Help talk:Citation Style 1/Archive 88

Possible invalid edits to live CS1 module
I'm not sure why modified a live CS1 module with no discussion or apparent testing, but I believe that one or both of these edits were invalid and should be reverted. The "empty unknown parameter" message currently renders as "Cite has empty unknown parameter : |fake=", which is clearly wrong. – Jonesey95 (talk) 04:06, 14 March 2023 (UTC)


 * Hi, Jonesey95. You're correct, I should have first tested this. I have fixed this error, while the second message is now displayed correctly (see e.g. here). --TadejM my talk 04:22, 14 March 2023 (UTC)
 * Because these modules are used in millions of pages, we nearly always accumulate fixes in the sandbox versions of the modules, test them adequately, and then deploy changes in batches every few months. – Jonesey95 (talk) 04:38, 14 March 2023 (UTC)
 * Noted, thank you. --TadejM my talk 06:07, 14 March 2023 (UTC)
 * @TadejM, you are new around these parts. Please generally avoid editing the live CS1 modules. They are transcluded sufficiently many times where sandboxed edits are the default action and even simple-looking edits may cause undesirable behavior. Izno (talk) 22:59, 14 March 2023 (UTC)
 * The first of those edits highlights an oversight on my part. The message key   is referenced only once in Module:Citation/CS1 at line 3705.  That line and the next line do nothing because   is not used after it has been set in 3705 and possibly modified in 3706.  Those two lines are leftovers from the transition from when some error messages appeared in the midst of the rendered citation to the current state where all error messages appear at the end of the rendered citation.  I have deleted the   k/v pair from Module:Citation/CS1/Configuration/sandbox and the two lines from Module:Citation/CS1/sandbox.
 * —Trappist the monk (talk) 14:05, 14 March 2023 (UTC)
 * —Trappist the monk (talk) 14:05, 14 March 2023 (UTC)

i18n date handling
At there is a link to an article at the Kannada wikipedia (kn:ಚದುರಂಗದ ನಿಯಮಗಳು). At the bottom of that article there are a couple of Lua script errors. These errors occur because of ೧೯೭೭ (1977) and ೧೯೯೨ (1992). At line 335 in kn:Module:Citation/CS1/Date validation, there is this:. That works fine so long as  is written as Arabic numerals. Kannada numerals are not Arabic numerals so the call to  returns. I think that I have fixed that in our sandbox and replacing  at kn.wiki with our version of the function seems to support that.

—Trappist the monk (talk) 17:29, 15 March 2023 (UTC)

Further to this, I have figured out how to get MediaWiki to supply non-English digits for wikis, like kn.wiki, that might write dates like ೧೬ ಮಾರ್ಚ್ ೨೦೨೩ (16 March 2023). This particular functionality is disabled at en.wiki.

—Trappist the monk (talk) 15:06, 16 March 2023 (UTC)

RFC on whether citing maps and graphs is original research
Please see Village pump (policy). Rschen7754 15:10, 19 March 2023 (UTC)

Edit request
Please add hatnote on the above template to distinguish between Cleanup press release, as noted in this RM. Thanks,  The Night Watch     (talk)   14:17, 20 March 2023 (UTC)
 * to Template:Cite press release/doc, which is not protected. GoingBatty (talk) 14:25, 20 March 2023 (UTC)

Original publisher
When a publisher purchases a work from another publisher and reprints it, should the citation show the original publisher? Does the answer depend on whether it is a reprint or a revision? Is it appropriate to use orig-date for the purpose, given that there is no orig-publisher parameter? Shmuel (Seymour J.) Metz Username:Chatul (talk) 19:38, 22 March 2023 (UTC)
 * I would list the publisher of the edition you've consulted. If you consult the original work from the original publisher, then cite that. If you've consulted the edition by the new publisher, cite them. WP:SAYWHEREYOUGOTIT applies.  Imzadi 1979  →   19:52, 22 March 2023 (UTC)
 * If it's a reprint, giving the original year and publisher might help readers in verifying the citation if they have access to the original work; e.g.: . -- Michael Bednarek (talk) 02:12, 23 March 2023 (UTC)
 * Give the citation to the work you are actually looking at. The reason is that the page numbering may be different on another edition or by another publisher. We cite the original date so people know how up-to-date the source is. If you want to help the reader find a copy, give them an OCLC. Hawkeye7   (discuss)  03:51, 23 March 2023 (UTC)
 * Is there a way to obtain the OCLC given the ISBN? --Shmuel (Seymour J.) Metz Username:Chatul (talk) 12:15, 23 March 2023 (UTC)
 * Special:BookSources; the 'Find this book at WorldCat' link under the Online databases heading.
 * —Trappist the monk (talk) 13:01, 23 March 2023 (UTC)
 * —Trappist the monk (talk) 13:01, 23 March 2023 (UTC)

Another generic title
Hello, can you add "Detect browser settings" as a generic title. Currently, 27 instances. Keith D (talk) 00:04, 24 March 2023 (UTC)

cite AV media: Audio media - Explanation and Examples, please
It would be helpful if more explanation and examples were provided on the Template:Cite AV media help page. It took a lot of trial, error, web searches, and time to use to cite a song on the B-side of a vinyl singles album on the page Truck Drivin' Man (Lynyrd Skynyrd song). I searched for more information on using the template but did not find any.

Questions I had included: (1) Should the composer and lyricist of the song be included, and if so, should their roles be placed in parentheses after their names? (2) Should the producer or others involved in creating the recording be named, and if so, should their roles be placed in parentheses after their names? (3) Should the fact that the song recording being referenced is a "single" be cited? (4) *How do I indicate the medium is a 7" vinyl record? (5a) If I were to cite a song from an album, how do I include the name of the album in the citation (and I'm assuming it should be included, right?)? (5b) And, how would I include the track number?

* Type: Under Parameters-Description-Title, the "type" parameter explanation says, "Use one of the following as applicable: Motion picture, Television production, Videotape, DVD, Blu-ray, Trailer, CD, Radio broadcast, Podcast." Options for vinyl or cassette are not listed. (I discovered that using "Vinyl" worked).

The first example of the parameters includes the parameter "people", then later in the article the parameter "others" as well as parameters including the term "author" are very briefly introduced. This is confusing and is related to the questions that I had.

-Thank you.

Dogsgopher (talk) 19:59, 8 April 2023 (UTC)

cite compare
I've tweaked. Description and examples at

—Trappist the monk (talk) 18:34, 10 April 2023 (UTC)

Hardcoded 2022
Is this code correct: local name = lang_obj:formatDate('F', '2022-' .. i .. '-1');			-- get long month name for each i		long_t[name] = i;														-- save it name = lang_obj:formatDate('M', '2022-' .. i .. '-1');					-- get short month name for each i		short_t[name] = i;

Why is '2022' hardcoded in Module:Citation/CS1/Configuration? --TadejM my talk 15:34, 11 April 2023 (UTC)
 * It is correct. The purpose of that code is to get long and abbreviated month names so the 'year' portion of the timestamp isn't all that important.  Compare:
 * —Trappist the monk (talk) 15:45, 11 April 2023 (UTC)
 * —Trappist the monk (talk) 15:45, 11 April 2023 (UTC)
 * —Trappist the monk (talk) 15:45, 11 April 2023 (UTC)
 * —Trappist the monk (talk) 15:45, 11 April 2023 (UTC)
 * —Trappist the monk (talk) 15:45, 11 April 2023 (UTC)

Great, thanks. --TadejM my talk 15:57, 11 April 2023 (UTC)

Potential improvement: automatic style choice
Our citation templates are smart enough to know when an article has a use mdy dates or use dmy dates directive, and automatically fix up date formats to be consistent with that. They are also smart enough to know how to change between citation style 1 and citation style 2, using the mode parameter. However, they don't seem to have a way of setting that mode for a whole article at once. I am constantly finding myself adding cs1-format citations to a cs2-format article, or cs2-format citations to a cs1-format article, and then having to go back and make another edit to make it consistent again. And I am even more frequently finding articles where other editors have not taken that care and the citations are not in a consistent format. Would it be possible to define some kind of article-wide directive, like the date format directives, that these templates could see and respect? —David Eppstein (talk) 21:59, 31 March 2023 (UTC)


 * I question why we don't merge the two styles together so that they provide the same output formatting. Then it wouldn't matter which template someone used to create a citation. To that end, I nominate using CS2's commas and capitalization scheme with CS1's terminal period. Then it wouldn't matter if someone used citation or cite book to create a citation to a book.  Imzadi 1979  →   22:18, 31 March 2023 (UTC)
 * I suggested that at using a template .  You also suggested it later in that same discussion.  I noted then that the response to those suggestions was underwhelming.  I periodically think about it for certain parameters: display-authors (and other display name parameters) so that all name lists are limited to n displayed names, mode as discussed above, and name-list-style so that all name lists have the same style.
 * —Trappist the monk (talk) 23:15, 31 March 2023 (UTC)
 * Thank you so much for your feedback. I have been studying the editing guidelines and will correct the errors when I identify them.  I am currently having difficulty typing in News articles (e.g. The Wall Street Journal) in italics. I am continuing to learn the how to do this. I hope you have patience.
 * thanks again LZPanther (talk) 13:03, 11 April 2023 (UTC)
 * I have hacked the sandboxen to implement global configuration setting via . Examples and some description at this version of my sandbox.  Keep?  Discard?  Other comments?
 * —Trappist the monk (talk) 15:31, 11 April 2023 (UTC)
 * The article appears in good shape and I appreciate your input. Thanks for taking the time to work with me. The Wikipedia page is referenced by the name Reid Lyon.
 * LZPanther1 LZPanther (talk) 14:30, 12 April 2023 (UTC)
 * It looks useful to me, at least until such time as Imzadi's merge proposal (which I like, in that form) gains enough consensus to be implemented. —David Eppstein (talk) 16:03, 12 April 2023 (UTC)
 * Thanks again LZPanther (talk) 16:30, 12 April 2023 (UTC)
 * Thanks again LZPanther (talk) 16:30, 12 April 2023 (UTC)

Calculated archive-date
Given that there is code now (I think, as discussed above) to look for a mismatch in archive-date and the date based on the archive-url, it would seem that the next step is to allow for the archive-date to not be present *if* the archive-date is calculatable from the archive-url. What would be issues with that? Would it slow down the pages? Naraht (talk) 13:55, 16 March 2023 (UTC)


 * Yes, it would slow down pages. We would need to benchmark on a representative page or two.
 * Besides that:
 * Izno (talk) 18:25, 16 March 2023 (UTC)
 * IMO this will cause unintended consequences not save much effort rather create extra efforts elsewhere. So many tools and processes depend on the existence of this argument. If the argument is missing is this because it's in the URL, and if so how do you parse the URL when there are 20+ archive providers with many variations - huge programming effort most won't bother with. Or is it because it's not in the URL and the missing argument is an error. Another hazzard is people will avoid adding the argument if they don't have to, and pretty soon they are not adding the argument even for URLs that require it - how do you teach people they need it for one case but not another - lots of errors will be introduced. -- Green  C  21:56, 16 March 2023 (UTC)
 * Possible step. An additional argument (archive-durl?) which can be used, which will be an alias to archive-url *but* will indicate that the date can be extracted by a bot which will replace archive-durl with archive-url and archive-date appropriately.Naraht (talk) 17:29, 28 March 2023 (UTC)
 * I would oppose this change. In addition to GreenC's comments, I can see a situation where a citation currently links to an archive at archive.today from date X, which would require the inclusion of the archive-date, but someone comes along and substitutes a better link from archive.org from date Y. Because the latter wouldn't require the explicit date, we could end up with a mis-match if that editor doesn't also remove or update archive-date from X to Y.
 * The proposal also presumes stability in the archive URL format from these third-parties, something we cannot assume now or in the future. The current scheme is the simplest and best option.  Imzadi 1979  →   18:04, 28 March 2023 (UTC)
 * Thanks for your consideration and realistic approach 73.28.235.193 (talk) 22:52, 15 April 2023 (UTC)

ISBN hypenation discussion
Interested editors are invited to look at WT:ISBN. – S. Rich (talk) 02:30, 16 April 2023 (UTC)

bibcode check needs update/exception
Category:CS1 errors: bibcode is populated exclusively by bibcodes related to Heliyon. For example fails a test, but is nonetheless a valid bibcode. &#32; Headbomb {t · c · p · b} 02:19, 13 April 2023 (UTC)
 * That bibcode fails the test because the last digit is, according to this, supposed to be the first letter of the last name of the first author – in this case  .  Here they have overflowed the four-digit page number with a five-digit article number (from which, presumably, the leading zero could have been removed).  For the nonce, and because Heliyon appears to be the only source using this sort of corruption, in the sandbox, I have enabled accept-as-written markup for bibcode:
 * – without accept-as-written markup
 * – with accept-as-written markup
 * —Trappist the monk (talk) 15:11, 13 April 2023 (UTC)
 * Yeah, but this likely will be required for every Heliyon bibcode out there. Manual bypass of every instance will be... tedious. &#32; Headbomb {t · c · p · b} 23:03, 16 April 2023 (UTC)
 * —Trappist the monk (talk) 15:11, 13 April 2023 (UTC)
 * Yeah, but this likely will be required for every Heliyon bibcode out there. Manual bypass of every instance will be... tedious. &#32; Headbomb {t · c · p · b} 23:03, 16 April 2023 (UTC)

Bump PMC limit
PMCs like are perfectly valid. &#32; Headbomb {t · c · p · b} 23:17, 16 April 2023 (UTC)

Help needed checking results of ReferenceExpander
Use of the ReferenceExpander bot without manually checking the results has led to problems affecting a potentially large number of articles. Help from people who like getting citations right is very much needed. XOR&#39;easter (talk) 22:15, 16 April 2023 (UTC)


 * @XOR'easter: Have you considered contacting the editors in your "Actor name" column to ask them for assistance with cleaning up the edits they made? GoingBatty (talk) 03:29, 17 April 2023 (UTC)
 * Have you considered contacting looking at the talk page history of the editors in the "Actor name" column to determine whether contact might have already been made and rebuffed? See e.g. Special:Diff/1147410661. —David Eppstein (talk) 05:30, 17 April 2023 (UTC)
 * @David Eppstein: Before making my previous post, I looked at the talk pages of two editors and didn't see any contact. I also looked at XOR'easter's recent edit history, and didn't see any user talk page edits about this topic.  Apologies if XOR'easter or you or others have already contacted the editors. GoingBatty (talk) 16:09, 17 April 2023 (UTC)

Template:Citation Style documentation/url leaves a Script warning
In Tandy Warnow, there is a citation that includes a viable archive-url, though the original url (this one, to seedmagazine.com) is now unfit / usurped due to being taken over by a website offering computer screen wallpaper. A previous editor has already followed the protocol, specifically defined at Template:Citation Style documentation/url, and set the url-status=unfit.

Which, in edit preview, causes the following to appear:
 * This is only a preview; your changes have not yet been saved! → Go to editing area
 * Script warning: One or more templates have maintenance messages; messages may be hidden (help).

Experimented a bit, found that url-status=usurped still produces the same Script warning. Also found that changing from "{{citation" to "{{cite journal" for the citation still produces a Script warning (albeit adjusted to "One or more {{cite journal}} templates..."). Also found that removing the url-status=unfit solved the Script warning issue - but it reactivated the citation's ridiculous link to the website offering computer screen wallpaper.

