Wikipedia talk:Citing sources

  1. Home
  2. Wiki

WMF seeking feedback on Reference Check

[edit]

Hi all! Relatively few newcomers know, let alone remember, to cite the content they are adding. In response, the WMF's Editing team has been working on developing Reference Check, a feature which prompts new editors to add citations before they publish an edit adding content to an article. We're hoping to bring it to English Wikipedia soon with an initial A/B test so that we can evaluate its impact and address any issues that arise. As the first stage of that, we wanted to reach out to gather your input on the feature and experiment's design.

Background

Reference Check is the first of a new set of features called Edit Checks, which aim to:

  1. Give newcomers guidance while they are in the process of making an edit to help them abide by Wikipedia's policies and guidelines.
  2. Reduce the amount of effort and attention experienced editors need to allocate toward addressing preventable damage.

In this case, it aims to address the problem that only 19% of edits adding content by new editors (those with ≤100 edits) include a reference. It has been in development since 2023, and is now deployed on every other language edition except for English.

How it works

Screenshot of Reference Check
Reference Check

Reference Check activates when a new editor adds a large amount of text in VisualEditor (on desktop or mobile) in an edit and clicks on publish without adding a citation. It creates a notice (see screenshot) prompting the editor to add a citation. If they choose to do so, it opens up the "Add a citation" box, and if they decline, it asks them to specify why, which is then recorded (and we intend will eventually be available for other editors to review).

Our research from other languages shows a positive impact: In an A/B test, editors who were shown the check were more than twice as likely to make an edit that included a reference, as well as less likely to be reverted and more likely to stick around and keep editing.

There have been several prior discussions where some editors have expressed interest in deploying it here, and we're now looking to follow up on those and complete the deployment in the near future.

Try it out

To test Reference Check, follow these steps:

  1. Go to https://en.wikipedia.org/w/index.php?title=Moon&veaction=edit&ecenable=1 (or any other article in VisualEditor with &ecenable=1 added to the URL)
  2. Create a new paragraph that is at least 50 characters long without adding a citation.
  3. Press the "Publish changes…" button.
  4. Interact with the prompt that appears.

Be sure not to click "Publish changes" (without the ...) from the "Save your changes" screen so that you don't actually publish the test edit.

Call for feedback

We’re hoping to run an A/B test showing the feature to a sample of newcomers who have made ≤100 edits so that we can evaluate its impact and address any issues that arise. We're interested to know what you think of this approach. Overall, does the check seem like something that would benefit the project? Is there anything you'd want us to look for or keep in mind during the test? (The current metrics we plan to track are documented on Phabricator.) We also continue to be open to more general feedback if there's anything else that testing it out brings to mind.

We'd also like to highlight that there are Community Configuration settings available for Reference Check, which allow any admin to adjust things like how many characters need to be added for the check to activate. Feel free to discuss existing settings options that you'd prefer, or to let us know if there are additional settings you'd like us to introduce.

We'll follow up after gathering input with next steps. And thanks as always for your collaboration!

Cheers, Sdkb-WMF talk 18:01, 25 September 2025 (UTC)[reply]

Notified: WP:Village pump (proposals), WT:Teahouse, WT:WikiProject Usability, Help talk:Referencing for beginners, WT:Reliable sources, WT:Verifiability. Sdkb-WMF talk 18:10, 25 September 2025 (UTC)[reply]
It looks for one citation per paragraph, which is perhaps the best that can be easily managed and certainly better than nothing, but will often be inadequate.
The four options offered when you decline to add a citation are insufficient. In particular, expanding from a preexisting source is quite common. Ideally when a new paragraph is used a NAMEDREF should be repeated, but newcomers may not expect that and an option that provides the opportunity to easily reuse an existing reference after an initial no is worth considering.
Some articles will not have any lead citations, bit of an edge case and maybe not worth the bother of an extra option but rewriting of leads is sometimes done by new users, especially if the existing one is self-evidently inadequate.
The other option offers no chance for further explanation which might be useful if for no other reason the data collection.
Because Moon is semi'd the instructions won't work for unregistered users, I employed [1] courtesy of Special:Random, but may be worth adjusting the instructions. 184.152.65.118 (talk) 19:18, 25 September 2025 (UTC)[reply]
Thanks for sharing these thoughts; they are all very helpful!
For making it easier to reuse references, we'll discuss this and follow up with any updates.
For articles without lead citations, one of the Community Configuration options, ignoreLeadSection, deactivates the check there.
For the other option, the team had initially decided not to prioritize that because of the complexity it'd introduce, but it remains a possible future enhancement. @PPelberg (WMF) has created phab:T405683 to document that work.
Cheers, Sdkb-WMF talk 22:50, 25 September 2025 (UTC)[reply]
It shouldn't trigger on disambiguation pages or redirects, which do not need references and asking for them may confuse new editors. Thryduulf (talk) 21:23, 25 September 2025 (UTC)[reply]
Probably also pages tagged as a set index, as for example many chemical compounds are 50 characters long even without an accompanying description. Also applies equally to certain other pages that are purely navigational like lists of lists though implementation may be too difficult in practice and users are allowed to select no when queried as whether a reference is needed so maybe not that big a deal though data is limited. 184.152.65.118 (talk) 22:28, 25 September 2025 (UTC)[reply]
The check will not run on list items, so it shouldn't show on disambiguation pages. The newly-added paragraph has to be a root-level paragraph. ESanders (WMF) (talk) 11:15, 26 September 2025 (UTC)[reply]
This seems like a very useful tool, thank you to all those involved in making it. Out of curiosity, what does it look like when an editor chooses not to add a reference? Is the edit tagged with the reason they gave? Toadspike [Talk] 22:33, 26 September 2025 (UTC)[reply]
Great questions, @Toadspike. Responses below. Please let me know what (if any) new questions this brings to mind or leaves unanswered...
...what does it look like when an editor chooses not to add a reference?
When someone elects not to add a reference, Reference Check will ask them to express why (see screenshot).
Reference Check mobile decline survey
Is the edit tagged with the reason they gave?
At present, no. The reason someone gave is not visible on-wiki. Although, we are considering re-introducing[1] this functionality in the future via T405132.
And I'm glad to know you see promise in the tool!
---
1. Emphasis on "re-introducing" because early in the development of Reference Check it was possible to see on-wiki why someone declined to add a Reference. Although, this tagging approach lost its meaning when it became possible for multiple Checks (of the same or different types) to become activated within a single edit. PPelberg (WMF) (talk) 23:32, 30 September 2025 (UTC)[reply]
Thank you, PPelberg. I think it would be really great if we could make the reason visible to other editors. Toadspike [Talk] 05:01, 1 October 2025 (UTC)[reply]
Yay to A/B testing this!
One of the recurring concerns is to what share of new contributions are improved vs what share are introducing more subtle mistakes. I dont expect too many problematic subtle mistakes here, but getting a list of edits in the two groups and allowing the community to compare might be nice. —Femke 🐦 (talk) 16:53, 15 October 2025 (UTC)[reply]
...getting a list of edits in the two groups and allowing the community to compare might be nice.
Good call. And you know what, I think generating the lists of edits you described above will be possible using the Edit Check edit tags that are already in place...
editcheck-references: ought to return a list of edits that meet the conditions for Reference Check to be shown had it been enabled
editcheck-references-shown: ought to return a list of edits in which Reference Check was actually shown to someone in the course of publishing an edit
@DLynch (WMF): can you please confirm the above is accurate? And assuming David confirms the above to be true, @Femke do these two tags sounds like they'll offer the info. you think would be helpful for volunteers to see?
Yay to A/B testing this!
We're happy to know the prospect of this is exciting to you too ^ _ ^ PPelberg (WMF) (talk) 17:09, 15 October 2025 (UTC)[reply]
Yes, that looks correct. (And thus, effectively, editcheck-references + editcheck-references-shown on the same revision should mean a revision where someone was asked to add a reference and chose not to.) DLynch (WMF) (talk) 17:15, 15 October 2025 (UTC)[reply]
Yes, that should work! —Femke 🐦 (talk) 20:10, 15 October 2025 (UTC)[reply]

Update

[edit]

Hi all! Based on the discussion above, we're planning to move ahead with the A/B test. The test is scheduled to run from Wednesday, 5 November 2025 to approximately 17 December 2025, after which we'll take some time to analyze the results, share them with you all, and decide with you how/if to enable the feature. Thanks for all your feedback above! We'll continue to monitor this thread, so if anything else comes to mind or you notice anything during the test, please let us know! Cheers, Sdkb‑WMF talk 21:31, 3 November 2025 (UTC)[reply]

In preparation for this, my colleague @DLynch (WMF) has suggested some community configuration settings that you might want to adopt. Any admin can implement them if they look good to you — courtesy pinging those from the discussion above, @Thryduulf, @Femke, and (not an admin when you commented above but one now!) @Toadspike. Cheers, Sdkb‑WMF talk 04:45, 4 November 2025 (UTC)[reply]
The synopsis section also doesn't require sources (as the work itself is considered the source). So that might be one to add too. —Femke 🐦 (talk) 07:58, 4 November 2025 (UTC)[reply]
Agree, we'll want some synonyms like "Plot" as well. Toadspike [Talk] 08:55, 4 November 2025 (UTC)[reply]
@Thryduulf, Femke, and Toadspike: Those settings (or whatever other ones you decide to adopt — the community configuration settings are yours to customize how you see fit!) are all fine by us. Feel free to implement them at will on the MediaWiki page. Cheers, Sdkb‑WMF talk 15:41, 5 November 2025 (UTC)[reply]
And like I said in that suggestion on the message's talk page, I'm happy to make the actual edit for you if you don't feel comfortable adding the JSON yourself -- I just don't want to step on anyone's toes by unilaterally running ahead with that. DLynch (WMF) (talk) 22:10, 5 November 2025 (UTC)[reply]
I've created the page MediaWiki:Editcheck-config.json. Please let me know if I've done it correctly. I'll add to it if/when I come across more synonyms. Am I correct in assuming that it's not case-sensitive? Toadspike [Talk] 15:36, 6 November 2025 (UTC)[reply]
Yeah, the section matching is case-insensitive. DLynch (WMF) (talk) 16:00, 6 November 2025 (UTC)[reply]
I'm assuming there's a way to exclude disambiguation pages and set index articles? More complicated will be name pages, which seem to use a wide variety of templates. Toadspike [Talk] 15:51, 6 November 2025 (UTC)[reply]
It should hopefully not show up on those pages, since (per Ed above) it doesn't trigger on list items. As a backup, we are exploring ways to suppress the feature on pages based on category (see phab:T347775), but that hasn't been implemented yet. Cheers, Sdkb‑WMF talk 22:16, 6 November 2025 (UTC)[reply]
Awesome. Are tables also excluded? I assume they are not "root level paragraphs". Toadspike [Talk] 07:34, 7 November 2025 (UTC)[reply]
That's right, table cell contents won't be included. DLynch (WMF) (talk) 17:55, 7 November 2025 (UTC)[reply]

Early results

[edit]

Hi all! I wanted to share that we have published some early data from the test; feel free to check it out on MediaWiki and let us know if you have any thoughts! We'll follow up next month to share the final report on the test once it's ready. Cheers, Sdkb‑WMF talk 23:05, 12 December 2025 (UTC)[reply]

Automated citation generation problems

[edit]

Somewhere there must be a bit of software that generates {{cite news}} citations from newspapers online, but it has a weakness. Here's a typical example (from Operation Prone). 

