Help talk:Citation Style 1

Citation templates

... in conception

... and in reality

Refactoring the code for the final rendering

Hello everyone! While translating the module into Russian, I encountered one issue - some participants do not want to use the CMS or APA style, but prefer styles that correspond to their own language. For Russian, this is GOST.

I attempted to fix the render [1], but it was very challenging because the entire rendering is scattered throughout the code and sometimes is completely illogical and opaque.

I believe that since individual languages and language projects have different standards for source representation (different component order, formatting, different sets of data), it would be very helpful to simplify editing this representation in a common module. Having the ability to remove italics and bold, as well as rearrange components, would be beneficial.

In ruwiki, there are specific templates that I want to transfer to this module, but it's necessary for them to be able to change the render when a certain parameter is entered. For example: citation_style = gost.

Source	{{cite journal |last1=Aries |first1=Myriam B. C. |last2=Newsham |first2=Guy R. |date=2008 |title=Effect of daylight saving time on lighting energy use: a literature review |url=http://archive.nrc-cnrc.gc.ca/obj/irc/doc/pubs/nrcc49212/nrcc49212.pdf |journal=Energy Policy |volume=36 |issue=6 |pages=1858–1866 |doi=10.1016/j.enpol.2007.05.021}}
APA/CMS	Aries, Myriam B. C.; Newsham, Guy R. (2008). "Effect of daylight saving time on lighting energy use: a literature review" (PDF). Energy Policy. 36 (6): 1858—1866. doi:10.1016/j.enpol.2007.05.021.
GOST	Aries Myriam B. C., Newsham Guy R. Effect of daylight saving time on lighting energy use: a literature review [PDF]. // Energy Policy. — 2008. — Vol. 36. — Is. 6. — P. 1858–1866. — doi:10.1016/j.enpol.2007.05.021.
DIN	Aries, Myriam B. C.; Newsham, Guy R.: "Effect of daylight saving time on lighting energy use: a literature review", Energy Policy, 2008, Vol. 36, Iss. 6, pp. 1858–1866. DOI: 10.1016/j.enpol.2007.05.021.

Iniquity (talk) 20:59, 19 February 2024 (UTC)

These templates are not CMS or APA but instead a mix of a few, and include other qualities that differ from either that we have added, so a bit of a false premise there.

Ignoring that, I have thought a few times about making more of the components of this system into 'libraries' that could be used from other citation styles. Maybe all the subpages can be today? I don't know and am pretty sure not based on some of the opinionated choices CS1 makes (mostly around "missing" information) which I'm pretty sure is in the subpages and not the primary module. If all the subpages were CS1 agnostic, that would get you most of the way there with potentially some duplication in the primary module. Izno (talk) 16:25, 18 March 2024 (UTC)

Questions about Category:CS1 maint: unflagged free DOI

The documentation at Category:CS1 maint: unflagged free DOI states that this issue can be resolved by adding |doi-access=free, but "Exception are pages with {{Academic peer reviewed}} or {{Cite Q}} on them, where the update needs to be done in Wikidata."

1. What specifically is the update that needs to be done in Wikidata?
2. Is it OK to simply add |doi-access=free to {{Cite Q}} templates? (e.g. this edit and this edit)

Thanks! GoingBatty (talk) 05:08, 23 February 2024 (UTC)

1. no idea. See this discussion.
2. yes, but ideally we would do something in Wikidata instead

Headbomb {t · c · p · b} 05:16, 24 February 2024 (UTC)

Cite book problem where "work" is not allowed

I'm not sure what was intended. I went to the archived source and it seems to be part of a book. Pearl Milling Company#cite note-BIA-1— Vchimpanzee • talk • contributions • 22:55, 23 February 2024 (UTC)

Looks like a junk citation to me:

{{cite book|last=Kern-Foxworth|first=Marilyn|title=Aunt Jemima, Uncle Ben and Rastus: Blacks in advertising, Yesterday, Today and Tomorrow|publisher=Connecticut and London: Greenwood Press|year=1994|url=http://testaae.greenwood.com/doc_print.aspx?fileID=GR5184&chapterID=GR5184-561&path=books/greenwood|archive-url=https://web.archive.org/web/20140424192836/http://testaae.greenwood.com/doc_print.aspx?fileID=GR5184&chapterID=GR5184-561&path=books%2Fgreenwood|archive-date=April 24, 2014|work=Public Relations Review|volume=16 (Fall):59}}

• template uses {{cite book}} but seems to be citing something (a review?) in Public Relations Review
• template links to what appears to be the book publisher: Greenwood Press
• at the bottom of the archive snapshot of the chapter(?) the publisher provides MLA and CMOS citations; neither mention Public Relations Review though the author does have an article in that journal that is used as a source for the 'cited' article/chapter/whatever. Note that the referenced article is is from the same journal issue as is mentioned in the junk citation:

  —— (Autumn 1990). "Plantation kitchen to American icon: Aunt Jemima". Public Relations Review. 16 (3): 55–67. doi:10.1016/S0363-8111(05)80069-4.

