Help talk:Citation Style 1
(Redirected from Category talk:CS1 Chechen-language sources (ce))
Latest comment: 50 minutes ago by David Eppstein in topic ?utm_source=chatgpt.com
| This is the talk page for discussing improvements to the Help:Citation Style 1 and the CS1 templates page. |
|
Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99Auto-archiving period: 20 days  |
 | To help centralize discussions and keep related topics together, the talk pages for all Citation Style 1 and Citation Style 2 templates and modules redirect here. A list of those talk pages and their historical archives can be found here. |
 | This help page does not require a rating on Wikipedia's content assessment scale. It is of interest to multiple WikiProjects. | |||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||
|
CS0
editWe have a slew of CS1-adjacent templates for identifiers. For purpose of documentation/style, I propose that we call those CS0 style. Specifically,
And possibly others from Template:Catalog_lookup_link#See also.
We could then bring error checking and other features from Module:Citation/CS1, which could share documentation and code, thereby facilitating maintenance etc...
We'd mirror the category scheme, so we'd have, for example
and the same for other categories, like Category:CS1 errors and its subcategories.
These would effectively have the same documention, and we'd just change "Citation Style 1 and Citation Style 2" to "Citation Style 0, Citation Style 1 and Citation Style 2" "CS1|2" to "CS0|1|2".
Thoughts? Headbomb {t · c · p · b} 16:43, 29 May 2025 (UTC)
- I object to "CS0" solely on the grounds that normal humans do not start counting at zero and "CS0" does not enlighten a casual reader. Let's not make this place look even more like a programmers-only exclusive club. I wouldn't be averse to a set of parallel categories with more human-friendly names like "Citation identifier templates: XXX errors". – Jonesey95 (talk) 16:52, 29 May 2025 (UTC)
- Lets clarify what CS1 and CS2 are, before we decide if we should call this set of tools "CS0".
- CS1 by default requires the user to specify what kind of source it is (book, web, journal, etc.) and by default separates the elements with periods.
- CS2 by default auto-detects the kind of source based on which parameters have values and which don't, and by default separates the elements with commas.
- In printed style guides, comma separators are typical of footnotes and endnotes.
- In printed style guides, period separators are typical of alphabetical bibliographies.
- In Wikipedia, endnotes predominate but period separators also predominate.
- The choice between CS1 and CS2 seems to be mostly based on whether the early editors of an article wanted to auto-detect the kind of source, with no concern about whether commas or periods were used.
- Considering what a mish-mash this is, I'm not sure we can make a sensible statement about what CS0 means. For me, the reasoning for this term must be all about making it more understandable for editors. If it's all about making the organization of the coding of the templates and modules easier, then it isn't a style at all; it's something like "citation utility template group" (CUTG). Jc3s5h (talk) 17:05, 29 May 2025 (UTC)
For purpose of documentation and categories
- CS1 templates are all the {{cite book}}, {{cite journal}}, {{cite magazine}}, etc. templates, with a full list here
- CS2 has one template: {{citation}}
For all documentations purposes and categorization purposes, CS1 and CS2 are identical. The only difference is one uses a period for delimiter (with a final period), the other uses a comma (with no final period).
- CS1: Baggaley, W. Jack (2000). "Advanced Meteor Orbit Radar observations of interstellar meteoroids". Journal of Geophysical Research: Space Physics. 105 (A5): 10353–10361. Bibcode:2000JGR...10510353B. doi:10.1029/1999JA900383.
- CS2: Baggaley, W. Jack (2000), "Advanced Meteor Orbit Radar observations of interstellar meteoroids", Journal of Geophysical Research: Space Physics, 105 (A5): 10353–10361, Bibcode:2000JGR...10510353B, doi:10.1029/1999JA900383
- You can make a CS1 template behave as CS2, and vice versa, by using
|mode=cs1or|mode=cs2. User:BrandonXLF/CitationStyleMarker.js is a useful script if you want to know at a glance which style is used in an article.
