Wikisource:Scriptorium

SpBot archives all sections tagged with {{section resolved|1=~~~~}} after 3 days and sections whose oldest comment is older than 30 days.

Proposals

Proposal: Modify the new WS:CSD criterion regarding non-English works

The new criterion at WS:CSD currently reads: "Author pages with no published English-language works by the author listed". However, there are many Author pages with published non-English works listed, that would be deleted under this criterion. These pages are not "empty", and they are not relevant to the cleanup of Category:Authors with no works for which the criterion was initially proposed.

The deletion of Author pages where no known English translations exist, is a separate deletion rationale than the deletion of Author pages that do not list any works. If non-English authors are to be speedied, I believe they should at least be listed under a separate criterion in WS:CSD.

As such, I propose that the new criterion be reworded as follows: "Author pages with no published works by the author listed". —Beleg Tâl (talk) 12:04, 18 October 2025 (UTC)

@Jan.Kamenicek @EncycloPetey courtesy ping —Beleg Tâl (talk) 12:05, 18 October 2025 (UTC)

I am OK with such wording. Just to make sure we all understand it the same: this wording means that empty Author pages without any list of works get deleted, but if there is a list of published non-English works, it cannot be a subject of speedy deletion, but instead some search whether published translations of these works exist is required. The Author page can still be nominated at the Proposed deletions and if no published translations of these non-English works are found, it gets deleted anyway. --Jan Kameníček (talk) 15:57, 18 October 2025 (UTC)

I am not in favor of throwing the burden on to the proposer for deletion to demonstrated they did the required search. We don't have the requirement in other contexts, e.g. if no source or no license it isn't on the proposer for deletion to do a thorough search and say, "I couldn't find a source". Searching for translations is not necessarily trivial as they might appear in a bunch of names, periodicals, etc. It's fine to require going through individual proposed deletion however as a way to allow curing and search by interested parties and to work through the backlog here systematically. MarkLSteadman (talk) 16:29, 18 October 2025 (UTC)

E.g. someone puts up a 1920s Ethiopian poet listing only titles in Ethiopian, I shouldn't have to prove that I exhausted proven the non-existence of a PD / free-licensed translation to nominate it for deletion. MarkLSteadman (talk) 16:33, 18 October 2025 (UTC)

Agree. It should be enough to nominate it for deletion and give the community a chance to find and list some translation. It nothing is listed in a reasonable time, it will get deleted. --Jan Kameníček (talk) 18:13, 18 October 2025 (UTC)

We do host some works that are transcriptions of letters, manuscripts, and other forms of unpublished documents for notable historical figures and historical documents. We also explicitly permit theses that have undergone review by committee, even if unpublished. I therefore would add the phrase "...nor any listed works permitted under WS:WWI. --EncycloPetey (talk) 19:37, 18 October 2025 (UTC)

Or just simplify, "Author pages with no hostable works listed" —Beleg Tâl (talk) 02:09, 19 October 2025 (UTC)

That returns us to the previous issue of having to evaluate the list prior to performing a speedy deletion, and permitting speedy deletion of Authors that do have listed works. A speedy deletion should not require that level of evaluation. --EncycloPetey (talk) 02:53, 19 October 2025 (UTC)

I'm not sure I understand you (genuinely). Do you want to avoid having to evaluate the list prior to speedy deletion, and also avoid permitting speedy deletion of Authors that have listed works? If so, I think the simplest wording would be "Authors with no works listed". —Beleg Tâl (talk) 14:33, 20 October 2025 (UTC)

I note that prior to mass-removing non-English language works that have long been listed on an Author page, one should probably check whether the appropriate foreign-language Wikisource is listing these works already, and, when in doubt, make sure to at least move the listing to the relevant Author Talk page so that the suitability of intweriki-transfering it can be explored. I have now done this for Archibald Pitcairne, where a number of Latin works were listed and were recently removed. (I've started a discussion to this end on Latin Wikisource at la:Vicifons:Scriptorium#Quaestio de Archibaldi Pitcarnii scriptoris operum recensione). Can we agree on this as standard policy going forward, so that information about extant works that may be appropriate for Wikisource as a whole is not inadvertently lost? --~2025-27371-40 (talk) 21:01, 19 October 2025 (UTC)

Well, if you wish to move it somewhere, you can move it directly to the appropriate language Wikisource. However, this discussion is about something different — we’re trying to find the best wording to reflect the result of the previous vote, so let's rather stay on topic. --Jan Kameníček (talk) 21:09, 19 October 2025 (UTC)

Since this discussion may or may not itself result in mass-removal of extant information about foreign-language works from English Wikisource, it's not obviously offtopic. Addressing your reply, I believe it would be inappropriate to expect en-WS users to "move [content] to the appropriate language Wikisource" as a matter of standard procedure, since interwiki transfers can be challenging for many reasons - e.g. they must be consistent with practices and policies on the destination Wikisource. Posting the content on the Author talk page, pending exploration of a suitable interwiki transfer, may then be a suitable compromise approach. --~2025-27371-40 (talk) 22:14, 19 October 2025 (UTC)

Since this discussion is about the deletion of Author pages, and since talk pages of deleted pages are also speedied under WS:CSD M4 "Orphaned talk page", moving content to the talk page wouldn't help for any of the pages affected by this policy. It's a good idea though, for any Author pages for which this is a concern. —Beleg Tâl (talk) 14:40, 20 October 2025 (UTC)

So, we have three kinds of wording of the speedy deletion provision suggested:

1. Author pages with no published works by the author listed.
2. Author pages with no published works by the author listed nor any listed works permitted under WS:WWI.
3. Author pages with no hostable works listed.

My personal understanding of these points is:

No. 1 is clear. It allows to speedy delete only pages not listing anything published. Thus pages with lists of works published in foreign languages cannot be speedied, but they can be nominated for deletion at WS:PD to give community some time to list their English translations if such exist.

No. 2 further restricts the range of pages eligible for speedy deletion, excluding pages that list works which were not published but are permitted to be hosted (such as theses that have been reviewed and approved by a thesis committee at an accredited university, or scientific research—whether peer-reviewed or not—if the author meets Wikipedia’s notability guidelines).

I am not in favor of this, as it makes the speedy deletion process unnecessarily complicated. It would require the proposer to verify details such as whether the university in question is accredited, among other things. Moreover, I find the idea of hosting lists of unpublished scientific research that might one day be accepted here unconvincing, as is the need to analyze an author’s notability in such cases. In my view, author pages of this kind should only be created if we actually host their works.

No. 3 allows to speedy delete also pages with lists of foreign language works, which is not allowed under no. 1 or 2.

If anyone wishes, additional points can be added to the list. To avoid any misunderstanding, it might be better to hold a regular vote this time. To keep things fair, I suggest starting the vote, for example, once there have been no new comments here for five days. --Jan Kameníček (talk) 22:22, 30 October 2025 (UTC)

Number 3 is too broad, and would allow for speedy deletion of pages that are explicitly permitted by policy and by consensus, such as notable persons. It also requires research on the part of the person performing the deletion beyond what should be necessary for a speedy deletion. Is Seneca's Hercules Oetaeus hostable? Yes, but you can't tell that without researching it. Is "Komarekiella atlantica gen. et sp. nov. (Nostocaceae, Cyanobacteria)" hostable? The title is in Latin, but the article is in English. It might be hostable, but you wouldn't be able to tell from a list simply bearing the title, and it would require an investigation into its copyright status. Is Lucan's "Pharsalia" hostable? Yes; there are several public-domain English translations. Strindberg's Fordringsägare? Yes, with at least two English translations in public domain. Menander's Dyskolos? Maybe; it depends on the date of the translation, since a big discovery of a new papyrus was made in 1959, and translations based on that text won't be in public domain yet. None of this could be evaluated as part of a "speedy" deletion. It would require feedback from people doing research, or who are well acquainted enough already with the literature. --EncycloPetey (talk) 01:10, 31 October 2025 (UTC)

