Help talk:Citation Style 1/Archive 89

This is an archive of past discussions about Help:Citation Style 1. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 85

←

Archive 87

→

Another generic author name

I've just cleaned up a few articles where the author field ended with "(View posts)" (sample diff) -- John of Reading (talk) 09:26, 13 July 2023 (UTC)

Translated Quote Parameter

Since there is a parameter for "translated title", and there is a parameter for "quote", I suggest a parameter for "translated quote". Currently I tend to write a translation of the quote in brackets after the quote, but this is probably not optimal, since the source itself is not the source of the translation (just as we usually cannot attribute a "translated title" to the source). Thiagovscoelho (talk) 12:26, 13 July 2023 (UTC)

If the quotation is important to the article, put it in the article body and cite it. Quotations require citations; citations do not require quotations. But, if you must: |trans-quote=.

—Trappist the monk (talk) 12:49, 13 July 2023 (UTC)

Oh, I didn't know we actually had this. I guess it was because it's not available from the visual editor 😅. Sometimes I used |quote= to give a fuller version of a quotation from the article itself, and sometimes it was to make clear how a source supported a claim, since it was a large webpage source and I can't give page numbers. Thiagovscoelho (talk) 13:30, 13 July 2023 (UTC)

Istro-Romanian-language sources

Hello. I frankly have no idea if this is the appropriate venue for this.

Istro-Romanian is one of the Balkan Romance languages. The others are Romanian, Aromanian and Megleno-Romanian. Adding a parameter |language=ro/rup/ruq in a citation template will produce (in Romanian/Aromanian/Megleno-Romanian), but |language=ruo does not produce (in Istro-Romanian). An example is reference 50 at Istro-Romanians. I fixed it manually with |version= but I don't see why Istro-Romanian should be excluded from Wikipedia's technical code, or whatever the root of this is. Note that there is already Template:Lang-ruo so it's not a problem of "ruo" or of the language not being integrated anywhere within Wikipedia's code.

Can this be fixed? Super Dromaeosaurus (talk) 20:17, 16 July 2023 (UTC)

Staff

Can "staff" be added to generic author names? There are 63 articles citing "Staff, Ars". 93.72.49.123 (talk) 06:39, 20 July 2023 (UTC)

in "cite magazine", how do I cite an issue name such as "Spring 2022"?

If I put this in the 'issue' field, it renders as "No. Spring 2022" which doesn't really seem correct. But it doesn't really seem like it should be in the date field like that either. –jacobolus (t) 15:49, 20 July 2023 (UTC)

Since you didn't say what magazine you are thinking of, I will suppose this is a typical magazine. Typically, "Spring 2022" would be the date of the issue and would be specified in the date parameter. The publisher may or may not specify an issue number, which for a quarterly magazine would probably be a number from 1 through 4. That may be specified with the issue parameter. Jc3s5h (talk) 16:00, 20 July 2023 (UTC)

I see from your edit history that the magazine in question seems to be Caltech Magazine. When accessing the issue online, there is an option to see the print version. In the online reproduction of the print version, on page 2, where the "masthead" is, it states "Spring 2023 Volume LXXXVI, Number 1". I would regard this as the official date, and the date at the top of the article as a garbage date. I would write a complaint letter to the magazine about providing a date other than the official date without explaining what it means. Jc3s5h (talk) 16:08, 20 July 2023 (UTC)

Great, thanks! –jacobolus (t) 16:09, 20 July 2023 (UTC)

Unix epoch

I'm wondering whether it would make sense to show an error when a date of 1 January 1970 (the Unix epoch) is supplied. Of course there would be false positives, but from a quick search most articles appear to be using this date in error. 93.72.49.123 (talk) 06:46, 20 July 2023 (UTC)

Perhaps it would be better to run the search for that date and recruit help to go check them all. –jacobolus (t) 15:50, 20 July 2023 (UTC)

Simple search results:

1 January 1970 – ~745
January 1, 1970 – ~500
1970-01-01 – ~905

—Trappist the monk (talk) 17:13, 20 July 2023 (UTC)

Is there a project for correcting issues? It would seem like a good place to organise something like this. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:22, 20 July 2023 (UTC)

Judging from a spot check, Category:CS1 maint: date eauals unix epoch may be indicated. I'll hit up mw:Citoid and see if they'd be willing to throw out this date for web sources. Folly Mox (talk) 20:35, 20 July 2023 (UTC)

Warning when adding duplicate arguments?

I fix a lot of script-generated citations, and about once every twenty or thirty, I'll end up duplicating a parameter that was actually already included but I didn't notice, causing the page to be added to Category:Articles using duplicate arguments in template calls and creating additional work as ~~other editors~~ User:Davemck and User:Ira Leviton clean up after me.

Is there a way for this to generate a warning message / for me to enable display of a warning message that is already generated? I am sometimes able to notice the category being added to the page on preview, but this typically only displays if I'm editing the full page rather than just a section, which accounts for maybe 5% of my edits. Folly Mox (talk) 13:14, 21 July 2023 (UTC)

No. MediaWiki can and does detect duplicate parameters but Module:Citation/CS1 gets only one of them (the last):

{{cite book |title=Title1 |title=Title2}}

Title2.

I use the generic text editor. When I preview this section, I see a warning message that looks something like this:

Warning: Help talk:Citation Style 1 (edit) is calling Template:Cite book with more than one value for the "title" parameter. Only the last value provided will be used. (Help)

—Trappist the monk (talk) 13:34, 21 July 2023 (UTC)

Oh, I do remember seeing warnings of that genre, but only in desktop mode. They must be restricted to that interface. Folly Mox (talk) 13:54, 21 July 2023 (UTC)

Resolved

Access-date gives an error for the 90's

Is this the expected behavior? Rjjiii (talk) 00:43, 12 July 2023 (UTC)

Where do you see this failure? Show an example.

—Trappist the monk (talk) 00:50, 12 July 2023 (UTC)

I get {{cite web}}: Check date values in: |access-date= (help) for dates until the year 2001 in these examples:

"ZDNet". Retrieved 4 July 1998. {{cite web}}: Check date values in: |access-date= (help)
"ZDNet". Retrieved 4 July 1999. {{cite web}}: Check date values in: |access-date= (help)
"ZDNet". Retrieved 4 July 2000. {{cite web}}: Check date values in: |access-date= (help)
"ZDNet". Retrieved 4 July 2001.
"ZDNet". Retrieved 4 July 2002.
"ZDNet". Retrieved 4 July 2003.
"ZDNet". Retrieved 4 July 2004.

I wasn't sure if the template just assumes an early access to be an error, Rjjiii (talk) 01:01, 12 July 2023 (UTC)

The first edit in Wikipedia's database was made on January 15, 2001. So, it is literally impossible for an access-date to be before that date. AManWithNoPlan (talk) 02:06, 12 July 2023 (UTC)

Oh, that makes sense. My thought was that material printed from the web had a date on the printout, but that's overly complicated. Thanks for the explanation, Rjjiii (talk) 02:42, 12 July 2023 (UTC)

I'm going to take the '[solved]' out of the title of this thread, because you would not believe how difficult it makes wikilinking to this discussion. The parser apparently just cannot handle it. FeRDNYC (talk) 12:07, 22 July 2023 (UTC)

Page numbers when citing journals

After adding references to Grand Sanhedrin, I noticed that Template:Cite journal creates somewhat obscure and (in my opinino) inaccessible output:

> Niles, H. (12 June 1830). "The Jews". Niles' Weekly Register. 38: 296.

The number 38 represents the "Volume" and "296" presentes the page number. In other citation templates, we state the page number after "pp.", such as when using Template:Cite book. I think applying this to journal citations would make it significantly easier for readers to find the facts in the source, and thus to help them read the surrounding information. This especially because the URLs tend to go to a place for the work as a whole, which places a heavy burden on understanding that 1) the page number is in fact given, and 2) which number is what.

