Wikipedia talk:Citing sources

From Wikipedia, the free encyclopedia

This is the talk page for discussing improvements to the Wikipedia:Citing sources page.
Archives: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21
Shortcut:
WT:CITE
This talk page is automatically archived by MiszaBot II.
Any sections older than 14 days are automatically archived. An archive index is available here.


Contents

[edit] Discursive notes

...earlier part archived...

...later part archived...

Doesn't look like this will happen without a wiki developer to sponsor it. Shame 'cos I thought it was quite a good idea. --SallyScot (talk) 17:08, 9 June 2008 (UTC)

[edit] Citation templates considered harmful

I have grown to loath the citation template system, which I believe is seriously degrading the source text of the wiki to the point of inedibility. I am not alone; when this discussion opened in the tech mailing list, it went on for days. But nothing came of it. I want to make sure people are aware of this problem, and hopefully get some sort of solution.

To be exact, it's not a single problem, its really a combination of two different issues. The first is that the CITE template is, itself, complex. Careful editing is required to ensure that a single-character problem doesn't render the entire article unreadable. But because there are so many "subparts" to the template, this sort of mis-edit is all too easy to make. The second aspect is that references are supposed to be placed in-line. Since the REF tag requires the CITE to be "inside it". The result is that, if one uses CITE + REF, the article source text can be made almost completely uneditable.

How big is this problem? Well a good example came up when I wanted to make a few minor edits to the excellent Cygnus X-1 article. Please take a moment and go read it over. It's fantastic. Now click Edit and try to find anything at all. It's completely indecipherable gobblygook. This is precisely the sort of technicality that can scare away the very editors we need to encourage on the wiki.

The general consensus last time this issue came up was that it was fixable through a change to the way REF works. If we could put the "body" of the CITE at the bottom of the article, where it belongs IMHO, the entire problem would go away. Not only that, but it would make finding, updating and editing references far easier than it is today. In a general sense, what we want to do is place all of the CITEs together, with whitespace of course, at the bottom of the article somewhere, likely inside the References section. Then we would "wrap" these somehow so they would not appear (a hidden DIV almost works). Finally we would place REF tags in the body text as we do today, except they would all be the "short form" that we have today <ref name=thingy/>. This makes everything much easier to read, easier to edit, and IMHO, easier to understand. There were several suggestions on how this could be accomplished.

I really believe that this is a significant technical problem that needs to be addressed. How do I go about trying to get some traction on this issue?

Maury (talk) 18:58, 9 May 2008 (UTC)

You're right, but the problem seems to be more how the editors of Cygnus X-1 used the cite template (with each term on a separate line) than with the template itself. It's entirely possible, and in my opinion preferable, to list the terms one after another, separated by the | symbol. In that case, the CITE template isn't much more of an obstacle to editing than the standard footnote entry -- and it does have the advantage of forcing stylistic consistency in the notes.
BTW, it took me a while to accept the CITE template; it lacks the flexibility historians like, but for most published works (which are what we cite in Wikipedia) it seems an acceptable solution. --SteveMcCluskey (talk) 19:32, 9 May 2008 (UTC)
I see your claim, and raise you a John Titor. As you can see in this example, the "inline CITE" doesn't really change the situation very much. I'm shooting for perfection here, and I'll happily accept half-way. Maury (talk) 19:39, 9 May 2008 (UTC)
IONO how many people came up with the shortened notes style independently, but I am one of them and I cranked out a lot of such-referenced articles last autumn to make it more familiar. I came to use this style (I found out later it was already being used by many SocSci people, but primarily in combination with Harvard cites and I usually don't do SocSci stuff) to be able to combine a clean referencing format with annotations, i.e. explanatory footnotes in the strict sense. (For me it's basically a Popperian consideration: I don't care how cool a referencing system works under ideal conditions; my aim was a way of referencing the most outlandish sources without breaking, that additionally looks tolerably well and codes cleanly.)
Shortened notes can be spiced up with whatever you like. I needed something that did not break easily, something to handle things like the poetry in Senna obtusifolia or the diversity of footnoteable information in Buteo or things like Passerine#cite_note-13, and I categorically needed alphabetically-sorted reference lists and so on. And there is perhaps one source in 150-200 where I cannot use it straightforwardly.
But I would not recommend to treat it as one of "the" referencing systems. Shortened notes are really more something like a building block or a foundation that can be used in combination with most all referencing systems to make the code neater and articles better-referenced and more informative. The advantage of the shortened notes style is that it can be used (and I can only highly recommend using it) as a baseline. It is clean and flexible, and it is code-wise nonintrusive, does not require templates or scripting except the barest necessity.
It provides a lot of benefits for the scholarly-minded editor: it is easy to see, at one glance, how much of the article is based on up-to-date sources, and what if any parts would benefit from digging up a newer, better source perhaps. References can be neatly sorted; it is very easy to see if one particular source is missing (this can be hard if the entire source is footnoted, as many editors cut corners to avoid overlong footnotes e.g. by just linking to an online fulltext). Dysmorodrepanis (talk) 14:01, 11 May 2008 (UTC)
I have found the template to be seriousöly lacking and do not use it. For one thing, it creates between 10 and 50% code overhead, and this is simply bad programming style. But more importantly, as soon as you e.g. cite a text with a lot of East Asian or Hungarian authors in rudimentary style (only initials, comma delimitation), you are likely to run into trouble. Ditto nobility-based names etc. The templates work well for most Germanic and many Romance authors; they are liable to produce ambiguous or wrong cites for anyone else. I have noticed this when I came across some citations that proved to be untraceable, because the shorthand style in combination with the template messed up the authors' names. Dysmorodrepanis (talk) 13:16, 11 May 2008 (UTC)

---

I also would like to see the changes suggested above to support the "body" of the CITE at the bottom of the article where it belongs. That said, it's probably worth bearing in mind how the Shortened notes method already described in the project page achieves this. I've copied relevant subsection below.

[edit] Clearer editing with shortened notes

Because footnotes work by placing the required content inside <ref> tags within the article text they necessarily break up the text to some degree when in edit mode. Article text can become difficult to read and maintain. In this respect well referenced articles can unfortunately suffer disproportionately in comparison to those not so well sourced. In any case the disruptive effect can kept to a minimum by using shortened notes.

See the "Example edits for different methods" page for some comparative examples using shortened notes and full length references in footnotes. These offer representations of edit mode views with examples of how they render to the reader.

