need a repetition of the references already extant in the respective Wikipedia articles? -The Gnome (talk) 16:25, 13 February 2013 (UTC)

Gnome, you keep responding as if people were telling you "Yes, absolutely, you must always provide inline citations no matter how stupid it seems!" So once again, let's run through this. There are exactly four situations that require inline citations on Wikipedia. They are (see WP:MINREF):

• A direct quotation: Do these lists contain direct quotations?
• Something that has been WP:CHALLENGED: Has anyone actually fact-tagged items? Left complaints on the talk page? Anything like that?
• Something that you believe is WP:LIKELY to get challenged: What's your guess about the probability, using your best editorial judgment? More than 50%?
• Contentious matter about BLPs: Anything about living people here that might upset folks?

If the answer is "no" to four out of four items, then no policy requires any inline (or other) citations, full stop. If the answer is "yes" to any one item, then whatever list entry triggers that "yes" must have an inline citation. WhatamIdoing (talk) 05:31, 15 February 2013 (UTC)

It seems to me that Gnome is referring to material that has been properly cited, but in another article. What he is talking about is where the references "are the Wikipedia articles in themselves." Particularly: "Why should such lists need a repetition of the references already extant in [other articles]?" [Emphasis added.] Gnome: that would be essentially citing Wikipedia, which we don't do. If some material needs to be cited — and I think we are all agreed as to what — then it has to be cited to the external source in each article. Each article has to stand on its sources on its own. Okay? ~ J. Johnson (JJ) (talk) 22:28, 15 February 2013 (UTC)

Then let's be clear: If the answer is "yes" to any one item, then whatever list entry triggers that "yes" must have an inline citation in that article, not just somewhere on the website. WhatamIdoing (talk) 02:01, 19 February 2013 (UTC)

Yes, let's be clear. W, it seems to me there has been absolutely no dispute that certain material requires references. (So why are you belaboring it?) The issue was, as Gnome clearly asked, and I repeated, whether there must be repetition of such references across articles. I think we can all agree that the answer to that is "yes", that all references needed within an article must be contained with in the article.

I believe we have consensus on this. Does anyone object? ~ J. Johnson (JJ) (talk) 23:35, 19 February 2013 (UTC)

Why am I belaboring the requirements? Because of the flip side. These four things are required to have inline citations on that page—and nothing else is required to have any citations anywhere. WhatamIdoing (talk) 00:36, 20 February 2013 (UTC)

Our policy on that "Lists, whether they are embedded lists or stand-alone lists, are encyclopedic content as are paragraphs and articles, and they are equally subject to Wikipedia's content policies such as Verifiability" to ensure that each item to be included on the list is adequately referenced and that the page on which the list appears as a whole represents a neutral point of view.Moxy (talk) 00:14, 20 February 2013 (UTC)

Yes, it's exactly the same rules. WhatamIdoing (talk) 00:36, 20 February 2013 (UTC)

SO WE ARE ALL AGREEED! Fine, nothing more to say, let's all go home. ~ J. Johnson (JJ) (talk) 02:28, 20 February 2013 (UTC)

Thanks to everyone who came forward with advice and/or opinion. Appreciated! -The Gnome (talk) 00:47, 21 February 2013 (UTC)

I just saw this discussion and I am late with this comment. Since I don't see the point made above, though, I'll make it now. WP:CIRCULAR says "Do not use articles from Wikipedia or from websites that mirror its content as sources, because this would amount to self-reference." Wikipedia articles are NOT reliable sources. Wtmitchell (talk) (earlier Boracay Bill) 02:41, 21 February 2013 (UTC)

...And it opens again! :-) Thanks for the input, and yours IMO is an important point. However, the difference here is this: It's not about a Wikipedia article that uses as references other Wikipedia articles. It's about a list, whereby each and every item listed directs to the respective Wikipedia article, properly referenced on its own. Clearly, as the consensus above indicates, there is no need to repeat in the list the references in those Wikipedia articles. Take care. -The Gnome (talk) 06:48, 22 February 2013 (UTC)

Not really. Lists are articles (unless the list's purpose is for navigation, e.g., a dab page). WhatamIdoing (talk) 02:18, 23 February 2013 (UTC)

Alright. When the list in question meets none of the four criteria listed in the relevant rule, then there is no need to repeat the citations already extant in the respective Wikipedia aerticles. A list, therefore, such as this, whereby BLPs are not involved at all, there are no direct quotations and nothing has ever been challenged nor is likely to be challenged, does not need to have citations. Thanks, again. -The Gnome (talk) 11:03, 23 February 2013 (UTC)

Unexpected ref error message

I made this edit [4]. It removed three named refs, defined elsewhere on the page. Still te page produces ref errors (at end of page). Why? I only removed a <ref name="..."/>. Not the ref link ifself (3x), that is still there. -DePiep (talk) 01:57, 3 March 2013 (UTC)

The article uses list-defined references, which means that the full bibliographic information is typed in the reference section near the end of the article. This system requires that each source contained in the list at the end be used at least once somewhere in the article, before the reference section, with a footnote such as <ref name="haaretz-pepperspray" />. Your edit removed the only footnotes referring to the sources with the names "haaretz-pepperspray", "haaretz-stabbing", and "guardian-sherwood", hence the error messages. You can fix them by removing the corresponding sources from the reference section. Jc3s5h (talk) 02:12, 3 March 2013 (UTC)

Short cites

Does anyone object to the clarification that a short cite can reference a full bibliographic citation that exists in either an earlier footnote or a separate references section? Since the issue is unsettled, I thought that both should be in the text and not just one point of view. --Bejnar (talk) 19:11, 3 March 2013 (UTC)

Yes, I consider it unwise for a short cite to refer to an earlier footnote, for three reasons. First, this might be seen as advocating the use of ibid., which can easily be disturbed by future edits. Second, material may be rearranged, so that the order of entries may be reversed. Third, there is a greater risk of deleting a citation without noticing that another citation depends on it. Jc3s5h (talk) 20:05, 3 March 2013 (UTC)

Your first one I don't understand since other places rail properly against ibid. The second is not a problem with harv, is it? The third is a problem regardless as anything that uses the "name slash ref" to hook up a previous cite has that danger. Are you against those as well? But my point was not whether it was a good or bad idea, but merely that as there was no consensus, that both points of view should be represented. --Bejnar (talk) 21:35, 3 March 2013 (UTC)

For sure, ibid is disallowed. But nothing here is advocating it. The main point is that a short cite always references a full citation — where ever it is. And as putting a full cite in a note is standard practice here, as well as fully acceptable by all (?) style authorities, it seems warranted to mention that. I haven't seen any reason why short cites require full citations in a separate references section. ~ J. Johnson (JJ) (talk) 21:55, 3 March 2013 (UTC)

Nobody has so far complained about the technique that I used for refs 2 & 6 of Boar's Head railway station. The essential features are there: the source is identified, and the page within that source is given. --Redrose64 (talk) 22:30, 3 March 2013 (UTC)

I think Redrose's example is a perfectly acceptable approach. Sure, there's a small chance that you might accidentally lose a full citation, but you can always go back in the history to retrieve it. WhatamIdoing (talk) 01:39, 4 March 2013 (UTC)

Yes. There are several good reasons for keeping the full citations in their own area, and I think that ought to be recommended. But lest any confusion develop it seems wise to mention that the full citations/references do not have be in a special section in order to use short cites. ~ J. Johnson (JJ) (talk) 21:42, 4 March 2013 (UTC)

General references

I do not think that this edit.

If the references in the References section in an article that uses short citations, are not "general references" then how do we describe those entries?

Does a general reference become something else simply because a short citation is added to an article that links to that general reference?

If there are entries in a References section some of which are linked to short citations and others that are not, are the ones that are not general references while those that do are not general references?

-- PBS (talk) 03:03, 30 December 2012 (UTC)

Someone changed this to describe all References sections, rather than a General references section, which is used without inline citations (and in fact is rarely if ever used now). So I changed it back. I'm thinking that we should probably remove it. This was discussed before, but I don't recall what the consensus was. SlimVirgin ^(talk) 03:08, 30 December 2012 (UTC)

I've clarified it to emphasize that the section is about the use of a general references section alone. [5] SlimVirgin ^(talk) 03:40, 30 December 2012 (UTC)

Any major changes that were made were to this section were agreed on the talk page. I do not agree with this change, because you have not first addressed the issue of what do we call the long citations in the general references section if not general references and if we do not call them general references then how to explain that they should be sorted in alphabetical order on author with any other references in that section?

So I don not think that "clarified it to emphasize that the section is about the use of a general references section alone" is constructive change until it is agreed on how to describe whatever we call the References section that includes the full citations when short citations are used.

There is no fire on these changes so lets agree what to change before rushing ahead with changes. -- PBS (talk) 10:57, 30 December 2012 (UTC)

Let's suppose there is an article with a general references section but no inline citations. Two of the sources listed in the references section are to books the third is to an online reliable source. Now suppose that one editor creates a Notes sections and adds a {{unreferenced}} template followed by a {{reflist}} template. Another editor comes along and notices this, and although (s)he does not have access to the off line sources, reads the online source and adds inline citations to the article. I do not think it helpful to change the name we use for the source in the references section to something else just because there is now a new section listing inline citations. If so the paragraph on the advise given has to be expanded to include the new term and to state that it and the "general reference" should be sorted by author, otherwise an editor reading this guideline might think that the sort only applies to "general reference" and not those sources which are linked to short citations. So for clarity and brevity I do not think using separate terms for entries in the references section is a good idea. However I am open to suggestions on this, but I suspect that the names "general references section" and "general reference" have taken on specific meanings and changing usage is likely to be confusing. -- PBS (talk) 13:51, 30 December 2012 (UTC)

I believe a good part of the confusion with citations arise from careless use of terms and concepts, and that we should strive to clarify terminology — especially where specific but unfortunate meanings and usage are starting to settle in. I suggest these definitions (from CMS):

• General reference: a source referred to "generally", not for any specific material.
• Full citation (or "full reference"): the full description of a source, containing the complete bibliographic details to aid in finding a source, and distinguishing between possible variants.
• Short citation (or "short reference"): a citation with sufficient detail to link to a full citation. (Typically used "in-line" to identify the source of specific material.)

Two points to note. 1) A short citation (or cite) implies a full citation (typically collected in reference list). 2) That an editor acknowledges, or even recommends, perusal of a source "generally" in no way precludes referring to it specifically. It is quite reasonable to have a source listed as a reference, and again in a separate (say) "recommended reading" list.

I hope that will help. ~ J. Johnson (JJ) (talk) 01:11, 31 December 2012 (UTC)

WP:FURTHER states "The Further reading section ... should normally not duplicate the content of the References section" -- PBS (talk) 23:48, 1 January 2013 (UTC)

A general reference is a citation (in a "References" or similar section) to a source was used while writing the article to find or confirm claims in the article, but the editor didn't provide an inline citation to the source. A "Further reading" section or similar section lists sources that were not used while writing the article, but which are recommended for related information. One source can be both a general reference, and the target of inline citations, if some of the claims that were written while in reliance are followed by inline citations, and other claims lack an inline citation. Jc3s5h (talk) 02:38, 31 December 2012 (UTC)

I think we are partly in agreement (e.g., a "reference" possibly being both general and specific), but I think you are confused on some points. E.g., a "reference" can mean the work itself — what I tend to call a source — such as one might find in the reference section of a library. It can also mean the entry — the full citation — that points (or refers) to the work (source), often found in a list of such "references".

Where I disagree with you is the contention that a general reference is a "citation" (ah, perhaps the source?) "used while writing the article" but not used "inline". I say that "general" means just that: a reference to a work (source) in itself, or its entirety. E.g.: in regarding (say) a "book that discusses all these theories" we would cite the book generally, not any specific passage. What makes a reference general is lack of a specificity in the citation, not lack of a citation. ~ J. Johnson (JJ) (talk) 22:26, 1 January 2013 (UTC)

OK, an editor writes a Wikipedia article, and to confirm certain claims that he put in the article, he reads books titled A, B, and C. The editor then creates a "References" section which contains full bibliographic citations to the three books. But the article contains no inline citations to show which claim came from which book. So the article contains only general references. An editor comes along, adds a new claim, and with it she puts an inline parenthetical citation, which reads "(A 2011)". This citation is inadequate, because it lacks any indication about where in the book the support for the claim may be found, but at least it indicates which book to look in. So the article now contains both general and inline citations, and A is the target of both kinds of citation. Jc3s5h (talk) 23:00, 1 January 2013 (UTC)

I think that Jc3s5h and I agree. A full citation to a source contains within it all the information needed to locate that source in a library (usually down to the page). That citation to a source may appear inline, or it may be placed in a general references section which we usually call "References" or "Sources" or "Bibliography" or some such similar name. I think "general reference" is a shorthand for "an entry in the general references section", as opposed to an "inline citation" which is placed within the body of an article (usually within ref tags). In other words it is the positioning of a citation to a source within a Wikipedia article which is being described when one says "general reference" not what information within an article that citation supports.

In a well developed article their might be, in the general references section, a citation to an article in a journal which is 10 pages long "pp. 89–99". In the body of the article there might be a short citation to the same journal "smith 2007, p. 95", and only that specific fact on page 95 is used to support information in the Wikipedia article. However that one fact has been extracted from page 95 to support the content of the article does not make the full citation to the article in the journal any less of "["an entry in the] general reference[s section]" -- PBS (talk) 23:48, 1 January 2013 (UTC)

Note that in short articles or articles based on a single source the "general references" can often be precise full citations (including page numbers). You can think of it as paragraph based inline citations of sorts, i.e. the article consists only of one (or a very few) paragraph(s), so that the author didn't bother to use footnotes, but rather listed the full citation for the "paragraph sized" article at its end. This is probably also a question of convenience and wikimedia format knowledge, as this approach simply requires less typing and can be used by unexperienced authors not being familiar with WM footnote system. Typical examples might be somthing like Gripenberg_Castle or Rule of Sarrus.--Kmhkmh (talk) 00:00, 2 January 2013 (UTC)

No, PBS and I don't agree. A general reference a full bibliographic citation to a source, but the editor left no information about which claims in the article are supported by that source. If every use of a source is associated with a particular claim by the use of inline citations, then the citations to that source, including both the inline part and, if separate, the full entry in the "References" section, are not general references. Jc3s5h (talk) 02:22, 2 January 2013 (UTC)

  We need a more clarification. First, a general reference can include page numbers, as in: Smith, 2007, "Article", pp. 89-99, Journal. That specifies only the location of the article, not any point in the article; it is a general reference. Likewise, the short cite "Smith, 2007", is also a general reference. But! adding a specification makes "Smith, 2007, p.95" a specific citation. These are usually seen as inline cites, but the defining criterion is not whether it is inline, but whether it is specific.

  Note that "general reference" is not shorthand for an entry in general references section, "as opposed to an inline citation". In fact, those entries are (or should be) general, as they identify works as a whole. Reference to (citation of) any specific parts should be done as a specification on the short cite. That the "full entry" (full citation) is general is irrespective of whether one or more short cites (presumably inline), whether general or specific, refer to it.

To take a swipe at answering Phil's questions (at the top):

1. If by "References section" you mean a bibliography: they are in the form of a general reference. As particular instances: "bibliographical entry"?
2. No: a general reference does not become specific ("something else") because something else links to it.
3. Huh? In general, a general reference (citation) is one that refers to a work as a whole, not to any specific part of it.

~ J. Johnson (JJ) (talk) 23:57, 2 January 2013 (UTC)

I don't know about that, I'm used to consider general references exactly as an entry in the (general) reference section as opposed to an inline citation. Maybe we have a language problem here as well, that is the term "general references" is used slightly differently by various authors for years now. The problem is, when you consider general versus specific, there are 2 perspectives to do that. One is looking only at the source itself and check whether it gives a specific page rather than just the "general" publication. This one you described above. But another perspective is to look at the specificity of content assignment to the source instead, i.e. is it assigned to a specific line (inline citation) or the whole content in general (in reference section at the end of the article). From this perspective a general reference is exactly a source given in the reference section (see the examples I've linked above).--Kmhkmh (talk) 00:54, 3 January 2013 (UTC)

Oh, no doubt about it, we do have a language problem here (reinforced by the years of slightly different usage); that is my starting proposition. But that is ultimately due to different concepts. My hope is that we can reform some of that. But I am running quite late today, so further comments must wait. ~ J. Johnson (JJ) (talk) 02:20, 4 January 2013 (UTC)

I think (though I may be more obtuse this moment than usual) your alternate perspective is where the the source is applied to the article in a general way, not for support of any particular quote, claim, etc. (Right?) An example would be where (say) two lines of sources conflict on some claim, and an editor uses some other sources to evaluate those claims. Or perhaps uses the source for background. But note that citations (even inline) do not have to be associated with specific material in the article. It is quite common to refer to such sources with a "general citation", usually at either the beginning or end, which acknowledges that. E.g.: "Jones and Brown (2001) provide valuable background." Note that such an article-general citation could be source-specific. E.g.: "This interpretation follows the comment of Jones...." In your examples the references are general in both use of the source and application to the article. (Though it might be better if they were specific.) ~ J. Johnson (JJ) (talk) 00:14, 5 January 2013 (UTC)

My understanding of how "general reference" has been used in Wikipedia guidelines is much different than J. Johnson's description. As I understand it, a general reference cites a work which generally supports the information in the article, but there is no indication about which claims in the article are supported by the work. Whether a specific part of the work is cited, or the entire work is cited, is a different matter. Jc3s5h (talk) 00:50, 3 January 2013 (UTC)

There is a problem with Kmhkmh's description. A "References" section may be an alphabetical list of full citations to sources, which are listed directly in that section (that is, there is no <references/> tag. There is often a "Notes" section that does contain a <references/> tag and which associates claims in the text with the works in the "References" section. The "References" section may also contain works that are not mentioned in any note. Those unmentioned works are certainly general references. Some of the works in the "References" might have been used both for claims with footnotes and claims without footnotes, so those works are both note targets and general references. Jc3s5h (talk) 01:07, 3 January 2013 (UTC)

Well yes, the inline citation/footnotes in reference section (if notes is not used) are of course no general references. In that sense general references are entries in the reference section other than footnotes/inline citations.--Kmhkmh (talk) 01:17, 3 January 2013 (UTC)

Sorry, I must mostly disagree. As I said above, the entries in a reference usually are general references (e.g., they don't cite specific locations or passages), but being in that location is not what makes them "general". Also, inline citations can be general references (when, again, they don't cite specific locations). There are some other good points here to address, but no time. Later. ~ J. Johnson (JJ) (talk) 02:21, 4 January 2013 (UTC)

  I would quibble with Jc about "unmentioned works" (not cited in the text or notes) being general references. The application of the work to the article might be in a general way (as I just described in the comment further up), but the reference is general or specific depending on whether the work is referred to in whole or in specific part. In practice this may not be much of a difference because (as I said before) bibliographic entry is general. But it could happen otherwise. ~ J. Johnson (JJ) (talk) 00:33, 5 January 2013 (UTC)

As I said above a g "general reference" is a shorthand for "an entry in the general references section" ie one that is placed into a references section usually in an alphabetic author list. Those that are placed inline and may be displayed through <references/> or {{reflist}} are inline citations but those are not general references. General reference is defined in Wikipedia:Citing sources#Types of citation. Whether the sorted list in references section is used to enhance inline citations, does not change the meaning of what they are.

Often works in the references section that are not coupled to inline citations should either be removed or moved into a "Futhrer reading" section, as they have often been added by editors to the references section, because there was no "Further reading" section rather than because they directly support the text. -- PBS (talk) 01:35, 6 January 2013 (UTC)

PBS and J Johnson are talking about the real world. Wikipedia is not the real world. On the English Wikipedia, the sole meaning of "general reference" is "any citation that is not an inline citation". It does not tell you anything about the ordering of the full citations, the title of the section (==General references== is quite unusual), or the vagueness of the citation. Here, it just means "not an inline citation", meaning not a "method that allows the reader to associate a given bit of material in an article with the specific reliable source(s) that support it". WhatamIdoing (talk) 04:11, 6 January 2013 (UTC)

  I mostly agree with your observations of what is. But I think it is also a valid observation that citation practice at Wikipedia is often poor, and generally confusing. I say this is largely because WP practice and conception has not only diverged from that of the rest of the world (which makes it bewildering for newbies), but become inconsistent in peculiar ways (which makes it difficult even for old salts such as present company). Which is why my viewpoint tends to be of what should be. If there are to be any improvements there will need to be some adjustments to concepts. And it would be preferable that such adjustments be consistent internally, and with the real world.

  "General reference" is a good example of the kinds of ambiguity plaguing us. As I showed above, a reference (citation, whether full or short) can be general in respect of the source used as support, or of the article content being supported. Example: an article might have a statement that some topic is "well studied", supported by citation of several works. That statement is specific content (thus warranting in-line citation) supported by references to several works as a whole (thus general references). Alternately, a reference (whether general or specific in regard of the source) might generally support the article as a whole where (say) an editor feels it provides important background not linked to specific content.

  Part of the problem here is this pernicious notion that content does not need specific (i.e., in-line) citation, that it suffices to say just "it's all in that book". Kmhkmh's example of Gripenberg_Castle shows this: this two paragraph stub relies entirely on just two references (a book and a web page) "generally", meaning no inline citations. Yet it makes makes at least ten specific assertions ("is a wooden manor house", etc.) which would be supported by specific passages in the sources. In this sense "general reference" is just an excuse to not support specific content with specific citation.

  I do not entirely agree with Phil's idea that "general" references should be removed (because I can see a type of general reference I believe is valid). But if we applied that to Gripenberg_Castle there would then be no references. I think a better view would be that content lacking specific citation (i.e., in-line) is subject to removal. If a "general" reference can be the basis for specific references, so much the better.

~ J. Johnson (JJ) (talk) 22:17, 6 January 2013 (UTC)

Well not everybody likes the "one line, one statement equals one inline citation"-approach, which is for many a somewhat formalistic and pointless overkill, which in addition creates visual clutter. Hence those authors used inline citation in more sparse fashion, often paragraph based or in short articles simply list them at the end. The whole point for sources & citations is merely, that checking content against sources is easy enough and is not requiring an unreasonable effort, which depending on the context can be achieved by different formats. The keywords here being "easy enough" and "not unreasonable", there is no requirement for the "easiest". I.e. we should not impose maximally inconvenient formal requirements on our authors, just to achieve the easiest verification option, instead the verification just has to be easy enough.--Kmhkmh (talk) 22:33, 6 January 2013 (UTC)

  I think we agree there is a trade-off in how much effort should go into making verification easier, but I disagree that we are imposing "maximally inconvenient formal requirements on our authors". Many editors scamp citation (because it is too confusing? or merely inconvenient?), but the fact remains: WP:V requires that material be supported by sources; that is part of the writing process. That editors find it inconvenient perhaps gets back to the problems we are trying to deal with. At any rate, as one of a seemingly small group of editors that do verify, I find verification of uncited material a LOT harder than adding a citation (with a page number or such) in the first place.

  To get back to Phil's point: if a "general" reference is removed, then it seems (to me) reasonable that any material thereby unsourced should also be removed. ~ J. Johnson (JJ) (talk) 23:37, 7 January 2013 (UTC)

There is no argument about the need for (specific) sources. The argument was about the format of those specific sources (placemenent of inline citations, general references). In connection with Gripenberg example all one has to do to verify the content is to read the 2 (short) general reference, wich is easy enough, i.e. no unreasonable effort is required.--Kmhkmh (talk) 01:09, 8 January 2013 (UTC)

  I am a little perplexed by your comment, so help me understand this. The first reference at Gripenberg is to a book of some 500 pages. (That it's also in German, not quite "easy enough" for me, let's overlook.) It also has a page reference, which is where the Google Book link takes us, which with my limited German suggests that our interest in this work is in a single passage, of less than a page. In regard of your previous comment I would agree that (generally) a page number is "easy enough", and for a specific reference at that. And that is what we have here: what I call a specific reference, which is not to a work as a whole, but only to a passage in the work. Where this reference is "general" is in applying to all of the points in this article (a condition perhaps better described as "universal").

  I suspect where you balk is at the prospect of citing every assertion in this article with an identical "Larsson et al., 1998, p. 54". (Which a certain bot would then bundle up as a single named-ref.) And I would agree that that would be tiresome. But I consider this a pathological case, resulting from relying entirely on a single source. However representative (unfortunately) this might be of many Wikipedia articles, it is a special case in respect of citation practice generally. I suspect that failing to recognize this difference accounts for some our differences of sentiments in this subject.

  With that in mind, perhaps you would explain what you mean about "format"? ~ J. Johnson (JJ) (talk) 21:10, 9 January 2013 (UTC)

By "format", Kmhkmh means whether the text is connected to the source by using <ref> tags, parenthetical citations, or one of the other less popular forms of WP:Inline citation, or whether it is typed after anasterisk at the bottom of the page.

That's what the ==General references== section is about: do you connect your source to your text, or not? Here's an example that should illustrate the issue:

General reference	Inline citation
The Moon is really big, but the Sun is even bigger.	The Moon is really big, but the Sun is even bigger.
References Expert, Alice. (June 2001) "On the Relative Sizes of the Sun and Moon". Fancy Magazine, p. 78.	References

See the difference? That difference is the only difference that we're discussing in this section. WhatamIdoing (talk) 22:24, 9 January 2013 (UTC)

  Yes, thanks, that clarifies at least where the confusion is. And I think we agree that "Inline citation" includes having the full citation in the text (with or without parentheses), or a short citation ("Harvard style" or not, with or without use of Harv templates).

  I say that the first example is really specific (in both senses I have described above) because it points to a specific part of a source that (presumably) supports a specific assertion in the article text. That in this case the source is not linked to the specific text is, I think, a failing. It still applies to only the specific assertion, not to the article as a whole. This reference therefore is not "general".

  As to whether "do you connect your source to your text": of course yes. That is a requirement of WP:V: to attribute "using an inline citation." ~ J. Johnson (JJ) (talk) 22:02, 10 January 2013 (UTC)

The left example given by WhatamIdoing illustrates a general reference. Alice Expert's article generally supports the Wikipedia article, but we don't know which part of the Wikipedia article is supported by Expert. That is how the term "general reference" has been sued in this guideline for a long time. Due to this history, the definition is too ingrained to change. If J. Johnson doesn't like it, he/she will have to propose a total rewrite that doesn't use the term "general reference" at all.

I very much disagree that it is a requirement to always use an inline citation. When an article contains claims that can easily be looked up in any of 20 textbooks on a well-settled subject, there is no need to provide inline citations. And certainly one could list some of those textbooks as general references. Jc3s5h (talk) 22:30, 10 January 2013 (UTC)

WP:V only requires inline citations for quotes and material this is challenged (or is likely to be). Is there something odd about including references not cited? Gimmetoo (talk) 22:35, 10 January 2013 (UTC)

"Is there something odd about including references not cited?" I don't think so. Sources that helped an editor get a general sense of the subject before writing the article, or that were used to confirm claims that could easily be located in the index of nearly any textbook on the subject, should be included as general references. Jc3s5h (talk) 23:52, 10 January 2013 (UTC)

That was a rhetorical question. I didn't expect an answer :) Gimmetoo (talk) 23:59, 10 January 2013 (UTC)

  But a good question nonetheless. One assertion in this multi-faceted discussion is the possibility of references that are truly general to the article as a whole. But I can see the oddness of a reference that lacks any mention of why it is included. The standard practice for such cases is to include (usually after the first sentence) a note which explains the matter. (E.g.: "For a basic background see ....") So even a truly general reference can be (and probably should be) cited. That the citation (footnote) applies not to specific text, but generally, may seem a little odd, but a general and accepted practice.

  In the example above (presuming that this line is not the entirety of the article, and that p. 78 in Expert 2001 supports this specific assertion), if a citation is not required for this assertion (perhaps per WP:SKYISBLUE), then there would be no need for the reference, and this entire discussion is moot. But in this case, where the reference is to a specific part of the source, and supports specific text: the reference is not "general", but specific. That the connection to the source is missing is beside the point: the reference is still specific. A missing connection does not make the reference general in its application, only ambiguous. ~ J. Johnson (JJ) (talk) 00:23, 12 January 2013 (UTC)

I don't actually care what you think this "should" be called. I'm telling you that this is called a "general reference" on the English Wikipedia. The section in question is about this thing, which happens to be called a "general reference" here.

(Gimmetoo, because of BLP's requirements, there are actually four types of material that require inline citations. They're listed at WP:MINREF. Nothing else is absolutely required to have any sort of citation, although it is both normal and good to exceed the rock-bottom minimum requirements.) WhatamIdoing (talk) 03:53, 15 January 2013 (UTC)

To extend WhatamIdoing's example

General reference	Inline citation	Short inline citation
The Moon is really big, but the Sun is even bigger.	The Moon is really big, but the Sun is even bigger.	The Moon is really big, but the Sun is even bigger.
References–1 Expert, Alice. (June 2001) "On the Relative Sizes of the Sun and Moon". Fancy Magazine, p. 78.	References–2	Notes–3 References–3 Expert, Alice (June 2001), "On the Relative Sizes of the Sun and Moon", Fancy Magazine

The thing in "References–1" and "References–3" is a general reference. The thing in "References–2" and "Notes–3" is an "inline citation" (the first an "inline citation" and the second one a "short [inline] citation"). The wording of Wikipedia:Citing sources#Types of citation that defines the usage of "general reference" makes this clear:

A general reference is a citation to a reliable source that supports content, but is not displayed as an inline citation. General references are usually listed at the end of the article in a References section. They may be found in underdeveloped articles, especially when all article content is supported by a single source. They may also be listed by author alphabetically in a References section in more developed articles as a supplement to inline citations.

It does not include the example in "References–2" because it is "displayed as an inline citation". In "References–1" it is "in [an] underdeveloped article" and in "References–3" it is "as a supplement to inline citation", so in both "References–1" and "References–3", because of its positioning in a References section, it is a general reference. -- PBS (talk) 12:36, 15 January 2013 (UTC)

  I disagree with PBS's nomenclature for the "Short inline citation" column. The short citation and the full citation together form the inline citation for the claim "The Moon is really big, but the Sun is even bigger." Only the "General reference" column contains a general reference. Jc3s5h (talk) 14:12, 15 January 2013 (UTC)

I also disagree. That's not a general reference. It's simply an alternate method of displaying inline citations. Try reading WP:Inline citation. It lists six or seven different methods of providing inline citations. WhatamIdoing (talk) 01:58, 19 January 2013 (UTC)

  Sort of agree, sort of not; which goes to show how complex this stuff is. "Inline citation" is another ill-defined term, but I would say that both of the links above (^[1] and ^[a]) are "inline" (as in "in the line of text"). (As would a short or full citation, in the line of text, with or without parentheses.) I could add additional examples, but they all would feature something in the text — a full (clunky!!) or short citation, or a link to somewhere else. A link typically (but not necessarily) goes to a note (created with <ref> tags), where there is either a full citation ("References-2"), or short cite ("Notes-3") which links to full cite ("References-3"). But what interests us here is only where there is nothing "inline" in the text. Which could be all of these examples if the bracketed links are taken out.

  Where I quibble with the definition of general reference is regarding the scope of the content it supports. The general and standard meaning of "general" is "applicable to the whole", "not confined by specializaton", "not specific or definite", "generic", etc. Our definition ignores that, and so the unfortunate custom has developed of calling "general" such references as are really specific, but lack an inline linkage to a citation. (Consider: remove the bracketed links, and all of the examples of above are "general" in the sense being (mis)used here.) Such a faulty definition does us no good, and puts WP in conflict with the ordinary meaning of "general".

  Slightly different issue: as I said way back, "general" references may tend to occur in a section named "References", but that is incidental; a reference is not general because of its "positioning". But I would reinforce (and expand) a point Jc3s5H makes: the "inline citation" should probably refer to whole apparatus of a link (if used), short cite (if used), and full citation. ~ J. Johnson (JJ) (talk) 00:45, 16 January 2013 (UTC)

I would agree with J. Johnson that a general reference should support a substantial part of the article. A reference that supports only one specific claim, but which is not linked to that claim by an inline citation, is really a citation error rather than a general reference. The problem is that since we don't know which claim is supported, it's hard to tell the difference between a general reference and a citation error unless the person checking the article is quite familiar with the source. Jc3s5h (talk) 02:32, 16 January 2013 (UTC)

Incomplete or broken may be better terms, as "error" suggests a positive misstatement, rather than lack. At any rate, recall what I said before that a (truly) general reference can be specifically cited "inline" (as long as a note explains that it is general), and probably should be. In this view even general references could be incomplete (i.e., uncited). The implication would be that all references should be cited in the text or a note, general references as well as specific. ~ J. Johnson (JJ) (talk) 22:54, 16 January 2013 (UTC)

Certainly a statement that is not a direct quote and not likely to be challenged could be followed by an inline citation to a particular page in a work that also supports many other statements in the article. But it is not required. "The implication would be that all references should be cited in the text or a note, general references as well as specific" has never been a Wikipedia requirement. If you require that every source mentioned in an article must be associated with some specific claim, even though that claim is not a quote and is not likely to be challenged, which claim would you pick to place an inline citation after? Would you require that every single claim that can be supported by the source be cited? That would be totally unworkable. Jc3s5h (talk) 23:08, 16 January 2013 (UTC)

  Your previous comment was about distinguishing between incomplete citations ("citation error", a reference not linked to specific material) and references truly general. My point above was that truly general references can be cited (mentioned in a note). If general references are cited, then there is no problem: incomplete citations are just what they are — incomplete. And this should not be a problem, because truly general references are rather rare.

  I believe the gist of your current comment is whether "every single claim that can be supported by the source be cited". This is a bit ambiguous. I often have claims that can be supported by more than one source. Often I will cite multiple sources, sometimes not; I certainly do not cite a source every time it could be cited. I believe you mean whether every claim should be supported by a source. (Right?) I am inclined to say yes, because that is what I do (well, mostly!), and I find it quite workable. However, the context here is where a source has been referenced, but not linked (cited) inline. (The common use of "general reference".) In such cases I would say that the hard work (of finding the source, and inserting a formatted full reference) is already done, and it is almost trivial to add a specific short cite inline. ~ J. Johnson (JJ) (talk) 21:05, 17 January 2013 (UTC)

First off, it has often been proposed that every claim must be followed by an inline citation, and this proposal has been repeatedly rejected. (See Wikipedia:Perennial proposals#Require inline citations for everything.) I think your comments about workability assume a particular scenario for writing or improving an article, but there are many scenarios where it is not workable. For example, I read a textbook to make sure I really understand a topic, and to get some ideas about how much detail to include, and what order to present ideas in. I then reorganize an article to remove extraneous details and present the remaining ideas in a better order. I find that all the quotes and all the claims that are likely to be challenged already have citations. Indeed, I find that all the claims where it would be helpful to the reader to provide a citation already have citations. So am I going to find some highly obvious claim, for example, in an advanced article on the solar system, that the Earth orbits the Sun, just so that when I add the textbook to the reference list, it will be the target of at least one inline citation? No. Why should I bother to find a claim that doesn't need a citation, then look up in the book where that claim is covered, just so I won't have a general reference? Jc3s5h (talk) 21:17, 17 January 2013 (UTC)

You misunderstand. To cite a reference used generally it is not necessary to find a specific claim to tie it to. Recall what I said before (12 Jan.): simply add a note (something like "For a basic background see...") and cite the source. Such notes are usually inserted at the end of the first sentence; the note explains that it applies to the whole article. I think you will agree that is hardly any bother at all. And such a note also explains why the source is referenced. ~ J. Johnson (JJ) (talk) 22:15, 17 January 2013 (UTC)

I agree it would be helpful to provide an indication of why a general reference was added to an article. If the Reference section is separate from the Notes section, there would be no need for a note and the comment would probably be at the end of the citation so as not to obscure the alphabetical order.

Sure, that is one way of doing it (annotating the reference). I would say that I prefer the annotation in the note, but I have a vague recollection of doing it somewhere just as you say. And of course (I hope this doesn't now confuse anyone), citing or annotating a reference as being used generally (generically) does not preclude it from being cited specifically elsewhere. ~ J. Johnson (JJ) (talk) 00:16, 18 January 2013 (UTC)

Recommended reading

To get back to a point Phil touched (30 Dec.) regarding "general references" as the title of sections that might also be called "recommended reading" or "for further study": such use of "general references" is confusing in itself. I suggest that such use should be avoided, and even deprecated, in favor of the more descriptive terms. ~ J. Johnson (JJ) (talk) 22:52, 18 January 2013 (UTC)

Added bolding to emphasize I am referring to use of "General references" as a section title, not as characterization of the contents of such sections. ~ J. Johnson (JJ) (talk) 22:13, 26 January 2013 (UTC)

I agree; references were used in writing the article; works listed in "For further study" or "Recommended reading" need not have been consulted while writing the article. Some editors might feel that listing a work in the "References" section implies it is recommended and so would not list it again under "Recommended reading"; other editors might see value in listing suitable works in both lists, when applicable. Jc3s5h (talk) 00:53, 19 January 2013 (UTC)

I agree with Jc3s5h: ==Further reading== is not (normally) supposed to include sources you used in writing the article. It turns out that WP:FURTHER (MOS guideline) and Wikipedia:Further reading (an old proposal) also agree with him. This has been settled practice for years. WhatamIdoing (talk) 01:58, 19 January 2013 (UTC)

Whoa, guys, you're going off on a tangent. Whether a work listed as a reference (source) can be included (or not) in some other list (e.g., "Further reading"), is not the point here. I am suggesting only that any such list should not be titled "General references". ~ J. Johnson (JJ) (talk) 23:25, 20 January 2013 (UTC)

There is an old Irish country saying when a person asks for directs and the description is complicated "If I was going there I would not start out from here". The issue of what to call the sections is described in WP:LAYOUT and the editors who lurk there are in favour of recommendation that follow practice and not prescribing a better solution. For better or worse the sections are know as "Notes", "References" and "Further reading" in that order and that is because that was once recommended and has now become the most common layout. The place to discuss changing the name of "Further reading" is at Wikipedia talk:Manual of Style/Layout. --PBS (talk) 12:14, 26 January 2013 (UTC)

I agree with Jc3s5h: ==Further reading== is not (normally) supposed to include sources you used in writing the article. I think that there is a problem here with short term thinking. The problem statement presupposed "you" are still around to know that "you" used an uncited source as a general reference when "you" wrote or expanded the article. I think that in a developed article, books that were read by someone who added text to an article but did not directly cite that book should be placed in "Further reading".

