Help talk:CS1 errors/Archive 3

Bibcode
Why there is 2010Nature....32..450P bibcode for doi 10.1038/nature09049? It is recognized as invalid by CS1 although cited on many places (Google search displays it, but sometimes there is 2010Natur.465..450S)... --Obsuser (talk) 20:46, 25 November 2016 (UTC)
 * Surely you aren't saying that Wikipedia is responsible for errors in cites at sites that are not controlled by Wikipedia? The first bibcode, 2010Nature....32..450P is clearly longer than the 19 characters allowed.  Click on the link and adsabs.harvard.edu also thinks that bibcode is not good:
 * Yet the second appears to be correct and also appears to match the doi:
 * So what is the question for us?
 * —Trappist the monk (talk) 21:11, 25 November 2016 (UTC)
 * No, just wasn't sure it is incorrect. Thanks.--Obsuser (talk) 22:10, 25 November 2016 (UTC)
 * —Trappist the monk (talk) 21:11, 25 November 2016 (UTC)
 * No, just wasn't sure it is incorrect. Thanks.--Obsuser (talk) 22:10, 25 November 2016 (UTC)

How do you confuse Needful with POV?
re: Help:CS1 errors...

It may be pov to you, but I prefer to consider it a breath of useful truth. THE LENGTH AND VERBOSITY of that Help page is a flunk-able offense to NEEDS of EDITORS donating time to improve articles, such as my fixing OTHER's Dates here. SO TELL ME, why should I be happy it takes five minutes to find the text on how to fix a simple and stupid problem because it has been buried on a page that should never have been assembled. (Note: MOST other types of cites errors SEEN here  are tag names, often misspellings, which are specific to the cite template used. Many of those could be combined like publication, journal, etc. if some elitist group wasn't trying to force this or that form of citation styles down our throats.) If you lot want a comprehensive page no one will ever read completely, then you've managed the goal.

If you want something useful and easy to maintain, then split that page into sections including subpages (by sections with apropos stand alone introductions nested in noinclude blocks and directly linkable), and have your errors point the sole subpage which is 'on topic'... NOT all over creation. In short, STOP WASTING MY TIME with such needless verbosity. I don't work for the Government. // Fra nkB 17:43, 1 December 2016 (UTC)


 * Calling people 'idiots' or the guideline "overblown verbose verbiage" is blatant POV trolling. The templates throw specific errors. If you're curious about why you get that error, the TOC is there. You might get a shortcut to the date table in the lead section if you phrase things neutrally, but you won't achieve much success with the attitude you currently have. Headbomb {talk / contribs / physics / books} 17:48, 1 December 2016 (UTC)
 * The page is long because editors make a wide variety of mistakes in using these complex templates. It has plenty of anchors, it is divided into sections with a TOC, and having it as one long page makes it so that the Find feature of your web browser works well once you have the page loaded. Also, errors in citations within articles always have a "help" link that takes you right to the appropriate section. (For example, in the page linked above, the help link goes to Help:CS1 errors). In that section, you can see "The access date (in |access-date=) is checked to ensure that it contains a full date (day, month, and year)...."


 * Do you have specific, constructive suggestions about how we could improve the content or navigability of this page while preserving the advantages described above? We make constructive improvements to the page quite often. Also, starting a discussion on the Talk page corresponding to the page you have questions about is the usual way to get those questions answers. Idiotically yours, – Jonesey95 (talk) 18:03, 1 December 2016 (UTC)


 * It might be useful to have two sections with the error listings (as appropriate) below: A "Common" and an "uncommon". And we can use these based on the current numbers of articles in the categories. This would probably help to pick out the problem when visiting the full page. --Izno (talk) 18:37, 1 December 2016 (UTC)


 * Perhaps some of the problem is your error handling doesn't get to the sections for which you have provided anchors. Further, there is no handling for an equivalent term if one cite template uses an variant found in another. This occasion gives a misspelling handling, in effect, so my suggestion there is direct such occasions to the section directly showing that compressed table of legal terms. That way we need not suffer wading through verbosity when we merely need a mental nudge (clue). Fra nkB


 * The only way I know to get to the page is from the help links in article error messages. These help links should take you to a section of the page that is relevant to that error message. Are there some help links that just take you to the top of the page, or are there other ways to get to the full page? – Jonesey95 (talk) 21:58, 1 December 2016 (UTC)
 * As best as I can now recall, this did occur with date errors. Put me at page top, so I had to get vexed trying to wade down looking for the key data. The general problem with the error messages should not be lost by you guys tending this knitting... looking up such is a distraction to what the editor is doing, trying to close, and keep straight. An interruption. The easier and quicker you can make it, the better for all. Hence my subpages suggestion... your handling goes directly to the page designed to show a quick answer (using noinclude blocking-it can display viewed quite directly than the version that gets included as a section. The text is maintained in just one place. )  Fra nkB  02:36, 2 December 2016 (UTC)
 * For the record, the text in Help:CS1 errors is maintained in just one place. The text is reused in the various error category pages.
 * —Trappist the monk (talk) 14:30, 2 December 2016 (UTC)
 * re: Do you have specific, constructive suggestions about how we could improve the content or navigability of this page while preserving the advantages described above?
 * Yes, have the error messages help links go to the content-error-type keyed coverage. In my experience, the date formats are the common problem (though oddly, the system as it stands now prefers the same format as my lifelong date preference-post naval service, at least! ... Until a few years ago. Now find YYYY MM DD more useful as a code, since is numerically unique each year.), ... followed by this or that template not taking equivalent terms that are legacy equivalents not used in the originals ( publication = journal = book, for example).
 * This as I'm sure you know could be handled either by a switch-default or Wikimarkup language's  including passing each alternative to a common parameter define in a intermediate (caller) template.
 * I used a pass template often and the subpage technique often in a Wikibook, if you want an example, I can re-visit and locate examples. // Fra nkB 02:36, 2 December 2016 (UTC)


 * Sorry, I simply do not understand the wandering, general statements that you are making above. It is a challenge to understand your written English; perhaps it is not your first language?
 * To respond about dates: many date formats are supported; all of them are in the Wikipedia Manual of Style.
 * As for help links: my experience with the help links is that they all link to specific sections that explain how to fix the problem.
 * If you could provide a specific citation number within a specific version of a specific article, and then explain what happens when you try to find a way to fix that citation, that would help. – Jonesey95 (talk) 03:43, 2 December 2016 (UTC)
 * I do not understand much of this editor's writing either. Editor : The burden is on you to show evidence that there is some cs1 error message that does not link to the appropriate section of Help:CS1 errors.  Without that evidence, nothing can be done to fix the problem.
 * —Trappist the monk (talk) 14:30, 2 December 2016 (UTC)
 * I have made a first edit to separate the common errors from the uncommon, using roughly ~10k articles in the corresponding categories as a threshold. Perhaps that will satisfy ? --Izno (talk) 13:27, 2 December 2016 (UTC)
 * For me, without evidence that there was something fundamentally wrong with the alphabetic organization of Help:CS1 errors, the changes you have made should not have been made. I think that this is a case of it-ain't-broke-so-don't-fix-it.
 * —Trappist the monk (talk) 14:30, 2 December 2016 (UTC)
 * I agree with Trappist here. The merging of 'identifier errors' does make some sense, although I didn't really see a need for it. Separating in common errors/other errors is unnatural and confusing though. Headbomb {talk / contribs / physics / books} 14:33, 2 December 2016 (UTC)
 * Why? This resolve the editor's issue, and I would be remiss if I did not also say that I've gotten lost looking for the common errors. It's also good UX to guide users to the more common issues first before providing an index of all issues. --Izno (talk) 15:15, 2 December 2016 (UTC)
 * A list of section links from the lede to the most common errors? That would seem preferable to rearranging the content and, presumably, the list of most common is variable so its easier to edit a simple list than it is to move whole sections.
 * —Trappist the monk (talk) 15:43, 2 December 2016 (UTC)
 * It may be variable at some point in the future, but the ones I picked out have over 10k pages affected (and some up to 30k). No-one is going to work through that backlog within the next year, if not larger time period. I think we can safely say those are the most common now and then re-evaluate later. list of section links from the lede to the most common errors Which would duplicate the TOC unnecessarily (if we just organized our content sanely!). --Izno (talk) 16:48, 2 December 2016 (UTC)
 * It too prefer summarizing in the lead rather than re-organizing the sections. Headbomb {talk / contribs / physics / books} 17:25, 2 December 2016 (UTC)
 * I have undone Editor Izno's changes because somewhere among those changes the markup that allows for text reuse by the various error categories was broken. The various categories should only display the text from Help:CS1 errors appropriate to the category's error message.
 * —Trappist the monk (talk) 14:49, 2 December 2016 (UTC)
 * I haven't looked into how Help:LST works, so I'm not going to troubleshoot. And especially since there seems to be some dissension about the change... --Izno (talk) 15:15, 2 December 2016 (UTC)

Thanks for undent. I'm afraid I've tended to avoid these meandering discussions since 2009, but for a couple on the commons. To be clear... MY ISSUE guys, is the CS errors page dumps you into a page that is way to large and hard to navigate. I'm often quite happy enough digging out cites and clear things such as cn tags others let go a long time, but the demands and differences between one template and another is often stupid to someone schooled in engineering vice concentrations in the liberal arts. I couldn't care less at which citation style is used on any article so long as others can follow along. But the larger point is, if you want a help page, write a help page. If you want an error message handler, which is what triggered we are discussing, write a KISS principle compliant helpful terse para for each type of needed error message, but keep its coverage short, (ideally less than a laptops screen in length--unless a table)... Don't link into a page which is written as comprehensive help and mostly off point. Link that, but as last thing! Another common issue is name mismatched, hence the aliasing of nested wikimarkup parameters. If you want a editor help link, give brief concise help with a link to a longer overblown concise guide there. In short, a page that long is one which wastes anyone's time 99% of the time. People running into cites error messages mostly just need just a spelling/correct-label nugget--not to search for a needle in a haystack. My suggestion was using sub and sub-subpages to directly show that brief message with customized includeonly, noinclude portions similar to the Template help system's /doc pages. the sub-subpages might have the 'quick lookup' of a legal/illegal table. Assuming your error detection points to the sub-page, the line which includes the sub-subpage/table could be nested in a noinclude block up high in the text to display when directly viewed. In other words, arranged for quick re-checks by editors, with leads elsewhere for newbies, so the terse important data first. (Upside down from now) Elsewhere, Text to display when included (such as a section of the current CS1_errors page) would permit total rearrangement of the data in a larger text context... so two occurrences of quick aids such as the dates table. How you handle section edit links depends on how you write and nest &  psuedo-templates... If the section titles are in the subpage, one can click the normal section-edit link to navigate to the section's subpage & edit the section. Since this is Wikimedia/Help namespace, one can assume (hopefully) only knowledgeable editors are editing. A bit awkward, but it's high time something on the website was organized to be mindful of editor's time and treat it as valuable-and refreshingly brief! // Fra nkB 00:04, 6 December 2016 (UTC) (Oh, and friend, sorry if you find my organization and interjections in responses to be less understandable than native English. Alas, for that thesis, I grew up near Pittsburgh and my college board score in English was 790 of 800 back in the day. Suspect that came from reading an average of about 5 books a night from the fourth grade on. I'll try to do better-with the organization, at least. Just for you. My English is far better than my Russian, German, Italian, Latin, and French. Never did have an ear for romance languages, so never bothered trying to learn Spanish when I had time, though now I wish I had or could!) Fra nkB

Clearly a date problem
Above, requested proof that something was fishy in Denmark. This section's verbosity clearly notes a date in the DD MMM YYYY format, yet ACCESS-DATE consistently fails to accept such, or at the least, fails on a date with a single number before the month. See the first cite herein, diff.
 * Upon further examination, accessdate or access-date formats probably aren't the error message generation and detection problem &mdash; after saving that page to document the problem here, I next substituted a whole series of 'accessdate=' variations, most I suspect were quite legal since they showed up properly in the output of the citation text string,... and got identical error messages.

