Template talk:Cite news/Archive 5

From Wikipedia, the free encyclopedia
Jump to: navigation, search
← Archive 4 Archive 5 Archive 6 →

Contents

Linking to disambiguation pages

Currently it doesn't seem to be possible for this template to link to pages that have disambiguation marks around them, e.g. The Nation (newspaper), in the "publisher" field. Is there any way to make this work? Perhaps a new field, 'publisherlink' (similar to 'authorlink') could be added to solve this. Sorry if this has been discussed here before. /Slarre (talk) 01:11, 4 July 2009 (UTC)

What is wrong with this?
"The title". The Nation (newspaper). July 1, 2009. p. 10. 
Crum375 (talk) 01:29, 4 July 2009 (UTC)
Because typically all links on Wikipedia, except on disambiguation pages, are shown without marks? Just like we link it John Smith instead of John Smith (actor) in the main text. /Slarre (talk) 01:38, 4 July 2009 (UTC)
You mean like this?
"The title". The Nation. July 1, 2009. p. 10. 
Crum375 (talk) 01:44, 4 July 2009 (UTC)
Ah, yes that's what I mean. It didn't work for me the first time, since I wrote [[The Nation (newspaper)|]] instead of [[The Nation (newspaper)|The Nation]]. /Slarre (talk) 13:45, 4 July 2009 (UTC)
Still wrong, though, because "The Nation" in this example should be in italics. Alarics (talk) 19:55, 13 September 2009 (UTC)
The error arises because you have put "The Nation" into the publisher field, whereas it should go in the "work" or "newspaper" field. That would have put the organ's title into italics, which is the correct form. -- Alarics (talk) 09:34, 15 September 2009 (UTC)

Database information

This template does not seem to have a field for database information. I retrieve many newspaper articles, for example, from LexisNexis. To properly cite this, I should say that obtained the article from LexisNexis (I have discovered, for example, that LexisNexis versions are sometimes different from the actual newspaper, so it is crucial to let readers know where the information is coming from). Could a field be added to facilitate adding this information? Awadewit (talk) 17:49, 5 July 2009 (UTC)

Specific the Digital object identifier, if available. --Adoniscik(t, c) 23:12, 21 July 2009 (UTC)

Insufficiently unique generated HTML id

The Wikipedia:Press_coverage_2009 page links to a lot of news articles using cite news. Problem is, the id generated by the template is not precise enough, so there are duplicates. For example, there are several articles by Noam Cohen, which produces several id="CITEREFCohen2009", when proper HTML requires the id attribute to be unique. So I can't link to the August NYT article reference on that page because Cohen wrote another article in March Wikipedia:Press_coverage_2009#CITEREFCohen2009. Is there a workaround for this? or could we just fix the template to at least put the full date (and prob. the full name and paper would be good too)? --Kai Carver (talk) 15:39, 25 August 2009 (UTC)

My workaround was to change the author from last=Cohen to author=Cohen, Noam and then to author=Cohen, N because there was already a Cohen, Noam article in January Wikipedia:Press_coverage_2009#CITEREFCohen.2C_N2009. This is of course not a satisfactory solution. --Kai Carver (talk) 15:55, 25 August 2009 (UTC)

The id is generated through {{Citation/core}}. You can manually set a unique id by adding a value for |ref=. ---— Gadget850 (Ed) talk 00:54, 26 August 2009 (UTC)
Thanks that's a better workaround: Wikipedia:Press_coverage_2009#CITEREFCohen2009-Aug-24 I fixed 5 cites on that page to make it validate and be more linkable (yes I'm that picky sometimes...). Still seems like a bug in the template, is there somewhere I could report it? --Kai Carver (talk) 07:25, 26 August 2009 (UTC)

{{Citation/core}} generates the id by:

  #if:{{{Ref|}}}
  |{{#ifeq:{{{Ref|}}}|none||id="{{anchorencode:{{{Ref|}}}}}"}}
  |{{#if:{{{Surname1|}}}{{{EditorSurname1|}}}
    |id="CITEREF{{anchorencode:{{#if:{{{Surname1|}}}
      |{{{Surname1}}}{{{Surname2|}}}{{{Surname3|}}}{{{Surname4|}}}
      |{{{EditorSurname1|}}}{{{EditorSurname2|}}}{{{EditorSurname3|}}}{{{EditorSurname4|}}}
    }}{{{Year|{{{Date|}}}}}}}}"

If |ref= is defined, then use it as the id; else build it as CITEREF + Surnames + EditorSurnames + Year (if defined) else Date.

However, although the Date is defined, the id includes only the year, This is because {{cite news}} extracts the Year from the Date:

  |Year={{{year|{{    <!-- attempt to derive year from date, if possible -->
             #if: {{{date|}}}
             |{{
                #iferror:{{#time:Y|{{{date|}}} }}
                |{{#iferror:{{#time:Y|{{{publication-date|einval}}} }}||{{#time:Y|{{{publication-date|}}} }}}}
                |{{#time:Y|{{{date|}}} }}
              }}
             |{{{publication-date|}}} <!-- last resort -->
           }}
        }}}

I don't understand the reasoning for this, other than an attempt to simplify the id. It generates the same id for any publication by authors with the same last name in the same year. This same markup is in {{cite book}} and probably others. ---— Gadget850 (Ed) talk 09:29, 26 August 2009 (UTC)

interesting, thanks! next step, change it... I wonder how we should go about this. I suppose it's a fairly major change that would break all existing ids. --Kai Carver (talk) 10:15, 28 August 2009 (UTC)
I vaguely remember this issue being brought up before someplace, but I never dug into it. Probably bring it up at WP:VPT for wider discussion. ---— Gadget850 (Ed) talk 11:28, 28 August 2009 (UTC)

publication date, newspaper

I have added "this is essential" to the "publication date" and "newspaper" (or "work") parameters because I am constantly finding articles where people have left these out. In particular, a lot of people include the access date but not the publication date, which is absurd, since if the publication date is included there is no need for an access date in most cases. Having been citing newspaper articles in references since at least 30 years before the internet was even thought of, it has been a bit of a shock for me to find that so many people don't realise that by far the most important pieces of information about a news article are (1) the date it appeared, and (2) the name of the publication. We clearly need to reflect this more clearly in the documentation. -- Alarics (talk) 20:22, 13 September 2009 (UTC)

Now done. The parameters are now divided into "essential" and "optional". Alarics (talk) 13:45, 15 September 2009 (UTC)

Request

{{editprotected}} To allow elimination of the last deprecated accessdaymonth and accessmonthday parameters, please add the line {{#if:{{{accessdaymonth|}}}{{{accessmonthday|}}}|[[Category:Cite web templates using unusual accessdate parameters|{{FULLPAGENAME}}]]}} inside the template, preferably right before the <noinclude> tag. Debresser (talk) 21:57, 13 September 2009 (UTC)

Yes check.svg DoneTheDJ (talkcontribs) 12:39, 14 September 2009 (UTC)

Accessdate

The documentation page for this template says "accessdate: Date when the item was accessed, if it was found on line. This is not needed when the article is itself clearly dated, as will normally be the case with an ordinary online newspaper article. In such a case, accessdate, if used at all, should be commented out so as not to be visible to the ordinary reader."

The addition about it not being necessary in certain cases to have an accessdate parameter, and the instruction to comment it out, is not found on any other citation template and is actually not logical in view of the fact that articles get updated sometimes days and months after the original post. In general, it seems wise to have the fullest possible range of information about any given link, including an accessdate. Practise and the documentation pages of other citation templates, certainly seem to indicate this.

I propose therefore to remove this addition. I actually did so, but was reverted, which is why I post here now. Note also that the reverting editor has added this sentence only two weeks ago without indication of consensus. And he did the same in this edit on Wikipedia:Citing sources, and when it was changed back a few days later, he reverted there also.

In all fairness I have to say that the subject has been discussed there at length. Nevertheless, as stated above, there is no precedent for the inclusion of this in the documentation of any citation template, nor do editors practise this. Debresser (talk) 00:11, 14 September 2009 (UTC)

Strongly disagree with hiding the access date. Access date does not refer to when the article was written, but when the URL was accessed, same as with cite web, and it should not be removed. And as it is a standard element of the majority of citing styles, I see no valid reason to hide it either.-- AnmaFinotera (talk ·contribs) 00:15, 14 September 2009 (UTC)
I have no idea what consensus that edit refers to. No discussion on the subject has ever gelled into any consensus for removal. The other changes are fine, but I recommend that the accessdate change be reverted. We might have a section on how to customize the viewing of references, such as hiding accessdate on an opt-in basis, which is simple. ---— Gadget850 (Ed) talk 01:09, 14 September 2009 (UTC)

Templates should adhere to the official guidelines. The one applicable here is Wikipedia:Citing sources. The access date issue has been discussed at great length on the guideline's discussion page, now to be found in the archive as Debresser helpfully pointed out. If, after reading that discussion, you still think you have any new arguments feel free to add them on the current discussion page. Here is not the right place. Thanks, --EnOreg (talk) 09:20, 14 September 2009 (UTC)

Caution re: Stability of sources

I don't want unnecessarily to excite the protagonists in this area, but having read various threads about accessdate, I suspect some editors have a mistaken impression of the "stability" of online news. Here are two examples of instability that I've come across recently:

  • The BBC used the same URL for their "US election - delegate totals" page (here) for at least three months (I tracked it from 6 March 2008 and it was last updated on 4 June 2008). The statistics (text and graphics) for Obama vs Clinton changed massively during the period.
  • The Sunday Times publishes an annual UK University Guide (here). The page is updated every year but the URL doesn't change. The result is that the historical "League table rankings" in University articles contain citations intended for earlier years that actually point to the current year's data.

Such examples should be included in the {{Cite news}} and perhaps Citing sources pages. In each case a publication date is ostensibly sufficient but an accessdate will more clearly highlight that the citation may now be out of date. The template documentation should address this. - Pointillist (talk) 01:24, 14 September 2009 (UTC)

The guideline appreciates this point but avoids visual clutter by asking for access dates in comments. If you disagree please take this discussion there. Thanks, --EnOreg (talk) 09:20, 14 September 2009 (UTC)
Calling "Retrieved on 12-24-2008." visual clutter is... peculiar. Debresser (talk) 09:56, 14 September 2009 (UTC)
Pointillist, your examples don't support the idea that there should be access dates in news cites as well as publication dates. In the first case, every BBC news online page has a "last updated" date at the top, and that is what we put as "publication date". If you go back to it later and find it has been updated, the dates won't match, so including the publication date in this instance serves the purpose claimed for "access date". In the Sunday Times case, that page clearly isn't a news article: it doesn't bear a date at all, so of course it is to be treated as a web cite and not a news cite (presumably it was actually published in printed form as well, but we can't cite that unless we happen to have the printed version to hand), so yes, an access date is the only one you can give in that instance. -- Alarics (talk) 14:59, 14 September 2009 (UTC)
Agree with Alarics on these two cases. I've seen a couple of even weirder cases, where a completely unrelated newspaper story replaces a previous one at the same URL. But again the change in publication date (and title) will show readers that there is a mismatch. And as for "Retrieved on 12-24-2008.", yes it is visual clutter when it closely follows "(11-23-2005)" or however the publication date is formatted. Also, the eye tends to gravitate towards what's at the end of the cite next to whitespace, the retrieval date, instead of the publication date which is somewhere in the middle of the cite. That's unfortunate, as the publication date is the more important of the two. Wasted Time R (talk) 11:34, 15 September 2009 (UTC)
I disagree with both your arguments. As for the "visual clutter" argument. That is a subjective argument. I for one disagree, and do not percieve this as clutter. Probably because the words "Retrieved on" separate the two dates. As for the case of a change in publication date. Not all newsarticles have a publication date, not all articles have a clearly visible publication date, but the main argument is that there are many references around here on Wikipedia with an accessdate but without a publication date. Debresser (talk) 12:06, 15 September 2009 (UTC)
But don't you see, that is exactly why promoting the idea of accessdate is such a bad thing in the case of news citations. Many casual editors don't understand about news sources, and they see that there is this space called "accessdate" so they fill it in and think they have done all that is needed. What they should have done is write the publication date (and the name of the publication, and the title of the article). As you say, the result is that there are lots of articles with hopelessly inadequate references. What to do about this is not to push the use of access dates for such references, but to do everything to encourage users to cite the three things that actually matter. When one comes across such a badly-done reference one has to change it so that it gives the publication date and the other essential information (name of publication and title of article) and comment out the access date.
If that is your argument, then update the documentation to stress the importance of the date paramater, along with stressing the optional character of the accessdate parameter. But in any case, your argument can not used to argue that the accessdate should be placed in remark tags. Debresser (talk) 12:34, 15 September 2009 (UTC)
You say "Not all newsarticles have a publication date, not all articles have a clearly visible publication date". I would say that if it doesn't have a clearly visible publication date, it's not a news article. -- Alarics (talk) 12:20, 15 September 2009 (UTC)
That is a matter of academic interest. :) Debresser (talk) 12:28, 15 September 2009 (UTC)
No, it isn't. It means that in such a case, what I am saying about news citations doesn't apply, and that your observation is therefore irrelevant. Alarics (talk) 12:31, 15 September 2009 (UTC)
You do understand that your statement "I would say that if it doesn't have a clearly visible publication date, it's not a news article" is ludicrous? Debresser (talk) 12:36, 15 September 2009 (UTC)
Not ludicrous at all. Having a date is part of the definition of a news item. I have said all along that my wish to comment out the access date does not apply in cases where the linked page does not have a date of its own. As I keep saying, I am talking only about those cases (which include all proper news items) where the article linked to has a clearly visible publication date, and therefore the access date is superfluous. Your mention of cases where there is NOT a visible publication date is therefore not relevant to what I am talking about. -- Alarics (talk) 12:43, 15 September 2009 (UTC)
But needs mentioning after all. Because others might (understatement) disagree with your assertation that a webarticle without a (clearly indicated) date is not a newsarticle fit to use {{Cite news}} on. Debresser (talk) 12:50, 15 September 2009 (UTC)

I'm not saying "not a news article fit to use {{Cite news}} on". I am saying "not a news article for the purposes of the discussion we are having", because I am talking only about sources that DO have a clearly visible publication date. Where they don't, nobody is objecting to the inclusion of the access date. -- Alarics (talk) 13:02, 15 September 2009 (UTC)