When looking through the archives of this "Help talk:Citation Style 1" page, I found these:
 * Help talk:Citation Style 1/Archive 68
 * Help talk:Citation Style 1/Archive 19
 * Help talk:Citation Style 1/Archive 84

Not really sure any of these explain why there should be a Script warning{{snd}}of any type{{snd}}when using url-status=unfit in the way explicitly defined by the documentation. Perhaps there is another discussion which explains it (lots of pages in Help talk:Citation Style 1/Archive). I have no doubt there is a logical explanation why the proper use of url-status=unfit produces a Script warning implying that it is not being used properly. Perhaps this is not the correct way to use it? If so, I can adjust the citation entry once I know what has been entered incorrectly.

Thanks, in advance, for the answers and guidance. Jmg38 (talk) 16:55, 17 April 2023 (UTC)

Season links in Template:Cite episode
The template currently supports series-link and episode-link, but curiously not season-link. Can this be added? Thanks in advance! InfiniteNexus (talk) 16:02, 18 April 2023 (UTC)


 * @InfiniteNexus: Hi there! Could you please provide an example of a citation where a season-link would be helpful?  Thanks! GoingBatty (talk) 16:43, 18 April 2023 (UTC)
 * It would be helpful for all series that have a standalone season article. For example, ref 37 on Infinity Stones cites the first episode of the first season of Loki. It seems all three of those links should be present in the template. InfiniteNexus (talk) 16:51, 18 April 2023 (UTC)

|subject-first= and |subject-last=
For who knows what reason, we do not have subject-firstn and subject-lastn or their given/surname aliases and their alternate enumerator position variants. Tweaked the sandbox so that these will work after the next update:

—Trappist the monk (talk) 14:48, 22 April 2023 (UTC)

abbreviated date range
We have a properties category. tracks articles with cs1|2 templates using date, publication-date, and year with abbreviated date ranges; for example, 2019–20. That is a valid but, discouraged(?), date format.

Something, ve? citoid? renders many journal dates in YYYY-MM format (with a hyphen). That date format can be ambiguous so we test for abbreviated year greater than the unabbreviated year (in the same century), abbreviated year may not be farther into the future than next year.

The live version of the module adds the properties category before the date is validated. That seems wrong to me so I have moved the property cat addition to after the validation; if the date isn't valid, no point in adding the article to the category.

This example, adds the category:

you can tell that it did by the  class in the  tag.

The sandbox for the same template does not:

Both correctly emit an error message because en.wiki does not support YYYY-MM date formats.

Rewriting the sandbox template to use a 'valid' abbreviated date range does add the properties category:

Of course, there is the other question: do we need this category? Is anyone using it? I don't have the time right now to research the whys and wherefores of that category but may do so tomorrow.

—Trappist the monk (talk) 01:17, 25 April 2023 (UTC)


 * @Trappist the monk: My bot and I are cleaning up many of these abbreviated date range issues from Category:CS1 errors: dates. I'm not using Category:CS1: abbreviated year range. GoingBatty (talk) 04:14, 25 April 2023 (UTC)

The whys-and-wherefores apparently rise from. @Editor Matthiaspaul: are you using this category?

—Trappist the monk (talk) 15:53, 25 April 2023 (UTC)

Unsuppress orig-year when date/year unspecified?
Hi! My content work often involves citing old secondary sources where there's only an estimated year of original completion (usually the Shiji), and online sources are not tied to a specific modern republication. year and date don't accept estimates, or years BCE, so I use orig-year for that. No problem. But if I'm going off an online source, I'll often leave the date parameter unspecified in the template call, and then there's no date shown at all. What downsides are there to displaying the orig-year if it's the only date parameter with a value? Folly Mox (talk) 21:31, 25 April 2023 (UTC)


 * works as expected for estimates (though yes, there is a bottom threshold on dates).
 * But if I'm going off an online source, I'll often leave the date parameter unspecified in the template call, and then there's no date shown at all. Why are you doing that? :) Izno (talk) 21:53, 25 April 2023 (UTC)
 * I no longer have a computer that can read the file system type of the hard drive I used in undergrad lo these many years ago to download illegal electronic versions of books I couldn't still can't afford. Plus the versions at wikisource accord with the versions at guoxue, ctext etc, and as often as not I'm improving the citations of other editors who cite with merely a link to a wikisource page and search string.There are modern punctuated editions that are tantamount to standard in the field, and which are likely the source of the hypertext versions, but I feel like including publication information for books I can't actually see violates WP:SAYWHEREYOUGOTIT. For my specific story, date does not play well with e.g. 80s or 90s BCE, but I understand it's an edge case. Folly Mox (talk) 23:06, 25 April 2023 (UTC)

Another generic title
Hello, can you add "Register &#124; British Newspaper Archive" as a generic title. Currently, 26 instances. Keith D (talk) 16:48, 26 April 2023 (UTC)

Author name suggestion
Hello! I admit this issue has come up rarely, but I thought it was worth flagging. Sorry if I'm in the wrong place, I've never made a suggestion like this before, and I probably rarely will in the future.

I've had two instances in recent months in which I couldn't tell what part of an author's name was the last name. Before you all think I'm an idiot, allow me to explain (at which point feel free to think I'm an idiot).

I'm aware of a few authors who use multi-word last names; sometimes, those names are hyphenated, but, sometimes, they aren't. So, when I see three names listed by an author, I have a moment of hesitation: Is this "Smith, John David" or "David Smith, John". (Is it "Ali, Safia Samee" or "Samee Ali, Safia"?)

The manual of style 99% of legal writers use, the Bluebook, says to, on first reference, use a "longform" citation that states the authors name as one usually would: First Middle Last. Something like: "John David Smith and Jane Ellen Cary, Article Title, Journal (date)." Granted, on second references, you're supposed to use a short form citation with only the last name (Bluebook uses supra for footnotes), so the issue comes up. But it's occurred to me that, on Wikipedia, it doesn't have to.

From what I've seen, most (not all) Wikipedia articles don't use shortform citations. They just use ref "names" in order to link to the original longform citation. So, why not just list author names in normal order rather than prioritizing last names? (Personally, I think prioritizing last names makes sense in a bibliography, that lists authors in alphabetical order and aligns with in-text citations that only refer to an author's last name ... but I don't see what the benefit is in an article that chiefly uses footnotes that aren't (generally) connected to an end-of-text bibliography.) So, I'd just like to ask if using the Bluebook's longorm approach to names would make sense?--Jerome Frank Disciple (talk) 13:55, 26 April 2023 (UTC)


 * @Jerome Frank Disciple: There are many citation styles that may be used - see WP:CITESTYLE. GoingBatty (talk) 14:45, 26 April 2023 (UTC)
 * Sorry! Maybe I should take the suggestion to the individual citation templates.--Jerome Frank Disciple (talk) 14:52, 26 April 2023 (UTC)
 * @Jerome Frank Disciple: With the citation templates, some people use author and others use lastfirst. Either use is acceptable per WP:CITESTYLE.  If you're proposing changes to the citation templates and/or their documentation, this is the place to do so. GoingBatty (talk) 18:44, 26 April 2023 (UTC)
 * Sounds good! I can hat this discussion, if you'd like.--Jerome Frank Disciple (talk) 19:02, 26 April 2023 (UTC)
 * The dedicated parameters for the name parts are in fact preferred and are not in fact covered by CITESTYLE, which applies to wide formats and not specific parameter use.
 * The real proposal here is "flip all 20 years of cite template use on its head", which is of course unrealistic; the more passive "let me choose my format" is also bad because one way to style authors is both easier to teach and fix bad use. Izno (talk) 02:02, 27 April 2023 (UTC)
 * Wait, what? I didn't say "let me choose my own format". Are you saying that GoingBatty's interpretation is wrong and that there is a set format (or that |author= shouldn't be used)? (Maybeee I should've hatted this once I realized it was the wrong place.)--Jerome Frank Disciple (talk) 02:06, 27 April 2023 (UTC)
 * You did not say that indeed. It was me commenting on a potential "middle ground" that occasionally pops up when discussing how things should work (with templates, MOS pages, etc.).
 * I am saying that firstlast is preferred to author, and that his assertion that CITESTYLE protects one or the other forms is accordingly wrong; but in addition, that it is wrong to attempt to invoke CITESTYLE in the context of specific parameters of a specific (series of) template(s).
 * This is not in fact the wrong place, so far as I can see. It is the centralized discussion page for every Help:CS1 template (and is generally preferred for existing Help:CS2 templates). You may have other templates in mind, but I do not think so, so you are in the right place. Izno (talk) 02:35, 27 April 2023 (UTC)
 * Ah, fair enough! I appreciate all the information! I currently default to last= first=, and, hey, I guess if I have to guess, I have to guess. :) --Jerome Frank Disciple (talk) 02:51, 27 April 2023 (UTC)
 * I used the word prefer for a reason; it means that if you don't know, feel free to use the lesser-preferred. :) Izno (talk) 03:23, 27 April 2023 (UTC)
 * Oh sorry, I thought the rule that it had to be consistent within the article meant I had to take a one-size-fits-all approach. Now I see what you were saying about the relevance (or lack thereof) of CITESTYLE. Thanks for the extra time!--Jerome Frank Disciple (talk) 12:34, 27 April 2023 (UTC)
 * @Izno: Hi there! I reread Template:Cite web and don't see where it says firstlast is preferred to author.  Could you please direct me to where the preference is specified?
 * If an editor has used firstlast on an article and another editor prefers author (or vice versa), my interpretation of WP:CITESTYLE (and its subsection WP:CITEVAR) is that the second editor shouldn't reformat the references just because they like a different format better. Do I have that wrong?
 * Thanks! GoingBatty (talk) 13:55, 27 April 2023 (UTC)
 * The reason I say that one is preferred to the other is because one produces better metadata than the other. Using last/first parameters splits the keys correctly in the COinS metadata (last and first go into last and first keys), while using author instead puts the entirety of the information into the last key. Which is obviously incorrect in the vast majority of people authors that we should be citing.
 * Do I have that wrong? Unequivocally yes. CITESTYLE/VAR apply at the level of actual styles, not for choices you can make within templates. Izno (talk) 14:45, 27 April 2023 (UTC)

Citations of manuals from online collections?
I frequently cite IBM publications that are organized into bookshelves or other collections. I would like to include a link to the web page for the collection, but [url collection] and [url collection] get CS1 error messages. Should there be website-url and work-url parameters? -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 01:23, 1 May 2023 (UTC)
 * Rather than trying to misuse one of the fields, simply add the additional information after the cite but still inside the ref tags. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 10:24, 1 May 2023 (UTC)

Cite AV media and the use of |others=
A comment above made me think of something I come across a lot, that is the maintenance message cite AV media notes: CS1 maint: others in cite AV media (notes). You will find this message repeated dozens of times on most discography articles. It's appears common pratice is to use |others= for the artist with no author fields being populated. Given that I'd like to suggest that this error is suppressed for "cite AV media notes", and it's alternative names. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 12:07, 19 April 2023 (UTC)


 * This also makes the tracking category for these messages pretty pointless, as nearly every article tracked is just this type of use. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 12:23, 19 April 2023 (UTC)
 * The most recent collation of relevant discussions and another discussion since then, besides this one now. Please feel free to review. Izno (talk) 19:30, 20 April 2023 (UTC)
 * I guess the sticking point, for me, is that all of those discussions have failed to produce anything beyond... discussion. In the "most recent collation", you write:
 * ...But, from the number of discussions that have already covered this ground, consensus seems to be that we're not going to deal with them, really. Which makes that maintenance category not actually a maintenance category at all, as there is no maintenance that can be performed on those entries, nor is that expected to change. (Maybe eventually. But the sun will burn out, eventually...) At best it's merely a tracking category, in the most fundamental sense, and nothing more.
 * But having that tracking category classified as CS1 maintenance causes side-effects not encountered with normal tracking categories. Editors get "Script warning" notices when using the template, and (as notes) those who've opted in to their display see green maintenance messages littering articles, despite those messages remaining inactionable. Which makes me think that keeping it there, just in some vague hope that a plan for dealing with those citations eventually emerges, is an example of the perfect being the enemy of the good. FeRDNYC (talk) 03:05, 4 May 2023 (UTC)
 * And a related discussion:
 * —Trappist the monk (talk) 19:41, 20 April 2023 (UTC)
 * And a related discussion:
 * —Trappist the monk (talk) 19:41, 20 April 2023 (UTC)

Unintentional double spaces in rendered output
CS1/2 templates produce double spaces that are (you can locate them in the rendered output by ctrl+f, depending on your browser, but they  as single spaces, not wide). It happens in the following scenarios: every time an editor + author/contributor is defined, and every time etal is used. It does not depend on whether the code has spaces or not, the spaces are always there. See for example: Surely this is unintentional? – Finnusertop (talk ⋅ contribs) 18:42, 2 May 2023 (UTC)


 * Is it a problem? This still renders ultimately as one space. Izno (talk) 18:49, 2 May 2023 (UTC)
 * It's not a big problem, it's a small one. I think code sanitization is definitely the forte of CS1/2, and if something can be fixed, it should be fixed. Aside from a miniscule increase in page size, a practical problem is clutter when searching for actual double spaces when making wikitext more readable. – Finnusertop (talk ⋅ contribs) 19:26, 2 May 2023 (UTC)
 * My browser compresses multiple space characters into a single space. To see the double spaces, I copied the example template into Special:ExpandTemplates and clicked Show raw HTML.  The live module shows doubled spaces between the name-list separator character (semicolon) and 'et' (3×) and also between the chapter title's terminator (a dot for cs1, a comma for cs2) and 'In' (cs1) / 'in' (cs2).
 * I think that I have found where these extra characters are coming from and have tweaked the sandbox a bit to remove them. It isn't much of a big deal for the 'et al' insertions, but I remember Editor Dragons flight having a bit of a struggle getting the author/date/editor/chapter to line up and fly right.  Since then I have been careful to avoid doing much of anything with that code so I'll stare at it tomorrow to make sure that I haven't buggered up some edge case that requires these extra spaces.
 * For your convenience, here is the sandbox version of the example template:
 * —Trappist the monk (talk) 00:12, 3 May 2023 (UTC)
 * I think I've fixed this. Here is a template that uses etal for all five name lists.  No double spaces:
 * Changing to or cs2 gives the same no-double-spaces result.
 * —Trappist the monk (talk) 13:54, 3 May 2023 (UTC)
 * Thank you, Trappist the monk. – Finnusertop (talk ⋅ contribs) 20:50, 3 May 2023 (UTC)
 * I think I've fixed this. Here is a template that uses etal for all five name lists.  No double spaces:
 * Changing to or cs2 gives the same no-double-spaces result.
 * —Trappist the monk (talk) 13:54, 3 May 2023 (UTC)
 * Thank you, Trappist the monk. – Finnusertop (talk ⋅ contribs) 20:50, 3 May 2023 (UTC)
 * Changing to or cs2 gives the same no-double-spaces result.
 * —Trappist the monk (talk) 13:54, 3 May 2023 (UTC)
 * Thank you, Trappist the monk. – Finnusertop (talk ⋅ contribs) 20:50, 3 May 2023 (UTC)

Periodical title parameter in cite book
I just stumbled across this page: WikiProject Academic Journals/Journals cited by Wikipedia/Statistics. Apparently these statistics are based on the presence of journal in templates. What struck me right off was the number of templates using journal. Is there any real reason that should support journal or any of the other periodical parameters?