The Nuttall Encyclopædia has many one line definitions and is ideal for putting in a one line stub see for example the entries under A:

• Wood, James, ed. (1907). "A" . The Nuttall Encyclopædia. London and New York: Frederick Warne.

Let us suppose that Nuttall was cited in the References section called "Sources" back in 2003, and let us suppose that that now a decade later it is now 35K article, with three dozen citation, some of which are short inline, some long inline and some general references. The current editors decide to bring it up to the standard of a good article (and that the initial editor is long since retired). Back in 2003 there was not a Wikisource article containing the Nuttall text, but there is now. If not one word of the initial Nuttal text exists in the current article why should Nuttall be listed in the references section and not placed in the external links section using {{Nuttall poster}} as is the usual way to display uncited Wikisource articles? It would look really odd (and not up to good article standards) to keep Nuttall in the References section while all the other encyclopaedia articles on Wikisource (about the article) are in poster format in the external links section.

There is some commercial and/or academic pressure to have works cited in Wikipedia so that it can be mentioned elsewhere (Some years ago I wrote an article called (Genocide definitions) and have been chuffed to see some academics consider it worth while adding their definition to the list). Keeping the list of references to those which are cited in an article helps editors with identifying such nefarious additions (that aid the sale of books rather than helping the reader understand the subject). -- PBS (talk) 13:43, 26 January 2013 (UTC)

Slow down guys! This is a tangent. Phil, I sympathize (and btw, thanks for splitting off this subsection), but the sole point I wanted to clarify in this section is that use of "General references" as a section title — e.g., as "== General references ==" — should be avoided, even deprecated, for the reasons previously discussed. I suggest that such sections should be titled "References", "Recommended reading", "For further study", or similar, as appropriate for the contents and use. I make no claim as to what is or should be included in such sections. ~ J. Johnson (JJ) (talk) 22:13, 26 January 2013 (UTC)

AFAIK "General references" is not recommended in any guideline a section header, what the guidance suggests is either "References" or "Sources" or a couple of others. What is being described as general references here are the things in the section, because in some articles there are no bullet points in the References section instead there is a list of inline citations generated by {{reflist}} -- PBS (talk) 02:09, 27 January 2013 (UTC)

I've never seen ==General references== recommended as a section heading, neither in this guideline, nor at LAYOUT, nor anywhere else. It is sometimes used, but it has never been officially recommended.

What exactly you're supposed to use if you have short citations, full citations for your short ones, explanatory footnotes, and general references all in the same article is a mystery. Nobody has ever come up with a good answer to that question. WhatamIdoing (talk) 06:15, 27 January 2013 (UTC)

Reprise

This discussion has explored deeply into several views of "general references". At this point I would like to summarize several points which I suggest would clarify matters:

1. By standard usage "general" means "applicable to the whole" of something, "not specific", "not particular".
2. A citation (full or short) is "general" in regard of the source when the source itself is referenced as a whole, not any specific passage. Example: "The theory of natural selection is described by Darwin in The Origin of Species."
3. A citation is "general" in regard of the article when the source, or some passage in the source, is applied to the whole article (or possibly a whole section). E.g., the article (or the editor writing the article) may depend on some source for general background on the subject, or the article implicitly follows some particular definition or interpretaton. Note that in such cases it is best to explain that, even if no specific quotations or claims are involved. This is usually done in a note at the start of the article which then cites the source "generally". A citation can be applied "generally" to a specific section, topic, or incident; for examples see Tuchman's The Guns of August.
4. A full reference does not support material in the text unless it (or a corresponding short cite) is cited inline (in the text, or in a note embedded in the text). Short articles that rely principally on a single source should explicitly cite the source in a note that explains it is being used "generally".
5. "General references" as a title for a section or list of works should be avoided due to conflicts with other concepts of the term.

Yes, I grant these go further than the current definition of "general reference" (and frequent usage). But wouldn't these clarifications would greatly reduce the current widespread confusion regarding this term? ~ J. Johnson (JJ) (talk) 23:54, 21 January 2013 (UTC)

Short answer: No.

Longer answer: There is a definition of general reference at the start of the guideline and as you say it is frequently used that way. "General reference" does not mean applied to the whole but is "["an entry in the] general reference[s section]" or as it is described in the guideline "A general reference is a citation to a reliable source that supports content, but is not displayed as an inline citation." There is nothing in that definition that supports your statement, you are also bending the meaning of cation to fit you argument because WP:CITEHOW states that inline citations should include page numbers or chapter for books (unless in unusual cases it is to the book and not to the page). I think you are looking at "general" from the wrong perspective. You are looking at it from the content of the citation, indicated by the formation of the citation, instead of its placement of the citation in the the article. For example I might have this citation the is very specific in pointing to the location in the source (actual citation taken from the Bosnian Genocide article -- for those who need reminding § means "section"):

• ECHR Jorgić v. Germany Judgment, 12 July 2007. § 43 citing the judgment of 19 April 2004 rendered by the Appeals Chamber of the ICTY, IT-98-33-A §§ 25,33

If one uses the definitions given in the section "Types of citation" in the guideline, then whether it is a general reference depends on its location in the Wikipedia article. If it appears inline then it is not a general reference. If it appears in a reference section then it is a general reference (and yes it can be both simultaneously). Whether or not it is a general reference does not depend on how precisely it guides the reader to the information in the source. -- PBS (talk) 11:58, 26 January 2013 (UTC)

As to your comment Short articles that rely principally on a single source should explicitly cite the source in a note that explains it is being used "generally". No! General references on their own are depreciated. If you look at new articles that use general references, stubs or not, many editors will place an {{unreferenced}} tag on the top (I have recently been admonish a user who has been doing this at the rate on average of 2 a minute because they were not attempting to look for citations themselves). If you look at the short history of the article Ernest Roume you will see a recent example of the creation of a one line stub, such tagging, and the removal of the tag with an inline citation.

WhatamIdoing has been fighting a rear guard action on this issue, but in recent years AFAIK each time this has been argued WaId has been in a minority as most editors do not consider that general references (used as defined in this guideline) on their own meet the requirements of the WP:V policy and this guideline (see this example (February 2011) in the archives of this page). -- PBS (talk) 11:58, 26 January 2013 (UTC)

  I am generallly with "most editors" that "general references" (as used here) are generally deficient in regard of WP:V. And I am particularly averse to some editors' claims "general reference" for what I would call in incomplete reference. Yet I am not with sympathy for some of Whatamidoing's claims. I think much of the problem is because of poor definition and interpretation.

  Please note that the point you object to is actually a way of resolving some of these differences. E.g., given that some reference is used "generally" in some way that a flock of inline cites would be tedious to the reader (note that I distinguish this from cases where an editor is simply omitting inline cites), then a single inline citation that explains the reference is being used generally makes that clear. This accommodates cases where a source is used "generally" as I define it [Whatamidoing: would you agree?], and does so in a way that I think would satisfy WP:V and WP:FACR because there is an inline citation. Strictly speaking the inline citation makes the reference not general (as defined here), and so lets the air out of that issue. ~ J. Johnson (JJ) (talk) 23:51, 26 January 2013 (UTC)

Actually, PBS is a bit unclear. It has always been my position that WP:V requires inline citations for three specific classes of material, and that BLP requires inline citations for a fourth. Those four are listed at WP:MINREF, which I wrote a couple of years ago, and whose section heading is ==When you must use inline citations==, so it should be perfectly clear to everyone that I believe that inline citations must be used in many instances. Where those four classes of material are concerned, general references are never acceptable.

But for material not in those four classes, WP:V does not require a citation of any kind. If no citation is required (e.g., WP:BLUE), then you may (and people do) list the sources that you consulted to acquire that information as general references.

Offhand, I can't remember using a general reference myself, ever. I don't really like them, and I believe that unless the article is very short and the source is freely available online, they contribute to sometimes insuperable maintenance challenges. I only recognize that the consensus of the community is that they may be used in some limited situations, that the reality is that they are used by many brand-new editors, and that we need to define the wikijargon here and tell people how to format them. WhatamIdoing (talk) 06:24, 27 January 2013 (UTC)

Can you give an example of the point you make: "If no citation is required (e.g., WP:BLUE), then you may (and people do) list the sources that you consulted to acquire that information as general references"? I don't think people do that. They used to, when WP first started. But I haven't seen it for years, except with stubs, where there are (e.g.) two sentences and one ref, and it's understood that the ref supplied the material.

But where you have a long article, which includes "the sky is blue" (for which no inline ref is required), no one would include a full citation to a source for "the sky is blue" in the References section. No ref required means no ref required. The References section is strictly a list of the references cited within the article, often in shortened form, which means full citations are required in the References section, as this page says. SlimVirgin ^(talk) 18:11, 27 January 2013 (UTC)

I don't think I'll be able to lay my hands on an example without a long search. You might go look at older versions of Elephant, though, which was an FA that I remember seeing with general references listed and with some material not supported with inlines.

I agree that "no ref required" means "no ref required". I do not agree that "no ref required" means "you are not permitted to add a citation anyway". The refs section is not a list of refs "cited within the article" (emphasis added). If that were the case, then general references could not be listed, because—by definition—they are not cited anywhere in the article. The refs section is a list of sources used to write the article. If you use a source, but it is not "cited within the article", then you are permitted but not required to list it with the references. WhatamIdoing (talk) 01:30, 29 January 2013 (UTC)

I concur: "not required" does not mean forbidden.

Examples! See International Code of Signals. The last Reference (Mead, 1934) is used "generally" (in both senses as I use it), but is also not cited in the text, thus qualifying under the current definition. Also look at the first note. "ICS 1969" is cited specifically for the quotation, but is also explicitly cited generally (as I mean it) for other material which is not specifically cited "inline". For some non-Wikipedia examples see Tuchman, The Guns of August. ~ J. Johnson (JJ) (talk) 00:28, 30 January 2013 (UTC)

Conclusion?

The above is too much to read. Is it okay if I go ahead and restore the edit that PBS reverted? I can't see the benefit of including the extra words.

Current

Proposed

A general reference is a citation to a reliable source that supports content, but is not linked to any particular piece of material in the article through an inline citation. General references are usually listed at the end of the article in a "References" section, and are usually sorted by the last name of the author or the editor. The appearance of a general reference sections is the same as those given above, in the sections on short citations and parenthetical references. A general references section may also be included in an article that will eventually use inline citations throughout if such citations have not yet been given for all the information in the article. In underdeveloped articles, a general references section may exist even though no inline citations at all have yet been added, especially when all article content is supported by a single source. The disadvantage of using general references alone is that text–source integrity is lost, unless the article is very short.

A general reference is a citation to a reliable source that supports content but is not displayed as an inline citation. General references are usually found in under-developed or very short articles, or articles that rely on a single source. They are listed at the end in a "References" section, sorted by the last name of the author or editor. The disadvantage of using general references is that, unless the article is short or there is only one source, text–source integrity is lost.

SlimVirgin ^(talk) 23:58, 26 January 2013 (UTC)

I would add another paragraph to the proposal:

When a general reference is placed in a good article it is often helpful to add a note explaining how the source was used, for example, as a model for the structure of the article.

Jc3s5h (talk) 00:13, 27 January 2013 (UTC)

Sorry, I don't think I understand what that's saying. Can you rephrase, and by good article do you mean a GA or something else? SlimVirgin ^(talk) 00:20, 27 January 2013 (UTC)

I don't necessarily mean an article that has gone through Wikipedia's good article process, I just mean a good article. A reference can be used in writing an article for purposes other than supporting specific statements. Such uses might be

• deciding on the general structure of the article
• deciding which topics were important enough to include
• deciding what notation to use in a math or physics article, even for equations based on other sources.

We should not word the guideline to suggest that articles that use sources in such ways are underdeveloped, or imply that such use should not be acknowledged. Jc3s5h (talk) 01:17, 27 January 2013 (UTC)

Hi, general references sections are used in under-developed articles. I haven't seen one for years that uses a general references section unless it's very short, or one of those articles that just copies a public-domain text. Do you have an example of another kind of use? SlimVirgin ^(talk) 17:56, 27 January 2013 (UTC)

I see no evidence you have read my posting; you just keep repeating yourself. Also, I disagree with your claim in the "Reprise" thread that " The References section is strictly a list of the references cited within the article...." According to an acceptable Wikipedia style, the Chicago Manual of Style page 687,

1. Full bibliography. A full bibliography includes all the works cited, whether in text or in notes, other than personal communications. Some particularly relevant works the author has consulted may also be listed, even if not mentioned in the text.... [cross-reference omitted, italics added]

Jc3s5h (talk) 18:42, 27 January 2013 (UTC)

As for the requested example, Gregorian calendar lists a 1999 work by Duncan in the "References" section but Duncan is not mentioned in any footnote. "Gregorian calendar" is a pretty good article; I would nominate it as a good article if religious controversy didn't doom its chances. Jc3s5h (talk) 18:49, 27 January 2013 (UTC)

Jc: editions of CMS vary. Could you give us a section number? ~ J. Johnson (JJ) (talk) 00:31, 30 January 2013 (UTC)

Section 14.59 of the 16th edition (2010).

SV, Yes I think it is a good idea to modify the "General references" section. But I think that what you are currently is producing a sawn-off version of the definition given in the section "Types of citation". For example a list of general references is common in articles other than undeveloped articles so "... are usually found in under-developed or very short articles" is not true, because many (most?) large articles (and all those that use short citations) have general references in a References section (see for example Battle of Stalingrad).

Hence I think the wording in the "Types of citation" section is appropriate: "They may be found in underdeveloped articles, especially when all article content is supported by a single source. They may also be listed by author alphabetically in a References section in more developed articles as a supplement to inline citations."

But if all we are going to do in the guideline section "General references" is regurgitate the paragraph in the "Types of citation" section we may as well just remove the section "General references" from the guideline lock stock and barrel. -- PBS (talk) 02:39, 27 January 2013 (UTC)

The combination of short citations and full citations is not general references. If there is any way for the reader to figure out which sentence or paragraph that source supports, then it is not a general reference. See the definition: A general reference is a citation to a reliable source that supports content, but is not linked to any particular piece of material in the article.

SV, your bit about "not displayed inline" is not going to work, because it's going to perpetuate the claim that Irish phonology (FA using parenthetical citations} uses three or four dozen general references. General references are "not connected" or "not linked" to content. The mechanics of how they are displayed is irrelevant to their nature. WhatamIdoing (talk) 06:32, 27 January 2013 (UTC)

The current version already says that. I'm proposing only a copy edit so it's not so wordy; if people want to change the content after that, that's fine. An inline citation includes parenthetical refs, so I don't think I follow your point. I've looked at Irish phonology and it doesn't use a general references section; it uses inline citations, and it lists the full citations in a References section, which is normal practice. That's an entirely separate thing from a general references section, which is just a list of texts used to create the article, without saying which text was used for which point. This page used to make that distinction clearer, but the text was extended and the difference muddied. SlimVirgin ^(talk) 18:03, 27 January 2013 (UTC)

No, the current version doesn't say that. The current version says "but is not linked to any particular piece of material in the article through an inline citation." Your proposal says "but is not displayed as an inline citation."

I grant that your version uses fewer words, but we need clarity, not brevity—and especially not the kind of brevity that gets people so confused that they put errors like "Examples of general reference sections are given above, in the sections on short citations and parenthetical references" into this guideline. WhatamIdoing (talk) 01:44, 29 January 2013 (UTC)

Could you link to a version of the guideline that used to make the distinction clearer?

"The good thing about standards is that there are so many to choose from". A general reference is a citation to a reliable source that supports content, but is not displayed as an inline citation. (from the section at the start of the guideline that contains the definition of a lot of phrases used in the guideline). The mechanics where they are displayed is relevant to their the definition used to describe what a general reference is. Taking the Battle of Stalingrad example there is a line in the Bibliography section (Biography being an alternative name for References) -- "Beevor, Antony (1998). Stalingrad. Viking, London"? I would describe it as a general reference, what would you describe it as?-- PBS (talk) 09:42, 27 January 2013 (UTC)

The entry in the "Bibliograpy" section of the "Battle of Stalingrad" article is the full citation that supports footnotes 12 and 23. I also note that footnote 66 is defective because it fails to specify a year, and so could refer to either of the works by Beevor. The bibliography entry is part of several inline citations. Jc3s5h (talk) 17:15, 27 January 2013 (UTC)

Why do the general references need to sorted in an alphabetical order?--Kmhkmh (talk) 04:44, 27 January 2013 (UTC)

They don't. You might choose to sort them by date, for example. They usually are sorted alphabetically by authors' last names, though. WhatamIdoing (talk) 06:33, 27 January 2013 (UTC)

That's right, they don't, that's just an incidental correlation that has nothing to do with the nature or definition of "general reference". Likewise for their location ("at the end") within a section, or the name of the section. Including such incidental observations only confuses matters as editors get the impression they are actual requirements, and it skews our concept of term. ~ J. Johnson (JJ) (talk) 21:07, 27 January 2013 (UTC)

That was basically my point, i. e. the reason for my question. Readers it may understand as a (new) requirement that has been sneaked in there.--Kmhkmh (talk) 21:30, 27 January 2013 (UTC)

It has been there for month probably years. Have you any examples of where (i) they are not sorted into an ordered list, (ii) not sorted by author (where available) -- PBS (talk) 08:17, 28 January 2013 (UTC)

How about anywhere that reference list is created by a {{reflist}} or WP:LDR? (For an ugly example see Climate_change_denial#References.) But that is really beside the point, which is: whether a reference is "general" is not dependent on whether the list or section it is in ordered (sorted) or not. ~ J. Johnson (JJ) (talk) 22:31, 28 January 2013 (UTC)

Actually no. The old/current formulation was "usually sorted" and the suggest new is "sorted". The former clearly just indicates the most common occurrence/practice but the latter is easily "misread" as requirement. I know that is is not the point we are trying to clarify here, but that's all the more reason not to introduce such (fishy) changes on the side.--Kmhkmh (talk) 23:40, 28 January 2013 (UTC)

OK now I understand! Would the addition of "usually" in front of the phrase "sorted by the last name of the author or editor" assuage you concern? -- PBS (talk) 11:41, 29 January 2013 (UTC)

Yes it would. The formulation with "usually" avoids unfortunate misunderstandings and with it hardly anybody would consider the alphabetic sorting a requirement.--Kmhkmh (talk) 00:34, 31 January 2013 (UTC)

Lets turn this on it head. The first definition of a general reference "is a citation to a reliable source that supports content, but is not displayed as an inline citation." That definition includes all of what Jc3s5h describes as "the full citation that supports footnotes ...". There are two problems with using Jc3s5h's description:

• It is inviting pedants to insist that the citation is not a "full citation" because it does not meet the requirements of "What information to include" (no page numbers), and so they will add {{page number}} to them to insist that they are added.
• (the more important one) It is also confusing because the last big edit to this page was to try to get away from the idea that the stuff in the References section were citations. This is because calling the lines in the References section citation encourages inexperienced editors then think that "general references (as currently defined)" with or without "short citations" are "full citations" and all that is required (as is often the case in secondary school education).

So if "the full citation that supports footnotes ..." are not "general references", and they are not "full citations", what name can be used to describe them? -- PBS (talk) 08:17, 28 January 2013 (UTC)

  First of all, citations and references, "full" or otherwise, do not support footnotes; footnotes (endnotes) are just a place to put stuff. Second, in regard of references ("citations") "full" means with regard to the bibliographical details (consult your favorite manual of style).

  I think what Phil means is: if (as currently defined here) "general" means not specifically cited in the text ("inline"), by what name do we distinguish the full references that are cited?

  I suggested before (16 Jan.) that where text that should be linked to ("supported" by) a reference, but is not, "incomplete" or "broken" might apply. But that is really about the lack of a linkage (inline cite). In regard of a reference I say that such distinction is pointless: the reference (full or otherwise) does not change. A "full" reference (citation) is always a full citation, specifically cite inline, or not. ~ J. Johnson (JJ) (talk) 23:50, 28 January 2013 (UTC)

Let's not, because what you call "the first definition" is simply wrong. WhatamIdoing (talk) 01:46, 29 January 2013 (UTC)

WaId. I accept that some (including you) think the first definition is wrong, but if it is wrong (and I do not think it is), what would you call an item in the list of works that appear as bullet points in a reference section of an article using short citations (because at the moment the first definition defines them as general references)? If that is to be changed then we need a short unambiguous name that can be used as a label on other talk pages to describe them, in the way that "general references" are currently used. -- PBS (talk) 11:32, 29 January 2013 (UTC)

  Waid, you're confused. "The first definition" is Phil's term, which he used just above, and is essentially the current definition. And perhaps I am confused to, as I thought you supported that (e.g., your example of 9 Jan., above). Could you clarify?

  Phil: Why do we need different names for references that are cited and references that are not cited? ~ J. Johnson (JJ) (talk) 00:00, 30 January 2013 (UTC)

Given the initial definition I don't think we do. For example 1,000 of articles have {{EB1911}} templates in their "References section"s with no inline citations (which they ought to have). I think it confusing to change the description of a bullet point to {{EB1911}} just because an editor added inline citations to an article, it is simpler to refer {{Cite EB1911}} it as a "general reference" whether or not it is enhanced with inline citations that appear in the body of the text and in a "Notes section", but without any visible change to the bullet point or to the rest of the References section.

However my reading of the above opinions such as the explanation given by Jc3s5h 17:15, 27 January 2013 and WaId's repeated statements indicates that they do not agree that works listed in a "references section" that "supplement" inline citations are general references but are something else because they form part of the inline citation. I am not convinced that making a distinction between bullet points that directly supplement inline citations and bullet points that do not, will not generate more confusion than it solves. But I may be wrong! and I am inviting those who think it desirable to make that distinction to come up with a simple short descriptor for "something else" that will not cause more confusion than exists currently.

-- PBS (talk) 15:25, 30 January 2013 (UTC)

  That's my point, that it is confusing (and pointless) to change the description of a full reference (I presume that is what you mean by "bullet point": the bulleted entry in a reference list) just because inline citations have been added. Regardless what the others think, perhaps you and I are in agreement on this point?

  As to whether such a distinction would generate more confusion than it solves: well, isn't that what we have currently? That is, the current definition does distinguish whether a reference is cited inline (or not), and it does generate more confusion (all of the above!) than it "solves". Right? ~ J. Johnson (JJ) (talk) 00:05, 31 January 2013 (UTC)

• We call this "third thing" a full citation. A full citation is one of the two parts of an inline citation system that uses shortened citations.
• We do not use this definition now. WP:CITE#General references says "A general reference is a citation to a reliable source that supports content, but is not linked to any particular piece of material in the article through an inline citation." The definition you were working from is SlimVirgin's proposed change, not the current one. WhatamIdoing (talk) 16:17, 2 February 2013 (UTC)

What do you mean by 'this "third thing"'?

The current definition (as you quoted it) does invoke (as I said) this distinction of "is not linked ... through an inline citation." Which is where the question arises of what to call a citation that IS so linked. (My point being that we do not need special terms for that distinction.) ~ J. Johnson (JJ) (talk) 22:36, 2 February 2013 (UTC)

By "third thing", I mean the full citations to sources that are cited inline with shortened citations of any type. We don't need a special name for it, because it already has a name. If you use {{sfn}} or parens, you have an inline citation ( like "(Smith 2001:25)") in the text and a full citation (like "Smith, Alice (2001) My Book. Oxbridge Press.") at the end. WhatamIdoing (talk) 16:23, 9 February 2013 (UTC)

I propose that whenever we have need of referring to full references that have not been cited in the text/notes ("inline") it is sufficient to say "uncited references". And the complement of that is adequately expressed as "cited references". (Note: this makes no inference about whether a reference should have an inline cite pointing to it.) Is this sufficient? Alternately, does anyone here feel there is any need to have a special term for either of these distinctions? ~ J. Johnson (JJ) (talk) 23:29, 1 February 2013 (UTC)

I oppose this. I oppose inventing a new term at all, because we already have term for this particular thing, and we shouldn't try to impose new jargon on the community. I oppose this particular suggestion, because it is an oxymoron ("Citing" does not mean "inline citing". You'd be citing a source, and then labeling it as uncited. The actual uncited sources as the ones that you secretly took material from and did not list anywhere at all) and is going to make the average editor believe that there is something seriously wrong ("uncited" is always bad). WhatamIdoing (talk) 16:17, 2 February 2013 (UTC)

W: It seems to me that you have totally missed what I am trying to do. I am not inventing a new term, I am trying to clarify and even simplify existing jargon. I am not proposing that "cited references" and "uncited references" should be adopted as specifically defined jargon. What I am saying is that where we have a reference for a source — and by this I mean the full reference (or full citation as it is sometimes called) that describes the source — and we want to distinguish whether that reference is linked or not linked "to any particular piece of material in the article through an inline citation", then it is sufficient to simply describe it: "that reference is linked [or not linked] to the text". Alternately: "that reference is cited in the text" (or not). Instead of using a seemingly ordinary term ("general reference") in a peculiar and confusing way that is specific to Wikipedia.

It appears that you took "citing" to mean the linkage from the full reference to the source. Please note that I did qualify my use of "cite" with "in the text/notes (inline)". Nor is there anything about this proposed usage that suggests that '"uncited" is always bad". ~ J. Johnson (JJ) (talk) 22:42, 2 February 2013 (UTC)

Perhaps that is resolved? So I will ask again: do we really need a special term for full references which lack inline cites linking to them? Or, when we need to refer to such cases, is it adequate to describe such cases (such as "full references not cited in the text or notes")? ~ J. Johnson (JJ) (talk) 00:46, 6 February 2013 (UTC)

I go away for a week and come back to check something... and find that the bickering continues. It's very simple: these are general references, and these are not. --Redrose64 (talk) 13:29, 6 February 2013 (UTC)

Not "bickering", but patient, thorough exploration of deeply held but somewhat discordant views which folks are often rather sensitive about. Despite the occasional testiness I think the apparent issues are not so deep but that clarifying some misunderstandings and allowing for some differences in point of view can lead to a general consensus.

However, it is necessary for folks to relax any death-grips on certain points. I respectfully suggest this includes the view "it's very simple". Actually, at a very basic level I would agree with that: the current definition of "general reference" is indeed very simple: a reference "... not linked ... through an inline citation." But that is also not simple in that it is a special definition peculiar to Wikipedia, not in accord with general usage, and not useful in that it leads to these other issues of inclusion in "References" versus "General references", what to call a reference that is linked via a citation, etc.

Citation on Wikipedia is more confusing than it needs to be. My hope is that if we can straighten out some of these kinks there will be less bickering overall. ~ J. Johnson (JJ) (talk) 22:10, 6 February 2013 (UTC)

It is never simple to change the community's terminology. Even if we did (and I point out that it has taken three years just to convince most editors that secondary is not a synonym for independent, even though every reliable source supports this claim), your initial proposal ("I propose that whenever we have need of referring to full references that have not been cited in the text/notes ("inline") it is sufficient to say "uncited references") doesn't work.

"Unlinked" might work. "Non-inline-cited" might work. "Orphaned" or "unconnected" might work.

But "uncited" doesn't. "Uncited" means "there is no citation at all". It does not mean "there is no inline citation to connect this source back to the text".

To expand on Redrose's examples, these are general references, and these are not, and this has uncited sources. WhatamIdoing (talk) 16:36, 9 February 2013 (UTC)

Please be more attentive. I have already explained that I am not urging adoption of a specific term, and that my use of "uncited references" was only an example. (Which I did qualify as referring only to use of an inline citation.) I also provided an alternate example using "unlinked", which is exactly what you have just suggested, along with some other alternatives, all of which I find adequate. My point is that, referring to "orphaned" (or unlinked or unconnected) references, ordinary language is sufficient; we do not need a specific term ("general reference"), defined with a specific meaning peculiar to Wikipedia. ~ J. Johnson (JJ) (talk) 22:28, 12 February 2013 (UTC)

I believe I have established that we do not need to give the term "general reference" a special meaning peculiar to Wikipedia, and resolved various confusions. Does anyone else have any questions or concerns about this? If not, I would like to proceed with a redefinition of "general reference" that removes the special meaning, and thereby remove the basis of a lot of confusion. ~ J. Johnson (JJ) (talk) 23:28, 14 February 2013 (UTC)

The community has a practical need for some way to refer to bibliographic citations that are unconnected to any particular material on the page. We have used the term general reference to identify these things for years. How exactly is the community supposed to (1) be able to discuss these things and (2) not use any terms for these things?

The worst possible solution is introducing a new term, which will inevitably have all the same problems as the old one (i.e., some editor will not understand, because his schoolteacher preferred a different term), plus the problem of no one actually using the new term except you, because everyone's used to the old one and nobody in the future will be able to make sense of the discussions on about four thousand talk pages.

The second worst possible solution is demanding that people replace a convenient term with a lengthy, cumbersome descriptive phrase. In practice, such a demand will simply fail. WP:Nobody reads the directions, and they certainly will ignore any direction that tells them to quit saying "general references" and instead type out something like "This is just a citation to a reliable source that supports content, but is not linked to any particular piece of material in the article through an inline citation" instead of "This is just a WP:General reference". WhatamIdoing (talk) 05:21, 15 February 2013 (UTC)

  Let's take this from the top. The current definition of "general references" is a problem because 1) it conflicts with the ordinary and standard meaning of "general" (confusing everyone who has had an English teacher, or merely uses a dictionary), and because 2) it is based on a pointless distinction that implies (among others) that a "general reference" does become something else when it is cited.

  You complain that I would replace a "convenient term" with "a lengthy, cumbersome descriptive phrase". But the convenience of your example is only in hiding the cumbersome part behind a wikilink. The lengthy descriptions have come in because (lacking a pre-packaged definition hidden behind a wikilink) we have to resolve differing conceptions of the very terms we are talking about (like the meaning of "citing").

  You complain about "introducing a new term", but that is a misstatement of what I am trying to do. (Is it really necessary to remind an experienced veteran such as your self of WP:TPG?) I am suggesting that ordinary terms — even such as you have suggested — can suffice. If we need to clarify what a term means (and I prefer we stick with ordinary and standard usages) or specify what terms should be used, fine, let's do that. But introducing a new term with a meaning peculiar to WP ("jargon") is the exact opposite of what I am trying to achieve.

  I don't know why you are so implacably opposed to my suggestion, other than it appears to touch on some matter of deep concern to you. I would reassure you even on that, but your caricaturizations are so absurd I hardly know where to start. I believe I have addressed your concerns reasonably (and patiently), and if that is not sufficient then it is time to move on.

~ J. Johnson (JJ) (talk) 20:50, 16 February 2013 (UTC)

Hypothetical example

Please give me an example of how you want to change the community's language. Pretend I'm in a discussion with another editor. He wants to take a section currently labeled ==Bibliography== (a section heading I dislike) and rename it ==Further reading==. I want to tell him, "we shouldn't do that; those are WP:General references, not WP:Further reading items."

You don't want me to use the term general references any more. So how exactly would you re-write that sentence? What words would you put in my mouth so we can get rid of a long-standing phrase that you hate? WhatamIdoing (talk) 02:06, 19 February 2013 (UTC)

Although not directed to me, I would change the sentence to say "we shouldn't do that; those are not WP:Further reading items because they were used while writing the article." As for what alternative titles could be used for the section, it would depend on how the works were used while writing the article. Jc3s5h (talk) 02:30, 19 February 2013 (UTC)

  First of all, it's not the term I "hate", but its bad definition and misuse. In your hypothetical situation my response would be as thus: As the [postulated] references in the ==Bibliography== section "support content" the section should be retitled ==References== (or ==Sources== or some such); ==Further reading== is for sources that go beyond the material covered in the article, as described in WP:Further reading.

  That is short enough, and I think clear enough. If there are any objections I expect it will be because I have not saluted the distinction of "not linked to any particular piece of material". But why is that needed? The references we are referring to are the full references (or full citations) with the complete bibliographic description of the source. Neither the source nor the reference describing it are changed in any way when invoked by an inline citation.

~ J. Johnson (JJ) (talk) 19:46, 20 February 2013 (UTC)

I do not agree that tripling the length of the sentence is "short", or even "short enough". WhatamIdoing (talk) 23:22, 21 February 2013 (UTC)

  You are being ridiculous. In the first place your hypothetical editor might validly say "But these references do have inline citations, they are not 'general references'." (I presumed this "Bibliography" was an ordinary reference list, as you did not specify otherwise.) Or he will look at other "Further reading" lists and notice that none of them have inline citations, so therefore they ARE "general references", and want to know why they are allowed. And then you will get bogged down in a long and frustrating discussion just like we are having now. All because of an absolutely pointless distinction.

  You are also starting to piss me off. It is ridiculous to say that one explanation is better than another solely because it is shorter. Especially (as I have stated before) when a major part of your explanation is hidden behind a wikilink. If you just want a pissing contest on who can come up with the shortest explanation, fine, I can come up with an explanation in just two words. Can you beat that? Or will you again change the parameters with some new objection? Or perhaps I should be asking: what is your problem? ~ J. Johnson (JJ) (talk) 22:10, 23 February 2013 (UTC)

No, I'm being practical, not ridiculous. People simply won't type long explanations if a short term will do the job. In fact, they often won't even type out the term itself, if they think they can get away with an abbreviation. Think about it: how often do you see people typing out "Wikipedia:Identifying reliable sources" in discussion? Does anyone use WP:IRS? It's almost always "You need WP:Reliable sources for that" or "You need a WP:RS". I'm sure that many people would prefer to type "That's a WP:GENREF" to save themselves a few keystrokes.

My example features true general references, as the term is used here on the English Wikipedia. I wouldn't say that they were general references if they were actually part of an inline citation system, after all. If he reads the links I provided in my short reply, then he'd know exactly what the difference is: General references were consulted by the people writing the article, and Further reading items were not. He will learn, for example, that Further reading entries do not always "go beyond the material covered in the article". They may be listed because of their historical value or simplicity.

What I want is nothing: I want no change. I want the term that is in active use to continue to be correctly defined on this page. You are free to avoid this bit if wikijargon yourself, but I don't want you to remove it and thus make it difficult for people to figure out what the other editors are talking about when they toss this term around. WhatamIdoing (talk) 01:20, 24 February 2013 (UTC)

So you asked how I would explain a certain point to another editor, and after I did so you come up with this criterion that it is not short enough. Which is ridiculous. Your rationale that people prefer to type abbreviations than full titles is beside the point, as you confuse a reference or pointer to an explanation with the explanation itself. (If you want to play by those rules, fine, and, as I said before, I can "explain" in just two words. How high on the wall can you piss?) What I am trying to work out is the explanation that would be behind a wikilink.

To get back to your hypothetical case: please tell me how the references in a "Further reading" list (per WP:Further reading), lacking inline citations, are not "general references" by the current definition. ~ J. Johnson (JJ) (talk) 22:15, 25 February 2013 (UTC)

GENREFs are always used to build the article. FURTHER items are normally not. WhatamIdoing (talk) 22:27, 25 February 2013 (UTC)

• Agree with what WhatamIdoing just said. --Redrose64 (talk) 23:55, 25 February 2013 (UTC)

  Additionally, this distinction has been explained on the page he linked for about two years now. It's the Wikipedia:Further reading#Relation to reference sections section. WhatamIdoing (talk) 02:53, 26 February 2013 (UTC)

But you did not include that link in your "short explanation" above. And it goes beyond the current definiton of "general reference"; your explanation is incomplete. (Too short!) In particular, how are we know whether a given source was read ("used") by an editor "to build article content"? If one of your general references lacks any notation that it was "actually used" in or supports material in the article, then how can any other editor determine whether a given source should be referenced as general or further reading? ~ J. Johnson (JJ) (talk) 23:57, 26 February 2013 (UTC) [Belated]

It is up to the editor to cite sources that he/she used while writing the article, on pain of being considered a plagiarist. In some cases, this can take the form of just putting the source in a list of sources with a heading such as "Bibliography" or "References". The presence of the source in such a list, in combination with the absence of any mention of the source in any inline citation, lets subsequent editors know it is a general reference. Jc3s5h (talk) 23:48, 26 February 2013 (UTC)

Using a source (perhaps for evaluting other sources) does not necessarily mean the source supports content; to cite the sources I have consulted ("used") without actually taking any material would double some of my artcles. But possibly you mean "used" in the sense of "took from", yes? ~ J. Johnson (JJ) (talk) 00:09, 27 February 2013 (UTC)

By used, I mean the editor wrote the article differently than he/she would have if he/she hadn't consulted the source. It could be taking a fact that can be looked up in any textbook on the subject, so doesn't really need an inline citation. But it could also be the order of presentation, or deciding what facts to include and which to omit. If the article includes formulas, it might be the names of the variables, if several different choices are common in the field. Jc3s5h (talk) 00:14, 27 February 2013 (UTC)

  So you are distinguishing between 1) where a source is used directly to support content (such as quotations, etc.), and 2) where an editor "used" (consulted) sources for background, to evaluate other sources, and as you have listed. But perhaps this does not matter, if we take the qualification from WP:Further reading#Relation to reference sections ("not used by editors to build the current article content") to mean used for content, or used otherwise. But this goes beyond the definition of "general reference", so Whatamidoing's explanation above is incomplete.

  To get back to my prior point: how is anyone else to know this key dataum, that a source not cited inline has not been "used to build the article"? You suggest "the absence of any mention of the source in any line citation"... but that's what I am talking about, sources not cited inline. Given such an uncited reference, how are we to know if some editor "used" it (for content, or otherwise)? ~ J. Johnson (JJ) (talk) 22:24, 28 February 2013 (UTC)

That depends on the context. In a large article with many sources we do not, however in small article with a very specific sources at the end we do (examples Partial_linear_space, Dinostratus'_theorem). More importantly it doesn't really matter whether the original editor used a given source or not. What matters instead is that the given source, be it an inline citation, general reference or whatever, supports the article's content, that is the content it can be associated with.

Another to keep in mind regarding the "how do we (or does a 2nd editor) know" is, that we never know for sure, unless we actually do read all the given sources ourselves.--Kmhkmh (talk) 00:50, 1 March 2013 (UTC)

  It does matter, to the extent that it is important (allegedly) for identifying "general references". (And the added requirement for "further references".) Yet, as you say, "we never know for sure". For the most part we can only rely on (as Jc said) "the presence of the source in such a list". Which is to say, if an editor puts a reference ("not cited inline") in a "Reference" list, we have to rely on that editor's inherent characterization (absent explicit indications otherwise) that it does support content. (And in "Further reading" the implication is that it neither supports nor was used.) So another (and shorter!!) response to W's hypothetical is: insufficent information to assess any change.

