Template talk:Citation/Archive 8

loc
One of the main reasons to use this template is to play nice with sfn and sfnp. However, right now, if the location of a passage is not described by pagination (as with references to chapters, sections, &c.), the footnote templates use a  field and this template uses an   field, producing needless errors and confusion. We type the footnote format much more often but you don't need to shift over to their usage. You should, however, support it. There's no reason the template can't process a  field the same way as an identical command to. — Llywelyn II   02:18, 6 September 2015 (UTC)


 * It could be confusing to support loc as an alias of at because it could be confused with location, which is an alias of place, the location of the publisher. Jc3s5h (talk) 12:39, 6 September 2015 (UTC)
 * said what I was going to say. location is used quite often. – Jonesey95 (talk) 14:10, 6 September 2015 (UTC)
 * Yes. I have tripped over that myself. "loc" and "location" are too similar. ~ J. Johnson (JJ) (talk) 20:55, 25 October 2015 (UTC)


 * I agree that loc would not be a good idea in the citation templates. However, the difference is annoying sometimes, and there seems no reason why sfn and sfnp could not have at as a synonym of loc. Editors could then consistently use the same parameter. Peter coxhead (talk) 07:02, 26 October 2015 (UTC)


 * Since sfn basically just creates the corresponding Harv link (wrapped in  tags), I think sfn's loc should correspond exactly with Harv's loc, which is used specify a location within a source, analogous to page numbers. This is pretty much what Citation's at does. But as noted, citation's location (alias to place) is about a characteristic of publication. I am feeling inclined towards (as Peter has suggested) using 'at=' consistently as the standard form across Sfn, Harv, and Citation for in-source locations, With a view eventually deprecating 'loc=' in Sfn and Harv. ~ J. Johnson (JJ) (talk) 22:52, 28 October 2015 (UTC)
 * I agree that this plan of deprecating loc in favor of at is a good idea. at is a more consistent choice with the same parameter in citation and the CS1 templates. —David Eppstein (talk) 23:00, 28 October 2015 (UTC)
 * Notification about this topic is probably in order at and.


 * There is a difference. In Module:Footnotes, loc, if set, is concatenated to the end of the rendered output:
 * and, for, is made part of the :
 * This is not, as far as I can tell, documented functionality, yet there it is, so I would not be surprised to find that editors have discovered it.
 * and, for, is made part of the :
 * This is not, as far as I can tell, documented functionality, yet there it is, so I would not be surprised to find that editors have discovered it.
 * This is not, as far as I can tell, documented functionality, yet there it is, so I would not be surprised to find that editors have discovered it.
 * This is not, as far as I can tell, documented functionality, yet there it is, so I would not be surprised to find that editors have discovered it.


 * cs1|2 does not work this way. at is an alias of page and of pages and so is not concatenated to the end of the other in-source location parameters.


 * —Trappist the monk (talk) 23:52, 28 October 2015 (UTC)


 * So change CS1|2. at is used for things like section and paragraph numbering, which do not replace page numbering. There has been discussion on this before. ~ J. Johnson (JJ) (talk) 22:42, 29 October 2015 (UTC)
 * It is not true that CS1/2 at, page, and pages are aliases. They go to the same part of the citation, and only one of them can be used, but (in some styles) page prepends "p." to its argument, and pages prepends "pp.", while at leaves it unchanged. —David Eppstein (talk) 00:22, 30 October 2015 (UTC)

Google
I spend a fair amount of time replacing gross google books urls with uses of the google books template. It occurs to me that life would be easier if we added a googleid (gid?) parameter to the cite and sfn template families. Something like: which would translate to something like Google apis allow pulling things like date and authors based on the id, making the cite building process even easier.

sfn could evolve, too: which would imply 123]

If this belongs elsewhere, directions appreciated! Lfstevens (talk) 05:00, 25 October 2015 (UTC)


 * Some of this is possible. It is not possible for cs1|2, or any template for that matter, to use Google's API.  Except in very limited cases (magic words and parser functions), templates do not have the capability  of seeing beyond their bounding   and.


 * It is possible to create a cs1|2 id handler for the Google id. If we do, editors would need to provide the whole value from the   query because Google uses other prefixes than just   and because the   query does not indicate ranges or comma separated page numbers.  Editors would also need to provide separate page numbers for page, pages, or at because alphanumeric page numbering is not prohibited.


 * Not the place to discuss but the same inability to see beyond its bounding curly braces will prevent it from doing as you have suggested.


 * —Trappist the monk (talk) 10:57, 25 October 2015 (UTC)


 * Your suggestion has a fair amount of sense for the various cite and citation templates, however I cannot see the point in modifying sfn and its cousins. sfn is meant to be simple, lightweight and unobtrusive in the main flow of the text. (I've tidied the layout of original post in this thread sightly, I hope that's OK) Martin of Sheffield (talk) 11:04, 25 October 2015 (UTC)


 * I have seen the Google books template used for this, like so:


 * Which yields:


 * It's not quite as simple, but it's already possible. Is this what you are already using? – Jonesey95 (talk) 11:19, 25 October 2015 (UTC)


 * It does appear simple to modify so that   has the same effect as  . This may or may not be worthwhile. Peter coxhead (talk) 11:47, 25 October 2015 (UTC)

Responses: Lfstevens (talk) 02:57, 31 October 2015 (UTC)
 * Thanks for all the good thinking on this.
 * I already use the google books template for urls. It's just tedious and I'm lazy.
 * The nice thing about changing sfn is that you can link to directly to the page in google books. If it's not possible to use the google apis to get author info (it's a little problematic because, e.g., do you cite the editors or the chapter author.) So I hereby amend the sfn proposal to look like:
 * Google book tool Converts bare url into cite book format  -- Moxy (talk) 13:08, 31 October 2015 (UTC)

Swedish LIBRIS database
Can someone please allow for the use of a template I just created, Template:LIBRIS. This allows linking to libris numbers, LIBRIS being the electronic catalog for the Royal National Library of Sweden. You can currently add it after the id tag, but I'd like to see this as a parameter, preferrably. Thanks - Letsbefiends (talk) 06:21, 19 December 2015 (UTC)
 * We already have far too many parameters for identifiers. Use id -- Red rose64 (talk) 08:29, 19 December 2015 (UTC)
 * Yes, I know - you just repeated what I said which was that you can add it as an id attribute. But why do you say that there are already too many id attributes? Is it a technical limitation? The wikilink you provide to the id section doesn't say that there should be a limit (not sure why you believe I don't know how the template works!) Why do you want to limit important IDs for national library catalogs? - Letsbefiends (talk) 11:32, 19 December 2015 (UTC)
 * I said that we have far too many because there have been several discussions over the last six years (perhaps more) suggesting that we reduce the number. There is no technical limit, but every additional parameter increases processing time. Don't know why you say "not sure why you believe I don't know how the template works" - where did I suggest that? I don't want to limit important IDs for national library catalogs per se, but remember that this is the English Wikipedia, and the Royal National Library of Sweden isn't really that relevant to us. It would carry much more weight if suggested at e.g. sv:Malldiskussion:Citation. -- Red rose64 (talk) 23:13, 19 December 2015 (UTC)
 * It was the way you repeated precisely what I had already said - it looked like you hadn't actually read my comment because I actually stated that I knew about how to use the id attribute ("You can currently add it after the id tag, but I'd like to see this as a parameter"). Anyway, whether it's the English Wikipedia or not is completely irrelevant. The sources for a lot of information in Swedish articles come from Swedish publications, and LIBRIS is that country's national cataloging system.
 * There is no discussion of adding LIBRIS codes to the citation template in the Swedish Wikipedia, incidentally, because it's already included in their citation template. Unless you mean they have suggested that the English Wikipedia use it, but why would they suggest that on the sv Wikipedia and not here? I can't see how that would ever happen, or even if that would even be appropriate!
 * Over the last six years there have been many improvements to template processing code, is number of parameters still a serious processing issue? If so, is there a bug listed about this, because I'd be interested in seeing it. And surely nested templates increase processing time also?
 * From a potential text processing/data mining POV, I would have though that having attributes in the template would be much easier to extract, rather than freeform text that can be added in any format. - Letsbefiends (talk) 03:05, 20 December 2015 (UTC)
 * I've looked carefully at both the source code and the documentation for sv:Mall:Citation and there is no mention of Libris at all. So, if it is already included in their citation template, it must have a different name.
 * As for nested templates, they only increase the processing time when they are actually used, and only on the pages that use that nesting. If code is added to a template in order to recognise a new parameter, that code is processed on every instance of the template on every page that uses that template, even if none of the instances on that page actually use that parameter. -- Red rose64 (talk) 16:52, 20 December 2015 (UTC)
 * sv:Mall:Bokref has this:
 * —Trappist the monk (talk) 17:00, 20 December 2015 (UTC)
 * sv:Mall:Bokref is the equivalent of our, not of our . -- Red rose64 (talk) 18:35, 20 December 2015 (UTC)
 * I'm clearly missing something. Why is this important? citation is used for all sorts of citations, which can include books. LIBRIS numbers are still important. Why would it not be important to include LIBRIS numbers?
 * You raise issues with processing time, but I'm not at all convinced that this is accurate. How do you know it increases processing time significantly? If it does, as I asked earlier, this sounds like a bug that should be tracked - so either it is currently being tracked by a bug ticket, or it needs to be demonstrated that there is a template related performance issue and a new bug logged. I'm happy to do this myself (in fact, I would like to do so if needed), but there must be a source for your template processing concerns, or you have actually measured the performance hit of too many parameters in the template. Could you please answer where you have either seen or have been advised of performance issues around this matter? - Letsbefiends (talk) 09:57, 21 December 2015 (UTC)
 * You said "it's already included in their citation template". It's not. It's included in their cite book template, or to be precise, sv:Mall:Bokref as pointed out by Trappist the monk. Whatever. I am yet to be convinced that this identifier would be a benefit to the English Wikipedia. -- Red rose64 (talk) 21:59, 21 December 2015 (UTC)
 * Consensus is that it's not needed yet, so I'm happy to go with this, if it becomes more widely used then I'll revisit the request. Unfortunately I'm still trying to understand your reasoning. The only reason I can see that might be at all valid is that it adds to template processing time. I've asked a few times now, but can you provide me with evidence that this is the case? - Letsbefiends (talk) 00:12, 24 December 2015 (UTC)
 * I'm not as opposed to this as Editor Redrose64 but, with only one article-space instance of (as I write this) and only 230-ish uses of the LIBRIS url in article space, I'm not convinced that it's time yet.
 * —Trappist the monk (talk) 12:36, 19 December 2015 (UTC)
 * Fair enough. Any tips on ensuring that LIBRIS is findable? - Letsbefiends (talk) 22:32, 19 December 2015 (UTC)
 * I just added it to Category:Catalog lookup templates — that might help. —David Eppstein (talk) 22:11, 21 December 2015 (UTC)
 * Thanks! I appreciate you doing that. - Letsbefiends (talk) 00:14, 24 December 2015 (UTC)
 * Thanks! I appreciate you doing that. - Letsbefiends (talk) 00:14, 24 December 2015 (UTC)

CGNDB Discussion
I thought some page watchers might be interested in a discussion for changing CGNDB into a redirect here. Thanks! Plastikspork ―Œ (talk)

Hyphens in page ranges and subsections within edited-volume chapters
In tabulation hashing, I have the following citation: This is suboptimal in multiple ways: —David Eppstein (talk) 00:33, 9 January 2016 (UTC)
 * . See in particular Section 11.1.1: Tabulation hashing, pp. 11-3 – 11-4.
 * . See in particular Section 11.1.1: Tabulation hashing, pp. 11-3 – 11-4.
 * The page numbers are formatted incorrectly. It is supposed to be a range of page numbers, where the page numbers are hyphenated: "11-1 – 11-23". Instead, the template incorrectly converts the hyphens into dashes: "11–1 – 11–23". How can I prevent this? This can be fixed by using at but it would be nicer if the template could recognize that the pages parameter already contains a dash and not try to make more of them itself.
 * It is necessary to cite three levels of hierarchy: the book and its editors, the chapter and its authors, and the specific section of the chapter containing the content I want to cite. As far as I know the citation template only allows two levels of hierarchy to be cited. Is there any way to cite all three within the template, or do I have to give up and use manually-formatted outside-the-template text as I have done above?
 * The book, on its front cover, actually distinguishes "edited by" (Gonzalez and Diaz-Herrera) from "editor-in-chief" (Tucker). Is there any way to indicate this in the citation?


 * The only cs1|2 template that supports three levels of hierarchy is and not all that well.  Here is a simplified example that somewhat mimics the form of your example:
 * But that isn't really satisfying because Editor is in the wrong place so this is probably not a solution.
 * But that isn't really satisfying because Editor is in the wrong place so this is probably not a solution.
 * But that isn't really satisfying because Editor is in the wrong place so this is probably not a solution.


 * For the hyphenated page number issue, as you've discovered the module is pretty mindless. I'll give some thought to how that might be improved.