The onus would be on the author creator to provide a link to the translation. So you want to create a page for Strindberg, you have to provide an instance of a translation. It's like CS:6 for Author pages listing all copyrighted work. If someone creates a modern author page listing 10 copyrighted books, it isn't on the admin to do an exhaustive search to see if they authored a freely-licensed journal article somewhere. MarkLSteadman (talk) 15:14, 31 October 2025 (UTC)

"Published" is a little bit tricky, between common usage (e.g. a book was printed by a publishing house) and copyright office use (e.g. "If an author places copies of their new short story in a library book exchange box at the end of their driveway this constitutes publication of that short story.") and things like "limited publication" (e.g. sending a book to a publisher and it being rejected). So e.g. are works that have been "limited" published, "published" acording to criteria 1 (which would already cover the theses example)? MarkLSteadman (talk) 15:01, 31 October 2025 (UTC)

Indeed. I honestly don't like any of the three wordings listed above. I prefer my original suggestion: "Author pages with no works by the author listed". --EncycloPetey (talk) 17:29, 31 October 2025 (UTC)

I'm in favour of this wording tbh —Beleg Tâl (talk) 23:10, 1 November 2025 (UTC)

Unless we discuss copyright, we usually mean the first meaning of the term, as it is also used in WS:WWI, and in this sense it is also meant here. The wording EncycloPetey has suggested here can be added to the voting list. --Jan Kameníček (talk) 23:00, 1 November 2025 (UTC)

• Jan Kameníček, I think your consideration for No. 2 is too far-fetched. I think, in such a case, that it is reasonable and desirable, in case of uncertainty as to whether a work qualifies for speedy deletion, to bring it under normal deletion. I think it is good that grounds for speedy deletion are limited. In a deletion discussion, the onus should be on the creator to just retention; but in the case of a speedy deletion, where the creator does not always have the opportunity to respond, then the onus should be on the person seeking deletion. TE(æ)A,ea. (talk) 16:45, 31 October 2025 (UTC)

  My issue with #2 is that it is too arbitrary. Rewording the suggested criterion slightly, "Author pages where all listed works (if any) are both unpublished and also fail WS:WWI" seems like a strange line in the sand to draw, especially since it's tangenital to the original purpose of the criterion (i.e. empty author pages). Also, I suspect that the few Author pages that list only unpublished non-hostable works are niche enough that I'd prefer they go through WS:PD regardless. —Beleg Tâl (talk) 15:16, 4 November 2025 (UTC)

Voting

Let's start voting to conclude the issue finally. --Jan Kameníček (talk) 15:37, 29 November 2025 (UTC)

1. Author pages with no published works by the author listed.
2. Author pages with no published works by the author listed nor any listed works permitted under WS:WWI.
3. Author pages with no hostable works listed.
   •  Support --Jan Kameníček (talk) 15:38, 29 November 2025 (UTC)
   •  Support — Alien  3
     3 3 09:25, 30 November 2025 (UTC)
   •  Support And opposing the options with unpublished works not being permitted, since unpublished works of notable people would be reasons to include them as author pages in my view. SnowyCinema (talk) 19:27, 29 December 2025 (UTC)
   •  Support —Beleg Âlt ^BT (talk) 20:06, 29 December 2025 (UTC)
4. Author pages with no works by the author listed.
5. Author pages where all listed works (if any) are both unpublished and also fail WS:WWI.

While the policy itself was accepted with a solid number of votes, the poll on the precise wording has attracted only two votes so far, so I will wait a while longer before closing it. --Jan Kameníček (talk) 23:58, 21 December 2025 (UTC)

Given the number of voters, it seems that the exact wording of the adopted policy does not matter much to contributors. If no further votes are cast, I will close the discussion with the result “Author pages with no hostable works listed.” Let’s give it one more week. --Jan Kameníček (talk) 19:16, 29 December 2025 (UTC)

I am closing the discussion aimed at refining the wording of the policy on speedy deletion of empty author pages, adopting the following preferred wording: "Author pages with no hostable works listed." --Jan Kameníček (talk) 20:27, 5 January 2026 (UTC)

Bot approval requests

• See Wikisource:Bots for information about applying for a bot status
• See Wikisource:Bot requests if you require an existing bot to undertake a task

Repairs (and moves)

Designated for requests related to the repair of works (and scans of works) presented on Wikisource

See also Wikisource:Scan lab

Index:Online Safety Act 2023 (Category 1, Category 2A and Category 2B Threshold Conditions) Regulations 2025.pdf

Please move to Index:The Online Safety Act 2023 (Category 1, Category 2A and Category 2B Threshold Conditions) Regulations 2025 (UKSI 2025-226 kp).pdf to match commons. ToxicPea (talk) 02:49, 19 December 2025 (UTC)

Index:Online Safety Act 2023.pdf

Please move to Index:Online Safety Act 2023 (UKPGA 2023-50 kp).pdf to match commons. ToxicPea (talk) 02:49, 19 December 2025 (UTC)

Index:Public Order Act 2023.pdf

Please move to Index:Public Order Act 2023 (UKPGA 2023-15 kp).pdf to match commons. ToxicPea (talk) 02:52, 19 December 2025 (UTC)

Index:The aborigines of Australia.djvu

Please finish migrating this to Index:The Aborigines of Australia (1988).djvu and fix the transclusion. It looks like the notice not to edit or proofread the 1888 scan was put there in 2011(!!!!), so this needs to be finished up. SnowyCinema (talk) 09:22, 6 January 2026 (UTC)

Other discussions

Are the Trump plans PD Edict?

Do Trump 28-point Ukraine–Russia plan (20 November 2025 version) and Trump 28-point Ukraine–Russia plan (23 November 2025 version) qualify under template:PD-EdictGov? These are stated by generally reliable sources to be what are implied to be word-for-word copies of specific versions of the original document, so they are various versions of a US government legal-type document. On en.Wikipedia, @Diannaa: removed the links to these two wikisource documents, stating in the edit summary that these are links to copyright violations, i.e. effectively suggesting that these are not PD.

Are they PD or not PD? Boud (talk) 00:36, 25 November 2025 (UTC)

• I would not trust English Wikipedia editors as to competence, especially on copyright issues (although in that respect I am more familiar with Wikimedia Commons). Neither of these are “edicts” as they are general or generic policy statements. The November 20 plan, which is a U.S. government work, is in the public domain for that reason; the November 23 plan is apparently a European government work, and so is likely copyrighted. TE(æ)A,ea. (talk) 21:15, 25 November 2025 (UTC)
  • The 20 Nov version as a US government work at least bypasses the question of whether they are edicts (by "question" I mean that I'm unfamiliar with the interpretation of what counts as an edict; I'm not saying that you're wrong).
    For the 23 Nov version, my understanding is that it's a derived version of the 20 Nov version, and I had thought that it was issued with US considered to be a co-author, i.e. it still is a US government document in that sense. However, that's not in the Reuters source, and currently I don't see any sources stating that the 23 Nov derived version is still a US government document (they're not interested in the copyright aspect), although they agree that it's a derived document, "By Sunday, a revision had emerged. It took the US plan as its basis and made steady amends, with several notable differences. "
    The Reuters source itself similarly says "... drafted by Europe's E3 powers of Britain, France and Germany, takes the U.S. plan as its basis but then goes through it point by point with suggested changes and deletions". Independently of whether it still qualifies as a US government document, the modified PD document probably takes on the additional copyright of the extra authors. Help:Official texts has no guide for France. Help:Official texts#Germany has "is in the public domain according to German copyright law [if] it is part of a statute, ordinance, official decree or judgment (official work) issued by a German federal or state authority or court" - which doesn't mention draft documents. Help:Official texts#United_Kingdom says "Since 2010, almost all information owned by the UK Crown is offered for use and re-use under the Open Government Licence by authority of The Controller of His Majesty's Stationery Office.".
    Is a modified US gov document still a US gov document (for the 23 Nov version)?
    Does {{OGL3}} apply for the 23 Nov version, since the UK Crown effectively "owns" a copy of the document? Boud (talk) 13:07, 26 November 2025 (UTC)
    • The 23 Nov version does not seem to qualify for any of the exceptions to OGL3 in UK. "Government policy is that public sector information should be licensed for use and re-use free of charge under the Open Government Licence (OGL) with only a few exceptions..." So OGL3 seems likely to apply. Boud (talk) 13:18, 26 November 2025 (UTC)

oh, they are PD. English admins are frequently wrong but never in doubt. and they don't play nice with sister projects. but a problem is, can't find a text source for the document, unlike Report On The Investigation Into Russian Interference In The 2016 Presidential Election. that would be an improvement. but this executive isn't known for text openness. --Slowking4 ‽ digitaleffie's ghost 23:36, 8 December 2025 (UTC)

Tech News: 2025-50

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Weekly highlight

• Anybody who wishes to secure their user account can now use two-factor authentication (2FA). This is available to all registered users of all Wikimedia projects. This is part of the Account Security initiative. Later, 2FA will be required for all users who can take security- or privacy-sensitive actions.