Debresser said: "If that is your argument, then update the documentation to stress the importance of the date paramater, along with stressing the optional character of the accessdate parameter." OK, I have now done this, dividing the list of parameters into "required" and "optional". This incidentally brings the documentation for this template into line with that for the "cite web" template.-- Alarics (talk) 13:06, 15 September 2009 (UTC)
I have changed a few things. Hope your fine with them. The most important of them is that in the case of {{Cite news}} these parameters are actually not required by the code and do not result in error messages, as opposed to {{Cite web}}. Debresser (talk) 13:17, 15 September 2009 (UTC)
Fine. I like "not required, just essential". It reminds me of the fellow who said that the situation was desperate but not serious. Alarics (talk) 13:23, 15 September 2009 (UTC)
Debresser said: "But in any case, your argument can not used to argue that the accessdate should be placed in remark tags." The argument for putting it in remark tags, if we have to have it all, was set out and debated at great length here. -- Alarics (talk) 13:13, 15 September 2009 (UTC)
So? Debresser (talk) 13:17, 15 September 2009 (UTC)
So please stop behaving as if the argument has not been made. Alarics (talk) 13:24, 15 September 2009 (UTC)
And you please stop ignoring the fact that a majority of editors here on the talkpage of template Cite news summarily disagree with that. In fact, all those who hadn't been a part of that discussion. You can't discuss things somewhere else, and then try to force editors here to agree and behave accordingly. Debresser (talk) 13:31, 15 September 2009 (UTC)
But User:EnOreg has already pointed out, further up this page, that the template must follow the guidelines for citation generally, and that the right place to discuss them is Wikipedia:Citing sources, not here. So according to him, what you have just said is wrong. And incidentally he disagrees with you on the substantive issue as well. The "majority of editors here on the talkpage", apart from yourself, comes to a grand total of two! Alarics (talk) 13:41, 15 September 2009 (UTC)
I count four (User:AnmaFinotera, User:Gadget850, User:Pointillist), and me. Against zero. Two, if I were to include editors who participated in the discussion on Wikipedia:Citing sources, which I don't. Because according to those same Wikipedia rules, the users of a specific page may decide to deviate from a guideline (not policy). Debresser (talk) 14:10, 15 September 2009 (UTC)
May I also quote Wikipedia:Policies and guidelines: "Guidelines are primarily advisory", "Whether a policy or guideline is an accurate description of best practice is determined by the community through consensus", and Wikipedia:List of guidelines: "Unlike policies, guidelines are usually more flexible and more likely to have exceptions and could be changed and improved more easily". Debresser (talk) 14:17, 15 September 2009 (UTC)
You seem to have forgotten to count me and User:EnOreg. Also on my side of the argument are User:Wasted Time R, User:Smith609, User:WhatamIdoing, and, at least up to a point, User:Rich Farmbrough and User:LouScheffer. You keep trying to compare a very brief discussion on this page, attached to a project page which attracts about 80 hits a day, with a months-long discussion at Wikipedia talk:Citing sources, attached to a project page which attracts about 3,500 hits a day. I am not interested in having a separate discussion that only relates to Template:Cite news, I am interested in what is the policy on this issue for all news citations, with or without a citation template. Alarics (talk) 16:16, 15 September 2009 (UTC)
I don't see these people here. Debresser (talk) 20:05, 15 September 2009 (UTC)
Let's not be childish, shall we? As you know perfectly well, they are either here, or at Wikipedia talk:Citing sources, or on your own talk page. It is all the same discussion, unfortunately split into at least three different places. Alarics (talk) 20:11, 15 September 2009 (UTC)
My point is that discussion about the Cite news template should take place here. If this were a general rule, it could be decided centrally. But this is a case of people deciding there, what we should be doing here. Sorry guys, but we here disagree. Debresser (talk) 20:31, 15 September 2009 (UTC)

I'd rather not be counted in either camp. The only point of view I am arguing is that the citation advice (here and at "citing sources") should be clearer, with better examples and guidelines for what to do in special cases. - Pointillist (talk) 16:22, 15 September 2009 (UTC)

  • As Alarics (talk · contribs) explained at the beginning of this section, material like the [Sunday Times University Guide http://extras.timesonline.co.uk/stug/universityguide.php] is better referenced with {{cite web}} rather than {{cite news}}. I think the guidelines should mention the possibility that cite news might not be the right template. It could go on to explain that other material on news websites may not be a reliable source for Wikipedia's purposes. I've seen discussion of Con Coughlin and Peter Hitchen's blogs in this context.
  • Newspaper/magazine/periodical/work is listed as one of the three Essential parameters in {{cite news}}. But, except for "foreign language news" none of these is used in any of the Examples.

In respect of dates:

  • All the {{cite news}} examples that use date also use accessdate, despite claims that accessdate is unnecessary in this situation.
  • The guidelines don't explain how to deal with multiple versions of a story. For example, this guardian story was first published online on 14 Sept, then in print on 15 Sept, then updated online on 15 Sept and is still "dated" 14 Sept online. IMO this situation is a good example of why accessdate is useful, in that the online story has been updated after the print edition went to press.
  • More prominence should be given to the idea that some types of online statistics are expected to change even though the URL doesn't. Examples include box office totals, votes, university rankings etc. For this reason editors need to recognize when archiving (e.g. with WebCite) might be appropriate. Unfortunately timesonline.co.uk, telegraph.co.uk and ft.com have "noarchive" settings, independent.co.uk doesn't technically restrict archiving, but has a notice saying it is not permitted (here). BBC News, guardian.co.uk and boxofficemojo.com seem to tolerate it. I don't know about US sources. - Pointillist (talk) 16:22, 15 September 2009 (UTC)
  • "I think the guidelines should mention the possibility that cite news might not be the right template." I wouldn't dissent from that, although the opposite problem (cite web being used when it should be cite news) seems to be vastly more prevalent.
  • "Other material on news websites may not be a reliable source." Yes, never mind Peter Hitchens's blog (a reliable source only on the subject of what Peter Hitchens thinks), even in The New York Times we surely draw a distinction between news reports and op-ed pieces. If that isn't sufficiently clear (and of course this is an issue not just for citation templates but for citations generally), it should be made so.
  • "Newspaper/magazine/periodical/work is listed as one of the three Essential parameters in {{cite news}}. But, except for "foreign language news" none of these is used in any of the Examples." You're right. In fact, every single one of those examples is a dud in one respect or another. Shall I fix them or will you?
  • "All the {{cite news}} examples that use date also use accessdate, despite claims that accessdate is unnecessary in this situation". Yes, but those are the claims that User:Debresser is resisting, so for the moment they are made at Wikipedia:Citing sources but not here.
  • "IMO this situation is a good example of why accessdate is useful, in that the online story has been updated after the print edition went to press." Granted, the behaviour of The Guardian in this particular case is a bit odd, but they do have a little "history of article" box you can click on to see if and when the piece was updated. In principle, if you are citing what is on the website rather than what appeared in the newspaper, you can cite the "latest update" date as the publication date, as with the BBC; it still isn't clear to me how an access date helps. I think the same applies to those newspaper websites (the Daily Mail is one, and the Straits Times is another) which have now turned into something separate from the paper whose name they bear, and have substantially different content, and in these cases I think we should perhaps give the newspaper title as Mail Online or Straits Times Interactive, and so on.
  • Archiving: How often in practice is this really a problem? The Times, Telegraph, BBC and Guardian all have online archives going back 10 or 15 years; many other papers can be found in things like Newsbank (if you have the right library card, of course not for citing as a web link). In the case of Times and Guardian you can go back to the 18th century if you have the right library card. I'm not sure that I think it would be worth as a general rule taking it upon oneself to archive stuff from mainstream newspapers separately from all that, just for the odd differences of text between versions. After all, the problem of different editions of a newspaper has always been with us, and I haven't noticed that anybody used to bother about that when citing them. You just cite whatever edition you find in the library. Likewise, with online material, you cite the version that ends up in the online archive being built up by the newspaper itself or on services like Newsbank.
-- Alarics (talk) 18:17, 15 September 2009 (UTC)
Thanks for your feedback, Alarics. I've started a fresh sub-section for us all to debate the next steps. - Pointillist (talk) 23:09, 15 September 2009 (UTC)
I'll second the motions that news sources without clear dates should be handled with {{cite web}} not {{cite news}}. And I agree that certain news sources are problematic about constantly evolving content for a certain period (CNN does this with their election stories), and so accessdate is appropriate for them (although strictly speaking you'd also need time-of-day to get it fully pinned down). But these are very much minority cases; the large majority of news sources are dated and fixed in content over time, and it's for those that I don't believe we need an accessdate. As for visual clutter, I agree that this is subjective and results will vary among different editors and different readers, but I definitely see accessdates as clutter. Wasted Time R (talk) 00:43, 16 September 2009 (UTC)
I agree the CNN case is very much a minority one, and I think it links in to a deeper question, which is the issue of "short-term transience" vs. "permanent information for posterity" as it relates to the difference between a newspaper and an encyclopaedia. I don't think WP should be attempting to keep up with breaking news stories; we are not supposed to be journalists; that is what Wikinews (and zillions of other on-line news services) is for. So an evolving election results story shouldn't concern us; we wait till all the results are in and have been analysed, and then record that for posterity. It is the final, stable version we want. I'd say that if the story is still developing, it is too early for WP to include the latest detail, though of course I realise that some broad-brush news stories (e.g. Palestine) go on for ever. (Likewise, I suspect weeklies are probably generally better than dailies as reliable organs of record: they are produced in slightly less of a hurry.) We should record what remains when the dust has settled, and by that time the source we cite should be stable anyway, much like the last edition of the newspaper being sent to the library and eventually getting microfilmed, where it will exist permanently long after web pages that might or might not have exactly the same text have come and gone. Viewed in this light, I really feel that the whole "access date" business becomes even more pointless, as it applies to news sources. Alarics (talk) 11:42, 16 September 2009 (UTC)

Improving guidelines for editors who try to cite their sources

The central objective here should be to ensure that editors who make the effort to cite their sources are given helpful advice about how to do so, and feel rewarded for their extra effort rather than penalised for technical errors. If you agree to that I see these areas for action, in increasing order (IMO) of difficulty - Pointillist (talk) 23:09, 15 September 2009 (UTC)

  • Fixing dud examples. I fully support this but if you (Alarics) care to do it you would probably improve them more than I can. Ideally examples should be based on real stories with real URLs. If you need a public-domain no-registration example of an historical news story, File:Tussauds story The Times Sat July 20 1889.jpg might serve.
  • Explaining how to reference statistics that get updated. I will try to draft something over the next few days, including a brief mention of appropriate use of archives. I'll submit the draft here for feedback.
  • Advising when to use {{cite web}} vs {{cite news}}. If you (Alarics) want to maintain a separation between them I have no idea how to present a general rule that editors can follow. The boundaries are often very hard to distinguish but at least we should explain that much of what is printed in a newspaper (and/or appears on an online news site) isn't news. Perhaps you could propose a text for this?
  • Suggesting how to decide whether a source is reliable. It would be helpful if we could go a step further and distinguish items in newspapers/on news websites that are expected to be reliable sources (like surveys) vs items that aren't (like blogs perhaps). I'd be happy to collaborate on this, but I expect it'll be fiercely controversial. I already had a run-in with the user who claims to be Peter Hitchens when I questioned whether his blog had to accord to the same editorial standards as a news story (and I was unexpectedly impressed by his apparent good faith in that context though we disagreed on the detail).
  • Getting a consensus on accessdate parameters. I see your point (e.g. what about the differences between print editions of newspapers in a single night) but (a) I've experienced situations where the accessdate has alerted me to significant problems with citations; (b) David Debresser is generally someone we can do business with so I'd like to include him in any consensus; and (c) as I'm sure you'll agree, a newspaper is under no obligation to preserve an archive of its mistakes, so the "last updated" date isn't a guarantee of anything. If I were a lawyer advising a newspaper I would certainly propose that they offer their own archive while denying archive rights to third parties. Perhaps we could put this to one side for a while?
I'll be happy to do business with any of you. :) Now for a short summary of relevant opinions of mine. 1. I agree that most news sources have a clearly indicated date. As they should have. 2. I am sure it will be impossible to define clear and logical instructions for when to use {{Cite web}} or {{Cite news}} in such a way that editors will stick to them significantly more than they do now. We may try, but should not try too hard. 3. I do not see the accessdate parameter as clutter and oppose hiding it and surely would not recommend to omit it altogether. Debresser (talk) 06:46, 16 September 2009 (UTC)
I am bit busy but I will try to do what Pointillist asks of me in the next few days. Alarics (talk) 11:45, 16 September 2009 (UTC)
I have now put in some new examples as promised, and hope these fit the bill. -- Alarics (talk) 13:24, 25 September 2009 (UTC)
Thanks, the examples are vastly improved, which puts the ball in my court to say something about statistics-that-change but have a static URL. Next week, I hope. - Pointillist (talk) 23:03, 25 September 2009 (UTC)
Template's documentation says "This template is used to cite sources in Wikipedia. It is specifically for news sources." Perhaps add "Online-only news sources should be cited using {{Cite web}}"? --EarthFurst (talk) 06:36, 25 October 2009 (UTC)
I don't understand why you say that. If it is a news item, what is wrong with using {{Cite news}} whether it is online-only, print-only, or both? -- Alarics (talk) 06:56, 25 October 2009 (UTC)

Please add Edition template argument

Many newspapers have multiple editions that are printed on a given day. I think it would be useful to include edition in the template. WilliamKF (talk) 03:12, 11 October 2009 (UTC)

Just so long as you are aware that many newspapers don't give that information, or even reveal to the public that different editions exist. -- Alarics (talk) 17:56, 11 October 2009 (UTC)
Fine, it should be optional, but for example, The New York Times does note it prominently. WilliamKF (talk) 22:29, 12 October 2009 (UTC)

Proliferation of error messages

What's going on this template? Most instances of the template are now displaying the words "Archived from the original" followed by an error message that says (in red) "Error: You must specify the date the archive was made using the |archivedate= parameter." This is happening for citations that are not archived at all, but use the original citation location (for an example, see the ref list on W. W. Herenton).

Links to the original location of a news article should not be getting flagged as archive links, nor should they be flagged as needing archive dates. --Orlady (talk) 16:10, 19 October 2009 (UTC)

Not anymore, apparently someone already purged it. More details on my talk page, I was too slow/made a mistake with a change of it, sorry. :\ The job queue will take care of it and it should all be cleaned up again soon. Amalthea 16:15, 19 October 2009 (UTC)
It's all fine now. It just took a while to switch to all the new templates. Debresser (talk) 16:24, 19 October 2009 (UTC)

Capitalised documentation page

See Template_talk:Cite_web#Upper_case_form_of_template_name. Debresser (talk) 06:38, 21 October 2009 (UTC)

archiveurl without original url old papers

I know this was discussed here already last year, but I have a question about Citing archives of an extremely old paper. There are a number of efforts underway to make available archives of out of copyright national newspapers, where the scanned copy and OCR text is stored. An example of how I would have (IMHO)thought best to reference these archives is:

{{cite news |title=Volume XVII, Issue 27 |newspaper=New Zealand Tablet |date=25 Oct 1889 |accessdate=13 Nov 2009 |location=Dunedin, New Zealand |page=18 |archiveurl=http://paperspast.natlib.govt.nz/cgi-bin/paperspast?a=d&d=NZT18891025.2.25}}

Of course it gives me this error:

"Volume XVII, Issue 27". New Zealand Tablet (Dunedin, New Zealand). 25 Oct 1889. p. 18. Archived from the original|archiveurl= requires |url= (help) on |archiveurl= requires |archivedate= (help). Retrieved 13 Nov 2009. 