What I'm proposing here is that for purpose of coding/documentation/categorization/error messages, we call CS0 those semi-templated citations that invoke those catalog lookup templates, and that they share code and documentation with CS1/2 templates when possible. If CS0 offends you, call it CS3 (or CS Platypus or whatever). Headbomb {t · c · p · b} 17:27, 29 May 2025 (UTC)
- Templates like {{doi}} do not implement a citation style, which is what "CS" stands for. An actual descriptive name would help both readers and editors. – Jonesey95 (talk) 17:40, 29 May 2025 (UTC)
- I know it's not a style. Not the point. The point is to unify and streamline the codebase, documentation, categories, etc. Headbomb {t · c · p · b} 17:44, 29 May 2025 (UTC)
- If it's not a style, don't call it a style. Jc3s5h (talk) 18:51, 29 May 2025 (UTC)
- I know it's not a style. Not the point. The point is to unify and streamline the codebase, documentation, categories, etc. Headbomb {t · c · p · b} 17:44, 29 May 2025 (UTC)
Returning to the nub of this discussion, there are issues that would need to be addressed somehow. Mostly it is the variety of parameters and options supported by the identifier templates:
{{arxiv}}– takes a single value; has support for|archive=parameter; the parameter is documented as deprecated and not apparently used in mainspace but is still supported in the template{{bibcode}}– takes a single value{{biorxiv}}– takes a single value{{citeseerx}}– host appears to be currently dead (502 bad gateway) – has support for up to nine identifiers; supports|type=with valuesdoiandpid{{doi}}– takes a single value{{hdl}}– takes a single value; supports|hdl-access=valuesfree,limited,registration,subscription{{isbn}}– has support for up to nine identifiers; supports|plainlink=,|link=,|leadout=,|invalid1=..|invalid9=,|template_name={{issn}}– has support for up to nine identifiers; supports|plainlink=,|link=,|leadout=,|invalid1=..|invalid9={{jfm}}– has support for up to nine identifiers; supports|leadout={{jstor}}– takes a single value; supports|stable=,|sici=,|issn={{lccn}}– takes a single value; supports|title=,|name=,|long=; uses Module:LCCN{{medrxiv}}– takes a single value{{mr}}– has support for up to nine identifiers; supports|leadout={{oclc}}– has support for up to nine identifiers; supports|leadout=,|show={{osti}}– takes a single value{{pmc}}– takes a single value{{pmid}}– has support for up to nine identifiers; supports|plainlink=,|leadout={{ssrn}}– takes a single value{{zbl}}– has support for up to nine identifiers; supports|leadout=
—Trappist the monk (talk) 19:42, 29 May 2025 (UTC)
- All of them should be brought in line with how they behave in CS1|2 templates. Corner cases can be either handled seperatly and offloaded to seperate templates (like multiple MRs/PMIDs being handled by a seperate templates).
- I also believe doi, hdl, both support
|<identifier>-access=. bibcode, jstor, osti, ssrn should support it too. Headbomb {t · c · p · b} 02:07, 30 May 2025 (UTC)- I've updated {{arxiv}}, {{biorxiv}}, {{citeseerx}}, and {{pmc}} to display green access locks by default (and recreated {{medrxiv}}). I've also updated {{bibcode}}, {{doi}}, {{hdl}}, {{jstor}}, {{osti}}, and {{ssrn}} to support
|<identifier>-access=free. Headbomb {t · c · p · b} 03:07, 30 May 2025 (UTC) - I've been poking at a module in my sandbox that supports, in whole or in part, all of the above named identifier templates. Some of the current templates use
{{Catalog lookup link}}. That template supports up to nine identifiers and the parameters|leadout=,|link=, and|plainlink=. The sandbox module supports these parameters and essentially unlimited numbers of identifier-values for all of the identifier templates. - Some identifier templates have other special features/parameters that are not supported by the sandbox module:
{{citeseerx}}– host seems to be mostly dead; most often returning 502 gateway errors. When marginally alive, doesn't seem to recognize dois in the form 10.x.x.x (where 'x' is some number of digits). Supports an undocumented parameter|type=which acceptsdoiandpidas values. Used in <5 articles; those specifying|type=pidappear to work when the host is working;pidtype identifiers not supported by cs1|2{{hdl}}– besidesfree, supports|hdl-access=valueslimited,registration, andsubscription; these parameter values do not appear to be used{{ISBN}}– supports:|invalidn== used in ~120 articles; can be replaced with accept-as-written markup((..))if rendered with the sandbox module|template_name=not documented; used to identify the template calling Module:Check isxn (a cs1|2 derived module to do error checking)
{{ISSN}}– supports|invalidn=used in ~10 articles; can be replaced with accept-as-written markup((..))if rendered with the sandbox module{{JSTOR}}– supports:|stable=used in ~5 articles; alias of{{{1}}}|sici=does does not appear to be used; cannot be used with{{{1}}}or|stable=|issn=used in <5 articles; cannot be used with{{{1}}}or|stable=or|sici=|no=does not appear to be used; alias of|issn=
{{lccn}}– has its own Module:LCCN; supports:{{{2}}}(a title or label) used in ~10 articles|long=used in ~15 articles
{{OCLC}}– supports|show=; used in ~270 articles; when used, WorldCat requires registration to view results
- Some testing of the sandbox module can be seen in my sandbox (permalink).
- Templates not currently supported by the sandbox module but might be are:
- Certainly the sandbox module can be used to transparently upgrade these templates:
- With a documentation tweak,
{{hdl}}can be upgraded. To upgrade{{isbn}}and{{issn}}(and{{ismn}}and{{sbn}}?) we must replace|invalidn=in instances of those templates; a relatively minor task. - That leaves us with these:
{{citeseerx}}{{jstor}}{{lccn}}{{oclc}}
- If we are to proceed with the notion of consolidating these identifiers with the sandbox module, what to do about these four.
- I suppose the more important question is: Should we consolidate these templates so that the supported templates use the cs1|2 module suite?
- —Trappist the monk (talk) 18:28, 16 June 2025 (UTC)
- These five templates were noted above as not supported by the sandbox module. All are now supported in whole or in part:
- Of them, two are problematic:
{{asin}}– used in ~4700 articles.- Of those:
- supports:
{{OL}}–used in ~140 articles- supports Internet Archive
ia:...identifiers (not currently used in any articles; not supported by cs1|2; the identifier can be converted to an Internet Archive url:{{OL|ia:workplacemanagem00onot}}→ OL ia:workplacemanagem00onothttps://archive.org/details/workplacemanagem00onot→ https://archive.org/details/workplacemanagem00onot
- supports Internet Archive
- My sandbox (permalink) has been updated to include example renderings of these five templates.
- —Trappist the monk (talk) 17:40, 19 June 2025 (UTC)
- Are we going to do anything with this? Should I create Module:Identifiers (or perhaps Module:CS1 identifiers would be a better name)?
- —Trappist the monk (talk) 23:06, 13 July 2025 (UTC)
- I've updated {{arxiv}}, {{biorxiv}}, {{citeseerx}}, and {{pmc}} to display green access locks by default (and recreated {{medrxiv}}). I've also updated {{bibcode}}, {{doi}}, {{hdl}}, {{jstor}}, {{osti}}, and {{ssrn}} to support
- As you've been updating these, shouldn't {{arxiv}}/{{cite arxiv}} support
|version=? (and|version-date=) -- 65.93.183.249 (talk) 15:15, 21 June 2025 (UTC)
No. If you're citing arXiv:0704.0001v1 specifically, then you use
- Balázs, C.; Berger, E. L.; Nadolsky, P. M.; Yuan, C. -P. (2 April 2007). "Calculation of prompt diphoton production cross sections at Tevatron and LHC energies". arXiv:0704.0001v1 [hep-ph].
If you're citing arXiv:0704.0001v2, then you use
- Balázs, C.; Berger, E. L.; Nadolsky, P.; Yuan, C.-P. (27 July 2007). "Calculation of prompt diphoton production cross sections at Fermilab Tevatron and CERN LHC energies". arXiv:0704.0001v2 [hep-ph].