This crude search finds about 2000 articles that have with journal – the search times out so who knows how may articles are involved. Examples from two of those articles:

from Early modern period:
 * It is not at all clear what this reference is citing. Britain in the Hanoverian Age, 1714–1837: An Encyclopedia is an encyclopedia published by Garland Publishing, New York and London; url links to a Google book facsimile; the doi points to a review in History: Reviews of New Books to which volume, issue (not rendered by ), and pages refer; the value assigned to publisher is the current publisher of the journal; isbn is the ISBN of the encyclopedia, and the named editor is the editor of the encyclopedia
 * It is not at all clear what this reference is citing. Britain in the Hanoverian Age, 1714–1837: An Encyclopedia is an encyclopedia published by Garland Publishing, New York and London; url links to a Google book facsimile; the doi points to a review in History: Reviews of New Books to which volume, issue (not rendered by ), and pages refer; the value assigned to publisher is the current publisher of the journal; isbn is the ISBN of the encyclopedia, and the named editor is the editor of the encyclopedia
 * It is not at all clear what this reference is citing. Britain in the Hanoverian Age, 1714–1837: An Encyclopedia is an encyclopedia published by Garland Publishing, New York and London; url links to a Google book facsimile; the doi points to a review in History: Reviews of New Books to which volume, issue (not rendered by ), and pages refer; the value assigned to publisher is the current publisher of the journal; isbn is the ISBN of the encyclopedia, and the named editor is the editor of the encyclopedia

from Clovis culture:
 * Here, journal really should be series; title names a chapter in a book called Climates, Landscapes, and Civilizations edited by Giosan et al. which title and editor name list are missing from the template; isbn points to the book title but that can confuse readers because the book title is missing in the citation; doi links to the chapter at the publisher's website
 * Here, journal is used to name the book title (without subtitle) and title is used to name the cited chapter; isbn points to the book title and doi points to the chapter at the publisher's website.
 * Here, journal really should be series; title names a chapter in a book called Climates, Landscapes, and Civilizations edited by Giosan et al. which title and editor name list are missing from the template; isbn points to the book title but that can confuse readers because the book title is missing in the citation; doi links to the chapter at the publisher's website
 * Here, journal is used to name the book title (without subtitle) and title is used to name the cited chapter; isbn points to the book title and doi points to the chapter at the publisher's website.
 * Here, journal is used to name the book title (without subtitle) and title is used to name the cited chapter; isbn points to the book title and doi points to the chapter at the publisher's website.
 * Here, journal is used to name the book title (without subtitle) and title is used to name the cited chapter; isbn points to the book title and doi points to the chapter at the publisher's website.
 * Here, journal is used to name the book title (without subtitle) and title is used to name the cited chapter; isbn points to the book title and doi points to the chapter at the publisher's website.

The first example should be rewritten to either cite a single source (the book or the journal) or split into two citations. In the latter two examples, the rendered result is wrong: chapter titles should be quoted in an upright font and book titles should be italicized.

It seems to me that we should prevent and  from accepting journal and the other periodical parameters in much the same way that we prevent periodical templates  from accepting chapter and its aliases.

Opinions?

—Trappist the monk (talk) 15:14, 12 May 2023 (UTC)


 * This is an excellent suggestion. I fully support it. &emsp;—&#8239;sbb&#8239;(talk) 17:40, 12 May 2023 (UTC)
 * @Trappist the monk: Support, as long as there's a tracking category and a corresponding addition to the Help:CS1 errors page. Should it start as a maintenance message before making it an error, in case there are patterns we could quickly fix before making it a visible error?  Thanks! GoingBatty (talk) 19:08, 12 May 2023 (UTC)
 * as long as there's a tracking category and a corresponding addition to the Help:CS1 errors page. Of course; cs1|2 doesn't do error messaging without matching categories and help text.
 * I think that this is an error condition and should be noted as such. If there are patterns that can be quickly fixed, the patterns will be just as apparent (maybe more so) with error messages as they will be with maintenance messages.
 * —Trappist the monk (talk) 21:58, 12 May 2023 (UTC)
 * @Trappist the monk - I just finished running a search against the May 1 database dump for  and found 7,789 matches. GoingBatty (talk) 22:29, 12 May 2023 (UTC)
 * Thanks. That's ~% of  articles.  Just out of curiosity, I ran the crude search for the other 'periodical' parameters:
 * magazine: ~50
 * newspaper: ~175
 * periodical: ~25
 * website: ~2350
 * work: ~5600
 * All of these searches time out. Results from the first three searches are, I think, negligible; the last two possibly not.  Editor GoingBatty: Can you do similar searches against the database dump for  using website and  using work?
 * —Trappist the monk (talk) 13:43, 13 May 2023 (UTC)
 * @Trappist the monk: 10,699 with website and 23,819 with work. GoingBatty (talk) 04:34, 14 May 2023 (UTC)
 * Thank you.
 * —Trappist the monk (talk) 12:13, 14 May 2023 (UTC)
 * I have hacked the sandbox. Here are examples of  with the various periodical parameter aliases:
 * I have hacked the sandbox. Here are examples of  with the various periodical parameter aliases:


 * and examples of with journal (the other periodical parameters produce the same sort of results); note the error messaging change:


 * and where the first three examples show that I haven't broken  when it renders encyclopedia or journal references; note the error messaging change:


 * —Trappist the monk (talk) 18:30, 13 May 2023 (UTC)
 * Support. I also can't imagine any citation to a book where  would be accurate or helpful. Glades12 (talk) 20:02, 13 May 2023 (UTC)

"Others" parameter in Cite episode?
The documentation for Cite episode provides the following example, where "(host)" gets appended into first:

I think it might be good if there were a others, like there is for Cite AV media, so that other people (e.g., writer, director, producer) can be added and without having to pretend that "(host)" is part of the given name:

Umimmak (talk) 17:45, 15 May 2023 (UTC)
 * does not support others because the wikitext version (before luafication) did not support others. Other in  was usurped to support transcript.  The  solution the others problem for, , and perhaps  is to create a list of acceptable 'role' parameters usable only in these templates.  Go to the en.wiki communities that use these templates and create a reasonably sized list of roles and bring that list here.  That would allow us to render those templates with role annotation without corrupting the citation's metadata.  Attempts to create such a list here have failed so it falls to the user community to create the list.
 * —Trappist the monk (talk) 18:45, 15 May 2023 (UTC)

Detect placeholder volume/issue/pages
If volume/issue/page/pages/at are equal to "n/a", "forthcoming", "in press", "to be published", or "0", it would be good if we had a maintenance category for this. "0" will net a few false positives, but the majority of volume=0 (and similar) are problematic to should be flagged. &#32; Headbomb {t · c · p · b} 12:29, 16 May 2023 (UTC)
 * Example of '0' as a false positive?
 * It seems to me that 0 or 0 is an error and should be marked as one. Without evidence to the contrary, I'm inclined to think that 0 or 0 is also an error.
 * How do,  ,  , and   comport with WP:SAYWHEREYOUREADIT?  Is a source marked with   a violation of WP:V?
 * —Trappist the monk (talk) 13:10, 16 May 2023 (UTC)
 * @Trappist the monk: 75 examples of 0 at these search results. GoingBatty (talk) 13:39, 16 May 2023 (UTC)
 * Thanks. I'd already done that search.  For the record:
 * issue: ~105
 * page and pages: ~290
 * at: ~15 but none were cs1|2 templates
 * and composite case-insensitive searches for:
 * : ~660
 * : ~45
 * : ~145
 * : none
 * —Trappist the monk (talk) 14:24, 16 May 2023 (UTC)
 * : none
 * —Trappist the monk (talk) 14:24, 16 May 2023 (UTC)

|location= without |publisher= (2)
A while ago I started a discussion which, like many discussions, wandered off into the weeds. That discussion was brought to mind when I recently encountered a handful of citoid-created cs1|2 templates ( for the most part) that had location but did not have publisher.

I have hacked the module sandboxen to add a maintenance category when and  templates have location, place, and/or publication-place without publisher. Similarly, without a periodical (work) alias and with or without encyclopedia will also emit the maintenance category when location (or an alias) is missing publisher.

book citations:





encyclopedia citations:





other citation types:
 * – not a 'book' cite so not checked for location-missing-publisher


 * – not a 'book' cite so not checked for location-missing-publisher

Keep? Discard? Other?

—Trappist the monk (talk) 14:42, 17 April 2023 (UTC)


 * Is this change in production? Where would find the maintenance category in question? Jc3s5h (talk) 14:49, 17 April 2023 (UTC)
 * Of course not. I wrote: I have hacked the module sandboxen....  When (if) the change is made to the live module, the category will be  (unless there is a better name?)  The maintenance messages are visible to those who enable them; see.
 * —Trappist the monk (talk) 14:55, 17 April 2023 (UTC)
 * It's a useful issue to track, I've seen it happen a lot when isbn has been used to auto-populate the book details. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:06, 17 April 2023 (UTC)
 * I re-read the original thread, where I wrote: I suspect that this requirement would turn out to be overly fussy, with many situations in which publisher information is unavailable. We don't want unfixable error messages. What would our recommendation be in that case? none would have to be accepted, but I think there would be complaints of the type "there was clearly a publisher of some kind, but the information doesn't exist, so 'none' is untruthful." If this change is approved by consensus here, I would like the change proposal to include new, explicit instructions for how to handle legitimate situations where publisher information is simply not available, and how that lack of publisher information would be displayed in the rendered citation. Those instructions would become part of the overall proposal to newly require publisher when location is present. I do not think we should backfill that sort of thing post-deployment of a change like this. – Jonesey95 (talk) 22:00, 17 April 2023 (UTC)
 * The original proposal was for an error message. The current proposal emits a maintenance message.
 * For older books, the publisher, if there was one, may not be known. I have seen examples where the book was printed for 'the Society' or for some nobleman but those are rare in comparison to the number of citations-sans-publisher that citoid churns out.  To constrain the categorization so that only 'modern' books and encyclopediae are categorized, we could tweak the test to require isbn or sbn (c. 1966 onwards) or perhaps require that date specifies a year after some defined point in time (1850 onwards for example – Macmillan Publishers was founded in 1843).
 * You don't like none? Suggest alternatives; perhaps unknown? missing?    has the advantage of familiarity because it is already used for postscript, ref, title, and type.  In the other discussion, I linked to three external style guides showing how those guides handled missing publisher data.  We could map none to one of those renderings – I like the Harvard '[no publisher]' because interpretation is not required.
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * I would suggest unknown, it's not as succinct as none but wouldn't suffer the quibbles that Jonesey mentioned. I like the 1850 idea as it would suppress most cases where the issue was invalid, while using isbn would result in a new maintenance message if an isbn was added (which would be annoying). No opinion on how |publisher=unknown should be displayed. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 11:56, 19 April 2023 (UTC)
 * I hacked an awb script to trawl through a small (25,000) sample of the ~1.56 million articles that transclude . The script inspected each page looking for  templates that:
 * have location, place, or publication-place an assigned value and  publisher (missing or empty)
 * meet the conditions in 1 above and have isbn or ISBN an assigned value
 * meet the conditions in 1 above and have date or year an assigned value where the year portion of the date is 1850 or later
 * Of the 25,000 articles in the sample:
 * test 1 above (unconstrained) found 2408 articles
 * test 2 above (requires an isbn) found 1490 articles
 * test 3 above (requires date 1850+) found 2195 articles
 * I guess that the result isn't all that surprising. Because isbn wasn't really in use before c. 1966, there are plenty of books that were published in the 1850–c. 1966 time period (115ish years v 57ish years).
 * Apply a constraint? Which one?
 * —Trappist the monk (talk) 23:35, 20 April 2023 (UTC)
 * Like so many, this discussion faded away without any resolution. So, I have hacked the sandboxen again to implement the date-based limit (1850+) and for none to suppress the maintenance message where appropriate.  You will notice that the examples at the top of this discussion do not show maintenance messages.  This is due to the date limitation; no date, no maint message.  Repeating them here with dates:
 * book citations:
 * – pre-1850; no maint message
 * – pre-1850; no maint message
 * encyclopedia citations:
 * – pre-1850; no maint message
 * – pre-1850; no maint message
 * other citation types:
 * – not a 'book' cite so not checked for location-missing-publisher
 * – not a 'book' cite so not checked for location-missing-publisher
 * When publisher is set to the common keyword, no maint message:
 * – maint message suppressed with none
 * – maint message suppressed with none
 * —Trappist the monk (talk) 21:48, 17 May 2023 (UTC)
 * Discard. There are cases where a publisher is either unknown or is omitted, and there's no reason to throw an error if someone chooses to include a publication location. Nikkimaria (talk) 03:14, 18 April 2023 (UTC)
 * I seem to recall reading in some style guide the direction that citations to older books omit the publisher but still contain the location. To follow on with that, at least one guide I saw recently encouraged the use of "n.p." in citations when the publisher is unknown.  Imzadi 1979  →   04:52, 18 April 2023 (UTC)
 * I think location= without publisher= may be legitimate on rare occasions, such as in old books whose publisher is not clearly stated on the title page. I've also sometimes seen location= used on conference proceedings to describe the location at which a conference was held, unrelated to the publisher or the publisher location, but I think that may be an abuse of the parameters. We should fix those, if we find them, but first we would need an appropriate other parameter to use to hold this metadata, which should not just be omitted (conference location is not useful for library lookups, but it may be relevant to readers in understanding the context of a conference publication). —David Eppstein (talk) 04:33, 18 April 2023 (UTC)
 * Yes, using location to name the conference venue city is a misuse.  has conference where venue city may be included.
 * The proposed categorization does not apply to :
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * – pre-1850; no maint message
 * other citation types:
 * – not a 'book' cite so not checked for location-missing-publisher
 * – not a 'book' cite so not checked for location-missing-publisher
 * When publisher is set to the common keyword, no maint message:
 * – maint message suppressed with none
 * – maint message suppressed with none
 * —Trappist the monk (talk) 21:48, 17 May 2023 (UTC)
 * Discard. There are cases where a publisher is either unknown or is omitted, and there's no reason to throw an error if someone chooses to include a publication location. Nikkimaria (talk) 03:14, 18 April 2023 (UTC)
 * I seem to recall reading in some style guide the direction that citations to older books omit the publisher but still contain the location. To follow on with that, at least one guide I saw recently encouraged the use of "n.p." in citations when the publisher is unknown.  Imzadi 1979  →   04:52, 18 April 2023 (UTC)
 * I think location= without publisher= may be legitimate on rare occasions, such as in old books whose publisher is not clearly stated on the title page. I've also sometimes seen location= used on conference proceedings to describe the location at which a conference was held, unrelated to the publisher or the publisher location, but I think that may be an abuse of the parameters. We should fix those, if we find them, but first we would need an appropriate other parameter to use to hold this metadata, which should not just be omitted (conference location is not useful for library lookups, but it may be relevant to readers in understanding the context of a conference publication). —David Eppstein (talk) 04:33, 18 April 2023 (UTC)
 * Yes, using location to name the conference venue city is a misuse.  has conference where venue city may be included.
 * The proposed categorization does not apply to :
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * Discard. There are cases where a publisher is either unknown or is omitted, and there's no reason to throw an error if someone chooses to include a publication location. Nikkimaria (talk) 03:14, 18 April 2023 (UTC)
 * I seem to recall reading in some style guide the direction that citations to older books omit the publisher but still contain the location. To follow on with that, at least one guide I saw recently encouraged the use of "n.p." in citations when the publisher is unknown.  Imzadi 1979  →   04:52, 18 April 2023 (UTC)
 * I think location= without publisher= may be legitimate on rare occasions, such as in old books whose publisher is not clearly stated on the title page. I've also sometimes seen location= used on conference proceedings to describe the location at which a conference was held, unrelated to the publisher or the publisher location, but I think that may be an abuse of the parameters. We should fix those, if we find them, but first we would need an appropriate other parameter to use to hold this metadata, which should not just be omitted (conference location is not useful for library lookups, but it may be relevant to readers in understanding the context of a conference publication). —David Eppstein (talk) 04:33, 18 April 2023 (UTC)
 * Yes, using location to name the conference venue city is a misuse.  has conference where venue city may be included.
 * The proposed categorization does not apply to :
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)