Updates for editors

• Following last week's deployments, the Add a link feature, which allows editors to add suggested links during editing, will be available to an additional 33 Wikipedias starting on 9 December. This expansion is possible thanks to the new prediction model that now supports all languages, including those that were previously not covered. While the feature has been available on most Wikipedias for some time, this rollout brings us closer to using the improved model everywhere. If you have any questions or would like more details please contact Trizek (WMF).
• Last week, the Search Platform team added transliterated as-you-type search suggestions to Georgian wikis. If there are only a few regular search suggestions, then queries in Latin or Cyrillic script are now rewritten into Georgian script to look for more matches. For example, searching for either bedniereba or бедниереба will now suggest the existing article about ბედნიერება ("happiness"). You can recommend other languages where transliterated suggestions would be useful on Phabricator for future development.
• Later this week, a controlled experiment will begin for editors on the 100 largest Wikipedias who are editing a section in the mobile web visual editor. 50% of these editors will notice a new "Edit full page" button that will enable them to expand their editing session to the whole page. This feature is intended to make it easier for people on mobile web to edit any article section, regardless of which section-edit icon they tapped to begin. The experiment will last ~4 weeks. You can find more details about the project.
• Later this week, the Reader Growth team will launch a mobile web experiment to expand all article sections by default (currently they are collapsed by default) and pin the section header the user is currently reading to the top of the page. The experiment will affect 10% of users on Arabic, Chinese, French, Indonesian, and Vietnamese Wikipedias. [1]
• The Wikipedia Year in Review 2025, a feature in the Wikipedia mobile apps (iOS and Android) that provides users with a personalised summary of their engagement with Wikipedia over the year, is now available on the iOS and Android apps. This edition includes expanded personalised insights, improved reading highlights, new donor messaging, and updated designs. Open the app to view your Year in Review and explore your reading journey from 2025.
• A recent software bug caused edits made with VisualEditor to make unintended changes to wikitext, including removing whitespace and replacing spaces with underscores in wikilinks inside citations. This was partially fixed last week, and further fixes are in progress. Editors who used VisualEditor between November 28 and December 2 should review their edits for unexpected modifications. [2]
• View all 23 community-submitted tasks that were resolved last week. For example, the incorrect handling of URLs copied from the address bar of Microsoft Edge users, has been resolved. [3]

Updates for technical contributors

• Starting this week, users of the "Improved Syntax Highlighting" beta feature will have CodeMirror as the editor for Lua, JavaScript, CSS, JSON and Vue content models, instead of CodeEditor. With this, the linters will be upgraded. This is part of a larger effort to eventually replace CodeEditor and provide a consistent code editing experience. [4]
• Developers are encouraged to take the 2025 Developer Satisfaction Survey, which remains open until 5 January 2026. If you build software for the Wikimedia ecosystem and would like to share your experiences or feedback, your participation is greatly appreciated. [5]
• There is no new MediaWiki version this week.

Tech news prepared by Tech News writers and posted by bot • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

MediaWiki message delivery 17:45, 8 December 2025 (UTC)

Inconsistent/misleading dates in Index:The Girl From the Marsh Croft-1916.djvu

The current PotM is a book published in 1916. It consists of a series of short stories (might be chapters, I haven't read it). It seems to have a problem, presumably linked to data in Wikidata, in that each short story has a date of 1910 in the header, which is clearly not consistent with the publication date of the book.

Also, I thought that for multi-chapter/section works, the publication date should only appear on the main space transclusion. Chrisguise (talk) 06:42, 9 December 2025 (UTC)

@Chrisguise: I fixed all of these instances, since you're absolutely right that these are incorrect uses of the Wikidata items. For subwork version items, within collections, the date should match the collection's date. The hierarchy: subwork version is part of collection version (not collection work). The original Swedish short story collection seems to have been from 1908, with the first English translation being from 1910 (according to Wikipedia), but this particular version/translation of the collection is from 1916, so all the subwork versions from it inherit that specific year. SnowyCinema (talk) 07:06, 9 December 2025 (UTC)

Confusion often occurs when the printing date and copyright date are different. What determines the date of an edition? Is in the initial print date? The copyright date? The date of the print run? And the answer to those questions might be different for different works / translations.

In the case of Loeb Classics, for example, the date on the title page is the date of the print run, even when the edition is identical in content and series volume number to earlier impressions. I have one such volume where the colophon indicates the initial print date, seven dates of reprints, the date the text was first emended, then two dates the emended text was reprinted. Most people would consider that list to consist of only two editions: the original text and the emended text. But there are nine additional impression dates. Wikidata has the means to recognize the two differing editions, but no setup for tracking the reprints without either making them independent editions or consolidating them with the editions they are part of. --EncycloPetey (talk) 08:56, 9 December 2025 (UTC)

Display differences between The Poetical Works of James Beattie/The Minstrel/Book 1 and The Poetical Works of James Beattie/The Minstrel/Book 2

Throughout this work ('The Minstrel') the stanza numbers are a different font size to the main text. I chose to format these numbers as, for example {{asc|XXVII}}. In the transclusion of Book 1, the stanza numbers display as small caps but in the transclusion of book 2 they do not. The only difference I can see between the two is that in book 1 'title = [[../../]]', whereas in book 2, title = {{auto parents}}. Anyone know why this difference appears to cause the display discrepancy? Chrisguise (talk) 23:15, 11 December 2025 (UTC)

• Chrisguise: {{asc}} calls a style sheet, without which it will not display properly. What matters is where the first instance is—which is in the name of the author in the quote before the poetry. Now, when you call a template within a Wiki-link, the style sheet is suppressed and does not work (and this failure carries on for the entire page, including through transclusion). Looking at the two examples, that is exactly what has happened: the Wiki-link to Virgil is inside {{asc}} in Book 1, while the Wiki-link to Horace encloses {{asc}} in Book 2. I have made this change, and the stanza numbers in Book 2 now display properly. TE(æ)A,ea. (talk) 23:42, 11 December 2025 (UTC)

  Well I'll go to the foot of our stairs! Chrisguise (talk) 02:53, 12 December 2025 (UTC)

Index:A Welsh Grammar, Historical and Comparative (IA welshgrammarhist00joneuoft).djvu

Newly uploaded. Can someone migrate the PNG based scans (already proofread over to it)? ShakespeareFan00 (talk) 08:48, 12 December 2025 (UTC)

 Comment Working on this now. SnowyCinema (talk) 22:10, 12 December 2025 (UTC)

Blanked talk pages

There are a good number of blanked talk pages, for example Talk:An Ariette for Music, which are blanked because the work is now scan-backed so doesn't need the {{textinfo}} from the previous non-scan-backed version. Consensus seems to be not to delete those talk pages, for historical reasons. But is there any objection to us making a template similar to {{Blanked userspace page}} to put on these talk pages so that editors, for example those looking to do maintenance on talk pages, know why this happened / to stop the talk pages from appearing in any technical backlogs for pages with 0 characters? SnowyCinema (talk) 14:13, 13 December 2025 (UTC)

It sounds like a good idea to me. —Tosca-the-engineer 19:32, 13 December 2025 (UTC)

Done, see {{Blanked talk page}}. SnowyCinema (talk) 18:10, 29 December 2025 (UTC)