Headbomb {t · c · p · b} 15:56, 21 June 2025 (UTC)
- I have created Module:CS1 identifiers and used it to replace Module:Doi in
{{doi}}. I will pick away at the other identifier templates. - —Trappist the monk (talk) 00:02, 30 July 2025 (UTC)
Picking away at this list; those with a 
mark have been converted to use Module:CS1 identifiers. No doubt, no doubt, template and category documentation needs improvement.
—Trappist the monk (talk) 22:44, 31 July 2025 (UTC)
- {{ProQuest}} ? —David Eppstein (talk) 00:33, 4 August 2025 (UTC)
- There is no cs1|2 parameter
|proquest=. - —Trappist the monk (talk) 17:35, 4 August 2025 (UTC)
- How about {{IETF RFC}}, {{Listed Invalid ISBN}}, {{doi-inline}}? --FlatLanguage (talk) 23:41, 12 August 2025 (UTC)
- I fixed {{Listed Invalid ISBN}}. --FlatLanguage (talk) 09:38, 24 August 2025 (UTC)
- How about {{IETF RFC}}, {{Listed Invalid ISBN}}, {{doi-inline}}? --FlatLanguage (talk) 23:41, 12 August 2025 (UTC)
- There is no cs1|2 parameter
Option to suppress full stop (period) at end of Title & other title-like parameters
editAgreement to make this so appears to have been reached back in 2013: of course a lot has changed since then!
Anyway, the issue resurfaced here (quite possibly, ofc, elsewhere, over the years?). How about it?
To reduce server load, I wouldn't suggest auto-detection of !, ?, ... => suppression, but rather creating a new parameter along the lines of |title-period=no
— Protalina (talk) 10:48, 29 July 2025 (UTC)
- This is a long time tracked feature request. That page probably doesn't get enough use. Izno (talk) 19:56, 21 August 2025 (UTC)
Citations wrapped in <cite>
edit
Currently, Wikipedia's citation templates wrap each citation in the <cite> tag. This is an improper use of the <cite> tag and should likely be replaced by <span> (although I don't really care what we replace it with, so long as we stop using it in this way).
According to the HTML specification, The
The spec then gives an example of a citation where cite element represents the title of a work.... The cite element is a key part of any citation in a bibliography, but it is only used to mark the title...cite is used to wrap the title of the work being cited, but not the whole citation.
I think the current state of affairs may have come about from there being a different description of the cite element's purpose in the W3C's specification back when they maintained a separate one, although I haven't checked to make sure. Regardless of the original reason, it seems clear that this should be corrected. Mr. Starfleet Command (talk) 06:50, 6 August 2025 (UTC)
- History:
<cite>...</cite>implemented at this edit per this discussion:
- preceding discussions (organized by namespace and archive number):
- Template talk:Citation/core/Archive 5 § Problem...
- Template talk:Citation/core/Archive 6 § Cite element, take two
- Template talk:Citation/core/Archive 12 § Separating content from formatting.