{{Cite news|url=https://www.nytimes.com/1988/08/09/world/aug-20-cease-fire-8-year-iran-iraq-war-southern-africa-pact-set-too-angola-truce.html|title=AUG. 20 CEASE-FIRE IS ON IN 8-YEAR IRAN-IRAQ WAR: SOUTHERN AFRICA PACT SET, TOO; Angola Truce Now|last=Times|first=Robert Pear, Special To The New York|date=9 August 1988|work=The New York Times|access-date=2017-07-11|language=en-US|issn=0362-4331}}</ref>

It looks to me as though it sees ", Special to the New York Times" in the 'author' slot of the source, and assumes from the comma that it's in (last, first) format, and then makes "Times" the last name and everything preceding that the first name (this clearly isn't a fully correct explanation though). There are many similar examples (not just for the NYT). Does anyone know anything about this? Is it just on older citations, or are they still being generated? Can we do anything about it? Colonies Chris (talk) 10:13, 26 September 2025 (UTC)[reply]

Yes it has to guess a bit, as this is not details provided by the article, so guessing is all it can do. —TheDJ (talk • contribs) 09:01, 1 October 2025 (UTC)[reply]
To be clear, the "Special to The New York Times" part shouldn't be in the 'author' tag of the original article, and isn't for newer NYTimes articles, but older articles are often shoved into a more modern straight jacket that doesn't account for all the variance in the history of the paper's publishing. —TheDJ (talk • contribs) 09:08, 1 October 2025 (UTC)[reply]
What software is this - is it something written for Wikipedia, or provided by an archival website? If it's ours, could it be tweaked to spot the phrase ", Special to xxxxx" at the end of the 'author' and deal with it more appropriately when building the citation? Colonies Chris (talk) 09:15, 1 October 2025 (UTC)[reply]
@Colonies Chris, good question. I believe the tool that generates the citations is mw:Citoid, which in turn uses Zotero to "translate" website metadata from a given URL into the appropriate fields.
Since Zotero is open-source, I believe the way to address this issue would be to go to its platform and try to fix it there. Sdkb talk 22:22, 5 October 2025 (UTC)[reply]
There's more than one tool doing this. WhatamIdoing (talk) 18:34, 11 October 2025 (UTC)[reply]
Taking a broader view, I think we do a terrible job of surfacing the fact that this is what's happening under the hood, which in turn makes it harder for people to report issues like the one you found. I'd be interested to hear from you and any editors with experience with Zotero: Are there places we could add this info? I'm thinking of something like a small notice in the editor after it autogenerates a citation with "Report a problem with this citation's formatting" that'd lead to somewhere that'd let editors work on fixing it. Sdkb talk 22:23, 5 October 2025 (UTC)[reply]
The problem is in the way that news outlet chooses to set up their HTML (the 'author' tag). Here's what any tool is starting with for that article:<meta data-rh="true" name="byl" content="By Robert Pear, Special To the New York Times"/> <a href="https://www.nytimes.com/by/robert-pear" class="last-byline css-ojhyzr e1jsehar0" itemProp="name">Robert Pear, Special To the New York Times</a>
With the website falsely claiming that this author has a name that's eight words long – and which a human, but not a bot, can see contains non-name information – there's not a lot that can be done by the tools, especially if you don't want to hand-code a thousand exceptions for all the different ways that all the different websites screw up their metadata. WhatamIdoing (talk) 18:42, 11 October 2025 (UTC)[reply]
Yes, it's not practical to try to catch all exceptions, but there are some frequently encountered and easily identified patterns such as this one that could surely be fixed rather than propagated into our articles? It's common enough that there are hundreds of articles with citations such as the one I quoted. Colonies Chris (talk) 10:58, 14 October 2025 (UTC)[reply]
Another common pattern is with items in the Google news archive, such as this one:
{{Cite web |title=Spokane Daily Chronicle - Google News Archive Search |url=https://news.google.com/newspapers?id=C5NYAAAAIBAJ&sjid=PvgDAAAAIBAJ&pg=6184,1800236&dq=long-island+golf+wright&hl=en |access-date=2022-11-23 |website=news.google.com}}
where the 'title' has been populated by the name of the publication, suffixed by 'Google News Archive', instead of the actual article title, which is nowhere to be seen, and the 'website' is somewhat misleadingly given as Google instead of the publication - and probably Google News Archive should be in the 'via' parameter. Colonies Chris (talk) 11:34, 14 October 2025 (UTC)[reply]
@WhatamIdoing, as tedious as hand-coding exceptions to screwed up metadata is, I am with Chris in that it's still better than the alternative: hand-coding every citation to a site with screwed up metadata. And that latter task is much larger, since there will be e.g. hundreds of citations screwed up with the "Special to the New York Times" issue and ditto for every other issue. If we could easily point people to a system that'd allow them to write translators to fix these issues, and make that system easy enough to use, I think there would be sufficient editor interest to make it useful. Sdkb talk 17:52, 14 October 2025 (UTC)[reply]
That assumption has been proven false in the past. For the mw:citoid service, the system is Zotero, and anyone who wants to create, update, improve, etc. for any website at all is welcome to do so. We even provide instructions on how to do it. But I think I've only seen two editors do this in the last decade, and at least one of those had non-wiki experience with Zotero.
While fixing the Zotero translator is the Right™ way to do this, and would benefit not just us but all Wikipedias and even non-wiki people all over the world, given our skill sets, I think that if you want an automated way to handle a hand-curated list of known problems, we'd be better off asking for help 'after the fact' at Wikipedia:Bot requests. WhatamIdoing (talk) 18:04, 14 October 2025 (UTC)[reply]
Re That assumption has been proven false in the past, I'd argue that the assumption has not been properly tested, since I included in my hypothesis that we'd easily point people to a system and make that system easy enough to use. Right now, we are not easily pointing people to it (as evidenced by the difficulty Chris had in finding it despite being highly motivated) and we have not made it easy to use (as evidenced by the 5000-word "help" page you point to, as well as by the fact I looked into making a translator myself in the past and gave up after deciding it was too complex to be worth the effort).
Granted, the work of making Zotero easier to use is on the Zotero community, and there are probably some inherent complexities that limit how simple it could be made. But still, I don't think the reason people aren't making translators is that there's no interest in the task. Rather, it's that they don't even know it's a possible task in the first place, let alone how to do it. Sdkb talk 18:41, 14 October 2025 (UTC)[reply]

Specific entry

[edit]

The article said:

  • If there are no page numbers, whether in ebooks or print materials, then you can use other means of identifying the relevant section of a lengthy work, such as the chapter number, the section title, or the specific entry.

But we don't seem to know what "..or the specific entry." means. @FaviFake @Peter coxhead Let's sort it out? Johnjbarton (talk) 17:39, 11 October 2025 (UTC)[reply]

I think quoting a bit of the content would effectively identify it. Johnjbarton (talk) 17:40, 11 October 2025 (UTC)[reply]
I agree that that sounds like a good method of identifying it. FaviFake (talk) 17:48, 11 October 2025 (UTC)[reply]
Thanks for the ping! Imo it means "quoting part of the specific entry verbatim", but since we're here, we might as well discuss the actual matter instead of guessing what someone meant a few years ago.
What should be another recommended way to cite a passage of material lacking page numbers (ebooks), besides the section title and chapter number? FaviFake (talk) 17:47, 11 October 2025 (UTC)[reply]
Quotations are a valid way to indicate a place, but they are by no means the only way to indicate a place. These are "specific entries" that do not use quotations:
  • For a dictionary, encyclopedia, or other book with short pieces, give the name of the short bit: "example". Little's Little Dictionary.
  • For a codified law, give the code numbers: U.S. Const. art. III, § 2, cl. 3.
  • For a table or image in an unpaginated book: Give the title or number of the desired piece: "Table 3: Comparison of formatting techniques", The Style Book.
These examples are what we meant by "specific entries". It tells you where to look, using some method other than page numbers, chapters (which may be titles or numbers), section headings, or quotations. WhatamIdoing (talk) 18:54, 11 October 2025 (UTC)[reply]
This sounds great! I think the second one is the same as chapter number, but the other ones are interesting. I think we should add them as additional examples. FaviFake (talk) 19:04, 11 October 2025 (UTC)[reply]
I have also interpreted the older wording to mean an entry in a reference work. Quotations, "code numbers", the "title or number of the desired piece" or "the name of the short bit" are probably all effective alternatives when a page number is not available. For example:
Page number(s)
Stevenson, Angus, ed. (2010). Oxford Dictionary of English. Oxford University Press. pp. 1199–1200. ISBN 978-0-19-957112-3.
Entry
Stevenson, Angus, ed. (2010). "Night". Oxford Dictionary of English. Oxford University Press. ISBN 978-0-19-957112-3.
Quotation
Stevenson, Angus, ed. (2010). Oxford Dictionary of English. Oxford University Press. ISBN 978-0-19-957112-3. night → noun 1 the period from sunset to sunrise in each twenty-four hours: a moonless night
Any of those would be enough for verification, Rjjiii (talk) 19:13, 11 October 2025 (UTC)[reply]
So
  • "... such as the chapter number, the section title, or the specific entry."
should read
  • "... such as the chapter number, the section title, or the name of a specific entry".
Johnjbarton (talk) 19:27, 11 October 2025 (UTC)[reply]
"U.S. Const. art. III, § 2, cl. 3." is not "the name of the specific entry". AFAICT U.S. Const. art. III, § 2, cl. 3. doesn't have "a name". I think we should just go back to the original wording, and if necessary, link to something that defines "dictionary entry". If you want to be more vague, then consider "the relevant part of the source".
(Also, Favi, "chapter" has a different meaning in laws. The US Constitution doesn't have chapters.) WhatamIdoing (talk) 19:37, 11 October 2025 (UTC)[reply]
Your example, "U.S. Const. art. III, § 2, cl. 3." is an example of a chapter number. The chapter. number is entirely adequate to identify it. Here is the specific entry:
Representatives and direct Taxes shall be apportioned among the several States which may be included within this Union, according to their respective Numbers, which shall be determined by adding to the whole Number of free Persons, including those bound to Service for a Term of Years, and excluding Indians not taxed, three fifths of all other Persons. The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct. The Number of Representatives shall not exceed one for every thirty Thousand, but each State shall have at Least one Representative; and until such enumeration shall be made, the State of New Hampshire shall be entitled to chuse three, Massachusetts eight, Rhode Island and Providence Plantations one, Connecticut five, New-York six, New Jersey four, Pennsylvania eight, Delaware one, Maryland six, Virginia ten, North Carolina five, South Carolina five, and Georgia three.
I submit that this "specific entry" is completely useless as a citation label. Johnjbarton (talk) 22:56, 11 October 2025 (UTC)[reply]
Document/article/section/clause is not a "chapter". Code/division/part/chapter is a chapter. Here, for example, is a chapter explaining, among other things, how to store ice cream scoops if there's a lull in customer traffic.
In terms of people understanding what's meant by "specific entry", I think that giving the example of a dictionary entry is probably the simplest. But we could just use other words. The point is to tell people that they need to use whatever is suitable and relevant for the source, even if the source does not have any chapter numbers or section titles (or if it has them but they are too large to be useful). If you can think of a better way to express this, then please feel free. What we don't want is for editors to think "Hooray, there's no chapter numbers or section titles, so I don't have to bother providing |at=details!" WhatamIdoing (talk) 04:17, 12 October 2025 (UTC)[reply]
First to the point: "If you can think of a better way to express this, then please feel free. " I have proposed two different solutions: 1) Omit the confusing phrase "..., or the specific entry." 2) suggest the obvious: "...or the name of a specific entry". Here's 3) "...or the headword of a dictionary entry or its equivalent". 4) "...or any identifying word or succinct phrase." Lots of options. I don't understand why you are fighting for the existing mess.
Next, to the argument: I disagree on both points.
First, document/article/section/clause is entirely equivalent to a chapter number. I believe most editors will understand that whatever numbering system the source uses should be adopted in citing the source.
Second, the meaning of an "entry" is something we can agree because we have reliable sources. For examples OED terminology or Interpreting a Dictionary Entry. Including the entry for the purpose of identifying the entry is not sensible. Johnjbarton (talk) 04:49, 12 October 2025 (UTC)[reply]
Uh, isn't naming the entry the obvious means of identifying the entry? Gawaon (talk) 07:28, 12 October 2025 (UTC)[reply]
This whole discussion is making me think we should use other words that are less vague than entry. FaviFake (talk) 10:53, 12 October 2025 (UTC)[reply]
Suggestions? Blueboar (talk) 12:19, 12 October 2025 (UTC)[reply]
"verbatim quote", "cited passage", "number of the table/image", "dictionary definition", "chapter/section number/name", "paragraph number"...
Any or all of the above come to mind. Just don't use "entry" alone. FaviFake (talk) 12:32, 12 October 2025 (UTC)[reply]
A "dictionary definition" would actually be a "verbatim quote", right? That sounds needlessly complicated when just specifying the entry (i.e. the headword) is sufficient. Gawaon (talk) 15:13, 12 October 2025 (UTC)[reply]
fine, then "dictionary entry"? as long as there isn't "entry" alone it's fine FaviFake (talk) 15:21, 12 October 2025 (UTC)[reply]
I think there is a fundamental misunderstanding of "dictionary entry". I gather that some editors believe that "dictionary entry", or "entry" for short, is the headword alone. However the word "entry" refers to all of the content that a dictionary gives for the headword. From that same source:

entry Entries are the primary building blocks of the dictionary. Each entry represents all the meanings of a given headword, throughout its recorded history. The entry is structured to show the evolution of meanings and uses over time. In most entries there is also a pronunciation section where relevant, an etymology section, and various other sections. Homographs are treated as separate entries.[2]

Thus "dictionary entry" won't solve this issue. We could use "headword of a dictionary entry", "name of an entry", or other wording that clarifies the intent to identify rather than quote the item. Johnjbarton (talk) 15:46, 12 October 2025 (UTC)[reply]
How about "specific headword"? That a headword is part of an entry seems clear enough. Gawaon (talk) 16:58, 12 October 2025 (UTC)[reply]
Works for me. Johnjbarton (talk) 17:02, 12 October 2025 (UTC)[reply]

For dictionary or glossary entries, 'Chicago Manual of Style' used to recommend s.v. which means sub verbo or sub voce, Latin for "under the word".[3]. However, §14.130 of the 18th edition abandons that recommendation, and suggests giving the headword of the entry in quotation mark. Some of their examples are

  • Encyclopaedia Britannica, 15th ed. (1980), under "salvation."
  • Dictionary of American Biography (1937), "Wadsworth, Jeremiah."