Orphaned and ungrouped .jpeg files of music scores

As part of the discussion here about the composer Author:Charles-Marie Widor, I came across a large number of scans of individual pages imported to Wikisource from Commons with separate File namespace pages for each jpeg image, and from my admittedly small sample size, they all seem to be orphaned (unlinked) pages too. The files are well-named; each composition has a unique name, and they all have page numbers, e.g. File:Délivrance ! - chanson de route - paroles de A. Chuquet ; musique de Ch.-M. Widor - bpt6k3852287 (1 of 6).jpg, but making separate Index pages for each image seems silly. Is there a procedure for re-associating an image with its counterparts when it has already been uploaded/imported? —Tosca-the-engineer 21:20, 13 December 2025 (UTC)

Though the example that you link is not in English - so I assume should not be here. I wonder how many of the others are like that. -- Beardo (talk) 22:58, 13 December 2025 (UTC)

Most of the scores don't have lyrics—the example I linked to seems to have been a fundraiser for the war effort (it's from 1916). I was actually thinking of translating it myself. —Tosca-the-engineer 23:12, 13 December 2025 (UTC)

Remember that it will need to be scan-backed on French wikisource before you can do a translation here. -- Beardo (talk) 02:04, 14 December 2025 (UTC)

Maybe the best way to do it (on Commons probably?) is to create a Category for these images? Extremely not sure tho ⸺ googoo0202 (he/him) (talk, contrib) 09:53, 17 December 2025 (UTC)

The images are all categorised appropriately on Commons (grouped into subcats of c:Category:Compositions by Charles-Marie Widor), so I think the Commons categorisation is entirely separate from the Wikisource categorisation. There's approximately 7000 jpegs, so adding them to a category one at a time would be super inefficient, but maybe if I install Cat-a-lot as a user gadget… —Tosca-the-engineer 10:06, 18 December 2025 (UTC)

Index:MU KPB 018 Comus by John Milton - Illustrated by Arthur Rackham.pdf & Index:MU KPB 018 Comus by John Miltow - Illustrated by Arthur Rackham.pdf

For some reason (possibly a renamed file), The Page: and Index: have become adrift. Can someone please site down and rebuild this so there is ONE file, Index: and pageset please? ShakespeareFan00 (talk) 11:52, 14 December 2025 (UTC)

• The file was moved from Miltow to Milton, but all of the work here (index, proofreading, and transclusion) was done on Miltow. Our Milton index needs to be deleted, and then everything needs to be migrated to Milton. TE(æ)A,ea. (talk) 15:51, 14 December 2025 (UTC)

Automatic removal of hyphens doesn't work in ws-export

When a page ends with a hyphen, the beginning and the end of the word will normally be combined automatically when transcluding the text into the main namespace. Example: "care-" from Page:Weird Tales Volume 6 Number 2 (1925-08).djvu/24 is combined into "carefully" in Weird Tales/Volume 6/Issue 2/The Oldest Story in the World.

However, this seems to have stopped working with the Wikisource export tool: When exporting the transcluded text, the resulting epub will display "care- fully" instead of "carefully". It may have something to do with the Parsoid library, as I can reproduce the error by adding "?useparsoid=1" to the url: https://en.wikisource.org/wiki/Weird_Tales/Volume_6/Issue_2/The_Oldest_Story_in_the_World?useparsoid=1.

The people at French Wikisource have also reported recent problems with the Wikisource export tool, but nobody seems to have mentioned the problem with hyphens.

• T408915
• T411303

Kåre-Olav (talk) 17:34, 14 December 2025 (UTC)

Thanks for the report. That does look like a parsoid regression. I've opened T412639. — Alien  3
3 3 19:05, 14 December 2025 (UTC)

The Poems of Henry Kendall/Araluen (1)

Can someone very SLOWLY explain WHERE the missing tag actually is, Because I am not seeing one, other than the linter apparently hallucinating one when it does the automated header.. It gets EXCEEDINGLY tiresome trying to seperate real issues with transcribed content, from 'phantom' issues, that arise because no-one seemingly has the time to actively ensure core functionality actually works PROPERLY!! ShakespeareFan00 (talk) 01:05, 15 December 2025 (UTC)

Tech News: 2025-51

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Updates for editors

• View all 18 community-submitted tasks that were resolved last week. For example, one of the fixes addressed an issue for temporary accounts adding an external URL, which triggered an hCaptcha request in more cases than intended, and did not display the required popup on the first attempt to publish the edit. [6]

Updates for technical contributors

• To improve database and site performance, external links to Wikimedia projects will no longer be stored in the database. This means they will not be searchable in Special:LinkSearch, will not be checked by the Spam Blacklist or AbuseFilter as new links, and will not be in the externallinks table on database replicas. In the future this may be extended to other highly-linked trusted websites on a per-wiki basis, such as Creative Commons links on Wikimedia Commons. [7]
• Detailed code updates later this week: MediaWiki

Tech news prepared by Tech News writers and posted by bot • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

MediaWiki message delivery 19:03, 15 December 2025 (UTC)

St._Nicholas/Volume_32/Number_3/Nature_and_Science

At the risk of looking like a complete idiot, this is blanaced up (tagwise) according to the original transcriber. However it's still showing up as having unblanced tags. My theory is that what is confusing the parsing, is that it's attempting to build using 2 pages tags, rather than Page directly. This means that an intermediate layer of HTML is being inserted, which is where the confusion arises...

Perhaps a third party can take a look from a technical perspective, as it was my understanding that you could not do the sort of jigsaw contruction intended? —unsigned comment by ShakespeareFan00 (talk) 15:43, 16 December 2025 (UTC).

Index:Abraham Lincoln - a history - the full and authorized record of his private life and public career (IA abrahamlincoln2116nico).pdf

After creating the index for this file from Commons, I saw that the rendered PDF was of poor quality and difficult for the editor to OCR. The DJVU from IA can be imported to Commons: would be the right move, seeing as it could be a redundant copy? If this was alright, the PDF index data can be moved to the DJVU index. A prior search for precedence informed my idea, at Wikisource:Scriptorium/Help/Archives/2024#Scan resolution (question for the technical people). Thanks. Overthrows (talk) 16:40, 17 December 2025 (UTC)

Categories

Can someone point me to what Categories to add, and what to avoid written out. I want to add the link to it in my user page for quick reference. RAN (talk) 01:22, 18 December 2025 (UTC)

Request for more opinions

I would like to draw attention to #Proposal: Modify the new WS:CSD criterion regarding non-English works, and to invite further opinions there. Thanks! -- Jan Kameníček (talk) 00:14, 21 December 2025 (UTC)

1769 KJV Bible—New Index and Transclusion

Now that I've gotten help in fulfilling my Scan Lab request for a new 1769 KJV Bible Index, I've been thinking that [[Bible (King James)]] is too generic of a page for the transclusion of one single edition of the KJV. I know I've already done most of my transclusion work on that page, but . . . I'm dealing with that later. Anyway, I'd like some discussion on what the best naming scheme could be for individual Bible editions, since the standard doesn't seem to exist yet. The naming should ideally be accurate, descriptive, and future-proof. A title like The Holy Bible is accurate (like, that is the actual title, sans-subtitle), but neither descriptive nor future-proof.

My working idea thus far is "The Holy Bible (King James, 1769; Blayney)". This retains the accurate title, describes enough about the exact edition of the book (version, year; notable editor), and seems to be future-proof; I don't think this title will clash with any similar edition any time soon. If the inclusion of an editor is too odd, it could instead list the publisher, like "The Holy Bible (King James, 1769; Oxford)".

Which seems best? Any additional insight on the order of descriptions or which to include?

—SpikeShroom (talk • contribs) 01:12, 21 December 2025 (UTC)

• Were there multiple, different editions of the King James version of the Bible published in 1769? If not, I don’t see the need for further disambiguation. Similarly, I find it hard to imagine there would be many (if any) different editions of the same version published in the same year. In addition, I prefer Bible over The Holy Bible as the top-level title. TE(æ)A,ea. (talk) 01:23, 21 December 2025 (UTC)

  I do agree that I prefer Bible over The Holy Bible for generic pages for the sake of professionalism, but when it comes to the title of a specific edition, the scan does say The Holy Bible, so I don't see why it shouldn't be included. I'm sure there are Bibles out there just titled Bible or something, and I'd honor those just the same, based on what the scan shows.

