Help talk:Citation Style 1

Add topic
Active discussions
Citation templates
... in conception
... and in reality

RfC: Should Citation bot use cite web, or cite magazine, or cite news?Edit

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
There is a strong consensus for Citation bot to use {{cite news}} and {{cite magazine}} in cases where online content doesn't appear in a print edition of a publication. ScottishFinnishRadish (talk) 23:32, 31 July 2022 (UTC)


If an article is published on a website associated with a magazine, or newspaper, but does not appear on the print edition of the publication, should you use {{Cite web}}, or {{Cite magazine}}, or {{Cite news}}? Sideswipe9th (talk) 23:55, 30 June 2022 (UTC)

Should you use {{Cite web}}, or {{Cite magazine}}, or {{Cite news}}?
Reasoning to use {{cite web}} Reasoning to use {{cite magazine}} or {{cite news}}
  • The websites are not the same as the physical publication.
  • Content published exclusively on the website of a publication is not the same as content published in the publication.
  • Many publications separate print editions, digital editions, and website content.
  • Specialised templates should only be used for print or digital editions of a publication. Not content on their websites.
  • Content published exclusively on the website of a publication is the same as content published in a publication.
  • The only difference between print editions, digital editions, and website content is the delivery mechanism.
  • Specialised templates should be used for any content published by the publication, via any delivery mechanism.
  • Using a specialised template ensures that the correct COinS metadata is embedded for reference management software.
  • For readers consuming the content via a browser, there is no difference between the generic or specialised templates.
The full past discussion on this can be found at here.
Example URLs for website only articles

Which citation template should be used for: https://ew.com/movies/doctor-strange-in-the-multiverse-of-madness-cast/

{{Cite web |last=Romano |first=Nick |date=December 10, 2020 |title=Doctor Strange sequel confirms cast, will tie into ''Spider-Man 3'' |url=https://ew.com/movies/doctor-strange-in-the-multiverse-of-madness-cast/ |url-status=live |archive-url=https://web.archive.org/web/20201211011901/https://ew.com/movies/doctor-strange-in-the-multiverse-of-madness-cast/ |archive-date=December 11, 2020 |access-date=December 10, 2020 |website=[[Entertainment Weekly]]}}

Romano, Nick (December 10, 2020). "Doctor Strange sequel confirms cast, will tie into Spider-Man 3". Entertainment Weekly. Archived from the original on December 11, 2020. Retrieved December 10, 2020.

{{Cite magazine |last=Romano |first=Nick |date=December 10, 2020 |title=Doctor Strange sequel confirms cast, will tie into ''Spider-Man 3'' |url=https://ew.com/movies/doctor-strange-in-the-multiverse-of-madness-cast/ |url-status=live |archive-url=https://web.archive.org/web/20201211011901/https://ew.com/movies/doctor-strange-in-the-multiverse-of-madness-cast/ |archive-date=December 11, 2020 |access-date=December 10, 2020 |magazine=[[Entertainment Weekly]]}}

Romano, Nick (December 10, 2020). "Doctor Strange sequel confirms cast, will tie into Spider-Man 3". Entertainment Weekly. Archived from the original on December 11, 2020. Retrieved December 10, 2020.

Which citation template should be used for: https://www.washingtonpost.com/technology/2022/04/25/twitter-elon-musk-deal/

{{Cite web |last=MacMillan |first=Douglas |last2=Siddiqui |first2=Faiz |last3=Lerman |first3=Rachel |last4=Telford |first4=Taylor |date=April 25, 2022 |title=Elon Musk acquires Twitter for roughly $44 billion |url=https://www.washingtonpost.com/technology/2022/04/25/twitter-elon-musk-deal/ |url-access=limited |url-status=live |archive-url=https://web.archive.org/web/20220425201853/https://www.washingtonpost.com/technology/2022/04/25/twitter-elon-musk-deal/ |archive-date=April 25, 2022 |access-date=April 26, 2022 |website=[[The Washington Post]]}}

MacMillan, Douglas; Siddiqui, Faiz; Lerman, Rachel; Telford, Taylor (April 25, 2022). "Elon Musk acquires Twitter for roughly $44 billion". The Washington Post. Archived from the original on April 25, 2022. Retrieved April 26, 2022.

{{Cite news |last=MacMillan |first=Douglas |last2=Siddiqui |first2=Faiz |last3=Lerman |first3=Rachel |last4=Telford |first4=Taylor |date=April 25, 2022 |title=Elon Musk acquires Twitter for roughly $44 billion |url=https://www.washingtonpost.com/technology/2022/04/25/twitter-elon-musk-deal/ |url-access=limited |url-status=live |archive-url=https://web.archive.org/web/20220425201853/https://www.washingtonpost.com/technology/2022/04/25/twitter-elon-musk-deal/ |archive-date=April 25, 2022 |access-date=April 26, 2022 |newspaper=[[The Washington Post]]}}

MacMillan, Douglas; Siddiqui, Faiz; Lerman, Rachel; Telford, Taylor (April 25, 2022). "Elon Musk acquires Twitter for roughly $44 billion". The Washington Post. Archived from the original on April 25, 2022. Retrieved April 26, 2022.