 * —Trappist the monk (talk) 01:37, 9 January 2016 (UTC)
 * It is easy enough to detect a range of hyphenated page numbers:
 * But, that breaks down when single hyphenated page designators are listed:
 * 11-1, 11-15, 23-7
 * We might concoct some sort of this-is-how-it's-meant-to-be markup perhaps like the doubled parentheses used by vauthors to allow corporate names:
 * ((11-1, 11-15, 23-7–23-10))
 * or perhaps we could 'escape' the hyphens in some way perhaps by doubling them:
 * ((11--1, 11--15, 23--7–23--10))
 * Opinions?
 * —Trappist the monk (talk) 12:12, 9 January 2016 (UTC)
 * I'm not convinced that the doubled paren or doubled hyphen solutions are any better than the at solution we have now. These are all going to require looking up the documentation to remember the syntax, and at is simpler. —David Eppstein (talk) 16:56, 9 January 2016 (UTC)
 * I'm not convinced that the doubled paren or doubled hyphen solutions are any better than the at solution we have now. These are all going to require looking up the documentation to remember the syntax, and at is simpler. —David Eppstein (talk) 16:56, 9 January 2016 (UTC)


 * Really, I think the solution is to treat hyphens as endashes by default everytime, and then let editors specifically override the template by typing hyphen or something in the few cases where hyphens are meant. Headbomb {talk / contribs / physics / books} 17:05, 9 January 2016 (UTC)
 * Someone's bot has been turning hyphenated pages 11-12 into the page range 11–12, so you have your wish by another's fiat.
 * Someone else suggested using a non-breaking hyphen (U+2011 / #8209;) for hyphenated pages. It's esoteric, and there isn't an HTML entity name for the character. It conveys the intention of a hypenated page. The non-breaking hyphen's height does not match the ordinary hyphen's height in Wiki's font (it's the same height as minus sign): (- &#8209; &minus; – —) It looks odd, but I must believe it is intentional.
 * (at= hyphen is untouched)
 * (page= hyphen is untouched)
 * (pages= converts hyphen to endash)
 * (pages= converts hyphens to endash) (wrong)
 * (pages= using non-breaking hyphen; position shifts)
 * Glrx (talk) 18:24, 11 January 2016 (UTC)

Link to Cite Patent template, question on templates versus WP:EL
I was trying to find info on the Template:Cite_patent template and didn't see any. Should there be a "see also" type note in this help article? Also, the citation template seems to be an exception in allowing external links in inline citations, versus creating a reference. Has that ever been discussed and agreed upon? Can't find any info in WP:EL either, and posted a question on the talk page for that article too.Timtempleton (talk) 19:36, 20 January 2016 (UTC)
 * What additional information are you seeking? Did you try asking on Template talk:Cite patent, the talk page for that template?


 * Some examples to illustrate your question about external links would be helpful. – Jonesey95 (talk) 19:42, 20 January 2016 (UTC)


 * I was looking at the article SHA-2 and saw that there was an external link in the body of the article. Here's the text:  The SHA-2 family of algorithms are patented in .  I clicked to edit it, thinking I'd be able to convert an in-line external link into a more appropriate reference, per WP:EL, and saw that the external link was generated by the template.  I always thought that external links were to be avoided in favor of references, and since this is a gap in my knowledge, I wanted to find out more.  It seems to me that there should be info about this and other possible other templated exceptions (if indeed others exist) on this Citation page as well as on the WP:EL page.  I can keep on looking - there's likely info elsewhere - but to make Wikipedia easier to use I don't think you can add too much helpful coding and standard policy info.  People who know all this intrinsically don't think to add info for the rest of us, which is why I'm bringing it up.Timtempleton (talk) 22:16, 20 January 2016 (UTC)


 * Thanks for the explanation. After reading WP:EL, which is pretty clear, I added a clarifying sentence to the cite patent documentation, and I moved that cite template into an in-line reference. Does that work for you? – Jonesey95 (talk) 23:57, 20 January 2016 (UTC)


 * Thanks - you've clarified something that may trip up others too. How about adding one more bit of clarification to drive it home?: This template should not be used in the body of an article; to cite online references to patents in the body of articles instead use the regular  template.  For more details, see the External Links guideline.Timtempleton (talk) 02:31, 21 January 2016 (UTC)


 * I don't think using cite web in an article body (by which I mean article prose) makes sense, because it also creates an external link. In general, citation templates are best used in references. I have also seen them used effectively in lists of sources, or in a list of an author's works, as a way of ensuring consistent formatting. I think that these latter uses are consistent with the spirit of WP:EL, even though they technically create external links in the body of an article. WP:EL is, in my view, trying to discourage links within article prose. – Jonesey95 (talk) 04:26, 21 January 2016 (UTC)


 * Right - I meant using it within a ref tag, as with Maybe I'm getting confused.  Will have to play around with ref tags and citation templates on my sandbox a bit.Timtempleton (talk) 09:04, 21 January 2016 (UTC)


 * If I were going to use the cite patent template, I would put it inside mark up just as if I were using a citation template or cite web. It is widely accepted that abbreviated citations to certain works are acceptable among scholars, such as bible verses or lines from Shakespeare's plays. It's up to you if you want to include patents in that category. I wouldn't, because I would want to be able to see the date and inventors without having to follow an external link; in the fields I follow, I'm personally acquainted with some of the inventors. Jc3s5h (talk) 14:21, 21 January 2016 (UTC)

Posters
There was not a template:cite poster so I am not sure what category it would fall under. In the last (lower-right) image in https://twitter.com/DaPatMac/status/701554145707356161 for example is a poster titled "Fast-Paced Animation" which announces that season 2 of Sonic Boom (TV series) is scheduled to debut Fall 2016 on Cartoon Network. How would you cite said poster?

In this specific case it's not needed since a news site wrote an article citing the poster/tweet (and this news article was in turn tweeted by a show producer, affirming it) but if that hadn't happened, how would one go about citing the poster as an official document from the company? What would it be classified as? 184.145.18.50 (talk) 17:54, 23 February 2016 (UTC)


 * How about cite sign? H:CS1 says that it is used for "signs, plaques and other visual sources". Wtmitchell (talk) (earlier Boracay Bill) 23:17, 24 February 2016 (UTC)


 * Given that this is the talk page for Citation, why not use Citation? Give it as much information as you have and it sorts out a sensible format.  The only reason I can see for the cite xyz templates is conformance with pre-existing usage. Martin of Sheffield (talk) 11:15, 25 February 2016 (UTC)

Vertical bars in titles
When there is a vertical bar ( | ) in the title of a citation, is there a preference to use  or  ? If there is a preference for one over the other, is it sufficient enough to include in, for example, the general fixes of WP:AWB? Thanks -crh23 (talk) 17:05, 28 March 2016 (UTC)
 * I would say that the best way to include a pipe in title is the  magic word because it produces the cleanest metadata.  Compare:
 * I have no opinion about such inclusion in AWB's general fixes because I never run AWB with that enabled.
 * —Trappist the monk (talk) 17:24, 28 March 2016 (UTC)
 * I have no opinion about such inclusion in AWB's general fixes because I never run AWB with that enabled.
 * —Trappist the monk (talk) 17:24, 28 March 2016 (UTC)
 * —Trappist the monk (talk) 17:24, 28 March 2016 (UTC)


 * Why should there be vertical bars in titles? Any real-world examples? ~ J. Johnson (JJ) (talk) 20:24, 28 March 2016 (UTC)
 * I see this on some cites, where an article has a multilevel title or when an author is included as part of a title. Things like "Local News|Smithtown" and "John Smith|Software Roundup". The former is analogous to an em-dash or colon (or else the first part is a section or chapter that could be omitted), the latter could just leave the author in its own specific field. DMacks (talk) 21:20, 28 March 2016 (UTC)
 * Copying and pasting of titles and other parameter values from web pages is one of the sources of . Sometimes removing the text after the pipe makes sense (e.g. Man lands on Moon) and sometimes it is part of a legitimate parameter value (e.g. C). – Jonesey95 (talk) 22:23, 28 March 2016 (UTC)
 * Who, this CNET? The pipe is in their logo, not the publisher's name. It's just a stylistic gewgaw. The publisher of CNET is CBS Interactive. Perhaps there's a better example, but that one isn't it. LeadSongDog come howl! 16:32, 31 March 2016 (UTC)
 * Seems to me like a very dubious practice. "Multilevel" titles would be those with one or more subtitles, and long established practice is to use colons, semicolons, or dashes. I suspect current instances of other characters are just people trying out new characters they have found on their keyboard. Where an author is included in the title, there is the question of who included it. If the publisher, then the name is part of the title, and a vertical bar is neither needed nor appropriate. (E.g.: "John Smith: My Life".) If a WP editor includes it, then its an error, pure and simple. Where someone uses it to (or any other symbol) make a sexier name use, well that could lead to problems, and I wonder just where they are to be found in the phone directory. More likely editors are just sucking up too much text without regard to just what that text is. ~ J. Johnson (JJ) (talk) 22:44, 28 March 2016 (UTC)
 * P.S. Catching this kind of stuff seems like an excellent reason for not having any real unnamed parameters in any of the citation templates. ~ J. Johnson (JJ) (talk) 22:46, 28 March 2016 (UTC)
 * In a web page's  element, a pipe is often used to separate the true page title from the website name - they might be in either order. Some websites use a hyphen, dash or similar for the same purpose - for example, this page has Template talk:Citation - Wikipedia, the free encyclopedia and Template talk:Citation; the first of these might be given on some websites in the form or as . On some websites, there might be two or more pipes, separating three or more steps along a breadcrumb trail.
 * The true page name should also be given in the first  element, and it is this which we should put in the title parameter - if there is only one pipe, the rest can go in website as it stands, but if there are two or more pipes, some care is needed to set website. -- Red rose64 (talk) 13:41, 29 March 2016 (UTC)
 * I think we can agree that the "true" title of a piece is exactly what should go into title, no less and no more. Where a source is aggregated in a larger work, or website, those elements should be named in their proper parameters. If someone feels a need to indcate some kind of nesting they can use the conventional colons, semicolons, angle brackets (">"), cheverons, or slashes. (Or even backward slashes if that's where they're at.) I am feeling stronger that when there is a vertical bar (however expressed) in a title the preference should be to tag it as an unspecified error requiring further investigation. ~ J. Johnson (JJ) (talk) 22:22, 29 March 2016 (UTC)
 * I think we should remember that titles are often filled in from software (for instance using the citation tool in the Visual Editor) rather than by humans who are familiar with (or want to learn to be familiar with) the intricacies of our citation templates. And as the discussion above makes clear, parsing the title to figure out what the proper treatment of the vertical bars should be requires human-level intelligence rather than being something a software system could be expected to do automatically. So we should certainly allow vertical bars – presumably, encoded as | – as a way of allowing the software to work and avoiding biting inexperienced users. They can be put in a cleanup category for someone more expert in citation titling to clean up, but should not immediately be flagged as an error. It is also even more important to avoid flagging vertical bars inside math markup as being any level of problem, since that's something that could happen legitimately. —David Eppstein (talk) 22:48, 29 March 2016 (UTC)
 * The title is whatever the title is in the source. If that title has a vertical bar in it, we should present that character to the reader. I admit that I am at a loss to find a good example on the web at the moment, but I believe that they are out there. That said, the vast majority of vertical bars in title parameters on WP will have non-title text after them, so they might be a fun thing to search for with AWB or in a database dump. – Jonesey95 (talk) 23:26, 29 March 2016 (UTC)
 * Here's one with pipes in the title element, forming a reverse breadcrumb trail: http://www.theguardian.com/uk-news/2014/aug/07/crossrail-hertfordshire-london-euston-tring-hemel-hempstead - the page source has these two elements:  so I would cite this as either of   giving:
 * I really don't see why the name of the website (in this case The Guardian) needs to go into the title parameter. -- Red rose64 (talk) 08:22, 30 March 2016 (UTC)
 * I agree, but this is difficult or impossible to automate, since the order of the trail varies and many sites don't use  and </h1> correctly (if it at all – a site I often use, the World Spider Catalog, has <h5 ></h5> as the highest level on many pages, and then for a generic phrase). So even if only an interim measure until a more experienced editor can clean up, we have to be able to cope with whatever gets put in tags which plausibly mark titles. Peter coxhead (talk) 08:43, 30 March 2016 (UTC)
 * Maybe I'm being obtuse, but I just can't see the issue. When displayed in my browser there is the simple title without any pipes.  If there is a clear displayed title why go looking for hidden markup?  I'd just cut-and-paste from the displayed title (as here: "New Crossrail route mooted from Hertfordshire into London") because that is what Topham or the Guardian wanted to be the title. Martin of Sheffield (talk) 16:40, 30 March 2016 (UTC)
 * It's not displaying pipes in any browser, for the simple reason that the only pipes that I added were properly used - as parameter separators. That is, I used three different parameters New Crossrail route mooted from Hertfordshire into London UK news The Guardian each preceded by a single pipe as required by the MediaWiki template parser. -- Red rose64 (talk) 21:04, 30 March 2016 (UTC)
 * You did, but look at the previous posing by Jonsey who appeared to be. The whole section is about pipes as part of the title, not as you showed with pipes as parameter separators. Martin of Sheffield (talk) 21:41, 30 March 2016 (UTC)
 * From what I've seen (here, and around), the vertical bar generally separates non-title elements, which some genuis thought ought to be displayed along with the title, and our editors don't quite understand its role as a separator. E.g., "Washington Post" is not an alternative to "the Moon" as a landing site.
 * Any citation tool responsible for putting vertical bars into title ought to be seriously bitten, and its developers as well. But regardless of why or how a vertical bar ends up in a title, the preference ought to be: no vertical bars. And informing inexperienced editors of that it is not necessarily "biting" them. To make that clearer perhaps we should have some kind of message like "your edit is provisionally accepted pending cleanup of a small problem or two". ~ J. Johnson (JJ) (talk) 20:32, 30 March 2016 (UTC)
 * From what I've seen (here, and around), the vertical bar generally separates non-title elements, which some genuis thought ought to be displayed along with the title, and our editors don't quite understand its role as a separator. E.g., "Washington Post" is not an alternative to "the Moon" as a landing site.
 * Any citation tool responsible for putting vertical bars into title ought to be seriously bitten, and its developers as well. But regardless of why or how a vertical bar ends up in a title, the preference ought to be: no vertical bars. And informing inexperienced editors of that it is not necessarily "biting" them. To make that clearer perhaps we should have some kind of message like "your edit is provisionally accepted pending cleanup of a small problem or two". ~ J. Johnson (JJ) (talk) 20:32, 30 March 2016 (UTC)

There are other items in the html/js of that Guardian article besides the. In this instance the "headline" and the "webTitle" are both given as: "New Crossrail route mooted from Hertfordshire into London", just as it is displayed in the browser. I'd also suggest, for what it is worth, that both "UK News" and "The Guardian" can be found (nearly literally) in the url filepath and second-level domain, separated from the date and title parts of the filepath. This is a very common formulation on news websites. Similar construction crops up on Bloomberg here, but using a spaced dash instead of a pipechar to separate the "Bloomberg" following the headline, so: <base href='http://www.bloomberg.com/'> <meta charset="utf-8"> U.S., China Seek to Prod Nations by Signing Climate Accord Early - Bloomberg is composited from <meta property="og:title" content="U.S., China Seek to Prod Nations by Signing Climate Accord Early" data-ephemeral="true"> or from <meta name="parsely-title" content="U.S., China Seek to Prod Nations by Signing Climate Accord Early" data-ephemeral="true"> It seems this sort of thing is not so difficult as to be infeasible to automate, though my coding skills are long out of use. LeadSongDog <small style="color:red; font-family:Papyrus">come howl! 17:34, 31 March 2016 (UTC)

20 minutes ago when I followed the link there were 14 entries in the category. There are now 5. All the others were trivial mistakes: missing "title", missing "=" or cut-and-paste into the title. Of the remaining five, one has a lot of errors, three look to be straight forward and one is possible tricky. Aren't we guilty of straining at gnats here? Martin of Sheffield (talk) 18:02, 31 March 2016 (UTC) ✅ All cleared in another 15 minutes work. It doesn't need automating, just the odd lunchtime. Martin of Sheffield (talk) 22:42, 31 March 2016 (UTC)

Does Template:Citation have all the same output parameters as Cite news and Cite web ?
I'd like to exclusively use Template:Citation in an article, similar to model at WP:FA quality The General in His Labyrinth, but I was reverted by another editor.