Jc3s5h (talk) 17:16, 12 October 2025 (UTC)[reply]

Also, either Help:References and page numbers should have a section on citing sources without page numbers, or there should be some separate page like Help:Citing sources without page numbers. Rjjiii (talk) 17:30, 12 October 2025 (UTC)[reply]
So @Jc3s5h would agree that we should change "specific entry" to "specific headword"? @FaviFake@WhatamIdoing? I just want to make some kind of infinitesimal progress. Johnjbarton (talk) 03:31, 13 October 2025 (UTC)[reply]
Yes, as well as other examples, if needed. FaviFake (talk) 04:19, 13 October 2025 (UTC)[reply]
I have changed it. Gawaon (talk) 07:17, 13 October 2025 (UTC)[reply]
Huh? What the f*ck is a “headword”? Policy should use terms that are readily understandable. Blueboar (talk) 12:17, 13 October 2025 (UTC)[reply]
Policy should also be precise as to not be misunderstood. I've linked the word to wdictionary. FaviFake (talk) 15:09, 13 October 2025 (UTC)[reply]
I don't really like "specific entry" because people are obviously confused by that. The passage I cited above from Chicago says "item". I'm not sure that would be understood in this context. I also agree with Blueboar's comment, 'What the f*ck is a “headword”?' Jc3s5h (talk) 15:59, 13 October 2025 (UTC)[reply]
"The headword is the main word at the top of an entry". Thus we could say
  • If there are no page numbers, whether in ebooks or print materials, then you can use other means of identifying the relevant section of a lengthy work, such as the chapter number, the section title, or the main word at the top of an named entry.
I added the word "named" because the OED source which is quoted above is specific for a dictionary and I think adding "named" clarifies the context for this third item on our list. Johnjbarton (talk) 16:24, 13 October 2025 (UTC)[reply]
"Item" is also appropriate for lists. WhatamIdoing (talk) 16:31, 13 October 2025 (UTC)[reply]
So how about
  • If there are no page numbers, whether in ebooks or print materials, then you can use other means of identifying the relevant section of a lengthy work, such as the chapter number, the section title, or the main word at the beginning of an named item.
Johnjbarton (talk) 17:23, 13 October 2025 (UTC)[reply]
i'm sorry but this really sounds wayy to convoluted to me. headword works FaviFake (talk) 17:24, 13 October 2025 (UTC)[reply]
I would just use “or other brief identifier”. Blueboar (talk) 17:38, 13 October 2025 (UTC)[reply]
I'd say "headword" wins over "main word at the beginning of a named item" and "brief identifier" for clarity, conciseness, and precision. Gawaon (talk) 07:46, 14 October 2025 (UTC)[reply]
Fully agreed especially after the link to wdictionary FaviFake (talk) 15:14, 14 October 2025 (UTC)[reply]
Meh… Even with the link, I find “headword” obscure and confusing. However, I can’t think of another word, so I won’t object strongly. Blueboar (talk) 15:28, 14 October 2025 (UTC)[reply]
Since many of the things that need to be pointed at don't have headwords, I don't think that's an adequate substitute. "...such as the chapter number, section title, headword, or other brief identifier" would work for me. There is no rule that says we're limited to three examples. WhatamIdoing (talk) 18:07, 14 October 2025 (UTC)[reply]
But since we already say "such as", there's also no need to strive for completeness. Gawaon (talk) 21:11, 14 October 2025 (UTC)[reply]

How is this a good idea?

[edit]

To avoid me getting into a big argument elsewhere, please would someone tell me whether they think this[4] edit is reasonable?

To me, it is entirely pointless to shorten the name of the publisher when the original reference writer either used a template or had the original work in front of them. ThoughtIdRetired TIR 22:12, 11 October 2025 (UTC)[reply]

It's a matter of taste but I think the edits are reasonable. "Penguin" is the part of the publisher name that actually identifies the publisher; "Penguin Books" would also be reasonable; "Penguin Books Ltd." adds information that is not so much part of the name but a description of what type of corporation they are. Editors could in good faith choose to use any one of those three forms. Prior to the edits the names were listed inconsistently ("Ltd." for one publisher, "Limited" for another, just the shortened form "Scribner's" for another not visible in the diff) so it would be difficult to argue that there is already a consistent style choice that should not be changed without discussion. —David Eppstein (talk) 22:36, 11 October 2025 (UTC)[reply]
I'd have kept it to Penguin Books myself, but Limited, LCC, Ltd., Private Ltd., GmBH, etc... are just corpolegal gibberish no one cares about. Headbomb {t · c · p · b} 22:44, 11 October 2025 (UTC)[reply]
Yeah, just "Penguin" or "Penguin Books" is entirely sufficient, there's no need for more. As I remember it, reliable style guides like the CMOS recommend omitting stuff like "Ltd." in such cases. Gawaon (talk) 07:31, 12 October 2025 (UTC)[reply]
IMO the company names are basically obviously superfluous. Books/Press/Publishing should generally be kept as it helps to distinguish and is also generally the most expected name. PARAKANYAA (talk) 22:45, 11 October 2025 (UTC)[reply]

What is interesting is to use the isbn to generate the reference with {{citebook}}. Then you discover that the publisher of the Bombing War is Allen Lane, an imprint of Penguin Group. So this is a case of impetus to impose a style failing on the accuracy of who the publisher is. The imprint name can often imply (or otherwise) the reliability of a source.

Incidentally, there is a difference between "Ltd" and "Limited" in the UK. This is not stylistic inconsistency, it is a matter of fact. It is always incorrect to swap one for another. Which is why I would resist omitting this element of a name. ThoughtIdRetired TIR 23:02, 11 October 2025 (UTC)[reply]

I'd agree with David Eppstein, this is a matter of taste. I wouldn't suggest going around enforcing one style or the other, but otherwise it's of little importance. -- LCU ActivelyDisinterested «@» °∆t° 23:10, 11 October 2025 (UTC)[reply]
A quick web search suggests that Limited and Ltd are the same in the UK, except that whichever the company uses, they're expected to be consistent about it. That doesn't sound like something we need to worry about, though. WhatamIdoing (talk) 04:23, 12 October 2025 (UTC)[reply]
The point is you don't distinguishing between Penguin Books Ltd, Penguin Books Limited, Penguin Books LLC, Penguin Books GmBH, etc. because these are neither meaningful distinctions, nor do they represent different entities than Penguin Books <nothing>. Headbomb {t · c · p · b} 05:15, 12 October 2025 (UTC)[reply]
All of those (assuming any of them exist) are the same as Penguin Books as far as we're concerned, but the first two in your list would refer specifically to the British corporation, the third to a US corporation, and the fourth to a German corporation. We don't care, but a contracts lawyer probably would (and shouldn't be using a citation in a Wikipedia article to figure out which entity they're dealing with, so again: we don't care, even if a very small fraction of our readers might,). WhatamIdoing (talk) 16:37, 13 October 2025 (UTC)[reply]

https://en.wikipedia.org/w/index.php?title=John_Kelynge&gesuggestededit=1# Iwanted to add a link to the source in the article

[edit]

I wanted to add a link to the source in the article, but it doesn't add and I'm not sure of its significance. However, in terms of citation, it matches word for word. Smart Andrew (talk) 20:33, 12 October 2025 (UTC)[reply]

@Smart Andrew, what's the URL you want to add? Could it be one of the many Wikipedia:Mirrors and forks? Do you get an error message when you try to publish your changes? WhatamIdoing (talk) 16:48, 13 October 2025 (UTC)[reply]
No, that's a different site. I already posted a link to it. Smart Andrew (talk) 09:28, 16 October 2025 (UTC)[reply]
The site you attempted to add appears to be https://www.pepysdiary.com/encyclopedia/11343/. Near the top of the page on that site this statement appears:

This text was copied from Wikipedia on 16 August 2025 at 4:10AM.

That means that page is a mirror of Wikipedia and must not be used as a source for a Wikipedia article.
See the verifiability policy and the Wikipedia:Citing sources Jc3s5h (talk) 14:20, 16 October 2025 (UTC)[reply]

Conflict between advice here and MOS:WEASEL

[edit]

Our advice here says that we should use formulations like "Researchers announced the new tissue type in 2012". Our advice at {{who}} says that we should not make 'attributions to vague "authorities" such as "serious scholars", "historians say", "some researchers", "many scientists", and the like', exactly in contradiction to this advice. And MOS:WEASEL adds that "A common form of weasel wording is through vague attribution, where a statement is dressed with authority, yet has no substantial basis". The existence of a citation after the sentence is not enough; it will generally support the claim that certain specific people announced the tissue type, but not that a general class of "researchers" did so. The attribution should be made specific, in-text, not indirectly through parsing what the footnotes might mean.

Is there some way of resolving this by choosing better examples here? Or is this a fundamental contradiction where the advice here really is what it seems to be, to change specific attributions into exactly the kind of attribution to vague authorities that MOS:WEASEL argues against? —David Eppstein (talk) 21:42, 14 October 2025 (UTC)[reply]

