Wikisource:Proposed deletions

WS:PD redirects here. For help with public domain materials, see Help:Public domain.

Index:The trail of the golden horn.djvu

This specific index is one of many such indexes; I nominate it as an example, but should the rationale be found sound, I will endeavor to make a list of all such indexes.

This index (and many others) were created by now-absent User:Languageseeker. My main concern is that the pages of these indexes have been added via match-and-split from some source, likely Project Gutenberg, which does not have a defined original copy. Because of this absence of real source, and the similarity of the text to the actual text of any given scanned copy, proofreading efforts would likely have to either not check the text against the original source or scrap the existing text entirely to ensure accuracy to the original on Wikisource. In light of this, I think the easiest approach is to delete the indexes and all pages thereunder; if there is organic desire to scan them at some point in the future, the indexes may be re-created, but I do not see a reason to keep the indexes as they stand. TE(æ)A,ea. (talk) 19:12, 9 December 2023 (UTC)

•  Comment Hmm. I don't see the Index: pages as problematic. But the "Not Proofread" Page: pages that were, as you say, created by Match & Split from a secondary transcription (mostly Gutenberg, but also other sources), I do consider problematic. We don't permit secondary transcriptions added directly to mainspace, so to permit them in Page: makes no sense. And in addition to the problems these create for Proofreading that TE(æ)A,ea. outlines, it is also an issue that many contributors are reluctant to work on Index:es with a lot of extant-but-not-Proofread (i.e. "Red") pages.
  We have around a million (IIRC; it may be half a mill.) of these that were bot-created with essentially raw OCR (the contributor vehemently denies they are "raw OCR", so I assume some fixes were applied, but the quality is very definitely not Proofread). Languageseeker's imports are of much higher quality, but are still problematic. I think we should get rid of both these classes of Page: pages. In fact, I think we should prohibit Not Proofread pages from being transcluded to mainspace (except as a temporary measure, and possibly some other common sense exceptions). --Xover (talk) 20:24, 9 December 2023 (UTC)
  • Xover: Assuming the status of the works to be equal, I would actually consider Languageseeker’s page creations to be worse, because, while it would look better as transcluded, it reduces the overall quality of the transcription. My main problem with the other user’s not-proofread page creations was that he focused a lot on indexes of very technical works, but provided no proofread baseline on which other editors could continue work—that was my main objection at the time, as it is easier to come on and off of work where there is an established style (for a complicated work) as opposed to starting a project and creating those standards yourself. As to the Page:/Index: issue, I ask for index deletion as well because these indexes were created only as a basis for the faulty text import, and I don’t want that to overlook any future transcription of those works. Again, I have no problem to work (or re-creation), I just think that these indexes (which are clearly abandoned, and were faulty ab origine) should be deleted. As for transclusion of not-proofread pages, I don’t think that the practice is so widespread that a policy needs to implemented (from my experience, at least); the issue is best dealt with on a case-by-case basis, or rather an user-by-user basis (as users can have different ways of turning raw OCR into not-proofread text, then following transclusion and finally proofread status). But of course, that (and the other user’s works, the indexes for which I think should probably be deleted) are a discussion for another time. (I will probably have more spare time starting soon, so I might start a discussion about the other user’s works after this discussion concludes.) TE(æ)A,ea. (talk) 02:28, 10 December 2023 (UTC)

    I'm not understanding what fault there is in the Index page. If the Page: pages had not been created, what problem would exist in the Index: page? --EncycloPetey (talk) 02:53, 10 December 2023 (UTC)

    • EncycloPetey: This isn’t a case where the index page’s existence is inherently bad; but the pages poison the index, in terms of future (potential) proofreading efforts and in terms of abandonment. TE(æ)A,ea. (talk) 03:07, 10 December 2023 (UTC)

      @TE(æ)A,ea.: Just to be clear, if the outcome here is to delete all the "Not Proofread" Page: pages, would you still consider the Index: pages bad (should be deleted)? So far that seems to be the most controversial part of this discussion, and the part that is a clear departure from established practice. Xover (talk) 07:40, 20 December 2023 (UTC)

      • Xover: Yes, I think those are also bad. They were created en masse for the purpose of adding this poor match-and-split text, and there is no additional value in keeping around hundreds of unused indexes whose only purpose was to facilitate a project consensus (here) clearly indicates in unwise. The main objection on that ground is that indexes are difficult to make; but that is not really true, and in any case is not a real issue, as a new editor who wishes to edit (but not create an index) can simply ask for one to be created. Another problem with these indexes is that they are not connected with other information (like the Author:-pages) that would help new editors find them. Insofar as they exist like this, the only real connection these indexes have to the project at large is through Languageseeker, who is now no longer editing. I don’t think that every abandoned index is a nuisance, but I do believe that this (substantial) group of mass-created indexes is a problem. TE(æ)A,ea. (talk) 21:10, 20 December 2023 (UTC)

• I support deleting the individual pages of the index. As for the Index page itself, I am OK with both deleting it as abandoned or keeping it to wait for somebody to start the work anew. I also support getting rid of other similar secondary transcriptions. If a discussion on prohibiting transclusion of not-proofread pages into main NS is started somewhere, I will probably support it too. --Jan Kameníček (talk) 00:39, 10 December 2023 (UTC)

 Comment I've always felt uncomfortable with the tendency of some users to want to bulk-add a bunch of Index pages which have the pages correctly labelled, but are left indefinitely with no pages proofread in them. I feel like a "transcription project" (as Index pages are labelled in templates) implies an ongoing, or at least somewhat complete, ordeal, and adding index pages without proofreading anything is really just duplicating data from other places into Wikisource. Not to say there's absolutely no value in adding lots of index pages this way, but the value seems minimal. The fact that index pages mostly rely on duplicate data as it is is already an annoying redundancy on the site, and I think most of what happens on Index pages should just be dealt with in Wikidata, so I think the best place to bulk-add data about works is there, not by mass-creating empty Index pages. I know my comment here is kind of unrelated to the specific issue of the discussion (being, indexes with pages matched and splitted or something), but the same user (Languageseeker) has tended to do that as well. I am struggling to come up with any specific arguments or policies to support my position against those empty index pages... but it just seems unnecessary, seems like it will cause problems in the future, and on a positive note I do applaud Languageseeker's massive effort—it shows something great about their character as an editor—but unfortunately I think their effort should have been more focused on areas other than the creation of as many Index pages as possible. PseudoSkull (talk) 04:15, 10 December 2023 (UTC)

Bulk-adding anything is probably a bad idea on Wikisource, because so much of what we do here requires a human touch. That being said, so far as I know the Index: pages Languageseeker created were perfectly fine in themselves, including having correct pagelists etc. This step is often complicated for new contributors, so creating the Index: without Proofreading anything is not without merit. It's pointing at an already set up transcription project onsite vs. just (ext)linking to a scan at IA for some users. The latter is an insurmountable effort for quite a lot of contributors. We also have historically permitted things to sit indefinitely in our non-content namespaces if they are merely incomplete rather than actually wrong in some way.