Side discussion about templates other than cite book or cite encyclopedia
In terms of books, the above makes sense. But I have often cited magazine and news websites in the form of with https://drive.com.au/blahblahblah, Drive and Australia. The publisher field seems unwarranted in the this situation but it is quite useful to know which country the reference comes from because each country often has a certain bias - eg, reports of a car getting a new engine often differ between countries because that engine was not offered in all countries but the report often talks only to their own readers so it sounds universal when it isn't.  Stepho  talk 22:28, 17 April 2023 (UTC)


 * Furthermore, it's common to include the location of a newspaper if it's not included in the title (compare "Daily News. New York." vs. "The New York Times.") without including the publisher for either of them. Thee first of my two examples shouldn't prompt an error.  Imzadi 1979  →   04:47, 18 April 2023 (UTC)
 * Yes, location is used to disambiguate newspapers, magazines, etc. The proposal does not include periodical templates  (see example under other citation types above),, .  Here are the other two:
 * No maintenance messages.
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * No maintenance messages.
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * No maintenance messages.
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)
 * —Trappist the monk (talk) 14:36, 18 April 2023 (UTC)

access-url
The documentation for url-access in Template:Cite book includes "Suggested values", but Template:Cite book indicates that only free, limited, registration and subscription are permitted. None of these match the case where the document is available as a purchased PDF. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 08:17, 18 May 2023 (UTC)

Issue number could be less ambiguous
For the periodical citation style, it's possible to specify issue number. This rather confused me today, as the issue number looked like a year of publication. I think it would be preferable to prepend 'issue' to disambiguate this; e.g. "(issue 1708)" rather than "(1708)". What do others think of this proposal? Seabass-labrax (talk) 20:08, 19 May 2023 (UTC)
 * Are you citing an academic or scholarly journal or are you citing a magazine, a newspaper, or some other sort of periodical? How cs1|2 renders issue depends on which periodical template is being used.  For academic and scholarly journals, cs1|2 follows the convention commonly used by those journals (issue number wrapped in parentheses):
 * For the other kinds of periodical, cs1|2 renders the issue number with a prefix:
 * —Trappist the monk (talk) 21:35, 19 May 2023 (UTC)
 * For the other kinds of periodical, cs1|2 renders the issue number with a prefix:
 * —Trappist the monk (talk) 21:35, 19 May 2023 (UTC)
 * —Trappist the monk (talk) 21:35, 19 May 2023 (UTC)
 * —Trappist the monk (talk) 21:35, 19 May 2023 (UTC)

ISSN parameter use and documentation
For the |issn= parameter, should the ISSN or ISSN-L be used?

And whichever one should be used or is preferred, it should be reflected in the template documentation. Obviously not making a hard and fast rule against using one or the other, but more of a 'should' or 'when possible' thing. OfTheUsername (talk) 13:41, 20 May 2023 (UTC)


 * Apologies if this has already been answered, I couldn't find anything. OfTheUsername (talk) 13:41, 20 May 2023 (UTC)
 * ISSN is not really a useful identifier because it isn't unique to a particular issue of a periodical. It might help to locate a library that holds some issues of the periodical – no guarantee that the library holds  that is being cited.  WP:SAYWHEREYOUREADIT applies.  If you think that an issn is important to the citation that you are creating, use the issn printed in the source that you are consulting.  For myself, I think that issn is low-level noise that, in most cases, we can do well enough without.
 * —Trappist the monk (talk) 14:16, 20 May 2023 (UTC)
 * —Trappist the monk (talk) 14:16, 20 May 2023 (UTC)

Protected edit request on 2 May 2023
Please make this module be accessable from other modules without need for use of,  , or similar -related tricks. Animal lover &#124;666&#124; 11:43, 2 May 2023 (UTC)
 * Red information icon with gradient background.svg Not done for now: Gonna take a bit more in effort than an edit request, though I support such a change. Izno (talk) 16:02, 2 May 2023 (UTC)
 * User:Izno, Where should I ask to get it done? Animal lover &#124;666&#124; 17:01, 2 May 2023 (UTC)
 * Here is the right place, it's just non-trivial work. Izno (talk) 18:48, 2 May 2023 (UTC)

website
The current description of the parameter website = is:

Title of website (when the website has a clear name, use that rather than the domain name)...

Can we strengthen this to say something like:

Title of website (Use the domain name only when there is no clear title)...

I ask this because I see many instances where they value of the parameter is the URL of the website, even when there is an obvious name. 76.14.122.5 (talk) 06:21, 21 May 2023 (UTC)
 * The template documentation is not protected. If you know how to make the documentation better, please do.  Just like almost everything else on en.wiki, WP:BOLD applies.
 * —Trappist the monk (talk) 11:36, 21 May 2023 (UTC)
 * I'd make it even clearer by saying: Title of website (Use the domain name only when that is the clear title)... SchreiberBike &#124; ⌨ 20:19, 21 May 2023 (UTC)

Unexpected CS1 maint from cite thesis location=
The citation produces. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 15:29, 21 May 2023 (UTC)
 * Third ombox from the top at (the one with the [[file:Stop hand nuvola.svg|30px]]) pretty much says don't do that.
 * As to why that category appears is because Module:Citation/CS1 looks for digits in the value assigned to location. In this case,  is expanded before  is processed so Module:Citation/CS1 sees the  expansion.  The expansion looks like this:
 * Institut d'informatique et mathématiques appliquées de Grenoble
 * the 85% font size and the html entities  and   cause the module to apply the maintenance message and category to the rendered citation.
 * Don't use in cs1|2 templates.  And shouldn't it be publisher that gets the interwiki link?:
 * —Trappist the monk (talk) 15:57, 21 May 2023 (UTC)
 * That sounds plausible; the citation is in ALGOL 60. However, the citation, already has publisher. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 16:25, 21 May 2023 (UTC)
 * Only plausible? Do you have a better explanation?
 * If you follow the link at ALGOL 60 ref 15 (permalink), it takes you to the source landing page. On the left there is a 'Cite' box which lists 'Université Joseph-Fourier - Grenoble I' as the publisher.  On the right is a blue box labeled 'Complete list of metadata' which, when clicked, opens an overlay that has as 'Establishment providing the course': 'Université Joseph-Fourier - Grenoble I'.  So that would be the value I would use.  location should get the city of the publisher but, in this case, because Grenoble is part of the publisher name, should be omitted.
 * —Trappist the monk (talk) 16:53, 21 May 2023 (UTC)
 * That sounds plausible; the citation is in ALGOL 60. However, the citation, already has publisher. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 16:25, 21 May 2023 (UTC)
 * Only plausible? Do you have a better explanation?
 * If you follow the link at ALGOL 60 ref 15 (permalink), it takes you to the source landing page. On the left there is a 'Cite' box which lists 'Université Joseph-Fourier - Grenoble I' as the publisher.  On the right is a blue box labeled 'Complete list of metadata' which, when clicked, opens an overlay that has as 'Establishment providing the course': 'Université Joseph-Fourier - Grenoble I'.  So that would be the value I would use.  location should get the city of the publisher but, in this case, because Grenoble is part of the publisher name, should be omitted.
 * —Trappist the monk (talk) 16:53, 21 May 2023 (UTC)
 * —Trappist the monk (talk) 16:53, 21 May 2023 (UTC)

Trial registration
Seems we should add support for trial registration for those articles with registration. Adding a parameter for trial registrations would be easier on authors than asking them to add a second template such as https://en.wikipedia.org/wiki/Template:ClinicalTrialsGov Badgettrg (talk) 16:33, 21 May 2023 (UTC)


 * That's what registration is for. &#32; Headbomb {t · c · p · b} 18:53, 21 May 2023 (UTC)

Mystery error
Hi, Jan and Herb Conn has a strange error reported by this template: ": Check date values in: |archive-date= (help)". Anybody knows how to fix it? Jarekt (talk) 01:37, 22 May 2023 (UTC)


 * It took me three edits because my hands are shaky and I've been using a different date format all weekend, but the parameter didn't like Feb. 23, 2008. 23 February 2008 would have worked, but the article uses MDY format so it wanted the comma back between the day and the year, as February 23, 2008. Folly Mox (talk) 02:44, 22 May 2023 (UTC)

Wikidata ID
It wound be nice to be able to tie a citation of a work to its Wikidata ID. --EncycloPetey (talk) 01:02, 9 May 2023 (UTC)


 * Why? We have too many useless junk ids cluttering up our references already, making it very hard for readers to guess which of dozens of links might actually lead them to the reference itself. How does adding one more junk id that definitely will not lead to the reference itself improve that situation? —David Eppstein (talk) 01:13, 9 May 2023 (UTC)
 * Because all the data for the citation has to be entered locally on each page where the citation appears, instead of pulling the citation data from Wikidata. It would be nice to be able to cite by simply using the citation data from Wikidata. --EncycloPetey (talk) 16:20, 10 May 2023 (UTC)
 * You can already autopopulate citation using DOI, URL, ISBN etc. Also having the details here allows editors to edit them without having to go to a separate site. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:33, 10 May 2023 (UTC)
 * That only works for books that have a DOI, URL, or ISBN. Most books I work with don't have those.  Having the details here means that all the information has to be put in on every page where the citation is made.  If it were on Wikidata, the citation would only require the WD ID, and any pages or chapter number cited.  --EncycloPetey (talk) 18:43, 10 May 2023 (UTC)
 * Or you could create a template for the work, and only have to enter the template name. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:50, 10 May 2023 (UTC)
 * @EncycloPetey@ActivelyDisinterested see Template:Cite Q. – Finnusertop (talk ⋅ contribs) 10:57, 24 May 2023 (UTC)
 * "Until the matter of transcluding Wikidata on Wikipedia is resolved (most likely with a huge and contentious RFC) usage of this template should be extremely vetted to ensure that all of the transcluded information is accurate.", as I have no interest in editting a separate project it's not something I would use. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 11:30, 24 May 2023 (UTC)
 * This has been previous discussed. Here are some simple searches: wikidata, QID. Izno (talk) 01:19, 9 May 2023 (UTC)

Looking for tools to fix harv, sfn, and bibliography citations
Hi, this may be an odd place to ask this question but I feel like the people who watch this page are likely to have the answer. I recently did a large update to the Piri Reis map article which previously cited Atlantis books as sources. I now need to check which sources in the Bibliography are no longer used.

Is there a tool to check a bibliography for unused sources? And also is there a tool to check for shortened footnotes pointed at nonexistent sources? I feel certain that I have seen other editor use something like that before, but I don't know where to look. Thanks in advance, Rjjiii (talk) 06:26, 29 May 2023 (UTC)
 * Yes, the script User:Trappist the monk/HarvErrors does what you are looking for. -- John of Reading (talk) 08:00, 29 May 2023 (UTC)
 * You're awesome! That helped a lot, Rjjiii (talk) 08:16, 29 May 2023 (UTC)

multiple names in |first=
See.

I have tweaked the module sandbox to detect commas and semicolons in firstn parameter values. It uses the same detector that is used to detect commas and semicolons in lastn parameter values with a different limit. lastn parameters allow one comma but firstn parameter allows none.

I know of no reason why commas should be allowed in firstn. Usually a comma is used to separate post nominals, degrees, ranks, affiliations, and other non-name stuff from the given name. MOS:JR proscribes the use of a comma separator before generational suffixes.

This search finds 2500+ articles where firstn has a comma. The search also times out so who knows how many articles are afflicted with this problem. As the code currently exists, commas detected in both firstn and lastn parameters are categorized in (currently holding  articles). Should we have separate given/surname subcategories?

Examples:
 * – violates MOS:JR
 * – a single comma allowed in last
 * – multiple commas not allowed
 * – multiple commas not allowed
 * – multiple commas not allowed

This test also applies to the other name lists (editor-firstn, translator-firstn, etc) and of course, to all of their aliases.

—Trappist the monk (talk) 12:06, 31 May 2023 (UTC)


 * I realize you're focussed on first and not last here, but your examples got me thinking (third time this year!), and I can't think of any examples where an actual last/family/surname would legitimately have a comma in it. So with some trepidation I ask the possibly heretical question: do we really want to allow for the inclusion of a first name in last? I'm always trying to make last and first so as to produce clean COinS metadata. Isn't that what we should be fostering? &mdash; JohnFromPinckney (talk / edits) 13:09, 31 May 2023 (UTC)
 * People put full names into last with some frequency, and if I recall correctly, have been advised to do so here. – Jonesey95 (talk) 13:45, 31 May 2023 (UTC)
 * In the best of all possible worlds, there would be no Surname, Given name or Surname, Given name, but, alas, author is an alias of last and editors know that the two can be used interchangeably so that's cows over the dam. At least with Surname, Given name or Surname, Given name the metadata aren't wholly corrupted as they are when multiple names or other extraneous text is included in last and/or first.  And, in the metadata, only the first author name, if given with last and first is split into   and   k/v pairs.  All other author names are assembled from lastn and firstn and placed in separate   k/v pairs.  For example:
 * Still, in the best of all possible worlds, names of persons in the name lists (Eastern name order excepted – we don't have a solution for them) use only the last/first parameter forms. Not holding my breath for that to happen...
 * —Trappist the monk (talk) 14:02, 31 May 2023 (UTC)
 * I have found that many of these invalid first values appear to be caused by an automated tool within the Visual Editor (and maybe other ways of editing WP pages as well). See this initial report; I or we may need to file bugs against Citoid or Zotero to get the source of these problems fixed. – Jonesey95 (talk) 13:45, 31 May 2023 (UTC)
 * The issue isn't just in VE, using the reftoolbar in source to autogenerate the reference has the same issue. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:19, 31 May 2023 (UTC)
 * I get this occasionally using Re/Fill, typically when populating citations from sites like CNN. For example, in Adam Schiff:  {{Cite web |last=Barrett |first=Clare Foran, Ali Zaslav, Ted |date= ... . BD2412  T 01:57, 2 June 2023 (UTC)
 * —Trappist the monk (talk) 14:02, 31 May 2023 (UTC)
 * I have found that many of these invalid first values appear to be caused by an automated tool within the Visual Editor (and maybe other ways of editing WP pages as well). See this initial report; I or we may need to file bugs against Citoid or Zotero to get the source of these problems fixed. – Jonesey95 (talk) 13:45, 31 May 2023 (UTC)
 * The issue isn't just in VE, using the reftoolbar in source to autogenerate the reference has the same issue. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:19, 31 May 2023 (UTC)
 * I get this occasionally using Re/Fill, typically when populating citations from sites like CNN. For example, in Adam Schiff:  {{Cite web |last=Barrett |first=Clare Foran, Ali Zaslav, Ted |date= ... . BD2412  T 01:57, 2 June 2023 (UTC)

what to do with ?
I would like to unhide the missing periodical error messages. The last time that I raised this topic, there were editors who objected because articles using, a long-time redirect to , would suddenly be showing error messages for references that are not to periodicals. At the time of the last discussion, had about 52,000 articles. That number has increased to 54,700+ as I write this.