Documentation on the template page says: "The Citation template generates a citation for a book, periodical, contribution in a collective work, patent, or a web page. It determines the citation type by examining which parameters are used."

Does this mean we can use it instead of multiple different cite templates on a page, like Cite news and Cite web ?

I ask because I'd like to increase standardization and uniformity on the article.

Would that be okay ?

Thank you for your expertise,

&mdash; Cirt (talk) 23:48, 6 April 2016 (UTC)
 * Please read this: "If the correct parameters are used, this template produces output identical to that of the Cite templates, such as Cite book and Cite web". I believe this is the definitive answer here. Let me know below if there is any more expertise or explanation anyone can provide. Thank you, &mdash; Cirt (talk) 23:59, 6 April 2016 (UTC)
 * As the article is now, it's not at all inconsistent. It might be argued that The Edinburgh Reporter cites should be  but that could go either way.  All of the templates are cs1 and there is nothing wrong with that.  Converting all to cs2 is considered by some to be a 'style' change and therefore a violation of WP:CITEVAR.  In your revert of the other editor's revert, your edit summary claims:
 * If the correct parameters are used, this template produces output identical to that of the Cite templates, such as and
 * Sort of. Compare these:
 * They are not exactly the same (inter-element punctuation, terminal punctuation, capitalization of static text, and automatic harv) To make all of the cs2 templates render like cs1 templates, set cs1.  That seems rather a waste of effort to my mind.
 * —Trappist the monk (talk) 00:22, 7 April 2016 (UTC)
 * Right, but what are the differences? Punctuation only? Commas and periods only ? &mdash; Cirt (talk) 00:24, 7 April 2016 (UTC)
 * Did you see that my edit summary was merely quoting verbatim from the text at the top of Template:Citation ? &mdash; Cirt (talk) 00:26, 7 April 2016 (UTC)
 * Yep, just as I wrote. Full stop (cs1) vs commas (cs2), terminal full stop (cs1) vs. no terminal punctuation (cs2), 'Retrieved' (cs1) vs 'retrieved' (cs2), ref (cs1) vs. harv (cs2).  Those are the primary differences.  Because there are 23ish cs1 templates,  cannot hope to support all of those output capabilities merely by looking at which parameters are provided (and without making the already very large code larger and more unwieldy).
 * Yep, saw that quote. The rest of that paragraph says pretty much what I've just written except that it leaves out the bit about capitalization.
 * —Trappist the monk (talk) 00:40, 7 April 2016 (UTC)
 * So, just to confirm, as opposed to the concerns at Talk:Jenna Fife about Citation, the same fields give the same outputs ? There is no cause for concern about a missing field, for example? &mdash; Cirt (talk) 00:43, 7 April 2016 (UTC)
 * Are you not reading what I have written? The complete rendering of  vs  or  is not identical.  Other cs1 templates produce output that  cannot produce so in these cases  may well omit information.  It is not wrong nor bad nor inconsistent to have all of the references in an article use cs1 templates.  This same is true for cs2 when it can accommodate the citation needs of the article.
 * —Trappist the monk (talk) 12:01, 7 April 2016 (UTC)
 * (e/c [many]) Hey Cirt. I *think* what the user who reverted you is getting at (with the edit summary "obscures detail about citations") is that, even though there's little difference in output – to the rendered article text – the use of cite web/book/journal, etc. provides some guidance in edit mode as to the type of source, which I guess they think is useful to editors. If I've captured what the user intended, I don't really see it. At best, it's a truly marginal benefit. On the other hand, I see no benefit (including "standardization and uniformity") to warrant changing the use of web/book/journal etc. to use of  instead and see arguments on the other side—but all of its boils down to matters of marginal benefit. Nor do I know of any preference of, or common criticism by, FA reviewers to switch to use of this template over all others. (I actually dislike one difference in this template's output, the commas in place of periods, but once again, we're talking about trivial stuff. It is superior in one respect though: you don't need to add ref=harv, as you do with the others). In sum, I see it as a complete wash on a cost/benefit analysis. That being said, I do think there was a valid basis for a revert: WP:CITEVAR. So my advice would be to walk away.--Fuhghettaboutit (talk) 00:45, 7 April 2016 (UTC)
 * As part of a quality improvement project, I want to switch to model used at WP:FA page, The General in His Labyrinth. Wouldn't that be easier if everything is just Citation ? &mdash; Cirt (talk) 00:47, 7 April 2016 (UTC)
 * You want to switch, but that appears to contravene WP:CITEVAR. You would need to gain consensus for switching the citation style on the article's talk page. – Jonesey95 (talk) 01:29, 7 April 2016 (UTC)
 * We're already using Citation for all cites. Is there a compelling reason to switch? &mdash; Cirt (talk) 01:35, 7 April 2016 (UTC)
 * You lost me. At the top, you said "I'd like to exclusively use Template:Citation in an article,", and three indents above the one I am writing in, you said "Wouldn't that be easier if everything is just Citation ?". And now you say that "We're already using Citation for all cites." Sorry for being dense, but I can't make sense of that sequence.
 * You want to switch, but that appears to contravene WP:CITEVAR. You would need to gain consensus for switching the citation style on the article's talk page. – Jonesey95 (talk) 01:29, 7 April 2016 (UTC)
 * We're already using Citation for all cites. Is there a compelling reason to switch? &mdash; Cirt (talk) 01:35, 7 April 2016 (UTC)
 * You lost me. At the top, you said "I'd like to exclusively use Template:Citation in an article,", and three indents above the one I am writing in, you said "Wouldn't that be easier if everything is just Citation ?". And now you say that "We're already using Citation for all cites." Sorry for being dense, but I can't make sense of that sequence.


 * Regardless of the current state of an article, if you want to change the citation style for an article, you need to comply with CITEVAR, which may mean that you would need to gain consensus for switching the citation style on the article's talk page. – Jonesey95 (talk) 02:54, 7 April 2016 (UTC)
 * Thanks, but that was not our dispute on the article, but rather another user thought the output was not the same, but I see now it is the same. Thank you, &mdash; Cirt (talk) 11:16, 7 April 2016 (UTC)
 * The other user was correct. The output is not the same.  Presumably, that is why you made the choice to leave  in the article:
 * But, that does not explain.
 * —Trappist the monk (talk) 12:01, 7 April 2016 (UTC)
 * But, that does not explain.
 * —Trappist the monk (talk) 12:01, 7 April 2016 (UTC)

Citation styles
Reading the documentation about citation, I'm getting increasingly confused. It is openly referred to as CS2 and is in the "Citation Style 2 templates" category (see the foot of the page). Why then does it boast which ultimately leads to  at the top of the page? Furthermore, when I view the source it ends up as #invoke:citation/CS1|citation|CitationClass=citation. So which is it? If you know which it is, or can explain the split terminology please post an explanation within the documentation. Thanks, Martin of Sheffield (talk) 09:11, 26 April 2016 (UTC)
 * In the beginning, there was no cs1 and no cs2. There was a collection of vaguely similar and independent templates.  In an attempt to make them more consistent with each other  was invented.  It was the driving engine behind all of cs1 and cs2.  In the early days of Lua, Module:Citation was started.  Module:Citation/CS1 came soon after.  Presumably, the intention was to keep the different styles separate and perhaps share those things that were common.  But, that intention, if it was the intention, did not happen.  All work soon stopped on Module:Citation, so now  and the cs1 templates all use Module:Citation/CS1.
 * All of what I have just written is mere speculation since I was not there at the time.
 * I have thought, from time to time, about untangling cs2 support from the Module suite but other, more pressing matters have intervened. Perhaps one day I will venture down that rabbit hole.
 * —Trappist the monk (talk) 10:35, 26 April 2016 (UTC)
 * I have thought, from time to time, about untangling cs2 support from the Module suite but other, more pressing matters have intervened. Perhaps one day I will venture down that rabbit hole.
 * —Trappist the monk (talk) 10:35, 26 April 2016 (UTC)


 * Thanks for that. If I understand you correctly   and its three children (  and  ) are the only distinguishing features between CS1 and CS2.  Is Module:Citation used anywhere?  I note that its revision history stops 3 years ago.  Likewise is  in use anywhere?  I had a quick look at cite book and it just calls Module:Citation/CS1 with no reference to /core.  Many thanks, and kind regards, Martin of Sheffield (talk) 11:32, 26 April 2016 (UTC)
 * Almost. There are five things that distinguish cs2 from cs1: the separator character: comma for cs2, period for cs1; postscript character: none for cs2, period for cs1; automatic harv for cs2; certain static text: 'retrived' for cs2, 'Retrived' for cs1; only the one template for cs2, twenty+ templates for cs1.  How static text is rendered is determined in multiple places in the code and uses the separator character to make the determination (comma = lower case, period = sentence case).  That could possibly be made part of   but that seems like a major pain to do.
 * I don't know of anything that uses Module:Citation.  is used by several non-cs1|2 templates.
 * —Trappist the monk (talk) 13:18, 26 April 2016 (UTC)
 * —Trappist the monk (talk) 13:18, 26 April 2016 (UTC)


 * I've added a sentence to alert other readers to the CS1/2 issue. Please have a look and feel free to hack it about.  Martin of Sheffield (talk) 13:37, 26 April 2016 (UTC)

Translation to a non-English language
How to explain using this template that the book referenced is a translation from one to another non-English language? Doyoon1995 (talk) 05:57, 28 April 2016 (UTC)
 * WP:SAYWHEREYOUGOTIT. Use  to identify your source.  If the language used in your source is not English then include language.  It is not necessary to identify the original source language.
 * —Trappist the monk (talk) 09:30, 28 April 2016 (UTC)
 * Then is there any citation style in WP like: (Translated from A to B (both not English) by ...) with or without a citation template? Doyoon1995 (talk) 03:27, 29 April 2016 (UTC)
 * It is not necessary to identify the original language of a source that you are consulting and neither cs1 not cs2 support such. There is a set of translator parameters:
 * translator
 * translator-last
 * translator-first
 * translator-mask
 * translator-link
 * So, if translator Bob translated author Mary's book Title from Arabic to Breton and you are consulting the Breton translation, then write:
 * —Trappist the monk (talk) 12:04, 29 April 2016 (UTC)
 * Thank you. Doyoon1995 (talk) 13:28, 29 April 2016 (UTC)
 * —Trappist the monk (talk) 12:04, 29 April 2016 (UTC)
 * Thank you. Doyoon1995 (talk) 13:28, 29 April 2016 (UTC)

Wikidata item ?
Where does the Wikidata item go? Jura1 (talk) 02:55, 7 June 2016 (UTC)
 * https://www.wikidata.org/wiki/Wikidata:Main_Page – Jonesey95 (talk) 05:06, 7 June 2016 (UTC)


 * Do you mean the link in the "Tools" that appears on the left hand side of the page at Citation? It leads to https://www.wikidata.org/wiki/Q6925554 which is an item that describes the wikimedia Citation template. Jc3s5h (talk) 16:15, 7 June 2016 (UTC)
 * No, I meant e.g. for the sample at Template:Citation, where can I indicate that the book has an item at Wikidata? Jura1 (talk) 19:04, 7 June 2016 (UTC)


 * The original question is written as if the does provide a way to link to the Wikidata item, and Jura1 just doesn't remember the syntax. But I don't think I've ever heard of a way to provide a link to the Wikidata item; does such a mechanism exist? Also, if there were a way to link to a Wikipedia article about the source, that would serve much the same purpose, because the Wikipedia article contains a link to the corresponding Wikidata item, but we don't provide a way to link to a Wikipedia article about the source. Jc3s5h (talk) 19:10, 7 June 2016 (UTC)
 * The book shown in that sample citation, All the Presidents' Names, does not really exist, rather the title was made up for the example. If the example had been a real book about which an article existed, i.e. The Origin of Species, then a corresponding Wikidata item Q20124 could be found. A recent event WikiCite 2016 discussed mechanisms for better linking citation data, but the report from that meeting is not yet complete. While I would anticipate that it will eventually be very useful, it is still a work in progress. We'll probably want a new identifier parameter in the citation templates, but it might not be necessary. LeadSongDog <small style="color:red; font-family:Papyrus">come howl! 21:31, 7 June 2016 (UTC)
 * I think the answer to the question is: if the book has its own Wikipedia article, then you should wikilink the title parameter, and the wikidata item can be found from that. If the book has a wikidata item but no Wikipedia article, then the best current way of linking its Wikidata item is by using the data item as an id parameter, e.g. Q20124. —David Eppstein (talk) 22:03, 7 June 2016 (UTC)

Wikilinking the title doesn't work if the work is also available online. The following example of a real book that has a Wikipedia article and is also available online illistrates the problem. (It also illustrates the problem of an author who only has a first name, not a last name. I had to lie and write code falsely claiming the last name is Bede.)