  Which illustrates the main point that I have been trying to communicate: despite the nicety with which WP:GENREF and other pronoucements have been applied, they do not really help us. ~ J. Johnson (JJ) (talk) 22:44, 1 March 2013 (UTC)

Ok you lost me there completely and right no i have no idea anymore what you're trying to communicate.--Kmhkmh (talk) 23:17, 1 March 2013 (UTC)

Ooops! Okay, a brief reprise. I have been claiming that the current definition of "general reference" is useless. In a hypothetical example regarding how to explain that some unspecified "general references" may not be included in a "Further reading" section, Whatamidoing asked for alternative language that does not use the term "general references". At this point I have asked how we are to distinguish a "further reference" from a "general reference". The key distinction seems to be whether an editor has "used" the source referred to, either directly in support of content, or in some other manner (as suggested by Jc3s5h). On this subpoint I agree with you that we (generally) can not know. Wherefore I conclude that the exacting manner in which editors try to apply these fine distinctions is useless, in that we lack information essential to the distinction. These distinctions do not help us. (Okay?) ~ J. Johnson (JJ) (talk) 23:06, 2 March 2013 (UTC)

So what term do you want to use for the type of sourcing of which i posted 2 examples above?--Kmhkmh (talk) 01:05, 3 March 2013 (UTC)

I believe that J. Johnson proposes that no term be given, and that every time you need to refer to this concept, you type out one or more sentences to explain it.

Figuring out whether a given source supports content can be impossible in some instances. However, in most instances, it's not so hard. You find the edit that added the source, and if it's by someone who didn't edit the article, then it's probably FURTHER (or maybe REFSPAM). If it was added by someone who also expanded the article, then it's probably GENREF. If you really can't figure it out, then you assume that the person adding the source correctly labeled it. Sometimes your assumption will be wrong, but that's the only practical solution, if it's unclear. WhatamIdoing (talk) 04:16, 3 March 2013 (UTC)

Yes. More particularly, I claim that the distinction made by the definition is not particularly useful, is not needful, and variant from the common and standard meaning of "general". If making that distinction is needful (such as in a discussion like this) it can be described. But in general it is not needed, not even in the hypothetical example. ~ J. Johnson (JJ) (talk) 22:08, 3 March 2013 (UTC)

Even as we go sliding off to another facet, I would point out that hypothetical explanation W provided could be simplified by omitting the reference to "general references". E.g., it could be cast as: The sources referenced in a "Further reading" section should comply with the guidelines in WP:Further reading. Referring to WP:General references is thus not necessary. ~ J. Johnson (JJ) (talk) 00:19, 6 March 2013 (UTC)

What is a citation?

WID your argument that general references are acceptable on their own is not the position that most editors support (I provide this example (February 2011) in the archives of this page higher up in this conversation). I frequently break up "Biography" sections into two "References" and "Further reading" and place articles one or the other depending on whether the sources is cited with the use of a short citation or long inline citations (also it is not infrequent to find cited sources listed in a Bibliography section or an "External links" section solely on whether it has a link to an online version -- a misreading of how to cite sources). The reasons for doing this are two fold. The first is that over time many articles that have little to no regular editorial oversight accumulate fluff with all but the kitchen sink being thrown into a general references section.

For example I have recently been through about 1500 articles that cite: Kidd's Debrett's Peerage and Baronetage (1990 edition) in the References section, (less than half a dozen were inline and not one had a page number). In nearly ever case that citation was accompanied either by a general citation to the home page of {{rayment}} or to Darryl Lundy's home page: The Peerage or both (see for example Charles Cocks, 1st Baron Somers). In practical terms it means that most of these articles have no citations to reliable sources (in the general references sections). For the moment I have just tagged them with in-line tags and added no-citations, but in the long run something will have to be done to fix the entries particularly those that are BLPs.

The second reason for splitting up list of sources into "References" and "Further reading" it helps other editors to identify people adding bogus entries in a references section for advertising purposes.

So I think that using "general reference" to mean "is a citation to a reliable source that supports content, but is not displayed as an inline citation" is more useful than changing its meaning.

The problem with using "full citation" is it implies that the bullet pointed references in the references section should include page numbers and other information that may instead be split between short and long citations.

I have reverted the changes that were recently made as there is no evidence that there is any sort of consensus for the change WhatamIdoing made.

-- PBS (talk) 13:18, 3 March 2013 (UTC)

I find it increasingly to see what you're arguing about. In fact I see no meaningful difference between your version and that of WhatamIdoing. They are 2 slightly different formulations of essentially stating the same thing. Nor can I see how that slight difference of formulation has any effect on your example of the Debrett's Peerage and Baronetage usage.--Kmhkmh (talk) 14:23, 3 March 2013 (UTC)

PBS, you've misused the tag here. {{Unref}} is for pages with no reliable sources named whatsoever. I recommend reading the instructions at the template's doc page. The template that you actually want is {{No footnotes}}, which is the one that says there's a list in the article, but no inline citations. I've fixed that page for you, but you need to go fix any others that you made that error on. WhatamIdoing (talk) 01:34, 4 March 2013 (UTC)

I am not misusing the tag (despite what is said on the documentation page of the template unreferenced). The wording of {{No footnotes}} is broken, as it implies that external links and "related [sic] reading" are sources for an article -- they are not (see the talk page of the template for more details on this). {{unreferenced}} simply says "This article does not cite any references or sources", which is an accurate statement. This is one of the reasons why I think the changes you made on 15 February are not helpful as it puts back into this guideline the discredited idea that general references are citations, which is not what WP:V says. -- PBS (talk) 10:34, 4 March 2013 (UTC)

I'm sorry to see that we're struggling with such basic material. The term citation is defined in the very first sentence of this guideline:

A citation, or reference, is a line of text that uniquely identifies a source:

Ritter, Ron. The Oxford Style Manual. Oxford University Press, 2002, p. 1.

Therefore, ==Further reading== and ==External links== are actually filled with citations—just not (unless someone has screwed up) citations to reliable sources that were actually used to write the article—as are any sections containing general references or inline citations. Our definition does not require that bibliographic citations be only inline.

WP:V makes no claim that citations are not citations. WP:V only says that non-inline citations are insufficient for the few things absolutely required to have an inline citation. WhatamIdoing (talk) 21:36, 4 March 2013 (UTC)

If an article includes no references to sources, then certainly that is a matter of {{Unreferenced}} (and its redirect, {{Unref}}). The problem is where there are references, but not inline citations (or other notes) to the references showing where or how they are applied. This is a problem because without perusing the source (and lacking other information) we have no way of knowing whether these are "general references", and thereby deemed okay, or are bogus. And therein lies the fundamental problem with this notion of "general references": the acceptance of references without inline citations. I have previously shown that sources used in a truly general manner can still be cited in a note. So there appear to be no reasons why any reference cannot be cited. And an excellent reason why they should be: to show how they are used. This "general reference" distinction is useful only for identifying references that should be cited inline. And actually harmful in condoning inclusion of sources without showing how they are used. ~ J. Johnson (JJ) (talk) 23:01, 4 March 2013 (UTC)

Slapping ref tags around the same citation and sticking it at the end of a sentence also doesn't tell you whether or not the cited source validly supports the content, either. A lot of refspam is completely bogus. Formatting cannot tell you whether a source is either reliable or properly used.

The problem in this discussion, though, is that PBS apparently thinks that citation is an exact synonym for inline citation, which is not true. According to PBS, Atomic semantics, a three-sentence stub complete with one bibliographic citation listed under the heading of ==References==, is just as unsourced as if that references section were blank. WhatamIdoing (talk) 23:44, 4 March 2013 (UTC)

Yes, you are quite right, the mere presence of a citation (no matter how nicely it is formatted) does not guarantee it validly supports the content. For that we rely on the good faith and competence of editor, just as we do regarding that there is such a source, it is reliable, s/he consulted it and interpreted it correctly, etc. Nonetheless, associating the citation at the particular material ("inline") it allegedly supports (and specifying the page/location in the source of the support) vastly reduces the effort to verify that support, because we can assume it supports that material; we don't have to see if it supports any material in the article.

However, not doing the whole inline citation for short stub articles seems to be an accepted practice, however sloppy. And you will have to admit that a three sentence article hardly leaves much confusion as to where it applies. Indeed, I would say that (although it is included under "References") the note at Atomic semantics is a general reference in the proper sense of the word: applicable to the whole.

BTW, note my comment in the previous subsection that referencing "General references" is not necessary in explaining "Further reading". ~ J. Johnson (JJ) (talk) 00:27, 6 March 2013 (UTC)

harvnb vs. sfn

I am a new editor. Another editor showed me how to use short citations with harvnb in, which I use in an article How to Create a Mind where I cite the same book over and over (maybe too much). It seems to work fine. But now I read "sfn" is newer? In Help:Shortened_footnotes the "render as" for both look identical. Are there any differences between the two? Is there any reason to swap over from harvnb to sfn? Or use one or the other in the future? More generally, is the citation style in my article appropriate? Thanks. Silas Ropac (talk) 13:26, 16 February 2013 (UTC)

The {{harvnb}} template is normally enclosed in <ref>...</ref> tags, e.g. <ref>{{harvnb|Smith|2008|p=12}}</ref>, whereas the {{sfn}} template constructs its own <ref>...</ref> tags, so you only need put {{sfn|Smith|2008|p=12}}. On the final page they are visually identical.

But besides being shorter to type, {{sfn}} scores when you cite the same page number (or range) twice - it detects the duplication and merges the references (see e.g. refs 1, 4, 5, 6 of NBR 224 and 420 Classes), whereas with {{harvnb}} you need to name the <ref> tag in order to perform the merge. For example, {{sfn|Smith|2008|p=12}} is exactly equivalent to <ref name="FOOTNOTESmith200812">{{harvnb|Smith|2008|p=12}}.</ref> --Redrose64 (talk) 13:50, 16 February 2013 (UTC)

Okay thanks, now I do see Help:Shortened_footnotes shows that {{sfn}} doesn't need <ref>...</ref> tags, I just didn't notice. It doesn't talk about combining page numbers that I see, but that will be nice because I do have some dups that look a bit silly right now. So yeah I think I will switch, and will use sfn going forward, good to know. Silas Ropac (talk) 14:13, 16 February 2013 (UTC)

@Redrose64 it is not quite the same. {{sfn}} places a full stop at the end of the citation. With the {{harvnb}} example you have given it does not. I have known pedants who do not want other editors to include new citations using {{sfn}} to argue that the addition of a full stop is a change in style. Personally I think that such arguments are less than helpful. -- PBS (talk) 10:46, 3 March 2013 (UTC)

You can remove or change the postscript using |ps=. See the {{sfn}} documentation. --— Gadget850 (Ed) ^talk 13:02, 3 March 2013 (UTC)

I happen to detest these automatic merges of footnotes, and "named refs" in general, as it means the citation gets buried somewhere else in the article, which can be a real hassle when you have to change something. So I recommend using Harv from the outset. But aside from that, I would say that if you have no other reason to change from one style to the other: don't. There can be reasons for changing, but none, I think, so compelling as to concern a new editor. ~ J. Johnson (JJ) (talk) 21:07, 16 February 2013 (UTC)

How is {{sfn}} "buried somewhere else in the article"? It's always right where you expect it. Look at NBR 224 and 420 Classes; find any ref marker, like , and locate those portions of text where that occurs (in this case it's "the first inside-cylinder 4-4-0 to run in Great Britain;" and "predated the G&SWR 6 Class"). Now edit those sections and you will see that the ref in both cases is immediately after the phrase, not "buried somewhere else" - it is given as {{sfn|Boddy|Brown|Fry|Hennigan|1968|p=5}}. If we later discovered that the information about either one of these claims was not on page 5 but on a different page, we simply need to amend that one instance of |p=5 to whatever the correct page was, and that action would not affect the one which had the correct page number. --Redrose64 (talk) 22:35, 16 February 2013 (UTC)

Yes, yes, you are quite right, and I apologize for being unclear and implying that {{sfn}} suffers from the sins of "named refs". The main point, which I believe we both can agree on, is that if an editor is satisfied with either {{harv}} or {{sfn}} there little reason to switch. ~ J. Johnson (JJ) (talk) 20:26, 17 February 2013 (UTC)

When you say "satisfied with either {{harv}} or {{sfn}} there little reason to switch". Do you mean using {{harv}} like this: {{harv|smith|2013|p=100}} --which produces an inline citation-- or <ref>{{harv|smith|2013|p=100}}</ref> --which produces a citation within a ref...tag footnote? -- PBS (talk) 10:46, 3 March 2013 (UTC)

Does it matter? And if are going to discuss fine points of where a citation is placed should we not do this in a new section? ~ J. Johnson (JJ) (talk) 18:50, 6 March 2013 (UTC)

There is a scenario where you would want to use both. You might use sfn for most of your footnotes, but have one or two that need some explanation as well as the short citation, in which case you would use harvnb. For example, you might want to both cite the source, and provide some additional details that would interrupt the flow of the article. Jc3s5h (talk) 19:05, 6 March 2013 (UTC)

Indeed. I am not at all shy about using {{harv}} with and without <ref> tags. But isn't this off-topic for this section? ~ J. Johnson (JJ) (talk) 21:54, 7 March 2013 (UTC)

Kindle for PC

Does anyone know how you cite books that have been downloaded to kindle for PC? The page numbering system is different and it uses ASIN rather than ISBN. Cheers.  Kitchen Roll  (Exchange words) 12:34, 24 February 2013 (UTC)

Resolved; I've found the wiki page on it, Template:ASIN.  Kitchen Roll  (Exchange words) 13:29, 24 February 2013 (UTC)

Most of the Citation Style 1 templates - {{cite book}} included - have the |asin= parameter already provided, so there should be no need to use {{ASIN}} separately. If there are actual page numbers, use |page= or |pages= as usual, but if there is some other means of indicating the position within the work, use |at= --Redrose64 (talk) 16:09, 24 February 2013 (UTC)

People keep asking about this, so we need to add the basic instructions (e.g., "if there aren't page numbers, then try the chapter title") to WP:Page numbers. WhatamIdoing (talk) 22:33, 25 February 2013 (UTC)

Thank you  Kitchen Roll  (Exchange words) 13:30, 27 February 2013 (UTC)

I've added these instructions. I offered examples of chapter numbers, section headings, standard divisions in plays and scholarly systems for standardized works. I think this probably gives people enough ideas, but feel free to improve. WhatamIdoing (talk) 22:25, 7 March 2013 (UTC)

Citing straight line distances on maps

I wish to use some straight line/great circle (as the crow flies) distance figures in an article, and have obtained them using the Google Maps distance calculator tool (part of the Maps Labs suite). Unfortunately, however, it seems that distances generated through this tool cannot be directly referenced by a link to Google Maps. I'm wondering what is the best way to get around this. Could I take a screenshot of the Google map with guidance as to which points are being measured so that others can go to the source and verify for themselves? Or is there some better way to cite straight line/great circle distances as measured on a map? Thanks  — Amakuru (talk) 12:55, 4 March 2013 (UTC)

Of course you can cite maps; otherwise, {{Cite map}} wouldn't exist. It's not necessary to provide a URL. WhatamIdoing (talk) 21:39, 4 March 2013 (UTC)

It's not the map he wants to cite, but the distance calculated from a map. Curious problem. The only problem with doing a calculation is if it is not obvious. I had a similar issue with an azimuth, and settled for putting in a note explaining the basis of the calculation and which on-line calculator I used. The idea is that there is enough information that someone else, using the same data and process, will get a similar result. ~ J. Johnson (JJ) (talk) 21:52, 4 March 2013 (UTC)

That was sort of my thought on the matter. I mean technically one could take a map and a ruler and measure off the distance straight off. That's original research maybe, but of a straightforward nature that anyone could reproduce. Similarly the reference could be "measure from this point to that point" which could be done using instructions for Google maps distance. Do you guys think that would satisfy the Wikipedia rules regarding citations? Thanks  — Amakuru (talk) 10:04, 8 March 2013 (UTC)

Technically you can run into all sorts of problems trying to do it yourself. (E.g., which projection does the map use? Are you measuring a great circle or rhumb line?) So I recommend using some standard calculator. If someone complains they get a different result (different calculator?) then you might have to dig into it. But unless and until there is some issue I would say just specify in a note that you plugged into this calculator these two coordinates and got this result. ~ J. Johnson (JJ) (talk) 22:12, 8 March 2013 (UTC)

In creating the article 1945 Japan–Washington flight, I used this tool which will give the distance between any two airports and between US zip codes. If this is about your userspace draft relevant to Rwanda then I don't know if it will help. Binksternet (talk) 23:05, 8 March 2013 (UTC)

Full citations

PBS has reverted several changes because he disagrees with the English Wikipedia's definition of WP:General references and apparently the normal definition of full citation. Full citation is used in multiple sources, including CMOS (e.g., 14.14 and elsewhere). It is defined in the Kaplan style guide like this:

"There are two basic types of citations. One is called a full citation, which appears on the references page, in the bibliography section, or a works cited page.... A full citation is just that—it provides complete information about a source so that a reader can look up the publication."

Full citation is the normal academic term for a bibliographic entry that contains all the information needed to identify the source. Full citation contrasts directly with shortened citation. The inline vs non-inline nature is irrelevant: if you have a full/complete/unabridged description of the source, then you have a full citation. Although any ==General references== section ought to be composed entirely of full citations, it is not the same thing, and using the term general references throughout has both been a source of confusion and also makes the guideline self-contradictory.

Does anyone else object to reverting PBS's change? WhatamIdoing (talk) 01:52, 5 March 2013 (UTC)

I did not make a change I reverted a change that you made!

A full citation would include page numbers and other things to identify the relevant text. That is not what usually goes into a general references section. Further we clearly make a distinction on Wikipedia about what is an acceptable citation method and your proposed change is a retrograde step because new editors who are unfamiliar with WP:PROVEIT are likely to believe that works placed in a general references section are acceptable as citations for Wikipedia articles when they are not. -- PBS (talk) 13:52, 6 March 2013 (UTC)

PBS, you made up "works placed in a general references section are acceptable as citations for Wikipedia articles when they are not." That is not the policy. The relevant policy from WP:V is "All quotations and any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation that directly supports the material." There are printed style guides that specifically allow including works that are not linked to any particular part of the citing work in a bibliography, so the allowance of general references in Wikipedia is not aberrant. Also, the version of WP:CITE favored by whatamidoing correctly expresses the long-standing agreement of what a general citation is (for Wikipedia purposes), and you have not demonstrated any change in the consensus. Jc3s5h (talk) 14:10, 6 March 2013 (UTC)

You write in the edit history PBS's version does not agree with long-accepted meaning of "general reference" in Wikipedia. but it is not my version I am reverting to the current consensus version. See here for a version at the start of this year and here for a version from the start of 2012 and here at the start of 2011 in all of them it is stated "A general reference is a citation to a reliable source that supports content, but is not displayed as an inline citation". So it is not I who need to show a change in consensus (as I am not changing the wording) it is those who wish to change the wording. -- PBS (talk) 15:17, 6 March 2013 (UTC)

• I strongly disagree with PBS's interpretation. Also, PBS said that the reversion was back to the last version by Vegaswikian. However, it appears that their was no such version as Vegaswikian did not edit it in the last three years that I could see. (Most recent Vegaswikian edit seems to be 22 June 2008 where it says Articles can be supported with references in two ways: the provision of general references – books or other sources that support a significant amount of the material in the article – ...) It appears instead that PBS reverted the article back to almost the last version by PBS. The location (placement) of the bibliographic information about a general reference does not affect its nature. --Bejnar (talk) 15:44, 6 March 2013 (UTC)

• • It was a version by Themeparkgc not (Vegaswikian) my mistake (See below for a fuller explanation). --PBS (talk) 09:03, 7 March 2013 (UTC)

• The 2011 version mentioned above by PBS does not confuse a general reference as something that could be referred to by a short footnote, although that confusion does occur in the 2012 version he mentioned.

The sentence "A short citation identifies the place in a source where specific information can be found, but without giving full details of the source – these will have been provided in a general reference" was introduced by Kotniski at 08:38, 9 September 2011, and in my view, that edit introduced the non-consensus idea that a general reference is general because it refers to the entire source rather than specific pages in the source, while the consensus view is that the quality that makes a reference general is that it is not linked to any particular part of the Wikipedia article.

Looking at a September 2011 version of the talk page, it seems that the participants were kind of saying they agreed with each other without really understanding each other's point of view. But I don't think we can use "general reference" to include a citation that is referred to by a short citation, because we need a phrase to describe a citation that is not linked to any particular part of the Wikipedia article. That is an important concept to be able to describe, because it is usually not a good idea, but is occasionally acceptable. I also believe the description of a general reference in the January 2011 version refelects the way the term has usually been used in the guideline and the talk page, and the September 2011 talk page discussion does not reflect the overall consensus.

Looking at other style guides, whenever I see descriptions of kinds of use that must be acknowledged to avoid plagiarism, it always seems to require that a specific passage in the article being written must have an inline citation to an appropriate part of the source. So general references, whether to an entire large work, or a page within a large work, are allowed by printed style guides, but they do not serve to prevent plagiarism. Similarly, an inline citation to a large work without giving a page or other means to locate the relevant part of the work does not serve to prevent plagiarism.

So a way forward might to be to do away with the concept of a general reference and just say there is no distinction between general references and further reading. A separate "Further reading" section may be provided, or the further reading may be interspersed with the "References" (a.k.a. "Bibliography"). The presence of full citations (with or without pages) that are not linked to a particular passage by an inline citation does not serve to prevent plagiarism. Of course, a few articles may not contain any quotes, paraphrases, or facts that are not well-known, and thus have no concerns with plagiarism. A stub may be so short that its are effectively inline citations, even they are not in that form. Jc3s5h (talk) 17:11, 6 March 2013 (UTC)

I my memory serves me well, Kotniski made his edit with full consultation on the talk page (see the archives). I disagree with the removal of the concept of general references from this guideline (whatever it is called) because they exist in 100,000 of articles. I think it is a very bad idea to encourage the combination of "References" and "Further reading" sections. -- PBS (talk) 09:03, 7 March 2013 (UTC)

• I have written a revision in my sandbox that tries to clear up the inconsistency and unnecessary distinctions. Sections of the guideline I would leave unchanged are omitted. My approach was:

• Treat full citation as all the information needed to find the relevant part of a source, whether that information is located all in one bibliography entry, or divided amongst one or more short footnotes and one bibliography entry.
• Highlight that citations not only support claims, but also prevent plagiarism.
• Since general references don't prevent plagiarism, tone down the distinction between general references and further reading.

Some might prefer to use the term reference list entry instead of bibliography entry; I don't have a strong preference either way. Jc3s5h (talk) 18:29, 6 March 2013 (UTC)

Overall, I don't think that we want to tone down the distinction between GENREFs and FURTHERs.

One of the oddities of the discussion is the insistence on page numbers. Atomic semantics is a good example: it has one full citation as a general reference. That citation contains page numbers. Does it truly need the page numbers? I don't think so: the source is an article in Distributed Computing (journal). The entire article has been cited. If you pulled the correct volume of the periodical, you could easily determine the page numbers. So here we have page numbers, but they're not actually helpful, and we have people saying that the absence of page numbers (or other details to identify precisely subsections of the source) makes the citation be something other than a full citation. Could you find this source without page numbers? Of course. The presence or absence of page numbers isn't a useful, valid, or verifiable distinction. Yes, it's best to include them if they're relevant, but they're not the defining feature of a full citation. WhatamIdoing (talk) 20:20, 7 March 2013 (UTC)

I think this is bit of misunderstanding/misconception, not page numbers as such matter but a sufficient specificity. Giving a page number is just the most common or simplest form to provide enough specificity. It reasonable to expect of some checking a source to browse/read a few pages rather than just a particular line or paragraph, since in doubt you have to do that anyhow to have some proper context information to understand the line or paragraph in question. Hence as specificity it can be considered good enough to narrow it down to few pages. Now most/many journal or newspaper articles only contain pages anyway and the article within the newspaper or journal can be found quickly through a table of content, index or an online/archive link, it may often be sufficient to simple state the title without specific page numbers. Similarly for a book it might be sufficient to state a chapter, provided that's short enough. However it not reasonable to expect reader to read a double digit page number or even a whole book to check some information, hence it those cases more specificity is required, which usually achieved by providing page number or a small range of page number. The bottom line is, providing page numbers is easiest (and most common) way, that always provides enough specificity.--Kmhkmh (talk) 21:08, 7 March 2013 (UTC)

I made a mistake. I now see the confusion my edit summary made. Here is the edit diff with my comment was "Reverted changes back to the last version by Vegaswikian. It is clear from the conversation on the talk page that there is no consensus for this change. I have let the insure ensure change in place." The version I reverted to was in fact Themeparkgc's revesion as of 06:06, 10 February 2013 (retaining insure ensure change). This was the edit immanently before WhatamIdoing edit as of 05:24, 15 February 2013 diff. Now that is clear I am reverting to the same stable version. While we discuss an proposed changes.-- PBS (talk) 08:42, 7 March 2013 (UTC)

With respect to PBS, Kmhkmh, and Whatamidoing: We need to be clearer about page numbers. In many cases, such as articles in journals, the source is incorporated into a larger work, and page numbers indicate the location of the source within the larger work. In this regard inclusion of page numbers in the full reference is entirley valid.

That usage should be distinguished from the use of page numbers for specification of a particular location or passage within the source ("to identify the relevant text", as Phil said). Where a source is cited only once and the full reference is in a note the specification be included in the same note. (The specification is not thereby part of the full reference, it is only sharing the same space in the note.) Where short cites are used the specification goes with the short cite.

Phil said: "A full citation would include page numbers and other things to identify the relevant text." I would say that a full reference — in the sense we have been using it here, as might be seen in a References list — is complete when it points to (describes) the source, and is not any fuller for identifying any contained material. However, from a previous comment I wonder if by "full" Phil means the complete linkage of short cite (including specification) and full reference. I agree we need all the parts, but wonder if "complete citation" would be a better term. ~ J. Johnson (JJ) (talk) 22:48, 7 March 2013 (UTC)

It is always valid to include page numbers (in the full reference), but not necessarily always needed. However using page numbers you are always on the safe side.--Kmhkmh (talk) 01:04, 8 March 2013 (UTC)

Sorry, no, not always valid to include page numbers in the full reference. It depends on what they are used for; the full reference would include page numbers that locate the source within a large source, but not page numbers that point to specific text within the source. ~ J. Johnson (JJ) (talk) 02:30, 8 March 2013 (UTC)

Which means it is nevertheless always valid to provide a page number. Again the purpose of the page number is provide enough specificity to locate the content in question easily, that applies to page numbers within the source itself or to page to locate the source within a larger equally. And again in the case of small journal article or a small book chapter it is valid but not strictly necessary, as a table of content allows you to locate the source with the larger source easily and hence the article or chapter name is already specific enough for a location.--Kmhkmh (talk) 05:04, 8 March 2013 (UTC)

That assertion deserves a {{Citation needed}} tag at minimum. If you're only using one sentence out of a 500-page book, then it's perfectly fine for your full citation to include that the page number for the relevant sentence. (Besides, if a full citation to a book can't include a page number, then what would you call a citation that includes a page number? A "really, really full citation"?) WhatamIdoing (talk) 03:31, 8 March 2013 (UTC)

Exactly if we use full citation only for the bibliographic reference of book, we would need a separate name for the citation providing enough specificity to actually locate the (sourcing) content in question. Which makes little sense to me. Furthermore a shorted citation (like author name, page number) would in some sense always provide more specificity than the full citation, which makes little sense either.

Essentially that problems stems from 2 different sourcing scenarios. "full citation" (including a page numer) within a footnote or shortened citation (including a page number) within a footnote and an associated "full citation" (without page number) in the reference section. You might argue in the second case it is not valid to provide a page number again, but I really don't see that either, it would be somewhat uncommon of sorts but certainly not "invalid".--Kmhkmh (talk) 05:20, 8 March 2013 (UTC)

It is precisely the problem with citations being split over two lines (with a short citation "the head") and the rest in a general references section "the tail" which is why calling a "general reference" "a citation to a reliable source that supports content" is less than helpful. I think that User:Jc3s5h/sandbox is on the right lines. My major objection to that draft is that there needs to be a clear distinction between a "references" section and a "further reading" section. The use of "Bibliography" should be avoided in this guideline for reasons given in WP:LAYOUT, because not only is "Bibliography" use as a term for a section containing the subjects own works, it confuses some editors into thinking that a section called "Bibliography" can only contain books, (not journals or web pages). -- PBS (talk) 11:22, 8 March 2013 (UTC)

Although "bibliography", literally meaning "list of books, has taken on a more general usage, it is subject to the problems you mention. Perhaps we all agree it is therefore not a suitable replacement for "References" (or similar). ~ J. Johnson (JJ) (talk) 22:30, 8 March 2013 (UTC)

Philip, please stop restoring your changes. [6] You're mixing up several different issues, as usual. Short cites aren't offered as general references. A general references section isn't linked to short or long cites, hence the term "general" references section. SlimVirgin ^(talk) 17:11, 8 March 2013 (UTC)

SV I think you are confused. I am not restoring changes, I am restoring the last stable version. -- PBS (talk) 17:53, 9 March 2013 (UTC)

You're restoring a version that people have objected to, and which is wrong. Please stop reverting to it. SlimVirgin ^(talk) 18:06, 9 March 2013 (UTC)

Why are you reverting to a version which introduces changes, when these changes are under discussion and to which objections have been raised? BRD not BRDR! -- PBS (talk) 18:15, 9 March 2013 (UTC)

I am reverting your changes, Philip, because they are wrong. You have misunderstood what a general reference is, just as you've misunderstood that full citations in References sections don't include page numbers. This is obvious, because a text might refer to the same citation multiple times, referencing different page numbers each time. Are you going to add them all to the References section? It concerns me that an editor who seems not to understand citation systems keeps editing this page. SlimVirgin ^(talk) 18:20, 9 March 2013 (UTC)

I am not making changes. I am reverting to the last stable version 06:06, 10 February 2013‎ by Themeparkgc. But reverting my revert you are reintroducing changes made by WhatamIdoing at 05:24, 15 February 2013‎ diff, while change is being discussed on the talk page. -- PBS (talk) 18:52, 9 March 2013 (UTC)

At to you comment on my talk page: "And short cites are not used together with general references; that's not what is meant by a "general reference" Well that depends on the definition of what a general references is that is what is under discussion. For more than two years this guideline has defined "A general reference [as] a citation to a reliable source that supports content, but is not displayed as an inline citation. ... They may also be listed by author alphabetically in a References section in more developed articles as a supplement to inline citations." -- PBS (talk) 18:52, 9 March 2013 (UTC)

That's correct. A general reference is NOT displayed as an inline citation; it is NOT linked to a cite in the text. But you are changing that to say that it IS linked, e.g. "The short citations and general references may be linked so that the reader can click on the short note ..." Perhaps you're not reading the version you keep reverting to. It contradicts the definition we offer. SlimVirgin ^(talk) 19:12, 9 March 2013 (UTC)

SV I am not changing anything I am reverting to the last stable version. The definition the I quoted is the one that has been in place for over two year. You have changed that definition with your revert please look at the diff on that sentence. It changes from that which has been here for two years: "A general reference is a citation to a reliable source that supports content, but is not displayed as an inline citation."; to: "A general reference is a citation to a reliable source that supports content, but is not linked to any particular piece of material in the article through an inline citation." As to the second sentence, if there is a short citations contains a {{harvnb}} template that is linked to a {{Citation}} template in the references section. What would you call the bullet pointed entry that is created by the {{Citation}} template? (Using the definition above that has been in place in this guideline for two years it is a general reference). -- PBS (talk) 19:39, 9 March 2013 (UTC)

Philip, please. Whether the text you are reverting to is five minutes or five years old, it is wrong. Someone added it and it wasn't noticed. It is a mistake. So people are reverting your restoration of it, and are objecting on talk. Therefore, you have to persuade us that the version you prefer is better. Sorry, I don't understand your questions. Rather than asking questions, could you explain why your version is preferable? SlimVirgin ^(talk)

Explain what part of my question was not clear and I will try to clarify it for you. -- PBS (talk) 20:35, 9 March 2013 (UTC)

Could you please explain what is better about your version, or drop it? I don't want to play question and answer with you. Just tell us why your version is better. SlimVirgin ^(talk) 21:10, 9 March 2013 (UTC)

Citation system components

I have compiled a table of what words some citation manuals use to discuss the components of a citation system. It doesn't look like we could ever find a set of words that would be compatible with all the pre-existing usage out there. We may very well have to set out our own definitions within the guideline.

If we did, the definitions would only apply to explaining citations; the titles of citation-related sections in articles would still be determined by the citation system chosen for the particular article. Jc3s5h (talk) 18:05, 8 March 2013 (UTC)

The first section of the guideline define the terms as used in the English Wikipedia, and that's precisely why we have to define our terms ourselves. WhatamIdoing (talk) 20:57, 8 March 2013 (UTC)

Yes, the "Types of citation" section defines some terms, but it doesn't explicitly state whether it is defining terms just for use in the guideline, or if it is trying to explain the terminology generally used in style manuals. The advantage to making definitions just for the guideline is we don't have to worry if our usage is different from some other publication. The disadvantage is people don't always read a guideline from beginning to end, and so might not be aware of a specialized meaning as they read some subsection. Jc3s5h (talk) 21:09, 8 March 2013 (UTC)

Yes, the table is a good idea, and a good start (thank you), though I think it needs revision and expansion. The key part is exactly as W says, defining the term (or concept). And we should be concerned about consistency with usages outside of this guideline, because people do carry these concepts over, and non-standard usage will only cause confusion. Once we sort that out the rest will follow fairly easily. ~ J. Johnson (JJ) (talk) 22:49, 8 March 2013 (UTC)

The words we use don't matter (editors can choose whatever titles they want for these sections, so long as they're clear), but the concepts are fairly uniform:

• a section for long or short cites linked in the text, with page numbers as appropriate, and for commentary (I call this Notes)
• sometimes a separate section for commentary (I can't remember what people usually call this)
• a section for full citations if short cites have been used in the text (I call this References)
• a general bibliography section containing items of interest (I call this Further reading)

SlimVirgin ^(talk) 18:20, 9 March 2013 (UTC)

Other style guides are pretty consistent about how each one refers to these components, although their terminology is somewhat inconsistent from one guide to another. So there is some merit in our guide being consistent as well. Consistency within our guideline should not extend into the articles; the articles can use any reasonable headings.

Some guides allow the bibliography and the further reading to be combined. Jc3s5h (talk) 23:46, 9 March 2013 (UTC)

If by bibliography you mean "works cited," yes. I used to be keen on keeping further reading (FR) and references separate; that is, anything cited in the text didn't go in FR. But increasingly I'm loosening that up. If the FR section contains a list of books about X, it seems absurd to leave out the main book about X, just because it's in the notes or references section. So nowadays I repeat that kind of thing in the FR section if it would look odd to leave it out. SlimVirgin ^(talk) 16:06, 10 March 2013 (UTC)

The "separate section for commentary" is usually called explanatory footnotes, to differentiate them from WP:Footnotes.

I'd be perfectly happy to have an introductory sentence at the top of WP:CITE#Types of citation that says "These are the definitions of the terms as used on the English Wikipedia". WhatamIdoing (talk) 20:10, 14 March 2013 (UTC)

Good idea, that would be helpful. SlimVirgin ^(talk) 20:14, 14 March 2013 (UTC)

This isn't the first time this has come up. Different help pages use different terms. User:Gadget850/Footnote glossary. --— Gadget850 (Ed) ^talk 20:26, 14 March 2013 (UTC)

Your citation comments removed?

Hi Jc3s5h If you removed your input from the Citation Talk page, so be it. I thought it was spot on. Centamia (talk) 09:15, 16 March 2013 (UTC)

Sorry, but I need a precise page name if I am to understand your comment. What is the "Citation Talk" page? Jc3s5h (talk) 18:01, 16 March 2013 (UTC)

Page numbers in the References section

Philip, just as a matter of interest, why are you moving Google page links from footnotes to the References section? [7] The References section is for the full citation, not for the page numbers. SlimVirgin ^(talk) 17:20, 8 March 2013 (UTC)

If its a full citation then it includes page numbers. I personally think it better to keep the google links out of the body of the text. Note that is not the same as moving the page numbers down into the references section. -- PBS (talk) 17:53, 9 March 2013 (UTC)

Page numbers are added to citations in footnotes (or in brackets), not to the References section. SlimVirgin ^(talk) 18:09, 9 March 2013 (UTC)

Page numbers

A number of questions raised above regarding page numbers and "full citation" can be resolved with a clearer concept of some fundamentals. Rather than go about it piecemeal I'd like to provide a whole explication here. Please bear with me.

First, please note that in the current context page numbers are used in two different ways. It is most important to distinguish these two ways.

The first way is in the full reference, which is the "entry" or description of the source that contains the "full" bibliographic details of the source. It will not have page numbers if the pagination of the source is entirely within the source, such as with a book. (Hold your horses! I'll get to "specification" in a moment.) It should have a page range if the source is found in a larger work — such as an article in a journal — with its own pagination. This is what goes into the |pp= parameter in {{citation}} or {{cite}}. The purpose of these page numbers is to show the location of the source within the containing work. But self-contained works (such as books) do not include a page range in the full reference.

The second way page numbers are used is to identify within the source the specific relevant text cited in support of some point (etc.). Note the different levels: the previous usage locates the source within a larger work, this usage locates the relevant text (etc.) within the source. This is what I call the specification. Note that while it is commonly a page number (sometimes a range), it can also be a section number, line number, or chapter and verse. This is what goes into the |p= or |pp= (alternately, |loc=) parameter of a {{harv}} or {{sfn}} template. It should not go into the similarly named parameter of the {{citation}} or {{cite}} templates.

Something else essential but seemingly overlooked: if more than one part of a source is cited, and these parts are in different locations, the specifications will be different. The specification therefore goes with the text being supported. If short cites ("Smith, 2001") are being used then the specification follows the short cite; it does not go with the full reference into a References list.

Whatamidoing says "it's perfectly fine for your full citation to include the page number for the relevant sentence." Yes, but only as a part of the citation, not as the full reference. If the full reference is in a note (typical of most articles, espcially if the source is cited only once) it may be followed by the specification of the relevant passage, and the whole (short or full reference, plus specification) can be termed the complete citation. They cohabit in the same note, but that does not make one part of the other.

