Wikipedia talk:Citing sources/Archive 30

Archive 25

←

Archive 28

Archive 29

Archive 30

Archive 31

Archive 32

→

Archive 35

How do I reference this?

If I found some information about Li Bai at this google book but at the bottom of the Li Bai section it clearly references that it got its data from books by Arthur Cooper and Arthur Waley, how should I footnote that? Do I need to mention either of the Arthurs?Active Banana (talk) 01:24, 29 May 2010 (UTC)

You should definitely cite the book that you obtained the information from, but I think it would also be helpful to readers if you provided information on the Cooper and Waley books. Perhaps you could phrase the footnote like this: "[Citation to book that you referred to], citing [citation to Cooper book]; [citation to Waley book]." — Cheers, JackLee ^–talk– 17:22, 29 May 2010 (UTC)

Thanks ! Active Banana (talk) 04:55, 30 May 2010 (UTC)

Question about Citing

Hi, I'm currently trying to do research for an article, and my problem is it's from the Associated Press but it's via yahoo, so I'm wondering who I cite as the publisher? I'm leaning on something like Associated Press via yahoo, but I'm not sure if that's right or not.--Deathawk (talk) 05:51, 1 June 2010 (UTC)

You got it from yahoo, and that is the source you should cite. You only have yahoo's word that it came from AP, so you should not say with certainty that it came from the AP via yahoo. You might say that is was published by yahoo, which cited the AP. At least that's my reading of WP:SAYWHEREYOUGOTIT. Wtmitchell (talk) (earlier Boracay Bill) 08:40, 1 June 2010 (UTC)

You can get better advice on this if you provide a URL. Here is an example citation of an AP article printed in the Tri-City Herald.

• Associated Press. "Ice deep-sixes deep-sea fix". Tri-City Herald. May 9, 2010. p. A1.

I hope this helps. – allen四names 01:31, 2 June 2010 (UTC)

Use {{cite news}} with |agency=; see the template documentation. ---— Gadget850 (Ed) ^talk 02:13, 2 June 2010 (UTC)

• Thanks --Deathawk (talk) 10:31, 2 June 2010 (UTC)

If you are editing an existing article, you should use the same method of citation as already in use in the article. If the article uses cite xxx templates, then Gadget850's advice is fine, but if not, you should not introduce those templates. Jc3s5h (talk) 13:26, 2 June 2010 (UTC)

If it's a story from the AP, we cite them, not Yahoo. SlimVirgin ^{talk contribs} 21:29, 3 June 2010 (UTC)

n:Wikinews:Style guide#Citing syndicated (wire agency) content

Many stories are provided by wire news agencies (e.g. the Associated Press (AP), Reuters, All Headline News (AHN), Agence France-Presse (AFP) or United Press International (UPI) that syndicate their content through other media outlets. Although the wire news agency writes the story, the carrying news media exercises editorial control in deciding whether or not to publish a story. Therefore, a report written by the Associated Press that appears in The Guardian should be credited as follows: the Associated Press as the author, and the Guardian as publisher. Where an AP author is cited this should be included. Where the abbreviation for the agency does not lead directly to a Wikipedia page (Eg w:AFP is a disambiguation page) the full name of the agency should be used (Agence France-Presse).

Whenever possible, choose the wire agency's site if the agency publishes its own stories. If this is not possible, try to pick a site that you think will have the story available online for the longest time, if you have more than one choice.

Articles from news sites which are initially from a wire service should have the wire service added to the author's name, or just the wire service if no author is given. For example, "author=Anne Gearan, AP".

---— Gadget850 (Ed) ^talk 15:41, 4 June 2010 (UTC)

Reference templates

Propose creating Category:Referencing templates for templates used to insert the citation into the document or to format the reference list. Currently, referencing and citation templates are all in Category:Citation templates. A similar proposal was made at Category talk:Citation templates#Subcategorizing footnote templates. For a list of referencing templates, see User:Gadget850/Reference templates. ---— Gadget850 (Ed) ^talk 12:08, 3 June 2010 (UTC)

Wiki doi is a covert advertising instrument

For a discussion of what is in my view a gross violation of our core principle Wikipedia:NOTADVERTISING, please see HERE. Gun Powder Ma (talk) 11:47, 11 June 2010 (UTC)

YouTube?

Is YouTube a reliable source? Acacia Ludwig was a backing vocalist for Red Hot Chili Peppers from 1995 to 1996, and there are numerous YouTube videos of her performing live with the band ([1], [2], and [3] for example). I need to cite her as a member on the List of Red Hot Chili Peppers band members, but I cannot find any other verification, other than one sentence on her father's band's MySpace. I'm not sure if MySpace can be used as a reliable source, either.. Any help would be greatly appreciated! WereWolf (talk) 17:29, 14 June 2010 (UTC)

1. Just because you think it's her, doesn't make that belief a reliable source. Is there captioning explicitly identifying her by name?

2. MySpace is a classic non-reliable source for a claim like that. --Orange Mike | Talk 20:58, 14 June 2010 (UTC)

By captioning, do you mean in the description box for the video, or text on the actual video? WereWolf (talk) 21:16, 14 June 2010 (UTC)

My take. It depends on the nature of the video. For example, if the video is a documentary, and features an interview with the band, including an identification of the vocalist in the video itself, that sounds okay with me. If it's just a video of the band playing, and you recognize the vocalist: no. That's WP:OR, essentially; it's your belief that the person depicted is who you think it is. If she's identified elsewhere on the page accompanying the video... well, that depends on who's doing that identification. If this is a YouTube channel that's the official RHCP outlet, or maintained by an otherwise reliable source, I wouldn't have a problem with it.

For this particular case, please see [4], where apparently the New Music Express, which I count as a reliable source, credits Ludwig as "Backing Vocals". I assume that this is not a fan-contributed section of NME.

Also, please see [5] and [6]. I don't know about their reliability, however. TJRC (talk) 21:19, 15 June 2010 (UTC)

Yes! I was going to use that NME article, but I wasn't sure about that. I'm pretty sure that is a non-fan created section, as well. Could I use that? WereWolf (talk) 22:01, 15 June 2010 (UTC)

Move

The following discussion is an archived discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the proposal was no consensus for the move. -- PBS (talk) 22:03, 18 June 2010 (UTC)

Wikipedia:Citing sources → Wikipedia:Manual of Style (citing sources) — Consolidating naming per Wikipedia_talk:Manual_of_Style#Poll Gnevin (talk) 16:23, 24 May 2010 (UTC)

Oppose; per my comment at the discussion there - this page is not primarily about style issues and its "master" page is WP:V, not WP:MOS. The primary message is, indeed, that citation style is a matter of only passing importance compared to citation completeness. Style details should be hammered out at the subpages related to specific referencing styles. Christopher Parham (talk) 18:50, 24 May 2010 (UTC)

Comment Then it should be removed from the MOS Gnevin (talk) 19:54, 24 May 2010 (UTC)

I'm not sure that's totally necessary...as it stands the page is part of the Manual of Style (because it contains some style information) but not a subpage of it (because it doesn't exist to provide further detail on a point of the broader page). This seems like a reasonable status quo to maintain. Christopher Parham (talk) 21:59, 24 May 2010 (UTC)

Pages that are part of the MOS should only provide stylistic details. I'll ask Wikipedia:Manual of Style to have a look Gnevin (talk) 22:42, 24 May 2010 (UTC)

The formatting for appropriate citing of sources clearly is a style issue. The questions of what sources to cite, when to cite them, and what information to put into the citation, are clearly not style issues. Maybe these should be separated out so that the style part can be included in MOS? —David Eppstein (talk) 22:48, 24 May 2010 (UTC)

I'd support this Gnevin (talk) 22:54, 24 May 2010 (UTC)

The style parts are included in MOS already, generally under WP:FOOT. Where useful, they are mentioned in this guideline as well. I'd be fine with not including any style advice here if that's what editors of the MOS would prefer, although there is very little beyond telling people to be consistent within articles, but I don't understand why this would be ideal. It makes sense to advise people of the few style rules we have at the same place we are telling them about when and where to cite. Christopher Parham (talk) 23:55, 24 May 2010 (UTC)

It assumes that <ref> is the most recent technique. It should also List-defined references, {{Harvard citation}} and {{sfn}} and their strengths and weaknesses. --Philcha (talk) 00:54, 25 May 2010 (UTC)

Since this is already covered at the Main MOS .I'm going to remove from the MOS Gnevin (talk) 08:54, 25 May 2010 (UTC)

Oppose This Guideline is only tangentially concerned with matters of style; it's primary concern is with policy related issues touching on Verifiability and Sources. If any change is needed, the suggestion of minimizing the style aspects and removing the links to MOS (Especially the header that says "This guideline is a part of the English Wikipedia's Manual of Style" seem most appropriate. --SteveMcCluskey (talk) 02:30, 25 May 2010 (UTC)

• Oppose not a style guideline. 70.29.210.155 (talk) 03:47, 25 May 2010 (UTC)

• Comment The main problem it that this page is both a guideline on why and when to cite sources and a how-to on presenting and formatting citations. ---— Gadget850 (Ed) ^talk 14:51, 25 May 2010 (UTC)
  • I have a hard time understanding why this is a problem...that seems to be a fairly coherent body of information. Christopher Parham (talk) 17:57, 25 May 2010 (UTC)
• Comment—Gnevin, you say "Pages that are part of the MOS should only provide stylistic details." I'm not sure I agree that there's a hard-and-fast boundary between style and other, related concepts. IMO, it will create more trouble than it's worth to insist on such a boundary. The MoS main page itself deals with matters that by some definitions are not stylistic. And some issues, take wikilinking, for example, involve stylistic and non-stylistic matters. Tony (talk) 11:28, 31 May 2010 (UTC)

The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Name of the publisher

WP:CITE#Citation styles says for books: "name of the publisher optional" but it ought to be provided if available, otherwise if a book is published on the same year on different sides of the pond, (not an unusual occurrence), the page numbers may not correspond in the different editions. If the ISBN is available then it could be argued it is not necessary, but that is true for all the other information as well. As it is not unusual for another well meaning editor at a later date to add the missing ISBN to a citation, one can not be sure that it is the correct one unless the other information in the citation can be cross checked against it and publisher is needed for that.

As publisher is regarded as part of the definition for a reliable source, "The word 'source' as used in Wikipedia has three meanings: ... and the publisher of the work (for example, The New York Times). All three can affect reliability." it rather mandates its inclusion here to help judge whether the citation is reliable and is useful for red-flagging a book as a self published source.

As many third party sources do not include publisher when they cite a source, it is a useful mechanism to find sources cited from a third party source in violation of WP:SAYWHEREYOUGOTIT. Of course all editors who know about WP:SAYWHEREYOUGOTIT would would follow that guidance, but not all editors have read the guidance, and including the publisher as standard would help other editors to find cases where the editor who adds a citation has not followed WP:SAYWHEREYOUGOTIT.

So I suggest name of the publisher should not be optional. Some archaic books do not have a publisher but in most of those cases they do have a "printed for in them" and where they have neither then that is a case of IAR because it is also true that not all books have an author or a publication date, but that does not stop us mandating their usage here. -- PBS (talk) 21:57, 18 June 2010 (UTC)

Re: Erectheum on acropolis in Athens - wrong photo

Re: Erectheum on the Acropolis in Athens

The wiki article on the Erechteum contains an old sepia photo that is stated to be of the Erechteum

This is incorrect - the sepia photo is of the Propylaea on the Acropolis not the Erectheum

Its a very beautiful photo - so it should be moved to the Wiki page relating to the Propylaea

BY the way - I think Wikipedia is fantasitic and have just provided this info to assist you.

Regards,

165.228.145.108 (talk) 07:01, 20 June 2010 (UTC)

Magazines Using Cite Journal

There is a disagreement at Wikipedia talk:Citation templates#Magazines [archived at Wikipedia talk:Citation templates/Archive 5#Magazines —Anomalocaris (talk) 17:03, 16 November 2016 (UTC)] regarding the use of the {{cite journal}} template or the {{cite news}} template when citing a magazine source in an article. Additional views would be useful. -- AnmaFinotera (talk · contribs) 20:50, 22 June 2010 (UTC)

What am I doing wrong?

I'm trying to add citations to the article Stonewall riots from a Village Voice story. I have a citation that reads as follows: <ref name="Truscott">{{cite news|url=http://news.google.com/newspapers?id=uuwjAAAAIBAJ&sjid=K4wDAAAAIBAJ&pg=6710,4693&dq=stonewall+inn&hl=en|title=Gay Power Comes to Sheridan Square|last=Truscott|first=Lucian|date=July 3, 1969|work=The Village Voice|page=1|accessdate=20 June 2010}}</ref> OK, fine, that appeared as a footnote. Then, I attempted to add <ref name="Truscott"/> and I got an error message. See [7]. I can't quite figure out what I'm doing wrong. Help! ScottyBerg (talk) 23:29, 20 June 2010 (UTC)

Never mind, I just figured out my mistake. I put the citation within another, lengthy citation. Arghhhh! ScottyBerg (talk) 23:58, 20 June 2010 (UTC)

The end of the error message referred you to the help page for that message. Did that help you? ---— Gadget850 (Ed) ^talk 00:04, 21 June 2010 (UTC)

No, because you see I had inadvertently placed my citation within the text of another citation! It was lengthy and I thought it was article text. I wouldn't expect that kind of situation to be covered by the help page. ScottyBerg (talk) 13:43, 21 June 2010 (UTC)

"This error will also show if <ref>...</ref> tags or a named reference are nested inside another set of <ref>...</ref> tags." ---— Gadget850 (Ed) ^talk 12:37, 23 June 2010 (UTC)

IMDb

Is the Internet Movie Database a reliable source? WereWolf (talk) 00:00, 27 June 2010 (UTC)

No— just like Wikipedia, additions can be made by anyone. {{imdb}} and similar templates are for the External links section. ---— Gadget850 (Ed) ^talk 11:28, 27 June 2010 (UTC)

It's a bit harder to add information to the IMDb; but not that much, in my experience: especially if you are yourself the subject of the listing. --Orange Mike | Talk 18:31, 29 June 2010 (UTC)

First editor's style set in stone?

You should follow the style already established in an article, if it has one; where there is disagreement, the style used by the first editor to use one should be respected.

Given an ongoing dispute between User:CBM and me, I was wondering how far this rule goes? For example, the article Bessel function, where CBM keeps defending the "style used by the first editor", although those very editors (User:Michael Hardy, User:Paul August) do not oppose the change from <references /> to {{reflist}}. Even the user who originally added <references /> to the article is in favor of the change to {{reflist}}.
I know this rule was a pragmatic compromise to prevent edit-wars over citation styles, but the way it is now (or at least the way User:CBM reads it), it puts an undue weight into one user's opinion: even if a dozen editors agree on the change, it only takes one person (could be someone who never contributed to the article before) to object, hence there's a "disagreement" and the established style remains. Is that how it was supposed to be? --bender235 (talk) 21:32, 5 July 2010 (UTC)

Omitted Information

There is no reference to Yusef Rahman Who was The arranger and Composer on Talking Book. Yusef is in the Book of "Who's Who of arrangers and composers"on this Album, I personally saw this and He introduced me to "Stevie" in LA in The Early 80's. I believe this should be added. He's higher in stature then a Recording Engineer. Thank you Randy Bluesman Hock —Preceding unsigned comment added by 72.184.206.165 (talk) 17:34, 6 July 2010 (UTC)

Ancestry.com as a reference?

Moved to WP:RS/N#Ancestry.com as a reference?

–xeno^talk 14:17, 9 July 2010 (UTC)

Citing BBC

There is a discussion concerning parameters which should be used for citing BBC News. Your input is appreciated. Beagel (talk) 11:06, 9 July 2010 (UTC)

personal witness evidence

It appears to me that an exclusive focus on academic footnotes has led to ignoring first hand witness statements... these are valid for court hearings so should be acceptable for Wikipedia... third party text sources are NOT accepted in court! But no protocol is cited for first hand witness statements. At present if I first write an article on my own web page then I can cite it because I can reference it...but as a first hand witness statement there has been no increase in validity! Even academic text allow for direct witness evidence or science articles reporting research would never get off the ground! —Preceding unsigned comment added by Napata102 (talk • contribs) 17:26, 13 July 2010 (UTC)

Wikipedia editors are either anonymous, or there is no independent proof of their identities. That is why their personal observations are considered unreliable and unworthy of inclusion in Wikipedia. I strongly suspect if you submitted an article to a science journal and signed the submission Napata102, with no return address and no institution affiliation, your submission would be rejected. Jc3s5h (talk) 18:31, 13 July 2010 (UTC)

The reason why first-hand evidence is admissible in court is because the witness is physically present in the courtroom, which enables the judge, jury, lawyers and members of the public to see that he or she is the person providing the evidence. In addition, the witness is subject to cross-examination so the accuracy of the evidence can be tested. Neither of these elements is present when someone inserts information into a Wikipedia article claiming: "Of course it's accurate, I was there when it happened." There is no way to ascertain whether the editor is who he or she claims to be, and no way to verify whether the information is correct or not. — Cheers, JackLee ^–talk– 18:46, 13 July 2010 (UTC)

Actually stop and think .. because what you are saying is quite UNTRUE. I never suggested that you could have a first hand witness that was anonymous...if someone wished to claim first hand witness then they would have to identify themselves and one could judge their reliability etc. ALL forms of evidence are capable of being challenged and actually a reference to a claim in another article is probably pretty weak when all is said and done...I really did not expect ad hominem remarks ... the fact that I am identified as napata102 is irrelevant because I am making an argument not a witness statement...when making an argument the person making it should be irrelevant to the evaluation of the argument .. we try to anonymise exam papers for this reason ...

My name is Dapo Ladimeji ... which adds nothing ..

There is no requirement for first hand evidence to be read in person to be accepted by a court. Either side can ask the witness to attend and of course if the evidence is disputed then the court would require that the other side have a chance to cross examine...(In fact in one court case the other side objected to a witness to giving oral testimony stating that the witness' statements were not disputed!) .. to cite this (unavailability for cross examination) as a reason is misleading because if I cite my own article you do not have a chance to cross examine me! The statement " In addition, the witness is subject to cross-examination so the accuracy of the evidence can be tested. Neither of these elements is present when someone inserts information into a Wikipedia article claiming: "Of course it's accurate, I was there when it happened." entirely misses the point by assuming that all first hand evidence will be disputed ... I don't think so. If the matter is disputed then an article reference is no better than a first hand witness statement ... also 'it must be true I was there ' ... is not a first hand witness statement and would not be accepted in court or by any legal process. There are many bigotted and untrue articles published every month.

The statement "I strongly suspect if you submitted an article to a science journal and signed the submission Napata102, with no return address and no institution affiliation, your submission would be rejected" is both insulting and untrue. Journals regularly allow anonymous comments, 'Foreign Affairs' famously published an article by X! An anonymous math article with proofs of an argument could easily be published in any scientific journal... As a matter of fact I have published an anonymous article years ago in an academic journal, I submitted it anonymously and it was published .. there was a witch hunt to try and find out who wrote it .. which is why it was anonymous in the first place .. but it was published...

The only point you are making ... which is quite obvious, is that first hand evidence must be shown to be first hand evidence and not an imposter and the only way to do that is by identifying the person claiming it.. First hand evidence is only along the lines ... I believe I saw/heard/smelt the following ... it cannot be I saw the X murder Y or I know that the reason B sacked D is F because I was there ... First hand witness can be mistaken, is subject to rebuttal BUT so are articles!

My direct reason for raising this is that I was invited to Obama's inauguration and I heard a song played almost all the time in the streets of Washington DC ... Clearly anyone else who spent time in Washington DC during the inauguration would also have witnessed this .. the song was Sam Cooke's 'A change is gonna come'... this seems to me an important part of the history of this song ...

Mutual respect and careful reading can go a long wa......y. —Preceding unsigned comment added by Napata102 (talk • contribs) 20:15, 17 July 2010 (UTC)

Multiple cites of a book but with different URLs

On List of fatal bear attacks in North America, there is an informative book that I'm using multiple times as a reference. I don't have much experience with this, but I know I could list it as one reference and then have a notes section for separate cites/page numbers. However, this book is available on google books, and for the benefit of the user, it is nice to link to the specific page view for more background information.

Is there anyway to do this besides having a separate reference each time I'm using this book? The book is Herrero, Stephen (2002). Bear Attacks: Their Causes and Avoidance.. you can see it used many times in the refs.

Thank you very much. --Omarcheeseboro (talk) 20:32, 13 July 2010 (UTC)

One common technique is shortened footnotes. With that technique, sometimes the full citations are removed to a separate article section and sometimes they are not. Another technique, not as common, is to use the {{Rp}} template. Wtmitchell (talk) (earlier Boracay Bill) 01:03, 14 July 2010 (UTC)

Google books

I feel as though this guideline should establish some clarity on how and when (if ever) it is appropriate to link to Google books. I for one think it is seldom appropriate to do so.

Deep links to Google books are only temporary. They expire after a fairly short period of time. Furthermore, content available on Google books itself changes periodically, and varies geographically. So a Google books deep link posted by someone in the United States may not work for someone outside the United States. There is also, as I recall, some concern over the legal implications of deep linking to copyrighted works under the DMCA, though experts would need to clarify this for the purposes of discussion. So, I think the bottom line is that deep links to Google books should be explicitly discouraged.

This leaves open the question of whether it is acceptable to link to the main page of the book (rather than deep linking to a particular content page). I don't see anything particularly wrong with this practice, except that it is obviously superceded by more robust methods of document retrieval, such as ISBN lookup (with which a book can be easily found on Google books, Amazon, WorldCat, etc.) Sławomir Biały (talk) 12:51, 14 July 2010 (UTC)

Personally, I link to the main page for the book when it's a full view (people can navigate to the page on their own), but will link to the specific snippet if all that's available is snippet view. I haven't had any problems so far with these links expiring but I'll look through my featured lists and see if any have. --Golbez (talk) 13:17, 14 July 2010 (UTC)

• I use The Wikipedia citation tool for Google Books, and it deep links by default. This raises questions I'd never considered. The US courts have not yet definitively weighed in on the question of linking to copyright violations; in Intellectual Reserve v. Utah Lighthouse Ministry they showed a tendency to take such concerns seriously, but no final judgment was rendered on that because the parties reached settlement. Unless there's been a new case with which I'm unfamiliar, there is not yet precedent here. I have been presuming that Google books (giant corporation and all) had achieved proper licensing for its displays. I had not read closely about the terms of its settlement but had vaguely developed an idea that it was all worked out. A bit of reading ([8], [9]) concerns me with respect to so-called "orphan works". It seems as though Google's "opt out" program is not so far different from YouTube's...in the absence of complaint, they'll make the contents available. Leaving aside the question of link stability, it seems like it might be a good idea to avoid deep linking to material that does not meet our own criteria for public domain (pre-1923, barring other evidence of PD) unless there is some final settlement securing Google's right to display it. --Moonriddengirl ^(talk) 13:40, 14 July 2010 (UTC)

I would like to see an additional parameter (say gbid) to add the Google Books ID, and then have {{Citation}} produce a stable Google Books link. Also, if parameter page has a value, the template could produce a stable deep link via the &pg=PA... parameter in the Google Books URL. It should look like this {{Citation |title=Wikis for Dummies |page=90 |gbid=5VXgXlU7g-YC |etc... }} produces "Wikis for Dummies, p. 90 (at Google Books)", or something like that. --bender235 (talk) 13:46, 14 July 2010 (UTC)

Good questions above re linking and deep-linking to Google Books. I'm located on a small island in the Philippines. I do a lot of wiki editing, and I place a lot of Google deep links crafted with the http://reftag.appspot.com/ tool, using {{Citation}}, {{Cite book}}, and plain wikicode. I also place Google deep links using shortened footnotes and the {{Rp}} tag — These issues reach beyond Template:Citation/core. Wtmitchell (talk) (earlier Boracay Bill) 23:45, 14 July 2010 (UTC)

I don't see a problem with this in general; we shouldn't be pre-emptively assuming that there is going to be some legal problem in the future if there isn't one now. If in the future one arises a bot could go through and fix links to go to the front page of the book or whatever. At the very least I can't see how it could be a problem to deep link to a public domain work on Google Books. Google gets an HTTP_Referrer header when someone clicks on a WP link; they can just shut off or redirect this kind of traffic - to any page they desire - if they want to have more control of the eyeballs.