Resurrecting this topic came to mind because of a discussion about Editor Qwerfjkl's bot at. The bot notifies editors when they make an edit that adds an article to a cs1|2 error category (only some categories, apparently); was chosen to be one of those because it is listed at. Because the error message is hidden, most editors are unaware that they have written a cs1 periodical template (, its redirects and, or , its redirect ) that requires a periodical parameter. Also, these same editors are unaware of preexisting missing-periodical errors because the error messages are hidden. Maybe unhiding the error message will help that...

Some history:

According to this search there are ~6900 articles that use the redirect ; the redirect is used in ~10 articles.

Of the ~6900 articles that use :
 * ~3700 use url
 * some use periodical parameters:
 * journal: ~20
 * magazine: ~2
 * newspaper: ~4
 * periodical: ~2
 * website: ~40
 * work: ~530
 * some use the various identifiers:


 * arxiv: ~30
 * asin: ~0
 * bibcode: ~25
 * biorxiv: ~0
 * citeseerx: ~400
 * doi: ~200
 * eissn: ~0
 * hdl: ~670
 * isbn: ~3
 * ismn: ~0
 * issn: ~15
 * jfm: ~0
 * jstor: ~100
 * lccn: ~2
 * mr: ~1
 * oclc: ~70
 * ol: ~10
 * osti: ~20
 * pmc: ~20
 * pmid: ~50
 * rfc: ~0
 * sbn: ~0
 * ssrn: ~230
 * s2cid: ~270
 * zbl: ~3

Since editors objected to showing the error messages while exists as a redirect to, we must create a real  template. Before we do that, we should attend to the instances of the redirect that exist in article space. The obvious first steps (to me) are: That being done, the only remaining templates should be those without url, identifier parameters, or periodical parameters. We can then replace the innards of with the innards from  (which temporarily becomes a redirect to ). Once all instances of in article space have been replaced with,  will be deleted.
 * create a placeholder template or some such that is a real cs1 template that renders like a periodical template (value from title is rendered upright quoted); requires publisher; does not support url (if a url is available use  or other appropriate template); perhaps the template accepts a subset of the usual cs1 parameters (we might start with the limited lists used by the preprint templates ( etc) augmented by certain identifier and other appropriate parameters)
 * convert the 600ish templates with periodical parameters to the appropriate periodical template or to
 * convert the templates that use identifier parameters to an appropriate template or to
 * convert the templates that use url to or other appropriate template (not to )

Will this work? I'm sure that there are things that I have not considered. What are those things? And, I guess finally, should we bother? Or, are we content to repeat the discussions listed at the top of this post ad nauseum?

—Trappist the monk (talk) 21:25, 26 May 2023 (UTC)


 * Cite document/paper should just be it's own thing. No different than a cite journal/magazine, but without the requirement that a 'work' parameter needs to be set. &#32; Headbomb {t · c · p · b} 21:44, 26 May 2023 (UTC)
 * My feeling is that cite document should be a generic template for citing anything, depending on what parameters are given to it, exactly the same as for citation but Citation Style 1 instead of Citation Style 2. It should not assume that the document is of some specific subtype. —David Eppstein (talk) 22:39, 26 May 2023 (UTC)
 * Cite document/paper should just be it's own thing. Yep, that is what I'm suggesting.  By [its] own thing I mean that it should fill the gap between something that can be cited using  (requires a url) and a periodical (requires a work parameter) or a book or an encyclopedia...  The purpose of the new  template is to cite published stand-alone sources that are not available through a url and are not published as part of a larger work (periodical, book, encyclopedia, proceedings, etc).  These stand-alone sources may be available online via a persistent identifier (doi in particular) so the new  must support a subset of our identifier list.
 * I cannot support the idea of the new being a -like template.  cs1 templates are specific to things: books, magazines, theses, conferences, signs, etc.  All of these templates are named accordingly; a -like  would violate that convention.
 * —Trappist the monk (talk) 14:30, 27 May 2023 (UTC)
 * There is no necessary reason for cs1 templates to be specific to things. It is merely unnecessary work for people who edit references, and very frequently incorrectly classified. It is entirely possible now to write in cs1 using unspecific templates (citation with mode=cs1) and making cite document do the same thing would ease that. The difficulty with defining what kind of specific thing cite document should be should be a sign to you that it is not actually significant. —David Eppstein (talk) 19:05, 28 May 2023 (UTC)
 * Perhaps there is no reason for cs1 templates to be specific to things but, interestingly enough, they are specific to 'things':  for books,  for ArXiv preprints,  for scholarly and academic journals,  for audio and visual works,  for conference proceedings,  for signs, plaques, gravestones, and other non-video visuals, etc.  So,  is created for published  papers and other documents that are not published in a book or a periodical or an encyclopedia or online (as written the proposed template is a bit gray on that last point because it will accept certain named identifiers: doi, etc).
 * It is entirely possible now to write in cs1 using unspecific templates (citation with mode=cs1) Well, sort of.  What you get is a rendering suitable for books but not correct for a standalone document:
 * cf :
 * —Trappist the monk (talk) 22:30, 28 May 2023 (UTC)
 * In what sense is it "correct for a standalone document" to use the article-within-larger-work double-quoted-upright style for its title rather than the unquoted-italic style that we use for other standalone documents, like books? These are made-up conventions with no justification for such dogmatism. —David Eppstein (talk) 22:58, 28 May 2023 (UTC)
 * MOS:MINORWORK. I did not make that up.  My experience with existing  templates, when another template would have been the better choice, is that the cited source is short.  There have been the rare occasions where  has been used to cite an entire book, of course such a source is better served with .  From that experience, and the lack of complaints that the current  isn't rendering the title in italic font suggested to me that the MOS:MINORWORK style is appropriate.
 * —Trappist the monk (talk) 00:50, 29 May 2023 (UTC)
 * @Trappist the monk: I like your proposal, but suggest using instead of, and suggest we could agree on an edit summary that would explain why we're changing from  to .  Would there be a benefit in seeing if Citation bot could convert some of the existing  templates with url to  with the appropriate identifier instead of a URL?  Thanks! GoingBatty (talk) 00:55, 27 May 2023 (UTC)
 * @Trappist the monk ...or convert them to with the appropriate parameters.  GoingBatty (talk) 01:31, 27 May 2023 (UTC)
 * Yeah, but... cs1|2 uses, for example,  for rendering a reference using the Module:Citation/CS1/sandbox suite.  I think that  is too close to that.  We could use  or some such.
 * If Citation bot can change an existing  to another cs1 template, great.  I emphasized correctly because I know that I've had to think about how best to convert  to some other template and it often requires adding to the chosen template.  A common edit summary is a good idea.
 * And, you really don't need to ping me here...
 * —Trappist the monk (talk) 14:30, 27 May 2023 (UTC)
 * Here are examples of the first hack to make a real cs1 template:
 * – should be
 * – should be
 * – should be
 * emits 'book'-type metadata format (the default for those references that don't have a specifically defined metadata format):
 * The template accepts parameters from these parameter lists in Module:Citation/CS1/Whitelist/sandbox:
 * (link)
 * (link)
 * (link)
 * (link)
 * Because this will be the only cs1 template that requires publisher there is a new error message and category:
 * Comments invited.
 * —Trappist the monk (talk) 15:32, 28 May 2023 (UTC)
 * I wonder if identifiers actually should be supported, as we can transform an identifier into a URL trivially, thus making it possible to use cite web. Izno (talk) 16:44, 31 May 2023 (UTC)
 * You both seem determined to make this template as useless as possible for those catch-all cases that don't fit into any of the other cite X templates. I don't understand why, but as long as you keep doing this I'm going to keep using citation. If you get started on making that one equally useless, I may have to fall back to manually-formatted html templates and an exclusion from all citation bot edits. These templates should be for the purpose of making it easy to get consistently-formatted citations, not for the purpose of fitting citations into some crazy Procrustean bed that exists in your imaginations of what an ideal citation must be. —David Eppstein (talk) 06:17, 2 June 2023 (UTC)
 * Then for a generic print source template, why not just make a cite text template that wraps citation with the "mode" parameter preset? Rjjiii (talk) 01:36, 4 June 2023 (UTC)
 * – should be
 * emits 'book'-type metadata format (the default for those references that don't have a specifically defined metadata format):
 * The template accepts parameters from these parameter lists in Module:Citation/CS1/Whitelist/sandbox:
 * (link)
 * (link)
 * (link)
 * (link)
 * Because this will be the only cs1 template that requires publisher there is a new error message and category:
 * Comments invited.
 * —Trappist the monk (talk) 15:32, 28 May 2023 (UTC)
 * I wonder if identifiers actually should be supported, as we can transform an identifier into a URL trivially, thus making it possible to use cite web. Izno (talk) 16:44, 31 May 2023 (UTC)
 * You both seem determined to make this template as useless as possible for those catch-all cases that don't fit into any of the other cite X templates. I don't understand why, but as long as you keep doing this I'm going to keep using citation. If you get started on making that one equally useless, I may have to fall back to manually-formatted html templates and an exclusion from all citation bot edits. These templates should be for the purpose of making it easy to get consistently-formatted citations, not for the purpose of fitting citations into some crazy Procrustean bed that exists in your imaginations of what an ideal citation must be. —David Eppstein (talk) 06:17, 2 June 2023 (UTC)
 * Then for a generic print source template, why not just make a cite text template that wraps citation with the "mode" parameter preset? Rjjiii (talk) 01:36, 4 June 2023 (UTC)
 * Comments invited.
 * —Trappist the monk (talk) 15:32, 28 May 2023 (UTC)
 * I wonder if identifiers actually should be supported, as we can transform an identifier into a URL trivially, thus making it possible to use cite web. Izno (talk) 16:44, 31 May 2023 (UTC)
 * You both seem determined to make this template as useless as possible for those catch-all cases that don't fit into any of the other cite X templates. I don't understand why, but as long as you keep doing this I'm going to keep using citation. If you get started on making that one equally useless, I may have to fall back to manually-formatted html templates and an exclusion from all citation bot edits. These templates should be for the purpose of making it easy to get consistently-formatted citations, not for the purpose of fitting citations into some crazy Procrustean bed that exists in your imaginations of what an ideal citation must be. —David Eppstein (talk) 06:17, 2 June 2023 (UTC)
 * Then for a generic print source template, why not just make a cite text template that wraps citation with the "mode" parameter preset? Rjjiii (talk) 01:36, 4 June 2023 (UTC)

Basic error checking for name fields?
There's a pretty big ongoing cleanup effort related to script assisted reference filling. There's an active AN thread and a related MfD. The takeaway is that script-assisted referencing has been producing a lot of garbage citations that users fail to double check, and there's pieces of the problem scattered in user scripts, in Citoid, and possibly all the way up to Zotero, although their processing of DOIs, PMIDs, etc is extremely reliable.I know that the citation templates do some basic error checking. For example, they hate it when I try to cite a work written in the BCE era, telling me to check the values in my date= field. Can some of this error checking be extended to the name fields? (By which I mean last, editorn-first, etc). Like, the reference filling tools will put stupid stuff there like addresses, phone numbers, "a ranger may not answer if they are already speaking with someone else" (see linked discussions for more examples).How difficult would it be to have the template throw an error if a name field contains, say, more than five whitespace characters (which should handle most corporate authors)? Or more than zero digits? It seems big improvements need to be made to a lot of code libraries, but warning users when parameters contain information they very obviously shouldn't seems like a reasonable first step towards reducing citation damage. Folly Mox (talk) 05:47, 6 June 2023 (UTC)
 * Names are difficult. cs1|2 does have these categories that are related to the name lists:
 * We could promote some or all of the maintenance categories to errors (at the risk of torches and pitchforks)...
 * —Trappist the monk (talk) 12:02, 6 June 2023 (UTC)
 * I just spot-checked about twenty articles from Category:CS1 maint: numeric names: authors list‎ and all seemed to be unambiguously erroneous (with the exception of the author listed by their twitter handle, a series of ✖️s, which I could make no sense of). Most of the articles in that category seem to be in the Draft: namespace as well.Maybe determining whether something is a complete sentence versus a concatenation of several names is difficult in the general case, but is there any way to separate out when this pattern shows up in author= and authors= (where it could be a list) and fields like first3=, where it is more likely an error? Folly Mox (talk) 12:18, 6 June 2023 (UTC)
 * I think that if you look closer at that you will find that most of the articles listed in that category are  in the   namespace.  Looks to me like someone has been slowly picking away at that category and has got as far as 'T'.
 * At the next update to the module suite, there will be support for multiple name detection in the various  parameters.  See.
 * —Trappist the monk (talk) 13:08, 6 June 2023 (UTC)
 * I stand corrected: most of the articles are in namespace 1. Spot-checking another sixteen or twenty, I found without exception every numeric author to be a genuine error, and in that sample each article had the date or part thereof duplicated in an author field. I think this category is safe to elevate from maintenance status to error status. Folly Mox (talk) 14:27, 6 June 2023 (UTC)
 * @Trappist the monk: I'd support changing Category:CS1 maint: numeric names: authors list‎ to an error, and asking to have their bot send out notifications for the errors.  Of course, fixing the tool(s) that are creating the bad data in the first place would be great! GoingBatty (talk) 13:52, 6 June 2023 (UTC)
 * We could promote some or all of the maintenance categories to errors (at the risk of torches and pitchforks)...
 * —Trappist the monk (talk) 12:02, 6 June 2023 (UTC)
 * I just spot-checked about twenty articles from Category:CS1 maint: numeric names: authors list‎ and all seemed to be unambiguously erroneous (with the exception of the author listed by their twitter handle, a series of ✖️s, which I could make no sense of). Most of the articles in that category seem to be in the Draft: namespace as well.Maybe determining whether something is a complete sentence versus a concatenation of several names is difficult in the general case, but is there any way to separate out when this pattern shows up in author= and authors= (where it could be a list) and fields like first3=, where it is more likely an error? Folly Mox (talk) 12:18, 6 June 2023 (UTC)
 * I think that if you look closer at that you will find that most of the articles listed in that category are  in the   namespace.  Looks to me like someone has been slowly picking away at that category and has got as far as 'T'.
 * At the next update to the module suite, there will be support for multiple name detection in the various  parameters.  See.
 * —Trappist the monk (talk) 13:08, 6 June 2023 (UTC)
 * I stand corrected: most of the articles are in namespace 1. Spot-checking another sixteen or twenty, I found without exception every numeric author to be a genuine error, and in that sample each article had the date or part thereof duplicated in an author field. I think this category is safe to elevate from maintenance status to error status. Folly Mox (talk) 14:27, 6 June 2023 (UTC)
 * @Trappist the monk: I'd support changing Category:CS1 maint: numeric names: authors list‎ to an error, and asking to have their bot send out notifications for the errors.  Of course, fixing the tool(s) that are creating the bad data in the first place would be great! GoingBatty (talk) 13:52, 6 June 2023 (UTC)
 * —Trappist the monk (talk) 12:02, 6 June 2023 (UTC)
 * I just spot-checked about twenty articles from Category:CS1 maint: numeric names: authors list‎ and all seemed to be unambiguously erroneous (with the exception of the author listed by their twitter handle, a series of ✖️s, which I could make no sense of). Most of the articles in that category seem to be in the Draft: namespace as well.Maybe determining whether something is a complete sentence versus a concatenation of several names is difficult in the general case, but is there any way to separate out when this pattern shows up in author= and authors= (where it could be a list) and fields like first3=, where it is more likely an error? Folly Mox (talk) 12:18, 6 June 2023 (UTC)
 * I think that if you look closer at that you will find that most of the articles listed in that category are  in the   namespace.  Looks to me like someone has been slowly picking away at that category and has got as far as 'T'.
 * At the next update to the module suite, there will be support for multiple name detection in the various  parameters.  See.
 * —Trappist the monk (talk) 13:08, 6 June 2023 (UTC)
 * I stand corrected: most of the articles are in namespace 1. Spot-checking another sixteen or twenty, I found without exception every numeric author to be a genuine error, and in that sample each article had the date or part thereof duplicated in an author field. I think this category is safe to elevate from maintenance status to error status. Folly Mox (talk) 14:27, 6 June 2023 (UTC)
 * @Trappist the monk: I'd support changing Category:CS1 maint: numeric names: authors list‎ to an error, and asking to have their bot send out notifications for the errors.  Of course, fixing the tool(s) that are creating the bad data in the first place would be great! GoingBatty (talk) 13:52, 6 June 2023 (UTC)
 * @Trappist the monk: I'd support changing Category:CS1 maint: numeric names: authors list‎ to an error, and asking to have their bot send out notifications for the errors.  Of course, fixing the tool(s) that are creating the bad data in the first place would be great! GoingBatty (talk) 13:52, 6 June 2023 (UTC)