Perhaps less importantly than the page numer is the volume. This because each volume tends to be its own work and thus its own ISBN and/or online entry page. This means the volume number isn't required for navigation within the work. Having said that, I wouldn't mind spelling that out as "Volume 38".

Would these changes be welcomed? What other ways might there be to address the usability issue? Or perhaps there exists documentation we expect people to find that explains this?

Krinkle (talk) 23:37, 12 July 2023 (UTC)

I think {{cite news}} or {{cite magazine}} is what you want for Niles' Weekly Register:

Niles, H. (12 June 1830). "The Jews". Niles' Weekly Register. Vol. 38. p. 296.

{{Cite journal}} is for academic journals. – Jonesey95 (talk) 23:49, 12 July 2023 (UTC)

I have to agree that the output of {{cite journal}} is obscure and for clarity needs to be changed, especially where there are a mixture of formats on a page. We should not expect readers to know a convention or have to guess what things mean. Keith D (talk) 11:47, 14 July 2023 (UTC)

It's only obscure if you've never encountered the notation before. It's an extremely common citation style. Headbomb {t · c · p · b} 13:08, 14 July 2023 (UTC)

I find the output intuitive enough, as well as pretty standard (APA, I think?). The bolded volume is clear and reduces clutter, plus basically every journal citation will link directly to the article in question. I wouldn't mind "p" or "pp" for the page numbers, but the template output should be legible to anyone who's ever read a footnote in a published academic work. Folly Mox (talk) 13:08, 14 July 2023 (UTC)

But the average reader in Wikipedia will not necessarily be too experienced in conventions in academia I believe. I think this is a great idea. Super Dromaeosaurus (talk) 20:24, 16 July 2023 (UTC)

Yeah, "this is how academic journals cite other academic journals" (or how they're cited in other works focused on academia, like research papers and science periodicals) shouldn't really matter all that much to Wikipedia, should it? It's odd that we'd cite (only) one particular class of references in "their native tongue", rather than citing them in ways that are more accessible to the target audience here. Which doesn't suddenly become academia simply because a journal was cited. FeRDNYC (talk) 11:29, 22 July 2023 (UTC)

It's how virtually all scholarship is cited in virtually all publications venues. Magazines, books, journals. This is what APA style and virtually all others mandate (some italicize the volume, most keep it plain). Headbomb {t · c · p · b} 13:30, 22 July 2023 (UTC)

I've been in favor of using the same style for journals as we do for magazines for quite some time. There are four points in favor of this.

Wikipedia is for a generalist audience, not just an academic one. Making academics read "Vol. 1 no. 2 p. 3" should not offend or confuse them, but making laypeople read "1 (2): 2" has a high likelihood for confusion.
Journals and magazines are both similar types of periodicals, yet one gets the terse in-source location format and the other does not.
The display of a volume number for a journal is also inconsistent with the format for the volume number in a book. Journals are the outlier here and should be adapted to conform to the rest.
Lastly, we already have guidance against abbreviating journal names, which is a standard practice in academic works, for accessibility to a general audience. Since Wikipedia is not distributed on paper, we don't need the space savings, however minimal that may be, of abbreviating the either the journal name or the volume/issue/page numbers. When the cost of saving space is comprehension, that savings should give way to comprehension.

For these reasons, I fully support the more verbose format from {{cite magazine}} over that of {{cite journal}}. If you need an external style guide for support, please see the Chicago Manual of Style, which has been used as an influence on CS1 in the past. Imzadi 1979 → 20:49, 22 July 2023 (UTC)

I like that journals and magazines format their volume numbers differently. It allows me to tell at a glance whether a source comes from a tradition of academic peer review or not. Also some magazines have a tendency to play fast and loose with their volume / issue numeration, whereas journals tend to have a set structure. It makes sense to differentiate them. And frankly if I were princess of citation template styles, I'd format book chapter numbers in bold right before the page numbers, similar to how journals are handled now.

I haven't seen anyone produce an argument where a reader – even never having encountered APA citation style before – could actually take the step of attempting to verify a citation (by clicking a link, copy-pasting into a search field, or whatever), and still come away from the process uncertain about how to locate the cited source. If people aren't bothering to check the source, and are just copypasting our sources into their own work, or whatever process includes seeing the citation written out but not actually looking at it, I don't feel we owe them hand-holding. Folly Mox (talk) 21:00, 22 July 2023 (UTC)

Cite tech report

With the move, the error messages and maint notices still point to cite techreport. Chandrasekhar Boyapati, William Beebee, Jr., Martin Rinard. A (Technical report).{{cite tech report}}: CS1 maint: multiple names: authors list (link) (Technical report). {{cite tech report}}: Missing or empty |title= (help) AManWithNoPlan (talk) 16:26, 25 July 2023 (UTC)

Fixed in the sandbox:

{{cite tech report/new | title=A |author1=Chandrasekhar Boyapati, William Beebee, Jr., Martin Rinard }}

Chandrasekhar Boyapati, William Beebee, Jr., Martin Rinard. A (Technical report).{{cite tech report}}: CS1 maint: multiple names: authors list (link)

{{cite tech report/new | title=}}

(Technical report). {{cite tech report}}: Missing or empty |title= (help)

—Trappist the monk (talk) 17:11, 25 July 2023 (UTC)

Trappist the monk, Can you also fix cite ssrn, vs the new cite SSRN also please. AManWithNoPlan (talk) 13:59, 26 July 2023 (UTC)

Done. Also {{Cite arXiv}}, {{Cite bioRxiv}}, {{Cite CiteSeerX}}, and {{Cite medRxiv}}.

—Trappist the monk (talk) 23:50, 26 July 2023 (UTC)

Translation link

For {{cite news}} or {{cite web}}, what parameter should be used to add a translation link to the original article (written in a foreign language) when both the URL and archive-URL have already been utilized? I don't see a translation-url or anything similar. Thank you.--TerryAlex (talk) 17:59, 25 July 2023 (UTC)

There is no |translation-url= parameter.

If you are citing the translation of something and not the original-language source, cite the translation and use |url= to link to it if it is available online. You might include |type=Translation. If you are citing the original-language source, cite that and use |url= to link to it if it is available online. In either case, you can always add an external wikilink to the 'other' (translated or original-language source) after the closing braces of the {{cite <whatever>}} template.

—Trappist the monk (talk) 18:14, 25 July 2023 (UTC)

Can you give me an example? For example, how to add a translation link to the following:

{{cite news|url=www.abc.com/newsarticle.html|archive-url=web.archive.org/20230724091214/www.abc.com/newsarticle.html|title=News Article|work=ABC|date=March 23, 2021|archive-date=July 24, 2023|access-date=July 24, 2023|language=foreign|url-status=live}}

. Thanks.--TerryAlex (talk) 20:32, 25 July 2023 (UTC)

I played around with the parameters and found that if I use the |at= parameter and format it like this

{{cite news|url=https://www.abc.com/newsarticle.html|archive-url=https://web.archive.org/20230724091214/www.abc.com/newsarticle.html|title=News Article|work=ABC|date=March 23, 2021|archive-date=July 24, 2023|access-date=July 24, 2023|language=foreign|url-status=live|at=[https://web.archive.org/20230725112323/www.translation.com/translationpage.html English Translation]}}

It yields no error and the citation shows up looking properly. Is this a good way to do it or does it violate the citation format in any ways? Thanks.--TerryAlex (talk) 00:05, 26 July 2023 (UTC)

No. Do not abuse parameters to do something for which they are not designed. Just because you don't get an error message does not mean that the use is acceptable. In your example, the wikilink assigned to |at= will corrupt the citation's metadata. What I meant in my first post was something like this:

<ref>{{cite news |url=https://www.abc.com/newsarticle.html |title=News Article |work=ABC |date=March 23, 2021 |access-date=July 24, 2023 |language=und}} [https://www.translation.com/translationpage.html English Translation]</ref>

—Trappist the monk (talk) 01:07, 26 July 2023 (UTC)

Got it. Thank you. I was initially confused and put the wikilink at the front and that syntax did not work.--TerryAlex (talk) 03:44, 26 July 2023 (UTC)

Use a single video source as a cite but multiple timestamps?