(But they don't do that, so I'm inclined to think that they don't mind deep linking or they even want deep linking. There's no reason they even needed to make deep linking possible in the application, but they did.)

It seems like a horrendous waste to me if you're actually doing your research on Google Books and you have the link immediately available to cut and paste, to not do that and force a WP reader to search through the book to find your reference. One of the major benefits of Wikipedia is often being able to immediately and rapidly check the citations of an article, IMO.

Also, I have not experienced the phenomena Sławomir Biały describes of these links being temporary; I just went and examined a sample of my links from two years and more ago and all but one of them worked. You may perhaps be talking about something I've seen on "Limited Preview" books with particular licensing conditions set, where some kind of licensing token appears in the URL as you browse. I've noticed that if you remove the licensing token from the URL you can't access the book any longer; so I wouldn't be surprised if the token has a time limit on it too. But I have only seen those licensing tokens appear on a very small percentage of the books I've looked at.

As for including the ISBN, I usually do that and the other bibliographical information in addition to the link, but it occurs to me that even if someone only leaves the link you could construct a bot to go in and fetch the other details from Google Books. --❨Ṩtruthious ℬandersnatch❩ 00:03, 15 July 2010 (UTC)

• This should apply not only to Google Books, but other hosts such as Scribd.
• Linking to a specific page precludes reuse of the citation; separate citations would need to be created for each use.
• {{Cite google book}} uses a bot to fetch the info, but it does not seem to be active yet.

---— Gadget850 (Ed) ^talk 08:44, 15 July 2010 (UTC)

Strange, regarding links expiring I just received a message that I reached my "viewing limit" for a book I've been using heavily in references (see my earlier post here). In another browser, it worked fine. This looks like another reason not to deep link. --Omarcheeseboro (talk) 14:09, 15 July 2010 (UTC)

That's something which only happens with "Limited Preview" books with particular licensing settings; so as I said above, this wouldn't happen with a public domain book at all. Also, I would expect that it's only you who reached the viewing limit, but other people / IP addresses who have never viewed the book can still follow your citations. If you're talking about the links to the book Bear Attacks: Their Causes and Avoidance I just went and clicked on a couple of them and they worked for me, for example. In addition you'll probably be able to go back and follow them yourself if you wait a few weeks or months.

I am beginning to think that the people who are opposed to deep links are the ones who are simply unfamiliar with the way that Google Books works. --❨Ṩtruthious ℬandersnatch❩ 21:05, 15 July 2010 (UTC)

Thanks, I do realize I reached some sort of limit. As I said, it worked in another browser. I've been referring to it a lot, so it's unlikely a user would also reach this limitation. However, tt would be jarring for the user if they encountered this. I still will use the deep links on this limited preview book, but it's something to keep in mind. --Omarcheeseboro (talk) 23:09, 15 July 2010 (UTC)

Since we're talking about Google Books I thought I'd mention an interesting situation I came across: as you may know a book can have an ISBN number but not really be "legit"; i.e. could be self-published or otherwise not a reliable source. I just came across one that really looked like a genuine academic work - and maybe it still is, I guess - but it wasn't until I looked for it on WorldCat that I found that the book had no WorldCat entry and that it's self-published. So it pays to cross-check. --❨Ṩtruthious ℬandersnatch❩ 03:28, 18 July 2010 (UTC)

Why and when to cite sources

I would like to add: Be sure to separate between what the source said and what actually transpired (and of which we know nothing). That way the reader can apply source criticism and form a personal judgment. The reason I would like to add this is because the wikipedia is positing an image of what is actual. We know not what is true, but often information is shown as a 'fact', while it is merely derived from a source, which may or may not be correct. It seems prudent to lay a bit more emphasis on this sine often sources are made into 'absolute truth'. —Preceding unsigned comment added by Faust (talk • contribs) 12:19, 8 July 2010 (UTC)

I don't think I agree. I want to say (with a reliable source, of course) that "X dug up an artifact in the desert," not that "News at 11 said that X dug up an artifact in the desert." My "reliable source" may be wrong. But it will serve until a better source can be found. And to qualify all statements would drive everyone nuts. Often, BTW, most sources are not even notable. So putting them in the text (if that is what you are asking) would be WP:PR for them and even considered WP:SPAM IMO. But maybe I misunderstood you. Student7 (talk) 15:41, 12 July 2010 (UTC)

You misunderstood me. My contribution was ment to counteract exactly what you are pointing to. I have seen many remarks at the wikipedia that are not true, but can be sourced. I might, for instance, give 100000000 sources that the world is flat. However, the world, in fact, is round. To prevent eager readers to think that the world is flat, a note saying that this is the source talking and not a fact is beneficial. To be able to judge the source will also reveal the irrelevance of certain sources and would, in effect, eliminate the problem you are raising. Should I choose different words, seeing as you misread me? --Faust, formerly Arjen (talk) 07:43, 16 July 2010 (UTC)

I think that "source said that..." should be very sparingly used. It is confusing IMO. WHY is the source being emphasized here? Because he alone says the world is round? Well, maybe useful then. But mostly should be omitted IMO because it just clutters article. I think you are saying it should be used when (and only when, I hope) that the source is offering an opinion which appears to differ from mainstream (what is widely understood, but wrong). You may have a point. Student7 (talk) 21:05, 19 July 2010 (UTC)

Hi Student, I am suggesting that the mainstream opinion is wrong 99,99% of the time (the world is oval, for instance). If the wikipedia is going to be a solid source of information, instead of disinformation, we should leave it up to the reader to decide if a source is correct or not. If we say something is x, the reader will not be thinking of source criticism and will stop thinking altogether, believing the final answer has been given. Do you see my point? --Faust, formerly Arjen (talk) 10:53, 22 July 2010 (UTC)

Again, if the statement reads "The world is a cube (cite)" this is going to be a bit hard to accept for the average reader. It would be better if the editor inserted "Albert Einstein said 'the world is a cube' (cite)." The reader is less likely to skip over the article figuring that it has been vandalized.

But most of the time, the facts we are presenting are mainstream. They may be "truncated" which is why they are not precisely accurate. I don't necessarily want to know all the details. A generation ago, people would accept that electrons travel in "orbits" around the nucleus of the atom. Physicists figured that this was easier to teach to the general population. Nowdays, our articles probably don't state this (!). But "thre truth" is a bit harder to explain, which is why it got truncated before. (not sure we know "the truth" BTW). Student7 (talk) 16:22, 27 July 2010 (UTC)

Hi Student, this is my point exactly. Is it not beneficial if we make sure that everybody understands that it is a source talking and not a 'fact'? That way a reader can accept the explanation given as a handhold, but that same reader will also know that there are probably more opinions out there and can decide later to do more research. Is that not the entire point of finding a number of different sources for a term anyway? --Faust (talk) 17:00, 27 July 2010 (UTC)

References and citations

I think in the context of Wikipedia "The word "reference" may refer to the citation, the source, or both." is not correct. I have raised this point before on this guideline talk page and I am raising it again because in my experience most editors use the word citation as a shorthand for source cited in an "in-line citation" and "reference" for a source cited in ==References== section.

So editors will say this article contains references but no citations. Templates such as {{unreferenced}} and {{refimprove}} by their very names indicate this.

I have just raised the issue of changing the wording of the unreferenced articles project. It is clear that the current wording supports the interpretation I have given above and not that currently given in this guideline. Changing it to fit this wording will make it needlessly more wordy as everyone currently reads references to mean the content of a references section (whether as a list generated by {{reflist}} of as a bullet pointed list).

So I suggest that we change the definition in this guideline to describe Wikipeida editors usage and not that of external sources. -- PBS (talk) 22:54, 15 July 2010 (UTC)

But, <ref>...</ref> tags, {{ref}} or some variation to present the in-text cite whereas the citation that shows in the reference list is often presented using a cite template. ---— Gadget850 (Ed) ^talk 01:14, 16 July 2010 (UTC)

This is a semantic question, of course, and the way a linguist solves a semantic question is with examples. Suppose I'm talking about a Wikipedia article and I say:

• "This article has really bad sources." I might say this if the source is a blog written by a high school student.
• "This article has really bad citations." I might say this if the article's citations don't appear in footnotes and consist of book titles in capital letters at the end of sentences, like so. STARSHIP TROOPERS, R.A.Heinlein.

But, in either case, I might say:

• "This article has really bad references."

Thus, "references" is ambiguous between source and citation.

I am not familiar with the usage you describe above. If you said,

• "The citations are great, but the references are terrible."

I would probably guess you are trying to say:

• "The citations are great, but the sources are terrible."

And I certainly would not think that you were trying to say:

• "The inline-citations are great, but the general references are terrible."

I would also like to note that our use of terms should reflect how most English speakers talk, not how Wikipedians have come to talk, if we expect these guidelines to be intelligible to anyone beside ourselves. ---- CharlesGillingham (talk) 05:26, 17 July 2010 (UTC)

It is more this "The articles has a list of references but it does not cite any sources". When people refer to references they mean a list of citations in a reference section. -- PBS (talk) 02:22, 18 July 2010 (UTC)

I would find that very confusing, if you said that. I would have to ask you to clarify what you mean. Anyone else? ---- CharlesGillingham (talk) 06:57, 18 July 2010 (UTC)

On second thought, I would interpret that sentence the same way you would. It would take a second, but it's the only interpretation that makes sense.

I think that the source of the disagreement is that you're hearing the word "citation" as short for "inline citation" (or what APA calls a "brief citation"). What you (PBS) are calling "references" I think of as "bibliographic citations". The full-length bibliographic citations that we usually list under References are citations as well, I think, or maybe more precisely they are part of the citation. At least that's how I hear the words.

I also think you are using the word "reference" in a sense that is more restricted than I do. I would have no trouble pointing to something like "(Turing 1950)" and saying "that's an excellent reference; I've used it before". (I would be referring to the source) I would be confused if you said that "(Turing 1950)" is not a reference. Would you say it's not a reference because it doesn't appear at the bottom of the article in the References section? ---- CharlesGillingham (talk) 07:16, 18 July 2010 (UTC)

Purdue has a nice APA guideline.[10] Looks like references are the fully formatted citations that appear in the reference list, whereas in-text citations are the short form.

Systems such as <ref>...</ref>, {{ref}}, {{harv}} and the like could be called "in-text citation systems" and {{citation}}, {{cite xxx}} and the like could be called "citation formatting systems". Templates such as {{reflist}} could be "reference list formatting systems".

You may be interested in Wikipedia:Village pump (proposals)#Citation templates: naming by style.

Probably the most popular practice is the use of the Cite in-text citation system in conjunction with Cite xxx templates (what I refer to as CS2). It is a bastardized hybrid, but none of the formal systems such as APA are designed for online publishing. ---— Gadget850 (Ed) ^talk 14:08, 18 July 2010 (UTC)

Okay, I read a few parts of APA guide, and I think I may be wrong about "citation", or at least not completely right. Sorry.

I'm still pretty certain that "reference" can refer to the source. If I say "the reference was written by Bill Clinton himself" you would probably think I was saying the Bill Clinton wrote a book that I'm using as a source. You probably wouldn't think that I was saying that Bill Clinton added a "citation" to the reference list at the end of a Wikipedia article. ---- CharlesGillingham (talk) 07:43, 19 July 2010 (UTC)

From Wiktionary:

• Source: The person, place or thing from which something (information, goods, etc.) comes or is acquired.
• Reference: (academic writing) A previously published written work within academic publishing, used as a source for theory or claims referred to which are used in the text.
• Cite: To list the source(s) from which you used information, words or literary or verbal context.
• Citation: Enumeration; mention; as, a citation of facts.

From Wikipedia:

• Citation: a reference to a published or unpublished source (not always the original source). More precisely, a citation is an abbreviated alphanumeric expression (e.g. [Newell84]) embedded in the body of an intellectual work that denotes an entry in the bibliographic references section of the work for the purpose of acknowledging the relevance of the works of others to the topic of discussion at the spot where the citation appears. Generally the combination of both the in-body citation and the bibliographic entry constitutes what is commonly thought of as a citation (whereas bibliographic entries by themselves are not).
• Reference: a previously published written work within academic publishing that has been used as a source for theory or claims referred to that are used in the text.
• Source text: a text (sometimes oral) from which information or ideas are derived.

---— Gadget850 (Ed) ^talk 13:27, 19 July 2010 (UTC)

Do you think Wikipedia is the most reliable source on this? ---- CharlesGillingham (talk) 23:33, 19 July 2010 (UTC)

If the articles are wrong, then it is going to be very hard to create guidelines. Wikipedia:Parenthetical referencing already references Parenthetical referencing. ---— Gadget850 (Ed) ^talk 03:03, 20 July 2010 (UTC)

I am not talking about what other places or style guides call them, I am referring to the usage in a Wikipedia talk page conversation: references are used to mean the list inside a "reference section" just as "external links" is taken to mean those things listed in the "External links section". If an external link is inside a ref-tab, editors don't usually refer to it as an external link on the talk page instead they talk about it being cited. -- PBS (talk) 04:52, 25 July 2010 (UTC)

As I said above, the usage of these words on Wikipedia's talk pages is irrelevant. If Wikipedians have begun to use some words in a peculiar way, they should stop, and speak English. We're supposed to be writers, after all. And more to the point, these guides are supposed to be intelligible to any educated reader. Our audience is new users, not long-time Wikipedians. If we start redefining words, we're sunk. We're a subculture.

(I hope that doesn't sound too harsh. Note that I don't think you are defending a truly peculiar usage of the word "citation". You're not wrong, about the semantics, not entirely. However, the underlying issue (about usage) is very clear cut for me.) ---- CharlesGillingham (talk) 06:56, 25 July 2010 (UTC)

"the usage of these words on Wikipedia's talk pages is irrelevant" So I take it you are not a strong believer that guidelines should reflect usage. It is quite usual for a group to develop their own specialised vocabulary to suit their needs, and proving it is used within the group (and not to confuse a general readership) is done all the time and is acceptable. For example the use here of the word guideline has a specific Wikipedia meaning as do many of the acronyms used in conversations on talk pages. Providing the words are defined and explained (for those not familiar with them eg new entrants to the group) and they are in common use among the group then they are perfectly acceptable. Indeed it is misleading for new editors not to be told what the common usage is. -- PBS (talk) 03:43, 26 July 2010 (UTC)

People can say whatever they want on talk pages, but the guideline (and our articles) should reflect English usage. These pages are for first time users who stop by to find the answer to a single question. They should not have to learn a new language to do that. We should not require that new users become wikipedians before they can navigate the guideline. The definitions are needed to separate the issues.

Having said that, I think we could use a definition of "citation" that indicates that it is attached to article text (with a footnote or parenthetical reference), i.e. that it is "inline". I'm just not sure how to say that clearly.

I think changing the definition of "reference" is a mistake. ---- CharlesGillingham (talk) 08:18, 27 July 2010 (UTC)

Other topics

May I add that I do not care for parenthetical citations? IMO opinion they distract from the article. This forces the perhaps unwilling reader to mentally "skip" over the parenthetical stuff. I really don't care that Harvard, Oxford and Albert Einstein all used them extensively. I find them distracting.Student7 (talk) 14:25, 20 July 2010 (UTC)

Hi, Student7. Parenthetical citations are like any other technique, they can be used well or badly. The harv... templates include options for every situation I can foresee (that's rash :-D) - for example {{harvnb}} is very unobtrusive, while e.g. {{harvcol}} is for cases where page numbers are required. --Philcha (talk) 15:10, 20 July 2010 (UTC)

That is getting off the topic of terminology here. ---— Gadget850 (Ed) ^talk 17:44, 20 July 2010 (UTC)

As long as we are airing personal preferences, let me just say that I find footnotes far more distracting than parenthetical references, because footnotes force the unwilling reader to jump down to the bottom of the "page" (which in long Wikipedia articles can be half a mile or more) and back again, just to see what all the fuss is about. However, I have a question for Philcha about these Harvard citation templates with "options for every situation": "Harvard ciitation" is of course not a specific style, but rather a catch-all for a miscellany of styles using author-date format. I have not been able to find a template that correctly formats the bibliographical data for Chicago Style in the List of References—they all seem to want to enclose the year of publication in parentheses, for example, or boldface journal volume numbers. Am I just looking in the wrong place?—Jerome Kohl (talk) 16:53, 20 July 2010 (UTC)

Again, wandering off topic. I am not aware of any Chicago style templates. See Wikipedia:Village pump (proposals)#Citation templates: naming by style. BTW: With an article formatted with in-text and citation templates, when you click the in-text link, the focus moves to the citation in the lists (and highlights most), pressing back (for most browsers) returns to the text. ---— Gadget850 (Ed) ^talk 17:44, 20 July 2010 (UTC)

Again, not sure that this remark should be in a separate subsection. I agree that any reference is better than no reference. That said, Wikipedia is written for the general population. There are no popular periodicals and few scholarly ones written that include parenthetical references as a matter of course. IMO we should be discouraging parenthetical references so Wikipedia will continue to be more readable. Mercifully, few editors use that anyway. I'm just sorry it is allowed at all. Student7 (talk) 22:22, 26 July 2010 (UTC)

(I am on a child-friendly computer which sometimes makes unexpected changes to text as it did above! Had no idea it was going to do that! I have been avoiding specific articles and titles with the 3 letter word starting with S**. I am not on this type of computer because I like it!  :) Student7 (talk) 22:45, 26 July 2010 (UTC)

Any

Just wondering if 'any' should be removed, per User:Tony1/Redundancy exercises: removing fluff from your writing#Removing one or two words, problem F. Kayau Voting IS evil 07:07, 28 July 2010 (UTC)

New refs with only author and year, without work title

Apparently one of our newer contributors thinks newly added references, with a last name and date (and occasionally a page number) are perfectly permitted by this policy. Is he correct? See Talk:State_church_of_the_Roman_Empire#Badly_composed_refs. I'm very suspicious of his whole endeavor. -- Kendrick7^talk 04:33, 5 August 2010 (UTC)

The citations are acceptable, provided that the work being cited appears in the alphabetical list of references. The few I spot-checked satisfied this requirement. This is standard practice in academic writing. Jc3s5h (talk) 05:00, 5 August 2010 (UTC)

Is it appropriate to edit refs to remove valid online links to the source material? For instance, should the bluelink to the excerpt at google books for the following quote be removed from the reference:

In 380 CE Emperor Theodosius issued the Edict of Thessalonica, which established Christianity as the official state religion, specifically the faith established by the Council of Nicaea in 325:

example Reference

μηδείς (talk) 17:13, 5 August 2010 (UTC)

I don't think there is any definite rule. An argument in favor of removing it (provided the link is available in the alphabetical reference list) is that one purpose of the shortened footnotes is to make the body of the article easier to edit, and having long URLs in the midst of the article body makes it hard to edit. Jc3s5h (talk) 17:27, 5 August 2010 (UTC)

OK, my mistake. I didn't realize the titles of the works appear in the external refs section. -- Kendrick7^talk 18:16, 5 August 2010 (UTC)

You seem to be referring to State church of the Roman Empire which uses shortened footnotes; See WP:CITESHORT. You should not have the link in the short note, as it is listed in the References section. The article uses long and short notes, and styles should not be mixed in articles. You can optionally link the short note to the reference using {{sfn}}. ---— Gadget850 (Ed) ^talk 19:40, 5 August 2010 (UTC)

formatting from source to citation

When it comes to citing web sources, I always maintain the formatting of the source's title. Occasionally I copy over an ALL-CAPPED source, and that gets dropped to lower cases and I generally disregard it. However, at Daniel Lakin, editor Kumioko (talk · contribs) has been changing "Medal of Honor Recipient" to "Medal of Honor recipient"; the former version is as presented at [11] by the US Army. Is there any SOP for the transcription of source formatting to the citation in the article? — pd_THOR ^{| =/\=} | 17:58, 5 August 2010 (UTC)

As I had mentioned to Thor my opinion was that this type of thing falls under Wikipedia:Manual of Style (capital letters) that the house style of WP is in natural caps and that full caps should be avoided unless under certain circumstances. The following is an exerpt copied from my talk page to help elaborate my reasoning for these changes. "I completely agree that we should carry the appropriate title across, thats one of the reasons I have been making so many edits to them [for the Medal of Honor recipients]. Because many do not have the accurate information. But I do not agree that the capitalization must remain the same. Using the same logic, if the whole title was in caps we would not leave it that way and I have been told by several that it is not necessary to maintain the capitalization of the reference title, just the title itself. A prime example is the individual names on the Medal of Honor website. I previously reported them as caps in the citation because thats the way it is in the website and was told by at least three other editors that maintaing the caps was not necessary and it was in fact preferred to use natural capitalization for citations because its easier for some readers/editors to read." I also support and agree with Thors decision to bring this here for discussion rather than the 2 of us hash it out individually and I am prepared to live with whatever concensus decides and am prepared to go back and correct these changes if needed. --Kumioko (talk) 18:17, 5 August 2010 (UTC)

There's no rule against using title case for titles in citations, or any rule requiring it. The MOS page specifies sentence case for Wikipedia's own contents, but does not require that it be used for citations. Each individual article is permitted to choose its own citation formatting system, including whether the citations will use title casing or sentence casing. WhatamIdoing (talk) 19:25, 5 August 2010 (UTC)

The key is to be consistent within an article. ---— Gadget850 (Ed) ^talk 19:33, 5 August 2010 (UTC)

With regards to consistency in that case, need that be consistence with regard to a particular capitalization style (i.e. title case v. sentence case), or can it simply be consistent in that all of the cited sources' formatting are copied verbatim? I generally prefer the latter because I feel it is imparting the intent and specificity of the original source, for however little it might impact. — pd_THOR ^{| =/\=} | 06:45, 8 August 2010 (UTC)

Use either title case or sentence case, but don't mix them. Use citation templates or hand-crafted citations, but don't mix them. Use the APA based templates or the Vancouver based templates, but don't mix them. Use long footnotes, shortened footnotes or parenthetical referencing, but don't mix them. ---— Gadget850 (Ed) ^talk 10:22, 8 August 2010 (UTC)

List-defined references and reflist

A quick note about this (relatively) new way of making references. It is an improvement of the current syntax and it does makes the syntax clearer. Slightly. But it's a lot of efforts to move every references manually. And the Usability Initiative is currently working on template folding. Template folding will be incredibly better than this reference hack. And when using template folding, it will become a nuisance to unfold a reference and find "<ref name="Moss-VH12002-07-22" />". Which will basically mean that you have to open a new tab and edit the last section of the article to finally see the damn reference. Template folding and reflist won't be compatible blend in nicely.