- MediaWiki talk:Common.css/Archive 15 § Restore default style for HTML cite element
- MediaWiki talk:Common.css/Archive 16 § Compensate for italic lean
- MediaWiki talk:Common.css/Archive 17 § The cite element needs to not auto-italicize any longer
- Module talk:Citation/CS1/Archive 3 § Optimization suggestions
- Help talk:Citation Style 1/Archive 6 § Lua module and css presentation
- Help talk:Citation Style 1/Archive 8 § HTML5 bait-and-switch: The cite element again
- —Trappist the monk (talk) 15:49, 6 August 2025 (UTC)
- Yes, that is how the affairs occurred. That said, I am of the mind simply to ignore the WHATWG's decision here. Izno (talk) 20:00, 21 August 2025 (UTC)
- When we wrapped the whole citation in
<cite>...</cite>tags in 2015, this was the state of the HTML5 recommendation that we were following. We appear to have been in compliance at that point. It looks like they changed their minds at some point in the last ten years, invalidating many formerly valid implementations. I am not privy to the conversations that led to a revision in the guidance, but I am not inclined to worry about it here. They will probably change the guidance again at some point. – Jonesey95 (talk) 20:27, 21 August 2025 (UTC)- The FAQ may illuminate. There's plenty elsewise about the element if you just google "cite WHATWG" or similar. Basically "it make italics so we make it title only" was it. Which totally ignores non-Latin language titles and totally ignores titles of what one might call lesser works such as chapters, sections, articles, and papers. Izno (talk) 20:51, 21 August 2025 (UTC)
- That FAQ contains essentially the opposite explanation of our 2015 discussion. The HTML gods in 2009 to 2013 tried to make
<cite>...</cite>apply only to titles, and HTML developers apparently revolted, so the HTML folks relented. Then, at some point, according to the FAQ, they made the same change again, possibly while nobody was looking, and I guess there wasn't a revolt, or developers had stopped listening, or something. A tempest in a teapot, followed by a calm teapot. Let's just have some tea and work on something else. – Jonesey95 (talk) 21:00, 21 August 2025 (UTC)- The some point is when the W3C ceded all authority to deciding how HTML works to WHATWG. Izno (talk) 00:36, 22 August 2025 (UTC)
- That FAQ contains essentially the opposite explanation of our 2015 discussion. The HTML gods in 2009 to 2013 tried to make
- The FAQ may illuminate. There's plenty elsewise about the element if you just google "cite WHATWG" or similar. Basically "it make italics so we make it title only" was it. Which totally ignores non-Latin language titles and totally ignores titles of what one might call lesser works such as chapters, sections, articles, and papers. Izno (talk) 20:51, 21 August 2025 (UTC)
- @Izno: May I ask what we gain by going against the spec? As a matter of principle, there's nothing wrong with an element for titles of works, and cite is just such an element. Although the name of the element and its original purpose seem to go against this, I see no compelling reason to use the element as was prescribed by a now-nonexistent specification.
- Of course, that's not to say that we need to start using the cite element for titles of works, at least for the moment; that's a separate issue. I just don't see what we gain by using it for citations. Mr. Starfleet Command (talk) 01:26, 27 August 2025 (UTC)
- Maybe they'll change their mind. Maybe I'm feeling pissy about the stupidity of the decision they made. Maybe I don't want to fix the two dozen places that also need sorting and a few more. Who's to know?
there's nothing wrong with an element for titles of works
I did not advance this opinion. I said how they came to the conclusion that this element should be used for titles was dumb. Izno (talk) 01:38, 27 August 2025 (UTC)Maybe they'll change their mind.
I find this extremely unlikely, considering they've stuck with the current definition for seventeen years.Maybe I'm feeling pissy about the stupidity of the decision they made.
Perfectly understandable, but this isn't a strong argument for going against the current standard.Maybe I don't want to fix the two dozen places that also need sorting and a few more.
I sympathize, but this is hardly an insurmountable roadblock. I'd be happy to fix any of those that aren't protected.[H]ow they came to the conclusion that this element should be used for titles was dumb.