  That being said, I'm inclined to agree that the last bit may be unnecessary. "The Holy Bible (King James, 1769)" seems to fit the bill. Or should the description be "(King James Version, 1769)"?

  —SpikeShroom (talk • contribs) 01:41, 21 December 2025 (UTC)

  Technically, it should really be called the "Authorised Version" (AV) rather than "King James Version" (KJV) as King James authorised it for use in the churches of England. However, KJV seems to be the vernacular title now for most editions outside of the UK. Beeswaxcandle (talk) 02:30, 22 December 2025 (UTC)

Tech News: 2025-52

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Updates for editors

• From January, edit filters can be set to automatically suppress their details such as rules and list of attempted edits and actions. This will help oversighters use edit filters to prevent doxxing or other suppressible material. [8]
• The next issue of Tech News will be sent out on 12 January 2026 because of the end of year holidays. Thank you to all of the translators, and people who submitted content or feedback, this year.
• View all 16 community-submitted tasks that were resolved last week. For example, the crash that occurred when tapping "First Steps" in the Wikipedia Android Year in Review has now been fixed, and the feature opens as expected. [9]

Updates for technical contributors

• Interface elements such as diffs and categories generated by MediaWiki used to have the attribute data-mw="interface" to distinguish from wiki content. The attribute has been replaced with data-mw-interface="", to avoid potential conflicts with other data-mw attributes, which are generated by Parsoid. [10]
• There is no new MediaWiki version this week or next week.

Meetings and events

• The Wikimedia Hackathon Northwestern Europe 2026 will take place on 13-14 March 2026 in Arnhem, the Netherlands. Applications just opened mid-December and will close in mid-January or earlier if capacity is reached. With space for approximately 100 participants, early application is encouraged.

Tech news prepared by Tech News writers and posted by bot • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

MediaWiki message delivery 21:45, 22 December 2025 (UTC)

Special:BrokenRedirects contains redirects to MetaWiki mirror user pages

I see a few items in Special:BrokenRedirects that seem valid to me. For example: https://en.wikisource.org/w/index.php?title=User:Jusjih&redirect=no linking to https://en.wikisource.org/wiki/User:TunnelESON (putting the URLs alone here so as not to ping them unnecessarily). What's going on here is that the user page it's redirecting to doesn't exist locally but mirrors from MetaWiki: "What you see on this page was copied from https://meta.wikimedia.org/wiki/User:TunnelESON." Do we think that those redirects to Meta-mirror user pages should be in Special:BrokenRedirects? SnowyCinema (talk) 18:34, 24 December 2025 (UTC)

Index:The Tribes of Burma

The underlying file at Commons has been deleted because it didn't have a licence. It's a 1910 work, the author died in 1948 and our transclusion says "This work is in the public domain worldwide because it was created by a public body of the United Kingdom with Crown Status and commercially published before 1975." -- Beardo (talk) 16:32, 25 December 2025 (UTC)

This work has been done as a set of png files, so it's only the Cover image that has been deleted at present. This doesn't appear to have been transcluded, so is of minor importance. However, it would be a good idea to check that there is a valid licence on the rest of the set of images to prevent them from going the same way. Beeswaxcandle (talk) 18:13, 25 December 2025 (UTC)

Template:Block indent

We have {{Block indent}}; could someone kindly make {{Block indent/s}} and {{Block indent/e}}, for use over page changes, please? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:29, 25 December 2025 (UTC)

Normally we use {{dent/s}} and {{dent/e}}, which perform the same function. --EncycloPetey (talk) 20:44, 25 December 2025 (UTC)

@Pigsonthewing Another option is using {{block left/s|style=margin-left:3em;}} and {{block left/e}}. Regards, TeysaKarlov (talk) 20:49, 25 December 2025 (UTC)

Working, thank you (both). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:47, 26 December 2025 (UTC)

Misplaced book on French WS

Hello all! We have on the French Wikisource a book that does not belong with us, as it is fully in English and the person who uploaded it long ago didn't embark on a translating project. We plan on deleting it per policy but wanted to give you a heads-up if anyone wants to transribe this delightful treatise Of the imagination, as a cause and as a cure of disorders of the body; exemplified by fictitious tractors, and epidemical convulsions. The corresponding file is on Commons but you may want to make use of the (admittedly few) already transcribed pages

Wikisourcely yours, Susuman77 (talk) 22:42, 29 December 2025 (UTC)

Hello! I think you can largely go on and delete it, as far as I can see. :) Cheers, — Alien  3
3 3 22:49, 29 December 2025 (UTC)

Noting for the record: the file on Commons is incomplete and missing about half the text anyway —Beleg Âlt ^BT (talk) 16:10, 30 December 2025 (UTC)

Un-validating pages

Index:The Story of the Life of Mackay of Uganda Told for Boys WDL9945.pdf says all the pages are validated, but they're just... obviously not. Basic stuff is often missing: lots of hyphenated words on line breaks haven't been put together, and nearly all the footnotes aren't set up as footnotes. I'm going through and fixing that now, but it would be easier for me if the finished vs unfisihed pages were marked. If I move a page back to "proofread" from "validated", that locks me out of validating it later on. Could someone else change pages 89–338 to "proofread" for me please? Eievie (talk) 06:29, 30 December 2025 (UTC)

@Eievie; pages 89 to 237 done. • M-le-mot-dit (talk) 16:03, 30 December 2025 (UTC)

Done • M-le-mot-dit (talk) 18:17, 30 December 2025 (UTC)

Double disambiguation pages

What is the community consensus regarding doubly-disambiguated disambiguation pages (i.e. disambiguation pages that are themselves disambiguated)? For example, Henry IV (Shakespeare) lists works by Shakespeare titled "Henry IV".

In my opinion, such pages are redundant; all works titled "Henry IV" by Shakespeare are already listed at both Henry IV and at Author:William Shakespeare (1564-1616). As such, Henry IV (Shakespeare) should redirect to one of those two locations (possibly with an anchor to the relevant section of the page).

We used to have a lot of double-disambiguation pages, but over the years they have all been merged into their parent pages. However, a few more have cropped up in the meantime. I've created a category for them here: Category:Double disambiguation pages. —Beleg Âlt ^BT (talk) 15:03, 30 December 2025 (UTC)

Probably shouldn't exist IMO, though I suspect a certain number of them should really be versions pages. — Alien  3
3 3 15:38, 30 December 2025 (UTC)

Of the pages we currently have, two are listings of editorials by a specific author, and this is improper use of a disambiguation page. Disambiguation should never be a list by the type of work. I agree that the other two seem superfluous --EncycloPetey (talk) 15:39, 30 December 2025 (UTC)

• These are useful, in certain situations, which is why I have created some of them in the past. Henry IV is a great example—needing to distinguish between the two plays of Henry IV is a far more likely use case than needing to distinguish between multiple different works which happen to have that title, two of which happen to be famous plays. The two lists of editorials are proper, because they are lists of works entitled “Editorial” by different authors. In addition, it would be unduly cumbersome to list several different short “editorials” together with all of the author’s other works, and it would be similarly cumbersome to have this list at “Editorial” or some other generic page. Thus, the middle ground, which avoids both of these cumbersomeness issues. TE(æ)A,ea. (talk) 02:44, 31 December 2025 (UTC)

  Would an anchored section not solve the issue just as easily, such as Henry IV#Shakespeare? (compare Sonnet#William Shakespeare) This removes the cumbersomeness without introducing redundancy, and also better follows our naming conventions for disambiguation pages (i.e. the title of the page is the title of the works being disambiguated). —Beleg Âlt ^BT (talk) 14:45, 31 December 2025 (UTC)