  • Problem one is that you can't use an archiveurl without already having URL. In this case I'm using the URL for the actual government archive of the old paper, which seems valid. Since a paper from 1889 would not have a URL. Is it better to use the url= for this sort of archived article, and ignore the archiveurl, since it's intended for the Internet Archive etc.?
  • Problem two is that there is no archivedate one can specify, yet it would be mandatory for archiveurl being used. I can't determine the archive date as it's an entire archive project. Surely archivedate could be optional? Lantrix ::Talk::Contrib:: 08:39, 13 November 2009 (UTC)
Yes, I have done this once or twice to cite articles in the free collection of very old (pre-1920 or whatever it is) New York Times items. Just use url= for the URL. There is no need for archiveurl or archivedate. They have no meaning in this context. -- Alarics (talk) 10:14, 13 November 2009 (UTC)
Exactly. The 'archiveurl' field is used to link to a third-party archive of a web page that existed previously and no longer does. (Internet Achive and Webcite are the typical archives used.) In this context, what is being archived is the old newspaper article, not a web page. It's an entirely different kind of "archive" from what the 'archiveurl' field was designed to handle. --RL0919 (talk) 14:09, 13 November 2009 (UTC)
Ok that makes sense. When citing old news sources in newspaper archives, I'll use url= for the actual url like this:
{{cite news |title=Volume XVII, Issue 27 |newspaper=New Zealand Tablet |date=25 Oct 1889 |url=http://paperspast.natlib.govt.nz/cgi-bin/paperspast?a=d&d=NZT18891025.2.25 |accessdate=13 Nov 2009 |location=Dunedin, New Zealand |page=18}}
Thanks for the response. Lantrix ::Talk::Contrib:: 02:11, 14 November 2009 (UTC)
Just one point about your example. "title=" should be the title (headline) of the article, not the issue number. -- Alarics (talk) 09:01, 14 November 2009 (UTC)
In this case there was no article title for this paper reference, so I used the issue number. Otherwise, this is understood as title is a mandatory option. Lantrix ::Talk::Contrib:: 07:24, 15 November 2009 (UTC)
Ah, I see. In such a case I would be inclined to put "(No headline)" followed by the first few words of the first sentence of the article. -- Alarics (talk) 09:09, 15 November 2009 (UTC)
That is because "archiveurl" is to archive a website. This is stated specfically in the documentation of all templates that allow this parameter. For an online archive of a paper source you should just use "url". Debresser (talk) 16:21, 14 November 2009 (UTC)

Still confusion around parameters

The parameters and instructions for {{cite news}} have been somewhat revised since I last took a close look, but there are still some very confusing points:

  • the instructions say that the "newspaper" parameter (or its aliases "magazine", "periodical" or "work") is "essential", yet only 2 of the 8 examples given use that parameter
  • it's still completely unclear what parameter to use for radio and television broadcast news sources (including their web sites) that don't get italicised, such as CNN, National Public Radio, BBC News, France 24, etc.

On the latter point, I've been using "publisher" for these, but the instructions are adamant that "publisher" is "not to be used for the name of the news source itself". So maybe a "broadcaster" parameter should be added, which does not get italicised, and the instructions modified to say that one or the other of "newspaper" or "broadcaster" should be used. Wasted Time R (talk) 12:19, 18 September 2009 (UTC)

Yes, I have promised to sort out the examples, which, as you say, don't match the instructions.
When you refer to broadcasters here, do you mean actual radio or TV broadcasts, or those broadcasters' news websites? They are two different things; I suspect most such citations are actually of the websites rather than to radio or TV programmes themselves, although I agree there are exceptions (such as where a programme transcript -- or I suppose nowadays a video clip excerpt -- has been put on line). To cite an item on the BBC News website, I call it BBC News Online, in italics, which seems to me to be the equivalent of the name of a newspaper. This differentiates it from citing a published transcript of something somebody said on BBC radio, and from an individual programme page on the BBC website, which includes a vast amount of stuff other than news. In the latter case, I'd put the programme name in italics, and BBC as publisher (and possibly use "cite web" rather than "cite news"). (I know the BBC itself doesn't call its website "Online" any more, but it used to do.) -- Alarics (talk) 12:53, 18 September 2009 (UTC)
I agree that most of the citing is done to broadcasters' websites, but I don't think yours is the right answer. I never see CNN referred to in italics, in any context, nor NPR or BBC or the others. The online arm of newspapers is still part of the newspaper, and stays in italics, while the online arm of broadcasters is still a broadcaster, and should stay in regular font. Wasted Time R (talk) 13:02, 18 September 2009 (UTC)
Where does it say that? -- Alarics (talk) 13:05, 18 September 2009 (UTC)
You see things written like, "The status of the official was unclear; The New York Times reported she would resign, while ABC News said she would stay on." That formatting holds true whether the Times first reported this in paper form or online, or whether ABC News first reported this on the air or online. Wasted Time R (talk) 13:08, 18 September 2009 (UTC)

We often see editors adding '' themselves. Instead, the use of italics should be a feature of the template only. Debresser (talk) 13:32, 18 September 2009 (UTC)

That's true. That would argue for adding a "broadcaster" parameter and letting the template decide whether to italicise it or not. Wasted Time R (talk) 13:51, 18 September 2009 (UTC)
Wasted Time R, the example you give doesn't seem relevant here; that is from a piece of prose, not a citation footnote. I still think I am right to italicise BBC News Online in a citation. It is something a good deal more specific than "the BBC", and it seems to me exactly equivalent to the name of an online newspaper. And when you say "The online arm of newspapers is still part of the newspaper, and stays in italics, while the online arm of broadcasters is still a broadcaster", I think that is ceasing to be meaningful, with all the multi-platform media convergence that's going on. A growing number of newspapers' websites are taking on a life of their own, with different content from what is in the paper; while the BBC increasingly behaves, and thinks of itself, as a media conglomerate and not just a broadcaster. I don't think the Guardian website and the BBC News website are materially different for our purposes. They are both news websites, doing to some extent different things from their parent bodies. There is probably more of a difficulty for us in the Guardian case, because ideally we ought to be distinguishing between guardian.co.uk (for a web-only item) and The Guardian (for something which has also appeared in the paper). The world is changing under our feet. I am not sure anybody has really thought all this through. -- Alarics (talk) 15:08, 18 September 2009 (UTC)

Citing a specific version of a story

Just continuing that thought, you can have a story that is first reported on guardian.co.uk, then appears in The Guardian the following morning, and is then updated on guardian.co.uk. In such a case it could be argued that we should only cite this as a print story if we have physically seen the printed version, and if we do cite it as a print story, we should only add the web URL if we are sure that the online text exactly matches the printed version – and then we'll need an accessdate/time too, in case the online version changes again. On that basis it seems better to ignore the print version and cite only the web one, using the "last updated" date as the date= parameter even if this is later than the date the story appeared in print. What do you think? - 17:00, 18 September 2009 (UTC)
No need to have physically seen the print version, because each Guardian web article now has a thing you can click called "Article history", which tells you if it also appeared in the printed paper, and if it has been updated since then (not common, I think - I can't find an example of it). However, other papers' websites don't provide this facility, and I have come across some (e.g. Daily Mail) where the story has been heavily rewritten in the web version (or maybe it is the other way round). Then again, stories can get heavily rewritten between editions of a dead-tree newspaper, so maybe we are worrying too much. It may well be that the substance is unlikely to be significantly different so our citation still makes sense, though I can see there might be a problem in the case of a direct quote from somebody which then gets edited out. -- Alarics (talk) 17:19, 18 September 2009 (UTC)
Actually, my example was based on a Guardian article history. My point is that a newspaper or news website is under no obligation to ensure that earlier versions of stories are retrievable—or even detectable—after they have been updated. Wikipedia editors should be aware of such risks when they cite online sources, particularly when stories are still breaking. - Pointillist (talk) 21:10, 18 September 2009 (UTC)
The New York Times has also had one archival version of a story. In the past, it was the one from the last edition published for the day, and it was that that went into their clip files, into the microfilm that got sold to libraries, etc. Those same archival versions are the ones that are now available online in their repository. I believe more recent articles still work the same way, except that now any future corrections published are also associated with the article. I think any web-only or web-still-updated content is distinguished as a blog or some other special type, so as not to confuse them with regular articles published in the paper. Wasted Time R (talk) 21:19, 18 September 2009 (UTC)
Ye-ee-es, but the central point is that while news is "breaking" a WP editor might cite an allegation that is later deleted by a news source that doesn't allow archiving. I'm concerned that we should resolve citation issues as soon as possible, rather than bequeathing them to the next generation of editors (who will be fewer in number and less motivated, because their work will increasingly be janitorial rather than inspirational). - Pointillist (talk) 21:56, 18 September 2009 (UTC)
I fully agree that trying to write history as it happens is very hazardous, with cites to changing online news stories being just one of the perils. It would really be better if everyone waited a few days to let things settle in before adding events to WP, but of course it doesn't work that way. Wasted Time R (talk) 22:23, 18 September 2009 (UTC)
Absolutely spot on. WP shouldn't be trying to keep up with breaking news stories at all. That's for journalism to do, not an encyclopaedia. Those who wish to do that should be doing it at Wikinews, not here. Wait till the dust settles, and record the final version. -- Alarics (talk) 22:29, 18 September 2009 (UTC)
Also, let's not lose sight of the purpose of citing sources in the first place: it is so that a reader can verify the statement or quotation being made in the WP article. If the information exists in a printed newspaper which somebody can (in theory at least) find in a library, that's fine, and of course it doesn't have to be on the internet. If it exists on a web page which remains in existence in a settled (final) version, that's fine too. If it existed fleetingly on the web but can no longer be found, we almost certainly shouldn't be citing it anyway. -- Alarics (talk) 22:38, 18 September 2009 (UTC)
I agree about telling editors not to try to deal with breaking news but how can anyone be sure that an online news page will "remain in existence in a settled (final) version". From a libel point of view, nothing online is final until everyone involved is dead which is probably not practical for Wikipedia. What I'm looking for is much better documentation – a "step change" – that illustrates such issues around citing news sources so editors "get it" and can police themselves. How can we achieve this? - Pointillist (talk) 00:31, 19 September 2009 (UTC)

Making a distinction between citing print and broadcast news sources

(od) Anyway, my primary suggestion is simply that the "newspaper", "magazine", "periodical" list be expanded to include more than print news sources, meaning adding "broadcaster". If you also want to add "onlinenewssite" and "blog" and whatever, that's okay too. I will continue to argue strongly that "broadcaster" shouldn't be rendered in italics, but that's a separate decision. Wasted Time R (talk) 21:30, 18 September 2009 (UTC)

Did any progress get made on this issue? I've always used "publisher" for BBC News but recently this has been questioned so I came here to check whether the issue had been discussed and found this, with no apparent resolution. Cordless Larry (talk) 15:18, 16 January 2010 (UTC)
It seems not. I still use "work" for BBC News Online and anything equivalent. The "publisher" parameter is meant for a different purpose altogether. The fact that the BBC is primarily a broadcaster seems to me irrelevant when what we are citing is not a broadcast but a BBC web page which is equivalent to news pages on the website of a newspaper. -- Alarics (talk) 15:45, 16 January 2010 (UTC)
WP:ITALICS states that italics should be used for periodicals, including newspapers, but I don't see anything there that suggests using them for websites. Cordless Larry (talk)
Indeed, we shouldn't use italics for the name of an ordinary non-news website, but when it is a news website I think we should because it is a news medium, by analogy with printed news media. (If it isn't a news website, it shouldn't be cited with "cite news" in any case.) -- Alarics (talk) 16:06, 16 January 2010 (UTC)
Which of the uses at WP:ITALICS would cover this though? Cordless Larry (talk) 16:08, 16 January 2010 (UTC)
Broadcast news should be cited using {{cite episode}}. In the case of the online websites of those broadcasters, {{cite web}} should be used. {{cite news}} should be used only for the printed news list. -- AnmaFinotera (talk · contribs) 16:30, 16 January 2010 (UTC)
Well {{Cite web}} states that "{{Cite news}} can also be used when citing a news source", but regardless of that, the italics issue still applies because cite web also italicises the "work" field. Cordless Larry (talk) 18:25, 16 January 2010 (UTC)
With cite web, you'd use the publisher field for this type of item, which is not italicized. -- AnmaFinotera (talk · contribs) 18:39, 16 January 2010 (UTC)
No, the publisher field is not meant for that, it is to enable you to give the parent company of a newspaper that nobody would otherwise have heard of, certainly not for the BBC. -- Alarics (talk) 23:06, 16 January 2010 (UTC)
{{Cite news}} is the preferred way to cite a news source such as BBC News or the New York Times, regardless of whether citing the online or the traditional format. There's nothing wrong with italicizing BBC News; some citations italicize it (as it's a work), and some people don't (as it's an organization), and both ways are fine. Eubulides (talk) 23:07, 16 January 2010 (UTC)
Quite so. In my view it is illogical to say you should italicise the name of an online news service that happens to be run by a newspaper, and not one that happens to be run by a broadcaster. -- Alarics (talk) 23:14, 16 January 2010 (UTC)
I agree that "both ways are fine", but the problem is that {{Cite news}} states that the "publisher" field is "not to be used for the name of the news source itself", which makes it hard not to use italics. Cordless Larry (talk) 11:11, 17 January 2010 (UTC)
Look for a copyright notice. It may be at page head; page foot; on the site's home page; or reached via a "About us" or "Contact us" link. If it says "© Xxxxx 2010", then the publisher is Xxxxx. The work is the overall title of the site. So taking this BBC news item, the work is "BBC News" and the publisher is "BBC". --Redrose64 (talk) 15:55, 17 January 2010 (UTC)
I understand that, but it doesn't help us with the issue of whether "BBC News" should be in italics or not, and if the latter then how to achieve this with the template. Cordless Larry (talk) 16:03, 17 January 2010 (UTC)
I don't agree that broadcasts should be cited using {{cite episode}} and {{cite web}}. They are news agencies just like print media. They have reporters and generate stories just like print. They have investigative reporters just like print media. To draw a distinction since the material is generally first reported on air and then the copy is added to a web cite does not make them any less of a news agency. Vegaswikian (talk) 21:42, 17 January 2010 (UTC)
Vegaswikian is right. -- Alarics (talk) 22:46, 17 January 2010 (UTC)
Thumperward (Chris Cunningham) proposed having specific templates for each of the main publishers (e.g. "cite BBC" etc) a year or so ago. The idea didn't take off at the time ...but perhaps something along those lines would improve the consistency of news citations and reduce the learning curve for editors. - Pointillist (talk) 23:30, 17 January 2010 (UTC)

add "page" as a common parameter?