In short: it's not worth the effort to change the references syntax just to change it back a few months later when template folding will be released. I suggest to remove it from the guideline for now. Or at least explicitly mention that reflist may not be a future-proof feature. Dodoïste (talk) 00:41, 8 August 2010 (UTC)

Erm, if template folding and reflist aren't compatible, then the usability team needs to take a strong look at what the heck they are doing. Reflist is one of the most widely used templates on the site. They're supposed to make things easier, not vastly more difficult. — Huntster (t @ c) 00:46, 8 August 2010 (UTC)

It would help if we were pointed towards discussions on this. As to "a few months", I'm sure there is an intent, but we've been down this road before. If the UI aims to make referencing easier, more intuitive and more standard, then it really needs to start afresh instead of adding one more hack to the pyramid of hacks that is the existing system. ---— Gadget850 (Ed) ^talk 00:57, 8 August 2010 (UTC)

I meant that the two features won't blend in nicely. It won't be a software issue. But usability-wise, the advantages of the refs at the end of the page becomes soon an inconvenient with template folding.

The relevant document would be Citron_Designs#Templates. Dodoïste (talk) 03:39, 8 August 2010 (UTC)

Having the references in the reference section rather than inline makes a world of sense. Template folding under the name of code folding has been around for years and will compliment list-defined references. Jack Merridew 16:02, 8 August 2010 (UTC)

Edit request from 203.45.100.59, 11 August 2010

{{editsemiprotected}}

203.45.100.59 (talk) 06:56, 11 August 2010 (UTC)

Not done: No request. --Stickee (talk) 07:05, 11 August 2010 (UTC)

inverting authors' names in footnotes

Unaware of the (to me) screwy examples of footnotes on the project page, I perpetrated this in one article, but was promptly corrected (?) with this. Which is what prompted me to look at the project page.

In the examples of footnotes, the authors' names are inverted. This looks bizarre to me, and I can think of no reason for it.

-- Not even convention (which for example dictates that "Chicago:" should be prefixed to "University of Chicago Press", thereby distinguishing that august institution from any of the other, upstart University of Chicago Presses, which in reality don't exist). --

The project page doesn't give any reason for the inversion of names in footnotes. What is the reason for it? -- Hoary (talk) 23:31, 11 August 2010 (UTC)

There is no preferred style, thus you need to work out on the article talk page what style is going to be used in the article. Styles such as Wikipedia Citation Styles 1 and 2, Harvard, APA style and the Vancouver system use last, first. ---— Gadget850 (Ed) ^talk 23:45, 11 August 2010 (UTC)

Some styles, especially those used outside Wikipedia, just give the name of the author and the year in parenthesis near the statement that the citation applies to. The full information about the authors is in an alphabetical list of references, which is sorted by the author's last name. This is known as parenthetical referencing. Since some popular templates can be used for parenthetical referencing, they put the last name first whether the article uses parenthetical referencing or not. Jc3s5h (talk) 01:04, 12 August 2010 (UTC)

The practice has arisen from the presentation of bibliographies in alphabetical order on the basis of the author's surname. Conventions vary from journal to journal, field to field, and individual to individual. I think either order is fine, as long as it's consistent within an article. Tony (talk) 09:32, 12 August 2010 (UTC)

Um, I do know the difference between author–date and notes. ¶ "APA" and "Harvard" are author–date, for which purpose you need a list of sources ordered by surname, for which purpose the inversion of (non-Hungarian) western names is helpful. I was unfamiliar with the "Vancouver" system; thank you for the tip. I suppose what I'm looking at is closest to this. But why does it invert names? I see no advantage to doing so here. -- Hoary (talk) 12:23, 13 August 2010 (UTC)

You would have to ask the folks who developed the Vancouver and other styles. I took a quick look at the Vancouver guides online and they use use last, first initial. We do have a series of Vancouver templates— the examples are using first last for some reason (I left a note on one of the talk pages). ---— Gadget850 (Ed) ^talk 13:05, 13 August 2010 (UTC)

Ah, but by contrast for a "Chicago" style, the reader of the Chicago Manual of Style wouldn't have to ask. The writers treat their readers as intelligent and sometimes curious adults, tend to explain the non-obvious, and very often give readers a choice between two options rather than dictating one. But I digress. -- Hoary (talk) 13:49, 13 August 2010 (UTC) rephrased Hoary (talk) 03:54, 14 August 2010 (UTC)

That's one of the main reasons why I think we should consider to at least trim down the available options for referencing to at a handful; allowing just about any standards available out there becomes a way to avoid meaningful consensus discussions and ignore complaints.

Peter ^Isotalo 14:54, 13 August 2010 (UTC)

The University of Chicago Press publishes works on a great variety of subjects and its style guide is adequate for these. I think its style guide is largely adequate for Wikipedia too, although Wikipedia has less of a space constraint than print does and more of a tolerance for non-roman scripts. It won't prescribe what many authors hope it will, but then it doesn't always prescribe what I wish either. -- Hoary (talk) 03:54, 14 August 2010 (UTC)

Citing multiple videos on a website using flash

[12] uses a bunch of videos and is flash-based so linking to each indivisual one through the |link= field isn't possible. Also given the number of videos, referencing which one is important. I'm not sure how to cite the specific videos though without making redundant citations to the same website as while there is a way for webpages to cite multiple pages, there is not one to cite multiple videos in the same page; these aren't seperate pages like is usual for seperate flash page (the |page= field lists Page X so simply using that won't suffice. Using the {{cite video}} template has the same problem because it doesn't have a parameter for such (|title= is for the page title for websites).陣内_Jinnai 19:23, 11 July 2010 (UTC)

Can i get some help with this. I'd really like to add the website to the article, but don't think I can right now due to limitations with {{citeweb}}, {{cite video}} or even the general {{citation}} templates due to (as far as i can tell) lack of citing ability. If anyone can see a way around this, I'd appreciate it.陣内_Jinnai 07:22, 20 August 2010 (UTC)

Another shortcut?

As the page has grown larger, it has made sense to provide {{shortcut}}s to commonly quoted sections. As a strong advocate of List-defined references, I'd appreciate a simple shortcut to that section for use in discussion (perhaps WP:CITEDEF?), but I'm loathe to do that unless others would also find it useful. Any thoughts? --RexxS (talk) 16:58, 13 August 2010 (UTC)

We have a shortcut to WP:LDR. In fact, we have this duplicated at:

• Wikipedia:Citing_sources#List-defined_references
• Help:Footnotes#List-defined_references
• Wikipedia:Manual of Style (footnotes)#List-defined references

---— Gadget850 (Ed) ^talk 04:14, 14 August 2010 (UTC)

Thanks Gadget, I hadn't realised we had the same guidance in three different places! I've added seealso for WP:LDR to the other two sections, as the MOS (footnotes) guidance contains extra advice. --RexxS (talk) 13:31, 14 August 2010 (UTC)

Authormask

The cite template hierarchy now support an optional authormask parameter, that is being used in a number of bibliographies, i.e. Michael E. Mann, Hans von Storch. The effect of this is replacing the name of one author, typically the subject of the biography, with a long dash. I'm honestly baffled why we would ever want to do this. I find it annoying and confusing, especially for people who are not aware of this convention. I see no advantage to this for a modern electronic reference work - it may have had a certain charm at a time when manuscripts were hand-typed and corrected with whiteout, but I see no reason to maintain this style today. It's certainly not a style used in any of the journals or conferences I publish in professionally.

Is there some existing discussion on this? If not, can we get to a consensus about how to handle this issue? --Stephan Schulz (talk) 16:45, 16 August 2010 (UTC)

I think Chicago style uses this, but not APA or most others. It also calls for use of hanging indent. ---— Gadget850 (Ed) ^talk 17:29, 16 August 2010 (UTC)

The template documentation says only for Bibliographies and it only works for the 1st author. For that case, it's brevity should help — which is why I did it. Easy to remove s/|authormask=1 //gc RDBrown (talk) 22:18, 16 August 2010 (UTC)

Yes, I found the template definition. And I agree that it's easy to undo. However, I think before we go adding and removing this (probably at the same time ;-), we should try to find a consensus on whether we want to use or avoid the authormask feature, and possibly get this consensus into the MOS. I guess my opinion is clear - I don't find the brevity argument compelling, as the gain is minimal, and bought at the price of usability. --Stephan Schulz (talk) 09:26, 17 August 2010 (UTC)

The Mann and von Storch biographies were experiments with authormask while I was doing cite journal conversions. For them, it doesn't really matter, the bibliographies are fairly short. If you look at say Václav E. Beneš there are well formatted long bibliographies that aren't using citation templates, but are mostly omitting the name. If converted to citation templates without authormask, how can you omit the first author and still permit the DOIBot to do it's work cleanly? For longer bibliographies and names, brevity IMO helps once the convention is understood, since the reader doesn't have to process the redundant, repeated, possibly varying information. I'd welcome a good rule of thumb for longer if authormask is contentious otherwise. RDBrown (talk) 13:06, 20 August 2010 (UTC)

RFC on citation style

There is an open RFC here which may be of interest.  pablo 11:56, 17 August 2010 (UTC)

Following on from that RfC, there is renewed discussion at WP:Centralized discussion/Citation discussion#Redux on the possibility of replacing citation templates with enhanced tags. All input would be welcome. --RexxS (talk) 17:29, 22 August 2010 (UTC)

RfC on in-text attribution

Fresh eyes would be appreciated on an RfC about whether, in using in-text attribution for sources on the Historicity of Jesus, we should include whether that source is an ordained minister or similar. See Talk:Historicity_of_Jesus#RfC_on_in-text_attribution. Many thanks, SlimVirgin ^{talk|contribs} 17:05, 26 August 2010 (UTC)

Citations in infoboxes

Is it good or bad practice to give citations in infoboxes, for facts already cited in the body of an article? Can this project page make clear the answer? Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 19:07, 10 August 2010 (UTC)

IMO, there is nothing wrong and everything right with using citations repeatedly for important information. I would not only like to see it, but would like it stated that lede material, as well as infobox, should have the same standard for citations, even when summarizing a cited longer version further down. An editor (or reader) should not be forced to read an entire article to see if the facts are really substantiated. It is not a magic trick! Credibility should be transparent, not hidden! Thanks for raising the issue! Student7 (talk) 02:19, 14 August 2010 (UTC)

I disagree. Those few people who want citations can find them in two seconds. The lede usually summarizes much more detailed information later in the article, and the longer section may have numerous citations which do not lend themselves to combining into one omnibus footnote, that is of no use to anyone. The footnotes, of course, are just as "hidden" in the lede as in the text--they're at the bottom. Book publishers often say theyr readers do NOT want footnotes on the same page as they're reading, they want them in the back.Rjensen (talk) 02:30, 14 August 2010 (UTC)

See the lead paragraph at WP:LEDE. Wtmitchell (talk) (earlier Boracay Bill) 02:48, 14 August 2010 (UTC)

Right. "...carefully sourced..." I don't think this disagrees with Rjensen's comment about unobtrusive footnotes being "in back." They are as far down as we can get them in the article per WP:LAYOUT. I agree that some people, even many people, are not interested. But unusual claims are more easily accepted, even if not read, if they are footnoted. I check them, but casual readers may not, and may not be interested, They do contribute to the overall credibility of the article, whether read by all readers or not. (The alternative is "top of the head" stuff that we have all seen everywhere, body, lede and infobox). Student7 (talk) 13:03, 16 August 2010 (UTC)

Steering us away from discussion of the lede and back to Infoboxes..? Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 13:26, 16 August 2010 (UTC)

I think that only information that is not referenced in the main text needs a separate citation in the infobox. — Cheers, JackLee ^–talk– 15:25, 16 August 2010 (UTC)

I second that. It seems like a consistent and reasonable solution. Having 20 or so footnotes in an infobox strikes me as being rather excessive and creates a needlessly messy typography. They should be treated the same as leads, ie primarily summaries of facts central to the article topic.

Peter ^Isotalo 15:54, 16 August 2010 (UTC)

While I agree with the "untidy" suggestion, it bothers me that too often I see figures in the infoboxes that disagree (or are hard to find) in the lede or body. Specifically, supposed census figures and numbers of students. If they disagree, I can untangle them if each is referenced. Harder to do, or I just forget it, when one differs substantially from the other and neither is referenced. And what's a little footnote, anyway? Once or twice in the infobox. Not that bad! Now, I can briskly rm infobox figures if they "don't seem right" and they are unreferenced. Not that easy to decide if they are not cited. Student7 (talk) 21:28, 18 August 2010 (UTC)

I don't see how adding notes fixes any of this. The only thing that fixes the problem are editors, and they only need a single reference to a source to fix synchronize several facts or figures.

Peter ^Isotalo 22:33, 18 August 2010 (UTC)

Cherrypicking sources in order to to synchronize presented facts or figures sounds to me like violation of WP:DUE. Wtmitchell (talk) (earlier Boracay Bill) 22:09, 19 August 2010 (UTC)

You're reading me too literally. The point is that there needs to be only one instance where you support a fact or figure with one or several references (in the article text) for a (NPOV-acceptable) synchronization to be possible. Placing reference(s) in more than one location to support the exact same fact or figure achieves little besides adding superfluous notes.

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

Here's an example of a info box change. In this case, it is not really that easy to verify that the box has been changed correctly. I just assumed (maybe wrongly) that the change was unexplained and most likely that the change was due to a wandering vandal which often happens. I don't know off the top of my head how many people were involved. This type of infobox typical of war/battle infoboxes. It's in there somewhere! If the user had left a cite, I might check it and change the text! Sometimes newbies know what they are talking about, but just aren't familiar enough with procedures to realize that it is in the text as well. Student7 (talk) 18:57, 21 August 2010 (UTC)

I agree that it is a problem, as we have a problem with it in serialized manga and anime whenever the next chapter/episode comes out the infobox is invariably updated without sourcing. However requiring the cite to prove it is likely to lead to copyright violations because the only RSes most English readers have access to are those scanned (and often unofficially translated) manga chapters and unofficially subbed episodes. Yet policing this to stop those updates will just cause massive headaches as once you undo one IP edit, another comes along and updates it so you'd have to have all ongoing manga and anime articles on permanent flagged revisions (goes against Wikipedia guideling policy for open editing for that large of a flagged revision) or you just got to let it go and realize eventually things will be sourced and in the meantime its not really that controversial 99.9% of the time.陣内_Jinnai 19:54, 4 September 2010 (UTC)

Template:No footnotes

There is a discussion over at Template:No footnotes over the issue of whether sources listed in "Further reading" or "External links" are just as valid a place to list sources as the "References" section. Or put another way there is no support in the WP:V for general references.

There is a proposal to alter the Template:No footnotes from

• This article includes a list of references, related reading or external links, but its sources remain unclear because it lacks inline citations.

to

• This article includes a list of references,related reading or external links, but its sources remain unclear because it lacks inline citations.

But as there is disagreement over where sources for an article may be placed there is as yet no clear consensus over whether the wording should be changed.

The conversation is split over two talk pages and a user's talk page

• Wikipedia talk:WikiProject Unreferenced articles#If the article has even one reference or external link -- This wording in project has since been modified
• Template talk:No footnotes#Clarification

It would help if some more people would get involved in the discussion in the section Template talk:No footnotes#Clarification. -- PBS (talk) 09:40, 28 August 2010 (UTC)

• If general references are allowed the existence of this template is questionable. It can be used as a de facto badge of shame. The usage of such template and the circumstances under which it is appropriate and under which it should be removed should be made clear. What does such a template wish to accomplish and in what circumstances does it work against its intent to improve the article? Such circumstances do exist. Lambanog (talk) 09:56, 2 October 2010 (UTC)

Page numbers

"When citing lengthy sources, you should normally identify which part of a source that you quote, paraphrase or cite."

This is a new sentence. I am not sure that it makes complete sense. Is cite the correct word here? -- PBS (talk) 12:30, 28 August 2010 (UTC)

Per wikt:cite: To list the source(s) from which you used information, words or literary or verbal context. ---— Gadget850 (Ed) ^talk 17:17, 28 August 2010 (UTC)

Well, PBS, do you think that it's useful to "identify" which "part" of a source you quote, paraphrase, or cite, if the source only has one part? For example, how would you identify the "part" of a two-paragraph news story you cite?

Adding "first of two paragraphs" to the citation seems more than a wee bit silly to me. We need page numbers (and similar details) only when the source is long enough to make them useful. WhatamIdoing (talk) 20:09, 28 August 2010 (UTC)

It was this construct I was asking about: "When citing lengthy sources, you should normally identify which part of a source you ... cite" -- PBS (talk) 02:18, 30 August 2010 (UTC)

Then what you have characterized as "new" is not new, since it was added back in December 2009. I had incorrectly assumed that your concern was with the only substantive change, i.e., that it's silly to provide page numbers for very short documents. WhatamIdoing (talk) 01:01, 1 September 2010 (UTC)

It may well be that it's silly to provide page numbers for very short documents, but the very first clause of the paragraph takes care of that "When citing lengthy sources,..." so there is no need explicitly to mention vert short short documents as this paragraph only covers "lengthy sources" (and there is more than enough wriggle room right there for a dispute without adding another sentence for people to squabble over).

I would still like some clarity on the first sentence no matter how old it is! -- PBS (talk) 13:33, 3 September 2010 (UTC)

Read the original source material yourself

I know where this has come from "It is often better to read the original source material yourself, in which case you can simply cite the original source."

Yes when one finds an unreliable source that for example quotes a secondary reliable source then yes one ought to read the original source to verify that what the unreliable source is true.

However there is a whole different usage where it is not better to read or cite the original material yourself. That is where a reliable secondary source, cites a primary source, or some other source that has not been published in a reliable secondary source, or some source that a non qualified expert could not be expected to understand and it needs the expert publishing in a secondary source.

So I think the sentence either needs removing or qualifying. -- PBS (talk) 02:43, 30 August 2010 (UTC)

Yes, and no. That is, yes, secondary sources are lovely, and usually shouldn't be discarded in favor of primary sources of equal quality. But, no, nobody should blindly trust a mass media story on a technical subject, even if Wikipedia (for certain purposes) calls a lightly edited press release a "secondary source" as soon as it is run in the newspaper. In that case, editors are better off with a good-quality primary source (e.g., the medical journal article that the press release is announcing).

Importantly, the "original source" mentioned here is not necessarily a primary source. A textbook or dictionary (for example) does not turn into a primary source merely because it's mentioned in someone's blog.

On another point, I'm tempted to expand the (a) and (b) reasons for citing your actual sources to include "(c) it's dishonest to say that you read X when you really read Y." WhatamIdoing (talk) 01:10, 1 September 2010 (UTC)

A history book such as Juliet Barker,'s Agincourt: The King, the Campaign, the Battle my well cite a primary source, I would not expect most Wikipedia editors to be able to read mediaeval French and work out the date of an event which will be given either as an ide or as a feast day of a year of the kings reign. What day does the year start on the day he was crowned, the 1st of January or in the English case 25 March? Working out such mediaeval dates into the standard Julian calendar is OR unless one is an expert! Therefore I would expect a Wikpedia article to cited Barker's book for a date of an event, noting the citation she gives if appropriate. I would take it as OR if someone when to the primary source and worked out the standard Julian date from it. So I think the sentence either needs removing or qualifying. -- PBS (talk) 13:25, 3 September 2010 (UTC)

Dead links - Linkrot - Restoring source that is no longer available on the web

In the course of trying to improve an article I added links to and downloaded pdf files from various web sources. Now I find that some of these pdfs are no longer available from the websites that previously hosted them and are appearing as dead links in the Wikipedia article I used them for as references. However, I still have copies of the pdfs. Is it possible to upload the copies I have anywhere and link to them or are there problems with that? Lambanog (talk) 14:39, 6 September 2010 (UTC)

I don't think it is acceptable for you to upload the PDFs to some other online repository, as that is likely to amount to unauthorized distribution of copyrighted material (unless the material is in the public domain). See if you can find an archived version of those web pages from The Internet Wayback Machine. In the future, archive web pages that you refer to using WebCite and insert the archive URL into citations using the parameters |archiveurl= and |archivedate=. — Cheers, JackLee ^–talk– 15:40, 6 September 2010 (UTC)

Yes, archive documents and pages as you go with WebCite, but also check http://webcitation.org/query - someone may have already archived those PDFs at their original location. --Lexein (talk) 19:03, 30 September 2010 (UTC)

Too long

It seems to me that this article is too long and detailed to be useful to someone new to the topic.

Is there a short form version available, along the following lines?

---

Example of citation of a journal: (show code and end result)

Example of citation of a book: (show code and end result)

Example of citation of a webpage: (show code and end result)

---

Thank you. Wanderer57 (talk) 04:52, 7 September 2010 (UTC)

There's certainly a lot of repetition of WP:V going on in the article. It might be a good idea to simply dump everything about the "why" and "when" of sourcing and stick only to the "how". At least that would be consistent with what the nutshell summary says.

Peter ^Isotalo 12:18, 7 September 2010 (UTC)

Hi, re: citations/verifications: have you included the Larousse? Must. Plus, any 18th and 19th cent. American cookbooks, with all the regionals. Pie is American, however, go to Europe (the Larousse) for some origin sourcing. When I write ad copy concerning pies, I go to the origins. If you are an encyclopedia (though what your reputation for that as yet is still out to jury), and you're asking for feedback on sourcing from the public, your best best bet would be food archives, the public library, and anyone's grandmother or great aunt from any town, u.s.a., the source for regional cooking. Thank you. Sincerely, —Preceding unsigned comment added by 67.174.231.109 (talk) 22:11, 19 September 2010 (UTC)

For a simplified version, try Wikipedia:Referencing for beginners. WhatamIdoing (talk) 02:46, 24 September 2010 (UTC)

Embedded links

I've bolded the existing deprecation of embedded links, to emphasize that they can be used, but are just not recommended at all. Also added 2 additional reasons not to use them. Discuss? --Lexein (talk) 03:37, 22 September 2010 (UTC)

I removed that they can be used in place of footnotes or Harvard refs, because in reality this is never acceptable nowadays. SlimVirgin ^{talk|contribs} 00:27, 24 September 2010 (UTC)

I personally agree, so let's see if it sticks. Guess I'm a gradualist. --Lexein (talk) 02:39, 24 September 2010 (UTC)

I normally agree with gradualism for guidelines, but embedded links haven't been used by regular editors for a long time, so this should be okay. Thanks for getting the ball rolling. SlimVirgin ^{talk|contribs} 02:46, 24 September 2010 (UTC)

So, should this steep for a while, or should broader consensus be sought/affirmed, or is it appropriate to apply now to articles like Project on Government Oversight? That was written back when the guideline was quite mild in its disapproval of embedded links. I'm recusing myself from editing embedded links in POGO, since I touched the guideline. --Lexein (talk) 15:33, 1 October 2010 (UTC)

If we're promoting deprecation of labeled inline external links (e.g., "Program areas include: Biodefense Security and Oversight, Nuclear Power Plant Security, ...", as used so much in the POGO article), IMHO we should clearly deprecate such usage in the guideline itself instead of (as currently) in the Wikipedia:Embedded citations essay.

Having looked at the POGO article with fresh eyes, I'm only a soft supporter on this. It seems to me that enforcing such deprecation there by converting the labeled inline links to footnotes expanded in a References section would not improve such articles.