SurveyEdit

  • Use {{cite magazine}} or {{cite news}}: I'm pretty firmly of the opinion that we should use {{cite magazine}} or {{cite news}} as appropriate for a source. I do not see a difference between content that appears in the print edition of, for example, Entertainment Weekly and content that appears on Entertainment Weekly's website. Sideswipe9th (talk) 23:55, 30 June 2022 (UTC)
  • Use {{Cite web}}: A website associated with a magazine or newspaper is a different entity from the print magazine or newspaper. Due to cost and space constraints, many articles are only published on the magazine or newspaper-associated website, but are not present on the print edition. As a result, it is inaccurate to state that an article that appears on a magazine or newspaper-associated is equivalent to an article from a print magazine or newspaper. Many print magazine and newspaper publishers also publish digital PDF-style editions of their magazines and newspapers, which would more accurately fit the description of an online magazine. A website associated with a magazine or newspaper is inherently a website, and websites should be cited using {{Cite web}}. Finally, {{Cite magazine}} and {{Cite news}} contain parameters not found on {{Cite web}} that are not applicable to articles published on magazine or newspaper-associated websites, such as |volume= and |issue=. There is therefore no substantial benefit to use those two templates over {{Cite web}}. InfiniteNexus (talk) 00:47, 1 July 2022 (UTC)
  • Use {{cite magazine}}/{{cite news}}: much like online journals are journals, which should be cited with {{cite journal}}, online magazines are magazines, and should be cited with {{cite magazine}} template. This emits the correct metadata, and respects the principle of least surprise. It is also useful for our WP:MCW compilation, just like {{cite journal}} is useful for our WP:JCW compilation. {{Cite web}} is for general websites and other online sources that aren't covered by the other templates, per its documentation, and should ideally not be used for magazines. The same applies for {{cite news}}. Headbomb {t · c · p · b} 03:47, 1 July 2022 (UTC)
  • Speedy close as WP:POINTy and malformed RFC by a small group of editors butthurt that their opinions were second-guessed by a bot, per earlier discussion. Also, the question is written in a way that presupposes that group's intended answer, that being published on a website is somehow different from being in "the print edition" of a magazine, as if such a thing always exists these days. And it fails to distinguish magazine content on the web site (for which the correct answer in my opinion is {{cite magazine}} regardless of print appearance) from other content that happens to be on the same site (like say an faq on subscriptions, which I think should use {{cite web}}) As such, the wording is too prejudicial to produce a meaningful result, and incapable of being answered in a way that would actually describe my position. —David Eppstein (talk) 05:56, 1 July 2022 (UTC)
Nice try. There was consensus in the discussion above for an RfC (save for opposition from two editors) since participants were equally split on the issue. Falsely describing editors who disagree with your views as WP:POINTy and disruptive is deeply uncivil and not assuming good faith. InfiniteNexus (talk) 16:31, 1 July 2022 (UTC)
A small group of people repeating the same points over and over and over and over again until everyone else gets tired and stops responding is not consensus for action, it is WP:BLUDGEONING. And this RFC is continued bludgeoning. —David Eppstein (talk) 19:02, 1 July 2022 (UTC)
Multiple users expressed approval for an RfC, with only two dissenters with rather weak arguments. There is a legitimate reason to start an RfC, because the discussion above ended up with no consensus. To quote WP:BLUDGEONING, To falsely accuse someone of bludgeoning is considered incivil, and should be avoided. InfiniteNexus (talk) 00:53, 2 July 2022 (UTC)
@InfiniteNexus: at User talk:Citation bot#Why_do_we_need_an_RFC? I count five editors (including me) who agree that the RFC is a pointless timewaster.
So @David Eppstein 's complaint is true, and you are falsely accusing David of a false accusation.
On 13th June, @Sideswipe9th presented a big table of data[1] showing how @Citation bot makes hundreds of changes of template type per day. The three sites to which changes are opposed by @InfiniteNexus and the rest of the angry dozen from WP:WikiProject Film/Marvel Cinematic Universe task force account for less than 1% of all template-type-changes ... while nobody else joined in the earlier discussion to endorse the angry dozen Marvelites.
That is classic bludgeoning. BrownHairedGirl (talk) • (contribs) 16:37, 4 July 2022 (UTC)
@BrownHairedGirl: The only users who have explicitly rejected the idea of an RfC are you and David. On the contrary, Adamstom.97, Favre1fan93, Sideswipe9th, Rlink2, and I were all explicitly in favor of one. You all had a whole month's time to suggest an alternate way to end this dispute, but you didn't. InfiniteNexus (talk) 17:07, 4 July 2022 (UTC)
Oh dear. I could dig out all the diffs to show that you are misrepresenting the discussion, but to avoid overloading I will just present one: [2], where @Sideswipe9th writes I'm very heavily leaning towards what BrownHairedGirl has said; "that a small number of editors are making a mountain out of a molehill", as not only is the total number of reverts for any reason tiny, the total number of reverts as part of this dispute is even smaller
The reason why Sideswipe9th proceeded with the RFC is simply that @InfiniteNexus and the rest of the Angry Dozen were unswayed by the evidence that they are in a tiny minority, and continued their WP:BLUDGEONING. BrownHairedGirl (talk) • (contribs) 17:27, 4 July 2022 (UTC)
The reason why Sideswipe9th proceeded with the RFC Somewhat this, but also of all the options that I saw available to resolve the issue, this was the least worst. Ignoring it was just going to keep folks angry at each other. All the references in all of the MCU articles getting tagged with {{cbignore}} also impacts on the work of IABot and I believe WaybackMedic, both of which are vital to ensure archives are added to sources to prevent link-rot. And if behavioural issues spiralled, I could foresee one or more highly acrimonious ANI threads being opened. I'd rather this issue gets resolved by a strong community consensus, than any of the other potentially worse options. Sideswipe9th (talk) 17:36, 4 July 2022 (UTC)
@BrownHairedGirl: Okay, but the rest of my statement holds true. Adamstom said, I think we are good to go per the latest comments at Sideswipe9th's talk page. Favre said, I still feel an RfC will be helpful in gathering consensus on how editors approach these sources and which citation templates they use. Rlink2 said, RFC it is then. Indagate said, Think RfC is probably best way to get a consensus, current discussion is lengthy and doesn't seem to be reaching a conclusion. And as a bonus, Gonnym said, They don't need your approval to start the RfC and this sub-section of yours is very condescending. InfiniteNexus (talk) 17:43, 4 July 2022 (UTC)
@InfiniteNexus: v fine cherrypicking.
You selected the statements which support holding an RFC to put an end to your bludgeoining, and omitted several which do not support you. I am not going to get into posting every diff, but your blatant misrepresentation is an ugly followup to your assumptions of bad faith (e.g. [3] I feel like some editors are just trying to stall the RfC from happening) and your outrageous choice to attack me for complaining about a malicious accusation that the bot was engaging in vandalism BrownHairedGirl (talk) • (contribs) 18:19, 4 July 2022 (UTC)
@BrownHairedGirl: Sigh, here we go again. I read through that discussion several times, and I reaffirm my prior statement that only two (or three, if you count Sideswipe9th) editors have explicitly said no to an RfC. I selected the statements which support holding an RFC because you and David claimed there was no consensus for an RfC, which is not true. I was setting the record straight, not cherrypicking. And once again, I never attacked you for your comments about Darkwarriorblake. My full quote was, Sure, it was a fair request to ask Darkwarriorblake to retract their claim about vandalism. But did you really need to add a snarky statement at the end? Notice how I conceded it was a fair request in the first half of the sentence? Constructive criticsm != attack. InfiniteNexus (talk) 18:37, 4 July 2022 (UTC)
I selected the statements which support holding an RFC because you and David claimed there was no consensus for an RfC
Oh dear, oh dear. I should not need to explain that if you want to demonstrate that there is a WP:CONSENSUS, then counting only the views on one side is not the way to do that. But indeed, here we go again: sadly, such basics do need to be explained in this case.
And again, your quoted comment about Darkwarriorblake reaffirms my point: you made no criticism of Darkwarriorblake, but chose instead to criticise my response to a nasty attack by Darkwarriorblake in which they repeated their bogus allegation of vandalism.[4]
My response[5] was OK, so either you have't read WP:NOTVAND ... or you have read it and want to make false allegations anyway. So much so that you repeat those false allegations. Not good conduct ... but you chose to call me snarky for noting that a repeated false allegation of vandalism is not good conduct. Boggle. BrownHairedGirl (talk) • (contribs) 19:03, 4 July 2022 (UTC)
@BrownHairedGirl: I should not need to explain that if you want to demonstrate that there is a WP:CONSENSUS, then counting only the views on one side is not the way to do that. – I agree. But to reiterate what I said, 5 users expressed approval for an RfC (with a legitimate reason) while only 3 did the opposite (by falsely claiming only a "minority" of participants wanted an RfC). You tell me if that's consensus.
Your quoted comment about Darkwarriorblake reaffirms my point – your "point" was I attacked you, not I criticized you. There's a difference. Yes, I criticized you, but I never attacked you.
You chose to call me "snarky" for noting that a repeated false allegation of vandalism is "not good conduct"  – apologies if I wasn't being clear, but I was referring to your comment that said, If you make utterly bogus allegations, no wonder you get laughed at. That sounds snarky to me, do you not think so? InfiniteNexus (talk) 19:40, 4 July 2022 (UTC)
  1. Two straw men. I did not claim that there was a consensus in that discussion, or that only a minority in that discussion wanted a RFC. I simply contested your false statement that only two editors objected to the RFC. And it's not three objectors; it's 4.
    I pointed out that the only edits being objected to were the few relating a few websites, so that it was clear that the vast majority of editors do not object to these changes.
    You however, continue to focus only on your group of a dozen angry editors from WP:WikiProject Film/Marvel Cinematic Universe task force, falsely assuming that the wee group who got each other all angry on one project page are representative of the wider community ... despite the evidence that hundreds of similar edits are made by Citation bot every day, and only the 12 angry Marvelites object. I took that as a clear indication that the 12 Angry Marvelites are in a small minority, and nothing I have since leads me to reconsider that assessment.
  2. Attack, criticise; whatever. But you made no criticism of the editor who made a bogus allegation, and who repeated it when challenged. And here you are again, still making no criticism whatsoever of your ally who made the bogus allegation, and who repeated it again when challenged. That is classic tag-teaming conduct. You have now had at least five opportunities to clearly condemn the bogus allegation of vandalism, and your continued failure to do so does not paint a good picture of you.
    And no, I do not think that it is snarky to explain to someone who complains of being laughed at that repeating a grave and wholly bogus allegation of misconduct is likely to make people laugh at you. My response was harsh, but the allegation was very serious. BrownHairedGirl (talk) • (contribs) 20:50, 4 July 2022 (UTC)
If you have a problem with Darkwarriorblake's comments, their talk page and ANI are both right around the corner. I'm not sure how my opinion on this matter is relevant. Per Sideswipe, silence is the weakest form of consensus. I also don't find it appropriate for you to disregard the opinion of a group of "Marvelites" just because they're from the same WikiProject taskforce. I am so fed up with this. InfiniteNexus (talk) 22:41, 4 July 2022 (UTC)
I am fed up with this too: your continued creation of straw men is wearing.
  1. Your opinion on Darkwarriorblake is relevant because you chose to criticise me rather than him. You can remedy that if you want to, but it seems that you are still satisfied to act as the protector of a smear-monger.
  2. I do not disregard the opinion of a group of "Marvelites" just because they're from the same WikiProject taskforce.
    However, I do note that their complaint and their anger seems to be shared by nobody outside that group, which is a strong indicator that the 12 Angry Marvelites have created a prolonged drama out of nothing. BrownHairedGirl (talk) • (contribs) 23:29, 4 July 2022 (UTC)
  • Use {{cite magazine}}/{{cite news}} in the cases likely to be of main interest: online magazines are magazines, and online newspapers are newspapers, just like online academic journals are online academic journals. The argument that a print newspaper is a different entity from its online version relies on distinctions that don't matter as far as our policies are concerned: their stories are by the same people, under the same editorial supervision, held to the same standards. {{cite web}} can be the better option for content that happens to share a domain name, as mentioned above (a "Contact Us" page isn't a news story, for example, and Forbes "contributor" blogs aren't Forbes magazine). I share the concern raised above that this is not a well-formed RfC, despite (or because of?) the lengthy discussion that apparently led up to it. XOR'easter (talk) 16:29, 1 July 2022 (UTC)
To test this out, I just ran Citation bot on my sandbox. Apparently, it doesn't even change Forbes or Scientific American citations from {{Cite web}} to {{Cite magazine}} even if it's an article, but for this non-article from the Entertainment Weekly website, it did. InfiniteNexus (talk) 16:50, 1 July 2022 (UTC)
It looks like the bot can't always make the right decision going by URLs alone. Film at 11. XOR'easter (talk) 17:05, 1 July 2022 (UTC)
I know Citation bot isn't technically part of this RfC question, but this whole issue came about because of Citation bot's automated changes. InfiniteNexus (talk) 00:53, 2 July 2022 (UTC)
  • Use {{Cite web}} by default, as the bot cannot work out whether the web article is actually part of a physical or digital magazine or if it is non-magazine web content. It should be on the user who is adding the source to determine whether {{Cite magazine}} should be used instead. I have already elaborated on my opinions a lot as part of the previous discussion, I won't repeat all of that here unless someone asks me to. - adamstom97 (talk) 02:35, 2 July 2022 (UTC)
  • Do not use {{cite web}} for any sources that can be otherwise classified. CS1, like most citation systems, cites sources by type of work, not by type of media. An online periodical work is a magazine. An online news agency or newspaper is a work of journalism. Outside of the present case, an online book/encyclopedia/image is a book, an encyclopedia or an image. Something found in a corporate/institutional/government website is information in a corporate/institutional/government promotional publication. And so on. Citations are structured to follow these conventions. That is why they include authors of works, dates of publication of works, etc. Citing by medium has no analog in the real world, except in the rare cases that cannot be classified otherwise. 68.174.121.16 (talk) 16:00, 2 July 2022 (UTC)
  • Change templates and/or citation guidelines per my comment below. If templates are web-exclusive, those are at the top of the visual hierarchy and print-exclusive at the bottom. If print-vs-web is tagged, the proper usage of such tags must be explicitly clear to every editor. Either way, for a site that prides itself on verifiability, the fact that we've been completely ambiguous in templates so far about whether we cite print or web news is inexcusable. SamuelRiv (talk) 17:08, 2 July 2022 (UTC)
Structured citations cite works, mainly by title, author and date, because that is how works are easily found. Any other relevant information, including the publishing medium, is secondary. The important primary citation elements are substantially related to the work type as magazine, book, image, recording or whatever. This general citation practice should, and largely is, followed by CS1, by its template applications, and by helper scripts such as this bot. 68.174.121.16 (talk) 20:06, 2 July 2022 (UTC)
I thought this was clear from below: The medium is critical if the substance of the source differs, which it almost always does in modern print journalism. The only fidelity you'll get is from digitized archives. Note also (browse ProQuest, say) that major papers such as NYT have separate archives for digitized print -- including modern articles -- and their online versions. Most editors will cite online versions, so most of the templates and instructions should be aimed toward that audience, but then explicitly telling editors how to cite print when they use print. This shouldn't be a foreign concept -- it's already widely understood that you must cite the correct ISBN for books for correct pagination (and edition/revision), especially in the case of EBooks. SamuelRiv (talk) 21:14, 2 July 2022 (UTC)
No, this is not how citations work. Primarily, sources are not classified by medium. Citations have to follow classification in the real world if sources are to be discovered easily. Emphasizing the publishing medium (at best a secondary or tertiary index) may potentially make the source harder to find and impede the speedy verification of wikitext. It also does not matter in this discussion whether newspapers have separate digital or print or whatever versions, any more than they have separate weekday and Sunday editions. Citations do not cite editions/versions. They cite works, with all other information being ancillary to the work. Such as, |edition=online. There is no "special" citation information in the real world about publishing media that could elevate that element to a point where templates should be named for it, no matter what some Wikipedia editors think. 66.108.237.136 (talk) 13:44, 3 July 2022 (UTC)
You do understand that the content in the online version of a modern newspaper is different than what is in the print version, correct? SamuelRiv (talk) 16:29, 3 July 2022 (UTC)
Sure. Citations have been dealing with different editions of works (which may have differences in content) practically since their foundation. 4.30.91.142 (talk) 21:46, 3 July 2022 (UTC)
Yes, and there are multiple checks in CS1 on a citation of a book, say, for the precise edition: edition=, isbn=, and sometimes publisher= and year. (And of course, an EBook version also lacks page numbers.) For general websites the check is date and access-date, and to my surprise most editors seem to have gotten used to recording access-dates. In both cases I've found that probably a majority of citations are verifiable, and if not the citation usually is fatally flawed in multiple ways. For journal articles there is a known rampant problem here (in parts of academia too, frankly) of not specifying whether one is citing the preprint or the version in the doi, because the only real check there is page number (and sometimes date). For online vs print news, there are about the same two potential checks: page number and date. Based on the examples of books and websites, a change in presentation and norms is for the most part all that's necessary, for the issue raised by this RfC in particular. SamuelRiv (talk) 22:07, 3 July 2022 (UTC)
  • Cite news or cite magazine. cite web says it is used to create citations for web sources that are not characterized by another CS1 template. Because we have cite news and cite magazine, cite web should not be used here. weeklyd3 (message me | my contributions) 17:31, 2 July 2022 (UTC)
  • Cite news or magazine - The feedback request bot brought me here. Looked at the documentation for each of the templates and cite web's documentation says it is a generic catchall for web citations that do not otherwise fit into other CS1 templates. Cite news, for example, is a CS1 template and web content is specifically listed as valid content for cite news. Since a news article would fit in cite news, cite web is, per its own documentation, not the appropriate template to use. The same with cite magazine. I admit that I'm guilty of using cite web when other templates would be more appropriate, but I think the answer to the questions raised by the RfC is that cite web should only be used when cite news or cite magazine would not apply; an online version of news is still news, and an online version of a magazine is still a magazine. - Aoidh (talk) 03:09, 3 July 2022 (UTC)
  • Use {{cite magazine}} or {{cite news}}: online newspapers are still newspapers, and online magazines are still magazines.
    This RFC is a timewasting drama created as the result of a few angry editors who got worked up about Citation bot's edits to refs to a small set of publications. They wouldn't take no for an answer, and made repeated allegations of bad faith: one of them even accused Citation bot of vandalism, and then others tag-teamed to attack me for denouncing the smear on the bot-owner.
    The core issue here is simple: the publishing industry is in transition to new media, but the nature of the publications remains largely unchanged.
    Take for example, The Guardian newspaper in London. For 100 years it was available only on paper. Some time in the 20th century, it also became available on microfilm, mainly for use in libraries. By the 1990s, it was also available on some commercial databases. In the mid 1990s, it began to publish some of its stories on the guardian.co.uk website, and by the early/mid 2000s the website's scope extended beyond that of the print edition: not just more frequent updates, but additional content. In 2011, it adopted a "digital first" strategy: in May 2021, former editor Alan Rusbridger described this as part of a revolution which had begun in 1993.
    Now The Guardian is published in several formats: as a print paper (tho less focused on breaking news), on its website theguardian.com, and as an app for mobile devices.
    But for all these technical changes, it is still a newspaper. It still publishes daily, still publishes news, still employs lots of journalists, still carries opinion pieces and editorial and reader's letters.
    Sure, its focus has shifted in response to the immediacy of the internet, but it has been through similar shifts before: a century ago, radio began to take over the role of breaking news headlines, and sixty years ago television became the primary medium for delivering pictures. A century before that, the arrival of the railways had allowed newspapers to expand their distribution way beyond one city. And along the way, several major advances in printing technology radically reduced the time involved to assemble sub-edited articles into a printed paper, pushing copy deadlines much closer to the moment when the paper reached the streets.
    Those who try to maintain the notion of some clear distinction between a print newspaper or magazine and its electronic formats are out of touch with the reality of 2020s journalism, and seemingly unaware that digital is just the latest step in a long journey of change. Their stance that it's not a newspaper article unless it is read on a printed page is an absurdity which denies the multiple-media nature of 2020s newspapers and magazines.
    Citation bot does a great job of deploying the various templates which format citations to various types of publications, and it is a great shame that so much energy has been wasted by a few editors whose fundamental error is to conflate the type of publication (newspaper, book, magazine, journal etc) with the medium of distribution. --BrownHairedGirl (talk) • (contribs) 16:11, 4 July 2022 (UTC)
  • Use {{cite magazine}} or {{cite news}} per existing guidelines as mentioned above. A magazine, whether online or in print, is still a magazine; mutatis mutandis about newspapers. Imzadi 1979  17:31, 4 July 2022 (UTC)
  • Use {{cite web}}, It's notable that despite me having raised this issue many times that the CitationBot groupies have not notified me of this vote because they know I will not go the way they want which is a deliberate attempt to sway the vote. Websites are not newspapers or magazines, articles that appear online do not necessarily appear in print and vice versa and it's bizarre to me how strongly the citationbot groupies are fighting against what clearly so many people outside their sphere do not want. It's an attempt to fix a problem that does not exist, even more bizarrely trying to force upon the wider Wikipedia something they did not ask for. There does not seem to be much notification about this RFC to the wider Wikipedia but funnily enough those directly involved with the bot, with a specific agenda, are all fully aware of it. As print inevitably dies, we will still be citing website only companies as magazines and newspapers? How would that make sense. Citation Bot should be fixing existing references, not modifying them to something different entirely which seems entirely beyond its purview or purpose. The citationbot groupies aggressively defend their position as can be seen on the talk page history, they are unwilling in anyway to see any POV that is not their own regarding this issue and so it must be forced upon them that they are not to be mass changing reference types by pretending that digital, online sources, are somehow a print magazine or newspaper. EDIT: TO put this in perspective, Entertainment Weekly ceased publication 3 months ago, and it's been a website since the 1990s. It's been a website almost as long as it was a magazine and it will be a website longer than it was a magazine. Now that any article it publishes is digital only, are we still meant to be citing it to a magazine?Darkwarriorblake / SEXY ACTION TALK PAGE! 08:12, 8 July 2022 (UTC)
    You have a watchlist just like the rest of us. Additionally "articles that appear online do not necessarily appear in print and vice" is completely irrelevant, and doesn't change an online magazine into a non-magazine. Headbomb {t · c · p · b} 08:31, 8 July 2022 (UTC)
    I don't watch the page as I don't like to see the constant bullying going on. But you have ignored that I've just said Entertainment Weekly, as a magazine, no longer exists, and it was posted online pretty much from the outset. Darkwarriorblake / SEXY ACTION TALK PAGE! 16:08, 8 July 2022 (UTC)
    It exists as an online magazine. Which is what you constantly forget, hence the need to repeat the obvious. Headbomb {t · c · p · b} 20:02, 8 July 2022 (UTC)
    Please note that Darkwarriorblake's accusations of bullying are malicious and unfounded.
    The only bullying which has occurred was that Darkwarriorblake accused @Citation bot (and by implication it maintainers @AManWithNoPlan) of vandalism. When pointed to WP:NOTVAND, Darkwarriorblake repeated te false accusation.[6]
    It is disgraceful that the bully Darkwarriorblake is trying to abuse this RFC as a place to try to invert history. BrownHairedGirl (talk) • (contribs) 14:57, 11 July 2022 (UTC)
  • Use {{citation}} That's what I do and I've never understood why we have a proliferation of media-specific templates when many sources are multimedia. For example, I get physical copies of The Times newspaper and especially look at its obituaries for notable deaths. I usually work from a clipping but, when citing this, add a URL for the web copy for the convenience of readers who can get past the paywall. This copy will then end up in various newspaper archives. And sometimes such obituaries are bound into books. A generic template which caters for such multiplicity seems best because it avoids arguments and provides alternatives for readers. Andrew🐉(talk) 09:18, 8 July 2022 (UTC)
    • That's a nice idea, but it means that people have to be very careful about parameter choices, which is its own can of worm. Also, some combinations of the paramters that make sense are not allowed. AManWithNoPlan (talk) 16:55, 8 July 2022 (UTC)
      I use the {{citation}} template routinely and have no trouble with it. Other experienced editors such as Johnbod don't use a template at all and just cite with plaintext in a natural way. The more the process of citation is complicated, the more difficulty it causes and the more of a barrier to new editors it becomes. See the KISS principle. Andrew🐉(talk) 10:14, 9 July 2022 (UTC)
  • Use {{cite magazine}} or {{cite news}} per existing guidelines as mentioned above. AManWithNoPlan (talk) 16:55, 8 July 2022 (UTC)
  • Use {{cite magazine}} or {{cite news}} according how the source describes itself. The medium is as irrelevant as whether the journalist wrote the story on a PC or a Mac. --John Maynard Friedman (talk) 17:02, 8 July 2022 (UTC)
  • Use {{Cite web}} or {{Cite news}} without rehashing all of my points listed in the past discussion at the CitationBot talk page, no where in current documentation for {{Cite magazine}} does it discuss "online magazines". In my opinion, based on the documentation of Cite magazine, that template should be used for your "actual" magazine issue publication and anything within it, be it in a print or digital "online" version. If you have an article published by the publication on their website (more so today than ever) that appears in no way shape or form in the magazine issue, Cite web or Cite news seems like the appropriate template per documentation that should be used. - Favre1fan93 (talk) 17:03, 8 July 2022 (UTC)
  • Use {{Cite web}}: basic rule of referencing is to cite the actual thing you've read, content beteen web and magazine can be different. Indagate (talk) 17:15, 8 July 2022 (UTC)
    • Not sure how that is relevant. Changing from {{Cite web}} to {{Cite magazine}} does not magically do a web search and replace the URL with a link to a library to pick up a copy. AManWithNoPlan (talk) 17:49, 8 July 2022 (UTC)
  • Use {{cite magazine}} or {{cite news}}. As others have noted, it's about the source, not the medium - the medium isn't relevant, because if it was, it would have some silly implications. Imagine two editors are adding content to a Wikipedia article, and both cite the same news article. One uses the physical medium version of an article, the other editor the web edition. If we want to be really strict, that means that the two editors should use/create distinct references, one to web and one to news/magazine, despite the source content being identical? That seems pointless, and worse, misleading, since the source is the same (barring some sort of online-only correction). Additionally, what about e-books? Those uncontroversially use cite book (I hope!). SnowFire (talk) 13:40, 9 July 2022 (UTC)
    • Obligatory proviso: Note that in the cases where a newspaper/magazine owns or sets up a website sufficiently separate from their main product, then {{cite web}} is fine, since that's more a website that happens to be published by a company that is also a newspaper / magazine. (e.g. FiveThirtyEight is not the same as ABC News, despite being owned by them.) I don't know what citation bot was doing, but a change like that would be bad (e.g. using the "corporate parent" too seriously). SnowFire (talk) 13:40, 9 July 2022 (UTC)
      • Note that in the cases where a newspaper/magazine owns or sets up a website sufficiently separate from their main product, then {{cite web}} is fine, since that's more a website that happens to be published by a company that is also a newspaper / magazine. This is a great comment that I think summarizes my feelings. And I'd take it a step further in asking (which is similar to a comment response I added below), if the website shares the same name as the newspaper/magazine, but what's being published on the website far exceeds what is included in said newspaper/magazine, should that be considered sufficiently separate to use separate cite templates? - Favre1fan93 (talk) 16:01, 9 July 2022 (UTC)
  • Use {{Cite magazine}}/{{Cite news}}. My reasoning stands not from the method of publication, but from the editorial process that is shared between the content published in print and on the web. I conceptualize {{Cite web}} as something published online with less editorial oversight than something reliable at RSP. E.g. WP:FORBESCON (ignore that it's generally unreliable--this is an example) I would use {{Cite web}}, NOT {{Cite news}}. I prefer {{Cite magazine}} over {{Cite news}} for magazines because the publication is a magazine. SWinxy (talk) 00:38, 11 July 2022 (UTC)
  • Use whatever. Online newspapers and magazines are web sources; they are also newspapers/magazines and news media. These categories are not mutually exclusive. As such, the jurisdictions of {{cite web}} and {{cite magazine}}/{{cite news}} overlap, with both being equally appropriate for online newspapers/magazines. They give exactly the same output as each other when used for online news sources under almost all circumstances (the only exception - and it's a rare one - is if you need to take advantage of a parameter that's available in one template and not in the others, in which case you should use the citation template that allows the use of said parameter). Seriously, people, don't we have more-important things to become mortal enemies about? 🤦‍♀️ Whoop whoop pull up Bitching Betty ⚧️ Averted crashes 07:39, 11 July 2022 (UTC)
  • That's exactly the point I tried to make in previous discussions, but the operators of Citation bot insisted on changing {{Cite web}} to {{Cite magazine}} even though there is practically nothing wrong with using {{tl|Cite web}). InfiniteNexus (talk) 16:36, 11 July 2022 (UTC)
  • cite book, cite journal magazine, etc. The following examples show that cite book, cite journal magazine, etc. are designed to display all the information that a reader is likely to want to locate the source in whichever form is available, or which the reader prefers. This is not the case for cite web.
Cite magazine comparison
Wikitext {{cite magazine|access-date=October 18, 2013|date=2008|doi=10.1016/j.enpol.2007.05.021|first1=Myriam B. C.|first2=Guy R.|issue=6|journal=Energy Policy|last1=Aries|last2=Newsham|name-list-style=amp|pages=1858–1866|title=Effect of daylight saving time on lighting energy use: a literature review|url=http://archive.nrc-cnrc.gc.ca/obj/irc/doc/pubs/nrcc49212/nrcc49212.pdf|volume=36}}
Live Aries, Myriam B. C. & Newsham, Guy R. (2008). "Effect of daylight saving time on lighting energy use: a literature review" (PDF). Energy Policy. Vol. 36, no. 6. pp. 1858–1866. doi:10.1016/j.enpol.2007.05.021. Retrieved October 18, 2013.
Cite web comparison
Wikitext {{cite web|access-date=October 18, 2013|date=2008|doi=10.1016/j.enpol.2007.05.021|first1=Myriam B. C.|first2=Guy R.|issue=6|journal=Energy Policy|last1=Aries|last2=Newsham|name-list-style=amp|pages=1858–1866|title=Effect of daylight saving time on lighting energy use: a literature review|url=http://archive.nrc-cnrc.gc.ca/obj/irc/doc/pubs/nrcc49212/nrcc49212.pdf|volume=36}}
Live Aries, Myriam B. C. & Newsham, Guy R. (2008). "Effect of daylight saving time on lighting energy use: a literature review" (PDF). Energy Policy. pp. 1858–1866. doi:10.1016/j.enpol.2007.05.021. Retrieved October 18, 2013.
Notice cite web omits the volume information. Jc3s5h (talk) 13:09, 11 July 2022 (UTC)
  • This RfC is about {{Cite magazine}} and {{Cite news}}, not {{Cite journal}}. There is no opposition against using {{Cite journal}} to cite journals. InfiniteNexus (talk) 16:39, 11 July 2022 (UTC)
    • Revised 16:46 UTC to reflect RFC asks about cite magazine and cite news. — Preceding unsigned comment added by Jc3s5h (talkcontribs) 16:47, 11 July 2022 (UTC)
      • Also note that the specialised parameters such as volume, issue, pages, etc. do not apply to web pages anyway. The only difference between using {{Cite magazine}} and {{Cite news}} instead of {{Cite web}} is that we have to put the website's name in the magazine or work parameters. - adamstom97 (talk) 21:22, 11 July 2022 (UTC)
        • That's not entirely true. As was discussed at User talk:Citation bot and is in the summary below the question, each of the different templates also outputs different COinS metadata. The use of {{cite magazine}} over {{cite web}} ensures that the correct metadata is embedded for that source. Information on how the metadata differs was detailed in this reply by Trappist the monk. Functionally there is a significant difference between all of the CS1 cite templates. Sideswipe9th (talk) 22:02, 11 July 2022 (UTC)
          • That would only be true if we agreed that a web article associated to a magazine should have the same metadata as an article that actually appears in a magazine, but that has not yet been determined since this RfC is still ongoing. - adamstom97 (talk) 22:26, 11 July 2022 (UTC)
        I disagree that the volume, issue, or pages parameters do not apply to web pages. It depends on the arrangement of the web page. It might be a landing page where one can select the issue one needs. Or it might be a PDF with a large number of pages, and the same pagination as the print version. I consider it easier to just use cite news, cite magazine, etc. whenever possible, rather than puzzle over all the subtle differences between these templates and cite web. Jc3s5h (talk) 23:00, 11 July 2022 (UTC)
  • Use {{cite magazine}} or {{cite news}} (Summoned by bot). Also use them in cases where print news and magazines publish online editions (or audible editions, or braille, or whatever) that may have different content. This is no different than the NY Times publishing several city editions, a couple of national editions, and an international edition. Online editions are merely an extension of this, and there is no need to treat the different content as "not a newspaper" or magazine, merely because it is published in a different medium or because the content may not be identical with a print edition. Mathglot (talk) 05:43, 13 July 2022 (UTC)