Shouldn't "page" be included as a common parameter? Also, what about including "column" under all parameters? Шизомби (Sz) (talk) 16:39, 17 January 2010 (UTC)

If you need to ref a specific column on a page, you can use the |at= parameter. Note that this cannot be used in conjunction with either |page= or |pages=. So, you could put
|at=p. 2, col. 3
The |at= parameter is not documented for {{cite news}}, but it functions in almost the same way when used in {{cite journal}} and {{citation}}, where it is documented. --Redrose64 (talk) 17:34, 17 January 2010 (UTC)
I've documented |at= now. --Redrose64 (talk) 18:09, 17 January 2010 (UTC)

Relevant discussion

There is a discussion occurring at Wikipedia:Centralized discussion/Wikipedia Citation Style. Your participation would be appreciated.
V = I * R (talk to Ohms law) 23:29, 18 January 2010 (UTC)

Newspaper section, page and column information

This template doesn't seem to be friendly for section and column information. Many newspapers use letters for the section, so those can be input as page= B-4, column 3 which yields p. B-4, column 3. However, the New York Times uses numbers for the sections. The current template forces that to read "p. 2-4, column 3", which is potential ambiguous and is not as easily understandable to a reader, forces one to put the section number after the page number (p. 2, section 4, column 3) which is counter-intuitive. How hard would it be to introduce section= and column= parameters within this template? Then output could read section 4, p. 2, column 3. Thanks. --Bejnar (talk) 20:54, 22 May 2009 (UTC)

You can use the |at= parameter for this. It will show in the same place that the page number normally appears. It cannot be used in conjunction with either |page= or |pages=, so you would put something like |at=section 2, p. 4, col. 3. --Redrose64 (talk) 14:18, 25 February 2010 (UTC)

2 parameters request

Could we add a "subscription" and "registration" parameter?Smallman12q (talk) 00:12, 28 February 2010 (UTC)

There are {{registration required}} and {{subscription required}} templates that can be added after the citation template. ---— Gadget850 (Ed) talk 00:49, 28 February 2010 (UTC)

Date

For date, do we use date created, or date last modified/updated?Smallman12q (talk) 01:34, 7 March 2010 (UTC)

That's up to you. If you want a template that can specify both dates, you can use {{vcite news}} instead; it has both |date= (for the original date) and |update= (for modified/updated date). Eubulides (talk) 01:47, 7 March 2010 (UTC)

What should be placed in the quote parameter?

What is the consensus of what quote parameters should be used for?

I started using this parameter, following the example of other's use of it. Another contributor recently characterized my use of the quote parameter as POV. After checking the documentation I don't really know whether my usage was according to the consensus as to how the parameter should be used, or even if there is a consensus.

Here is the diff.

Thanks! Geo Swan (talk) 21:43, 7 March 2009 (UTC)

I personally don't use it much. My thoughts about it are that it's a good feature for either offline news sources, or news sources that may become unavailable online in the future, to provide the quote that supports the text in the article; particularly if the verifiability of the text might be an issue.
In the case here, the quote "To detain anyone, anytime, anywhere in the world, indefinitely, under any rules they devise, that just can't be -- must not be -- the law of the land." is not attributed to its speaker, which suggests that it is attributed to the news article's author, Carol Leonnig herself, which seems unlikely; and it's a footnote to the text George Brent Mickum IV is an American lawyer. The quote is a non sequitor with respect to the article text, and does not belong. It's an expression of a particular individual's (probably Mickum's) view of the law, clearly not NPOV. I agree with its removal here.
The other quote is "The hypocrisy that we cannot discuss Zubaydah's treatment, but the government can admit to waterboarding him and claim that it is legal is rather astounding," said Mickum on Tuesday. At least this one is properly attributed to Mickum, but it's still got nothing to do with the text it purports to support: More recently Mickum was one of the first two lawyers to see Abu Zubaydah, one of the three captives the CIA acknowledges waterboarding, when he was in one of their "black sites." The quote does not belong and its removal was appropriate.
My $0.02, anyway. TJRC (talk) 16:42, 8 March 2009 (UTC)
Sorry to return to this a eyar later. I missed this repsponse. No offense, but I do not believe the comment above that the 1st quote could not be attributed to Mickum is accurate. I will reproduce the quote below, in context. I don't believe there is any question that it is a direct quote from Mickum.
But lawyer George Brent Mickum IV, who represents detainees seized in Africa, questioned how the president's authority could be limitless around the globe.
"To detain anyone, anytime, anywhere in the world, indefinitely, under any rules they devise, that just can't be -- must not be -- the law of the land," he said.
I will thank the writer above for their response. After a year I will not ask them to return to more fully explain themselves. They didn't explain their comment about why the second quote doesn't belong. After a year, with this the only reply. I can't help wondering whether the purpose of the quote field remains undefined. 173.206.10.154 (talk) 19:47, 13 March 2010 (UTC)

Dates showing up twice!

It seems that cite news is now showing the date TWICE when it lists a reference in the "Reflist" section of an article. Have a look at Auckland Council (as random example I was working on). The initial dates in (brackets) weren't there shortly before! Ingolfson (talk) 08:23, 20 March 2010 (UTC)

I can only guess something was changed at Template:Citation, which this one is based off of. This template itself hasn't changed lately, but Citation/Core was changed a few hours ago. I've left a note with the admin who made the change asking them to look into your question. -- AnmaFinotera (talk · contribs) 09:03, 20 March 2010 (UTC)
I'm pretty sure it was this change to {{citation/core}}. --Redrose64 (talk) 13:57, 20 March 2010 (UTC)
It's widespread. I think we'd better centralise the discussion of these related problems, I suggest the talk page of the template which is at fault, ie at Template talk:Citation/core#Last change breaks Editor fields. --Redrose64 (talk) 14:14, 20 March 2010 (UTC)

Author or Agency

For certain New York Times articles, the author is given as "The Associated Press"...should we put that in agency and leave the "author" blank(or vice versa)?Smallman12q (talk) 00:28, 12 March 2010 (UTC)

I would say put |agency=Associated Press and leave |author= blank. Rjwilmsi 14:22, 31 March 2010 (UTC)

Section parameters

{{editprotected}} I'd like to gather concensus for adding a "section" field. In many cases with output I get something like the following: The New York Times December 14, 1986, Sunday, Late City Final Edition BEST SELLERS: DECEMBER 14, 1986 SECTION: Section 7; Page 38, Column 2; Book Review Desk . In this case, page 38 is within Section 7. It would be much easier for those trying to track down these news articles if there was something along these lines. The current output for the above is:

"Best Sellers: December 14, 1986". The New York Times. 1986-12-14. p. 38. 

Something like "Best Sellers: December 14, 1986". The New York Times: Section 7, p. 38. 1986-12-14. Thoughts? Der Wohltemperierte Fuchs(talk) 22:54, 18 April 2010 (UTC)

Editprotected disabled, as there is no "complete and specific description of the request", which would include the changes to markup.
This template supports |at=, which should do what you want:
  • at: Position within the newspaper when page/pages is unsuitable, e.g. |at=p. 2, col. 3, |at=p. 2, para. 4 (when the page number is not sufficiently specific), or at=no. 456 (something in a numbered list). This parameter is ignored if either page/pages is specified.
---— Gadget850 (Ed) talk 23:52, 18 April 2010 (UTC)
Ah, totally missed that in the documentation. Thanks, Der Wohltemperierte Fuchs(talk) 01:49, 19 April 2010 (UTC)

Accessdate documentation

A recent edit to the documentation added |accessdate= to the "Commonly used parameters". My impression is that its use isn't that common compared to the other parameters, and judging by the discussion in #Accessdate above (and in the citing sources talk page), its use in {{cite news}} is controversial. I suggest removing it from the commonly-used parameter list, and adding the following text to the documentation for |accessdate=:

"This parameter is useful when referring to news sources that regularly update or move their pages; it is not necessary for news sources whose pages are stable once published."

Eubulides (talk) 20:32, 2 December 2009 (UTC)

It should always be used when citing an online copy, as there is generally no such thing as a site "whose pages are stable once published". They may appear to be stable for several years, but all it takes is one new department head, and the entire website gets scrapped and redone from scratch. I say this as both an editor and a web professional who has a) had this done to a site I managed immediately after I left, such that virtually every deep link to any content in the site just broke one day and never was fixed; and b) been hired to do this (more carefully, with redirects for old content locations), for a client. Belief in anything on the Web, or the Net more generally, being somehow permanent is most unwise. :-) — SMcCandlish [talk] [cont] ‹(-¿-)› 21:21, 2 December 2009 (UTC)
I think it is used often enough to be called "commonly used". Debresser (talk) 22:18, 2 December 2009 (UTC)
Having argued in favour of using accessdate more frequently in {{Cite news}}, here are a couple of suggestions that didn't occur to me last time we debated it:
  • If accessdate= is used in {{Cite news}}, the publication date must also be filled in. If an online story has no publication date, then {{Cite web}} is preferable to {{Cite news}}.
  • When accessdate is needed for a rapidly-evolving story, the hh:mm should be added, with the timezone (accessdate simply uses a string parameter, so "accessdate=2009-12-02 22:43 UTC" will work fine). This technique means that if the story subsequently changes, there's an audit trail. A newspaper that corrects something libellous in an online story won't want to leave any evidence of the previous version, will they? - Pointillist (talk) 23:41, 2 December 2009 (UTC)
Respectable newspapers have lawyers, who will edit the libel out before it reaches the stage of being published, either on the web or in print. So all of this is pretty theoretical. When has the question ever actually arisen? Anyway, I don't think Wikipedia should be trying to cover "rapidly-evolving stories". WP is not a news service. That is what Wikinews is for. We have more than enough on our plates being an encyclopaedia, without trying to compete with news sites. Wait till the dust settles and then write what has definitely happened.
Consequently, as one who believes accessdates are often unecessary, I naturally think there is even less need for the time of day as well as the date. We are making life far too complicated here.
However, I agree that the publication date of a news item is much more significant than the date somebody accessed it. It is amazing how many people seem to think that one can cite a news article without bothering to mention the publication date. I also agree that if the source doesn't show a publication date, it isn't a news item, and "cite web" should then be used instead of "cite news". -- Alarics (talk) 10:26, 3 December 2009 (UTC)
Alarics you should add that to the documentation. I would also say that in general the documentation is lacking in making it clear which citation template should be used or might be preferable given or absent certain data. As for the accessdate parameter I think it is necessary and used commonly so I don't think the proposed comment should be included. There is a whole world out there of information sources and one cannot assume what sources will be necessary to use and how reliable they might be in keeping the information available. In foreign countries the most authoritative source of news and information may not have the same resources and capacity to store information. That is already a source of systematic bias. That should not be compounded further with rigid templates. There should be some flexibility. 2 edits. [Deleted sentence 1st edit. Maybe a strikeout should be used but it was only up for a couple of minutes. Second edit for clarification.] Lambanog (talk) 04:44, 18 December 2009 (UTC)
Decided there was no need to wait to have someone else add my suggestion in previous comment, so I did it myself and have edited the {{cite news}} documentation accordingly. Did not act on the matter raised in this section though. Lambanog (talk) 05:08, 18 December 2009 (UTC)

Re-opening discussion, June 2010

Not to rehash old news, but I re-added "accessdate" to common uses after having to go through an article for WP:FAC and add the accessdate to several citations (which came from the FAC reviewer, not me). All of the citations missing accessdate were done using Cite News since the accessdate was removed from common use and all had publication dates and valid urls. If the url is present, then the accessdate needs to be present from what I've gathered here at Wikipedia and in sourcing outside of Wikipedia as well. Doesn't seem to be any consensus here that accessdate isn't common use and after FAC I don't see how it isn't when the url parameter is. No url = no accessdate and vice versa; you delete it if you don't need it. I agree that the publication date is far more important for news, but accessdate can tell if someone has looked at the article recently, which is especially important when news sites alter their web address (which has happened to me more than once) or update a story without updating the publishing time. --JonRidinger (talk) 21:00, 15 June 2010 (UTC)

One suggestion that may or may not have already been brought here (lots of info on the subject I see) would be to have a common use section for printed-only news (that is missing "url" and "accessdate" but has "page" or "pages" parameters) and a common use for online-only and printed news that is also available online (that has "url" and "accessdate" parameters) much like Cite web has the different date format versions. Every academic source I've seen and now FAC here has the accessdate with any web url present (though an academic publication wouldn't likely cite a printed news story/publication with a url at all even if one were available; it would simply be cited as a printed news article). Trust me, if FAC didn't want the accessdate on all web uses, I wouldn't be here right now. --JonRidinger (talk) 21:21, 15 June 2010 (UTC)
I also think its use is common enough to have it in the more common form. BTW, JonRidinger, your edit was undone. Debresser (talk) 22:42, 15 June 2010 (UTC)
Thanks...I am aware it was undone, which is why I came here rather than start an edit war. I always use accessdate whenever I have a url in any citation template as that is common practice. If there is worry that accessdate will be used inappropriately (like using it for a printed, offline source) then why have the url parameter? Again, FAC wants an accessdate with any url, otherwise I wouldn't care. --JonRidinger (talk) 23:14, 15 June 2010 (UTC)
I made an example of what I mean in regards to common forms for online and offline news sources. --JonRidinger (talk) 23:27, 15 June 2010 (UTC)
I see no possible reason to not have the accessdate always. Debresser (talk) 05:42, 16 June 2010 (UTC)

I think it's pretty straightforward: any example in the documentation that includes url= should also include accessdate=. accessdate means that someone has checked that URL and that the source existed there at the time stated, and confirmed that it suppotred the assertion for which it is cited as of the date of access. TJRC (talk) 00:12, 16 June 2010 (UTC)

Well OK, but I still believe it should be made clear that the date of publication is essential whereas the access date is merely desirable. I have too often come across citations referring to news items that give the access date but not the publication date! Alarics (talk) 05:38, 16 June 2010 (UTC)
This template uses {{citation/core}}, so if |accessdate= is provided to {{cite news}}, it is not displayed unless any one of the following five are also provided: |archiveurl=, |chapter-url=, |chapterurl=, |contribution-url=, |url=. It therefore does no harm to have it present.
My own preference is that where either blank templates or examples show any one of these |url=-type parameters, they must also show the |accessdate=. --Redrose64 (talk) 12:09, 16 June 2010 (UTC)