 * Jc3s5h (talk) 10:21, 8 June 2016 (UTC)
 * Don't worry too much about the first/last name - you can't apply 21C standards to 8C. In such a case (I've just been citing Plato) I use author instead. Martin of Sheffield (talk) 12:14, 8 June 2016 (UTC)


 * I think it might be worth adding a specific field for the Wikidata item (such as "item=" or "wikidata=") as most works cited should eventually have Wikidata items and many other identifiers already have dedicated fields. Jura1 (talk) 10:56, 8 June 2016 (UTC)
 * Recently, there was a meandering and, as far as I can tell, inconclusive conversation on a similar topic at CS1.
 * —Trappist the monk (talk) 12:56, 8 June 2016 (UTC)
 * Reading the start of it, it seems to be about the author, not the source itself. Not sure if I'd do that. For the source, should we use "item=" or "wikidata=" ? Jura1 (talk) 14:14, 8 June 2016 (UTC)

Accessdate
It is not clear from the documentation whether the access date for an archived source refers to the original or the archived version. I've assumed the latter but can anyone confirm this is the correct interpretation? If so I'll update the documentation. Martin of Sheffield (talk) 11:28, 27 March 2016 (UTC)


 * The original, surely? There's no need for an access date to an archived version since the text will not change, whereas a non-archived version may change after it was accessed. Peter coxhead (talk) 11:30, 27 March 2016 (UTC)


 * This came about because I was checking the work of a bot. I assumed that the date was confirming the accessibility of the text, but you make a good point about changes.  Would not the archive date indicate this though? Martin of Sheffield (talk) 16:18, 27 March 2016 (UTC)
 * access-date applies to url and is used to indicate the date on which that url supported the article's text. It is important that follow-on editors be able to identify that date if, for whatever reason, the current content of the page that url points to no longer supports the article text.  Knowing that date, editors may more easily locate an appropriate archived copy of the page.  archive-date identifies the date on which the archive was created.
 * —Trappist the monk (talk) 16:49, 27 March 2016 (UTC)
 * Precisely. Peter coxhead (talk) 18:32, 27 March 2016 (UTC)
 * It should be the original, but occasionally the archived version is used from the very beginning. Hawkeye7 (talk) 01:56, 22 June 2016 (UTC)

Duplicated
Occasionally you get a warning like this:
 * Warning: Some Article is calling Template:Cite book with more than one value for the "parameter" parameter. Only the last value provided will be used.

Good luck finding it! Is there any way that it can put out some more information to allow you to locate the faulty citation amongst the hundreds of references? Hawkeye7 (talk) 02:09, 22 June 2016 (UTC)
 * This is not an error message from the cs1|2 templates but from MediaWiki. The template and Module:Citation/CS1 code cannot see multiple parameters in the template's wiki text.  There are some tools listed at ; perhaps they will help.
 * —Trappist the monk (talk) 02:48, 22 June 2016 (UTC)
 * Add the following to your skin (e.g. vector.js) file:

//Find duplicate template parameters importScript('User:Frietjes/findargdups.js'); // User:Frietjes/findargdups.js
 * Click Edit on an article, then click "Find dups" in your left-hand "Tools" bar. It will tell you which template has duplicated parameters. – Jonesey95 (talk) 04:10, 22 June 2016 (UTC)
 * Thanks for that! Hawkeye7 (talk) 13:54, 26 June 2016 (UTC)

Why year instead of date when using Simple citation?
Using date is recommended unless several conditions are met, and on most pages they are not. Unless I'm missing something date is better since one can state either the full date or just the year, as needed. --Middle 8 (t • c &#124; privacy • COI) 12:19, 14 August 2016 (UTC)
 * Because it's simpler to provide a year than a full date. Hence "simple citation". – Jonesey95 (talk) 16:54, 14 August 2016 (UTC)
 * However, if 'date' can take a year value and not throw an error, simplicity suggests using only one field. Wonder if there is a way to display 'year' and mean that you are adding a value to the 'date' field? --User:Ceyockey ( talk to me ) 01:26, 1 October 2016 (UTC)

Display of access date
Would anyone support a proposal to hide (but not remove) the access date field? The field is useful to editors, but not so much to the average reader, so it tends to clutter up references unnecessarily. They also tend not to reflect the actual last access date, just the last time an editor bothered to update it. As a result, I, and I'd suspect a number of other editors, often actually remove access dates when I'm cleaning up references. --Dr. Fleischman (talk) 21:18, 30 September 2016 (UTC)
 * Not a good practice to just remove access dates because you don't like them. They have value else they would not have been made part of cs1|2 as they are are similarly used by the published style guides that informed cs1|2 development.  If you don't want to see them in rendered articles, see Help:Citation_Style_1/accessdate.
 * —Trappist the monk (talk) 21:28, 30 September 2016 (UTC)
 * Not only is it "not a good practice", it's contrary to WP:CITEVAR if there's a consistent citation style. Removing access dates is quite unacceptable. Peter coxhead (talk) 01:08, 1 October 2016 (UTC)
 * Don't mean to pile on, but I agree, removing this information isn't a good idea. I do agree, though, that the value of the information is very variable.  It would be really cool if we had a link-check-bot that slowly ran through all of the external links and updated the last access date or, as appropriate, labeled a link as dead.  I seem to remember that there is, in fact, something like this available.  Anyone know more about this? --User:Ceyockey ( talk to me ) 01:22, 1 October 2016 (UTC)
 * Bots should not update accessdate, because they can only tell if the link is still active, they can't tell if the web page still supports the claim in the article. Jc3s5h (talk) 01:29, 1 October 2016 (UTC)

Registration or subscription needed
Currently I understand the &#8220;subscription&#8221; parameter accepts only a true/false value, correct? But many scholarly sources, especially medical journals, provide a free abstract but require payment to view the full text.

Therefore, I propose a new value &#8220;fulltext&#8221; (aliases: &#8220;full text&#8221; or &#8220;full-text&#8221;) for the &#8220;subscription&#8221; parameter. If somebody puts &#8220;,&#8221; then it should say that payment is required to view the whole article. 2600:1003:B015:319D:0:50:8948:D601 (talk) 16:27, 8 August 2016 (UTC)
 * I think that updating the documentation to say that a preview or an abstract may be available, but that access to the full source requires subscription or registration, would be a better enhancement. – Jonesey95 (talk) 02:02, 9 August 2016 (UTC)
 * I agree with Jonsey95, but in broader terms. Typically, the real situation is that there is a paywall, that you don't need a subscription to buy the article.  Nowadays, 'subscription' rarely applies to online resources but remains very true for offline sources.  I would actually like to see the introduction of 'paywall' as a yes/no value.  Now, the broader context - often for pay resources you might not, in fact, get an abstract, but the first 100 words; this is true of newspaper archives and some live newspapers that require subscription or payment for timed article access.  Another type of 'only the first one is free' is that some news outlets allow 3 to 5 articles free each month and all after that require subscription or payment-as-accessed.  So, might be good to have an overall review of the most common non-free access schemes and see what might be the best combination of current or new fields would be good to use to accommodate communication of these to end users in an easily understandable manner. --User:Ceyockey ( talk to me ) 01:34, 1 October 2016 (UTC)
 * I should have included the distinction between free and paid subscriptions as well. --User:Ceyockey ( talk to me ) 01:38, 1 October 2016 (UTC)

url and chapterurl, with archiveurl etc.
Both the main url and a chapterurl are set. Both items were long gone, but are archived in wayback. But there is only one archiveurl but needs also archivechapterurl (or chapterarchiveurl).

Use of deadurl doesn't make any difference. And even though both url and chapterurl are gone, the legend "archived from |the original" has the original url is 'live'. 213.205.252.119 (talk) 22:49, 5 August 2016 (UTC)
 * Perhaps the best solution to this problem is to split the single citation into two so that you can have original and archive urls for both. That might also allow you to use the other parameter correctly: 'Census of Population (1995, 2000 and 2007)' is not an author name; don't put it in an author name parameter.  Remember that cs1|2 templates are designed to render a single citation for a single source.
 * —Trappist the monk (talk) 11:02, 6 August 2016 (UTC)
 * Personally, I think that only the most specific url, if archived, would be sufficient. The aim is to provide a citation supporting information included in the encyclopedia, not to provide a convenient set of links to information resources.  That being said, I wouldn't be opposed to having 'archive' urls for each of the specific url types included in various citation templates. --User:Ceyockey ( talk to me ) 01:43, 1 October 2016 (UTC)

page-url
Could a page-url or a similar parameter be added? I'd like to link the section number in Smyth, and the template does not currently allow it. — Eru·tuon 01:26, 3 October 2016 (UTC)

Clarification: Smyth's Greek Grammar for colleges is organized in three chapters. I'm using chapter for the chapters, and since chapter and section are considered the same thing, there is no  parameter left for the section numbers. Hope this makes sense... — Eru·tuon 01:28, 3 October 2016 (UTC)
 * I don't see why you need two URLs. Use chapterurl. -- Red rose64 (talk) 07:05, 3 October 2016 (UTC)
 * , that's not what I meant: I wanted the section to be linked. But I decided to just put the link directly into the at parameter, so the problem is sort of solved. Still, it seems like the template should have a URL parameter corresponding to the page or pages parameter... — Eru·tuon 07:16, 3 October 2016 (UTC)
 * I don't see us doing this without we create enumerated pagen and pagesn with matching enumerated page-urln and pages-urln, and deprecate the current use of pages where heretofore editors have been able to list multiple pages in a free-form fashion: 2, 13–15, 19, 21. To which of these would page-url apply?
 * —Trappist the monk (talk) 10:21, 3 October 2016 (UTC)
 * Yes, it may seem a good idea to have page-url, but actually it's not. There may be a case for at-url, though, since in this case it would be clear that the whole of the value of at would be the text of the link. Peter coxhead (talk) 14:20, 3 October 2016 (UTC)


 * , I see what you mean, because I have encountered this problem trying to convert refs in Ancient Greek grammar to the template Smyth. When there's a range of pages, there isn't an appropriate URL to link to. — Eru·tuon 18:00, 3 October 2016 (UTC)

Protected edit request on 23 November 2016
Please add (full stop) after the phrase terminal punctuation is a period (under "mode") in the documentation at Template:Citation—both for the sake of differing US-UK terminology and to make the relevant part more easily searchable in the page (since searching for period also yields all instances of periodical). —Sangdeboeuf (talk) 19:13, 23 November 2016 (UTC)
 * Template:Citation/doc is not protected, so this does not need to be handled as a protected edit request. —David Eppstein (talk) 19:21, 23 November 2016 (UTC)

Accessdate
Regarding the 'accessdate' parameter: if I am translating a citation from another wiki would I use the date that wiki first retrieved the source, or when I first use it in the English Wiki? I know it's probably a stupid question, but I'm just curious. Psychotic  Spartan  123  23:00, 11 January 2017 (UTC)
 * The stupid question is the one that you don't ask. The date specified by the original editor is just as valid in your translation as it was in the original language.  If you check the source and prove to yourself that it still supports the translated article today, then you can use today's date if you wish.
 * —Trappist the monk (talk) 23:06, 11 January 2017 (UTC)

doi=
Thanks to changes in the crossreference site, some doi's are broken. The template may need to accommodate indicating broken doi's. See details. — Preceding unsigned comment added by 50.66.189.2 (talk) 17:06, 1 February 2017 (UTC)
 * Use doi-inactive-date as described in this template's documentation. – Jonesey95 (talk) 18:04, 1 February 2017 (UTC)

Paid subscription required
I am looking at an OED citation (to which I do not have access), and I see the helpful legend "(Subscription required (help))". Wondering what sort of help might possibly be provided, I hover over "Help", and get a popup which says: "The site requires a paid subscription to access this page." I find this irritatingly silly: the only conceivable extra information is that the "subscription" requires payment. I suggest that the legend should be changed to "(Paid subscription required)", and the popup removed, giving the same information in two fewer characters. Imaginatorium (talk) 04:35, 18 June 2017 (UTC)

Possible bug when 3 authors, no date
The sfn-style citation:

Text *

comes out as:


 * Text



I would have expected the reference to be called "Carrera, Jiménez & Viteri.", which is what happens if I add a year:


 * Text



Aymatth2 (talk) 21:40, 18 June 2017 (UTC)
 * It's not a bug. The date is required. When you use it the way you are using it, it thinks "Viteri" is the date. —David Eppstein (talk) 21:42, 18 June 2017 (UTC)
 * The date is not always given by the source, as in this example. Aymatth2 (talk) 23:52, 18 June 2017 (UTC)
 * If the cited source has no date, you must tell that there is no date by including a 'no-date' parameter (accepted values are   and  ) and similarly, you must tell  (or any of the cs1 templates) that there is no date by setting date to match the date parameter used in the matching  template.
 * —Trappist the monk (talk) 21:56, 18 June 2017 (UTC)


 * See Template:Sfn and an associated discussion here. Martin of Sheffield (talk) 22:25, 18 June 2017 (UTC)

O.k. – I see what is happening, but dislike the appearance given by  or. The result looks odd to me and would puzzle most readers. My solution below:


 * Text.