Archiving hundreds of healthy (live) sources
Is there any need for this type of edit? Shouldn't we just archive dead or unfit sources? Unlike the former, this edit actually makes sense because it did rescue sources. SLBedit (talk) 11:35, 7 June 2023 (UTC)


 * My thought was that the archiving process isn't automatic, so pre-empting the links becoming dead means there is always a viable link in the article. But this might be a question better suited to WT:Link rot -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 14:03, 7 June 2023 (UTC)
 * This is the wrong venue for asking about an approved bot performing link-rot prevention. See Link rot and User:InternetArchiveBot for an explanation of why this is being done. – Jonesey95 (talk) 15:49, 7 June 2023 (UTC)

Discussion on connecting reference properties to CS1 in the Wikidata template
You are invited to join the discussion at Module talk:Wd § References mapping. &#123;{u&#124; Sdkb  }&#125;  talk 20:51, 7 June 2023 (UTC)

Incorrect language display for ISO 639 code
The ISO 639 code fkv (Kven language) displays as "in Kvensk" rather than the expected English-language text "in Kven". For reference "kvensk" is the Norwegian term for the language. I'm not familiar with where this data is collected from, so I have no clue how to fix this. I'd appreciate some help with this issue. ArcticSeeress (talk) 05:29, 8 June 2023 (UTC)
 * Using the MediaWiki feature  at mw:Help:Magic words,   produces  which says "Kvensk" when I write this. fkv says "Kvensk" in all languages I have tried, also in other scripts. For example,   gives . For comparison with a translated language name,   gives  while   gives . I don't think we should try to make overrides of   here at the English Wikipedia so it would have to be fixed at Phabricator or maybe the CLDR database. I found an old 2016 request at T151269 which asked for English names for some languages including fkv: "Kven". But the alleged fix 424556 (click "LocalNames/LocalNamesEn.php") says  . PrimeHunter (talk) 11:18, 8 June 2023 (UTC)
 * At the cldr ticket linked from T151269 is this: there is no intent for CLDR to have the English names of all languages ... and point to http://www.iana.org/assignments/language-subtag-registry/language-subtag-registry as a source for any extra ones that people need. Module:Lang uses data from the IANA registry so:
 * → ('Kven Finnish' is overridden to 'Kven' in Module:Lang/data)
 * We can, and have in the past, overridden the MediaWiki language names in cs1|2 and can probably do that for this language. The preferred name for   is 'Kven'?
 * —Trappist the monk (talk) 12:19, 8 June 2023 (UTC)
 * Yes, the English-language name for fkv is Kven. ArcticSeeress (talk) 03:29, 9 June 2023 (UTC)
 * Fixed in the sandbox:
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)
 * —Trappist the monk (talk) 13:13, 9 June 2023 (UTC)

Check |oclc= value
The use of https://worldcat.org/title/1127958624 raises an error: Why? How to fix it? -- Michael Bednarek (talk) 13:26, 9 June 2023 (UTC)


 * There's an invisible right-to-left mark (\u200f) at the end of the example you gave, which the module is seeing and getting confused by. Removing that should fix it. Aidan9382 (talk) 13:41, 9 June 2023 (UTC)

Double quotes in titles and quotes
(continuing conversation from Help talk:Citation Style 1/Archive 87)

It's not too hard to write code that does what a human editor would do in the vast majority of cases. (I already have code that checks for balanced quote marks for spell-check purposes.) For example, both  and   should display as. The algorithm would have to ignore apostrophes inside of words (so words like  and   don't break it) and before flipping " and ' all the way down make sure that the quote marks are balanced and nested cleanly and don't do weird things like "-inside-". (Otherwise it could display an error and make no changes.)

A more manual solution would be to detect outer double quotes and show a warning on preview. That would probably cut down a lot on people introducing these errors and encourage them to be fixed slowly over time, especially if these landed in a hidden tracking category.

Since there hasn't been any activity in several months, I'll start writing code to slow repair the wikitext based on my best guess of affected templates. -- Beland (talk) 07:08, 6 June 2023 (UTC)
 * I already have code that checks for balanced quote marks for spell-check purposes. Share it?
 * —Trappist the monk (talk) 12:11, 6 June 2023 (UTC)
 * For outer double quotes, all I do is substitute the regex /"\S[^"]{0,1000}?\S"|"\S"|""/ to the empty string. If the remainder contains a double quote mark, there is an imbalance. I could whip up some Python to handle more layers if that would be helpful; it looks like there are hundreds of thousands of instances where outer double quotes are used, and I'm not looking forward to trawling through that. -- Beland (talk) 05:12, 7 June 2023 (UTC)
 * The proposed automatic code would also have to account for apostrophes outside of words, like "Review of 'In My Parents' House', now in theaters". – Jonesey95 (talk) 14:06, 6 June 2023 (UTC)
 * It might be difficult to do that reliably; I think it would be fine to in those cases ask for manual review. Unless that happens an awful lot, in which case, yeah, we could code in some intelligence regarding English plural possessives. -- Beland (talk) 05:01, 7 June 2023 (UTC)
 * You mean they should both display as ? —  Qwerfjkl  talk  14:37, 6 June 2023 (UTC)
 * Whoops, yes, fixed. -- Beland (talk) 05:01, 7 June 2023 (UTC)


 * Edge cases, off the top of my head: cruisin' for a bruisin', down 'n dirty, rock 'n roll (which can also be rock 'n' roll or rock n' roll), git'r'dun, kickin' ass 'n takin' names. jp×g 09:50, 10 June 2023 (UTC)

Another generic title
Hello, can you add "Archived" as a generic title. Currently, around 61 instances. Keith D (talk) 22:05, 9 June 2023 (UTC)
 * Another generic title is "Error" currently 475 instances. Keith D (talk) 20:10, 12 June 2023 (UTC)

Genre parameter for reviews and press releases
I'm wondering about the best parameter to use for genre of a work. I realize the genre is not usually necessary to be included, but I find it helpful specifically in the cases of reviews and press releases. In the past, I have sometimes used the format parameter, and I have recently sometimes used the department parameter. Neither of these seems correct. Is it best just to put the genre in brackets at the end of the title parameter? I am interested in following best practices. Daask (talk) 12:52, 14 June 2023 (UTC)


 * Why do you need to put the 'genre' in the first place? What is 'genre' anyway? &#32; Headbomb {t · c · p · b} 12:55, 14 June 2023 (UTC)
 * exists. It presets type.  To cite a review you can Review.  Do not add stuff that is not part of the source's title to title.  Do not use format for anything other than the source's electronic file format:  ok;  not ok.
 * —Trappist the monk (talk) 13:04, 14 June 2023 (UTC)
 * —Trappist the monk (talk) 13:04, 14 June 2023 (UTC)


 * Thanks! I had forgotten about Help:Citation Style 1 and somehow missed it when I went looking today. Thanks again! Daask (talk) 18:53, 14 June 2023 (UTC)

deprecated parameters
This to note that sometime within the past 24 hours, went empty for the first time since 2021-01-03. It wasn't me who did that. There were some objection to the deprecation of lay-date, lay-format, lay-source and lay-url, especially from WP:MED, so I stopped cleaning that category. Still, slowly over time, other editors have removed articles from the category. I intend to wait a while to see what happens, but I anticipate that the category will remain empty. If it does remain empty until say, this time next month, I'll remove support for the lay-* parameters.

Since it was deprecated, I have not seen transcripturl appear in Category:CS1 errors: deprecated parameters so I have removed support for that alias of transcript-url from the sandboxen.

—Trappist the monk (talk) 14:24, 23 April 2023 (UTC)
 * The category has remained empty so I have removed support for lay-date, lay-format, lay-source and lay-url from the sandboxen.
 * —Trappist the monk (talk) 13:20, 23 May 2023 (UTC)
 * I was going to add a transcipt url to a citation today but it was missing. Can it be undeprecated and brought back? I think it would be more useful than most things if a transcript was available. I don't know coding so I could be all wrong about what I'm talking about, but I thought it was important to say something somewhere on the talk page. Let me know. Kire1975 (talk) 21:02, 15 June 2023 (UTC)
 * transcripturl is deprecated; transcript-url is not.
 * —Trappist the monk (talk) 21:58, 15 June 2023 (UTC)

ProQuest (or similar) as merely an alternative
I don't have a subscription to the NYT; thanks to the Wikipedia Library, I can read the NYT via ProQuest. Therefore in this edit, I changed | url=https://www.nytimes.com/2019/10/09/arts/jill-freedman-dead.html | to | url=https://www.proquest.com/docview/2302935851 | via=ProQuest | Readers with access to ProQuest but not otherwise to the NYT (and surely there are many such readers) will thank me.

Readers with access to the NYT but not to ProQuest (and surely there are many of these too) will not thank me.

Various other approaches come to mind. I could have left the reference untouched, and let readers figure out for themselves that this article is somewhere at ProQuest. Or I could have left just the content of the "cite news" template untouched but added immediately after it (and within the reference) "Also available at .", or similar. However, I suspect that this matter has already been exhaustively discussed somewhere. Perhaps somebody could point me to such a page (or of course make a suggestion here). -- Hoary (talk) 07:59, 22 May 2023 (UTC)


 * I would suggest leaving the reference as it was, as the reference was for the NYT. Simply add "Also available via proquest " after the cite, but before the ending ref tag, or use proquest in the |id= field (as |id= ). -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 11:35, 22 May 2023 (UTC)
 * "id=": Yes, that does it. Thank you, ActivelyDisinterested. -- Hoary (talk) 00:05, 23 May 2023 (UTC)

Can you add "remap" support for Urum language [uum]
There are citations referencing the Urum language [ISO 639-3: uum], but they don't work either as the name or the code. Is it possible to add entries for it in the language "remapping" tables (to make it work for the time that MediaWiki still does not offer any localization support for it)? That language is not dead, it originates from Crimea (from Pontic/Greek orthodox Christians that lived there all around the Black Sea before islamisation of the Ottoman Empire, written there in Cyrillic, also traditionally in Greek) and remains used in Turkey, southern Ukraine (Donetsk/Azov region), Russia and parts of Georgia. verdy_p (talk) 22:53, 10 June 2023 (UTC)
 * The purpose of the remapping is to override MediaWiki language tag definitions so that the rendered language names are in line with en.wiki expectations.
 * but they don't work either as the name or the code. I dispute that:
 * A couple of quick cirrus searches did not find any cs1|2 templates using uum or Urum:
 * 
 * 
 * —Trappist the monk (talk) 23:47, 10 June 2023 (UTC)
 * The above shows CS1 maint: unrecognized language. Gonnym (talk) 16:56, 16 June 2023 (UTC)
 * Of course, because cs1|2 can't tell the difference between an unknown language name and a misspelled language name:
 * —Trappist the monk (talk) 17:38, 16 June 2023 (UTC)
 * The above shows CS1 maint: unrecognized language. Gonnym (talk) 16:56, 16 June 2023 (UTC)
 * Of course, because cs1|2 can't tell the difference between an unknown language name and a misspelled language name:
 * —Trappist the monk (talk) 17:38, 16 June 2023 (UTC)
 * —Trappist the monk (talk) 17:38, 16 June 2023 (UTC)
 * —Trappist the monk (talk) 17:38, 16 June 2023 (UTC)

Multiple OCLCs in one reference
Probably sounds daft. But what if you want to use Cite book to refence a range of books, with multiple volumes; for example 2 volumes, 1936–1937. But each volume would have its own identifier (there being no overarcing ID for the series as a set). See whaty I mean? Thanmks for any help! SN54129 14:19, 16 June 2023 (UTC)


 * When I've had this come up I've just referenced each volume separately, with its own template. Does that work in your use case? Mackensen (talk) 14:58, 16 June 2023 (UTC)
 * Cheers :) I had a nasty feeling someone would say that! Thanks for confirming though, at least I won't have to waste time experimenting. Have a good weekend!   SN54129  16:23, 16 June 2023 (UTC)

create a |department= alias |title-note=?
I have now been part of two recent discussions that question the correctness of department when used to do other than name the department of a (typically periodical) publication (Sports, Op Ed, Entertainment, etc). In both discussions, the intent was to include some sort of descriptive text that doesn't fit the 'department' mold. In the days of, department was assigned to the metaparameter TitleNote; the Module:Citation/CS1 metaparameter   continues to do that. I had thought that there was a similarly named parameter titlenote, title_note, title-note but apparently not. I wonder if we should create an alias that matches the metaparameter title-note. Of course, type can also be used; the differences between these two parameters are positioning and that the value assigned to type is wrapped in parentheses:

Neither of type and department contribute to a citation's metadata

So, should we create an alias title-note?

—Trappist the monk (talk) 14:44, 14 June 2023 (UTC)
 * There having been no comment: added.
 * —Trappist the monk (talk) 13:44, 19 June 2023 (UTC)

Cite report issue parameter not displaying
Cite report used below does not appear to support displaying the  number if no   is supplied. I am not sure if  in the source below means volume 1, issue 08, so I had skipped adding the volume parameter. Any suggestions or insights on how to correctly fix the citation?