I want to reference a single video but different timestamps, ideally using SFN style where that uses page numbers, is there a way of doing this for time stamps? Darkwarriorblake (talk) 16:03, 26 July 2023 (UTC)

Not a cs1|2 question. Does {{sfn}} parameter |loc= not do what you want?

—Trappist the monk (talk) 16:24, 26 July 2023 (UTC)

Maybe, I will take a look, the talk pag for Cite AV Media just brought me here. Darkwarriorblake (talk) 16:30, 26 July 2023 (UTC)

How about^[1] this^[2] or this?^[3] – Jonesey95 (talk) 01:09, 27 July 2023 (UTC)

References

^ Smith (2000). AV media source.
^ Smith 2000, 23:05.
^ Smith 2000, 40:50.

support trans-series

{{cite book|last1=Hiriart-Urruty|first1=Jean-Baptiste|last2=Lemaréchal|first2=Claude|author-link2=Claude Lemaréchal|year=1993|chapter=XII Abstract duality for practitioners|title=Convex analysis and minimization algorithms, Volume II: Advanced theory and bundle methods|series=Grundlehren der Mathematischen Wissenschaften |trans-series=Fundamental Principles of Mathematical Sciences|volume=306|publisher=Springer-Verlag |location=Berlin |pages=136–193 (and bibliographical comments on pp. 334–335)|isbn=3-540-56852-2 |mr=1295240}}
Hiriart-Urruty, Jean-Baptiste; Lemaréchal, Claude (1993). "XII Abstract duality for practitioners". Convex analysis and minimization algorithms, Volume II: Advanced theory and bundle methods. Grundlehren der Mathematischen Wissenschaften. Vol. 306. Berlin: Springer-Verlag. pp. 136–193 (and bibliographical comments on pp. 334–335). ISBN 3-540-56852-2. MR 1295240. {{cite book}}: Unknown parameter |trans-series= ignored (help)

Headbomb {t · c · p · b} 15:05, 2 July 2023 (UTC)

Any progress on this? Headbomb {t · c · p · b} 20:50, 1 August 2023 (UTC)

Rename category Category:CS1 maint: Zbl to Category:CS1 maint: Zbl format

For consistency with the other format categories (Bibcode, MR, PMC). Headbomb {t · c · p · b} 04:04, 7 July 2023 (UTC)

Category:CS1 maint: Bibcode format does not exist. Category:CS1 maint: MR format and Category:CS1 maint: PMC format exist to catch cases where editors write |mr=MR1234567 or |pmc=PMC12345 when they should have written |mr=1234567 or |pmc=12345. Category:CS1 maint: Zbl exists to catch cases where the specified identifier is temporary so that editors can come back later and provide the permanent identifier.

—Trappist the monk (talk) 12:17, 7 July 2023 (UTC)

Maybe it could be renamed Category:CS1 temporary Zbl label or some such, to be more clear about its non-label-format purpose? —David Eppstein (talk) 16:11, 7 July 2023 (UTC)

It's at Category:CS1 maint: bibcode, which should be shifted to Category:CS1 maint: bibcode format too. Unless we want both to be Category:CS1 maint: temporary bibcode and Category:CS1 maint: temporary Zbl. Headbomb {t · c · p · b} 17:14, 7 July 2023 (UTC)

Speaking of Category:CS1 maint: bibcode, I notice that maint_bibcode isn't defined in the Module:Citation/CS1/Configuration error_conditions table — that's why the category page is showing "Pages with this condition are automatically placed in unknown error_conditions key: maint_bibcode." instead of its own name. FeRDNYC (talk) 17:28, 13 July 2023 (UTC)

Yes, it's because the template hasn't been updated in eons. Headbomb {t · c · p · b} 20:08, 1 August 2023 (UTC)

Cite committee meeting and cite committee report

This morning, I tried repeatedly without success, first to archive on the WayBack Machine, then secondly to cite the two relevant webpages of the meeting on 19 July 2023 of the House of Commons Business and Trade Select Committee (1,2). The WayBack Machine failed to archive the two relevant Committee webpage. Using the easy citation tool in Wikipedias's Visual Editor, I was unable to cite either webpage of the Committee. Hence, I now need to cite the two webpages using the Wikipedia generic default of {WebCite}. I'm not sure what citation protocol best to follow. I've tried without success to look up the APA guidelines for citing committee meetings and committee reports. In my view, it would be helpful to have wikipedia citation templates specifically for citing committee meetings and citing committee reports.

[1] https://committees.parliament.uk/work/7774/food-and-fuel-price-inflation-will-prices-come-down-this-year

[2] https://committees.parliament.uk/event/19035/formal-meeting-oral-evidence-session/

Humanity Dick (talk) 11:59, 24 July 2023 (UTC)

Sorry, {Cite web} not {WebCite}

Humanity Dick (talk) 12:06, 24 July 2023 (UTC)

There are editors at en.wiki who think that there are too many cs1|2 templates...

Wayback machine and Visual Editor failings are not in the Citation Style 1 remit. I can't speak to Wayback machine but for VE, it seems likely that no one has written a Zotero translator for the UK Parliament committees' websites.

I will not use the visual editor so were I citing these two web pages, using the source editor I might write something like this:

{{cite web |author=((Business and Trade Committee)) |date=19 July 2023 |title=Food and fuel price inflation: will prices come down this year? |website=UK Parliament |url=https://committees.parliament.uk/work/7774/food-and-fuel-price-inflation-will-prices-come-down-this-year}}

Business and Trade Committee (19 July 2023). "Food and fuel price inflation: will prices come down this year?". UK Parliament.

{{cite web |author=((Business and Trade Committee)) |date=19 July 2023 |title=Food and fuel price inflation: will prices come down this year? - Oral evidence |website=UK Parliament |url=https://committees.parliament.uk/event/19035/formal-meeting-oral-evidence-session/}}

Business and Trade Committee (19 July 2023). "Food and fuel price inflation: will prices come down this year? - Oral evidence". UK Parliament.

—Trappist the monk (talk) 12:26, 24 July 2023 (UTC)

Thanks, Trappist. Very helpful. I'm surprised you won't use Visual Editor. When I used to edit Wikipedia anonymously, much of my time was taken up trying to add reputable sources to support uncited assertions made by other editors. Because I'm partially disabled, I can't use a PC for very long before my back and neck become too painful. Hence, these days I'm forced to do all the work I once used to do sitting at a PC lying on my back pecking away with my right forefinger on the touch screen of my 2017 iPad resting on my abdomen. Using {Web Cite} takes me forever, in part because I also have various cognitive and visual disabilities that mean I'm very inept at even the most basic stuff I need to input, frequently making frustrating mistakes and errors I then have to try to fix, often succeeding only after multiple attempts. I never seem to learn the basics, making the same mistakes again and again. Hence, for me, and I suspect for many other disabled or partly disabled contributors, the Visual Editor and the citation templates are the best thing that have ever happened on Wikipedia Humanity Dick (talk) 16:33, 1 August 2023 (UTC)

Humanity Dick: Re: archiving, the page has a lot of content buried behind links such as PDFs behind tabs. Archive.today works for single pages such as https://archive.ph/0Qv4n but would need to archive every sub-link individually. Another method is sign up for https://conifer.rhizome.org/ which allows one to interactively view links and it will record real time creating one giant archive. This is the best method for that kind of page. If you use Conifer and add it to Wikipedia, be sure to also include a {{cbignore}} otherwise IABot will remove it (c.f. T321146)-- GreenC 14:21, 24 July 2023 (UTC)

Thanks, GreenC. Also very helpful and useful. I'll have to read and reread your suggestions many times before, if ever, I'm able to understand your advice. I'm always the tortoise not the hare. I'll try to follow your suggestions and see how I get on. Humanity Dick (talk) 16:38, 1 August 2023 (UTC)

Wikipedia:Wikipedia Signpost/2023-08-01/Tips and tricks

This gives an overview of various citation tools out there. I figured many of you would get something out of this. Headbomb {t · c · p · b} 05:52, 1 August 2023 (UTC)

Page numbers

URLs get embedded in various places of the template which makes link rot repair difficult to manage for bots and scripts. Typically they are skipped. For example Help_talk:Citation_Style_1#Documentation_needed_for_linking_multiple_urls_within_the_'pages'_parameter shows the free-form nature of how we do it. Another way is more systematically like we do for |author= + |author-link= separating the metadata from the link.