When to set location parameter

I am in disagreement with another editor (PeeJay2K3 (talk · contribs)) over the use of |location= within {{cite news}}. In this instance we disagree about cites of BBC News – I understand from the {{cite news}} documentation that for such citations it's appropriate to set |location=London since the news source ("BBC News") doesn't contain that location. I've been setting the location as part of completion of other parameters (|date= in particular). Is that the right interpretation of the documentation? Thanks Rjwilmsi 23:11, 25 March 2010 (UTC)

I do not believe it is necessary to include the "location" parameter for BBC references, simply because it is not relevant. BBC already stands for "British Broadcasting Corporation", so the location is implicit. – PeeJay 23:28, 25 March 2010 (UTC)
In the case of local news stories which are also covered at national level, the |location= parameter might be useful as an additional indication of where the story came from. So, a story concerning a premier league football club in financial difficulties will make the national news, but the local station are likely to give the story more depth. So, in the case of Portsmouth F.C.'s recent difficulties, this went national, but got more extensive coverage on BBC South; if the latter were to be cited, |location=Southampton might be appropriate. --Redrose64 (talk) 00:07, 26 March 2010 (UTC)
True enough. That said, the locations of the BBC's regional offices are not common knowledge outside those regions, and so it might simply be enough to put |work=BBC North East Wales |publisher=British Broadcasting Corporation. – PeeJay 03:01, 26 March 2010 (UTC)
I don't think |location= is at all essential in the case of the BBC, but on the other hand it doesn't do any harm. There might be people somewhere for whom it is helpful to be reminded that the BBC is where it is, without having to interrupt themselves to go and look at the WP article about the BBC. I don't feel strongly about this one way or the other. But the BBC is perhaps a bit of a special case. Where I do think |location= is important is for newspapers whose title gives no clue as to their whereabouts -- in the last couple of days I have come across references citing the Times-Picayune and The Telegraph and Argus without further explanation that these are based in New Orleans and Bradford, respectively. Admittedly there probably isn't any other Times-Picayune, but there are many places that have a "Times" or a "Guardian" or a "Telegraph", and I don't like the Britocentric assumption that people should take these to mean the London ones unless otherwise stated. Alarics (talk) 05:33, 16 June 2010 (UTC)
Indeed. It's not so long ago that The Guardian was headquartered in Manchester, although a national daily newspaper. --Redrose64 (talk) 12:37, 16 June 2010 (UTC)

Letter to the Editor

How to format a letter to an editor correctly. I can't figure out which parameter to use for the address, of a letter written by a reliable source eg:

Beevor, Antony (25 September 2005). "Letter to the editor: A Woman in Berlin". New York Times. 

where Beevor gives his address as "London". Any other improvements would be welcome to how I have formatted this would be welcome.

I also think from following on from the #Newspaper section, page and column information a "section=" parameter would be useful, as many weekend newspapers in the English speaking world have their output sectioned up. -- PBS (talk) 01:05, 16 June 2010 (UTC)

Looks fine to me. I don't think it's relevant that the lettewriter lives in London. Alarics (talk) 06:20, 16 June 2010 (UTC)
Traditionally letter writers give name and address. I suppose it helps identify if we mean Sir John Smith of Foggy Bottom instead of Sir John Smith of Dingley Dell. But what ever the reason traditionally in British newspapers the address is usually supplied as part of the identifier of the letter writer. -- PBS (talk) 08:05, 16 June 2010 (UTC)
Regarding section: as mentioned earlier (here and here), you can use |at= for this purpose. I've expanded the documentation to cover it. --Redrose64 (talk) 12:19, 16 June 2010 (UTC)

Magazines using cite journal

To quote User:AnmaFinotera: "There is a disagreement at Wikipedia talk:Citation templates#Magazines 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." --Karnesky (talk) 21:39, 22 June 2010 (UTC)

Include editors like Cite-book?

BusinessWeek is including the names of the editors of articles as well as their writers. Please incorporate editor parameters here like at Template:Cite book to encourage this excellent transparency. -- Jeandré (talk), 2010-06-28t13:10z

Seems a bit over the top to me. Don't we want to keep things as simple as possible? Alarics (talk) 14:47, 28 June 2010 (UTC)
Is there a style guide that includes editors when citing news articles? ---— Gadget850 (Ed) talk 17:47, 28 June 2010 (UTC)
Not sure, but I am sure it is not a standard thing to do. Alarics (talk) 22:24, 28 June 2010 (UTC)

Backup URL

Would it be possible to add a backup URL field? Such a field would be useful in case the original reference has been removed but a copy of the reference exists at a different URL. I would hate to use the URL for the copy in the URL field, as there is no way to know for sure if the copy is 100%. Victor Victoria (talk) 21:52, 24 September 2010 (UTC)

I can't think of how or why we would display that in a citation. You could add it as a comment either using a dummy field such as |backupurl= or within an HTML comment. ---— Gadget850 (Ed) talk 02:40, 25 September 2010 (UTC)

annotation field

There should be an annotation field. -- Tomdo08 (talk) 17:20, 29 September 2010 (UTC)

See Template talk:Cite journal#Annotation field. --Redrose64 (talk) 11:38, 30 September 2010 (UTC)

Proposed modification of Template documentation in relation to The Times

I propose the following change to documentation:

One
To change
  • location: Place of publication, e.g., Toronto for The Globe and Mail. Should be included if the city of publication is not part of the name of the newspaper. For student newspapers, include the name of the school.
to
  • location: Place of publication, e.g., Toronto for The Globe and Mail. Should almost always be included if the city of publication is not part of the name of the newspaper. For student newspapers, include the name of the school. For The Times published in London inclusion of location data is a stylistic choice.
Two
To change
Using page to identify the reference by page; news item in library but not on line
  • {{cite news | first=Arthur | last=Reed | title=Four-rate plan in air-fare 'jungle' | page=3 | date=9 December 1976 | newspaper=The Times | location = London }}
    Reed, Arthur (9 December 1976). "Four-rate plan in air-fare 'jungle'". The Times (London). p. 3. 
to
Using page to identify the reference by page; news item in library but not on line
  • {{cite news | first=Arthur | last=Reed | title=Four-rate plan in air-fare 'jungle' | page=3 | date=9 December 1976 | newspaper=The Times }}
    Reed, Arthur (9 December 1976). "Four-rate plan in air-fare 'jungle'". The Times. p. 3. 

Discussion

I propose this due to bots imposing the documented style in relation to The Times, and relying on the documentation as it stands for an essentialloy stylistic choice.

The unique nature—due to age, venerability, and common knowledge—of The Times which is published in London produces this situation.

Citation style stylistic choices supported by cite news include provisions against the citation of location data in relation to The Times. The citation styles suggesting this stylistic decision do so on the following grounds. It makes the encyclopaedia look like an ass to cite The Times as The Times (London). Other newspapers such as The New York Times are properly named as such. Other newspapers called simply The Times are simply insufficiently renowned to be confused with The Times, and necessarily take location data due to the possibility of their being confused with The Times. The inverse, of mistaking The Times (Podunk, Province, Nation) for The Times should not occur to a reader of international English sufficiently accultured in English usage to be able to read a work containing citations.

As editors responsible for bots are relying on this documentation to impose a stylistic choice above another stylistic choice in relation to citation style requirements, I propose the above changes to template documentation. Fifelfoo (talk) 14:51, 8 September 2010 (UTC)

I completely disagree. It is very Britocentric to suppose that every reader, in any part of the world, who may not have English as mother tongue, will be certain that we mean the Times of London. Even in the UK there a few other newspapers called The Times. Across the world there are many dozens. I strongly oppose this proposal. Alarics (talk) 08:20, 9 September 2010 (UTC)
(Fifelfoo and I discussed this before (s)he posted here). I also disagree with this change and think the current documentation and examples are correct. Rjwilmsi 17:36, 11 September 2010 (UTC)
After a month of this being undiscussed, I have boldly changed it. I find it incredibly Americo-centric to have prescriptivist documentation being relied upon so editors can bot impose a stylistic decision on articles. Fifelfoo (talk) 03:39, 6 October 2010 (UTC)
I have reverted the above bold change. In the above discussion, two editors have opposed the proposal and nobody has supported it apart from the proposer.
It's nothing to do with bots, and it's absolutely nothing to do with being Americo-centric (I am British and I have never even been to America). It's to do with having a consistent style for citing newspapers. I think it is quite absurd to write into the rules a special dispensation for one specific newspaper, out of all the newspapers in the world. Alarics (talk) 08:39, 6 October 2010 (UTC)
The special reason is bound up with longevity, influence and obviousness. In addition the special reason is that other newspapers with geographic determiners in their names, named themselves after the influence of The Times. I came across this practice while conducting a citation review for the Military History project where the awkwardness of The Times. (London) stuck out like a sore thumb. The edit producing this stylistic choice was a result of a bot editor who produced an argument from authority citing the examples here. Fifelfoo (talk) 08:47, 6 October 2010 (UTC)
I have not yet come across anyone else who has a problem with this. I would not be prepared to see the policy changed in the way suggested without a much wider debate. Alarics (talk) 12:36, 6 October 2010 (UTC)
  • Oppose Comprehension is the most important thing. There are hundreds, if not thousands, of newspapers called The [Something] Times across the world. There are also newspapers called The Times that aren't The Times. Nobody is helped by this change - nobody at all for any reason - but Wikipedia's readers are liable to be confused by dropping the mention of London. Editors here seem to forget very easily that Wikipedia exists for the benefit of us readers, not for the benefit of its editors, who are vastly outnumbered by us readers. If the proposer still want to make this odd change, then they need to use a more public venue than this page to discuss it: WP:RFC, for instance. 80.176.233.6 (talk) 14:35, 6 October 2010 (UTC)

'quote' field

I'm now seeing misuse of the 'quote' field for reasons of blatant partisanship. When something in a source can't be included in the body of an article for various 'not encyclopedic' reasons, it's now being included in the quote field. This of course makes it obvious to anyone reading the article without requiring them to click on the link to read it in context. Here's one example. I understand the reason for the field's existence, as stated in the documentation, and this isn't it. Any thoughts on how to prevent this? Perhaps a bot to flag its use so we can at least correct the most egregious uses? (This close to the mid-term elections, I expect it will become even more 'popular'.) Flatterworld (talk) 16:08, 18 October 2010 (UTC)

Agency, newspaper, and location

Currently, if {{Cite news}} is given an agency, a newspaper, and a location, it will list them in the following order:

Newspaper. Agency (Location).

However, it would make more sense to list the location immediately following the newspaper -- the location refers to the where the newspaper is published, not where the agency is located. Where the agency goes in this case is less of a priority to me, as long as it is put somewhere logical. But the other two items should be displayed in this way:

Newspaper (Location).

Thanks to anyone who can make this possible. --Metropolitan90 (talk) 21:41, 22 August 2010 (UTC)

The order of display is determined by {{citation/core}}, and its relevant parameters are: |Series= (for agency); |Periodical= (for newspaper); |Place= and |PublicationPlace= (for location).
{{citation/core}} uses some rather complicated logic to determine just which order of output to use, depending on the presence or absence of various parameters, not just these. --Redrose64 (talk) 21:57, 22 August 2010 (UTC)
In that case, I'm going to go over to Template talk:Citation/core to make the same request, because I doubt I will ever be able to figure out this sort of thing on my own. --Metropolitan90 (talk) 22:08, 22 August 2010 (UTC)
I completely agree with Metropolitan90 on this. I have noticed quite a few cases where e.g. "Agence France Presse" is made to appear to be in, say, Singapore, when it is the newspaper using the agency's material, and obviously not AFP itself, that is in Singapore. Alarics (talk) 22:23, 22 August 2010 (UTC)
We've really missed the point here. For news stories |place= should really be the place of the author's filing, as shown on the dateline, not the place of publication. That's what |publication-place= is for. LeadSongDog come howl! 19:25, 25 August 2010 (UTC)
I don't see much need for the dateline to be included in a citation to a newspaper article. I'm not aware of any major citation style that requires that. I have seen citation styles that require the location of the newspaper (where not already indicated by the title of the newspaper). --Metropolitan90 (talk) 03:45, 26 August 2010 (UTC)
That is correct. The dateline is completely irrelevant. The place of publication is what is meant by "location", as part of the identification of the source. For example, to differentiate between The Guardian (London) and The Guardian (Dar-es-Salaam). Alarics (talk) 22:25, 27 August 2010 (UTC)
That would be a correct use of |publication-place=London or |publication-place=Dar-es-Salaam, but an incorrect use of |place=London or |place=Dar-es-Salaam. LeadSongDog come howl! 19:58, 19 October 2010 (UTC)
Those are not the parameters we are talking about. What is being discussed here is the parameter "location". That clearly means, in the case of a newspaper, the city of publication. Alarics (talk) 20:14, 19 October 2010 (UTC)
Drat! I was thinking of the parameters at {{citation}}, where the example shows:

{{Citation | last=Smith | first=Joseph III | author-link=Joseph Smith III | title=Last Testimony of Sister Emma | newspaper=The Saints' Herald | publication-place=Plano, IL | volume=26 | issue=19 | date=October 1, 1879 | year=1879 | month=October | page=289 | url=http://www.sidneyrigdon.com/dbroadhu/IL/sain1872.htm#100179 }}
to render
Smith, Joseph III (October 1, 1879), "Last Testimony of Sister Emma", The Saints' Herald (Plano, IL) 26 (19): 289  It is rather sad that we can't seem to use the same parameter names for all these templates, though.LeadSongDog come howl! 20:40, 20 October 2010 (UTC)

Which Title?