Some of you (e.g., Phil and Kmhkmh) seem to be using "full citation" both in the sense I use "full reference", but also in the sense of all the pieces (full reference, short cite, and specification) used to link material to its source, what I would call the complete citation. This is confusing; it would be better to distinguish these different usages.

I reiterate what what I have said before: a reference is full when it carries complete information about the source. It does not get "fuller" specifying the page numbers of each passage being cited, or for that matter the phase of the moon or next week's winning Lotto number, because those are irrelevant in describing the source. Whether a reference has page numbers depends on whether that helps to find or identify the source. Page numbers used in a specification identify text (etc.) within a source.

My thanks to everyone who patiently reads through this. ~ J. Johnson (JJ) (talk) 23:43, 9 March 2013 (UTC)

This may seem old-fashioned in the age of the Internet, but in the past it was common to go to one's local library and put in a request for a paper copy of a journal article from another library that held that journal. Knowing the page numbers would facilitate the work of the distant librarian who didn't want to read the article, just copy it. Jc3s5h (talk) 23:52, 9 March 2013 (UTC)

That's a separate issue, though. It's fine to include a page range, in the References section, of an article in a journal; it helps people to find it, and it tells you how long it is, which can be useful in deciding whether to read it. What's at issue here is PBS adding page numbers, to the References section, that are used to support specific points in the text. Those are never part of a full citation in the References or Works cited section. SlimVirgin ^(talk) 00:00, 10 March 2013 (UTC)

Yes. But it should be understood that your "full citation in the References" — equivalent to my "full reference" — is more restricted than how Phil uses "full citation". Which I suspect is how he might come to a slightly different reading. (Phil?) ~ J. Johnson (JJ) (talk) 22:10, 10 March 2013 (UTC)

Whatever we call it, page numbers aren't added to a "works cited" section, either on Wikipedia or elsewhere, unless it's the page range of a paper or book chapter. But never specific page numbers linking to material used as a source. That just makes no sense, for obvious reasons, namely that if 200 pages were cited, we would have to list all 200, pointlessly. It's really important not to try to reinvent the wheel on this page, and certainly not without knowing that the wheel already exists. SlimVirgin ^(talk) 17:12, 16 March 2013 (UTC)

Well we have no "works cited" sections but "Reference" sections instead. At least in case of short stable articles based on very few sources, where you simply might add them at the end of the article rather than using footnotes, it is perfectly fine to provide an exact page number. Note the possible content of one page in a book is often more than that of a short article. Let's say you have a biography article of half a page in WP, and it is based on an one page long entry in some biography collection, which is simply listed at the end. Then of course you'd provide a single page number with the "full citation". Imho the above discussion bothers way too much with "how I or somebody else is (or should be) using a particular term (reference/citations/works cited/etc." rather than bothering with the actual practical scenarios occuring in WP.--Kmhkmh (talk) 17:54, 16 March 2013 (UTC)

You've misunderstood. Philip is doing this:

Blah. Blah. Blah.

Notes

References

• Rawls, John. A Theory of Justice. Harvard University Press, 1971, p. 3, p. 4, p. 5.

SlimVirgin ^(talk) 18:09, 16 March 2013 (UTC)

I have no problem with such an approach though arguably it looks rather odd and imho it would better to put the online links into the shortened citations/footnotes. However above I was replying to you and the notion that you seem to express, that single page numbers never make sense under references. In some cases they do as in the given scenario in my posting above.--Kmhkmh (talk) 18:39, 16 March 2013 (UTC)

You have no problem with an approach that lists in Works cited/References: Rawls, John. A Theory of Justice. Harvard University Press, 1971, pp. 1, 3, 6, 9, 10, 14, 17, 50, 56, 79, 81, 157, 220, 344, etc etc etc etc?

Look, I know we can do whatever we want so long as it's internally consistent. But a huge problem on Wikipedia – and not only on this page – has always been people thinking they can reinvent the wheel, without bothering to check whether it has already been invented. I think it is really problematic that people are editing this guideline when they have no idea about the way references are handled by publishers, and (more importantly) why they are handled that way. That is, we need to educate ourselves about the basics before trying to give advice to others. SlimVirgin ^(talk) 19:38, 16 March 2013 (UTC)

I think the problem is that Kmh kmh's statement 'Well we have no "works cited" sections but "Reference" sections instead" is meaningless. First of, I don't know who "we" is. Is it supposed to mean Wikipedia, or a few editors engaged in this discussion, or what? The meaning of "works cited" is pretty clear, because all the printed style guides that use that as a section title use some form of short citations, whether they be parenthetical or footnotes, and the "Works cited" section is always an alphabetical listing of works that referred to by a short citation, and never include any works that are not referred to by a short citation. But what is a "Reference" section? It could be anything. Jc3s5h (talk) 19:48, 16 March 2013 (UTC)

"We" stand for wikipedia (of course). What you described as "the problem" was the whole point, since section which is usually entitled references may not just contained "works cited", that is works appearing in footnotes, but it can contain additional works (not cited explicitly) and in short articles it can also serve as replacement for footnotes (see the example in my posting above). In other words while the meaning of "works cited" is pretty clear, is the (implicit) assumption that this is all there is in reference section completely false.--Kmhkmh (talk) 22:56, 16 March 2013 (UTC)

Even as we build our WP specific wheel, I fully agree with SlimVirgin that we should not totally reinvent it, but should heed the practices and standards that have evolved, and, as said, why.

  The example above is to a book, which would not include a page range. A similar case, to an article, might look like this:

• Smith, John, "Summarizing Rawls", Journal of Philosopy, March 1975, pp. 20-30; p. 23, p. 24, p. 25.

  The "pp. 20-30" is where the source (the article) is found (as I described above), and the other page numbers are (presumably) citations to specific pages (what I call the specification).

  As Kmhkmh says, it does look odd. This might be tolerable in a stub article of only two or three sentences, with only one source, and perhaps only three passages cited, and the short cites ("Notes") are left out. But only because the article is so short there is little doubt as to what material is being supported. Otherwise: where the specific page numbers are included in the short cite, it is redundant, and pointless, to include them in the full reference. A full reference describes the source, not the individual passages cited within the source. The exception would be if the full reference was used in place of each short reference in the notes, but I doubt if anyone is in favor of that. ~ J. Johnson (JJ) (talk) 21:40, 16 March 2013 (UTC)

I think more than one footnote per source that gives the complete bibliographic information (together with a specific page that supports the claim) mostly happens in an article where nearly all of the footnotes are for unique sources, but a few sources support claims in more than one spot within the article. I think that scenario is legitimate.

Another situation is where an article started out with unique sources, but over time, as citations were added, it developed into an article with several full footnotes for each of several sources. This is a case of no one wanting to go to the trouble of creating separate "Notes" and "Works cited" sections (or equivalent) and going through the article and redoing the citations to follow the new structure. It's pretty time consuming. Jc3s5h (talk) 23:18, 16 March 2013 (UTC)

Yes. Though I think it's more like the full reference gets put into a note (<ref>), and then subsequent citations to the source are linked to the full reference using named refs. The problem with that is that the subsequent citations/references can't be given a specific location (unless with the ugly {{rp}}). The problem with trailing the specific locations after the full reference is we don't know which piece of text goes to which location in the source. ~ J. Johnson (JJ) (talk) 20:37, 17 March 2013 (UTC)

See what I put at #Short cites above, 22:30, 3 March 2013. None of the {{rp}} there. --Redrose64 (talk) 21:22, 17 March 2013 (UTC)

@Simvirgin: As i said it looks rather odd to me and certainly wouldn't use such a format myself nor recommend it. Nevertheless it doesn't cause any harm nor does it bother me in particular, so if some author feels strongly about using that format in articles where he's the primary author, then that's fine with me or at least i wouldn't bother enforcing a different format.--Kmhkmh (talk) 23:01, 16 March 2013 (UTC)

Wikipedia "policy"on secondary sources: no means offered here for good citing.

How does one cite a Wikipedia "secondary source?" An important issue, see this to make the point: WP:PSTS

The best offered here is:

Smith, John. Name of Book I Haven't Seen, Cambridge University Press, 2009, p. 1, cited in Paul Jones (ed.). Name of Encyclopedia I Have Seen. Oxford University Press, 2010, p. 2.

Is this the best format for corroborating a secondary source?

Worrisome, the comment that follows this example dilutes the demand for verifying a secondary source:

"However, if you have read Smith's book yourself, you may cite it directly; there is no need to give credit to any sources, search engines, websites, library catalogs, etc., that led you to that book."

If so, then secondary sources are irrelevant? There is no need to provide corroboration? This seems pernicious of the Wikipedia project.

Centamia (talk) 07:34, 10 March 2013 (UTC)

A secondary source is a work by (an) author(s) who have read a number of primary sources (and perhaps some secondary and tertiary sources as well) and created a work that considers the views in all the sources that were read. A review article in an academic journal would be an example. A situation where you quote or describe a source you haven't read based on a quote or description in a source you have seen is a different situation; I'm not aware of a short phrase for that situation. Jc3s5h (talk) 12:30, 10 March 2013 (UTC)

Hi Centamia, as Jc3s5h says, that's not what's meant by secondary source. This page deals only with how to write citations, not with what kind of sources to use. To discuss primary/secondary sources, the best page is WT:NOR, because the primary/secondary issue is part of that policy, or WT:RS, the reliable-sources talk page. A secondary source is an uninvolved, independent one. A primary source is involved with or close to the subject. So if you're writing about a clinical trial, a research paper by those involved in the trial is a primary source of information about that trial. A review article or newspaper article about the trial is a secondary source, if written by uninvolved people. Wikipedia articles should for the most part rely on secondary sources, especially for anything contentious. SlimVirgin ^(talk) 16:08, 10 March 2013 (UTC)

Hi Jc3s5h, The ideal citation outlined in your example is spot on!: a knowledgeable primary source referencing a fully-cited secondary source. Citations are the basis for reliability: To pretend that citations and reliability are separable is moonshine.

So, consider this case: http://www.census.gov/compendia/statab/2010/tables/10s0075.pdf The U.S. Census Bureau cites and publishes a Trinity College survey of 175,000 respondents.

According to Wikipedia policies, the citation can only be for the Trinity College survey. 'But,' not citing the Census Bureau dilutes the reliability. How should such a citation be handled so as to make clear that it is vetted through a reliable primary source? Centamia (talk) 09:11, 16 March 2013 (UTC)

Centamia, if you want to discuss that particular source, the best thing to do is post to the reliable sources noticeboard, explaining what you want to use it for. SlimVirgin ^(talk) 17:06, 16 March 2013 (UTC)

I second the suggestion to move to RSN. Among the issues: Wikipedia policies directly say you are permitted to WP:USEPRIMARY sources; primary sources don't cite secondary sources; the Census Bureau report could be either primary or secondary; and if you got the information from the Census Bureau, then you must tell the honest truth and say where you got it, not where your source got the information from. WhatamIdoing (talk) 03:26, 17 March 2013 (UTC)

I know what Centamia wants to use this for. It's for Christian Science. Secondary sources say there are under 100,000 church members worldwide, while the church says 400,000; both figures are in the article (see 4th paragraph of this section). Centamia has found this survey, which gives figures that are difficult to interpret, but which seem to imply a significant rise in membership over the last few years, when everyone (church included) says membership is falling. So for all these reasons -- that it's not clear how to interpret the survey, that it's a primary source, and that it contradicts other sources -- there's consensus so far on talk not to use it. SlimVirgin ^(talk) 21:14, 17 March 2013 (UTC)

Hi SlimVirgin. My question here is only about making sure that readers know the reliability of both the primary and secondary sources. Regarding the variance in the tabular data that you find so troubling, that is easily explained because each year is adapted from a different primary source. Your concern only begs the question I am asking about: is this a situation where both the primary source PLUS secondary source is warranted.

Hello UTC and SlimVirgin: The question is strictly about a citation procedure that best serves the purpose of making it easy for a reader to judge the reliability of the citation source.

In statistical tables, a citation method is often used that is different from that used on Wikipedia. In the case at hand, here is what it looks like:

U.S. Census Bureau (2010) Compendia, Table 75, 'Self-Described Religious Identification of Adult Population: 1990 to 2008.' 2008 data adapted from Barry A. Kosmin and Ariela Keysar (2009) Religion in A Free Market: Religious and Non-Religious Americans, Who, What, Why, Where. Trinity College, Hartford, CT. Retrieved 3/18/2011 http://www.census.gov/compendia/statab/2010/tables/10s0075.pdf

(The phrase 'adapted from' would not be in bold.) Would this be best? Your final thoughts? Centamia (talk) 08:02, 18 March 2013 (UTC)

Hi again, I'm not entirely sure what you're asking now; the issue is not how to write the citation, but whether the source is reliable, and whether it's needed. You can either continue the discussion at Talk:Christian Science, or if you want to bring in uninvolved people you can post a request for input at the reliable sources noticeboard, asking if people think it's a reliable source, and if so how best to use it. But this page isn't the right place to discuss this any further, so I hope you'll continue the conversation elsewhere if you feel it's something you want to pursue. SlimVirgin ^(talk) 19:07, 18 March 2013 (UTC)

He initially asked how to cite a secondary source, as if that would differ for primary or tertiary sources. Further on it seems he is asking how to indicate the reliability of a source in the citation. Perhaps it would help to point out that citations only identify sources; they don't grade them. An unreliable source is cited the same way as a reliable source. ~ J. Johnson (JJ) (talk) 21:31, 18 March 2013 (UTC)

Reprint dates in works cited

I feel sure I've seen this point discussed multiple times, but can't find anything in the relevant pages; perhaps someone can provide me with a link. The general rule as I understand it is to cite the edition you have, not the first edition of that work, and I understand the reasons for this. However, I think there are a couple of cases where it would be useful to give the reader more information about the date of the text. For example, I'm about to cite something from page 187 of Damon Knight's In Search of Wonder, a collection of critical essays; my edition was published in 1974. This is a reprint (without significant changes) of the 1967 second edition; the first edition appeared in 1956. I'm citing from chapter 20, "New Stars"; the copyright statement indicates that portions of that chapter have been reprinted from various magazines printed between 1953 and 1956. I have those magazines and could probably dig out the original version, if necessary.

My question is: should the reader be given the information that the text cited was not composed in 1974, but possibly as early as 1953? Or that the text is essentially just a reprint of the 1967 edition, but might differ from earlier versions? What ought we to tell a reader in this situation, and what would such a citation look like? Or should we just leave at as "1974", and make no comment? Mike Christie (talk - contribs - library) 00:42, 19 March 2013 (UTC)

Use |origyear= to show the year when the first edition was published; use |year= for the year of reprinting; use |edition= for everything else. For example, |origyear=1956 |year=1974 |edition=2nd ed. 1967 (reprinted) which produces:

• Knight, Damon (1974) [1956]. "Chapter 20, New Stars". In Search of Wonder (2nd ed. 1967 (reprinted) ed.). p. 187.

It does look a bit clumsy because "ed." appears twice, but all the essential information is there. Don't worry about giving information pertaining to those 1953 magazines; but if you do have those, and want to use them as source material, you should build a second reference, but using {{cite magazine}} instead of {{cite book}}.

The "golden rule" is to give details of the work which you actually consulted. --Redrose64 (talk) 10:10, 19 March 2013 (UTC)

Thanks -- this is exactly what I was looking for. Mike Christie (talk - contribs - library) 12:15, 19 March 2013 (UTC)

Is a ártical a sours

— Preceding unsigned comment added by 72.68.251.130 (talk) 01:02, 23 March 2013 (UTC)

Long page ranges

Hello,

in Fyodor Dostoyevsky we see Frank footnotes with very long page ranges. People have complained about that, because it does not seem to improve our readers' understanding and they may lose their trust in what is written. However, I think long pages are fine since it is a proof that a particular information is supported by references. I use Joseph Frank's five-volume biography as a general reference, so I would like to ask you if they should be removed (except particular in-line citations)? Regards.--Tomcat (7) 10:42, 23 March 2013 (UTC)

You mean like "Frank 1997, pp. 42–183"? To my mind, those should be split up into more specific indications; that is, the range should exclude the numbers of any pages which do not support the text. I don't have Frank's book, so I can't suggest anything exactly, but hypothetically I would put "Frank 1997, pp. 42–44, 53, 66, 137–140, 145, 182–183" or similar. --Redrose64 (talk) 16:34, 23 March 2013 (UTC)

There is no general rule, because it depends on the details. You might summarize and cite an entire book in support of a single sentence ("Richard Nixon coordinated a series of crimes against political opponents that resulted in his impeachment" and cite a book, Richard Nixon's Political Crimes and Their Results). So since we accept whole books, it would be silly to have a rule against citing less than a whole book.

In my experience, when someone complains about citing chapters or long passages, it's an inexperienced editor who doesn't like the content and can't believe that anyone would actually read a hundred pages and try to produce an encyclopedic summary of them, rather than just copying one or two sentences out of the book. In other words, they're hoping to prove that your content is wrong without having to go to the trouble of actually reading more than a few paragraphs in the cited source.

But if Frank actually wrote 120 pages about something (e.g., about Fyodor's "Idiot" and the fate of the manuscript), and those 120 pages are the actual basis for a sentence or paragraph, then it is normal practice on Wikipedia and in academia for you to actually cite those 120 pages. It is better to cite the whole section than to end up with a footnote that says "For the word idiot, see page 235. For the word burned, see page 267. For the date, see page 278..." WhatamIdoing (talk) 19:06, 23 March 2013 (UTC)

Fact-plucking like that may fall foul of WP:SYNTH. --Redrose64 (talk) 20:06, 23 March 2013 (UTC)

Broken links to Google books

There is a Village pump discussion about how citations should be formatted. Apparently an alteration to the citation template has broken some of the Google links when they are included in the "page" parameter. Anyway, there seems to be a conflict between a particular citation format and the technical side of things. The discussion is at Wikipedia:Village_pump_(technical)#Linking_problem_to_Google_books. Betty Logan (talk) 17:47, 25 March 2013 (UTC)

Creating section to hold footnotes before adding footnotes to be placed there

In this edit, I've WP:BOLDly reordered and reworded some material on the project page. My intent here was to have the project page explain how to create the References section (if such a section is needed to hold footnotes) before explaining how to use Refs to place footnotes into this section. I think that this will probably be noncontroversial but, if I'm wrong about this, feel free to revert or improve what I've done. Wtmitchell (talk) (earlier Boracay Bill) 20:21, 24 March 2013 (UTC)

I am not entirely convinced this is the better ordering. Seems to me there is no need to add a reflist or Notes section prior to adding footnotes (they're not going to wander off). And doing so is somewhat hypothetical until there are some footnotes to populate the list. ~ J. Johnson (JJ) (talk) 21:09, 24 March 2013 (UTC)

Ref'd inline footnotes do indeed "wander off" if no <References /> or {{reflist}} is created for them, at least in the sense that the footnoted content does not appear in the rendered wikitext and in the sense that the inline link to the footnote content is dead. This situation also generates an error notice in the rendered wikitext of WP articles. It is not a good idea to put Ref'd inline footnotes into an article which doesn't have a section with a <References /> or {{reflist}} somewhere near the bottom. If Ref'd inline footnotes are placed in an article, it is a good idea that the article have a <References /> or {{reflist}} somewhere near the bottom to as a place for the list of footnotes to appear. It is not unusual for me to add missing {{reflist}}s to articles while WP:Gnomeing (e.g., [8], [9], [10], etc. I guess it depends on what one considers "not unusual."). That said, as I said above, feel free to revert or improve what I've done. Wtmitchell (talk) (earlier Boracay Bill) 22:57, 24 March 2013 (UTC)

For sure, refs without a reflist/references would be kind of pointless. What I meant was that refs (notes) won't go missing if they are added prior to adding a reflist. And it seems to me (I await correction!) that the tutorial is a little more straightforward if first we explain how to add a note, and then show, if they don't already show up somewhere, how to add the reflist. Though I wonder: does adding a ref without a reflist log an error somewhere, such that it clogs the logs with transient errors? That would convince me to add a reflist first. And I allow that, at a slightly deeper level, when adding short cites [ Harv templates] to an existing article I tend to put the full reference in first so that the readers don't get even a transient flash of red. So I raise the issue, but I am not yet convinced either way. ~ J. Johnson (JJ) (talk) 21:37, 25 March 2013 (UTC)

Most of the time, the reflist/references and the initial ref tag are added in the same edit, so the question of the impact of adding one without the other does not arise. The tutorial is aimed at editors for whom this is not second nature -- editors trying to figure it out -- we both understand that. The tutorial has "Footnotes" and "Parenthetical referencing" as subsections of "Inline citations". A main difference between the two is that footnotes are gathered into a list in a section near the bottom of the article. It seems to me that it makes sense to start the "Footnotes" subsection off with an explanation of how to create the section where the list of footnotes is to appear and then, once that is done, to explain how to place individual footnotes into that list of footnotes which the reader of the tutorial then understands how to create. Wtmitchell (talk) (earlier Boracay Bill) 00:40, 26 March 2013 (UTC)

In regard of new articles I think there is much to be said for providing an example to copy in that provides the basic structure, including a References section with a reflist template. Then, when we get to adding text and citations, we don't have to take a detour to deal with structure. In regard of existing articles... well, there is the question of whether a reflist/references already exists. If it does, then arguably the editor really should check it prior to adding any citations/references, just to see what form/style s/he should be conforming to. If one is not found, then we deal with adding one. Okay, I think I've convinced my self that the way to go is: check for reflist, add if necessary, then do references/notes. ~ J. Johnson (JJ) (talk) 20:27, 29 March 2013 (UTC)

Please add a reference to DEADREF

Could somebody please add the following reference to the Wikipedia:Citing sources#Preventing and repairing dead links section?

<ref name="Internet Archive">
{{cite web
 | url = http://archive.org/about/faqs.php#103
 | title = Internet Archive Frequently Asked Questions
 | author = IA staff
 | year = 2013
 | work = [[Wayback Machine]]
 | publisher = [[Internet Archive]]
 | quote = '''Q:''' Why are there no recent archives in the Wayback Machine? '''A:''' It generally takes 6 months or more (up to 24 months) for pages to appear in the Wayback Machine after they are collected, because of delays in transferring material to long-term storage and indexing, or the requirements of our collection partners. 
 | accessdate = March 31, 2013
}}
</ref>

So that it looks like this.

Markup	Renders as
''Note:'' Most archives currently operate with a delay of ~18 months before a link is made public.<ref name="Internet Archive"> {{cite web | url = http://archive.org/about/faqs.php#103 | title = Internet Archive Frequently Asked Questions | author = IA staff | year = 2013 | work = [[Wayback Machine]] | publisher = [[Internet Archive]] | quote = '''Q:''' Why are there no recent archives in the Wayback Machine? '''A:''' It generally takes 6 months or more (up to 24 months) for pages to appear in the Wayback Machine after they are collected, because of delays in transferring material to long-term storage and indexing, or the requirements of our collection partners. | accessdate = March 31, 2013 }} </ref> As a result, editors should wait ~24 months after the link is first tagged as dead before declaring that no web archive exists. Dead URLs to reliable sources should normally be tagged with <code>{{Tlx|dead link|date{{=}}{{CURRENTMONTHNAME}} {{CURRENTYEAR}}}}</code>, so that you can estimate how long the link has been dead. }}	Note: Most archives currently operate with a delay of ~18 months before a link is made public. As a result, editors should wait ~24 months after the link is first tagged as dead before declaring that no web archive exists. Dead URLs to reliable sources should normally be tagged with {{dead link|date=November 2024}}, so that you can estimate how long the link has been dead. Notes

Thanks very much. 64.40.54.208 (talk) 02:25, 31 March 2013 (UTC)

Why? We don't normally provide reliable sources to "prove" that our policies and guidelines really are our policies and guidelines. WhatamIdoing (talk) 05:24, 31 March 2013 (UTC)

In general, to be helpful to end users. Specifically, because people keep asking about why they have to wait for 2 years before removing a dead link. Here is the most recent example. But if you think edit would not be helpful, then please remove the {{SPER}}. I'm fine with the edit request being being denied. Cheers. 64.40.54.208 (talk) 07:23, 31 March 2013 (UTC)

I agree that the link to the archive.org FAQ is helpful.--SaskatchewanSenator (talk) 19:58, 2 April 2013 (UTC)

I've marked the request as answered (and not made the change) because I don't see a reason to add a reference (which is really what it is since it's already explained in-text, and also serves as a bit of low level advertising for Wayback Machine) proving a guideline. Callanecc (talk • contribs • logs) 06:40, 3 April 2013 (UTC)

Is an entry in the "Further reading" section called a "reference"?

The "Further reading" section is documented at WP:Layout#Further reading.  Is it correct or at least reasonable to refer to entries there as "references"?  If not, what is a proper or at least reasonable term?  Unscintillating (talk) 04:37, 6 April 2013 (UTC)

No, because they were not used to support the statements made in the article, therefore are not references. --Redrose64 (talk) 14:36, 6 April 2013 (UTC)

It depends on the context. For the purposes of the "unreferenced" tag, both further reading and external links count as "references". The idea is that the name of the section does not determine whether a source could be used as a reference - after all the "further reading" could just be merged into the "references" section, and so the difference between the two is mainly organizational. We should not think worse of an article because someone happened to use an alternate name for the general references section. — Carl (CBM · talk) 15:13, 6 April 2013 (UTC)

To begin with, this guideline allows any consistent citation style to be used. So "Layout" only indicates where citation-related sections should be placed within the article. The headings do not need to use the words mentioned in "Layout". An article might be following a style guide that calls for further reading to be merged with references, or might list further reading under a different title.

If an article isn't following any particular citation style, and contains a "Further reading" section, presumably the meaning is as stated in "Layout". "Layout" says "this section is not intended as a repository for general references that were used to create the article content." It also says "the Further reading section should...normally not duplicate the content of the References section, unless the References section is too long for a reader to use as part of a general reading list." So I would say that if a work appears only in the "Further reading" section, it was not consulted while the article was being written, and the editors who wrote the article take no position about whether the work supports the content of the article or not. Jc3s5h (talk) 15:43, 6 April 2013 (UTC)

I don't think you've addressed Unscintillating's question: The thing that's typed into a ==Further reading== section (i.e., the line of text that includes an author's name, the title, the date, etc.) is usually called a bibliographic citation or citation in our guidelines, although many editors also call it a reference. All of these terms have been used by outside style guides, so they're all "correct". We avoid calling it a reference in the guidelines because ==References== is a popular section heading specifically for listing works that were used to create article content, and ==Further reading== is usually and at least mostly a list of works that were not used (yet) to create article content. WhatamIdoing (talk) 16:21, 6 April 2013 (UTC)

Depends on what you mean by "reference". (Gee, do we need a definition here? :-0) However that term is to be used, I will point out that a work — be it a book, website, or whatever — is absolutely no different for being consulted as source than merely for "further reading". Nor is the "line of text" that uniquely identifies such a work any different; there is no reason for a "References" entry to be any different from a "Further reading" entry. The difference is not what the entries are, and I would say not even in how they are used, because in both cases the "entry" is used for the same purpose: identifying the work. The only difference is the where they are used (the context). Is that important enough to have separate terms for identical form and usage? ~ J. Johnson (JJ) (talk) 21:32, 6 April 2013 (UTC)

Proposed definitions

Hoping that we have resolved some of our conflicting interpretations, I propose the following definitions.

1) Citation is the linkage of specific material (quotations, claims, etc.) in the text of an article to specific passages in a source that support that material. A citation should always incorporate a full reference, possibly a short reference, and usually a specification.

1 2) A full reference contains the "full" bibliographic details of the source, sufficient to identify and locate the source, and to distinguish it from similar sources. Full references may be located in the text, in notes, or elsewhere in the article; they can be generated with the {{citation}} or {{cite}} templates. A full reference usually appears only once in an article, to avoid having to duplicate the bibliographic details.

2 3) A short reference is used in place of a full reference in the text or in a note; it has just enough information (usually the author and date) to identify or link to a full reference located elsewhere in the article. When followed by specification (below) it may be called a short citation or short cite. Short cites with automatically generated links to a full reference are conveniently assembled using the {{harv}} and {{sfn}} templates.

3 4) A specification is the page number, section number, line number, verse, or such other information as will identify the specific location within the source that is being cited. It usually follows a short reference; it may a follow a full reference used in the text (or a note) to directly support the text. It should be distinguished from a page range used in a full reference to identify the location of a source (such as an article) within a larger work.

Citation of specific material "inline" requires inclusion in the text being supported, or in a note embedded in (linked from) the text, either a full reference, or a short reference linking to a full reference (with or without parentheses), followed by a specification. A citation is incomplete if it lacks an inline element necessary to connect the material supported with the source material that supports it.that lacks any of these elements is incomplete.

These definitions do not conform exactly to everyone's sensibilities (which is impossible), but I believe these definitions require the least adjustment overall to everyone's sensibilities, as well as addressing several specific concerns. And are clearer, more consistent, and more workable than the current mishmash. ~ J. Johnson (JJ) (talk) 19:57, 14 March 2013 (UTC) Renumbered. ~ J. Johnson (JJ) (talk) 21:12, 15 March 2013 (UTC)

This is a proposal to massively change the established wikijargon. I have two questions for you:

• Do you think that it's wise to try to evict the longstanding terminology from this guideline and replace it with your own?
• Given that WP:Nobody reads the directions anyway, do you think it would stop people from using the old terms (and then have new editors be unable to find a definitions of the old terms, because you evicted them from this guideline)? WhatamIdoing (talk) 20:14, 14 March 2013 (UTC)

  If WP:Nobody reads the directions then surely it would make no difference at all what the directions were or how they were changed, right? In fact people do read directions (as I suspect we all did when we first approached this topic), and they will stick with them if they are helpful, and not confusing. The problem is that the current directions/definitions are ambiguous, and so editors come away with conflicting interpretations — as we have seen. E.g., the existing definition says that a "full citation fully identifies a reliable source and, where applicable, the place in that source (such as a page number) where the information in question can be found." This confuses the two different uses of page numbers (as we discussed above). Why shouldn't this be clarified?

  My proposal is not so much to "massively change" the jargon, but to clarify it. And what I propose is not as massive or radical as you seem to fear; these definitions mainly clarify the edges of what is meant. E.g., my definition of "short citation" (as a short reference plus a specification) works out to essentially the same as the existing definition, but is clearer both as to the constituent parts, and the relation with other elements, and avoids unnecessary qualifications.

  I think I cannot stop you (or anyone else) from using the old, confusing terminology. But why should you want to persist in that? ~ J. Johnson (JJ) (talk) 23:21, 15 March 2013 (UTC)

Not a massive change?

At (1), you re-defined citation from being a thing—a line of text—to being an action, namely the action of linking material to sources. Furthermore, you implicitly define the bibliographic citations used for WP:General references and WP:Further reading and all of what you call an incomplete citation as being non-citations, because these bibliographic citations are (a) not linking article material to sources and/or (b) not sufficiently specific (e.g., containing exact page numbers).

(Your text, BTW, is inconsistent: you define citation as being an action, and then you use the word to refer to the "line of text" meaning throughout.)

You introduce new terms for nearly everything else. What we've called "short citations" for years is re-titled "short references". You've made up this new "specifications" idea. I can't imagine why you think that re-defining (1), picking a distinctly less-common term for (2) and (3), and inventing a completely new one for (4) is "clarifying" the community's own language rather than an attempt to "massively change" the meaning and terms used in every single sentence of your proposal. WhatamIdoing (talk) 00:55, 16 March 2013 (UTC)

  W: It may be clearer to address first your last complaint, that I have "invent[ed] a completely new" term, "specification". Please recall in the discussion above regarding page numbers that various statements as to whether they are allowed in a "References" list, or are even useful — your own comment being "they're not actually helpful" (20:20 7 March) — really depend on which of two ways they are used (as described by both Phil and myself). There has been much dissension because people were not clear as to which context applied, or possibly were not aware that there was another context. So I am suggesting a way to distinguish these. And not by some never before seen term that I have invented, but an ordinary term in its ordinary meaning. This does not change — "massively" or otherwise — any existing term used on WP, it is merely a suggestion that we use an ordinary term to clarify what we are talking about.

  You complain that I have picked the "distinctly less-common" terms full reference and short reference. Full reference is hardly uncommon, and is indeed the term used by CMS-13 (§17.2). That it is "less-common" on WP is all to the better, because there is less entrenched mis-usage to "massively change".

  I did shorten CMS-13's shortened reference (shorter is better, remember?); this short/shortened equivalency is entirely in accordance with practice here (e.g., see WP:CITESHORT). CMS-13 also calls this a basic reference (§15.7), and their usage (which I adopted unchanged) is conformable to WP practice. I also shortened the descriptive "specific page, section, equation, or other division of the cited work" (§15.8) to merely "specification".

  Defining short citation as "short reference + specification" is entirely conformable with the description given at WP:CITE#Short citations, which specifies "with a page number". If you rely on the definition with the messy "inline citation ... specific information ... given either as footnotes or as parenthetical references within the text" — ah, guess what, it is ambiguous, doesn't mention "page number" at all. (See why we have so much bickering?) The definition I propose is resolves the ambiguity, is cleaner, and much shorter.

  As to your complaint that I redefined "citation": I think you are confused, and perhaps you would like to reconsider. ~ J. Johnson (JJ) (talk) 21:19, 17 March 2013 (UTC)

I don't think that your proposals are going to gain any support at all, so I'm not sure that we're doing anything other than beating a dead horse here. WhatamIdoing (talk) 19:35, 19 March 2013 (UTC)

  Since you have no better response may we presume that (possibly aside from reference/citation) my proposals do not "massively change" anything?

  If you are done distorting what I am proposing and raising spurious complaints perhaps you would stand back and let the rest of us discuss the merits of the proposals? ~ J. Johnson (JJ) (talk) 00:27, 20 March 2013 (UTC)

Aside from the explanatory paragraph following my proposed definitions, and aside from whether "reference" and "citation" should be distinguished as suggested below, does anyone (aside from Whatamidoing, who wants no change) have any problem with the proposed definitions? I believe it is quite evident that we do have a problem with our terminology here, and to explicitly state what we mean will reduce some of the misunderstandings. I believe these definitions are pretty solid, but if they need to be tweaked then let's do so. At any rate, let's define what we are talking about. ~ J. Johnson (JJ) (talk) 22:41, 22 March 2013 (UTC)

Okay, I take it that there are no major objections (aside from WhatamIdoing) with these definitions. I am not proposing specific text yet as I think there are other issues to address in that first section ("Types of citation"). ~ J. Johnson (JJ) (talk) 20:35, 29 March 2013 (UTC)

I agree with WhatamIdoing that there is not need to define new jargon. For most purposes, words like "reference" or "citation" are used interchangeably. Sometimes (especially in this guideline) it is necessary to distinguish between "citations", "short citations", "full citations", "references", "sources", and (don't forget) "footnotes", the distinction must be discussed in the paragraph that needs it. Otherwise, the reader will be lost -- you should be able to read any sub-section in the guideline and understand it without training in our jargon.

Remember, these guidelines are primarily descriptive, not prescriptive -- this guideline describes common practices, using the terminology that most Wikipedians use. We document the terminology in use. We don't make it up ourselves. ---- CharlesGillingham (talk) 02:27, 5 April 2013 (UTC)

  Part of the problem is that all these terms (including "general reference") are used variously (differently and interchangeably), and thereby confound communication. Distinguishing these terms as needed fails because we make different distinctions; experience (i.e., this extended discussion) shows the futility of trying resolve these differences as we proceed. Note that I am not trying to define new jargon — I am trying to get our various understandings of our existing terms all pointed the same way.

  You say that these guidelines should be descriptive, not prescriptive. That is not guidance, that is a lack of guidance. In fact the definitions here are cited by many editors (including those present) as prescriptive. What I am suggesting is that we should offer a standard, consistent conception of citation, rather than leave everyone to make it up as they go. ~ J. Johnson (JJ) (talk) 21:37, 6 April 2013 (UTC)

I disagree very strongly that this guideline is primarily prescriptive. This is a guide to common practices and solutions to common problems. This explains why we document so many different techniques. Sadly, we do not have the power to create "conception of citation". The conception of a citation has already been created by the community as a whole. At most, we can document what already exists. But our highest priority is to communicate clearly about citation techniques. This is not a matter of definition but good writing in individual paragraphs. ---- CharlesGillingham (talk) 01:22, 8 April 2013 (UTC)

No amount of "good writing in individual paragraphs" can overcome bad definitions. As I said below: you CSS. ~ J. Johnson (JJ) (talk) 23:05, 9 April 2013 (UTC)

Reference vs. citation

In the definitions above I would distinguish between "reference" and "citation", a point which is underdiscussed. This is a point on which eminent authorities are ambiguous, so I have previously tried to treat them as synonyms. (Even in this discussion; e.g. 23:50 28 Jan..) But I have been swayed (it happens) by several comments (Phil's?) to consider individual citations to be the complete point-to-point linkage (or connection) from article text to source text, such linkages being comprised of the link components of full reference, short reference, and specification (as defined above). — Preceding unsigned comment added by J. Johnson (talk • contribs)

I believe citation and reference are always used interchangeably, including to mean what you're calling the point-to-point linkage. SlimVirgin ^(talk) 23:47, 17 March 2013 (UTC)

I think there is a general tendency to do so, as I also have done. But I also detect some significant non-overlap. Given that it would be useful to distinguish the entire linkage and state from the constituent parts — and I am convinced that is very necessary — then I think these terms would fit the purpose as proposed. Which is not to give either term a new meaning, but to be more precise in which of a broad range of meanings is meant. ~ J. Johnson (JJ) (talk) 21:15, 18 March 2013 (UTC)

Can you cite a source that uses them to mean different things? I've looked around but couldn't see one. I don't mind if we come up with our own definitions, but they would have to be very clear, and they would also have to be useful to Wikipedians. If they're not useful distinctions, people will just ignore them.

I'm not clear myself what distinction you want to draw. Are you saying a reference is: "John Smith, Name of Book, Name of Publisher, 2013," and a citation is "John Smith, Name of Book, Name of Publisher, 2013, p. 1," and Smith 2013, p. 1"? That is, it's a citation if it includes page numbers that point to the precise source text, and a reference if it doesn't? SlimVirgin ^(talk) 21:25, 18 March 2013 (UTC)

Yes. A full reference identifies and points to the source (work). This is in accord with the work itself (not the text describing it) commonly being called a reference, such that we speak of "reference works", and of "consulting a reference", even "reference desk", whereas we don't refer to "consulting a citation", etc. Also, "to refer to" has a sense of pointing to something in the distance, whereas "to cite" still has some of the original sense of summoning (or bringing) something here, to this point of text. Which is what we do with inline citations.