It would be |page1= ("page=42") + |page1-url= with perhaps support for up to 5 or 10 pages. Note |page= would be an alias for |page1=.

For page ranges use |pages1=42-44 (display: "pages=42-44") + |pages1a-url= + |pages1b-url= where "a" is for 42 and "b" for 44. More page ranges can be added for example |pages2=47-50 (display: "pages=42-44, 47-50") + |pages2a-url= + |pages2b-url=. This way one can have a mix of single page and page ranges, all with their own URLs, contained in parameters. -- GreenC 16:03, 10 July 2023 (UTC)

Did this change at some point? I remember some time back that wikilinking pages was a no-no. That the url link should always supposed to be to the most specific part of the source. I looked but haven't found any restriction about it now but I seem to remember there was in the past. I kind of like this idea. Jason Quinn (talk) 04:57, 12 July 2023 (UTC)

@Jason Quinn: The "most specific part of the source" advice is more useful in dissuading editors from just linking to the title page of the work they're citing, rather than the chapter/heading/page where the cited information can be found. Citations that span multiple specific locations — like a page 4 newspaper article that's "continued on page 81" halfway through the sentence being cited, for example — are kind of their own deeper problem beyond that.

I'm not sure how I feel about links in the |page*= parameters, but I definitely recognize that there are times when only linking to a single location risks leaving the user fumbling around to figure out how they can reach the rest of the content being cited.

Though, now that I say this, a quick search reveals that there are a lot of citations with a specific, single link in |page=, alongside a more generic |url= linking to the source as a whole. ...I was also under the impression that was incorrect: the |url= parameter should simply hold the link from |page=, shouldn't it?

...Unless my own understanding of how these params are meant to be used is incorrect / outdated? FeRDNYC (talk) 07:45, 12 July 2023 (UTC)

The "quick search" times out due to the size but I was able to get as many as 37,000 with some refinements, though there are probably a lot more. -- GreenC 14:49, 13 July 2023 (UTC)

Yeah, sorry, that was a "quick search" as in "I spent about 12 seconds defining it"... its performance on the server being inversely proportional seems about right. FeRDNYC (talk) 17:02, 13 July 2023 (UTC)

33,000 for "pages" and 30,000 for "page" is 63,000 (depending on server load). Both time out, there are some false positives, and false negatives, in keeping the regex simple. Also weird more pages than page. It will require a different method of search. -- GreenC 18:30, 13 July 2023 (UTC)

It is often desirable for the link on the title of a book to go to the Wikipedia article on the book, using |title-link=. When that is done, and there is no named |contribution= within the book to cite, the only way to provide a convenience link for specific content within the book (and the way that has been repeatedly recommended) is to put the link into the page or pages parameter. —David Eppstein (talk) 20:26, 13 July 2023 (UTC)

Often |section-url= will do what you need. Normally I use both links unless the relevant text is on the first page of the section. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 21:17, 13 July 2023 (UTC)

Hmm, that's a fair point, I'd forgotten how |title-link= and |url= interact with each other. (Or more to the point, that they don't, and result in a CS1 error when used in the same citation.)

That seems like an even stronger argument for formalizing page links via some sort of |pageN-url= scheme like GreenC is proposing. FeRDNYC (talk) 02:45, 15 July 2023 (UTC)

On that topic, the examples in Template:Cite book/doc include at least one demonstrating "title with a piped wikilink":

Three authors, title with a piped wikilink, edition
Markup	`{{cite book \|last1=Bloggs \|first1=Joe \|author-link1=Joe Bloggs \|last2=Smith \|first2=John \|last3=Smythe \|first3=Jim \|title=[[A Thousand Acres\|1000 Acres]] \|edition=2nd}}`
Renders as	Bloggs, Joe; Smith, John; Smythe, Jim. 1000 Acres (2nd ed.).

Should that be changed to use |title-link= instead of piping? I'll be bold and fix it (plus any others), if we agree |title-link= is always preferable now. I feel like the documentation should demonstrate best current practices, not outdated ones. (And we wonder how editors pick up / retain bad habits?) FeRDNYC (talk) 05:06, 15 July 2023 (UTC)

The best way to establish best practices is by showing. People tend to copy-cat what they see in the real world. As such in this case, a WP:BOTREQ to convert linked |title= to |title-link= could go a long way. And/or, a feature request to Citation bot (if it doesn't already have). -- GreenC 01:33, 17 July 2023 (UTC)

While I don't disagree with any of that, it feels entirely tangential to my question about the documentation for {{Cite book}}. 😃 But I'll take it to imply that, indeed, |title-link= is always preferred, and update the docs. FeRDNYC (talk) 10:54, 22 July 2023 (UTC)

Done I updated two examples, and also added notes to the parameter documentation to the effect that |title-link= and |url= should not be used together. FeRDNYC (talk) 11:11, 22 July 2023 (UTC)

@GreenC and FeRDNYC: A follow-on question of sorts: what about archived links for individual pages? I have been citing an old magazine article which is available online, but the hosting site could be gone tomorrow. When I link to individual pages of the article I have been linking directly to the archived url - or will we have |page1-url= and |page1-archive-url= and |page1-archive-date= fields for every individual page? Mr.choppers | ✎ 16:35, 2 August 2023 (UTC)

I would prefer directly linking the archive URL in page1-url -- GreenC 19:23, 2 August 2023 (UTC)

Proper fields for republished online news articles

I'm trying to {{Cite news}} an article that is republished from another work and the original work is attributed in the article. The original article is paywalled so I want to cite the freely available copy.

https://www.inforum.com/news/minnesota/minnesota-historical-society-nixes-renaming-historic-fort-snelling

Kelly Smith, Star Tribune staff writer. Republished same day with no changes to Inforum. Should I say via=Inforum? And for work say Star Tribune? Or work=Inforum and agency=Star Tribune? Pingnova (talk) 18:29, 2 August 2023 (UTC)

https://en.wikipedia.org/wiki/Module:Citation/CS1/Identifiers

'^[^1-5]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 60000+); accepts: 10000–59999 needs to now be 1-6 instead of 1-5 since doi:10.60082/2817-5069.2017 exists. AManWithNoPlan (talk) 00:39, 27 July 2023 (UTC)

Fixed in the sandbox, I believe:

Cite journal comparison
Wikitext	`{{cite journal\|doi=10.60082/2817-5069.2017\|journal=Journal\|title=Title}}`
Live	"Title". Journal. doi:10.60082/2817-5069.2017.
Sandbox	"Title". Journal. doi:10.60082/2817-5069.2017.

If I did it wrong, someone will correct it. – Jonesey95 (talk) 01:14, 27 July 2023 (UTC)

Template:Cite report

Shouldn't {{Cite report}} wrap the output of |title= in quotes? Its TemplateData says it should, at least. –MJL ‐Talk‐^☖ 17:32, 3 August 2023 (UTC)

History:

Help talk:Citation Style 1/Archive 6 § Should Template:Cite report be listed on this Help page?