DiscussionEdit

Note: So far I've only added the tech topic. I'm not sure if any of the other topics are appropriate, but if they are feel free to add em or let me know and I'll add them. We'll also want to notify any relevant WikiProjects if anyone has a list of those handy, and the village pump about this discussion, as it is relevant to a great many pages across enwiki. Thanks. Sideswipe9th (talk) 23:55, 30 June 2022 (UTC)

Notified: Help talk:Citation Style 1, WikiProject Citation cleanup, WikiProject Academic Journals, WikiProject Magazines, WikiProject Newspapers Sideswipe9th (talk) 00:10, 1 July 2022 (UTC), updated Sideswipe9th (talk) 16:24, 1 July 2022 (UTC)
Note: Updated the notices after the move, and I've struck the one for this page because that's now where we're holding the RfC. Sideswipe9th (talk) 16:12, 2 July 2022 (UTC)
Comment: Before I do so, do editors believe it would be considered WP:CANVASSING to notify Wikipedia:WikiProject Film/Marvel Cinematic Universe task force of this RfC, as that is where this issue first arose? InfiniteNexus (talk) 00:49, 1 July 2022 (UTC)
Yes, because this is not specific to that WikiProject. WP:JOURNALS and WP:MAGAZINES should be notified though, since these are the projects associated with journals, magazines, etc... Headbomb {t · c · p · b} 03:50, 1 July 2022 (UTC)
I've now notified WikiProject Academic Journals, Magazines, and Newspapers. Sideswipe9th (talk) 16:24, 1 July 2022 (UTC)
What about Wikipedia:Bots/Noticeboard? InfiniteNexus (talk) 00:53, 2 July 2022 (UTC)
@InfiniteNexus: I've been thinking about this one for the past few days. While the discussion did spin out from Citation Bot's talk page, due to issues that were originally raised there, the RfC question itself is somewhat broader because it's asking specifically about the citation templates and not the bot's edits. Because of this and the header at BOTN, I'm not entirely sure this is within the scope of that noticeboard. I'd like to hear from others though on this.
Conversely, what do people think about notifying one of the Village Pumps? I'm not sure which of them to notify though, otherwise I likely would have done it already. Sideswipe9th (talk) 15:37, 8 July 2022 (UTC)
@Darkwarriorblake: if you feel as though this RfC has not been adequately publicised, please feel free to suggest additional pages where we can provide that notification. There's still at least 23 days until the RfC tag expires, so that is plenty of time to provide further notifications. Sideswipe9th (talk) 15:33, 8 July 2022 (UTC)
Featured Articles should be asked, since the changes that this bot makes that introduce inconsistencies between citations have been cited as an issue in several FAs I've put forward. But requesting comments from the groups whose templates would BENEFIT from enforcing the improper use of cite newspaper and magazine to cite websites is not a fair RFC. Darkwarriorblake / SEXY ACTION TALK PAGE! 16:12, 8 July 2022 (UTC)
WP:FA has many pages, though I can't immediately see a noticeboard. Is there a centralised FA discussion noticeboard that I'm missing, or do you have a specific talk page where such a notification would be best placed? I'll make no more comment than I have above in the survey about whether or not one of the options is an improper use of the various citation templates, because ultimately that is what is under discussion here. Sideswipe9th (talk) 16:19, 8 July 2022 (UTC)
I probably should have done this earlier, but to address Darkwarriorblake's concerns I will go ahead and ping all those who participated in the prior discussion but have yet to vote here (which should not be considered canvassing per WP:APPNOTE, which allows notifying Editors who have participated in previous discussions on the same topic (or closely related topics)): @Abductive, AManWithNoPlan, Eurohunter, Favre1fan93, Gonnym, Indagate, John Maynard Friedman, Rlink2, Trailblazer101, Trappist the monk, and ZooBlazer. InfiniteNexus (talk) 16:50, 8 July 2022 (UTC)

Physical print media will always require a different citation than the same item posted online (unless perhaps if it's an image scan). Anyone who has ever looked up a recent news article online can see the "Updated on xx-xx-xxxx" date and know that there will be some, however slight, difference from whatever went out in print. That said, WP citations from print media will likely be extremely rare relative to online-access media (with the diminishing exception of long-form books). So I ask what is the point of having "news", "journal", and "magazine" citation templates presented so prominently when it would seem for all online publication the proper template would "cite web" (or some variant for journals – remember print can be different there too). Really this requires rethinking how the template options are presented to editors: almost always the sources will be either print books or web-accessible. After those are introduced, then you can show the exceptional cases, such as A/V media and from-print citations. SamuelRiv (talk) 03:10, 1 July 2022 (UTC)

I suppose this could be accomplished with just a type=Print source or similar specification tag, but the rules on how to cite, use citation templates, and for these types of tag have to be made crystal clear. Verifiability is not an MOS issue. SamuelRiv (talk) 19:17, 1 July 2022 (UTC)
@SamuelRiv writes Anyone who has ever looked up a recent news article online can see the "Updated on xx-xx-xxxx" date and know that there will be some, however slight, difference from whatever went out in print.
That is partially true. Yes, online stories can be modified indefinitely; but many newspapers go through several editions each day of the printed paper. Sometimes the front page and several other pages of the final print run can be almost entirely different from that day's first edition.
Some newspapers, such as the UK's Daily Mail and Daily Express run radically different version of the same story in different regions, with the English edition denouncing as a outrage something which the Scottish edition cheers in the same front page slot.
So the variability applies to both media, just to a different degree.
The way to deal with this is simple for online sources: to include with each ref an archived copy of the source as used for the article. I have been doing this for years, and it should be standard practice; the use of an unarchived (and hence modifiable) sources should be strongly deprecated. BrownHairedGirl (talk) • (contribs) 16:23, 4 July 2022 (UTC)
Right, obviously you cite the edition -- if something other than the national/syndicated edition -- when you cite print news. It is a known historical problem that most newspapers only microfilm one of their daily editions, which excludes some articles, though libraries sometimes do have different editions archived. I'd say it's one of those cases where the internet took a problem that was more of an idiosyncrasy in the past, limited just to big papers and simply resolved with a library visit, and made it far more pervasive. Now even the crummiest local or school papers update their stories many days after publication, so edition control is no longer just an issue for urban giants. Of course thanks to iarchive it's often possible to recover several intermediate versions if necessary. I'm not implying the old days were better -- I mean they were for journalism, but not at all for this reason. But this is getting quite off topic.
The point is that if someone, for whatever reason (e.g. a lot of freelance work is not archived online per Tasini), is citing a print source for WP to the exclusion of an electronic version, the procedure and template has to be clear for that purpose. The misunderstanding between the two sides that is driving this RfC is in part that one side is saying essentially, "What are cite news and cite magazine for if not for print sources," while the other is saying, "How does that make sense with how the templates are used in practice?" Given that the number of citations from print sources (aside from books) is a tiny minority, I think the variety of templates offered to the users should reflect that. But the instructions for use should be presented clearly -- "If you are citing print, this is the method" -- and the templates themselves should in turn be less ambiguous in their documentation.
And if your argument is that we should prefer online-accessible sources, then I completely agree. Of course if an editor is replacing original citation information they have to check that the citation still verifies what is attributed to it (which is just good practice for citation maintenance anyway because the prevalence of citation decay and plain misattribution is pretty unsettling). And of course the reason this RfC is here as far as I understand are enthusiasts who are using print sources and don't really have an adequate alternative. And then you have all the print books that are used, the plethora of English-language print from just this century that still hasn't been digitized. And if you want any good historical material for anywhere else in the world, it's going to be a long long time before anyone has the inclination to digitize that stuff. It's just something that has to be addressed. SamuelRiv (talk) 20:13, 4 July 2022 (UTC)
@SamuelRiv: no, I am not arguing that we should prefer online-accessible sources. Many fine scholarly sources are not online, and in my view Wikipedia is far too tolerant of low-quality online sources.
I am simply noting that while the move online has (as we seem to agree) massively increased the modification of articles, most of the problems can be resolved by the simple good practice of archiving the page as cited. Citation decay is rampant, but we already have all the tools in place to prevent decay of most online refs. Sadly, many editors think it is acceptable to add a bare URL as a ref, and there is no systematic effort to discourage that. Few editors routinely archive the citations they add, and I have several times faced angry responses from editors who think that the archive data is clutter.
The variability of printed newspapers is less easy to track. It's not just a matter of national/syndicated versions; each edition may change in the course of a print run. As a young policy wonk in the political system, I often used to note how the first edition copy I purchased on my way home at midnight was significantly different from that which my housemates purchased in the morning. And most newspapers are far from clear in identifying what edition the reader holds in their hands.
After a year of working full time on refs, my conclusion is that WP:Verifiability does not in practice carry anything remotely like the weight that it should, or that it purports to carry. It's not just that we tolerate widespread use of newspapers and magazines rather than insisting on scholarly sources; we are very lax in how sources are cited.
In my first few weeks at a university humanities course, our twice-weekly tutorial papers were mercilessly shredded for failures in citation. They were returned in front of the whole (smallish) class, with fierce criticism of any failure to attribute or any lack of completeness in the citation. It was brutal, but it worked. It seems to me that very few en.wp editors have ever received such invaluable training in the centrality of citations to scholarly writing.
Here on Wikipedia, we gently treat wholly a unsourced addition as a minor lapse which might merit a {{citation needed}} tag, while bare URLs and other vague waves at sources are rarely reproached, and the widespread practice of citing a long book or PDF without a page number is almost never rebuked. Instead of telling newbies firmly that full and accurate citations to high-quality sources are the absolute basic standard required of any contribution, WP:BITE is thrown at those who try to uphold good practice. Heck, this whole RFC arose out of some editors focused on refs to three magazines: Entertainment Weekly, Rolling Stone, and Billboard. They are better than blogs, but would not be acceptable as sources of fact in academic work. Heck, we even have hundreds of of thousands of articles with references to user-generated sites such as IMDB and Ancestry.com, and to social media.
So long as we tolerate such low standards, I fear that the energy involved in precise attention to the nuances is misplaced. BrownHairedGirl (talk) • (contribs) 21:40, 4 July 2022 (UTC)
I applaud your sentiment; people should not get through high school without being able to cite properly. Alas, adding a archive reference to every link is not currently possible. I have tried to persuade archive.org to archive pages off the Wikipedia feed (and when that did not work, I tried to get WMF to buy archive.org). I do feel that we should be able to speedily delete unsourced articles, but WP:PRESERVE is the policy that says that we must gently treat a wholly unsourced addition as a minor lapse which might merit a {{citation needed}} tag. If you want to start an RfC to repeal or rewrite it, you have my support. Hawkeye7 (discuss) 00:06, 6 July 2022 (UTC)
Thanks, @Hawkeye7, but it seems to me that the culture of patronising newbies by tolerating unsourced or poorly-cited additions is very deeply ingrained. So I don't see much point in starting an RFC which would just descend into an acrimonious bout of shouting from those who don't want to be held to anything remotely near basic scholarly standards. BrownHairedGirl (talk) • (contribs) 16:51, 6 July 2022 (UTC)