Have you discussed this with the editor who created the junk citation? (perhaps this edit at Aunt Jemima?) What is it that they are really trying to cite? A chapter in a book? A book review? A journal article?

Regardless, the citation is junk so the error message is correct and not the fault of {{cite book}}.

—Trappist the monk (talk) 00:07, 24 February 2024 (UTC)

They've combined two refs that are for different works. One is for a book called "Aunt Jemima, Uncle Ben, and Rastus: Blacks in Advertising, Yesterday, Today, and Tomorrow" and the other is an article titled "Plantation kitchen to American icon: Aunt Jemima". The article ref was being used to support a quote that is no longer used in either article, so the |work= and |volume= details can just be removed. -- LCU ActivelyDisinterested «@» °∆t° 00:55, 24 February 2024 (UTC)

I forgot I had even asked this question. So what should be done with the ref?— Vchimpanzee • talk • contributions • 16:19, 29 February 2024 (UTC)

Figure out which of the two sources best supports the en.wiki article text and then adjust the the template accordingly.

—Trappist the monk (talk) 16:35, 29 February 2024 (UTC)

So far the source that was linked to seems to support the content but there are six places it is used. "Aunt Jemima, Uncle Ben, and Rastus: Blacks in Advertising, Yesterday, Today, and Tomorrow" by Marilyn Kern-Foxworth but I can't tell what the publisher or work might be.— Vchimpanzee • talk • contributions • 17:36, 29 February 2024 (UTC)

I can't find "rice flour and corn sugar" in the source.— Vchimpanzee • talk • contributions • 17:40, 29 February 2024 (UTC)

Exclude errors in citations at AfD

I'm going through Category:CS1 errors: generic title and a decent chunk of them are in old, closed AfDs, e.g. Wikipedia:Articles for deletion/Indiavision news. I can't think of any case where I'd want to actually edit these. Is there a way to exclude them from display in the tracking category? It isn't actually a big deal, there's a grand total of 11 out of almost 3,000 pages, but as someone with vague aspirations to clear the category it itches. Rusalkii (talk) 22:22, 24 February 2024 (UTC)

Add |no-tracking=yes to the offending template(s). Error messaging will remain but the page will not included in tracking categories.

—Trappist the monk (talk) 22:48, 24 February 2024 (UTC)

To the individual AfD pages, or the AfD template as a whole? If the former, I don't see where I would add it. Rusalkii (talk) 23:16, 24 February 2024 (UTC)

to the offending template(s). So, in your example case, change:

{{cite web|url=http://wayback.archive.org/web/*/indiavision.com|title=wayback machine}}

to:

{{cite web|url=http://wayback.archive.org/web/*/indiavision.com|title=wayback machine |no-tracking=yes}}

—Trappist the monk (talk) 23:24, 24 February 2024 (UTC)

That would have been the obvious interpretation, wouldn't it. Thank you! Rusalkii (talk) 23:31, 24 February 2024 (UTC)

Module access

Is there a reason why Module:Citation/CS1 does not accept access from modules that pass "frame.args" and insists on requiring "frame:getParent().args" ? Are you open to allowing others to add such an feature ? Snævar (talk) 14:26, 25 February 2024 (UTC)

In cs1|2, each template {{#invoke}}s Module:Citation/CS1 with at least one parameter needed by the module to identify the calling template: {{cite book}}, {{cite journal}}, {{cite news}}, {{cite web}}, {{citation}}, etc. For example, {{cite book}} has this:

{{#invoke:citation/CS1|citation
|CitationClass=book
}}

|CitationClass= is passed to the module in the frame object. To get the parameters from this {{cite book}}:

{{cite book |title=Title |date=2024 |publisher=Publisher}}

the module must consult the parent frame object.

This was how the module has worked more-or-less from its inception. The primary and overriding purpose of the module is to support the cs1|2 templates. To the best of my knowledge, no one has proposed expanding that purpose to other modules. Why do you want to do it?