Help talk:Citation Style 1/Archive 6 § Cite report

Template talk:Cite report/Archive 1 § Update to citation/core

Template talk:Citation/core/Archive 13 § Backwards specification and ownership behaviour over Template:Cite report

I wonder if we ought to revisit the no-markup 'decision'... Opinions?

—Trappist the monk (talk) 19:57, 3 August 2023 (UTC)

@Trappist the monk: I mean, I guess I should ask what {{cite report}} is primarily used for? I use it for things I would otherwise use {{cite journal}} for because there is nothing I could've put in the |journal= parameter. For example:

example

{{cite report |last1=Baumer |first1=Matt |last2=Kephart |first2=Curtis |title=Aggregate dynamics in a large virtual economy: Prices and real activity in Team Fortress 2 |id={{hdl|10419/125549}} |publisher=University of California, Economics Department |location=Santa Cruz |type=Working Paper |year=2015 |language=en}}

to output:
Baumer, Matt; Kephart, Curtis (2015). Aggregate dynamics in a large virtual economy: Prices and real activity in Team Fortress 2 (Working Paper). Santa Cruz: University of California, Economics Department. hdl:10419/125549.

As you can see, it's a working paper. It's not "published" in any true sense, but it does have a stable identifier and publisher. I could use {{cite web}}, but that would require me to remove the |id= parameter since it's more-or-less redundant.
If I'm using it wrong here, then I could see why original maintainers would've prefer to remove the markup. However, it doesn't make sense for the above use-case. –MJL ‐Talk‐^☖ 16:20, 4 August 2023 (UTC)

The upcoming {{cite document}} would suite that reference:

{{cite document/new |last1=Baumer |first1=Matt |last2=Kephart |first2=Curtis |title=Aggregate dynamics in a large virtual economy: Prices and real activity in Team Fortress 2 |id={{hdl|10419/125549}} |publisher=University of California, Economics Department |location=Santa Cruz |type=Working Paper |year=2015 |language=en}}

Baumer, Matt; Kephart, Curtis (2015). "Aggregate dynamics in a large virtual economy: Prices and real activity in Team Fortress 2" (Working Paper). Santa Cruz: University of California, Economics Department. hdl:10419/125549.

See the discussion about {{cite document}}.

—Trappist the monk (talk) 16:38, 4 August 2023 (UTC)

Don't use |id={{hdl|foobar}}. Use |hdl=foobar directly. Headbomb {t · c · p · b} 22:32, 5 August 2023 (UTC)

Numerology

One of the citations on this page shows an error message when the SSRN value that the citation bot added is correct and greater than 4500000. Achmad Rachmani (talk) 07:37, 7 August 2023 (UTC)

Per Help:CS1_errors#bad_ssrn: "If the value is correct and larger than the currently configured limit of 4500000, please report this at Help talk:Citation Style 1, so that the limit can be updated."

@Trappist the monk can you up the limit? Nobody (talk) 07:53, 7 August 2023 (UTC)

Should be line 2115 on Module:Citation/CS1/Configuration. Nobody (talk) 07:58, 7 August 2023 (UTC)

"Illustrator" parameter

Wikipedia articles on illustrated books -- especially children's books -- could really use an "illustrator" parameter, to go along with (for example) "editor." Jhlechner (talk) 15:52, 8 August 2023 (UTC)

You can use the |others= parameter for this, |others=Illustrated by John Smith for example. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 16:00, 8 August 2023 (UTC)

Another generic title

Hello, can you add "Facebook" as a generic title. Currently, 859 instances. Keith D (talk) 22:00, 9 August 2023 (UTC)

But work=Facebook with title=Something specific would probably be non-erroneous. —David Eppstein (talk) 00:06, 10 August 2023 (UTC)

I was looking at |title=Facebook. Keith D (talk) 12:02, 10 August 2023 (UTC)

Wizard needed: can a cite be generated from an ISBN?

I know that ReFill works such that <ref>https://example.com/whatever</ref> gets converted into a properly formatted cite template with the correct metadata. Does there exist anything that does this for ISBNs? I just noticed that I am probably wasting a lot of time typing in book citation information when there's already a gigantic unified metadata tracking system that is based on ISBNs. jp×g 23:40, 11 August 2023 (UTC)

@JPxG: Yes

Make a citation with {{cite book}} and fill out the ISBN parameter. You'll have something like this:
. ISBN 978-0736426701. {{cite book}}: Missing or empty |title= (help)
Run Citation Bot on the page: https://citations.toolforge.org/
It will give you something like this when finished:
Disney, R. H. (5 January 2010). Walt Disney's Alice in Wonderland (Disney Classic). National Geographic Books. ISBN 978-0736426701.

Rjjiii (talk) 06:09, 12 August 2023 (UTC)

@JPxG: WP:Wikipedia Signpost/2023-08-01/Tips and tricks#Citation Expander cough cough... Headbomb {t · c · p · b} 06:12, 12 August 2023 (UTC)

Cite Gaia EDR3

So, on TRAPPIST-1 we use the Template:Cite Gaia EDR3 (a wrapper for a {{cite journal}}) and we also need to add {{sfn whitelist}} because otherwise it throws sfn errors. Is there a way to fix the {{Cite Gaia EDR3}} template so that we don't need to hide bogus harv errors? Jo-Jo Eumerus (talk) 05:38, 12 August 2023 (UTC)

See Module:Footnotes/whitelist Headbomb {t · c · p · b} 06:10, 12 August 2023 (UTC)

You mean, just list the template on the whitelist? JoJo Eumerus mobile (main talk) 07:43, 12 August 2023 (UTC)

I've requested someone add the template to the whitelist here Module talk:Footnotes#Cite Gaia EDR3. Once someone adds it you won't need {{sfn whitelist}} anymore. I can't guarantee how long that will take though. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 11:16, 12 August 2023 (UTC)

JoJo Eumerus mobile The "Cite Gaia EDR3" template has been whitelisted for "Brown 2021". You can remove the "sfn whitelist" template from TRAPPIST-1. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:42, 12 August 2023 (UTC)

Can archive links, especially for still-living pages, be made more compact?

The current "Archived from the original on YYYY-MM-DD" phrasing at the end of many citations takes a pretty significant amount of space, especially since we now have bots indiscriminately adding these pre-emptive archive links to every citation. Is there any way we could cut that down to just "Archived YYYY-MM-DD" or the like? (If it were up to me we would entirely leave off archive links for still-living pages, as these typically have no direct benefit to readers.) –jacobolus (t) 17:57, 31 July 2023 (UTC)

I'd support removing from the original where url-status=live (where else would it be archived from?). I'd also support suppressing display of ~~/ bot removal of~~ |access-date= where an |archive-date= is present and url-status=dead, since if the link is dead and there's an archive, there's no point. Ditto for |access-date= for all {{cite book}} and {{cite journal}}, since if they've gone to print there won't be an update to the information. Folly Mox (talk) 20:03, 31 July 2023 (UTC)

Sometimes the archive added is no good, the access-date can come in useful to finding a valid link. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:04, 31 July 2023 (UTC)

Exactly, humans should remove the access-date - if the archive is good - but no bot should do such things. AManWithNoPlan (talk) 21:58, 31 July 2023 (UTC)

Fair points, both, although I've never encountered a no good archive that wasn't either a. totally unsalvageable or b. rescuable with the earliest archive available. But now that I'm considering it, mass removal of |access-date= parameters by bot seems disruptive and annoying. Probably easier just to suppress display in cases where it's of no use to readers, only editors tryna repair an archive. Folly Mox (talk) 08:46, 1 August 2023 (UTC)

Earliest archive available is usually a decent choice, though sometimes personal web pages (as compared to e.g. newspapers) had meaningful later updates, with authors starting them as drafts or treating them as living documents. –jacobolus (t) 15:59, 1 August 2023 (UTC)

I like the idea of suppressing display of access-date if an archive URL exists. -- GreenC 16:02, 1 August 2023 (UTC)