* I CONCLUDE it is the missing day field in the date after recollecting a couple of citations wherein I was forced to 'gimmack' that date parameter in a couple edits last week... using similar historical sources. Fra nkB
 * That leads to the suspicion and conclusion * that the issue is the date= test is complaining that a works' date (i.e. issues by month, in this case September 1899) are missing information. Sorry, many older docs don't give a day date, so the detection for these and the date=year forms are both false positives, the MOS be damned.  The world is what it is, not what the MOS wants. I shouldn't have to commit intellectual perjury to satisfy some blindly idealistic format in a date when such is not available. //Fra nkB  03:41, 6 December 2016 (UTC)
 * WP:MOSDATE does not allow for the comma, and accessdate take a complete and either current or past date, not the future. --Izno (talk) 03:56, 6 December 2016 (UTC)
 * What Editor Izno said.
 * But, your reply doesn't answer my request. You wrote that the error handling doesn't get to the sections for which [we] have provided anchors. That is the claim that I have asked you to prove.
 * —Trappist the monk (talk) 04:04, 6 December 2016 (UTC)
 * , if you click on the "help" link in the error message, you will be taken to section of the Help page that includes a helpful table listing dates that are out of compliance with MOS and how to fix them. The case of "September, 1899" that you linked to above is listed as "Comma in month and year", with an incorrect format followed by the correct format. There is a similar example for "Access date in future". I'm not sure what "gimmack" means (it wasn't on my SAT exam), but reading the help page should help you resolve any of these date errors. If there is an example missing, you are welcome to add it to the table or inquire here.
 * , if you click on the "help" link in the error message, you will be taken to section of the Help page that includes a helpful table listing dates that are out of compliance with MOS and how to fix them. The case of "September, 1899" that you linked to above is listed as "Comma in month and year", with an incorrect format followed by the correct format. There is a similar example for "Access date in future". I'm not sure what "gimmack" means (it wasn't on my SAT exam), but reading the help page should help you resolve any of these date errors. If there is an example missing, you are welcome to add it to the table or inquire here.


 * As for writing comprehensible English, when I am advising writers and copy editors, I always suggest reading prose aloud. It is often possible to locate errors that way that your eyes skip over during silent reading. – Jonesey95 (talk) 04:06, 6 December 2016 (UTC)
 * This first, since it's a LOL event, to me--I have an eyesight problem just closing edits half the time. My visual acuity can just flat out disappear in the span of 5 minutes, so I'm often pushing to finish proofing an edit, struggling to see it. Then again, not so sure that 'Out Loud' technique will go over too well with the Mrs. watching her news programs on the next couch. Then, again, on closing an update,... there are currently 17 windows open on this desktop, a programmer's editor with 19 tabs, a PDF reader with four open references, at least six Windows of multiple tab browsers which have 'open edits' for parallel updates, and most of the rest being reference browsers with such histories as the one with September, 1899 as a date. Four Windows have Maps in one stage or another preparatory to upload in support of one of those edits. So LOL, I edit when I have time, and often work parallel topics integrating pages, esp. in industrial history.  But I do my best, time and failing eyesight permitting!
 * re: :::... if you click on the "help" link in the error message, you will be taken to section of the Help page that includes a helpful table listing dates that are out of compliance with MOS and how to fix them.... As helpful as that seems, that's the link which started this whole dance. SO I MADE my edit to add an anchor to link the table, not the preamble. My last on this topic is since news organizations and various professional Journals use the date formats with embedded Commas as part of issue dates, the MOS is wrong to force a wrong date/quote in the  case. Be well, all. I've gotten my two year fix of Wikipolitics and likely won't be back! // Fra nkB 01:03, 7 December 2016 (UTC)
 * Until I read your last sentence, I was going to suggest taking up MOS objections at WT:MOSDATE, but having spent some time there, I can't recommend highly enough that you stay away. It's a challenging place to visit, let alone stay for a while. – Jonesey95 (talk) 06:44, 7 December 2016 (UTC)
 * Now we're cooking! Nice to see someone get it! Thanks & LOL with a giggle! Fra nkB  20:28, 7 December 2016 (UTC)

URLs in non-Latin scripts
Why are URLs in non-Latin scripts not allowed? Bever (talk) 23:55, 5 December 2016 (UTC)
 * One conversation in one place please.
 * —Trappist the monk (talk) 00:01, 6 December 2016 (UTC)

dual month issue
The template is throwing an error on the following ref. The citation is accurate. What is the correct thing to do here?

-- Jytdog (talk) 04:13, 18 March 2017 (UTC)


 * Use an endash, not a slash. Headbomb {talk / contribs / physics / books} 06:11, 18 March 2017 (UTC)
 * let's see ...

yep! Thanks! Jytdog (talk) 08:46, 18 March 2017 (UTC)
 * Your reference is now producing a Category:CS1 maint: Date format maintenance message which apparently indicates yours was a hyphen? Weird. Like so uses a correct n-dash: --Izno (talk) 10:27, 18 March 2017 (UTC)

Date with range in both season and year
I'm trying to cite an issue of a journal whose date is listed as "Autumn/Winter 1993/94". But when I try  or   I get an error message. Both  and   work, but it doesn't seem to let me have ranges for both. Am I missing something? Umimmak (talk) 20:19, 19 April 2017 (UTC)
 * This was recently discussed at Help talk:Citation Style 1/Archive 31. --Izno (talk) 20:31, 19 April 2017 (UTC)
 * Ah thanks for finding that . Well that's unfortunate that there's no way to accurately cite the issue's date, but also good to know that I'm not the only one with this problem or that I wasn't missing anything obvious l. Umimmak (talk) 20:58, 19 April 2017 (UTC)
 * We are somewhat at the whim of publishers here, and they sometimes do strange things. If I were putting this date in a citation, I'd pick a publication year (presumably 1993) to put in the date field, then put the issue date in issue. – Jonesey95 (talk) 22:06, 19 April 2017 (UTC)
 * Well if you're putting the seasons in the issue, you can have a year-range in the date, like so: ? (or for issue did you mean something like 2 Autumn–3 Winter?) I think I'll just have Autumn–Winter 1993, to follow what did in Pioneer Helmet as per the earlier discussion. Umimmak (talk) 22:52, 19 April 2017 (UTC)

Bibcode error
Bibcode 2002A&A...393..897 is valid, but generates an error in the citation template because it is the wrong length. Praemonitus (talk) 19:02, 24 April 2017 (UTC)
 * That's because it is not valid. The actual bibcode is . Yours is missing an R at the end. Headbomb {t · c · p · b} 19:23, 24 April 2017 (UTC)


 * Just because a malformed bibcode 'works' does not mean that it is 'correct' or 'valid'.
 * —Trappist the monk (talk) 19:24, 24 April 2017 (UTC)
 * Got it. Thanks. Praemonitus (talk) 19:39, 24 April 2017 (UTC)

no title in a
I'm getting this error in a number of situations associated with historic newspapers, which don't follow the modern convention of putting headlines onto all articles, particularly snippets of local news (major stories do have headlines). It also occurs when you cite a page as a whole (rather than a separate article). I don't think titles should be expected in cite news when the original source does not have a title. Kerry (talk) 04:17, 15 May 2017 (UTC)
 * There has been some previous discussion but nothing conclusive enough to act upon.


 * Discussions related to changing how cs1|2 works is best done at Help talk:Citation Style 1.
 * —Trappist the monk (talk) 10:11, 15 May 2017 (UTC)

Spaces are counted
In the last section of the article, it says "Space characters between the assignment operator (the '=' sign) and the parameter value are not counted." This is confusing... do you mean spaces between the "=" and the very first character (ie "title= Title causing error" with a space after '=' vs "title=Title causing error" with no space after '=") or spaces in the line of text itself? It seems to imply the second (as in spaces among the value) and this is incorrect. Thanks. —МандичкаYO 😜 02:11, 20 June 2017 (UTC)
 * Module:Citation/CS1 accepts only named parameters. Editors may, and do, write stuff like this (especially when writing these templates in vertical format so that all of the assignment operators and parameter values are vertically aligned all pretty-like):

| title    =     Wind in the willows
 * MediaWiki trims leading and trailing whitespace from named parameter values so what the module gets is more like this:

|title=Wind in the willows
 * The module begins counting at 'W' because it does not and cannot know how many space characters are present in the wikitext. So, for my example, the five space characters between   and   are not included in the count reported in the error message.  Space characters used to separate words in a parameter value are counted because they do not lie 'between the assignment operator ... and the parameter value' but rather, lie within the parameter value.
 * —Trappist the monk (talk) 03:14, 20 June 2017 (UTC)

Years and Classical publications from Antiquity
This seems to throw an error when using dates of Classics

















It appears to need a three digit absolute magnitude value or greater. This is clearly an error in the processor. -- 70.51.46.15 (talk) 06:57, 11 October 2017 (UTC)
 * Duplicate of a post at Help_talk:Citation_Style_1. Answered there.
 * —Trappist the monk (talk) 11:23, 11 October 2017 (UTC)

ISBN check sum error
I'm getting an error on the ISBN-13 value for a source, even though I copied it directly from the Google scan. Praemonitus (talk) 16:15, 4 January 2018 (UTC)
 * I corrected an issue where you were using dashes rather than hyphens. The checksum error remains. It's probably the case the book was printed with a bad checksum. --Izno (talk) 16:31, 4 January 2018 (UTC)
 * Do not use the minus sign (U+2212); use hyphens:
 * Even so, that ISBN seems to have been misprinted in the book. If you cannot find the real and true ISBN for the book, add yes.
 * —Trappist the monk (talk) 16:38, 4 January 2018 (UTC)
 * Yes, I first tried it with dashes, then with no dashes, then with &amp;minus; to see if it would make a difference. Thanks for the ignore option. Praemonitus (talk) 18:37, 4 January 2018 (UTC)
 * —Trappist the monk (talk) 16:38, 4 January 2018 (UTC)
 * Yes, I first tried it with dashes, then with no dashes, then with &amp;minus; to see if it would make a difference. Thanks for the ignore option. Praemonitus (talk) 18:37, 4 January 2018 (UTC)

Overzealous software
The bot flagged the access date in a reference I inserted today as impossible because I'm in a time zone ahead of the bot (GMT +7) and it is already June 1 here (but still May 31 in California). Software needs to be tweaked to prevent this presumptuous type of flagging -- the software can already detect location so it shouldn't be hard to link that info with a time zone calculation.Martindo (talk) 02:27, 1 June 2018 (UTC)
 * Is it this template from The Ugly American (film):
 * If so, the error is because of 2015--31 which should read 2015-01-31. I think that yours is the first complaint with regard to access-date and time zones. The rarity of such complaints given the plethora of cs1|2 templates with access-date would suggest that Module:Citation/CS1/Date validation is not overzealous.  The module operates in UTC time so the time in California (PST/PDT) is irrelevant.
 * —Trappist the monk (talk) 02:57, 1 June 2018 (UTC)
 * Why does a timestamp (an integral part of the URL in, a footnote in México en la Piel (album)) have to be 14 digits? An error message is generated, but AFAICT there's no error.  Mini  apolis  23:55, 12 June 2018 (UTC)
 * Because, for archive.org, a 14-digit timestamp uniquely identifies the moment that a web page was archived. One of the two (as I write this) archive urls looks like this:
 * where the timestamp  decodes to 2018-05-15 17:47:02 (15 May 2018 at 5:47:02pm).  But, there is something peculiar about that page.  If I copy the above url into my browser's (Chrome win7) address bar and watch it after I press the enter key, I see it change to this:
 * which is a malformed archive-url – it is missing this bit: .  This is not the fault of cs1|2 but rather is the fault of some process external to en.wiki – javascript in the source page perhaps?  I don't know.
 * So, the fix at México en la Piel (album) is to replace the url in archive-url with the complete archive.org url above.
 * —Trappist the monk (talk) 00:49, 13 June 2018 (UTC)
 * which is a malformed archive-url – it is missing this bit: .  This is not the fault of cs1|2 but rather is the fault of some process external to en.wiki – javascript in the source page perhaps?  I don't know.
 * So, the fix at México en la Piel (album) is to replace the url in archive-url with the complete archive.org url above.
 * —Trappist the monk (talk) 00:49, 13 June 2018 (UTC)
 * So, the fix at México en la Piel (album) is to replace the url in archive-url with the complete archive.org url above.
 * —Trappist the monk (talk) 00:49, 13 June 2018 (UTC)

Unknown parameter ignored
Seeing "Unknown parameter |1= ignored (help)" in the Template:Royal Collection at Crown Jewels of the United Kingdom for no apparent reason. Here is an example:. Firebrace (talk) 12:01, 10 June 2018 (UTC)


 * Caused by a stray pipe in Royal Collection, I fixed it, but those errors shouldn't be visible in the first place when  is empty. Headbomb {t · c · p · b} 12:26, 10 June 2018 (UTC)
 * Yes, it was strange and only started happening today. Thanks for fixing the error. Firebrace (talk) 16:23, 10 June 2018 (UTC)

