User talk:Citation bot/Archive 35

One for the "to do sometime" list: clean up trivially invalid ISBNs
Could the bot be enhanced sometime to correct trivial errors like this one? Wikipedia talk:ISBN. 𝕁𝕄𝔽 (talk) 11:17, 19 April 2023 (UTC)
 * It now does that within CS1/2 templates. AManWithNoPlan (talk) 12:43, 19 April 2023 (UTC)
 * fixed AManWithNoPlan (talk) 13:30, 20 April 2023 (UTC)

Articles with bare URLs for citations backlog
Hello. I was wondering if Citation bot could help go through the backlog at Category:All articles with bare URLs for citations, if the bot is not already working on it. I do realize that the bot cannot fix all of them, such as PDF files. However, I believe the bot could help make a huge dent on this backlog that is currently over 76,000. Thanks! MrLinkinPark333 (talk) 18:17, 18 April 2023 (UTC)
 * I think that User:BrownHairedGirl is doing this. AManWithNoPlan (talk) 18:24, 18 April 2023 (UTC)
 * Flag to archive. notabug AManWithNoPlan (talk) 15:00, 23 April 2023 (UTC)

zero comes and goes with each edit

 * pubmed sends issue=0 0 AManWithNoPlan (talk) 13:40, 25 April 2023 (UTC)

An error trying to fix an error
Citation bot recently fixed an typo of mine where the 'year' parameter in a citation was given as '022' instead of '2022' (diff page). The original typo did not throw any errors (perhaps because the template accepts years with three digits?), but when the bot changed it to '22' date format errors appeared. I don't know if there is anything that can be done about this kind of typo + fix error, but I thought I'd bring it up here for you to look at. Citation bot is an awesome tool regardless. Junglenut &#124;Talk 10:21, 24 April 2023 (UTC)
 * Module:Citation/CS1/Date validation/sandbox tweaked:
 * —Trappist the monk (talk) 13:15, 24 April 2023 (UTC)
 * —Trappist the monk (talk) 13:15, 24 April 2023 (UTC)
 * —Trappist the monk (talk) 13:15, 24 April 2023 (UTC)


 * Will be fixed in the templates. AManWithNoPlan (talk) 12:54, 28 April 2023 (UTC)

Bot incorrectly changed article titles
In this edit from a year ago the bot incorrectly changed the title of two articles to "Riemann Surfacese and Related Topics (AM-97)" and "Visions in Mathematics."

Is it ok if I block the bot from editing pages where I've already been careful on the references? I've found that it usually adds useless or incorrect info (as in the linked edit) Gumshoe2 (talk) 03:44, 28 April 2023 (UTC)


 * That bug has been fixed (the use of the rare parameter of book-title). AManWithNoPlan (talk) 12:54, 28 April 2023 (UTC)
 * notabug anymore. AManWithNoPlan (talk) 17:24, 30 April 2023 (UTC)

Jezik in slovstvo
Hello. This correction was unnecessary as the correct spelling of the source title is Jezik in slovstvo. Thank you. --TadejM my talk 20:50, 1 May 2023 (UTC)


 * AManWithNoPlan (talk) 12:43, 2 May 2023 (UTC)

lang
I had read that adding a language tag would prevent the bot from attempting to correct the capitalisation of foreign articles. This is a very useful bot but it combines many functions together, and for capitalisation it can cause more maintenence than it performs.

Also if you stop the bot from language overreach until it is capable, it would be good to stop it from time overreach: prevent it from operating on titles published before standardised spelling. Or add to citation templates a parameter to prevent the bot from trying to correct capitalisation.

Sorry for the trouble!


 * Adding the language tag will do nothing. AManWithNoPlan (talk) 12:44, 5 May 2023 (UTC)

Jstor links and the Jstor parameter
Can the bot be changed to stop adding URLs to places where is populated and to move Jstor URLs into the  parameter? Having it duplicated as it currently does seems only to pad citations, both for the viewer and in the markup. Ifly6 (talk) 06:54, 30 April 2023 (UTC)
 * Example please? AManWithNoPlan (talk) 12:10, 30 April 2023 (UTC)
 * Flag as, since removing URLs by the bot is generally banned. I have added code so that jstor URLs will not be added by the "add a free url" code, so that is  .  AManWithNoPlan (talk) 00:22, 3 May 2023 (UTC)

Head's up
T336298 is probably of interest to this page. Izno (talk) 23:00, 9 May 2023 (UTC)
 * They already shutdown the ISSN API. Just another brick in the wall.  AManWithNoPlan (talk) 23:03, 9 May 2023 (UTC)
 * for the heads up. AManWithNoPlan (talk) 23:04, 9 May 2023 (UTC)

Suggestion: Add to the reference template that Bot has already edited
If the database is correct, the bot should edit only once, and when the bot made a mistake, they usually refer to the wrong database. SilverMatsu (talk) 02:30, 15 May 2023 (UTC)
 * This is highly counterproductive, as citations get updated all the time, get covered in new databases, etc. &#32; Headbomb {t · c · p · b} 13:48, 15 May 2023 (UTC)
 * ❌ AManWithNoPlan (talk) 00:28, 23 May 2023 (UTC)

Nomination of The Daily Campus for deletion
A discussion is taking place as to whether the article The Daily Campus is suitable for inclusion in Wikipedia according to Wikipedia's policies and guidelines or whether it should be deleted.

The article will be discussed at Articles for deletion/The Daily Campus until a consensus is reached, and anyone, including you, is welcome to contribute to the discussion. The nomination will explain the policies and guidelines which are of concern. The discussion focuses on high-quality evidence and our policies and guidelines.

Users may edit the article during the discussion, including to improve the article to address concerns raised in the discussion. However, do not remove the article-for-deletion notice from the top of the article until the discussion has finished. Chances last a finite time (talk) 14:00, 23 May 2023 (UTC)


 * Flag to archive ❌ AManWithNoPlan (talk) 14:21, 23 May 2023 (UTC)

IETF Datatracker is not a newspaper
The IETF Datatracker (https://datatracker.ietf.org/) is where the IETF has links to its various documents. It is not at all a newspaper: it is a portal for links.
 * When citing IETF RFCs, use, not.
 * —Trappist the monk (talk) 23:05, 13 May 2023 (UTC)
 * That seems like a good thing for the bot to do. Paulehoffman (talk) 15:02, 14 May 2023 (UTC)
 * Without the newspaper parameter, the template generates an error. AManWithNoPlan (talk) 14:59, 14 May 2023 (UTC)
 * Then we should clean up the template usage, not pretend something is a newspaper when it is not. Paulehoffman (talk) 15:02, 14 May 2023 (UTC)
 * So long as you insist on simply reverting the bot when it edits an improperly constructed cs1|2 template, nothing will be improved. Since you apparently care about the article, you should correct the malformed templates so that the bot doesn't attempt to fix the junk.  I have assembled a list of the obvious citations that you should fix:
 * These should use :
 * ref 6 – incorrectly uses
 * ref 7 – incorrectly uses
 * ref 10 – incorrectly uses
 * ref 13 – first in a bundle of two – incorrectly uses
 * ref 13 – second in a bundle of two – incorrectly uses
 * ref 14 – incorrectly uses
 * These should use :
 * ref 2 – correctly uses ; source linked from url has moved on to v.12 so either the value in title or the value in url needs an update
 * ref 3 – incorrectly uses ; source linked from url has moved on to v.16 so either the value in title or the value in url needs an update
 * —Trappist the monk (talk) 15:56, 14 May 2023 (UTC)
 * Many thanks: I didn't realize that the problem came from the source, not the bot. Now fixed.
 * A related question: is it appropriate for me to update the other references that point to tools.ietf.org to datatracker.ietf.org, which is now considered the proper long-term place for all URLs? Paulehoffman (talk) 21:41, 14 May 2023 (UTC)
 * Unclear. IETF seem to be sort of schizophrenic.  You can get to an RFC through its   doi which takes you to  .  The   url redirects to   so which is the real link target?  At https://datatracker.ietf.org/ right at the top is this message:
 * The IETF Datatracker is the day-to-day front-end to the IETF database for people who work on IETF standards.
 * It contains data about the documents, working groups, meetings, agendas, minutes, presentations, and more, of the IETF.
 * The primary public face of the IETF is at www.ietf.org.
 * To me, that implies that they would prefer that we not link to datatracker because we don't work on IETF standards. This discussion topic should probably continue elsewhere because it isn't a bot related topic.
 * One last thing: yeah, you changed and  to  but not  (ref 3)?  Why?  Also ref 2 and ref 3 still have the title / url-target version mismatch. which you really ought to fix.
 * —Trappist the monk (talk) 00:18, 15 May 2023 (UTC)
 * The DOIs were created before it was clear that the Datatracker would be the one preferred way; they also keep a separation between the RFC Editor and the IETF. There is a long-running disagreement about the One True Way to refer to RFCs, but most people are in favor of the Datatracker because it gives immediate access to valuable metadata that the RFC Editor does not.
 * Your interpretation about the wording from the Datatracker is interesting: I hadn't thought of it that way. I'll bring this to the attention of the IETF Tools Team. I don't think they meant it that way, and they certainly don't want people linking to tools.ietf.org any more (although they will keep the redirects to the Datatracker forever).
 * I left that reference to (ref 3) as news by accident. I'll now do a full cleanup with everything being soon. Thanks again for the pointers! Paulehoffman (talk) 02:34, 15 May 2023 (UTC)
 * The DOIs were created before it was clear that the Datatracker would be the one preferred way; they also keep a separation between the RFC Editor and the IETF. There is a long-running disagreement about the One True Way to refer to RFCs, but most people are in favor of the Datatracker because it gives immediate access to valuable metadata that the RFC Editor does not.
 * Your interpretation about the wording from the Datatracker is interesting: I hadn't thought of it that way. I'll bring this to the attention of the IETF Tools Team. I don't think they meant it that way, and they certainly don't want people linking to tools.ietf.org any more (although they will keep the redirects to the Datatracker forever).
 * I left that reference to (ref 3) as news by accident. I'll now do a full cleanup with everything being soon. Thanks again for the pointers! Paulehoffman (talk) 02:34, 15 May 2023 (UTC)