I don't like the idea of suppressing the access date because not all webpages are static documents. Even in the case of an archive, I'd like to know when that page was accessed. Imzadi 1979 → 18:43, 1 August 2023 (UTC)

User:Imzadi1979, if you're that interested in finding out the relative values of whenever the editor who used the page to add sourced content versus whenever the page was archived, you could always drop into editing mode and have a look. Meanwhile for every use case outside that it would save close to a full line of vertical cruft on mobile, for everyone else who only cares about when the page was archived (if that). Folly Mox (talk) 19:11, 1 August 2023 (UTC)

If it takes up too much space for you, maybe the solution is to wrap that information in a class so you can suppress it through your personal CSS. I shouldn't have to drop into editing mode while reading an article. Imzadi 1979 → 19:49, 1 August 2023 (UTC)

Already available: see Help:Citation Style 1/accessdate. I presume that it works. The html rendering of |access-date= looks like this:

<span class="reference-accessdate">. Retrieved <span class="nowrap">2023-08-01</span></span>

—Trappist the monk (talk) 19:56, 1 August 2023 (UTC)

@Imzadi1979: not all webpages are static documents. Even in the case of an archive, I'd like to know when that page was accessed. But, archives are static documents — even if the page changes over time, the archive won't. It's a snapshot of a particular moment, and the date of that snapshot "trumps" the access date... no matter when the archive was accessed, the version they got is the one archived on the date indicated. An |archive-date= (and |url-status=dead) makes the access date completely irrelevant. FeRDNYC (talk) 02:01, 16 August 2023 (UTC)

@FeRDNYC: you're assuming the archive was consulted for the citation. It may be added after the fact and wasn't what was actually consulted by the editor adding the citation. The archive may be added long after the citation, so the original access date is not redundant, especially when it doesn't match the access date. If I have a webpage consulted on January 1, but not archived until January 8, the original webpage may have changed in the interim between citation and archiving. When they don't match, especially when that archive link was added much later, the original access date is not irrelevant. Imzadi 1979 → 03:21, 16 August 2023 (UTC)

I like the 'suppress one of the dates' idea too. Thumbs up. –jacobolus (t) 16:00, 1 August 2023 (UTC)

I'm against. If you checked the page on January 23, and the bot archived on February 28, there could be a lot of differences between the two. The only time I'd be in favour would be if both dates are the same, and the message could then be shortened to "retrived and archived on DATE" Headbomb {t · c · p · b} 20:56, 1 August 2023 (UTC)

This, exactly this. Even with the same date, there could be updates, but that would be rare. Imzadi 1979 → 21:26, 1 August 2023 (UTC)

It's a possibility, but 99.9& of readers would benefit if the access-date was suppressed by default when an archive date is present. It's only those of us who are trying to check and double check references who benefit from this, and we will likely be looking at the code in any case. Personally, I think that we (not bots) ought to remove |access-date= if the archived page has been checked and the content matches. Mr.choppers | ✎ 16:17, 2 August 2023 (UTC)

I agree, most users doing verification who need the access-date know it exists in the code and where to find it, for the vast majority it clutters and complicates the already heavily dense reference section. Even if we suppress, there is a possibility of making it appear by default, using the same method above that makes it disappear by default. -- GreenC 16:02, 5 August 2023 (UTC)

Yeah, I understand the use case for hanging onto the access-date for dead links with working archives. In fact I had to use that bit of information just yesterday to verify a claim where the archive predated the access-date for some reason. But that is a vanishingly scant proportion of how citations are used. I might need that information once in a thousand edits, and if I'm doing the verification work I'm already in the source. Meanwhile everyone who reads the article, including the vast majority who do not have accounts and therefore the option to suppress the display with custom css, are stuck skimming through a bunch of extra details they'll never need or care about. Folly Mox (talk) 07:54, 7 August 2023 (UTC)

...Won't archives often predate the access? The only time I'd expect it not to is when it's added after-the-fact. (And actually, even then it's probably better to find a snapshot that's as close to the |access-date= as possible, ideally.) But if someone submits an |archive-url= with a citation (I try to do so, if I think it's likely to vanish at some point), I'd certainly expect it not to be an archive created after that point! FeRDNYC (talk) 02:07, 16 August 2023 (UTC)

That's a good point. A fair proportion of editors are not as thoughtful, and do not provide archives at the time of edit. I think I would characterise the frequency of archives being added only after the original link is dead as "pretty frequent". In the case I alluded to above, the archive predated the access by sufficient time that the claim citing it was not true as of the archive date, so I fetched a more recent one. Folly Mox (talk) 02:40, 16 August 2023 (UTC)

not as thoughtful, and do not provide archives at the time of edit – personally I submit pages to be saved by the internet archive but then intentionally ("thoughtfully") do not include the archive link on the page (or sometimes even remove archive links that were added by the IA bot), because to me the proliferation of archive links for still living pages seems like a space-wasting spammy distraction. –jacobolus (t) 05:01, 16 August 2023 (UTC)

Thank you; that is indeed also thoughtful. I apologise for implying thoughtlessness from those who don't add archives to live links, which I know is a deeply contentious matter. I was trying to be kind, and ended up offensive. Really, today, I am the thoughtless one. The heat melted them all. Folly Mox (talk) 05:52, 16 August 2023 (UTC)

No worries, I wasn't offended, and didn't mean to sound testy. I just hope we can collectively consider the costs as well as benefits of displaying additional piece of metadata. :-) –jacobolus (t) 16:14, 16 August 2023 (UTC)

ISSNs in citations

https://en.wikipedia.org/wiki/Wikipedia:ISSN includes "An ISSN is particularly helpful in the following circumstances" which include "In a citation to an article that is not available online except behind a WP:PAYWALL" and "In a citation to an article that is not available online in full text". How are they helpful in those circumstances? When I've seen paywalled articles cited I've sought/created archived versions. Should I also add ISSNs? To date I've only noticed an ISSN of one Wall Street Journal article. Mcljlm (talk) 18:42, 12 August 2023 (UTC)

Well that was a load of nonsense. I've removed the dubious advice. ISSNs in citations are nearly worthless. There's some corner cases, like you're not sure of the full journal/periodical title, or that the title is ambiguous and could refer to several publications. But if you have URLs or identifiers like DOIs, then the ISSN adds nearly nothing. Most citation styles omits them entirely. Headbomb {t · c · p · b} 21:29, 12 August 2023 (UTC)

Glad to see we're on the same page, @Headbomb. Copying my original reply from the TH if it's useful for anyone:

Personally, my experience is that, because ISSNs are for publications rather than for individual articles, they're much less helpful than e.g. DOIs, and they're often overused because VisualEditor's citation tool seems to like them. Basically, all you really want to have is at least one identifier in each citation. That could be a URL, or a DOI, or an OCLC, or an ISSN, but just plain text is non-preferable. And all of this only matters if you're trying to get an article's references to a featured-quality level. Hope that helps! Cheers, {{u|Sdkb}} ^talk 18:12, 12 August 2023 (UTC)

If someone wants a project, investigating why the VisualEditor always tries to add ISSNs and getting it to...not do that...would have a huge impact on this issue. {{u|Sdkb}} ^talk 21:37, 12 August 2023 (UTC)

There have been attempts to get ve/citiod to omit WorldCat urls that duplicate OCLCs; that effort failed. I suspect that any attempt to get ve/citoid to omit ISSNs would be equally unproductive.

—Trappist the monk (talk) 21:41, 12 August 2023 (UTC)

Indeed the VE team has been spectacularly unconcerned with the concerns of the community on basic aspects of style and what's appropriate to include by default (ISSNs and Publishers being the particularly egregious items). 21:47, 12 August 2023 (UTC) Headbomb {t · c · p · b} 21:47, 12 August 2023 (UTC)

|language=en for citations on the English language Wikipedia is my favourite totally unnecessary automated reference add, which I'll usually delete when cleaning up other people's work to add page numbers etc.