Deprecated in
I have no idea what to do to clean these uses: Faraway, So Close!, Monty Python and the Holy Grail, The Devil's Advocate (1997 film), and City of Angels (film). Thoughts requested. --Izno (talk) 16:36, 10 June 2018 (UTC)
 * Clicking the "help" link usually takes you to useful instructions. It is possible that the instructions were updated after you started seeing the errors; they currently explain that in should be replaced with language. – Jonesey95 (talk) 00:08, 11 June 2018 (UTC)
 * My mistake, you were asking about some pathological cases. I should have looked at them first. I changed in to title and title to chapter. – Jonesey95 (talk) 00:10, 11 June 2018 (UTC)
 * Yes, I worked through the batch in the category yesterday and none of them had to do with a language. Just goes to show you that our original logic for deprecation was correct. ;) --Izno (talk) 12:31, 11 June 2018 (UTC)

Issues with valid date format
While the documentation states the date format " accessdate= October 2015 " is valid, it is producing the following error message: Check date values in: |accessdate= (help). For example, see John Wesley Hardin#cite_ref-WDE1_43-0. — btphelps (talk to me) (what I've done) 18:46, 11 September 2018 (UTC)
 * Where is it stated that the documentation states the date format " accessdate= October 2015 " is valid? October 2015 gives the error that you see because the date does not include a day so is not a 'full date' as is stated in the template documentation.
 * —Trappist the monk (talk) 18:56, 11 September 2018 (UTC)

Display loop
Please what display was broke in the category?. –Ammarpad (talk) 09:54, 5 October 2018 (UTC)
 * At this writing, nothing is broken. The category shows only the text that it extracted from access-date= requires |url=.  It is unclear to me what your 'fix' was intended to do but it included the entirety of Help:CS1 errors from §|access-date= requires |url= to the end.
 * So now, your turn. What do you mean by prevent loop display in the cat?
 * —Trappist the monk (talk) 10:13, 5 October 2018 (UTC)
 * If it was unclear to you what you should do was ask. The text Pages with this error are automatically placed in Category:Pages using citations with accessdate and no URL.[a]" should appear in the Help:CS1 errors page where it was written but not in the category page. That's loop display, unintentional reference to current page caused by transclusion and that's why tags exist to exclude unnedeed text either in the destination page or in the page hosting the text. –Ammarpad (talk) 10:37, 5 October 2018 (UTC)
 * What you was unclear; what resulted was not unclear so the action I took was correct.  I wrote most of this page and I wrote the   parser function at the bottom of access-date= requires |url= (and every other error message section in this page).  It is not unintentional reference to current page caused by transclusion.  I intended to include that text in the category page.
 * —Trappist the monk (talk) 11:03, 5 October 2018 (UTC)
 * I actually mostly agree that this text shouldn't display on the category page as it doesn't add a whole lot of value from what I can see. I tweaked the implementation. (I am not at all attached to the change.) --Izno (talk) 14:43, 5 October 2018 (UTC)
 * I think this change should be reverted so that the page works the same as the other 44 CS1 error category pages. – Jonesey95 (talk) 14:54, 5 October 2018 (UTC)
 * So... consistency? That's not a great argument for reversion at all. I'm not attached, but it should be reverted on the substance, if there is such. --Izno (talk) 15:00, 5 October 2018 (UTC)
 * You wrote (some part of) it, but you don't own it; that's first. When I said "unintentional reference" I am referring to the standard way of transclusion and extending or not extending portion of text to the destination page, I am not referring to you, for it's ridiculous to do so. I am referring to the text in whole. And now, as you can see, the earlier no sense display loop was no longer there and that's all I wanted to do from the beginning. –Ammarpad (talk) 15:06, 5 October 2018 (UTC)
 * I presume that you are talking to me. Please do not put words into my mouth that I have not spoken; I have made no claim to ownership.  It is true that I wrote most of the help text; it is true that I created the parser function found there.  Neither of those assertions are claims to ownership.  I wrote what I wrote because you claim that the reference-to-the-category-page-on-the-category-page is 'unintended'; you are mistaken, it is not unintended.  If it can be done better then by all-mean propose a better way that retains or improves upon the current functionality.
 * —Trappist the monk (talk) 15:42, 5 October 2018 (UTC)
 * Consistency with what? with meaningless back reference to current page?. It's the other pages that need sanitization to get rid of this. Not otherwise. –Ammarpad (talk)
 * Because you are not attached to your change, I'm going to revert. Each of the error message sections in Help:CS1 errors has that parser function code; it is the same for each section intentionally so that all of the sections and all of the associated categories have the same look and feel.  This common mechanism also makes the creation of new sections a bit easier.  I am also reverting, at least partially, your changes at  because all of the categories should be the same and because I think that cs1|2 error categories should retain the footnote detailing the excluded namespaces.  If it is important that 'Pages with this error ...' text be removed from the cs1|2 error categories then we should discuss what, if anything, should replace that text and how that should be implemented across all of the categories.  We should not piece-meal tweak one or another according to our individual likes and dislikes.
 * —Trappist the monk (talk) 15:16, 5 October 2018 (UTC)
 * Reverts done. Let us now discuss.
 * —Trappist the monk (talk) 15:42, 5 October 2018 (UTC)
 * Consistency among similar pages is a reader-friendly practice, and it is a good argument for reversion of a change to just one of those pages. I am happy to hear recommendations for changes to some or all of the pages that would keep all of the CS1 error category pages consistent with one another. – Jonesey95 (talk) 15:50, 5 October 2018 (UTC)
 * The improvement I meant was clear, and now even more so since at least one person now understand it. It remains ether to keep no sense back referencing to current page because of "consistency"/you don't like it or  to update the other categories so that they longer refer back to themselves. That's the entire essence of tags and parser functions that provide the functionality. –Ammarpad (talk) 16:02, 5 October 2018 (UTC)
 * The improvement I meant was clear, and now even more so since at least one person now understand it. It remains ether to keep no sense back referencing to current page because of "consistency"/you don't like it or  to update the other categories so that they longer refer back to themselves. That's the entire essence of tags and parser functions that provide the functionality. –Ammarpad (talk) 16:02, 5 October 2018 (UTC)

What should I do about multiple ISBNs?
I'm quoting a book that has two 13-digit ISBNs listed inside the cover (and two 10-digit ones, but that's besides the point). I tried putting both of them in with a semicolon to separate them, but of course it thinks there is an error. How can I list both, or if that's "against the rules" according to Wikipedia, which one should I pick? Hannahshipman (talk) 05:48, 8 November 2018 (UTC)
 * Put one or both of them outside the citation template. – Jonesey95 (talk) 07:14, 8 November 2018 (UTC)
 * Are the two disambiguated in some way? Hardcover? Softcover? ... When there is disambiguation, choose the one that best matches the source that you consulted.  When there is no disambiguation, choose one to put one inside the cs1|2 template so that that number is included in the citation's metadata; put the other outside.
 * —Trappist the monk (talk) 09:24, 8 November 2018 (UTC)
 * —Trappist the monk (talk) 09:24, 8 November 2018 (UTC)


 * Or we could have a proper fix for this. Headbomb {t · c · p · b} 18:06, 8 November 2018 (UTC)

"Missing or empty |url=" with ASINs
Hello, I'm writing to this talk page specifically to address one concern: templates with an assigned asin= parameter but no url= value. It is to my understanding that web citations with a valid asin= value automatically generate the desired Amazon.com link in the citation. Therefore, a reference to an Amazon.com page with a correspoding ASIN should only include, at bare minimum, the ASIN, and the URl parameter should be excluded to avoid redundancy. However, this is not the case in practice. An article I have devoted years on, Dexter's Laboratory, has citation errors for reflinks that include an ASIN without a URL parameter. What should I do in this situation? Do I add redundant Amazon.com links in the url= parameter? Or, do I ignore the citation errors? Or something else? Paper Luigi T • C 12:08, 13 November 2018 (UTC)
 * Since on 20 March 2013,  has required url to have a valid assigned value so that it can link the value in title.  It is not the purpose of the named identifiers to do that linking (pmc is the singular annoying exception).  The error message that you see reflects that requirement.  Yes, asin does create an external link to Amazon (most named identifiers create external links to related hosts) but that does not remove the requirement for url to have an assigned value.
 * The minimal correct solution is to add a value to url. If you wish to retain the asin id and avoid redundant links consider using this form:
 * —Trappist the monk (talk) 13:12, 13 November 2018 (UTC)
 * Or a more reasonable
 * to give
 * Headbomb {t · c · p · b} 16:04, 13 November 2018 (UTC)
 * Or a more reasonable
 * to give
 * Headbomb {t · c · p · b} 16:04, 13 November 2018 (UTC)
 * Headbomb {t · c · p · b} 16:04, 13 November 2018 (UTC)
 * Headbomb {t · c · p · b} 16:04, 13 November 2018 (UTC)

Date without day
Need guidance on what corrective action to take when encountering an  that does not include a day. It would be wrong to just make up a day, so besides completely redoing the reference, there seems to be no simple fix at all. Personally I would prefer if a day could be entered as "00" to indicate 'unspecified' and get rid of the auto-generated error message. —DIV (1.144.110.230 (talk) 09:30, 25 May 2019 (UTC))
 * Yeah, no simple solution which is a large part of why lists so many articles.  The correct remedy is to confirm that the linked source still supports the article text. When it does, update access-date to the current whole date.
 * There was a time when cs1|2 2_date_parameter_values|considered supporting edtf date formats which, at the time defined  or   as a digit replacement character so   would mean some day in May 2019 (the replacement character has been changed to  ).  Had we adopted that date format, I think it unlikely that we would have allowed the format in access-date because we want day-precision dates there (online source can change at any time at the whim of their owners).  Except for internal representation of seasons, the edtf format experiment was removed; see this discussion.
 * —Trappist the monk (talk) 10:42, 25 May 2019 (UTC)
 * You could also check the article history to see when the reference was added, and use that date. MeegsC (talk) 20:41, 29 May 2019 (UTC)
 * You could also check the article history to see when the reference was added, and use that date. MeegsC (talk) 20:41, 29 May 2019 (UTC)

Error in PMC check
The Alcmonavis article is showing a CS1 error for the PMC value of one reference — but the number is correct, and the linked article shows up just fine. I think somebody may need to update the code, as it appears article numbers are now past 6000000. MeegsC (talk) 20:39, 29 May 2019 (UTC)
 * Fixed in the sandbox; see Help talk:Citation Style 1.
 * —Trappist the monk (talk) 20:46, 29 May 2019 (UTC)

Two digit years
Greetings and felicitations. Regarding two digit years, the Help page states


 * Does not handle years before 100 AD, including BCE/BC dates. Try using parameter orig-year instead.

Unfortunately



yields



(See Transfiguration of Jesus.)

I.e., no year at all. Suggestions? —DocWatson42 (talk) 11:34, 25 July 2019 (UTC)
 * Are you actually putting on the white cotton gloves and consulting a 78 CE manuscript? If you are not, then you must be consulting a later edition.  Use the date from the later edition.
 * —Trappist the monk (talk) 11:57, 25 July 2019 (UTC)

Bimonthly published magazines
Some magazines are published bimonthly and give their publication date as e.g., "August/September" - how to handle this? Currently this creates an error message. FOARP (talk) 09:03, 8 August 2019 (UTC)
 * Search for "Slash in date range (use en dash)" on the help page;  should work, using an ndash. -- John of Reading (talk) 09:19, 8 August 2019 (UTC)
 * Thanks FOARP (talk) 11:13, 8 August 2019 (UTC)

Check |url= value error for o.bike
I noticed that there are some references on OBike generates the Check |url= error, with the links generally being starting with https://www.o.bike/ . The domain is dead for a while, but had been cited when the startup was still in operation. From this help page, it seems that single-character second level domains are generally being flagged with some exception. What's the remedy for this? 1. report to somewhere to add o.bike as an exception? If so, where? 2. delete/replace reference (but also not sure which other pages have o.bike as reference)? robertsky (talk) 04:07, 21 August 2019 (UTC)
 * I'm not inclined to add support for a specific url without there are a of them used in cs1|2 templates; there are not.  Because o.bike is dead, one might set unfit or usurped.  That will hide the original url and its error message.
 * —Trappist the monk (talk) 11:14, 21 August 2019 (UTC)
 * Thanks for the advice! I have updated the refs to unfit. robertsky (talk) 16:11, 21 August 2019 (UTC)