Which title should be used if the title for the print and online versions are different? (At the moment I have been using the print title.) Reubot (talk) 12:39, 20 October 2010 (UTC)

I would suggest the online title. First, that's what people will see if they follow the link, and it's a bit disconcerting to see a different title and wonder if it's the correct article. Second, if the news source changes the link, then searching for the actual online title will make it easier to find the new link. Flatterworld (talk) 14:07, 20 October 2010 (UTC)

Consistency with dates when author is or is not present

I have an article (Richie McCaw) where I relied heavily on newspaper articles for sourcing. Some have authors and some do not. The date formatting is different in each case and I was hoping there was an easy way to make it appear consistent. AIRcorn (talk) 11:27, 30 December 2010 (UTC)

This is deliberate and is based on style guides. It is invoked through {{citation/core}} and is applied to the series of templates. ---— Gadget850 (Ed) talk 12:14, 30 December 2010 (UTC)
This has been discussed before, and I still don't like it. It looks quite silly and I agree with User:Aircorn that consistency would be much better. Why cannot the citation core (whatever that may be) be changed, if that is the problem? -- Alarics (talk) 12:22, 30 December 2010 (UTC)
This and other templates are based on the core meta-template. To change how this works, you really need to discuss it on the core talk page. ---— Gadget850 (Ed) talk 15:03, 23 February 2011 (UTC)

Editorials

Is there a preferred way to add e.g. [Editorial] or Editorial. to a news citation, for either a signed or unsigned editorial? --Pnm (talk) 21:10, 9 January 2011 (UTC)

archivedate= as a prerequisite for archiveurl=

Not all archived news articles specify a date for when the article was archived. It's frustrating to edit a page and discover that nasty error message in the references section, when there's no way at all to obtain the archive date of the archived article. Making the archivedate= parameter available is entirely understandable; but making it required? Even after reading through the Discussion archives here, I'm at a loss as to why archivedate= is a prerequisite to archiveurl=. And to avoid the obvious question, Did you make every possible endeavor to determine the archive date? – the answer is Yes, unequivocally. (Archive in question).—Biosketch (talk) 10:27, 23 February 2011 (UTC)

Where are you getting these archived news articles? Pl;ease provide an example. ---— Gadget850 (Ed) talk 12:30, 23 February 2011 (UTC)
That doesn't really look like an actual archive site, an archive site meaning that a full copy (perhaps without images) of the original is available independent of the original. In this case the yasni page just seems to embed the scostman page – if the original went yasni would not have its own copy. Rjwilmsi 13:39, 23 February 2011 (UTC)
Was in a bit of a hurry and didn't see the link. Yes, this is not an archive, it embeds the original page. Looking at the page source, it leads us to the original at http://living.scotsman.com/features/A-woman-of-peace.2796109.jp. The date is very important, as there can be different versions of an archive. For example, see the Wikipedia main page at Archive.org.
Firstly as an aside, thank you for posting the direct url to the original article. I tried clipping the yansi url all kinds of ways but couldn't figure out a way back to the source page. So thanks for the help on that front. As to the topic at hand – even though it's somewhat beside the point now – although I can certainly appreciate the value of having an archive date to go with the archive url, as the Wikipedia example compelling illustrates, I do wonder if making the date parameter a prerequisite is constructive given that there may be cases (yansi excluded) where the archiving site for whatever reason didn't specify the archive date. But presumably this has yet to be the case.—Biosketch (talk) 18:09, 23 February 2011 (UTC)

Agency report without news site

This source is an AP News article stored on the site The Free Library, which I couldn't find published in any other news outlet. Can cite news be used for this somehow? Cite web doesn't have all the same parameters. --Rontombontom (talk) 23:54, 2 March 2011 (UTC)

You should use cite news for this. Cite web should be used only when the source doesn't fit one of the other templates. ---— Gadget850 (Ed) talk 00:01, 3 March 2011 (UTC)
In that case, who is the "newspaper"? The "newspaper" parameter is required, indeed the citation doesn't display properly without it. Doesn't AP News belong into the agency parameter, and The Free Library in the publisher parameter? --Rontombontom (talk) 06:39, 3 March 2011 (UTC)
I should be inclined to put AP News under "work" in this case, since there isn't a separate "newspaper" or "work" to put. -- Alarics (talk) 09:47, 3 March 2011 (UTC)
Is it okay to not indicate The Free Library in any way? --Rontombontom (talk) 10:07, 3 March 2011 (UTC)

Also, a similar issue. Focus Taiwan is a website run by Taiwan's Central News Agency (CNA), publishing its wires. In this case, I guess "Focus Taiwan" can be entered in a "work" parameter, but what about CNA: should it be entered under "publisher" or "agency"? (Factually it's both.) --Rontombontom (talk) 08:50, 3 March 2011 (UTC)

Surely it's "agency". "Publisher" in "cite news" is meant only for the case where "newspaper" (or "work") is a small local paper that is part of a larger group. "Publisher" should not normally be used at all. I rather wish it didn't exist, since people are constantly putting the name of the newspaper or source under "Publisher" instead of under "newspaper" or "work". It matters because "newspaper" or "work" produces the name in italics, as required by convention. This doesn't happen with "publisher". I must have wasted many hours putting this right all over the place. -- Alarics (talk) 09:43, 3 March 2011 (UTC)

"newspaper" parameter

Should this be renamed "Work"? As already mentioned, "You may also use journal, magazine, periodical, or work, but do not use publisher for this." So when I cite news.bbc.co.uk or foxnews.com, there will be no confusion. And the template is cite news, not cite newspaper. --GalupK (talk) 01:59, 13 April 2011 (UTC)

For the record, I faced this issue while making this edit. --GalupK (talk) 02:01, 13 April 2011 (UTC)
Newspaper is just one kind of news source. It's easy to remember, if what you are citing really is a newspaper. "Work" does exactly the same job so it doesn't matter which you use. If you are citing a news source that is not a newspaper, then just use "work" anyway. The main point about all of these is that it produces the name of the source in italics, which is the appropriate style for the name of a publication. -- Alarics (talk) 07:24, 13 April 2011 (UTC)
Journal, periodical, newspaper, magazine and work are all aliases for the same parameter. If you try to use more than one, only one will show; i.e. if you use journal and work, only the value for journal will show. ---— Gadget850 (Ed) talk 11:40, 13 April 2011 (UTC)
Hmm. I get it now, thanks! --GalupK (talk) 00:32, 14 April 2011 (UTC)

Problem with access date parameter

In this edit I did a whole bunch of things, but among them I added this template to cite an old newspaper article (look at the end of the first paragraph in the "Construction" section). As usual, I included the access date, but it did not show up in the citation when I saved the page. I'm sure the template was formatted correctly without typos. Is this common or is there a problem with the template documentation? Thanks. Jsayre64 (talk) 01:14, 31 May 2011 (UTC)

Missing |url=. ---— Gadget850 (Ed) talk 02:53, 31 May 2011 (UTC)
Yes check.svg Done Aha! Thank you. Jsayre64 (talk) 03:27, 31 May 2011 (UTC)

When to use authorlink= paramter?

Please see Template talk:Citation/core#authorlink paramater on a question of when the |authorlink= should be used and when the |author= and |editor= fields can be linked directly. —  HELLKNOWZ  ▎TALK 12:17, 16 June 2011 (UTC)

Request to add deadurl= parameter

Per Wikipedia:Requests for comment/Dead url parameter for citations, could the following change please be made: Sandbox Source (revision diff).

This request is one of several follow-ups on implemented change to Citation/core. The request adds an optional |deadurl= parameter that is passed down to {{Citation/core}} as |DeadURL=. Additionally, if |deadurl=no then the main |IncludedWorkURL= link sent to Citation/core will be the original link and not the archived one. This will (in Citation/core) change the archive text to lead to the |archiveurl= (see RfC or sandbox below for preview).

Here are all the test cases I could think of: User:H3llkn0wz/Sandbox3. —  HELLKNOWZ  ▎TALK 20:27, 18 June 2011 (UTC)

Yes check.svg Done — Martin (MSGJ · talk) 12:48, 20 June 2011 (UTC)

"Location" of newspaper vs. "Location" of byline

Summary of request: Can you move the location parameter ahead of the agency parameter -- and add a separate parameter "AgencyBylineLocation" or "AgencyLocation" that would be displayed after the agency/wire service parameter?

In some citations for wire stories, the location indicated in the byline, if any, is different from the location where the newspaper is published -- but both may be useful to placing the source in context.

For example, I just added a citation [1] to the Pine Tar Incident article for an AP story that's published in the Junction City Kansas Daily Union newspaper.

Unfortunately, "Union City, Kansas" is listed after "Associated Press", so the reader gets the false impression that the AP byline is from that city, when that's actually where the newspaper is published.

I do want the reader to know the city where the paper is published -- and if the AP story had a byline from New York, then it should be indicated after the agency name.

Thanks. Infoman99 (talk) 08:51, 30 June 2011 (UTC)

I agree that the agency information appears (in the fairly rare cases when anyone bothers to include it) in the wrong place in the citation, and something should be done about it. But I don't agree that we need a separate location parameter for the agency. The byline of the story is not included when there isn't an agency involved, so there is no reason for including it when there is. -- Alarics (talk) 12:44, 30 June 2011 (UTC)
Also, the page number parameter should be placed before the agency parameter. The page number applies to the placement of the story in the particular newspaper, not in the agency news wire. Thanks. Infoman99 (talk) 21:06, 30 June 2011 (UTC)

Identifier update

Somehow this template wasn't updated to handle the various identifiers like the other templates. Could you replace

with

Thanks. Headbomb {talk / contribs / physics / books} 16:42, 13 July 2011 (UTC)

Yes check.svg Done. Removed the code from your request to avoid clutter. — Martin (MSGJ · talk) 11:31, 14 July 2011 (UTC)

Wrong?

It seems here there's something wrong with this template:

http://en.wikipedia.org/wiki/Moody%27s#cite_ref-10

Curb Chain (talk) 10:17, 6 August 2011 (UTC)

From the documentation: "Authorlink works either with author or with last & first to link to the appropriate Wikipedia article. Does not work with URLs." There is no parameter to link to an author's website. ---— Gadget850 (Ed) talk 11:28, 6 August 2011 (UTC)

Archiveurl and regular url

I don't understand why this template returns an error message when you specify an archive url without a standard url. Generally people only provide an archive url when the original is a dead link. Who cares about the old dead link that doesn't work? D O N D E groovily Talk to me 15:44, 26 August 2011 (UTC)

What happens if the archived URL goes dead? With the original URL, you still have a point to find a new archive. ---— Gadget850 (Ed) talk 16:46, 26 August 2011 (UTC)
Good point. D O N D E groovily Talk to me 17:28, 26 August 2011 (UTC)
You should always provide an |url=, because that is the original source of the material. Citation cites the source, and never an archived copy. It can only use an archived copy as explicitly marked means of accessing now unavailable source. —  HELLKNOWZ  ▎TALK 16:25, 28 August 2011 (UTC)
I don't see why this should be such a rigid requirement. Apart from anything else, some news citations are not on line at all and refer to print-only publications. In other cases the original online source might have long since completely disappeared, or conceivably, who can now tell, it was never on line at all, but an archived version of it exists on some other reliable website. Surely that is a great deal better than nothing. -- Alarics (talk) 21:49, 28 August 2011 (UTC)

Question about deadurl parameter

If you call the deadurl=yes and don't provide an archive url, does this template mark it as a dead link and categorize the page? If not it should. Or maybe a bot could do that D O N D E groovily Talk to me 17:31, 26 August 2011 (UTC)

It should (see discussion), but it is not implemented yet. —  HELLKNOWZ  ▎TALK 16:20, 28 August 2011 (UTC)

Linking to publisher and work entity articles

Hello, there is a bot approval request at WP:BRFA/H3llBot 9 for adding wikilinks to work/publisher fields where the entity can be unambiguously identified from a pre-selected list. Comments welcome, thanks. —  HELLKNOWZ  ▎TALK 16:18, 28 August 2011 (UTC)

Publisher parameter

Would anyone mind if I removed this? It's never needed, and indeed the template documentation notes that. Newbies are spending time trying to find out who the publisher is, and yet adding it only causes clutter. Also, when that parameter is filled in with the name of the newspaper, as often happens, it mean's it's not italicized. So all in all, it's not a good parameter to include. SlimVirgin TALK|CONTRIBS 01:48, 14 May 2011 (UTC)

IIRC there are a couple of web newspapers for which the publisher field makes sense [Staff] (1997-02-01) "Buy more homeloans" The Online Realestate shill newspaper (Wyong, NSW: Notnews Unlimited). In these cases it is because the Newspaper is barely a newspaper, and certainly not a newspaper of record, whereas the publisher is a widely known media conglomerate with a definite POV. Fifelfoo (talk) 02:40, 14 May 2011 (UTC)
The times when it might be of any use are so very few that, in my view, any benefit gained from keeping the parameter is massively outweighed by the confusion it causes, and the time some of us have to waste trying to correct references where the purpose of the parameter has been misunderstood. I strongly support the proposal by SlimVirgin to abolish this parameter. -- Alarics (talk) 05:52, 14 May 2011 (UTC)
Certainly there is widespread misuse of |publisher= versus |work= and |newspaper=, the same is true for {{cite web}}, and this is a frustrating thing to deal with if you're trying to correct it. This is something that I've been wondering about addressing for a while. However, |publisher= is useful for small newspapers/periodicals, so I don't think it can be removed outright. I would be happy to propose a bot to clean up incorrect use of |publisher= with your support Slim (it could operate on a known list of publications rather than changing all occurrences?). However, haven't we got to allow that certain pages have a "citation style" to use |publisher=, such as on Nick Griffin? [I'm not personally 100% in favour of each article having a special citation style, but I think I'm right in saying you've stated many times that this is permitted, and I know there are others of that opinion] Rjwilmsi 07:31, 14 May 2011 (UTC)
I'm not sure what the Nick Griffin issue is; could you say a bit more about it?
The issue isn't citation style; it's that it's almost never helpful to add the name of the publisher of a newspaper. We don't need to know in a citation that the publisher of The New York Times is Arthur Ochs Sulzberger Jr. When it's not a person, but a company, it's regularly added instead of "work," because people get the name of the newspaper confused with the name of the company that produces it. Or people add The New York Times in the publisher parameter, which means it doesn't get italicized. Plus, newbies complain that they spend time tracking down the publisher, thinking it's needed, only to be told off when they add it because it's clutter.
So this parameter really is a bit of a nuisance, caused by its creators getting confused between Book title (when name of publisher is needed) and Newspaper title (when it isn't). SlimVirgin TALK|CONTRIBS 23:03, 14 May 2011 (UTC)
Slim, yes, I've understood the problem and agree with you that there is a problem of widespread misuse of |publisher= for news/magazine/periodical citations, and I too would like to see the situation improved and standardised across en-wiki. (If editors followed the template documentation we wouldn't have a problem.) On the example of Nick Griffin the major contributor has chosen to set |publisher=nytimes.com rather than |newspaper=New York Times, and while I personally don't like it, it's what might be called a permitted "citation style", so the solution to the problem would not be as simple as dropping the |publisher= from cite news as you suggest, or making it behave like |newspaper=.
So I agree with your description of the problem, but not your proposed solution. I have no desire to create a complex solution, but just dropping the parameter won't work. I think a bot task to do some cleanup, and clearer/stronger wording in the template documentation are much closer to what's needed. I'd be happy to collaborate on both of those. Rjwilmsi 00:12, 15 May 2011 (UTC)
But the references in Nick Griffin are simply wrong and need changing! The name of the publication doesn't appear in italics, and it is a WP-wide requirement that names of publications appear in italics (see WP:ITALIC). (They are also, in my view, incorrect in often giving as the publication title the overall URL of the paper's website, but that's a separate issue.) "Clearer/stronger wording in the template documentation" is insufficient because, as you say yourself, "if editors followed the template documentation we wouldn't have a problem". Dropping the parameter altogether is exactly what we need to do. -- Alarics (talk) 07:51, 15 May 2011 (UTC)
Why don't we add a hidden tracking category at least temporarily. Then we can see how may articles use |publisher= and if it is actually misused? ---— Gadget850 (Ed) talk 11:40, 15 May 2011 (UTC)
If you need to do that, please do it. Or you could just take my word for it. I constantly find many cases where people have put the name of the newspaper under "publisher". -- Alarics (talk) 11:46, 15 May 2011 (UTC)
I have it in sandbox if anyone wants to test it. ---— Gadget850 (Ed) talk 15:21, 15 May 2011 (UTC)
Please just do it, or explain to me how to do it. -- Alarics (talk) 14:24, 17 May 2011 (UTC)