--SallyScot (talk) 19:55, 9 May 2008 (UTC)

Actually I like the "Shortened notes with wikilinks" version very much. However I find the internal portion "too complex", I would prefer a "fully shortened" version like you outline above. The downside to that version, which is also on that page, is that you have to hunt around for the unlinked reference. But that said, I think that it might be easier, technically, to introduce these changes because all it really needs is an additional tag in the CITE and REF to refer to each other...
<ref cite=smith pages=99-100/> would be the "placeholder" in the text, and by adding a similar "name=smith" in the CITE would allow the system to be completely automated. I threw in the page numbers because everyone complains about that :-) Maury (talk) 20:33, 9 May 2008 (UTC)

One can actually simulate, although not perfectly, this concept. See this version of the water memory article for instance. I used a "trick", and if you look at the source text I think you'll agree that it's very editable. It seems quite reasonable to suggest that there will be some sort of marker where the numbered note will appear in-line, and using the short refs makes it almost trivial. Maury (talk) 20:05, 9 May 2008 (UTC)

The solution proposed by you here has been suggested before (about 6 months ago). One of the reasons, I recall listed against it, that it may be instable in certain situations/editors. I do agree however that some kind of "reference library" somewhere at the end of the article, only using pointers inside the maintext would be a great development. This, apparently, needs some hardcore code rewriting (which I don;t know anything about). Arnoutf (talk) 21:13, 9 May 2008 (UTC)
That makes two of us! I did look into the code, and in spite of a relatively knowledgeable background in programming in general, I was completely stumped by the syntax in question and decided it was best left to the experts. :-( I am quite heartened to see that I am not entirely alone in believing this needs some attention though! Maury (talk) 23:27, 9 May 2008 (UTC)
Another argument which came up against this "trick" is that all the footnotes produced contain a nonfunctional backlink. In the example article, the final backlink in each footnote is nonfunctional. Placing the hidden div (<div name=cites style="display:none;"> ... </div>) containing the footnote declarations at the beginning of the article instead of at the end provides the additional capability for the editor to control the ordering of the footnotes by declaring them in that hidden div in the their desired appearance order. In that case, however, the nonfunctional backlink in each footnote would be the initial backlink instead of the final one. Bugzilla:12796 is a proposed mediawiki enhancement which would provide this functionality without resorting to the hidden div "trick" and without producing a nonfunctional backlink. Code to implement this is included there. This was submitted to Bugzilla on 2008-01-26, and is tagged "need-review". -- Boracay Bill (talk) 01:58, 10 May 2008 (UTC)

There's a simpler way to do this: use author-date/Harvard referencing. In fact, if you use anchors and then the cite tag at the bottom, you completely avoid these hassles. Now, you can clutter the visible page up some with authors and dates, but I believe that as an electronic encyclopedia we don't have to do that necessarily. We could just do away with the author and date in-text if we wanted and anchor page numbers to the source; if there were no page numbers to reference we could do the author or some sort of title. We're electronic; the author-date isn't necessary in-text. The CMOS basically says: ultimately do what makes the most sense for your reader, and in this case I think we could use an entirely new system, for the benefit of the readers and writers. Also, I like raw citations. The templates are terrible to read, and terribly wordy as well. ImperfectlyInformed | {talk - contribs} 03:37, 10 May 2008 (UTC)

---

I've got nothing against author-date referencing, but it isn't really much simpler than Shortened notes. Shortened notes can be coded as pretty much the same thing, only with the opening bracket of the author-date reference replaced with a <ref> tag, and the closing bracket with a closing </ref>.

An author-date example:

The Sun is pretty big (Miller 2005, p.23),
but the Moon is not so big (Brown 2006, p.46).
The Sun is also quite hot (Miller 2005, p.34).
== References ==
*Brown, R (2006). "Size of the Moon", Scientific American, 51(78).
*Miller, E (2005). "The Sun", Academic Press.

Same example using Shortened notes:

The Sun is pretty big,<ref>Miller 2005, p.23.</ref>
but the Moon is not so big.<ref>Brown 2006, p.46.</ref>
The Sun is also quite hot.<ref>Miller 2005, p.34.</ref>
== Notes ==
{{reflist|2}}
== References ==
*Brown, R (2006). "Size of the Moon", Scientific American, 51(78).
*Miller, E (2005). "The Sun", Academic Press.

--SallyScot (talk) 10:19, 10 May 2008 (UTC)

And author/date doesn't really work well when you're using a reference for a single item. I really hate having to jump from section to section to read it. Maury (talk) 11:50, 10 May 2008 (UTC)

but how well does "when you're using a reference for a single item" relate to WP, where you are not the only editor of the article at issue?
A-D is less repetitive, however. With your shortened footnotes you would have to glance down at the bibliography, while with A-D you would be taken immediately to the source. I'm not sure I'm following when you say "jump from section to section". If you anchor your A-D references, how is it any different from footnotes? ImperfectlyInformed | {talk - contribs} 15:51, 10 May 2008 (UTC)
Yes, author-date in-text references, if wikilinked, do allow the reader to link straight to references rather than two steps as with shortened notes. The disadvantage with parenthetical systems is that they break the flow of the text for the reader more than less intrusive footnotes references. The effect that this has varies from article to article depending on the number of references of course. Author-date references are fine for articles about uncontroversial subjects which may not require so many references. A footnote system (of some sort) comes into its own the more references are used. --SallyScot (talk) 19:17, 10 May 2008 (UTC)
In a controversial article, readers might or might not find the author-date system more intrusive. If the reader is checking the source of each claim, and the reader becomes familiar with the various source authors, it is more convenient to have the source of the claim right there in the text. --Gerry Ashton (talk) 19:29, 10 May 2008 (UTC)
The option to use inline author-date referencing is available. No technical barrier exists which would prevent use of both inline author-date and footnoted author-date references. Without closely re-reading the current version of this project page, my recollection is that it has little or nothing to say on the subject of inline author-date vs. author-date in numbered footnotes vs. both intermixed. The project page probably should provide some guidance in this area. In practice, when I see author-date, my impression is that it's usually either inline or in numbered footnotes — seldom if ever intermixed — and the choice of inline vs. footnoted seems to be made by the creator of the initial ref, and changing from that initial choice seems to be by consensus (either through prior discussion or by discussion of a reverted WP:BOLD change) on an article-by-article basis. I tend to use footnoted author-date myself, as most pages I edit have preexisting footnoted citations. -- Boracay Bill (talk) 23:24, 10 May 2008 (UTC)
Instead of saying inline, perhaps you should differentiate between parenthetical author-date and footnote author-date (I should have changed "Harvard referencing" to "Parenthetical referencing"). I suspect that's what you mean. Structuring references by author and date is a very simple and useful way to structure references in general because the author and the date are two of the most relevant facts, whether the inline is a footnote or not. I think that both parenthetical in-line cites and footnotes should be allowed on any one page. That way you can avoid repetitive footnotes and give people freedom. The Chicago Manual of Style does not appear opposed to using both for references, although I don't know that I've ever seen it professionally. WP, as electronic, should not be constrained by style: it should be constrained by functionality. And, as I said before, we don't necessarily need to use the author and date inline when your citation is anchored; could just use (p. 32) and let people click to see the reference. I know this sounds radical, but it makes sense. Of course, that does reduce the advantage when you're familiar with authors, and it makes it useless in print...but Wikipedia should not be printed anyway. ImperfectlyInformed | {talk - contribs} 23:37, 10 May 2008 (UTC)
At present, there does not seem to be any citation template that uses the Modern Language Association (MLA) style guide. The MLA guide is widely used and is the most commonly used style guide worldwide in the area of the social sciences which represent much of the Wikipedia material. I find that APA guides are often assigned at university as a "simplified style" most often associated with the sciences. What is often difficult to determine in the APA style is the use of multiple editions as the date is tied to the author note. As well, dropping the place of publication also leaves a gap as many publishing houses operate international offices. Due to these limitations, I often "scratch" catalog or reference source using the MLA guide of: Author, title, place of publication, publisher and date format. The APA citiation template also has certain limitations in its format including the use of ISO dating only which often introduces a jarring element in the article as invariably, two date conventions are in play. I would almost (repeat, almost) accept the APA guide as a standard if the ISO dating could be altered to a more readable m-d-y or d-m-y format. I had earlier asked a question as to why no MLA template exists? Again, why not? FWiW Bzuk (talk) 14:25, 11 May 2008 (UTC).
MLA is certainly not the most commonly used style for the social sciences. It is mainly used in the humanities. Since Wikipedia does more scientifically-oriented documentation than humanities documentation, we haven't added the MLA format. However, I think we should change the "author-date referencing" to "parenthetical referencing"; which is a more global and descriptive term. Under that system author-date is the most common method, but we should also mention MLA style. ImperfectlyInformed | {talk - contribs} 22:17, 11 May 2008 (UTC)
Since I write mainly in history-related articles, I would see the need for the use of the MLA system, especially for editors unfamiliar with the conventions of setting up a reference notation or citation, and rely on templates. One of the common issues that I see is that the information provided from templates are rife with errors, mainly due to inputs being made inaccurately. Again, I tend to "scratch" catalog and do not use the templates, but I can see the need to have templates to suit diverse subject areas, especially in the humanities. Wikipedia had primarily scientific-oriented articles? Really? maybe my reading of articles is too limited. FWiW Bzuk (talk) 03:04, 12 May 2008 (UTC).