Also, we'll need a new template something like e.g., {{embedded links}} or {{linkimprove}} for insertion of a cleanup hatnote and to add tagged articles to Category:articles needing cleanup, right? If so, let's pick a template name.Wtmitchell (talk) (earlier Boracay Bill) 21:33, 1 October 2010 (UTC)

(I actually meant deletion/rewrite/RS-paragraph-citation, not conversion: POGO embedded links are just a blizzard of primary sources, and WP isn't a TOC for other websites. Secondary sources, not primaries, get to grow up to be references.)---Lexein (talk) 08:41, 2 October 2010 (UTC)

On WP:Embedded citations, I would go further and recommend that that essay be revised to expressly make clear that the practice of embedding links is a Bad Thing. At present, it's a bit wishy-washy: "Embedded citations that fail to include the full citation part are better than no citation and are easy to implement, but the use of embedded links for inline citations is not particularly recommended as a method of best practice." TJRC (talk) 21:48, 1 October 2010 (UTC)

Agreed, remembering that there is intentional slack for (only) newcomers because of "don't worry about making mistakes, other editors can help you fix things later" philosophy espoused in WP:Welcome to Wikipedia, as well as WP:Your first article, based on "anyone can edit." This sentiment, if echoed here, should be expanded to: "Embedded links are only permitted temporarily, during article development by newcomers, and should be simply converted to inline citations with the use of <ref> and </ref> tags. Other editors will help you, and encourage you, to learn to correctly format citations, point you to tutorials, and point you to the various semiautomatic tools which will convert bare urls within <ref></ref> tags to full inline citations. -Lexein (talk) 08:41, 2 October 2010 (UTC)

Hmmm... In my comment above, I was thinking about Style. However, this project page is a guideline about Content, not a guideline about Style. By deprecating embedded citations in a guideline about Content, we'd effectively be saying that we prefer no citations at all to citations embedded inline. Contrary to what I said above, I don't think I could support that. One of the things I do in my otherwise empty moments is convert inline embedded links into fleshed-out refs&cites; the starting point for that is existing inline embedded link content in a style which can be improved.

If this guideline were to be recast as a guideline about Style, or the Content and Style portions separated, that'd be another matter. Wtmitchell (talk) (earlier Boracay Bill) 01:47, 2 October 2010 (UTC)

Potential conflict with APA – The recommended change in the guideline conflicts with a recommendation of the APA manual 6th edition that when citing an entire website it is appropriate to simply make a citation inline in the manner of (www.example.com) and have no reference entry. Lambanog (talk) 03:06, 2 October 2010 (UTC)

Ah, www.example.com is a nonclickable domain name (no protocol part) of the homepage of a website, as inline prose but not technically a link, and is not disputed here. This discussion, and thank you for clarifying, is about http://www.example.com/dir/mil, [13], and Military spending, each an example of a live deep full URL (with protocol) link, named or not, as in the POGO links. --Lexein (talk) 08:51, 2 October 2010 (UTC)

Thank you for clarifying. The distinction should be addressed. Perhaps I missed it but I don't think it appears in the essay WP:ECITE and it is possible to create the citation (www.example.com) with a link or (http://www.example.com). Then I presume it would be disputable? But the question still remains: "Why not link?" Lambanog (talk) 10:27, 2 October 2010 (UTC)

It's partially stylistic (visual distraction), and partially procedural (let citations be citations). Wikilinks are transparent, but embedded links are distracting with the big pointy non-prose arrow. Due to WP:LINKROT an embedded link cannot suffice as a citation, and further lacks supporting identification information: title, authorship, date, publisher, etc. --Lexein (talk) 11:44, 2 October 2010 (UTC)

I now believe that the essay WP:ECITE WP:Embedded citations should be top-noted as retired and obsolete, and/or heavily amended. I was just about to paste this in:

Embedded links as citations:

1. Are never permitted in WP:Good Articles or WP:Featured Articles, to which every article should aspire.
2. Require manual editing effort to create and maintain association of numbered links to entries in the full citation list: this effort is a burden for multi-page articles, whereas <ref></ref> and <references/> or {{reflist}} perform this references list assembly automatically.
3. Impose a double burden due to WP:LINKROT, since the embedded part and the full citation need to be updated if/when link rot occurs.
4. Should be considered to be temporary, and to be candidates for unavoidable, if eventual, conversion. Newcomers using this method should be thanked, and gently, civilly directed to use <ref></ref>, in the style consistent with that present in the article. Where used in long-standing articles, notice of pending conversion to full inline citations should be placed in the article's Talk page. Once between <ref></ref>, automated tools such as WP:Reflinks can greatly ease the burden of citation completion, requiring only date format touchup to match the article's established style.
5. Wikipedia is first and foremost an online encyclopedia, with aspirations to be as scholarly and rigorous as the best encyclopedias on earth, not first and foremost a generic wiki with links. This places the convenience of one-click access to online sources below the importance of the rigorousness of its sourcing, and the least error-prone and easiest-maintained documentation of those sources, even if that imposes a two-click burden to access a source.

--Lexein (talk) 18:47, 2 October 2010 (UTC)

1. Actually, your assertion is wrong with respect to GA, as the actual GA criteria impose exactly zero requirements of any kind on the formatting style for GAs. If the reviewer can figure out what the source is, then even bare URLs are acceptable. (NB that I'm not encouraging their use, for practical reasons, such as linkrot, but that there is no GA requirement.)
2. ECITE is no more complicated or burdensome than WP:PAREN or WP:LDR.
3. This seems like a trivial complaint, as the same search-and-replace feature used for the one can be used for the other.
4. If ECITE (or anything else) is the established format for an article, then you need consensus to change it.
5. ECITE doesn't recommend links; it uses links like [14]. But all of this is a red herring, as the quality of the source isn't affected by the formatting of the citation. A high-quality source is a high-quality source, even if no URL is provided. WhatamIdoing (talk) 20:40, 2 October 2010 (UTC)

0. Probably would have been better if you had just rewritten the block to suit your taste, but what the heck, at the risk of being WP:LAME:

1. I sorta had to write that to get this narrowed down. Ok, I take your point that embedded citations accompanied by full citations, are ok (even though I think they're a pain in the ass). But plain embedded links without full citations are still out, aren't they? And what's your position on the many, many primary source links in Project on Government Oversight? Can a Wikipedia article, be a convenient directory of embedded deep links to other websites?

2. In embedded citations, aren't the numbered links supposed to correspond with the numbered full references? If a new claim is added in the middle of an article, isn't extra manual care required to maintain numeric sync? My point was that <ref></ref> avoids that extra effort.

3. Search/replace is wonderful, but there are a broad variety of editors with a range of skills and intentions. If a newcomer reads an article, clicks an embedded link and it's dead, that editor will likely only fix that link, not seeing the full reference at the bottom of the page: a search is unlikely to be done. I tend to advocate simpler as better, single copies over multiple, and so on.

4. No red herring intended. So ECITES, if done correctly, are fine, but embedded links in lieu of full citations are still not ok, right?

5. So, what about Project on Government Oversight where all the embedded links are deep links to the same primary source, not 3rd party? Such a blizzard of primary source inline citations would be reverted as spam and undue emphasis, and I would support that. So what about a blizzard of embedded links? --Lexein (talk) 02:06, 3 October 2010 (UTC)

1. My primary point is that GA imposes zero requirements for citation formatting. Even bare URLs -- just [http://example.com], with nothing else at all, is technically acceptable. No matter how long you search through the actual GA criteria, you won't find anything that sounds like "nicely formatted citations in an approved style".
2. If you choose to follow the recommendations at WP:ECITE closely, yes. But editors are free to make up their own, and perhaps someone would make a different choice.
3. Note that I'm not encouraging anybody to do this; I'm only saying that it's not actually prohibited. Ideas can be technically not prohibited without being anything other than bad ideas.
4. Yes, as is any other style the editor chooses. We've been avoiding the "but my academic field says that..." holy war. (For an example, see this list, which I got so tired of repeating that I made a handy copy.)
5. Embedded links, like all reliable sources, need to support actual article content. I'm fond of the example at WP:EL#cite_ref-2 (note that the URLs are different between "yes" and "no").
   IMO the URLs spammed throughout Project on Government Oversight (see top of this section) look a lot more like WP:External links than WP:Reliable sources being used to support article content, and I'd remove them entirely on those grounds -- not merely change their formatting and bury them in the footnotes. The "Reports" section might additionally be a violation of WP:NOTDIR. In short, I think all of those links should be completely removed as WP:REFSPAM, or perhaps as just plain old WP:SPAM, rather than reformatted in the style used in that article. WhatamIdoing (talk) 03:15, 3 October 2010 (UTC)

Excellent overall response. Stepping back a bit has been helpful. I agree about GA, in that method is less important than actual citations. I've come to see the value of the various academic citation systems I've seen, but have happily settled on the simplest, most-automated, template-based formats, of late. Your input about the POGO links is the most helpful here. 1. I agree that the links closely resemble misplaced external links more than references, and should be excluded as WP:REFSPAM or even simple WP:SPAM. 2. The Reports section does seem to collide with WP:NOTDIR - but might merit rewriting as prose, since some of the material is RS supported. Those two main points seem like an action plan. --Lexein (talk) 18:21, 5 October 2010 (UTC)

Update request

please update

1. Michigan: 877-304-38 2. Texas: 845-321-34 3. Notre Dame: 837-291-42 4. Nebraska: 827-341-41 5. Ohio State: 819-308-53 6. Alabama: 813-316-43 7. Penn State 812-351-43 8. Oklahoma: 796-305-53 9. Tennessee: 783-333-55 10. USC: 774-307-54 —Preceding unsigned comment added by 160.69.1.253 (talk) 15:20, 24 September 2010 (UTC)

Footnote 3

The discussion on deprecated citation systems reminds me that Wikipedia:Footnote3 is still in use. It is certainly deprecated for in-text citations, but it is highly used in tables. I can understand such use, as ^{[Note 1]} uses a lot more space than ^[a]. Should this be discussed?

Note that such use in tables would be deprecated if/when cite.php is ever updated to support custom link labels; see Wikipedia:Manual of Style (footnotes)/Cite link labels. ---— Gadget850 (Ed) ^talk 03:16, 25 September 2010 (UTC)

Separate References and Footnotes sections

If you take a look at Role-playing video game, you can see that some of the footnotes link to another item located in the References section. Is there a standard way using templates to create these sorts of footnotes? SharkD  Talk  20:31, 25 September 2010 (UTC)

Template:Harvnb. See also Wikipedia:Citing sources#Shortened footnotes. postdlf (talk) 20:43, 25 September 2010 (UTC)

Should there be a period at the end? Also, what if there are multiple authors, or if the same author is cited with multiple works published in the same year? SharkD  Talk  21:42, 25 September 2010 (UTC)

OK, I see now what to do about multiple authors, but not about multiple publications. SharkD  Talk  21:43, 25 September 2010 (UTC)

Nevermind about the last two questions. SharkD  Talk  21:44, 25 September 2010 (UTC)

Also, what if pages aren't specifically numbered, and only have titles instead (such as web pages)? SharkD  Talk  22:10, 25 September 2010 (UTC)

Are general references discouraged?

I note that there was a recent modification that states general references are unusual. This seems to be discouraging their use. Is this now policy? Flexibility in referencing is important given that the project relies on volunteers to create articles. Creation of new articles should not be discouraged solely on the basis that they aren't cited to GA or FA standards. There are also instances where a general reference makes the most sense. Most encyclopedias I have seen use general references. Guidance on this should be clear: either it is allowed or it is not allowed. Lambanog (talk) 03:20, 28 September 2010 (UTC)

How would you prefer to word it, Lambanog? It currently says: "In under-developed articles editors sometimes add a general reference at the end, without using inline citations, but this is increasingly unusual." We could just remove the last part about it being unusual. SlimVirgin ^{talk|contribs} 03:23, 28 September 2010 (UTC)

I think that would be better. I will change it to that in a couple of hours if there are no more comments. Lambanog (talk) 03:30, 28 September 2010 (UTC)

That's fine by me. SlimVirgin ^{talk|contribs} 04:01, 28 September 2010 (UTC)

SlimVirgin, I know you dislike citation templates, but they are the best way to tell newcomers how to produce citations. For example without the guidance of a template a newcomer may provide the name of a book (do you know how many are called "Invertebrate Zoology"?) with no more identification and no page number(s). A citation builder like User:Mr.Z-man/refToolbar makes the job easier and provides a checklist of the items to be added. I'm not sure which to recommend - Wikipedia:RefToolbar 2.0 may be right soon, but I would go safe with Wikipedia:RefToolbar 1.0 for now. --Philcha (talk) 13:55, 28 September 2010 (UTC)

Is that connected to general references, or a separate point, Phil? SlimVirgin ^{talk|contribs} 09:46, 29 September 2010 (UTC)

Hi, SlimVirgin. This time the crux is the 2nd sentence, saying that a book called e.g. "Invertebrate Zoology" with no more identification and no page number(s) is useless. I admit that I think the way to resolve this is teach newcomers to use the tools to build citations. --Philcha (talk) 12:37, 30 September 2010 (UTC)

I've had recurring problems on articles which require dense, high quality citations (also they require dense, high quality sourcing) where users aren't aware that chapter authors and chapter titles need to be cited for chapters contained in edited works and similar things. Given that you seem to be aware of this Philcha, how well do citation templates handle prompting for works contained in other works?—Preceding unsigned comment added by Fifelfoo (talk • contribs) 22:01, September 28, 2010

The question wasn't directed to me, but I'll contribute an illustration:

*{{Citation|title=Title|chapter=Chapter|work=Work|publisher=Publisher|month=Month|year=Year|origyear=Original publication year|author=Author|editor=Editor|series=series}} yields:

• Author (Year) [Original publication year], Editor (ed.), "Title", Work, series, Publisher {{citation}}: |author= has generic name (help); |chapter= ignored (help); Check date values in: |year= (help); Unknown parameter |month= ignored (help)CS1 maint: year (link)

There are other parameters available, of course. See {{Citation}}. The {{Cite book}}, {{Cite journal}}, etc. templates are similar in functionality, but different in detail; most/all of them use the core template {{Citation/core}} to do the donkey-work. Generally, Citation offers similar functionality to the various {{cite xxx}} templates (e.g., in Citation, the journal parameter is an alternative for the work parameter), with some differences in detail. Wtmitchell (talk) (earlier Boracay Bill) 19:07, 28 September 2010 (UTC)

They're discouraged on anything other than under-developed articles, and problematic regardless. If they don't become in-line citations they'll eventually be removed as the article develops, because it becomes increasingly hard to determine what a reference might verify, if anything. --Ronz (talk) 20:44, 28 September 2010 (UTC)

General references are not usually capable of verifying the material in a Wikipedia article. I think Lambanadog is wrong about this. WP:Verification, as I understand it anyway, means that it is possible for a normal person to choose any sentence in any article, use the citation to find the relevant part of the source, and prove to themselves that Wikipedia is accurately reporting information and analysis from the source.

General references simply make it impossible (in practical terms) to make the connection between sentence and source. You're forcing the reader to "trust us". Unfortunately, a cautious reader simply can't trust us without inline citations.

This is not how citations are normally used in academic writing, or in standard encyclopedias. Wikipedia is not like any other kind of publication. Unlike normal encyclopedias our readers have no reason to believe what we write; we have no authority. The only authority Wikipedia has derives from the authority of its sources. If there is no citation that directly ties together text in the article with text in a reliable source, then the text shouldn't be trusted by the reader and needs to be rewritten with inline citations. General references won't do the job. ---- CharlesGillingham (talk) 09:29, 29 September 2010 (UTC)

I agree, but at the moment I believe it says "In under-developed articles editors sometimes add a general reference at the end, without using inline citations," which is true. It's not a recommendation. SlimVirgin ^{talk|contribs} 09:47, 29 September 2010 (UTC)

Without acceptance of general references it is theoretically possible to take things to an extreme and claim that any article not submitted and cited to at least GA standards can be erased or eviscerated. There are enough self-appointed guardians of quality here that do not contribute content that it is a legitimate concern. The fixation Wikipedia it seems has grown to have with inline citations is idiosyncratic. Other references generally held to be superior do not rely on inline citations for their perceived quality and get by just fine with general references. Wikipedia will always be a second rate reference along those lines. Even an article soaked in references line by line can be held to be suspect as long as the quality of its editors is suspect and they are unaccountable. Of course Wikipedia is a first rate reference in other respects. Too bad there don't seem to be many that appreciate that enough or the reasons that is the case. Lambanog (talk) 12:26, 29 September 2010 (UTC)

Here's what I know:

• General references are no longer accepted for FA.
• They are permissible at GA. (GA actually requires inline citations for only five specific types of material; if you could write an article without including any of those five statements, then you could have a Good article that contains only general references.)
• They are uncommon in articles that are longer than about two paragraphs.
• The most recent version of the MLA stylebook (the most widely used academic style book in language and arts) has dropped inline citations altogether, and recommends relying exclusively on general references. WhatamIdoing (talk) 15:15, 29 September 2010 (UTC)

MLA really doesn't apply here, as pointed out above. --Ronz (talk) 15:40, 29 September 2010 (UTC)

Wikipedia permits editors to choose any style they think is appropriate, or to make up a completely new one. If editors choose to follow MLA, then that's okay. They're not required to choose MLA, but they're not prohibited from choosing it, either. So IMO the point about MLA's style is highly relevant. WhatamIdoing (talk) 15:59, 29 September 2010 (UTC)

Content needs to be verifiable, which requires we use inline citation solutions. The policies of WP:V and WP:IAR trump any style that would make it difficult or impossible to verify article content. --Ronz (talk) 16:06, 29 September 2010 (UTC)

Ronz, the placement of your comment at 16:06, 29 September 2010 (UTC) implies that you think the MLA style makes it "difficult or impossible to verify article content". This is not the case. This style has been used by scholars for decades to facilitate the verification of citations.

WhatamIdoing, I don't own the MLA style manual, but your statement that it no longer calls for inline citation seems to be contradicted by this web site and does not agree with the Wikipedia article MLA Style Manual. I suggest you cite the exact edition in which this change appears and make appropriate changes to MLA Style Manual. You might want to discuss it here to make sure it isn't just a difference in meaning between Wikipedia's use of inline v. general, compared to the Modern Language Association. Jc3s5h (talk) 16:35, 29 September 2010 (UTC)

Inline citations are required by the sourcing policy for anything challenged or likely to be challenged, and for quotations. That covers just about anything. It's true that undeveloped articles may use general references, but this is not a good thing, and we can't encourage it. There's no point in comparing us to academic texts and the MLA, because our needs are very specific. We have no authority as a publication except in terms of the sources we cite, and so we need text-source integrity. SlimVirgin ^{talk|contribs} 20:09, 29 September 2010 (UTC)

General references are preferred for science articles that go into details, instead of merely giving a general overview. You simply cite some general textbook or article while presenting the theory here on Wikpedia. There is no way you could give inline citations to every nontrivial statement in such a case. Count Iblis (talk) 16:23, 29 September 2010 (UTC)

Yes, that can be a problem (I have actually seen something like, "The human hand usually has four fingers and a thumb."^{[citation needed]}), and a general reference is not actually any less useful than spamming a footnote to the same textbook after each and every sentence in an article.

Importantly, the policy is that information must be verifiable, not verified with the least possible effort. If a reliable source has ever been published, then the material is verifiable. WP:CITE is about our convenience -- about making it quick and easy for me to find out whether a relevant RS has previously been located. Uncited material may be entirely verifiable. WhatamIdoing (talk) 16:28, 29 September 2010 (UTC)

I do actually agree with Count Iblis' point, which is why I said "they are usually not capable". If 90% of the material in an article can be found in a ten-page section of a textbook, then a general reference is appropriate. But I think that (1) the general reference should contain page numbers (2) the other 10% it doesn't cover should have footnotes or parenthetical references, and (3) there should be only a very few general references. Otherwise, it is not feasible to verify it.

I stand my ground that it must be feasible to verify an article. If verifying the article requires checking 100 statements against 3000 pages of text in 10 sources, then the article needs more specific citations.

Again, this is Wikipedia, and the reader has no reason to believe what is written here, unless we provide them with a method to verify it. It doesn't matter what MLA does or other publications do. Our citations have to provide WP:verifiability. Theirs don't.

Finally, as with everything in this WP:Guideline, we are only making recommendations for editors to follow. Nobody should delete any articles or remove any work based on a guideline. At best, we should try to fix articles with insufficient sources, not delete them. At the least, we may WP:tag an article, just to let the reader know that the article can't be verified. ---- CharlesGillingham (talk) 19:32, 29 September 2010 (UTC)

Basically, I think I disagree with everything you've said, which is strange, because I think we'd make the same choices in an article.

• IMO the primary purpose of providing citations is to help editors (for example, by making it easy to figure out which existing material was previously [supposedly] verified by someone else), not to increase our readers' confidence. (If our readers see a little blue number at the end of the sentence, and blindly assume that it means the information is in the cited source, much less that the cited source is a good one, then woe is the world! In that case, we might be better off with zero references, so that the reader kept a healthy skepticism alive. But perhaps I'm biased, after removing "citations" to editors' personal experiences from Involuntary commitment recently.)
• Even if an article would benefit from more precise citations, "good idea" and "required" are very different things -- and any experienced editor knows that what we gently "suggest" in pages like this will quickly be turned into "absolute requirement" by overzealous editors. (Example: an editor quick-failed a GA nomination last month... because one image was one pixel wider than MOS:IMAGES's "usually" recommendation.) The recent suggestions here have IMO moved too far away from the tone of "recommendations" or "options", and too close to "requirements".
• It's silly to provide page numbers for a general reference that is short (e.g., articles in academic journals), or when you're citing the entire thing. We use the same standard for general refs as for inline refs: provide them if appropriate, and don't bother if not.
• Material that can't be verified should be deleted as an unverifiable hoax. Material that is verifiable (that is, you are confident that reliable sources indicate that the human hand typically sports four fingers and a thumb, or whatever it is that the sentence says) shouldn't be spammed with tags as badges of shame, even if the entire page lists exactly zero references. The point about the tags is to get people to improve articles, not to "warn the readers". WhatamIdoing (talk) 22:18, 29 September 2010 (UTC)

References play two complementary roles:

• To support specific facts in the article
• To provide the reader with a list of sources to look at to learn more about the topic

Particularly for articles on subjects that are well-covered by textbooks, the second goal can be served particularly well through general references. Unless we add a footnote such as "General references include: ...", the reader has a hard time telling which footnotes actually make good general references. Cf. recursion theory which has a lot of inline references (Harvard style), footnotes (for side comments and online-only references), and a list of general references. A few sections could use additional references, but the overall structure is clear. — Carl (CBM · talk) 02:41, 30 September 2010 (UTC)

I think we should update this guideline to indicate the best situations for using general references. Recursion theory is not my idea of an "underdeveloped article". WhatamIdoing (talk) 20:55, 2 October 2010 (UTC)

Based on this discussion, I've changed that section. I'm sure that it can be improved, and invite other editors to do so. WhatamIdoing (talk) 20:27, 12 October 2010 (UTC)

Combining web references for one author's multiple pages

I am just starting to rewrite the article Tornado (robot), having successfully rewritten the related article Razer (robot) to GA-standard last month. However, I would appreciate your advice on an area of referencing with which I am unfamiliar: the best way to combine references to webpages with different URLs, but which are all from the same source - i.e., one website. For instance, on the aforementioned Tornado article you'll see that every reference so far has the same author, website name and access date - there's a lot of redundancy.