Having four entries doesn't seem a big issue one way or the other. Personally, I would like to see more examples to see whether they are useful or are becoming a problem and a hindrance. MarkLSteadman (talk) 02:53, 31 December 2025 (UTC)

There used to be a lot more. I don't think they are useful, but I don't think they are a problem or hindrance either. They're just redundant. —Beleg Âlt ^BT (talk) 14:38, 31 December 2025 (UTC)

I would say we should delete these, and move their contents to Henry IV. We want one disambiguation page per keyword as much as possible. SnowyCinema (talk) 04:18, 31 December 2025 (UTC)

The information is already also listed at Editorial and Henry IV and I don't think those pages are too cumbersome. I think it would be better to eliminate these pages as unnecessary - though to accept that such a page might be useful if the parent disambiguation page does become too cumbersome. -- Beardo (talk) 04:57, 31 December 2025 (UTC)

I think this specific one is useful. Where else should we link "Henry IV" from Page:Shakespearean Tragedy (1912).djvu/22 and similar? These cannot be linked to Henry IV. Anchor does not solve it, as the readers would be taken to a page full of other irrelevant works which makes it confusing for them. --Jan Kameníček (talk) 16:12, 31 December 2025 (UTC)

A different example. The version page Austria and Europe (The Bohemian Review, 1918) was created per Wikisource:Scriptorium/Archives/2024-08#Serialized works in periodicals (voting). As such it is also categorized in Category:Serialized works. Thus it cannot be included in Austria and Europe, as it 1) would be against the accepted policy and 2) could not be properly categorized. --Jan Kameníček (talk) 19:35, 31 December 2025 (UTC)

I don't understand—Austria and Europe (The Bohemian Review, 1918) is an edition, not a disambiguation page, and it is included among the editions listed at Austria and Europe —Beleg Tâl (talk) 01:06, 1 January 2026 (UTC)

True, sorry for the confusion. --Jan Kameníček (talk) 01:11, 1 January 2026 (UTC)

I personally don't have a problem with disambiguating something like Sonnet as it grows and if there starts to be large numbers of 100s of Sonnets by Poet X, moving them to there own subpage or breaking it up into subpages by author last initial or something. MarkLSteadman (talk) 20:53, 4 January 2026 (UTC)

User:333Bot and preventing tagging for deletion nominations

Noting here for general awareness (I've only now belatedly documented it at the bot's userpage, though it's been possible for a while): When you need for some reason or other to make the bot not tag a page listed for deletion, add <!--notag--> to the page in question (avoids having to revert again and again). — Alien  3
3 3 20:26, 1 January 2026 (UTC)

Is there a way to decrease the space between the author's name and his title?

Is there a way to decrease the space between the author's name and his title to more closely match the original? See: The Philadelphia Inquirer/1974/6 in Family on Trial in Trooper Battle. RAN (talk) 20:46, 1 January 2026 (UTC)

I tried something. As a general principle, don't use multiple {{c}}s consecutively. Then you can for instance use a {{br}} instead of a paragraph break. — Alien  3
3 3 20:59, 1 January 2026 (UTC)

• Thanks! Just what was needed. --RAN (talk) 22:31, 1 January 2026 (UTC)

Should I remove the "sic" where there is an error, or should the explanation be restored?

See: "Edward Schneider [sic]" at New York Daily News/1940/12/24/Cheated Death In Air Battles, Dies In Crash. Should we remove the "sic" pointing out an error, or should the explanation of the error be restored? Having an error pointed out and not explaining the error is confusing. Which is best practices? The person who removed it wrote: "the information is partly included in the header and author's page, and partly belongs to Wikipedia, not here". I am not sure that reading a Wikipedia biography will help people work out what the error is in the article. We shouldn't expect people to read a biography to figure out why a name or a date is incorrect in a newspaper article. I always assumed the "notes" section, was for … notes. RAN (talk) 21:00, 1 January 2026 (UTC)

Our text should reflect what the original said. If you want to point out an error, you can use {{SIC}} with the correct information. -- Beardo (talk) 21:05, 1 January 2026 (UTC)

• You are stating what I already know. The text already "reflects what the original said", I asked about the "notes=" portion of the header. I assumed it was for adding notes about the text. If it isn't for housing notes, perhaps change the name, or eliminate it. This isn't a matter of typed word versus presumed word, it is more complicated. --RAN (talk) 22:28, 1 January 2026 (UTC)

  Did the original text included [sic] ? -- Beardo (talk) 22:34, 1 January 2026 (UTC)

The documentation for the {{Header}} template states that the Notes field is for "notes to explain the work, to add context, or to impart concise information that adds value to the reader; for example, use of {{listen}}". Thus, it's not for explanations about misspellings in the text itself. Beeswaxcandle (talk) 18:19, 2 January 2026 (UTC)

• This is not a misspelling, it is an error of fact. My inquiry was about the removal of the note explaining an error of fact in the article. The note read: His legal name was "Eddie Schneider" but some sources incorrectly formalized it to "Edward Schneider". He wrote in Look Out, Lindbergh - Here I Come in 1931: "because people are always asking me, my name is really Eddie: I was christened that way. It isn't very dressy, but it serves the purpose." It certainly "add[ed] context [and] impart[ed] concise information that add[ed] value to the reader." Shouldn't it be "value for the reader" and not "value to the reader". The reader is not now more valuable, the text is now more valuable. I think it was removed for esthetic reasons, not because it did not add value for the reader. --RAN (talk) 12:32, 4 January 2026 (UTC)

The original text does not have any "[sic]" note included, so neither should our transcription. If you feel that the error should be pointed out, you can use the template {{SIC|Edward|Eddie}}. However, there is a little problem that it is also linked, which causes that the dots of the tooltip under the text are not displayed. This can be worked around with placing <templatestyles src="Template:Tooltip/styles.css" /> somewhere at the top of the text. --Jan Kameníček (talk) 21:30, 2 January 2026 (UTC)

• I temporarily restored the explanation of the error of fact in the notes section and also added it to the sic template. Which is less intrusive? I don't expect readers to research why the article used the wrong name, when a concise explanation can be included. We only need one version, the one in the sic template does not allow a link to his biography. I want to thank Jan Kameníček for the amazing formatting of the entry. --RAN (talk) 12:47, 4 January 2026 (UTC)

  I think the SIC template is enough, we usually do not use the notes in the header for explanation of such nuances. One of the reasons is that unlike Wikipedia our small community does not have enough volunteers who could keep checking accuracy and sources of such information. Instead, we provide links to Wikipedia articles in "sister projects" section of the headers, where anything can be explained in detail and where the community has well established fact checking processes. --Jan Kameníček (talk) 13:01, 4 January 2026 (UTC)

• We only need one, but the section is for "notes to explain the work, to add context, or to impart concise information that adds value to the reader", so maybe its time to rewrite the definition of notes=, or eliminate the notes= section. --RAN (talk) 14:10, 4 January 2026 (UTC)

  The notes section is for information adding context for the whole text, not individual bits of information contained in the text. It can bear publication information or details about the print history that can't be otherwise easily encoded in the header, such as indicating a volume is the 1927 printing of the 1922 edition. Or that the text is the US / UK edition of a book published in both countries. Or that a particular poem was previously published in another location, or that the following edition of that poem contains a proem added by the editor. So the notes section serves a very useful function for some publications. --EncycloPetey (talk) 14:27, 4 January 2026 (UTC)

 Comment I just want to add that a good place for this type of commentary would be the work's talk page —Beleg Tâl (talk) 19:33, 4 January 2026 (UTC)

Strange things in the Wikisource:Scriptorium/Help/Archives/2025

Something strange is happening at Wikisource:Scriptorium/Help/Archives/2025, where some archived topics seem to have disappeared. If you look at the previous revisions, you may find there e.g. the topic "Deceased Wikisource contributors page?". Then, some new topics were added by the Wikisource-bot, and the following revision does not contain that topic anymore. The strange thing is that the diff does not show any deletions from the page. Any explanation? -- Jan Kameníček (talk) 00:23, 3 January 2026 (UTC)

• Jan Kameníček: I have fixed it. Arcorann’s comment in the “footnotes, endnotes, and refs in general” section had <ref> without <nowiki>. The next </ref> happened to be in ShakespeareFan00’s comment in the just-now-archived “Page:A voyage to Abyssinia (Salt).djvu/359 &Page:A voyage to Abyssinia (Salt).djvu/360” section. By archiving this section, it closes the <ref>, which incidentally hides a hundred or so other sections. TE(æ)A,ea. (talk) 00:39, 3 January 2026 (UTC)

  Oh, that is funny :-) Thanks for finding the problem out and fixing it! --Jan Kameníček (talk) 00:52, 3 January 2026 (UTC)

Should Versions pages have license tags?