I would also like to endorse Shortened notes. They easily resolve the problems detailed at the beginning of this thread and are quite easy to use and edit, even for novices. Madman (talk) 19:34, 15 May 2008 (UTC)

[edit] A method that doesn't require big changes

I don't foresee using the proposed shortened notes for articles I am typically involved with. It might be suitable for low-editor-activity articles that are stable. One un-described method (in this thread) is what I would describe as "formatted references & text" which separate the text from the references, and actually do make it easier to edit both text and references, and have been used on very actively edited articles. The good result of my technique for reference-heavy articles, is that the editor can locate the end of any reference easily, since the closing </ref> is on the first column of the editing window. This should make it easier to add or edit the text of the article while noticing where the end and start of each reference is. Here is the format, with a two-reference example:

some text at the end of the sentence.NOSPACE<ref>NEWLINE
body of reference material hereSPACE-NEWLINE
</ref><ref>
body of second citation reference materialSPACE-NEWLINE
</ref>
Start of next sentence.

The result looks like this while editing:

some text at the end of the sentence.<ref>
body of reference material here
</ref><ref>
body of second citation reference material
</ref>
Start of next sentence.

Recapitulating:

  • The intended result is that the editor can easily find:
    - the end of any reference,
    - the beginning and start of each reference when there are several together at the end of a paragraph or sentence,
    - and not least, easily find the start of the sentences,
    - plus generally all sentences are set-off from the references, instead of all being run-on together.
  • The run-on aspect of references and text in a reference-intensive article--is avoided. The run-on aspect is challenge for most editors.

This is not a theoretical item. It's in use here: Dismissal_of_U.S._attorneys_controversy. -- Yellowdesk (talk) 04:06, 16 May 2008 (UTC)

[edit] Return to the "shortened notes" topic

I agree with User:Madman, User:SallyScot, User:Dysmorodrepanis who lean towards shortened notes. The short footnotes keep the source as clear as I can imagine and they don't clutter up the text for the non-academic reader. When they are also linked (with anchors or with {{Harv}}) there is very little functionality lost. In this system, the original complaint (that templates take up too much space in the source) is moot. ---- CharlesGillingham (talk) 08:18, 17 May 2008 (UTC)
I think we should stick to the current system of letting the primary editors use what works best for them. I personally find the shortened notes system ugly, non-user friendly, and a pain to use. I much prefer using cite templates within a ref tag, as do many others. Why try to force other people to use such a system when its clear that Wikipedia allows and endorses several different methods. Shortened notes seems to primarily be preferred in topics with primarily academic type editors. Let them be used there then, but I see no reason to try to force all other articles to use it. Collectonian (talk) 08:31, 17 May 2008 (UTC)
I think you might have misread me. I think we're just making suggestions here -- exploring some of the possibilities that an editor could choose. I never intended to suggest that we should "force" anyone to do anything. There are articles where shortened notes are appropriate and articles where they are not. Shortened notes solve a few problems (for example, if you are citing several different pages from the same book, it saves you from repeating the citation for each new page reference). They are useless for articles that use a different source for every paragraph (like articles on current events) or articles that use mostly online sources (no page numbers).
More to the point, shortened notes clutter the source a little less, and cluttered source text is the problem that this discussion was trying to solve. ---- CharlesGillingham (talk) 09:50, 17 May 2008 (UTC)
Okay...cause it read to me like this was an attempt to make this the preferred method. Collectonian (talk) 05:33, 18 May 2008 (UTC)
Just to bring up an opinion (that can be a disturbing image, I agree), contrary to Collectie's reversion to the shortened notes, I think they are actually more readable and less obtrusive then full reference citations. Again, only IMHO. LOL Bzuk (talk) 13:25, 18 May 2008 (UTC).

I don't like shortened notes (too repetitive), but I do want less of the references cluttering up the edit window. Why can't we have some sort of way to build a special footnote reference (the "base" reference) without turning it into a footnote, and instead having it display the reference in full? That way we could give it a name and reference it all we want with the given name in the article, with each reference adding an a b c d ect to the main reference. If anyone's not following me, let me know and maybe I could make it more clear. Impin | {talk - contribs} 19:56, 18 May 2008 (UTC)

  • You seem to describe the current status quo, with the use of named references, such as <ref name ='some convenient name'/> . Not sure what your point is. -- Yellowdesk (talk) 02:07, 19 May 2008 (UTC)
    • The difference is that in the current system, the "base" reference creates a footnote. What has been repeatedly requested is a way to put all the base references outside the actual text, for instance in the reference section, without creating anything visible at that location. Christopher Parham (talk) 04:54, 19 May 2008 (UTC)
      • ...such as the cite.php extension submitted in January of 2008 as Bugzilla:12796, including code to implement it, still marked "need-review". -- Boracay Bill (talk) 06:04, 19 May 2008 (UTC)

Why don't we get people to vote for this? I just voted. By the way, I'm inclined to put this in Persistent when it's archived to make sure it gets noticed. I really want this to happen, and it seems to have consensus. Perhaps I could just put a summary of it, with a reference to your Bugzilla fix? ImpIn | (t - c) 22:49, 30 May 2008 (UTC)

I thought I'd stop back here and cite (so to speak !) Wrigley Square as an example of the non-shortened-notes citations making it very difficult to read (in edit mode) the text of the article. The frequent long citations, many of which containing random-ish strings of numbers, just overwhelm the actual text.
And, no, I don't want to force anyone to use shortened notes, but I'm thinking that it should perhaps be the suggested format. Thanks, Madman (talk) 18:44, 6 June 2008 (UTC)

[edit] Please cosider voting for Bugzilla:12796

A fair amount of people were involved with this discussion. It appears that there is a relatively easy way to hide all of the reference junk at the bottom of the article. Boracay Bill has done much of the work, it just needs to be glanced over and plugged in. Please vote so that we don't have to deal with all this junk when we edit; it's terribly distracting. If you disagree with the change, please let us know. ImpIn | (t - c) 08:54, 5 June 2008 (UTC)

[edit] Retrieval dates for online versions of old printed sources, again

I know this has been discussed a couple of times in these Talk archives, but I want to bring it up again. What is the rationale for requiring access/retrieval dates for online versions of past printed materials?