Though these two terms used in ways not entirely equivalent (as I have shown), I don't recall seeing any authority explicitly distinguishing them. Which I take to be part of the problem. ~ J. Johnson (JJ) (talk) 22:38, 18 March 2013 (UTC)

Past efforts to differentiate these terms have been rejected, and I don't see neither any new arguments in favor of the change nor any value in repeating a discussion that is doomed both in theory (because there is so little support on this page) and in practice (because WP:Nobody reads the directions, so nobody will use JJ's very fine distinction). Additionally, since many of our sources are webpages and short articles, for which no page number is sensible, and since whole books can and are cited in their entirety, the distinction is largely irrelevant. On the rare occasions when it matters, people can use more specific and immediately comprehensible terminology, like "We need a page number" rather than "We need a citation" or "We need a specification". WhatamIdoing (talk) 19:41, 19 March 2013 (UTC)

JJ, I don't think that distinction would fly. We use the terms reference, citation and sometimes source interchangeably, and it doesn't really cause problems. SlimVirgin ^(talk) 23:23, 19 March 2013 (UTC)

W: I think you are being ridiculous again, or at least irrelevant. Also, I would love to see any prior discussions you think relevant; could you please provide links?

SV: I agree that "reference" and "citation" are often used interchangeably ("source" is dubious). But not always, and certainly they are not identical, and that does cause some problems where people are running on non-parallel meanings. However, my motivation here is not so much concern about all that as about the confusion regarding the underlying concepts. I think it would be useful to distinguish the complete point-to-point linkage (connection, process, state, etc.) from the components; the bridge, as it were, from the arches, piers, and abutments. And I think these terms would be appropriate, as they already carry some of the proposed meaning.

~ J. Johnson (JJ) (talk) 01:10, 20 March 2013 (UTC)

Perhaps you should search the archives for this talk page if you really want to know what discussions have happened in the past. WhatamIdoing (talk) 00:08, 23 March 2013 (UTC)

W: Regarding your edit summary please note that I have NOT said that you are ridiculous. I said that you being ridiculous. E.g., you said: "many of our sources are webpages and short articles, for which no page number is sensible ..." — ah, perhaps you were thinking of SV's example, which is a citation of a book, for which a page number is certainly "sensible". That a page number is not "sensible" for a web page — so what? We were not talking about web pages. Your syllogism seems to be: page numbers are non-sensible with web pages, therefore use of page numbers anywhere else is irrelevant. Is that not ridiculous? (Or did you leave something out?) If you asked for a comparable example for a web page I could give you one, but no, you just make absurd statements.

As to searching the archives: I don't know which of many past discussions you consider relevant. If you want to invoke some prior discussion please tell me which one. ~ J. Johnson (JJ) (talk) 20:53, 24 March 2013 (UTC)

Again, I think it is important to remember that our job here is to be useful. This is not legal document or a philosophy text. It's a guideline. The primary goal should be to write paragraphs that make sense on the first reading by an editor who has never added a citation to Wikipedia before. If we need to make finer distinctions, it must be in the paragraph where it is relevant. The text of the paragraph must distinguish the concepts. Defining specific jargon just serves to hide the concepts behind opaque words. ---- CharlesGillingham (talk) 02:35, 5 April 2013 (UTC)

It is not a guideline if it does not guide. If the basic concepts are defined at the lowest levels, then each paragraph has its own definition. Which leads to mishmash nearly as bad as every editor having their own definition. Citation would be facilitated if our terms and concepts were consistent across the board, and (as far as possible) with non-WP usage. ~ J. Johnson (JJ) (talk) 21:43, 6 April 2013 (UTC)

I think that a philosopher, linguist or legal scholar would disagree that your "lowest level" actually exists at all. There is no bottom to definitions. This is why legal definitions tend to get longer, more complicated and more confusing as time passes and more exceptions and finer distinctions come to light. These kind of precise definitions are necessary in the law, but are not necessary to accomplish the goals of this guideline. Paragraphs using overly precise definitions require training just to be understood. This is poor fit for this guideline because all of our readers are "lay readers". We don't need definitions, we need explanations. ---- CharlesGillingham (talk) 01:28, 8 April 2013 (UTC)

I think you misunderstand. By "lowest level" I mean where terms are defined at each specific point of usage, typically the level of a paragraph. The reason certain kinds of legal definitions get longer, etc., is that they get defined at the case level (i.e., in specific instances), but in different ways, which the appellate courts then have reconcile. Similarly here, as existing practice: we have editors using (or taking) these terms in different, and even nonsensical, ways. It is because our editors are "lay" — that is, are not experts — that our usage and definitions are so piss poor. If those of us at this page can work out better definitions and put them at a top level, then they won't have be redefined (and likely mal-defined) every time a term is used. Explanations can certainly be added, but, as we used to say in the service, you can't shine shit. Explanations should start from good concepts. And clear, precise concepts require less explanation. ~ J. Johnson (JJ) (talk) 22:57, 9 April 2013 (UTC)

sfn citation format

I am beginning to see more and more articles with a double level of sources which are unclickable, do nothing, and don't contain the relevant sentence of text. Is this some new innovation? In ictu oculi (talk) 06:32, 9 April 2013 (UTC)

Fixed. That article uses Shortened footnotes. Problems:

• The full citations need to have the anchors enabled, in this case by adding |ref=harv.
• One of the shortnotes had "ed.", but the longnote anchor is only formed from the editor last name and year.
• One of the shortnotes had the author last name misspelled.

If you install the script listed at Help:Shortened_footnotes#Errors, then these errors are detected and a message displayed. --  Gadget850 (Ed) ^talk 10:02, 9 April 2013 (UTC)

Hi Gadget, with this system is it possible to display a sentence of text from a book like note 4 in this example.
And also display a sentence from p10 of a book for note [1], then a different sentence from same book p20 for note [2]? In ictu oculi (talk) 10:19, 9 April 2013 (UTC)

The {{sfn}} and similar templates supports page numbers. See Shortened footnotes. --  Gadget850 (Ed) ^talk 10:23, 9 April 2013 (UTC)

Oh I see what it's done, it lifted the sentence into a note box. That means the Book, the page number, the sentence are now in 3 different places. In ictu oculi (talk) 10:24, 9 April 2013 (UTC)

What an awful system - is there a system that keeps all 3 together? other than just existing manual ref brackets. In ictu oculi (talk) 10:25, 9 April 2013 (UTC)

See NBR 224 and 420 Classes (13:32, August 1, 2011) for a good example of this system. Footnotes do not use manually placed brackets. For other ways of including page numbers, see Help:References and page numbers. --  Gadget850 (Ed) ^talk 10:37, 9 April 2013 (UTC)

Okay, thanks. NBR 224 and 420 Classes (13:32, August 1, 2011) seems appropriate when one is listing a large number of page numbers from paper books, on trust that the book actually says that and assuming people will go away and use a paper library. The sfn system doesn't seem appropriate to where quoting giving a hardwired sentence from the book to show the edit is not original research. In ictu oculi (talk) 10:45, 9 April 2013 (UTC)

You don't need to include a quote for every reference. --  Gadget850 (Ed) ^talk 11:20, 9 April 2013 (UTC)

In fact, it is unusual to include quotations. Unless someone has a good reason for that particular quotation, I suggest removing it entirely. WhatamIdoing (talk) 23:57, 9 April 2013 (UTC)

I only recall suggesting a quote when the source was an eBook with no in-source locators. --  Gadget850 (Ed) ^talk 00:35, 10 April 2013 (UTC)

Hi WhatamIdoing, "In fact, it is unusual to include quotations" - that is unfortunately true, and when followed to a paper copy many references don't say what the text before the ref says.

The reason for including quotations in controverted subjects is because of the need to verify what is edited is backed up by a sentence in the source that says what is claimed. Steam engines may not be a controversial subject where WP:SYNTHESIS and WP:OR are a problem. Depending only on linking to eBooks is a bad idea due to international access to Google Books varying, link rot, and occasional removal of Google Books. And just plain hassle and inconvenience to the reader. And also you can only link to a page, not the relevant sentence. In ictu oculi (talk) 01:25, 10 April 2013 (UTC)

Directly quoting material to indicate availability

From Wikipedia:Citing sources#Indicating availability (emphasis mine):

If a citation without an external link is challenged as unavailable, any of the following is sufficient to show the material to be reasonably available (though not necessarily reliable): providing an ISBN or OCLC number; linking to an established Wikipedia article about the source (the work, its author, or its publisher); or directly quoting the material on the talk page, briefly and in context.

This policy is only relevant if someone has already questioned the availability of a source. When a source's availability is already in question, I don't believe quoting material supposedly from that source is sufficient to establish availability. If the source is reasonably available, some other evidence of that can be found; if it isn't reasonably available, one editor's assertion of the text shouldn't be sufficient.

As such, I propose removing the emphasised text.

—me_and 00:23, 10 April 2013 (UTC)

Relevant discussion from the talk page archives, which (I think) discusses the original addition of this text. —me_and 00:25, 10 April 2013 (UTC)

Problems with current definition of "general reference".

The definition of "general reference" (defined at WP:CITE#Types of citation and repeated WP:GENREF) can be parsed as follows:

A general reference is

a) a citation

b) to a reliable source

c) that supports content,

d) but is not linked to any particular piece of material in the article

e) through an inline citation.

Various problems with this definition are summarized as follows.

a) If "reference" and "citation" are synonymous then "reference is a citation" is redundant. If not synonymous, the definition fails to explain the distinction, and is thereby ambiguous.

b) By this definition citations to unreliable sources are excluded. But reliability cannot be determined without some information about the source, which is external and unavailable in the context where the term is used. Reliability may also depend on how the source is used. This qualification is unnecessary, and makes the definition indefinite.

c) "Supports content"? Presumably there is no reason to have citations that do not support content. Possibly this qualification was added to distinguish a "general reference" from a "Further reading reference", the latter presumably not supporting content. But even that might not be true: a "Further reading reference" could support content (which might be why it has been suggested for further reading) even though it was not cited in support.

d) Non-particularity is the essence of generality. But what this definition does not clarify, and where many editors are confused, is that there are two different, and independent, contexts here for applying generality. To illustrate: a particular passage in a source might be applied generally to a whole article (or section of an article). Or a particular point or claim in an article might be supported by citing a whole source generally. The current definition of "general reference" does not include the latter usage, although the term is often used that way; usage and definition conflict.

Perhaps most damning for the current definition is in failing to distinguish when there should be a linkage to particular material. If there is particular material from a source, it should be linked. The lack of a link constitutes an omission, not a "general reference", and the failure to distinguish this promotes incomplete citation.

e) The "inline citation" (as currently defined) is the linkage from particular material to a source. It is unclear as to what other linkages there might be, though the lede refers to citations "indicated by a superscript number...." This suggests that use of Harv templates in the text are not "inline", and therefore the full references so linked could be "general".

The existing definition is fundamentally flawed in trying to attribute generality to a full reference. References can be invoked multiple times, and using a reference generally does not preclude using it particularly. It is the usage of a reference, not the reference itself, which may be general. A citation (as I would define it, and including "inline citations") might be usefully be called "general", but the existing definition makes not such a distinction.

These problems show that the current definition of "general reference" is flawed, and requires revision. ~ J. Johnson (JJ) (talk) 19:00, 30 March 2013 (UTC)

Looking at GENREF, and thinking of what I've seen while WP:GNOMEing, I'm thinking of a "general reference" as a reliable source which is cited in support of the article generally, but is not cited in particular in support of individual assertions or items of information in the article. I often see such sources listed in a list of sources. Some, all, or none of such listed sources might be referred to by parenthetical references or footnoted inline citations. Sometimes a bulleted list of such sources is placed immediately after a {{reflist}}, sometimes it is placed in a separate article section.

I sometimes see edits which remove a particular reference to a listed source. If the reference thus removed is the only one in the article which particularly refers to that source, removing it would seemingly convert that listed source into a general reference if that source is regarded as still supporting the article generally -- or, possibly alternatively, removal of the final particular reference to a source might suggest that that source should be reclassified as a Further reading item.

I agree that this is messy and needs attention, but specific suggestions don't come immediately to mind. Wtmitchell (talk) (earlier Boracay Bill) 21:12, 30 March 2013 (UTC)

I believe that J. Johnson needs to WP:Drop the stick and back slowly away from the horse carcass. He's spent a lot of time over the last couple of months first trying to convince editors to redefine the term according to his initial instinct about what it "ought" to mean, and then to convince them that this term should be eliminated. There's no consensus to do this. There's no hope of a consensus to do this. All of us would be better served by letting the dead horse lie in peace, and finding something productive to do. WhatamIdoing (talk) 00:15, 31 March 2013 (UTC)

  Bill: I am working on specific suggestions, but at this particular point my focus is examine the specific points of why GENREF needs to be changed. Your comments illustrate some of the difficulty we labor under for lack of well-defined terms. E.g., I interpret your "edits which remove a particular reference to a listed source" as removal of an inline citation that links to a full reference, but others might read that somewhat differently. I am hoping that when I make suggestions I won't have to go back and argue why they are needed.

  W: I think the dead horse is the existing definition of "general reference". Of course I have spent a lot of time on this — I have been chewing on the general problem of WP citation for much longer than a "couple of months" — because it is a significant issue which warrants full and careful consideration. The stinking carcass is evident; your claim of no consensus to make any changes is basically you claiming that there is no consensus, and no hope of consensus. I would address your concerns, but if all you can do is to reiterate "no consensus", or make nonsensical arguments (such as "WP:Nobody reads the directions, so nobody will use JJ's very fine distinction"), there is not much to address. You have stated that you "want no change", which is likely founded deeper in your pysche than I have competence to address. Okay, so be it. Your sentiment (no change, none at all) is duly noted. As you have nothing else pertinent to say, please allow the rest of us to consider how this page might be improved.

~ J. Johnson (JJ) (talk) 19:35, 31 March 2013 (UTC)

Sounds like you don't really understand what a so-called "general reference" is, J.j. ... so I guess that it is an argument to rewrite the paragraph. Feel free.

General references are commonplace in articles that cover well-understood topics that might found in any textbook, for example Chain rule (probability). An article on a subject like this doesn't need a citation on every line, indeed, it does not need inline citations at all. Nothing is likely to be challenged, and all the material in the article can be found within a few pages in any textbook that covers it. "General reference" is a poor term, I agree, but I doubt that we would be served by making up a new term. The paragraph needs to make the concept clear. This is the problem. ---- CharlesGillingham (talk) 02:52, 5 April 2013 (UTC)

Ah, are you a bit off-topic here? As a matter of fact I think I do understand what a "general reference" is in common usage. But perhaps you quibble with my analysis above? Perhaps you could be more specific? ~ J. Johnson (JJ) (talk) 21:48, 6 April 2013 (UTC)

In this guideline, section one, we are trying to make a quick list of the types of objects that this guideline discusses. One of the items in the list is citations like the one used in Chain rule (probability). This is what you are trying to define. ---- CharlesGillingham (talk) 01:32, 8 April 2013 (UTC)

 Huh?? How about staying on topic? The specific question I raise in this section is various problems with the existing GENREF. In this section I am not trying to define anything, I only examine the problems. And these problems exist regardless of what we are trying to do in the guideline.

  To help you refocus, tell me this: are "reference" and "citation" (as used in the existing definition, above) synonymous? And in that case is the phrase "reference is a citation" redundant, and can the latter part be removed? If not synonymous, how should they be disambiguated?

  (I will address "what we are trying to make" below in #Proposed redefinition of "general reference".)

~ J. Johnson (JJ) (talk) 23:42, 8 April 2013 (UTC)

I have WP:BOLDly rewritten the section, in hopes that it is less ambiguous or confusing. I avoided the term "general reference" since, really, it means nothing, and we don't need it. What's important is that some articles don't need inline references, why they don't need them, how they look, and what the risks are. ---- CharlesGillingham (talk) 15:12, 10 April 2013 (UTC)

You're braver than me. I don't entirely agree with the changes, but I'll give you points for boldness. I think we do need to address "general references", because 1) there is a valid use of the term, and (I think this was W's point a ways back) 2) it has been used, and that prior usage needs to be explained. But I also think we can sort that out at leisure, and perhaps better after we get a better handle on what a "genref" really is. ~ J. Johnson (JJ) (talk) 20:26, 10 April 2013 (UTC)

Newspaper article with author?

An article with a single author, Joe Smith, appears on page 5 of a 1932 newspaper. When using short footnotes, is "Smith 1932, p. 5." acceptable as the inline reference? The page number seems appropriate for magazines and journals, but I just wanted to make sure for newspapers that it was the same. Thanks. – Kerαunoςcopia^◁_galaxies 03:30, 12 April 2013 (UTC)

It's acceptable, but not necessary. WhatamIdoing (talk) 04:48, 12 April 2013 (UTC)

So "Smith 1932." would be acceptable as well (if I understand you correctly)? (By the way, the actual {{cite journal}} template would have the page 5 in it. – Kerαunoςcopia^◁_galaxies 05:37, 12 April 2013 (UTC)

I would say it is strongly preferable to include specific location (e.g., page number). But not in the {{cite journal}} template (or similar templates); that's used for identifying the location of a (say) article within the containing work. ~ J. Johnson (JJ) (talk) 22:18, 12 April 2013 (UTC)

Most newspaper articles are less than one page long. The normal Harvard-style citation for this news story, assuming you'd actually seen it on newsprint, is

Rich, Motoko (11 April 2013). "Texas Considers Backtracking on Testing". The New York Times. p. A12.

in the bibliographic list at the end, and (Rich 2013) inline. You are permitted to make up a style that results in (Rich 2013 p. A12) if you want, but that's not the most common approach. WhatamIdoing (talk) 00:43, 13 April 2013 (UTC)

Getting mixed signals, but I think WhatamIdoing's suggestion makes sense. I'll keep the page number in the cite journal template (otherwise why is there a page parameter) and the inline will be similar to "Rich 2013". If I take the article to GA or further, any requested changes will be made. Thanks so much! – Kerαunoςcopia^◁_galaxies 01:15, 13 April 2013 (UTC)

There is a difference. Consider: if you had a multi-page article, and you cited it at two places, to two different pages: well, the individual short cites would be need different page numbers. (Which is not a "made up" style, but very common.) If you were using {{harv}} templates you would use the |p= parameter in that template (not in {{cite}}). But don't worry about it. If what you have done works for you, good enough. Though you might consider using {{cite newspaper}}. ~ J. Johnson (JJ) (talk) 00:07, 14 April 2013 (UTC)

But I want to do what's right, not what works for me. It makes sense that I should be using the {{cite news}}(paper) template, I just got caught up in using the journal template. The article I'm citing appears on only one page, which is really why I was confusing myself. (I'm still confused.) Although the article appears online, it appears within the entire scanned newspaper, so a search for that specific page is necessary anyway (I wasn't able to get the URL to "zoom in" to that page). What I'm wondering is if I should say "Smith 2013, p. 5." and also mention "p. 5" in the {{cite news}} template, or is it one or the other. It's these stupid little things that bother me, but I appreciate everyone's input immensely. – Kerαunoςcopia^◁_galaxies 00:39, 14 April 2013 (UTC)

Two points:

• There is no "right" way to do it. There are dozens of "right" ways to do it. Whatever you do in this instance is likely to line up with at least one of them.
• GA does not require perfection in citation formatting or page numbers. See Wikipedia:What the Good article criteria are not#(2) Factually accurate and verifiable. WhatamIdoing (talk) 01:27, 14 April 2013 (UTC)

I think it may become clearer if we see how the methods you've described would extend to a more complicated situation. Let's say you're citing an article titled "Business activity" by Jane Smith on page 1 and 8 of the Smallville Telegram for January 6, 2095, and the article uses Citation style 1 and short footnotes. So the bibliography entry at the end of the article cites the newspaper with {{cite news}} and includes the parameter "pp = 1, 8". At a point in the article where a claim is supported by the part of the newspaper story that appears on page 8, you use {{sfn|Smith|2095|p=8}}. Jc3s5h (talk) 01:47, 14 April 2013 (UTC)

I am pleased to state that I fully agree with WhatamIdoing that 1) there is no one right way, and 2) GA doesn't require perfection. As to what way might be better — ah, opinions vary on this, and we are still working it out. (Don't stand too close to the machinery!) In your instance I say the page number could be validly used in both the short cite and the full reference (though for different reasons). It might seem a little odd, but that's because you have kind of special case. Not a big deal. ~ J. Johnson (JJ) (talk) 19:15, 14 April 2013 (UTC)

See Shortened footnotes. And the style you would use is the style that is already in use in the article. --  Gadget850 (Ed) ^talk 20:38, 14 April 2013 (UTC)

Thank you everyone for your kind help! – Kerαunoςcopia^◁_galaxies 20:52, 14 April 2013 (UTC)

Google books

Hello! Another editor and I are wondering whether a citation for a book, which includes a link to a Google book page, should include a retrieval date or not. The "webpages" section seems to indicate it should, while the Google books section makes no mention of it. Can anyone clarify? Thanks! MeegsC (talk) 03:26, 11 April 2013 (UTC)

The purpose of the citation is to identify the source. Style guides such as Chicago and APA state to include an access date only for web pages that change (i.e. Wikipedia main page) or for web pages with no date. Otherwise, the date you accessed the page really doesn't mean anything. A Google Book isn't going to change. --  Gadget850 (Ed) ^talk 07:36, 11 April 2013 (UTC)

Actually the Google Book link might change in the sense that it can become unaccessible (happens occasionally) and the theory google ould remove a link altogether. However the book which is actually cited remains the same and the Google book link is just a convenience link to an online copy anyhow.--Kmhkmh (talk) 11:17, 11 April 2013 (UTC)

No idea how I ended up here, but anyway wanted to say that the widely used Google Books citation tool automatically inserts access dates. Perhaps we should point this out on the help page. — MusikAnimal ^talk 02:28, 12 April 2013 (UTC)

If the book has an ISBN, it shouldn't have a link to Google Books in the citation at all, since you can follow through to Google Books (and a variety of other sources) from the ISBN link. From WP:ELNO (links to be avoided): "Sites already linked through Wikipedia sourcing tools. For example, instead of linking to a commercial book site, consider the "ISBN" linking format, which gives readers an opportunity to search a wide variety of free and non-free book sources." —me_and 09:12, 15 April 2013 (UTC)

Linking full or restricted preview in Google doesn't fall under the "commercial book links" to cite above. As long as you just want to link full or restricted preview you cab do that of course. The only thing you shouldn't do is providing the Google link merely for identification rather than for a preview, in that case the ISBN linking for is preferred.--Kmhkmh (talk) 11:09, 15 April 2013 (UTC)

The prohibition in WP:ELNO is against "Sites already linked through Wikipedia sourcing tools", which includes Google Books. The comment about linking to a commercial site is only an example. Provide the ISBN and a page number, and the reader can look it up in their choice of sources, which may be Google Books, or may be somewhere else entirely. —me_and 12:13, 15 April 2013 (UTC)

WP:ELNO doesn't forbid full or restricted view Google book links, this has been discussed ample of times on various project sites, nvm that the ISBN is providing the same functionality anyhow.--Kmhkmh (talk) 13:13, 15 April 2013 (UTC)

WP:ISBN providing the same functionality is why we don't need to link to Google Books directly in the citation. What's your interpretation of "Sites already linked through Wikipedia sourcing tools" that means it doesn't apply to Google Books?

That said, if there is an existing consensus, I'm willing to bow to it (although I disagree)—can you point me at somewhere where this has been established? I'm wary of project pages establishing local consensus.

—me_and 15:07, 15 April 2013 (UTC)

No WP:ISBN is not providing the same functionality. I'm talking about directly online copies of the source (as you get can from archive.org, Google books, arxiv, etc.) and not about a list where you might find the book or in which catalogs it is it contained. WP:ISBN delivers the latter and not the former. Linking online copies as convience links (be it on archive.org, Google, arxiv or similar option) is long established practice, which has different purpose and will certainly not be abolished because of WP:ISBN nor was there ever a consensus for such a measure.--Kmhkmh (talk) 16:22, 15 April 2013 (UTC)

ELNO applies to external links, not to citations that support article content. I believe the guideline currently says that six times, including in a footnote at the top of ELNO, but if we need to make it a seventh, then please let me know. WhatamIdoing (talk) 16:08, 15 April 2013 (UTC)

yes that was apparently ignored by Meand as well.--Kmhkmh (talk) 16:26, 15 April 2013 (UTC)

Okay, you've got me on that one. Six times, perhaps, but I still managed to miss them all. Not sure how you could make it much more clearer, to be honest. I'll go back in my box now.

I do object to being accused of ignoring things, though—that implies I saw those points and took no notice. This is sheer incompetence on my part, not malice. —me_and 17:32, 15 April 2013 (UTC)

You aren't the first, and you definitely won't be the last. That's why it's in the guideline six times already. I think I'll add another copy of the footnote directly to that item. That will at least give the next person a better chance at seeing it, especially if he's sent to that item with a direct link to WP:ELNO#EL15. WhatamIdoing (talk) 20:41, 15 April 2013 (UTC)

It might be also worthwhile to note there that the isbn link is no replacement for directly online copies (which for instance could be in further reading section or similar as well).--Kmhkmh (talk) 21:57, 15 April 2013 (UTC)

Proposed redefinition of "general reference"

Because of the demonstrated faults of the existing definition, I propose redefining "general reference" as follows. This is based on the common and standard meaning of "general" as "applicable to the whole", and "not specific or particular".

A general reference is the use of a source, or of a reference to a source, in a general manner, meaning either 1) the referencing of an entire source as a whole, or 2) citation of a source, whether as a whole, or any particular part thereof, for an entire article or section.

This definition would be followed by explanation that full references (such as uniquely identify a source but not any particular part of it) are inherently general in regards of the source, but that the use (or citation) of a reference may be general or specific depending on how it is used. It will be further explained that citation of a source generally or in part is independent of whether the citation supports an article (or section) generally, or some part thereof; therefore a citation can be "general" and "particular" simultaneously.

It should be further noted that support of particular material (quotations, claims, etc.) requires (per WP:V) a citation "inline" (in the text near the material), and that citation of particular material within a source should specify the location of the material, but neither the presence nor absence of an inline citation or a citation specification determines whether a citation is "general". ~ J. Johnson (JJ) (talk) 19:15, 4 April 2013 (UTC)

• This guideline has no business specifying what material "requires" an inline citation. We have policies that do this.
• I oppose your ongoing efforts to redefine these long-standing terms to mean anything other than what the community has used them for over the last 5+ years. WhatamIdoing (talk) 23:51, 4 April 2013 (UTC)

  Neither this proposed definition nor the accompanying explanation specify what material requires citation; it only states what is required by WP:V. But I will explicitly include "per WP:V" so there is less chance of confusion.

  Your implacable opposition to any change has already been noted. Perhaps we might add that you think the existing formulation is absolutely perfect and admits of no improvement? ~ J. Johnson (JJ) (talk) 21:59, 5 April 2013 (UTC)

J.J., I wonder if you could centralize your discussion in a single section? I've replied to this above. I think you've misunderstood the concept. ---- CharlesGillingham (talk) 02:54, 5 April 2013 (UTC)

I was trying to address the various concerns separately so they would be less entangled. I will review the previous discussions (and your other comments), but perhaps you could point to any specific reply you have in mind? ~ J. Johnson (JJ) (talk) 21:59, 5 April 2013 (UTC)

  Charles: I was under the impression you had commented on this topic before, but I did not find any comments from you in the most recent archive (34). On the current page your only comments were on 5 April, at #Proposed definitions, #Reference vs. citation, and #Problems with current definition of "general reference". I have responded to each in its place.

  You have objected that these guidelines should not be prescriptive, but describe how the terms are actually used. In fact the existing definition is cited by many editors, including some present now, as prescriptive. But if you will carefully examine the proposed definition you will find it generally more conformable than the existing definition with the various examples that have been raised. Please explain if you think I have misunderstood the concept. ~ J. Johnson (JJ) (talk) 22:01, 6 April 2013 (UTC)

I commented two sections above, which also about the definition of (what we currently call) a "general reference". Your definition above is very poor because, as I said above "In this guideline, section one, we are trying to make a quick list of the types of objects that this guideline discusses. One of the items in the list is citations like the one used in Chain rule (probability). This is what you are trying to define." Your definition at the top of this section does a poor job of this. ---- CharlesGillingham (talk) 01:32, 8 April 2013 (UTC)

Your definition covers "some of the many ways a reader might fail to understand what we're talking about when we say 'general reference'". It does not communicate, it has the opposite effect -- it creates confusion and misses the point. Context-free definitions are actually harmful here. You need a definition that describes the context as well as the term. For example:

In some articles all of the material may be found in a single source, often in a small section of the source. These articles are typically on narrow academic subjects or are "stub" articles with just a few paragraphs. In cases like these, inline citations are often not necessary and a few references at the bottom of the article is all that is required. Specific page numbers are usually attached to the full citation at the bottom of the page. This guideline calls these citations "general references."

We're defining a term for a specific concept in a specific context, the context of the guideline. Without the context, your definition will be unreadable, confusing, arguable (in the sense that it can never be completely correct) and harmful (in the sense that it fails to get across the information we need to get across in the guideline). ---- CharlesGillingham (talk) 01:41, 8 April 2013 (UTC)

Where does it say "we are trying to make a quick list of the types of objects that this guideline discusses"? A quick search indicates that "quick list" and "types of objects" are novel terms never before used on this page. (It appears you are trying to redefine our purpose here.) You also refer to "[o]ne of the items in the list" — ah, what list?

It seems to me the deeper issue here is that you (and one or more other editors) want to protect a certain usage: citations like the one used in Chain rule (probability). But your concerns are entirely misplaced: to the extent that your citations are "use of a source ... for an entire article or section", hey, that is explictly allowed in my proposed definition!

You claim that the proposed definition "is very poor", "does not communicate", and "fails to get across the information we need....", but you have not demonstrated any this. Sure, it does not communicate any covert messages, but you do not need that. So why all the resistance? ~ J. Johnson (JJ) (talk) 23:49, 8 April 2013 (UTC)

I suspect that his "resistance" is based on the same point as mine: I do not believe that your proposal would improve this page.

I'm not going to bother explaining why and how it would be worse than what we've got, because I am invincibly convinced that you are incapable of hearing that, and therefore there's no point in wasting my time with an explanation. WhatamIdoing (talk) 00:14, 9 April 2013 (UTC)

  You disrespect me in declaring that I am incapable of hearing. In fact I heard you loud and clear just above (23:51, 4 Apr) when you claimed that 'This guideline has no business specifying what material "requires" an inline citation.' As I noted then, this definition does NOT specify what material requires inline citations. I also modified the proposed definition to explicitly mention WP:V to make it clearer. But perhaps you have a hearing problem?

  I also heard you where you oppose changing anything ("I want no change"), despite demonstrated problems with the existing use and definition of terms. Your implacable opposition is the "dead horse" here. Your real motivations are indeed a mystery to me. But not for lack of hearing on my part, but for lack of explaining on your part. You certainly have not explained why this definition would be worse. ~ J. Johnson (JJ) (talk) 22:02, 9 April 2013 (UTC)

The "quick list" is the first section of the article, titled Types of Citations. (I said that, didn't I?) That's the part of the article you're thinking of editing, am I right? You want to add a more precise definition into that section, correct? More or less obviously, the section Types of Citations is a list of the types of objects treated by this article. (Not a big stretch on that point, I think.) The question at issue is this: does it help the reader to add the definition you've composed above to this section?

Ah, it wasn't clear. I didn't see that as list, and I think "objects" is not quite the term we want here. "Concepts" seems better. -JJ

Am I wrong about any of that? (I don't like your comment that I am somehow "redefining our purpose" here.)

It seemed to me you were, but if you're not — fine. -JJ

Our purpose here is create a guideline that helps editors to add citations to articles. I think your definition doesn't do that. I think my paragraph does. I just want to communicate clearly to the reader. We need to tell them about articles like "Chain Rule", so they know they exist. The term "general references" is a means to this end. That is its only purpose. Otherwise, we wouldn't mention "general references" at all. ---- CharlesGillingham (talk) 10:23, 10 April 2013 (UTC)

I agree as to our purpose. Comments on the rest later. ~ J. Johnson (JJ) (talk) 20:48, 10 April 2013 (UTC)

I have WP:BOLDly rewritten WP:GEN, in hopes that it is less ambiguous or confusing. I avoided the term "general reference" since, really, it means nothing, and we don't need it. What's important is that some articles don't need inline references, why they don't need them, how they look, and what the risks are. ---- CharlesGillingham (talk) 15:12, 10 April 2013 (UTC)

I oppose the change that CharlesGillingham has made. It does not improve the advice to editors. It contain unstated assumptions that will not be clear to readers. The pre-existing formulation while not perfect was much better than this. I am on a real clunker of a terminal, so I will expand on this later. --Bejnar (talk) 17:03, 10 April 2013 (UTC)

For sure, Charles' changes aren't perfect, and I would have been less bold. But the previous version had problems (discussed above), so perhaps we could focus on what changes are better? ~ J. Johnson (JJ) (talk) 20:41, 10 April 2013 (UTC)

One of the problems is that it doesn't cover the common case of a general reference being used in addition to inline citations. WhatamIdoing (talk) 04:51, 12 April 2013 (UTC)

"It" being Charles' changes (which have been reverted), or the existing definition? Either way, I would say that any discussion about how so-called "general references" are used really needs a better definition. (For the reasons listed above.) ~ J. Johnson (JJ) (talk) 22:30, 12 April 2013 (UTC)

Benjar: Not clear on what "unstated assumptions" were. Can you state them?

WhatamIdoing: You are right, I did leave out the case of mixed inline and general citations. In this case, can the reader assume that the general reference is the source for "everything not otherwise cited that is likely to be challenged"? Is that how you understand it? ---- CharlesGillingham (talk) 13:00, 16 April 2013 (UTC)

If it is WP:LIKELY to be challenged, then it needs an inline citation. You can only use GENREFs for material that is unlikely to be challenged or that is additionally supported by an inline citation. WhatamIdoing (talk) 21:17, 16 April 2013 (UTC)

Are we not confounding WHAT a general reference is with WHERE it might be used and WHEN it must (or not) be used? The latter are good questions, but really should be dealt with separately. ~ J. Johnson (JJ) (talk) 23:36, 16 April 2013 (UTC)

WhatamIdoing: I thought that any material that is unlikely to be challenged does not require a citation at all. Am I wrong?

JJ: On the contrary, a "general reference" (in this guideline) is defined by where and when it is used. It is a technical term, a bit of jargon, unique to this guideline. It only describes this one particular that way that citations are sometimes used in Wikipedia. ---- CharlesGillingham (talk) 10:55, 20 April 2013 (UTC)

  I think your statement would be clearer if the emphasis was rearranged to state that a genref is "defined by where and when it is used." (Yes?) Which is the very point I just raised: so it does appear we are coming at this from different directions. And like I said below (17 April), it seems to me you are trying to define the term inductively to fit a certain context. Okay, let me ponder on that for a couple of days.

  BTW, I wonder if WhatamIdoing's concept is that "general references" are a sort of "citation-lite": better than nothing, but not sufficient where a citation is required. That seems rather dubious. ~ J. Johnson (JJ) (talk) 20:24, 20 April 2013 (UTC)

Yes, Charles, you're right: any material that is unlikely to be challenged does not require a citation at all. You may optionally provide either an inline citation and/or a general reference to a source that supports this kind of material, but it is not required. You may equally leave it blank.

The list of situations for which inline citations are required is at WP:MINREF. General references are never required. Outside of those four named situations, no form of citation is required.

Especially in a very short article, a gen ref is often much better than nothing. But it is never sufficient for those four situations for which WP:V or WP:BLP requires an inline citation. WhatamIdoing (talk) 01:27, 21 April 2013 (UTC)

W: I find myself concurring with you here. But would quibble a little bit that 1) just because a citation is not strictly required by V or BLP doesn't mean that leaving out a citation is fully acceptable, and 2) to the extent that a so-called "genref" is merely an incomplete citation, being better than nothing is also short of fully acceptable. However, I would point out that we are off-topic here. ~ 22:24, 21 April 2013 (UTC)

Comparison of proposed and alternate definitions

Charles: In your comments ca 01:41 8 Apr you state that my proposed definition "does not communicate", is unreadable, confusing harmful, etc. This rather bewilders me, as you have not explained any of these points, only offered an alternative definition. So let us examine your alternative. It starts by describing some articles, such as use a single source, "typically on narrow academic subjects", where "inline citations are often not necessary". All this describes where examples may be found, but does not identify which instances are general references, or something else. I could easily write "some" article on a narrow academic subject, not requiring (per WP:V) inline citations, and yet with a distinctly non-general reference.

The operative part of your definition is this: "Specific page numbers are usually attached to the full citation at the bottom of the page. This guideline calls these citations 'general references.'" It is a fair reading of this to conclude that general references are full citations at the bottom of the page with attached page numbers. An immediate objection to this is: are these page numbers the specification that locates the material cited within the source? Or are they the page range that locates the source within a larger work (such as an article within a journal)? (Of course you recall the discussion about this at WT:Citing sources/Archive 34#Page numbers in the References section.) This language also leads right to the same "arguable" weakness of the current definition, in that it suggests that the presence or absence of "page numbers" determines whether a "full citation" is a general reference.

Your definition does not explain (let alone define) what a general reference is; it only gives us a few clues that are weakly correlative, and unhelpful and confusing as often as not. My definition is an explicit definition, based on the common and standard meaning of general, and avoids ambiguous correlations. ~ J. Johnson (JJ) (talk) 20:15, 11 April 2013 (UTC)

I hope you understand that I have, all along, been arguing against the need for definitions. The problems with your definition above are a side issue for me. If the text of the guideline is written well, there should be no need to define anything. If confusion arises, it should be dealt with at the point in the article where the confusion arose.

Having said that, and if we must have a definition, this is closer I think:

A general reference is a citation that which identifies a source for any material in the article that does not have an inline citation and is likely to be challenged. General reference typically appear at the bottom of the page. ... blah blah etc.

Poorly written, I guess, but correct, I think. What do you think?