From what I understand, the Foundation has a single contractor responsible for all of Citoid, who is currently tasked with integrating it into Wikidata for some reason instead of fixing any of its issues. Folly Mox (talk) 21:52, 12 August 2023 (UTC)

I know it seems dumb, but it's very helpful when translating articles between different language wikis. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 00:29, 13 August 2023 (UTC)

Yeah, I concur on that. {{u|Sdkb}} ^talk 00:39, 13 August 2023 (UTC)

Oh I don't disagree that the |language= parameter is very valuable, but it doesn't display anything when it matches the language of the project it's being used in. Aren't people checking the references whenever they translate articles? I'd think that they'd be able to fill out the language parameter manually during that process. It's hardly a blip of effort when the overall time investment of translation is taken into account. Folly Mox (talk) 02:22, 13 August 2023 (UTC)

Autotranslation, and if editors are checking crosswiki because of errors in a translation it's helpful. I'll say little on how much some editors check when translating, but I do have to go crosswiki to find missing bits of referencing quite often. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 12:11, 13 August 2023 (UTC)

User:Skdb, getting any of the automated reference tools to look at any part of their output at all aside from verifying it's returning a URL would be a huge step in the right direction. Folly Mox (talk) 21:55, 12 August 2023 (UTC)

It looks like VE is under the supervision of the editing team, so courtesy pinging @ESanders (WMF) and @PPelberg (WMF) — we remain open to working with you to improve citations in VisualEditor. Better automatic formatting of citations would improve both readers' and moderators' ability to verify references, in the latter case making it easier for them to approve/provide feedback on drafts. {{u|Sdkb}} ^talk 22:05, 12 August 2023 (UTC)

See also Wikipedia:WikiProject Citation cleanup/Repairing algorithmically generated citations for common issues, and mw:Talk:Citoid for recent suggestions and improvement requests. Folly Mox (talk) 02:30, 13 August 2023 (UTC)

Thanks everybody, especially Sdkb for suggesting I asked here. Until reading the responses above I assumed I was missing something significant about their use. Maybe I should delete the ISSN from the WSJ citation. Mcljlm (talk) 23:38, 12 August 2023 (UTC)

I remove ISSNs from globally known newspapers on sight, and haven't had any pushback. Same for |language=en, except for multi-lingual pages, e.g. |language=de,en,fr. -- Michael Bednarek (talk) 01:53, 13 August 2023 (UTC)

At first reading Michael Bednarek it appears to be obvious which newspapers are "globally known" but then I wondered if there's an objective list of those newspapers to which the term applies? Some published in specific countries are known to most people living in those countries but not necessarily abroad. I haven't heard about most of those listed in List of national newspapers - which omits newspapers I would have expected to be included. Mcljlm (talk) 03:48, 14 August 2023 (UTC)

Cite book template change?

Coincidentally, I had just yesterday gone through every entry on Bibliography of works on Davy Crockett to tidy up references, etc. Looked great yesterday. Imagine my surprise when I opened that list today and found red messages on numerous cite book entries such as "{{cite book}}: Cite has empty unknown parameter: |1= (help)" Etc. etc. It looked perfect yesterday. Whatever it is, also affected the drop-down cite book template - in that if you had an ISBN number, you would put it in the template and click on that little magnifying glass to the right of it, it would fill out the rest of the template. Now it just sits there and does nothing if you try that. So, I've been randomly going through other lists and it seems that only the cite book template triggers that. What happened between yesterday and today that affected the cite book template? — Maile (talk) 01:49, 13 August 2023 (UTC)

Oh, now I see Category:CS1 errors: empty unknown parameters, is a pretty good indication it isn't just me. — Maile (talk) 01:54, 13 August 2023 (UTC)

Me too, i'm seeing the same error as you! Roberth Martinez (talk) 03:37, 13 August 2023 (UTC)

Mark Robert Parris About 4 or 5 hours ago. Citation Bot added this wording to an existing source:

" {{cite book}}: Empty citation (help): |website= ignored (help)" "22:01, August 13, 2023‎ Citation bot talk contribs block‎ 3,818 bytes −24‎ Alter: template type, url. URLs might have been anonymized. Add: date, newspaper, authors 1-1. Removed parameters. Some additions/deletions were parameter name changes. | Use this bot. Report bugs. | Suggested by Eastmain | #UCB_webform 25/30 " — Maile (talk) 03:14, 14 August 2023 (UTC)

The bot didn't add that source, it was already there. And the issue was that someone put nonsense in a website field. Headbomb {t · c · p · b} 04:36, 14 August 2023 (UTC)

Good information to have. Thanks. — Maile (talk) 12:18, 14 August 2023 (UTC)

Problem with {{SLS Q}}

I have encountered a problem in Elias Lönnrot: two references using {{SLS Q}} yield a maintenance error message: "location missing publisher" but the Wikidata items (Swedish Writings volume 1 (Q113396160) and Swedish Writings volume 2 (Q113396181)) have both publisher (P123) and place of publication (P291).-- Carnby (talk) 11:56, 13 August 2023 (UTC)

The template {{SLS Q}} has |publisher=unset, which is going to block an publisher setup on wikidata. I suggest asking the creator of the template about it. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 12:17, 13 August 2023 (UTC)

Weird.-- Carnby (talk) 15:04, 13 August 2023 (UTC)

Category:CS1 errors: archive-url

Could we add Template: namespace? For example in European Union Referendum Act 2015 there is Template:UKEU2016Result which has the error. If it was in the tracking category my bot would find it automatically. As-is I have to manually track them down. In the first 200 articles of the category, randomly sorted, there were 3 articles with templates that required a fix (1.5%) -- GreenC 00:26, 14 August 2023 (UTC)

At the bottom of that template page is a link to Category:CS1 errors: archive-url. Categorization is not instantaneous. I did this simple search, did not find the template. I then refreshed the template and repeated the search and now the template is in the category.

—Trappist the monk (talk) 00:59, 14 August 2023 (UTC)

It is likely to take a few days to a month or more for all affected pages to be null-edited by the job queue. – Jonesey95 (talk) 02:20, 14 August 2023 (UTC)

Pages are now transcluding themselves

Something in the latest update is causing a page with a reference to transclude itself. This has the downside that templates that are otherwise unused, are now hidden from reports. Is this something that can be fixed? A setting that can be set to ignore on template pages? I remember a previous change a while back that also had this issue. Gonnym (talk) 09:23, 14 August 2023 (UTC)

I think that I have fixed this. I proved that to myself by adding a call to error() in the namespace test that calls title_object:getContent() (this is what counts as self transclusion). Previewing an article with the modified code showed an error for every cs1|2 template in the article (the expected result). I then previewed {{cite Grove}}; no errors so the module is not calling title_object:getContent() in template namespace. Still, as I write this, in Special:WhatLinksHere for {{cite Grove}}, the template appears to still be transcluding itself. Is there a WhatLinksHere lag as there is for categories?

—Trappist the monk (talk) 13:33, 14 August 2023 (UTC)

Yes, sometimes they need to be null edited to refresh transclusions, but in this case I think the /doc page is actually trasncluding the template in an example. I checked Template:Arrow (rail service)/detailed which was transcluding itself in the morning, and it doesn't anymore. Gonnym (talk) 13:41, 14 August 2023 (UTC)

Is year really discouraged

year: Year of source being referenced. The usage of this parameter is discouraged; use the more flexible |date= parameter instead unless both of the following conditions are met: Currently the Citation Bot adds years for journals, etc. Should we be using date? AManWithNoPlan (talk) 18:30, 1 August 2023 (UTC)

If it is true that journals are usually issued on monthly, quarterly, seasonal schedules, then |date= is most appropriate.

|year=2023 – ok because parameter name matches parameter value

|year=Winter 2023 – not ok because semantic conflict

|date=2023 – ok because parameter name matches parameter value

|date=Winter 2023 – ok because parameter name matches parameter value

If Citation Bot does not do CITEREF disambiguation then, for ease of coding, it should probably just use |date=.