For example, editors are beginning to link book cites to Google Books. Thus, editors are putting "Retrieved on" on their cites, in addition to the usual author, title, publisher, year, ISBN, and page information. It looks very strange to see a book being "retrieved" ... such a link is just a convenience link (problematic too, given the semi-random way Google Books' "limited view" works); the content of the book is unchanging. If the link goes bad, the rest of the cite remains: an unchanging reference to an unchanging book.

Another case are old newspaper and magazine articles. If a cite gives a 1983 New York Times story's publication date, title, and author, and also gives a convenience link to the NYT archive, what is the value of having the retrieval date for this? The content of the story is fixed and unchanging, and is defined by the print/microfilm version. Again, if the archive goes away, the rest of the cite remains, an unchanging reference to an unchanging story. If the archive gets moved, one would re-lookup the online version by the published date/title/author information; knowing the old retrieval date wouldn't tell you anything.

And there is a real cost to having retrieval dates in place everywhere: to us they take up article edit space, to browsers they increase output HTML space, and to readers they clutter up the cite and can be visually confused with publication date. I understand that retrieval dates are necessary for web pages without publication dates, and arguably necessary for dated news stories originally published online (CNN, current NYT, etc.), but I just don't see the rationale for them in the above cases. Wasted Time R (talk) 23:08, 10 May 2008 (UTC)

It's useful to be able to refer to that date in the WayBack Machine at archive.org. In the case of the NYT archive, we can be fairly certain that those will always remain, but other links won't. It's quite possible that some print sources could be basically impossible (or rather expensive/time-consuming) to track down. People will increasingly rid of print archives. However, if you're crunched for time, do what you can. If it's a podunk town newspaper, put the date; if it's the NYT, don't worry about it. That's my take at least. ImperfectlyInformed | {talk - contribs} 23:40, 10 May 2008 (UTC)
The most common cause of newspaper links going bad is that articles get moved behind pay/subscriber walls. Is the WayBack machine able to show the article anyway, or are they enjoined from making free what is otherwise supposed to be charged for? Wasted Time R (talk) 23:58, 10 May 2008 (UTC)
One of the issues with the citation template is that the nomenclature of "retrieved on" is tacked on automatically and now has become part of the architecture of the citationa as judged by the amount of citation templates in place. I agree that the term looks arcane but with its widespread use, it is hard now to incorporate a "found," "accessed" or "located" tag as an alternative. FWiW Bzuk (talk) 14:25, 11 May 2008 (UTC).
To clarify, my issue is not with what word is used here. I don't think books or old newspaper articles should be listed as "found", "accessed", or "located" either. Those printed sources are unchanging over time; it doesn't matter if you "find" a 1976 book in 1988 or 2008, it's the same book. Wasted Time R (talk) 12:04, 12 May 2008 (UTC)
Absolutely agree on that point, sources that are "fixed" in time, do not require a location date. FWiW Bzuk (talk) 19:26, 13 May 2008 (UTC).

The "retrieved date" merely refers to the convenience link to the online version, and may be safely removed on any cite that is not an online link. That's all. (And if the link goes bad, the dead-tree portion of the cite remains valid.) -- Yellowdesk (talk) 00:40, 16 May 2008 (UTC)

But what's the purpose of a retrieval date for an online version that's just mirroring a print original? What usefulness does it have? What does it tell anyone? Wasted Time R (talk) 04:29, 16 May 2008 (UTC)
  • On more than a few occasions I have used the retrieval date for munged references to rediscover the orginal edit that created it, and on more ephemeral sources, search for likely new location for the changing location of the convenience link. In some cases a retireval date indicates when the (changing) source was viewed and relied upon, occasionally important, when the source has changed. It's not superflous, but I would consider it optional.
    Who's to say that even a supposedly fixed archival convenience link will stay that way, and what harm comes from using the access date even there, such as in this example:
    "New Hampshire: Nomination of Bainbridge Wadleigh for United States Senator at the Republican Caucus.", New York Times, June 14, 1872, pp. 1. Retrieved on 2008-05-05.  -- Yellowdesk (talk) 14:29, 16 May 2008 (UTC)
  • The harm is that the "Retrieved on" takes up extra space (a real issue for our longer, heavily-cited BLP articles) and moreoever is visually confusing — the reader sees two dates, instead of the expected one, and has to figure out what each means, which a possible risk of mistaking the retrieval date for the publication date. In the example of this old NYT story, if the link stops working, it's because the NYT moved its archive or changed its for-free policy on this time period or something like that. If you need to find where they moved it to, you'll do a lookup within nytimes.com using the article's title and publication date; when someone last retrieved it won't matter one way or another. And would you really use a retrieval date for a book, that someone happened to look up in Google Books instead of at a physical library? That really seems offbase to me. Wasted Time R (talk) 21:44, 16 May 2008 (UTC)
  • Yes, I would, and have. Especially on heavily edited articles. For the reasons I stated further above: an indicator of when the convenience link worked. I do consider it optional. For example, if some link has an old retrieval date, and apparently not findable by search, then I tend toward deleting the convenience link. For more recent dead links, I'm less likely to remove the link--perhaps the publisher/source is in process of revising the link/location. Essential? No. Useful? Yes. The "retrieved on" is in english, and if using a template, the template does indicate through the parameters how to properly use it. Say more about the confusion you've encountered. (I have to remark, there's plenty of other confusion on articles surrounding refs, such as puctuation, quotations, where to place it and so on, and I've done a fair big of cleaning up other's typos and misplacments on that score. Is this that much different?) -- Yellowdesk (talk) 05:23, 18 May 2008 (UTC)
I think this largely depends on the 'dependability' of the on-line source. For the NYT above, the accessdate is not really needed. On the other end of the spectrum, here is where someone (it's not even clear who) added sections of a (very) small town newspaper from the first half of the 1900s. It's true that this is on-line copy of a print original, but I think it would be rather difficult for even a motivated researcher to find that original. So in practice, the web copy is all that exists, its maintenance is unknown, and an accessdate tag is appropriate. As to how this might be implemented in practice, I think there could be a list of sources that are considered stable enough that accessdate tags are not needed (major newspapers, academic journals (DOIs are an explicit attempt to address this here), arXiv and other pre-print servers, and so on). LouScheffer (talk) 17:51, 26 May 2008 (UTC)
(off topic) You deserve a barnstar if you've been cleaning up refs. I'm surprised you haven't run off screaming. :) -- Fullstop (talk) 19:14, 26 May 2008 (UTC)