Need blue tick Asifsayyad07 (talk) 20:47, 9 October 2019 (UTC)

deprecated dead-url
At Toyota RAV4 I am getting a lot of  messages. The help link merely says "use a supported parameter". and say that dead-url is still supported. Even other parts of this error help page say the dead-url is still supported (with limited values, which are used). What gives?  Stepho  talk 11:30, 3 September 2019 (UTC)
 * Yeah, I have the same problem with many articles I've worked on. What am I supposed to replace it with? And taking it out doesn't seem to be a good option, as doing so automatically causes the archived url to appear first; as if the link is dead, even though that isn't the case. PanagiotisZois (talk) 11:45, 3 September 2019 (UTC)
 * It seems url-status is the new term, according to Help:CS1_errors. Might wait for a bit to see what kind of bot cleanup happens. Quuux (talk) 11:51, 3 September 2019 (UTC)
 * Yes, you all have figured it out. See also 2 module suite after 2 September 2019. --Izno (talk) 12:26, 3 September 2019 (UTC)
 * Sheesh, I don't want to be ungrateful for your hard work (which I do appreciate). But releasing the documentation only after complaints came rolling in is a recipe for confusion. Even better if a bot had been let loose to make this simple and mechanical change before making it mandatory and leaving practically every article with a sea of red references. My recommedation is to turn off then error reporting for a month or so while a bot makes the changes for us. Then turn the warnign back on when the bot has done the bulk of the work for us. Why make people do the simple and boring work when the machines can do it for us.  Stepho  talk 12:36, 3 September 2019 (UTC)
 * A bot could not have been run before the change because then you would have been complaining about unrecognized parameter errors (because the previous version of the module suite did not know about url-status).
 * Bots/Requests for approval/Monkbot 16
 * —Trappist the monk (talk) 13:35, 3 September 2019 (UTC)
 * Was this something that really needed changing? I've never heard of url-status, is it mass used? Get a bot on this please. Darkwarriorblake / SEXY ACTION TALK PAGE! 13:53, 3 September 2019 (UTC)
 * Was it sensible to make this change without at least a transitional period when the module will recognise either the old or new parameter? --David Biddulph (talk) 13:59, 3 September 2019 (UTC)
 * This is the transition period. The deprecated parameter error message is there to educate editors that the deprecated parameter is going to go away.  Both dead-url (and deadurl) and url-status are both functional.  This process is not new and has been how we have handled deprecated parameter for several years.
 * —Trappist the monk (talk) 14:12, 3 September 2019 (UTC)
 * There is also an ongoing discussion at Administrators' noticeboard. 114.159.158.244 (talk) 14:23, 3 September 2019 (UTC)
 * I have only just become aware of the death of . I have used this parameter absolutely everywhere on Wikipedia articles. Is there a bot going around fixing these? Because there is no way I or anyone else is going to go back and manually re-edit literally thousands of articles to comply with this surprise change. Cnbrb (talk) 14:46, 3 September 2019 (UTC)
 * yes, link above.
 * —Trappist the monk (talk) 14:50, 3 September 2019 (UTC)
 * Thank you. I suggest maybe you post a notice about it on the main help page to avoid confusion (and so you don't have to answer the same question again tomorrow!). Cnbrb (talk) 00:08, 4 September 2019 (UTC)
 * Thank you. I suggest maybe you post a notice about it on the main help page to avoid confusion (and so you don't have to answer the same question again tomorrow!). Cnbrb (talk) 00:08, 4 September 2019 (UTC)

Help pages need to be updated. WP:LINKROT says to use "dead-url", for example. I found myself confused by contradictory information on "dead-url" vs "url-status" as I was trying to fix some references in an article I was updating a few days ago (Anontune) and had to figure out which worked by trial and error. Just today, I ran across another article with a lot of error messages due to "dead-url" no longer working (ARPANET). Documentation needs to be correct and consistent when changes like this are implemented. I'd correct WP:LINKROT myself, if I was sure that this change was going to stick. Carl Henderson (talk) 04:53, 4 September 2019 (UTC)
 * I have updated the LINKROT page. Thanks for the note. – Jonesey95 (talk) 14:16, 4 September 2019 (UTC)

Cite letter
I've come across a Template:Cite_letter template that's getting flagged with the error "Cite news requires |newspaper=" (see Machine_Identification_Code for the actual error). The error references this help page, but the advice to add a periodical doesn't make a ton of sense in this context. I think Template:Cite_letter should be exempted from this error, since a letters are often standalone entities and not part of a larger body of work. It seems like this is probably happening because Template:Cite letter says it is descended from Template:Cite newspaper - GretLomborg (talk) 17:33, 3 September 2019 (UTC)
 * I addressed that at Template talk:Cite letter. Anyone can make the fix.
 * —Trappist the monk (talk) 17:40, 3 September 2019 (UTC)
 * Fixed, I believe. – Jonesey95 (talk) 18:23, 3 September 2019 (UTC)

Problematic change
Suddenly italic or bold wikimarkup is not allowed in publisher and periodical parameters including publisher=, journal=, magazine=, newspaper=, periodical=, website=, work=. There are times when such markup is needed. For example the publisher of Vulture.com is New York magazine. Without being able to italicize that in the publisher field, we wind up with a grammatical error — a non-italicized magazine title.

The MOS says that not all rules fit every single case exactly, and that we're to use common sense. I could give other examples, but the point is that there are going to be future examples that we can't envision yet. This stricture needs to be loosened so that we can make common-sense adjustments as the MOS allows. --Tenebrae (talk) 22:05, 3 September 2019 (UTC)


 * If New York is the publisher (e.g. for an 'About Us' page), then you put it in the publisher field, where it shouldn't be italicized, as per rules for publishers. If it's the publication (e.g. you are citing an article written for the magazine New York), then New York is italicized, as per rules for magazines. &#32; Headbomb {t · c · p · b} 01:32, 4 September 2019 (UTC)

CS1 errors: missing periodical
Hello, fellow Wikipedians!

Several days ago, when I checked our WikiProject's [to-do-list], suddenly hundreds of new articles that need to be solved occurred and I confused by these things. Usually, it gives us some task every some period. But this time, the CS1 errors flocked the list, especially CS1 errors: deprecated parameters and CS1 errors: missing periodical. The first one seems OK because there's a way how to solve it, but I don't know about the second one. Therefore, any help from you will be appreciated. Thank you.

Sincerely, Samuelsp15 (talk) 13:45, 14 September 2019 (UTC)
 * null edit the pages. this should clear the missing periodical category from the pages. robertsky (talk) 14:27, 14 September 2019 (UTC)
 * See Help Talk:CS1 for long discussions about these error messages. Almost all of the deprecated parameter errors are being fixed by a bot; since there are so many of them, it will take a month or more for the bot to traverse the category, but there is no need to fix the errors yourself. Most of the missing periodical errors will also disappear by themselves.
 * The missing periodical errors, if they are still present in an article when you look at it, can be fixed by looking at the help text linked from the error message. As a short example, cite journal requires the parameter journal to have something in it. You'll have to use judgement to determine the best solution. In Bank of America Tower (Manhattan), for example, there is a citation to a document by Richard A. Cook that uses cite journal without journal. Looking at the PDF document, I see that it is a paper from a conference presentation, so cite conference, with appropriate parameters, is probably a better template to use. – Jonesey95 (talk) 15:09, 14 September 2019 (UTC)
 * The missing periodical errors, if they are still present in an article when you look at it, can be fixed by looking at the help text linked from the error message. As a short example, cite journal requires the parameter journal to have something in it. You'll have to use judgement to determine the best solution. In Bank of America Tower (Manhattan), for example, there is a citation to a document by Richard A. Cook that uses cite journal without journal. Looking at the PDF document, I see that it is a paper from a conference presentation, so cite conference, with appropriate parameters, is probably a better template to use. – Jonesey95 (talk) 15:09, 14 September 2019 (UTC)

unknown language code - de
Is the German really unknown language? I used "de" for script-title and got this error.
 * The error is correct. German is a language that uses a Latin script so script-title is inappropriate.  script-title and its companion parameters are for languages that are written with scripts that are not Latin: Chinese, Japanese, Russian, Greek, Arabic, Hebrew, etc.
 * —Trappist the monk (talk) 12:35, 2 October 2019 (UTC)
 * Thanks!--Nicoljaus (talk) 16:40, 2 October 2019 (UTC)

Help fix errors in references on sdwiki article
Hi, I tried to fix the references errors in one of the Sindhi Wikipedia article Rudhrama Devi (film) but the problem of the errors in references still exists, I request your help to fix references errors in article. thanks JogiAsad  Talk   12:51, 6 January 2020 (UTC)
 * Umm, both of those links are to empty pages ... Actual working links would be helpful if we are to help you.  This page: رڌراماديوي (فلم)?
 * I think, though I can't tell from the histories because I don't read Sindhi, but I think that you are using a very old (2016) version of the cs1 module suite with some newer elements (November 2017) in your sd:Module:Citation/CS1/Configuration that are not supported by the 2016 version of sd:Module:Citation/CS1/Date validation. Don't mix old and new; there is no guarantee that doing so will produce correct results.
 * My recommendation to you is to wait a couple of weeks after en.wiki makes its next update (expected 11–12 January 2020) then, import en.wiki's live module suite into your sandbox; make whatever changes to configuration are necessary, test your changes and then update your live module suite from your sandbox.
 * —Trappist the monk (talk) 13:29, 6 January 2020 (UTC)
 * Yes رڌراماديوي (فلم) is that page with ref errors JogiAsad Talk   15:17, 6 January 2020 (UTC)
 * Then after updates which you have mentioned should I create modules on sdwiki? JogiAsad Talk   15:17, 6 January 2020 (UTC)
 * I have moved your comments out of my post. At en.wiki, commenting within another editor's comments is discouraged.
 * Yes. The purpose of the two-week-wait is to allow time for us to find an fix any bugs that were not detected during development (real life it the best text suite).  So, your TODO list:
 * wait two weeks
 * import:
 * en:Module:Citation/CS1 → sd:Module:Citation/CS1/sandbox
 * en:Module:Citation/CS1/Configuration → sd:Module:Citation/CS1/Configuration/sandbox
 * en:Module:Citation/CS1/Whitelist → sd:Module:Citation/CS1/Whitelist/sandbox
 * en:Module:Citation/CS1/Date validation → sd:Module:Citation/CS1/Date validation/sandbox
 * en:Module:Citation/CS1/Identifiers → sd:Module:Citation/CS1/Identifiers/sandbox
 * en:Module:Citation/CS1/Utilities → sd:Module:Citation/CS1/Utilities/sandbox
 * en:Module:Citation/CS1/COinS → sd:Module:Citation/CS1/COinS/sandbox
 * en:Module:Citation/CS1/styles.css → sd:Module:Citation/CS1/sandbox/styles.css – note different name format; these css pages are sanitized css content model; changing content model on sd.wiki may require someone with administrator privileges
 * en:Module:Citation/CS1/Suggestions → sd:Module:Citation/CS1/Suggestions/sandbox
 * make configuration changes to ~/Configuration/sandbox
 * test
 * copy all of your sandbox modules over the live modules:
 * sd:Module:Citation/CS1/sandbox → sd:Module:Citation/CS1
 * etc
 * Yes, its a big task. It will be easier if editors at sd.wiki monitor changes here and keep sd.wiki's cs1 module suite nominally in sync with en.wiki.
 * —Trappist the monk (talk) 15:40, 6 January 2020 (UTC)
 * These all above modules are already existed on sdwiki, I'll update Module:~/Configuration ASAP its being updated on enwiki.Thanks JogiAsad Talk   09:29, 7 January 2020 (UTC)
 * I do not think that we are communicating. According to this search, there are no sandbox versions of the cs1 modules on sd.wiki.
 * —Trappist the monk (talk) 13:04, 7 January 2020 (UTC)
 * Yes Sandbox are not available but those modules are available on sdwiki. JogiAsad Talk   14:52, 8 January 2020 (UTC)
 * These all above modules are already existed on sdwiki, I'll update Module:~/Configuration ASAP its being updated on enwiki.Thanks JogiAsad Talk   09:29, 7 January 2020 (UTC)
 * I do not think that we are communicating. According to this search, there are no sandbox versions of the cs1 modules on sd.wiki.
 * —Trappist the monk (talk) 13:04, 7 January 2020 (UTC)
 * Yes Sandbox are not available but those modules are available on sdwiki. JogiAsad Talk   14:52, 8 January 2020 (UTC)