But why not automatically display a reference formatted like this if the last parameter is not a 4-digit number? Aymatth2 (talk) 23:52, 18 June 2017 (UTC)
 * We began to do that in October 2016 but did not complete the task. See Template talk:Sfn.
 * —Trappist the monk (talk) 12:04, 19 June 2017 (UTC)
 * I think it would be an improvement, although there is certainly no urgency. I looked back and found a similar example in an earlier article that I had not spotted because I tend to just glance at the references and citations to make sure there are no red warnings:


 * A reader might find the reference slightly odd looking. Picky stuff, but a minor improvement is worthwhile if it improves several thousands pages. Aymatth2 (talk) 14:47, 19 June 2017 (UTC)
 * Documentation for this situation is at Template:Harvard citation. – Jonesey95 (talk) 04:54, 19 June 2017 (UTC)
 * I suppose the situation is uncommon and the workaround above is simple. Still, formatting differently if the last field does not look like a year seems easy enough, if not exactly high priority. Aymatth2 (talk) 10:42, 19 June 2017 (UTC)

Wikidata
Can wikidata be used for this template? -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  13:46, 14 August 2017 (UTC)
 * takes wikidata (as 'reliable' as we all know source is) and renders a cs2 citation using.
 * —Trappist the monk (talk) 13:53, 14 August 2017 (UTC)
 * Thanks a lot. This is somewhat I was looking for. But the issue is that, editors are generally not aware of Qids. Can something easier be used like ISBN? -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  14:04, 14 August 2017 (UTC)
 * That facility already exists in the reftoolbar at the top of the wikisource edit window (and, I presume, also in ve) does it not? Click Cite, choose 'cite book' from the Templates dropdown, enter the ISBN in the ISBN field and click the quizzing glass to populate the rest of the form.  I don't know where the data come from but if they are retrieved from WorldCat, then the result is no more reliable than wikidata so editors must still copy edit the final template.
 * —Trappist the monk (talk) 14:34, 14 August 2017 (UTC)
 * That's for those who use that. I was talking about something like you just mentioned. Same book is referred on different pages using different templates and values. Some are even incomplete cites, while others use incorrect values. If we could just replace that with something like or, and the cite auto populate, then these errors can be avoided. This could also pave way for a bot to clean up cites on various articles. -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  15:06, 14 August 2017 (UTC)
 * There is no guarantee that an auto-populated cite will be a correct cite. Both wikidata and WorldCat are known to be unreliable.  For the isbn case, isbns may be shared among different editions.  It is possible that the edition listed in whatever database the auto-populator chooses to use will not be the same as the edition that the editor consulted.  The requirements at WP:SAYWHEREYOUGOTIT apply.  The auto-populator cannot know if the source it is using is the same as the source that you the editor are using.  For these same reasons, it is not a good idea for bots to convert cs1|2 templates to  or some other similar template.
 * —Trappist the monk (talk) 16:03, 14 August 2017 (UTC)
 * I posted the similar query on Wikidata as well and one of the editors referred me to Wikidata:WikiProject Source MetaData. This is in line with what I am talking about. The citations should be organised data which is understandable to machines rather than merely text wrapped in templates. -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  16:41, 14 August 2017 (UTC)
 * Further, the doesn't seem to work on even a single source listed on Mahavira or Jainism. We need something that can work on Wikipedia articles. -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  16:51, 14 August 2017 (UTC)
 * If only we lived in the best of all possible worlds. Alas, we don't.  For example, WikiProject Source MetaData does not exist.
 * You don't provide an example of not working so I cannot help with that.  If it is a matter of no support in wikidata for the sources you tried, of course,  won't work.  If it is something else, perhaps the best place to raise that issue is at the  talk page.
 * —Trappist the monk (talk) 17:28, 14 August 2017 (UTC)
 * —Trappist the monk (talk) 17:28, 14 August 2017 (UTC)

I don't know how to link Wikidata pages. Here's the external link -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  18:11, 14 August 2017 (UTC)
 * → d:Wikidata:WikiProject Source MetaData
 * See Help:interwiki linking
 * —Trappist the monk (talk) 19:05, 14 August 2017 (UTC)
 * Thanks, here is the relevant discussion on wikidata - d:Wikidata:Project chat/Archive/2017/08 -- Pankaj Jain Capankajsmilyo (talk · contribs · [//tools.wmflabs.org/xtools-ec/?user=Capankajsmilyo&project=en.wikipedia.org count])  01:29, 15 August 2017 (UTC)

|chapter-format= and |format= small glitch with "pdf" vs. "PDF"
I think I ran across this a long time ago and didn't say anything about it. It came to my attention again recently. It's not a big deal, but it does seem to be a small error. I think the displayed value should be consistent, regardless of the input. Since the display is "(PDF)" when no format is input but the url ends in ".pdf", that should be what's always displayed. I didn't investigate whether this applies for other values for chapter-format or format; I'm not aware of what the common ones in use are. I also don't know, or have forgotten, whether fixing it for will cascade into  and  templates, but the issue is there as well. Examples below:

[Edit: fixed small errors in examples.]&mdash; D'Ranged 1 &#124;  VTalk  :  23:40, 12 September 2017 (UTC)
 * It looks to me like the relevant code is:

local function style_format (format, url, fmt_param, url_param) if is_set (format) then format = wrap_style ('format', format);		-- add leading space, parentheses, resize if not is_set (url) then format = format .. set_error( 'format_missing_url', {fmt_param, url_param} );	-- add an error message end elseif is_pdf (url) then				-- format is not set so if url is a pdf file then format = wrap_style ('format', 'PDF');		-- set format to pdf else format = '';					-- empty string for concatenation end return format; end
 * In the first bit of code, format is set to the value of the parameter, unmodified, if the parameter has a value. In the second bit of code, format is set to PDF (all caps) if format is not filled in but the URL ends in "pdf" or "PDF".


 * The question appears to be: do we want to keep this situation as it is (preserve formatting); convert all values of format to all caps; or something else?


 * For reference the CS1 documentation says this about format:
 * format: Format of the work referred to by url; for example: PDF, DOC, or XLS; displayed in parentheses after title. (For media format, use type.) HTML is implied and should not be specified. Automatically added when a PDF icon is displayed. Does not change the external link icon. Note: External link icons do not include alt text; thus, they do not add format information for the visually impaired.
 * The TemplateData report for Template:Cite web says that format is used over 210,000 times in that template, with more than 50 unique values. It is reasonable to assume that not all values of format are file extensions, so if we go with all caps, we will end up with output like "(HARDCOVER)" or "(SECOND EDITION)". The same is true of the other CS1 templates. Template:Cite book has many parameter values that do not comply with the documentation (do a find on the page for "chapter-format"). Thoughts? – Jonesey95 (talk) 01:32, 13 September 2017 (UTC)
 * I have removed the templates from this discussion heading so that watch page link works.
 * Also applies to archive-format, conference-format, lay-format, and transcript-format.
 * I think that the only place where cs1|2 pays any attention to capitalization is in date-holding parameters where month-names are expected to be properly capitalized and in vauthors and veditors where the Vancouver Style requires uppercase initials. For all other parameters, cs1|2 templates are indifferent to case so editors are free to write the content of the various format parameters however they choose or as is appropriate for the source's file format.  We did, at one time force upper case for all format values but removed that functionality as a result of this discussion.
 * —Trappist the monk (talk) 10:26, 13 September 2017 (UTC)
 * I think that the only place where cs1|2 pays any attention to capitalization is in date-holding parameters where month-names are expected to be properly capitalized and in vauthors and veditors where the Vancouver Style requires uppercase initials. For all other parameters, cs1|2 templates are indifferent to case so editors are free to write the content of the various format parameters however they choose or as is appropriate for the source's file format.  We did, at one time force upper case for all format values but removed that functionality as a result of this discussion.
 * —Trappist the monk (talk) 10:26, 13 September 2017 (UTC)


 * In light of reviewing the tools link and seeing the scope of what is actually entered in these sorts of parameters, I concede that there's no way to code any sort of cleanup. My OCD will be triggered often; that's nothing new! Thank you both, Jonesey95 and talk:Trappist.
 * &mdash; D'Ranged 1 &#124;  VTalk  :  02:00, 14 September 2017 (UTC)
 * I don't know how to code it, but it might be possible to look for "pdf" and convert it to "PDF". It certainly would not be an elegant way to approach the problem, though, because then do we go looking for "doc", "xls", etc.? Where would the manual cherry-picking end?
 * If you're looking to avoid triggering your ocd, perhaps Wikipedia is not a good place to hang out. This place is fun and interesting, but consistent it is not. – Jonesey95 (talk) 03:40, 14 September 2017 (UTC)
 * An early draft of my response actually included a list of file extensions that could be listed for conversion. I started thinking about what file formats could be cited; I came up with text, image, video, and audio files. Those categories alone would yield quite a list of extensions to be "cherry-picked", as you say. Then I realized that lower-case might be a better default than converting everything to upper-case, as file extensions are lower-case by design; hence my conclusion that it's not a feasible idea. (Getting consensus on upper- versus lower-case to start with would take ages!) I've been around long enough to realize that WP is far from consistent; I just have to ignore my OCD triggers here. (That's more difficult to do than it used to be, which is one of the reasons I spend less time here.)
 * &mdash; D'Ranged 1 &#124;  VTalk  :  05:23, 14 September 2017 (UTC)
 * One might write:
 * &mdash; D'Ranged 1 &#124;  VTalk  :  05:23, 14 September 2017 (UTC)
 * One might write:
 * One might write:

<pre style="margin-left:9.6em">	if is_set (format) then format = format:gsub ('^[Pp][Dd][Ff]$', 'PDF');
 * Any variation of 'pdf' becomes 'PDF'; all other values of  are unmolested.  I'm not unsympathetic to the idea of rendering this format value in uppercase so that it will be consistent with the automatic rendering.  I am not enthusiastic about making and maintaining lists of file formats.
 * —Trappist the monk (talk) 09:59, 14 September 2017 (UTC)
 * The file extension case is an operating system/user preference. In a case-sensitive OS like *nix, common extensions are usually lowercase, and native tools will often (lazily) assume that.  In a case-insensitive OS like Windows then the extensions can be either.  (Indeed in older versions of VMS they had to be uppercase only.)  The format name however is an initialisation and should be treated like any other initialisation in English.  Hence you might see: "the file bid.pdf is a PDF file containing our bid".  TXT and DOC are abbreviations, so probably ought to be in lower case.  XLS stands for "Excel Binary File Format" so make of that what you will! Martin of Sheffield (talk) 16:15, 14 September 2017 (UTC)
 * The file extension case is an operating system/user preference. In a case-sensitive OS like *nix, common extensions are usually lowercase, and native tools will often (lazily) assume that.  In a case-insensitive OS like Windows then the extensions can be either.  (Indeed in older versions of VMS they had to be uppercase only.)  The format name however is an initialisation and should be treated like any other initialisation in English.  Hence you might see: "the file bid.pdf is a PDF file containing our bid".  TXT and DOC are abbreviations, so probably ought to be in lower case.  XLS stands for "Excel Binary File Format" so make of that what you will! Martin of Sheffield (talk) 16:15, 14 September 2017 (UTC)

Your last sentence is indicative of the problems that would most likely ensue if an attempt were to be made to convert all file extension values for  and its kin to either upper- or lower-case. There are valid arguments for both. Given that the parameter is often used for explanatory text, rather than just a bare file format (see Jonesey95's link to examples above for Template:Cite book), it would be a bit of a nightmare to try and extract the file extension from the parameter and manipulate its case.

I had a quick look at a list of common Windows file extensions. Other extensions that are also acronyms and might possibly appear in a citation template include: CSV: Comma-Separated Values; DIF: Data Interchange Format; EPS: Encapsulated Postscript; EPUB: Electronic Publication; GIF: Graphics Interchange Format; JPG/JPEG: Joint Photographic Experts Group; MIDI: Musical Instrument Digital Interface; PNG: Public Network Graphic; QT: QuickTime; RA: Real Audio; RTF: Real Text Format; SVGL Scalable Vector Graphics; and TIF/TIFF: Tagged Image File Format. There are others that are not true acronyms, but might be misconstrued as such: DBF: dBase II, III, IV (Data Base File); MDB: Microsoft Access (Microsoft Data Base); PPT/PPTX: PowerPoint; TXT: Text; and XLS/XLSX: Excel.

I'm sure I've missed some and included others that are unlikely to ever be used in a citation template. None of the above produce default rendering if the file extension is present in the citation's  parameter and the   parameter is not present or is empty. That said, I think it's entirely your decision as to whether or not to implement the pdf→PDF fix you offered above to conform with the default rendering. I would equally support leaving it as is.

&mdash; D'Ranged 1 &#124;  VTalk  :  01:25, 15 September 2017 (UTC)

|location= not behaving as documented
I was trying to get the location in the template to display in parentheses, per the documentation for the template:


 * When set, work changes the formatting of other parameters:
 * title is not italicized and is enclosed in quotes.
 * chapter does not display in this citation template (and will produce an error message).
 * location and publisher are enclosed in parentheses.
 * page and pages do not show p. or pp.
 * edition does not display.





I thought it might be specific to the alias "work", but that's not the case, either:





Shouldn't the location be shown in parentheses unless there isn't a value for work? I haven't explored the validity of the other changes that the documentation lists, but it seems that title should be italicized if work has no value, unless I'm reading the documentation incorrectly.





&mdash; D'Ranged 1 &#124;  VTalk  :  00:37, 17 September 2017 (UTC)
 * This is the talk page for . For issues concerning  and the other cs1 templates, the talk page is Help talk:Citation Style 1.


 * See here and here.
 * —Trappist the monk (talk) 02:57, 17 September 2017 (UTC)

Dates in template data
Shouldn't the source date and access date parameters be considered date instead of string? Daylen (talk) 05:10, 2 October 2017 (UTC)


 * Please give an exact location of where this issue arises. Is it in a template? Citations are handled with a series of nested template invocations. Give which template and which line number in that template you are concerned about. Jc3s5h (talk) 12:48, 2 October 2017 (UTC)


 * Hi . I am referring to the TemplateData (Template:Citation), which is used by VisualEditor. I have went ahead and implemented the changes to Template:Citation/doc. Cheers, Daylen (talk) 15:34, 2 October 2017 (UTC)


 * I have reverted your change because you have not clearly explained the full implications of your change and gained consensus for it. Maybe you're different from other editors who make programmeer-type changes related to dates, but most programmer/editors I've encountered have an insufficient knowledge of dates for Wikipedia purposes.