The perspective from someone who just learned from this topic that he's been doing it wrong the whole time: I was surprised to learn there is a difference between publisher and newspaper. When an article appears in the Washington Post, the publisher is the Washington Post, not some dude who happened to buy the paper. Washington Post published the article, not Joe Investor. I think the fix is to make publisher behave identically to newspaper, and have a new "owner" to replace what publisher used to do, just for news cites. D O N D E groovily Talk to me 04:08, 27 August 2011 (UTC)

No, the Washington Post is the publication, not the publisher. I would have said the publisher was the Washington Post Company, but under the strange US terminology for these things the publisher is actually somebody called Katharine Weymouth. But we don't need to know either of those things. We just need to know that the item appeared in the Washington Post, and those two words need to be italicised in the reference. I do wish somebody would now fix this. -- Alarics (talk) 07:54, 27 August 2011 (UTC)
Well, I AM an American, I would never think the publisher is Katharine Weymouth. The Washington Post is not owned by an individual, it's a public stockholder company, so the current CEO isn't the publisher or owner by any stretch of the imagination. You could legitimately list the Post as publisher for Slate, or list Gannett as the publisher for a variety on local newspapers, but the CEO isn't the publisher of anything, she doesn't own the paper, she just works for them. That's why it's best to delete the publisher tag altogether, or make it duplicate the function of newspaper. Not to mention, a lot of users are using publisher for news not published by newspapers, such as television news (CNN and the like). D O N D E groovily Talk to me 14:20, 27 August 2011 (UTC)
  • Strong oppose: This field exists for a reason. Where it isn't helpful (e.g. because the newspaper and the publisher are the same corporate entity) don't use it. Where this is not the case, it is often crucial information, especially with regard to liberal or conservative bias. It's makes a very, very big difference to a large number of people whether something published by, say, Murdoch or Fox. PS: This template is used for more than newspapers. PPS: And of course the parameter is not for an individual with a job title, it's for an organizational entity. If the /doc needs clarifying, then clarify it. Don't remove functionality simply because you personally prefer not to avail yourself of it. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:34, 24 February 2012 (UTC)
The publisher may be an individual, or a body corporate: but whichever it is, that is the entity upon whom a writ is served in the event of a libel action being brought. That is why it is a legal requirement for newspapers to state their publishers. It's not a legal requirement for us to give the publisher in a citation: but it does help to distinguish between two similarly-named newspapers where the location is either ambiguous or not stated. --Redrose64 (talk) 13:20, 24 February 2012 (UTC)
If " Newbies are spending time trying to find out who the publisher is", when no publisher is needed, then that suggests that the documentation needs to be improved. It's not a case for removing a useful parameter from the template. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:40, 24 February 2012 (UTC)
That is all very well in theory, but in practice the majority of editors, I would say at a guess 98% of them, are not consulting the documentation before using the template. They see the parameter names and assume they are supposed to fill them all in. That's the problem. -- Alarics (talk) 22:29, 24 February 2012 (UTC)
Where are they seeing the parameter names, if not the documentation? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:16, 25 February 2012 (UTC)
Citation tools, I expect. There are plenty about. Some of them offer a form, the user fills in the boxes, these get turned into template parameters. If they enter garbage into one of the boxes, garbage is saved in one of the parameters. --Redrose64 (talk) 13:16, 25 February 2012 (UTC)
Citation tools, indeed. One of these (I think it might be the one that appears if you select "Show edit toolbar" in Preferences) no longer has the problem, the "publisher" parameter in "cite news" having been removed from the box that appears, thankfully, after various people including me pointed out that it was being misunderstood. But with other such tools the problem evidently remains. What editors put in it isn't actually garbage, usually, it is simply the information that ought instead to be entered in "newspaper" or "work". -- Alarics (talk) 13:53, 25 February 2012 (UTC)
Again, why not fix the documentation of, and included in, those tools? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:22, 27 February 2012 (UTC)
We might not have the means to do so. The websites of some newspapers, publishers and libraries provide a "cite this article" link (or similar), which when clicked produces a selection of citation methods over which we have no control. For example, see this, which yields {{Citation | title=Buttrose, Ita (1942-) | journal=Trove | publication-date=2008 | url=http://nla.gov.au/nla.party-487813 | access-date=undefined }} which uses the deprecated |access-date= parameter. --Redrose64 (talk) 15:52, 27 February 2012 (UTC)
Even if there is some documentation hidden away somewhere, there is no reason why it would occur to somebody coming upon a set of boxes to be filled in that there is any need to consult any documentation, or indeed that there might be any documentation to consult. They just fill in the boxes as seems appropriate to them, and they misunderstand what is meant by the box marked "publisher". -- Alarics (talk) 16:04, 27 February 2012 (UTC)
If there is a box marked "publisher", then that single word is the documentation, and is clearly inadequate. That can be fixed. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:53, 27 February 2012 (UTC)
Well then I wish somebody would fix it. - Alarics (talk) 09:41, 28 February 2012 (UTC)
Redrose - thanks for providing that example. I've just updated WP:AWB/RTP to convert |access-date= to |accessdate= for the {{citation}} template. GoingBatty (talk) 04:49, 28 February 2012 (UTC)

Accessdate plus real date?

I thought that when we had a real date of publication, that the accessdate for the URL was supposed to be suppressed. This template doesn't seem to do that. Is there some reason why it displays both dates to the reader for a newspaper article? WhatamIdoing (talk) 15:42, 15 August 2011 (UTC)

It makes sense to show both. Date is when the article was written. Access date is when someone visited the web site and it worked. If the link goes dead, this gives people a date when the web site worked, which makes fixing dead links using archives easier. D O N D E groovily Talk to me 15:42, 26 August 2011 (UTC)
The access date is optional. The publication date is essential. You can put both if you like, but the publication date is the one that matters. -- Alarics (talk) 08:09, 27 August 2011 (UTC)
Access date would matter to bots and verification when linkrot kicks in. The content may have been updated/changed since publication. Although, that affects {{cite web}} more than this template. —  HELLKNOWZ  ▎TALK 16:26, 28 August 2011 (UTC)
Only marginally, and decreasingly, as most newspapers have an online version, and people tend to use this tempate to cite them. That being why it has a url parameter, and all. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:36, 24 February 2012 (UTC)

Inconsistency between cite news and cite web

Unresolved: There is an actual problem here.

If {{cite web}} is given a |url= but no |title=, it throws an error. This is not the case with {{cite news}}:

Would it be possible to harmonise these? --Redrose64 (talk) 19:18, 8 September 2011 (UTC)

{{cite news/sandbox |url=http://www.example.com/}}. http://www.example.com/.
Admins: Replace the code on Template:Cite news with this. — Z 20:11, 8 September 2011 (UTC)
You actually have to use the {{editprotected}} template to get admins to fix stuff like this. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:29, 24 February 2012 (UTC)
@SMcCandlish: I didn't use {{editprotected}} because I wasn't asking for a specific amendment - I wanted a discussion about what action to take. I dislike inconsistency - if {{cite web}} throws an error but {{cite news}}, given exactly the same parameters, lets it straight through without complaint - this is an inconsistency and should be resolved one way or the other. It might be that {{cite web}} has the incorrect action: why do we require titles for one, but not the other? --Redrose64 (talk) 13:06, 24 February 2012 (UTC)

Note that other changes have been made since this was posted, but the code should be easy to figure out. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:29, 24 February 2012 (UTC)

Whilst the documentation does say that |title= is an essential parameter, from the examples given above it seems to work without so there may well be uses of this template out there that do not specify the title. I don't know how many of these there are, but for those uses this would be a breaking change. Whilst there could be the argument that having loads of red error messages everywhere might encourage people to specify a title, I'm not sure if this is the best way to go about that. It would be better to hold off making this change until we know what impact it would have. Tra (Talk) 09:38, 24 February 2012 (UTC)
Problem is, this sat here for more than 5 months with no one even noticing. No one is going to find out "what impact it would have", which is surely having red "fix me" text pop up in a number of places, without actually making the change. I filed the editprotected for someone else who evidently didn't know how, so I don't have a big vested interest in this one, but it makes sense to me. If there's a case where we're citing something without a title, someone needs to provide examples of why this is needed/appropriate. If it is demonstrably needed, then we need a | parameter that can be used to explicitly indicate that the work has no title, otherwise it is impossible to tell if the lack of a title is an error or is a correct. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:14, 25 February 2012 (UTC)
I had a look at why this happened: Both templates used to require |title= until they were updated to use {{citation/core}}. Cite web kept the error message but cite news lost it. This was all three years ago and since that time, people may have added citations without a title. I'm not willing to make a change that will affect loads of pages just to see what impact it would have. Although having these extra error messages would draw attention to where additional information needs to be supplied, they would make the articles look worse. Other options are to add a tracking category to pages with citations missing |title= or perhaps doing something with ParserFunctions to make the error message only come up if the accessdate is after e.g. today so that only future usage of the template is affected.
As for why the citations may be missing a title, I think one reason may be to save time. It's one less thing to copy/paste over. Ok, it's not a very good reason but I wouldn't be surprised if this a cause. This sort of usage would definitely need to be fixed but I would much rather avoid putting error messages all over the place to do this. Tra (Talk) 05:10, 25 February 2012 (UTC)
@SMcCandlish: regarding "I filed the editprotected for someone else who evidently didn't know how" - this edit summary makes me assume that you are referring to me. I do know how to use {{editprotected}}, thank you very much, and have done for something like two years. If I hadn't known the procedure, I would hardly have made this statement just one month after starting this thread, nor would I have made posts like these: 4 September 2011; 24 August 2011; 23 August 2011, all at dates earlier than this thread. --Redrose64 (talk) 14:11, 25 February 2012 (UTC)
I agree it makes sense to me for you to not use edit protected since you were just asking about harmonising the templates and you weren't proposing a specific edit. On the other hand, when SMcCandlish used edit protected, there should have really been a more specific description of the required edit (rather than asking for the code to be 'figured out') but I'll accept that this approach is fairly commonplace judging by the various requests I've seen recently. In any case, this is just a technicality so I wouldn't dwell on it too much. Tra (Talk) 14:57, 25 February 2012 (UTC)

Accessdate deprecated?

Has the accessdate parameter been deprecated? I ask because Citation bot 1 is removing it from this and other citation templates. Or is this an error with the bot that should be passed along to its owner? ElKevbo (talk) 06:06, 26 September 2011 (UTC)