Why are 3 initials discouraged?
Why is this edit discouraged by the text "in case of more than two initials, list only the first two" under Vancouver style error. When one goes to the source, it's apparent that Webster KLW is a more accurate representation of the person's name. (FYI, I used to generate the citation.) Also, Webster KLW is how they are identified in PubMed, for example. Why would we need to remove an initial? Why shouldn't we instead add parentheses to get rid of the error? Biosthmors (talk) 21:09, 13 January 2020 (UTC)
 * In this citation:
 * you told cs1|2 that the author name-list uses Vancouver style. cs1|2 tries to adhere to the basics of that style.  The style is defined in:
 * Portions of that are linked from the error message help text at . Not linked, though perhaps it should be is Citation Rules with Examples for Journal Articles § General Rules for Author which has as it fourth bullet point:
 * Convert given (first) names and middle names to initials, for a maximum of two initials following each surname
 * Yes, NLM sometime does not obey their own style guide. cs1|2 has no control over what NLM does.  If you want NLM to do things differently, you must talk to them about those things.  Until the standard changes, cs1|2 will continue to emit error messages when there are more than two initials in an author's name.
 * The  is intend specifically for corporate and institutional names and for mixed-case romanizations of non-Latin initials.  Using the markup as you are doing to quell a legitimate error message is a misuse of that markup.
 * —Trappist the monk (talk) 00:04, 14 January 2020 (UTC)
 * Yep. And this is yet another reason to not impose off-site, "reader-hateful" citation styles on Wikipedia. Just because WP:CITEVAR the use of any attested citation style (in the interests of avoiding tedious "style-warring") doesn't mean that every one of them is a  to use here.  Anything that produces confusing output, or interferes with the ability of readers checking sources to identify sources and their authors, is a bad idea. The main reason CITEVAR exists is so that professionals in various fields can contribute material in formats they are entirely used to from their work, without some wiki-nanny hassling them about cite formatting trivia (we want subject-matter experts to improve articles and add good sources to them, more than we want them to format citations in the most perfectly WP-reader-friendly way). If a citation style is unhelpful at an article (e.g. because it is resulting in the confusion of someone listed at PubMed as "Webster KLW" with someones different listed there as "Webster KL" and "Webster KLB" and "Webster KLR" or whatever, by truncating them all to an ambiguous "Webster KL" in that citation style), then this is a very good reason to propose changing the citation style at the article's talk page (or just boldly changing it if the article is still a stub, or isn't one but doesn't have a long-established and consistently applied citation style). Short of that, it's also a good rationale to WP:IAR and just use "Webster KLM" in the citation even if the publisher of the off-site cite style wouldn't approve, since we know that off-site publications that nominally purport to follow that citation style also do exactly the same thing in the real world.  — SMcCandlish ☏ ¢ 😼  13:51, 16 January 2020 (UTC)
 * If you want to violate Vancouver style, the solution is simply: don't use Vancouver-style related parameters. &#32; Headbomb {t · c · p · b} 14:02, 16 January 2020 (UTC)
 * —Trappist the monk (talk) 00:04, 14 January 2020 (UTC)
 * Yep. And this is yet another reason to not impose off-site, "reader-hateful" citation styles on Wikipedia. Just because WP:CITEVAR the use of any attested citation style (in the interests of avoiding tedious "style-warring") doesn't mean that every one of them is a  to use here.  Anything that produces confusing output, or interferes with the ability of readers checking sources to identify sources and their authors, is a bad idea. The main reason CITEVAR exists is so that professionals in various fields can contribute material in formats they are entirely used to from their work, without some wiki-nanny hassling them about cite formatting trivia (we want subject-matter experts to improve articles and add good sources to them, more than we want them to format citations in the most perfectly WP-reader-friendly way). If a citation style is unhelpful at an article (e.g. because it is resulting in the confusion of someone listed at PubMed as "Webster KLW" with someones different listed there as "Webster KL" and "Webster KLB" and "Webster KLR" or whatever, by truncating them all to an ambiguous "Webster KL" in that citation style), then this is a very good reason to propose changing the citation style at the article's talk page (or just boldly changing it if the article is still a stub, or isn't one but doesn't have a long-established and consistently applied citation style). Short of that, it's also a good rationale to WP:IAR and just use "Webster KLM" in the citation even if the publisher of the off-site cite style wouldn't approve, since we know that off-site publications that nominally purport to follow that citation style also do exactly the same thing in the real world.  — SMcCandlish ☏ ¢ 😼  13:51, 16 January 2020 (UTC)
 * If you want to violate Vancouver style, the solution is simply: don't use Vancouver-style related parameters. &#32; Headbomb {t · c · p · b} 14:02, 16 January 2020 (UTC)

dates
The acceptable formats for dates seemin some case absurdly strict, because the computer should be perfect capable of normalizing many of them, for Example
 * October, 1913 should work as well as October 1913  -- what possible ambiguity is there?
 * in the other direction, February 28 1900should work as well as the form with the comma.
 * January 04, 1987 should work as well as January 4, 1987 again, what's the possible ambiguity?
 * Full Caps for month should work -- ditto
 * Febr. should work for February, -- ditto

I recognize that some of the errors are true ambiguities, like 00--00-0000, and others are plain errors.  DGG ( talk ) 19:06, 28 January 2020 (UTC)
 * The acceptable date formats are a community consensus, as displayed at MOS:DATEFORMAT. The best venue for any concerns with this consensus is Wikipedia talk:Manual of Style/Dates and numbers. – Jonesey95 (talk) 21:58, 28 January 2020 (UTC)
 * You need to manually insert a space before and after the endash; Spaced en dash won't work, and I don't think a hard space (after the first date, per the MOS) will either.  Mini  apolis  18:01, 4 March 2020 (UTC)

Check PMID value
I checked it - it is correct... All the best: Rich Farmbrough  (the apparently calm and reasonable) 22:32, 1 February 2020 (UTC).


 * Updated at Module:Citation/CS1/Identifiers/sandbox. We sometimes deploy this fix out of cycle, but I do not have admin rights to make this change to the module. – Jonesey95 (talk) 02:43, 2 February 2020 (UTC)

Unknown parameter |xxxx= ignored
Do you have a suggestion on how to include more specific information and/or references? In this case, I was trying to document how a reverted page might use a deprecated parameter like  and how to deal with it.

My feeling is eventually something like the Python Enhancement Proposal (PEP) process to document changes might be needed. PEP's document the discussion, possible implementations, pros, cons, changes, and migration from old to new features. PEP 238 -- Changing the Division Operator, as an example, was a major change from Python 2 to Python 3.

I was too specific with my edit, but I thought "This help page is a how-to guide. It details processes or procedures[...]"

I would like to find a way to include this detailed technical information, in addition to the English instructions.

Here's my |xxxx=_ignored revision to the end of the section:


 * Reverting a page to a previous version may generate this error as Wikipedia may now process pages differently. For instance, the previously valid parameter  now generates this error and should be replaced with the   parameter. The deprecation and replacement looks like this:


 * See dead-url deprecation Help_talk:.

This was removed and concisely replaced by modifying the introductory paragraph from:

to:

I feel this English text leaves someone who has followed the "help" link on the invalid citation with insufficient information to resolve the citation's problem.

Happily, at least, I figured out that &#x7C;&#x7B;&#x7B;strikethrough&#x7C;dead-url&#x3C;nowiki&#x3E;&#x3D;&#x3C;&#x2F;nowiki&#x3E;no&#x7D;&#x7D; produces | Lent (talk) 06:14, 16 February 2020 (UTC) :-)


 * On the automated side of things, I made a suggestion for an automated suggestion for the now removed, previously deprecated, dead-url and deadurl, to url-status.
 * See: Village_pump_(technical) --Lent (talk) 10:06, 16 February 2020 (UTC)
 * Suggestion has been added to Module:Citation/CS1/Suggestions/sandbox.
 * I'm not sure that I understand how you would have us use PEP as a model. As I understand it, PEP is a formalized  process used to manage enhancements to the Python programming language.  But here, you appear to be discussing our help-text.  PEP needs to be technical and is read by people with an aptitude for technical subjects; our help text is for people who are not technical; too much technical information for them is mind-numbing and pointless.
 * —Trappist the monk (talk) 16:24, 16 February 2020 (UTC)
 * Perhaps help that ended with a link or button for more information, or a pointer to similar past fixes might be helpful. Or to a step-by-step walk-through for the process. Or a place to ask for help beyond what they have found?
 * Thanks for making the swift changes to Module:Citation/CS1/Suggestions/sandbox! I was happy to see that other changes, and a whitelisted item got folded into the change as well.
 * Yeah, I guess the PEP would be overkill, but I do like the "all the details in one place" result of a PEP.
 * I understand the "mind-numbing" TMI. When I went to look at Phabricator as a way to suggest the change, I was overwhelmed.
 * --Lent (talk) 08:29, 18 February 2020 (UTC)
 * Doesn't the current mechanism already do most of what you are asking?
 * At the end of the rendering is a brief error message with a link [to] more information; at the link's target page there are walk-through instructions on how to repair the error; at the link's target page there is another link to a place to ask for help: WP:HELPDESK.
 * I'm not sure what you mean by a pointer to similar past fixes might be helpful. The purpose of the error messaging and of this cs1|2 help page is to show editors that there is something amiss with a particular cs1|2 template and to provide the editor with sufficient information to correct the problem.  Yeah, it isn't perfect because sometimes the documentation and help text lags implementation (as is the case with dead-url).
 * —Trappist the monk (talk) 12:44, 18 February 2020 (UTC)
 * Right on all counts. I don't fault documentation and help text lags when attempts can be made to "fix" perceived problems quickly, and revised swiftly by calmer minds who leave an explanation.
 * I disregarded the WP:HELPDESK button, much like the test users for Windows 95 missed the Start button.
 * I guess I wanted a more before and after example. Like:
 * If URL is supplied, then the title cannot be wikilinked.
 * The URL must include the URI scheme in order for MediaWiki to recognize it as a link. For example:  vs..
 * or
 * For example, the "publisher" parameter should be omitted in these examples:
 * Amazon.comAmazon Inc.
 * The URL must include the URI scheme in order for MediaWiki to recognize it as a link. For example:  vs..
 * or
 * For example, the "publisher" parameter should be omitted in these examples:
 * Amazon.comAmazon Inc.


 * CNNWarnerMedia


 * For a pointer to similar past fixes might be helpful, a link to a diff of a change, like Government spin-off: Difference between revisions would give a contextual idea of the change. I suppose the best example would be one which has the most, non-Bot fixable, errors. That way a specific example is provided to those who might do the most good.
 * Thanks for helping me think more deeply about this issue and also truely see what is already available.
 * --Lent (talk) 19:59, 18 February 2020 (UTC)
 * I guess the reason I am so interested in links is I have watched what happened after I gave a suggestion in a focus group. I said "It would be really helpful to be able to go to the Dell service website, type in the service tag, and get the original shipped configuration of the PC. Users may have modified the machine with more (or less) memory, graphics cards, or other hardware. Dell already ships the configuration on paper with the PC. If it was available online that would really help." With this there came a huge thump from behind the big mirror, and the muffled sounds of excited discussion. Two months later the feature I asked for was part of Dell's website. Years later, Dell integrated the system into their "Order Dell replacement parts for your system website". See.
 * Ideas made better by the other ideas keep me very hopeful for the future. Thanks!
 * --Lent (talk) 20:26, 18 February 2020 (UTC)

date error messages
Need help to understand where the problem lies with the following articles: Many thanks in advance. --  Ohc  ¡digame! 13:43, 9 March 2020 (UTC)
 * Ras Shorty I
 * Rebecca Levene
 * Richard Dixon (footballer, born 1990)
 * For the first two, the "access dates" pre-date the founding of Wikipedia. They can't be the dates on which an editor accessed these sources to verify the articles. The third one must have been a control character of some kind, as it disappeared when I deleted and re-entered the date. -- John of Reading (talk) 13:54, 9 March 2020 (UTC)
 * no-break-space at the end. See
 * —Trappist the monk (talk) 15:01, 9 March 2020 (UTC)
 * How do I remove the nbsp using my script? --  Ohc  ¡digame! 18:01, 10 March 2020 (UTC)

I've tried fixes relating to all the above with Charles Henry Juliá Barreras, but there's still an error. --  Ohc  ¡digame! 18:01, 10 March 2020 (UTC)
 * Founding date of Wikipedia is 15 January 2001. The access date at Charles Henry Juli%C3%A1 Barreras is June 28, 2000 which falls before Wikipedia founding date so is invalid.  This error is documented in the error message help text.
 * —Trappist the monk (talk) 18:29, 10 March 2020 (UTC)

This should be reported as a DOI error
&#32; Headbomb {t · c · p · b} 03:09, 5 April 2020 (UTC)

—Trappist the monk (talk) 11:06, 5 April 2020 (UTC)