David Eppstein I wanted to address and query one point you raised above with respect to the RfC being malformed, specifically the question written in a way that presupposes that group's intended answer. As is clear from my contributions in the past discussion, I drafted a not insubstantial amount of the RfC and how it was presented. I did this because, per my comments in #Why do we need an RFC? I saw no other way to address an intractable dispute between editors that was spilling over into the article space. I was open to, and acted upon feedback from many contributors both in favour of the transformations of the bot and who are opposed to it. As there was a period of 30 days between the posting of the second draft, and the opening of the RfC, why did you not voice any concerns about the question being leading or presupposed towards an intended answer? I would happily have tried to address those concerns during the drafting period, as while my opinion is firmly in that the bot's edits are fine and correct, I wanted to be as fair as I possibly could to both sides of the argument. Sideswipe9th (talk) 19:19, 1 July 2022 (UTC)

@Sideswipe9th: - For the record the feedback request bot brought me here so that question was my first interaction with what's going on in the discussion. I think it's very neutrally worded and after getting a better perspective on what the various talking point are, is a good summation. - Aoidh (talk) 03:11, 3 July 2022 (UTC)
@Aoidh: thank you for the kind words. While I obviously have my own views on the situation, I wanted to be as fair as possible to both sides of this disagreement. It's also why I'm trying to be proactive in addressing issues as they arise, as ultimately I want this to be the strongest possible consensus no-matter the outcome, because I want to see this issue resolved without further acrimony where possible. Sideswipe9th (talk) 16:28, 8 July 2022 (UTC)

Comment: I've seen a few editors opine on this above, and wanted to address this directly. The issue of whether or not the website of a publication, eg Entertainment Weekly or The Guardian, is the same as or different to the print edition of the publication when it comes to citations is fundamentally the locus of this dispute. This is an important question to consider across enwiki as print media continues to decline, and more traditionally print only sources like newspapers and magazines become website only. It is clear from the discussion above that there are editors who believe that using {{cite magazine}} for Entertainment Weekly, or {{cite news}} for article content on their websites is appropriate, and it is also clear that there are editors who believe that only {{cite web}} is appropriate. This RfC is best thought of, in my opinion, as a measuring stick to gauge what the community consensus is on this issue, and it is not an appropriate discussion to be remarking on behavioural concerns. Sideswipe9th (talk) 16:40, 8 July 2022 (UTC)

I agree. That in my eyes is what I hoped to gain clarity on. Sure, an "online magazine" is a magazine, but where does the classification distinction come up, when you have more and more content released from these publications that aren't actually featured in their "magazines"? - Favre1fan93 (talk) 15:57, 9 July 2022 (UTC)
Especially in the cases where the publication has an "online magazine" that is separate from their website and the content between the two is different, which is basically where this discussion started. Many of the editors who has responded to the RfC have not addressed this part of the issue, which was a concern of mine when drafting the RfC question. - adamstom97 (talk) 02:01, 10 July 2022 (UTC)
That seems almost a philosophical question. Do you consider the website of a publication to be separate from the publication? Or is the website of a publication another distribution mechanism for the publication's readers? Does is strictly matter from our perspective of citing a source if one format of a publication has content exclusive to one of it's delivery mechanisms? Sideswipe9th (talk) 22:05, 11 July 2022 (UTC)
Also I think it's possible to infer from some of the answers. Take SWinxy's contribution, it's pretty clear that if you were citing an article on the website of a magazine or newspaper where the website shares the same editorial standards, you would use {{cite magazine}} or {{cite news}} as appropriate.
I could also reverse the question. Using the EW example above, why do editors like Favre1fan93 and yourself see that as not being published by the magazine? Why is the website of the publication different from the publication itself? Sideswipe9th (talk) 22:13, 11 July 2022 (UTC)
Because if you are to define what a magazine is (a publication of various articles "bound" in issues or volumes), if a web article published by someone like EW doesn't appear in that "bounded" magazine (being it on physical print paper or a digital/scanned version), how can we call that a "magazine"? - Favre1fan93 (talk) 17:26, 12 July 2022 (UTC)
I don't think that's the sole universally accepted of a magazine, and it pretty much excludes online magazines that are delivered via website instead of PDF or some other ebook container format. And while all print magazines will likely have issues (typically numbered or date delineated), not all will have volumes. Are print magazines without volumes less of a magazine than one that does? Sideswipe9th (talk) 18:05, 12 July 2022 (UTC)
Perhaps my quick definition wasn't the best, but there is, and I think should, be a distinction between material presented within an issue publication (choose your "container" method of delivery), and then a web article that simply is not part of such publication. - Favre1fan93 (talk) 18:39, 12 July 2022 (UTC)
a web article that simply is not part of such publication I can't agree with that. Sticking with the EW example, content on ew.com is subject to the same editorial standards as their former print edition, and is published by the same organisation. This is also true for content on Variety, Billboard, and TIME. As such I do not see any distinction between an article that is published only on that publication's website, and an article that is published only on that publication's print or ebook container. It is fundamentally the same publication, written by the same authors, edited by the same editors, and held to the same standards. Sideswipe9th (talk) 18:48, 12 July 2022 (UTC)
But there is a difference to them, they are deciding what content goes in their actual magazine (physically published or digitally published) and what content will just go on their website. We aren't saying there is a difference in quality, just that there is a difference, and using {{cite magazine}} with the |magazine= parameter to source information that they didn't put in their magazine just seems incorrect. As pointed out above, {{cite magazine}} and the other more specific cite templates include parameters that are specific to their respective media, they are designed for sourcing information from that media. If I have never read EW magazine and I am just getting information from ew.com that was never included in any (physical or digital) magazine publication and does not need any of the special magazine parameters that {{cite magazine}} was designed to provide, then why should I use {{cite magazine}}? The answer seems to be that the metadata in {{cite magazine}} is required for such a citation, but that metadata is incorrectly saying that the information comes from a magazine. - adamstom97 (talk) 00:47, 13 July 2022 (UTC)

Note As the RfC tag has now expired, and the last contribution to the discussion was on 13 July, I've made a closure request at WP:CR so we can wrap things up. Sideswipe9th (talk) 01:18, 31 July 2022 (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.

unsupported and deprecated parameter cleanupEdit

I am reminded that, per Editor Jonesey95, we should have a formal discussion to fully remove support for:

  • |booktitle=
  • |chapterurl=
  • |episodelink=
  • |mailinglist=
  • |mapurl=
  • |nopp=
  • |publicationdate=
  • |publicationplace=
  • |serieslink=
  • |transcripturl=

All of these, if used in a cs1|2 template, will emit the unknown parameter error message, |transcripturl= excepted which is deprecated in {{cite episode}}.

The only vestiges of these parameters are found in Module:Citation/CS1/Configuration except for |transcripturl= which is also in Module:Citation/CS1/Whitelist.

Shall we remove support for the above named parameters?

Trappist the monk (talk) 14:25, 7 July 2022 (UTC)

I support this cleanup. To elaborate on the above description, the function of the canonical versions of these parameters is not being removed; we are only removing the unhyphenated aliases of the parameters. We have slowly standardized on hyphenation of multi-word parameters over the years. In other words, |chapter-url= will continue to work fine, but |chapterurl= will not. The above parameters currently emit an "unknown parameter" error message in preview as well as assigning Category:CS1 errors: unsupported parameter. That category contains only 69 pages currently, so finally removing the above parameter aliases should not affect the display of articles beyond those 69 pages. Trappist, please correct any of this if it is incorrect. – Jonesey95 (talk) 16:07, 7 July 2022 (UTC)
This search should find articles in Category:CS1 errors: unsupported parameter that use any of the above listed parameters except |transcripturl= for which use this search.
Trappist the monk (talk) 18:56, 7 July 2022 (UTC)
One other category where use of these parameters might be found is Category:CS1 errors: empty unknown parameters. Use this search to look there (includes |transcripturl=).
Trappist the monk (talk) 21:35, 9 July 2022 (UTC)
I've gnomed that category down to empty and none of the errors that I fixed were caused by the parameters in question. Tangent alert! There were, however, a lot of errors caused by |deadurl=. Is there currently a bot fixing these, and if not might Monkbot be interested in them? They're a bit tedious to do by hand but a bot should find them tasty. Best, Wham2001 (talk) 09:07, 8 July 2022 (UTC)
Support the cleanup, per Jonesey95. 68.132.154.35 (talk) 17:04, 7 July 2022 (UTC)
The documentation for Citation Style Vancouver suggests the opposite convention to CS1|2 – not using hyphenated parameter aliases – for the sake of getting out every bit of performance. Will CSVAN's documentation change to encourage using to hyphenated parameter names? Is CSVAN still even maintained? SamuelRiv (talk) 17:57, 7 July 2022 (UTC)
Those templates use a mishmash of parameter styles: |trans_chapter=, |chapter-url=, and |chapterformat= are listed in the documentation for {{Vcite book}}. CS1 started moving away from this mishmash long ago, and is almost done. {{Vcite book}} has only 123 transclusions, and {{Vcite web}} has only 93. It would be an interesting exercise to determine if those templates could be converted to {{cite book}} with appropriate style parameter values without changing their rendered appearance. If so, it would probably make sense to merge them with their corresponding CS1 templates. – Jonesey95 (talk) 18:36, 7 July 2022 (UTC)
Small-caps is going to be a hard stop. Izno (talk) 18:51, 7 July 2022 (UTC)
Is CSVAN still even maintained? No, and the user who created it mostly by himself has been blocked for a long time now. Izno (talk) 18:50, 7 July 2022 (UTC)
I think that I would oppose merging Citation Style 1 with Citation Style Vancouver. cs1|2 is complex enough that attempting to support CSVAN would be a nightmare. Perhaps best to simply let existing {{vcite ...}} citations fade away and then remove support for the templates via TFD.
Trappist the monk (talk) 18:56, 7 July 2022 (UTC)
@Izno: the user who created it mostly by himself has been blocked for a long time now Who might that be? The templates {{Vcite book}}, {{Vcite journal}}, {{Vcite news}} and {{Vcite web}} were all created by Eubulides (talk · contribs), who is no longer around, this is true; but their block log is clean. --Redrose64 🌹 (talk) 19:40, 8 July 2022 (UTC)
I'm thinking of Wikid, who I recall pushing these as if they were sliced bread. Izno (talk) 21:13, 8 July 2022 (UTC)
Who also has a clean block log, and moreover, zero edits in template space. --Redrose64 🌹 (talk) 20:48, 9 July 2022 (UTC)
@Redrose64: I think "Wikid" refers to Wikid77. * Pppery * it has begun... 21:17, 9 July 2022 (UTC)
Performance may have been one of the reasons that the {{vcite ...}} templates were created but that 'advantage', if it were an advantage, pretty much went away when MediaWiki enabled mw:Extension:Scribunto and Lua.
Trappist the monk (talk) 18:56, 7 July 2022 (UTC)
I'm only now trying to learn this stuff and searching through the forums gives all the discussions of the old performance issues of COinS but only hints as to what extent that was resolved (other than people stopped talking about it). CSVAN still features prominently in {{Wikipedia referencing}} which is on the CS1 Help page, so if it's old and dead (as are a lot of citation templates) then some cleanup is warranted just to let people know what the current standards are. I'm still curious if it's possible to "switch off" the metadata tags with say a wrapper for CS1|2-based templates which may have some fields that might not be suitable for inheritance, such as if the data in the field will likely always be considered polluting. SamuelRiv (talk) 19:12, 7 July 2022 (UTC)
Use of the {{vcite ...}} templates in mainspace:
These numbers suggest that the popularity of the {{vcite ...}} templates is on the wane. No doubt the primary reason is that bots, visual editor, RefToobar, Diberri Boghog template filler, etc create cs1|2 templates. If I'm not mistaken, WP:MED used to be one of the primary users of the {{vcite ...}} templates. The above searches suggest that WP:MED has moved away from {{vcite ...}}.
Wrapping cs1|2 templates gives you the benefits of the cs1|2 module suite which includes metadata, error checking, presentation, etc. If you don't want metadata or if you think that the metadata provided by a wrapped cs1|2 template would be bogus, it would probably be best to write your template from scratch. If you expect such a template to be used in articles that are predominantly cs1|2-cited articles, be mindful of presentation because editors care about that; that includes differentiating between cs1 and cs2 presentation.
Trappist the monk (talk) 22:07, 7 July 2022 (UTC)
No. Significant opposition to the proposed changes was raised in the RfC, and it does not appear that anything has changed since then.
Pinging participants in that RfC who have not yet commented here, apologies if I've missed anyone: @Pppery, Thryduulf, Imzadi 1979, Kusma, MichaelMaggs, Fram, Wham2001, Hawkeye7, Gonnym, Tol, Levivich, JohnFromPinckney, Andrew Davidson, Finnusertop, and MGA73:. Nikkimaria (talk) 01:38, 8 July 2022 (UTC)
What is the objection to this specific change, which will have zero effect on any articles? Deprecation of the above parameters happened in February 2021, all of the parameters were updated shortly thereafter, and any stray instances of those parameters have been generating error messages (and have been quickly fixed) since April 2021. Additional factual information appears in this discussion. – Jonesey95 (talk) 13:45, 8 July 2022 (UTC)
The problems with this effort were outlined in the two previous RfCs, plus there is the concern raised by Hog Farm/Nosebagbear below. Nikkimaria (talk) 17:21, 9 July 2022 (UTC)
  • Support removal—since these unhyphenated forms don't do anything anymore except generate errors, why are we keeping them around any longer. Imzadi 1979  02:52, 8 July 2022 (UTC)
  • Thank-you for the ping Nikkimaria – my watchlist is currently an unusable tire-fire so I would have missed this discussion otherwise. I support (if we're doing bold !voting) removal of these aliases largely per Jonesey95. They appear not to be in use anywhere in the encyclopedia (unlike, say, |accessdate= which was controversial in the RFC). As such I think that standardisation on the hyphenated variants is only beneficial. Best, Wham2001 (talk) 09:14, 8 July 2022 (UTC)
  • Support removal - it is much easier to maintain (and copy to other wikis) if there are not a bunch of variants that are really not needed. And thank you for the ping. --MGA73 (talk) 16:24, 8 July 2022 (UTC)
  • There is a rough consensus that non-hyphenated parameters should not be deprecated in citation templates (option C). That's April 2021. What did I miss? Levivich[block] 16:55, 8 July 2022 (UTC)
    It is confusing. What you missed is the link to a subsequent January 2022 RFC at the top of this thread; in that RFC, the above parameters remained deprecated, but complete removal of support for them was pushed to a new discussion. This is that discussion. What you and many participants in that April 2021 discussion missed is that the above parameters did not exist in article space at the time of the discussion. They had already been deprecated, and removal of support for them was to be a mere formality. Unfortunately, they were one of the babies thrown out with the bathwater and remained in a sort of limbo. This discussion is finishing that cleanup work in the formal way required by the January 2022 RFC closure. Support for or deprecation of the six remaining active unhyphenated parameter aliases, |accessdate=, |airdate=, |archivedate=, |archiveurl=, |authorlink=, and |origyear=, is not being contemplated in this discussion, and I beg everyone here not to mention them again in this thread, lest it go off the rails. – Jonesey95 (talk) 06:36, 9 July 2022 (UTC)
    I am confused. Thanks for the link, but what you linked to does not appear to be an RFC. What I linked was an RfC from a year ago where we decided not deprecate any unhyphenated parameters. So where is the RfC that changed that consensus? Because from where I'm sitting, since April 2021, there are no deprecated unhyphenated parameters. I don't think there's been another RfC since there, has there? Even this isn't tagged RfC. So what gives? Why are editors referring to deprecated parameters after the April 2021 RFC? Levivich[block] 02:48, 10 July 2022 (UTC)
    The January 2022 RfC here. -- GreenC 03:01, 10 July 2022 (UTC)
    Where in that RFC was it decided to deprecate anything? Can you quote it for me? Levivich[block] 03:07, 10 July 2022 (UTC)
    That RFC found no consensus to remove deprecated parameters, and though it's not stated in the closing, the reason is because there aren't supposed to be these deprecated parameters in the first place. chapterurl should be an alias for chapter-url, booktitle should be an alias for book-title, and so forth. That's what the April 2021 RFC decided, and as far as I can tell, that hasn't changed. People are pointing to a (non-RFC) discussion in Feb 2021 as the place where it was decided to deprecate these parameters, but that local consensus was overridden by the global consensus of the April 2021 RFC that said "There is a rough consensus that non-hyphenated parameters should not be deprecated in citation templates." That global consensus wasn't changed by the January 2022 RFC. This is all highly irregular. Levivich[block] 03:10, 10 July 2022 (UTC)
    The January 2022 RFC said that there is no consensus on removal of deprecated parameters in this discussion, and further discussion will be necessary to roll that part out. This is that necessary discussion, a year and a half after the January 2022 RFC. The parameters above have been deprecated since February 2021, have not been used anywhere since March or April 2021, and the proposal now is to remove them from the citation module code entirely. This is how deprecation works, when it works; parameters or methods are deprecated, then removed from use, then discontinued entirely. Historically, with these modules, the process has happened much more smoothly and with a lot less drama. – Jonesey95 (talk) 06:17, 10 July 2022 (UTC)
    What does "non-hyphenated parameters should not be deprecated", from the April 2021 RFC mean, and why doesn't that override the earlier Feb 2021 decision to deprecate? I'm sorry that my questions are too "drama" for you, I'll try to ask them in a less dramatic way. Can we just do this where you pretend I'm not stupid and explain in plain English why the words "non-hyphenated parameters should not be deprecated" doesn't mean that non-hyphenated parameters should not be deprecated, and why you're saying that any non-hyphenated parameters are deprecated if there was an RfC that resulted in consensus that non-hyphenated parameters should not be deprecated. Because that's how RFCs work. Levivich[block] 06:37, 10 July 2022 (UTC)
    When the closer said that "non-hyphenated parameters should not be deprecated" in April 2021, they were referring to the remaining six unhyphenated multi-word parameters: |accessdate=, |airdate=, |archivedate=, |archiveurl=, |authorlink=, and |origyear=. Those six parameters are not at issue here; the beginning of mass bot modification to hyphenated parameters prior to a deprecation discussion about those six was the main cause of that RFC and the main objection of participants. Maybe that is the main source of your confusion? That would be understandable, because that discussion got pretty convoluted.
    As of April 2021, the parameters at the top of this discussion were already deprecated and had been removed from all namespaces that assign CS1 tracking categories. They were not undeprecated by the April 2021 discussion. Now, go forward in time to January 2022. In that month's RFC, there was a long list of code changes proposed, including remove support for previously deprecated parameters |booktitle=, |chapterurl=, |episodelink=, |mailinglist=, |mapurl=, |nopp=, |publicationdate=, |publicationplace=, |serieslink=, |transcripturl= (note that this is the list of ten parameters, deprecated since February 2021, at the top of this July 2022 discussion). Those changes were generally approved for implementation, but there is no consensus on removal of deprecated parameters in this discussion, and further discussion will be necessary to roll that part out, so those ten parameters remained in the deprecated state. Now we are having that further discussion, with the goal of moving those ten parameters from deprecated, non-standard, and unused (where they have been since February 2021) to completely unsupported, i.e. removed from the code. – Jonesey95 (talk) 07:07, 10 July 2022 (UTC)