From recent Featured Articles of the day:
Clownfish
"A 2005 study found that anemones grew and regenerated faster in the presence of clownfish groups, and attributed this to ammonium from clownfish waste."
Tell es-Sakan
"The archaeologists who led the excavations at Tell es-Sakan, Pierre de Miroschedji and Moain Sadeq, proposed that there were three areas of Egyptian expansion into the southern Levant during the late 4th millennium BCE, and Tell es-Sakan was one of the major settlements in the region."
Spaghetti House siege
"Peter Waddington, in his study of policing, writes that the police's "reputation for restraint received dramatic vindication by the way in which two highly publicised sieges were handled by the Metropolitan Police"."
Hurricane Ophelia (2005)
"The Climate Prediction Center determined four primary factors driving the season's activity: the Atlantic multidecadal oscillation, the reduction of atmospheric convection in the tropical Pacific, record-high sea surface temperatures in the tropical Atlantic and Caribbean, and conducive wind and pressure patterns across the western Caribbean and Gulf of Mexico."
Redshift
"Arthur Eddington used the term "red shift" as early as 1923, which is the oldest example of the term reported by the Oxford English Dictionary."
Through the Looking-Glass
"Among more recent comments on the book, Daniel Hahn in The Oxford Companion to Children's Literature (2015) writes that sentimentality plays a larger part in Through the Looking Glass than in Alice's Adventures in Wonderland."
Rjjiii (talk) 02:02, 15 October 2025 (UTC)[reply]
I agree that
  • Researchers announced the new tissue type in 2012.
is a both a weasel and news. An encyclopedia should include dated items a part of a history which requires context to explain significance (which would of course be available in the secondary reference). All of the examples posted by RjjIII are better. Johnjbarton (talk) 02:20, 15 October 2025 (UTC)[reply]
It's also at risk of falling afoul of WP:MEDSAY, assuming Wikipedia:Manual of Style/Medicine-related articles applies. WhatamIdoing (talk) 17:44, 16 October 2025 (UTC)[reply]
We should always attribute inline for biased statements of opinion as WP:INTEXT says. For other types of statement I think it's up to editor discretion and consensus at the relevant talk pages. If the wording used implies that an opinion is widespread, but in fact that is not the case, then that has to be fixed, and attribution is one way to fix it, but it's not the only way. A relevant recent discussion about an edit to CLOP is here, and the discussion it refers to at WT:CITE is here. I think the force of WEASEL should be that omitting attribution is a tactic that can (and should not) be used to inappropriately give authority to some statements, not that omitting attributions is always a bad thing. Mike Christie (talk - contribs - library) 02:41, 15 October 2025 (UTC)[reply]
Even for biased statements of opinion, we shouldn't always attribute inline if there are big groups holding the opinion. I often use wording like "arguments in favour are X" and "arguments against are Y", without name dropping like our policy seems to suggest. —Femke 🐦 (talk) 16:56, 15 October 2025 (UTC)[reply]
For something that's really an opinion (e.g., "This is a good book", "This politician is important"), then I think the hierarchy ought to be something like this:
  • Ideal: Talk about established groups associated with known viewpoints.
    • Consequentialists say that the ends justify the means, but deontologists disagree.
    • Republocrats say that this bill solved the fiscal problem, but Demicans opposed it as merely postponing a problem.
    • The rock opera was praised by critics as a light-hearted romp, but was criticized as irreverent by religious organizations.
  • Good enough: Identify a couple of individual people and their views.
    • Alice Expert said that the book was "a valuable contribution to our understanding of just how big the Sun actually is".
    • Bob Business said that blue-green widgets are a new preppy aesthetic.
  • Bad idea #1: Hyping the people who are given as examples.
    • Book expert Prof. I.M. Portant, chair of the Learned Department at Big University, said that it is a good book.
    • Chris Celebrity, who is famous for having cameo parts in television shows, said that it is a good video.
  • Bad idea #2: Using WP:INTEXT to draw WP:UNDUE attention to small facts.
    • According to a 2016 non-randomized internet-based survey sponsored by Good Organization of people who self-report having autism and who were recruited primarily through social media advertisements, the organization's own membership, and word of mouth, most people with autism like watching videos online.
WhatamIdoing (talk) 18:08, 16 October 2025 (UTC)[reply]
Bad idea #3: Asserting opinions with no poll taken, eg
  • Wikipedians commonly engage in long arguments.
This form I see a lot. Johnjbarton (talk) 18:12, 16 October 2025 (UTC)[reply]
Polls aren't the only way to determine that kind of fact. It should be cited but doesn't require any particular method. WhatamIdoing (talk) 20:07, 16 October 2025 (UTC)[reply]
Oh, and to be clear, the frequency at which something does/doesn't happen is actually a fact instead of an opinion. An opinion would sound like "Wikipedians spend too much time on long arguments" or "Wikipedians have unreasonably long arguments over unimportant things". WhatamIdoing (talk) 20:08, 16 October 2025 (UTC)[reply]
Like Johnjbarton I see this form a lot, often unrelated to Wikipedia, in examples that say some position, terminology, or methodology is common or widespread but sourced only to one or more individual publications that take that position and do not say anything about how common it is. —David Eppstein (talk) 21:42, 16 October 2025 (UTC)[reply]
I've seen that mistake, too, but that's a problem of {{failed verification}} (the source is an example of the claim, but does not WP:Directly support any claim about the commonness of the claim) rather than a problem of WP:INTEXT attribution. WhatamIdoing (talk) 01:40, 17 October 2025 (UTC)[reply]
Concrete proposal to address the original post: replace the example
  • ☒N In an article published in The Lancet in 2012, researchers announced the discovery of the new tissue type.[3]
  • checkY Researchers announced the new tissue type in 2012.[3]
with
  • ☒NIn his book The Mathematical Theory of Relativity Arthur Eddington first used the term "red shift".[3]
  • checkYArthur Eddington used the term "red shift" as early as 1923.[3]
Johnjbarton (talk) 18:23, 16 October 2025 (UTC)[reply]
We should use a more recent (last couple of years) example, and a 'smaller' source (an article instead of a scholarly book). WhatamIdoing (talk) 20:06, 16 October 2025 (UTC)[reply]
So change
  • checkY Researchers announced the new tissue type in 2012.[3]
to
  • checkY The discovery of the new tissue type was reported in 2012.[3]
Johnjbarton (talk) 01:50, 17 October 2025 (UTC)[reply]
 Done WhatamIdoing (talk) 04:03, 17 October 2025 (UTC)[reply]

"Wikipedia:Unreferenced" listed at Redirects for discussion

[edit]

The redirect Wikipedia:Unreferenced has been listed at redirects for discussion to determine whether its use and function meets the redirect guidelines. Readers of this page are welcome to comment on this redirect at Wikipedia:Redirects for discussion/Log/2025 October 27 § Wikipedia:Unreferenced until a consensus is reached. -Samoht27 (talk) 19:32, 27 October 2025 (UTC)[reply]

url-status=unfit

[edit]

Copied from User talk:Graeme Bartlett after I realized there was a better place to ask:

You gave advice to someone here and I wanted to see how it worked for myself. It did not.— Vchimpanzee • talk • contributions • 22:54, 27 October 2025 (UTC)[reply]

Hmm, it should work, per Template:Cite_web, but you're right, it fails in your sandbox example. 'Usurped' isn't working per the documentation either. Schazjmd (talk) 23:00, 27 October 2025 (UTC)[reply]
Did you read the template documentation?
—Trappist the monk (talk) 23:03, 27 October 2025 (UTC)[reply]
I did now, but it doesn't seem to help.— Vchimpanzee • talk • contributions • 23:09, 27 October 2025 (UTC)[reply]
Really? Are you sure? Here is the reference from your sandbox:
{{cite news|url=https://www.opb.org/article/2025/10/19/a-peaceful-day-of-no-kings-protests-across-oregon-ends-with-a-show-of-force-in-portland/|url-status=unfit|title=A peaceful day of No Kings protests across Oregon ends with a show of force in Portland|work=[[Oregon Public Broadcasting]]|date=October 19, 2025|access-date=October 26, 2025}}
The template documentation says:
'requires url and archive-url' – emphasis in original
Your template does not have an archive snapshot so there is nothing for the control-switch parameter |url-status= to switch to. But, if you add |archive-url= and |archive-date= then you get summat that looks like this:
{{cite news |url=https://www.opb.org/article/2025/10/19/a-peaceful-day-of-no-kings-protests-across-oregon-ends-with-a-show-of-force-in-portland/ |archive-url=https://web.archive.org/web/20251027231843/https://www.opb.org/article/2025/10/19/a-peaceful-day-of-no-kings-protests-across-oregon-ends-with-a-show-of-force-in-portland/ |archive-date=2025-10-27 |url-status=unfit |title=A peaceful day of No Kings protests across Oregon ends with a show of force in Portland |work=[[Oregon Public Broadcasting]] |date=October 19, 2025 |access-date=October 26, 2025}}
"A peaceful day of No Kings protests across Oregon ends with a show of force in Portland". Oregon Public Broadcasting. October 19, 2025. Archived from the original on 2025-10-27. Retrieved October 26, 2025.
If |url-status=unfit is working, you should see that the citation title links to OPB's main page; there is no link to the original url.
—Trappist the monk (talk) 23:29, 27 October 2025 (UTC)[reply]
I missed the archive-url prerequisite too, my mistake. Schazjmd (talk) 23:40, 27 October 2025 (UTC)[reply]
All I saw was what came after that.— Vchimpanzee • talk • contributions • 16:33, 30 October 2025 (UTC)[reply]

Splitting the references into two sections: Primary and secondary sources

[edit]

What do folks think about this effort to indicate primary sources with a symbol, and place them in their own separate section? It's like a little ghetto for primary sources. As far as I know, no other editor has used this style. I don't think it serves the reader who will end up questioning whether the primary sources are suitable. In my view, the references should be presented all together, no matter whether they are primary, secondary, tertiary or whatever. The job of questioning whether a primary source is appropriate should be the responsibility of the Wikipedia editor who cites it, not the reader. Binksternet (talk) 17:38, 4 November 2025 (UTC)[reply]

I don't understand the point of segregating sources by these definitions and it doesn't seem to be a useful distinction for readers. (Readers who aware of the types of sources can identify them for themselves.) Schazjmd (talk) 17:55, 4 November 2025 (UTC)[reply]
I adapted this style from The Amazing Digital Circus, which is a Good article so I believed it to be a good example of a sourcing style to follow. Popturtle (talk) 18:17, 4 November 2025 (UTC)[reply]
Doesn't make sense to me in that article either. @Skyshifter:, why did you choose that organization of citations in The Amazing Digital Circus? Schazjmd (talk) 18:24, 4 November 2025 (UTC)[reply]
I don't think the article even gets the classification of primary or secondary correct. For example, the first so-called primarysecondary source is basically the author, Gerald Dih, giving his opinion about whether the album is any good. That makes it a primary source. Jc3s5h (talk) 19:22, 4 November 2025 (UTC)[reply]
I want to mention that splitting primary and secondary sources in the bibliography is common in articles on classical studies. I would even say that doing so is a best practice. Ifly6 (talk) 19:51, 4 November 2025 (UTC)[reply]
OTOH, it probably makes more sense in an academic discipline with a clear and useful distinction between the types, versus Wikipedia where "primary" too often means "I don't like this source so I want to denigrate it in the hope it will be removed" or "I don't like this article, so I'll claim the sources are primary which means it doesn't pass WP:N". Anomie⚔ 01:51, 5 November 2025 (UTC)[reply]
Yea it's also often the case in academic classical studies that the bibliography includes only modern sources. You're just supposed to know what the citations refer to for the ancient ones. Ifly6 (talk) 03:26, 5 November 2025 (UTC)[reply]
I agree the distinction makes indeed a lot of sense when dealing with older historical or literary topics, where "primary" effectively means something like "published more than 300 (or 1000) years ago". It makes less sense when referring to recent topics, where the distinction between primary and secondary is much less sharp. Gawaon (talk) 08:38, 5 November 2025 (UTC)[reply]
In any case, regardless of whether or not one groups them in different sections in the bibliography, this marking of primary sources with something like a dagger symbol seems highly unusual and I can't think of any good purpose it could serve. If it's supposed to signal "that's a primary source, don't trust it!" then, of course, it's actually the editors' job to figure out whether a source seems reliable enough to use in the article or not. If yes, then no special "warning" symbol is needed; if no, then it shouldn't be used at all. Gawaon (talk) 08:43, 5 November 2025 (UTC)[reply]
I'd just like to add that in the sciences the majority of references you'll see in academic writing are, in fact, primary sources. A secondary source, e.g. a review article, is generally indicated as such. For example, after the title you'll often see "(Review)" added in parentheses. Personally I'm used to seeing mostly primary sources, but I can see the rationale of not emphasizing them here since it's a more general audience. BetsyRogers (talk) 00:02, 17 November 2025 (UTC)[reply]
No, that's not how these terms are used, check out WP:PSTS if you didn't yet. Any academic piece of writing must be written with some distance to the subject matter, which makes it secondary. If it's primary (say the autobiography of someone directly involved), then it's not strictly speaking academic writing. Primary literature usually isn't peer-reviewed, while academic (secondary) literature is. As usual, there are borderline cases, but the general distinction is clear enough. Gawaon (talk) 08:23, 17 November 2025 (UTC)[reply]
No, that's not how these terms are used You left out "on Wikipedia". We have co-opted these terms and given them somewhat different meanings and implications. Anomie⚔ 13:25, 17 November 2025 (UTC)[reply]
You may have a point there, especially when it comes to STEM fields where "primary sources" in our sense don't really exist. However, I think that our usage is close enough to the usage established in history, literature, and other humanities, where primary sources are generally non-academic (say a novel about which researchers may write, or letters and diaries from people directly involved in a historical event). Gawaon (talk) 15:59, 17 November 2025 (UTC)[reply]
We still have differences, and we don't even agree on them. For example, some consider editorially reviewed news articles about current events as secondary, while others insist they're primary and that secondary sources won't show up until years later. Personally I think we'd do best to demote WP:PSTS to an essay and have the actual policies and guidelines focus on what really matters like "significant coverage in reliable sources that are independent of the subject" instead of ever mentioning the words "primary" or "secondary". But that's unlikely to ever happen, since too many people like being able to say "it's primary" to mean "I don't like this source for some reason" rather than actually looking into reliability and independence. Anomie⚔ 17:24, 17 November 2025 (UTC)[reply]
Apparently there are field-specific definitions of a primary source. But using Wikipedia's definition of a primary source (which I checked out a long ago), any published original research is a primary source.
From #Primary, secondary, and tertiary sources:
"Primary sources are original materials that are close to an event, and are often accounts written by people who are directly involved." 
If someone does the research themselves, analyzes the data, draws conclusions from it, and publishes it, then they are definitely directly involved. I've asked other editors about this to be sure I understood Wikipedia's definition of primary sources, and everyone I've asked so far has concurred that research papers are a primary source. BetsyRogers (talk) 01:48, 18 November 2025 (UTC)[reply]
But by that interpretation, wouldn't a historian who researched an event or period also produce "primary sources" because they did the research themselves (visiting archives, studying original sources from those involved or from the time it happened, etc.), then drawing their conclusions and publishing their findings? This would make the distinction between "primary" and "secondary sources" that's usual in history and other humanities completely disappear (merging them both into "primary sources"), so consider me unconvinced. Gawaon (talk) 08:17, 18 November 2025 (UTC)[reply]
I'm not trying to convince you of anything, just explaining what constitutes a primary source in scientific fields. I guess if you traveled back in time to directly observe an event or period, then traveled back to the present and published your findings, that would be a primary source. (Alternatively, and with much less effort, you could just Google "what is a primary source in science"). BetsyRogers (talk) 08:33, 18 November 2025 (UTC)[reply]
This is useless and I don't see a reason for it but I also don't see a reason to prohibit it. PARAKANYAA (talk) 01:53, 5 November 2025 (UTC)[reply]
Under WP:CITEVAR, I think it would be difficult to prohibit it (as a general case; getting it removed from a single article would require only an ordinary, consensus-oriented discussion on the article's talk page).
But mostly I wonder: @Popturtle, when you classified the sources, were you remembering that Secondary does not mean independent, and that All sources are primary for something? Because when I look at the lists, it seems to me that you've actually split Wikipedia:Independent sources vs non-independent sources, rather than primary vs secondary. WhatamIdoing (talk) 05:39, 5 November 2025 (UTC)[reply]
I was not thinking about that and more so grouped them based on independence, I'll combine the sources back together in one list then. Popturtle (talk) 09:17, 5 November 2025 (UTC)[reply]