—Trappist the monk (talk) 15:28, 25 February 2024 (UTC)

It is quite common for modules to have access for other modules, both here on the English Wikipedia and on other sites like Wikimedia Commons. Citation/CS1 is going against the wind on this one. One module, or a set of them in this case, does not set the norms, the majority of the modules do. As for usecases, they are allready here, there are a lot of modules and templates due to this module access not being in place. These modules and templates are: Module:Cite arXiv, Module:Cite bioRxiv, Module:Cite book, Module:Cite conference, Module:Cite document, Module:Cite encyclopedia, Module:Cite episode, Module:Cite interview, Module:Cite journal, Module:Cite magazine, Module:Cite mailing list, Module:Cite map, Module:Cite medRxiv, Module:Cite news, Module:Cite newsgroup, Module:Cite podcast, Module:Cite press release, Module:Cite report, Module:Cite serial, Module:Cite sign, Module:Cite speech, Module:Cite tech report, Module:Cite thesis and Module:Cite web. As for the templates it is a bunch of substitution templates with similar names, although I do know the reason for their existance is a bit different and unlike the modules might be justified. I am not trying to offend anyone by saying this, but this list of modules feels like an garbage dump. The only difference between these modules is the CitationClass and what title they return.

@User:Pppery: I think you are going to like this. Is it not worth getting rid of 24 nearly identical modules, by merging them into one? Snævar (talk) 05:10, 1 March 2024 (UTC)

"The only difference between these modules is the CitationClass and what title they return." There's a lot more differences than class and 'title', whatever you mean by that. The different modules accept different parameters, and format them differently. {{cite arxiv}} for instance, does not support |journal/volume/issue/isbn= etc... Headbomb {t · c · p · b} 20:31, 16 March 2024 (UTC)

Two-part news article

Hello! I'm trying to optimize and thought I would consult w you guys. I do a lot of citations from newspaper archives. In some case the article starts on one page and then is "continued on page 35" or whatever. I like to include links to both clipping URLs in one reference because they're one entity in my mind. I usually end up with something like this:

<ref name="Kathman-1990">{{Cite news |last=Kathman |first=Janice |date=1990-04-26 |title=Popularity of Bottled Water Can Be Traced Back to Alhambra Valley Springs [part 1 of 2] |url=https://www.newspapers.com/article/martinez-news-gazette-popularity-of-bott/142039526/ |work=Martinez News-Gazette |location=Martinez, California |page=1 |volume=132 |issue=82}} & {{Cite news |title=Alhambra Water [part 2 of 2] |url=https://www.newspapers.com/article/martinez-news-gazette-alhambra-water/142039410/ |page=2}}</ref>

Is there any more refined way to do this, other than continuing to put [part X of Y] in brackets manually?

Thanks in advance for any guidance you can offer. best, jengod (talk) 04:00, 26 February 2024 (UTC)

@Jengod: what I usually do, using your example, is something like:

<ref name="Kathman-1990">{{Cite news |last=Kathman |first=Janice |date=1990-04-26 |title=Popularity of Bottled Water Can Be Traced Back to Alhambra Valley Springs |url=https://www.newspapers.com/article/martinez-news-gazette-popularity-of-bott/142039526/ |work=Martinez News-Gazette |location=Martinez, California |pages=1, [https://www.newspapers.com/article/martinez-news-gazette-alhambra-water/142039410/ 2] |volume=132 |issue=82}}</ref>

One citation for what is really one source, and yet both pages' clippings are linked. Imzadi 1979 → 05:30, 26 February 2024 (UTC)

@Imzadi1979 ooh I like that--very clean! TY!! jengod (talk) 05:34, 26 February 2024 (UTC)

Inconsistent handling of ref=sfnref on desktop v mobile

If a citation uses a quoted name in sfnref (e.g., |ref={{sfnref|"Technical Report"|2021}}, it works fine and looks fine in desktop view. But viewed on mobile, it generates a harv error: the citations see no target source and the source sees no incoming citations. For a real world example, see https://en.wikipedia.org/w/index.php?title=East_West_Rail&oldid=1210714109 . Removing the quotes fixes the problem. (In case it matters, "mobile" means Android+Chrome, "desktop" means ChromeBook+Chrome.)

Presumably this needs reporting somewhere, but where? 𝕁𝕄𝔽 (talk) 11:00, 28 February 2024 (UTC)

That's bizarre. For convenience, here's a link that shows the offending behaviour via "Mobile view" on a desktop. I notice however, that the click-function 'References -> Sources' works as expected. -- Michael Bednarek (talk) 13:03, 28 February 2024 (UTC)

I don't see any error on desktop or mobile, but I'm only using the basic error messages rather than using a acript. This could be an issue with script your using, I suggest reporting the error at User talk:Trappist the monk/HarvErrors. -- LCU ActivelyDisinterested «@» °∆t° 14:40, 28 February 2024 (UTC)

(edit conflict)