Can we stop littering S2CIDs everywhere?
Semantic Scholar is a product from (I think?) a for-profit company trying to insert itself into people’s scientific research process. Its links are essentially useless line noise, clicking them does not show anything that a search of the web or any citation index wouldn't also find, does not in my experience ever turn up other materials that aren't easier to find by other means. Wikipedia does not need to indiscriminately mirror every bit of corporate-produced metadata to be found on the web about every citation. All of the other miscellaneous identifiers that Citation Bot adds are also probably a net-negative, but at least they arguably sometimes include reviews etc., but the Semantic Scholar links are a total waste of space. –jacobolus (t) 23:02, 11 March 2023 (UTC)


 * Semantic Scholar is hugely useful and a great backup for many resources when their servers are down. If you don't like it, don't use it. But don't let your dislike of something get in the way of others' use of it. (Edit: Also the Allen institute is a non-profit.). &#32; Headbomb {t · c · p · b} 23:27, 11 March 2023 (UTC)
 * Which "down" servers are you thinking of? What part of it do you find useful? My concern is for readers who are overwhelmed by a huge pile of redundant identifiers that Citation Bot crams into every reference on Wikipedia. These take up a lot of space and are visually distracting. We should consider carefully whether each type is useful/necessary, because every additional one added imposes an attention cost on every reader, the vast majority (99.9%?) of whom get no benefit. –jacobolus (t) 23:32, 11 March 2023 (UTC)
 * Any of the others. DOI servers can be down. Bibcode servers can be down. Pubmed servers can be down. &#32; Headbomb {t · c · p · b} 07:53, 12 March 2023 (UTC)
 * What percentage of the time is the DOI server down, and how long do those outages last? I have personally never experienced this and I click through a lot of DOIs. –jacobolus (t) 08:24, 12 March 2023 (UTC)
 * For a concrete example, Citation Bot just added an SC2 link here on the page Gregory's series. The link points at https://api.semanticscholar.org/CorpusID:121433151 which includes absolutely nothing useful that isn’t already at Springer's site (which was already linked from the DOI https://doi.org/10.1007/BF02838013). The generated summary for the paper is Shailesh Shirali has beenat the Rishi Valley School(Krishnamurti Foundationof India), Rishi Valley,Andhra Pradesh, for morethan ten years and iscurrently the Principal. Hehas been involved in theMathematical OlympiadProgramme since 1988. Hehas a deep interest intalking and writing aboutmathematics, particularlyabout its historicalaspects. He is alsointerested in problemsolving (particularly in thefields of elementarynumber theory, geometryand combinatorics). This is just a poorly formatted copy/paste of the author's bio, unrelated to content the article. The only entity benefitting from including links like this on Wikipedia is Semantic Scholar itself, at the expense of Wikipedia readers. I'd revert the edit as unhelpful but Citation Bot will just put it back again later, unless explicitly blocked from the page. –jacobolus (t) 23:39, 11 March 2023 (UTC)
 * If there's not a PFD copy of the paper at S2CID, the bot shouldn't add the S2CID, i'll agree there. &#32; Headbomb {t · c · p · b} 04:22, 12 March 2023 (UTC)
 * The bot has added the s2cid to other citations when the semantic scholar page doesn't have a PDF. There were some earlier discussions about adding the s2cid here and here. --Whywhenwhohow (talk) 05:42, 12 March 2023 (UTC)


 * I will look into what the API can tell us. Also, it seems to me that URLs without PDFs should be removed and replaced with a S2CID, since they are lof little use.  AManWithNoPlan (talk) 12:09, 12 March 2023 (UTC)
 * Some URLs without PDFs are very useful. No bot should ever be automatically removing URLs without a human inspecting them. –jacobolus (t) 17:47, 12 March 2023 (UTC)
 * there would still be an s2cid url. AManWithNoPlan (talk) 22:39, 12 March 2023 (UTC)
 * I think I am misunderstanding you. Are you just talking about removing URLs pointing to Semantic Scholar specifically? I thought you were talking about removing general URLs to other sites.
 * If it were up to me SC2IDs would never be included, and Semantic Scholar links would only be included as URLs in the event that the PDF of the paper is not freely available from any other source. The SC2IDs are an eyesore. –jacobolus (t) 22:46, 12 March 2023 (UTC)
 * Here is an example where the added s2cid doesn't provide a PDF link. --Whywhenwhohow (talk) 04:12, 13 March 2023 (UTC)
 * Just to pile in, I don't like the s2cid template field either. Whether you think it is commercial or not, it is just clutter in almost every case I've seen.  There is always an alternative that isn't down and any edit that Citation Bot comes up with that consists mainly of s2cid additions doesn't get committed.  As for the url field, that's a whole different can of worms that has been discussed before up to a very high level, but might still merit visiting again.  Lithopsian (talk) 14:22, 13 March 2023 (UTC)
 * Add me to the list of people who think of s2cid's, and of s2cid-adding edits, as spammy useless clutter. —David Eppstein (talk) 21:46, 16 March 2023 (UTC)
 * By the way, article of Semantic Scholar says that: Semantic Scholar is free to use and unlike similar search engines (i.e. Google Scholar) does not search for material that is behind a paywall. --SilverMatsu (talk) 08:26, 19 March 2023 (UTC)
 * What does “doesn't search for material behind a paywall” mean? In practice it indexes plenty of copyrighted items not available online except as paywalled scans. –jacobolus (t) 15:14, 19 March 2023 (UTC)
 * Semantic Scholar is produced by Allen Institute for AI.
 * Citation counts and database contents will differ between S2 and say Scopus. Therefore additional sources of citation information are useful. Invasive Spices (talk) 14:27, 20 March 2023 (UTC)
 * But for anyone who cares: If you have the DOI, and the author name, title, and journal name, how much extra work does it take you to look up the paper in Semantic Scholar for yourself if Wikipedia doesn't include the link directly. Maybe 5 seconds? –jacobolus (t) 15:18, 20 March 2023 (UTC)
 * The same is true for all variables in s and it's impossible for anyone who doesn't know SS exists. Invasive Spices (talk) 20:43, 24 March 2023 (UTC)
 * Yes, and most of the variables in cite templates are mostly a useless waste of space. But the others arguably have marginal redeeming value. Semantic Scholar IDs are almost pure noise. If it were up to me we would also get rid of bot additions of all of these other identifiers. For instance, MathSciNet and zbMATH IDs link to paid-subscriber-only reviews/summaries of various books and papers which are useless to the vast majority of Wikipedia readers and only marginally useful even to subscribers (except perhaps in cases where the original source is not available online).
 * I think parameter of s2cid is useful when the author's personal web page has a PDF file, but in most case, that PDF file (Preprint) is also uploaded on arXiv. --SilverMatsu (talk) 05:56, 25 March 2023 (UTC)
 * If the author's personal page has a PDF, can't we just put it as the "url" field? –jacobolus (t) 15:39, 25 March 2023 (UTC)
 * zbMATH is open to the public, not paid-subscriber-only. —David Eppstein (talk) 07:08, 25 March 2023 (UTC)
 * Thanks for the correction. Do you find the paywalled MathSciNet reviews useful / worth the space? –jacobolus (t) 15:38, 25 March 2023 (UTC)
 * They're not paywalled for me; my employer has a site license. Yes, I find them useful. For one thing, they describe the main results of an article, often more clearly than the article abstract. For another, they are a good source for bibtex entries for the publications they list. —David Eppstein (talk) 16:40, 25 March 2023 (UTC)
 * I am a little late here, but I agree that MathSciNet and ZbMath are both good to link, and that s2cid are wastes of space. I also don't see the point of linking to PubMed ID, PubMed Central ID, bibcode, and/or jstor when there's already complete info and doi links, e.g.:
 * vs the bot's update:
 * Will it cause any problems if I undo some such edits by the bot? — Preceding unsigned comment added by Gumshoe2 (talk • contribs) 03:54, 28 April 2023 (UTC)
 * Will it cause any problems if I undo some such edits by the bot? — Preceding unsigned comment added by Gumshoe2 (talk • contribs) 03:54, 28 April 2023 (UTC)
 * Will it cause any problems if I undo some such edits by the bot? — Preceding unsigned comment added by Gumshoe2 (talk • contribs) 03:54, 28 April 2023 (UTC)