This seems an impossible suggestion to me. Many empirical empirical articles start with a background/literature section making use of that part a secondary source; while it continues presenting new empirical data which makes use of that part a primary source. Thus a single paper can be used in the same wiki article as both a secondary and a primary source. It becomes even more complicated when we look at synthesis from the literature in such articles which can be argued to be either primary (especially if creative combinations are made) or secondary if it is a structured logical addition of what is already known. My head already hurts thinking of all the endless discussion this idea is going to provoke. Arnoutf (talk) 18:22, 17 November 2025 (UTC)[reply]

Preservation of original punctuation?

[edit]

Greetings and felicitations. I thought I recently came across a passage in the MOS or similar document that preservation of a reference's title's original punctuation was not necessary (at least to some degree), but I can't find it. Did I hallucinate it or conflate it with something else? —DocWatson42 (talk) 05:44, 12 November 2025 (UTC)[reply]

We are certainly allowed to convert curly quotes to straight quotes. More of a gray area: If a work has a title and a subtitle, are we allowed to standardize the punctuation between them (common choices I've seen include a colon, period, or dash, but sometimes they are just separated by whitespace)? What about (the case I think DW has in mind) changing double quotes to single quotes to avoid nested double quotes? —David Eppstein (talk) 06:50, 12 November 2025 (UTC)[reply]
The first and last cases I do automatically (patrolling "quote=" fields for mistakes is a small specialty of mine). For subtitles (in English) I assume that the punctuation is a colon unless the source itself states otherwise. I maintain archaic punctuation (e.g. semicolon em dash) when I come across it. I also convert hyphens to em and en dashes where appropriate (I know that German uses en dashes for ranges, but I don't know enough about French punctuation to make changes), and the same for "x" to multiplication signs. I'm wondering about any other cases (and your thoughts on the preceding types), and if there actually is or was such a statement, or a statement against it. —DocWatson42 (talk) 07:10, 12 November 2025 (UTC)[reply]
And if corrections are permitted (as per Wikipedia:Quotations#Formatting: "Trivial spelling or typographical errors that do not affect the intended meaning may be silently corrected."). (I'm opposed to correcting the spelling of titles, so that the reference is searchable/reproducible.) —DocWatson42 (talk) 07:22, 12 November 2025 (UTC)[reply]
That passage refers to quoted text, it doesn't refer to the titles of cited works and I'd say that it shouldn't apply there. However, changing the type of quotation marks and dashes used within titles is trivial, happens all the time, and is nothing to worry about. (Indeed, changing curly to straight quotation marks is needed per our MOS and changing double to single ones is often needed to ensure proper nesting.) Gawaon (talk) 08:17, 12 November 2025 (UTC)[reply]
Pardon me—I meant in the same way as Wikipedia:Quotations#Formatting, not that that applied directly. —DocWatson42 (talk) 08:52, 14 November 2025 (UTC)[reply]
MOS:CONFORMTITLE may be what you have in mind? That section also explicitly states that MOS:CONFORM applies to titles too. Gawaon (talk) 08:21, 12 November 2025 (UTC)[reply]
Basically, yes. Thank you. ^_^ —DocWatson42 (talk) 08:53, 14 November 2025 (UTC)[reply]

These guidelines say that "general references" (vs. citations) belong in References?)

[edit]

In the "Citation types" section, under "general references" it says: "A general reference is a citation that supports content, but is not linked to any particular piece of material in the article through an inline citation. General references are usually listed at the end of the article in a References section."

Is this right? I've never heard of putting a "general" reference (i.e. not a source used to verify specific content in the text) in a References section with other citations. Where in the numerical list would it even go? Everything else that I've seen/read indicates that any material that isn't a citation actually linked to the article should go in a separate section (further reading, external links, etc.).

Can anyone clarify? Is it possible that this is outdated information? BetsyRogers (talk) 23:00, 16 November 2025 (UTC)[reply]

You see them a lot on older articles. It is better than nothing so is not actually deprecated. PARAKANYAA (talk) 23:27, 16 November 2025 (UTC)[reply]
Thanks. (I was editing my question when you replied, just to clarify a bit, but it's still the same question.) But is that still a current recommendation? If not, couldn't it be updated and include a footnote saying basically what you said? BetsyRogers (talk) 23:50, 16 November 2025 (UTC)[reply]
See the following sentence about "underdeveloped articles" clarify - it's better than nothing but not typical for higher quality articles. As WP:GENREF notes, people will generally rework them as an article improves. Nikkimaria (talk) 04:42, 17 November 2025 (UTC)[reply]

how to add this insert to my draft

[edit]

how to add this insert to my draft (for example, the king of England from such to such a year is the predecessor/ successor) if something is not the king but the baron, I make an example of the king to make it clearer what I'm talking about Smart Andrew (talk) 07:58, 27 November 2025 (UTC)[reply]

@Smart Andrew, if you still need an answer to this question, please ask at the Wikipedia:Teahouse. WhatamIdoing (talk) 04:09, 10 December 2025 (UTC)[reply]

Sources with long titles?

[edit]

It used to be a common style for newspaper headlines to be quite long. A good example would be:

"BIG MOTOR RACE MAY BE REPEATED; Contestants in Morris Park Event Believe Record Can Be Beaten. HOW MILEAGE WAS FIGURED Track Will Be In Better Condition a Month Later -- Race Was Notably Free from Serious Accidents". 1907-09-09. Retrieved 2025-12-01.

Any thoughts on how to cite these? What I've been tending to do is just cite the first fragment, i.e.:

"Big Motor Race May be Repeated". 1907-09-09. Retrieved 2025-12-01.

Does that seem reasonable? Maybe add a ... to the title? RoySmith (talk) 15:48, 1 December 2025 (UTC)[reply]

@RoySmith Trim it to what you say. This happens sometimes when there are multiple different level headings. If you look at the original article the “Big Motor Race May Be Repeated” is the large, obvious heading. PARAKANYAA (talk) 17:32, 1 December 2025 (UTC)[reply]
I looked up published citation guides' advice on this a couple of weeks ago when I came across a very old book where the title took some double digit number of line on the front page which seemed absurd to offer in the citation. Some guides (APA) said to cite the whole title no matter how long for the full citation and some (Chiacgo) said to shorten long titles where "appropriate" but I don't recall seeing a specific cutoff point. Rjjiii (talk) 02:46, 2 December 2025 (UTC)[reply]
Yes, it's very reasonable. Subtitles do not generally have to be included; when the title including subtitles is too long, I'd use just the main title. This article seems to have four subtitles in addition to the main title. How many of them to include is ultimately up to you, but just giving the main one should indeed be sufficient. Gawaon (talk) 07:01, 2 December 2025 (UTC)[reply]
I think it is allowed to omit subtitles, but not required. Sometimes I'll include a subtitle when it makes more obvious what is in the reference and why it's relevant to the article, such as a subtitle that includes the subject's name for a biography. Four subtitles is definitely too many to include, and they look more like a table of contents than subtitles anyway. So in this case, just the main title would be best. Another situation where this comes up is very old books, where it might not even be clear where the main title ends and the subtitles start, but you have to truncate it somewhere. —David Eppstein (talk) 07:48, 2 December 2025 (UTC)[reply]

Does WP:CITEVAR prohibit adding metadata to citations?

[edit]

Over at my talk page, @David Eppstein claims that WP:CITEVAR prohibits Citation Bot from adding metadata identifiers (specifically bibcode and ISBN) to citations because Editors should not attempt to change an article's established citation style... without first seeking consensus for the change. I have always considered this to fall under WP:CITEVARYES point 1, improving existing citations by adding missing information. I'd love to hear what other people think about this. Jay8g [V•T•E] 23:51, 9 December 2025 (UTC)[reply]