Please quote where in the April 2021 RFC it specified certain parameters and not others. The RfC question was Should non-hyphenated parameters be fully removed from the CS1/2 family of templates? The background section said In 2014 an RFC determined that all parameter names in Citation Style 1 templates should have an alias...This meant, for example, that access-date= would be shown as the preferred parameter while accessdate= was shown as acceptable but discouraged from use. In the following years there have been various trends and discussions formally deprecating many of the non-hyphenated parameters, from a small handful (2019 example) to the current list which contains over 70 entries. Many of these are the non-hyphenated variants of the preferred/hyphenated versions, which are being removed to decrease the maintenance burden and increase the uniformity between templates (i.e. "ease of use")...In October 2020, all non-hyphenated parameters were added to the "current list" linked above. Option C was Option C: Non-hyphenated parameters should not be deprecated; deprecation should not be continued and bot approval should be revoked. This will also mean that the deprecated parameter list will need to be updated to remove the non-hyphenated parameters. And the close was: There is a rough consensus that non-hyphenated parameters should not be deprecated in citation templates (option C). This means that existing hyphenated (e.g. access-date) and non-hyphenated (e.g. accessdate) parameters should both continue to be supported by the templates. Bot removal of non-hyphenated parameters from transclusions, i.e. Monkbot task 18, does not have community consensus. Where does any of that specify some deprecated parameters and not others? In October 2020, all non-hyphenated parameters were added to the "current list" and This will also mean that the deprecated parameter list will need to be updated to remove the non-hyphenated parameters. means, to me, that ALL non-hyphenated parameters were un-deprecated in April 2021. @Joe Roe: am I misreading this close? Levivich[block] 14:39, 10 July 2022 (UTC)

It is clear to me that my multiple attempts to use words to explain the timeline have not been effective for you, so I will either leave it to someone else to help you, or leave it to the person who summarizes this discussion to determine what the next step should be. Happy editing! – Jonesey95 (talk) 04:02, 11 July 2022 (UTC)
  • Support removal - been deprecated for a long time, and are essentially unused now. Time to clean them up. --PresN 19:25, 8 July 2022 (UTC)
  • Support removal it's high time to cleanup. Headbomb {t · c · p · b} 20:00, 8 July 2022 (UTC)
Support removal. Was pinged to this discussion. These parameters are not used and the convention is to use hyphenated versions. Gonnym (talk) 23:18, 8 July 2022 (UTC)
Support removal for these few as they've been essentially trace fossils for quite some time, with the understanding that this discussion should not be claimed as a consensus for removal of other non-hyphenated parameters. Hog Farm Talk 06:40, 9 July 2022 (UTC)
Concur with removal for these without precedent - as with Hog Farm, I back the removal here without wanting the risk for it to act as a precedent across all paramaters. Nosebagbear (talk) 12:32, 9 July 2022 (UTC)
Oppose This is trying to justify something that the community agreed should never have been done based on a fait accompli * Pppery * it has begun... 18:44, 9 July 2022 (UTC)
  • Support removal and thank you for your efforts in maintaining citation templates. 19:42, 9 July 2022 (UTC) — Preceding unsigned comment added by Renata3 (talkcontribs)
  • Oppose because this is going to be used as evidence to remove the other non-hyphenated parameters --Guerillero Parlez Moi 14:11, 11 July 2022 (UTC)
    That is not a valid reason for any opinion, pro or con. A discussion result is not evidence of anything except of the existence or absence of consensus among participants of said discussion. Apart from that anything can be used to promote or oppose anything. And that is irrelevant. 50.75.226.250 (talk) 15:54, 11 July 2022 (UTC)
  • Support removal of these parameters for better standardisation. Tol (talk | contribs) @ 18:58, 12 July 2022 (UTC)
  • Support only for those that generate errors - What is the point of keeping them if they generate errors? I opposed ditching the hyphenated parameters in the previous RfC, and still think there are too many hyphenated parameters that people actively use. I don't mind deprecating them, but they shouldn't be removed. Just have a bot/AWB/whatever go around and clean them up if they're functional-but-not-ideal. The idea is to to maximize user functionality. Lots of people have used hyphenated parameters for a long time, so getting rid of them... is just going to generate errors or not show up. Of course, if some are already only generating errors, that ship has sailed. — Rhododendrites talk \\ 12:29, 13 July 2022 (UTC)
    @Rhododendrites: Do you mean non-hyphenated? No one is suggesting to get rid of the hyphenated ones. –MJLTalk 06:48, 14 July 2022 (UTC)
    All of the parameters listed at the top of this discussion generate error messages and error-tracking categories. They have been deprecated since February 2021. Deprecated (and unsupported) parameters in CS1 templates generate error messages and error-tracking categories. – Jonesey95 (talk) 07:41, 14 July 2022 (UTC)
    Yes, I understand that, but they're not hyphenated (eg. |booktitle= is unhyphenated). –MJLTalk 19:17, 14 July 2022 (UTC)
  • Oppose removal, after thinking about it a bit, because instead of removal, what should happen is that these unhyphenated parameters should be aliases to their hyphenated counterparts. |booktitle= should be an alias to |book-title=, |chapterurl= should be an alias to |chapter-url=, and so on. That these parameters currently give errors isn't a reason to remove them, it's a reason to fix them and return them to the aliases they used to be. The April 2021 RFC found consensus that non-hyphenated parameters should not be deprecated in citation templates and that the deprecated parameter list will need to be updated to remove the non-hyphenated parameters. Assuming that was done, there should be no non-hyphenated parameters on the deprecated parameter list; they should not be throwing off errors; support for these parameters should not be removed; they should be returned to functioning aliases. Levivich (talk) 02:28, 17 July 2022 (UTC)
  • Comment: I am not sure what the best option is here, since I can come up with decent arguments in favor of either. On one hand, removing them would be kind of a pain in the ass. Having the templates break entirely when given unhyphenated parameters would cost us something, in that everyone trying to insert a reference would have to go back and dick around with the template to fix huge red error messages. In a big article with a lot of references, this can be a significant amount of work. On the other hand, keeping them would also be kind of a pain in the ass -- it would populate the citation error category, requiring someone to go through and gnome the errors out. On the third hand, though, this seems a lot easier than having people fix the parameters while writing the article (you could probably just have a bot go through every few days and fix the errors). In conclusion... who knows. jp×g 20:15, 17 July 2022 (UTC)
    It is possible that JPxG misunderstands the proposed change and its effects. These unhyphenated multi-word parameters, like |chapterurl=, have been deprecated for a year and a half, currently display error messages, do not (or should not, at least; we may have missed a couple) appear in the documentation, and are not in use in the spaces that are categorized by the citation templates. Nobody is going to have to do any work to implement or clean up after this change. Standard hyphenated multi-word parameters, like |chapter-url=, will continue to work without any problems. – Jonesey95 (talk) 23:35, 17 July 2022 (UTC)
Yeah, sorry, I wrote this ass-backwards. I meant to say "unhyphenated". jp×g 07:14, 18 July 2022 (UTC)

need to automate doi-access change to 'free', based on dateEdit

Is this worth adding to the module, or should I create a new citation template for JIPA?

We will be citing several hundred illustration of the IPA in JIPA, linked via DOI to their host publisher site at CUP. They are subscription for the first 3 years, then free public access. It would be nice if the template could handle this. E.g., something like doi-access= {{#ifexpr:(({{CURRENTYEAR}} - {{{date}}}) > {{{freeage}}})|free}}, where we could set freeage to 3. (Since for the URL link 'free' is the default, there shouldn't be any confusion that 'freeage' applies to the DOI link.)

See Qaqet language#Further reading, where I used the 'translation' parameter to add a note about the article. It is -- or will be -- in the 'Illustrations of the IPA' section of the journal, and that is the standard wording used in the lit to refer to these JIPA articles, so that phrasing should be included but isn't part of the title. Also there is supplementary material (sound files) online which are a good publicly accessible resource and should be mentioned, but I don't see a param designed to handle that info.

Please ping. Thanks, — kwami (talk) 21:48, 12 July 2022 (UTC)

For pmc identifiers we have |pmc-embargo-date=. But, embargoed pmcs are relatively common and not constrained to any particular journal. Are there other journals or publishers that have similar doi embargos? I'm not enthusiastic about creating and maintaining code in the cs1|2 module suite that will support only one journal. If doi embargos are a common practice among multiple journals/publishers then we might consider creating |doi-embargo-date=.
|trans-title= has a defined purpose. Do not misuse cs1|2 parameters for purposes that do not fit with the parameters' defined purposes. Use |department= for the 'Illustrations of the IPA' section of the journal.
Trappist the monk (talk) 11:04, 13 July 2022 (UTC)
@Trappist the monk: Okay, thanks. That meaning of "department" isn't in my vocabulary. What param should I use for notes/comments, such as 'includes supplementary material'? 'Pages', maybe? — kwami (talk) 01:54, 18 July 2022 (UTC)
|doi-embargo-date= has some merit as an idea. The issue is mostly that this is a very hard thing to keep track of manually, since it varies a lot by journal and publisher. Headbomb {t · c · p · b} 02:18, 18 July 2022 (UTC)
cs1|2 does not have any parameters for notes/comments. |pages= has a defined purpose. Do not misuse it for something else. If you want to append notes/comments to a citation, they can go at the end – after the template's closing }} and before the reference's closing </ref>.
Trappist the monk (talk) 11:39, 18 July 2022 (UTC)

url-status=live without an archive-urlEdit

I've added this to Template:Citation Style documentation/doc:

Note: if |url-status=live is included but there is no |archive-url=, an error message is displayed when the link is hovered over, and a warning shown when the edit is previewed, although the reference shows normally in the list.

It is a pain to find and remove these instances; I'd suggest that |url-status=live be allowed without generating an error message, it seems harmless.

I added this comment to the documentation boldly; if it's thought inappropriate, or needs rewriting, go ahead.

Best wishes, Pol098 (talk) 11:45, 13 July 2022 (UTC)

This is about this edit at Template:Citation Style documentation/url‎.
A maintenance message is not an error message. To display maintenance messages, editors must enable them using the css specified at Help:CS1 errors § Controlling error message display. Are you using Navigation popups? That tool does not obey css rules and so displays the maintenance message on hover. The generic popup does obey css rules so the normally hidden maintenance messages are not displayed in the popup on hover unless they have been properly enabled.
I will undo your edit at Template:Citation Style documentation/url‎.
Trappist the monk (talk) 13:14, 13 July 2022 (UTC)
Thanks for response and revert. The documentation doesn't make any suggestion about whether adding |url-status=live without an |archive-url= is appropriate; should there be any preferred option? I won't do anything else. Best wishes, Pol098 (talk) 14:02, 13 July 2022 (UTC)
Perhaps. At the time of original insertion. all URLs are presumed live, so adding the parameter then may or may not be deemed overkill. Establishing a relationship between |url-status= and |archive-url= when the URL is not live may prompt editors to add an archived version. But I don't see any utility in generating any kind of message, maintenance or otherwise, when there is no |archive-url= and there is |url-status=live. 50.75.226.250 (talk) 14:49, 13 July 2022 (UTC)
@Pol098: It should be documented that using any value of |url-status= is incorrect, unless the citation includes an |archive-url=. (I thought it was, somewhere, but I'll have to look.) The reasoning is that Wikipedia can never make statements about the "current status" of a URL in citations (since that status can change at any moment, and the parameter would not be updated), so in fact we don't do that because it's not what |url-status= means.
|url-status=live actually has nothing to do with the status of the URL itself, really, its sole purpose is to communicate to an |archive-url= parameter that it should allow the |url= parameter to remain the primary link for the citation, rather than being relegated to a "the original" link with the |archive-url= displayed as primary. So |url-status= is merely a secondary argument to |archive-url=, like |archive-date=, that controls how the citation is formatted. The same way an |archive-date= without an |archive-url= would be a mistake (it makes no sense whatsoever), a |url-status= without an |archive-url= is a mistake for the same reason.:::|url-status=live/dead is probably misnamed, and should have been called |archive-status=backup/primary or something similar. (With |archive-status=primary being the default, corresponding to the default |url-status=dead today.) That would better communicate both the true meaning of the parameter and its relationship to |archive-url=. That'd be a hard thing to change now, but might be worth it anyway just to clear up the confusion that results from the generic |url-status= label. FeRDNYC (talk) 15:06, 14 July 2022 (UTC)

Thanks @FeRDNYC and Trappist the monk for useful contributions. FeRDNYC's explanation makes so much sense, though I don't know any of the background, that I propose another edit to the documentation in place of what I originally said. I will consider adding it if there are no comments to the contrary, and nobody else has done it (probably better).