Citation bot damages existing references by renaming source
The fix isn't impossible -- it is easy to detect this case, and then not make any edit. Just because the reference style is easily broken doesn't mean the bot should go around breaking it. -- Mikeblas (talk) 14:06, 5 May 2023 (UTC)
 * How is is easy to detect? AManWithNoPlan (talk) 15:34, 5 May 2023 (UTC)
 * Just the same way a human would:
 * Count the number of errors on the page
 * Make an edit
 * Preview the results of the edit
 * Count the number of errors on the page
 * If the number of errors is now greater, don't save the edit
 * Or, by searching the references on the page for footnotes of this form. Meanwhile, why is it necessary to change "google.co.uk" to "google.com", in the first place? -- Mikeblas (talk) 08:50, 13 May 2023 (UTC)

Here are two more edits where User:Citation bot exercised this bug: and. -- Mikeblas (talk) 08:56, 13 May 2023 (UTC)

Here is one more: -- Mikeblas (talk) 09:01, 13 May 2023 (UTC)


 * These are not bugs, these are internationalizing links. There's no reason why you should impose a UK google interface to someone that prefers a Canadian one. &#32; Headbomb {t · c · p · b} 11:23, 13 May 2023 (UTC)
 * The bug is that Citation Bot damages the article by breaking references. Nobody notices because the work the bot does isn't carefully monitored, so it's up to human editors to eventually find the problems and fix them. Not sure who "you" is in your example -- can you explain what you mean for me, please? -- Mikeblas (talk) 18:31, 13 May 2023 (UTC)


 * Skipping pages with Template:Sfn could be a fix for this. Enhancing999 (talk) 05:24, 14 May 2023 (UTC)

journal=SpringerLink
SpringerLink is a distribution plaftorm similar to ScienceDirect. SpringerLink should never be added to anything. &#32; Headbomb {t · c · p · b} 13:11, 26 May 2023 (UTC)
 * And journal (or magazine, newspaper, periodical, website, work) should never be added to  or an encyclopedia citation (, when it has encyclopedia, or ) ...
 * —Trappist the monk (talk) 13:21, 26 May 2023 (UTC)

Crelle's journal

 * In case this may not be obvious from the diff: Crelle's journal = Journal für die reine und angewandte Mathematik. Also the linked url for the article in question clearly states "Band 32" (= Volume 32), both in the displayed metadata and spelled-out on the facsimile title page of the journal. —David Eppstein (talk) 05:57, 28 May 2023 (UTC)
 * The volume number is 1846 in the publisher provided meta-data https://api.crossref.org/v1/works/10.1515/crll.1846.32.119 AManWithNoPlan (talk) 15:30, 28 May 2023 (UTC)
 * And https://www.degruyter.com/document/doi/10.1515/crll.1846.32.119/html so yeah, bad data. AManWithNoPlan (talk) 15:33, 28 May 2023 (UTC)
 * This is not a mistake, they like that meta-data https://www.degruyter.com/journal/key/crll/html AManWithNoPlan (talk) 15:44, 28 May 2023 (UTC)
 * Degruyter can make up whatever metadata they want, but Citation Bot should not run around copy/pasting it into Wikipedia, clobbering Wikipedians' directly expressed (more correct) preferences. –jacobolus (t) 16:02, 28 May 2023 (UTC)
 * Interestingly enough, the older ones do have issues. Page 119 starts issue 2 https://www.digizeitschriften.de/id/243919689_0032%7Clog19?tify=%7B%22pages%22%3A%5B4%5D%2C%22pan%22%3A%7B%22x%22%3A0.465%2C%22y%22%3A0.726%7D%2C%22view%22%3A%22info%22%2C%22zoom%22%3A0.511%7D AManWithNoPlan (talk) 23:46, 28 May 2023 (UTC)
 * Yes, the facsimile title page from the diff already states that it consists of four issues. Degruyter also claims ownership of the article and charges US$42 to see it despite that it has long been out of copyright. I don't even see how they can reasonably be called "the publisher"; at this point, they are just a copyright-squatter. —David Eppstein (talk) 23:51, 28 May 2023 (UTC)
 * I have versions for sale for half the pice of Degruyter. :-) AManWithNoPlan (talk) 00:24, 29 May 2023 (UTC)

I do not like spaces

 * That diff has no random spaces &#32; Headbomb {t · c · p · b} 10:52, 28 May 2023 (UTC)
 * There is a near universal agreement that new bot added content should match existing content, but with two exceptions. |x=y| is very hard to read and new things will be |x=y |.  |x = y| is very hard to read and new things will be | x=y |. AManWithNoPlan (talk) 19:29, 29 May 2023 (UTC)

Bogusremovalofspaces
Bad CrossRef Meta-Data. I have flagged it on the page. Thank you for noticing. AManWithNoPlan (talk) 12:53, 1 June 2023 (UTC)

Problem with citations
On June 4, 2022, the Citation Bot made a change in an article I am working on in my Sandbox (Sandbox 1), which destroyed many of the citations that had been legible before (see the difference between day before: and the day after.

While there were a few red-link problems the day before, most of the citations for the most part then were legible--but afterward most are now a red-letter mess! I went through a lot of trouble to create a special kind of Harvard notes that are able to show several different references simultaneously. I want to enter the new material from my sandbox into the main article, but I want to make sure that I have citations that look right. Could someone please help me by either reverting the citations to their previous state, or find suitable corrections. Perhaps someone could also help me fix whatever red-letter citations were there before? I know I should have let you know about this earlier, but I have not had much time over last couple of years. Admittedly, I am a writer, not tech person. I would greatly appreciate somebody's help. Thanks, Garagepunk66 (talk) 21:54, 5 June 2023 (UTC)


 * The red links are a result of errors. The bot simply made them obvious: they should have been there before.  In fact, sometimes when I load the previous page they are there and sometimes not.  I do not understand.  AManWithNoPlan (talk) 22:39, 5 June 2023 (UTC)
 * Thank you so much with your kind assistance--you really saved to day for me!!! Garagepunk66 (talk) 22:59, 5 June 2023 (UTC)

Auth
what exactly happens?

PayWall
Courtesy pings to,, who both suggested this bot edit. Flibirigit (talk) 10:57, 12 June 2023 (UTC)
 * The versions you used are proxy links, which will work for no one but you. They don't work for anyone else. "authentication failure" just means you haven't logged in the site properly because newspaperarchives.com is a paywalled service. &#32; Headbomb {t · c · p · b} 11:12, 12 June 2023 (UTC)
 * The same error persists even when logged in. Please explain in more detail. Flibirigit (talk) 12:40, 12 June 2023 (UTC)
 * Please note, I get the same error whether logged in or logged out of Wikipedia or newspaperarchive.com. I fail to get any article with link made by the bot. Flibirigit (talk) 12:44, 12 June 2023 (UTC)
 * Try them now. I've removed "access." from them. &#32; Headbomb {t · c · p · b} 12:51, 12 June 2023 (UTC)
 * The bot now fixes that also. AManWithNoPlan (talk) 14:18, 12 June 2023 (UTC)
 * I've tried the changes logged in and out, using different browsers, and it works each time now. Thanks for the changes. Flibirigit (talk) 16:44, 12 June 2023 (UTC)