CITEVARYES covers adding missing essential information necessary to identify a source. Elements like bibcodes are optional, so an established consistent style that doesn't include them is covered by CITEVAR. Nikkimaria (talk) 00:15, 10 December 2025 (UTC)[reply]
Specifically, Jay8g is insisting on adding identifiers that do not provide a unique resource beyond the metadata that is already in (or should already be in) the article. If you go to the page linked by these identifiers, you only see the same information you already saw in the reference. There is no copy of the reference itself to be found there, nor a review, abstract, or other additional information. (This is true for some bibcodes, but not all; I have no objection to adding those bibcodes that provide full reference text.) The only function of these identifiers is to sit there as incomprehensible lumps in the reference and to make it difficult for readers to find the links that might lead them to actual copies of the reference. Also, since Jay8g moved this from User talk:Citation bot without providing the context there, here are some past links to similar discussions copied from the Citation bot discussion: User talk:Citation bot/Archive 41#Useless bibcodes redux and User talk:Citation bot/Archive 42#Bad pmid —David Eppstein (talk) 00:22, 10 December 2025 (UTC)[reply]
I'm not "insisting on" doing anything, and I think it's about time to apply WP:AGF in this and other Citation Bot-related discussions. I'm attempting to find a consensus, which no one has been able to show so far (those threads don't have anything close to a consensus to not include these identifiers; if they did, Citation Bot would have been changed to stop including them long ago. Jay8g [V•T•E] 00:28, 10 December 2025 (UTC)[reply]
Surely the onus should be to find consensus to always include this useless cruft, not to force them in without consensus to exclude them. So tell me: where is that consensus? —David Eppstein (talk) 00:35, 10 December 2025 (UTC)[reply]
Considering that the bot (and editors) have been doing this for years without significant opposition, it seems to have been a case of WP:SILENCE. Now that you have brought these concerns to my attention, I'm trying to determine that consensus. But I can't just ask for a major change to the way Citation Bot operates because one person has an issue with it; no one's going to be willing to implement that, especially considering that the first discussion you linked includes significant support from other editors to include this information. And one bot's talk page isn't the right place to have this discussion, which is why I brought it to a venue that is hopefully more well-attended. Jay8g [V•T•E] 00:42, 10 December 2025 (UTC)[reply]
If you think this is one person and that there has been no past opposition then you haven't even read the linked past discussions. Others with similar opinions there include Jacobolus and Johnjbarton. You might also notice that Citation bot has not been adding s2cid identifiers to articles for quite a while, as a result of similar discussions. It also stopped adding bibcodes pointing only to arXiv preprints, again as a result of similar past discussions. In any case, the argument that we've been doing this useless thing so long that we have to keep doing it comes across to me as pretty weak. —David Eppstein (talk) 00:57, 10 December 2025 (UTC)[reply]
I was trying to avoid WP:CANVASS, but per that discussion, I'll also ping @Headbomb, @Modest Genius, @Jo-Jo Eumerus, and @21.Andromedae so that we ping everyone involved in that discussion.
The point isn't that we have to keep doing it. The point is that you keep implying there is consensus against doing it, which does not currently appear to be the case. Jay8g [V•T•E] 01:10, 10 December 2025 (UTC)[reply]
I oppose automatically adding (or removing) bibcodes. Some (older?) astronomy articles are archived on the bibcode site, but otherwise the links are counterproductive for me and churning the page for that purpose is more wasting time. Johnjbarton (talk) 01:14, 10 December 2025 (UTC)[reply]
My subjective personal impression is that in the long term on pages where humans are also paying attention Citation Bot is about 30% substantively useful changes, 30% harmlessly trivial changes, 10% mistakes, and 30% spam. I consider it good general practice to block the bot from pages which are being actively curated by humans, especially if the bot had previously done a pass through the page, because it routinely screws up or adds new spam nonsense, and carefully checking its changes is mostly a waste of time.
Rarely bibcodes contain a full-text copy of the source; these should be included on pages. (Sometimes the bibcode even leads to the only online copy of a source.) The vast majority of bibcodes, PMIDs, etc. added by Citation Bot are pure spam, a pointless distraction for humans containing no information which wasn't already included in the Wikipedia citation. They might be useful to someone trying to build a machine-readable database of citation metadata, but Wikipedia isn't the right venue – perhaps try adding it to Wikidata or something. It's not reasonable to call this "missing information", and these should be left out, especially where there's a local editor preference against them. I would be in favor of a policy change to the effect that all bibcodes ever added to English Wikipedia by bots should be removed, and the privilege of adding bibcodes to citation metadata should be reserved for humans. –jacobolus (t) 01:13, 10 December 2025 (UTC)[reply]
I think that Wikidata was looking at porting all citations over but it would be roughly the size of the present Wikidata, so they can't. Until a Wikicitation is created, it would be best to keep bibcodes alive on Wikipedia. Most readers are not upset by some extra characters in the citations at the bottom. Abductive (reasoning) 02:58, 10 December 2025 (UTC)[reply]
@Abductive The issue discussed here is not "should bibcodes be removed", but "should bibcodes be automatically added by bots". They have very limited value so why add more? Johnjbarton (talk) 03:09, 10 December 2025 (UTC)[reply]
I'm okay with limited value, and I'm also okay with editors applying deny=citation_bot to articles that they care deeply about. Abductive (reasoning) 03:11, 10 December 2025 (UTC)[reply]
Maybe we can get the bot to add deny=citation_bot. Johnjbarton (talk) 03:36, 10 December 2025 (UTC)[reply]
I only added it myself following the second time the bot was run, by the same editor, producing the same erroneous citation, after I had reverted it the first time and reported the error. (On circle packing theorem, which I am in the middle of a major expansion of. I might consider making the exclusion more local to the specific problematic citation once the article is more stable again. The bigger problem was actual incorrect metadata, not merely useless ids.) —David Eppstein (talk) 06:10, 10 December 2025 (UTC)[reply]
I feel you. Just last week I asked over at the Village Pump about the GNIS 4 template being embedded into infoboxes of US locales. This serves no useful purpose at all, and GNIS is riddled with errors. Nobody responded to me. At least here editors can deny the bot. Abductive (reasoning) 10:23, 10 December 2025 (UTC)[reply]
That would Meta:Shared citations ~ 🦝 Shushugah (he/him • talk) 11:56, 13 December 2025 (UTC)[reply]
Did you really mean: Meta:WikiCite/Shared Citations?
—Trappist the monk (talk) 12:18, 13 December 2025 (UTC)[reply]
@Trappist the monk yes thank you! This can also be voted on in Meta:Community Wishlist/W240 ~ 🦝 Shushugah (he/him • talk) 12:26, 13 December 2025 (UTC)[reply]
To answer the question in the section heading:
No, WP:CITEVAR doesn't prohibit the addition of various id numbers and other metadata about sources. CITEVAR neither demands nor prohibits these things, just like it neither demands nor prohibits the inclusion of an author's full name, and just like it neither demands nor prohibits having the citations presented in green text. CITEVAR, like its cousins WP:ENGVAR and WP:STYLEVAR, is trying to prevent edit wars. Therefore, the CITEVAR rules are:
  • Thou shalt not edit war.
  • If you believe that an edit will change the style in a potentially disputable fashion, then you should discuss your planned changes on the talk page first, and consensus wins. CITEVAR provides a short list of style changes that you should expect to be disputed, as well as some that the community would find surprising if someone objected to them.
  • Zero shall be the number of edit wars that thou shall count.
  • If you believe that the change to the style will not be objected to (i.e., by editors actually working on that article, not with someone saying "I dispute that for every single article on Wikipedia"), then you can Wikipedia:Be bold.
  • The number of edit wars shall be zero.
  • If an actual objection appears, the citation style should be discussed (like civilized human beings, or if that seems too difficult, like Santa Claus is watching you right now) on the Talk: page, and consensus wins (include, exclude, change, don't change – CITEVAR does not care, so long as an agreement is reached).
  • Neither shalt thou tempt thy fellow editors to edit war.
  • If editors realio, trulio cannot reach an agreement (a fairly rare occurrence), then the CITEVAR specifies a default formatting scheme.
  • Do not edit war.
I would suggest that the existence of a bot, and therefore of a bot approval, is some evidence that the community supports adding these id numbers as a general thing. But if an objection is raised for a particular article, and consensus formed against it in that particular article, then it would be ideal if there were some way of excluding the bot, because we don't ever want a bot edit warring against consensus. WhatamIdoing (talk) 04:31, 10 December 2025 (UTC)[reply]
To my knowledge, the addition of bibcodes (and of the other identifiers David Eppstein mentions above as having been disputed) were never explicitly approved via BRFA for this bot. Do you have knowledge otherwise? Nikkimaria (talk) 04:49, 10 December 2025 (UTC)[reply]
I have zero knowledge of the paperwork – just a belief that there ought to have been some. WhatamIdoing (talk) 05:04, 10 December 2025 (UTC)[reply]
There is a way to exclude the bot from editing specific articles (and it has been used in the article in question), but some users seem to think the bot should magically know not to edit certain articles even without being specifically being told not to edit those articles.
I also feel like CITEVAR is being used to mean that people should never, ever edit citations once they've been added to an article. The whole thing has a tinge of WP:OWN to it, to be honest. Jay8g [V•T•E] 06:12, 10 December 2025 (UTC)[reply]
I do kind of feel that if you personally initiated a bot run on an article and it was reverted for buggy behavior then you should not personally initiate another bot run on the same article less than 24 hours later with the bug report still unaddressed. I don't think magic need be involved to figure out that might be an annoying thing to do. —David Eppstein (talk) 06:24, 10 December 2025 (UTC)[reply]
Do you really expect people to remember every minor cleanup edit they've ever made and notice if it's been reverted (when it's not from their account so they don't get a notification that it has been reverted)? Anyway, if you had just fixed the bugged part instead of reverting the whole edit (and therefore re-introducing CS1 errors/maintenance messages), the page wouldn't have popped back up in the Category:CS1 maint: unflagged free DOI queue. Jay8g [V•T•E] 06:35, 10 December 2025 (UTC)[reply]
You are not a bot, so you should not act like a bot. Take responsibility for the edits you make and pay attention when they are disputed rather than just running the bot automatically and then immediately forgetting what you have done like a bot would do.
See also: User talk:Citation bot#Bot limited to only single page request: The bot is currently forbidden from making whole-category runs. This is for a reason. You should not be working around this by manually triggering the bot on entire categories. —David Eppstein (talk) 07:34, 10 December 2025 (UTC)[reply]
If I may expand a little on this; the bot is not forbidden, as in prohibited by an administrator from whole-category runs. The limitation has been implemented by the maintainer(s) to support the development of the code and bug fixing - It was not a part of the unblock requirements, as can be seen in the unblock reasoning. The max pages has been set to 1 right now - technically a category with only one page in it might work.
Agreed that editors should pay attention to edits made by their requests. As far as I know, the idea is not that a user monitors all pages and sees if the bot has been reverted; it is that the user checks the edits right after they have been made - Sometimes the bot indeed has bad code and it was correctly blocked by an admin for that; however, sometimes the external data is bad itself, something the bot can't know, and then the user should block the bot from the specific citation (or the page) and report it on the talk page, so the external data can be added to a block list. All things considered there is a reason that the bot doesn't run 24\7 on its own across all pages like other maintenance bots do.
There should not be a need to run the bot multiple times on the same page within a short time. There also is no need to revert a whole edit if only a small part was bad. Both of these are valid points I think. Redalert2fan (talk) 08:31, 10 December 2025 (UTC)[reply]
I'm commenting only because I was pinged. I cannot see anything in WP:CITEVAR that prohibits a bot adding identifiers to references. Addition of a DOI, Bibcode, ISBN or similar identifier to one or more references is not a change in "established citation style". CITEVAR is part of the WP:CITESTYLE section - that's clearly what it is referring to. Claiming that adding a Bibcode changes the overall style (and is therefore banned) is a perverse reading of the guideline that I cannot support. The choice of which identifiers are worth including is another matter; for Citation Bot we've had various discussions which supported the addition of ISBNs and Bibcodes but not some other identifiers (e.g. s2cid). I largely agree with WhatamIdoing above and implore editors not to waste time and effort arguing or edit warring about such a trivial issue. Modest Genius talk 13:34, 10 December 2025 (UTC)[reply]
I tend to think that CITEVAR is about changing things like sfn to ref or list-defined references to other types i.e not merely appending some information to a citation template. One key consideration is that while picking a citation style is something one does consciously i.e for a reason, typically people omit parameters from citations because they don't know them, don't realize they could be added or expect the bot to add them and save them some manual work. While there are circumstances where a parameter was omitted for a reason, I think that's much less common than a parameter being omitted for one of the other three reasons. Jo-Jo Eumerus (talk) 13:37, 10 December 2025 (UTC)[reply]
In this specific case we're talking about an example when information was deliberately not "appended", and the goal was to save frustration and hassle for human readers, not anything about avoiding "manual work".
But in general, this gets the answer backwards: Additional information should only be added for a reason, and deliberately not adding extraneous information which does not benefit readers is obviously better than adding it merely because a bot will do so unthinkingly. Wikipedia is not a database of citation metadata, and we should not add human-irrelevant clutter to our articles for clutter's own sake. –jacobolus (t) 14:01, 10 December 2025 (UTC)[reply]
How much benefit to readers is necessary? Because if it's less than 1%, we should omit citations entirely.
On average, only 1 in 300 page views results in a reader clicking through to one source. 99.7% of the time, readers don't care about the sources at all. If we take Circle packing theorem as an example, there are 46 little blue clicky numbers in the article. The article gets almost 9K page views per year. If we assume that the article behaves like a spherical frictionless chicken, then a reader will look at one ref every two weeks, and each footnote will be checked once every 21 months.
Maybe adding that "human-irrelevant clutter" would make it possible for certain readers to find the source (e.g., via their university's library system) and actually read it. WhatamIdoing (talk) 19:03, 10 December 2025 (UTC)[reply]
The bibcodes in question typically contain no information whatsoever that isn't already in our citation. That is, they list information such as the article title, author name, journal name, year, page number, and a DOI link to the publisher's website and/or and arXiv link to a preprint, but no additional information (and no copy of the full text). Can you elaborate about how you imagine how this would help "certain readers find the source", when those readers can already see precisely the same information and links from the Wikipedia article? –jacobolus (t) 21:46, 10 December 2025 (UTC)[reply]
I do find bibcode useful when there are no other identifiers. The problem though is that we need more than "this is useless/useful in some special case" to argue about adding bibcodes more generally. If there is evidence that bibcodes routinely mislead readers, waste valuable server space or bring Wikipedia in disrepute, or more wiki-wide arguments more generally, then we could contest their addition as a general matter. If not, we need a way to flag a parameter as not worth adding - do citation templates and bots play nice with commented-out notes "please don't add this" being put in the parameter's place? Jo-Jo Eumerus (talk) 09:38, 13 December 2025 (UTC)[reply]
Yes, the way I get citation bot to stop adding one is to make the parameter say something like bibcode=<!-- useless bibcode: XYZ -->, which is then visible in the source but not shown to readers. –jacobolus (t) 10:16, 13 December 2025 (UTC)[reply]
Agree with Modest Genius's comment. What is happening is analogous to in a Chicago-style citation where when a publisher is present someone adds a location (eg "New York" for "Oxford University Press"). The style of the citation hasn't changed and trying to fit the addition of IDs into WP:CITESTYLE clearly outside its scope. Ifly6 (talk) 20:01, 10 December 2025 (UTC)[reply]
If there's an established consistent style that doesn't include locations and you add locations, that's a style change too. Nikkimaria (talk) 23:56, 10 December 2025 (UTC)[reply]