That's not to say that all these Index: pages are necessarily golden, but imo those that are problematic (if any) should be dealt with individually. Xover (talk) 09:08, 10 December 2023 (UTC)

Oh, also, what we host on Wikidata vs. what's hosted locally in our Index: pages is a huge and complicated discussion (hmu if you want the outline). For the purposes of this discussion it, imo, makes the most sense to just view that as an entirely orthogonal issue. If and when (and how and why and...) we push some or all our Index: page contents somewhere other than our current solution, it'll deal with these Index:es as well as every other. Xover (talk) 07:33, 20 December 2023 (UTC)

 Comment I do not support creating them, but since they exist, I try to make good use of them. I usually proofread offline for convenience and when I add the text I check the diff. If anything differs, it is an extra check for me as I could be the one who made mistakes. So I would keep them.

BTW, nobody forbids to press the OCR button and restart. Mpaa (talk) 18:35, 10 December 2023 (UTC)

While that is true, my experience is that the kinds of errors introduced by a mystery text layer is insidious, and most editors are unaware of the issue, or fail to notice small problems such as UK/US spelling differences, changes to punctuation, minor word changed, etc. So, while a person could reset the text, what would alert them to the fact that they should, rather than working from the existing unproofed page?

H. G. Wells' First Men in the Moon is a prime example. A well-meaning editor matched-and-split the text into the scan. Two experienced editors crawled through making multiple corrections to validate the work, yet as recently as this past week we have had editors continue to find small mistakes throughout. Experience shows that match-and-split text is actually worse for Wikisource proofreading than the raw OCR because of these persistent text errors. --EncycloPetey (talk) 18:51, 10 December 2023 (UTC)

In my workflow, I start from OCR, then compare what I did with what is available. It is an independent reference which I use for quality check. The probability that I did the same error is low (and the error would be anyhow there). It is almost as if someone is validating my text (or vice-versa). For me it is definitely a help. I follow the same process when validating text. I do not look at what is there and then compare. Mpaa (talk) 19:21, 10 December 2023 (UTC)

Right. You do that, and I work similarly. But experience shows that the vast majority of contributors don't do that; they either don't touch the text due to the red pages, or they try to proofread off the extant text and leave behind subtle errors as EncycloPetey outlines. Xover (talk) 19:35, 10 December 2023 (UTC)

We could argue forever. I do not know what evidence you have to say that works started from match-and-split are worse than others. I doubt anyone has real numbers to say that. IMHO it all depends on the attitude of contributors. I have seen works reaching a Validated stage and being crappy all the same. If you want to be consistent, you should delete all pages in a NotProofread state and currently not worked on because I doubt a non-experienced user will look where the text is coming from when editing, from a match-and-split or whatever.

Also, then we should shutdown the match-and-split tool or letting only admins to run it, after being 100% sure that the version to split is the same as the version to scan.

I am not advocating it as a process, I am only saying that what is there is there and it could be useful to some. If the community will decide otherwise, fine, I can cope with that. Mpaa (talk) 20:32, 10 December 2023 (UTC)

I do not know what evidence you have to say that works started from match-and-split are worse than others. Anecdotal evidence only, certainly. But EncycloPetey gave a concrete example (H. G. Wells' First Men in the Moon), and both of us are asserting that we have seen this time and again: when the starting point is Match & Split text, the odds are high that the result will contain subtle errors in punctuation, US/UK spelling differences, words changed between editions, and so forth. All the things that do not jump out at you as "misspelled". Your experience may, obviously, differ, and it's certainly a valid point that we can end up with poor quality results for other reasons too.

Your argumentum ad absurdum arguments are also well taken, but nobody's arguing we go hog-wild and delete everything. Languageseeker, specifically, went on an import-spree from Gutenberg (and managed to piss off the Distributed Proofreaders in the process), snarfing in a whole bunch of texts in a short period of time. All of these are secondary transcriptions, and Languageseeker was never going to proofread these themselves (their idea was almost certainly to either transclude them as is, or to run them in the Monthly Challenge).

For these sorts of bulk actions that create an unmanageable workload to handle, I think deletion (return to the status quo ante) is a reasonable option. The same would go for the other user that bulk-imported something like 500k/1 mill. (I've got to go check that number) Page: pages of effectively uncorrected OCR. For anything else I'd be more hesitant, and certainly wouldn't want to take a position in aggregate. Those would be case-by-case stuff, but that really isn't an option for these bulk actions. Xover (talk) 07:17, 11 December 2023 (UTC)

 Comment I am agianst deleting the Index. Indexes are one of the most tedious work to do when starting a transcription. Having index pages prepared and checked against the scan will save a lot of work. Mpaa (talk) 21:46, 10 December 2023 (UTC)

•  Keep the Index, but  Delete the pages. None of the bot-created pages have the header, which is a pain to add after-the-fact unless you can run a bot. The fact that they were created by match-and-split, instead of proofreading the text layer is poor practice. --EncycloPetey (talk) 19:15, 20 December 2023 (UTC)

  There are many recently added "new texts" with no headers. Mpaa (talk) 22:00, 23 December 2023 (UTC)

  • What percent of editors want headers; and what percent do not care? Do you have data? --EncycloPetey (talk) 22:03, 23 December 2023 (UTC)

    No, I am only stating is not a good argument for deletion in my opinion, unless it is considered mandatory. Mpaa (talk) 22:22, 23 December 2023 (UTC)

    • It is a good argument if most potential editors want to include the headers, and are put off working on proofreading by the fact that pages were created without the headers in place. There are works I've chosen not to work on for this reason. --EncycloPetey (talk) 23:04, 23 December 2023 (UTC)

    I agree that on its own the lack of headers is not a good argument for deletion. But I read it here to be intended as one additional factor on the scales that added together favour deletion. Which I do think is a valid argument (one can disagree, of course). Xover (talk) 23:57, 23 December 2023 (UTC)

  • Mpaa: That is the result of the efforts of one user, who has declared headers superfluous. I was going to start another discussion on that topic after this one (only one big discussion at a time for me, please). I think that, for all editors who want headers (most of them), not having them (because of the match-and-split seen here) is bad. Also, in response to your other comments above about proofreading over existing text, I usually do that as well, but I prefer proofreading on my own, without needing to check against a base—that’s why I focus on proofreading, not validation. For that same reason, I avoid all-not-proofread indexes like those at issue here. TE(æ)A,ea. (talk) 23:22, 23 December 2023 (UTC)

    I was thinking the same about headers, it would be good to have a consistent approach about works, in all their parts/namespaces. Mpaa (talk) 09:47, 24 December 2023 (UTC)