The |url-status= parameter should not be used (with any value) unless a (dated) |archive-url= has been provided. Wikipedia never knows the current status of a URL, whatever it was when a citation was created.

Best wishes, Pol098 (talk) 15:38, 14 July 2022 (UTC)

You misunderstand the nature and purpose of |url-status=. It is a secondary helper parameter, and its value, following the norm of CS1 templates, was never expected to be automated or dynamic. It is a static value (including a default static value), valid at the time of a related URL-based edit, that may trigger a change in the presentation of some URL-based parameters and other associated parameters. Granted that is not documented clearly, but your proposal makes things worse, and misrepresents the issue. 68.132.154.35 (talk) 19:27, 14 July 2022 (UTC)
I understand what url-status is for; I don't see how to word this very briefly in documentation. My wording won't actually make usage worse - it should discourage anyone who reads it from using it for non-archived references, but I'm the first to agree that others can do better. Describing it as a " a secondary helper parameter", while correct, requires quite a bit of explanation to an editor seeking simple guidance. How do you, 68.132.154.35, or anyone suggest this should be worded in documentation with the purpose of discouraging misuse? "It should be said this way" is useful; "it shouldn't be said that way" is not. Beat wishes, Pol098 (talk) 19:36, 14 July 2022 (UTC)
It can definitely be worded better. However, the responses only referred to your proposed edit. Also, is it worth anybody's time and effort getting involved in this any more than cursorily? For instance, imo your proposal approaches this from the wrong angle. A rational design would simply make the parameter dependent on the presence of any URL-based parameter, not just one of them, and condition its value range accordingly. The current blinkered design and implementation is reflected in the documentation and obviously generates equally confused/confusing contributions such as this section or the RFC to rename the parameter, as if the name was the fundamental problem here. Good luck in your endeavors! 50.75.226.250 (talk) 15:38, 15 July 2022 (UTC)
Thanks. It can definitely be worded better. - so how about some actual suggestions, or even edit the documentation appropriately? is it worth anybody's time and effort getting involved in this any more than cursorily? Cursorily is my aim, add a sentence to the documentation to reduce the instances of unneeded live statuses, not the "proper solution" proposed.

Where I'm coming from: editing articles I find warnings in the edit preview. There is no information on which of the huge number of references is at fault, or how, and many warnings don't show as red messages in the references. While I suppose there are better ways to find the errors, I haven't gone about this systematically, but simply tried hovering over each reference number in the preview, which is tedious. It would be a minor improvement if editors were encouraged not to add unnecessary statuses, rather than think that it is "helpful". Best wishes, Pol098 (talk) 16:24, 15 July 2022 (UTC)
There is no information on which of the huge number of references is at fault. Not in the yellow preview-message box, but every cs1|2 template that contributes to the green messages in the yellow preview-message box emits its own warning message where the template is rendered. This template emits the warning message for |url-status=live without |archive-url=:
{{cite book |title=Title |url-status=live}}
Title.{{cite book}}: CS1 maint: url-status (link)
If you do not see the maintenance message at the right end of the line above, you need to enable message display for which, see Help:CS1 errors § Controlling error message display.
Trappist the monk (talk) 16:45, 15 July 2022 (UTC)
@68.132.154.35: It is a secondary helper parameter, and its value, following the norm of CS1 templates, was never expected to be automated or dynamic. Well... I mean, except for the bot we have running around, making automated changes to that and other parameters. FeRDNYC (talk) 03:56, 19 July 2022 (UTC)
What CS1 is about, and what other scripts do with it, are different subjects. In this case, the particular script is helpful. As stated above, the problem is with the design and implementation of |url-status=, not with the archive parameters or any script that applies them. In any case, "dependent variable" does not necessarily mean "dynamic variable". 50.75.226.250 (talk) 15:23, 19 July 2022 (UTC)

@FeRDNYC:@Trappist the monk: I have added the following unexplained instruction to the documentation, which I think follows the spirit of this discussion without going into unnecessary detailed explanation. I am sure it will be deleted or corrected if there is disagreement. The |url-status= parameter should only be included if an |archive-url= is set.

Thanks for the suggestions on revealing CS1 messages. Best wishes, Pol098 (talk) 15:17, 10 August 2022 (UTC)

RFC: Rename url-status parameterEdit

The following discussion 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.

The meaning of the |url-status= citation parameter is frequently misunderstood by editors. (See the discussion immediately above this one, url-status=live without an archive-url, for just one example of literally dozens, if not hundreds, over the years.)

I would argue that this confusion is our fault, because the current parameter name is just terrible. |url-status= has nothing whatsoever to do with the current status of the citation URL. (An unknowable, since it can change at any moment.) The only effect of |url-status=live is to prevent the citation's |archive-url= from being promoted to the primary citation link, the way it otherwise is with the default |url-status=dead in effect. Like |archive-date=, NO value of |url-status= has any meaning unless an |archive-url= parameter is also provided.

As such, the name of the parameter should clearly start with archive-. Above, I proposed |archive-status=primary/backup, but that doesn't account for the third possible value of the existing parameter, |url-status=usurped. So upon further reflection, I instead propose that we deprecate |url-status= and replace it with |archive-display=primary/backup/usurped.

|archive-display=primary
Would be the default, where the |archive-url= is the primary citation link and the |url= becomes a "the original" link. Same as the current default |url-status=dead.
|archive-display=backup
Would disable the replacement of the |url= as the primary link, to support preemptive archival linking of non-dead citations. (Same as |url-status=live today.)
|archive-display=usurped
Would, like |url-status=usurped today, disable all linking to the citation's original |url=, for situations where the original link has changed without becoming dead.

I'm fully conscious of what a colossal PITA it will be to make this change, but I feel it can be done in stages (and with some bot assistance) to ease the pain. More importantly, I feel it's ultimately worth the pain because the current parameter name, |url-status=, is just bad. It bears almost no relation to the actual meaning of the parameter, in ways that will continue to mislead and confuse editors until we finally fix our stuff. (I'll cross-link this proposal to Village Pump (technical) as well.) FeRDNYC (talk) 18:15, 14 July 2022 (UTC)

I ended up posting at Wikipedia:Village_pump (proposals)#Discussion on renaming Citation Style 1 template parameter url-status instead. FeRDNYC (talk) 18:27, 14 July 2022 (UTC)
Strong oppose. You said that url-status= has nothing whatsoever to do with the current status of the citation URL when in fact it literally does. The whole point of url-status is to determine if the link is dead. You can have url-status=dead without there being an archive link. That way, other people can add the archived link later.
Besides the amount of disruption and work required for a basically cosmetic change is not worth it. Rlink2 (talk) 18:30, 14 July 2022 (UTC)
Comment I have no opinion even though it will result in a lot of broken bots and user tools that would need to retool (if ever gets done, some poorly maintained but active tools remain nameless). Also "primary" and "backup" are coded terms that are not in themselves explanatory. Suggest "first" and "second" ie. archive URL displayed first ie. archive-display=first .. is a lot more intuitive and easy to remember what it means. -- GreenC 18:36, 14 July 2022 (UTC)
Strong oppose—this parameter was renamed/reworked not that long ago, and renaming it again would be highly disruptive. Imzadi 1979  18:39, 14 July 2022 (UTC)
History:
Trappist the monk (talk) 18:42, 14 July 2022 (UTC)
Oppose Let's not have more of this sort of churn please. * Pppery * it has begun... 18:44, 14 July 2022 (UTC)
Oppose pointless churn, big hassle for editors and bot developers, no benefit to readers. —David Eppstein (talk) 18:58, 14 July 2022 (UTC)
  • Oppose, no benefit. Proposal is worse than the status quo. Editor should not have attempted to start an RFC before starting a discussion to get the general consensus, which is clearly opposed. – Jonesey95 (talk) 19:14, 14 July 2022 (UTC)
  • Oppose I don't see how this parameter is confusing. InfiniteNexus (talk) 04:11, 19 July 2022 (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.

Accept zero padding in datesEdit

Currently the template does not accept a date such as 09 July 2022, giving an unhelpful error message. Instead one must write 9 July 2022. I think it's silly to bother editors with this, the template should accept both styles and silently reformat it. I thought it would be better to ask here before changing it because this template is so widely used. Tercer (talk) 12:52, 16 July 2022 (UTC)

See MOS:DATESNO: "Do not zero-pad day". – Jonesey95 (talk) 20:20, 16 July 2022 (UTC)
You misunderstand me. I'm saying that we should always display 9 July 2022, but allow editors to input 09 July 2022. Tercer (talk) 20:28, 16 July 2022 (UTC)
Why not get people to do it right in the first place? If they get into the habit of using it in cite templates then they will use it in other places in the text which will not get autocorrected. Keith D (talk) 22:27, 16 July 2022 (UTC)
Tercer, I understand what you are saying, but it is a bad idea, as Keith D explains. For the same reason, we also emit error messages when editors type 1/15/2003 or 01-15-03 or 2004-05. Editors should type valid dates so that there is a better chance that the date they type is correct and will not have to be checked or fixed by others. – Jonesey95 (talk) 00:19, 17 July 2022 (UTC)
Dates in M/D/Y format, two-digit years, or "2004-05" are ambiguous (though technically speaking, 1/15/2003 isn't). 09 July 2022 isn't ambiguous, though. (On a side note: Module:Citation/CS1/Date validation seems to have the code already written to support leading zeros, though it is commented out.) isaacl (talk) 20:56, 17 July 2022 (UTC)

Frankly, annoying editors with a pointless error message in order to teach them a minor stylistic point? You guys are a lost cause. I'm sorry I tried to help. Tercer (talk) 08:12, 17 July 2022 (UTC)

You are correct, and such pointless error messaging has caused problems in the past. Also agree this increasingly looks like a lost cause. 104.247.55.106 (talk) 13:43, 17 July 2022 (UTC)

I entirely agree with @Tercer. I do a lot filling of refs, and for accuracy's sake I copy-paste the date from the article where possible. Many date formats are unambiguous, but need to be re-formatted to avoid error error messages. These other formats are easily parsed; why trigger an error message when the date is unambiguous?

e.g. for 9 December 2014, the templates should accept:

  • 09 December 2014
  • 9 December, 2014
  • 9 DECEMBER 2014
  • 9 DECEMBER, 2014
  • 9 DEC 2014
  • 9 Dec. 2014
  • 09 Dec., 2014
  • 09th DEC, 2014
  • ... and many other variants.

The current demand for conformity is just a make-work. --BrownHairedGirl (talk) • (contribs) 08:03, 22 July 2022 (UTC) By all means trigger a warning, but not an error.

When updating the URL, should the access-date and archive-url parameters also be updated?Edit

One of the pages I'm wanting to edit uses the {{cite news}} template, but the news article URL that the page links to 404s - the site in question has changed its layout. I've found the new URL on the site, and I'm assuming (but am not completely sure) that the access-date parameter should be changed in the template, since the new URL was accessed on the date I'm editing. Is this correct?

Obviously, the old archive-url link still works, since it's, well, an archive. Should I leave the archive URL as-is, or update it to match the new URL? --TheSophera (talk) 23:48, 17 July 2022 (UTC)

However you like, so long as the links verify the cited fact. If there no fact being cited, IMO prefer the newer link in both url and archive-url .. the main thing is WP:V -- GreenC 23:55, 17 July 2022 (UTC)

AV identifiersEdit

For your consideration:

May have enough critical mass to be included in the Identifiers group. 50.75.226.250 (talk) 16:09, 18 July 2022 (UTC)

Cite document emits a missing periodical maintenance message. It shouldn'tEdit

Case in point:

Pinging User:Keith D. Headbomb {t · c · p · b} 12:28, 22 July 2022 (UTC)

It also does not have the page identifier (pp.) Keith D (talk) 12:32, 22 July 2022 (UTC)
{{cite document}} is not a cs1|2 template. It is merely a redirect to {{cite journal}}. As such, cs1|2 sees the template as a {{cite journal}} template and renders whatever it has been given as a journal-style citation. {{cite journal}} requires |journal= so when that parameter is empty or omitted, cs1|2 emits the error message. Pagination rendering follows the {{cite journal}} format (<colon><space><pagination>).
The source in the example template is clearly not a journal so were it me, I'd rewrite the example template:
{{cite web |last=Amos |first=David |url=https://www.academia.edu/40640025 |title=Controversies in Coal: Internment, Impoundment and Intrigue at Harworth Colliery (1913–1924) |date=January 2017 |pages=10–11 |access-date=5 October 2020 |website=Academia}}
Amos, David (January 2017). "Controversies in Coal: Internment, Impoundment and Intrigue at Harworth Colliery (1913–1924)". Academia. pp. 10–11. Retrieved 5 October 2020.
Trappist the monk (talk) 13:07, 22 July 2022 (UTC)
cite document is meant for things that aren't journals and should behave accordingly. The solution can't also be to redirect it to cite web instead of cite journal because it should also not require a url. Headbomb {t · c · p · b} 13:16, 22 July 2022 (UTC)
Documents should be cited only when published. Unless published as pamphlets, i.e. standalone which would fall under {{cite book}}, they are mostly published as parts of other works, including collections, repositories and websites. The citation should follow the conventions of the including entity, so {{cite web}} seems proper here. The logical treatment would be to delete so-called citation templates that contravene these simple guidelines, or at a minimum to strongly emphasize that editors should not expect support for such templates here if they insist on using them. This page should not waste time every time an editor gets a fancy redirect idea. 50.75.226.250 (talk) 14:53, 22 July 2022 (UTC)
I agree with Headbomb, a document is not necessarily contained in a journal, and it does not necessarily have a URL. Two CS1 templates that seem close are {{cite report}} and {{cite techreport}}. "Cite report" documentation says

[it] is used to create citations for reports by government departments, instrumentalities, operated companies, etc. Examples include: government printed reports which lack ISSN or ISBN numbers, and reports from major semi-governmental instrumentalities that are freely circulating and available for verification, but which lack a formal ISBN/ISSN publication process.