[edit] Hide the access date

In order to find the content of broken links in archives it would be sufficient to store the retrieval date in a comment that is not visible to a reader, only to editors. This is an approach I would support.
Otherwise, I second the notion that (visible) retrieval dates for off-line media are visually irritating, cluttering and superfluous. --EnOreg (talk) 05:50, 20 May 2008 (UTC)
  • Fair enough. How do you propose to obtain uniform use of the proposed standard? -- Yellowdesk (talk) 04:44, 21 May 2008 (UTC)
A partial solution would be modify the citation templates to store info generated by the accessdate= parameters as a in an HTML comment that is not visible to a reader, only to editors. That would quickly handle a large percentage of retrieval dates. Many thousands of articles would need to be individually edited to bring the handcrafted cites into line. -- Boracay Bill (talk) 05:41, 21 May 2008 (UTC)
Bill suggests what I had in mind: Leave the parameters of the citation templates as they are, just modify their implementation to not display the access date (except cite web). And adjust the WP policy pages to reflect this change. --EnOreg (talk) 01:27, 22 May 2008 (UTC)
Wouldn't it be easier to just use a field that is visible to people editing the page but not to people viewing the page? But that function is available now in all templates: just use a field that the template does not itself already use. E.g. invisible-retrieval-date= ... —David Eppstein (talk) 16:23, 26 May 2008 (UTC)
Exactly. By removing any mention of {{{accessdate}}} in the template implementation, the data would remain, but wouldn't be parsed by the server, so the casual reader's display wouldn't be cluttered. I'd support that for {{cite journal}}, at the very least, as with this template the accessdate is of no real utility when rendered. Smith609 Talk 16:59, 26 May 2008 (UTC)
Not sure sure I follow. Sounds to me like we violently agree. What's the difference between your proposal and Bill's? --EnOreg (talk) 18:19, 26 May 2008 (UTC)
To "hide" the access date, the templates only have to not parse accessdate= parameter. No HTML comment is necessary, nor is it necessary to rename the parameter. After all, its still in the source.
But "hiding" the access date only addresses the symptoms. It does not fix the underlying problem, which is the misconception that a source on the web is a web source.
As such, merely hiding the access date (however that hiding occurs) for all but {{cite web}} will not be much use -- {{cite web}} is being used for virtually everything that editors happen to find on the "web".
The source of this misconception is of course the {{cite xyz}} farrago. That a source on the web must be cited with {{cite web}} is merely a "logical" continuation of that nonsensical paradigm. That is the real problem (and living proof that caring about sources has zero priority).
But hiding accessdate is a start, even if its only a band-aid. Next step other insane linking (e.g. google books, amazon, jstor and so on). In the long run we must teach editors how to cite properly, how to quote properly, and why it is necessary to do both.
-- Fullstop (talk) 19:14, 26 May 2008 (UTC)

The retrieved date allows a reader to understand the age of the online link. In the past, I have done a manual link check and have updated those retrieved dates to show that the links were still valid as of that date. The CheckLinks tool checks links, updates to archived links on dead links and now optionally updates the retrieved dates. --—— Gadget850 (Ed) talk - 15:30, 27 May 2008 (UTC)

  • I invite someone to apprise those who watch the various "cite" templates to put a notice on each of the cite-template talk pages, that this conversation is occurring. -- Yellowdesk (talk) 05:19, 28 May 2008 (UTC)
I put a notice there already some days ago. Anything else we can do to invite feedback? --EnOreg (talk) 15:48, 28 May 2008 (UTC)


I do not see how an accessdate on sources which do not change - such as journal articles - is beneficial. However, on sources which may change, such as web content, it helps clarify which version of a page is being cited. Therefore I feel it ought to be displayed only in the cite web template. I don't think anyone has disagreed with this feeling here, so I suggest that someone bold goes ahead and proposes or enacts the change at all non-"cite-web" templates. People have had the chance to complain if they feel otherwise! Smith609 Talk 23:12, 3 June 2008 (UTC)

I appreciate what is being discussed here. In my opinion there are two issues popping up:

  1. Print sources of which you get a copy from the web (JSTOR etc) should be referred to as their print version. Access date is irrelevant as the content is not dependent on the web, nor will it ever change. For such sources the use of citeweb should discouraged, and access date not listed or removed
  2. True web sources, which are rarer than most editors seem to think is another issue altogether. Websources are not permanent, and even if they are long term the content may dramatically change. Therefore it is not only essential that access date is recorded and reported, but also that when updating text for such sources a critical reflection whether the text is still covered by the website has to be applied. In printed articles, this is not so much an issue as you refer to the website once, and your text will not change, even if the website content does. As both Wikipedia and referred to websites change this is very complicated indeed. Arnoutf (talk) 06:18, 4 June 2008 (UTC)

Consensus: It indeed seems we have consensus that access dates for online copies of offline sources, while helpful as a comment in the source, should be hidden from the reader. I have removed the RFC (style) tag and will modify the policy. Anybody who is competent to adapt the citation templates, please do so. Thanks everybody, --EnOreg (talk) 08:41, 5 June 2008 (UTC)

HTML comments are stripped out by the Mediawiki software, so these won't be visible except in the original template call. I've included one here, for instance: Would it be better to hide the date with CSS? — Omegatron (talk) 17:28, 7 June 2008 (UTC)

That's a good idea. We can also assign an ID to it in case people want to make it visible with user css. --Karnesky (talk) 18:22, 7 June 2008 (UTC)
I've responded to all the editprotected requests that are up at the moment by wrapping the "retrieved on..." text in a CSS class (reference-accessdate), so it can be hidden in either personal or sitewide CSS while still being accessible for those that want to see it. You can personally hide the accessdates yourself by adding 
.reference-accessdate {display: none}
to your monobook.css. If there is a real and extensive consensus to hide these data, adding the same code to MediaWiki:Common.css would have the same effect for all users who didn't override it in their own monobook. Happy‑melon 17:54, 8 June 2008 (UTC)
Thank you. Options are better than hard coding here. Where do we document this? --—— Gadget850 (Ed) talk - 18:36, 8 June 2008 (UTC)
No idea :D. From a technical end, I've added to the catalogue at WP:CLASS; where and how you note the new feature is the balliwack of people on this page. As an ultimate goal, we ought to be working towards encapsulating all the similar reference 'facts' in suitable css (reference-title, reference-volume, etc) and defining their appearance centrally in Mediawiki:Common.css. That greatly facilitates updating and standardisation between cite templates (I shouldn't have had to edit five templates to implement this change), and instantly circumvents the "data X should have formatting Y because it's the standard of source Z": we can just say: go on then, add foo to your monobook and the problem is solved. Ultimately, I have yet to see a good reason why a properly-built {{cite meta}} is not possible, to centralise and de-duplicate the considerable amount of code (the CoinS tags, for instance) that is almost identical across all the cite templates, and needs to be maintained in the same way in each. But that's another story. Happy‑melon 19:32, 8 June 2008 (UTC)

[edit] Access date for newsgroups and mailing lists

I don't see any strong consensus to hide this parameter for templates where the availability of material might be ephemeral. I think it should stay visible on, at least, the generic citation template, the mailing list template, the newsgroup template. --Karnesky (talk) 13:22, 5 June 2008 (UTC)