I agree with is said by Jo-Jo Eumerus, Modest Genius and WhatamIdoing, and do not understand why would the bibcode field be useless. For astronomy articles at least, it provides a copy of the abstract, allows you to see other works from the same author (a feature absent in many astronomy journals, or which just redirects to ADS/Google Scholar), provide direct links to the preprint and published version, links to the SIMBAD database for large catalogues where the full data is contained in an external table, and in many older papers, the only full-text copy available in the web. I am not sure about other areas, so i need examples of 'useless bibcodes' and the occurence rate of such bibcodes. I have seen some pages like HAT-P-67 using invisible text to prevent citation bot adding a non-existent page number, which may work to prevent such bibcodes. 21 Andromedae (talk) 19:02, 10 December 2025 (UTC)[reply]

Here are examples of useless bibcodes. I compiled a table (User:Johnjbarton/sandbox/bibcodes) with all of the bibcodes and DOIs from an article I have been working on, History of atomic theory. Eleven of the 16 bibcodes where only the citation. 5 had the abstract, but in every case the DOI had the same or more information. Also many more DOIs were in the article with no corresponding bibcode. Bibcodes are specific to astronomy and we should not be automatically adding them all over wikipedia.
Now my challenge to you: can you find any examples outside of the field of astronomy where a bibcode gives more information than the DOI? Johnjbarton (talk) 19:57, 10 December 2025 (UTC)[reply]
I largely only edit astronomy-related articles, so i can't provide such examples. 21 Andromedae (talk) 00:47, 11 December 2025 (UTC)[reply]
@Johnjbarton, I don't think we should expect an "identifier used by several astronomical data systems" (to quote our article on Bibcode) to be used much "outside the field of astronomy". WhatamIdoing (talk) 03:07, 11 December 2025 (UTC)[reply]
Indeed, hence my proposal in the next topic. Johnjbarton (talk) 16:26, 11 December 2025 (UTC)[reply]
I don't have a problem with including bibcodes for astronomy/astrophysics papers; similarly, I don't have a problem with PMIDs for medical research papers. But when bibcodes or PMIDs get attached to e.g. history, math, or social science papers, they typically don't contain any additional useful information. But the bot adds them anyway, indiscriminately, just on the basis that the identifier exists rather than for any concrete benefit to readers. –jacobolus (t) 03:28, 11 December 2025 (UTC)[reply]
The difference between a doi and a bibcode is the bibcode is link to a database of papers and citations. The database is being expanded into something called SciX which will index a broader range of subjects: "SciX covers and unifies the fields of Earth science, planetary science, astrophysics, heliophysics, and the NASA-funded biological and physical sciences.", so don't think of it as limited to astronomy. Interestingly for very new papers the bibcode works while the doi is too new to work, but a short time cures that. StarryGrandma (talk) 03:54, 11 December 2025 (UTC)[reply]

Proposal to stop adding Bibcode values to citations unless they provide information

[edit]

In the middle of the preceding very long discussion about WP:CITEVAR the question of the value of bibcodes was raised. That topic got mixed up in the other one. So I want to open the question about bibcodes cleanly. Here is my proposal:

  • Bibcode values should be added to a citation only when the linked page provide unique information unavailable through other links in the citation.

In my experience DOI has effectively replaced bibcode. The bibcode typically only shows the same info as the citation. I've compiled some examples here. The only exception I've seen is some older astronomy sources which are archived at the bibcode site and in these cases a bibcode would be fine. The proposal would apply to bots as well. Johnjbarton (talk) 23:16, 10 December 2025 (UTC)[reply]