 Comment in the future, if anyone feels blocked for the lack of headers, or wants to add headers, please make a bot request.Mpaa (talk) 09:47, 24 December 2023 (UTC)

 Comment I am proofreading this specific text. This discussion can be as reference for the other indexes, as TE(æ)A,ea. mentioned at the beginning of the discussion. BTW, a list would be useful, so I can fetch before a (possible) deletion. Mpaa (talk) 12:53, 2 January 2024 (UTC)

• I have removed your discussion-closure notice, EncycloPetey, because I created this discussion as a general issue, not tied to the specific index at hand. Mpaa, just to be clear, you would like a list of the indexes with match-and-split pages, or would you like a more general listing of that user’s indexes? TE(æ)A,ea. (talk) 01:09, 10 January 2024 (UTC)

  Per the description at the top of this page: "This page is for proposing deletion of specific articles on Wikisource in accordance with the deletion policy, and appealing previously-deleted works." This discussion was closed because the decision was to Keep Index:The trail of the golden horn.djvu now that it is proofread. Do you disagree with that decision? Because this page is not for discussing general issues. That needs to happen in the Scriptorium, not here. --EncycloPetey (talk) 02:47, 10 January 2024 (UTC)

  @TE(æ)A,ea. those with match-and-split pages that could be deleted. Mpaa (talk) 20:48, 10 January 2024 (UTC)

The Picture in the House (unknown)

Duplicative of Weird Tales/Volume 3/Issue 1/The Picture in the House, starting discussion to decide whether to remove or migrate the librivox recording. MarkLSteadman (talk) 06:43, 29 December 2023 (UTC)

Gah. Tough call.

The two texts are not the same. Both Weird Tales in 1924 and the 1937 reprint use … the antique and repellent wooden building which blinked with bleared windows from between two huge leafless oaks near the foot of a rocky hill, but the unsourced text uses … elms …. LibriVox for once actually gives a source, and in the case of File:LibriVox - picture in the house lovecraft sz.ogg that source is The Picture in the House (unknown) (modulo a page move after the fact here), and the audio narration does match (uses "elms"). The change to "elms" seems to be a later innovation, possibly applied by an editor as late as 1982 (Bloodcurdling Tales of Horror and the Macabre, the earliest use of "elms" there I could find right now), and the likely ultimate source of our text. The texts differ in other ways too, but up to this point the difference could be explained by transcription errors, lack of scan-backing and validation, etc.).

So… I don't think we can move the LibriVox file over to our new text (different edition). And because the nominated text is from an indeterminate edition and we have a scan-backed version of this work, we should  Delete The Picture in the House (unknown) too.

But it's really annoying that when LibriVox for once both gives the source text they have used for their reading and actually links back to us, we have to delete the page. I wish they'd coordinate more with us on issues like this so we could get the maximum benefit out of our respective volunteer efforts. Xover (talk) 08:38, 29 December 2023 (UTC)

I guess that the LibreVox versions dates to when this was the only version available. Can we put the LibreVox link on The Picture in the House ? -- Beardo (talk) 18:33, 29 December 2023 (UTC)

Hmm, no, I don't think so. We can't start amassing random multimedia versions of texts at the dab pages. Eventually we want spoken-word versions of our texts automatically linked from data on Wikidata, and that requires control over which specific edition the spoken-word version is from. Xover (talk) 10:49, 31 December 2023 (UTC)

•  Comment The earliest copy I found with "...elms..." is a Best Supernatural Stories of H.P. Lovecraft on Google books, p. 122, which the site dates to 1945. --EncycloPetey (talk) 20:29, 30 December 2023 (UTC)

weak  Delete - it would probably be better for us to just start from scratch, although I recognize its value as being linked to from LibriVox, so maybe it could just be redirected to the current scanned version instead of outright deleted. SnowyCinema (talk) 03:30, 8 March 2024 (UTC)

• But start from scratch using what? The issue is that our scan-backed copy has a different text from the LibriVox recording. The text of the nominated copy can be attested, but not (yet) from a volume dated before 1945. Ideally, we would find a PD volume with the current text. --EncycloPetey (talk) 04:35, 18 March 2024 (UTC)

Official Journal of the European Union, L 042I, 23 February 2022

Excerpt of just parts of the title page (a pseudo-toc) of an issue of the journal of record for the EU. Xover (talk) 11:29, 11 February 2024 (UTC)

Also Official Journal of the European Union, L 078, 17 March 2014 Xover (talk) 11:34, 11 February 2024 (UTC)

Also Official Journal of the European Union, L 087I, 15 March 2022 Xover (talk) 11:35, 11 February 2024 (UTC)

Also Official Journal of the European Union, L 110, 8 April 2022 Xover (talk) 11:36, 11 February 2024 (UTC)

Also Official Journal of the European Union, L 153, 3 June 2022 Xover (talk) 11:37, 11 February 2024 (UTC)

Also Official Journal of the European Union, L 066, 2 March 2022 Xover (talk) 11:39, 11 February 2024 (UTC)

Also Official Journal of the European Union, L 116, 13 April 2022 Xover (talk) 11:39, 11 February 2024 (UTC)

•  Keep This isn't an excerpt; it matches the Contents page of the on-line journal and links to the same items, which have also been transcribed. The format does not match as closely as it might, but it's not an excerpt. --EncycloPetey (talk) 04:52, 12 February 2024 (UTC)

  That's not the contents page of the online journal, it's the download page for the journal that happens to display the first page of the PDF (which is the title page, that also happens to list the contents). See here for the published form of this work. What we're hosting is a poorly-formatted de-coupled excerpt of the title page. It's also—regardless of sourcing—just a loose table of contents. Xover (talk) 07:09, 13 February 2024 (UTC)

  I don't understand. You're saying that it matches the contents of the journal, yet somehow it also doesn't? Yet, if I click on the individual items in the contents, I get the named items on a subpage. How is this different from what we do everywhere else on Wikisource? --EncycloPetey (talk) 16:35, 13 February 2024 (UTC)

  They are loose tables of contents extracted from the title pages of issues of a journal. They link horizontally (not to subpages) to extracted texts and function like navboxes, not tables of contents on the top level page of a work. That their formatting is arbitrary wikipedia-like just reinforces this.