Should Versions pages have license tags, such as {{PD-US}}, or should that only be done in the actual transcribed versions they list? Example: The Stolen Body SnowyCinema (talk) 04:37, 4 January 2026 (UTC)

• They should not; license tags only make sense at the level of a published volume (as they are currently implemented, anyway). TE(æ)A,ea. (talk) 04:54, 4 January 2026 (UTC)

  Agree. Each listed volume may include additional material that would not fall under a license tag applied at the version page level, which would be confusing. This concern applies even more strongly to translation pages. --Jan Kameníček (talk) 13:44, 4 January 2026 (UTC)

They should, since they represent the intellectual work, and outside the US, it is the intellectual work that enters public domain based on the date of the author's death. Only in US law is the copyright date tied to date of publication and thus a specific edition. It also permits a shortcut with translations, allowing us in a central location to say that the original is in public domain, rather than restating that fact on each translation. --EncycloPetey (talk) 14:21, 4 January 2026 (UTC)

This can be confusing to readers, who may expect that the licence given at the version page applies to all the volumes listed there. But that often is not true, because the volumes besides the work itself may include illustrations by different authors, prefaces, introductions, notes, footnotes etc. The volumes can also have different editors. As a result each of the volume often falls under a different copyright license not only in the US, but outside the US too. For this reason it makes much more sense to give the copyright tags only with the actual transcribed versions. --Jan Kameníček (talk) 19:40, 4 January 2026 (UTC)

If someone if confused, then they're not reading the full tag on the versions page: "This work was published before January 1, 1931, and is in the public domain worldwide because the author died at least 100 years ago. Translations or editions published later may be copyrighted." Our license on the versions pages makes that point clear.

And those individual license tags on particular volumes pertain to the additional material in the specific copy, such as illustrations, front matter, notes, and the like, not to the base work. The license on such editions are not for the work, but for the additions to the specific copy. The purpose of tagging the versions page is to give the license for the underlying work. The license on one of our editions does not provide that information. --EncycloPetey (talk) 19:52, 4 January 2026 (UTC)

Well, e. g. {{PD-US}} template and others do not say anything about later editions as you quoted. There are many different license tags. In some cases the license tag that would be given at a version would not apply to any of the versions listed. And that can be confusing for some readers no matter how precise the message would be. If each version has its own copyright tag, having one more at the version page is imo quite redundant. --Jan Kameníček (talk) 20:17, 4 January 2026 (UTC)

If our license templates are inadequate, then we should improve them. Inadequacy of our templates is not a rationale for making general policy decisions. We can always improve the templates. --EncycloPetey (talk) 21:27, 4 January 2026 (UTC)

Agree that license templates can be improved. However, it would not change my main point: it is not much use to put there a license tag that may not apply to the listed items anyway. --Jan Kameníček (talk) 21:32, 4 January 2026 (UTC)

That doesn't help itself when the versions page itself lists versions that are copyrighted, especially if they are heavily revised versions of the same work as opposed to just new material, or have different publication dates. If the first edition is published 101 years ago and a much revised and a much expanded edition is published 5 years later, and that is the version transcribed, which publication date to stick in "This work"? This is common with say textbooks where later editions can have many new chapters, e.g. Mashall's Principles between its first edition in 1890 to the ninth in 1960. MarkLSteadman (talk) 20:19, 4 January 2026 (UTC)

E.g. isting "This work was published before January 1, 1931, and is in the public domain worldwide because the author died at least 100 years ago. Translations or editions published later may be copyrighted" on a page that only lists say the 20th (ed. died 1964) and the 24th (ed. died 1960) of Gray's Anatomy. Note the 43rd edition is coming out this year. MarkLSteadman (talk) 20:46, 4 January 2026 (UTC)

We shouldn't be listing editions that are copyrighted, since copyrighted items are beyond scope, without tagging them using a clear line tag or template. --EncycloPetey (talk) 21:24, 4 January 2026 (UTC)

• Non-U.S. copyright is irrelevant to enWS, and should not be a concern whatsoever. We have a generic disclaimer to that effect. Your claim about to what copyright adheres is also incorrect; in other countries, just as in the United States, it adheres to individual editions of works. The different is just when the copyright expires, not its existence. Even looking to other countries’ copyright laws, different editions of the same work may have different copyright terms, such as where later editions have different (and later-deceased) illustrators or editors. In addition, the license tag on a versions page has no meaning; after all, a versions page has no text of the work in question, just a list of different copies of that text. There is nothing to license, so a license tag is pointless. Now, translations pages are different from versions pages (which was the original topic). However, those are even more inappropriate; after all, we do not host the original-language version, and thus have nothing to license. As for your later statement, “The license on such editions are not for the work, but for the additions to the specific copy”, this is also incorrect. For any individual work, the license tag covers the copyright of the entirety of that work, which includes the residual copyright from an earlier edition and the new copyright of that edition. We cannot host a later, public-domain edition of a work if an earlier edition, the basis for that later edition, is still copyrighted. You also have previously stated that listing copyrighted editions is acceptable. TE(æ)A,ea. (talk) 01:14, 6 January 2026 (UTC)

"Only in US law is the copyright date tied to date of publication". This is false as even a basic look at https://cdn.nationalarchives.gov.uk/documents/information-management/crown-copyright-flowchart.pdf and https://cdn.nationalarchives.gov.uk/documents/information-management/non-crown-copyright-flowchart.pdf shows. MarkLSteadman (talk) 02:51, 6 January 2026 (UTC)

Yet all of our license templates state that "This work is in public domain ..." without any mention of editions. --EncycloPetey (talk) 18:35, 6 January 2026 (UTC)

This is caused by the fact that we sometimes use the expression "work" generally, meaning all its versions, and sometimes in a narrower sense, more or less synonymous to "edition". The text of the templates used under the editions speaks about "work" and "longest living autor of the work", but actually refers not only to the expired copyright of the author of the work, but also of the illustrator, editor, author of the preface/introduction/foreword, author of the notes..., all of which are connected with the specific edition only and not with the work in the general sense. And if the same author publishes a new version of the same "work", with some changes of the text, the new and different text of the template will still speak about "work" again, although it would in fact refer to the new edition only. --Jan Kameníček (talk) 19:29, 6 January 2026 (UTC)

Also an artifact of our history of having unsourced / second-source text. We see this where we replace unsourced / secondary-source "works" in place with proper defined "editions." Without proper edition information we don't things like the exact publication date of an edition / reprint so it tends to get blurred. MarkLSteadman (talk) 21:43, 6 January 2026 (UTC)

Given the point made several times in this discussion that the licensing is confusing, should be improve the template wording? --EncycloPetey (talk) 22:00, 6 January 2026 (UTC)

If so, it should probably be done in a new discussion. --Jan Kameníček (talk) 22:19, 6 January 2026 (UTC)

If that's the case, then I would say any decision in this discussion should be pended. Our license wording currently says ""work", but we're arguing that "work" pages should not have the license. The two issues are very thoroughly entwined. --EncycloPetey (talk) 22:42, 6 January 2026 (UTC)

I do not think so, if majority discussing contributors think that copyright templates should not be used on version pages, they simply should not. So unless the request for pending the decision is supported by more contributors, I do not think it is really necessary. I personally do not see the wording of the templates so problematic that it really needs any urgent adjustment, and at this moment it is not certain that such adjustments will take place at all. Or the adjustments might take a long time, but they are not an obstacle for stopping using the templates at version pages if the community decides so. --Jan Kameníček (talk) 23:04, 6 January 2026 (UTC)

From WS:WD, "Work items should hold information true to all editions, versions, manifestations, or implementations of the work." Very clearly copyright info does not belong on WD for work items if that is your question. This is clearly documented on that page. We could cleanup our terminology "Works on Wikidata are split into two data items: Work items: For the work in general. Edition items: For each specific version or edition." but also this has been true for ages...