Invalid category in edit summary
On 6 December 2022, Citation bot performed this edit to the Perfect (Exceeder) article, with an invalid category in the edit summary: Category:Singlechart usages for Wallonia Dance (missing space). The correct spelling is Category:Single chart usages for Wallonia Dance. I found neither categories in the article's edit history, after, or immediately leading up to said edit; only found in the edit summary. &#8212;&#160;CJDOS,&#160;Sheridan,&#160;OR&#160;(talk) 00:15, 13 June 2023 (UTC)
 * ❌ It was valid when the job was run. https://en.wikipedia.org/w/index.php?title=Category:Singlechart_usages_for_Wallonia_Dance&action=edit&redlink=1 AManWithNoPlan (talk) 00:19, 13 June 2023 (UTC)
 * Thank you. Seeing the category red linked, I copied it into Wikipedia's search box, thinking I would be taken straight to the page as if I had clicked on the red link itself (a 'go here' assumption). Thus, I wasn't presented with the information you provided. My mistake, sorry. &#8212;&#160;CJDOS,&#160;Sheridan,&#160;OR&#160;(talk) 00:45, 13 June 2023 (UTC)
 * We appreciate any and all bug reports. Even those that turn out to be nothing.  Much better than the "The bot has been making this mistake for ten years and have reverted it 37 times and I have finally decided to mention it" complaints.  AManWithNoPlan (talk) 12:55, 13 June 2023 (UTC)

JSTOR raw links
That should be in addition to the link, but yes. Do you have a diff? &#32; Headbomb {t · c · p · b} 22:13, 18 June 2023 (UTC)
 * Can be tested on Rufus Elefante and Tammany Hall. &#32; Headbomb {t · c · p · b} 22:16, 18 June 2023 (UTC)

"Accept Terms and Conditions on JSTOR" is a bad title and the URL is salvage garbage
Accept Terms and Conditions on JSTOR should be marked as a bad title and replaced, and links which are like url should be changed to https://www.jstor.org/stable/pdf/resrep29468.8.pdf or https://www.jstor.org/stable/resrep29468.8. This is caused by Citoid in ViualEditor being run on that pdf-link to generate an autofilled reference. See https://en.wikipedia.org/w/index.php?title=Women%27s_rights_in_Saudi_Arabia&diff=prev&oldid=1160807260 Jonatan Svensson Glad (talk) 22:26, 18 June 2023 (UTC)
 * Currently it breaks citations by removing all information in the URL after the question mark, see https://en.wikipedia.org/w/index.php?title=Swedish_iron-ore_industry_during_World_War_II&diff=prev&oldid=1117090534 Jonatan Svensson Glad (talk) 22:45, 18 June 2023 (UTC)
 * AManWithNoPlan (talk) 01:03, 19 June 2023 (UTC)

Redundant "journal" added instead of converting "work"
In this edit, the bot added a journal parameter to a citation that already had the journal as "work", and introduced a violation of MOS:NOITALIC. Perhaps the fact that everything was hidden with "script" and "trans" was confusing the bot. I have tried to fix this by converting to script-journal and trans-journal. —Kusma (talk) 07:07, 18 June 2023 (UTC)


 * Something to do with script-work and trans-work. &#32; Headbomb {t · c · p · b} 18:44, 18 June 2023 (UTC)
 * by adding trans-work detection. AManWithNoPlan (talk) 15:11, 19 June 2023 (UTC)

journal and series
Seems to be GIGO, since it is not a journal, but a book series. I have fixed the citations AManWithNoPlan (talk) 15:01, 19 June 2023 (UTC)

Citation dated added to L. K. Madigan is one day off

 * That's not a bug in Citation bot. The page source has:
 * "dateCreated": "2011-02-24T03:58:03Z"
 * That "Z" at the end is "Zulu time", UTC, like Wikipedia talkpage timestamps.Your browser may have rendered it as a local timezone. Citation bot took it exactly as stated. –  . Raven  .talk 06:54, 22 June 2023 (UTC)

lost stuff
can you specify what text exactly. i cannot tell on my phone. AManWithNoPlan (talk) 11:19, 23 June 2023 (UTC)
 * The bot only added things, nothing was removed. &#32; Headbomb {t · c · p · b} 12:00, 23 June 2023 (UTC)
 * The long-ago added wrong PMID sure made a mess of things though. AManWithNoPlan (talk) 12:28, 23 June 2023 (UTC)

Remove publisher when journal=publisher
Likewise for website=publisher, magazine=publisher, work=publisher, etc... &#32; Headbomb {t · c · p · b} 17:52, 25 June 2023 (UTC)
 * The bot already does this, when it has good reason to prefer one over the other. The problem is that often it is journal that needs removed. AManWithNoPlan (talk) 18:00, 25 June 2023 (UTC)
 * I've yet to see such a case. &#32; Headbomb {t · c · p · b} 18:13, 25 June 2023 (UTC)

Random source title
Bug: This bot putting title name randomly. for url: https://www.jstor.org/stable/44147540

activity diff;

1. Special:MobileDiff/1162321446

2. Special:MobileDiff/1157359979

MaxA-Matrix&#128172; talk 13:21, 28 June 2023 (UTC)


 * ❌ - that is the correct title for AManWithNoPlan (talk) 13:33, 28 June 2023 (UTC)
 * solved, it was actually the missing key of JSTOR.
 * Thanks MaxA-Matrix&#128172; talk 13:36, 28 June 2023 (UTC)

dates

 * That is a problem when the website reports its published date as the last time is was published including small changes like reviews.  AManWithNoPlan (talk) 12:56, 28 June 2023 (UTC)

Request for removal of feature
For, "work" is a shorter alternative for "newspaper". This is more of an unnecessary feature than a bug report. The additional five characters don't make much of a difference on short pages, but on long pages citing hundreds of news articles they add up. Space4Time3Continuum2x (talk) 11:35, 25 June 2023 (UTC)
 * The difference is that newspaper is clear, and work is not, and we favour clarity. &#32; Headbomb {t · c · p · b} 11:37, 25 June 2023 (UTC)
 * Who are "we"? Quoting another editor: "|work= and its aliases all do exactly the same thing, so the aliases exist only to (1) add a degree of unnecessary complication to citation coding, as if editors needed more complication of anything, and (2) create more things for editors to argue about, as we see in this discussion. That's just poor system design in my opinion, and poor system design should be corrected. If I had my way, the aliases would be deprecated and the citation bot would be modified to convert all existing occurrences of the aliases to |work=." Space4Time3Continuum2x  (talk) 12:20, 25 June 2023 (UTC)
 * The quote above comes from.
 * —Trappist the monk (talk) 12:35, 25 June 2023 (UTC)
 * This really a battle for the CS1/CS2 talk pages. AManWithNoPlan (talk) 13:41, 28 June 2023 (UTC)

magazine=Wired publisher=Wired
Same for the other magazines. &#32; Headbomb {t · c · p · b} 22:22, 30 June 2023 (UTC)

Missing page number
Could you write a special url/doi parsing for these, where the last thing after the final dot in 10.1103/.83.084019 gets recognized as the article id / page number? &#32; Headbomb {t · c · p · b} 01:39, 2 July 2023 (UTC)

Glucagon-like peptide-1 receptor agonist
One of the journal citation templates shows an error message because a PMC you added, 10303335, is greater than 1030000. Achmad Rachmani (talk) 03:39, 3 July 2023 (UTC)
 * I think this error is related to the template rather than the BOT, so it might be better to report it at the WT:CS1. --SilverMatsu (talk) 04:07, 3 July 2023 (UTC)
 * notabug, but bug in template outdated sanity checks. AManWithNoPlan (talk) 12:40, 3 July 2023 (UTC)

Bot will not expand if existing title is garbage
The diff showed the changes I saved, the Bot added Series, Volume & ISBN, and usefully SCID. I suspect that using harvnb as a title parameter to a cite book with a chapter isn't correct, though it cuts down repetition, when multiple chapters from conference proceedings or the like are referenced. If there's a better way, please let me know. Springer providing DOIs allowed the Bot to work. Not processing the reference if the title is a harv (or sfn?) template would be nice. RDBrown (talk) 12:28, 3 July 2023 (UTC)


 * Harnvb for title is exceedingly wrong. This is not a bug, this is parameter (and harvnb) misuse. &#32; Headbomb {t · c · p · b} 12:32, 3 July 2023 (UTC)
 * Can you suggest how an abbreviated reference can be done using citation templates? If cite document needs a journal, it would give the same issue & you need a URL for the chapter for cite web. RDBrown (talk) 13:11, 3 July 2023 (UTC)
 * —Trappist the monk (talk) 13:27, 3 July 2023 (UTC)
 * —Trappist the monk (talk) 13:27, 3 July 2023 (UTC)

TNT title/chapter=SpringerLink or SpringerLink - Journal Article or SpringerLink – Resource Secured
SpringerLink/Springerlink, -/–

&#32; Headbomb {t · c · p · b} 14:15, 3 July 2023 (UTC)

Verifying citations?
In the unlikely event that you are looking to diversify CBOT's scope, you may "enjoy" this a short OpEd by (London) Observer columnist, Prof John Naughton. He gives a nice name check to Wikipedia and our verification standards. But what worries me is fake but convincing citations: if I see a statement cited to Nature, I am almost certain to assume good faith. No doubt you have heard of the legal case where the lawyer relied on ChatGPT to find "suitable" precedents that turned out to be complete fabrications. We can't just take the easy assumption that anything published by Nature is ipso facto reliable without first checking that they did in fact publish it! Is that ever likely to be possible?--𝕁𝕄𝔽 (talk) 16:54, 5 July 2023 (UTC)