Sanglahi86 (talk) 18:01, 28 May 2023 (UTC)
 * You're right, does not support issue because reports are not periodicals.  If your source is publisher periodically, as it appears to be since the publisher has attached a 'number' to it, you might rewrite like this:
 * Or, keep it as and put the report number in id:
 * —Trappist the monk (talk) 22:56, 28 May 2023 (UTC)
 * It's a bit confusing because Cite techreport works with number, but Cite report uses docket, which Cite thesis uses as a synonym for id. In a perfect world, Cite techreport and Cite report would both use number and place the input within the parenthetical that follows the title, so for the following input:
 * you'd get:
 * Technical Manual of Airship Aerodynamics (PDF) (Report TM 1-320). Washington, D.C.: United States Department of War. February 11, 1941. Retrieved June 19, 2023.
 * instead what you get when you use docket or id:
 * Technical Manual of Airship Aerodynamics (PDF) (Report). Washington, D.C.: United States Department of War. February 11, 1941. TM 1-320. Retrieved June 19, 2023.
 * The report number is getting lost at the end. Having it in the parenthetical would better match my expectations for keeping information in the proper, relevant order, and it would be more logical for the editor creating the citation. &mdash;Carter (Tcr25) (talk) 16:22, 19 June 2023 (UTC)
 * You can override the automatic Report applied by the template by writing: Report TM 1-320:
 * For this citation, perhaps it is better to use (really just a redirect to ) because the source calls itself a 'manual', not a 'report':
 * —Trappist the monk (talk) 16:49, 19 June 2023 (UTC)
 * Thanks for the workaround. I still think it's worth rethinking how this parameter works on cite report and cite technical report, but this helps. &mdash;Carter (Tcr25) (talk) 18:33, 19 June 2023 (UTC)
 * The report number is getting lost at the end. Having it in the parenthetical would better match my expectations for keeping information in the proper, relevant order, and it would be more logical for the editor creating the citation. &mdash;Carter (Tcr25) (talk) 16:22, 19 June 2023 (UTC)
 * You can override the automatic Report applied by the template by writing: Report TM 1-320:
 * For this citation, perhaps it is better to use (really just a redirect to ) because the source calls itself a 'manual', not a 'report':
 * —Trappist the monk (talk) 16:49, 19 June 2023 (UTC)
 * Thanks for the workaround. I still think it's worth rethinking how this parameter works on cite report and cite technical report, but this helps. &mdash;Carter (Tcr25) (talk) 18:33, 19 June 2023 (UTC)
 * —Trappist the monk (talk) 16:49, 19 June 2023 (UTC)
 * Thanks for the workaround. I still think it's worth rethinking how this parameter works on cite report and cite technical report, but this helps. &mdash;Carter (Tcr25) (talk) 18:33, 19 June 2023 (UTC)
 * —Trappist the monk (talk) 16:49, 19 June 2023 (UTC)
 * Thanks for the workaround. I still think it's worth rethinking how this parameter works on cite report and cite technical report, but this helps. &mdash;Carter (Tcr25) (talk) 18:33, 19 June 2023 (UTC)

What if the archive-url is an earlier version of the source that does not support the material it's attached to?
We've got a situation where a news story was published, then it was archived at IA, then later it was updated by the publisher. We use the updated story in a WP article, but the material (including a quote) isn't in the archive-url, it's only in the updated story (at the original url). Is it appropriate to use that archive-url even if it doesn't support the material? This came up at Talk:2023 Titan submersible incident. GA-RT-22 (talk) 04:11, 20 June 2023 (UTC)


 * Now I understand what you meant. What you should do in this case is archive the new version of the URL using the Wayback Machine and then add that to the article (as I have done here). Nythar  (💬-🍀) 04:19, 20 June 2023 (UTC)

Template:Cite web documentation - url-status prerequisite inconsistency
About halfway through the Template:Cite web documentation it says. Clear enough. The problem is that right near the top of the page, in the Full parameter set in vertical format table of the Usage section, the only prerequisite listed for url-status is. Should that not be changed to ? Could this be contributing to the confusion on the correct usage of url-status? Let's face it, many editors are not going to be reading the documentation beyond the Usage section. - Wikkiwonkk (talk) 16:25, 20 June 2023 (UTC)
 * No doubt the template documentation can be improved. I think, however, that the inconsistency that you describe is not the cause of the misuse of url-status.  I suspect that much of the misuse is caused by visual editor because the  lists url-status, archive-url, and archive-date as 'suggested' parameters.  This search finds about 34300 articles with url-status archive-url archive-date (where url-status may have an assigned value of   or   or nothing; the others are present but empty).  This parameter sequence is the sequence listed in the TemplateData under  ; this edit for example.
 * The real fix, is to fix ve and/or TemplateData so that ve doesn't add empty parameters to a template (even when suggested); is mindful of prerequisites; can handle enumerated parameters without TemplateData having a bazillion authorn entries where the only thing that changed is the enumerator; etc. Are you holding your breath waiting for these fixes?  Don't.
 * I suppose that the 'documentation' (such as it is) for url-status in TemplateData might be tweaked so that its  begins with a statement of requirements: "Requires |url= and |archive-url=; If set to 'live', the title display is adjusted ..."  Similarly, the description for archive-url and archive-date might begin with their prerequisites...  Will this do any good?  I'm skeptical, but feel free to try.
 * —Trappist the monk (talk) 17:24, 20 June 2023 (UTC)
 * @Wikkiwonkk - I fixed the issue with Template:Cite web/doc that you suggested, and leave Trappist's suggestions for you. GoingBatty (talk) 17:54, 20 June 2023 (UTC)
 * Thanks. I'm not going to touch TemplateData though, modifying a template that is used so widely is way above my current editing risk tolerance. - Wikkiwonkk (talk) 03:53, 21 June 2023 (UTC)
 * @Wikkiwonkk - I fixed the issue with Template:Cite web/doc that you suggested, and leave Trappist's suggestions for you. GoingBatty (talk) 17:54, 20 June 2023 (UTC)
 * Thanks. I'm not going to touch TemplateData though, modifying a template that is used so widely is way above my current editing risk tolerance. - Wikkiwonkk (talk) 03:53, 21 June 2023 (UTC)

Citing sections?
There doesn't seem to be a template for citing standards, e.g., FIPS PUB 60-1, ANSI X3.53-1976. Is cite report appropriate for that purpose? --

— Preceding unsigned comment added by Chatul (talk • contribs), 8:32, 18 May 2023 (UTC)


 * Your question contradicts your section heading. If you are asking about citing standards, then see Cite techreport. If you are asking about about citing a specific section within a source, maybe try the chapter or at parameters in most any citation template. —  void  xor  15:59, 22 June 2023 (UTC)

Google drive pages: which url-access
I am finding references to Google Drive pages that say, "You need access", for example: This came up in Puracé. I believe the cite template needs a url-access parameter; the three choices are registration, limited, and subscription. Which one should be used? —Anomalocaris (talk) 09:23, 19 June 2023 (UTC)


 * The citation appears to sugget that Google drive is the only place to find the map, in which case I don't believe it's usable for referencing (as it hasn't been published). Also I doubt having links to random Google drive files is a good idea from a security perspective. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 09:56, 19 June 2023 (UTC)
 * I had a quick look around online. It's possible that this https://recordcenter.sgc.gov.co/B4/13010010024433/mapa/Pdf/0101244331300002.pdf PDF file is the map being referenced. If it covers the details being referenced simply switch it out for the Google drive link. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 10:00, 19 June 2023 (UTC)
 * ActivelyDisinterested: Thanks. I used your link in Puracé and made a few similar replacements in a few other articles. Then I got to thinking that there might be a lot of Google Drive links in Wikipedia, so I did a search:
 * : link
 * which has 4,854 results. The first one happens to be N. H. Wilson, which has 3 drive.google.com links
 * all of which work fine, without any need to gain access first. Do you see any problem retaining these links? Should Wikipedia create a project of replacing or eliminating Google drive links in 4,854 articles? —Anomalocaris (talk) 17:48, 22 June 2023 (UTC)
 * Just as a data point, 2022 Guamanian general election has a link to Google Drive, but it's the link provided by the Guam Election Commission on its homepage for the official results. That would be a case where the Google Drive link (which isn't restricted access) may need to stay, since there's no other page for the results. There may be other instances were an alternative link to an otherwise allowable source isn't available. &mdash;Carter (Tcr25) (talk) 18:11, 22 June 2023 (UTC)
 * I think though that quite are few are going to be copyvio, the three above certainly look like it. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:16, 22 June 2023 (UTC)
 * Totally agree. I just wanted to point out an instance (perhaps the only one) where such a link might be both valid and desirable. &mdash;Carter (Tcr25) (talk) 18:19, 22 June 2023 (UTC)
 * The item in The New Rhodesia is now over 72 years old. The other two are over 61 years old. How long do these copyrights last? 75 years? —Anomalocaris (talk) 18:33, 22 June 2023 (UTC)
 * 70 years after the death of the author, at least in the US/UK. Without knowing the date of the authors death it's best to be cautious. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:57, 22 June 2023 (UTC)
 * Totally agree. I just wanted to point out an instance (perhaps the only one) where such a link might be both valid and desirable. &mdash;Carter (Tcr25) (talk) 18:19, 22 June 2023 (UTC)
 * The item in The New Rhodesia is now over 72 years old. The other two are over 61 years old. How long do these copyrights last? 75 years? —Anomalocaris (talk) 18:33, 22 June 2023 (UTC)
 * 70 years after the death of the author, at least in the US/UK. Without knowing the date of the authors death it's best to be cautious. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:57, 22 June 2023 (UTC)

Documentation needed for linking multiple urls within the 'pages' parameter
When citing newspapers I often want to link to both the front page article and the "continued on page X" second page. Using newspapers.com clips, the second page might not be accessible to people without an account/subscription unless explicitly linked. This has been discussed/requested in the past:


 * Help talk:Citation Style 1/Archive 83
 * Help talk:Citation Style 1/Archive 78
 * Help talk:Citation Style 1/Archive 78
 * Help talk:Citation Style 1/Archive 60

From the last of those links I discovered that it is possible to link to multiple urls using the  parameter, like so:

(visible at article)

I don't think this is well explained in the documentation. Could support for multiple URLs in the  parameter be explained somewhere in this documentation? Perhaps at the  parameter documentation and elsewhere.

PK-WIKI (talk) 18:19, 25 June 2023 (UTC)


 * @PK-WIKI: Thanks for the suggestion! You should be able to be bold and update the documentation.  GoingBatty (talk) 15:08, 26 June 2023 (UTC)

QID parameter in Cite_Report?
Cite report currently doesn't have a parameter for QIDs. For this reference type, it'd actually be quite useful to include it as a possible identifier and link to additional metadata (perhaps similar to the display of ). I realise that consensus has been against includeion of QIDs in CS1 templates for other reference types, but I think that a case cacn be made for reports, ccine their relevant metadata is so hetreogenous. T.Shafee(Evo &#38; Evo)talk 05:58, 27 June 2023 (UTC)

Requested move 22 June 2023

 * The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion.

The result of the move request was: moved. (closed by non-admin page mover) – Material  Works  ping me! 17:36, 29 June 2023 (UTC)

– "Techreport" isn't a word. There's no reason that we need to mash words together to avoid one additional space; the sister templates like Cite AV media, Cite mailing list, and Cite press release don't.
 * Template:Cite techreport → Template:Cite tech report

—  void  xor  16:53, 22 June 2023 (UTC)


 * @Voidxor Note that templates such as Cite AVMedia, Cite mailinglist, Cite pressrelease are redirects to the templates with a space. I wouldn't mind if cite techreport redirects to cite tech report, which is the opposite of the current arrangement. GoingBatty (talk) 20:22, 23 June 2023 (UTC)
 * Yes, I understand the use of redirects. I'm not sure what your point is beyond that. This is about where the template itself belongs, not whether redirects are available. And yes, my intent is to reverse the redirect. —  void  xor  21:22, 23 June 2023 (UTC)


 * Support per nom. Gonnym (talk) 18:48, 23 June 2023 (UTC)
 * Support. It's completely necessary. : Tarantula TM (speak with me) (my legacy) 20:01, 23 June 2023 (UTC)
 * Support per nom. &#32; Headbomb {t · c · p · b} 20:24, 23 June 2023 (UTC)
 * Support for understandability and consistency with other citation templates. "Tech report" with a space is easier to parse than scriptio continua. Glades12 (talk) 10:39, 24 June 2023 (UTC)
 * Support per nom. 〜 Festucalex  •  talk  07:01, 26 June 2023 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Quotation marks within the title= parameter of Template:cite web
Since cite web puts double quotation marks around titles, the documentation for this template should specify the replacement of double quotes within the title by single quotes per MOS:QINQ. —  Peter Brown (talk) 17:15, 28 June 2023 (UTC)
 * The documentation for the cs1|2 templates is not protected. If you believe that the documentation can/should be improved, please do so.
 * —Trappist the monk (talk) 17:28, 28 June 2023 (UTC)
 * Trappist the monk or whoever: I am really struggling to figure out which article I should edit to effect this change. Can you help? Template:Cite Web consists of other templates. is explicit that it doesn't apply to cite web. I've been thrashing around for some time now, and I'm close to giving up. Peter Brown (talk) 21:41, 28 June 2023 (UTC)
 * I have boldly attempted to implement this suggestion at . Please correct as needed. -- Visviva (talk) 23:42, 28 June 2023 (UTC)
 * That's helpful, but I had really hoped to modify, which is unchanged. Here, I corrected the quotation marks in two instances of Cite web and I expect that the editor did consult Template:Cite web before proceeding. I, at least, do study the documentation for all but the simplest templates before using them, while I don't look at Help:Citation Style 1. I had hoped to use something like:
 * title: The title of the source page on the website, usually found at the top of your web browser, will display with double quotation marks (") added. If the title itself contains quotation marks, change double quotation marks to single and vice versa, per MOS:QINQ.
 * I have not been able to figure out how to incorporate this text into, however.
 * Peter Brown (talk) 02:11, 29 June 2023 (UTC)
 * I think Template:Citation Style documentation/title is what you're looking for, but you'll note that the relevant passage is implemented as a switch, in what almost looks like a deliberate attempt to render it functionally uneditable. One could in theory edit in the desired text for the "cite web" case, but I will leave that to your judgment because I definitely don't understand the thought process behind this setup. -- Visviva (talk) 02:18, 29 June 2023 (UTC)
 * Yes, I finally got that far, too, but then decided it looked like some kind of tar baby, and I've decided not to poke it. Sorry. &mdash; JohnFromPinckney (talk / edits) 02:30, 29 June 2023 (UTC)
 * I have edited the relevant documentation pages. Feedback is welcome. – Jonesey95 (talk) 06:37, 29 June 2023 (UTC)
 * Splendid! That's all that I was after. Visviva and JohnFromPinckney might be interested in a more detailed account of what you did; I don't expect that I would understand. Peter Brown (talk) 14:12, 29 June 2023 (UTC)
 * I edited /title and /web. There are probably more places where similar text could be provided. – Jonesey95 (talk) 14:21, 29 June 2023 (UTC)
 * Looks great! My ParserFunctions skills are evidently even rustier than I thought, as that simple solution didn't even occur to me. -- Visviva (talk) 14:37, 29 June 2023 (UTC)
 * Thanks, Jonesey, for both the edits and the pointers. There is still a non-trivial likelihood that I will leave future, similar improvements to you (or others); there are rather more meta-levels than I can grok, I think. &mdash; JohnFromPinckney (talk / edits) 18:06, 29 June 2023 (UTC)

New url access level needed
Can we get a new url access level of  or   or   (maybe aliases of each other) or similar? I was able to get a 1936 journal article from de:Historisches Jahrbuch through an WP:RX request that was answered by someone who had access to it via German university database, here. You will get some basic metadata at that link, plus 'access denied' for the text in the main window. I have sufficient information now to cite this fully, and I would like to include the link as it helps WP:Verifiability, but I would like to include &lt;something> but the values and  don't fit, and although the word  sounds good as far as the English meaning goes, the description "free access is subject to limited trial and a subscription is normally required" is completely inapplicable for this case. For example, there are many resources online that are not available to the public, or through WP:TWL, or public library databases, that are available to university students/staff/alumni, and I could imagine other categories of institution (government agencies, think tanks, museums, documentation centers, archives, etc.) that have online repositories with restricted access available to some with proper credentials, but that still meet WP:SOURCEACCESS and are thus WP:Verifiable. It would be good to have value(s) of url-access that we could use to indicate this access level.