It isn't spelled out what it is about "cite report" that make it unsuitable for any report, published by anybody, that isn't in the domain of one of the other CS1 templates. If it's a paper report published by the Committee to Use Customary American Units of Measure (which I just made up) why would and somehow manages to satisfy WP:IRS, why couldn't you use "cite report"? Jc3s5h (talk) 15:02, 22 July 2022 (UTC)
Although all reports are documents, not all documents are reports (technical or otherwise). CS1 definition is narrow, but correct. Published government reports are generally easily available and are usually subject to higher scrutiny before and after publication, at least in countries with some semblance of working democratic institutions. There are specific reasons why government publishers may not care for an ISBN, ISSN or any identifier other than their own, or perhaps an id from a catalog of record which in the US is the Library of Congress. Like published court opinions, government reports are authoritative records, not just reliable ones. 172.254.222.178 (talk) 00:46, 23 July 2022 (UTC)
Can we just redirect {{cite document}} to something other than {{cite journal}} as journal appears to be the wrong target. Keith D (talk) 11:40, 23 July 2022 (UTC)
The nomenclature {{cite document}} is wrong, because it suggests that documents are self-contained, stand-alone publications. From the purview of the reader they rarely are, they are locations in another source. When they are published as stand-alone they fall (as a citable publication) under {{cite book}}. Perhaps the fit is not 100%, but it is good enough. We are discussing templates (forms) not custom-tailored solutions with a perfect match. Also, anyone is free to not use templates and structure a citation as they see fit. Or maybe a non-CS1 template can offer a better application. 68.173.78.83 (talk) 13:01, 23 July 2022 (UTC)
{{cite document}} has existed as a redirect to {{cite journal}} ever since it was created way back in 2010. If it were to be altered to redirect to another template, this would require a WP:RFD first. This would not be necessary if somebody were to alter it to be a full CS1 template in its own right. --Redrose64 🌹 (talk) 13:24, 23 July 2022 (UTC)
Been there, done that, got the bloodstained t-shirt: see Help talk:Citation Style 1/Archive 81#"Cite document" needs its own template. Redirecting to cite journal is illogical and unfriendly. I think the conclusion was broadly that it is less than ideal but it is a complete can of vipers [sic] and filed under "too difficult". --John Maynard Friedman (talk) 15:27, 23 July 2022 (UTC)
Where is the difficulty? As was suggested in the previous conversation (and similar, much older ones) there is no such thing as a "document" citation class in the real world. A "document citation" template is as untenable as an "article citation" or a "song citation". As for the misdirected redirect, anyone can redirect a template to a citation template. That does not make the underlying template responsible for any misalignment, or for the waste of time these recurring discussions are. 68.132.154.35 (talk) 15:44, 23 July 2022 (UTC)
So by that logic, the United States Declaration of Independence is a collective hallucination of millions of Americans since 1776. It is clearly not a journal. It is clearly not a web page. Nor a book. So Magna Carta is obviously a hoax! And Luther never wrote the Ninety-five Theses? Need I continue? --John Maynard Friedman (talk) 23:17, 23 July 2022 (UTC)
Ok, that is your interpretation, which is fine, but it has nothing to do with citing sources in general, or using structured citations in particular. 104.247.55.106 (talk) 13:03, 24 July 2022 (UTC)
The full publication information is on page 4: This article is a result of the research and subsequent book produced as part of the AHRC funded project ‘Controversies in Coal: Interment, Impoundment and Intrigue at Harworth Colliery (1913-1924) through the Centre for Hidden Histories at the University of Nottingham. The Project ran from September 2016 to April 2017. So find the book bib info and then you have several options: you could do a construction like: "{cite report}, published in {cite book}"; or you could put the two together in cite journal or cite encyclopedia, though the latter might be more appropriate since its effectively a collection of articles. (By the way, speaking of redirects, we really need a {cite collection} redirect and |collection= alias for encyclopedia, since that makes up a huge amount of (intended) use cases, which is not obvious at all from the documentation.) SamuelRiv (talk) 05:12, 24 July 2022 (UTC)
This does not seem correct. As quoted above, and in the article, this was not "originally published in", and neither is it a research report. It is an article resulting from both, published in an unrelated platform, the Academia website. The closest existing matching template, if one wishes to use CS1 templates, is still {{cite web}}. We cannot make up a probable imaginary provenance for it and retrofit into any other template. 204.19.162.34 (talk) 17:13, 24 July 2022 (UTC)
Imaginary provenance? The wording in the report is not precise -- to do diligence you would look up the book, and if it's an article in a collection then the above is the appropriate citation format. If a standalone article that later is published in a collection can't be cited in this way, that goes against the running consensus for why we allow preprints to be linked for journal citations and Google Books excerpts from eBooks to be linked for print books.
That being said, I looked up the book and in this particular case the book seems to be cited less than the paper. It must have been an institutional publication, because I've only found two citations to the book and some mentions on museum websites. Also for some reason it's missing on AHRC's projects page, though the project lead is there. So at the end of the day I couldn't find even a table of contents to verify if the book is the same as or contains the article, though they have the exact same title, author, and date. SamuelRiv (talk) 17:47, 24 July 2022 (UTC)
It seems you presume there was no due diligence related to the previous post? You would be wrong. You duplicated the steps taken to formulate the previous post, and all that is still irrelevant. The article is not a "report", "preprint", "paper" or whatever other label is wrongly assigned. It is based on the book (and the related research), according to the author. It is not a reprint or an extract from that book, and it should not be assigned to that source. It is content of a webpage in a website that includes similar items. Making it something that is not, will likely make it harder to find for readers who want to verify whatever wikitext it supports. 71.105.141.131 (talk) 22:02, 24 July 2022 (UTC)
As an experiment, I edited 50 of the articles that used the {{cite document}} redirect. I replaced the redirects with:
{{cite arxiv}}
{{cite book}}
{{cite citeseerx}}
{{cite conference}}
{{cite encyclopedia}}
{{cite journal}}
{{cite news}}
{{cite periodical}}
{{cite report}}
{{cite ssrn}}
{{cite tech report}}
{{cite thesis}}
{{cite web}}
By far the most common replacement was {{cite web}} (24×) followed by {{cite citeseerx}} and {{cite ssrn}} (7× each). It is unlikely that there is a simple 'automated fix' that converts a {{cite document}} redirect to some other, more correct template. There may be some small subsets of {{cite document}} use that are amenable to fixes-by-script but, to be fixed properly, most {{cite document}} redirects will need humans to do the fixing.
Trappist the monk (talk) 19:32, 24 July 2022 (UTC)

Generic titleEdit

Hello, can "Page Title" at the start of a |title= be added to the generic title list. Currently about 25 occurrences. Thanks, Keith D Keith D (talk) 23:04, 22 July 2022 (UTC)

Section ≠ chapterEdit

Is there a proper way to cite a particular section from a particular chapter from a particular part of a particular book, which is a particular supplement to a particular volume of a particular encyclopedia? Namely, there is a NASA publication called "Computers in Spaceflight: the NASA Experience", formally being "Volume: 18, Issue: Supplemet 3" of "Encyclopedia of Computer Science and Technology", but in fact it is a book and is also published on the Web, with separate webpages for each section. The structure is:

I need to cite that section, and preferably without losing the rest of the structure information (chapter, part, book title and its location within the encyclopedia). Can this be done? {{cite book}} doesn't even allow specifying |chapter= and |section= at the same time. — Mikhail Ryazanov (talk) 11:59, 26 July 2022 (UTC)

Since the chapters are sequential across the parts (of which there are only 3), and I don't think the parts would be published in separate print volumes, I'd suggest omit the part name and just do |title=(book title), |chapter=[Chapter 6: Distributing ...], and |at=(section title) (it appears the sections are each bold title, not each separate webpage). SamuelRiv (talk) 15:30, 26 July 2022 (UTC)
(Sections there are separate webpages, as can be seen within the "Part II" link above. Bold titles within each section webpage correspond to subsections – which are not important here.)
Using |at= looks strange:
and |issue= is not shown at all (without any warnings). — Mikhail Ryazanov (talk) 21:53, 26 July 2022 (UTC)
The pdf version at archive.org uses the term 'issue' 15 times; none of which refer to an issue of an encyclopedia. The term 'Encyclopedia' is used only once (in §Foreword):
The Editors have taken the unusual step of devoting an entire Supplement volume of the Encyclopedia to a single topic: "Computers in Spaceflight: the NASA Experience."
Comparing these google-books scans, Volume 17 Supplement 2 and Volume 18 Supplement 3, it seems obvious that Supplement n is merely a subtitle of the volume number.
Two options I think, If you are citing the html version, {{cite web}}:
{{cite web |title=Voyager - The flying computer center |website=NASA History |url=https://history.nasa.gov/computers/Ch6-2.html}}
"Voyager - The flying computer center". NASA History.
if you are citing the book-length pdf, cite it as a book and forget about the encyclopedia:
{{cite book |last=Tomayko |first=James E. |date=March 1988 |chapter=Distributed Computing On Board Voyager and Galileo §Voyager-The Flying Computer Center |title=Computers in Spaceflight: The NASA Experience |page=173 |id=NASA-CR-182505 |url=https://strives-uploads-prod.s3.us-gov-west-1.amazonaws.com/19880069935/19880069935.pdf?AWSAccessKeyId=AKIASEVSKC45ZTTM42XZ&Expires=1600217623&Signature=jw3KXryBL%2B0kfI4iPyx5Z8Eg%2B1w%3D |archive-url=https://web.archive.org/web/20200916005338/https://strives-uploads-prod.s3.us-gov-west-1.amazonaws.com/19880069935/19880069935.pdf?AWSAccessKeyId=AKIASEVSKC45ZTTM42XZ&Expires=1600217623&Signature=jw3KXryBL%2B0kfI4iPyx5Z8Eg%2B1w%3D |archive-date=2020-09-16}}
Tomayko, James E. (March 1988). "Distributed Computing On Board Voyager and Galileo §Voyager-The Flying Computer Center". Computers in Spaceflight: The NASA Experience (PDF). p. 173. NASA-CR-182505. Archived from the original (PDF) on 2020-09-16.
Trappist the monk (talk) 22:49, 26 July 2022 (UTC)
@Mikhail Ryazanov To add to Ttm, the encyclopedia can still be named in {{cite book}} per the |series= parameter (its usage in the template documentation is at "Complex usage showing effect of using volume parameter and lastauthoramp parameter (with volume and lastauthoramp)"). I agree that in this case "Supplement 3" belongs as part of the volume parameter. You should limit yourself to linking to either the pdf (from the NASA citations page or the one Ttm found) or the the web version, as both are free and readily accessible and you generally want to minimize ambiguity of which source was originally used. The |at= parameter is supposed to be for any kind of pinpoint (except page), so you would have to denote "sec.", "subsec.", "§", "para.", or whatever you need to do to (ideally) get as precise or as broad a pinpoint as is best for verification. The form Ttm shows is fine too. SamuelRiv (talk) 23:47, 26 July 2022 (UTC)
Well, since {{cite book}} also complains about external links in |section=, the best I could do was to use |section=chapter § section and |section-url=URL (instead of more logical |section=chapter § [URL|section]). But it would be much better if the template could handle |chapter= and |section= together, with corresponding |...-url=. Regarding "limit to linking to either the pdf or the the web version", the problem is that the PDF is really huge, so linking to a lightweight web page makes much more sense, but the web version doesn't have proper bibliographic information, thus it also makes sense to link to the NASA bibliographic page for the whole book. After all, {{cite journal}} allows multiple simultaneous links (doi, pubmed, jstor, ...) and wiki-linking the journal itself, and {{cite book}} also has |isbn= "linking" to the whole book record even when |url= is much more specific. — Mikhail Ryazanov (talk) 18:14, 28 July 2022 (UTC)
The approach is incorrect. Instead consider this: when a wikitext claim is made, it is hopefully supported by a source known to the editor. The only thing citations are concerned with is revealing that source to the reader in the least-complicated way possible. The location of bibliographic information is not entirely relevant, as what is cited in this case is not a bibliographic record, but the specific content and way(s) to find it. The reason behind multiple content identifiers is redundancy with as little clutter as possible. Classification/marketing identifiers such as ISBN or OCLC help in finding the content, not exposing it, and serve a different purpose. And do not forget that in many cases, online PDFs allow page-linking. 24.103.63.182 (talk) 12:22, 29 July 2022 (UTC)

infobox and citation tag translation across wiki'sEdit

Has there been any thought to making sure the various infoboxes and citation tags across different wiki's are the same? I've translated a few items from and to English using the translation tool and more often than not it makes a mess of the citation tags being translated (to a lesser extent, infoboxes). It's a pain in the butt to try and figure out how to fix the citation tags after doing a translation. Oaktree b (talk) 14:50, 26 July 2022 (UTC)

Infoboxen are out of scope here. What do you mean by citation tags? Did you mean to write 'citation templates'? What citation templates? What languages? We do have some crude 'translation' templates that are automatically subst'd to a more-or-less appropriate cs1|2 template. For example, the German template de:Vorlage:Literatur is a redirect ({{Literatur}}) to {{cite book/German}}. If you copy a {{Literatur}} template from de.wiki to en.wiki and save, User:AnomieBOT will subst that template with a 'translated' equivalent ({{citation}}). No guarantee that such translations are correct, they often aren't. For other available 'translators' see {{Non-English citation templates}}.
Trappist the monk (talk) 15:21, 26 July 2022 (UTC)
Such things can only be harmonised across all language Wikipedias if all can first agree on a common policy reagrding verifiability. German Wikipedia, for example, has nothing matching our {{BLP sources}} and {{Citation needed}} tags. Does that mean that they insist on sources for everything - or that they don't care if something is sourced or not? Looking at sections like de:Michael Schumacher#Kontroversen, I suspect the latter. --Redrose64 🌹 (talk) 18:38, 26 July 2022 (UTC)
I think the topic of how to write a citation is more or less separate from whether a citation is needed or not.
The citation templates in the English Wikipedia are based on English-language citation guides such as the Chicago Manual of Style and APA Style. I don't know how much citation customs in other languages resemble English-language customs. Jc3s5h (talk) 19:45, 26 July 2022 (UTC)

automatic live v. sandbox suggestions-module selection tweakEdit

Unlike all other modules in the cs1|2 module suite, the list of suggestions (Module:Citation/CS1/Suggestions or Module:Citation/CS1/Suggestions/sandbox) is only loaded when it is needed. Editor Tholme points out that the mechanism that decides whether to load ~Suggestions or ~Suggestions/sandbox needs improvement. That editor offered one solution that I knee-jerk reverted. Correctly, my revert was reverted and I have since improved on Editor Tholme's fix.

Trappist the monk (talk) 18:42, 30 July 2022 (UTC)

Thanks, your fix was off course a lot better :) Would it also be possible to move the sandbox i18n variable to Module:Citation/CS1/Configuration? This way we can import the Module:Citation/CS1 to other languages without doing any changes in this module. Tholme (talk) 19:45, 30 July 2022 (UTC)
I don't think that that is possible because Module:Citation/CS1/sandbox doesn't know to load the sandboxen submodules until it examines its own invoke. So to do what you suggest, Module:Citation/CS1/sandbox would need to load Module:Citation/CS1/Configuration to know that it needs to load Module:Citation/CS1/Configuration/sandbox. An alternative might be to do something like I have done at this edit to Module:Citation/CS1/sandbox. This requires all templates that invoke Module:Citation/CS1/sandbox to have a new parameter |SandboxPath=. At {{cite book/new}} the new parameter is |SandboxPath=/sandbox. At no.wiki the templates that invoke no:Modul:Citation/CS1/sandkasse are listed here.
Trappist the monk (talk) 22:24, 30 July 2022 (UTC)
Yes, that is off course a problem... Thanks for your suggestion with a new parameter, but I think this is more work than just updating sandbox = '/sandkasse' in Module:Citation/CS1. I do think it would be ok if this module checked for variable sandbox in Module:Citation/CS1/Configuration and not Module:Citation/CS1/Configuration/sandbox. There would be need to be a comment that the sandbox variable needs to be correct in the live module and not only in the sandbox. This would also mean that the sandbox module will need to load Module:Citation/CS1/Configuration first, and then afterward read all the other configuration from Module:Citation/CS1/Configuration/sandbox. I think this should work. Would you be ok with an implementation like this? Tholme (talk) 23:41, 30 July 2022 (UTC)
Or maybe load it from cfg['sandbox-subpage'] in Module:Documentation/config? It seems like most wikis uses this. Tholme (talk) 00:13, 31 July 2022 (UTC)
Interesting notion, that. But, at present, 187 MediaWiki wikis use Module:Citation/CS1 (see Module:Citation/CS1 (Q15403807)) but only 111 use Module:Documentation/config (see Module:Documentation/config (Q15506579)). Yeah, updating 12 templates at no.wiki (36 here) is a bit of extra work but that work only needs doing once – there are wikis out there that do not sandbox their cs1|2 module suite so no extra work for them ... I'm not so much in favor of borrowing code from outside of cs1|2 because that code is not beholden to us so can be modified to suit its primary purpose which might break cs1|2.
I've tweaked my prospective code a bit to protect against |SandboxPath= (parameter present without assigned value which returns an empty string; which would make sandbox module name same as live module name):
local sandbox = ((config.SandboxPath and '' ~= config.SandboxPath) and config.SandboxPath) or '/sandbox';
The way the prospective code is written, when a sandbox {{cite <whatever>}} template doesn't have |SandboxPath=, Module:Citation/CS1 will use whatever default sandbox name is provided in the module.
Trappist the monk (talk) 11:58, 31 July 2022 (UTC)
It occured to me that something like this might work:
local sandbox = frame:getParent():getTitle():match ('/.+$');
This returns whatever text follows that last / in the invoking template's name. But, there is no guarantee that whatever text follows the last / will be the 'sandbox'. For example, at en.wiki we have a series of (very) crude translator templates that have the form {{cite <whatever>/<country name>}} so the above will return /<country name> to the live module. Not a good idea...
Trappist the monk (talk) 16:07, 31 July 2022 (UTC) 16:20, 31 July 2022 (UTC) (strike)
Well that's not quite true. {{cite <whatever>/<country name>}} wraps a cs1|2 template and it is that template name that is returned by frame:getParent():getTitle() so if the wrapped template is a sandbox template then perhaps this mechanism will work. At en.wiki there is of course the {{cite <whatever>/new}} variant that we have used for a very long time in place of {{cite <whatever>/sandbox}}.
Trappist the monk (talk) 16:20, 31 July 2022 (UTC)
Thanks for your efforts. I think I will just stick with changing the sandbox variable in Module:Citation/CS1 after import. Tholme (talk) 21:39, 31 July 2022 (UTC)

ISBD referenceEdit

International Standard Bibliographic Description. May be used as an authoritative reference in discussion of structured citations.