 * Where is this "date" type documented. Is what is allowed in a date type identical to a string? Does it allow "Spring 2017"? Does it allow "c. 1554"? Does it ever change a date entered like "9 December 1745", which in the right context is known to be a Julian calendar date, to "1945-12-09", which would be contrary to WP:Citing sources? Jc3s5h (talk) 16:13, 2 October 2017 (UTC)


 * I found some documentation. It says the date type is for ISO 8601 dates. The documentation for the citation templates, Help:Citation Style 1 defers to Manual of Style/Dates and numbers § Dates, months and years for acceptable date formats. That guideline calls for publication dates to be in the same format as the article body, or an abbreviated format with the same order among year, month, and day. So if "July 4, 2017" is used in the article body, then "September 4, 2017" or "Sep 4, 2017" would be acceptable publication dates; "4 Sep 2017" and "2017-09-04" would not be acceptable as publication dates in that article. Jc3s5h (talk) 16:30, 2 October 2017 (UTC)

author#
Using a "#" as a number sign is a North American English usage. It will be meaningless to most editors from the UK and Ireland (not sure about the other Commonwealth Countries) as it is known as hash (see Number sign). So I suggest that either a footnotes is added to explain this regional usage, or it changed to something else that has a more universal meaning. -- PBS (talk) 07:58, 22 August 2017 (UTC)
 * Forgive my skepticism. I guess that I would be very surprised to find an English speaking editor who can read and understand the COinS section and yet cannot understand the meaning of the # symbol in author#.
 * —Trappist the monk (talk) 10:27, 22 August 2017 (UTC)
 * Hence my use of "most editors from the UK and Ireland" not "all editors" For those who are trying to find out what the parameters are, but are not programmers, it is unlikely that they will know what # represents to Americans. The only place they come across it is as a "# tag" called a hash tag and not a pound tag (which would be "£ tag") and 99% of those would not know that its origins in HTML. -- PBS (talk) 14:44, 22 August 2017 (UTC)
 * I go with Trappist on this. We are discussing the talk page of a template.  I would suggest that the only readers who come here are fairly computer literate and are well aware of the use of a hash as a number sign.  It's been on the front covers of imported comics, for instance, since I was a schoolboy - and that was way before computers would fit on a large desk!  I do agree that calling it a pound sign is ambiguous.  Although there are instances of it being used in that way, most seem to stem from the introduction of video terminals. Martin of Sheffield (talk) 15:32, 22 August 2017 (UTC)
 * Getting into the weeds here, but it's a fun bit of techno-history. A careful comparison of the Baudot code (Fig-N) and Murray code (Fig-H) will reveal the source of the conflation. The same 5-bit code (Baudot 11110 or Murray 10100 when shown with the lsb on the right) was used on early teletype machines for the hash symbol in the American configurations as was used for the pound symbol in the UK and for the No character in Europe. This situation pertained for many years, and trans-Atlantic messages conflated the two symbol's meanings. The confusion was continued later when, in the US, ASCII 0x23 was assigned to the hash symbol a few years before ratification of the very similar ISO/IEC 646 supported national variations, in which many variations used 0x23 for the hash symbol, several used it for the pound symbol, and a few made other assignments. LeadSongDog <small style="color:red; font-family:Papyrus">come howl!  19:25, 22 August 2017 (UTC)
 * Ah! Earlier than I thought then. I can remember collecting printout from the computer centre in the 1970s and you could tell which printer had been used depending upon £/#.  I don't remember the issue with golfball terminals, presumably they had the appropriate programming (#) or cash (£) golf ball fitted.  BTW, this was EBCDIC not ASCII yet the same confusion occurred. Martin of Sheffield (talk) 19:48, 22 August 2017 (UTC)

The gulfs had different balls. And still no movement on the documentation. I suggest a footnote on the first usage. -- PBS (talk) 16:42, 2 October 2017 (UTC)

Archival Resource Key parameter
Has anyone considered implementing an Archival Resource Key parameter? It's just another kind of ID comparable to the existing doi or jstor parameters, and would presumably be abbreviated ark. Daask (talk) 16:35, 24 January 2018 (UTC)

Some books have more than one ISBN
Some publishers produce their books in a hardcover edition for libraries, and in a paperback edition for the general public and bookstores. But besides the binding, the book is the same. The markup in price of the hardcover compared to the paperback provides extra revenue for the publisher, and is less than an individual hardcover binding commissioned by the library. The two editions have different ISBNs as the book trade needs to be able to distinquish both editions because of their different price. On the other hand, the same ISBN may be used for subsequent printings labelled as new edition even if the only change is the printing number and year in one of the first, identifying pages. I would like to add both ISBNs to a CITE BOOK or generic CITATION. Is this possible, if not could it please be made possible? --L.Willms (talk) 12:25, 4 December 2017 (UTC)
 * No. The cs1|2 templates are not intended, nor designed, to document multiple sources.  WP:SAYWHEREYOUGOTIT requires you to cite the source of information that you add to a Wikipedia article.  If you got it from the paperback edition, cite that and only that.  You can always add  outside of the cs1|2 template and still inside the <ref ></ref> tags.
 * —Trappist the monk (talk) 12:48, 4 December 2017 (UTC)
 * For citation purposes it is not relevant, if the cover of the book is hard or flexible. The source is the printed text which is page by page and line by line identical in every respect between hard and soft cover versions. The information in the REF is to help the user find the book either in the bookshop or in libraries. When only one of the two ISBNs is given, the user misses out half or more of the possibilities to find the source of the quote. On the other hand, you complained that identical reprints of a book have the same ISBN. Wrong. If the book is identical between the covers it is the smae source of the content. ISBN has been created as a tool for the book trade. For the book trade a different price is important, not the identical content. Identical content should be offered as one and the same soure, not as different ones. I don't know if you only program this "citation" macro or if you actually work with ISBN and other tools of the book trade one of referring Wikipedia readers to sources. I do, and I therefore care. Cheers, L.Willms (talk) 23:41, 4 December 2017 (UTC)
 * I agree, cover construction is irrelevant for a citation. Editors here and at Help talk:Citation Style 1 have reported that pagination and other content is ofttimes different between different covers (search the archives of these talk pages).  It is for this reason that cs1|2 do not support multiple isbns.
 * You wrote: "On the other hand, you complained that identical reprints of a book have the same ISBN." I have made no such claim nor complaint, please do not put words into my mouth that I have not spoken.
 * —Trappist the monk (talk) 01:13, 5 December 2017 (UTC)
 * —Trappist the monk (talk) 01:13, 5 December 2017 (UTC)

On a related point, some books have multiple ISBNs for the exact same physical printed volume. Springer-Verlag is known for assigning a 0-387-prefix (English) and 3-540-prefix (German) ISBN to the same book. (These prefixes also have the same modulo-11 checksums so the ISBNs have the same check digits.) For example, Proceedings of the European Conference on Computer Algebra (EUROCal '85) has ISBN 3-540-15983-5 and ISBN 0-387-15983-5. 23.83.37.241 (talk) 09:25, 24 December 2017 (UTC)

The practical solution for including multiple ISBNs, until the template handles them, is to use the ISBN template directly for the additional ISBNs, placing it after the main citation, and using the punctuation control in the citation template to handle the final punctuation as desired. Of course many ISBNs are added after the original reference, so there is usually no way to tell which version was originally consulted. On the other hand, if a single book has 8 or 10 different ISBNs for different printings and countries, including all of them may seem strange - in practice it is not a big problem to just choose one. &mdash; Carl (CBM · talk) 14:19, 24 December 2017 (UTC)

Book citations can (and most often should) indicate relevant page numbers. I've seen instances where the hardback and paperback editions of the same edition of a book had and different page numberingss, making them different variations of that book edition. Individual ISBNs are assigned to each edition and variation (except reprintings) of a book. Audiobook and E-book issuances of that same book edition would also have individual ISBNs. A citation should indicate the particular variation of the book edition where the information being cited was found (and most often should indicate where the information was found therein). It is outside the purpose of a citation to identify other variations of the cited book edition variation. Wtmitchell (talk) (earlier Boracay Bill) 22:36, 24 December 2017 (UTC)


 * It may however be within the scope of web citations, where space is not an issue but dead links and lack of physical libraries certainly are, to include alternate links should the editor wish. Currently I just include it in commented code for an editor to replace in case a link goes dead, but another convention might be briefly linking the alternate sources following the Harvard-style citation, using a common phrase like "Alternative locations". The difference would be that one could include it within the template itself, instead of having editors adding the phrase manually after the template. The markup might look like:
 * "Alternative locations: ISBN 1234, test[→], Your local library."
 * Thoughts? SamuelRiv (talk) 17:09, 25 March 2018 (UTC)
 * Thoughts? SamuelRiv (talk) 17:09, 25 March 2018 (UTC)

New identifier requests
The DOI foundation recognizes several international libraries with their own identifiers. There are tens of thousands articles in en.wp that cite Chinese and Japanese papers, so support for CNKI, JaLC, etc. would be useful. Code might be available already on zh.wp.

Also supporting the above poster in adding ark identifier from historical collections scanned at the Internet Archive (1921 example 13960/t12n61m3f). SamuelRiv (talk) 16:45, 25 March 2018 (UTC)


 * Additionally requesting Uniform Resource Name urn support, which among other things is used for several universities' classics catalogs. SamuelRiv (talk) 16:56, 1 April 2018 (UTC)
 * I think this request (and the similar one above it) is premature. The pathway to getting these supported is to start using them in the id parameter. Once we get a large body of citations using certain types of ids, it will become clearer that there is a need to support them directly. —David Eppstein (talk) 17:38, 1 April 2018 (UTC)

Unexpected citation check message
As can be seen below (as of the post date), the following reference: gives a 'check' message for the doi value. However, the doi link works properly. Praemonitus (talk) 15:14, 6 July 2018 (UTC)
 * bibcode is missing its pipe . There is also a missing pipe maintenance message that links to  – instructions on how to see those kinds of messages are on the category page.
 * —Trappist the monk (talk) 15:18, 6 July 2018 (UTC)
 * Ah, okay. Sorry for the bother then. Praemonitus (talk) 15:50, 6 July 2018 (UTC)

Regarding social media posts.
At current I am using as an example |publisher=Certain Football Club with |website=Twitter.com, quite often I am working out what the |title should be for the citation, I was wondering if we need a new element for social media posts that might help cover the title, Govvy (talk) 19:40, 6 July 2018 (UTC)
 * Hasn't this all been worked out with ?
 * —Trappist the monk (talk) 20:02, 6 July 2018 (UTC)
 * Okay, then maybe you can mention it in the main template article, because I read through the whole thing and there was no mention of cite twitter. Cheers. Govvy (talk) 20:08, 6 July 2018 (UTC)

Controlling whether "eds." or "ed." is added to editors parameter
In the specific-source citation template, which cites the Flora of North America among other resources from eFloras.org, I added Flora of North America Editorial Committee, eds. 1993+ when the FNA has been selected. But  is added, redundantly, when there are no authors.

I am also running into this problem in, where I am considering using National Plant Data Team (NPDT), eds. to avoid the typographical error of two consecutive parentheses:. Currently,  ends up being appended. But another solution is to remove the abbreviation.

Could a way be provided to remove  when it is not needed?

One idea is to make it so that  or   is only added if the pattern   is not found in the editors parameter. That would match "ed.", "eds." after non-letters (not a-z, A-Z, if the basic string functions are used). Or the frontier sets could be changed to, to match after spaces or punctuation. But maybe it would be a bad idea to automate it. — Eru·tuon 18:54, 14 July 2018 (UTC)
 * The obvious solution is to get rid of the redundant "(NPDT)". Headbomb {t · c · p · b} 19:10, 14 July 2018 (UTC)
 * Yes, as I said. But what about the first case I mentioned, Flora of North America Editorial Committee, eds. 1993+ in ? (Here I'm trying to follow the recommendation at http://floranorthamerica.org/cite.) — Eru·tuon 19:13, 14 July 2018 (UTC)


 * Again, get rid of "eds." (and likely 1993+ as well). Headbomb {t · c · p · b} 19:17, 14 July 2018 (UTC)
 * I suppose that is the only option if the proposed feature is not added. Still, it is somewhat frustrating that the citation templates cannot accommodate the recommended way of citing the editors. — Eru·tuon 19:36, 14 July 2018 (UTC)
 * Each publication sets its own citation style (in the case of Wikipedia, the editors of each article, by consensus, set the citation style). Sources do not set the citation style. A few sources give suggestions on how to cite them, and those are somewhat useful if they give you a few different suggestions for different styles (Chicago, APA, etc.). But those are nothing more than mild suggestions which you should feel free to ignore. Jc3s5h (talk) 19:41, 14 July 2018 (UTC)
 * If the editors are in the editors parameter, the text "eds." in the parameter's value is redundant. Omit it. – Jonesey95 (talk) 19:43, 14 July 2018 (UTC)
 * I'm thinking 1993+ is the number of editors. In that case, I'm not sure it would make sense without  before it. — Eru·tuon 19:54, 14 July 2018 (UTC)
 * 1993+ refers to the years of publication of the 20 volumes of the series. Headbomb {t · c · p · b} 19:58, 14 July 2018 (UTC)
 * Oh, I guess so. Somehow I got the idea that wasn't the case. — Eru·tuon 20:02, 14 July 2018 (UTC)
 * In that case, I withdraw this request. — Eru·tuon 20:10, 14 July 2018 (UTC)