Right, I'm afraid this hasn't been discussed properly, yet. To make this clear: I don't advocate removing the access date, only hiding it from the reader. Unlike most web pages, posts to mailing lists and newsgroups carry a "publication" date that doesn't change. Therefore, the additional access date doesn't add any value for the reader. It can, however, make it easier for editors to recover a link that has become unavailable. That's why we should keep it in the page source as a comment. Note that mailing lists and newsgroups are being replicated and archived in so many different places that it is much easier to find a post than a copy of an arbitrary web page. --EnOreg (talk) 13:49, 5 June 2008 (UTC)
I understand what you are advocating, but I think that it should stay visible for content that might not be locatable or might have changed at some future date.
As a reader, I've printed out articles & retrieved the references from them (both physical sources & online sources), and the accessdate is useful for sources that might change URLs, disappear completely (some usenet posts have requests not to archive, for example), etc. The parameter's utility is greater than any aesthetic objections. At bare minimum, the accessdate should be visible when the publication date parameter is not given. But I think it should always be visible for sources that don't have physical manifestations. --Karnesky (talk) 14:21, 5 June 2008 (UTC)

I'd like state that I'm strongly opposed to this idea for any template that may cite any kind of online material. For Cite book, Cite paper, etc, that are only used to cite physical or "permanent" publications (even if it may be found online and linked to in a particular template), then so be it, Accessdate isn't necessary. But to hide it in Cite news, Cite press release, Cite map, etc etc (which more and more may cite a document online that *cannot* be found in print) is doing a grave disservice to anyone who doesn't want or know how do delve into the edit page and figure things out, yet still may want information that will allow them to access a website that has been lost over time. That is precisely what Accessdate is useful for; not to mention, even for webpages that are still existent, it says precisely when data was originally pulled from the source. "Accessed on..." or some variant of it is an almost universal standard for citation formats outside of Wiki...I see no reason why we should be the oddballs and not use them in a citation display. — Huntster (t • @ • c) 14:25, 6 June 2008 (UTC)

Hiding the date for one template such as {{cite news}} without changing all of the templates is going to cause some inconsistency. There are already enough differences among the cite templates. There are opinions on both sides of the issue as to show or hide the accessdate— why not allow editors who don't want to see the accessdate to be able to hide it? We should be able to come up with a script to do this and get it approved as a gadget. --—— Gadget850 (Ed) talk - 19:17, 7 June 2008 (UTC)

[edit] Linking to full-text articles copied by non-mainstream sources

I'd like some input on this edit. Basically we're linking to the full text of an article. I undid Shot info's edit and added the NYT archive link, but left the full-text for convenience. Petergkeyes claims to have seen the full text, and the NYT archive looks the same for what we can see. I often see links to the full-text of for-pay articles for convenience. Is this wrong? ImpIn | {talk - contribs} 17:20, 23 May 2008 (UTC)

My input: Wikipedia:Copyrights says, "... if you know that an external Web site is carrying a work in violation of the creator's copyright, do not link to that copy of the work." This link should be removed or tagged {{copyvio link}}. -- Boracay Bill (talk) 00:40, 24 May 2008 (UTC)
Makes sense. Thanks. ImpIn | {talk - contribs} 01:03, 24 May 2008 (UTC)

For clean NYT links, see User:Gadget850/New York Times. --—— Gadget850 (Ed) talk - 15:51, 27 May 2008 (UTC)

If you search for the link you provided back on that article, however, they ask to charge for it. Are you sure you're not kind of overcoming that and breaking copyright? ImpIn | (t - c) 01:15, 31 May 2008 (UTC)
I considered that, but links like these come up in Google searches. I registered with the NYT, but I don't have a subscription. As I noted, the URL comes up in different ways— I don't know why. --—— Gadget850 (Ed) talk - 10:01, 31 May 2008 (UTC)
http://www.fluoridealert.org/health/accidents/kennerly.html appears to be a copyright-violating link. If it is, it should not be linked from WP. Period. End of story. -- Boracay Bill (talk) 11:51, 31 May 2008 (UTC)

[edit] Blogs as a source

Are blogs an acceptable source?(Bonzai273 (talk) 11:29, 1 June 2008 (UTC))

Generally not. See WP:SPS. -- Boracay Bill (talk) 11:42, 1 June 2008 (UTC)
Depends on whose (am I using the correct word?) blog it is , as SPS notes. RealClimate has been used as a RS. By the way, Bill, did you see my comment on getting the vote out on that code improvement for footnotes? ImpIn | (t - c) 05:09, 2 June 2008 (UTC)
Yep—never say never. I saw the comment and the vote on bugzilla:12796. Thanks. That makes two. -- Boracay Bill (talk) 05:32, 2 June 2008 (UTC)

[edit] alphabetizing authors with different naming traditions

Cambodian names follow a surname + given name pattern. Given names are the names most commonly used in formal discourse. In an alphabetized reference list, should it be Given name, surname or Surname given name? (e.g. name is Chea Phalla [Chea=surname, Phalla=given name], should it be Phalla, Chea or Chea Phalla?) Thanks, Mangostar (talk) 05:29, 3 June 2008 (UTC)

Even when English language names would appear as "Smith, Michael", it should appear as "Chea Phalla". If you are alphabetizing a list of authors, the entry would be alphabetized with the "C"s. This is true in all systems I am familiar with. Christopher Parham (talk) 06:08, 3 June 2008 (UTC)
A follow-up: I have changed it so now I'm not alphabetizing, but I'm doing last names first in in-line cites for Western names, and first names first for Cambodian names. But now the issue is how to do op cits for Cambodian names. If you were addressing Chea Phalla in polite company you might say something along the lines of Mr. Phalla, or if you were writing an encyclopedia you would write Phalla. Should I then use Phalla in the op cit refs? It has the disadvantage of being visually confusing, since it's not the leftmost name (since we're righting "Chea Phalla", it is harder to skim for Phalla than if we were looking for "Jones, Smith" based on a Jones reference). Mangostar (talk) 09:19, 3 June 2008 (UTC)
By the way, the relevant article is Ratanakiri Province. Mostly now it's just the Sith Samath one that's tripping me up. (And I guess Phat Palith.) Mangostar (talk) 09:21, 3 June 2008 (UTC)
Personally I would prefer to stick to surnames and use "Chea, op. cit" but you would have to consult the style guide you are trying to adhere to for a definitive ruling. Christopher Parham (talk) 22:51, 3 June 2008 (UTC)

[edit] Best reliable sources

In case of multiple possible references for a statement, the "best" reliable sources should be used. This line from WP:WHEN indicates that best practice is not to cite every possible reference for a statement, but to select responsibly among them; otherwise people are often tempted to string five or more citations together for a single proposition. I believe this is an important balancing guideline for WP:CITE#When to cite, that needs to be stated to affirm best practices; where and how would it best be inserted? JJB 08:09, 4 June 2008 (UTC)

I think this is something which should be left up to the judgement of the editors of a particular article. Some subjects are non-controversial and hardly need a reference at all. Other subjects require intensive sourcing. (For example, when there is a constant flow of original research and fringe points of view. If several major sources make the same point, it proves that it is an essential part of the subject matter.) Sometimes a few extra sources doesn't hurt. ---- CharlesGillingham (talk) 09:28, 4 June 2008 (UTC)
Agreed. I don't think we need a statement about that on this page. It is part of the NPOV policy, and this page doesn't need to repeat everything in the policies. Multiple references for a statement (I mean more than three) are not as common as you imply, and the phenomenon is usually related to talk page controversies of some sort. qp10qp (talk) 12:42, 4 June 2008 (UTC)

[edit] When/How Does "Unreferenced" Get Removed?