68.132.154.35 (talk) 14:20, 31 July 2022 (UTC)

Protected edit request on 1 August 2022Edit

Please change the link ISBN (identifier) in {{cite book}} to ISBN. Kailash29792 (talk) 11:01, 1 August 2022 (UTC)

This would require a full RFC not just for ISBNs, but all identifiers. Headbomb {t · c · p · b} 13:18, 1 August 2022 (UTC)
I could be mistaken but I think Kailash29792 is asking for the wikilinks in the documentation to be changed from ISBN (identifier) to ISBN. Not sure why though as the identifier page is already a redirect to ISBN. Sideswipe9th (talk) 14:43, 1 August 2022 (UTC)
I interpreted the request to change the output of the template, not the documentation. The templates intentionally link through the redirect so that they can be filtered out in the Special:WhatLinksHere page for the article on ISBN. (Ditto all of the other identifiers.) Linking directly to the article in this way would break this filtering technique. It would require an RfC to overcome the past decisions on this topic. Imzadi 1979  17:06, 1 August 2022 (UTC)

PMC numbers have gone beyond 9300000Edit

See tecovirimat where a citation to pmc=9300470 is now showing an error. Please increase limit in the validation code. Thanks. Mike Turnbull (talk) 11:20, 1 August 2022 (UTC)

  Done Thanks, whoever fixed that. Mike Turnbull (talk) 13:34, 1 August 2022 (UTC)

Demeter's Manual of Parliamentary Law and ProcedureEdit

Whats the issue with this citation? Headbomb {t · c · p · b} 13:05, 1 August 2022 (UTC)

ALL citations seems to be broken.AManWithNoPlan (talk) 13:05, 1 August 2022 (UTC)
Maybe this? https://en.wikipedia.org/w/index.php?title=Module:Citation/CS1/Configuration&diff=1101715557&oldid=1096144888 AManWithNoPlan (talk) 13:12, 1 August 2022 (UTC)
I thought that, but if you see the section above the error preexists that edit. -- LCU ActivelyDisinterested transmissions °co-ords° 13:18, 1 August 2022 (UTC)

Whatever it was, it seems fixed now. Headbomb {t · c · p · b} 13:17, 1 August 2022 (UTC)

Well that was nasty, but the sea of red I was seeing across all articles using sfn is now gone. SandyGeorgia (Talk) 13:20, 1 August 2022 (UTC)
Sea of red still there for me, for example on Cycloheptatriene. Mike Turnbull (talk) 13:22, 1 August 2022 (UTC)
@Michael D. Turnbull: Purging the page resolved the issue. GoingBatty (talk) 13:25, 1 August 2022 (UTC)
GoingBatty Yes, thanks. Now you only have to purge 6 million+ other pages! (I jest). Mike Turnbull (talk) 13:26, 1 August 2022 (UTC)

Wikipedia:Wikipedia Signpost/2022-08-01/Tips and tricksEdit

Just a heads up that my little handy dandy guide on Citation bot finally got in the Signpost, as intended years ago. Feel free to leave a comment (or make suggestions for other guides on different topics). Headbomb {t · c · p · b} 20:04, 1 August 2022 (UTC)

allmusicEdit

I have added AllMusic to the generic name list; presently occurs in |author<n>= and |last<n>= parameters in ~740 articles.

{{cite web/new |author=AllMusic Review by [reviewer] |title=Title |url=//allmusic.com}}
AllMusic Review by [reviewer]. "Title". {{cite web}}: |author= has generic name (help)
Trappist the monk (talk) 17:01, 3 August 2022 (UTC)
@Trappist the monk: I have BattyBot working on the common cases, and manual work will be needed on the rest. GoingBatty (talk) 15:06, 4 August 2022 (UTC)

url-status parameter invalidEdit

The instructions indicate the use of |url-status=deviated, and when the original URL has been usurped for the purposes of spam, advertising, or is otherwise unsuitable, setting |url-status=unfit or |url-status=usurped. And, at Category:CS1 errors: invalid parameter value it says that these values are valid. However, at Category:CS1 maint: unfit URL it does not accept these values. Can this be fixed so that green hidden "One or more {{cite news}} templates have maintenance messages" for their use can be avoided? --Bejnar (talk) 17:48, 3 August 2022 (UTC)

See also Category talk:CS1 maint: unfit URL which hasn't been addressed. --Bejnar (talk) 18:11, 3 August 2022 (UTC)
It is not clear to me what it is that you are saying. |url-status=deviated, |url-status=unfit, and |url-status=usurped are all valid:
{{cite book |title=Title |url=//example.com |archive-url=//archive.org |archive-date=2022-08-03 |url-status=deviated}}
Title. Archived from the original on 2022-08-03. – the rendered citation links to the url assigned to |url= as a secondary link
{{cite book |title=Title |url=//example.com |archive-url=//archive.org |archive-date=2022-08-03 |url-status=unfit}}
Title. Archived from the original on 2022-08-03.{{cite book}}: CS1 maint: unfit URL (link) – the rendered citation does not link to the url assigned to |url=
{{cite book |title=Title |url=//example.com |archive-url=//archive.org |archive-date=2022-08-03 |url-status=usurped}}
Title. Archived from the original on 2022-08-03.{{cite book}}: CS1 maint: unfit URL (link) – the rendered citation does not link to the url assigned to |url=
Were these parameter values invalid, cs1|2 would return an error message:
{{cite book |title=Title |url=//example.com |archive-url=//archive.org |archive-date=2022-08-03 |url-status=USURPED}}
Title. Archived from the original on 2022-08-03. {{cite book}}: Invalid |url-status=USURPED (help) – all keywords used in cs1|2 templates must be lowercase
What do you mean by: at Category:CS1 maint: unfit URL it does not accept these values? There are 21,195 articles in that category so something must be happening.
Trappist the monk (talk) 18:27, 3 August 2022 (UTC)
@Trappist the monk: See, for example, this citation:

Masi, Alessandria (14 December 2015). "Saudi Coalition, Houthi Rebels Intensify Attacks In Yemen Ahead Of Proposed Ceasefire". International Business Times. Archived from the original on 4 October 2016.{{cite news}}: CS1 maint: unfit URL (link) [Cite news|last=Masi |first=Alessandria |date=14 December 2015 |title=Saudi Coalition, Houthi Rebels Intensify Attacks In Yemen Ahead Of Proposed Ceasefire |newspaper=International Business Times |url=http://www.ibtimes.com/saudi-coalition-houthi-rebels-intensify-attacks-yemen-ahead-proposed-ceasefire-2223830 |archive-url=https://web.archive.org/web/20161004172106/http://www.ibtimes.com/saudi-coalition-houthi-rebels-intensify-attacks-yemen-ahead-proposed-ceasefire-2223830 |archive-date=4 October 2016 |url-status=unfit ] at December 2015 Taiz missile attack which yields an error. I marked it as unfit, because the article is occluded at the original website. If "unfit" needs to be checked, then how is the error cleared? --Bejnar (talk) 18:37, 3 August 2022 (UTC)

Maintenance messages are not error messages.
For me, the url http://www.ibtimes.com/saudi-coalition-houthi-rebels-intensify-attacks-yemen-ahead-proposed-ceasefire-2223830 is not unfit. There is no mechanism to mark urls that are geographically limited – is that your issue? I think that we have discussed this before; if we have, we did not decide to do anything about geographically limited urls. Marking a geographically limited url as unfit does not seem to me to be a good idea; the url is fit to be viewed – it isn't spam or porn or a phishing site, etc – so should not be labeled as unfit.
Trappist the monk (talk) 19:04, 3 August 2022 (UTC)
@Trappist the monk: It was not geographically limited. When access was attempted this morning, the article was occluded by a "please subscribe" overlay with no "close" function. Now when I access it, it displays properly, and the url-status can be reset to live. But that was just an example, my question still is, as modified, if maintenance is required for those 21 thousand articles, how is it to be accomplished? That is, what is the review procedure that affirms a valid use of the value "unfit" for that parameter? --Bejnar (talk) 22:19, 3 August 2022 (UTC)
If the site requires a subscription then |url-access=subscription is the correct parameter to use.
There is no required action for most maintenance messages. Alas, some maintenance categories (Category:CS1 maint: archived copy as title (201,333),‎ Category:CS1 maint: uses authors parameter‎ (13,800), Category:CS1 maint: multiple names: authors list (38,107)) should emit error messages but those categories apply to so many articles that making them error messages would cause a torches-and-pitchforks uprising; we've seen that kind of uprising before and it isn't pretty.
Trappist the monk (talk) 23:09, 3 August 2022 (UTC)

url-status = deviatedEdit

What is the recommended action when a cite with |url-status=deviated is converted to a dead link. Normally it is flipped to |url-status=dead. -- GreenC 18:08, 3 August 2022 (UTC)

If a deviated url becomes dead (returns 404) then |url-status=deviated should be removed. No need to replace it with |url-status=dead because cs1|2 assumes that the url in |url= is dead when |archive-url= has an assigned url.
Trappist the monk (talk) 18:32, 3 August 2022 (UTC)

pmid limitEdit

"If the value is correct and larger than the currently configured limit of 35900000, please report this at Help talk:Citation Style 1, so that the limit can be updated."

https://pubmed.ncbi.nlm.nih.gov/35918775/ Aethyta (talk) 19:47, 3 August 2022 (UTC)

Date bugEdit

Minor thing, but the template complains if the date has a leading zero – date=09 August 2021GhostInTheMachine talk to me 10:01, 4 August 2022 (UTC)

Solution: Remove the leading zero, per MOS:DATE/MOS:DATESNO. Headbomb {t · c · p · b} 10:53, 4 August 2022 (UTC)

Template: Cite Web change (might apply to others to)Edit

I notice {{Cite web}} has a field |trans-title. For usability could this field be renamed translated-title or en-title? The use of the word Trans is confusing because of its use in the gender-spectrum and it isn't entirely clear that this field exists or what it is used for. >> Lil-unique1 (talk) — 21:31, 8 August 2022 (UTC)

Support non-controversial parameter rename. Suggest translated-title, as this will be easier to ... translate/apply in non-English wiki CS1/2 iterations. Also en- could be mistaken for language code rather than abbreviation. 68.132.154.35 (talk) 23:40, 8 August 2022 (UTC)

ref=harv in VE template parameter descriptionEdit

Description of ref parameter in VE for {{Cite journal}} (and possibly other Cite variants) give 'harv' as valid value, while it's deprecated (according to the Template:Cite_journal#Anchor) and causes display of Invalid |ref=harv error.

Edit: This concerns also the TemplateData description.

Discovered while editing:

Bibliography

https://en.wikipedia.org/wiki/Analytical_Engine

(Side note: Template refbegin/end under VE should be searchable, finding one specific citation block can be tedious). MarMi wiki (talk) 21:48, 9 August 2022 (UTC)

@MarMi wiki: It's valid, but it's not useful because simulating |ref=harv is the default action when the |ref= parameter is blank or omitted. --Redrose64 🌹 (talk) 23:47, 9 August 2022 (UTC)
Doesn't look like valid value:
Rojas, Raul (January–March 2021). "The Computer Programs of Charles Babbage". IEEE Annals of the History of Computing. 43 (1): 6–18. doi:10.1109/MAHC.2020.3045717. {{cite journal}}: Invalid |ref=harv (help) MarMi wiki (talk) 19:10, 11 August 2022 (UTC)
The wording of the message is invalid, not the value. It should say something like "superfluous (harv is default)". It is properly a maintenance message, not an error, and of a fairly low priority. Imo, unnecessary. If it is documented (it is), the module could well stay silent/ignore this. 108.176.2.154 (talk) 12:36, 12 August 2022 (UTC)
The reason and consensus it is an error now was discussed at Help talk:Citation Style 1/Archive 76#ref = harv. Izno (talk) 02:19, 15 August 2022 (UTC)
This means the TemplateData needs to be adjusted/fixed. Izno (talk) 03:22, 15 August 2022 (UTC)

Gale problemEdit

While trying to improve the article Morris Bishop, I am, for the first time, using Gale online (more specifically, via the Wikipedia library). And having a slight problem of referencing.

In Wikipedia Library, I select Gale, and then: "Gale in Context: Biography" | search for "Morris Bishop" | click on "Bishop, Morris" | "Biographies (1): Morris Bishop; From:Gale Literature: Contemporary Authors" -- and I arrive at the page I'm now using and wanting to cite. Briefly, this one page is laid out: "About this person | Personal information | Career | Awards | Works | Sidelights", etc. (Contrary to what one might expect, "Sidelights" is by far the longest and most interesting section.) Its URL is, for me, a monstrous thing starting "https://go-gale-com.wikipedialibrary.idm.oclc.org": obviously unusable for a reference to an article. Near the foot of the page, we read "Gale Document Number: GALE|H1000008881". I add "{{Gale|H1000008881}}" to my reference. Mediawiki interprets the business part of that as "<a rel="nofollow" class="external text" href="https://go.gale.com/ps/anonymous?id=GALE%7CH1000008881">H1000008881</a>".

Using a different browser (with no Wikipedia or Library log-in) to click on that link, what I get is truncated: "This is a preview. Get the full text through your school or public library." This is to be expected. But what I don't expect is truncation at both ends: the humdrum stuff near the top of the page, above "Sidelights", is all cut. It looks as if I've incompetently given the wrong reference (not to "Morris Bishop" but instead to "Morris Bishop: Sidelights") or am providing a bogus "source".

Of course Wikipedia can't control the method by which Gale chops off material that's not free to everyone for the asking. But I wonder if apparently mistaken links is a known irritation, one for which people brighter (or more caffeinated) than me can make good suggestions. (Template:Gale/doc is terse and says nothing relevant.) -- Hoary (talk) 23:15, 12 August 2022 (UTC)

This is a complex issue and can be frustrating. I believe that it is probably caused by the nature of Gale's "In Context" products, which basically merge information from different databases. Also, another reason is that many providers store (or extract-on-the-fly) abstracts/previews in different properties. In this case, the information you want lives in Gale literature: Contemporary Authors, which is then presented via Gale: Biographies In Context (Gale product id "BIC") per your choice. This is a paywalled resource. The free preview however is stored in Gale: Academic OneFile (Gale product id "AONE"). Notice that the citation helper at the bottom of the Academic OneFile preview does not include a URL, unlike the full resource at Biographies in Context. If you actually log in to Gale, you can click at "Access Through Your Library" at the end of the preview text. This leads nowhere (404 error) because the actual full info is in a different database that is accessed differently. If you compare the Biography in Context URL with the 404 page at Academic OneFile, you will notice the query term &prodId (product ID) is "BIC" for your full reference, but "AONE" for the target of the preview's bad link to the full source.
Not much can be done. If you are technically inclined you can attempt to construct a custom OpenURL, or perhaps use other ways. Gale has relevant documents at "Gale Support→Tools→Technical Documents" (e.g. "How do I construct an OpenURL for Gale content?") with a bunch of info on accessibility and discovery. But this is also a paywalled resource. 68.132.154.35 (talk) 17:18, 13 August 2022 (UTC)
Although you provide no simple solution (and of course don't claim to), IP, what you write explains a lot. Many thanks. Come September, I may be able to visit an actual university library and there plug myself into Gale in a way that's perhaps more expensive (to taxpayers, if not to me) -- and I could even just look in, and cite, the relevant codex among the hundreds making up Contemporary Authors. -- Hoary (talk) 02:05, 14 August 2022 (UTC)