Access Date error for no URL= should not happen if the DOI is specified, which generates a link
Tom Stannage is currently reporting a CS1 error of "|access-date= requires |url=", but the ref has a DOI listed, which generates a URL. That should be considered before this error is raised. The-Pope (talk) 16:04, 9 April 2020 (UTC)
 * The error message is telling you that access-date requires a URL in url. This is because the content at a URL is often ephemeral so access-date identifies that point in time when the source supported the text in an en.wiki article and serves as a starting point for editors seeking an archived copy of that source.  The so-called identifier-generated urls are not ephemeral so access-date does not apply to them.
 * —Trappist the monk (talk) 16:15, 9 April 2020 (UTC)
 * Okay.
 * 1) This error will appear when someone inappropriately adds access-date to accompany one or more of doi et al. People should learn not to. (In the {cite...} instructions.)
 * 2) This error will appear when a person or bot deletes a url without deleting the subsidiary access-date.
 * 2.1) A bot deletes each url that specifies the same URL made by doi in the same {cite...}. I assume such bot also deletes the access-date.
 * 3) The explanation of this error doesn't tell people what to do when access-date seems justified by one or more of doi et al. People have been "commenting out" the access-date parameter, rather than delete the potentially "valuable" information. For example, this 2016-06-09 edit made 24 out of 30 spurious access-date parameters into comments, in this case placing them between the  and the  . (These access-date parameters were added spuriously (accompanying doi and not url) while the article was in sandbox in 2012.) - A876 (talk) 23:22, 16 July 2020 (UTC)

Invalid-yet-correct-doi
The above DOI works, even if in the wrong format. An error-checking bypass is needed. &#32; Headbomb {t · c · p · b} 15:28, 26 April 2020 (UTC)


 * If this idea goes ahead there are a few examples of valid DOIs with // instead of / in them. I can find examples if helpful. Rjwilmsi  16:04, 26 April 2020 (UTC)
 * That might be useful yes. But // isn't flagged as an error. &#32; Headbomb {t · c · p · b} 17:00, 26 April 2020 (UTC)


 * Did you mean to edit your example above? Anyway, is an example of a DOI that has // and resolves OK, and the version with just / doesn't resolve. (I have assumed, but maybe it's just my assumption, that DOIs don't normally contain //).  Rjwilmsi  19:49, 26 April 2020 (UTC)
 * Didn't mean to, no. You're correct that DOIs don't normally contain //, but we don't check for that. It would be interesting to see if broken DOIs with // get fixed if / is used instead though. &#32; Headbomb {t · c · p · b} 19:52, 26 April 2020 (UTC)

Twitter titles
what would be the appropriate way to treat citations for Twitter which throw up errors for "url in title field"? --  Ohc  ¡digame! 15:58, 18 June 2020 (UTC)
 * ? Alas, it isn't a simple change-the-name fix.
 * —Trappist the monk (talk) 16:06, 18 June 2020 (UTC)

Please beef up error-checking at &lt;param>-link=_value
Please beef up error-checking at 26. &lt;param&gt;-link= value in order to check for missing colon on interwiki links. The module should check for presence of an interwiki prefix (a two- or three-character code for foreign Wikipedias; see list), and require presence of a leading colon before the Wikipedia language code, to prevent confusing downstream problems. For a complete description of the problem, please see WT:CS1. Thanks, Mathglot (talk) 03:47, 3 August 2020 (UTC)

why not robots?
It seems like rather than pollute thousands of articles with bright red warning errors "Cite uses deprecated parameter x", a dedicated bot could simply replace, say, all deprecated instances of "editorlink=" with the agreeable and copacetic "editor-link=". Let's focus on keeping Wikipedia a high quality work, not scrawl red warnings on it for rather trivial reasons. How bout, it, bot lords? --Animalparty! (talk) 04:13, 12 October 2020 (UTC)
 * Why at all? We have authorlink, so what's wrong with editorlink? -- Michael Bednarek (talk) 02:01, 22 October 2020 (UTC)
 * We will not always have authorlink. We are moving to hyphenated multiword parameter names from the non-hyphenated forms so that all cs1|2 parameter names have a consistent style.  At present, authorlink is used in about 270k articles; comparatively few when compared to accessdate at 2.8 million articles.
 * —Trappist the monk (talk) 10:19, 22 October 2020 (UTC)
 * Why can we not permanently allow both the new hyphenated forms and the old unhyphenated forms? --RichardW57 (talk) 12:06, 22 October 2020 (UTC)
 * We moved away from underscore-separated parameter names, away from capitalized parameter names, away from camel-case parameter names. Moving away from all-run-together names is just another step in the process of standardization.  Standardizing the parameter name style makes writing and maintaining cs1|2 templates easier for editors and automated tools.  Yeah, until we get this sorted, there will be churn but we will all benefit once the standardization is complete.
 * —Trappist the monk (talk) 13:55, 22 October 2020 (UTC)
 * Standardisation with what? The latest fashion? --RichardW57 (talk) 19:10, 22 October 2020 (UTC)
 * This is a nice history lesson, but back to the question at hand: why can't a bot simply replace the thousands or millions of instances of troublesome text? --Animalparty! (talk) 18:58, 29 October 2020 (UTC)
 * We can; I did. Why not just do it ahead of time?  WP:COSMETICBOT.  Also, because doing it ahead of time may break cs1|2 templates because the current module suite may not know about the new parameter names; bots breaking citations is not looked upon favorably by the community.  So we deprecate.  That creates error messages that need to be fixed which avoids the strictures of WP:COSMETICBOT.  I restricted Monkbot task 17 so that another editor could piggyback cosmetic changes he wanted to make to a certain group of articles that might otherwise have been in violation WP:AWBRULES rule 4.  It is all a game.
 * —Trappist the monk (talk) 19:38, 29 October 2020 (UTC)

Help explaining doi errors
I was doing doi error cleanup and stumbled across Special:PermanentLink/1010187246 n5 and pt7. Since they both resolve correctly, I used ((accept-as-written)) markup to suppress the error. I assumed the error was being triggered by there being dashes in the doi, but Special:PermanentLink/1010309336 n1 has dashes in the doi without issue. Can someone explain what is triggering the doi error in these two examples, and whether it is something that can be remedied or just needs to be accepted-as-written? Sincerely, InsaneHacker (💬) 21:06, 4 March 2021 (UTC)
 * The help text correctly identifies the problem that the current version of the module suite sees with those dois (invalid registrant code). The next update fixes that:
 * Were it me, I would not employ the accept-this-as-written markup without you remember to remove that markup after the next update
 * —Trappist the monk (talk) 23:47, 4 March 2021 (UTC)
 * I see. I'll revert the edits I made to those dois in anticipation of the next update. Thanks for the explanation! Sincerely, InsaneHacker (💬) 08:09, 5 March 2021 (UTC)
 * —Trappist the monk (talk) 23:47, 4 March 2021 (UTC)
 * I see. I'll revert the edits I made to those dois in anticipation of the next update. Thanks for the explanation! Sincerely, InsaneHacker (💬) 08:09, 5 March 2021 (UTC)

What to do about et al errors.
The help page describes et al. errors, but does not state how to solve them.--TZubiri (talk) 23:16, 19 March 2021 (UTC)
 * This sentence from doesn't answer that question?
 * Use the display-authors parameter instead (or other applicable display-&lt;names> parameter), as documented at Help:Citation Style 1.
 * —Trappist the monk (talk) 23:26, 19 March 2021 (UTC)

‎Lua error in Module:Citation/CS1/Utilities at line 127: Called with an undefined error condition: invalid_param_val
Probably caused by adding a ref with invalid value subscription. (Note that that edit also includes duplicate pages params in the added ref, but that doesn't normally throw a module error.) The version throwing this error is rev. 1013109707 of article Attraction to transgender people (diff ). Fixed in a later revision by not using invalid value "subscription", but shouldn't this be caught and handled explicitly and not throw a module error? Mathglot (talk) 02:37, 20 March 2021 (UTC)
 * That's been fixed in the sandbox:
 * —Trappist the monk (talk) 02:50, 20 March 2021 (UTC)
 * Much better, thank you! Mathglot (talk) 10:15, 20 March 2021 (UTC)
 * —Trappist the monk (talk) 02:50, 20 March 2021 (UTC)
 * Much better, thank you! Mathglot (talk) 10:15, 20 March 2021 (UTC)

Volume errors
Is there a way to treat the volume parameter differently for books and journals? I understand not wanting a literal "Volume" in a journal listing, but making "extra text" an error is causing some serious issues with books. Just putting a bare number after a book title makes no intuitive sense at all. And what about books that say something like "Volume 16: Tanagers to New World Blackbirds"? Do you suggest that we now add that to the book title instead of putting it in the volume field? (Talk about unintuitive!) Suggestions please. WP:BIRD now has 1200+ articles showing error messages. MeegsC (talk) 12:21, 13 April 2021 (UTC)
 * —Trappist the monk (talk) 13:01, 13 April 2021 (UTC)
 * Related discussions:
 * Help_talk:CS1_errors (about how to get rid of the message)
 * Help_talk:Citation_Style_1 (about how to possibly add a "Vol." prefix to the cite book output)
 * --Matthiaspaul (talk) 10:14, 27 August 2021 (UTC)
 * --Matthiaspaul (talk) 10:14, 27 August 2021 (UTC)

Issue and Number
This template does not allow both issue and number. I am looking at the Shadis magazine in archive.org (https://archive.org/details/shadismagazine/Shadis/Shadis%20Magazine%20%2319/page/42/mode/2up). They have issue and number fields. archive.org lists the article by issue number. But if you look at the end of the article you only see the number. These are different. Slimy asparagus (talk) 12:46, 7 August 2021 (UTC)
 * archive.org lists the article by issue number. But if you look at the end of the article you only see the number. Is that what you really meant to write?  Yeah, archive.org lists the  by sequence number.  But, it looks to me like each magazine typically lists volume and number at the bottom of even-numbered pages.  In the handful that I looked at, none of the articles reverted to the sequence number at the end of the article.
 * Of course, Shadis isn't wholly consistent. The magazine listed by archive.org as #23 doesn't identify volume and issue in the masthead (presumably it is Volume IV, number 5) but lists number 'twenty-three' at the bottom of even-numbered pages, and 'Volume IV' on odd-numbered pages with the date.  There may be other inconsistencies.
 * My advice? Pick one form; choose the sequence number (on the cover and listed in the masthead) or choose the volume/number pair (at the bottom of even numbered article pages and listed in the masthead) and use the chosen form consistently in our article.
 * Technically, cs1|2 cannot simultaneously support both issue and number because the metadata standard used by all of the cs1|2 templates has a slot for only one of those parameter aliases so Module:Citation/CS1 allows only one of the two parameters to have an assigned value.
 * —Trappist the monk (talk) 19:39, 7 August 2021 (UTC)
 * See also:
 * MOS:NUMBERSIGN
 * Help:Citation_Style_1, Template:Cite_journal, Template:Cite_magazine, Template:Cite_news
 * Help_talk:Citation_Style_1/Archive_29
 * Help_talk:Citation_Style_1/Archive_37
 * Help_talk:Citation_Style_1/Archive_42
 * Help_talk:Citation_Style_1/Archive_62
 * Help_talk:Citation_Style_1/Archive_72
 * Help_talk:Citation_Style_1/Archive_77
 * Help_talk:Citation_Style_1
 * --Matthiaspaul (talk) 10:14, 13 August 2021 (UTC)
 * To sum it up, editors should use either issue or number depending on what is actually used in the source. (For templates displaying prefixes in front of the issue number, the chosen prefix "Iss." or "No." may depend on which parameter was used.)
 * At present, it is not possible to provide both parameters at the same time, because they are treated as aliases internally. Therefore, if both values are given in the source, both values must be combined into one of these parameters. If one of them is an absolute number and the other a volume-relative number, the absolute number should be prefixed with #. This is not necessary, when only one number is given in the source.
 * Adding support for both issue and number as independent parameters has been discussed various times and basically agreed upon. Metadata for both will have to be combined into the single COinS key.
 * --Matthiaspaul (talk) 11:54, 27 August 2021 (UTC)
 * At present, it is not possible to provide both parameters at the same time, because they are treated as aliases internally. Therefore, if both values are given in the source, both values must be combined into one of these parameters. If one of them is an absolute number and the other a volume-relative number, the absolute number should be prefixed with #. This is not necessary, when only one number is given in the source.
 * Adding support for both issue and number as independent parameters has been discussed various times and basically agreed upon. Metadata for both will have to be combined into the single COinS key.
 * --Matthiaspaul (talk) 11:54, 27 August 2021 (UTC)

last3 with no last1/author1 doesn't generate error
I'm surprised that having last3 without last1 or author1 doesn't generate an error:





Would it be worthwhile to include scenarios like this in future error checking? Thanks! GoingBatty (talk) 20:51, 26 August 2021 (UTC)