Unlikely that this a cs1|2 or Module:Footnotes issue because neither distinguish mobile view from desktop view. I suspect that this is the same problem described at phab:T348928 where MediaWiki is incorrectly url-encoding the short-form link when it should be anchor-encoding the link:

{{urlencode:CITEREF"Technical Report"2021}} → CITEREF%22Technical+Report%222021

{{anchorencode:CITEREF"Technical Report"2021}} → CITEREF"Technical_Report"2021

—Trappist the monk (talk) 14:45, 28 February 2024 (UTC)

PMID limit increase

Please see the page 2024 in arthropod paleontology - the PMID limit should be increase, as it gives false positives on PMID values that are over 38400000, such as 38401545. Maxim Masiutin (talk) 23:54, 29 February 2024 (UTC)

Edition ordinals

This is a repeat of a proposal that has been made twice before:

• Help talk:Citation Style 1/Archive 11 § Suggestion for edition= parameter to treat raw numbers
• Help talk:Citation Style 1/Archive 64 § Proposal to enhance edition= parameter to support special numerical symbols

I quote Matthiaspaul:

This would help to further decouple semantics (which edition?) from presentation (f.e. "3rd ed."). It would not only make it easier to add common edition information, but also improve readability, maintainability and translatability, and it would allow to centrally change the rendering in the future, would this become necessary ("3rd ed.", "third ed.", "third edition" etc.), depending on the output device (f.e., display the abbreviated form "3rd ed." on the small display of a mobile device, but "third edition" on a desktop or printout), or target language (e.g. "third edition", "dritte Ausgabe", etc.).

Jc3s5h suggests that sometimes, particularly for software, there is a "3 edition" which is distinct from a "3rd edition". As a native English speaker I don't perceive a difference between "3rd edition" or "edition 3". They are essentially interchangeable. "3 edition" is just wrong. If the edition name is "3", then it should be displayed as "3rd ed.".

Trappist the monk raised concerns about "other languages where the cs1|2 module suite is used". While I want to respect the challenge of maintainability of the CS1 suite across Wikipedia sites in various languages, and am wholly unaware of the systems in place for this, I don't think this needs to hold us back here. Changing this could be as simple as adding the following to Template:Cite book when it invokes the module, without even modifying the module to be English-specific.

{{Ifnumber|{{{edition|}}}|{{Ordinal|{{{edition}}}}}|{{{edition|}}}}}

I find Matthiaspaul's description of the benefits to be compelling, and don't understand the downsides to this feature. Daask (talk) 21:33, 1 March 2024 (UTC)

I don't know the context of the line of code provided by Daask, or what programming language it is in. But if "Ifnumber" does what I think it does it would be unacceptable. Here is an example of a citation that would be a problem:

Cite book comparison

Wikitext	{{cite book|date=2010|edition=2011|isbn=978 1 11 154223 8|location=Quincy, MA|publisher=National Fire Protection Association|title=NFPA 70: National Electrical code}}
Live	NFPA 70: National Electrical code (2011 ed.). Quincy, MA: National Fire Protection Association. 2010. ISBN 978 1 11 154223 8.

Jc3s5h (talk) 22:12, 1 March 2024 (UTC)

The proposed (pseudo?)code does not reflect the discussion. As far as I can tell from reading the linked discussion, that citation would not be affected, since the edition number is greater than 99. – Jonesey95 (talk) 20:49, 2 March 2024 (UTC)

Problem with Template:Haber kaynağı

Was helping a user with Draft:Halkalı-Bahçeşehir Rail System and noticed that the bot didn't exactly subst the template properly. Is this an issue with the wrapper or with the original placement of the template? Primefac (talk) 13:53, 2 March 2024 (UTC)

Oneida is missing from the recognized languages

Oneida has the ISO 639-3 code one. Can this be added so Oneida sources no longer show up under Category:CS1_maint:unrecognized language?

References

Snowman304|talk 05:22, 7 March 2024 (UTC)

Not recognized by MediaWiki:

one ← {{#language:one|en}}

so you'll have to spell it out: |language=Oneida.

—Trappist the monk (talk) 13:04, 7 March 2024 (UTC)

Hi Trappist, thanks for the quick response. I'm looking at Morris Swadesh, and I can't figure out what the unrecognized language would be except for Oneida, which is spelled out. Snowman304|talk 19:22, 7 March 2024 (UTC)

That's the one (pun not really intended). cs1|2 emits the maintenance message because Oneida is not a language name recognized by MediaWiki. The whole list of MediaWiki-recognized languages and their tags is at Template:Citation Style documentation/language/doc.

See Help:CS1 errors § Controlling error message display to show the cs1|2 maintenance messages.