(I still think that, without knowing what kind of articles use general references, it's easy to think you can just add a general reference to any old article and still preserve text-source integrity. This isn't the case.) ---- CharlesGillingham (talk) 14:16, 16 April 2013 (UTC)

  So how can any guideline be considered well written if there is not a clear and unambiguous understanding of the key terms? And where ambiguity and confusion abound, how else can we proceed without explicit definition? You state that any confusion "should be dealt with at the point in the article where the confusion arose." The problem with that is we then get local definitions, often randomly variant. It would be like when every railroad had its own guage, or own time. Like I said above (at #Reference vs. citation), we need good definitions that can be used across the entire project.

  I think your redefinition suffers from the same problems as before. (E.g., why should a citation be "general" based on whether the material is likely to be challenged?) But your comment reminds of your earlier comment that we "need a definition that describes the context as well as the term". Indeed, it seems to me that you (and W) have been trying to define genref by its context. (Possibly definition by induction?) Perhaps we should open another section to examine that approach? ~ J. Johnson (JJ) (talk) 00:08, 17 April 2013 (UTC)

argument from authority

Collapsed off-topic ramble

this 2010 post brings up a good point, the policy of citing only sources from 'authorities' on matters (we can't even cite newspapers, we have to have 'experts' interpret them for us to be able to refer to them) is pretty much a massive argument from authority experiment. For Wikipedia to be neutral on issues, the ideal stance would be that nobody is an expert until they are proven to be experts. Really to call anyone an expert is an interpretation, and if Wikipedians are not supposed to independently, we reach an impasse on defining expertise. Who has the authority to define who is an expert and who is not? Other experts? Who defines them? The loop we go on here should be obvious, inevitably we are going to be making assumptions about people's expertise if we define ANYONE as an expert. The same with 'reliable source'. The default assumption, just as we assume people to be non-experts until argued to be expert, is that sources aren't reliable until argued to be reliable. But I wonder, does our establishment of reliability rely on subjective opinions from Wikipedians, or of 'experts', in determining that status? If a source is truly notable and verifiable, every source we use would have their own Wikipedia article. I question: how do we allege that a source is reliable, that a person is an expert in their field, if they aren't even notable enough to have an article? In many cases we cite the names of people who write papers who don't have Wikipedia articles. I do not assert that every expert has a Wikipedia article, that would be silly, but I think if we are to use people as experts, we should make it a priority to make articles about them by redlinking a researcher's name in a reference tag whenever it is used. If someone is expert enough to be used as a citation, surely they are notable enough for an article. It is a bit of a double standard if we assert "X is noteworthy enough to base articles on their writings" but not "X is noteworthy enough to base articles on them". The problem with appeal to authority is basically that authority is not something everyone agrees on, and by saying 'X is authority' we are deviating from neutrality. 'X is considered authority by Y' would be more neutral, for example. It gets worse when we simply present claims by authorities as facts by wording their claims as truth. Surely there should be more rigorous criteria for fact. Even for widely agreed-upon facts with consensus we can briefly note whom it is a consensus among. This gives context to claims, so they will be more transparent. I mean sure, in reference brackets we link to external links and stuff, but wouldn't it be ideal if we could include internal links as well? External links are not neutral and open to criticism the same way internal links are via their articles talk pages. Ranze (talk) 20:43, 24 April 2013 (UTC) You're on the wrong page. This page is about how to format the bibliographic citations for sources that you've already decided to use, as in "Where does the author's name go?" or "Do I need to include page numbers?" Your questions are answered at WP:V (the WP:SPS section defines expert) and WP:RS. Also, we do cite newspapers; in fact, it appears that newspaper articles have been cited more than half a million times on Wikipedia. WhatamIdoing (talk) 22:28, 24 April 2013 (UTC)

Citing Film/TV/Music Video

On the article, it says cite "approximate time at which event or point of interest occurs, where appropriate". Can someone please tell me what the tag is to cite the exact time in the TV/Music video production? I looked at WP:CITET and I don't see it. Thanks. Lightspeedx (talk) 08:18, 29 April 2013 (UTC)

You need the bizarrely titled {{Cite AV media}}, and you can set the time= parameter. Betty Logan (talk) 00:32, 2 May 2013 (UTC)

A general reference is a reference to a reliable source

SV wrote in the edit histoy "rv restored "citation"; otherwise it's not clear (a reference is a reference)" -- Not what it say: "A general reference is a reference to a reliable source ..." is not a defining reference but is defining a "general reference" and so the sentence is not stating that "a reference is a reference".

Using citation here ought to be avoided as it can be read to imply that general references are acceptable as citations -- they are not. -- PBS (talk) 13:52, 26 April 2013 (UTC)

General references are acceptable as citations if the article is short enough, and the list of general references is short enough, that the reader can associate the material in the article with the source. Jc3s5h (talk) 14:44, 26 April 2013 (UTC)

For the, what, fifth time now?, let me paste the definition of the term citation from the top of this guideline onto the talk page, in the hope that you will read it and understand it:

"A citation, or reference, is a line of text that uniquely identifies a source:

Ritter, Ron. The Oxford Style Manual. Oxford University Press, 2002, p. 1.

A general reference is a citation. It is not an inline citation, but it does count as "a line of text that uniquely identifies a source", which we define in the very first sentence of this guideline as being "a citation". SV is therefore correct. In case you forget that we have defined citation this way in the future, I wish you luck in remembering that this term is defined at the very top of the guideline, so that you can go look it up instead of making the same mistake over and over and over. WhatamIdoing (talk) 15:49, 26 April 2013 (UTC)

First I would remind you that until recently there was a definition for general reference which was defined in the first section which has now been changed. Second, the change I made did not say that a general reference is an inline citation it said that a "general reference is a reference to a reliable source" as a reference and a citation are both defined at the top as "is a line of text that uniquely identifies a source" the change I made removes potential confusion over whether a general reference is an acceptable as a citation -- it is not because it lacks text-source integrity. Jc3s5h It is no use saying "General references are acceptable as citations if the article is short enough" because without reading the reference (which may not be to hand) or looking through the history of an article it is impossible to tell if the current text in an article is all covered by the general references. What happens if there are two general references how does one know which facts are covered by which general reference? What about three? four etc. General references have been depreciated as a support for verification for some years now because of the problems with text-source integrity (which is why WP:CHALLENGE requires inline citations. -- PBS (talk) 22:42, 27 April 2013 (UTC)

A general reference is acceptable as a citation. By definition. A general reference is a citation. When something is something, by definition, then that something is acceptable as something. A red apple is an apple; a red apple is acceptable as an apple. The fact that green apples and yellow apples also exist does not make red apples unacceptable.

You are correct that a non-inline citation is not acceptable as an inline citation, but that's irrelevant. WhatamIdoing (talk) 22:59, 30 April 2013 (UTC)

I think Phil is saying (in part) that "general references" are a subset of all possible "references". But I think we should drop the "to a reliable source" qualification. Of course we expect reliable sources, but that is covered in WP:RS, and need not (should not!) be duplicated here. It also opens up that bucket of worms of what to call a reference otherwise "general" that points to an unreliable source.

While we are here: how about if we replace the "line of text" with "a description"? Of course the description is textual (has anyone ever suggested that a reference could be a picture of a book?), and certainly not limited "a" single line. ~ J. Johnson (JJ) (talk) 19:58, 26 April 2013 (UTC)

Yes I am saying that "general references" are a subset of all possible "references", I think that theis guideline should not drop the "to a reliable source" qualification from this guideline as it is based on WP:SOURCE not WP:RS. It calls for the "general references" to an unreliable source either to be removed from the article, or moved into "Further reading" or "External links". -- PBS (talk) 22:42, 27 April 2013 (UTC)

Yes, and now can you bring yourself to admit that general references are equally a subset of all possible citations? WhatamIdoing (talk) 22:59, 30 April 2013 (UTC)

Possibly no one objects to replacing "line of text" with "description", so I may screw up my courage and do that. As to "reliable source": I agree that an unreliable source (though this depends on the use) should be removed. But I don't think it is the role of this guideline to enforce that requirement. (And regardless of whether the requirement is derived from WP:SOURCE or WP:RS.) It could also be said that references must be written in English (this being the English Wikipedia), else be removed, but we don't explicitly incorporate that into the definition of "reference". And as I have said before, if the definition of "reference" depends on it being to a reliable source, then we have to evaluate that source to determine what is a "reference" (general or otherwise). Sure, we may have evaluate it to determine whether to keep it or not, but we should not have to do so in order discuss format, placement, etc. It is not useful, and causes confusion, if we mix what a reference is with the requirements on using it. ~ J. Johnson (JJ) (talk) 21:02, 28 April 2013 (UTC)

The line of text was fussed over a while back. I don't think that description is actually better. "A mass-market paperback with a red cover" is "a description", after all. WhatamIdoing (talk) 22:59, 30 April 2013 (UTC)

  Of course the domain of all possible descriptions is much vaster than those that constitute valid references. So what? The domain of all possible "lines of text" is even greater. In either case the range of possible instances is greatly reduced by the condition "uniquely identifies a source". The problem with "text" is that it implies free-form narrative text (such as the text of the article), whereas citations/references are highly structured to be more economical. E.g., we do not say: "This source is a book, the author is Jones, the year of publication is 2000, the title is 'Big Blue Book'", etc. We just say: Jones, Big Blue Book, 2000 ...." Furthermore, the "line of" adds no useful information at all. Does "a" line of text mean only one line? Is that a "physical" line (terminated at the margin of the page)? Or is it a "logical" line, terminated by some special "end of line" notation, such as a period? Oops, can't use a period for eol, as some styles of citation use that internally.

  QED, "line of text" is a can of worms. I do think "description" is better. As it appears no one else objects (aside from WhatamIdoing, who objects to all changes), I am inclined to make this change. ~ J. Johnson (JJ) (talk) 01:23, 1 May 2013 (UTC)

J. Johnson wrote "The problem with 'text' is that it implies free-form narrative text (such as the text of the article), whereas citations/references are highly structured to be more economical. E.g., we do not say: 'This source is a book, the author is Jones, the year of publication is 2000, the title is "Big Blue Book"', etc." Well, sometimes we do incorporate the source in the narrative. For example, "President Obama in his 2013 State of the Union address claimed...." It is frequently done in articles about movies and books, where all the necessary bibliographic information is given in the body of the article, and some of claims in the article are obviously supported by the work that is the subject of the article. But that doesn't really weigh on whether "line of text" or "description" is better.

About the "to a reliable source": There should be neither general references nor inline citations to unreliable sources. You might have a bibliographic citation to an unreliable source under ==Further reading== (and one might also be listed under ==External links==), but you shouldn't encounter unreliable sources under ==References==. WhatamIdoing (talk) 23:16, 30 April 2013 (UTC)

Do you not understand what I said? Of course there should not be references/citations to unreliable sources. But if "a line of text" is a reference (of any kind) only if it points to a reliable source, then we can't determine whether reference-like lines of text are truly "references" until we evaluate the source. Which is likely not at hand. And then there are sources which are reliable in some regards, but not in others, so the reference-like lines of text might be references, or not, depending on what they are being used for. You are trying to make the definition do more than define; the result becomes indeterminate, and even absurd. A definition should stick with definition, and not try to enforce other policies. ~ J. Johnson (JJ) (talk) 01:29, 1 May 2013 (UTC)

Nope. A citation is the line of text that describes some publication or another. It does not matter in the least whether that publication is useful, reliable, relevant, or garbage. Stop thinking "a citation is something that properly supports material in an article". That's not our definition (nor, in fact, is it anyone's definition, since style manuals don't prohibit people from citing nonsense). A citation is an entry in a list of publications, or some other method of telling people "Look, this publication, not just any old publication". WhatamIdoing (talk) 04:01, 1 May 2013 (UTC)

"It does not matter in the least whether that publication is useful, reliable, relevant, or garbage"? Ah, W, isn't that essentially what I said? Are we actually in agreement on this? ~ J. Johnson (JJ) (talk) 23:09, 1 May 2013 (UTC)

Can someone sum up what is wrong with "a citation .... is a line of text that uniquely identifies a source"? SlimVirgin ^(talk) 02:16, 1 May 2013 (UTC)

I think nearly everyone will understand what the guideline is getting at. But strictly speaking a citation is not necessarily a line of text. It may be several lines of text. It may be spread around, with the page number in an inline superscript and the title in a full citation. It may even be implied, as an article about a book that contains descriptive statements about the book, which can be verified by reading the book. Jc3s5h (talk) 02:38, 1 May 2013 (UTC)

Slim, didn't I do just that? (Scroll up to my comment of 1 May.) I point out that part of the problem is just as Jc3s5h just said, that a citation in the fullest sense can be a combination of parts. But even if we restrict this to just a full reference (aka "entry in a ref list"), the nature of the entry is not that it is textual (we assume that), but that it is descriptive. The "line of" serves no purpose whatsoever, and the "a" suggests restriction to a single line of text. If so, then how do we define "line of"? ~ J. Johnson (JJ) (talk) 23:12, 1 May 2013 (UTC)

I suppose the point is that the guideline is written for people who may have no idea what a citation is. To argue about "a line of text," because it might be two lines, seems to me to be dancing on the head of a pin. But I don't mind if we say instead: "a citation .... uniquely identifies a source."

If you want to be very precise and very succinct, that's great, but given that a lot of the rest of the guideline is a mess, there are other sections we could be making more succinct with some urgency. SlimVirgin ^(talk) 00:25, 2 May 2013 (UTC)

  I agree that other parts need work. But this is the basis of the rest; we need to have a good grasp of the concepts and terms before we can use them.

  My proposal is to replace "line of text", which really adds nothing to the definition, with "description". As an alternate proposal I suppose we could just leave that out, without explaining (at this point) just how "a citation uniquely identifies a source." Anyone have an objections to that? ~ J. Johnson (JJ) (talk) 21:28, 2 May 2013 (UTC)

Here's a "description" of the book I just put down: The book is a hard cover, almost an inch and a half thick. The cloth is a faded red. There's a dust jacket with red and white diamonds on the spine.

Here's another "description" of the same book: The protagonist is depressed loner who is making plans for a dramatic suicide, except that the dust jacket copy says he'll decide against it. I can't decide whether the author likes him. I also can't decide whether it's worth my time to find out.

I don't think that's the sort of "description" you have in mind.

Removing line of text means that we no longer have a definition ("A citation is this thing") but instead proceed directly into a description of what it does ("A citation serves this function"). That usually produces a steady level of complaints from inexperienced people, which we may deem unimportant, but it also makes it harder to deal with people who have trouble believing that the bit of text in the {{quotation}} box is actually called "a citation", and who misinterpret major policies as a result (e.g., declaring a page that contains a list of citations to sources actually used to create the article content to be {{unreferenced}}). We need a plain, indisputable definition that this thing—the semi-standardized line of text that reports the basic facts of author, title, date, etc.—is called by these two names on Wikipedia. It's not clear to me whether that needs to be in the first sentence, but it is a reasonable enough place.

Also, specifically for the purposes of this particular guideline, I'm not sure that an in-text description of a source is "a citation". Something like "According to Alice Expert in her 2012 book, Important Facts..." is technically an inline citation (and is even officially preferred by some major academic style for certain famous works, e.g., Biblical quotations and Plato's works), but it's not the kind of citation that this particular guideline deals with. It is easy enough to say that this isn't a (separate) line of text, and therefore not what we're talking about, but it would certainly be "a description", and even "a description that uniquely identifies a source". WhatamIdoing (talk) 01:09, 4 May 2013 (UTC)

Since this guideline allows any consistent style, and calls out by name several printed style guides, some of which allow the entire citation to be in the running text (for certain kinds of works), I would say that such a citation is specifically covered by this guideline. Jc3s5h (talk) 01:59, 4 May 2013 (UTC)

Yes. ~ JJ

That would certainly be an inline citation that is sufficient for WP:V's purposes, but I'm not sure that it's what we're talking about here. Most of this guideline is about how to format list entries. It's about page numbers and publishers and links. Some of the bits that aren't about that probably don't belong here at all. WP:NOCITE, for example, belongs on some other page entirely. So I'm not sure that the core of this guideline is really about "citations" that amount to nothing more than typing "In Plato's Republic..." WhatamIdoing (talk) 20:13, 4 May 2013 (UTC)

I would say the guideline repeats information that could be found in most printed style guides, perhaps for the benefit of people who don't own one. It also suggests how to deal with the the fact that Wikipedia and many of the sources used are electronic, while some printed style guides are oriented to writing paper papers and citing paper sources. How to cite a source by describing it in running text is certainly within the scope of a citation style guide. Jc3s5h (talk) 15:31, 5 May 2013 (UTC)

W: your descriptions do not uniquely identify a book. (Unless you add something like the red book "next to my desk". Which, unless you are acquainted with the books next to my desk, does not help you find a copy.) But, as Jc3s5h says, there is no prohibition against narrative descriptions. In fact we do prefer, for reasons of economy and clarity, a briefer, more structured textual form, and this can be addressed in the guideline. As you have said yourself (do we still agree on this?), it does not matter whether the source is "useful, reliable, relevant, or garbage"; I say that the form of how this unique identification is made is also irrelevant to the definition. The various requirements, caveats, recommendations, etc., need to be explained, but not in the definition itself. ~ J. Johnson (JJ) (talk) 19:45, 5 May 2013 (UTC)

When and why to cite sources

Consider this sentence in an article I might write: [[George Washington]] was the first President of the United States.

The "When and why to cite sources" states: By citing sources for Wikipedia content, you enable users to verify that the information given is supported by reliable sources, thus improving the credibility of Wikipedia while showing that the content is not original research. You also help users find additional information on the subject; and you avoid plagiarising the source of your words or ideas by giving attribution.

But that is exactly what the link to the Washington page does. My adding a citation in such cases is bloat, redundant, clutter, obscuring citations that are significant. I suggest the "When and why to cite sources" include a description of linked pages as sometimes being sources.50.136.247.190 (talk) 04:29, 2 May 2013 (UTC)

The test of whether a citation is necessary is whether the material is challenged or likely to be challenged. In this case, the material is not likely to be challenged because it can be trivially verified (including by checking the sources at George Washington or Presidents of the United States). Thus there's no need to add a citation. —me_and 18:25, 2 May 2013 (UTC)

Technically, you also need an inline citation if the text is a direct quotation or if it is about a living person. Neither of these apply either, so no citation is needed. (Full list at WP:MINREF for the curious.) WhatamIdoing (talk) 19:31, 2 May 2013 (UTC)

Strictly speaking a wikilink is not a citation, in part because WP does not cite itself. But linking to (say) the article on George Washington would cover for the kind of "obvious" information such as being the first President (see WP:BLUE) that is generally not cited. ~ J. Johnson (JJ) (talk) 22:15, 3 May 2013 (UTC)

List of cities in China by urban population

Could someone have a look at the recent edits at the above article. Recent reverts have resulted in cited material being removed, but I don't want to get involved in an edit war. Thanks. Eldumpo (talk) 17:47, 4 May 2013 (UTC)

marking dead sources

Wouldn't it make sense to have a template to mark a source as dead link? I often follow a source and the page got deleted or so. However, I don't want to follow the steps to tread a dead source immediately, so I'd like to mark the source as dead at least. There could also be bots doing this. --88.73.110.22 (talk) 20:31, 18 May 2013 (UTC)

I think there already is and there already are. -- Alarics (talk) 20:34, 18 May 2013 (UTC)

{{Dead link}}. It's listed at the end of step #2 in WP:DEADREF. WhatamIdoing (talk) 19:40, 19 May 2013 (UTC)

RfC on new library search tool for Wikipedia

We have a new tool, Forward to Libraries, which helps readers find books at their local library related to the articles they are reading. The tool can also be used by editors to cite reliable sources. There is an RfC at Wikipedia:Village pump (proposals)#Linking subjects to books at your local library (Forward to Libraries) to determine how this tool should be used on Wikipedia. Interested users may wish to comment there. 64.40.54.57 (talk) 01:19, 20 May 2013 (UTC)

Inconsistent date presentation in cite templates

Users and developers of citation templates, please see my question at Help talk:Citation Style 1/Archive 3#Date location. Jc3s5h (talk) 12:27, 22 May 2013 (UTC)

Distinction of "full" and "short" citations

If definition of reference/citation as identifying a source should keep (I am not entirely happy with "uniquely", but that can wait), we might further distinguish, in accord with probably all authoritative sources, between "full" and "short" citations/references in accordance with their scope as follows:

• A full reference contains the full bibliographic details for locating and identifying a source in the world at large.
• A short reference (or short citation or short cite) contains enough information — often only the author's last name and the year of publication — to identify within an article the corresponding full reference.

Presumably these would be followed by examples. ~ J. Johnson (JJ) (talk) 20:09, 5 May 2013 (UTC)

Is it your intention that these replace the existing definitions, which are:

• A full citation fully identifies a reliable source and, where applicable, the place in that source (such as a page number) where the information in question can be found. For example: Rawls, John. A Theory of Justice. Harvard University Press, 1971, p. 1. This type of citation is usually given as a footnote, and is the most commonly used citation method in Wikipedia articles.
• A short citation is an inline citation that identifies the place in a source where specific information can be found, but without giving full details of the source – these will have been provided in a full bibliographic citation either in an earlier footnote, or in a separate section. For example: Rawls 1971, p. 1. This system is used in some articles; the short citations may be given either as footnotes, or as parenthetical references within the text.

I'm not wild about the existing definitions, which seem to wrongly preclude using a short citation to refer to a whole work, as well as using the word where (a reference to a physical location) instead of if, but I'm not sure that yours are perfect, either. WhatamIdoing (talk) 20:12, 6 May 2013 (UTC)

Yes, as to replacement. Nor am I certain my definitions are perfect, but I think pretty close, and certainly better than the current defs. One problem is that (as we have discussed before) putting extraneous material, such as examples and other conditions, into the actual definition, tends to distract from the essence of the def. This could also be said of the hyphenated part of my second definition. That short cites are overwhelmingly (though not universally) in the author-date form seems to justify directing the reader's attention to that right from the start. I am not entirely easy with that, but uncertain how else to handle it. ~ J. Johnson (JJ) (talk) 21:16, 6 May 2013 (UTC)

That's actually a problem with your definition: short citations do not have the single function of identifying the full citation. They normally also provide greater specificity, e.g., the exact page number that the material can be found on. WhatamIdoing (talk) 20:41, 7 May 2013 (UTC)

Please pay closer attention: my definition does NOT say that a short cite has the single function of identifying the full citation. Whether the specification of where the material is found within the source (e.g., a page number) should be considered an integral part of the short cite (e.g., everything in the brackets: "[Ritter, 2001, p. 1]") or simply an addition (e.g.: "[Ritter, 2001], p. 1"), is a separate issue, which is not decided, need not be decided, and which my proposed definition does not attempt to decide. How a short cite should be used, where, in what format or style, etc., etc., are neither needed nor useful to the definition itself; such details can be addressed in the subsequent text. ~ J. Johnson (JJ) (talk) 22:02, 7 May 2013 (UTC)

When we say "A short citation contains enough information to identify the full citation", some people will read this as meaning "A short citation only contains enough information to identify the full citation". This is wrong; this is not what you mean. However, it is what a non-trivial minority of people will believe was intended.

Policy writing is hard. It's not enough to say what you mean, but you also need to avoid as many predictable misunderstandings as you can. This is a predictable misunderstanding, and your proposal does not avoid it. WhatamIdoing (talk) 00:47, 8 May 2013 (UTC)

If you want to complain that some people — apparently including yourself — are so inattentive as to misread plain text, sure, that is a problem. But that problem is NOT with the definition. Do we need to handle the problem of people wildly misinterpreting? Sure. And also world hunger, global warming, etc., but just as every possible problem of the world should not be dealt with here, neither should a definition deal with all possible misinterpretations. That, as well as caveats, explanations, and examples, follows in the explication. Trying to put everything that can be said about something into its definition does not avoid misunderstanding, it creates it; it confuses the definition with issues that are remote and peripheral. Definitions need to be succinct. Mine are, yours are not. ~ J. Johnson (JJ) (talk) 19:05, 8 May 2013 (UTC)

Brevity is not a virtue in a definition when that brevity leads to misunderstandings. WhatamIdoing (talk) 16:13, 10 May 2013 (UTC)

  So you would require that a definition cover, in the very first sentence, EVERY conceivable misintepretation that some inattentive editor can squeeze out of it? Then you really ought to object to "a citation ... is a line of text" because, hey, "a" means one, and if I squeeze my browser window down to 40 characters your Ritter example wraps around to a second line. Therefore it is not a citation.

  And this from the same editor that rejected a hypothetical example because it wasn't short enough. You complain of a possible misunderstanding when the language is brief, and when a point is covered you complain of lack of brevity. That, and your persistent misreading of what is actually said, is not good-faith discussion, that is obstructionism. ~ J. Johnson (JJ) (talk) 23:02, 10 May 2013 (UTC)

Since this keeps coming up, I guess it's worth explaining: A line of text is defined by a line break (computing) or a "hard return", not by how line wrapping makes it fit on your device. Your entire comment immediately above this reply is two lines, no matter what size screen you view it on. It is extremely unusual for editors to use a multi-line citation style on Wikipedia. WhatamIdoing (talk) 19:38, 19 May 2013 (UTC)

  No, no, that is not what the existing definition says, that is not it at all. That is only your personal interpretation of you think it should mean.

  But even if (in the spirit moving the discussion forward) we take your new and (in this context) novel interpretation as a proposed definition, it simply is not adequate. And even incorrect. E.g., the "line of text" that WP delivers to our browsers (quite aside from any matter of line-wrapping) is delimited not by "hard returns", but by <dd> tags. And as any "hard returns" (did you mean ASCII "end of line" chrs.?) are ignored, their presence is entirely irrelevant.

  Did you perhaps mean the use of newlines in the specific context of Wikimedia edit boxes? Well, that is not what you said (need I remind you of your own comment that brevity is no good if it leads to misunderstanding?), but in the interest of salvaging something useful from your explanation let us presume some textual epicycle that specifies "as mediated by the Wikimedia software". Does that help us?

  Not really. In the Wikimedia context newlines delimit entire paragraphs. In that sense I would definitely agree with you that "multi-line" (= multi-paragraph) citations are "extemely unusual". Indeed, I rather doubt there ever has been a single instance of such. And being so extremely rare why are we wasting bits specifying "a line" of text? There is utterly no showing that "a line" is useful, let alone necessary. (For that matter, even "text" seems quite unnecessary.)

  But wait, line breaks are allowed inside templates. E.g.:

{{Citation |last1= Smith |first1= James ... }}

  In this usage we do have a multi-line citation. Which your definition, and even the current definition, exclude. And to think that you accused me of being radical!

  There are other issues (like, can a line have more than one citation?), but, as I have shown, your attempt to salvage "a line of text" requires so much additional definition that it is really not worth the effort. This useless, meaningless, quite unnecessary qualification is dead. Say goodbye, and let's take it off of life-support. ~ J. Johnson (JJ) (talk) 21:42, 20 May 2013 (UTC)

The citation templates all produce a single line, no matter how they are formatted in the current (and soon-to-be-replaced) edit window. The citation is what that the reader sees, not what the editor types. What the reader sees is almost always one line of text (which may wrap a dozen times, depending on the screen size/font size/layout choices). WhatamIdoing (talk) 00:35, 21 May 2013 (UTC)

  I belief Visual Editor is an alternative to the edit window, not a replacement. But no matter, it only buries actual ASCII eol chrs even deeper in irrelevance. And what the various citation templates produce is a string of HTML "text", usually not a line delimited by eol chrs.

  So now you want to define citation as "what the reader sees", but ignoring all formatting aspects, which you say "is almost always one line of text". But this takes us back to the start: what is a "line" of text? "Hard returns", <dd> tags, etc., are not visible to the reader; all that s/he has to go on is the formatting. Which you disallow.

  In summary: citation as a "line of text" has never been adequately defined, and, as I have shown, is useless. I suggest we let that go, and define citation in terms of its core purpose or essential characteristic. E.g., something like:

* A citation connects material in an article to a source.

  And I will head off your many objections that this core definition does not include or exclude all of the specific or rare cases you want to address, nor address other requirements or ways that inattentive editors might misconstrue it, by saying that the core concepts should be kept simple, with any necessary explanations added at appropriate levels. ~ J. Johnson (JJ) (talk) 22:55, 21 May 2013 (UTC)

Nope: As you've been told repeatedly, an inline citation connects material in an article to a source. The citation itself is the text that tells you what the source is. It is still a citation even if (as in the example at the top of this guideline) it is wholly unconnected to anything. WhatamIdoing (talk) 23:31, 23 May 2013 (UTC)

  What is "inline" (that is, in the text of an article) is typically only a bracketed number, which connects to note (<ref>). Or what's "inline" could be a short citation (e.g.: "Kerr 1990"), which does not connect to a source (insufficient info to resolve), but only to a full citation which (hopefully) has sufficient information to identify and locate the source. Unless your "inline citation" incorporates the entire "citation" (as you seem to mean it) "inline", it is incomplete, and it does not connect to a source. And if you want quibble about "connects" please note that the same objection can be said for "uniquely identifies", and even "reliable source".

  Your problem here (aside from an inability to pay attention) is restricting "citation" to a thing, the piece of text giving the "full bibiliographic details", such as I have defined at the top of this page full citation (or full reference). You leave off all the other pieces (and then have to stick them in however you can, to never ending confusion). I think it would be best to define citation broadly to encompass the entire state or process of connecting material to sources, and then we will have a consistent basis for fitting everything together. ~ J. Johnson (JJ) (talk) 21:17, 25 May 2013 (UTC)

By my count "a line of text" has been inserted into the definition of citation four times, NEVER with any explanation of what the need or benefit is, or what it adds, nor any satisfactory explanation of what it means. The inadequacies of this term have been laid out by Jc3sh5 and myself, without any response that can justify the term.

Slim Virgin: you have twice restored this term, without discussion. So I call on you to pay up your arrears of discussion and justify why citation should be defined (in part) as "a line of text". ~ J. Johnson (JJ) (talk) 21:12, 23 May 2013 (UTC)

I see no discussion supporting the inclusion of "a line of text". Not now, when discussion is requested, nor any of the four times this phrase was inserted into the page. In that it serves no useful purpose, nor has any definable meaning, does anyone else concur that it should be removed? ~ J. Johnson (JJ) (talk) 21:34, 25 May 2013 (UTC)

In-line citations and general references

Hello,

in Talk:Maria Sharapova/GA5, the reviewer failed the article because he felt there should be in-line citations, but there is already a general reference which supports certain information. Could you say who was right and who not? If he was right, why should there exist redundant in-line citations which supports the same information as the official site? Thanks. Regards. --Tomcat (7) 10:40, 22 May 2013 (UTC)

The article is being worked on today, and I'm not going to waste my time making remarks that will be outdated in a few minutes. Therefore I will discuss this version from May 19. There are a great many inline citations, and one general reference. So a reader, seeing a fact that he wants to check or explore in further detail, has to decide whether to look in the general reference, or maybe there is a footnote at the end of the paragraph that might or might not be the source. So there are likely to be two sources to choose from. That isn't that bad. Also, a reader could probably guess what kind of data could be found in a WTA (World Tennis Association?) profile, and would be able to make a pretty good guess about which facts in the article came from that profile. So in this case, I wouldn't have failed the article. Jc3s5h (talk) 11:59, 22 May 2013 (UTC)

The justification for "general references" usually hinges on an article being a stub; there seems to be an implicit view that where an article is short and relies on only one source the editor shouldn't have to go to the trouble of an explicit in-line citation. But this case is a fully-developed article. So is there any reason why this "general" reference cannot be explicitly cited? ~ J. Johnson (JJ) (talk) 21:44, 22 May 2013 (UTC)

The reviewer needs to follow WP:GACR. He may rightly object to any specific, identifiable statement that includes:

• direct quotations,
• statistics,
• published opinion,
• counter-intuitive or controversial statements that are challenged or likely to be challenged, and
• contentious material relating to living persons.

The reviewer should be able to say something like, "The sentence about the BLP being a mutant lizard person from Mars is 'contentious material relating to living persons' and therefore must have an inline citation per GA criteria." He may not object to the mere existence of a general reference, and he may not object to any sentence that doesn't fall into one (or more) of these five categories being wholly unreferenced. You may wish to read WP:GACN. WhatamIdoing (talk) 23:38, 23 May 2013 (UTC)

I believe your argument is that in any instance where there is no requirement for a citation, there is no requirement for a complete (inline, perfect, etc.) citation. Does this also relax the requirement that sources be reliable?

However, you miss my point. I was asking: is there any reason this general reference cannot be explicitly cited? ~ J. Johnson (JJ) (talk) 20:34, 24 May 2013 (UTC)

I think Johnson is missing the point of a good article evaluation. It needs to be adequate, not ideal. Maybe it would be a slight improvement to cite certain facts inline to a source that at present is only a general reference. But if there no claims in the article that lack a required inline citation, then the presence of a general reference is not a reason to fail the article.

Also, people sometimes complain about overcitation. If it is apparent that certain facts could most likely be looked up in the general reference, it could be seen as overkill to insert a dozen inline citations to this source. Jc3s5h (talk) 00:58, 25 May 2013 (UTC)

This debate reminds me of the scene in Catch 22 where an operations officer having briefed the bomber crews asks "any further questions" and members of the crews start to ask questions about all sorts of things but not about operation. What he meant of course was "any further questions about the operation?". It seems to me that support for the use of general references is similar to the "mistake" that the operations officer made. -- PBS (talk) 10:41, 25 May 2013 (UTC)

A general reference is one which was used to obtain or confirm claims in the article for which no inline citations were provided. A full bibliography entry that was included only as the target of a footnote number or shortened citation is not a general reference. We've had trouble finding a satisfactory phrase for what it is. Jc3s5h (talk) 12:25, 25 May 2013 (UTC)

We can call a "A full bibliography entry" a "reference to a source". A general reference can not usefully be used to "to obtain or confirm claims in the article for which no inline citations were provided" either the information is general knowledge or the claim needs a inline citation for verification. -- PBS (talk) 11:04, 26 May 2013 (UTC)

Sometimes a general reference can be used to confirm a claim that is not general knowledge. If the claim is obviously one that pertains to a certain field, and the general reference is a survey of that field with a good table of contents and a good index, the reader can figure out that is an appropriate source to try, and the index and TOC will let the reader find the information about the claim. It isn't as definite or easy as an inline citation, but it isn't useless either. Jc3s5h (talk) 11:34, 26 May 2013 (UTC)

In such circumstances a general reference can not meet the verifiability policy requirements, so from the point of view of meeting policy requirements it can not be used to "confirm a claim that is not general knowledge" even though the claim may be generally known by experts in that field that the general reference covers. -- PBS (talk) 19:41, 26 May 2013 (UTC)

In my view PBS has lost this policy fight but refuses to retire from the field, so I will just ignore PBS. Jc3s5h (talk) 19:58, 26 May 2013 (UTC)

The only text that does not need inline citations is text that does not need general references. In practice, given the widespread distribution of English speakers who's general education varies from country to country, most "common knowledge" needs a citation as it is really not that common at all (this is quickly shown if one plays the game of trivial pursuits in a country other than ones own). -- PBS (talk) 10:41, 25 May 2013 (UTC)

A general reference is too loose a phase and they are frequently useless for fulfilling the requirements of the Verification policy. If a book has multi-volumes, then it may well be a general reference, but it is next to useless as a verifiable citation. For example Wikipedia have about 10,000 articles were {{EB1911}} is given as a source with no other details. On most of the pages where it sits it is a general reference (often the only reference) to the subject of the Wikiepdia article, but without volume and article it is as good as useless as a citation to support the text as what is being stated is "that somewhere within the 29 volumes of the Encyclopaedia Britannica (11th ed) there is something on this subject which supports part or all of this Wikipeida article, but we are not going to tell you where in the EB it is, or what parts of the Wikipedia article it supports".Most of these 10,0000 articles have text copied from EB1911 and will need inline citations as they are in breach of the Plagiarism guideline without them. -- PBS (talk) 10:41, 25 May 2013 (UTC)

As soon as two or more general references are added to a Wikiepdia article, Text source integrity is lost and so general references do not meet the Wikiepdia verification policy requirements. I think it makes more sense, as it is clear to the reader of the article, to use inline citations with a general references section if needed support inline citations, and move "general references" which do not support inline citations, into a "Further reading" section. -- PBS (talk)

There some interesting issues here, including whether a citation that is not required is also not required to be good, complete, etc. However, you all have missed my point. Note it is perfectly possible to cite "inline" a reference that is used generally. (E.g., add a note: "Smith 2000 is used generally.") What I have asked — and quite aside adequacy versus ideal, or however one feels about "general references" — is whether an objection to a "general" reference cannot be answered by simply adding a suitable inline citation.

PBS: I think you and I are largely in agreement. Do check how Barbara Tuchman handles some of references in The Guns of August. ~ J. Johnson (JJ) (talk) 19:48, 25 May 2013 (UTC)

Interesting. Let us look at one of the references in that article :footnote 12 (which contains Tuchman, Barbara W. (1981). Practicing History. New York: Albert A. Knopf. ISBN 0-394-52086-6.") and it is placed immediately after the place in the text where the name of the book is mentioned, and I think for that specific requirement is adequate. However I would have thought that inline citation needs to be moved to the end of the sentence and given page numbers to support the fact the follows the mention of the book: "traveling from Constantinople to Jerusalem on August 29th, 1914, to deliver funds to the Jewish community there".-- PBS (talk) 11:04, 26 May 2013 (UTC)

My apologies!! I meant how Tuchman handles some references in her book, not the article about the book. ~ J. Johnson (JJ) (talk) 20:43, 26 May 2013 (UTC)

The only text that does not need inline citations is text that does not need general references.

Correct, but irrelevant to the argument: the fact that the genref is not required (and it never is, because it doesn't fulfill any of the minimum requirements) does not mean that genrefs are prohibited. You may supply them, just like you may a superfluous inline citation for the number of fingers normally found on the human hand. WhatamIdoing (talk) 01:33, 26 May 2013 (UTC)

While general references are not prohibited (and there is no reason that they should be), there is a clear disadvantage to including them in a references section (confusion, etc), so that in anything other than a very short stub with just one general reference, if they are not to be be part of an inline citation (under whatever descriptive name is to be used), then I think such references are better off in a "Further reading" or "External links" section. -- PBS (talk) 11:04, 26 May 2013 (UTC)

If a reference is used in an article — "generally" or not, by any definition of general — then it really should be in the "References" section (however actually called). That a reference (source) is not specifically cited in an article might be a problem, but deporting the reference is not a proper fix. As I have asked: why not just add an inline citation?