Citing municipal ordinance
I need to cite a municipal ordinance, which has a title, chapter, and section. When I use the citation template, I get an error message telling me that chapter and section are duplicate entries. How can I enter all three parameters? Comfr (talk) 02:14, 20 October 2018 (UTC)
 * Here's one I did earlier (FYI, "1 & 2 Geo 5" refers to the Parliamentary year, each act therein forms a chapter):


 * Regards, Martin of Sheffield (talk) 08:01, 20 October 2018 (UTC)
 * According to the index, "Ventilation" encompasses §§29–31 so if citing §29 wouldn't it be better to write:
 * According to the index, "Ventilation" encompasses §§29–31 so if citing §29 wouldn't it be better to write:

<pre style="margin-left:4.8em">
 * The content of id is not included in the citation's metadata so those who consume this citation using reference management software will not get "1 & 2 Geo 5, chapter 50".
 * —Trappist the monk (talk) 11:26, 20 October 2018 (UTC)
 * Given that you are a leading template author I will bow to your judgement. It does seem a little odd though to have the chapter title (Coal Mines Act) and number (50) not in the "chapter" parameter, which was why I avoided it.  I'll go back and correct it shortly. Martin of Sheffield (talk) 19:56, 20 October 2018 (UTC)  ✅ Martin of Sheffield (talk) 21:43, 20 October 2018 (UTC)
 * I must not have been writing clearly. I asked a question; I did not declare that the form suggested in my question is somehow better or the best way to write such citations.  What I did write about the metadata is true and was not a question.
 * I don't know what '1 & 2 Geo 5' is or how it relates to Coal Mines Act; I gather that it has something to do with George V first and second year? If 'chapter 50' of 1 & 2 Geo 5 is "Coal Mines Act" then perhaps the citation might be written:
 * I don't know what '1 & 2 Geo 5' is or how it relates to Coal Mines Act; I gather that it has something to do with George V first and second year? If 'chapter 50' of 1 & 2 Geo 5 is "Coal Mines Act" then perhaps the citation might be written:
 * I don't know what '1 & 2 Geo 5' is or how it relates to Coal Mines Act; I gather that it has something to do with George V first and second year? If 'chapter 50' of 1 & 2 Geo 5 is "Coal Mines Act" then perhaps the citation might be written:

<pre style="margin-left:9.6em">
 * Because cs1|2 templates are general purpose templates, they are not suitable for all citation needs. We can try to coerce usable citations from them when the source doesn't fit the mold of Title "Chapter" in-source-location but we should not try to force the issue and write the citation by hand or use a template that is specifically designed to cite that source.
 * —Trappist the monk (talk) 12:17, 21 October 2018 (UTC)
 * For information (and thereby drifting a bit off topic): UK (and before that English) Statutes become law on the Royal Assent. Historically all Acts, proclamation and charters were related to the regnal year.  As Parliament developed acts were catalogued by the parliamentary year, but that in its turn was defined by the regnal years.  1 & 2 Geo 5 means the parliamentary year spanning the first and second years of the reign of George V.  An earlier example is the Habeas Corpus Act 1679: 31 Cha. 2 c. 2 (31st year of the reign of Charles II).  Within each session each act forms a chapter, hence the Bill of Rights Habeas Corpus Act was the second act passed that year, and the Coal Mines Act was the 50th Act passed in its year.  Regards, Martin of Sheffield (talk) 12:48, 21 October 2018 (UTC)
 * In law at any level (national, state/province, municipal) words such as title, chapter, and section have specific meaning within a particular jurisdiction. The laws themselves often use these words in ways that directly affect the meaning of the law. (For example, Act 160 of the Vermont 2017-2018 General Assembly contains "(H) relates to a matter subject to Title 4, 12, 13, 15, 18, 20, 23, or 33 of the Vermont Statutes Annotated").
 * I therefore believe it would be confusing to create metadata that labels anything as a title or chapter unless it actually is regarded by the relevant jurisdiction as a title or chapter. I'm not sure whether it would be better to create a citation that has the same format as a CS1 citation but isn't, or citations in the style suggested by Manual of Style/Legal. Jc3s5h (talk) 20:14, 20 October 2018 (UTC)
 * I therefore believe it would be confusing to create metadata that labels anything as a title or chapter unless it actually is regarded by the relevant jurisdiction as a title or chapter. I'm not sure whether it would be better to create a citation that has the same format as a CS1 citation but isn't, or citations in the style suggested by Manual of Style/Legal. Jc3s5h (talk) 20:14, 20 October 2018 (UTC)

I would like to cite a city ordinance in a standard way so that it conforms to the rules of Wikipedia, and is searchable using reference management software. The parameters needed are:

There are two problems with the citation template:


 * 1) It will not allow use of both chapter and section.
 * 2) title, chapter, and section have both a number and a description

Either the citation template should be fixed to accommodate more parameters, or another template should be created. I do not feel competent for either of these tasks. How can I get this in the queue? Comfr (talk) 23:38, 21 October 2018 (UTC)comfr
 * You might write it this way:

<pre style="margin-left:3.2em">
 * Title page of the document says that this document is titled Portland Zoning Code. I left out the ordinance because that is not a number that identifies this document.  Citing a living document is a pain.  The original date from the title page is 1 January 1991, the current version date is 22 August 2018, and the section is dated 24 May 2018.  Because you are linking to it, I presume that you are citing this version of the document so it seems to me that the whole-document date from the title page is the date you should be using in the citation.  Citing another version? Use the date from that version and link to that version.  Citing this version? Consider archiving it at archive.org or one of the other archive sites; provide archive links in the template.  The purpose of at is to identify in-source locations for which page and pages aren't suited.
 * —Trappist the monk (talk) 00:29, 22 October 2018 (UTC)
 * —Trappist the monk (talk) 00:29, 22 October 2018 (UTC)


 * I cited as you suggested. Thanks for leading the way.  Comfr (talk) 01:46, 22 October 2018 (UTC)

Automatic change - to –
It could be really nice if the template/module just changed - to – when adding page numbers. It gives a lot of noice when AWB-people afterwards run their bots. Christian75 (talk) 10:27, 21 October 2018 (UTC)
 * Not possible because Modules and templates cannot, are not allowed to, modify wiki text.
 * —Trappist the monk (talk) 11:35, 21 October 2018 (UTC)
 * I know. Maybe I was not clear enough. When somebody use this template with pages=1-100 the module could change the - to – in the output, and afterwards, no AWB-user has to fix the article. Christian75 (talk) 12:43, 21 October 2018 (UTC)
 * If you mean that the module should render a hyphenated page-range with an ndash in place of the hyphen, it does that:
 * There is nothing in that process that can prevent AWB users from replacing the hyphen in wikitext with an ndash.
 * —Trappist the monk (talk) 12:54, 21 October 2018 (UTC)
 * Why does the doc for "pages" param say "Separate using an en dash"? This seems unnecessary and inconvenient. And it gives justification to those bot users. Can we just remove this? Kendall-K1 (talk) 00:30, 22 October 2018 (UTC)
 * The doc says to use an en dash because MOS:NUMRANGE specifies an en dash for page ranges and other similar ranges. – Jonesey95 (talk) 04:47, 22 October 2018 (UTC)
 * NUMRANGE doesn't apply here. That's for text that appears in articles, not for input to templates. Kendall-K1 (talk) 05:50, 22 October 2018 (UTC)
 * NUMRANGE doesn't apply here. That's for text that appears in articles, not for input to templates. Kendall-K1 (talk) 05:50, 22 October 2018 (UTC)

Error in hiwiki
Please have a look at hi:साँचा:Citation/testcases and help in resolving the error. Thanks Capankajsmilyo(Talk 03:45, 6 November 2018 (UTC)
 * Not really an error. The templates/modules are responding correctly to what they have been given.  An editor at hi.wiki has done some work toward translating hi:Module:Citation/CS1/Configuration/sandbox so  hi:Template:Citation/sandbox should probably   hi:Module:Citation/CS1/sandbox (unless you are explicitly testing the code in hi:साँचा:Citation/sandbox).
 * —Trappist the monk (talk) 10:00, 6 November 2018 (UTC)

Italics in title
Does anyone know how to remove these? I'm working on an article that uses this template for news articles, so we need to use title= without italics. I could use chapter=, but it will look odd to other editors. There's a desire on the page to use this template, rather than cite news. Can anyone advise? SarahSV (talk) 18:35, 8 April 2019 (UTC)
 * Since forever, has treated any title as a book title when the template does not use any of: journal, magazine, newspaper, periodical, website, work or encyclopedia.  You are citing a news source, name the source with the appropriate parameter.
 * —Trappist the monk (talk) 19:01, 8 April 2019 (UTC)
 * Or . Headbomb {t · c · p · b} 19:05, 8 April 2019 (UTC)
 * Thank you both. I usually use cite news etc, so I'm not used to working with this one. How would you remove the italics in the title when publisher= is used (e.g. BBC News, Associated Press)?
 * SarahSV (talk) 20:15, 8 April 2019 (UTC)
 * Were it me, I wouldn't use BBC News because  is the work published by the British Broadcasting Corporation (which I would omit because it is so similar to to 'BBC News') so I would write:
 * —Trappist the monk (talk) 20:36, 8 April 2019 (UTC)
 * Work= places BBC News in italics, but BBC News isn't a work; it's not a work/title like The Guardian. If you don't like BBC News as an example, try Associated Press. That's the problem with the template. There's no obvious way to produce a non-italicized title and a non-italicized publisher. SarahSV (talk) 20:53, 8 April 2019 (UTC)
 * BBC News is the work published by BBC just as The Guardian is the work published by Guardian News & Media Limited. But, I recognize that some don't hold that opinion so I will not argue it any further – it's a glass-half-full-glass-wrong-size argument that will never end (though editors at each article  decide amongst themselves how to resolve the question for that article).
 * If we're talking about Decline in insect populations and this citation:
 * then perhaps it might be converted to like this:
 * In this version  moves to agency just as it would were you citing the same article in a newspaper.
 * But, because cannot be all things to all people, we can rewrite  so that it has the same style as, perhaps like this:
 * In this version Phys.org serves to alert readers that they aren't going to the AP website to read the article – Phys.org would do the same thing.
 * —Trappist the monk (talk) 22:15, 8 April 2019 (UTC)
 * Trappist, thank you, this is very helpful. The article's creator prefers to use citation, so I'm trying to make it work for all the different source categories. Another question: is there a way with that template to create a final period at the end of the citation? SarahSV (talk) 00:29, 9 April 2019 (UTC)
 * —Trappist the monk (talk) 00:41, 9 April 2019 (UTC)
 * Thank you! SarahSV (talk) 02:38, 9 April 2019 (UTC)
 * In this version  moves to agency just as it would were you citing the same article in a newspaper.
 * But, because cannot be all things to all people, we can rewrite  so that it has the same style as, perhaps like this:
 * In this version Phys.org serves to alert readers that they aren't going to the AP website to read the article – Phys.org would do the same thing.
 * —Trappist the monk (talk) 22:15, 8 April 2019 (UTC)
 * Trappist, thank you, this is very helpful. The article's creator prefers to use citation, so I'm trying to make it work for all the different source categories. Another question: is there a way with that template to create a final period at the end of the citation? SarahSV (talk) 00:29, 9 April 2019 (UTC)
 * —Trappist the monk (talk) 00:41, 9 April 2019 (UTC)
 * Thank you! SarahSV (talk) 02:38, 9 April 2019 (UTC)
 * In this version Phys.org serves to alert readers that they aren't going to the AP website to read the article – Phys.org would do the same thing.
 * —Trappist the monk (talk) 22:15, 8 April 2019 (UTC)
 * Trappist, thank you, this is very helpful. The article's creator prefers to use citation, so I'm trying to make it work for all the different source categories. Another question: is there a way with that template to create a final period at the end of the citation? SarahSV (talk) 00:29, 9 April 2019 (UTC)
 * —Trappist the monk (talk) 00:41, 9 April 2019 (UTC)
 * Thank you! SarahSV (talk) 02:38, 9 April 2019 (UTC)
 * Thank you! SarahSV (talk) 02:38, 9 April 2019 (UTC)

Citation parameter
Hello, The template citation in conjunction with sfn could use an additional parameter to bypass the auxilliary set of references.

I use shortened notes linked with sfn and citation templates, styled Sudirman, (the featured article on the day I joined). The benefits I find using sfn are no large interruptions in prose when editing, and the alphabetization of sources using citation. The only detriment is the auxilliary set of references in addition to sources (see Current, Citations and Sources). - NorthPark1417 (talk) 08:55, 9 March 2019 (UTC)

patent parameters appearing in citation templates
Recently I have been seeing issue-date in templates. These are occurring because patent citation parameters are listed in Template:Citation. They don't belong there because their presence in TemplateData makes ve think that they are legitimate parameters.

Why is this you ask? Initially I suspected that the patent part of was created to provide a cs2 counterpart to the cs1-styled. But, when is given any one of seven inventor surname aliases, or country-code, or , then the   is bypassed and the appropriate parameters from the template call are passed to  for rendering –  does the rendering for  so renderings are the same. ve doesn't know about the bypass, so it will happily add patent parameters to templates.

Here are a few comparisons of and  renderings:
 * – cite patent
 * – citation
 * – cite patent
 * – citation

this example shows that while both and  have rendering problems when url has a value, the  version ignores the value in url and uses the default url:
 * – cite patent
 * – citation

With the execption of the url issues noted above, the renderings of all three examples are identical.