—Trappist the monk (talk) 20:18, 7 March 2024 (UTC)

OCLC limit hit

See for example OCLC 10146270069, which exceeds the current limit of 10100000000. Headbomb {t · c · p · b} 01:54, 11 March 2024 (UTC)

Same for PMC limits, which now exceed 10900000. Headbomb {t · c · p · b} 02:00, 11 March 2024 (UTC)

Nonsensical error when title=none is set

The following gives a "CS1 maint: untitled periodical" maintenance message, but clearly the periodical is named here.

• Torrence, Eve (October 2019). Journal of Mathematics and the Arts. 14 (3): 283–284. doi:10.1080/17513472.2019.1666459. S2CID 209985329.{{cite journal}}: CS1 maint: untitled periodical (link)

Headbomb {t · c · p · b} 02:12, 11 March 2024 (UTC)

The warning message is unfortunately worded. What it should say is "article title missing in periodical" or something similar. -- Michael Bednarek (talk) 02:26, 11 March 2024 (UTC)

Which would also be an error, because the title is willingly bypassed. Headbomb {t · c · p · b} 04:41, 11 March 2024 (UTC)

S2CID still needs an update

Looks like someone last mentioned this a few months ago. The S2CID limit is still too low. For example, https://en.wikipedia.org/wiki/Pry_(novel) has a working article with an ID of 268071715. Can we maybe bump it up to 269000000 (or 270000000, if we're feeling cheeky)? Snowman304|talk 07:27, 11 March 2024 (UTC)

url-status when archive-url is also dead

When you set url-status to "dead", it signals to use the archive-url by default. But what do you do when the archive-url is also dead, and you can't immediately find a working archive? Mokadoshi (talk) 12:49, 13 March 2024 (UTC)

The best solution would be to find a new source, but otherwise I would suggest removing the archive parameters (|archive-url=, |archive-date= and |url-status=), and adding {{dead link}} with the |fix-attempted=yes parameter after the cite. Defunct archive URLs are pretty worthless. -- LCU ActivelyDisinterested «@» °∆t° 15:51, 13 March 2024 (UTC)

• You'd also need to update iabot.org otherwise IABot will likely re-add the non-working archive URL. -- GreenC 16:12, 13 March 2024 (UTC)

  Are you sure that IABot would do this on a link that is marked permanently dead? And if so, how exactly do I make this fix? Mokadoshi (talk) 16:29, 13 March 2024 (UTC)

I agree. Thanks! Mokadoshi (talk) 16:18, 13 March 2024 (UTC)

extended definitions for Latn character set

This template fails because the character 'ạ' (U+1EA1 ạ LATIN SMALL LETTER A WITH DOT BELOW) is not recognized by the Vancouver tests (Vancouver allows only Latin characters):

Cite book comparison

Wikitext	{{cite book|first=Lê Cao|last=Đại|name-list-style=vanc|publisher=Vietnam Red Cross Society|title=Agent Orange in the Vietnam War: History and Consequences|year=2000}}
Live	Đại LC (2000). Agent Orange in the Vietnam War: History and Consequences. Vietnam Red Cross Society.
Sandbox	Đại LC (2000). Agent Orange in the Vietnam War: History and Consequences. Vietnam Red Cross Society.

The fix supports characters from Latin Extended Additional (U+1E00–U+1EFF).

—Trappist the monk (talk) 23:26, 13 March 2024 (UTC)

module suite update 23–24 March 2024

I propose to update cs1|2 module suite over the weekend 23–24 March 2024. Here are the changes:

Module:Citation/CS1

• removed temporary Julian–Gregorian uncertainty categorization; discussion
• combine extra-text tests for |volume= and |issue=; discussion
• fix bug related to hyphenated given names when reducing to initials for vancouver style; discussion
• add |script-encyclopedia= and |trans-encyclopedia=; discussion
• allow |mode=cs1 and |postscript=none in {{citation}}; discussion
• fix long-term-sleeping bibcode/postscript interaction bug; discussion
• fix archive.today timestamp check; discussion
• cleanup tcommon assignments; discussion
• extend latn char definition; discussion

Module:Citation/CS1/Configuration

• add doi free registrants: 1045 - D-Lib Magazine; 1074 and 1194 - American Society for Biochemistry and Molecular Biology; 1096 - FASEB; 4249 - Scholarpedia; 5210 - University of Illinois Libraries; 7759 - Cureus; 14256 - Croatian Association of Civil Engineers; 15347 - Wikijournals; 22323 - SISSA
• removed temporary Julian–Gregorian uncertainty categorization
• combine extra-text tests for |volume= and |issue=
• add |script-encyclopedia= and |trans-encyclopedia=
• use tabular data file at commons for identifier limit values; discussion
• removed doi free registrant 3410 - F1000; discussion
• extend latn char definition;