Regarding the other question of whether a citation not required is thereby not required to be "good": are the requirements of WP:RS and WP:WEIGHT also thereby relaxed? ~ J. Johnson (JJ) (talk) 21:05, 26 May 2013 (UTC)

I think perhaps a description may help of why it may appear that a reference to a source may appear to be a general reference when it is not, or at least do not add anything to the article that requires an intline citation. Let us suppose that there is a 2004 Wikipedia article that is a copy from {{EB1911}} and while it is attributed to EB1911 no inline citations are provided (this is very common for articles of that age). Over the next nine years other encyclopaedia articles on Wikisource are added to the references section without inline citations (a classic example would be {{Nuttall}} which mainly consists of short overview articles). Other editors come along during those nine years and add other Wikisource links to the External links section. Eventually someone comes along and adds inline citations as required by WP:PLAGIARISM and in doing so realises that all of the text in the Wikipedia article is a copy from EB1911, should the other references to sources (which give a general overview, but were never used to improve the text) be left in the References section or moved into "External links" (which is where Wikisource links related to a subject but are not used in the Wikipedia article are usually placed)? A person who reads an article and is not an editor of Wikiepdia is likely to conclude--not unreasonably--that there is more significance to the placement of references to a source in a section than just on the choices made by different editors over a nine year period. -- PBS (talk) 10:15, 27 May 2013 (UTC)

Interesting. EB1911 seems to be a special case of where an article was simply copied as a whole. There is no point in citing individual parts because it all goes back only to EB; they are the whole and sole authority, and a single reference suffices for all. If other parts are added, then they ought to be cited as exceptions to what EB1911 is responsible for.

In general I say that all sources actually used in an article, but only those, should be referenced (typically in a "References" section), whether they are cited "inline" or not. Sources an editor uses to understand the topic, etc., do not get cited/referenced, unless the editor feels they are important enough to mention, which typically is done in a note, or at the end of the "Notes" section. Sources not used in article, but possibly of interest to a reader, go into "Further reading" (or some such). ~ J. Johnson (JJ) (talk) 00:58, 29 May 2013 (UTC)

JJ there are lots of other articles based on other PD sources eg DNB, (over 5,000) and CE1913 (over 6,000). PD text must have inline citations and further attribution to meet plagiarism guideline requirements (which at the same time fulfils WP:V requirements). This is important both morally (as described in the plagiarism guideline) and for the internal politics of Wikipedia as a sizeable number of influential editors do not approve of copying PD text into Wikiepdia articles, and only just about tolerate it as a practice if the text is attributed as laid out in the plagiarism guideline. -- PBS (talk) 20:06, 29 May 2013 (UTC)

EB1911 is an unusual problem, in part because the community's long-standing and unchanged practice there diverges so significantly (and on more than 10,000 articles) from the written guideline as to call the accuracy of the guideline into question. The real policy is always what the community actually does, not what's written on the page, and the community does not seem to agree with that guideline where EB1911 is concerned. Consequently, I'm not convinced that discussions around EB1911 articles are going to help us determine what the community actually wants to do. WhatamIdoing (talk) 01:41, 29 May 2013 (UTC)

The whole point of this guidance is that it give guidance on how to bring all articles up to a standard that meets the minimum requirements of WP:V whether or not they originate from PD sources or summaries of copyright sources. The majority of article sourcing currently do not meet the requirements of WP:V. I only used a PD source as an example, because the copied text makes it clear that the additional "general references" have not been placed in the references section to meet WP:V needs -- something which is more difficult to prove articles based on text summaries -- but rather as further reading texts (and/or possible sources for expansion), and I broadly agree with the second half JJ last posting to this thread on where such sources should be placed. -- PBS (talk) 20:06, 29 May 2013 (UTC)

I would point out that there is often a slight divergence between "what the community actually does" and "actually wants to do"; the community generally feeling rather lost in the wilderness and wanting guidance. And although EB1911 is, for sure, an unusual problem, I think it helps to illuminate some of the thinking re general references, and thus of somewhat helpful. In all other respects I am pleased to note that I agree with you here. ~ J. Johnson (JJ) (talk) 19:27, 29 May 2013 (UTC)

Tags for language and "subscription only"?

I'm currently having a discussion with User:Brianboulton on WP:FAC (more specifically, at FAC:Jürgen Ehlers) about the advisability of adding certain kinds of tags to sources such as journal articles (and, equivalently, books). I think it would be good to have a more general discussion about this topic here and, if there is a consensus one way or the other, to add some sentences to WP:CITE for clarification. The two tag issues are independent of each other.

The first issue concerns language tags, such as "(in German)" or, a bit longer, "(in German; English translation of title: "<Whatever the title is>"). Should those be recommended by our style guidelines?

The second issue is, in my eyes, a bit more complex. It concerns subscription tags. If the online version of a journal article is not available freely, but requires a subscription, should we require the citation to include (or [strongly] suggest that it include) a tag "(subscription required)"?

I'll be adding my own opinions in a separate section below; this is just the statement of the two questions. Markus Pössel (talk) 16:26, 1 June 2013 (UTC)

My opinion on the additional tags is that any additional tags make the reference section more crowded, and that tags are only advisable if they add unambiguous information that helps the reader.

The information provided by language tags would be helpful to anyone not fluent in that particular language, but willing to track down somebody who can translate. I'm undecided on whether or not it should be recommended for a translation of the title to be included.

The information provided by subscription tags is ambiguous; it would be available even without the tag with just one additional click; for most readers interested in tracking down the source, it wouldn't even be an additional click; also, the information is likely to change over time, requiring updates. Concretely:

• The online version of an article as linked by the Digital Object Identifier might require subscription, but a perfectly serviceable e-print might not. Not using the (subscription required) tag in that case would be misleading, as it would convey false information about the official online version; using it would be misleading because it would imply that any only version, including e-prints, requires subscription.
• Subscription policies are so diverse as to further diminish the information contained in a subscription tag. Even if I'm a reader that is only interested in sources accessible without me paying anything, the tag might mark an article where, as in the case of a number of JSTOR articles, I might be able to get free access simply by registering (or not, if I have used that option before). So the subscription tag does not save me the necessity of clicking the given online link, in other words: contains no usable information.
• Whether or not a subscription is required is subject to change - a number of journals will make articles older than a number of years available without subscription. Introducing a subscription tag would introduce substantial additional work in keeping an article up to date.
• In the great majority of cases where an online link is provided by DOI, the first page reached by clicking that link contains all the necessary information. Even if we leave out the subscription tag, not much is lost - one click, and whoever wants information about subscription information will have that information.
• In practice, that is not even an extra click - whoever is interested in the article is quite likely to click on the DOI in any case, simply to retrieve the abstract.

In summary, I think subscription tags add extra work without giving much usable information in return, and I am opposed to recommending them in WP:CITE. Markus Pössel (talk) 16:26, 1 June 2013 (UTC)

The fact remains that the "subscription required" tag does exist and we need to know when to use it, or not. Maybe this is more relevant to newspaper citations. I think we should use it for The Times (London), which has been completely behind a paywall for some while. But what do we do about The Daily Telegraph (London) -- to which there are squillions of references in many WP articles on UK subjects -- which has always been free until now but, as of a week or two ago, now allows the reader a certain number of views per month but after that you have to take out a paid subscription? Should we have a separate tag for this situation, called something like "restricted access"? -- Alarics (talk) 18:10, 1 June 2013 (UTC)

The mere existence of the tag doesn't carry much weight one way or the other. The example of The Daily Telegraph is a good one for two reasons: It shows that there are various ways of "subscription required", so the mere tag doesn't carry all that much information, and also that if use of that tag should be encouraged, then every time the subscription model changes, we face squillions of necessary changes. Both are, I think, good reasons for discouraging use of that particular tag.

It's also a good example because it shows another difference: For a link to a subscription-only newspaper article, clicking without a subscription will be frustrating because you essentially end up in a dead end. For scientific journals, you usually at least get some additional information by clicking, even if the complete content is subscription only: you at least get the abstract. Another reason not to lump the two cases together. Markus Pössel (talk) 18:31, 1 June 2013 (UTC)

Use of the "subscription required" tag does provide information to the reader. We need not be concerned about crowded reference sections. Most readers refer to the section for a specific reference, and we are not type-setting.

• (1) The "subscription required" tag does not imply anything about any other versions, it is strictly about the link (URL) provided.
• (2) The "subscription required" tag does not stop one from clicking through, it just informs the reader that they may not be able to view the complete document when doing so.
• (3) While occasionally vendors will make articles older than a certain number of years available without subscription, this not the usual case, excepting the six-month to one-year blackout period adopted by some journals upon publication, in which case the "subscription required" tag would not be appropriate.

As to language tags, they help when I don't recognize the language, or when the title does not disclose it, such as Latin titles. And yes, I want to know if a volume has been translated, and by whom, and the original title. If something is a reprint with a title change, I'd like to know the original publication date and title as well. --Bejnar (talk) 19:44, 1 June 2013 (UTC)

In view of all the above, my inclination, until somebody tells me otherwise, is (a) not to add the "subscription required" tag to any new references, (b) not to cite the URL at all in the case of The Times, since readers who do have a subscription to it will know how to find it online and for everyone else it is pointless and frustrating, (c) cite the URL but without a tag in the case of The Daily Telegraph, in the hope that most readers will not reach the maximum number of articles allowed, and (d) add the "subscription required" tag to existing citations for The Times that do include the URL. Does this seem reasonable? -- Alarics (talk) 07:49, 2 June 2013 (UTC)

What you "should" do is follow the style already present in the article, whatever that is. Are they adding links? If yes, then add links. If no, then don't. Are they tagging sources by language? If yes, then tag the sources. If no, then don't.

Are you establishing the citation style yourself? Then do whatever you want, and talk about it on the article's talk page if you want to involve any other editors at that article.

What you shouldn't do is go to FAC with some idea that there is a right way and a wrong way to do this. Per WP:CITEVAR, it is our official guideline that editors at every single individual article gets the right to establish the citation style that is best for that article, without being told that "my English teacher said" or "The only good stylebook says" or "All the other Wikipedia articles do this other thing". WhatamIdoing (talk) 23:00, 5 June 2013 (UTC)

Issue with multiple refs to the same source (Restatement of)

Here's a different problem, or the same problem, stated a completely different way, depending on how you see it:
Often, an editor improving an article using {{cite}}-family citations (typically Citation Style 1) comes to https://en.wikipedia.org/wiki/Wikipedia:IBID#Citing_multiple_pages_of_the_same_source, and wants to know how to cite the same source multiple times, what they see is:

When an article cites many different pages from the same source, most Wikipedia editors use short citations in footnotes. Other methods include short citations in parenthesis and the template {{rp}}.

So of course they are going to click on short citations - and find no instructions on how to make short citations to the a source already used in the article and cited with a {{cite}}-family citation. That's a problem. --Elvey (talk) 22:34, 9 June 2013 (UTC)

Seeming contradiction between sections

Our "convenience links" section says that it's helpful to provide links to content that's being hosted by reliable third parties with permission. Two sections later, we're told not to provide links to content being hosted by databases, even though they're reliable third parties that host with permission — for example, JSTOR isn't the original publisher of any of the journals that it hosts. At the same time, anybody can read the first page of most JSTOR articles for free, so this section effectively permits JSTOR just a few sentences after telling us not to link it under normal circumstances. All that being said, why would it ever be a problem to have a link to JSTOR or other databases in a citation? Lots of people have access to major databases, so including a link is clearly helpful for them, while it provides no problems whatsoever to people who don't have access to major databases. Imagine that I'm reading an article about a topic I've never heard of, and I want to see the source — it's much simpler if you give me the JSTOR link than if you don't, since I'll know exactly where to go instead of going to the library, searching ProQuest, searching Ebscohost, and finally finding it in JSTOR. Nyttend (talk) 01:47, 10 June 2013 (UTC)

I think you're right. WhatamIdoing (talk) 22:13, 10 June 2013 (UTC)

I agree. The second section was added here in December 2010, without anyone noticing (including the editor who added it) that there was an inconsistency with another section, which is easily done. I wouldn't mind seeing the whole paragraph removed. SlimVirgin ^(talk) 01:50, 11 June 2013 (UTC)

I don't see the contradiction it generally is not important to cite a database such as ProQuest, EbscoHost, or JStor (see the list of academic databases and search engines) or to link to such a database requiring a subscription or a third party's login. does not say don't. Don't add a URL that has a part of a password embedded in the URL I think is correct advise. -- PBS (talk) 07:30, 11 June 2013 (UTC)

There is no reason to specifically discourage links to JStor el al., as long as enough information is given to locate the article without the use of the database. Indeed it is ironic to discourage providing a link in these cases, when (1) the paper can be found in print as well and (2) the URL is not likely to expire, when we make no effort to discourage links to news stories that are only published online and which may become unavailable (or subscription only) at any point in time. — Carl (CBM · talk) 15:53, 11 June 2013 (UTC)

RFC: Consistent date location in citation templates

Please see Help talk:Citation Style 1#RFC: Consistent date location where I ask if the location of the date in the most popular citation templates should always be consistent. At present, the date is immediately after the authors if authors are given in the citation, but near the end if not. Jc3s5h (talk) 10:44, 13 June 2013 (UTC)

This is something to raise on the template pages for cs1. Fifelfoo (talk) 10:52, 13 June 2013 (UTC)

If the maintainers of the citation templates are not willing to maintain them as a set and accept requests in a central location then they do not form a consistent style and should not be allowed. Jc3s5h (talk) 11:00, 13 June 2013 (UTC)

Talk for the Citation Style 1 templates is centralized at Help talk:Citation Style 1. --  Gadget850 ^talk 22:27, 13 June 2013 (UTC)

Multiple references to the same source

When a single source is cited multiple times in the same article, getting back from the references section to the right section of prose requires guesswork. This is (imo) a bad thing and I have prosed to change it.

For a detailed explanation see Wikipedia:Village pump (technical)#Multiple references to the same source. You are invited to participate in the discussion there. Thryduulf (talk) 00:36, 16 June 2013 (UTC)

Clarification: the issue is with the "a b c" links when the <ref name="..."> syntax is used. But we can keep comments centralized on the other page. — Carl (CBM · talk) 01:56, 16 June 2013 (UTC)

Proposed replacement for "This page contains...."

I propose that the third paragraph of the page, starting "This page contains information ...", be replaced with the following text:

Citation can be done in several ways and many variations, with the merits of each often ardently debated. Therefore Wikipedia has not set any standard method or style of citation, requiring only that citation be consistent within each article; this is the policy often cited as WP:CITEVAR. If you add citations to an existing article that has a consistent citation style, you must conform to that style. If you wish to change that style you must first seek the consensus of other editors. If you start a new article you may select a method and style as you prefer, but it is recommended that you stay with standard methods that other editors will understand. This guideline explains several ways of doing citation on Wikipedia.

~ J. Johnson (JJ) (talk) 23:34, 4 June 2013 (UTC)

It's better to keep the writing tight. I think the current third paragraph is fine, and it makes clear that citations don't have to be written perfectly. SlimVirgin ^(talk) 23:59, 4 June 2013 (UTC)

I'm still thinking about the overall idea, but the phrasing of "you must conform to that style" is wrong. The citations must (eventually) conform; you need not be the person who makes them conform. This is particularly important because of the effect on new users: we should not be reverting their expansions or warning them about violations if they don't get the citations correctly formatted on the first try. Somebody needs to deal with the citation formatting before the WP:DEADLINE, but it doesn't matter who does it or how soon it happens. WhatamIdoing (talk) 23:55, 5 June 2013 (UTC)

I actually agree with you somewhat. In particular, "must" seems rather strong (I faintly recall there was some discussion of this way, way back); "should" is probably better. But there is a fundamental problem in this idea of "you need not bother doing it right because surely someone else will fix it." (Yes, not exactly what you said, but allow me a little artistic license here, as that is the way it often turns out.) It gives free license to people who won't do citations (in any form). As to new editors that may have something to contribute but are not yet acquainted with the details of citation: I don't object to adding something about doing what one can, but we can't just leave it at that. And certainly experienced editors should not be allowed to evade WP:CITEVAR on the grounds that someone else can cleanup. Would you care to propose such additional text? ~ J. Johnson (JJ) (talk) 21:24, 6 June 2013 (UTC)

We can't prevent people adding citations in whatever way they feel is easiest for them, and indeed we want to encourage that. All we can ask is that they allow others to make them consistent, and that they give consideration to making them consistent when they first add them (if they know how to). SlimVirgin ^(talk) 22:10, 6 June 2013 (UTC)

So you are saying that WP:CITEVAR is entirely optional? That those of us who would like citation to be consistent are free to make it so, but no one else need bother to do so? ~ J. Johnson (JJ) (talk) 22:46, 6 June 2013 (UTC)

WP:CITEVAR is not a policy, it is part of a "content guideline" and it says "editors should attempt to follow" what it says (with further qualifications). In saying "Editors should not attempt to change an article's established citation style" it is certainly recommending that you should make an effort to fit in with what is there already. I wouldn't describe that as "entirely optional" but neither is it a breach of policy (or necessarily of the guideline) to fail to do that. On the other hand, there is no requirement or advice whatever that you should go around making existing citations compatible within articles (though it says that to do so is helpful). If you want to do that, it's great. But that aspect is entirely optional. Thincat (talk) 23:14, 6 June 2013 (UTC)

So the bottom line for CITEVAR is: ... entirely optional. Right? ~ J. Johnson (JJ) (talk) 20:36, 7 June 2013 (UTC)

JJ, I think your approach here isn't helpful. You know that CITEVAR isn't optional within this guideline, but you also know that whether to follow the guideline is optional, and that disputes about it can go either way. The point of CITEVAR is to say to editors: (a) if you know what you're doing, and (b) if you want to be decent to your fellow editors, then (c) please follow the existing style, and (d) if you don't or can't at least let others fix it. SlimVirgin ^(talk) 20:53, 7 June 2013 (UTC)

I do not see where CITEVAR (as has been interpreted) is anything but optional. That consistency is nice, but someone else will clean up after you. By the way, catch the related discussion at WT:V (skip down to the end). ~ J. Johnson (JJ) (talk) 22:51, 7 June 2013 (UTC)

CITEVAR is exactly as optional as the entire rest of the guideline. But rather than us all repeating this again, why don't you go read Wikipedia talk:Citing sources/Archive 33#Reverts_based_on_Citation_variation, where we told you exactly the same thing at length just a few months ago? WhatamIdoing (talk) 04:11, 8 June 2013 (UTC)

If it helps, I think of this as being rather like driving a car. There are some rules which are absolute - you don't drive on the wrong side of the road, for example, even when you're a learner. On the wiki, you don't vandalism articles - doesn't matter who you are, even if you're new. There are some guidelines which are important, but are about what people should do. You should drive your car, for example, at an adequate speed. If you consistently drive well below the speed limit unnecessarily in the UK, causing problems for others, some other road users will eventually start to honk their horns, shout, gesture rudely and get angry with you. Eventually, some foolish ones will take risks and try to overtake. But it's not illegal to drive slowly on most roads. Learner drivers do it all the time, and we don't honk our horns at them - we know they're learning. Lost drivers will also do it, as they try to work out where they are. CITEVAR is just like that - you should follow it, and there are likely consequences if you don't - but it is not an absolute rule. Hchc2009 (talk) 06:56, 8 June 2013 (UTC)

I fully agree about not yelling at newbies. But! we also have non-newbie editors who simply refuse to bother with any kind of adequate citation, who persist in driving down the wrong side of the road (and throwing beer bottles out the window), who accept no responsibility for fixing anything. Consequences? Nope, haven't seen any. ~ J. Johnson (JJ) (talk) 19:10, 8 June 2013 (UTC)

If your idea of "adequate citation" is "properly formatted exactly the way I would have done it myself", then you're right: failing to type out a full, properly formatted citation is not a blockable offense. If your idea of "adequate citation" is "can't even be bothered to add a bare URL for material that absolutely requires a citation", then you're wrong, and, if you want proof that your personal experience is incomplete, then I suggest that you go ask Kww about how many people he's personally blocked for this. WhatamIdoing (talk) 20:21, 9 June 2013 (UTC)

The latter of course. And thanks for the suggestion. Hopefully I'll remember it the next time the issue comes up. Allowing that my personal experience is incomplete, I am still inclined towards having something to the effect that the "newbie exception" isn't a free ride for violating CITEVAR. ~ J. Johnson (JJ) (talk) 22:46, 9 June 2013 (UTC)

---

I think this is a bad idea. The priories expressed are wrong. "What matters most is that you provide enough information to identify the source." is true, and established consensus. But this proposed wording would violate that consensus and make deleting a cite seem acceptable solely because its formatting is inconsistent or bad. It isn't acceptable. And... --Elvey (talk) 22:24, 7 June 2013 (UTC)

Dubious

... Speaking of a "standard method or style of citation, I just added a {{dubious}} tag, and that's why I'm here. It seems like I hardly see these (short citations with footnotes) any more, and citation templates are used most. Is that because I'm not editing a representative article sample? --Elvey (talk) 22:24, 7 June 2013 (UTC)

Short refs are common, Elvey. They mean you don't have to keep repeating the full cite. So you say once (in the references section, or on first reference in the notes section): Smith, John. Name of Book. Name of Publisher, 2013. And thereafter Smith 2013. SlimVirgin ^(talk) 03:32, 8 June 2013 (UTC)

I agree that the "standard methods" bit is wrong. Editors may make up whatever style they want, even if nobody else has ever done anything similar before.

Short cites are common in the single instance of lengthy works being used repeatedly (so that you need to vary the page numbers). Most of our articles don't use long works at all, much less repeatedly. WhatamIdoing (talk) 04:05, 8 June 2013 (UTC)

Shortened footnotes are pretty common. If you are really interested, you could go through Category:Author-date citation templates and total up the transclusions. --  Gadget850 ^talk 15:36, 8 June 2013 (UTC)

I very frequently see short cites being used in featured articles because books are often used there as reference sources and there are many precise citations. Also, the editing is technically to a high standard and care has been taken over presentation. Regarding your edit summary about short cites becoming less popular than citation templates. These aspects are not really in competition. Short cites may be done using templates (Template:sfn, for example) and the full cites they point to may also be done with or without templates. Thincat (talk) 21:13, 8 June 2013 (UTC)

I think I know what short citations are and this edit I had just made shows me fixing problems with Harvard and named references! A reference I added there is "<ref name="Guibault 2006 xx" />{{rp|1}}" and according to short citations, that's NOT a 'short citation', even though it's a citation that's short. That suggests a confusion over terms that's creating an apparent disagreement. Per SlimVirgin's definition, OTOH, what I added IS a short citation. As I see it, both definitions of short citation can't be right. --Elvey (talk) 19:53, 9 June 2013 (UTC)

It looks like I didn't express myself clearly. (Sorry!) First: the text I'm challenging says "most Wikipedia editors use short citations in footnotes." It's the word MOST that I'm challenging. Second: When I wrote "citation templates" I meant to refer more specifically to the use of {{Cite}}, so Thincat: short cites as defined at WP:CITESHORT and use of {{Cite}} are in competition. Certainly short cites are still 'common', but that's a far cry from the claim that 'most' editors use them. If 'most' is to stay, it needs to be verifiable, and I'd bet it isn't. "Wikipedia editors commonly use..." would be something of an improvement. But if we don't know what is most common, we mustn't claim to. --Elvey (talk) 19:53, 9 June 2013 (UTC)

Ideas for moving forward :
OPTION A) (revised) Change the definitions in the article and at WP:CITESHORT and Shortened footnotes as needed to clearly include the style of the citation I added (named references) in their explanations for how to make short citations.
OPTION B) Change what we say here to be true/verifiable without such changes, while referring to the MOST common citation style.
OR OPTION C) Change what we say here to "...use short citations or named references ..." --Elvey (talk) 19:53, 9 June 2013 (UTC)

Elvey does not understand what a "short citation" is. But that goes precisely to the point I have been trying to make, and I believe he is trying to make: the confusion over terms endemic in this corner of Wikipedia. ~ J. Johnson (JJ) (talk) 20:19, 9 June 2013 (UTC)

(Such a kind, helpful, constructive comment, JJ. Well done.</sarcasm>) Really? Precisely what part of "A reference I added there is "<ref name="Guibault 2006 xx" />{{rp|1}}" and according to short citations, that's NOT a 'short citation', even though it's a citation that's short." do you disagree with? --Elvey (talk) 21:12, 9 June 2013 (UTC)

Elvey, you need to read the whole sentence. It does not say "most Wikipedia editors use short citations". It says "When an article cites many different pages from the same source, most Wikipedia editors use short citations". See the difference? "Most editors" don't use short citations because "most editors" never cite many different pages from the same source. But when an article cites many different pages from the same source, {{sfn}} gets used slightly more than twice as often as {{rp}}, and plain hand-formatted short cites are even more common than that. WhatamIdoing (talk) 20:26, 9 June 2013 (UTC)

If I may: since this section isn't trying to limit or reinforce one way or the other (is it?), could all parties be made happy by replacing "most editors" with "some editors"? -- JHunterJ (talk) 20:41, 9 June 2013 (UTC)

What makes you think I didn't read the whole sentence that I quoted from? I don't think it's true that "When an article cites many different pages from the same source most Wikipedia editors use {{sfn}}" (Note: I changed the end of the sentence to make it clear what I'm referring to.) Your evidence doesn't support the claim that it does. What I see most often when an article cites many different pages from the same source is that Wikipedia editors use {{cite}} (including {{cite web}}...), but unfortunately, don't use {{rp}}, but rather use multiple long citations to the same work. (I assume these couldn't become FAs unless improved, but the logical way to improve them would be to use {{rp}}, not {{sfn}}.) --Elvey (talk) 22:26, 9 June 2013 (UTC)

CITESHORT does not limit the concept of short citations to uses of {{sfn}}. We define short cites at the top of the guideline, and we explicitly say that there are multiple methods of creating them. Using {{cite}} to produce a short citation is one of many options.

I agree with you that some articles that use one source only a couple of times will repeat the full citation, but that is not a case of "many different pages". I've yet to see someone repeat a full citation two dozen times in an article; have you? Sometimes the repetition is even deliberate (e.g., someone who hates named refs, or who is worried that the other citation will get removed and leave a short cite with no full citation to support it). But it is not what most editors choose. WhatamIdoing (talk) 21:00, 9 June 2013 (UTC)

I didn't say it did limit the concept that way. Do you claim that CITESHORT clearly includes, in the concept of short citations, cite or rp style refs? Where? I find this: "Forms of short citations used include author-date referencing (APA style, Harvard style, or Chicago style), and author-title or author-page referencing (MLA style or Chicago style)." (It doesn't explicitly exclude them; I'm not saying that. But when you include a long list of them and don't include cite, what's one to think?) 24x? Straw man. IIRC, we have a bot that recovers if someone deletes the master copy of a named ref used multiple times. Oh and you say, "Using {{cite}} to produce a short citation is one of many options." So then what's wrong with option A, above? --Elvey (talk) 22:26, 9 June 2013 (UTC)

I'm sorry, but I've lost track of what "option A, above" is. WhatamIdoing (talk) 22:10, 10 June 2013 (UTC)

Seriously? Search for "A)". [Placeholder for charitable comment on your seeing and searching skills.] --Elvey (talk) 06:43, 23 June 2013 (UTC)

Here's the actual definition from the guideline:

A short citation is an inline citation that identifies the place in a source where specific information can be found, but without giving full details of the source – these will have been provided in a full bibliographic citation either in an earlier footnote, or in a separate section.

I don't really see anything in this statement that says {{rp}} is absolutely not a short citation. After all, the page number is "an inline citation that identifies the place in a source where specific information can be found, but without giving full details of the source". The system might be better described as a hybrid, with the {{rp}} part being the "short" part and the little blue number leading to the "full" part. WhatamIdoing (talk) 21:35, 9 June 2013 (UTC)

Wow. [Placeholder for charitable comment on your seeing and searching skills.] No, a page number (alone) is not that. But I can't force you to see what's already been put right in front of you. In any case, what you have NOT done is address the dubiousness of the claim I tagged. --Elvey (talk) 06:43, 23 June 2013 (UTC)

Elvey: if you want to argue that there is a confusion of terms here (no?) then I am in agreement with you. But regarding your own evident and demonstrated confusion I concur with WhatamIdoing. And that in itself should be a strong sign that you are way off base. ~ J. Johnson (JJ) (talk) 22:37, 9 June 2013 (UTC)

Elvey, a short cite is just that: a citation that has been shortened. So it could be Smith 2013; Smith 2013, p. 1; Smith 2013, chapter 1; Name of Book; Name of Book (2013); Name of newspaper, 9 June 2013; and so on. And these can be produced with or without citation templates.

The full citation is then given somewhere in the article, usually only once, and usually at the end in the References section. SlimVirgin ^(talk) 00:07, 10 June 2013 (UTC)

One point that seems to be confusing me is "named refs". Rhabdomyolysis uses named refs. It does not contain a single short citation. Breast cancer also uses named refs. Breast_cancer#cite_ref-Olson102_116-0 is an example of a named ref that is a short citation and is reused; Breast_cancer#cite_note-Olson1-117 is a named ref that is a short citation and is not reused. WhatamIdoing (talk) 22:10, 10 June 2013 (UTC)

[Replies above.]--Elvey (talk) 06:43, 23 June 2013 (UTC)

So what you'd like is this (newly labeled) option:

OPTION A) (revised) Change the definitions in the article and at WP:CITESHORT and Shortened footnotes as needed to clearly include the style of the citation I added (named references) in their explanations for how to make short citations.

Now what I need to know is why we should consider this:

<ref name=CritCare2005>{{cite journal |author=Huerta-Alardín AL, Varon J, Marik PE |title=Bench-to-bedside review: rhabdomyolysis – an overview for clinicians |journal=Critical Care |volume=9 |issue=2 |pages=158–69 |year=2005 |pmid=15774072 |doi=10.1186/cc2978|url=http://ccforum.com/content/9/2/158 | pmc=1175909}}</ref>

to be a short citation. That's a WP:Named reference from Rhabdomyolysis. It is not short. WhatamIdoing (talk) 07:28, 26 June 2013 (UTC)

That's not in the style of the citation I added. My patience is exhausted. (Congrats?) --Elvey (talk) 08:16, 26 June 2013 (UTC)

We can't put "the style Elvey added one time" into the guideline. No matter what you meant, you actually asked for WP:Named references to be added as a style of short citations. They aren't. Did you mean to ask for something else? WhatamIdoing (talk) 09:12, 26 June 2013 (UTC)

Inconsistent citation style (course of procedure?)

In my effort to improve consistency of citations, I came across this article, which had a fairly inconsistent citation style: some refs used Harvard style, some Chicago style, some used citation templates, others didn't. The "Further reading" section itself had five different styles of citation going on. So then I did what according to WP:CITEVAR is "generally considered helpful": imposing one style on an article with incompatible citation styles: an improvement because it makes the formatting consistent. I did this by gradually implementing citation templates, until my efforts were reverted by User:ColonelHenry.

So far, so good. I brought this topic to the article's talk page, in order to find consensus. Whether we implemented the citation style I suggested, or some other, was to be determined by discussion, I believed. However, on insistence of ColonelHenry and User:CBM, the citation style did not merit any debate, because—according to CBM's interpretation of WP:CITE—if "anyone objects", it must not be changed. I asked CBM on where WP:CITE states the existence of such a veto right, but he did not reply.

So my question is: what is the course of procedure in cases like this? Is the citation style, like any other aspect of an article, subject to potential debate? Can it be changed after consensus has been established? --bender235 (talk) 17:18, 11 June 2013 (UTC)

The article in question was established with references that do not use citation templates. The merits of citation templates have been discussed endlessly on this page, and we all know it is an unsolvable disagreement whether citation templates are better or worse than non-templated citations. As an experienced editor, Bender235 should have known it wasn't appropriate to convert all the references in the article he had never touched before to use citation templates. More importantly, when someone objected to the change, he should have left the page as it was, instead of badgering the user about why the user objected [11]. This looks to me like an example of WP:IDIDNTHEARTHAT with respect to preserving the established citation styles of articles. A basic moral of this page is that editors should not have to spend time defending the absence of citation templates in an article. — Carl (CBM · talk) 17:28, 11 June 2013 (UTC)

Contrary to what CBM claims, the article in question did use citation templates before, although only in some references. Overall, it had an inconsistent citation style. One could argue whether it would be better to use a consistent citation style w/ or w/out templates, but that should be subject to a debate. I don't see any reason why some mysterious veto rule should apply.

Also contrary to what CBM claims, I did in fact "leave the page as it was", not starting an edit-war or anything. Instead, I turned to the article's talk page, starting a discussion on how a consistent citation style for the article in question should look like (what CBM calls "badgering" is what other Wikipedians refer to as WP:Discussion). However, like stated above, I was told that changes of citation style were not a legitimate subject for debate on Wikipedia according to some unwritten rule, apparently.

To be clear: this is not about the use of citation templates. I could have implemented a consistent citation style on the article in question w/out using templates, too. It would have been the same kind of edit: a user changing the current citation style. --bender235 (talk) 17:50, 11 June 2013 (UTC)

I think that "if anyone objects" is a bit strong. Certainly my strong objections to a change at Breast cancer awareness were overruled, mostly by people who hadn't contributed even a paragraph to it.

However, it appears from this discussion that the question might be more about whether to use citation templates, not about whether it's good to have one style. There should be, at minimum, one style in the output. (Whether the underlying source code needs to "match" is not actually something we've had very strong agreement about in the past.) WhatamIdoing (talk) 19:35, 11 June 2013 (UTC)

It was not about whether to use citation templates, but rather on how the course of procedure looks if one suggest a change in citation style. Should it be possible to discuss such things on the respective article's talk page, or is it considered "badgering"? Again, it is not about citation templates. User:ColonelHenry did exactly what I did, only without templates: he chose a citation style, and imposed it on the entire article. Should his edit now be reverted, too? --bender235 (talk) 20:00, 11 June 2013 (UTC)

The title of the section you started on the talk page was "citation templates" and the question you asked was "Per WP:BRD: is there any reason why this article should not use citation templates?". The substantive objection to the changes you made was that you converted the article to use citation templates. So I think it is safe to say that the discussion is about whether to use citation templates. — Carl (CBM · talk) 20:08, 11 June 2013 (UTC)

I distinguish between the discussion here and the discussion at article's talk page, and so should you. Regardless, would you please answer these questions:

(i) Are you aware of the fact that the article in question did use citation templates before, although only in some references?

(ii) Had I imposed a style similar to the one produced by citation templates, without actually using citation templates, would things be different? --bender235 (talk) 20:21, 11 June 2013 (UTC)

The article was established without citation templates; here is a typical version from the end of 2011 [12]. That is what matters for WP:CITEVAR. I believe one editor first added some templated citations in an edit in 2012 [13]. The correct response, if you had adequately investigated the page history before editing, would have been to replace the templates with non-templated references, bringing the citations into agreement with the previously established style. And, yes, if you have simply tidied up the references some without also converting them all to templates, removing templates that were mistakenly added, it is not clear that anyone would have objected. But you know all this already. — Carl (CBM · talk) 20:28, 11 June 2013 (UTC)

The fact that you arbitrarily pick one version of the article as "typical", and some other as atypical, is somewhat amusing considering the nature of Wikipedia.

Also, I'm puzzled by your notion of an "established style" you claim this article had, because it did not. As I pointed out on the article's talk page, there were five different styles alone in the seven entry "further reading" section. The only thing they had in common was they were obviously not produced by an underlying template.

So when you claim the "correct response" to one editor's addition of a source with citation templates would have been to remove those templates and to transform those citations to the "established style", everyone but you wonders which style that ought to be? I suppose anything but template, but that's actually not a citation style, but a paradigm.

By the way, I'm curious, if it isn't actually the citation style those templates produce (since you said same style sans templates would be okay), what is it that makes you oppose them? --bender235 (talk) 21:05, 11 June 2013 (UTC)

If you couldn't tell what the established style was, then you weren't really in a position to unilaterally edit the citations until you did more investigation. But the more important thing in your behavior on the article was that, even after an editor objected to the introduction of templates, you continued to press for them. The direct language of CITEVAR is, "if there is disagreement about which style is best, defer to the style used by the first major contributor". Since the article existed for years without citation templates in the footnotes (until 2012), whatever the style used by the first major contributor was, it certainly didn't involve templates in the footnotes. — Carl (CBM · talk) 21:08, 11 June 2013 (UTC)

Sorry if I wasn't clear enough in my choice of words, but this article did not have an established style. The only thing "established" was that it used (almost) no citation templates. But other than that, every citation looked different in terms of punctuation, etc. Even now, after one user imposed his citation style onto the article (which he, unlike me, is entitled to), the article's citation style is far from being consistent. I just pointed out that there are still five different versions of style.

• (i) Locke, Frederick W. "Dante and T. S. Eliot's Prufrock." in Modern Language Notes. (1963) 78:51-59.

Period after article title, period after publication title, no comma after year.

• (ii) Stepanchev, Stephen. "The Origin of J. Alfred Prufrock" in Modern Language Notes. (1951), 66:400-401.

No period after article title, period after publication title, comma after year.

• (iii) Soles, Derek. "The Prufrock Makeover" in The English Journal (1999), 88:59-61.

No period after article title, no period after publication title, comma after year.

• (iv) Luthy, Melvin J. "The Case of Prufrock's Grammar" in College English (1978) 39:841-853.

No period after article title, no period after publication title, no comma after year.

• (v) Sorum, Eve. "Masochistic Modernisms: A Reading of Eliot and Woolf." Journal of Modern Literature. 28 (3), (Spring 2005) 25-43.

Meh.

For journal articles alone! There are more inconsistencies if you also include books and book chapters. So, I'm asking again: which one do you think is the "established style" of this article, and how do you determine? Because I really have no clue how to determine that. --bender235 talk) 21:33, 11 June 2013 (UTC)

Browse the page history until you find "the style used by the first major contributor". In this case I believe it is this edit: [14]. — Carl (CBM · talk) 21:42, 11 June 2013 (UTC)

Ah, I see. So it is like this:

• "On 'The Love Song of J. Alfred Prufrock'", from Blasing, Mutlu Konuk. 'American Poetry: The Rhetoric of Its Forms'. New Haven: Yale UP, 1987.

Then again, this one is totally different from the one that has recently been imposed by User:ColonelHenry.

• Blasing, Mutlu Konuk, "On 'The Love Song of J. Alfred Prufrock'", in American Poetry: The Rhetoric of Its Forms (New Haven: Yale University Press, 1987).