  The linked texts should strictly speaking also be migrated to a scan of the actual journal, but since those are actual texts (and not a loose navigation aid) I'm more inclined to let them sit there until someone does the work to move them within the containing work and scan-backing them. Xover (talk) 08:35, 20 February 2024 (UTC)

  So, do I understand then that the articles should be consolidated as subpages, like a journal? In which case, these pages are necessary to have as the base page. Deleting them would disconnect all the component articles. It sounds more as though you're unhappy with the page formatting, rather than anything else. They are certainly not "excerpts", which was the basis for nominating them for deletion, and with that argument removed, there is no remaining basis for deletion. --EncycloPetey (talk) 19:41, 25 February 2024 (UTC)

Translation:La Serva Padrona

There is no scan supported original language work present on the appropriate Italian Wikisource, as required by Wikisource:Translations. -- Jan Kameníček (talk) 09:50, 28 March 2024 (UTC)

•  Comment There is an 1862 Italian copy of the libretto on IA, with just 26 pages and no music score to transcribe, if any is willing to transcribe it on it.WS. --EncycloPetey (talk) 17:28, 28 March 2024 (UTC)

  I've started the Italian transcription at it:Indice:La serva padrona - intermezzi due (IA laservapadronain00fede).pdf --EncycloPetey (talk) 20:20, 19 April 2024 (UTC)

  I've set up the English translation match at Index:La serva padrona - intermezzi due (IA laservapadronain00fede).pdf --EncycloPetey (talk) 21:26, 19 April 2024 (UTC)

•  Comment In trying to match the translation to the Italian text, I've run into large difficulties. The text lacks most stage directions, and the one version of the opera I've now watched does not match either our translation nor the Italian text. Correctly translating the text is therefore near impossible, since I have no context for many of the lines. They appear to depend upon the "business" on stage for understanding what is meant, who the line is spoken to, and there may be some significant differences between 19th-century Italian and modern Italian. There are certainly some culturally dependent context phrases, such as "chocolate" probably meaning "hot chocolate drink" and not simply a confection. I do not think I am likely to finish work on this; is anyone else willing and able to make an attempt? --EncycloPetey (talk) 22:07, 8 June 2024 (UTC)

Contracts Awarded by the CPA

Out of scope per WS:WWI as it's a mere listing of data devoid of any published context. Xover (talk) 12:53, 31 March 2024 (UTC)

 Keep if scan-backed to this PDF document. Since the PDF document is from 2004, a time when the WWW existed but wasn't nearly as universal to society as today, I find the thought that this wasn't printed and distributed absurdly unlikely. And the copyright license would be PD-text, since none of the text is complex enough for copyright, being a list of general facts. Also, this document is historically significant, since it involves the relationships between two federal governments during a quite turbulent war in that region. SnowyCinema (talk) 14:25, 31 March 2024 (UTC)

(And it should be renamed to "CPA-CA Register of Awards" to accurately reflect the document.) SnowyCinema (talk) 14:32, 31 March 2024 (UTC)

It's still just a list of data devoid of any context that might justify its inclusion (like if it were, e.g., the appendix to a report on something or other). Xover (talk) 19:51, 13 April 2024 (UTC)

Maybe I should write a user essay on this, since this is something I've had to justify in other discussions, so I can just link to that in the future.

I don't take the policy to mean we don't want compilations of data on principle, or else we'd be deleting works like the US copyright catalogs (which despite containing introductions, etc., the body is fundamentally just a list of data). The policy says the justification on the very page. What we're trying to avoid is, rather, "user-compiled and unverified" data, like Wikisource editors (not external publications) listing resources for a certain project. And if you personally disagree, that's fine, but that's how I read the sentiment of the policy. I think that whether something was published, or at least printed or collected by a reputable-enough source, should be considered fair game. I'm more interested in weeding out research that was compiled on the fly by individual newbie editors, than federal government official compilations.

But to be fair, even in my line of logic, this is sort of an iffy case, since the version of the document I gave gives absolutely no context besides "CPA-CA REGISTER OF AWARDS (1 JAN 04- 10 APRIL 04)" so it is difficult to verify the actual validity of the document's publication in 2004, but I would lean to keep this just because I think the likelihood is in the favor of the document being valid, and the data is on a notable subject. And if evidence comes to light that proves its validity beyond a shadow of a doubt, then certainly. SnowyCinema (talk) 00:03, 20 April 2024 (UTC)

Evidence of validity: The search metadata gives a date of April 11, 2004, and the parent URL is clearly an early 2000s web page just by the looks of it. My keep vote is sustained. SnowyCinema (talk) 00:16, 20 April 2024 (UTC)

The Athenaeum

This has been an empty page since it was created in 2015. --EncycloPetey (talk) 00:06, 25 May 2024 (UTC)

Normally I would suggest speedy deletion (no notable content or history). However, we do appear to have two articles from The Athenaeum that should be moved to subpages of that work: Folk-lore (extracted from The Athenaeum 1846-08-22) and Folk-lore (extracted from The Athenaeum 1876-08-29). —Beleg Tâl (talk) 01:07, 25 May 2024 (UTC)

I've done some checking. The Athenaeum is not a unique title. There is also a student paper from Acadian University by this name that has been published since the late 19th century; there was also a (now defunct?) publication from Yale by this title; and there is a well-known Brazilian novel with this as its English title. So at the very least, any hub page for the London literary publication would need to be placed under a disambiguated title. --EncycloPetey (talk) 18:18, 26 July 2024 (UTC)

Looks like we do have justification to keep the page and convert it into a base page for the periodical (though the subpage convention might need to be worked out -- the periodical doesn't have numbered volumes and but does number issues continuously). While we're here, do we have conventions on how to handle "æ" in work titles (since The Athenæum was always written as such)? Arcorann (talk) 04:09, 27 May 2024 (UTC)

We don't have a strict convention, there are benefits to both arguments (faithfulness vs accessibility). Ether way, redirects should be created so that both spellings direct you to the correct work. —Beleg Âlt ^BT (talk) 13:24, 27 May 2024 (UTC)

Bongbong Marcos' Second State of the Nation Address

This address has large sections with multiple paragraphs at a time not in English. It is a mixed language speech for a bilingual audience and should therefore reside on mul.WS. --EncycloPetey (talk) 15:36, 22 July 2024 (UTC)

 Support, agree that it belongs on mulWS —Beleg Tâl (talk) 16:39, 22 July 2024 (UTC)

 Keep Although there are some parts not in English, most of the text is in English and so for me it is OK to host it here. I think that we should try to avoid moving texts to mul if it is at least a bit justifiable for them to be hosted here, because mul: is not really a user friendly site. --Jan Kameníček (talk) 18:46, 26 July 2024 (UTC)

The challenges present on other Wiktionaries are not a reason for keeping a work here. --EncycloPetey (talk) 20:24, 27 July 2024 (UTC)

• Keep. This is an English-language text. In a formal transcription the Tagalog-language text can be proofread on the appropriate language version and trans-language section-included here, as is the case for other works with multiple languages. This logic would require all dictionaries on all versions to be deleted and move to old Wikisource, which is not useful. TE(æ)A,ea. (talk) 20:19, 27 July 2024 (UTC)

  Dictionaries are a straw-man argument, since the dictionaries have a target audience that speaks English. But the target for this speech was a bilingual audience, not an English-speaking one. --EncycloPetey (talk) 20:23, 27 July 2024 (UTC)