In the article on stubs, it says, "Be bold in removing stub tags that are clearly no longer applicable." But what about an "Unreferenced" tag? Clearly, I would think, more circumspection is required for the latter than the former, but surely at some point, after a certain critical mass of supporting citations has accrued the article, it must become appropriate to remove the latter as well.

So, then:

  • What is the critical mass?
  • By whom should the removal be done?
  • Through what process should he or she go, prior to removing?

208.252.192.131 (talk) 10:00, 5 June 2008 (UTC)

As with most things here, it is going to be up to the judgment of individual editors. It is not only quantity, but also quality. Add or remove any of the tags as needed. Remember: edit, revert, discuss; If you make a change and it is reverted, discuss it before getting into an edit war. --—— Gadget850 (Ed) talk - 14:59, 6 June 2008 (UTC)

[edit] Findability

The citation should clearly, fully, and precisely demonstrate that the source text is reasonably findable, such as by external link to the source website. Providing an ISBN or OCLC number, linking to an established Wikipedia article about the source, and directly quoting brief context on the talk page also each assert findability sufficiently.

Above text I inserted was reverted by Crum375. Variations of this have also been tried at WP:V and the suggestion was made to try it here. A couple editors seem to have misunderstood the intent.
The question that arose at WT:V was: when have I in good faith discharged my duty to ensure my source can be found, if it's not on the web? I should be able to have guidelines for knowing this at time of insertion, without having to wait for the potential challenge. The answer we developed was: it's sufficient for "findability" if some other link that indicates the source can be found by a reasonable editor. If you cite an ISBN or OCLC number, a link arises to demonstrate that some libraries or archives do in fact contain the source. If you quote the context in talk, that demonstrates that you have the book and (assuming good faith) have summarized it correctly in mainspace. The same is true if you wikilink the source from the citation, as is routinely done: if the publication, author, or publisher has sufficient independent notability to have a WP article, that indicates that the source can be found. For instance, my local paper is reliable but not widely famous. If I quote something from print which is unavailable online (which I have done), the fact that the paper has its own WP article is sufficient to demonstrate that my quotation is findable.
The whole point is that if someone quotes a rare book and leaves, we may well have a full cite with page numbers, but we can't prove that the book and author exist without some findability check. So if no link is provided [add the obvious: or found by another editor's reasonable search], the next editor is justified in deleting the alleged source as unverifiable. If a link is provided, then discussion can turn to the other topics, such as what the source actually says (more V), whether it's reliable (RS), and whether the summary matches the source (NOR), and so on. I think the misunderstanding arises from the idea that this passage confers automatic verifiability on all sources that happen to be mentioned by WP. No, it confers findability. Would anyone else like to comment on the worth of this guideline for settling the question raised? JJB 16:20, 6 June 2008 (UTC)
We depend to a large degree on good faith, so one should not delete a citation to a rare source just because it seems inaccessible to us or because the editor neglected to add an OCLC number, or whatever. If the source cannot be found, the approach should be to seek to verify the information through alternative sources. If you cannot do so, you could then argue that the information itself is too obscure to go in the article and replace it (citation and all) with something more widely known. But that is a different thing to shooting difficult-to-find sources on sight.
Although a comment on a source in the talk page is courteous and useful when adding citations to rare sources, it has no verifying effect on the article text itself, because most readers won't think of checking in the talk-page archives. And a source shouldn't be cut because an editor neglected to add details on the talk page. qp10qp (talk) 17:25, 6 June 2008 (UTC)
Sorry, I may have introduced another possible misimplication in my latest explanation, now refactored. Your first paragraph is exactly what I meant. The issue is that if I have a reliable source so rare that zero information exists anywhere on the web about it (beyond my using it in cites at WP), I should describe some method that someone besides me can access the book directly (such as by directly quoting it, taken in good faith). If I haven't, then a challenger is acting in good faith to request that access method be supplied. That's all. JJB 18:26, 6 June 2008 (UTC)
Discussion at WT:V#Demonstrably findable? and WT:V#Extended protection sounds like consensus that the description of findability should go on this page. I am refactoring it based on the need not to make it seem like a demand unless another editor finds the text to be inaccessible, and spelling it out a bit more because guideline not policy:

The citation should state, as clearly, fully, and precisely as possible, how a reader can find the source material, such as by external link to the source website. If the material is not findable online, it should be findable in reputable libraries, archives, or collections. If a citation without an external link is challenged as unfindable, any of the following is sufficient to show the material to be reasonably findable (though not necessarily reliable): providing an ISBN or OCLC number; linking to an established Wikipedia article about the source (the work, its author, or its publisher); or directly quoting the material on the talk page, briefly and in context.

JJB 20:25, 6 June 2008 (UTC)
Inserting due to lack of discussion. JJB 16:57, 9 June 2008 (UTC)

[edit] Embedded links

These break the connection between the place a link is used and the full citation, which is not only a problem for editors, but perhaps more so for readers, since they can't simply click their way to the full citation. I propose to drop this citation method entirely. -- Shinobu (talk) 19:32, 20 November 2007 (UTC)

What is the accepted wikpedia policy on external links as references? There seems to be conflict (either explicit or confusion) about what the proper reference style is. If I were to link to an external site as a reference to a fact, is the proper usage to [1] link it with a numbered link, or to put that link with full information inside a <ref> tag so that the link appears in the reflist and only a numbered link to the reflist is placed? I always thought numbered external links (the first way) was discouraged in favoure of reference lists, but Wikipedia:Embedded citations seems to suggest this as the proper way. TheHYPO (talk) 18:49, 18 December 2007 (UTC)
Embedded reference are tolerated, but full bibliographical information is better, because if the link goes dead, there will be better clues to help figure out where the information might have moved to, or where to find equivalent information. The full bibliographical information can be included with <ref> tags or with Harvard referencing. --Gerry Ashton (talk) 19:49, 18 December 2007 (UTC)

---

With regard to Embedded links (discussion points above from Archive 19 - see also here) I would like to include the following sentence at the end of the section...

Because of the difficulties in associating them with their appropriate full references, the use of embedded links for inline citations is not particularly recommended as a method of best practice.

I'd like to hear any reasonable counterarguments beforehand. Thanks, --SallyScot (talk) 23:09, 7 June 2008 (UTC)

[edit] Last Name First in Footnotes?