 * That's really not related to the bot, though I suppose if you add a spurious citation to nature it won't be able to find it's DOI, and if you use a fake DOI, it won't match the information found on the page, so it's easily detectable so long as you follow the sources. &#32; Headbomb {t · c · p · b} 17:11, 5 July 2023 (UTC)


 * The bot has flagged some bogus refs and I have manually cleaned the articles. AManWithNoPlan (talk) 17:52, 5 July 2023 (UTC)
 * it's easily detectable so long as you follow the sources True, but would it occur to anyone that they needed to do it, unless the cited claim was so off the wall that you would feel obliged to verify that the article did indeed say that. I had in mind more insidious cases. I have no doubt, for example, that the precedents generated by ChatGPT were sufficiently convincing to take in a professional lawyer.
 * That's great news. Hopefully it won't become an overwhelming problem. I would be astonished if we don't already have articles entirely or at least substantially written by an LLM, complete with lots of beautifully formatted citations and reaching GA standard at first try. --𝕁𝕄𝔽 (talk) 19:32, 5 July 2023 (UTC)


 * Flag as, since it is not really the direct job of the bot. AManWithNoPlan (talk) 23:38, 5 July 2023 (UTC)

series cleanup
Seems to be a problem with linked "journals/series" &#32; Headbomb {t · c · p · b} 19:41, 5 July 2023 (UTC)

Camel case and italics
I notice very often that CamelCase words get weird/random spaces and italics sometimes. What gives? &#32; Headbomb {t · c · p · b} 14:20, 1 July 2023 (UTC)


 * That is this code. It is because the original written was in this field.  AManWithNoPlan (talk) 01:34, 2 July 2023 (UTC)

function restore_italics (string $text) : string { // tags often go missing around species names in CrossRef if (str_ireplace(array('arxiv', 'ebay', 'aRMadillo', 'imac'), '', $text) !== $text) return $text; // Words with capitals in the middle, but not the first character return safe_preg_replace('~([a-z]+)([A-Z][a-z]+\b)~', "$1 $2", $text); }
 * Aggressively working on. Flagged as fixed since significant progress is being made AManWithNoPlan (talk) 21:16, 4 July 2023 (UTC)

One time run... journal = Advances in Cryptology
These are book series. If you could have a one-off run TNTing chapter/title/journal on those citations, that would be a huge boon to cleanup efforts.

The affected pages are

• #A5/2

• #Accumulator (cryptography)

• #Ali Baba and the Forty Thieves

• #Alternating step generator

• #Amit Sahai

• #Birthday attack

• #Blind signature

• #Block cipher

• #Boolean function

• #Boomerang attack

• #Brent Waters

• #Center for Research on Computation and Society

• #Cryptographic multilinear map

• #Cryptography

• #DRE-i with enhanced privacy

• #Dan Boneh

• #Data Encryption Standard

• #Decoding methods

• #Deterministic encryption

• #Diffie–Hellman key exchange

• #DigiCash

• #Dual EC DRBG

• #Efficient Probabilistic Public-Key Encryption Scheme

• #Elliptic-curve cryptography

• #Feistel cipher

• #Fiat–Shamir heuristic

• #FourQ

• #Garbled circuit

• #Group signature

• #HElib

• #Hashcash

• #Helios Voting

• #Homomorphic secret sharing

• #Homomorphic signatures for network coding

• #Hugo Krawczyk

• #Indistinguishability obfuscation

• #Information-theoretic security

• #International Data Encryption Algorithm

• #Key generator

• #Lattice problem

• #Lattice-based cryptography

• #MD4

• #Michael J. Freedman

• #Microsoft SEAL

• #Niederreiter cryptosystem

• #Non-interactive zero-knowledge proof

• #Non-malleable code

• #Non-repudiation

• #Open vote network

• #Ouroboros (protocol)

• #Paillier cryptosystem

• #Post-quantum cryptography

• #Prince (cipher)

• #Proof of work

• #Provable security

• #Pseudorandom permutation

• #RSA (cryptosystem)

• #Ran Canetti

• #Randomness test

• #Ring signature

• #SHA-2

• #SM9 (cryptography standard)

• #Searchable symmetric encryption

• #Seny Kamara

• #Structured encryption

• #Subset sum problem

• #TWINKLE

• #Taher Elgamal

• #Talk:CryptoNote

• #Trapdoor function

• #Unbalanced oil and vinegar scheme

• #User:BillHPike/sandbox

• #User:Ilmari Karonen/sandbox/SIV mode

• #User:Matěj Grabovský/sandbox

• #Verifiable random function

• #Verifiable secret sharing

• #Zero-knowledge proof

• #Zvika Brakerski

&#32; Headbomb {t · c · p · b} 19:15, 6 July 2023 (UTC)


 * fixed. AManWithNoPlan (talk) 21:34, 6 July 2023 (UTC)
 * Not fully... 73 pages still have some type of journal=Advances in Cryptology ... &#32; Headbomb {t · c · p · b} 22:00, 6 July 2023 (UTC)
 * , took multiple passes. But all done now. AManWithNoPlan (talk) 11:40, 7 July 2023 (UTC)

Caps: npj
Old bug report.

This should affect all journals starting with npj (linked or not). &#32; Headbomb {t · c · p · b} 03:34, 9 July 2023 (UTC)

last1/first1 harmonization
For some reason the bot missed that one. &#32; Headbomb {t · c · p · b} 03:16, 9 July 2023 (UTC)

Camel case and italics, part 2
And I have a trouble ticket with CrossRef to see if they can fix this. AManWithNoPlan (talk) 13:15, 10 July 2023 (UTC)

Better handling of ScienceDirect books
TNT'd into a cite book with the isbn= taken from the URL (sciencedirect.com/book/9780128173428/... ; sciencedirect.com/science/article/pii/B9780128161371000052). &#32; Headbomb {t · c · p · b} 13:31, 7 July 2023 (UTC)
 * If you don't TNT, you need to have some cleanup tweaks (+) &#32; Headbomb {t · c · p · b} 13:50, 7 July 2023 (UTC)

&#32; Headbomb {t · c · p · b} 13:49, 7 July 2023 (UTC)
 * insource:/url *= *https?:\/\/(www\.)?sciencedirect\.com\/book\/978/ will reveal the first URL type
 * insource:/url *= *https?:\/\/(www\.)?sciencedirect\.com\/science\/article\/pii\/B978/ will reveal the second URL type


 * I will work this, even though it is Friday Night.  AManWithNoPlan (talk) 01:41, 8 July 2023 (UTC)
 * Well, you still have your choice of It's Friday, Friday, and many others. &#32; Headbomb {t · c · p · b} 01:43, 8 July 2023 (UTC)

Bot keeps giving me problems
Hello. I had already been here because the bot was adding a number on the citation as the issue when it was already cited as the volume. I was told to add the note. The same is happening again with another citation. Is there a solution to this, or would I have to just pay attention to my watchlist to see which of the articles I wrote are getting their citations altered?

At this point though I'm also questioning whether I know the difference between a volume and an issue. Super Dromaeosaurus (talk) 08:09, 11 July 2023 (UTC)


 * Since most of wikipedia calls it "volume", I have added code to make the bot prefer that to "issue". Some journals are loose with "volume" vs "issue".  Especially those that are not in English.  AManWithNoPlan (talk) 12:26, 11 July 2023 (UTC)
 * Thank you so much. Super Dromaeosaurus (talk) 12:43, 11 July 2023 (UTC)
 * - flag for archiver. AManWithNoPlan (talk) 12:54, 11 July 2023 (UTC)
 * Just to make life entertaining, some journals and (especially) magazines bind a year's worth of issues into a volume. So, for example, we might have volume=2023, issue=July. 𝕁𝕄𝔽 (talk) 13:00, 11 July 2023 (UTC)

What is the policy on Google books links?
Recently I noticed that the bot changes links to a book that is available on Google books from the link to the title page that I put in, to a link that gives a number of snippets inside the book. Example from Jacques-Louis Comte de Noyelles My intent  became. (Please click the links to see the difference in result.) The difference is that &source=gbs_navlinks_s was eliminated, but what results is very unhelpful: a number of disjointed snippets. Now, I suppose I am to file a bug report, but I am a tech ignoramus. I don't know how to provide useful information to the repair people, except by providing this example. Please help. Meanwhile I am just putting the old links back. Ereunetes (talk) 22:02, 10 July 2023 (UTC)


 * The bot normalizes links, which makes them consistent for all users. If you want to link to just a book, without search terms, then all the stuff after the Book ID should be deleted. AManWithNoPlan (talk) 01:35, 11 July 2023 (UTC)

Bot down?
Running the bot gives this No webservice

The URL you have requested, https://citations.toolforge.org/process_page.php?edit=toolbar&slow=1&page=Intelligence_amplification, is not currently serviced. If you have reached this page from somewhere else...