Kamoliddin Tohirjonovich Kacimbekov's statement

No source, no license, no indication of being in the public domain —Beleg Tâl (talk) 17:22, 7 August 2024 (UTC)

Found the source: [1] — Alien333 (what I did & why I did it wrong) 19:54, 7 August 2024 (UTC)

The text of the source does not match what we have. I am having trouble finding our opening passages in the link you posted. --EncycloPetey (talk) 19:58, 7 August 2024 (UTC)

(At least, a sentence matched). @EncycloPetey: Found it, the content that corresponds to our page starts in the middle in the page 44 of that pdf, though the delimiting of paragraphs seems to be made up. — Alien333 (what I did & why I did it wrong) 20:00, 7 August 2024 (UTC)

That means we have an extract. --EncycloPetey (talk) 00:39, 9 August 2024 (UTC)

• No, it appears that the PDF is a compilation of several different, thematically related documents. His statement (English’d) is one such separate document. TE(æ)A,ea. (talk) 00:53, 9 August 2024 (UTC)

  In which case we do not yet have a source. --EncycloPetey (talk) 00:55, 9 August 2024 (UTC)

  • No, that is the source; it’s just that the PDF contains multiple separate documents, like I said. It’s like the “Family Jewel” papers or the “Den of Espionage” documents. TE(æ)A,ea. (talk) 00:58, 9 August 2024 (UTC)

    Sorry, I meant to say that we do not have a source for it as an independently hosted work. To use the provided source, it would need to be moved into the containing work. --EncycloPetey (talk) 01:55, 9 August 2024 (UTC)

    Well these document collections are bit messy, they were originally independent documents / works but they are collected together for release, e.g. because someone filed a FOIA request for all documents related to person X. I don't think it is unreasonable if someone were to extract out the document. I wouldn't object if someone was like I went to an archive and grabbed document X out of Folder Y in Box Z but if someone requested a digital version of the file from the same archive they might just get the whole box from the archive scanned as a single file. Something like the "Family Jewels" is at least editorial collected, has a cover letter, etc., this is more like years 1870-1885 of this magazine are on microfiche roll XXV, we need to organize by microfiche roll. MarkLSteadman (talk) 11:17, 9 August 2024 (UTC)

    @EncycloPetey since this PDF is published on the DOD/WHS website, doesn't that make this particular collection of documents a publication of DOD/WHS? (Genuine question, I can imagine there are cases -- and maybe this is one -- where it's not useful to be so literal about what constitutes a publication or to go off a different definition. But I'm interested in your thinking.) -Pete (talk) 20:11, 9 August 2024 (UTC)

• • Why would a particular website warrant a different consideration in terms of what we consider a publication? How and why do you think it should be treated differently? According to what criteria and standards? --EncycloPetey (talk) 20:23, 9 August 2024 (UTC)

    Your reply seems to assume I have a strong opinion on this. I don't. My question is not for the purpose of advocating a position, but for the purpose of understanding your position. (As I said, it's a genuine question. Meaning, not a rhetorical or a didactic one.) If you don't want to answer, that's your prerogative of course.

    I'll note that Wikisource:Extracts#Project scope states, "The creation of extracts and abridgements of original works involves an element of creativity on the part of the user and falls under the restriction on original writing." (Emphasis is mine.) This extract is clearly not the work of a Wikisource user, so the statement does not apply to it. It's an extract created by (or at least published) by the United States Department of Defense, an entity whose publishing has been used to justify the inclusion of numerous works on Wikisource.

    But, I have no strong opinion on this decision. I'm merely seeking to understand the firmly held opinions of experienced Wikisource users. -Pete (talk) 20:42, 9 August 2024 (UTC)

    You misunderstand. The page we currently have on our site is, based on what we have so far, an extract from a longer document. And that extract was made by a user on Wikisource. There is no evidence that the page we currently have was never published independently, so the extract issue applies here. We can host it as part of the larger work, however, just as we host poems and short stories published in a magazine. We always want the work to be included in the context in which it was published. --EncycloPetey (talk) 20:55, 9 August 2024 (UTC)

    OK. I did understand that to be TEaeA,ea's position, but it appeared to me that you were disagreeing and I did not understand the reasons. Sounds like there's greater agreement than I was perceiving though. Pete (talk) 21:36, 9 August 2024 (UTC)

    I am unclear what you are referring to as a "longer document." Are you referring to the need to transcribe the Russian portion? That there are unreleased pages beyond the piece we have here?. Or are you saying the "longer document" is all 53 sets of releases almost 4000 pages listed here (https://www.esd.whs.mil/FOIA/Reading-Room/Reading-Room-List_2/Detainee_Related/)? I hope you are not advocating for merging all ~4000 pages into a single continuous page here, some some subdivision I assume is envisioned.

    Re the policy statement: I am not sure that is definitive: if someone writes me a letter or a poem and I paste that into a scrapbook, is the "work" the letter, the scrapbook or both? Does it matter if it is a binder or a folder instead of a scrapbook? If a reporter copies down a speech in a notebook, is the work the speech or the whole notebook. etc. I am pretty sure we haven't defined with enough precision to point to policy to say one interpretation of "work" is clearly wrong, which is why we have the discussion. MarkLSteadman (talk) 05:36, 10 August 2024 (UTC)

    The basic unit in WS:WWI is the published unit; we deal in works that have been published. We would not host a poem you wrote and pasted into a scrapbook, because it has not been published. For us to consider hosting something that has not been published usually requires some sort of extraordinary circumstances. --EncycloPetey (talk) 15:53, 10 August 2024 (UTC)

    From WWSI: "Most written work ... created but never published prior to 1929 may be included", Documentary sources include; "personal correspondence and diaries." The point isn't the published works, that is clear. If someone takes the poem edits it and publishes in a collection its clear. It's the unpublished works sitting in archives, documentary sources, etc. Is the work the unpublished form it went into the archive (e.g separate letters) or the unpublished form currently in the archives (e.g. bound together) or is it if I request pages 73-78 from the archives those 5 pages in the scan are the work and if you request pages 67-75 those are a separate work? MarkLSteadman (talk) 17:18, 10 August 2024 (UTC)

    I will just add that in every other context we refer to a work as the physical thing and not a mere scanned facsimile. We don't consider Eighteenth Century Collections Online scanning a particular printed editions and putting up a scan as the "published unit" as distinct from the British Library putting up their scan as opposed to the LOC putting up their scan or finding a version on microfilm. Of course, someone taking documents and doing things (like the Pentagon Papers, or the Family Jewels) might create a new work, but AFAICT in this context it is just mere reproduction. MarkLSteadman (talk) 05:37, 12 August 2024 (UTC)

    In the issue at hand, I am unaware of any second or third releases / publications. As far as I know, there is only the one release / publication. When a collection or selection is released / published from an archive collection, that release is a publication. And we do not have access to the archive. --EncycloPetey (talk) 17:34, 12 August 2024 (UTC)

    We have access, via filing a FOIA request. That is literally how those documents appeared there, they are hosted under: "5 U.S.C. § 552 (a)(2)(D) Records - Records released to the public, under the FOIA," which are by law where records are hosted that have been requested three times. And in general, every archive has policies around access. And I can't just walk into Harvard or Oxford libraries and handle their books either.