Module:Citation/CS1/Whitelist

• add |script-encyclopedia= and |trans-encyclopedia=

Module:Citation/CS1/Date validation

• removed temporary Julian–Gregorian uncertainty categorization

Module:Citation/CS1/styles.css

• revise access and wikisource icon css; discussion

—Trappist the monk (talk) 14:27, 16 March 2024 (UTC)

I've specified that 3410 is F1000 for future discussion. It's still a free registrant, it's just doing some fuckery with its access pages. Headbomb {t · c · p · b} 20:06, 16 March 2024 (UTC)

Re: Help talk:Citation Style 1/Archive 93#Limits: the way to incorporate this into Wikidata for each identifier would be to use property constraint (P2302) > range constraint (Q21510860) > maximum value (P2312) & minimum value (P2313), per d:Help:Property constraints portal/Range#Example 1. Regarding fragility & vandalism, I think using Wikidata would be superior to c:Data:CS1/Identifier limits.tab, if only for the # of people already watching, for example, OCLC control number (P243), etc.   ~ Tom.Reding (talk ⋅dgaf)  11:54, 17 March 2024 (UTC)

Thanks for that. If I understand (not saying that I do), OCLC control number (P243) would get a new constraint as a Qid to be used on only seven identifier properties. The new Qid might be something like 'cs1 limit constraint'. Each of the seven identifier properties would get the new Qid constraint with a maximum value (P2312) constraint property. That would require us to go on bended knee to the Masters of Wikidata to plead for the new Qid. Pleading for a new Qid is the 'politics' to which I referred in the original discussion; politics that I would prefer to avoid. Tabular data at commons doesn't require a plea to the Masters of Commons; it is one file and it works.

—Trappist the monk (talk) 16:17, 17 March 2024 (UTC)

No new Qid and no pleading required! I can add all the appropriate min/max constraints to Wikidata. I just need to know the minimum values allowed for all of the identifiers @ c:Data:CS1/Identifier limits.tab.   ~ Tom.Reding (talk ⋅dgaf)  00:21, 18 March 2024 (UTC)

I've added the appropriate maximum value (P2312) to OCLC control number (P243), OSTI article ID (P3894), PMCID (P932), PubMed ID (P698), RfC ID (P892), SSRN article ID (P893), & Semantic Scholar paper ID (P4011), effectively duplicating c:Data:CS1/Identifier limits.tab in Wikidata.   ~ Tom.Reding (talk ⋅dgaf)  17:43, 18 March 2024 (UTC)

I don't think that was a wise thing to do. I don't think that we should apply a constraint that isn't qualified to be a cs1|2-only constraint. What you have done, I think, is apply the cs1|2-only limits as constraints for any use of those identifier properties. That is why I said that I think to use wikidata we must have a cs1|2 Qid and to do that we must petition the Masters of Wikidata on bended knee.

—Trappist the monk (talk) 17:59, 18 March 2024 (UTC)

What you have done, I think, is apply the cs1|2-only limits as constraints for any use of those identifier properties. - correct. I don't see why OOB values should exist on Wikidata (unless that limit has yet to be updated after a recent bump). If there are no legitimate uses for OOB values, then I don't see why those limits shouldn't be in Wikidata, regardless of whether or not cs1|2 uses them. If there are legitimate reasons, either I or someone else will remove them.   ~ Tom.Reding (talk ⋅dgaf)  18:13, 18 March 2024 (UTC)

I'm still not convinced but that may be a moot point. Unless I'm missing something, Wikibase does not provide a mechanism for getting a property of a property that isn't part of an entity (Qid). See Wikibase Lua doc. Yeah, if you know the title of the work and it matches the label used in Wikidata, you can get a Qid (mw.wikibase.getEntityIdForTitle ('The Decameron') → Q16438 but the name in Italian: mw.wikibase.getEntityIdForTitle ('Il Decamerone') → nil). We must always be able to get to the identifier limits; we can do that easily with commons tabular data.

—Trappist the monk (talk) 23:19, 18 March 2024 (UTC)

@Izno: Whatever it is that you did to Module:Citation/CS1/sandbox/styles.css has broken the css for minerva (icons too big), monobook (icons clipped), and timeless (icons too big) skins. See these testcases at my sandbox (permalink):

• minerva
• monobook
• timeless
• vector
• vector 2022

—Trappist the monk (talk) 17:15, 18 March 2024 (UTC)

Yes, I know. I'm not done. Izno (talk) 17:16, 18 March 2024 (UTC)

Sorted. Izno (talk) 17:21, 18 March 2024 (UTC)

Ummm, really? Minerva, monobook, and timeless still looked borked to me.