Fair enough; it wasn't my intention to misrepresent your argument, and I apologize for inadvertently doing so.- I ask again: What do we gain by going against the HTML specification? Mr. Starfleet Command (talk) 23:58, 30 August 2025 (UTC)
- That's stating the question back to front. I'm neutral on if this is necessary (it's not my area), but the real question is what do we gain from making the change and is it worth the effort required? It's the typical question of any such work. -- LCU ActivelyDisinterested «@» °∆t° 00:11, 31 August 2025 (UTC)
- The other cost, beside the effort, is that some software may expect us to continue doing things the way we have been doing them and break if we change. I don't know this to be the case but it wouldn't surprise me. —David Eppstein (talk) 00:16, 31 August 2025 (UTC)
- That's stating the question back to front. I'm neutral on if this is necessary (it's not my area), but the real question is what do we gain from making the change and is it worth the effort required? It's the typical question of any such work. -- LCU ActivelyDisinterested «@» °∆t° 00:11, 31 August 2025 (UTC)
- When we wrapped the whole citation in
More generic cite titles
editI have collected some more generic cite titles related to bot protection. Below are the ones with the most results:
- Client Challenge (88 results), used by Fastly Client Challenge
- You are being redirected... (331 results), used by Sucuri WAF
- ShieldSquare Captcha (129 results), used by Radware (formerly ShieldSquare) Bot Manager
There are also other ones with only a couple results, if anyone wants them. OutsideNormality (talk) 18:38, 16 August 2025 (UTC)
module suite update 30–31 August 2025
editI propose to update the cs1|2 module suite over the weekend 30–31 August 2025. Here are the changes:
- maint cat to track
{{cite journal}}templates misusing|page=for|article-number=; discussion - fix url access parameter application on wikipedia library url errors; discussion
Module:Citation/CS1/Configuration:
- add script-lang tkr;
- change color of maint warning message in preveiw box from #3a3 to #085 to match cs1|2 maint message color;
- tweak tabular data fetch; discussion
- maint cat to track
{{cite journal}}templates misusing|page=for|article-number= - add 10.1016/j.patter to free DOI prefix recognition (Patterns); 10.9778 to free DOI recognition (CMAJ Open)
- maint cat for post 2007 arxiv format without
|class=; see discussion - add 'updated' as bad author name; discussion
- fix url access parameter application on wikipedia library url errors
Module:Citation/CS1/Date validation:
- make
|date=Jan 0500,|date=0999–1000,|date=090–100error, accept|date=Jan 500; discussion - make
|date=2000-10-01,|year=2010error; discussion
Module:Citation/CS1/Identifiers:
- maint cat for post 2007 arxiv format without
|class= - tweak wikidata identifier article name fetch; discussion
- fix 8-digit only medRxiv test; discussion
—Trappist the monk (talk) 19:16, 24 August 2025 (UTC)
- Before this is rolled out, could we get support for
|proquest=? Usage is certainly high enough to warrant it. Headbomb {t · c · p · b} 09:55, 25 August 2025 (UTC)
Also
- Add 10.11158/saasp to free DOI prefix recognition (Systematic and Applied Acarology Special Publications) [1]
- Add 10.11646/megataxa to free DOI prefix recognition (Megataxa) [2]
- Add 10.11646/mesozoic to free DOI prefix recognition (Mesozoic) [3]
- Add 10.22073/pja to free DOI prefix recognition (Persian Journal of Acarology) [4]
- Add 10.37520/fi to free DOI prefix recognition (Fossil Imprint) [5]
- Add 10.53562/ajcb to free DOI prefix recognition (Asian Journal of Conservation Biology) [6]
- Add 10.35249/rche to free DOI prefix recognition (Revista Chilena de Entomología) [7]
Headbomb {t · c · p · b} 08:43, 29 August 2025 (UTC)
To note for future reference, I posted to WP:VPT about this update after people requested announcements for updates to this module. I don't know if it resulted in anybody dropping by to check over changes, though, Rjjiii (talk) 17:21, 31 August 2025 (UTC)
Revisiting https
editWe talked about detecting "hhttps" and similar malformed URL starters in 2023 and before that in 2020, but the module was not changed to detect them.