I did preview the page using {{Sfn}} instead. However, the issue is that whilst six of the eight links are accessible from a single web URL (a frame-based page presenting the information as a chaptered online diary - please see [15]), two are not. Instead, these are other pages from the Tornado website not linked from the diary. Because the author of these pages is the same as for the diary, as and there are no dates to distinguish when these pages were published, {{Sfn}} cannot function optimally.

My question is: can I remove the redundancies from the references, whilst still retaining all the necessary distinct URLs? Any and all advice would be richly appreciated. Thank you, CountdownCrispy 12:51, 3 October 2010 (UTC)

I think your current method is best. You are essentially referencing chapters of a a publication. BTW: you should check out {{reflist-talk}}. ---— Gadget850 (Ed) ^talk 12:11, 4 October 2010 (UTC)

I agree that each individual page is like a chapter of a publication. --Philcha (talk) 13:41, 4 October 2010 (UTC)

Thank you both for replying - I'll carry on as I started. Also, thanks for bringing {{reflist-talk}} to my attention - I've altered my original post to work with it. Best wishes, CountdownCrispy 14:38, 4 October 2010 (UTC)

References

Citing sources from ebooks

Do we have a preferred method for citing sources from ebooks? In particular, for books on a Kindle? The Kindle has locations which one can use but they don't correspond to pages in any nice way. JoshuaZ (talk) 21:22, 3 October 2010 (UTC)

Good question. Doing a search for how to cite kindle books apa reveals that this question has been asked before. It appears that locations are dependent on the selected font size, so they are not very useful. ---— Gadget850 (Ed) ^talk 12:04, 4 October 2010 (UTC)

You should probably use |format= to indicate the reader format and |id= for the publisher id; Amazon uses ASIN. For example:

{{cite book |last=Einsten |first=Albert |title=The World as I See It |format=Kindle |year=1933 |publisher=Citadel |id=ASIN B002MCZ52M}}

Einsten, Albert (1933). The World as I See It. Citadel. ASIN B002MCZ52M. {{cite book}}: |format= requires |url= (help)

I don't see any way to properly include page numbers, which are important. ---— Gadget850 (Ed) ^talk 14:54, 5 October 2010 (UTC)

There's no way to corollate the Kindle information with a physical copy so it's impossible to check referenced information unless one has a Kindle as well. I don't know if the page number issue also applies to IPad and Nook. This seems to be a fairly serious issue to me. I don't expect all reference books to exist on the Internet (a la Google Books), but there should at least be a way to verify the information without having to resort to buying a Kindle and then having to buy the e-book being referenced. At least at the present time, I can't even check for the info at a local library (lIke I can if I don't own a physical copy of a book or magazine, etc.). Shearonink (talk) 15:37, 5 October 2010 (UTC)

There are many ways to correlate unnumbered or varying page format works:

Albert Einstein, The World as I see It Citadel: 1933; republished by Kindle [undated]; material cited from "Chapter" at ¶This is the start of the cited paragraph… Fifelfoo (talk) 06:58, 6 October 2010 (UTC)

But there isn't a WP:MOS for it...at least not that I could find. I know e-books are the wave of the future, but I guess I didn't realize that the Kindles/Nooks/IPads didn't come with some type of equivalency program that gave where to find the information in known printed versions. Checking the information by having to search for the "¶This is the start of the cited paragraph" within Chapter X might be the only way, but it still annoys me. I'll have to spend time searching for that exact paragraph in a physical copy. The accepted way of citing references makes it much easier to check information, the reader has a fairly exact GPS to point them on their way. Doing it with just the 'Chapter' & a paragraph-start is more like having to ask for directions after you've gotten on the road. Both will get you there, but one will probably take longer. Shearonink (talk) 13:26, 6 October 2010 (UTC)

It is the only way to indicate where in the text which was read (an ebook) where the material which was read found though. You try comparing a page number citation from the US edition to the US edition, or from Publisher X's hardback to Publisher Y's softback. Same problem, a general location and a piece of text to visually search for. It is more generalisable for people reading a paper book than "Page 15 on the ePub of the text bought through XXX service when viewed on an iPhone 4 with small font size in Times New Roman." Fifelfoo (talk) 13:39, 6 October 2010 (UTC)

I didn't say it was unmanageable, just that I personally don't like it. Trying to flit between the e-book versions (various readers, different fonts, etc.) and physical copies when assessing references is a challenge we all have to face. It would be nice if it had been possible to build an equivalency-program into the e-book readers. Shearonink (talk) 14:06, 6 October 2010 (UTC)

As for Kindle, I read Kindle books on a PC using the free software Amazon provides. You do NOT have to buy the Kindle reader....but you do have to buy the book to read it. Happily Amazon allows text searches for the paper editions of Kindle books. Locations don't work--ouch. The solution may be to cite the opening words of the passage and use SEARCH. Rjensen (talk) 15:28, 6 October 2010 (UTC)

So maybe I should just include in the footnote a direct quote of the relevant material? JoshuaZ (talk) 02:48, 8 October 2010 (UTC)

If you're quoting a short phrase or clause or sentence, why not? "John Johnson, My Life [originally published as London: Pressly Books, 1970], Online: Amazon/Kindle, last retrievable 7 October 2010, Chapter 7 by text search reads, "This is the example quote of relevance." However, if you're quoting paragraphs or multiple sentences "John Johnson, My Life [originally published as London: Pressly Books, 1970], Online: Amazon/Kindle, last retrievable 7 October 2010, Chapter 7 by text search at paragraph beginning, "This is the example quote..." and four following paragraphs. How does that sound to help you? Fifelfoo (talk) 02:54, 8 October 2010 (UTC)

Web source, url unavailable

This url which appear in List of National Treasures of Japan (ancient documents) has become unavailable and is not present on the internet archive or webcite either. How to deal with cases like this? Do I have to remove the reference from the article? (I am planning to nominate the article at WP:FLC.) bamse (talk) 22:09, 5 October 2010 (UTC)

Google still has the page cached, if that helps. bamse (talk) 23:40, 5 October 2010 (UTC)

Citations should not be removed from articles when links become dead, any more that citations to books should be removed when the book goes out of print. Jc3s5h (talk) 01:22, 6 October 2010 (UTC)

See Wikipedia:Link rot. ---— Gadget850 (Ed) ^talk 06:48, 6 October 2010 (UTC)

Thank you. bamse (talk) 09:11, 6 October 2010 (UTC)

Conversation about accessdates in the external links section

There is a conversation regarding the use of accessdates in the External links section here Wikipedia talk:External links#Question about accessdates. Your comments would be greatly appreciated. --Kumioko (talk) 14:17, 7 October 2010 (UTC)

Question on general references again

I feel like a broken record on this but the bureaucratic red tape being thrown up as I am writing the article List of Philippine restaurant chains is getting ridiculous and it isn't even a controversial subject. This guideline says that general references are possible. However, I have an admin challenging use of general references on the talk page of the aforementioned article and has removed them citing the first line of this guideline namely: "The policy on sourcing is Verifiability, which requires inline citations for any material challenged or likely to be challenged, and for all quotations. The policy is strictly applied to all material in the mainspace—articles, lists, captions, and sections of articles—without exception; in the event of a contradiction between this page and the policy, the policy takes priority, and this page should be updated to reflect it."

If the above is to be accepted then the implication is that nothing that isn't cited line-by-line is truly safe from being questioned. What makes it really silly is that I had earlier provided two websites that can confirm the items I included in the list and I also provided their official homepages to boot but they were stripped off because one editor expressed the opinion the website was the kind that shouldn't be linked to for the former while another editor believed the official sites were external links masquerading as references. The impression I'm getting is that as long as there is an editor who is willing to question something—anything—the only way to escape some sort of complaint is to cite every line in which case the idea of start class articles would appear to be useless. Can people take a look? There are a series of diffs on the talk page which show the various stages of the article with different ways of referencing used. Lambanog (talk) 18:29, 11 October 2010 (UTC)

Anyone who is asking for an inline citation on a list like that should be ready to give a good reason why they are looking for that specific citation. General references do meet the verifiability requirements for general lists (it's not as if "this is a restaurant in the Phillipines" is likely to be controversial). So anyone who is challenging verifiability of individual entries there should have already consulted the general references before doing so, or else they deserve a WP:TROUT for not assuming good faith. Similarly, removing general references because they are not specific enough misses the point that some reference is better than none.

With that being said, I think there is a genuine concern that the topic of the article ("list of restaurants in the Philippines") is unsuitable for Wikipedia. However, the issue is not verifiability. Editors who want feel the article is unsuitable should nominate it for deletion, which will resolve the issue quickly enough. — Carl (CBM · talk) 19:25, 11 October 2010 (UTC)

Hi CBM, the problem is that Lambanog has been asked by another editor and myself to say which, if any, entries the general references support, and so far hasn't. If there was a web link, then obviously I could look myself, however that hasn't been provided either. Personally, I think it's entirely legitimate to require an independent source for all the entries, and disagree with the views you've expressed. PhilKnight (talk) 23:18, 11 October 2010 (UTC)

I don't see how removing the references helps that problem though. Certainly the reference is a propos the topic of the list and might be of interest for someone reading here. That's independent of the question whether it supports any restaurant on the list at all. — Carl (CBM · talk) 00:39, 12 October 2010 (UTC)

General references shouldn't really be used, Lambanog. This guideline notes that they are sometimes used in under-developed articles, but for anything that's growing, including lists, inline citations are expected so that there's text-source-integrity. SlimVirgin ^{talk|contribs} 23:26, 11 October 2010 (UTC)

If a reference would be of interest to a reader, including it is often worthwhile. Also, see recursion theory for a non-underdeveloped article that appropriately uses some general references. — Carl (CBM · talk) 00:39, 12 October 2010 (UTC)

I'm with Carl. General references are permitted in any article except FAs (which SlimVirgin is involved in, and that may be coloring her view of 'normal'). Furthermore, I think it's really silly to see a list that says one,^[1] two,^[1] three, ^[1] four,^[1] five,^[1] six,^[1] seven,^[1] eight,^[1] and so forth, all the way to the end of the page (which would be sixty-two times, in this instance). WhatamIdoing (talk) 23:38, 11 October 2010 (UTC)

I think where we disagree is that I'm unconvinced the general references provide the necessary information. For example this search implies the first entry isn't contained in the general reference. PhilKnight (talk) 00:13, 12 October 2010 (UTC)

The policy says that inline citations are required for anything challenged or likely to be challenged, not only in FAs. The material here is being challenged. There is surely a way to ensure text-source integrity without repeating the footnote multiple times. SlimVirgin ^{talk|contribs} 00:16, 12 October 2010 (UTC)

If that is the case what on earth is this page for: Wikipedia:Version 1.0 Editorial Team/Assessment? Is policy now to delete Stub, Start, C-Class, and List articles? It seems Wikipedia cannot seem to get its message straight. In previous versions of the article, I provided refs of the chains' official websites [16] (something also done I might note in the B-class rated McDonald's, KFC, and Burger King; the GA-class rated Ben's Chili Bowl; and the FA-class rated BAE Systems) as well as an outside website like ClicktheCity.com to check them against and corroborate [17]. But the editing hoard could not be sated—they wanted a book! So I gave a couple of books. Now the hoard seems nonplussed they cannot verify them on the instant (why do you think I provided the websites first?) and cannot even be bothered to check both general references given before entering into another round of questioning. For some reason this search also turns up nothing. But if I give my magic keyboard another whirl: look at the third entry and say "ta da". So are people now going to stop going on about this? Or am I better off contemplating do pigs fly? Lambanog (talk) 11:33, 12 October 2010 (UTC)

I've added the 'The Food and Agriculture Centennial Book: 100 Years of Philippine Agriculture' as a cite verifying 'Max's' to the article. Regarding the other book, if you could let me know if it verifies any of the entries in the article. If you can't provide a web link, then a page number and quote would be perfectly acceptable. PhilKnight (talk) 12:09, 12 October 2010 (UTC)

Lambanog, I can't see what the problem is. If someone is able to add a general reference to the end of the article (Smith, John. Name of Book. Cambridge University Press, 2010), they can surely add it instead as an inline citation (<ref>Smith, John. ''Name of Book''. Cambridge University Press, 2010, p. 100ff.<ref> There are no mixed messages. Inline citations have been required by policy for all articles for several years. SlimVirgin ^{talk|contribs} 20:53, 12 October 2010 (UTC)

SV, we keep telling you that your assertion is not true. What does it take to convince you? Picture an article that contains (1) one non-controversial sentence, (2) one general reference, (3) zero direct quotations, (4) no sign of any challenges. Now please either show me an exact quotation in a current version of any content policy that says such an article absolutely must use an inline citation format -- or please quit spreading this misinformation. WhatamIdoing (talk) 20:57, 12 October 2010 (UTC)

I don't know who "we" refers to, because I only see you telling me that. The only assertion I'm making is that inline citations have been required by the sourcing policy, WP:V—for quotations and anything challenged or likely to be challenged—for several years, which is demonstrable fact. Please read it for yourself. You're not doing people any favors by telling them general references are okay, because as soon as something is challenged, an inline citation is needed. SlimVirgin ^{talk|contribs} 21:06, 12 October 2010 (UTC)

You said, "Inline citations have been required by policy for all articles for several years." -- not "all articles containing direct quotations and material that has been challenged", but "all articles", as in "all articles, without exception". WhatamIdoing (talk) 21:09, 12 October 2010 (UTC)

I don't know what you're trying to achieve here. All articles that contain anything that's ---> likely to be challenged <--- need inline citation for those points. It's just as easy to do that as it is to add a general reference at the end, so I can't see what the fuss is about. SlimVirgin ^{talk|contribs} 21:13, 12 October 2010 (UTC)

Because editors have different opinions about what's likely to be challenged. You apparently think that "The human hand typically has four fingers and one thumb" is likely to be challenged (except as a joke). Most of the community doesn't, and thus most of the community thinks that general references are acceptable in certain situations -- and some of the community even thinks a general reference is preferable to a long, redundant string of ^[1], if one source supports an entire page. WhatamIdoing (talk) 21:19, 12 October 2010 (UTC)

Why are you putting words in my mouth about the human hand?

If you write something and you truly believe it's not likely to be challenged, don't add an inline citation. But if someone does challenge it, you need to add one. The difference between a general and inline cite is just one of formatting and placement, so again I can't see what the fuss is about. SlimVirgin ^{talk|contribs} 21:23, 12 October 2010 (UTC)

So SlimVirgin in your view if I go to List of restaurant chains in the United States or List of newspapers in the United States and demand an inline citation for each one coming from a secondary source–even McDonald's, KFC, Burger King, The New York Times, The Washington Post, and USA Today–you consider that reasonable? Lambanog (talk) 21:35, 12 October 2010 (UTC)

With all our policies there's an unwritten assumption that they'll be applied reasonably, and not in a POINTy way. Requesting inline citations for "Paris is in France" isn't reasonable. Where to draw the line differs from person to person, but over time you come to know when it's faster to add an inline cite than argue about it. SlimVirgin ^{talk|contribs} 21:53, 12 October 2010 (UTC)

I'll make it easier on you: Show me a direct quotation from a policy that prohibits the reference system currently used in 69870 Fizeau (the first article I encountered with Special:Random). That gives you a three-sentence stub to work with. WhatamIdoing (talk) 21:00, 12 October 2010 (UTC)

I do think that asking for a reference for every entry of List of Philippine restaurant chains is quite similar to asking for a reference for every entry of List of newspapers in the United States (where far from every entry is referenced). As SlimVirgin says, the assumption is that people will not make requests like that without some good reason. While it can be faster to simply add references, in other cases it's better to simply point out that the request is unreasonable and point them at WP:TROUT.

Often these general "reference everything" requests are motivated by concerns about notability, rather than verifiability, and this seems to be the case with the list of restaurants in the Philippines. The issue of adding general references is not really related to the issue of whether individual list items have inline citations, since general references and inline citations can coexist perfectly well. — Carl (CBM · talk) 22:46, 12 October 2010 (UTC)

Any method is acceptable

Somehow, this page lost the fundamental point that any system of citations is acceptable, and no method is preferred. That is a core aspect of this page, which has been stable for years, and has become a core editing principle on Wikipedia. Language explaining it needs to remain in the guideline in some form, in case people might be confused that there is some small number of systems from which an author has to choose. — Carl (CBM · talk) 22:56, 12 October 2010 (UTC)

Similarly, the page somehow said that only certain systems of parenthetical referencing "may be" used, but in reality we allow authors to choose a system that works for the article at hand. They are not required to follow any particular style guide. — Carl (CBM · talk) 23:00, 12 October 2010 (UTC)

Your edits look fine to me. SlimVirgin ^{talk|contribs} 23:07, 12 October 2010 (UTC)

Citing a work within a work

A niggling question I've been meaning to ask for ages. When citing a work within a work (e.g. a paper within a book), how do we handle last name/first name?

• Smith, John. "Name of Paper" in Paul Jones (ed.). Name of Book. Routledge, 2010l; or

• Smith, John. "Name of Paper" in Jones, Paul (ed.). Name of Book. Routledge, 2010.

I think the first is correct, and I believe it's MLA style. For example, see their example here: Allende, Isabel. "Toad's Mouth." Trans. Margaret Sayers Peden. A Hammock beneath the Mangoes: Stories from Latin America. Ed. Thomas Colchie. New York: Plume, 1992. 83-88.

I know some style manuals, e.g. Chicago (and perhaps all), differ regarding last name/first name depending on whether it's in a footnote or in the references section, which complicates things further. Does anyone know for sure whether it's first name/last name for editors? SlimVirgin ^{talk|contribs} 18:39, 26 October 2010 (UTC)

There are many versions of correct, and wikipedia is allowed to establish a variety of versions of correct. cite book currently renders the ugly Smith, John (2010). "Name of Paper". In Jones, Paul. Name of Book. Routledge. {{cite book |last1=Smith |first1=John |chapter=Name of Paper |title=Name of Book |publisher=Routledge |year=2010 |editor1-last=Jones |editor1-first=Paul}} Fifelfoo (talk) 22:26, 26 October 2010 (UTC)

Thanks for the reply, FF. SlimVirgin ^{talk|contribs} 17:16, 29 October 2010 (UTC)

Problems with a reference

I'm not sure this is the correct place for this question, but I've got a problem with a reference. http://www.dfb.de/index.php?id=500014&tx_dfbnews_pi1[showUid]=25237&tx_dfbnews_pi1[sword]=Regionalliga reform&tx_dfbnews_pi4[cat]=212 just won't display right, probably because of the []'s. How can this be fixed, any ideas? Calistemon (talk) 09:27, 28 October 2010 (UTC)

Replace the left brackets with "&#91;" and the right brackets with "&#93;". — Cheers, JackLee ^–talk– 10:20, 28 October 2010 (UTC)

Mate, you are a genius! Thanks for that! Calistemon (talk) 12:04, 28 October 2010 (UTC)

I encountered the same problem recently. Generally, if wikitext chokes because of the use of certain symbols, a simple solution is to replace the symbols with ASCII codes. — Cheers, JackLee ^–talk– 14:05, 28 October 2010 (UTC)

Citing to a magazine insert

I recently created an animal species stub based entirely on a fold-out insert in National Geographic. The insert was bound in the magazine, not a loose map or poster like Nat Geo sometimes does, but it lacked page numbers and was perforated for tear-out. The species info is only found on the insert, not repeated within the text of the article that the insert was a part of. The facts I derived from the insert could ultimately be sourced elsewhere, but let's pretend not. For now, I've just cited to the article generally, giving the page range, but is there an established way for giving a pinpoint cite to such an unpaginated supplement? Thanks, postdlf (talk) 16:18, 17 October 2010 (UTC)

...Nothing? Guys? postdlf (talk) 00:25, 2 November 2010 (UTC)

Is it large enough that the lack of page numbers would make it difficult for anyone to find the information if you just cited it as "Insert included with National Geographic, year, issue so-and-so" or "Insert included beween pages x and y in" etc."? If it is unpaginated but still has pages (since you are saying it wasn't a map or poster), you could cite it as "unpaginated page [4]" or just "page [4]" (or whatever page in the insert it is that you are citing), and possibly include a remark in the bibliography to the effect that it is unpaginated. BTW, I find the effect of the citation template confusing; I think the "August 2010" part should go with the journal rather than the author (as it identifies the issue, not just the time of publication) and put in a bibliographic section. That way you can just cite the author's name in the footnote(s), or use "Todhunter (2010)", or Todhunter (2010a), if you would have several works by the same author. (But then, I neither like citation templates nor named refs very much... If anything, citation templates should be limited to the bibliography and banished from the footnotes.) --Hegvald (talk) 01:19, 2 November 2010 (UTC)

Typically, any subordinate title with its format can be listed in quotation marks with the main title in italics. The cite, if an author is noted would read as: Smallwood, Joe. "Bahaman Caracara (insert)." National Geographic magazine, Volume 33, Issue 12, No. 444, December 2010. FWiW Bzuk (talk) 01:25, 2 November 2010 (UTC).

I knew there was someone out there...  ; ) Thanks for your comments. The fold-out insert itself is bound into the magazine, but perforated so it can be torn out, which is in contrast to the typical map or poster insert in Nat Geo that is just folded and loose inside the magazine. The insert does not have any page numbering, nor is it reflected in the page count; the article's page numbering ignores the insert so it isn't simply that page numbers aren't printed on it. The poster does have a title of sorts, so I think I've decided this is the best way to give all the key information: Tumas, Alejandro; Hobbs, Amanda (August 2010), Todhunter, Andrew (ed.), "Deep Dark Secrets", National Geographic, 218 (2): insert {{citation}}: |chapter= ignored (help). I'd like to also give the page range for the article it's stuck in the middle of, but I guess you can't have everything. postdlf (talk) 14:18, 2 November 2010 (UTC)

how to cite if most users need different links for access

This policy problem arises under WP:SAYWHEREYOUGOTIT. When a source is available in several versions or media (hardcopy, online, reprint in a collection, etc.), we're supposed to cite principally the one we used. I'd like to clarify the policy.

I often want to cite a source available in more than one way but which I read in an online database. The database is available to many people, but the link is specific to one institution where I accessed the database. Anyone who's unknown to that institution and trying that link only gets a home page. Putting the URL into an article footnote is annoying to users. But I didn't read the source in another format, and I don't want to when that would likely entail hours for retrieval and for comparing literal texts, e.g., seeking an old newspaper in hardcopy. I try to provide alternative URLs to give users a choice, but some of those are equally problematic. This applies to sources that are unquestionably citable in WP; e.g., they're reliable sources. This isn't a case of free vs. nonfree; functionally, access may be free but a user must qualify, such as by having a public library card. Many of the URLs I've been citing have been parameterized only for the library where I accessed them, so that, even though the database may be available worldwide, users may need local URLs to access them via subscribing institutions and I don't have a permalink or a uniform URL for worldwide access even if all other qualifications are met (when I see such uniform links I generally provide them). Often a DOI or another uniform identifier doesn't exist for a given source; and I found a case where a DOI linked to a database with even less access than the database URL I had used.

I've seen an indication that a single database may be available in different versions to different subscribers. For example, a periodicals database might offer more articles to a subscribing institution that pays more, even though the database has the same name at all institutions subscribing to it. If this is so, a Wikipedia reader trying to verify a source through a database may not be able to because of the institution through which they access the database, and may not be able to tell the difference. In that case, the URL for the particular article is likely very relevant specifically because it identifies the subscriber, although that could change if the URL is tried much later than when it was copied and the subscriber has changed what it's subscribing to, and at any rate that doesn't solve the problem of not having the login.