—Trappist the monk (talk) 17:33, 18 March 2024 (UTC)

Your revisions are stuck in parser cache. Review my user page instead in the various skins. Izno (talk) 18:12, 18 March 2024 (UTC)

Just for ease:

• vector
• 22
• timeless
• minerva
• monobook

Izno (talk) 18:14, 18 March 2024 (UTC)

Yep.

—Trappist the monk (talk) 23:19, 18 March 2024 (UTC)

Best practice for when to use "n.d." vs. leaving blank

The documentation for CS1 currently includes When a source does not have a publication date, use |date=n.d.. However, for citations of things like web pages that don't have a date, I find it much more common to just leave out the date parameter rather than specifying that it has been omitted.

I'm curious to hear from others, what do you think the best practice should be for this? Sdkb ^talk 21:35, 17 March 2024 (UTC)

(I see in the archives that there has also been discussion about possibly using "undated" instead of "n.d." or adding a tooltip. If folks want to pick that up, please open a subthread so that we can keep everything organized. Cheers, Sdkb ^talk 21:37, 17 March 2024 (UTC))

Blank is definitely common, but blank can mean, "there was a date, but I didn't fill it in" or "it is undated". An "n.d." clarifies, so is better. ―Justin (koavf)❤T☮C☺M☯ 23:19, 17 March 2024 (UTC)

I've seen |date=<!--no date-->, which at least clarifies it for editors. Sdkb ^talk 23:42, 17 March 2024 (UTC)

I guess the thing that makes me uncomfortable about it is that we don't call out the absence of any other element in CS1 citations that I know of. If there's no DOI, or issue number, or second author, we don't designate that, but rather just leave the space blank. I'm open to being persuaded, but my intuition is that it'd be most consistent to adopt the same stance here (or at least to specify something like that "n.d." should only be used in situations where readers might normally expect there to be a date). Sdkb ^talk 23:48, 17 March 2024 (UTC)

In principle, everything that is published is published on some date (except maybe something like search results that are generated on a certain date, but if we're talking that, the access date is identical), but virtually any other field is something that could not be there (not URI for a web page, of course). ―Justin (koavf)❤T☮C☺M☯ 00:01, 18 March 2024 (UTC)

@Indagate: Please undo your edit to the template and participate here. ―Justin (koavf)❤T☮C☺M☯ 13:50, 18 March 2024 (UTC)

Didn't see this before reverting you there, seems best to leave the status quo for now until consensus is established for its inclusion. It's far from standard practice for web pages without a date to include n.d., pages are updated without any date being updated so the access-date in WP articles should be the latest date the information was updated if updated properly by editors. Including n.d. doesn't give readers any more information so seems redundant. Indagate (talk) 14:22, 18 March 2024 (UTC)

Did you see the instructions at Template:Cite web? There is nothing redundant here. ―Justin (koavf)❤T☮C☺M☯ 14:25, 18 March 2024 (UTC)

Yes, I read the parts about date there, but don't think I've seen any cite web using date=n.d. for websites without a date like Rotten Tomatoes so it doesn't reflect standard practice so doesn't seem to have consensus already Indagate (talk) 14:39, 18 March 2024 (UTC)

There are a lot of rules that are broken or ignored: that doesn't justify breaking or ignoring them further. If there is some reason why we should not use "n.d.", then that should be incorporated into the template instructions. ―Justin (koavf)❤T☮C☺M☯ 15:19, 18 March 2024 (UTC)

When there is a conflict between what the rules appear to say and what general practice is, as here, it's reasonable to keep the status quo while we work to resolve the underlying issue and figure out what guidance should say. This discussion is for figuring that out; let's stay focused on that. Sdkb ^talk 15:42, 18 March 2024 (UTC)

Don't forget that {{sfnp}} and related need something in the date field, even if it is means having to use {{sfnp|Doe|n.d.|page=123}}. The notation n.d. is reasonably well recognised and I don't know of any other conventions that are. But I guess it can be left to whoever is pulling together multiple citations of the same source to backfill it. --𝕁𝕄𝔽 (talk) 17:08, 18 March 2024 (UTC)

Category:CS1 maint: unflagged free DOI

This one. What is the purpose of this category? Jo-Jo Eumerus (talk) 17:14, 18 March 2024 (UTC)

Does the documentation at Category:CS1 maint: unflagged free DOI not answer your question?

—Trappist the monk (talk) 17:17, 18 March 2024 (UTC)

Only the technical purpose, not why there is a maintenance category instead of automatically flagging all free DOIs. Jo-Jo Eumerus (talk) 10:49, 19 March 2024 (UTC)

The implementing discussion is at Help talk:Citation Style 1/Archive 90 § Recognize free DOI prefixes.