This URI is managed by the citations tool, maintained by AManWithNoPlan, Dbarratt , Kaldari , Mattsenate , Maximilianklein , Smith609.

That tool might not have a web interface, or it may currently be disabled.

If you're pretty sure this shouldn't be an error, you may wish to notify the tool's maintainers (above) about the error and how you ended up here. If you maintain this tool

You have not enabled a web service for your tool, or it has stopped working because of a fatal error. Please check the error logs of your web service.

&#32; Headbomb {t · c · p · b} 20:28, 11 July 2023 (UTC)


 * There is something weird with a CrossRef Meta-data change and it will be down until I fix the damaged pages, and the then the source code. AManWithNoPlan (talk) 20:36, 11 July 2023 (UTC)
 * Do you have an ETA for the fix, by curiosity? &#32; Headbomb {t · c · p · b} 20:39, 11 July 2023 (UTC)
 * I have fixed the problem, and I am now cleaning up. AManWithNoPlan (talk) 21:06, 11 July 2023 (UTC)
 * Not fully fixed. Back off. AManWithNoPlan (talk) 21:29, 11 July 2023 (UTC)

503
when i try to use citation bot via the website, it results in a 503 Service Unavailable error, why is it happening? Notrealname1234 (talk) 23:02, 13 July 2023 (UTC)


 * I received the same 503 Service Unavailable error when I attempted to use the citation bot.Skilgis1900 (talk) 23:04, 13 July 2023 (UTC)
 * Found out that the user talk says this. "A 503 error means that the bot is overloaded and you should try again later – wait at least an hour." Notrealname1234 (talk) 23:12, 13 July 2023 (UTC)


 * Thank you for the effort and the information!!Skilgis1900 (talk) 23:18, 13 July 2023 (UTC)

JSTOR with letter 'e' before numbers
That appears to be an entire issue of a journal (volume 43, issue 1, of The Great Circle). Is that really what you want to cite? If so, cite journal is not well-suited for that; it is for journal articles, not journal issues, and requires a title of an individual article in a journal. —David Eppstein (talk) 22:24, 14 July 2023 (UTC)
 * Oh, that's true...didn't notice that. Perhaps at least cite web would be fine, but with jstor-parameter.
 * seems to work.
 * --Jonatan Svensson Glad (talk) 22:50, 14 July 2023 (UTC)
 * But, I'll see what article was actually wanted to be used as a source here.... Jonatan Svensson Glad (talk) 22:51, 14 July 2023 (UTC)
 * there is a special template for referencing journal issues. i do not remember which one. AManWithNoPlan (talk) 00:20, 15 July 2023 (UTC)
 * And yes, the bot explicitly ignores the E jstors for this reason -- 99% of the time, they are wrong thing, and someone really wants an article. AManWithNoPlan (talk) 00:21, 15 July 2023 (UTC)

The bot changes "work" parameter to "newspaper"
Previous thread: User_talk:Citation_bot/Archive_35. I just came across the template page for "cite news" and noticed that the examples for citing news articles all use "work" (Template:Cite_news). Isn't that enough justification for not changing "work" to "newspaper"? Space4Time3Continuum2x (talk) 12:59, 10 July 2023 (UTC)


 * Yea I also noticed that. Still, I would prefer "newspaper" over "work". If you use Cite news, can the source be anything else than a newspaper? Take care, Manifestation (talk) 13:13, 10 July 2023 (UTC)
 * @Manifestation: Yes! Template:Cite news states "This Citation Style 1 template is used to create citations for news articles in print, video, audio or web." GoingBatty (talk) 01:59, 11 July 2023 (UTC)
 * That's strange, because further down, it reads:
 * can be used for offline (paper) sources whereas generates a missing URL error when no URL is provided
 * So it was my understanding that Cite news would be used to cite physical newspapers, and Cite web for the online stuff. Manifestation (talk) 10:54, 11 July 2023 (UTC)
 * @Manifestation: I read that sentence as "cite news can be used for offline sources whereas cite news cannot be used for offline sources", not "cite news may ONLY be used for offline sources". Note that cite news supports the url parameter. GoingBatty (talk) 00:13, 12 July 2023 (UTC)

This part if notabug AManWithNoPlan (talk) 19:23, 18 July 2023 (UTC)

Optica Publishing Group
Optica Publishing Group should be marked as a bad title. See diff.

Also, https://opg.optica.org/ome/viewmedia.cfm?uri=ome-5-11-2459&html=true should use and not. Is there a way to scrape the doi etc. from that page? Jonatan Svensson Glad (talk) 22:52, 17 July 2023 (UTC)


 * I have added it to the list of bad titles. Very strange the the DOI is not grabbed, since it is sitting right there in the META-DATA.  Zotero really misses the boat on this one.  AManWithNoPlan (talk) 13:28, 18 July 2023 (UTC)


 * AManWithNoPlan (talk) 19:24, 18 July 2023 (UTC)

Citation bot adding editor as author
In article Tikki Tikki Tembo, Citation bot added "|author1=柳田, 国男, 1875-1962 |author2=日本放送協会" at Special:Diff/1164796262 and after being reverted with a proper edit note, keeps re-adding at Special:Diff/1165812619. As the metadata in https://dl.ndl.go.jp/pid/1124189/1/1 states, 日本放送協会 is the editor (編), not author, which I already wrote in the "editor=" field along with "editor-link=NHK". How can I stop this? Note also that 柳田国男 is credited as "supervisor (監修)", not author. Wotheina (talk) 01:55, 18 July 2023 (UTC)
 * The meta data they uploaded to doi.org clearly states that this is the author. So, japanlinkcenter.org has uploaded bad data. I will modify the page to block the bot on that item.  AManWithNoPlan (talk) 13:34, 18 July 2023 (UTC)


 * AManWithNoPlan (talk) 19:24, 18 July 2023 (UTC)

Better handling of IEEE books
Achieved by TNTing title/journal (to trigger on the condition that IEEE doi (10.1109/...) or ieeexplore.ieee.org URL + ISBN?) &#32; Headbomb {t · c · p · b} 08:28, 21 July 2023 (UTC)
 * I will think about when not to trigger it also (if ever) and what all to TNT (series, title, etc). AManWithNoPlan (talk) 13:37, 21 July 2023 (UTC)

See also + tnt fix &#32; Headbomb {t · c · p · b} 18:20, 21 July 2023 (UTC)

Weird italics
These are all logged and I got back and fix them by hand. The code is now good enough that the "guesses" are now on average bad (it almost never guesses now). I have just switched the code to by default NOT apply the guess. I still log them, since sometimes they are right. Also, then I add the phrase to the list of "fix this". AManWithNoPlan (talk) 12:13, 22 July 2023 (UTC)
 * And the new and improved method for getting titles does not work with books. Thus the IEEE runs are fraught with errors.  AManWithNoPlan (talk) 12:18, 22 July 2023 (UTC)

Citation bot didn't ignore a citation with cbignore template
See edit. Izno (talk) 18:44, 23 July 2023 (UTC)

❌ - that template is for other bots. AManWithNoPlan (talk) 22:04, 23 July 2023 (UTC)

Page appears to be empty
When running the bot in a single page it says "!Page Example_page from en.wikipedia.org appears to be empty" about 5% of the time. When run again it finds the page. Abductive (reasoning) 00:53, 23 July 2023 (UTC)
 * Wikipedia seems super slow right now. Even adding the comment took long.  AManWithNoPlan (talk) 01:27, 23 July 2023 (UTC)
 * There was an outage/service slowdown, most likely due to DDoS attack. Now it should be fixed. So with this tool. Jonatan Svensson Glad (talk) 15:10, 24 July 2023 (UTC)

Nature.com PDF links
https://www.nature.com/articles/npre.2008.2088.1 and https://www.nature.com/articles/npre.2008.2088.1.pdf?proof=t has the same Nature article ID (npre.2008.2088.1), so the bot should be able to expand the PDF-link as well based on that info. See edit. Jonatan Svensson Glad (talk) 17:01, 24 July 2023 (UTC)


 * Cool. Code added and it works great.  AManWithNoPlan (talk) 17:08, 24 July 2023 (UTC)

convert ResearchGate deref link to doi
Specifially, convert https://www.researchgate.net/deref/http%3A%2F%2Fdx.doi.org%2F10.22270%2Fajprd.v7i2.463 to 10.22270/ajprd.v7i2.463. But since that was already present, remove it.