MarkLSteadman (talk) 23:39, 6 January 2026 (UTC)

I'm not sure how that page is particularly relevant to the discussion here, as (a) it has no counterpart at Wikidata, and (b) it is neither a policy nor community-accepted guideline for Wikisource. Even if Wikisource adopted the page as a policy or guideline, the page concerns activity on Wikidata, and so it would carry no weight there, as Wikidata is a separate project from Wikisource. I personally agree with what it says there, but the context is what belongs on Wikidata, not what belongs on Wikisource.

I also note that, among the "invariant" things to be listed on the work data item are the title and author, but even those values can change from edition to edition, not only for textbooks, but also for translations, and for UK/US editions. --EncycloPetey (talk) 00:17, 7 January 2026 (UTC)

If it is not relevant to discussions about what we think nominate it for deletion or propose placing a disclaimer on it that it is not relevant to anything, e.g. some innocent editor following it and getting slapped for doing ti wrong because unconveyed reasons. My point is that it isn't just my coming out of thin area (just like a court might site books about topic X, saying it's not law, not relevant isn't the point). It provides evidence of what we think belongs on the page. MarkLSteadman (talk) 02:54, 7 January 2026 (UTC)

And we have a specific policy statement on resolving exactly this title question anyways on the versions page policy. I am happy to propose adding similar language for author or leaving blank (like anonymous works). Problem solved. Or are we going to start complaining that some untitled works don't have titles so we shouldn't have titles for pages? 03:01, 7 January 2026 (UTC) MarkLSteadman (talk) 03:01, 7 January 2026 (UTC)

My general point is that we've drifted into discussing something not relevant to the current thread using a page that applies only practice on another site. Note that we have no versions policy; Wikisource:Versions has never been made policy. -EncycloPetey (talk) 12:11, 7 January 2026 (UTC)

You've missed the context and jumped to deletion as an option. The section you've quoted opens with "Works on Wikidata are split into two data items..." so it's clear from context that the section is discussing practices on Wikidata, not on Wikisource. We don't delete pages simply because you've applied them out of context. If someone quoted a poem here out of context, that doesn't mean we ought to delete it. Instead, we should acknowledge that the quote was taken out of context, not jump to deletion.

And in this instance our page is at odds with recommendations on Wikidata itself. See d:Wikidata:Property proposal/copyright status. When the copyright status property on Wikidata was proposed and set up, it was explicitly applied to creative works, literary works, etc., and not to editions. The current type constraints listed for d:Property:P6216 reinforce that the property is intended for works and makes no mention of editions. --EncycloPetey (talk) 12:16, 7 January 2026 (UTC)

I don't really see that versions pages need licenses. The line for where editing a work gains copyright is unclear. US law requiring a copyright notice forced an editor to at least declare that there was new copyright, if they didn't have to be specific. An abridgment certainly would, and possibly, if the underlying source text was complex enough, forming a new distinct text with creative choices for inclusions and omissions. (The Frankenstein that underlies PG's edition apparently merges both editions, which would probably qualify.) As others have said, nonfiction books that have run through a number of editions get complex enough that one license doesn't really help. The Elements of Style has a 1935 edition by a new author, a 1956 edition by a new author (which I don't think is derivative of the 1935 edition), and John Woldemar Cowan wrote a 2006 revision of the 1918 original, released under the CC-BY-SA. They're versions of the same work, but come under different licenses.--Prosfilaes (talk) 00:40, 6 January 2026 (UTC)

Action to take

Based on the above discussion, I'm going to start this "action to take" thread on the assumption that no major new issues will be rasised, though acknowledging that the above discussion has been over only three days, and so new ideas might enter that discussion.

Assuming therefore that we intend to not have license tags on versions pages, and considering that user confusion was raised as an issue: Do we want to (a) simply remove all license templates from versions pages, and leave no information in its place; or (b) develop a new template for versions pages (not a license) that advises roughly what we've said above: that license information is placed on the pages of individual editions rather than the general page, since the text and images, authors and editors, can vary from edition to edition, affecting the copyright status. --EncycloPetey (talk) 12:32, 7 January 2026 (UTC)

Brace2 display issue

Page: namespace

Mainspace

I am getting different display of {{brace2}} apparently depending upon namespace. At right are images showing the same section of text as displayed in the Page namespace and Mainspace. In Mainspace, the brace is displayed at roughly half the height, even though it is the same content.

The locations are Page:George Bernard Shaw - The Apple Cart.pdf/81 in Page; and The Apple Cart/Act I#45 in Main. --EncycloPetey (talk) 14:16, 4 January 2026 (UTC)

@EncycloPetey: I remark that the problem occurs only with layouts 2 and 4, not 1 and 3. • M-le-mot-dit (talk) 15:22, 4 January 2026 (UTC)

But not consistently so. The same contextual syntax (and layout) is used on Page:Coriolanus (1924) Yale.djvu/12 and Coriolanus (1924) Yale/Text, but on those pages the display of the brace is consistent. --EncycloPetey (talk) 15:28, 4 January 2026 (UTC)

@EncycloPetey: In fact, the brace size depends on the width of the cell. That's why is seems to depend on the layout. You may see with layout 1 how the brace height changes with a narrower window. • M-le-mot-dit (talk) 16:03, 4 January 2026 (UTC)

That is quite unpleasant, and it must be some new bug, because I did not experience it in the past. Is there any way to work around it? --Jan Kameníček (talk) 20:00, 4 January 2026 (UTC)

@Jan.Kamenicek I do not believe this a new bug, as I have experienced it for a while (unless your definition of new was the past year or two at least). For whatever reason, layout or otherwise, if the cell containing the brace2 does not have sufficient width to display the full height brace, then the entire brace shrinks. As in @M-le-mot-dit's edit, one workaround is just to fix the cell width using {{ts|width:Xem}}. Regards, TeysaKarlov (talk) 22:22, 4 January 2026 (UTC)

Game books

Consider the Consequences!, whose transcription I have just completed, is a "game book" (a kind of "chose your own adventure"—in fact, reputedly the first ever).

I have put it into Category:Games, for now, but do we have one that is more precise? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:38, 4 January 2026 (UTC)

Thank you. I don't know the answer to your question, but I have also added it to Portal:Games. By the way, I think for your second link, you meant w:Consider the Consequences!. -- Beardo (talk) 01:16, 5 January 2026 (UTC)

We don't appear to have one that's more precise at the moment, but if we get more books of this type we can always create one later! —Beleg Âlt ^BT (talk) 14:35, 5 January 2026 (UTC)

rh and rvh

If a page has been created using the template rh, is there any advantage in changing it to use rvh ? -- Beardo (talk) 02:34, 5 January 2026 (UTC)

Not as far as I know. Just different ways of working. — Alien  3
3 3 11:48, 5 January 2026 (UTC)

The difference is that rvh is dynamically making a display decision based on the page number (in the scan?) so I would think that in a proofread page, we'd rather keep the rh if it already exists. We sometimes need to modify scans to insert or remove pages, and rvh in such a situation could lead to reversing the header format across all affected pages. --EncycloPetey (talk) 12:25, 7 January 2026 (UTC)

Mediawiki request

Could an administrator please import the MediaWiki:Cite link label group-lower-alpha from Mediawiki? This will allow us to create ref-tag references without numbers – specifically, <ref group="lower-alpha">Text</ref> would produce ^[a]. This would be helpful. Thanks, Cremastra (talk) 01:37, 7 January 2026 (UTC)

Per policy (Help:Footnotes and endnotes) our style here is to use numbered footnotes, regardless of what the style the original book used. This is why this label group hasn't been imported. Beeswaxcandle (talk) 09:12, 7 January 2026 (UTC)