I have noticed that the Citation template lists the author's last name first, first name last (e.g., Shakespeare, William, 134). This makes sense in an alphabetized bibliography, not in a footnote. Why is it set up this way? Is there a style guide which approves this format? Thanks.Editor437 (talk) 02:07, 9 June 2008 (UTC)

The citation templates are designed primarily for making full citations in reference lists. That said, their use in footnotes has become common. I don't think any readers are likely to be confused by this. Christopher Parham (talk) 03:02, 9 June 2008 (UTC)
I don't think there is a style guide that approves the format. It would be nice to have more sophistication, though. I haven't been able to find a correct bibliographical form: full stops, rather than commas. Also, the last-name/first-name reversal repeats with the co-authors, which is redundant. qp10qp (talk) 10:48, 9 June 2008 (UTC)

[edit] Making sure to point out the page numbers

One advantage of parenthetical citing is that you are forced to point out page numbers. It seems fairly common on here not to do that. Now, the Find function (CTRL-F) is powerful, but not everyone is aware of it. I'm curious: where do I fit these notes in on a footnote citation template, say the cite journal one? I would think location, but the examples have places (e.g. Berlin) in location. ImpIn | (t - c) 02:51, 9 June 2008 (UTC)

There is a "pages=" field in most of the citation templates. Christopher Parham (talk) 03:03, 9 June 2008 (UTC)
Err, yeah. I was thinking of a more specific place to put the particular page with the statement rather than all the pages of the article, but I suppose I should just add see <particular pages> to that section. ImpIn | (t - c) 03:09, 9 June 2008 (UTC)
Well I would assume that the field could be used either to show all the pages of an article/chapter/etc., if you were making a general reference, or a specific page if you were citing a particular fact. I can't imagine when you would want to show both a range of pages and a specific page within that range, though. Christopher Parham (talk) 03:12, 9 June 2008 (UTC)
Hmm, yeah. You make a good point. I guess this was sort of a dumb thing to bring up. But when you cite to a particular page, you can't use the reference again, and have to do a new footnote. And that's a hassle. But when you cite generally, people have to search. Personally, I think that many papers could be cited by page number alone in-text; if there's an anchor, the author and date is unnecessary. ImpIn | (t - c) 03:35, 9 June 2008 (UTC)
You can solve this problem by creating a subsection under References to house those general citations, and then use inline cites with page numbers to reference them. See an example of this at University of Tennessee Forensic Anthropology Facility. So long as the article doesn't use multiple sources from the same author written in the same year, the results are very easy to interpret. — Huntster (t • @ • c) 04:09, 9 June 2008 (UTC)
You're describing shortened notes. See Wikipedia:Citing sources#Shortened notes, or (for a longer list of examples) see Wikipedia:Verification methods#Shortened notes. This is the most popular method for citing several pages of the same source. (There are a few other methods, such as {{Rp}}). ---- CharlesGillingham (talk) 04:49, 9 June 2008 (UTC)

I don't like shortened notes at all. {{Rp}} is much better, and sort of what I was looking for. For large works used often in a complex article, however, in-text referencing of the page numbers is better in my mind because you end up creating less of a mess of footnotes, and you point them directly to the work in question. ImpIn | (t - c) 00:55, 10 June 2008 (UTC)

[edit] Where's the help? Just awful as basic guidance

I make a few Wikipedia edits a week. I know citing sources is important. So why is the "help" for doing this so garbled and disorganized? I have to troll through paragraphs on different citation styles, lots of WP:This and WP:That, links to subsections, links to absurdly complex tables of reference templates. Nothing explains the difference between cite and ref and footnotes, or summarizes the most useful templates. So every single time I give up and just copy the wiki text of a nearby citation (or reference?), I have no idea if the one I copied is done right or not.

Please, provide editors one simple guide to doing the right thing, and make sure all the other pages feature a link to that simple guide early on.

  • Help:Citations quick reference is too bare-bones. The examples don't look like decent references, and the column "In References" doesn't explain how to make a reference.
  • Wikipedia:Referencing for beginners is promising, but it doesn't give any guidance on standard formats or templates, it's flagged as "just an opinion", and it's longer than Help:Footnotes.
  • Wikipedia:Embedded citations adds insult to injury by saying "For details about the other inline citation methods see Wikipedia:Citing sources." I don't want details, I want the basics. 95% of Wikipedia editors agree! [citation needed]

I shouldn't have to understand the difference between cite, ref, and footnote just to respond to the endless "citation needed" exhortations. If I do, then explain them to me, don't assume anything.

I think it's as simple as, if the document already has a References section, just add <ref>{{some standard citation template|its params...}}</ref> after the text. But I sure didn't learn that by reading all these pages.

Thanks for listening, now I'm off to make a citation reference footnote, badly -- Skierpage (talk) 04:25, 10 June 2008 (UTC)

I agree, there's a lot of noise on the page. This is true for almost all Wikipedia documentation...I believe it may be difficult to change, but let's try. ImpIn | (t - c) 04:38, 10 June 2008 (UTC)
I agree that this page is too long and poorly organized, especially for the beginner. ---- CharlesGillingham (talk) 12:21, 10 June 2008 (UTC)
Agree with this well-formed request, working this, slowly. JJB 15:34, 10 June 2008 (UTC)

Suggestions:

  1. The overhead banner should be shortened somewhat. The WP:CITING should be changed to WP:CITEWIKIPEDIA or WP:CITEWIKI; WP:CITING should redirect here.
  2. The intro paragraph should have a description of the basic ways to cite, anchored links to "how to write" each of these, and perhaps an anchored list to "why to cite". At the moment it is redundant.
  3. Why sources should be cited should be moved down to after the basic rundown of how to write ect.
  4. "Use of terms" section can be relegated to a footnote.
  5. In the intro, it should be noted that citations are not always necessary, with an anchored link to the "When to cite sources"; however, this is common sense, and thus that section should be put after the practical details.
  6. The "How to cite sources" section can be cleaned up significantly into a to the point, practical explanation.
  7. Tools should be emphasized immediately in the How to Cite section, as they are very useful. Nobody has time to hand-write really good citations when they're volunteering. I use the Google Scholar Wikify tool a fair amount. Attempts should be made to give different tools short, distinct, and descriptive names.
  8. Structuring citation templates so that they cover much of the page irritates me (this is how the above tool does it by default). If possible, we should come to some sort of consensus on this. I'll take this moment to remind people that we can remove citation templates from the prose itself by voting for Bugzilla:12796. ImpIn | (t - c) 00:55, 11 June 2008 (UTC)
Could you make each of these suggestions under a different topic? I.e., make a headline for each of these? It's hard to comment on a things in a numbered list. ---- CharlesGillingham (talk) 12:26, 11 June 2008 (UTC)
A headline for each? Don't you think that's excessive? I'll start with the intro paragraph ImpIn | (t - c) 00:59, 12 June 2008 (UTC)

[edit] Revising the intro paragraph: currently redundantly goes over style

Currently the intro paragraph redundantly repeats the information in the headline: that this is a style guideline. It would be better if it summarized the article and succinctly pointed people to the different methods, how to use them, and the tools available. It should be noted briefly that citations are not always necessary, with an anchored link to the "When to cite sources"; however, this is common sense, and thus that section should not be given undue weight. The focus here should be on telling people how to cite, rather than explaining why. ImpIn | (t - c) 00:59, 12 June 2008 (UTC)

[edit] Reference section fold up

Does Reference section in a big article will be fold up to occpying fixed height whatever how much.219.68.144.162 (talk) 07:00, 13 June 2008 (UTC)

Don't do it, because it won't print. This was discussed before. --Gerry Ashton (talk) 12:09, 13 June 2008 (UTC)
Yep. see Wikipedia:Citing_sources#Scrolling_lists. -- Boracay Bill (talk) 02:58, 14 June 2008 (UTC)

[edit] It's all true

    First time I've seen His lineage on the internet.  —Preceding unsigned comment added by 69.248.25.243 (talk) 02:56, 14 June 2008 (UTC)

[edit] Language

I hope i am not sounding biased or bad or anything. I am just wondering whether if references have to be or at least should be in english\english translation. Looking at the recent 2008 Iwate earthquake, a good number of the references are in Japanese, for example. Simply south (talk) 01:02, 15 June 2008 (UTC)