Complicating this is that at least one apparently-large-circulation database publisher includes a subscriber's password in the URL, albeit probably only part of the password. When I discussed this with one institution's staff the discussion turned bizarre, so I inquired whether they'd mind if I published their password, which turned the conversation even more bizarre. The database publisher confirmed that they include the password and say it's because they're using older technology for compatibility with customers. (I erased the password string I saw and didn't and don't plan to publish it ever and recommended that they stop the practice but what's important here is that passwords may turn up anyway. It's slightly possible that I've unknowingly published similar passwords in the past, since I don't usually amuse myself by parsing long URLs in my head, but I don't know where to find any older ones. I'll settle for not posting any in the future.)

Some databases have what they call an accession number (approximate term) but I don't know enough either to ensure uniqueness even within the database reporting it or to translate it into a URL or into an identifier useful to WP readers. I'm not sure if any of the databases support searching by those accession numbers.

Solutions possible:

• providing several or many links as alternatives;
• providing only one link, and hoping it's not too annoying;
• not linking but stating that access was via a ProQuest, EbscoHost, JStor, or other identified database;
• not linking but stating that access was via a database, without identifying it;
• providing URLs but only in a comment element that does not display to readers, and is visible only to editors;
• not citing a source if it would need one of these solutions; or
• not citing a source if it would need one of these solutions and not putting its supported content into an article, even though an editor read the source at the URL and drafted content that the source would support

Any suggestions or preferences?

Thank you. Nick Levinson (talk) 13:29, 1 November 2010 (UTC) (Corrected grammar: 13:43, 1 November 2010 (UTC))

There is no requirement that souces be available online or that online sources be available to all readers; see WP:PAYWALL. See also {{registration required}} and {{subscription required}}. ---— Gadget850 (Ed) ^talk 14:56, 1 November 2010 (UTC)

The issue I have is that Nick keeps adding links like this, which are useless to anyone not currently logged into a proquest session. When I click on it, it takes me here. The proquest link gives the impression that the source is available, when it's not. I don't mind providing the DOI or pubmed link - those are accessible to everyone, in all cases, as they are publicly accessible. I've also never pushed for any links to be removed because they could only be accessed through proquest, I just didn't think the proquest links were worth including when they won't be useful to 99% of editors. SAYWHEREYOUGOTIT means citing the citation - author, year, title, journal, volume, issue, pages, and if it's citing a source in another citation, that should be mentioned. Nick has provided the citation, this isn't SAYWHEREYOUGOTIT, this is "say which search engine you used to get it" which isn't necessary or useful. The proquest citation is to an original article, not to an article citing another article. WLU (t) (c) Wikipedia's rules:^simple/_complex 23:03, 1 November 2010 (UTC)

Saying where you got it doesn't mean saying where you found a book or article. It just means don't cite a source that you haven't read yourself. SlimVirgin ^{talk|contribs} 23:16, 1 November 2010 (UTC)

Actually citation style manuals recommend it. Use the form "Retrieved from ..." to say which database the article was taken from. The question here is do we link to it or not? Another question that could be asked is do we link to a copy of an available source online provided by a third party when the original publisher seems to be restricting access? Lambanog (talk) 23:32, 1 November 2010 (UTC)

Some do, and others don't. You won't find any such recommendation in the Uniform Requirements, for example. WhatamIdoing (talk) 23:37, 1 November 2010 (UTC)

That's a separate issue, Lambanog, and if you want to start a section about it, feel free. But the section "say where you read it" is about not citing a source unless you've seen it yourself. SlimVirgin ^{talk|contribs} 01:41, 2 November 2010 (UTC)

Didn't we deal with a year ago? Didn't we go through the whole "Yes, I know that you read this bestseller on Googlebooks, but, no, editors don't need to say "the copy scanned into Google Books, which, in some alternate universe, might be noticeably different from the book of the same ISBN that someone else will find at the bookstore or the local library", and after establishing a clear consensus here, clarify that section appropriately?

(Fundamentally, the confusion arises from people confusing the name of the shortcut with the contents of the relevant section.) WhatamIdoing (talk) 23:37, 1 November 2010 (UTC)

Textual differences are not so rare:

• They occur more when scanning depends on OCR (optical character recognition) rather than, say, photography, and those differences occur about as often as typos occur in hand-typed text, or at least did with older OCR scannings, which are probably still being accessed. For instance, "O" when scanned can become "()" and a black speck fallen on a gap can turn "i" into "l". Error types tend to be characteristic of how they're made, such as by OCR.
• Public domain works, especially popular older ones appearing in multiple sites and media, probably differ in many minor ways, due to editorial decisions about whether to update spelling, syntax, and whatnot. I'd guess the Bible and Shakespeare fall into this category. The U.S. Constitution definitely does, especially since there's more than one original written in 1789 and some errors introduced in later copies have apparently been compounded over the centuries.
• Journalism seems often subject to correction online that doesn't apply to print. The N.Y. Times seems to do this, judging from tag lines sometimes appearing online next to articles. Whether third-party databases capture those updates is questionable; that probably depends on their contracts with the news media.
• Paperbacks usually come after hardbound copies and sometimes differ. One author, a professor, told me his paperback had one change from the hardcover, and it likely wasn't obvious to skimmers and nonreaders. Within a given binding, editions differ and printings may.

To say "accessed from ProQuest" or "retrieved from EbscoHost" without a link looks like advertising, although I'm willing given the issues around specifying links that most readers can't use. But that looks outside the policy consensus.

When a link is likely to work for most people, and we read the online copy, I assume linking is expected of us. So, if a book in the public domain is available to anyone online, and we read the online copy, that link should be cited, but if we read it in hardcopy, then the link should not be provided, because it's clutter or spam and we could easily come up with half a dozen links for one source.

Adding {{subscription required}} or {{registration required}} probably wouldn't slow the deleting of the links with these characteristics.

I guess one solution is for me to keep a link offline at home in case anyone challenges my citation, but I probably won't, because of the overhead in keeping them.

But I'm now getting the sense that the say-where-you-read-it policy excludes distinguishing between hardcopy and an online database that is reliably engaged in accurate and stable reproduction, even if errors may occur and even if some databases like Google Books show different partial texts at different dates/times, so that the linking is not wanted. If that's correct, maybe I'll draft a clarification for the policy, although a recent edit to it is a movement in that direction already.

Thanks. Nick Levinson (talk) 04:17, 2 November 2010 (UTC) (Corrected "whatot" to "whatnot" and clarified wording by adding "offline at home": 04:29, 2 November 2010 (UTC))

I would say always to provide a link if source material is online, to make it easy for others to check it. But I'm not entirely clear about the rest of your points. SlimVirgin ^{talk|contribs} 04:30, 2 November 2010 (UTC)

I guess, then, if the link would not make it easy for most readers to access the online text, such as if a subscription is required, then you'd agree with editors who don't want the link presented? That's what seems to be the policy consensus. Nick Levinson (talk) 04:49, 2 November 2010 (UTC)

We don't discriminate against sources that require subscription. Academic journals invariably require subscription of some kind, though lots of editors can access them via university libraries. If there's a link that will take readers directly to the source you're using, by all means add it, and let them decide whether they want to pay for it. SlimVirgin ^{talk|contribs} 06:42, 2 November 2010 (UTC)

But then they can't access it and are stopped at a prior point, such as a library's home page, which is WLU's and other editors' complaints. And often the article's URL varies by subscriber or institution, which means someone in Beijing needs a Brooklyn login for the URL I provided. Uniform identifiers like DOIs don't exist for all sources. It looks like I shouldn't link in such cases, but if I should then what link should I offer? Or should I write that the source is in ProQuest without linking (but that looks like advertising)? Or should I write simply "(database)" without saying which one, as long as I've cited title, author, etc., to distinguish from hardcopy? Nick Levinson (talk) 16:32, 2 November 2010 (UTC)

What style manual are you following? APA, MLA, Chicago, Vancouver, something else? Lambanog (talk) 16:53, 2 November 2010 (UTC)

Nick, can you give an example? SlimVirgin ^{talk|contribs} 21:45, 2 November 2010 (UTC)

stating that source is in, say, ProQuest should be adequate. It is not "advertising" (ProQuest is not a retail product), but it will help readers. Rjensen (talk) 22:56, 2 November 2010 (UTC)

Compare references 4 (Nick versus mine), 9 (Nick versus mine) and 41 (Nick versus mine).

Nick, IMO even though there may be differences between versions, it is rare that they are so substantially different due to these minor variations that the text being verified is affected. This isn't a legal case, unless the differences between editions are substantive and deliberate the minor variations aren't normally an issue for editors or readers. If the differences are substantive and deliberate, that should be captured by precision in the edition information (for books), through ad hoc accommodations, and if especially important - probably discussed in the text. I've never encountered a situation in my four years on wikipedia where it must be noted that Proquest is the source. Perhaps other have. We are always citing the "Platonic source" rather than the specific version digitized; if only the specific, digitized version can be used to verify the text, that's actually a problem since in principle the sources should be verifiable by anyone with access to a reasonable library. The database used to retrieve a source is irrelevant, it is the reference itself that is sourced. As far as I can tell this is the equivalent of "I found this on Web of Science/Google scholar/Pubmed/Yahoo/Ask Jeeves". I see Proquest as a database and archiving service, while we're citing the actual article. Though perhaps it is sufficiently different that it should be noted and the ignorance is mine. In any case, if is should be noted (which I still don't see as necessary), I would think it be noted as "From ProQuest" rather than a url to nowhere. WLU (t) (c) Wikipedia's rules:^simple/_complex 23:11, 2 November 2010 (UTC)

Just looking at the first link, which says "Porno Violence (letter to the editor), in The Wall Street Journal, Apr. 24, 1986 (Eastern ed.) (ISSN 00999660), as accessed in ProQuest Jan. 24, 2010." That's never necessary. It's enough to give the publication, and if possible a link to the article or letter. And lots of editors offer a retrieval date if they've added a link to the newspaper. But saying that you read it via ProQuest isn't needed. I would write "Porno Violence," The Wall Street Journal, April 24, 1986. And I would add the name of the writer if I had it. SlimVirgin ^{talk|contribs} 00:15, 3 November 2010 (UTC)

In case I wasn't clear before, that's been exactly my point in previous discussions, and my exact conclusion as well. I've never seen another citation formatted to include proquest. WLU (t) (c) Wikipedia's rules:^simple/_complex 01:05, 3 November 2010 (UTC)

Some points and then a proposed consensus summary:

A different kind of example is this pair: the link I provided and the resulting equivalent denying access to most users. The access denial doesn't even get as far as the database's front end.

Differences between source versions and relevant to the article should be noted or discussed, at least significant ones, but in most cases I never do the comparison in the first place and I doubt most people do. We don't have the time. The comparison would implicitly occur if editor #1 quotes or paraphrases source version #1 and editor #2 reads source version #2 and says the article misquoted or misrepresented the source when what really happened is that the source in two versions itself differed and no one noticed that, only noticing the consequence in WP. I have no case in which I found such a difference while preparing content for WP, but that's because I read one version and almost never read any other version of the same original. I suspect that's true among most people, including most WP editors.

Under WP:SOURCEACCESS, "[t]he principle of verifiability implies nothing about ease of access to sources", so access to a rare book or a specialized library may be required. Likewise, a particular version may be cited where the evidence is that only one particular version will do, but, in general, no such distinction is evidenced when preparing WP content.

Which style manual doesn't matter. This is for WP. If we link or not, a style manual has something to say, but I don't think they address the concern of this topic.

I don't know if anyone was confused between got, found, or read. In all three cases, reading is just as necessary. I was concerned about how specific was the meaning of where.

To answer an earlier query, "do we link to a copy of an available source online provided by a third party when the original publisher seems to be restricting access?", we do not link to copyright infringement or to sites that facilitate it, but if the third-party source is legitimately providing the copy then even if the original publisher restricts access to the publisher's own copy WP can still link to the legitimate copy elsewhere.

I start with the assumption that I should be thorough as part of being careful and precise, but it seems I take it too far, in the judgment of others. I can certainly provide briefer citations, if that's preferred in WP.

Google Books is ignored here, since it's in a separate topic on this talk page, not yet resolved.

What I'm getting as consensus is, in general:

• not to link unless the link displays the document or an abstract without a login or subscription,
• a uniform identifier such as a DOI should be supplied if available,
• (to return to a point I raised in the beginning) not to link if the URL includes someone's password or part of it, and
• if no link is supplied not to mention that it was sourced via a database.

Is that right as consensus?

Nick Levinson (talk) 04:17, 3 November 2010 (UTC)

I think that's about right. If you cite a newspaper, you cite the author, the name of the article, the name of the newspaper, and the publication date. If the article's online, supply a link that goes directly to it, and some editors like to add the access date. If it's a book, then it's name of author, name of book, name of publisher, year of publication, page number. Plus an optional link to the book or page number and ISBN. And for journal articles doi where applicable (not necessary but a good idea).

There's no need to say which library we got the book from, or which store we bought the newspaper in, whether we're talking about real libraries/stores, or online databases. The only thing that matters is the text, not how we happened to find it. SlimVirgin ^{talk|contribs} 04:26, 3 November 2010 (UTC)

A good rule of thumb would be pick an appropriate citation template and fill it out as much as possible (a bit of common sense applies as some fields are harder to fill in and less relevant than others; I rarely use the issn = field, but it can be useful if a journal changes its name or something and I don't think I've ever included publisher = for a {{cite journal}}). WLU (t) (c) Wikipedia's rules:^simple/_complex 10:54, 3 November 2010 (UTC)

I'd rather write brief citations without the template, now that I know that WP wants to avoid the longer cites and given some problems with the templates.

I agree on not citing a journal's publisher, perhaps the only exception being if two journals have the same name, which is rare.

In a separate talk section, I'm proposing an edit to WP:SAYWHEREYOUGOTIT to cover what to omit.

Thank you. Nick Levinson (talk) 05:36, 5 November 2010 (UTC) (My last edit was to correct a link & this edit is to mark when: 05:55, 5 November 2010 (UTC))

Webcite outage? (?)

This may be sorta "related" to THIS#1 and This#2 (old Talk:-Page entries); but, since those are in an archive, it says to put new posts in the current place (that is, here).

The following may be (sorta) "original research" (reporting on my personal experience, today), -- but that's OK, right? Since this is just a "Talk:" page, not an actual article in the "main" Wikipedia space. (right?).

My apologies if this should have gone on to Talk:WebCite instead of here. (Should I put a link to this, from Talk:WebCite?)

What I saw today

Today I tried to go to http://www.webcitation.org/ -- I was intending to "archive" something. (The web page I was going to "archive" was "John Q. Public" URL); and I got a squawk that seemed to suggest that, either the server was down, or -- (worse?) -- that the domain-name was no longer "set up", as it had been, before. The Google "cache" entry for it, did still contain a valid page, and its "header" stated that it was "a snapshot of the page as it appeared on Nov 6, 2010 10:31:04 GMT." That as-of date is several days "old". (How often does that "webcache" service update their "cache" pages?)

Then, I tried with a different browser, (Mozilla Firefox, instead of Google Chrome) -- a browser that was set up with "Yahoo" instead of "Google" for its default "search" engine. This time, it also gave a squawk -- a similar squawk. ((The webpage "www.webcitation.org" cannot be found / DNS error occurred. Server cannot be found. The link may be broken.))

Is this just another temporary outage?

Anyone know what is going on? I hope this is just temporary, like the June/July 2009 events mentioned at WebCite#Outages.

I like (the idea of) being able to use Webcite for archiving stuff -- both for Wikipedia and for "non-" Wikipedia purposes. I will really miss it, if it no longer accepts new entries. ...and, if OLD entries were to disappear, then that would be even more disappointing -- on the order of, say, if the Wayback Machine were to start "forgetting" stuff. Just my 0.02. --Mike Schwartz (talk) 21:37, 9 November 2010 (UTC)

complex statement with one source and anticipating addition

A problem: Say an editor creates:
Statement A1. Statement A2. Statement A3.<ref>Source for all type-A statements.</ref>

This displays as:
Statement A1. Statement A2. Statement A3.^[1]
1. ^ Source for all type-A statements.

But say we anticipate that later someone will insert something (that's now unknown) into the main text:
Statement A1. Statement A2. Statement B. Statement A3.<ref>Source for all type-A statements.</ref>

This will require:
Statement A1.<ref name="A">Source for all type-A statements.</ref> Statement A2.<ref name="A" /> Statement B.<ref>Source for statement B.</ref> Statement A3.<ref name="A" />

This will yield:
Statement A1.^[1] Statement A2.^[1] Statement B.^[2] Statement A3.^[1]
1. ^ ^{a b c} Source for all type-A statements.
2. ^ Source for statement B.

But the insertion of the later statement is only anticipated, so, when we edit ahead of that time, we edit either for present display simplicity that becomes wrong later or for future growth with present display complexity.

If opting for present display simplicity:
Statement A1. Statement A2. Statement A3.<ref>Source meant for all type-A statements but that's not obvious for earlier ones.</ref>

Adding Statement B will break the citation for some of the rest, with this yield:
Statement A1. Statement A2. Statement B.^[1] Statement A3.^[2]
1. ^ Source for statement B.
2. ^ Source meant for all type-A statements but that's not obvious for earlier ones.

That looks like there are statements without sources, when all are fully sourced.

If opting for future growth with present display complexity:
Statement A1.<ref name="A">Source for all type-A statements.</ref> Statement A2.<ref name="A" /> Statement A3.<ref name="A" />

This will yield:
Statement A1.^[1] Statement A2.^[1] Statement A3.^[1]
1. ^ ^{a b c} Source for all type-A statements.

That looks like too many footnotes, inviting editing downward.

Both choices are problematic for readers. Perhaps we need a new provision in the guideline. Nick Levinson (talk) 12:45, 10 November 2010 (UTC) (Corrected format by adding line breaks: 12:55, 10 November 2010 (UTC)) (Clarified wording in one sentence: 13:07, 10 November 2010 (UTC))

I don't see how you can remove an form of ambiguity in a reasonable manner. A particular footnote may refer to part of a sentence, a sentence or a paragraph. Usually you can use the exact placing and spacing to indicate which of the 3 cases is meant. Also a small degree of ambiguity is something we can leave with, in doubt you simply need to check the referenced source to see what they verify. As a general rule other editors should avoid moving references around that they have not checked or read themselves.--Kmhkmh (talk) 13:56, 10 November 2010 (UTC)

P.S. Another thing that might be useful to consider here, is to avoid "overreferencing". It has become rather common in many WP articles to use inline citations for every single single fact or line, no matter whether the line is actually likely to be disputed or contains or more uncommon or unexpected knowledge. In many cases such extensive inline citations are not really required. In other words one way of avoiding possible ambiguities is to refrain from excessive inline citations, when there's no real need for them.--Kmhkmh (talk) 14:07, 10 November 2010 (UTC)

Overreferencing is off the point. Obviously, Nick intends us to assume that Statements A1, A2, A3 and B are likely to be challenged. ---- CharlesGillingham (talk) 18:31, 10 November 2010 (UTC)

Obviously? However even in that case one could simply use inline citations for the for the paragraph instead of for each individual statement. Probably not optimal but it still works.--Kmhkmh (talk) 18:45, 10 November 2010 (UTC)

I would argue "present simplicity" is best. Two reasons: (1) B may never appear. (2) I think it is the responsibility of the editor adding "B" to check that the entire article is still verifiable. (3) In real life this will probably not happen, so we wind up with A1 and A2 unsourced. Eventually someone will flag these as unsourced and/or eventually an editor who has "source A" will realize that they are covered by A and fix the footnotes. This third point is an example of the fact that WP:Wikipedia is a work in progress.

I am glad that we are starting to think and talk about WP:INTEGRITY seriously. ---- CharlesGillingham (talk) 18:31, 10 November 2010 (UTC)

When writing material with multiple points from multiple sources, while allowing for future growth, citation bundling is a good solution. For example:

In his lifetime Wittgenstein published just one book review, one article, a children's dictionary, and the 75-page Tractatus Logico-Philosophicus (1921)—25,000 words of philosophical writing published when he was alive, and three million unpublished. Professional philosophers have ranked his posthumously published Philosophical Investigations (1953) as the most important book of 20th-century philosophy.