Wait, I think I figured it out. This parameter is only used when the referenced source is online, right? That seems to be the case although I don't really agree with that approach; it should be optional but never removed if provided because it might be important in some instances. ElKevbo (talk) 06:10, 26 September 2011 (UTC)
Could you give any examples where |accessdate= is important for non-online sources? I can't really think of any off the top of my head. —  HELLKNOWZ  ▎TALK 08:00, 26 September 2011 (UTC)
Print sources are updated and revised, too, so it may be useful to know when a source was actually consulted. I know it's a small corner case, it just seems silly to remove what *might* be useful or interesting information that isn't being displayed or otherwise harming anything whatsoever. ElKevbo (talk) 08:30, 26 September 2011 (UTC)
For print sources the publication date and edition are sufficient. If I buy a book today but don't read it until this time next year, the content of my copy won't change, although those sold in the shops might get revised. --Redrose64 (talk) 09:41, 26 September 2011 (UTC)
Because publishers never make changes to published books, magazines, and newspapers, right? ElKevbo (talk) 15:05, 26 September 2011 (UTC)
|accessdate= only shows when |url= is set. ---— Gadget850 (Ed) talk 15:39, 26 September 2011 (UTC)
When publishers make changes in a published book, it will have a new publication date and a new edition. --Conti| 17:31, 26 September 2011 (UTC)
The accessdate originated as citation metadata for online sources, but is not limited to them. No bot should ever be removing non-broken accessdate parameters from any citation template. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:27, 24 February 2012 (UTC)
Accessdate is indeed for online sources only. Are you seriously proposing to include an "access date" for a book that you find in the library? What purpose would that serve? What if you find the book on your own bookshelf? This is nonsense. -- Alarics (talk) 22:20, 24 February 2012 (UTC)
No one said it would be mandatory. It appears to be moot, anyway, since the date doesn't show if there's not URL. It should because as ElKevbo observed, "publishers ... make changes to published books, magazines, and newspapers", and as others have observed here and elsewhere it it is metadata that can be useful for other reasons., such as tracking trends in sourcing. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:19, 25 February 2012 (UTC)
No, it shouldn't, because as Conti, Alarics and myself have all pointed out, it is the publication date which is important for print sources, because that is updated for revisions and new editions.
Example: There is a book titled The Story of London's Underground, by John R. Day, first published in 1963. This is now in its tenth edition (pub. 2008, ISBN 978 1 85414 316 7), and I possess five of those (1st, 2nd, 4th, 6th, 10th). I wish to draw a fact from this book, and have found it in some of the earlier editions but can't check it in the current one because I've lent it to somebody else. Therefore I cite the earlier edition, making sure that I fill in |edition=6th|year=1979|origyear=1963 - it's no use citing the current edition because the layout has changed at least twice, and the pagination has changed with every edition. The date that I looked it up is completely irrelevant because the text of the sixth edition has not changed one iota since 1979. I could look it up any time in the future, it would still be exactly the same. --Redrose64 (talk) 13:43, 25 February 2012 (UTC)
Exactly right. Very well explained. -- Alarics (talk) 13:57, 25 February 2012 (UTC)
That reasoning only holds true when the edition is explicit, which is not 100% of the time. As I and ElKevbo have already made clear to you. And to repeat myself again, "as others have observed here and elsewhere, it is metadata that can be useful for other reasons, such as tracking trends in sourcing". Please stop playing WP:IDHT games. No one has suggested that the date "you find the book on your own bookshelf" is relevant; that's an obvious straw man. The date the citation was added to the Wikipedia article is be useful as metadata. I have no huge issue with it not being visible in absence of a URL being specified, really, if people are going to pitch a fit about it. But there are many use cases for it.
One example of (human, not bot) metadata use, without picking on any real, particular article: Let's say the "theory of flux capacitors" is published in 2010, and a WP article about is written and sourced mostly from 2010 to early 2011. In late 2011, a refereed journal publishes a paper that indicates that the earlier research may have been faulty, and this generates some controversy in the academic press. Someone adds this fact and a citation in late 2011. In early 2012, someone looks at the article and wonders how current it is. The publication dates of the cited refs aren't all that relevant, because any refs about the stated facts of the theory are still going to date from when they were published. The editor can see a 2011 citation, so it's been "touched" since the dispute, but how much? The accessdates, however, indicate only one major edit, in late 2011, that addresses the questioned veracity of the theory, indicating that the issue is in fact under-explored in our outdated article. So the editor raises the issue on the talk page and goes and looks for more sources to update the article with. Yay. Without the accessdate, the editor would be prone to assume that the issue was adequately covered, because they'd see a late 2011 update, but also see an apparent consensus to use and keep earlier sources and the position they present in the article, as if the 2011 issue hadn't had much effect, when in reality it might have a had a huge effect and the article's just out of step with reality. This sort of thing happens all the time, especially for the reason that I wrote WP:SOURCEMINE about: people have a tendency to "drive-by cite". They want to add something, usually something very narrow, to an article, find a soure, and cite it for that fact and that fact alone, then move one, without "mining" the source for the rest of the facts and context that it contains, and how they relate to other sources, both used and not yet used. Back to the example: In the other direction, if some of the older cited works about the theory have recent accessdates, and an older version of the article from before the controversy shows those sources cited with older accessdates or none, this is clear evidence that someone's been checking them out again in light of the new citation, and affirmatively re-citing them as reliable, suggesting that the report of controversy was a WP:RECENT thing, and that our article addresses it well enough for now. I update accessdates for exactly this reason every time I verify an existing citation. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 02:12, 26 February 2012 (UTC)
Sorry, not persuaded. You want to see when the article had major edits? That's what the edit history is for. -- Alarics (talk) 09:35, 26 February 2012 (UTC)
You mention metadata. This doesn't contain the accessdate, even when a URL is present. This edit of mine, which uses {{cite news}} twice, both having |url= and |accessdate=, produced the following metadata for each citation:
ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Oxfam+shop+founder+dies+aged+88&rft.jtitle=BBC+News&rft.date=2+October+2007&rft_id=http%3A%2F%2Fnews.bbc.co.uk%2F1%2Fhi%2Fuk%2F7024049.stm&rfr_id=info:sid/en.wikipedia.org:Oxfam
ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Dead+at+88%2C+the+man+who+sold+us+charity+shops&rft.jtitle=Mail+Online&rft.aulast=Koster&rft.aufirst=Olinka&rft.au=Koster%2C%26%2332%3BOlinka&rft.date=2+October+2007&rft.pub=Associated+Newspapers+Ltd&rft_id=http%3A%2F%2Fwww.dailymail.co.uk%2Fnews%2Farticle-485197%2FDead-88-man-sold-charity-shops.html&rfr_id=info:sid/en.wikipedia.org:Oxfam
The character sequence "2012" does not occur, so there is no access date. By contrast, the article dates are both in there - look for the sequence "2007".
Now, to the other point. Let's say that I visit the Bodleian Library and consult a book. It's a first edition, because the Bodleian has first editions of everything published in the UK going back hundreds of years. I can't borrow it though because it's not a lending library. However, the book was published in the last few years so is probably still in print. I pop across the road to Blackwell's, find the book and buy it. I get it home, and find that the text I was interested in differs in some aspect, and no longer supports the Wikipedia edit that I wished to make; turning to the copyright page I see that it's a "revised edition". Whether I cite the first edition, which I consulted earlier today, or the "revised edition", which I have in front of me now, the access date is the same. How is an access date going to help anybody here? --Redrose64 (talk) 23:12, 26 February 2012 (UTC)

authormask

Can this parameter be added so that the reference list at Yooper dialect won't repeat Ms. Remlinger's name on the third reference cited. The first two use {{cite book}} so |authormask=—— suppressed the display on the second, but not the third source by her. Imzadi 1979  23:48, 6 November 2011 (UTC)

Authormask is intended for bibliography lists, not for references. ---— Gadget850 (Ed) talk 06:52, 7 November 2011 (UTC)
Cite news currently doesn't support authormask. Fifelfoo (talk) 07:08, 7 November 2011 (UTC)
Authormask currently works in cite news. Fifelfoo (talk) 11:00, 7 November 2011 (UTC)
Authormask was observed to work correctly in Yooper dialect. yours, Fifelfoo (talk) 11:01, 7 November 2011 (UTC)
It does not fix the basic problem of a lack of in-text citations for the article. When this is resolved, then authormask will need to be removed, as the citations will be ordered by use, not by author. ---— Gadget850 (Ed) talk 12:44, 7 November 2011 (UTC)
Not necessarily, they could use parenthetical references with a full bibliography. Or short or full notes with a full bibliography. Fifelfoo (talk) 12:27, 7 November 2011 (UTC)

Edit Request: support for authormask

Add line 3: |AuthorMask = {{{authormask|{{{author-mask|}}}}}}
such that the first four lines now read
 {{Citation/core
   |Citation class=news
   |AuthorMask = {{{authormask|{{{author-mask|}}}}}}
   |Surname1 = {{{last|{{{surname|{{{last1|{{{surname1|{{{author1|{{{author|{{{authors|{{{author|}}}}}}}}}}}}}}}}}}}}}}}}

This will allow cite news to support author-mask. Fifelfoo (talk) 07:08, 7 November 2011 (UTC)
Yes check.svg Done—An  optimist on the run! 10:51, 7 November 2011 (UTC)


Testing for borken:
  • 1900-mix-a-lot (2011-13-32). "bananaman". the clariontimesdailyhustler. p. 14. 
  • —— (2011-13-32). "bananaman". the clariontimesdailyhustler. p. 14. 
  • —— (2011-13-32). "bananaman". the clariontimesdailyhustler. p. 14. 
  • bacon, francis (2011-13-32). "bananaman". the clariontimesdailyhustler. p. 14. 
  • bacon, francis (2011-13-32). "bananaman". the clariontimesdailyhustler. p. 14. 
It works! Fifelfoo (talk) 10:59, 7 November 2011 (UTC)

Edit request: New logic and category for invalid paramaters

I am requesting an admin to implement the code I just added to the sandbox here. I added some code and created a new category to identify and rename some invalid parameters. This is just a sampling to get the logic right and if it works I will add a much larger group as well as adding this function to the other cite templates. Once the backlog is eliminated the code will be shortened greatly or eliminated entirely at the end of the run. Depending on the size of the list of articles affected we may implement a bot to do some of the work. Please let me know if there are any questions. --Kumioko (talk) 02:27, 28 November 2011 (UTC)

Trying to figure out details here:
  • Added AuthorMask— that is really for bibliographies, but OK
  • Removed check for no title— huh?
  • Added checks for a bunch of parameter misspellings
Where did the list of misspellings come from? And yes, a bot would be a lot more efficient. And it could remove unused parameters while it was at it. ---— Gadget850 (Ed) talk 03:22, 28 November 2011 (UTC)
If helpful, I've already got a database scan of unrecognized parameters in cite templates. Rjwilmsi 09:08, 28 November 2011 (UTC)
Mmmh, I think it would be wiser to create/modify a bot/AWB check if there are misspelled parameters and then fixing this automatically (if possible, otherwise creating a list of "unknown/don't know how to proceed"). No need to bloat that template even more. And the bot/AWB could also check the other citation templates...
@Gadget850: that diff shows that the authormask is already implemented...
mabdul 12:04, 28 November 2011 (UTC)
Never mind; I'm so used to updating the sandbox from current and saving then applying changes. ---— Gadget850 (Ed) talk 12:59, 28 November 2011 (UTC)
I'm about six months ahead of you on the bot idea: Wikipedia:Bots/Requests_for_approval/RjwilmsiBot_7. Rjwilmsi 12:35, 28 November 2011 (UTC)
User:Rjwilmsi/BadCiteParams seems quite exhaustive and much more than we can ever stuff into a template check. ---— Gadget850 (Ed) talk 12:59, 28 November 2011 (UTC)
Rjwilmsi already did 99% of them with his bot. Your right there is no way we want to put all of these checks in the template at once. We were going to break them into chunks a few at a time and then remove some of the less possible ones from the Rename parameter list now built into AWB. I would have implemented myself but I don't have admin rights to edit a protected template. I would have then removed the group that and replaced it with the next group. I also created the invalid parameter category to narrow the focus to only those that were a problem. There are just so many possibilities and many of them are probably one offs and don't need to be there. As for the other changes that were already previously implemented. I copied the live code to the sandbox before I made any changes to the sandbox so if there are other changes besides what I asked for then they were already live but the sand box wasn't updated. --Kumioko (talk) 14:18, 28 November 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── The current diff seems to indicate at least two line breaks have been added to the code. Additionally, it's not clear why this is being done anyway: presumably this is only useful above what the bot does if these checks are left in permanently, but that would add dozens of extra conditionals to one of the most widely-used templates on the project. Disabling for now as I'm a little wary of pushing this live 12 hours after the proposal. Chris Cunningham (user:thumperward) (talk) 15:33, 28 November 2011 (UTC)

Take the line breaks out then if you want. Its been 12 hours because I left it at about 10 PM last night. If knowone wants to add the code then I'll just have to go through all the cite linked articles manually. That means that I will have to pull them all in and take up precious time and resources skipping all the ones I don't want to review but thats fine I guess it will just take me away from other tasks. The reason I am doing this is because there was a new page created to Rename some of these invalid parameters here based on Rj's bot. Its a great addition to the AWB logic and Rj spent a lot of time and effort in creating the list. In the template rename page there are hundreds and hundreds of lines of parameter checks that are on very few if any actual articles and although many of them need to stay in case of actual spelling typos a lot of them are just taking up space. This was to try and eliminate some of the unlikely ones so we can clean up and shorten the list of parameter checks. Why I'm doing this is because this page is full of templates to display the parameters so we can't add too many more parameters without hitting the Templates per page exceeded threshhold of WP. And I have a very long list of parameters I want to add (relating to Infoboxes and WikiProject templates mostly).
As I stated above, I am not intending to add them all at once but a few at a time to make sure we eliminate as many as possible and clean up the list. I would have done it myself but I don't have admin rights and it seems like all the templates are protected these days. I didn't think I needed to leave a longwinded explaination of the purpose of this change and usually knowone reads them anyway. --Kumioko (talk) 15:52, 28 November 2011 (UTC)
One of the reasons that this template is fully protected is because it's very heavily used. Making any change to the template - no matter how small - will invalidate the cache for every page which transcludes it, including those unaffected by the change: potentially hundreds of thousands of pages will need rebuilding. So, we don't undertake changes lightly. --Redrose64 (talk) 17:00, 28 November 2011 (UTC)
Fair enough! --Kumioko (talk) 17:03, 28 November 2011 (UTC)

If anyone is interested here is the full list of invalid spellings for the parameters in Cite news as found on the AWB Rename template parameters page. This doesn't even include the list of parameters for Cite web and the others.

Many of these are very likely to occur repeatedly such as sate, publiser, quthor, etc. however many of these are unlikely to occur again and I do not have the time to mine through hundreds of thousands of articles to determine which ones can be removed. That is why I asked for these to be added a few at a time to the template so we could limit the list to only those articles that actually had problems. If anyone wants to do this task efficiently by adding a dozen or so of these at a time to the code so we can fix them let me know. This would also make it a whole lot easier for the bot to find them too by the way rather than using resources to mine through all the articles. Otherwise I have a lot of other projects to work on. --Kumioko (talk) 02:02, 30 November 2011 (UTC)


"Quote" Parameter

The current reference is as follows...

  • quote: Relevant quotation. Adding a quotation can help locate online copies of the item using a text search, especially if the original link goes dead.

I am soliciting some comments/clarification on the following. If a lengthy, online-accessible newspaper article is utilized as a source for multiple WP article content, should a separate citation containing a "relevant quotation" be constructed for each citation? Thanks. JakeInJoisey (talk) 16:27, 6 February 2012 (UTC)

No. I think the "quote" parameter is overused anyway. It should not normally be necessary to include quotes. They clutter up the reference block and can usually be avoided by wording the main text better. -- Alarics (talk) 21:56, 6 February 2012 (UTC)
I respect your opinion as quite, quite legitimate...and one that might be shared by many. However, I think you'll agree that the current template guidance allows for some rather wide latitude in its interpretation...perhaps too much to establish a standard for {{cite news}}. I also recall seeing some discussion somewhere as to tightening up "citation" policy in this regard...and I believe {{cite news}} might benefit from a less vague explanation. JakeInJoisey (talk) 22:20, 6 February 2012 (UTC)
Yes, I should be happy to see it tightened up, especially if it can be reworded to encourage editors to write the body text in such a way that use of the "quote" parameter is not generally necessary. -- Alarics (talk) 22:28, 6 February 2012 (UTC)
I think we must part company there as it has always been my impression that the purpose of the "quote" citation provision was to facilitate a search for a new source in the event the cited source became unavailable online...which is still reflected in the explanation...but now with the qualifier "relevant quotation" added. Quite frankly, I was somewhat surprised when I saw that since my last recollection of the explanation contained no such language (I could, of course, be wrong) and I have always built citations utilizing a portion of the first sentence (at least since I was last castigated by another editor many moons ago for ostensibly introducing POV "cherry-picked" text...which I still see being done to this day...and sometimes at GREAT length).
Sooooooo...which is it? A means to specify "relevant" source content (which I could easily support) or simply a search-assist entry? I think we need to make a clear determination one way or the other...and then clarify the guidance. JakeInJoisey (talk) 23:14, 6 February 2012 (UTC)
Anecdotal, but I've been here over 6 years, and I have never once seen that rationale before. — SMcCandlish   Talk⇒〈°⌊°〉 Contribs. 00:23, 24 February 2012 (UTC)

A space missing

Hi, I've spotted that in the last reference (Planinski vestnik, 1980) in the article Triglavka, there is no space between language and journal: (Slovene)Planinski vestnik. I think it should be written with a space. Also, a period should probably be written after the language (at least this is how it was done in other templates). Can someone please correct this? --Eleassar my talk 15:17, 23 February 2012 (UTC)

Well, the only ref that seems to fit your description is this:
but I don't see the problem described. --Redrose64 (talk) 19:38, 23 February 2012 (UTC)