Alternatively (yuck), it could be converted to https://doi.org/10.22270%2Fajprd.v7i2.463.

&#32; Headbomb {t · c · p · b} 11:44, 25 July 2023 (UTC)

IEEE books, again
The new stuff works pretty well, but there's still a few kinks. &#32; Headbomb {t · c · p · b} 20:04, 21 July 2023 (UTC)
 * In this edit where it moved  from title to chapter it should have changed url to chapter-url (as long as the URL corresponds with the chapter title provided of course, as it did in this case). Jonatan Svensson Glad (talk) 21:39, 21 July 2023 (UTC)
 * Also, in the same edit, it should also remove IEEE International Conference on Plasma Science (ICOPS) since it added 2022 IEEE International Conference on Plasma Science (ICOPS) (Not sure what's correct though regarding the year, keeping it as year or including it in the title. Jonatan Svensson Glad (talk) 21:42, 21 July 2023 (UTC)
 * (See manual edit: https://en.wikipedia.org/w/index.php?title=Nonthermal_plasma&diff=prev&oldid=1166484750) Jonatan Svensson Glad (talk) 21:44, 21 July 2023 (UTC)
 * A lot of this comes down the the horribly redundant/duplicate CS1/2 parameters. AManWithNoPlan (talk) 21:48, 21 July 2023 (UTC)
 * More code added. AManWithNoPlan (talk) 22:04, 21 July 2023 (UTC)
 * This will really pay off in the august edition of the WP:JCW/DOI/10.1100 compilation. The cleanup is too late for this dump, but in August... yowza! &#32; Headbomb {t · c · p · b} 23:17, 21 July 2023 (UTC)

See edit. The bot is not consistent when using url or chapter-url. Also, it added 1 if using the cite conference template, but not the book template. (Another question: Should cite book really be used, and not cite conference?) Jonatan Svensson Glad (talk) 00:19, 22 July 2023 (UTC)
 * Also see edit where the bot used cite web instead (and no mention of book/conference or authors).
 * That has no DOI, so citoid is all you get. AManWithNoPlan (talk) 14:47, 22 July 2023 (UTC)
 * It can't figure the ISBN? &#32; Headbomb {t · c · p · b} 15:03, 22 July 2023 (UTC)
 * Does not seem to be able to. AManWithNoPlan (talk) 15:16, 22 July 2023 (UTC)
 * Bummer &#32; Headbomb {t · c · p · b} 15:25, 22 July 2023 (UTC)
 * Seems like a Zotero problem, since it is not hidden/encoded. AManWithNoPlan (talk) 15:33, 22 July 2023 (UTC)
 * Mind helping reporting it upstream? Jonatan Svensson Glad (talk) 15:11, 24 July 2023 (UTC)
 * I have no idea where to report such things. AManWithNoPlan (talk) 21:32, 25 July 2023 (UTC)
 * Conference proceedings are books, so either are fine. Cite book is much more standard however. Cite conference is filled with garbage. &#32; Headbomb {t · c · p · b} 00:40, 22 July 2023 (UTC)

I've been seeing (on my watchlist) the bot convert a bunch of conference papers, formatted as journal papers with the name of the conference incorrectly set in the journal parameter, into properly formatted conference papers. Thumbs up. Seems to be working well, and a useful thing to do. —David Eppstein (talk) 07:11, 22 July 2023 (UTC)


 * Ok, but Special:Diff/1166612964 is a mistake: it was a valid journal citation with an incorrect conference doi, which the bot tried to replace with the corresponding conference citation. Not really a bot bug (except for the way the citation ended up mangled as a result); the correct fix would have been to use the journal doi instead of the conference doi. Would someone else fix this, please? I have a COI. Maybe User:Headbomb, since it was your suggestion to run the bot on this article? —David Eppstein (talk) 18:15, 22 July 2023 (UTC)
 * If something's blatantly incorrect, anyone can fix it, COI or not. GIGO is GIGO. &#32; Headbomb {t · c · p · b} 19:15, 22 July 2023 (UTC)
 * I was just chided this evening for refusing to create an article on a colleague (who deserves one) because of my COI. I do make COI edits sometimes, but on this article I prefer to stay strict. That way I can justifiably say that any shortcomings of the article are not my problem. —David Eppstein (talk) 06:35, 23 July 2023 (UTC)

Proceedings of the AAAI handling
See also for other potential cleanup, though those might be much harder/impossible to automate. &#32; Headbomb {t · c · p · b} 22:21, 22 July 2023 (UTC)

None of those have DOIs, other than one that claims to be a journal. https://api.crossref.org/v1/works/10.1609/aaai.v33i01.33017627. It will require some logic at the end to just stiff-arm the citation into being a book. AManWithNoPlan (talk) 22:30, 22 July 2023 (UTC)


 * Yup. Hence why it might be impossible. And the ACM doi is fake. &#32; Headbomb {t · c · p · b} 22:35, 22 July 2023 (UTC)
 * Gotta love those 10.5555 DOIs, just like the 123456789 HDLs. AManWithNoPlan (talk) 22:44, 22 July 2023 (UTC)

Boxed sister project templates in reference section
According to WP:MOSSIS, boxed sister project templates do not belong in § References, due to the undesirable white space it creates. It also states that a § External links should not be created if a boxed sister project template is all that would populate it; instead, it allows using an inline template. Please review the recent revert to the Dairy cattle article, and update Citation bot to account for this situation. I have raised the issue with the user operating the bot responsible for the revert, offering an alternative plan which utilizes the inline template rather than the box template, and avoids creating § External links. &#8212;&#160;CJDOS,&#160;Sheridan,&#160;OR&#160;(talk) 23:14, 27 July 2023 (UTC)

Oh—I see what you mean. The citation bot was the editor prior to me, hence why it appeared in the revert edit summary. &#8212;&#160;CJDOS,&#160;Sheridan,&#160;OR&#160;(talk) 00:33, 28 July 2023 (UTC)
 * Are you sure that Citation bot was involved doing that? I can't find any diffs where Citation bot touched that template. Jonatan Svensson Glad (talk) 23:20, 27 July 2023 (UTC)
 * It's possible that's it's an inaccurate edit summary, or a misinterpretation of WP:MOSSIS. I wasn't about to revert the revert without raising the disagreement with the editor, and offering a compromise. Perhaps WP:MOSSIS due for an update... or am I giving too much leeway over the revert? &#8212;&#160;CJDOS,&#160;Sheridan,&#160;OR&#160;(talk) 00:25, 28 July 2023 (UTC)
 * Yeah, exactly. The user who reverted you restored back to the revision which Citation bot had done. Jonatan Svensson Glad (talk) 00:34, 28 July 2023 (UTC)
 * It seems I've made an ass of myself. Sorry. Thank you for your patience. &#8212;&#160;CJDOS,&#160;Sheridan,&#160;OR&#160;(talk) 03:17, 28 July 2023 (UTC)
 * Better to risk being a butt and ask than to risk a bot having a bug. AManWithNoPlan (talk) 13:30, 28 July 2023 (UTC)
 * ❌ - flag for archiving. AManWithNoPlan (talk) 13:30, 28 July 2023 (UTC)

Caps: 42nd Cospar Scientific Assembly

 * It also adds 44Th Annual Lunar and Planetary Science Conference (should be 44th) when run on http://adsabs.harvard.edu/abs/2013LPI....44.2854S Jonatan Svensson Glad (talk) 23:39, 27 July 2023 (UTC)
 * This is true in general for all of #st, #nd, #rd, #th. &#32; Headbomb {t · c · p · b} 00:05, 28 July 2023 (UTC)

Caps: NIST
NIST is not a journal so switching from to  is certainly the wrong thing to do.

The initial state of that template is junk anyway:

wherein title has all sorts of stuff that is not the article title: a date that belongs in September 30, 2016, a date that belongs in November 17, 2019, something that might go in NIST and the Nobel.