—Trappist the monk (talk) 18:49, 1 August 2023 (UTC)

When it's a year, use |year=. When it's a full date/date with month, use |date=. Headbomb {t · c · p · b} 20:49, 1 August 2023 (UTC)

Why not just use |date= for either case? I see no benefit in having two types of fields for this information. Mr.choppers | ✎ 16:09, 2 August 2023 (UTC)

The specific situation where just date wouldn't get the job done in Citation Style 1 is when two sources by the same author, published in the same year, are cited. Lets say one is an article in IEEE Spectrum in the June 2022 issue, and one is an article in Games: Research and Practice, also in June 2022. Since all the identifiers normally used to link from the little number in the text to the endnote are the same, the citation system can't tell them apart. This is solved by putting "year = 2022a" in one citation, and "year = 2022b" in the other. Due to the design of the citation system, even when the dates are not identical, the system can't tell them apart if the year is the same. So articles published by the same author in June 2022 and September 2022 would need the year parameter with "2022a" and "2022b". Jc3s5h (talk) 16:26, 2 August 2023 (UTC)

Thanks Jc3s5h - I just entered 1979a and 1979b in date parameters a few hours ago; I had better go fix that I suppose. Mr.choppers | ✎ 16:37, 2 August 2023 (UTC)

Fixing that is not required and, if I had my druthers, they would not be fixed. Hiding the CITEREF disambiguators may make it more difficult for readers to determine which of your 1979 sources is being cited. For example, someone reading a printed copy of an en.wiki article will not be able to mouse-over the short form citation to see which one gets highlighted. The only case where |year= is required is the one I described above. Don't hide the CITEREF disambiguators.

Because the redundant |year= parameter is not necessary, when |date=1979 and |year=1979a appear in the same cs1|2 template, Module:Citation/CS1 emits a maintenance message and adds the article to Category:CS1 maint: date and year.

—Trappist the monk (talk) 17:11, 2 August 2023 (UTC)

When I've had to cite multiple works by the same author in the same year, I keep |date= as just the year, but set |ref= to use 2007a, 2007b etc. Is this recommended practice? Folly Mox (talk) 19:50, 2 August 2023 (UTC)

That is confusing the reader. Common practice is to append the distinguishing letter visibly to the year: Doe, Joe (1973a). A Title.; Doe, Joe (1973b). A Different Title.. -- Michael Bednarek (talk) 01:26, 3 August 2023 (UTC)

Yes, I concur; distinguishing letters directly appended to the year, and use |date=, because it saves you the two seconds pondering which one you ought to be using. Mathglot (talk) 06:48, 3 August 2023 (UTC)

Ok I've changed the full citation templates to use the a/b formatting consistent with the {{sfnp}}s on the articles I can remember. I think I assumed the citation templates would throw an error if the |date= fields weren't exactly dates; glad to know that isn't the case. Folly Mox (talk) 13:31, 3 August 2023 (UTC)

@Mr.choppers: Those diffs look good. I think it's only required for this kind of special case: Help:Shortened footnotes#Date Rjjiii (talk) 05:17, 3 August 2023 (UTC)

Those diffs look good. I disagree. At §Notes there are three Francillon 1979 short-form references. In §Bibliography there is no Francillon 1979 and one each of Francillon 1979a and Francillon 1979b. Which of the three Francillon 1979 short-form references goes to Francillon 1979a and which goes to Francillon 1979b?

—Trappist the monk (talk) 11:58, 3 August 2023 (UTC)

Thank you Rjjiii and Trappist the monk: Those short-form references were added long ago by someone else who didn't clarify which work they were referring to, which is why I was changing the bibliography in the first place. I intend to investigate the page history to determine what's what; if that doesn't work I'll have an excuse to buy one of those books. Mr.choppers | ✎ 13:34, 3 August 2023 (UTC)

To address this specific point, almost certainly all three point to 1979a (as the cites have page numbers in the 400s and 1979b isn't that long).Nigel Ish (talk) 15:26, 17 August 2023 (UTC)

Are "Please check ISBN" and "Articles with invalid ISBNs" obsolete?

Am I right in thinking that {{Please check ISBN}} and Category:Articles with invalid ISBNs are obsolete? (The former places articles in the latter.)

There are currently only 10 pages in that category, and all but one of them are also in Category:CS1 maint: ignored ISBN errors (which itself contains 382 pages). In the only one that isn't, Rodney Hallworth, the template marks the text "ISBN B001ALS2EY" in a hand-written book reference. The last non-minor edit on the category's talk page is from 2015; the last and only edits on the template's talk page are from 2012.

So it looks as if these have fallen out of use and been replaced by Category:CS1 maint: ignored ISBN errors and Category:Pages with ISBN errors (perhaps due to the switch from magic links for ISBNs to {{ISBN}} and isbn= parameters), and the template is only being used sporadically. Of the 10 current cases, 9 were already being tracked elsewhere, and the remaining exceptional one could have been handled by adding the ISBN template (which should be done anyway) and then the invalid ISBN would have shown up in Category:Pages with ISBN errors.

So it seems that this template and category should be decommissioned? Or am I overlooking a reason for keeping them? Joriki (talk) 20:32, 17 August 2023 (UTC)

No current opinion about the category or template; they do not fall within the cs1|2 bailiwick. That ISBN B001ALS2EY is really an ASIN so should read: 'ASIN B001ALS2EY'.

—Trappist the monk (talk) 20:52, 17 August 2023 (UTC)

I think {{Please check ISBN}} is still a valid template, and is not replaced by Category:CS1 maint: ignored ISBN errors or Category:Pages with ISBN errors, because it has a different meaning. If {{Please check ISBN}} places things in Category:Articles with invalid ISBNs as you say (looks like it does), then that category seems poorly named. Maybe that category used to list some other things that have now automatically moved to the other two categories? The template seems to mean an editor thinks there is a problem with how the ISBN is presented but cannot verify the correct form and wants someone else to check. I'd expect:

Most things in Category:CS1 maint: ignored ISBN errors to be correct as marked up, added automatically to the page, but possibly contain some ISBNs or markup that need changes.
Basically all things on Category:Pages with ISBN errors need some fix, and be automatically added to the category.
{{Please check ISBN}} to have its own maint category, but only for manually flagged ISBNs that need more attention. I can imagine cases where it could be used on valid, invalid, or invalid-as-printed ISBNs., i.e. totally independent of the other categories.

Sometimes it is easy to verify an invalid-as-printed ISBN and mark it up as such, or make a correction to a invalid ISBN, sometimes it's not. {{Please check ISBN}} is a indicator to the editor who added the ref, or is watching the article to maybe double-check and mark up the ISBN correctly. Good question though, apologies if it's not 100% cs1|2. Salpynx (talk) 22:07, 17 August 2023 (UTC)

I've just been through and check every ISBN where it was used and cleared them. The majority were invalid ISBNs but are the ISBNs supplied by the publisher, and had already had (()) and so appeared in Category:CS1 maint: ignored ISBN errors. This doesn't appear to be a valid use of the template. The rest were some other kind of code inappropriately added to the ISBN field, ASIN / UPC / etc. If this template has a valid function it would be useful if it displayed somekind of message in the article, as it is unless you happen to know that Category:Articles with invalid ISBNs exists they are not going to be checked.

There was one use of the template where the ISBN was a valid ISBN, so maybe the editor who added the template just wanted someone else to check. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:46, 17 August 2023 (UTC)

[1] Smith (2000). AV media source.

[FOOTNOTESmith200023:05-2] Smith 2000, 23:05.

[FOOTNOTESmith200040:50-3] Smith 2000, 40:50.

[1]

[2]

[3]