    My point isn't that can't be the interpretation we could adopt or have stricter policies around archival material. Just that I don't believe we can point to a statement saying "work" or "published unit" and having that "obviously" means that a request for pages 1-5 of a ten report is obviously hostable if someone requests just those five pages via FOIA as a "complete work" while someone cutting out just the whole report now needs to be deleted because that was released as part of a 1000 page large document release and hence is now an "extract" of that 1000 page release. That requires discussion, consensus, point to precedent etc. And if people here agree with that interpretation go ahead. MarkLSteadman (talk) 03:16, 18 August 2024 (UTC)

    For example, I extracted Index:Alexandra Kollontai - The Workers Opposition in Russia (1921).djvu out of [2]. My understanding of your position is that according to policy the "work" is actually all 5 scans from the Newberry Library archives joined together (or, maybe only if there are work that was previously unpublished?), and that therefore it is an "extract" in violation of policy. But if I uploaded this [3] instead, that is okay? Or maybe it depends on the access policies of Newberry vs. the National Archives? Or it depends on publication status (so I can extract only published pamphlets from the scans but not something like a meeting minutes, so even though they might be in the same scan the "work" is different?) MarkLSteadman (talk) 03:45, 18 August 2024 (UTC)

    If the scan joined multiple published items, that were published separately, I would see no need to force them to be part of the same scan, provided the scan preserves the original publication in toto. I say that because there are Classical texts where all we have is the set of smushed together documents, and they are now considered a "work". This isn't a problem limited to modern scans, archives, and the like. The problem is centuries old. --EncycloPetey (talk) 04:21, 18 August 2024 (UTC)

    So if in those thousands of pages there is a meeting minute or letter between people ("unpublished") then I can't? MarkLSteadman (talk) 13:57, 20 August 2024 (UTC)

This discussion has gone way beyond my ability to follow it. However, I do want to point out that we do have precedent for considering documents like those contained in this file adequate sources for inclusion in enWS. I mention this because if the above discussion established a change in precedent, there will be a large number of other works that can be deleted under similar argument (including ones which I have previously unsuccessfully proposed for deletion). —Beleg Tâl (talk) 13:14, 13 August 2024 (UTC)

for example, see the vast majority of works at Portal:Guantanamo —Beleg Tâl (talk) 13:15, 13 August 2024 (UTC)

(@EncycloPetey, @MarkLSteadman) So, to be clear, the idea would be to say that works which were published once and only once, and as part of a collection of works, but that were created on Wikisource on their own, to be treated of extracts and deleted per WS:WWI#Extracts?

If this is the case, it ought to be discussed at WS:S because as BT said a lot of other works would qualify for this that are currently kept because of that precedent, including most of our non-scan-backed poetry and most works that appeared in periodicals. This is a very significant chunk of our content. — Alien333 (what I did & why I did it wrong) 09:29, 14 August 2024 (UTC)

Also, that would classify encyclopedia articles as extracts, which would finally decide the question of whether it is appropriate to list them on disambiguation pages (i.e., it would not be appropriate, because they are extracts) —Beleg Tâl (talk) 13:23, 14 August 2024 (UTC)

Extracts are only good for deletion if created separately from the main work. As far as I understood this, if someone does for example a whole collection of documents, they did the whole work, so it's fine, it's only if it's created separately (like this is the case here) that they would be eligible for deletion. Editing comment accordingly. — Alien333 (what I did & why I did it wrong) 15:00, 14 August 2024 (UTC)

We would not host an article from an encyclopedia as a work in its own right; it would need to be part of its containing work, such as a subpage of the work, and not a stand-alone article. I believe the same principle applies here. --EncycloPetey (talk) 15:36, 14 August 2024 (UTC)

Much of our non-scan backed poetry looks like this A Picture Song which is already non-policy compliant (no source). For those listing a source such as an anthology, policy would generally indicate the should end up being listed as subworks of the anthology they were listed in. I don't think I have seen an example of a poetry anthology scan being split up into a hundred different separate poems transcribed as individual works rather than as a hundred subworks of the anthology work.

Periodicals are their own mess, especially with works published serially. Whatever we say here also doesn't affect definitely answer the question of redirects, links, disambiguation as we already have policies and precedent allowing linking to sub-works (e.g. we allow linking to laws or treaties contained in statute books, collections, appendices, etc.). MarkLSteadman (talk) 02:57, 18 August 2024 (UTC)

They are non-policy compliant, but this consensus appears to have been that though adding sourceless works is not allowed, we do not delete the old ones, which this, if done, would do. — Alien333 ( what I did &
why I did it wrong ) 07:55, 18 August 2024 (UTC)

Pages of Index:Historical and Biographical Annals of Columbia and Montour Counties, Pennsylvania, Containing a Concise History of the Two Counties and a Genealogical and Biographical Record of Representative Families.pdf

OCR is mess and all over the place, Just throw the whole thing out and start again, unless someone has the time to calmly realign all the pages. ShakespeareFan00 (talk) 22:44, 18 August 2024 (UTC)

• Delete. The OCR is far too bad to be useful. Definitely better to just delete it. TE(æ)A,ea. (talk) 22:52, 18 August 2024 (UTC)

Please make sure that you tag the Index with the deletion notice. I see that not only are the created Pages full of OCR errors, but many of those created Pages have content that does not match the scan in any way for the side-by-side comparison. There may be deeper issues with the PDF. --EncycloPetey (talk) 22:53, 18 August 2024 (UTC)

The index itself is fine. There's no mechanism for mass noming a batch of pages though ShakespeareFan00 (talk) 22:58, 18 August 2024 (UTC)

Then when you said "throw the whole thing out", you did not mean the Index (which is what is listed)? --EncycloPetey (talk) 23:00, 18 August 2024 (UTC)

I meant the Page: s , the actual Index: page itself isn't bad. ShakespeareFan00 (talk) 06:38, 19 August 2024 (UTC)

Index:Dick Sands the Boy Captain.djvu

The file is missing two pages, and a number of additional pages have poorly scanned pages which would need replacement. In addition, the actual scan quality itself is poor, and doesn’t serve easy proofreading. We already have better scans (an illustrated one here and one from a collection here). The images would also be difficult to extract, owing to the same issue. The OCR is poor, and the text added to the pages isn’t useful either. The index and the pages should be deleted. TE(æ)A,ea. (talk) 00:36, 22 August 2024 (UTC)

Undelete FBI eMails concerning Moussaoui

In closing the discussion EncycloPetey has taken a stance on a matter currently of much debate: the “original source” for the purposes of items returned pursuant to FOIA requests. While the discussion in regards to this item did not touch upon the subject, the closing comment necessarily falls down on one side. Here are the two opinions:

1. The original source is the work as originally created. This means that each individual e-mail, witness statement, report, &c., within a FOIA request is its own separate item, and a source file can be extracted to fulfil the purpose.
2. The original source is the work as released. This means that each PDF returned from a FOIA request is a separate source, and the source file is the entire returned item.