Properly the template might be rewritten like this:

—Trappist the monk (talk) 22:54, 27 July 2023 (UTC)
 * You are totally correct about the specific citation. However, this "report" was mainly regarding the fact it cited NIST as Nist, but a more in-depth look at all issues are appreciated. Jonatan Svensson Glad (talk) 22:59, 27 July 2023 (UTC)

Can't run on "Meat tenderness"
For some reason, I am unable to run the bot on Meat tenderness. Not using  nor "the gadget button" under the editor window. Jonatan Svensson Glad (talk) 23:19, 27 July 2023 (UTC)
 * It does run, just very slowly for some reason trying to verify the DOIs. The webbrowser/websserver connection gets dropped.  It does not make any changes, so it looks like it failed.  AManWithNoPlan (talk) 13:41, 28 July 2023 (UTC)
 * notabug, but annoying. AManWithNoPlan (talk) 13:45, 28 July 2023 (UTC)

Moved URLs
Seems like he bot can't expand from  anymore? The link redirects to https://www.pnas.org/doi/full/10.1073/pnas.1420949112 - anyway we can "follow" the redirect and attempt to expand from the new source (if not dead). Jonatan Svensson Glad (talk) 23:20, 24 July 2023 (UTC)
 * It does seem that way. Not sure why Citoid/Zotero fails.   Refill will re-write urls and do it.  But as always be careful and make sure that it does not rewite it into https://SuperSpamDomain.com/pnas.org or https://www.pnas.org/404.html  AManWithNoPlan (talk) 00:22, 25 July 2023 (UTC)
 * notabug in our code, so wontfix AManWithNoPlan (talk) 23:40, 28 July 2023 (UTC)

Bot making cosmetic-only edits

 * bot needs stopping please. Nikkimaria (talk) 02:01, 30 July 2023 (UTC)
 * New clean-up added, but did not get non-consmetic done right. AManWithNoPlan (talk) 02:03, 30 July 2023 (UTC)

Multiple caps

 * Algae
 * Code4Lib
 * CovertAction Quarterly
 * JCO Clinical Cancer Informatics
 * Kunstforum International/Kunstforum
 * MedPage
 * MIT Sloan Management
 * PEN International
 * Saga-Book

&#32; Headbomb {t · c · p · b} 18:53, 31 July 2023 (UTC)

Multiple caps (2)
&#32; Headbomb {t · c · p · b} 19:04, 31 July 2023 (UTC)
 * Cathair na Mart
 * Pain Practice
 * Revista Universum

Multiple caps (3)
&#32; Headbomb {t · c · p · b} 19:01, 31 July 2023 (UTC)
 * Babesch
 * Bilim ve Teknik [and all 've']
 * D-Lib Magazine
 * Deutsches Ärzteblatt International
 * GeoArabia
 * GeoJournal
 * IT Professional
 * JCI Insight
 * Kinema
 * NASSP Bulletin
 * Nederlands Tijdschrift voor Geneeskunde [and all 'voor']
 * Time Off
 * USSR
 * Zeitschrift für ausländisches öffentliches Recht und Völkerrecht

SpringerLink clean-up
However in the last diff above, I removed the content of the parameter, but left an empty title before running the bot. For some reason the bot added an empty chapter when adding the title again (see https://en.wikipedia.org/w/index.php?title=User%3AJosve05a%2Fsandbox&diff=prev&oldid=1168204340) - why? Jonatan Svensson Glad (talk) 11:50, 1 August 2023 (UTC)
 * The Meta-Data gives a chapter I assume, but then the Bot during clean up probably notices that it and the title are the same (or something similar) and removes it. AManWithNoPlan (talk) 13:31, 1 August 2023 (UTC)
 * but it then forgets to remove the stray empty parameter, which should also be cleaned up in that case? Jonatan Svensson Glad (talk) 13:50, 1 August 2023 (UTC)

Issues fetching titles for journals on springer link
Temporary issue, or something changed? Jonatan Svensson Glad (talk) 01:41, 29 July 2023 (UTC)
 * I removed this section earlier since I thought it was temporary, but now it is happening on like very other article I try to use the bot. Running the bot on : https://en.wikipedia.org/w/index.php?title=Diseases_and_epidemics_of_the_19th_century&diff=prev&oldid=1168190851.
 * Why is only Springer-links affected? Can we fetch title through Crossref instead or something? Jonatan Svensson Glad (talk) 09:25, 1 August 2023 (UTC)
 * https://en.wikipedia.org/w/index.php?title=Religious_intellectualism_in_Iran&diff=prev&oldid=1168192138 Jonatan Svensson Glad (talk) 09:55, 1 August 2023 (UTC)
 * I think the issue is more general, and is due to 10.1023/A:... type of DOIs having bad metadata. &#32; Headbomb {t · c · p · b} 12:09, 1 August 2023 (UTC)
 * The Bot could use Zotero/Citoid on the springer link URL for the title. http://link.springer.com/10.1023/A:1012973118615 since the DOI has no title data.  Currently, if the DOI "works", then Citoid is not used, since it is usually worse date (and in this case, there is no explicity URL)  AManWithNoPlan (talk) 13:29, 1 August 2023 (UTC)

should citation bot modify 'doi-broken-date'?
This edit seems like a bug. Isn't the whole point of 'doi-broken-date' to identify the earliest known time when the DOI was broken? Arbitrarily changing it to a later value seems like it removes whatever useful information that attribute contained. –jacobolus (t) 16:07, 1 August 2023 (UTC)
 * It only updates it, if it is over six months since it has last been checked. About every 9 to 12 months, I run a special job on all broken DOI pages and update the dates.  That is what just ran.  Anyway, the data is a "last checked" date, and not a "first seen" date.  AManWithNoPlan (talk) 16:10, 1 August 2023 (UTC)
 * I will add that the six months is because the bot used to update the date with all edits, and that got a bit disruptive, since basically anytime the bot was run, the page got edited. AManWithNoPlan (talk) 16:12, 1 August 2023 (UTC)
 * Maybe the citation documentation should be changed then. It says: If the doi link is broken, then use of doi-broken-date unlinks the doi value, indicates when the doi-problem was first noticed, and will also add the page to "CS1 maint: DOI inactive as of Mmmm YYYY" (tracking category Category:CS1 maint: DOI inactive) –jacobolus (t) 16:18, 1 August 2023 (UTC)
 * That documentation is very wrong. The de-linking was removed a long time ago.  AManWithNoPlan (talk) 16:26, 1 August 2023 (UTC)
 * Is there an explanation in documentation someplace of what doi-broken-date is supposed to do then? I had never heard of it until now, went to look it up, and it seemed like citation bot behavior didn't match the documented purpose. –jacobolus (t) 16:33, 1 August 2023 (UTC)
 * All that parameter does is flag the DOI as broken (as of date X). &#32; Headbomb {t · c · p · b} 17:10, 1 August 2023 (UTC)
 * the documentation. AManWithNoPlan (talk) 19:56, 1 August 2023 (UTC)

Months and ISSN
I've noticed that Citoid adds months (and sometimes incorrectly dates, by specifying the first day of a specific month) as well as ISSN for journals, which is not something that Citation bot does (it only adds years and no ISSN). Are there any best practices if months should be specified for journals or not (it is available through Crossref for many journals) as well as ISSN for publications? Jonatan Svensson Glad (talk) 07:14, 1 August 2023 (UTC)


 * ISSNs are effectively useless identifiers that do not identify the article being cited. There are best omitted. &#32; Headbomb {t · c · p · b} 07:19, 1 August 2023 (UTC)
 * Perhaps a question for the template's talk page (or the CS1/CS2 talk page) but why even display them at the end like doi's? Should be directly after the publication imo, like " in my opinion, since it is the publication's identifier, and not the article's. But, II digress... Jonatan Svensson Glad (talk) 07:23, 1 August 2023 (UTC)
 * The ISSN is useful when it is the only identifier and the journal/magazine/newspaper is obscure. But, it seems to me that almost universally, the ISSN is used for well known things like WSJ and NYT or with full plat of DOI/PMID/S2CID/PMC/etc.  It is rare for it to be helpful.  AManWithNoPlan (talk) 13:33, 1 August 2023 (UTC)
 * Citation Bot does not trust months and do not add ISSN, since it is junk (other than a few very specific cases). ❌  AManWithNoPlan (talk) 17:52, 1 August 2023 (UTC)

Add publisher to cite book
We ignore that data, since it seems to be not very reliable. No idea when this code was added, or why. AManWithNoPlan (talk) 13:38, 1 August 2023 (UTC) // Possibly contains dud information on occasion // $this->add_if_new('publisher', str_replace("___", ":", $xml->dc___publisher));


 * GitHub is back up and see that this comment is at least 8 years old. AManWithNoPlan (talk) 13:47, 1 August 2023 (UTC)


 * Maybe limit to cases where it can find an ISBN from the google book link? &#32; Headbomb {t · c · p · b} 15:38, 1 August 2023 (UTC)
 * probably if (GoogleHasISBN and Ref does not have a PMID/PMC/DOI/arxiv/journal/magazine) AManWithNoPlan (talk) 16:15, 1 August 2023 (UTC)

volume/issue with supplements
I'm not sure if having the supplement in the issue value is correct or not, but it seemed weird that the bot added in the parameter without removing it from the volume name, this could potentially be a GIGO scenario, not certain though. --Lightlowemon (talk) 05:14, 1 August 2023 (UTC)
 * Putting it in issue is normal, but yes it should be removed from volume if so. &#32; Headbomb {t · c · p · b} 07:10, 1 August 2023 (UTC)