Can anyone revert his changes now, or is only User:Samael775 allowed to veto, according to your interpretation of the guidelines? --bender235 (talk) 21:48, 11 June 2013 (UTC)

At this point, I believe that your best option is to learn WP:How to lose with a little more grace, but CBM is essentially correct: the fact that the article was totally screwed up a couple of weeks ago does not mean that there never was "an established style", and even if there isn't, then which style to impose is not necessarily a first-come, first-served question (unless the article is so neglected that nobody actually cares). WhatamIdoing (talk) 00:26, 12 June 2013 (UTC)

Okay, so let's assume User:Samael775's citation style was the article's "established" one. Is it then any different to impose a completely new citation style, like ColonelHenry did, from imposing a completely new citation style with help of templates, like I did? To me, it is exactly the same. If you begin with:

• "On 'The Love Song of J. Alfred Prufrock'", from Blasing, Mutlu Konuk. 'American Poetry: The Rhetoric of Its Forms'. New Haven: Yale UP, 1987.

where is the difference between changing it to:

• Blasing, Mutlu Konuk, "On 'The Love Song of J. Alfred Prufrock'", in American Poetry: The Rhetoric of Its Forms (New Haven: Yale University Press, 1987).

or changing it to:

• Blasing, Mutlu Konuk (1987). "On 'The Love Song of J. Alfred Prufrock'". American Poetry: The Rhetoric of Its Forms. New Haven: Yale University Press.

Both is absolutely the same type of edit. I don't see any reason to treat them differently just because one used templates and one didn't. --bender235 (talk) 00:46, 12 June 2013 (UTC)

Bender, can you make the style consistent without using templates? It seems that would solve the problem (if I've read the above correctly). SlimVirgin ^(talk) 01:21, 12 June 2013 (UTC)

Sure, that would be easy to do. But then again, what's the point of doing that? Plus, it also eliminates one of the key features of citation templates: that is the creation of a hidden COinS element (check the source code behind the line I posted above), which is a machine-readable version of the citation. --bender235 (talk) 01:34, 12 June 2013 (UTC)

I took a look at the article, and this, for example, is the kind of edit CITEVAR seeks to avoid. Templates were added to manual cites that were already properly formatted, except perhaps for the occasional inconsistency, which can easily be fixed. Even after the templates were imposed, there were inconsistencies (at least one cite with pp, others without).

If the article is developed, the templates will introduce even more inconsistency, which is hard to fix (e.g. some templates put date in brackets after author, but date at the end if there is no author). And when there a lot of them, load time slows down. For all these reasons, CITEVAR asks people not to add them when there are manual refs that are properly formatted. SlimVirgin ^(talk) 01:45, 12 June 2013 (UTC)

@SV. It has been pointed out that the refs were inconsistently formatted, so in this case that argument can be placed to one side. Speed is not usually a consideration: do you have any figures for the load speed difference on an article like this? Citation templates do not introduce inconsistency as they are consistent. If the consistent positioning of dates in different places depending on whether an author is present is a concern to you then place to raise that is on the cite template core talk pages.

@CBM years ago I used not to use citation templates, mainly because I could not be bothered to learn the syntax of them, but having done so I think that the time was well spent. Are you really saying that if an article was created by copying EB1911 and a {{1911}} template was added at it creation in 2004, that today even though there is no text from EB1911 in the article (and none since 2006 when the {{1911}} template was removed) and every citation since has been formatted in a consistent style, someone is at liberty to insist that those citations are replaced with templates because back in 2004 the {{1911}} template was used?

@CBM, I have created many articles. Can I insist as the primary author that although I did not use to use citation templates when I wrote am article that article should be converted to use citation templates under "If you are the first contributor to add citations to an article, you may choose whichever style you think best for the article." That would seems like ownership to me.

@CBM, many of the citation templates add hidden categories to articles to aid maintenance, are you really saying that if a person adds text copied from the DNB they should not use the {{DNB}} template if there is no other citation template currently in use for other citations? If so why, as not doing so hinders development of the project?

-- PBS (talk) 13:12, 12 June 2013 (UTC)

This thread is about edits such as this. The article was established and well referenced without the use of citation templates from 2006 through mid 2012. The fact that there are minor inconsistencies between citations, or that someone might have accidentally used a citation templates in 2012 without looking to see if the were already present, are not sufficient justification for an editor to unilaterally convert the entire article to use citation templates. There are some particularly subtle points of CITEVAR that must be discussed on an article by article basis, but the particular case under discussion is not subtle and doesn't require solving them. — Carl (CBM · talk) 14:16, 12 June 2013 (UTC)

Leaving aside the fact that your notion of "minor inconsistencies between citations" is plain laughable, where was the "sufficient justification for an editor to unilaterally convert the entire article" to his citation style in this case? As you pointed out above, ColonelHenry's citation style wasn't the one originally used in this article. So when are you going to revert his edit? --bender235 (talk) 18:15, 12 June 2013 (UTC)

• Will you give it a rest already? The citation style generally in use before was largely Chicago, and as established it needed minor cleanup. MINOR. That's commas, making sure the dates are in the right place, parentheses (something Chicago's different versions is inconsistent on). I did that minor cleanup after your intolerably incessant whining. There's nothing that says someone can't do a minor cleanup when it is minor as was the case here. You changed it from non-template to template which is clearly discouraged by policy. That is put in very unambiguous language. There's a whole chasm of difference between what is reality and the nonsensical interpretation of events you are persistent in asserting. The fact that you are persist in being obtuse and not accepting this simple truth is wasting everyone's time here. MOVE ON ALREADY!!!!!!!!!!!!! Your continued persistence is becoming an affront to WP:DISRUPTPOINT that I'd applaud seeing you blocked. Seriously, MOVE ON. --ColonelHenry (talk) 03:36, 13 June 2013 (UTC)

I will move on. I agree with WhatamIdoing that I should "do something more useful than argue with stubborn people". It is obvious that you just hate citation templates for no rational reason (I asked you repeatedly to explain your objection, you answered with pure hatred). I realize you claim ownership over this article, and there's nothing I can do about it since you're backed by an administrator in CBM. I attempted to start a constructive disussion on Talk:The_Love_Song_of_J._Alfred_Prufrock, but I was rebuffed. I give up. Continue to add that wonky citation style you're favoring, I don't care anymore. The article is yours.

What I do care about, however, is that for the sake of Wikipedia we do get rid off this ridiculous misinterpretation of WP:CITEVAR, in which every citation style, no matter how different, is summed up as "non-templates style" and therefore considered "consistent". --bender235 (talk) 09:19, 13 June 2013 (UTC)

Thank you, I'm glad you decided FINALLY to move on. It isn't about ownership, it isn't about winning. I just wish you'd accept that policies likes WP:CITEVAR exist solely to keep people like you from aggravating other editors through such intransigently obstructive stances by which you INSISTED upon imposing your own desired personal style preferences and proceded to IGNORE voices (yes, including mine) who would work with you if you weren't so stubbornly INSISTENT, and where there were no reasonable circumstances demanding a NECESSITY to change things from non-templates to templates. I'll keep an eye on your contributions in light of this, and I'm sure others will too, as you've done this several times before (i.e. ignore WP:CITEVAR, etc.), and such continued action is potentially disruptive and pointy. --ColonelHenry (talk) 10:41, 13 June 2013 (UTC)

P.S. Sure, I can say I hate cite templates all I want. I am not compelled to like them. But, I'm open to a persuasive argument for cite templates and other templates (i.e. the usefulness of infoboxes if appropriate for a specific article) should one come up...however, no one in all my editing has offered me a persuasive argument to applaud their merits. When they do, I'll reconsider. Until then, I'll continue to hate cite templates. Based on your intransigence, Bender235, I don't like you either, but eventually when you stop being insistent and see the error of your ways, I could be persuaded otherwise.--ColonelHenry (talk) 11:54, 13 June 2013 (UTC)

This is just incredible. I have no idea what else you expect from me. I already came to you, apologizing for us starting off on the wrong foot, and asked what is it that you don't like about templates. Your reply was: "I have better things do with my time than to justify the way I feel about templates to you. I hate them. PERIOD." And now you say you're "open to a persuasive argument"? What shall I take from this? That you want to hear arguments, unless I am the one presenting them to you?

To this day, I have no idea what went wrong in this incident. I made changes to the said article, you reverted them, I started a discussion on the talk page. There was no edit war, no personal attacks, no nothing. And still you complain about my "disruptive actions", my "intransigence", my "ignorances". What in the world have I done to you? --bender235 (talk) 14:35, 13 June 2013 (UTC)

• simple, and mentioned several times. you continued to insist. you continued to badger. no one wants to hear a sales pitch after they've already told the salesman "no, i'm not interested". Seriously, when someone says "back off," you back off. You don't question the reasons they say "back off" when the reasons are clear (i.e. policy that says clearly "DON'T DO THIS"). You don't try again and again and again and again. Stop badgering people. If you wonder what you've done, count how many times you were told to "back off" or "move on" and you keep coming back and back and back again. You aggravated me with your relentless pursuit of this--long after you were told several times nicely and CLEARLY to back off. Obviously, the problem is your lack of comprehension of a simple request. For the nth time, MOVE ON. --ColonelHenry (talk) 14:43, 13 June 2013 (UTC)

I think with your mindset you're wrong to participate in a project like Wikipedia. Basically this whole debate went like this:

• "Should we use citation templates on this article?"
• "No, I don't like them. Now back off!"
• "What is it you don't like about them?"
• "Stop badgering! You're being ignorant just to prove a point. Move on!"

It was basically the epitome of the complete opposite of WP:DR. In the process of this incident I have been called "a dick", compared to a two-year old, accused of "badgering" and worse. All for asking a simple question. Unfortunately, all of this hideous behavior was backed by an administrator, CBM. I have no idea what to do now. One simply can't debate with someone who's only answer to everything is "Move on!". In nine years of working on Wikipedia, this is the saddest moment I ever experienced. Shame on ColonelHenry, and shame on CBM for letting this happen. I give up. --bender235 (talk) 15:00, 13 June 2013 (UTC)

• Well, if you didn't ignore, insist, badger and persist, after you got an answer you disagreed with, you wouldn't have brought it upon yourself. I'm typically a nice guy. But you chose to ignore, insist, badger and persist. Keep poking a sleeping tiger with a stick after being told not to, the tiger will eventually bite back. You shouldn't expect the tiger to keep taking it. I don't go out of my way to start shit with others, but I certainly don't take shit from others. And frankly, you were wrong for ignoring, insisting, badgering and persisting that I should. You reap what you sow. Au revoir. --ColonelHenry (talk) 15:09, 13 June 2013 (UTC)

• Perhaps if you recognized that your behaviour in this dispute has been completely unacceptable, ColonelHenry, the issue wouldn't exist. Bender235 made the citations consistent, and added information that was not currently in them. You then proceeded to object without providing any rational basis for your objection whatsoever. Why should anyone care what your opinion on a topic is if you refuse to discuss it rationally?—Kww(talk) 18:08, 16 June 2013 (UTC)

The point of CITEVAR

The following comment is a perfect example of the reason we have CITEVAR, ENGVAR, and the other MOS policies about stability:

Find something more constructive to do with your time instead of making it harder for me to work on the things I want to work on. Instead of contributing to articles today as I planned (I wanted today to finish preparing List of colonial governors of New Jersey for WP:FLC), I've wasted hours bickering with you. [15]

Experience shows that discussions about whether to use templates or not tend to go on without end, wasting large amounts of productivity, for no concrete benefit. Therefore, the policy is to simply leave some things as they are, rather than going around trying to convert articles from one optional method to another because of personal preference. — Carl (CBM · talk) 14:24, 12 June 2013 (UTC)

The problem you keep on ignoring is "leaving things as they are" would mean leaving an article with incomplete, inconsistent citations. However, citations are a key part of Wikipedia.

One way or another, this article needed a debate on how citations should consistently look like. Because there simply was no consistent citation style. There still is not (see above). So if in that debate a majority favored citation templates, why not implement them? --bender235 (talk) 17:01, 12 June 2013 (UTC)

...what's the point of doing that?

The point is to end the dispute so you can do something more useful than argue with stubborn people. If your efforts to add citation templates aren't wanted at that article, then feel free to either use a style that doesn't waste your time with arguing, or to let someone else clean up the mess. There are thousands of articles out there whose editors would be pleased to have your help. Why are you wasting your time on that particular one? WhatamIdoing (talk) 17:30, 12 June 2013 (UTC)

Because this sets a precedent. If CBM is allowed to continue to abuse CITEVAR in this blunt misinterpretation, it is bad for Wikipedia in the long run. For some reason, CBM is of the absurd believe that "no templates" qualifies as a "citation style", when in fact this is just bullshit. I'm sorry, but at this point I'm loosing my temper. See how he maliciously took ColonelHenry's quote out of context above: if you read the whole thread on ColonelHenry's talk page, you realize I approached him with good intent, asking whether I could help him with templates. I was answered with pure hatred. I was called "a dick" for pointing out that ColonelHenry's attempt to "make citations consistent" made the situation worse than it was before. And on top of it, CBM accused me of badgering. This is just evil. --bender235 (talk) 18:10, 12 June 2013 (UTC)

To be clear: the problem wasn't that ColonelHenry reverted my initial edits to the article. The problem was that after I started a discussion on the article's talk page in order to see what the consensus was, CBM immediately replied the potential addition of citation templates was no subject to be discussed, citing some obscure unwritten veto right entitled (for whatever reason) to ColonelHenry. It is that latter part I have a problem with. If there is consensus within the article's editor group to not use templates, I'm fine with that. But not based on some obscure ownership/veto right. --bender235 (talk) 18:33, 12 June 2013 (UTC)

No, the problem was ignoring whatever we said (WP:IDIDNTHEARTHAT), stubbornly insisting on imposing your personal preferences on an article you don't do anything with and saying "well, I'll just bide my time and do it anyway." (comments on my talk page) Seriously, find something better and more constructive to do with your time than wasting mine and others time with a blatant disregard for WP:CITEVAR and WP:CITECONSENSUS. I'm not interested in wasting my time further with this nonsense. You've badgered me, you're continuing to badger CBM and others and refusing to move from your stubborn position, and you're ignoring plainly written policy and choosing to fight it out instead of moving on. There will be no consensus. Several of us don't agree, you won't budge...seriously, stop wasting your time, their time, my time, and find some place else you haven't contributed to yet where your services are appreciated.--ColonelHenry (talk) 21:32, 12 June 2013 (UTC)

Also, thank you for expecting me to assume good faith when you've had this discussion going two days now, bantered my name around, and yet haven't informed me of it (I had to find out from a friendly concerned email from one of the participants above). Quite disingenuous, I must say.--ColonelHenry (talk) 22:10, 12 June 2013 (UTC)

"stubbornly insisting on imposing your personal preferences on an article" is exactly what you did. As CBM pointed out, the original citation style look like this:

• "On 'The Love Song of J. Alfred Prufrock'", from Blasing, Mutlu Konuk. 'American Poetry: The Rhetoric of Its Forms'. New Haven: Yale UP, 1987.

You changed it to this:

• Blasing, Mutlu Konuk, "On 'The Love Song of J. Alfred Prufrock'", in American Poetry: The Rhetoric of Its Forms (New Haven: Yale University Press, 1987).

The problem isn't that you changed it (because everybody is allowed to do that). The problem is you claim to have a special right to finally decide, because you contributed to the article more than I did. With is nothing else but claim of ownership.

"You've badgered me, you're continuing to badger CBM and others and refusing to move from your stubborn position, and you're ignoring plainly written policy and choosing to fight it out instead of moving on."

See, the problem right from the beginning was the you claim there is "written policy" that gives you a veto right to determine this articles citation style, when there is not. CITEVAR does not say such thing anywhere.

"There will be no consensus. Several of us don't agree, you won't budge..."

As far as I see, on Talk:The Love Song of J. Alfred Prufrock, there's you and CBM in favor of your citation style, and then there's WikiParker and me in favor of mine. So your claim "there will be no consensus" is a bit far-fetched, isn't it?

"Also, thank you for [...] haven't informed me of it"

I didn't thought it would be necessary since CBM told me to bring this question to WT:CITE (which is here) all the time. You followed our discussion, didn't you? --bender235 (talk) 07:31, 13 June 2013 (UTC)

Typically it's good form to advise other editors of a discussion elsewhere that involves their previous interaction on a topic. You failed to do so, but continued to banter my name around, which leads me to assume your act of bad faith. Thankfully someone saw fit to inform me.--ColonelHenry (talk) 11:31, 13 June 2013 (UTC)

I agree that this is the kind of discussion CITEVAR is meant to avoid. ColonelHenry wants to improve an article that needs improving, so please let him do it without sidetracking him. The refs appear to be consistent now, so there's no problem. SlimVirgin ^(talk) 22:59, 12 June 2013 (UTC)

He could continue to improve the article by adding content, and let others handle the copyediting, such as citation style. Wasn't this type of collaborative approach the original idea behind Wikipedia? --bender235 (talk) 07:33, 13 June 2013 (UTC)

yes, and WP:CITEVAR exists to prevent editors like you from showing up at an article only to dictate to the article's contributors your personal style preferences without any other contribution to it. You don't seem to grasp that the entire conversation hinges on the issue that an editor such as yourself is not supposed to UNILATERALLY switch style from non-template to template based on personal preferences--especially at an article you have had no involvement in. It is not a valid WP:BRD position. Who could care less if I used parentheses, or spelled out University Press instead of keeping the incorrect Yale UP, or that I evoked in instead of from--you are distracting from the main issue because you are WRONG on the main issue. That you're being obtuse is becoming disruptively pointy.--ColonelHenry (talk) 11:31, 13 June 2013 (UTC)

Now that I have had a chance to look at some of Bender235's recent contribs, he has made similar inappropriate changes to other articles such as [16] and [17]. The general principle of CITEVAR is that these changes should not be made in the first place. The same would apply if someone went around just removing templates from numerous articles, of course. — Carl (CBM · talk) 23:10, 12 June 2013 (UTC)

Do you think that the citations in those articles had a consistent style before the edits by Bender235? What evidence do you have that Bender235 edits were done "merely on the grounds of personal preference"? If not how does CITEVAR apply? -- PBS (talk) 07:24, 13 June 2013 (UTC)

I do those edits very frequently. What might surprise you is they are almost unanimously welcomed everywhere. In almost nine years at Wikipedia, I can remember only two case (including this one), in which my work had been reverted, and I remember both involving you.

Also, as an explanation: when I find an article that has various types of citation styles going on, like the ones above you mentioned, WP:CITEVAR actually encourages "imposing one style on an article with incompatible citation styles". That means, in any case, I'd have to pick one style over the others and implement it. And why then not use Harvard style via templates? --bender235 (talk) 07:31, 13 June 2013 (UTC)

CITEVAR certainly does not encourage taking an article with no citation templates at all, as the two above, finding some trivial difference between two citations, and using that as a pretext to convert the entire article to use citation templates. To put it very simply, so you cannot claim to be unaware of the rule any longer: do not convert articles don't use citation templates to use citation templates, and don't convert articles that do use citation templates so that they no longer use citation templates. CITEVAR encomasses other aspects of citation as well, but the use or non-use of templates is one of its core aspects. — Carl (CBM · talk) 09:26, 13 June 2013 (UTC)

Such rule does not exist. Regardless of your claims. I will not obey to a rule that only exists in your imagination. --bender235 (talk) 10:08, 13 June 2013 (UTC)

By the way: you're probably the only one who considered the differences in citation style in this article to be "trivial".

• McCoy, Kathleen, and Harlan, Judith. English Literature From 1785 (New York: HarperCollins, 1992), 265-66.
• Southam, B.C. A Guide to the Selected Poems of T.S. Eliot. Harcourt, Brace & Company, New York 1994, p. 45.

One uses brackets, the other doesn't. One has location: publisher, one has publisher, location.

Same goes for journal articles:

• Sorum, Eve. "Masochistic Modernisms: A Reading of Eliot and Woolf." Journal of Modern Literature. 28 (3): 25-43. Spring 2005.
• Walcutt, Charles Child. Eliot's "The Love Song of J. Alfred Prufrock". (1957). College English, 19, 71-72.

Just completely different. Only in your eyes these are consistent (since all are "non-template"). But that's just ridiculous. --bender235 (talk) 10:15, 13 June 2013 (UTC)

Well, Bender, when you bring 38 citations into conformity, anyone might miss one or two. BIG F&%$#@G DEAL. Seriously, you're wrong on the main issue (i.e. switching to templates unilaterally) so you'll now nitpick on an overlooked comma, etc. Give it up already. Seriously, my two-year old accepts when she's wrong and moves on. Move on.--ColonelHenry (talk) 11:38, 13 June 2013 (UTC)

Bender235, the articles linked to by CBM do not use Harvard citations. Harvard citations give a little information about the source in parentheses immediately after the statement supported, such as giving the author and year; full details of the source are given in a bibliography or equivalent section. Harvard citations can be accomplished with or without citation templates. Jc3s5h (talk) 10:00, 13 June 2013 (UTC)

I'm aware what Harvard citation is. The article in question used ColonelHenry's interpretation of Chicago style. --bender235 (talk) 10:08, 13 June 2013 (UTC)

Nothing is ever perfect, Bender, and typically when I do a massive overhaul of an article (as for several months I intend to do on Prufrock to bring it possibly to GA or FA status), the references are something that I fix or bring into conformity as I go along, and then check again after. Sorry, but while I was bringing them into conformity with minor changes (i.e. commas, date placement, parentheses, etc.) I didn't have a style guide next to me for reference and did it off the cuff. So, I concede, I might have missed a comma or not properly addressed the volume number. So, eventually, down the road, I'll repair the remaining stray comma issues as I go along. It is not of absolute necessity to rush to fix a benignly misplaced comma--sorry you consider it to be an impertinence, I consider it a petty distraction on your part to deflect from how wrong your main thesis is. But, I would reiterate: it is definitely not a necessity and indeed is quite the impertinence to waste everyone's time with a drastic and PRESUMPTUOUS non-template to template format changes against WP:CITEVAR, and to stubbornly insist upon it despite WP:CITEVAR.--ColonelHenry (talk) 11:20, 13 June 2013 (UTC)

It is not clear to my ColonelHenry why you do not use citation templates, or why you would object to someone else using them in this case. If you were to use citation templates when you do a "massive overhaul of an article" the you do not need to worry about "minor changes" of "commas, date placement, parentheses, etc." as that will be taken care of by the citation templates and you do not have to have a "have a style guide next [you]" as that is done automatically by the templates. If you do use citation templates you will not miss a comma and volume numbers are just another parameter. So why do you not use citation templates? -- PBS (talk) 12:26, 13 June 2013 (UTC)

Cite templates are a bitch to edit around, moreso IMHO than ref tags. We each edit in our own way, and I find if there's an article with cite templates on a topic I'd like to edit, I find myself more often than not walking away from the article because of it. I find it easier to control and edit citations in ref tags. I find my productivity sharply decreases if I have to deal with cite templates and other hinderances that I could just do manually. As SlimVirgin stated above, the cite templates are not consistent in results, and there's little control should they prove inconsistent in results. It's not a matter of not being familiar with them or the coding, I am quite familiar with them, I just find it more effective for me as an editor doing it manually. Lastly, it's not a matter of ownership, but I detest editors who do drive-by unilateral stylistic/formatting changes to an article (as in the present case) without offering anything else--especially for something so petty as saying "well, I like templates and will put them in anyway" despite doing nothing else. Any article on which I do a major overhaul I will post on the talk page and state what my intentions are, what improvements I'd like to add, I check the page's edit history, sometimes if there are recent contributions I contact other editors. It's just courtesy. My opinion on the matter is very much akin to showing up to insist upon putting an infobox in an article despite that article's major contributors already choosing not to. To steamroll your own formatting preferences (i.e. bender and cite template conversions) despite the history, disregarding input from an article's contributors, it's arrogant presumption--the type that WP:CITEVAR was implemented to avoid. And we've wasted more than enough time because one editor insists on ignoring what has clearly been stated in WP:CITEVAR (whether he knew about it before or not, he knows now, and still insists). --ColonelHenry (talk) 12:34, 13 June 2013 (UTC)

I'm going to pile on to ColonelHenry's answer. Until the creation of Template:Citation/core, the editors of each citation template did their own thing. Imagine if Chicago assigned a one editor to the section on citing journals, another editor to the section on citing books, and so on, and didn't require the editors to cooperate. Chicago would have been a flop and the only place you could find it would be a rare book shop. Now that templates are being converted to Lua, it's possible the bad old days might return, but I'm not sure. See Template:Citation/core#Consistent date location & absence of central discussion.

On a different point made by ColonelHenry, I see no objection to someone imposing the citation style of the editors choice on an article that lacks any consistent style, when no consistent style can be found in the article history, even if the editor isn't interested in adding substance to the article. If an editor is in the process of updating the citation style and doesn't want work-in-progress to be mistaken for no interest in citation style on the part of the editors, the Under construction template is available. Jc3s5h (talk) 12:59, 13 June 2013 (UTC)

@Jc3s5h: To your point on the related matter above, if Citation/core ever makes it work and makes it consistent in a way I'm satisfied using, I would consider reexamining my stance on using citation templates. Right now though, I will continue to refrain from using them and do it manually. As for changing citation styles unilaterally, this matter would have been better off had Bender posted on the talk page for Prufrock and after looking through page history sought out the input of the handful of recent editors who have been involved in the article (and maybe looked at the related T.S. Eliot articles for others to contact since it's a tight community). Instead, he gets ticked off because a recent contributor (me) reverted his edits as violative of WP:CITEVAR (then doubles down on his disagreement with the policy), and the subsequent argument has only forced me to draw my efforts away from preparing another article I've been focused on recently for the WP:FLC process, to wasting time arguing here and adding material to Prufrock that I hadn't planned to get to as a priority for another week or so. Next time, he'd avoid a larger issue by contacting contributors to an article he wants to convert and respecting that he should move on when they object. If it's a neglected article, have at it, but post on the talk page first and wait a day or two. I don't go into another person's home and start painting their bedrooms because I don't like their selected colour. It's just an expectation of courtesy. --ColonelHenry (talk) 13:43, 13 June 2013 (UTC)

The most common citation templates use Module:citation/CS1 which has big advantages over the the previous core code. So expect all citation template to be using it before long (and so considerations of "Imagine if Chicago assigned a one editor" are not relevant (and thanks to core have not been for some time).

ColonelHenry you write above "I find it easier to control and edit citations in ref tags". With this edit left in place {{reflist}} why not replace it with <references/>? In the same edit you introduced a citation template {{rp}} which was not used before you started editing the article and that is a much larger change in style than the {{cite web}} and {{cite book}} templates that you removed. So who was "for changing citation styles unilaterally"? -- PBS (talk) 15:41, 13 June 2013 (UTC)

I'm wondering that, too. Expect to be recommended to "move on" for that question.

@ColonelHenry:

"[...] this matter would have been better off had Bender posted on the talk page for Prufrock and after looking through page history sought out the input of the handful of recent editors who have been involved in the article (and maybe looked at the related T.S. Eliot articles for others to contact since it's a tight community). Instead, he gets ticked off because a recent contributor (me) reverted his edits [...]"

That's funny because after you reverted, I did just what you recommended: turned to the talk page and asked "why not use templates". The one who "ticked off" then was not me, but you, insulting me in basically every reply you gave. All under the eyes of an administrator, which is a pathetic display of discussion culture on Wikipedia.

"I don't go into another person's home and start painting their bedrooms because I don't like their selected colour. It's just an expectation of courtesy."

See, there's your basic misconception. A Wikipedia article is not your bedroom. It is not your property, regardless of how much you contributed to it. Nobody owns any article. --bender235 (talk) 17:15, 13 June 2013 (UTC)

But see WP:OAS. When someone is looking after the bedroom, it's rude to throw them out, shouting "You don't own this room!" SlimVirgin ^(talk) 19:32, 13 June 2013 (UTC)

True. But then again, there's a difference between yelling "You don't own this room, I'll decide how we color it!" and "You don't own this room, so let's figure out we to do, let's find a consensus." I did the latter. --bender235 (talk) 23:06, 13 June 2013 (UTC)

• You seriously must be deluded if you think that's what you did. It was more like "Hey, I like this room, haven't been here before but let's paint it red", then "sorry, we don't like it red, red doesn't work well and the rules say not to insist on red" and you reply "let's paint it red. Why don't you like red?" about another 50 friggin' times. You don't get it. You are an impertinence.--ColonelHenry (talk) 23:48, 13 June 2013 (UTC)

ColonelHenry, in the exchange above you seem to have missed the two questions I asked you. If you do not like citation templates why leave {{reflist}} in place? Isn't introducing the {{rp}} template a larger change in style than the templates you removed in the same series of edits? Personally I think that introducing the {{rp}} onto a page is as large a change as altering citations from Harvard style to footnote style, and in comparison whether a citation template such as {{cite web}} is used in place of a similar line of text is trivial. Yet you write "You don't seem to grasp that the entire conversation hinges on the issue that an editor such as yourself is not supposed to UNILATERALLY switch style from non-template to template based on personal preferences" placing unilaterally in capital letters ("bold"). (1) Introducing citation templates such as {{cite book}} does not usually alter a consistent style (it did not in this case). (2) Now that it has been pointed out to you that you unilateral "switched styles" by introduction of the {{rp}} template, do you think that your sentence with its bold word is reasonable? -- PBS (talk) 07:53, 14 June 2013 (UTC)

Having read the above, I think it is clear that WP:CITEVAR is being used to justify squatting on an article and preventing it from being edited. The original purpose of this guideline was to prevent an article from looking inconsistent to the reader: that it not be some bizarre mishmash of different styles of footnotes. Using it to give a single editor veto power over the use of templates is not what was intended. No articles that were created in 2003 had citation templates: they didn't exist at the time. Making wholesale changes that go against consensus is a problem, but I don't really see that Bender235 did that. There is a strong consensus that Colonel Henry will object to the templates, but there's no consensus that the templates are better or worse for this article. I don't assign any particular weight to Colonel Henry's opinion: he's just one editor among many. It seems to me that the only way to prevent this guideline from allowing a single editor to get a deathgrip on the article is to have an RFC on the preferable citation style for the individual article, with it being made clear that Colonel Henry's opinion on the topic is just that: Colonel Henry's opinion on the topic.—Kww(talk) 17:53, 16 June 2013 (UTC)

If you trace the history of CITEVAR you will see that citation templates have always been one of the key issues. I like them, but there has always been a significant number of editors, many of whom are active in featured content, who strongly dislike them. For that reason, many years ago consensus was formed that articles should not be unilaterally converted. For example see this version from 2008: [18]. By the end of 2009, the rule to defer to the first major contributor had been established [19]. — Carl (CBM · talk) 22:45, 16 June 2013 (UTC)

Separately, the last thing we need is to have hundreds of RFCs open all over the place about whether to switch each article to use citation templates on each one. They will all feature the same arguments, which have been rehashed on this talk page many times. This is why, as with British and American English, we defer to the first contributor as a practical way of making an otherwise arbitrary decision. — Carl (CBM · talk) 22:49, 16 June 2013 (UTC)

The issue with citation templates?

Something I always wondered: besides WP:JUSTDONTLIKEIT, what are the cons for citation templates? I'm always hearing the slowing the article's loading speed, but do they really? I'm wondering because part of WP:WCC is adding unique document identifiers such as DOI, PMID, JSTOR, etc., which if not added via the respective citation template parameter, one usually adds them with {{DOI}}, {{PMID}}, {{JSTOR}}, which then of course sums up to three templates per citation, instead of just one. So, is loading speed really an argument here, and whether or not, what are other arguments against citation templates? --bender235 (talk) 17:08, 13 June 2013 (UTC)

See Wikipedia:Centralized discussion/Citation discussion. Jc3s5h (talk) 18:49, 13 June 2013 (UTC)

Holy wall of text, Batman! All of it, or a particular section? Just realized it's this section. Of the ten points presented there, most are void, however. First of all, points 1, 2 ("they're too easy to use, people will add to many refs") contradict points 9, 10 ("too hard to use, can't fix errors quickly enough"). All other points can easily be fixed. The only on that actually remains is No. 4 ("They slow down load time, sometimes considerably."). Is there actual evidence for that? What is the latency difference between these two versions? --bender235 (talk) 19:02, 13 June 2013 (UTC)

Concerns with citation templates are sprinkled throughout. Jc3s5h (talk) 19:00, 13 June 2013 (UTC)

• Most of the technical issues have been resolved:
  • 4. There were a lot of performance issues when the CS1 templates were heavily used on a page. The speed issues are mostly resolved with the Lua implementation of the Citation Style 1 templates.
  • 5. The templates that use a meta-template have now been codified as Citation Style 1, which uses APA/Chicago adapted for the web.
  • 6. Templates are not changed on a whim. There is a high level of conservatism amongst the maintaining editors. We make changes only when there is a clear problem or after consensus is reached on a style change.
  • 7. The date formats are clear in the template documentation. There is a feature request to validate dates.
  • 8. "When they're added to a References section, every entry in the list starts with 'c'" I have no idea what this one means.
  • 9. The Lua based templates now have more error checking, and every error message links to a help page.
• For more, see Module talk:Citation/CS1/Updates. --  Gadget850 ^talk 19:17, 13 June 2013 (UTC)

On sorting a list of references

• The point about every line starting with "c" is that when you're writing an article you often need to glance down in edit mode and grab a ref from the References list. Templates slow that down (when there are a lot of them), because every entry begins with a "c," so finding something from an alphabetical list is more awkward. SlimVirgin ^(talk) 19:46, 13 June 2013 (UTC)

So they not only can be fixed, but already were. --bender235 (talk) 19:22, 13 June 2013 (UTC)

Gadget850, I am not convinced that particular list or section captures all the citation issues mentioned on that page. In any case,

• 8. When they're added to a References section, every entry in the list starts with 'c'

means that when editing a bibliography, all the entries start with "{{cite " which makes it more difficult to see if the list is in alphabetical order. I personally would add that since the parameters that control the order, the authors and the date, could be anywhere in the template, it is virtually impossible to tell if the list is in order. One must have two windows open, the edit window, and a reading window, to put a list in order. Jc3s5h (talk) 19:40, 13 June 2013 (UTC)

I frequently need to keep two windows open when editing, for example when alphabetizing a table. It is not that much of an increased burden. --Bejnar (talk) 20:55, 13 June 2013 (UTC)

For you (and for me). But it turns out that not all users are just like you. For example, many of them are editing from a mobile device, which usually makes two windows impossible. WhatamIdoing (talk) 23:21, 16 June 2013 (UTC)

Jc: possibily you have in mind doing references as a series of lines like:

{cite book|last1=Smith |year= ... }

{cite book|last1= Jones |year= ... }

and then sorting them. Right? An interesting idea (I admire it) but not really workable. E.g., 'cite journal' breaks it, and (depending on your sort tool) the extra space could break it. There are better approaches to alphabetizing, but we should discuss those somewhere more appropriate. ~ J. Johnson (JJ) (talk) 22:04, 13 June 2013 (UTC)

I'm discussing, and I think the people in the 2010 discussion were thinking of, sorting the list manually. If templates are not used, the cites would look something like

*Smith, A. B. (2013)...

*Jones, C. D. (1853)...

*Smith, A. B. (1999)...

Those would be fairly easy to sort manually. The following examples from the Universal Time article are not so easy:

*{{cite web|url=http://www.usno.navy.mil/USNO/earth-orientation/eo-info/general/date-time-def|title=Date and Time Definitions|publisher=United States Naval Observatory|ref=harv | accessdate = 3 March 2013}}

*{{cite book|authorlink=Peter Galison |last=Galison|first=Peter |title=Einstein's clocks, Poincaré's maps: Empires of time|location=New York|publisher=W.W. Norton & Co|year=2003|isbn=0-393-02001-0|ref=harv}} Discusses the history of time standardization.

*{{Anchor|ERC}}{{cite web|title= Earth Rotation Variations Due to Zonal Tides | publisher = Earth Orientation Center | location = Paris | accessdate = 2 October 2011 | url = http://hpiers.obspm.fr/eop-pc/models/UT1/UT1R_tab.html |ref=harv}}

Jc3s5h (talk) 23:16, 13 June 2013 (UTC)

  You did pick a tough one! But your example is tough largely because two of your three items have neither a cited author nor a date. (Which is quite typical for web citations.) Even if you wrote these out as text it would be hard to alphabetize them; you really have an unequal comparison here. And if you manually include the url links your text gets intricate. That is, you're not dealing with (as your first example implies) "Galison, P. (2003)...", but "[[Peter Galison|Gallison, P.]] (2003)...". Not quite so simple after all.

  Having said that, I agree that manual sorting of plain text is generally easier for not having all that meta-packaging. But there are ways of making templates easier to handle. E.g., using vertical (multi-line) rather than horizontal format. And always putting author(s) and date at the top (front). It is also possible to write a tool that would do the sorting, and in that event the ease of sorting would be totally in favor of templates. (I'm tempted.) ~ J. Johnson (JJ) (talk) 21:27, 14 June 2013 (UTC)

I'd like to add that inconsistent citation style examples can lead to link rot. I noticed that this page, Wikipedia:Citing sources, has a section on "Preventing and repairing dead links", but another section "What information to include: Webpages" has an example with a bare url. I believe it should read: "Citations for World Wide Web pages typically include: archived URL of the webpage. ~ Devolicious77 (talk), 15 June 2013 (UTC)

Incompleteness leads to link rot. Using multiple different styles, each of which present full information, just in a different order or with different punctuation, doesn't cause link rot. WhatamIdoing (talk)

I'm missing something here- where would you sort references other than the Shortened footnotes bibliography. --  Gadget850 ^talk 01:02, 17 June 2013 (UTC)

You would sort references for any bibliography that is used in conjunction with any kind of shortened footnotes or parenthetical citations, whether templates are used or not. Jc3s5h (talk) 01:06, 17 June 2013 (UTC)

To sort a bibliographic list that is formatted using templates, you are going to have to do it by hand. Does that sum this up?

I clean up the list to put it into author date order then use http://sortmylist.com/ to sort. Then manually sort to adjust for book/journal/news. --  Gadget850 ^talk 08:25, 17 June 2013 (UTC)

"... have to sort it by hand"? NOT ANY MORE!! If you have Perl available use my nifty script. It also generates sample Harv templates. ~ J. Johnson (JJ) (talk) 23:10, 26 June 2013 (UTC)