It avoids lots of little blue numbers. It's easy to read, write, and understand. And both the text and the list of sources can be extended without disturbing what's already there, preserving text-source integrity. SlimVirgin ^{talk|contribs} 09:13, 11 November 2010 (UTC)

I am inclined to favor the simplest presentation of the present sources. The expectation that additional statements or sources will be added later suggests the subject is a bit volatile; when additional information becomes available, it might not lead in the anticipated direction and a complete revamp of the paragraph may be needed. Jc3s5h (talk) 16:58, 11 November 2010 (UTC)

RfC on the relationship between the sourcing policies and guidelines

Input would be appreciated at an RfC to ask whether the sourcing guidelines (such as CITE, IRS, MEDRS) should make clear that the core content polices take priority by saying something like: "In the event of inconsistencies between this page and the policies, the policies take priority, and this guideline should be amended accordingly." Please see the RfC here at IRS. Many thanks, SlimVirgin ^{talk|contribs} 23:58, 11 November 2010 (UTC)

In-text attribution

It seems to me that "in-text attribution" has nothing to do with citing sources. It may be there is a more appropriate guideline or policy where it can be placed but it is cluttering up this guideline and should be removed or compressed to one sentence. -- PBS (talk) 20:31, 13 November 2010 (UTC)

People often ask about in relation to CITE: whether citing involves intext attribution, and if not when is appropriate. It makes sense to refer to it here. I'm not sure how you can say it has nothing to do with citing sources. Intext attribution is the citing of a source. SlimVirgin ^{talk|contribs} 20:56, 13 November 2010 (UTC)

While I understand in general it can be argued that an in-text attrition identifies a source (meaning a person) and so is a citation, I do not see in-text as a "citation" as meant and defined in this guideline, and I think that the wording of the sentence "An inline citation should follow the attribution, either after the phrase, sentence, or paragraph in question." makes that clear. -- PBS (talk) 21:11, 13 November 2010 (UTC)

Intext attribution is a way to cite your source, and this guideline is about how to cite sources, so it's appropriate here. In addition a person could write "John Rawls (1971) argues that ..." at which point inline citation and intext attribution become intertwined.SlimVirgin ^{talk|contribs} 21:24, 13 November 2010 (UTC)

Agree with Slim here. See also {{harvtxt}} (the template for in-text attribution) which is definitely something covered by this guideline. ---- CharlesGillingham (talk) 21:30, 13 November 2010 (UTC)

As I understand it from reading the section this is not about Parenthetical referencing which covers the use of {{harvtxt}} because it says "An inline citation should follow the attribution" and we use Parenthetical referencing as a form of in-line citation. If you and I read this section in such a different way, then there is clearly something wrong with the section. If this is being put forward as a hybrid system then it needs moving down below the sections "Footnotes" and "Parenthetical referencing". First I guess we need to decide what it is that the section is meant to be about. -- PBS (talk) 20:21, 14 November 2010 (UTC)

No, you're right. I was just trying to point out that it's very much like a citation. ---- CharlesGillingham (talk) 06:23, 15 November 2010 (UTC)

I think it's appropriate for this page to include that. I think it would be silly to say that editors must write things like "In his famous 1971 book, The Sun is Really Big, John Rawls argues that ... (Rawls, 1971)." The point of inline citations is to let the reader know where the information came from; in this kind of sentence, they already have that information. I do not believe that our readers are so stupid that they need a short citation to repeat the information already in the sentence. WhatamIdoing (talk) 04:32, 15 November 2010 (UTC)

Text-source integrity

Since I last looked here the section "Text-source integrity" seems to have been plonked in front of "How to write [footnotes]" which seems odd because how is someone who needs to know how to write them meant to understand their ordering before it is explained to them how to write them? Also "Text-source integrity" seems to be at best a subsection of the older Bundling citations and at worst a reiteration of the same information. I suggest that the text is moved down to Bundling citations as a subsection and then integrated into Bundling citations. -- PBS (talk) 21:00, 13 November 2010 (UTC)

It's not a subsection of bundling. It advises people to make sure they place citations in a way that makes clear which text the cites support. One of the recent plagiarism problems people have been talking about was caused by that not being done. SlimVirgin ^{talk|contribs} 21:03, 13 November 2010 (UTC)

Without opening old wounds, if that is the intent then why not give a demonstration of placing the citations within a sentence? At the moment the only explicit example given has the advise "Where you are using multiple sources for one sentence, consider bundling citations at the end of the sentence or paragraph". Let us assume for the moment that it is a section worth keeping then I think the advise should be moved down to after "How to format and place inline citations", eg just before "Dealing with unsourced material" -- PBS (talk) 21:21, 13 November 2010 (UTC)

I don't know what the old wounds are. The section does say that they can be placed after the phrase, sentence, or paragraph. SlimVirgin ^{talk|contribs} 21:24, 13 November 2010 (UTC)

old wounds: ref tags and punctuation. -- PBS (talk) 20:28, 14 November 2010 (UTC)

<Slim, I liked the structure you had a few weeks ago, where you placed a number of topics under Background. Although this was a bit of a potpourri, it helped first-time readers to jump straight into the "how-to" section. Could we revive the "background" and place integrity in there? That way integrity would still be prominent but not obstructive. ---- CharlesGillingham (talk) 21:27, 13 November 2010 (UTC)

Just in case it's not clear: the issue here is that the current structure buries the first time reader in details that will only begin to make sense when they have understood the basics. ---- CharlesGillingham (talk) 21:33, 13 November 2010 (UTC)

Okay, I'll play around with it along those lines. SlimVirgin ^{talk|contribs} 21:34, 13 November 2010 (UTC)

I just tried it and it looked even more top-heavy to have integrity in the background. Which bits do you think should be at the top and/or most obvious? SlimVirgin ^{talk|contribs} 21:41, 13 November 2010 (UTC)

1. Numbered list item

Move "In-text attribution" and "Text-source integrity" below or to the end of "How to format and place inline citations". -- PBS (talk) 20:36, 14 November 2010 (UTC)

Keeping in mind that my top priority is that new editors find the information they need without getting confused and giving up, I like this structure:

1. How to format and place inline citations (with the paragraph that used to be under inline citations)
   1. Footnotes
      1. How to write them
      2. Bundling
      3. Shortened footnotes
      4. List-defined references
   2. In text attributions
   3. Parenthetical references
2. When and why to cite sources
3. Text source integrity
4. Dealing with unsourced material
5. Say where you got it

etc...

n Use of terms

I'm thinking of "in text attribution" as type of citation, and it belongs in the section describing types of citations. I'm thinking of the next set of things as philosophy: important to understand and to point out to other editors, but not something you're looking for the first time you open this page. ---- CharlesGillingham (talk) 23:25, 14 November 2010 (UTC)

I think it better to place "In text attributions" after "Parenthetical references" If indeed it is meant to be hybrid method as I think placing "In text attributions" implies that can not be used with parenthetical references. -- PBS (talk) 23:41, 14 November 2010 (UTC)

Nothing in the guideline says that you can't use them together ... you might need to add a sentence to the section "inline citations" that says "in text attributions are used in articles that use footnotes and also in articles that use parenthetical references." This is not awkward at all.

Note that I am just rearranging things here. I would be WP:BOLD with this, except that I want to hear what Slim says. ---- CharlesGillingham (talk) 03:57, 15 November 2010 (UTC)

The logic for moving "In text attributions" after "Parenthetical references" is the same as moving it down in the first place (describe the method before in this case PR, before an application of its useage). -- PBS (talk) 21:16, 16 November 2010 (UTC)

Hi Charles, I wouldn't mind keeping inline citations, in-text attribution, and text-source integrity near each other. I was thinking of adding something to the effect of INCITE + INTEXT + INTEGRITY = three rules of thumb when citing your sources. Haven't thought of how to word it yet, or where to place it. Do you think it's a good idea?

Maybe you could go ahead and make your BOLD edit, and I could try to work around it. SlimVirgin ^{talk|contribs} 21:29, 16 November 2010 (UTC)

Ok, I did it. (Keep in mind that my main interest is in presenting topics in an order that is most useful to the first-time reader. I think it's important to keep in mind that this is a guideline and not a legal document or a computer program. The most important thing is that new editors find it useful, not that its logical structure is airtight or that it covers every caveat and detail every time a subject comes up.)

Anyway, now I hope the first section of the article provides a quick how-to for the new reader. ---- CharlesGillingham (talk) 01:37, 19 November 2010 (UTC)

Slim, I like the "three rules of thumb" idea. Section 2 of the article? Maybe Citation principles? ---- CharlesGillingham (talk) 01:37, 19 November 2010 (UTC)

order of named refs

A problem: Say an editor creates: Supported statement.<ref name="main-source" /><ref name="less-important-source" />

Say this displays as: Supported statement.^[2][1]

Along comes a bot or editor who rearranges the references into numerical order. Then they'll be in less than ideal order, because the less-important source will precede the main source.

I don't have much of a suggestion, except that editors should not count on the order in which references appear. Perhaps this should be addressed in the guideline. Nick Levinson (talk) 12:12, 10 November 2010 (UTC)

Manually assigning references, such as <ref>cite1; cite2</ref> would be superior in this case. Fifelfoo (talk) 12:16, 10 November 2010 (UTC)

but you want to reuse a particular reference several times, so combining is not always a good choice. It also makes it harder to maintain references (replace them by better ones or remove them, since you lose the Identifikation by a unique index/footnote.--Kmhkmh (talk) 12:21, 10 November 2010 (UTC)

I think most readers/editors won't consider the order of the references as an order of priority or importance. Though imho it makes little sense to apply such a scheme, since it will not be transparent to others.--Kmhkmh (talk) 12:18, 10 November 2010 (UTC)

I agree: the order of the refs is just not that important. WhatamIdoing (talk)

Don't use named references. They force you to make considerations such a this one. Use short references in the footnotes (<ref>Smith 2001, p. 49</ref>) and put the source info in a bibliography at the bottom of the page. Combine references in a single note when that seems logical and explain in the footnote if the sources support different aspects of the statement or claim. --Hegvald (talk) 12:50, 10 November 2010 (UTC)

That only works if your sources are lengthy (e.g., books) and you're not citing the same page several times.

A short citation of "Smith 2001, p. 1 (out of 1)" would be kind of silly. WhatamIdoing (talk) 00:36, 12 November 2010 (UTC)

Why would anyone write "p. 1 (out of 1)"? If the source is a single page one can just write "Smith 2001". The bibliography will explain it as referring to 'Smith, Alfred, "A short note on something", Journal of Somethingology, Vol. 76 (2001), p. 1.' And you can cite "Smith 2001" as many times as you need to. --Hegvald (talk) 02:37, 12 November 2010 (UTC)

Why would anyone create a short citation of just "Smith 2001" when they could use ref name to go directly to the full citation? WhatamIdoing (talk) 04:27, 15 November 2010 (UTC)

Because in many cases, one will want to cite several sources in the same footnote, or add explanatory content to the footnote. (See the example by Headbomb below, and the comment by Rjwilmsi below that.) The latter you cannot do if you use named refs consistently, and doing the former by just piling on footnotes after a sentence leaves the curious reader and later potential contributors wondering why you are citing all these sources.^{[1][67][2][12][3][5]} Does each source support a separate part of the preceding statement? In that case, which part? Or is it just a case of some non-discerning author adding any random source they find through Google that appears to support the claim they are making? I don't see any advantage to being able to "go directly to the full citation" as opposed to scrolling down to a bibliography. --Hegvald (talk) 14:57, 20 November 2010 (UTC)

The order of refs is a matter of editorial judgment. It is standard in my field for references to be presented out-of-order, because they are sorted by author at the end of a paper and then referenced by number. So you might see "This is a claim [5,2,7]", which indicates that 5 is the best source, then 2 then 7. However, if someone in another field is used to sorting the inline notes by number, that's fine too. It's a matter of the citation style in an individual article. — Carl (CBM · talk) 00:58, 12 November 2010 (UTC)

I like the [5,2,7] option (as an option, so editors who ignore order are still fine), but to keep the option of arbitrary order how do we stop or discourage the bot or editor that insists on rearranging in various articles? I don't remember which articles I saw it in, so I can't track who's doing it. If it's a bot, then I guess WP knows. How do we find out so we can keep our option? Nick Levinson (talk) 05:49, 12 November 2010 (UTC)

Carl and Nick, I agree with you, and on the pages I watch it is WP:AWB that does this, a good place to start would be to ask the programmers to disable that feature. -- PBS (talk) 20:51, 13 November 2010 (UTC)

I've posted regarding AWB. Thanks for the clue. Nick Levinson (talk) 05:47, 20 November 2010 (UTC)

Sources should be in numerical order. If someone wants to say a source is the best, it should be done in the note itself, like this Using some obscure reference ordering system lost on the reader is not a good practice. Headbomb {talk / contribs / physics / books} 05:56, 20 November 2010 (UTC)

Example note

I don't think you can assume that readers will infer anything about importance/relevance/scope of the references from their ordering, so if one source is more important than the other and you want readers to know about it, then make it clear in the reference text as suggested by Headbomb. Rjwilmsi 09:25, 20 November 2010 (UTC)

But it's completely established practice in some fields to sort juxtaposed references by importance. You can't make a general principle about it, nor can you assume it will be lost on readers.

The more important thing is that if some editor has explicitly gone out of their way to sort the references, they shouldn't be resorted by some quasi-automated process. One difficulty with longer notes (e.g. Headbomb's example) is that they wouldn't fit well in an article that only uses short footnotes.

If the editors of a particular article decide they want to sort the footnotes numerically, good for them. But if editors have decided to sort the notes by importance, those editors shouldn't have to worry about drive-by changes to change the order they have established. Our general principle is that reference style is established on a per-article basis. — Carl (CBM · talk) 13:19, 20 November 2010 (UTC)

The guiding principle of the Manual of Style is "write so you cannot be misunderstood". Out-of-order referencing are completely lost on the reader so the only thing that achieves is giving a look of sloppy editing and the vast majority of editors. Headbomb {talk / contribs / physics / books} 15:03, 20 November 2010 (UTC)

It can't be misunderstood in any critical way - at worst, the reader still has two references with no idea which is more important. But the readers who do know that the order reflects importance will be able to interpret the order correctly. If anyone actually complains that it looks "sloppy", it's no different than if someone complains that Gynaecology is misspelled. It's an opportunity for them to learn something they didn't know before. Given that there is no real issue of confusion, it's best to leave it up to the editors of each individual article. — Carl (CBM · talk) 21:17, 20 November 2010 (UTC)

And then the next time they see a random ^{[2][3][1][14][4]} they'll think it means they are sorted in terms of importance, while they have no meaning. These implicit system only work when they are consistently followed, which they aren't. Headbomb {talk / contribs / physics / books} 21:27, 20 November 2010 (UTC)

Yes, of course I view those as sorted by importance :) No system will work when the editors of an article don't do what they mean. But we don't want to undo intentional work of some editors because of mistakes by others. By analogy, we don't want to replace "Gynecology" with "Gynaecology" in an article that uses US English, just in the articles that use British English. Similarly, an editor has to spend time reading the references to know which ones are more important than others for a particular claim. Just looking at the footnote numbers tells nothing, and in general the number should just be viewed as opaque labels for references. — Carl (CBM · talk) 21:35, 20 November 2010 (UTC)

Google Books

If the discussion about Google Books comes up again, you can now point people to User:Uncle G/On common Google Books mistakes. Uncle G (talk) 21:01, 21 November 2010 (UTC)

I'm not sure how helpful that site is overall. It's by no means nor complete as far as the somewhat vague technical aspects are concerned.--Kmhkmh (talk) 23:41, 22 November 2010 (UTC)

Discussion in The Scientist of citations

• Sprecht (2010-10-20). "Opinion: Mutations of citations - Just like genetic information, citations can accumulate heritable mutations". The Scientist. patsw (talk) 03:45, 23 November 2010 (UTC)

Ha, interesting, especially "The high incidence of wrong citations reflects the fact that the contained information is to a certain extent redundant ..." :) SlimVirgin ^{talk|contribs} 04:30, 23 November 2010 (UTC)

Linking to Google Books pages

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

There seems to be clear consensus that Google Book page links (page links, not general Google Books links) are appropriate so long as preview is available, so I'll add words to that effect to the guideline. SV

---

Should WP:CITE say that Google Books page links are allowed in footnotes, although not required, and that editors should not go around removing them?

Note: this RfC is about whether we're allowed to add URLs that go directly to a specific page of the book, where preview is available. It is not about adding Google Books links in general. SlimVirgin ^{talk|contribs} 18:18, 29 October 2010 (UTC)

Comments

• Support page links being allowed. No one should be required to add them, but in the same vein no one should be going around removing them either. I only recently became aware, when I tried to add them to an article, that some editors object and are systematically removing them. I don't mean links to the main book entry on Google Books, but links to the actual pages being cited. These can be added in a number of ways. For example:

• Rawls, John. A Theory of Justice. Harvard University Press, 1971, p. 200.

• (edit mode) Rawls, John. [http://books.google.ca/books?id=kvpby7HtAe0C&printsec=frontcover&pg=PA200 ''A Theory of Justice'']. Harvard University Press, 1971, p. 200.

• Rawls 1971, p. 200
• Rawls 1971, p. 200.

Or with a citation template:

• Rawls, John (1971). A Theory of Justice. Harvard University Press. p. 200.
• There's also a special template: {{Google books}}

There seem to be four arguments against them: (1) they add clutter to footnotes; (2) they increase load time; (3) they may not be visible in all countries; and (4) they are unstable.

1. The clutter argument seems weak: this link http://books.google.com/books?id=kvpby7HtAe0C&printsec=frontcover&pg=PA200 adds no more clutter than a newspaper link.
2. I can't judge the load time argument, as I'm not a technical person. But I've never heard editors complain about newspaper links being added because of load time.
3. The argument that they may not be visible in all countries ignores that they are visible to millions of our readers. We don't remove from Wikipedia everything that isn't visible in China, and anyway the reach of Google Books is increasing with time.
4. The view that they are unstable may be anecdotal. Perhaps it was true in the past, but stability is likely to increase as agreements with publishers are reached.

I'm aware of a few discussions about this before, which concluded they're allowed, but as it keeps being raised, I'd like to determine consensus. See earlier discussions at Village pump, Nov 2007; FAC talk, May 2009; and FAC talk, Oct 2010. SlimVirgin ^{talk|contribs} 18:21, 29 October 2010 (UTC)

• Support permitting but not requiring them. Where has this been discussed more fully? I'm not really seeing the arguments against. RJC ^Talk_Contribs 20:42, 29 October 2010 (UTC)
• Support allowing but not requiring per nom, and without prejudice to the specific free online book preview service used (although I'm not personally aware of any, but if there are analogous services to GBooks, editors should be free to use them instead). Making citation verification only a click away rather than a library trip away is a very useful thing. --Cybercobra (talk) 20:43, 29 October 2010 (UTC)
• Support SV's 3rd argument (visible to many) is most compelling to me. Sasata (talk) 20:45, 29 October 2010 (UTC)
• Support. Slim's rebuttal of arguments against their inclusion are very convincing, in my mind. And Sasata's comment about saving a library trip is also quite true. Not requiring them (or allowing any other such service that may now or in the future exist) also makes the fundamental aspect true: It's the book that's being cited, not a webpage. The link just makes the book itself more accessible. oknazevad (talk) 20:50, 29 October 2010 (UTC)
• Support them being allowed, totally oppose any effort to make them compulsory. – iridescent 20:55, 29 October 2010 (UTC)
• Allow them but don't make them compulsory per my comments at the lengthy recent discussion at Talk:Featured article candidates on this topic. Johnbod (talk) 21:29, 29 October 2010 (UTC)
• Support but don't make them a requirement of any kind. Quadzilla99 (talk) 21:52, 29 October 2010 (UTC)
• Support them being allowedSwitch to oppose per hamiltonstone and Elcobbala, both of whom raise good points; strong oppose them being required; strong oppose them being imposed on any page where there is not consensus to do so; oppose them being used in addition to links to the main book entry. Nikkimaria (talk) 22:03, 29 October 2010 (UTC)
• Oppose "editors should not go around removing them" verbiage, but support other aspects. Not all Google Book entries have the work available online (e.g. those with no preview or only snippet view). To use the current TFA as an example, the first referenced work has a GB entry, but no viewable content. It would be utterly pointless to add a GB link to this reference, but the purposed wording would preclude removal of the meaningless link if that were to happen. Wording should address the necessity of adding a GB link. I'd like to see something in the spirit of ENGVAR and citation format consistency (i.e. the existing state of affairs should be respected unless there is a compelling reason for the link). There are articles, for example, that reference no online material; adding GB links would be clutter in those examples (a GB link is not equivalent to a newspaper link when the latter isn't there in the first place). Эlcobbola _talk 22:06, 29 October 2010 (UTC)
• Oppose While possibly helpful, this can lead to internal inconsistency for even the same reference work where it may be available for pg 100 but not pg 200. It also puts a lot of weight on an non-standard external source (unlike, say, ISBN number databases). In a more ideal world when you can guaranty every page of every book would be available from that service, this makes complete sense, but not with the patchwork way GBooks works now. --MASEM (t) 22:20, 29 October 2010 (UTC)
• There are two issues here:
  1. Are links to Google Books acceptable (ever, or occasionally)? I strongly oppose anointing Google Books as a preferred source. The fact that I personally (usually) prefer the commercial source Google Books to the non-commercial source Gutenberg Press is no excuse for me inflicting my personal preference on readers. If the Google Books link isn't offering the reader more than the dozens of personalizable options available through the ISBN magic word, then it shouldn't be included (and, yes, should frequently be removed: A link to "snippet view" is inappropriate). If, however, the link leads to something that can't be achieved any other way -- say, a link directly to the specific page being cited, or a book without an ISBN -- then I would keep the link.
  2. Should the community's current views be documented on this page? Oppose. Fundamentally, I don't think that this page needs to mention it one way or the other. WhatamIdoing (talk) 22:34, 29 October 2010 (UTC)
• Support Google Book links properly used can be an invaluable asset for verification by broad editor and reader base, which is essential to the WP quality management. However if another (non commercial) online source of the reference is available, its use is preferable to Google. There are justified caveats against Google books links such as the overly reliance on a commercial company and that the access may vary from country to country. However imho as long as a large portion of our readers and editors can access such links, the benefits (quality improvement, verification, reliability, etc.) outweigh the caveats. As far as the different variation above are concerned I think that the use of templates should be preferred, since this allows a central management of such links and makes it easier to react to potential changes (for instance like Google changes the format of the url).--Kmhkmh (talk) 22:41, 29 October 2010 (UTC)
• Support - linking to a specific page, which confirms the reference supports the text, is helpful to readers and other editors. PhilKnight (talk) 22:50, 29 October 2010 (UTC)
• Oppose linking to a specific page using a long search string such as this which goes to a book without a page view: http://books.google.com/booksid=_t0EAQAAIAAJ&q=A+serious+character:+the+life+of+Ezra+Pound&dq=A+serious+character:+the+life+of+Ezra+Pound&hl=en&ei=Ssm_TMGJFYv9nAe7_f34CQ&sa=X&oi=book_result&ct=result&resnum=1&ved=0CCgQ6AEwAA. If the article uses multiple pages from a text, then add page numbers to footnotes and a convenience link to the Gbook URL in the references. The long search string clutters the edit window, (which in this case moves off the page) makes editing difficult, and moreover it's not impossible to add an error to the string in the course of editing, rendering the URL inoperable. Support single convenience link per source in reference or source sections. Truthkeeper88 (talk) 23:28, 29 October 2010 (UTC) Strike comment. I was the person going around removing and don't consider this an appropriate place to comment regarding the cause of my action. Truthkeeper88 (talk) 00:32, 31 October 2010 (UTC)

• Comment I agree that urls such as that are quite ugly. However, there is an alternative. While I often find my way to material in a Google book with such a search, I then copy the url into the Google Book citation tool which makes a much nicer link, usually managing to get the proper title, author etc, although just as with any tool one shouldn't use it mindlessly.--SPhilbrickT 00:23, 30 October 2010 (UTC)

For example, if you enter the url used about in Slimvirgin's example, it generates a proper cite, even including isbn.