 * It would be yes. In general, foobarx+1 without a corresponding foobarx should all be reported as errors. &#32; Headbomb {t · c · p · b} 22:07, 26 August 2021 (UTC)
 * I agree. And in fact it does check for holes in the enumeration for as long as their size is one. It does not detect errors if the size of the holes is larger.
 * According to Help:CS1_errors this appears to be "by design", but it would certainly be better if holes of any size would be detected.
 * --Matthiaspaul (talk) 07:32, 27 August 2021 (UTC)
 * Why is this discussion here and not at Help talk:Citation Style 1 where it belongs?
 * —Trappist the monk (talk) 11:42, 27 August 2021 (UTC)
 * Because it's not clear to me what topics belong here and what topics belong at Help talk:Citation Style 1. Please feel free to move it and clarify the instructions at the top of this page.  Thanks!  GoingBatty (talk) 14:14, 27 August 2021 (UTC)

Question about volume has extra text error
Jasminum sambac reference #29 has the follow error:



Template:Cite book has a similar example, with a similar error:

Should the error logic be tweaked to allow this format, or should this example be removed/changed? Thanks! GoingBatty (talk) 02:37, 27 August 2021 (UTC)
 * In both cases, the string "Volume" is triggering this error message:
 * Volume 8, Flowers
 * Volume I: The Early Successes Against Italy (to May 1941)
 * There are two ways how to solve this: In most cases, the best way to deal with such errors is to just remove the offending pattern:
 * 8, Flowers
 * I: The Early Successes Against Italy (to May 1941)
 * Per the current consensus, that cite book should display volumes without any prefix (like cite journal), this gives the correct output. However, there have been complaints that it should better behave like cite magazine, that is, prefix the volume value with "Vol.", so it is quite likely that this will be changed in the future. (Potential changes like this are among the reasons, why such prefixes should be generated by the template, if appropriate, and not be made part of the input value itself.)
 * In cases where the volume designation entered must, by all means, be reproduced as given, the error detection can be bypassed by using our "accept-this-as-it-is" syntax:
 * ((Volume 8, Flowers))
 * ((Volume I: The Early Successes Against Italy (to May 1941)))
 * However, this may cause a problem when cite book (or cite journal) would actually start to display a prefix "Vol." in front of the given value, because then, in the current model, the output would look like:
 * "Vol. Volume 8, Flowers".
 * This is a case we would have to think about, and a possible solution could be that, if an entry was given using the accept-this-as-is syntax and it actually contains one of the offending patterns like "Vol.", "Volume" etc., the template would suppress the generation of the prefix. Such a scheme could work for extra text given as prefix (in the volume case) or as suffix (in the edition case), but it might not work well for other kinds of "extra text" in the middle of a given value.
 * Nevertheless, these are the two options you have right now. And, yes, we should change the example accordingly.
 * --Matthiaspaul (talk) 09:59, 27 August 2021 (UTC)
 * Thanks for your reply. There's another option on the cite book documentation: Include "Volume I: The Early Successes Against Italy (to May 1941)" in the title field.  GoingBatty (talk) 14:17, 27 August 2021 (UTC)
 * Yeah, that's true. I just saw it as well when I fixed the example. However, I think putting it into the title stresses the purpose of the title parameter a bit, can cause an inconsistent look of citations, and creates suboptimal metadata. Still better than to omit the information altogether.
 * What could be useful to have for such cases is a volume-title parameter. Even if the template would internally just combine it with the volume information, it would stop questions where to put volume titles, if volume is for semi-numerical values only or not, and if to include "Volume" prefixes or not, and at the same time it would give us flexibility to render the information consistently according to some common standard instead of every editor having to invent their own style. (Similar issue for edition, issue and number, which also sometimes take titles rather than numbers only.) Also related, we are still lacking a part parameter; not in your example, but some of what we have to treat like "volumes" right now are actually "parts", and there even is a dedicated COinS key for parts (which we do not take advantage of at present).
 * --Matthiaspaul (talk) 15:11, 27 August 2021 (UTC)
 * Why is this discussion here and not at Help talk:Citation Style 1 where it belongs?
 * —Trappist the monk (talk) 11:43, 27 August 2021 (UTC)
 * Because it's not clear to me what topics belong here and what topics belong at Help talk:Citation Style 1. Please feel free to move it and clarify the instructions at the top of this page.  Thanks!  GoingBatty (talk) 14:15, 27 August 2021 (UTC)

missing title and |subject=
@Editor Paine Ellsworth: With, you added this text:
 * Sometimes the title parameter is tied to another parameter, such as subject. To resolve this error, provide a value for the tied parameter, for example (name of subject).

That doesn't make any sense to me. subject and subjectn are aliases of author, authorn, last, and lastn. None of those have anything to do with missing title detection. Can you provide an example to show that a missing subject parameter causes this error?

—Trappist the monk (talk) 10:40, 24 October 2021 (UTC)
 * yes, that is the result of an edit request at Template talk:Cite letter. In that citation template, the subject parameter is tied to the title parameter so that if the mandatory subject is omitted, then the title error is produced. Since the documentation doesn't mention the title, only the subject parameter, I thought it best to include the blurb on this page. If it's too confusing then by all means clean it up or erase it, whatever you think best.  P.I. Ellsworth &numsp;- ed.  put'r there 10:48, 24 October 2021 (UTC)
 * Thanks. I mentioned the error message at .  I will revert your edit at Help:CS1 errors.
 * —Trappist the monk (talk) 11:08, 24 October 2021 (UTC)

References with a one-letter second-level domain URL
One-letter second-level domains like d.tube cause Check   value  errors when used in references, but are supported and becoming more common with the newer top-level domain registries, and seem to be appearing more often throughout the encyclopedia. I created a temporary example here. Is there a way to easily rectify this error on Wikipedia, or is there a table within the wiki, or code within the MediaWiki software itself that would need to be changed? Thank you! —  W ILD S TAR  talk 15:58, 27 October 2021 (UTC)
 * Fixed in the sandbox:
 * In your example, live is meaningless without archive-url so should be omitted.
 * —Trappist the monk (talk) 16:14, 27 October 2021 (UTC)
 * , yeah... it's the VE's Cite web that automatically includes the empty parameter live —  W ILD S TAR  talk  17:37, 27 October 2021 (UTC)
 * —Trappist the monk (talk) 16:14, 27 October 2021 (UTC)
 * , yeah... it's the VE's Cite web that automatically includes the empty parameter live —  W ILD S TAR  talk  17:37, 27 October 2021 (UTC)

How to link a ref to it's own wiki page?
I am referencing De Re Atari, which has its own wiki page here. I am pointing to the physical copy at archive.org. Attempting to link to both causes a wikilink error. This seems... wrong. Is there any reason an inline could not be provided by a second parened link like (local article) or similar? Maury Markowitz (talk) 14:52, 9 November 2021 (UTC)
 * See WP:CIRCULAR. We do not link to Wikipedia articles in references. – Jonesey95 (talk) 18:26, 9 November 2021 (UTC)
 * Sure we do, author_link does precisely that. Maury Markowitz (talk) 20:32, 9 November 2021 (UTC)
 * You may add a link to either the 'physical copy' or to the 'local article' after the citation – between  and .  If you are citing the actual De Re Atari (the physical copy is the cited source), then that is what you should use in the citation template.  Put the link to the en.wiki article ( the cited source) outside of the citation template.
 * —Trappist the monk (talk) 20:59, 9 November 2021 (UTC)
 * But I'm using {{ 's and sfns, where do I put it in that case? Maury Markowitz (talk) 21:50, 9 November 2021 (UTC)
 * Umm, what? What does {{tlx|sfn}} have to do with linking to the 'physical copy' or linking to the 'local article'?  Show an example of what it is that you are trying to accomplish?
 * —Trappist the monk (talk) 23:05, 9 November 2021 (UTC)

Access date
I attempted to cite a source (List of Malaysian electoral districts citation 8 as of the time of writing) with the access dates of 23 and 27 November 2021. Saw that only endashes are accepted. Proceeded to make it 2327 November 2021. Read the help page that says that the date must have only one day. So I removed the day. Then read that it must have a day, a month is insufficient. Now, is there something I'm not getting about citing with an access date of two days? — Fredrick Campbell (talk) 09:26, 27 November 2021 (UTC)
 * The documentation for access-date states Full date when the content pointed to by url was last verified to support the text in the article, so if you checked the source on two different dates, just give the later of the two. -- John of Reading (talk) 09:52, 27 November 2021 (UTC)
 * Thank you. — Fredrick Campbell (talk) 04:07, 28 November 2021 (UTC)

Date in YYYY-MM format was invalid
I was copying over a reference from another language and found out that I needed to convert "1907-12" to "December 1907" to be compliant with the parser on the English language Wikipedia. Since "YYYY-MM-DD" is already supported, could the English language Wikipedia's parser be updated to match? --Saledomo (talk) 09:37, 4 January 2022 (UTC)
 * First you must persuade editors at WT:MOSDATE to make this change to MOS:DATES from which cs1|2 takes its guidance. This has been argued before so you will likely need a new and startlingly rational argument to overcome opposition to YYYY-MM.
 * —Trappist the monk (talk) 13:10, 4 January 2022 (UTC)
 * Since "1907-12" could mean "1907 through 1912" or "December 1907", we define the YYYY-MM to be invalid. GoingBatty (talk) 13:58, 4 January 2022 (UTC)
 * If we're not going to follow ISO 8601 in assuming that a date is YYYY-MM when 4 digits are followed by a hyphen and two more digits, can we at least get support for 2022-03-XX which the docs here claim is supposed to work but doesn't? It looks really ugly to be reading a list of citations with all beautifully correct ISO-formatted dates and then have a “month YYYY” jump out. Xenophore ; talk 22:35, 4 March 2022 (UTC)
 * YYYY-MM-XX (EDTF) was briefly supported because MediaWiki giveth and MediaWiki taketh away (see phab:T132308). The EDTF date format should not be documented anywhere on en.wiki.  If it still is, please say where that is so that the documentation can be fixed.
 * —Trappist the monk (talk) 22:51, 4 March 2022 (UTC)
 * Indeed, it is on this very page. Here's the link to the section where it can be found: Check date values in: |&lt;param1>=, |&lt;param2>=, ... It's in the table called "Examples of unacceptable dates and how to fix them"
 * Xenophore ; talk 03:41, 5 March 2022 (UTC)
 * Thanks. Fixed.
 * —Trappist the monk (talk) 03:50, 5 March 2022 (UTC)
 * I'm responding because you pinged me, not because I can do anything about YYYY-MM-XX. You could convert your beautifully correct ISO-formatted dates to beautifully correct DD MMM YYYY dates or beautifully correct MMM DD, YYYY dates to avoid the jump out.  GoingBatty (talk) 03:16, 5 March 2022 (UTC)
 * That would violate WP:DATERETAIN.  Stepho  talk 09:33, 5 March 2022 (UTC)

Seasonal dates spanning multiple years
I've found that cite book and cite magazine don't recognize seasonal dates spanning multiple years, such as  (see e.g., note 48). I'm not seeing this date format mentioned on the help page or under MOS:BADDATE. Is there a way around this? Thanks. --Sangdeboeuf (talk) 02:00, 1 March 2022 (UTC)
 * Umm:
 * —Trappist the monk (talk) 03:33, 1 March 2022 (UTC)
 * Weird. I guess it works with a spaced dash, but not an unspaced one. Thanks. --Sangdeboeuf (talk) 06:56, 1 March 2022 (UTC)
 * yesterday by BattyBot in . GoingBatty (talk) 16:10, 1 March 2022 (UTC)
 * yesterday by BattyBot in . GoingBatty (talk) 16:10, 1 March 2022 (UTC)

Parse error
Hello. I am receiving a parse error for archive-url in this edit. The url does work but I think ?=&t=o-que-e is a problem for the template. I am unable to guess why this would not also yield an error in url but it does not. Because the url is live this is not urgent. Invasive Spices (talk) 15 March 2022 (UTC)
 * You mean this edit? That error is because you wrote:  .  Don't do that.  Use archive-date instead.
 * —Trappist the monk (talk) 21:34, 15 March 2022 (UTC)
 * Yes. I gave you the wrong diff number and I have no clue how I did the other thing. Storchy fixed it anyhow. Thank you. I suppose I need use findargdups or such on my own edits. Invasive Spices (talk) 16 March 2022 (UTC)

Controlling error message display
To display maintenance messages, the page says to include .mw-parser-output span.cs1-maint {display: inline;} /* display Citation Style 1 maintenance messages */ in one's common.css. Entering that into the editor, however, gives the message Warning: Element (span.cs1-maint) is overqualified, just use .cs1-maint without element name. Xenophore ; talk 17:22, 17 March 2022 (UTC)