—Trappist the monk (talk) 14:04, 19 March 2024 (UTC)

Add publication-date parameter

The {{cite book}} template already implements a separate publication-date parameter from CS1. Some journals and their articles have different publication-date, date and orig-date values. The latter is more flexible, but insufficient when there are three different dates, i.e. an article "orig-date=written 1930", published in a journal edition dated "date=2004-2005" and published "publication-date=2006". Ivan (talk) 18:00, 19 March 2024 (UTC)

Then the date is 2004-2005. The others are irrelevant. Headbomb {t · c · p · b} 07:47, 21 March 2024 (UTC)

In the case I have given, sources referencing it casually (i.e. without citation) will sometimes reference it with "1930", others with "2005", others with "2006", and if it was published in an issue of a periodical in 2004 whose contents were later published together with all other issues in 2006, then one will also encounter "2004". All for the same article.

For citations, it might not matter much. But for bibliography articles, precision helps the reader avoid confusion. I am not a template editor. How would one go about incorporating the parameter for {{cite journal}}? This is the correct page to ask? Ivan (talk) 17:55, 21 March 2024 (UTC)

Cite what you read. If you read an article in a periodical that was published in 2004, the date is 2004. If you were told that the article was republished in a set of bound things that contained all the articles the journal ever published, and that set was published in 2006, you don't care, you didn't read it. If I obtain the set of bound things and read the article, I'll put a publication date of 2006 in my citation. I could also add a parameter orig-date=2004. Jc3s5h (talk) 01:58, 22 March 2024 (UTC)

Cite what you read. Pardon my confusing wording above, but the need does not stem from the requirements of references but of bibliographies, where the works in question are the subject of the article. Some citation styles prefer the date of printing/publication (i.e. 1931), while others prefer the year published for (i.e. 1929/1930). It is not for the bibliographer to decide which must be used, and including both can help resolve confusion from the reader over which date to use given the style required of them. Posthumously published works excepted, most citation styles do not take the date of writing or reception into account, which would free up the orig-date parameter for the publication date, but in bibliographies that parameter is still needed to provide useful information to the reader; for example that the discrepancy between the publication date and that of a familiar author's death is due to posthumous publication. Ivan (talk) 03:02, 22 March 2024 (UTC)

There's no different requirement in biographies. Something written in 1923 but published in 2012 has a date of 2012. Headbomb {t · c · p · b} 03:08, 22 March 2024 (UTC)

I am not referring to bibliography sections at the end of articles, but to annotated bibliography articles. The purpose is different. Something published 2012 but with a year number 2010/2011 can have a date of 2012, 2011, or even 2010 in the case I referred to in the second paragraph of this section, all depending on the citation style required in the context. But more scientific bibliography articles do not make the choice for the reader. And the only reason I mentioned date of writing is to show you why we cannot always simply add =published to the orig-date parameter.

One of a dedicated bibliography article's purposes is to help the reader find the source detailed therein regardless of the system used in their library's catalogue, or any parametric restrictions imposed by said catalogue, or whichever means they end up using to find it. And to make them aware of any republications, English translations, and so on. I mentioned annotated bibliographies to give you an idea of how extensive the citations in those bibliographies can be on Wikipedia, just in case you are unaware of their existence on the project. Ivan (talk) 03:30, 22 March 2024 (UTC)

It sounds like this article is a rare case that needs additional annotations outside of the citation template. Citation templates are not intended to handle all possible edge cases. – Jonesey95 (talk) 13:11, 22 March 2024 (UTC)

Why is archive-date required when archive-url has the date?

Maybe a stupid question, but why is |archive-date required and then checked for matching |archive-url even when the latter already contains the date, like archive.org urls normally do? Wouldn't it make sense to forgo the separate date and just take the one from the url? Or, if it's expected by some processing somewhere else, at least automatically populate the date upon submission instead of rising an error? Since subst exists, this can't be because there's a policy against modifying the user-submitted wikitext. 82.131.19.61 (talk) 14:02, 22 March 2024 (UTC)

There may be archiving sites that format their URLs differently to archive.org. Certainly other archive organisations have been used in the past.Nigel Ish (talk) 16:21, 22 March 2024 (UTC)

Well, I think what I said would also apply to any other widely used archive that has timestamps included in its urls. With all of them, the date could be taken from the url. I'm not sure if the date is actually checked with any others besides archive.org, but it's this "we must check the date against the url if entered, but we won't populate the date from the url" one-way logic that I don't quite get. I seems like inconveniencing the user for no good reason (that I can think of). If the data is available, why error out and nag? 82.131.19.61 (talk) 16:55, 22 March 2024 (UTC)