There appear to be about 100 articles using |url=hhttp at this time. Is it worth detecting these? – Jonesey95 (talk) 13:06, 25 August 2025 (UTC)
- It's probably worth detecting known misspellings, but not flagging every scheme you don't recognize; new ones are added periodically. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 13:49, 25 August 2025 (UTC)
DOI, title, URL and Citation bot
editThe |url= parameter causes the title to link to the provided URL, which |DOI= does not. If you specify both |doi= and |url=, Citation bot removes |url=, leaving a link from the DOI but not from the title. Is that a bug or a feature? If the bot is legitimate then there should be a warning. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 18:58, 26 August 2025 (UTC)
- Typically a DOI url (
|url=https://doi.org/10..../...) that matches the value assigned to|doi=(|doi=10..../...) is hidden behind a paywall and is redundant so I can understand why Citation bot would delete the|url=parameter. As to whether Citation bot should perform that deletion is not a topic for this venue. - Are you suggesting that redundant
|doi=and|url=parameters should emit some sort of maintenance message when they match? - When
{{cite journal}}has a|doi=that links to a free-to-read copy of the source, setting|doi-access=freewill cause{{cite journal}}to link|title=to https://doi.org/10..../... as if|url=https://doi.org/10..../...had been set. This automatic process yields to|url=with an assigned value. - —Trappist the monk (talk) 19:14, 26 August 2025 (UTC)
- I'm suggesting both that the combination of
|doi=and|url=parameters should emit some sort of maintenance message when they match, and that the|DOI=and|URL=documentation warn that|URL=may be removed by a bot if both are specified. -- Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:31, 26 August 2025 (UTC)- The community is opposed to the bot removal of such redundant URLs, sadly. Only when
|doi-access=freeis set should citation bot remove the doi.org url|url=. At least as of the last RFC on the issue. Headbomb {t · c · p · b} 22:20, 26 August 2025 (UTC)
- The community is opposed to the bot removal of such redundant URLs, sadly. Only when
- I'm suggesting both that the combination of
AI-generated sources
editPlease see Wikipedia_talk:WikiProject_AI_Cleanup#Publifye_AS. Citation templates should emit an error (similar to Help:CS1_errors#generic_name) when an author's name contains "AI" (e.g. [8][9]). This should preferably be a new category so that it is easier to swiftly remove AI slop from articles. Kovcszaln6 (talk) 11:52, 27 August 2025 (UTC)
- Since this is Kovcszaln6's suggestion, I ask that Kovcszaln6 investigate whether all author-like parameters such as author & editor accept accept-as-written notation. The name of a corporate author might contain the word "AI". Jc3s5h (talk) 12:18, 27 August 2025 (UTC)
Extend Category:CS1 maint: ignored DOI errors to other identifiers, e.g. Category:CS1 maint: ignored Bibcode errors
editTo track bypassed bibcodes like the one in Betty Klepper (Bibcode:2008AgrJ..100..-53M).
This should also be extended to the CS0 efforts above too. Headbomb {t · c · p · b} 04:19, 30 August 2025 (UTC)
|id= doi-inline emits extra punctuation?
editCompare
- Naveca, Felipe Gomes; et al. (2021). "COVID-19 in Amazonas, Brazil, was driven by the persistence of endemic lineages and P.1 emergence". Nature Medicine. 27 (7): 1230–1238. doi:10.1038/s41591-021-01378-7. PMID 34035535. Preprint.
with
- Naveca, Felipe Gomes; et al. (2021). "COVID-19 in Amazonas, Brazil, was driven by the persistence of endemic lineages and P.1 emergence". Nature Medicine. 27 (7): 1230–1238. doi:10.1038/s41591-021-01378-7. PMID 34035535. Preprint.
Headbomb {t · c · p · b} 13:38, 31 August 2025 (UTC)
- From your example:
{{doi-inline|10.21203/rs.3.rs-33554|Preprint}}→[https://doi.org/10.21203%2Frs.3.rs-33554 Preprint]
- The trailing semicolon is seen by cs1|2 as extra punctuation. The
was added to{{doi-inline}}at this edit. Perhaps the better place for that is immediately before the green lock file link. - —Trappist the monk (talk) 14:12, 31 August 2025 (UTC)
?utm_source=chatgpt.com
editWe should add a maintenance category when the URL of a citation includes ?utm_source=chatgpt.com and similar AI-generated URLs. I think it would help editors better vet AI-generated sections. Maybe it would make sense to have a bot add the {{AI-generated}} tag when it detects those particular tracking parameters? Snowman304|talk 20:48, 31 August 2025 (UTC)
- It's caught by User:Headbomb/unreliable. But I wouldn't want gnomes removing the tag without removing the AI slop that came with it. —David Eppstein (talk) 21:38, 31 August 2025 (UTC)