Support (Obviously, but I want to encourage other editors to be clear with respect to the proposal). Johnjbarton (talk) 23:45, 10 December 2025 (UTC)[reply]
  • First a disclaimer: I have no idea what a bibcode is… but based on the above discussion… if a bibcode simply repeats information already in the citation, I see no reason to include it. Exceptions can be made for specific citations where the bibcode does provide additional information (like a link to an abstract). That would probably mean adding it by hand, rather than by bot. Blueboar (talk) 01:21, 11 December 2025 (UTC)[reply]
Support Please yes. I waste a not insignificant amount of time checking these junk identifiers added by citation bot and removing them. Went through the same thing with s2cid back in the day. Esculenta (talk) 01:04, 11 December 2025 (UTC)[reply]
Support. Makes sense. Additional information (metadata) is in my opinion usually helpful and should not just be forbidden because "some editors don't like it", but repeating information already accessible via other existing metadata (such as the DOI) adds nothing of use. Gawaon (talk) 02:44, 11 December 2025 (UTC)[reply]
I've no opinion about the value of Bibcode, as I don't use it myself – it is, after all, an "identifier used by several astronomical data systems", so only people editing astronomy-related articles are likely to see it – but the proposal creates some fortune-telling problems for editors:
  • Alice adds a citation with a Bibcode in 2024.
  • Bob adds a different identifier in 2025. The new identifier duplicates all the Bibcode content (and maybe adds something else).
  • Now Alice's addition of the Bibcode should be reverted, because the "the linked page" no longer "provide[s] unique information unavailable through other links in the citation"? Or maybe we all just need to make a mental note that this now-duplicative Bibcode should stay, because when it was "added" there were no "other links in the citation"?
WhatamIdoing (talk) 03:04, 11 December 2025 (UTC)[reply]
so only people editing astronomy-related articles are likely to see it. One could argue that those are the people most likely to see useful bibcodes, but because it has been incorporated into Citation bot's editing, bibcodes appear widely outside of the astronomy context, in everything from Abortion to Zoonosis.
In answer to your question, it would make sense to remove the bibcode with or after Bob's edit. Nikkimaria (talk) 03:09, 11 December 2025 (UTC)[reply]
Exactly. I expect astronomy articles will continue use bibcode, but not elsewhere. Johnjbarton (talk) 03:20, 11 December 2025 (UTC)[reply]
Would it be possible to set the bot up so it only triggers adding bibcodes when on astronomy articles (defined broadly/loosely)? Or alternatively only on citations to a broad whitelist of known astronomical publications. That might avoid a lot of the lower value ones, while keeping them for the contexts in which they are ore straightforwardly useful. Andrew Gray (talk) 10:13, 14 December 2025 (UTC)[reply]
The proposal as worded doesn't apply to this case, where the bibcode was there first. It's only about adding bibcodes when all the supplied information is already there. And nobody has suggested a mass removal of existing bibcodes. Gawaon (talk) 03:30, 11 December 2025 (UTC)[reply]
If we adopt this rule, then there will be someone suggesting mass removal. WhatamIdoing (talk) 04:01, 11 December 2025 (UTC)[reply]
Well, should that happen, we can discuss it. Gawaon (talk) 06:02, 11 December 2025 (UTC)[reply]
  • Support per Blueboar. Nikkimaria (talk) 03:10, 11 December 2025 (UTC)[reply]
  • Support. More generally, I would support similar wording to WP:ELNO #1 for all reference identifiers (not just bibcodes): they should only be added when they provide a unique resource beyond what the reference itself already contains or should contain. So identifiers that provide article text (jstor, doi, most hdl, some bibcode) are generally ok, as are identifiers that provide reviews (mr, zbl) but metadata-only identifiers (most bibcode) are not. —David Eppstein (talk) 17:09, 11 December 2025 (UTC)[reply]
    That wording would ban ISBNs. WhatamIdoing (talk) 17:32, 11 December 2025 (UTC)[reply]
    I think not. ISBNs link through a Wikimedia page that, among other things, leads to Google Books pages for the book that might provide previews.
    It might ban ISSNs but I'm fine with that. —David Eppstein (talk) 18:39, 11 December 2025 (UTC)[reply]
    I only find ISSN useful in the case where a journal is obscure, especially if it has a common name and there is no publisher link or DOI for each paper. If the journal is called something like the Journal of Psychology but is the one published by a college in the Philippines instead of the one published by T&F, then throwing in the ISSN (2423-2084) could improve clarity and help a reader find the source (though disambiguating the journal title like Journal of Psychology (SDCA) or SDCA Journal of Psychology might also be a good idea). Adding the ISSNs to every citation seems unhelpful. –jacobolus (t) 19:21, 11 December 2025 (UTC)[reply]
    I also find ISSNs useful because it's more efficient to search Scopus with an ISSN than with a name. With an ISSN, you know that you're getting the right journal and not one that has almost the same number. WhatamIdoing (talk) 17:11, 13 December 2025 (UTC)[reply]
    An ISBN is a "reference identifier". If you restrict "all reference identifiers", then you restrict ISBNs. You can argue that Special:BookSources is "a unique resource", but others can argue that it's not, especially when it's not linked (example, example). They can also argue that a DOI is not "a unique resource" compared to a direct URL since, in the optimal case, they both end up at the same place.
    More generally, I think it's misguided because it assumes that there is only one valid way to use these "references identifiers". I can walk into a book store with an ISBN and order the book. Is that "a unique resource" within the meaning of WP:ELNO#EL1? I don't think so, but it's still a valid use of the reference identifier. Similarly, I suspect that just as I gravitate towards PMIDs instead of DOIs, other readers may have a preference for various identifiers (e.g., if they're using citation management software that plays well with Bibcodes but not with PMIDs). I don't see any reason to give them "my" preferred identifier and hide theirs. WhatamIdoing (talk) 17:18, 13 December 2025 (UTC)[reply]
Possible support with different wording. A bibcode links to a database, so by definition it gives more information than a doi alone. It is also being expanded to cover fields other than astronomy. The new SciX replacement coming has a neat field indicating whether the paper as been refereed. The question is what kind of information do we need for references? I propose including the bibcode only if it links to a full copy of the paper with no subscription required and the doi does not, or if the paper does not have a doi.
The bibcode and ADS have become so successful we may not need them for many astronomy articles anymore. The American astronomy journals have gone open access and also put their pre-1998 archives into ADS. A doi for one of these old astronomy articles is translated into a bibcode on access, and ends up at the database. See doi:10.1086/115657. StarryGrandma (talk) 20:52, 12 December 2025 (UTC)[reply]
If ADS hosts a full copy of the paper, I think it's fine to include a bibcode, irrespective of what the DOI has. If there is a DOI and a bibcode which lead to the same webpage, I think we should just use one of them (the bibcode is probably better in that case, but either would be fine). –jacobolus (t) 22:03, 12 December 2025 (UTC)[reply]
You have a good point. I just ran into a small journal where the full copy in ADS downloads easily but the journal site keeps timing out. StarryGrandma (talk) 23:09, 12 December 2025 (UTC)[reply]
In my opinion these cases are covered by the proposal. This is proposal is not against using bibcode when they are uniquely useful. It is against using them routinely even when they offer nothing but a waste of time. Johnjbarton (talk) 00:45, 13 December 2025 (UTC)[reply]
Why do say the bibcode would be better in such a case? DOIs are far more widespread, so they should be the better choice when in doubt. Gawaon (talk) 03:08, 13 December 2025 (UTC)[reply]
In many cases when full copies are available through multiple sources, they are behind different paywalls and different copies might be accessible to different readers. This often happens to me with articles on JSTOR (which I can read courtesy of my employer) for which the official doi version points elsewhere. The same is likely true with some ADS full-text versions. So I am happy to keep both bibcodes and dois for those ones; its the ones that don't provide full text that are problematic. —David Eppstein (talk) 03:22, 13 December 2025 (UTC)[reply]
But my reply was about the situation where both "lead to the same webpage", excluding scenarios such as the one you mention. It's more or less agreed that if the bibcode provides additional useful information (such as another full-text link), it's helpful to add and keep it. Gawaon (talk) 03:34, 13 December 2025 (UTC)[reply]
Let me be more precise:
  • If there is a DOI for the publisher's website which points somewhere different than the bibcode, and the bibcode contains unique information (e.g. has a full-text scan, or some data files or code associated with the paper), then they should both be listed.
  • If the only available DOI is just a redirect to the bibcode page, we should only keep one of the two; I think it's preferable to keep the bibcode in that case, because it gives readers a better idea where their click will go, but either one or the other would be fine.
  • If there is a DOI or other link to a publisher's site or JSTOR or comparable and the bibcode for the paper leads to a page which does not contain unique information (e.g. only has basic citation metadata, or only citation metadata + abstract but the abstract is also available at the publisher's site) I think we should leave out the bibcode.
–jacobolus (t) 05:09, 13 December 2025 (UTC)[reply]
That seems largely equivalent to Johnjbarton's original proposal if one prefers the DOI when in doubt. And too complicated a solution won't work anyway, so that's probably for the best. Gawaon (talk) 09:02, 13 December 2025 (UTC)[reply]
"A bibcode links to a database, so by definition it gives more information than a doi alone." Please provide one example. I provided multiple examples of the opposite. Johnjbarton (talk) 00:43, 13 December 2025 (UTC)[reply]
The example doi:10.1086/115657 is registered by the American Astronomical Society itself, see https://api.crossref.org/works/10.1086/115657 . Which makes me wonder, is there any case where a bibcode leads to a fulltext which is not provided by the corresponding DOI's destination? Last time I checked, such cases were exceedingly rare (say, about 100 records out of 26 million ADS records), but right now I can't find a single one. There are some cases like records for 1970s articles in soviet journals which claim to have a PDF but actually don't. (Not sure if the translation allegedly attached to the record went missing in one of the recent pointless redesigns.) Nemo 14:03, 22 December 2025 (UTC)[reply]
I have seen cases where there was a scan in ADS and also a separate paywalled publisher link which did not provide free access to the full text (I don't have any examples at hand though). I have seen cases where there was a scan in ADS and a different scan freely available from the publisher. I have also seen cases where there was a scan available in ADS and as far as I know no DOI at all (e.g. Bibcode:1984QJRAS..25..126S). This proposal would support keeping bibcodes in all such examples. –jacobolus (t) 14:38, 22 December 2025 (UTC)[reply]
  • Question, aren't DOI's subject to a lot of linkrot? This search shows 8,336 articles with broken-doi notes. Abductive (reasoning) 14:57, 13 December 2025 (UTC)[reply]
    Well, in theory they are meant to last forever. In practice, of course, things aren't always quite as neat as in theory. Gawaon (talk) 15:17, 13 December 2025 (UTC)[reply]
    As far as I can tell from
    • Singh, H., West, R., & Colavizza, G. (2021). Wikipedia citations: A comprehensive data set of citations with identifiers extracted from English Wikipedia. Quantitative Science Studies, 2(1), 1-19.
    in 2020 there were around 1 million DOIs in wikipedia, so that puts the link rot around 1%. In my experience that is low compared to other issues with citations. In any case, the proposal here is not affected one way or another. Johnjbarton (talk) 18:11, 13 December 2025 (UTC)[reply]
    Naw, I think that broken DOIs never get fixed and are removed from the citations over time. And also people shouldn't count on the PMID and PMC identifiers staying up, as they are US government websites currently controlled by anti-vaxxers. During the shutdown the US Census and GNIS websites went down. Abductive (reasoning) 21:25, 13 December 2025 (UTC)[reply]
    DOIs are still valid unique identifiers and remain so forever. If broken, they can be restored at a later date (I have seen that happen after someone inquired about one), and I wouldn't advice removing even a broken DOI from a reference. It's less useful than a working one, but not completely useless. Gawaon (talk) 02:43, 14 December 2025 (UTC)[reply]
    Some broken DOIs are fixable, they are simply edit errors. Examples: [5] [6] [7] [8]. I think the citation bot could repair these. Johnjbarton (talk) 03:30, 14 December 2025 (UTC)[reply]
    I have also encountered broken dois from still-active journals when they rearranged their web site. Contacting the publisher can often get them to fix these. —David Eppstein (talk) 08:37, 14 December 2025 (UTC)[reply]
    Yes, I've run into a few of those. In a few cases that can make the source inaccessible; it is better to have other alternatives, when possible. Praemonitus (talk) 05:50, 22 December 2025 (UTC)[reply]
    This remains irrelevant to the current proposal, which is not about alternative ways of accessing the source. When a DOI is broken, a bibcode which leads to an ADS page containing nothing but redundant article metadata and a "publisher link" consisting of a DOI is not going to be of any additional use to readers. If they click the bibcode and then click through to a broken DOI, they'll end up at exactly the same 404 page as if they clicked a broken DOI directly from Wikipedia. –jacobolus (t) 05:59, 22 December 2025 (UTC)[reply]
    What jacobolus said. But I'll answer the question anyway: it's pretty rare for DOIs to fail, although it happens more often with some incompetent legacy press. Some mistakes take longer than others to solve. A DOI will nearly always return at least the metadata record associated with it. There can be some complicated situations where it's not clear who "owns" a DOI (or not), e.g. if a publisher is failing but not completely dead yet; or when a publication is sold to a new publisher and remains in a limbo for a while. The full text is often not archived, but when it is the DOI eventually directs to an archive when the publisher fails. Nemo 14:03, 22 December 2025 (UTC)[reply]
  • Oppose - I think that it is valuable for readers who look at citations (I grant that such readers may be in the minority) to have as many methods of getting to the source material as possible. This allows readers to exercise their own choice in terms of their preferred method of looking things up, and guards against future failure of other links. What harm does an additional link do? Cheers, SunloungerFrog (talk) 10:11, 14 December 2025 (UTC)[reply]
    We normally edit article to remove two sentences that say essentially the same thing, saving the one that provides the best explanation. We should apply that principle here, for the same reason. The vast major of bibcodes waste editor time when verifying sources. The ADS site is often slow and the page it delivers is almost always pointless. The bibcodes are extraneous junk characters except in few cases. By asking editors to only include bibcodes that provide value we make them much more valuable. Bibcodes are not an alternative lookup method as I demonstrated with my example above. Johnjbarton (talk) 16:45, 14 December 2025 (UTC)[reply]
    I think that there is a big difference between removing redundancy in article content, which I agree is a good idea, and removing redundancy of possible access to article sources, which I think is a bad idea. Cheers, SunloungerFrog (talk) 17:11, 14 December 2025 (UTC)[reply]
    The bibcodes under discussion here do not provide access to the sources. They only contain metadata and redundant links to the publisher's site. –jacobolus (t) 01:39, 15 December 2025 (UTC)[reply]
  • Support: bibcode is the first or second worst offender in terms of cruft added to citation templates. It could be ok to add it to the wikitext if templates are modified to not display it. When the bibcode leads to a full text, the direct link to the PDF should preferably be in the URL parameter. Nemo 17:54, 14 December 2025 (UTC)[reply]
  • Question Somewhere in the prior discussions about bibcodes, it was pointed out that the ADS website gives information on which articles cite which. It was argued that their citation graph is too incomplete for this to really be useful information. I'm not sure of that and am open to being convinced either way. (I think the relative completeness might be a field-by-field thing that probably depends on how much that corner of science uses the arXiv. Speaking very informally, the citation counts in ADS seem to be lower in my experience than those from Google Scholar, which isn't always a bad thing because GS picks up random PDFs from all over the place, and sometimes ADS catches things that GS misses.) Is there a consensus that the ADS citation graph is not useful information? Stepwise Continuous Dysfunction (talk) 21:37, 14 December 2025 (UTC)[reply]
    There are a large number of citation graph websites. I don't think linking (any or all of) them is useful enough to justify such links, especially since going to the site and typing in the DOI or other metadata takes only a few seconds for any reader who cares (or faster using a bookmarklet or browser extension), and then they can pick between Scopus, Web of Science, Google Scholar, Internet Archive Scholar, CiteSeer, OpenAlex, Semantic Scholar, ADS, OpenCitations Corpus, or whichever other, based on personal preference. Cross-linking all of the citation graphs seems like another thing that could be done at Wikidata or some similar venue, but it doesn't seem relevant to Wikipedia. –jacobolus (t) 01:48, 15 December 2025 (UTC)[reply]
    I updated my list of bibcode examples to include the number of citations reported by ADS and Google Scholar. Scholar, in my experience, is inflated a bit, including some sources that don't cite the root article but I guess the biggest issue is that the ADS database is limited and its citation tree is only among the articles in its database. Johnjbarton (talk) 16:45, 15 December 2025 (UTC)[reply]
    The ADS database is limited, and other citation-graph services exist, but does that make the citation information brought up by clicking a bibcode useless information? It's still information that no other link in the reference would immediately provide. Stepwise Continuous Dysfunction (talk) 22:54, 15 December 2025 (UTC)[reply]
    There has to be sufficient benefit to balance the extra distraction caused by adding each new identifier. I could plausibly imagine that for an article about astronomy or astrophysics, where the relevant part of the citation graph is mostly included in the database, many sources may have full text and/or extra data files available, and readers in the field may be heavy users of ADS, it might be worthwhile to include bibcodes for every citation which has one. But for Wikipedia articles about other topics, the content hosted at ADS is entirely redundant with the citation metadata already included in Wikipedia, the database only contains an arbitrary small fraction of the citation graph, and typical readers (including experts) are unfamiliar with bibcodes and destined to be disappointed or even confused if they click one. –jacobolus (t) 09:01, 16 December 2025 (UTC)[reply]
  • Comment I could get on board with a rule that bibcodes generally shouldn't be included when arXiv IDs are, since the ADS page is only a click away from the abstract page that the arXiv serves up. Stepwise Continuous Dysfunction (talk) 22:55, 15 December 2025 (UTC)[reply]
  • Oppose Bibcodes are useful. That you don't personally find them useful is not a reason to ban them. Headbomb {t · c · p · b} 23:22, 21 December 2025 (UTC)[reply]
    Please read the proposal: bibcodes are not being banned. Johnjbarton (talk) 23:33, 21 December 2025 (UTC)[reply]
    Although about a different context, WP:USEFUL is also relevant: to whom are these particular bibcodes (the ones we are discussing, the ones that do not provide information) useful, and why is that an encyclopedic purpose? —David Eppstein (talk) 00:26, 22 December 2025 (UTC)[reply]
  • Comment: in many cases the ADS page linked by the bibcode provides a free ADS_SCAN, giving an accessible copy of the article. That's reason enough to retain bibcodes. But I'd be okay with the bots only checking articles under the 'category:astronomy' tree. Probably what the bots should be doing is checking for citation urls that point to the bibcode link and replacing that with the bibcode parameter. Praemonitus (talk) 01:31, 22 December 2025 (UTC)[reply]
    Yes, this is why the proposal specifically calls out cases where the ADS page adds value. In my experience ADS add value for older astronomy sources but not across the entire spectrum of wikipedia. The proposal is not about removing bibcodes. My sole goal is to stop edits that add them when they are just noise. Johnjbarton (talk) 01:57, 22 December 2025 (UTC)[reply]

Small improvement to "source capitalization" sentence?

[edit]

WP:CITESTYLE includes this guidance: "Preserving the capitalization style of each individual source is not considered a consistent style." That seems a bit ambiguous, since editors could interpret "preserving" as:

a) The cap style found in a "References" section when the editor encounters the article (guidance similar to MOS:STYLEVAR); or
b) The cap style found within the source (that is, how the source styles its own title)

The RfC from early 2025 that established this guideline clarifies that it is the latter. MOS experts that participated in the RfC may not see any ambiguity, but this style question comes up a lot in reviews (GA, PR, FA) so making it crystal clear to non-MOS experts may be helpful. Would it be an improvement to re-word this green sentence to something like: "Preserving the capitalization style that each individual source uses for itself is not considered a consistent style."? Noleander (talk) 13:52, 11 December 2025 (UTC)[reply]

Sounds acceptable, though I'm not sure it's really needed. Gawaon (talk) 14:50, 11 December 2025 (UTC)[reply]
I'm fine with that. Firefangledfeathers (talk / contribs) 14:51, 11 December 2025 (UTC)[reply]
We could also link to the RFC, maybe under the words "not considered". WhatamIdoing (talk) 17:33, 11 December 2025 (UTC)[reply]
I believe the sentence already has a footnote that links to the RFC. Not sure if that satisfies your desire or not. Noleander (talk) 17:38, 11 December 2025 (UTC)[reply]

Seeing no objections, I made the change. I also moved the footnote (which linked to the RfC that established the guidance) from the middle of the sentence to the end. Noleander (talk) 15:47, 12 December 2025 (UTC)[reply]

Discussion on references vs. reflist

[edit]

There is a discussion at Help talk:Footnotes § Tag or template preference which may be of interest. -- Beland (talk) 05:28, 18 December 2025 (UTC)[reply]