Problem with |archive-url= requires |archive-date= CS1 error
I have been attempting to fix a CS1 archive-url error on Monty Oum, to no avail. I have tried two different types of valid MOS:DATEFORMAT, but the error on Monty Oum continues to show :   requires   (help). I have purged the page cache, as well. Any advice and/or assistance would be appreciated. Thanks! Top5a (talk) 22:56, 1 April 2022 (UTC) Fixed transclusion issue. Nevermind! — Preceding unsigned comment added by Top5a (talk • contribs) 23:01, 1 April 2022 (UTC)

Adding script into my CSS
Can someone explain how I can add the appropriate "Error and maintenance messages" script so the maintenance message would show? I read it on the main page but am confused. Thanks.--TerryAlex (talk) 23:03, 7 April 2022 (UTC)
 * Help us improve the help text. What is it that is confusing?
 * —Trappist the monk (talk) 23:07, 7 April 2022 (UTC)
 * Actually, I got it now. Just not a technical person. Thanks!!!--TerryAlex (talk) 23:23, 7 April 2022 (UTC)

How should this be corrected?
Hi, I am adding information to a page and when citing sources of the info, I got a Script warning telling me that: One or more “cite web” templates have errors, and when I got down to the citation area, there this: cite web: line feed character in |title= at position 24 So how can this be corrected? I have very few experiences in citing sources on Wikipedia, would appreciate assistance as soon as possible. Thank you! — Preceding unsigned comment added by Medical droid (talk • contribs) 13:47, 8 April 2022 (UTC)
 * You can see where it's happening in this diff. Find that text and remove the newline that exists between 'Port of ' and 'Hong Kong'.
 * —Trappist the monk (talk) 14:14, 8 April 2022 (UTC)
 * Thanks for your reply, I will now know how to correct this type of errors in the future. Medical droid (talk) 08:40, 9 April 2022 (UTC)

common CSS page
Hi, there is a confusing instruction in the Controlling error message display section that advises those who want to display maintenance messages, include some text in your common CSS page. Could somebody please explain what this means. Thanks Davidelit (Talk) 10:32, 2 May 2022 (UTC)
 * Which instruction do you find confusing? Specific questions beget specific answers.
 * —Trappist the monk (talk) 11:21, 2 May 2022 (UTC)

What's happening here?
Hi, I can't find what CS1 error is present here.--Carnby (talk) 18:39, 23 May 2022 (UTC)
 * There was some sort of invisible character in the access-date parameter value. It was a tricky one. I resolved it by erasing the access date, along with the = and the } on either side, and retyping them. – Jonesey95 (talk) 18:44, 23 May 2022 (UTC)
 * Thank you!--Carnby (talk) 18:49, 23 May 2022 (UTC)

not helpfull page
Somebody who knows what they are doing should have a look at this page. I tried several ways of copy/pasting the boxed text at "error and maintenance messages" on my CSS page but was unsuccesfull. --Dutchy45 (talk) 10:16, 28 May 2022 (UTC)


 * What are you trying to do exactly? Because I see nothing in your recent contributions that indicates you tried to do something with css. &#32; Headbomb {t · c · p · b} 10:40, 28 May 2022 (UTC)


 * Hello @Headbomb, yeah since I couldn't do it, I exited without doing anything. I came across a template that said a maintenance messages was hidden from me. I clicked and was taken here. I clicked on the common.css link and found out I needed to create a new page there;User:Dutchy45/common.css. I did that and tried copy/pasting .mw-parser-output span.cs1-maint {display: inline;} /* display Citation Style 1 maintenance messages */ like it tells me to do. The result was the little yellow triangle with the exclamation mark and the text:Warning: Element (span.cs1-maint) is overqualified, just use .cs1-maint without element name. I tried removing several different things from the copy/pasted text hoping to stumble on this mysterious element name that I needed to get rid off. Alas, I was unsuccesfull and gave up. Dutchy45 (talk) 02:14, 1 June 2022 (UTC)
 * It just gives you a warning that an element is overqualified (which means there's redundancy/overspecificity in the code, technically,  isn't needed in   and you could just have  ). You can ignore it and save anyway. &#32; Headbomb {t · c · p · b} 02:23, 1 June 2022 (UTC)
 * Thanks @Headbomb, I've done it. I still feel the correct text i.e.  should be on the help page, but I'm not gonna mess with that. Dutchy45 (talk) 02:39, 1 June 2022 (UTC)

chinese language parameter
I notice that pages using the zh now display red categories at the bottom of articles in which they are used, namely (Categories: CS1 Chinese (Hong Kong)-language sources (zh-hk) and CS1 Traditional Chinese-language sources (zh-hant)) (see Witman Hung partygate. Is this deliberate?  Ohc  revolution of our times 06:42, 2 July 2022 (UTC)
 * I close this thread as this issue seems to have been solved. --  Ohc  revolution of our times 09:54, 4 July 2022 (UTC)

ISBN error
It may be worth noting that in some cases an ISBN error may be because it’s actually an 9-digit SBN number from the early 70s. SBN has its own option etc. jengod (talk) 03:47, 15 July 2022 (UTC)

Controlling the display of hidden error and warning messages needs to be easier
I was brought to this help page because an article I was editing showed these warnings in the preview stage:

Script warning: One or more templates have maintenance messages; messages may be hidden (help). Script warning: One or more templates have maintenance messages; messages may be hidden (help).

The help links brought me to the first section of this help page, "Controlling error message display", which has this advice:

To display maintenance messages, include the following text in your common CSS page or your specific skin's CSS page (common.css and skin.css respectively): ...

This is not helpful to the typical editor. When I click on the links in the text above, "common.css" and "skin.css", the following nonexistent pages open: https://en.wikipedia.org/wiki/User:Teemu_Leisti/common.css and https://en.wikipedia.org/wiki/User:Teemu_Leisti/vector.css. (The name of the latter doesn't even match the displayed name of the link, "skin.css".) What am I to understand I should do here?

I humbly suggest that it would be a good idea to automate the procedure this text is suggesting, so that an editor would only have to toggle a switch to see the hidden maintenance messages. Teemu Leisti (talk) 11:42, 27 July 2022 (UTC)
 * It is not possible to automate the procedure because only you or those few editors with interface editing privileges can edit .css page(s).  Apparently you figured it out.  If you can think of a better way to describe what needs to be done, edit the current description to make it better.
 * —Trappist the monk (talk) 13:22, 27 July 2022 (UTC)

So, I tried creating the nonexistent pages referred to above, with the following content:

I then saved the pages, and clicked the "Show preview" button again in the article I was editing. Nothing in the displayed content changed: the warning messages I referred to above were still displayed, and I saw no warnings in the references section, where all the mentioned cite-templates were defined. Yes, I had bypassed the browser cache, as instructed on page https://en.wikipedia.org/wiki/User:Teemu_Leisti/common.css after I created it. Teemu Leisti (talk) 12:03, 27 July 2022 (UTC)
 * The preview messaging always shows when cs1|2 has emitted error or maintenance messages regardless of your .css settings. The green maintenance messages are often hard to find in a reference section unless you are looking right at them.  I use an in-browser search (Ctrl+F in windows OS) for 'cs1' to find them.  For the red error messages, I search for '(help' – with the opening parenthesis.
 * You didn't name an article where you are having this problem, but your edit history suggests that the article is Tampere. In that article, there are three maintenance messages (all are CS1 maint: url-status (link)) at refs 4, 16, and 61.
 * —Trappist the monk (talk) 13:22, 27 July 2022 (UTC)
 * —Trappist the monk (talk) 13:22, 27 July 2022 (UTC)

Thanks for the replies, Trappist the monk. I edited section the section in question to give more help to the inexperienced editors. I've also repeated my suggestion above as Village_pump_(technical). I suggest that further discussion be continued there. Teemu Leisti (talk) 15:15, 27 July 2022 (UTC)


 * Can someone point me to an example sandbox where this can be tested? — xaosflux  Talk 16:36, 27 July 2022 (UTC)
 * Where can be tested?  It is, after all, just text in a user's preferred .css page.  Is there something about your .css pages that preclude 'testing' there?
 * —Trappist the monk (talk) 17:04, 27 July 2022 (UTC)
 * Category:CS1 maint: url-status has plenty of example articles if you add the user CSS. Your browser probably has a feature to inject CSS directly in a viewed page without having to save user CSS in your account. PrimeHunter (talk) 17:20, 27 July 2022 (UTC)

Purpose of AAW markup in title
Hey! I was recently looking into Category:CS1 errors: generic title, and found that situations where the citation module had misflagged a title as generic could be supressed with the AAW ( () ) markup to get around this. However, I noticed this behaviour is not mentioned in either the help page for generic title, nor the general section about what AAW is used for. I was wondering - is it entirely safe to use AAW to get around these errors, as I noticed it has other functionality when used within the title, and if so, should it be mentioned on either the generic title section or AAW section? Aidan9382 (talk) 15:19, 12 August 2022 (UTC)

Confusing message
If I edit Future Nostalgia and then do a preview of the edit, I see this warning:  The "(help)" link goes to this page, which has confusing instructions about hacking different little files with weird CSS and JavaScript fragments. How can I tell if the warnings are displayed or not? How can I easily make them be displayed so I can correct the article? Can't this be made at least a little bit more user friendly? -- Mikeblas (talk) 15:32, 19 November 2022 (UTC)
 * CSS only; no javascript. At User:Mikeblas/common.css you added the necessary css when you created that page.
 * At Future Nostalgia you should see a green maintenance message at this citation. When looking for maintenance messages, a CTRL+F for   usually finds cs1|2 templates with maintenance messages (there are exceptions: template is hidden in a collapsed section, etc).
 * Documentation can almost always be improved. If you know how to make the instructions more user friendly, please edit  to make it so.
 * —Trappist the monk (talk) 16:24, 19 November 2022 (UTC)
 * "How can I tell if the warnings are displayed or not?" If you see the warnings, they are displayed. Maintenance messages are off by default, so if you never enabled them, you shouldn't see them. Error messages however, you should always see. &#32; Headbomb {t · c · p · b} 19:24, 19 November 2022 (UTC)
 * This issue keeps coming up, and I think we should fix it. It is confusing for readers and editors, especially when they go to our help page and the instructions do not work. The problem is with the message that shows at the top of a Preview view: Script warning: One or more { {cite AV media notes} } templates have maintenance messages; messages may be hidden (help). The message appears even for logged-out editors, and it does not appear to be controllable by CSS. This is different from the error and maintenance messages, controllable by CSS per our instructions, that appear in the rendered citation templates. – Jonesey95 (talk) 23:31, 20 November 2022 (UTC)
 * It is not clear to me just what you would have us fix; the instructions? the preview messaging code? something else?
 * the instructions do not work Is there evidence that this is true?  If the instructions are flawed, how should they be fixed?
 * Messaging in the preview warning box cannot be controlled by css; Module:Citation/CS1 does the styling for those messages manually.
 * The message appears even for logged-out editors How is that different from the preview warnings from, for example, Module:Check for unknown parameters?
 * —Trappist the monk (talk) 23:49, 20 November 2022 (UTC)
 * I think the confusion comes from the Preview box wording that says "messages may be hidden". That phrase has two possible meanings: "you might not see these messages", and "you may hide messages if you wish". I think people are interpreting the phrase as having the second meaning, and then being disappointed when they modify their CSS and find that these maintenance-related messages are still displayed in the Preview box. I think it would be helpful if it said something like "messages are not visible to all editors" (for maintenance messages only). I have also tweaked the documentation to clarify a few things about the display of these messages. – Jonesey95 (talk) 16:29, 27 November 2022 (UTC)
 * Ok, how about ?
 * —Trappist the monk (talk) 17:42, 27 November 2022 (UTC)
 * That is quite nice. I still think that the ambiguous "messages may be hidden" text is the source of some of this confusion. – Jonesey95 (talk) 20:27, 27 November 2022 (UTC)
 * —Trappist the monk (talk) 23:49, 20 November 2022 (UTC)
 * I think the confusion comes from the Preview box wording that says "messages may be hidden". That phrase has two possible meanings: "you might not see these messages", and "you may hide messages if you wish". I think people are interpreting the phrase as having the second meaning, and then being disappointed when they modify their CSS and find that these maintenance-related messages are still displayed in the Preview box. I think it would be helpful if it said something like "messages are not visible to all editors" (for maintenance messages only). I have also tweaked the documentation to clarify a few things about the display of these messages. – Jonesey95 (talk) 16:29, 27 November 2022 (UTC)
 * Ok, how about ?
 * —Trappist the monk (talk) 17:42, 27 November 2022 (UTC)
 * That is quite nice. I still think that the ambiguous "messages may be hidden" text is the source of some of this confusion. – Jonesey95 (talk) 20:27, 27 November 2022 (UTC)