• John Rawls (1999). A theory of justice. Harvard University Press. p. 200. ISBN 9780674000780. Retrieved 30 October 2010.--SPhilbrickT 00:28, 30 October 2010 (UTC)

• Support as proposed. --JN466 00:42, 30 October 2010 (UTC)
• Conditional Oppose You forgot to say "ADDING or removing them" . Plus "go around" is informal, but that's a trivial matter. changed to O per Hamiltonstone via EC. • Ling.Nut (talk) 03:01, 30 October 2010 (UTC)
• Kind of oppose Google Books is not the only organisation / facility through which books can be linked. I think if WP:CITE is going to document this at all, it should be to say "web-based external links to books are allowed in footnotes, but are not required" or (simpler, and per Elcobbola) "web-based external links to books are not required". I also inclined to agree with Elcobbola that a case can be made for removing links to GB if the linked page provides no preview of any sort. Those cases actually seem to me a form of WP:linkspam and should be actively removed, because they effectively take someone to a particular platform yet don't provide any more information than would the full text cite or the ISBN link. The principle should be platform neutrality in respect of the level of information available. A GB link is allowed if it gives access to information not provided through, eg, the ISBN or the cite itself, but otherwise it is linkspam. Does that make sense? hamiltonstone (talk) 03:12, 30 October 2010 (UTC)
  • As the people at WP:External links have to spend a fair bit of time explaining the difference between a citation that supports article content, and an external link (which does not), I would sincerely appreciate it if the words "external links" did not appear in any sentence on this page that applies to proper reference citations or the further reading section. (Try "links" or "URLs" or some other term -- just not "external links", please.) WhatamIdoing (talk) 23:08, 31 October 2010 (UTC)
• Oppose. I don't want any rule saying we can't remove links to things that Google and its partners sell. I don't like the idea that Wikipedia is being used to advertise items for any specific companies. I think we ought to try and link to books that are totally free, like the books on Archive.org. Free sources should always come first, then commercial.--Brianann MacAmhlaidh (talk) 04:12, 30 October 2010 (UTC)
• Oppose we shouldn't be linking to what is basically a platform for promoting books, and which has geographical biases. If I find a Google books link that doesn't work for me in the UK, I can't see why it should remain Jimfbleak - talk to me? 05:35, 30 October 2010 (UTC)
• Support. Don't disallow them. Don't encourage them. Don't discourage them. Do discourage their removal. Wtmitchell (talk) (earlier Boracay Bill) 06:43, 30 October 2010 (UTC)
• Support Wiki has millions of students who would not otherwise know how to get the very valuable information that books.google provides. As for the "commercial" aspect, if publishers can't sell books they will go out of business (as is happening to newspapers) and then where would Wikipedia get its RS? Rjensen (talk) 08:21, 30 October 2010 (UTC)
• Don't regulate, don't advertize, don't ban. One of the reasons of keeping these links is cover your ass tactics against template warriors who aren't happy even with inline footnotes; this crowd needs instantly viewable proof. East of Borschov 11:12, 30 October 2010 (UTC)
• Support They generally allow ready access to what the book actually says. It is an added impulse to keep our stuff in line with what the source actually says. I wouldn't link to a book page if I wanted to indulge in some word twisting or cherry picking etc. But at the same time, I would say that not linking to a page does not mean that I am doing OR.-Civilizededucation^talk 15:51, 30 October 2010 (UTC)
• Support - Google books links are used in actual practice and always will be. Google books is also a piss-poor substitute for actually citing hardcopy, since usually only fractions of relevant content can be accessed by this means. But a link beats no link. Carrite (talk) 16:57, 30 October 2010 (UTC)
• Oppose Not needed and unnecessary promotion of GB. This is a convenience link if it works and includes the required pages; otherwise I see no merit in it. Just because some editors are "going around" doing something there is no consensus for them to do, doesn't mean we need a guideline banning their actions. Colin°^Talk 18:39, 30 October 2010 (UTC)
• Support per the rationale of Wtmitchell: Don't disallow them. Don't encourage them. Don't discourage them. Do discourage their removal. -- Quiddity (talk) 20:24, 30 October 2010 (UTC)
• Support per the rationale of Wtmitchell: Don't disallow them. Don't encourage them. Don't discourage them. Do discourage their removal. --InaMaka (talk) 20:54, 30 October 2010 (UTC)
• Partly neutral I see both sides of the issues concerning favoring one company over others, etc., and don't have a strong view either way. Partly oppose: However I strongly disagree with the suggestion that using the {{Google books}} template alone could be regarded as a proper citation.   Will Beback  talk  21:53, 30 October 2010 (UTC)
• Support. The compromise position SlimVirgin proposes here is eminently sensible. Links to google books are neither required nor discouraged, but wholescale removal of information that can be useful to many readers is not in the spirit of Wikipedia. I would also likely support variants of the proposal which take into account the style of the first principal contributor. Geometry guy 22:22, 30 October 2010 (UTC)
• Support proposals of SlimVirgin and Wtmitchell. — pd_THOR ^{| =/\=} | 00:07, 31 October 2010 (UTC)
• Support per SlimVirgin and Wtmitchell. Jayjg ^(talk) 01:05, 31 October 2010 (UTC)
• Support SlimVirgin. These links should be encouraged, not removed. The more information in a cite the better. We want to make it as easy as possible for readers to verify the text. -- œ^™ 04:15, 31 October 2010 (UTC)
• Oppose unnecessary standards creep. Fifelfoo (talk) 12:43, 31 October 2010 (UTC)
• Support a modified version that gives a generic description of free online republishers, without naming Google specifically. Jc3s5h (talk) 14:08, 31 October 2010 (UTC)
• Other: I certainly agree with Hamiltonstone's position that not all GB links are equal, and we should enforce a baseline of "no link unless there is actually some text there" - a Google Books link without even partial readable content is basically useless as further reading. This may seem a bit tangential to specific page numbers, but we don't currently mention GB at all - being definite about what we are and aren't encouraging is essential. Generalising this to cover all online book services, as Brianann suggests, is sensible - where we're looking at publicly viewable content, GB is on a par with the open books in Hathi, say, or on archive.org, and it's odd to have a rule for one specific case. Shimgray | talk | 15:36, 31 October 2010 (UTC)
• Support keeping links to source material. I see this as an extension of a tendency of certain enforcers to perceive any external link as advertisements. It's getting to be a ridiculous dogmatic position that is contrary to Wikipedia's interests. Why shouldn't Wikipedia link to Google Books? From what I can tell Wikipedia benefits quite a lot from the way Google searches seem to favor Wikipedia articles in Google search results. Wikipedia became what it is because it is open. I see trends at work steering the project towards abandoning principles that have made it successful. This case is but one symptom. Lambanog (talk) 17:18, 31 October 2010 (UTC)
• Oppose. We cannot guarantee that the information will be available. This will mislead people into thinking, "oh boy, I can get that information", when a great deal of the time they can't. Karanacs (talk) 16:46, 1 November 2010 (UTC)
  • comment This is a problem for any form of link. If one buys into this argument one should never link to anything since the link might drop out. JoshuaZ (talk) 18:49, 1 November 2010 (UTC)
    • It's one thing to wager that the link isn't going to change. It's another matter entirely to deliberately link to a website where you know that the majority of the content will not be available to the average reader. Karanacs (talk) 18:54, 1 November 2010 (UTC)
      • Do you need the majority of the content? If the relevant pages are there then what's the issue? Is this any different then people having links to journals which require pay to read more than the abstracts? JoshuaZ (talk) 18:59, 1 November 2010 (UTC)
      • I think there may be a misunderstanding. This RfC isn't about adding general Google Books links, but about adding URLs to specific pages. Those URLs can't be generated (so far as I know) unless preview is in place. So if it's there at all, it will indeed be visible to most of our readers. Who is not able to see page 200 here—Rawls 1971, p. 200? That's helpful to millions of readers, and to other editors who need to check that the material matches the source, but doesn't match it too closely. SlimVirgin ^{talk|contribs} 19:06, 1 November 2010 (UTC)
        • I can't, for one. That link takes me to page 202, and when I try to get to page 200 I get a message saying the page is unavailable. I don't know why that happens, but it's certainly not the only time it's happened for me. There are undoubtedly misunderstandings happening on both sides of the debate. Nikkimaria (talk) 19:18, 1 November 2010 (UTC)
          • There's no reason this shouldn't be available to you, NM. Can you take a snapshot of what you see, showing the URL you used, and email it to me, please (slimvirgin at gmail dot com)? I'd be interested to see what happened in case we can work out how to fix it. SlimVirgin ^{talk|contribs} 19:27, 1 November 2010 (UTC)
            • I don't have a working camera available at the moment. I just clicked on the link you provided above, and the message "You have either reached a page that is unavailable for viewing or have reached your viewing limit for this book". Nikkimaria (talk) 19:32, 1 November 2010 (UTC)
              • You don't need a camera, NM, just the snipping tool. SlimVirgin ^{talk|contribs} 19:37, 1 November 2010 (UTC)
                • The what now? What's "the snipping tool"? Nikkimaria (talk) 19:39, 1 November 2010 (UTC)
                • Actually, never mind. I think I've figured out what the problem was, and can now see page 200. Of course, now I can't see 202...and as we all know, context is important ;-). Nikkimaria (talk) 19:47, 1 November 2010 (UTC)
                  • Funny; with me it was exactly the same thing as happened to Nikkimaria: The first time I couldn't see anything, on trying again it worked, but page 202 is/was? blank now. Buchraeumer (talk) 21:03, 1 November 2010 (UTC)
                    • Google page views are idiosyncratic at best. Sometimes (often!) clearing your browser cache or using another browser will allow you to see a page that is "reached your viewing limit for this book." I've gotten a lot of mileage out of switching between two browsers to view pages that are not viewable, after I've reached my vague page limit on the first one. First Light (talk) 04:35, 3 November 2010 (UTC)
• Oppose. As is the case with ISBNs, in practice this will only lead to pressure on editors to include links, even when they are working with real life printed books. People can very easily copy and paste the author/title or ISBN and go to google book search themselves. Where I live I can't use google books without most of the book becoming "access denied" incredibly soon, even if the book is on preview, which happens seldom enough. Also, sometimes these conditions change: A book I could use at google in January was suddenly at "no preview" in May. Buchraeumer (talk) 17:55, 1 November 2010 (UTC)
• Support. A link to the the Google books page is merely a convenience and does no harm. It would help readers and editors to verify text more quickly in any case where it worked. It cases where it did not succeed (China, I suppose), all the normal methods of verification are still available. ---- CharlesGillingham (talk) 18:45, 1 November 2010 (UTC)
• Support permitting but not requiring, as outlined above.--brewcrewer (yada, yada) 18:55, 1 November 2010 (UTC)
• Conditional support—I think we need to differentiate between two kinds of Google Books links—ones that contain a view of the passages we're sourcing from, and ones that don't. I am strongly in favor of linking to the former, and strongly opposed to linking to the latter. Sometimes these things change and the links should be updated/added/removed accordingly. From my point of view, linking to Google Books pages that don't contain what we're citing is no different from linking to a commercial site selling the book (e.g. Amazon), while linking to pages that do creates much more transparency in verification, because most serious articles make heavy use of book sources. —Ynhockey ^(Talk) 19:17, 1 November 2010 (UTC)
• To the question as put: No. Uncle G (talk) 22:26, 1 November 2010 (UTC)
• HEAVY SUPPORT: all sources that can be verified but use of the source should be acceptable. FWiW Bzuk (talk) 01:28, 2 November 2010 (UTC)
• Support --occono (talk) 16:55, 2 November 2010 (UTC)
• Support Allow Google Book links. They help.Kmarinas86 (Expert Sectioneer of Wikipedia) ^{19+9+14 + karma = 19+9+14 + talk = 86} 19:28, 2 November 2010 (UTC)
• Strongest Support. I think that any URL which yields specific printed page results should be allowed, such as this one from archive.org, a page scanned from the U.S. Army Register from 1946 which I used inside a cite book template at the biography article Caleb V. Haynes. I am a huge fan of Google books, Internet Archive, and any other organization which scans pages and offers them online. They all add to human knowledge, not take away from it. Binksternet (talk) 21:35, 2 November 2010 (UTC)
• Support encouraging them as a courtesy to other editors. Also support strongly discouraging their removal. No, they don't always show up for every user. Yes, they are a wonderful courtesy for editors and readers who would like to confirm the reference or learn more about the subject. First Light (talk) 02:59, 3 November 2010 (UTC)
• Support allowing such links - as an online resource it's a fdact of life that users and editors will benefit from direct links to the quickest way to check factual claims in articles. I haven't heard anyone claiming Google Books is an unreliable source. If it's not there so be it, but if it is, then (as for any media link - newspaper, academic peer reviewed paper, or any other source) cite where it can be found and try to help the reader. FT2 ^{(Talk | email)} 04:00, 3 November 2010 (UTC)
• Support the spirit of the proposal, but obviously we don't need to explicitly state that editors shouldn't go around removing the links. Once we state that the links are allowed and can be helpful then no one will remove them, right? Or if they did they could be referred to this guideline. Yaris678 (talk) 17:55, 3 November 2010 (UTC)
• I agree that we shouldn't remove these links. Anyone systematically removing useful links like this who refuses to stop should be blocked for disruptive editing. The argument that "I can't read it so it should be removed" is deeply parochial - not all links to references need to be immediately accessible to all readers with no charge. Fences&Windows 21:11, 4 November 2010 (UTC)
• Support for citations only. The purpose is not to advertise for books in random links. But as part of a citation, it might be helpful to allow users to follow up with the secondary source. It helps readers understand the verifiability standard on Wikipedia more easily, and makes us seem more reliable. Shooterwalker (talk) 20:14, 22 November 2010 (UTC)
• Support - This allows for easier access to verify citations and in cases where there is either a full view or at least substantial amounts allowed in the preview, it is worthwhile linking for Further reading sections. This isn't advertising. I use it a lot but have never bought a publication based on seeing it at Googlebooks. You find out-of-print materials that aren't for sale commercially.
  ⋙–Berean–Hunter—► ((⊕)) 23:20, 22 November 2010 (UTC)
• Support. Would we censor useful, free, realiable(=verifiable) references (I am talking about pages that can be viewed)?--Wickey-nl (talk) 16:48, 24 November 2010 (UTC)

Inappropiate cites may always removed. Adapt the last part: If appropiate, such footnotes should not be removed.

--Wickey-nl (talk) 10:27, 25 November 2010 (UTC)

Threaded discussion

I don't see any reason to disallow these links with a blanket rule; we allow pretty much any reference style. On the other hand, as usual, if an article has an established style that doesn't include these links, we should generally respect that, just like we respect any other citation style. So new editors on a particular article should generally defer to the established style if there are complaints. — Carl (CBM · talk) 23:48, 29 October 2010 (UTC)

Good point. Changing the established reference style - whether to accommodate these links or due to personal preferences - should not be done. Nikkimaria (talk) 23:52, 29 October 2010 (UTC)

Well, if several people are all working on one article and they all agree to the change, that could be OK. So I don't think it's completely black and white. But if I start going to random articles outside my usual editing area simply to insert links, that's a different story. A complete reference without links is still a perfectly good reference, after all. I can see some merit in the argument that very long google book links can make the source code hard to edit in the same way that citation templates can. — Carl (CBM · talk) 23:59, 29 October 2010 (UTC)

Adding a link to a footnote isn't changing the citation style. For example, changing this:

Deboick, Sophia. "Poland's faith divide", The Guardian, 28 October 2010.

to this:

Deboick, Sophia. "Poland's faith divide", The Guardian, 28 October 2010.

is not a style change and is always helpful. Google Books page links are no different, and they are short when done properly, e.g. http://books.google.ca/books?id=kvpby7HtAe0C&pg=PA200, which leads directly to page 200 of that book, and is a shorter link than the one I posted above for The Guardian. We can't have a situation where a small number of editors are running around removing links for some reason, perhaps because they don't like Google (I'm still not entirely clear what the reasons are). It's a commercial company just like The Guardian or Cambridge University Press, and we don't remove links to their publications. When we supply sourcing information we're doing it to make things easy for the reader, and if a link might do that, that's reason enough to add it. SlimVirgin ^{talk|contribs} 17:36, 30 October 2010 (UTC)

I agree that people should not go around removing links where the established style of the article includes them, but similarly I would not like to see links added "by force" if the established style of the article is to leave them out. If none of the references has links, and that was a conscious choice, then adding the links is certainly a style change. A reference without a link is still a perfectly good reference, and especially for book references we really want people to have the book in their hands anyway. — Carl (CBM · talk) 19:56, 30 October 2010 (UTC)

It would be a major policy change to say that adding newspaper, journal or book page links to a citation was a style change. We don't want people to be forced to have the books in their hands. We want them to be able to check that what we claim the source said really is what the source said. If they can do that by clicking on a link rather than going to a library (which is not an easy thing for everyone to do), that's a good thing. SlimVirgin ^{talk|contribs} 20:22, 30 October 2010 (UTC)

It is a style change to add links - whether that's for better or for worse in all cases is part of what's being debated here. Nikkimaria (talk) 20:52, 30 October 2010 (UTC)

There's a list of citation styles here. Whether a courtesy link is supplied is not connected to what style is used. SlimVirgin ^{talk|contribs} 01:09, 31 October 2010 (UTC)

Slim, could you point out to me which of the citation styles in that list include courtesy links the way you propose? In any case, to add them in is to change the style, whether or not you choose to explicitly connect them to the style in use. Nikkimaria (talk) 01:27, 31 October 2010 (UTC)

I think we need to be more specific regarding what we mean by Google links, since their appearance and usefulness greatly differ. There are

• a) no preview
• b) snippet
• c) restricted preview (page of interested is accessible)
• d) full preview

a) and b) have little or no value for verification or further information to readers or editors and hence can bee seen as purely promotional links for a particular company (aka spam). However c) and d) can be very valuable for verification and further information. I suspect that most supporters have mainly c) and d) in mind, while some opposers are particularly annoyed by a) and b). So it might be useful to state the undesired arbitrary removal of Google book links refers to the c) and d) scenario or more precisely taking the geolocation issue into account refers to links for which c) or d) is true for a large part of our readers/editors.

Another thing is the exact intent of the added text. Is its purpose to generally block people from removing Google links or is its purpose to block people from going on a anti Google rampage in "other people's" articles? This also connected to respecting style or preferences of the main authors or maintainers of an article. Concerning the latter one might even wonder, why we would need a specific "lex google" at all, since this seems to be covered by general guidelines/practices anyway. The problem here is that individual editors might adopt (or apparently have done so in the past) a very particular notion of Google books such as considering them generally as undesired spam. From that perspective they could overrule any "respecting the style of other authors or articles" by claiming a guideline violation such as WP:SPAM. Hence it might be useful, that if the community at large does not share such a specific viewpoint of Google books links, it will get stated explicitly to block such a line of argumentation in the future. Also Google books seems to be a of a "hot issue" over which editors repeatedly clash, therefore it makes sense to deal with it explicitly.

The current guideline contains already a section for convenience links (i. e. online copies of a cited source). I think this section could be expanded a bit to deal with the Google books explicitly to be clear which kind of behaviour or approach towards Google books links is accepted and which is not and how it relates to other guidelines (such as WP:SPAM). We probably should state there as well that other non-commercial convenience links (archive.org, gutenberg, university sites offering free online copies, etc.) are preferable to Google books links (if available).--Kmhkmh (talk) 09:17, 30 October 2010 (UTC)

I use snippet all the time, especially to verify statements and quotations from the book and to get page numbers and short substantive info like dates. It's unusually valuable and there is no alternative short of inter-library loan 95% of the time. Rjensen (talk) 09:24, 30 October 2010 (UTC)

Well imho snippets is borderline scenario, they can indeed be valuabe for verification for a very short and invidual fact (such as a date or a very short quote). However often the snippet is so short that contrary to a page access one lacks contextual information that often is important to fully understand the meaning of the line in the snippet. So snippets have a much higher potential of being misleading than a page access and hence their careless might actually cause quality/verification issues as well rather than helping with the verification.--Kmhkmh (talk) 10:46, 30 October 2010 (UTC)

Personally I'd never link to a snippet/no view, though of course that might be visible as "preview" somewhere else, so without knowing that, they can't be banned while other googler links are allowed. Johnbod (talk) 12:01, 30 October 2010 (UTC)

• A side point, but I strongly disagree with " I think we ought to try and link to books that are totally free, like the books on Archive.org. Free sources should always come first, then commercial." by Brianann MacAmhlaidh above. The vast majority of "totally free" books are public domain = really old = probably not RS, & certainly not to be preferred. Johnbod (talk) 12:01, 30 October 2010 (UTC)

Well i think we can all agree that the decision for source itself (being reputable, reliable, current, etc.) comes first and after the source is picked and we are just considering a convenience link for it, then we can consider a preference a free link over a commercial one.--Kmhkmh (talk) 12:10, 30 October 2010 (UTC)

There was a proposal some time ago to add to WP:V that links to free sites were preferred over commercial ones, if the sources were of equivalent value, and it was rejected. Almost every publication we cite is commercial, including all the academic sources, many of which lie behind paywalls. That has never been a consideration for Wikipedia, and Google Books is free to the reader anyway. SlimVirgin ^{talk|contribs} 17:42, 30 October 2010 (UTC)

I'm not quite sure how to read this. Are you agreeing with me or not? There is a difference between the choosing the source and choosing the convenience link. The debunked proposoal was about preferring free versus commercial regarding the choice of the source and here we are talking about free versus commercial regarding the choice of the convenience link.--Kmhkmh (talk) 18:37, 30 October 2010 (UTC)

I'm not sure whether I'm agreeing with you. :) I think we should choose the best source, and the best convenience link, regardless of any other consideration. The only service that currently allows links to specific pages is Google Books and it's free, so that satisfies all concerns. Whether something is commercial or a charity (both aim to make money in some way, just for different purposes) isn't something I think we should consider. SlimVirgin ^{talk|contribs} 19:40, 30 October 2010 (UTC)

It does matter about "free as in freedom" as opposed to "free as in beer". Google puts their content into hard to access formats in terms of re-use and acquisition. This is by design for their "clients" and due to the demands put on them by publishers, but it does pose some problems for end-users. I think it is an ethically sound argument to suggest that Wikipedia ought to be encouraging sites like Project Gutenberg and Archive.org which have genuinely free content that can be used without having to go through proprietary access tools. If there is a book that can be obtained from one of these other sources and referenced in this manner, it ought to be preferred even at the expense of removing a link to Google Books in that instance. It isn't just how somebody makes money, but how the end-user can access the content and if there might be some sort of licensing agreement or other limitation to the content once you have accessed it. It doesn't cover content under copyright that you can't obtain from sites like this, but for those situations where you can obtain the content from free sites, it ought to be not only preferred but to me preferred at the exclusion of "non-free" sites. Google is certainly one of those "non-free" sites. --Robert Horning (talk) 21:57, 30 October 2010 (UTC)

(reply to Johnbod) I meant that if a particular book is available on GoogleBooks for sale, and also available legitimately for free on another site, we ought to link to the free one. That's all I meant. It's not right that we are about to make a policy forcing articles to flog books for GoogleBooks when we don't need to. It'll be a race to the finish: as soon as one editor links every cite to a book on GoogleBooks, whether it's available for free on other sites or not, we won't be allowed to switch it because of this rule. Obviously recent sources are gonna be more valuable than outdated ones. Common sense man. Whose gonna argue with you about that? What I meant with my comment was that when we've got the choice, the same book one being sold on GoogleBooks, the other totally free, go for the free one. Exactly how Robert Horning put it above me.--Brianann MacAmhlaidh (talk) 04:31, 31 October 2010 (UTC)

Well ok, but it didn't read like that, & your example was probably not well chosen in that case. There is actually quite a problem in many areas of people using old 19th PD sources that are not RS when they could get much better information from more modern ones on Google books or elsewhere. Plus the google books search facility (within a book) is often much easier to use in practice. Johnbod (talk) 05:49, 31 October 2010 (UTC)

• Reply to Hamiltonstone. Just to note that we're talking here about links to Google Book pages, where these exist. Not links to Google Books in general. So this kind of link where it goes directly to the page: Rawls, John. A Theory of Justice. Harvard University Press, 1971, p. 200. SlimVirgin ^{talk|contribs} 20:28, 30 October 2010 (UTC)

• Thanks SV. I am comfortable with that kind of linking, since it clearly infers that the Google book in question has the page viewable. In my experience, most Google Book links are not made to a particular page, but a general link to the book, and it is those ones I would not want to explicitly support, in the cases where the linked book cannot be previewed in GB. Incidentally, I actually think there is a high degree of common ground between most of us commenting here. I liked Robert Horning's observations above too, and agree with Johnbod re watching out for the use of old sources just because they happen to be free - I think most editors who have participated in the discussion here are not the sort of people who would do that, but i have certainly come across it, and one wouldn't want to encourage that laziness :-) hamiltonstone (talk) 23:01, 31 October 2010 (UTC)

In this list of responses, I can't figure out what these support and oppose terms refer to. There's more than one question, and I think a few editors may be using the opposite terms to label the same positions. Does "oppose" mean "I oppose including links" or "I oppose removing links" or "I oppose saying anything about the links here"? WhatamIdoing (talk) 23:27, 30 October 2010 (UTC)

The proposal was clear enough to me, as are most of the support votes: we do not want wholesale removal of google books links. Apart from that, the position that we neither encourage nor discourage the inclusion of such links appears to be the status quo. It would require a separate discussion to change that status quo, and as far as I can judge, most editors (including myself) do not wish to change it. Geometry guy 23:53, 30 October 2010 (UTC)