I am of the former opinion; EncycloPetey is of the latter. This has come up with more force in the discussion above on “Kamoliddin Tohirjonovich Kacimbekov's statement.” In that case, which is as good an example as any for this sort of discussion, the FOIA release is an 83-page PDF containing numerous documents relating to a military trial of a detainee. One of those documents, found on pp. 43–45 of that document, is the statement which is presented as an individual work here. The question, then, is what is the original source for the statement? Is it the three-page document which contains the statement, or is it the 83-page PDF which contains the statement and other items released pursuant to the FOIA request? I use the statement as an example because nothing of it has yet been deleted, but the same logic applies to the e-mail in dispute. TE(æ)A,ea. (talk) 19:31, 22 August 2024 (UTC)

 Oppose The key point in closing this deletion request was that the "FBI eMails concerning Moussaoui" was neither an individual work, nor an element of the source found, but was a compilation selectively assembled by a Wikisource user. The fact that this was a user-assembled compilation was one of the reasons Jan Kameníček nominated the work, and we have previously deleted such compilations. The compilation was also annotated, which violates our Wikisource:Annotations policy.

We also had no source for the collection of emails. In the discussion, TE(æ)A,ea. pointed to a web page claiming it as a source, but I was not going to click through a hundred links on that page to try to hunt down the source. We have a claim above that there is a PDF as a source, but no link to this PDF has been provided. Since we had (a) a compilation, (b) annotated, with (c) no source, there was more than sufficient reason to delete, and there is no reason to undelete. --EncycloPetey (talk) 19:42, 22 August 2024 (UTC)

Translation:Exinct sincerae devotions affectus

Not in accordance with WS:T: "A scan supported original language work must be present on the appropriate language wiki..." -- Jan Kameníček (talk) 21:19, 28 August 2024 (UTC)

This appears to be a fairly important document, and otherwise not readily available in English. I think it's worth trying to salvage. I've uploaded File:Boletín de la Real Academia de la Historia, tomo 15.djvu to Commons; once I figure out how to navigate esWS I'll start transcribing this section. (Note, even though the work itself is in Latin, it is published in the Spanish periodical es:Boletín de la Real Academia de la Historia, so I'm assuming that its proper location is esWS rather than laWS) —Beleg Tâl (talk) 13:34, 29 August 2024 (UTC)

Why wouldn't a work in latin go to la? As far as I'm aware, only the language of the text means anything inclusion-wise. — Alien333 ( what I did
why I did it wrong ) 14:04, 29 August 2024 (UTC)

Because the Latin text is only a couple of pages embedded within a work written in Spanish. It's similar to how the Maori text of Translation:Ka Mate is hosted on here on enWS. —Beleg Tâl (talk) 14:41, 29 August 2024 (UTC)

But then wouldn't this constitute an excerpt translation? — Alien333 ( what I did
why I did it wrong ) 15:36, 29 August 2024 (UTC)

I suppose an argument could be made for treating such works as "excerpt translations" ... but Translation space is weird like that sometimes. Some rules don't apply there, like the rules against self-published original works, or the rules against evolving texts. But if the community is concerned about it, I can start setting up Translation:Boletín de la Real Academia de la Historia for the page to reside in. —Beleg Tâl (talk) 17:44, 29 August 2024 (UTC)

Template:Commons top icon

The following discussion is closed and will soon be archived:

Deleted. Unused and unneeded.

Now-unused template that puts a Commons logo in the indicators area. It used to be used on the file description pages of files hosted on Commons (through hooks provided by MediaWiki), but is now unused. It has no use outside that context, and the only reason it exists as a distinct entity is that templates (vs. Lua) do not support modularization through subroutines and similar (otherwise this whole template would have been a function inside a larger template). Xover (talk) 07:02, 31 August 2024 (UTC)

 This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. --EncycloPetey (talk) 01:22, 9 September 2024 (UTC)

Translation:Constitution of the Esthonian Republic (1934)

The following discussion is closed and will soon be archived:

Kept. User-original translation is now tied to a scan-backed original at ee.WS.

This translation has two key problems: (1) It is based on an English translation, not on a copy in another language. (2) There is no original backed by a scan on the parent language Wikisource. --EncycloPetey (talk) 15:47, 2 September 2024 (UTC)

Note: it claims to be a translation of the preceding constitution, which has different text. — Alien333 ( what I did
why I did it wrong ) 16:07, 2 September 2024 (UTC)

How did you get that from the document? It claims to be based on another English translation, to which it links. --EncycloPetey (talk) 16:28, 2 September 2024 (UTC)

And that translation is titled: Constitution of the Esthonian Republic (1920), not 1934. (It's the same one that's linked in the previous header link.) I checked with the copyvio detector, and it is indeed a different text, with the '34 one having a few rephrasings/additions. — Alien333 ( what I did
why I did it wrong ) 16:49, 2 September 2024 (UTC)

What is relevant for the purposes of a user-created Translation on Wikisource is that it must be a translation from a non-English original language. Translation from modern English to modern English is not a Translation. Being based on a public domain English translation isn't relevant here, particularly when the editor of the page has previously made errors in posting dates. --EncycloPetey (talk) 17:14, 2 September 2024 (UTC)

Yes, yes, I know, this was just a note passing by,  Delete anyways. — Alien333 ( what I did
why I did it wrong ) 17:15, 2 September 2024 (UTC)

But your note is incorrect. It does not claim to be a translation of the 1920 constitution; it claims to use the language from that translation for a translation of the 1934 constitution. Hence: based on. --EncycloPetey (talk) EncycloPetey (talk) 17:18, 2 September 2024 (UTC)

Sorry. — Alien333 ( what I did
why I did it wrong ) 17:19, 2 September 2024 (UTC)

Because the 1934 constitution isn't a fully-fledged constitution, but an amended version of the 1920 constitution. Hence, the parts of the 1920 version which were not affected by the 1934 amendments were kept as-is. Glide08 (talk) 19:06, 7 September 2024 (UTC)

 This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. --EncycloPetey (talk) 06:22, 10 September 2024 (UTC)

Index:Notes & Queries for Somerset and Dorset v4.djvu

The following discussion is closed and will soon be archived:

Deleted. BAdly damaged scan with missing and unreadable pages. Also a duplicate of Index:Notes and Queries for Somerset and Dorset - Volume 4.pdf

While it says that p. 278 is missing, that’s not actually true; it is there, just so poorly scanned as to be useless. (The back of the plate facing p. 360 and p. 361 are missing.) But that list isn’t complete: here are all of the pages with scans so poor as to be useless: 398, 44, 59, 120, the image in between 132 and 133, 145, 188, 224, 236, 244, 246, 250, 274, 278, 317, 338, 376, 382, vi, and many pages at the end which were the title pages of individual parts of the volume. TE(æ)A,ea. (talk) 21:47, 2 September 2024 (UTC)

 This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. --EncycloPetey (talk) 22:51, 10 September 2024 (UTC)

Index:Notes & Queries for Somerset and Dorset v9.djvu

This has a similar problem to volume 4. The following pages are either not scanned, missing, or were scanned such that some or all of the page is unreadable: viii–2, 4, 10, plate facing p. 35, 62, 74, 84, 106, 108–110, 114, 117, 225, 229, 237, 247, 255, 257, 276, 282, 303, 309, 341–342, 344–345, 347, 361, 379, 383, 385, back of Part LXXII. TE(æ)A,ea. (talk) 02:19, 3 September 2024 (UTC)

"Collective Edition of 1883"

The following discussion is closed and will soon be archived:

Deleted. Long-abandoned table of contents with no backing scans.

Empty TOC, no content after 10 years —Beleg Tâl (talk) 14:43, 3 September 2024 (UTC)

 Delete as above. —CalendulaAsteraceae (talk • contribs) 23:52, 8 September 2024 (UTC)

 This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. --EncycloPetey (talk) 22:53, 10 September 2024 (UTC)

Template:Decade years navbox

Unused template; redundant to and less flexible than {{categories by date}}. —CalendulaAsteraceae (talk • contribs) 02:11, 9 September 2024 (UTC)

Several Maccabees

The following:

• Translation:2 Maccabees
• Translation:4 Maccabees
• Translation:5 Maccabees
• Translation:6 Maccabees
• Translation:7 Maccabees
• Translation:8 Maccabees
• Translation:9 Maccabees

All of these translations have sat unfinished (some with only a header) and abandoned since 2012. There is no designated scan-backed original for any of them. As far as I can tell, there isn't any copy for any these books on el.WS, whether scan-backed or not. --EncycloPetey (talk) 22:11, 10 September 2024 (UTC)

Boykisser (song)

These are lyrics from YouTube, which is effectively self-publication, and is not allowed under WS:WWI. --EncycloPetey (talk) 21:25, 12 September 2024 (UTC)

 Delete per nom. I appreciate the good faith contribution, but I don't see any wiggle room here. —Beleg Tâl (talk) 23:23, 12 September 2024 (UTC)

The sourced video is at nearly a million views. Many works we accept only sold copies in the thousands or less during their time. Its subject appears to also be moderately culturally relevant in modern times. Our transcription of it also got 52 page views this month, not at all an insignificant number. I tend to pay close attention to our page view statistics, in the pursuit of my personal interest in "Wikisource recruitment", and I'll just tell you a whole lot of very academically notable works get a lot less views than this.

 Abstain for now. We should consider that perhaps our site could get some badly needed attention if we used modern consumption trends (social media, video games, and we've already been doing movies extensively) as a tactic to gain new editors and readers through exposure. The reality is that social media defines the modern ecosystem, whether we like it or not. I certainly don't, but it is a reality, and all I'm saying is Wikisource could benefit if we somehow took advantage of that.

I also understand the sentiment that we don't need to become a dumping ground for whatever garbage someone wants to throw on TikTok or YouTube. It'd be ideal to draw some sort of arbitrary measure for where to draw the line with "social media", but I don't know what that margin should be. If it has a Wikipedia article about it (like Me at the zoo), that's usually pretty uncontroversial to include already, but obviously this "Boykisser" video doesn't, so it wouldn't apply to that rule. This can be subject for a broader policy discussion, one of these days. (Wiktionary has relatively recently changed its rules to allow Internet sources instead of just traditional paper sources due to similar kinds of lexical restraints they were having, and maybe we could consider a similar path.)

I'm just brainstorming, don't take it too seriously. SnowyCinema (talk) 03:11, 13 September 2024 (UTC)

Without some objective criterion for inclusion, one that isn't subject to whim or trends, the point is moot. --EncycloPetey (talk) 03:25, 13 September 2024 (UTC)

• Keep. I don’t think that a strict application of the self-publication rule is appropriate in this case. Historical factors have led to self-publication as a viable alternative to what would have been formal publication, without the correspondent negative associations which prevailed at the time the policy was enacted. TE(æ)A,ea. (talk) 16:03, 13 September 2024 (UTC)
•  Delete per nom. Should we accept such sources under some circumstances, we need to define such circumstances and modify our rules first, not vice versa. --Jan Kameníček (talk) 16:08, 13 September 2024 (UTC)

Official Toki Pona Dictionary and Notes on lipu pu

These are extracts from Toki Pona: The Language of Good (2014) and Toki Pona Dictionary (2021) respectively, and hence fail WS:WWI. It is perhaps worth noting that these extracts have been released under a CC license, while the remainders of these works have not (and are not available online at all). —Beleg Tâl (talk) 13:36, 13 September 2024 (UTC)

• Keep for the reasons you have given. Insofar as they have been separately released, they are whole works. Notes on lipu pu is already scan-backed, and it shouldn’t be too hard to get a copy of the 2014 book for other scan-backing purposes. TE(æ)A,ea. (talk) 16:03, 13 September 2024 (UTC)

  But the first text is pulling content from a copyrighted work and slapping a CC license on it. That's a serious problem, because I don't see how you can claim CC on copyrighted material. The second work appears to be a response to the original text, quoting from it, and does not appear to be an extract. --EncycloPetey (talk) 16:13, 13 September 2024 (UTC)

  • EncycloPetey: From a technical standpoint, obviously, one must have copyright on certain material in order to release it under a license. I assumed (without looking in to the matter) that the license was legitimate. I found a forum post which says that the dictionary was released into the public domain in the original book; if this is the case, the license is not an issue. Given that statement, I have ordered the book and will scan the dictionary if it has actually been released. TE(æ)A,ea. (talk) 19:50, 13 September 2024 (UTC)

•  Delete, noting that the "scan" of the Notes on lipu pu seems to be just some self-published pdf. --Jan Kameníček (talk) 16:20, 13 September 2024 (UTC)
  • Jan Kameníček: No, that’s just a copy of the “notes” section from the 2021 book, which has been published. TE(æ)A,ea. (talk) 19:50, 13 September 2024 (UTC)

    I see. I am striking my vote for now and will think about it again. --Jan Kameníček (talk) 09:42, 14 September 2024 (UTC)