How to fix this? Perhaps this: This list may be incomplete or even wrong but is a starting point.
 * 1) remove patent parameters from TemplateData to stop ve creation of invalid templates ✅
 * 2) replace patent templates in article space with  (~60 instances)
 * 3) remove patent support from
 * 4) fix the url rendering issue in – this could probably be done at any time

—Trappist the monk (talk) 11:49, 16 April 2019 (UTC)
 * I think that it is outdated and unintuitive to support patents using dedicated code in citation, and that we should deprecate and discontinue that usage in favor of cite patent. If the insource search above is returning valid results (always something to wonder), citation is used only 60 times to cite patents, and cite patent is used over 2,700 times, showing that editors prefer the obvious choice to the obscure one. Doing a bit of enhancement on cite patent would be welcome as well. – Jonesey95 (talk) 14:30, 16 April 2019 (UTC)
 * This TemplateData issue was noted in 2013 though it is unclear if anything came of that brief discussion.
 * If the patent bypass in goes away then the templates that  uses that are subpages of  should be moved to become subpages of .  These tasks would be:
 * delete currently unused [ list of Template-space transclusions]
 * move to new  [ list of Template-space transclusions]
 * move →  [ list of Template-space transclusions]
 * The other template that uses,  shall remain where it is because it is used by more than just  [ list of Template-space transclusions]
 * With all of this, I am thinking that it would be best to run this through WP:TfD.
 * —Trappist the monk (talk) 23:19, 17 April 2019 (UTC)
 * I agree with TfD and with all of the above. I will be on a wikibreak for a while, but I will contribute when I return if the discussion is still open. – Jonesey95 (talk) 04:31, 18 April 2019 (UTC)
 * —Trappist the monk (talk) 23:19, 17 April 2019 (UTC)
 * I agree with TfD and with all of the above. I will be on a wikibreak for a while, but I will contribute when I return if the discussion is still open. – Jonesey95 (talk) 04:31, 18 April 2019 (UTC)


 * —Trappist the monk (talk) 14:29, 18 April 2019 (UTC)

Bibcode
The adsabs site linked by the Bibcode entry now has a warning at the top that it is deprecated in May and is going way in October of this year. They provide two links to alternatives. Praemonitus (talk) 14:56, 1 May 2019 (UTC)
 * Duplicate of post at Help talk:Citation Style 1 . Edit: two topics merged in to Help talk:Citation Style 1. Answered there.
 * —Trappist the monk (talk) 15:12, 1 May 2019 (UTC) 15:36, 1 May 2019 (UTC)

patent citations
this template no longer supports patent citations; see this TfD.

alas, the insource: search results identified in the TfD turnout to be misleading; there are more articles that use the →  bypass than were expected. Because the bypass is no longer in place, emits a bunch of unknown parameter errors when it see the parameter specifically defined for. These show up in. I have an awb script that fixes the errors and I will troll through that category periodically to clean it out. For those who wish to repair these citations manually, simply change  to. For those who have awb access, the search string that I've been using is:

and the replace is:

—Trappist the monk (talk) 18:40, 6 May 2019 (UTC)

Citing norms and standards
There does not seem to be neither any documentation on how to cite a formal norm or standard with a number, optional part number and release date, issued by one of the (inter)national standardization bodies like ISO, CEN/CENELEC (EN), IEC, IEEE, ETSI, EBU, ANSI, BS, DIN, JIS etc. nor any support for those identifiers in the template code, only IETF RFCs are supported by a special parameter  (used as   in Template:Citation/identifier).

I could probably use generic, but I wish there was a way to make it simpler and generate more harmonized formatting. I found cite ISO standard, which seems very limited, and tried to add something similar to the respective sandbox by introducing a new identifier. I would like to see some general advice and automation.

Standards are reliable sources, of course, but only few are available for free, which makes them less popular references. On the other hand, several standards do have a distinguished Wikipedia article, e.g. ISO 9, or provide a redirect to a more generic description of their topic, e.g. Romanization of Cyrillic. The following mockup is overly verbose, but it shows metadata that could be used. — Christoph Päper 07:43, 24 April 2019 (UTC)


 * This is the talk page for Template:Citation so discussion about doesn't really belong here.  Since,  does not exist, if you think that there is sufficient need for such a template, write it.  If you do, post a notification here.
 * —Trappist the monk (talk) 12:02, 24 April 2019 (UTC)
 * I'll take it to Help talk:Citation Style 1. — Christoph Päper 10:01, 8 May 2019 (UTC)

Geographical restriction on web access
It would be useful to have an access-indicator parameter to indicate that a URL is geographically restricted. For example, some US websites (including the Baltimore Sun) are not available in many European countries, probably because of the EU Directive on Copyright in the Digital Single Market, and some BBC content is only available in the UK. Verbcatcher (talk) 22:38, 11 May 2019 (UTC)

RfC regarding italicization of the names of websites in citations and references
There is a request for comment about the italicization of the names of websites in citations and references at Help talk:Citation Style 1/Archive 72. Please contribute. SchreiberBike &#124; ⌨ 04:52, 18 May 2019 (UTC)

Subscription keyword
It appears that the subscription keyword has been deprecated, but how then can I indicate that a subscription is required? Hawkeye7  (discuss)  00:31, 23 June 2019 (UTC)
 * Does the help text in the deprecated parameter error message not answer that question?
 * —Trappist the monk (talk) 00:43, 23 June 2019 (UTC)
 * —Trappist the monk (talk) 00:43, 23 June 2019 (UTC)
 * —Trappist the monk (talk) 00:43, 23 June 2019 (UTC)

Where should links to url or doi params appear in the citation?
re User_talk:Nemo_bis and an ongoing mass removal of url in favour of doi. This is having the side-effect of making the link disappear from a prominent position in reference titles to where it's only displayed around doi etc, which is much less familiar to most readers.

IMHO, this needs a change to citation, cite book, cite journal,  cite web et al. If any parameter generating an external link is available, one of them should be used to generate a prominent link, such as from the title of the reference. This would emulate the previous and established behaviour, when url is provided. Andy Dingley (talk) 18:32, 19 July 2019 (UTC)


 * There is a related discussion here: Template_talk:Cite_book --Matthiaspaul (talk) 00:27, 14 August 2019 (UTC)

How do I cite a work which was edited, and then re-published by another editor?
I have found an odd situation: The book Gifts of Power: The Writings of Rebecca Jackson, Black Visionary, Shaker Eldress is a collection of autobiographical writings compiled and edited by Jean McMahon Humez. Where this gets complicated is that one of the original autobiographical writings was also edited and annotated for publication. Is it possible to distinguish between the editor of the autobiography (since he is a primary source), and the editor who put the collection together as an academic publication?--<b style="color:navy">3family6</b> (<u style="color:black">Talk to me &#124; <small style="color:purple">See what I have done ) 03:59, 20 August 2019 (UTC)
 * [Original pub], as cited in [New pub]. &#32; Headbomb {t · c · p · b} 04:15, 20 August 2019 (UTC)
 * So there is no "original editor" function? I supposed that "original publication" would work.--<b style="color:navy">3family6</b> (<u style="color:black">Talk to me &#124; <small style="color:purple">See what I have done ) 13:29, 20 August 2019 (UTC)
 * The difficulty is that the original documents were never published, they only exist in draft form. There was no publisher, only an editor who prepared and annotated the documents for publication.--<b style="color:navy">3family6</b> (<u style="color:black">Talk to me &#124; <small style="color:purple">See what I have done ) 13:33, 20 August 2019 (UTC)

Update to omit deprecated parameters ?
This template still lists deadurl= in the "Template parameters" table, about four-fifths of the way down the page. It is deprecated. Should it be removed from the table? —RCraig09 (talk) 16:36, 3 September 2019 (UTC)
 * P.S. Does anyone know if there is or will be a bot that will replace the zillions of cites that used deadurl? —RCraig09 (talk) 16:36, 3 September 2019 (UTC)
 * Yes, there will be a bot, if someone from the BAG takes action on it: Bots/Requests for approval/Monkbot 16. – Jonesey95 (talk) 17:00, 3 September 2019 (UTC)

WP:AN discussion regarding recent edits causing millions of red links
Here is a link to the discussions regarding the recent edits that are causing millions of redlinks to appear: There is also a brief VPT discussion (here) that points to the longer AN discussion linked above. Mathglot (talk) 06:03, 4 September 2019 (UTC)
 * WP:AN
 * Help talk:Citation Style 1
 * Bots/Requests for approval/Monkbot 16

ndash in dates
I just spent 20 minutes trying to figure out why  throws an error. Turns out only hyphen is accepted, not ndash. I don't want to trigger anyone's hyphen-ndash PTSD, but ndash is actually correct, and while I don't suggest rejecting hyphen, ndash really should be accepted. <b style="color: red;">E</b><b style="color: blue;">Eng</b> 02:03, 8 September 2019 (UTC)
 * I presume that you mean .  Yep, that does not work but the ndash character does:
 * Hyphens are converted to ndash characters and the article added to
 * Instruction to not use the html character entity is mentioned in the error message help text along with instructions on how to make an endash character.
 * —Trappist the monk (talk) 02:12, 8 September 2019 (UTC)
 * Hyphens are converted to ndash characters and the article added to
 * Instruction to not use the html character entity is mentioned in the error message help text along with instructions on how to make an endash character.
 * —Trappist the monk (talk) 02:12, 8 September 2019 (UTC)
 * Instruction to not use the html character entity is mentioned in the error message help text along with instructions on how to make an endash character.
 * —Trappist the monk (talk) 02:12, 8 September 2019 (UTC)
 * —Trappist the monk (talk) 02:12, 8 September 2019 (UTC)

block-level elements within quote
Is there any good way at all to do something like this:

(and somehow position the closing quotation mark at the end of line #2)? The actual example I'm looking at has  elements with manual numbering on a single line (which avoids the "line feed character" warning), e.g.  quote = Rules: 1. You do not talk about Fight Club. 2. You do not talk about Fight Club. whereas I'd like to get away from that. The actual text of each numbered line is long enough to wrap around, so left-margin (relative to the numbers) becomes a factor. ―cobaltcigs 21:10, 27 September 2019 (UTC)
 * I don't think so. I have always believed that quotes, if they are important enough for the article, should be in the article as quoted text and cited there or, in separate end notes and cited there; not part of the citations themselves.
 * In cs1|2, quotes are wrapped in <q ></q> tags so that css can be used to render quote marks appropriate to the language of the wiki where the cs1|2 module suite is used (Module:Citation/CS1/styles.css).
 * If you must have the quotation with the reference, the quotation can be placed between the cs1|2 template's closing  and the reference's  – quote marks can be omitted then or manually added as necessary.
 * —Trappist the monk (talk) 21:56, 27 September 2019 (UTC)
 * If you must have the quotation with the reference, the quotation can be placed between the cs1|2 template's closing  and the reference's  – quote marks can be omitted then or manually added as necessary.
 * —Trappist the monk (talk) 21:56, 27 September 2019 (UTC)

Agreed, and it wasn't my idea to put it there. Just trying to cleanup the output without upsetting whoever did put it there. I guess I'll put it after the template as suggested. ―cobaltcigs 22:21, 27 September 2019 (UTC)

Italic or bold markup not allowed
I see a new nag has appeared in citations across Wikipedia overnight, resplendent in reprimanding red:. Whatever the reasoning here, would it not be a simpler solution to strip out unwanted markup either at edit submission or at runtime? This error now relies on editors with the time and motivation to go around manually correcting it - if there is a Bot doing this job, I have yet to see it. Cnbrb (talk) 14:01, 28 September 2019 (UTC)
 * No. Very often, editors use wiki markup, particularly italic markup, to change the visual rendering of the value assigned to work and publisher parameters.  For example, it is common to see in  something like:
 * The New York Times
 * The correct form is:
 * The New York Times
 * which automatically italicizes the newspaper's name.
 * Additionally, all of the periodical templates (,, , and ) produce COinS metadata using that standard's journal object. The journal object does not have support for publisher metadata (though cs1|2 can and does display the value assigned to publisher).  When editors misuse publisher to hold the periodical name, they do a disservice to readers who consume cs1|2 citations by way of the metadata because those readers do not get the (rather important) periodical name with the rest of the citation's bibliographic detail.
 * The correct repair for these errors is to first make sure that the template is using the correct parameter (rename the parameter if required) and then remove the wiki markup.
 * Monkbot/task 14 can repair some of these however, that task is currently deferred while Monkbot/task 16 is clearing.
 * —Trappist the monk (talk) 14:26, 28 September 2019 (UTC)
 * Monkbot/task 14 can repair some of these however, that task is currently deferred while Monkbot/task 16 is clearing.
 * —Trappist the monk (talk) 14:26, 28 September 2019 (UTC)
 * —Trappist the monk (talk) 14:26, 28 September 2019 (UTC)

RFC regarding the scope of RfC regarding italicization of the names of websites in citations and references
Pursuant to a request by the closer:

There is a request for comment to definitively determine how widely the RFC Italics of websites in citations and references – request for comment should be applied. Please contribute.

—Trappist the monk (talk) 14:13, 6 October 2019 (UTC)

ISBNs and line wrapping
See screenshot. Apparently avoiding a line-break between the abbreviation "ISBN" and the first digit was deemed more critical than avoiding a line-break between digits. Seeing a partial ISBN at the end of the line seems a lot worse than seeing a complete ISBN at the beginning of the line (and not immediately realizing it's an ISBN).

Recommend something like: ISBN 978-0-520-04128-8 ↑                                         (regular space) with css: span.isbn { white-space: nowrap; } ―cobaltcigs 03:58, 8 October 2019 (UTC)
 * As far as I know, identifier names (e.g. ISBN, OCLC) are typically separated from the identifier itself (i.e. the numbers) by a non-breaking space, as you have observed. If I am reading Module:Catalog lookup link correctly, an nbsp is inserted after the identifier label. See Template:Catalog lookup link for a list of templates that use this convention. Citation also follows that convention, AFAIK, though it does not use Catalog lookup link. If you are successful at Template talk:Catalog lookup link in persuading editors to change the wrapping behavior, come on back here and let us know. [Disclaimer: I am making a series of assumptions here, one or more of which may be incorrect, in the interest of providing a helpful and reasonably timely answer. Other editors may need to correct my inaccuracies.] – Jonesey95 (talk) 05:10, 8 October 2019 (UTC)