One quick & dirty solution occurs to me: rewrite the documentation so that limited is redefined as meaning, "membership or institutional credentials required" to cover the case under discussion, and rewrite subscription to bundle the existing meaning along with the current meaning of "limited", so that we'd have, "subscription required, or limited free trial and subscription". The q&d is not ideal, as we lose the distinction between "subscription" and "free trial+subscription", but it saves having to make any change to the software. Mathglot (talk) 00:10, 29 June 2023 (UTC)


 * I would rather redo the documentation versus adding different types, because there are so many scenarios. -- Green  C  01:20, 29 June 2023 (UTC)
 * Which scenarios can you envision that would need different treatment? Afaic, web pages restricted to members/staff/alumni of universities, government agencies, think tanks, museums, documentation centers, or archives could all share one new access level: credentials, as I don't see anything different among any of them. Can you see some other scenario that is essentially different in some way? Mathglot (talk) 01:45, 29 June 2023 (UTC)
 * "registration" is a good catch-all for these types of things. There are pros and cons with clumping vs. splitting. Generally unless there is a compelling reason clumping is preferable to reduce complexity. We could get into a clumping/splitting debate if you want, but it has no right answer. Like, one needs to be a "registered student" at a school library to access the link, is a form of registration. Registration can mean multiple things. I'm also not convinced the distinction of "member" vs. "registration" is going to mean much to end users, they both mean essentially the same thing in terms of access, you need some kind of credentials, that's all we need to say. If you want to get into the specific method of access then yes there are many possibilities. -- Green  C  02:47, 29 June 2023 (UTC)
 * I'm okay going with "registration" (especially if we can add an alias like "credentials") but even with no change to the software if we go with "registration", then we would have to change the doc at Template:Citation Style documentation/registration to mention the new case, so users know which one to apply. I have no objection to updating it myself, but it seems like it's a significant enough change it should get buy-in. Or, I could just make a bold edit, link back here in the summary, and see if anybody squawks. Mathglot (talk) 03:32, 29 June 2023 (UTC)
 * I would rather not have to categorize all the links in all of the references I edit by all of the different ways that access may be available to some people from some addresses but not available to other people from other people with other addresses, require subscriptions from a third class of people with a third class of accesses, be available for a one-time fee for a fourth class of people from a fourth class of addresses, and be available only if it's among their first three views for a fifth class of people from a fifth class of addresses. That way lies madness. —David Eppstein (talk) 01:48, 29 June 2023 (UTC)
 * David, I don't envision multiple classes of addresses; rather, I'm faced with a concrete issue I'm trying to resolve, namely, to write the citation for a resource I have. Here's what I have so far:
 * What value should I put in place of the x's? I think would be good, but that doesn't exist. Thanks, Mathglot (talk) 02:21, 29 June 2023 (UTC)
 * You may not envision multiple classes of addresses, but they're present in that example. When I try it from different addresses I get different behavior. "Nothing" should always be an acceptable answer for how to fill in a field that does nothing to help readers find the reference and cannot be filled in a way that is correct and meaningful to all readers. —David Eppstein (talk) 04:07, 29 June 2023 (UTC)
 * I don't understand "multiple classes of addresses", nor "trying it from different addresses", so I'll just let that part of it go. "Nothing" is indeed an acceptable answer, and the fields lang, date, url, and pages could also be left out as they are not required by WP:V. However, I like to use any params that might help another editor locate and verify the content. So, yes: you could just leave out url-access entirely, and that would be a valid option. Not one I would knowingly choose, but valid. Mathglot (talk) 04:14, 29 June 2023 (UTC)
 * For how to get different addresses without using different computers or being different people, see VPN. —David Eppstein (talk) 04:41, 29 June 2023 (UTC)
 * For how to get different addresses without using different computers or being different people, see VPN. —David Eppstein (talk) 04:41, 29 June 2023 (UTC)

Julian vs. Gregorian
Unsurprisingly, all the dates in Giuseppe Tomasi di Lampedusa use the Gregorian calendar. I don't offhand see any from an era when this would be in question. How can I get rid of Category:CS1: Julian–Gregorian uncertainty? - Jmabel &#124; Talk 21:53, 29 June 2023 (UTC)
 * This one has a date that is in the uncertainty realm (1 October 1582 and 1 January 1926).
 * The category was created because there are editors out there who questioned whether cs1|2 templates are creating bogus (read: illegitimate) dates in the COinS metadata; there was a 2017 rfc: . So far as I know, those editors have never once used that category for anything.  If that is true, I think that the category can / should go away because it is merely clutter.  I solicit opinions...
 * —Trappist the monk (talk) 22:10, 29 June 2023 (UTC)
 * Does the generation of metadata do anything differently if an article is in that category? It's not tracking errors but potential errors, that would be better handle by an inline template that editors can add if they are concerned with a potential ambiguous date. It's likely the vast majority of the articles being tracked are fasle positives. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:46, 29 June 2023 (UTC)
 * The creation of metadata has nothing to do with the en.wiki article that holds the cs1|2 template. Because this is the English wikipedia, and because Great Britain adopted the Gregorian calendar in 1752, and because most sources used here are written in English, the 'window of uncertainty' for those sources is narrower: 1582–1752, so yeah English-language sources dated 1752–1926 are false positive and are listed in .  So far as I know, no one has ever attempted to determine how many of the categorized templates are false positive.
 * For the metadata, dates in the Julian calendar (before 1582) are reduced to year-only regardless of precision in date. I should probably tweak that so dates before 1 October 1582 are reduced to year only; that was the intent; don't know why it didn't happen.
 * —Trappist the monk (talk) 00:13, 30 June 2023 (UTC)
 * Module:Citation/CS1/Date validation/sandbox tweaked to properly render 1582 Julian and Gregorian dates:
 * —Trappist the monk (talk) 16:56, 30 June 2023 (UTC)
 * Also tweaked Module:Citation/CS1/sandbox so that invalid dates are not placed in.
 * —Trappist the monk (talk) 17:01, 30 June 2023 (UTC)
 * —Trappist the monk (talk) 16:56, 30 June 2023 (UTC)
 * Also tweaked Module:Citation/CS1/sandbox so that invalid dates are not placed in.
 * —Trappist the monk (talk) 17:01, 30 June 2023 (UTC)
 * —Trappist the monk (talk) 16:56, 30 June 2023 (UTC)
 * Also tweaked Module:Citation/CS1/sandbox so that invalid dates are not placed in.
 * —Trappist the monk (talk) 17:01, 30 June 2023 (UTC)
 * —Trappist the monk (talk) 16:56, 30 June 2023 (UTC)
 * Also tweaked Module:Citation/CS1/sandbox so that invalid dates are not placed in.
 * —Trappist the monk (talk) 17:01, 30 June 2023 (UTC)
 * —Trappist the monk (talk) 17:01, 30 June 2023 (UTC)

medrxiv?
Should we support MedRxiv in the same way that we support BioRxiv? Both have similar identifiers so adding medrxiv would make use of code already used for biorxiv. The start date for the validator would be 2020-01-01 instead of biorxiv's 2019-12-11. The medrxiv validator would not accept a shortened identifier as biorxiv does.

Do it? Don't do it?

—Trappist the monk (talk) 16:39, 31 May 2023 (UTC)


 * medRxiv, EarthArXiv, PsyArXiv, etc... should all be supported. It might be easier to have a generic cite preprint handle those automatically however. &#32; Headbomb {t · c · p · b} 20:04, 31 May 2023 (UTC)
 * I have implemented medrxiv and created . Here is a journal citation using medrxiv:
 * and to cite an earlier version of the preprint mentioned in the above:
 * After the next module update, can be updated to support.
 * —Trappist the monk (talk) 22:48, 1 June 2023 (UTC)
 * I think it would be useful to have medrxiv support for cite journal template. Occasionally, people cite the preprint doi instead of the published article doi, which can lead to confusion about what identifiers to include and whether the article is peer-reviewed. I've run into confusion editing around this, so it would be a nice fix. &#12296; Forbes72 &#124; Talk &#12297; 23:30, 1 July 2023 (UTC)
 * After the next module update, can be updated to support.
 * —Trappist the monk (talk) 22:48, 1 June 2023 (UTC)
 * I think it would be useful to have medrxiv support for cite journal template. Occasionally, people cite the preprint doi instead of the published article doi, which can lead to confusion about what identifiers to include and whether the article is peer-reviewed. I've run into confusion editing around this, so it would be a nice fix. &#12296; Forbes72 &#124; Talk &#12297; 23:30, 1 July 2023 (UTC)
 * —Trappist the monk (talk) 22:48, 1 June 2023 (UTC)
 * I think it would be useful to have medrxiv support for cite journal template. Occasionally, people cite the preprint doi instead of the published article doi, which can lead to confusion about what identifiers to include and whether the article is peer-reviewed. I've run into confusion editing around this, so it would be a nice fix. &#12296; Forbes72 &#124; Talk &#12297; 23:30, 1 July 2023 (UTC)

Discussion at Wikipedia:Bots/Noticeboard § Further steps?
You are invited to join the discussion at Wikipedia:Bots/Noticeboard § Further steps?. – Novem Linguae (talk) 18:12, 4 July 2023 (UTC)

PMID numbers exceeding configured limit
I understand that CS1 templates have a constraint on PMID numbers, limiting them to 37400000. New articles published as of 3 July 2023 are exceeding that number. See for example. DontCallMeLateForDinner (talk) 18:35, 5 July 2023 (UTC)

Use of title and corrections of said title
So under the parameters I see for title (which is required for cite web) we are to use Template:Cite_web the "Title of source page on website." Ok... I usually see the title of the actual article on the source page but no matter. But per the template it seems that the citation: is ok. Sure it could be better but it is not an error? If someone then changes it to: would you think that the person who changed it should be required to add the new title instead of leaving it with a red error code of "missing title" in the ref section? I'm all for making things better but if you go to the trouble of putting in the "work=" parameter (only required by "cite journal" and "cite magazine" then it seems you should also re-add the removed title= parameter instead of leaving it for others to fix. Am I missing something? Fyunck(click) (talk) 23:46, 9 July 2023 (UTC)


 * @Fyunck(click): Each editor working on the reference should have used the proper values in the appropriate parameters. The editor who changed the reference to have no title parameter should have seen the red error code, and should have received a bot notice on their user talk page reminding them to fix the reference.  Since you didn't provide the Wikipedia article name, I hope you'll fix the reference if it is still incorrect.  Thanks! GoingBatty (talk) 00:03, 10 July 2023 (UTC)
 * This discussion is a WP:TALKFORK of this discussion. The article is incorrectly using Cite web when it should be using cite news, and was misusing the title parameter in those incorrect citation templates. So much to learn. – Jonesey95 (talk) 01:49, 10 July 2023 (UTC)
 * Per the things I am reading "cite web" was not incorrect... not if it was from a web page like newspapers.com. You use if it is from an actual newspaper that is not online. This was just a curious post here to make sure I'm getting the correct info from you. I thought I was but now I'm not so sure as this is a very grey area. . I was giving an editor advise based on listening to editors Jonesey95 and Colonies Chris. In digging around the templates and wikipedia articles I'm now thinking I may have been given incomplete info. Hence I asked here what other experts think. Articles are going from no errors in the ref section to dozens and dozens of errors in the ref section. If it was one it's easily fixed and i would have just done it.Fyunck(click) (talk) 03:57, 10 July 2023 (UTC)
 * User:Fyunck(click), I was going to mention that the wrong citation template was being used in your example, then I figured I might as well go fix it myself instead of being pedantic about it, then couldn't find the link using an insource= search across all namespaces, and ended up taking no action. Newspapers.com is a hosting service, and if anywhere other than the URL it should go in . The should be given the value of whatever headline was at the top of the news story you're citing. cite web is way overused. Folly Mox (talk) 04:11, 10 July 2023 (UTC)
 * The title of the actual news story is what I would generally use myself, but the doc in the wikipedia template does not say that at all. Very confusing. And simply removing the title= creates dozens of red errors in articles that likely may not get fixed for years. Fyunck(click) (talk) 04:18, 10 July 2023 (UTC)
 * All the best practices surrounding citations have been a continuing journey for me. There's a lot of different stakeholders, with the template editors here working with the Wikidata people for precise metadata, and the Internet Archive people and their link rot prevention efforts, while the Foundation understandably wants to keep the barrier of entry as low as possible for new people to edit, but doesn't seem interested in improving their own citation tools, and the maintainers of the volunteer written citation tools don't have time to maintain them....The good news is that I think you have to have a setting enabled to see template errors anyway, so the people who see them are likelier than average to try to address them, but I might be wrong about the enabling a setting bit. Folly Mox (talk) 04:42, 10 July 2023 (UTC)
 * Obviously it's a journey for me too. The wild ways of wikipedia can oft times be confusing. Fyunck(click) (talk) 06:23, 10 July 2023 (UTC)

Volume titles in "Cite book"
Greetings and felicitations. I have had a recent discussion regarding the italicization of volume titles in the  field of the  template. Citing that discussion, volume titles are italicized per


 * The Chicago Manual of Style 17th Ed. §14.117 though §14.122 (pp. 808–810 of the hardcover)
 * the MLA Handbook, Ninth Ed., §5.117 "Multivolume Works" (pp. 215–216 of the hardcover)

and also


 * New Oxford Style Manual (2016), §18.2.7 (p. 357 of the hardcover)

The APA 6th Ed. does not address this; I can also check the minor style guide Words into Print if desired. Per Nardog's request (as linked at the top of this section) for me to make my case for italicizing all (book) volume titles, can we agree to (at least) add one or more italicized volume titles as examples to the Cite book template, or even make it explicit in the template's documentation? —DocWatson42 (talk) 15:35, 10 July 2023 (UTC)


 * Could you maybe provide an example (here or in one of the other places you're having this discussion) of what you're looking for, formatted manually the way you envision it. I've looked at three "discussions" but still can't tell what the "before" and "after" formatting is supposed to look like. &mdash; JohnFromPinckney (talk / edits) 10:54, 12 July 2023 (UTC)