Please do not post any new comments on this page. This is a discussion archive first created in , although the comments contained were likely posted before and after this date. See current discussion or the archives index.
Hi all. I'm very new here, so if I'm asking this in the wrong place, I apologise. I have been working on proofreading Facts and Fancies about Our "Son of the Woods", Henry Clarence Kendall and his Poetry to the best of my ability. After skipping over a few pages and parts of pages due to complications I am at the 21st page and it seems that... the errata information is covering up half of the page. Is there any way I can remedy this? Does this part of the book have to be re-scanned? Again, I hope this is the correct place to ask. Any help is appreciated, cheers. KoolKidz112 (talk) 22:02, 18 April 2023 (UTC)
A good scanner will do that: scan the Errata, then the "back" of the errata slip, then the page after the Errata. But there are books where they didn't take such car. I'm glad to hear this instance does not have a problem. --EncycloPetey (talk) 23:39, 18 April 2023 (UTC)
The two are not really related. Phetools runs on the Toolforge servers and queries the replica databases, while {{Graph:PageViews}} shows data from the WMFs statistics service in production. There are certainly things that could take down both at once but they are relatively unlikely to occur in practice. In any case, it appears the Phetools problem was a transient issue since it's back now.The PageViews problem is because the extension was disabled due to a security issue, cf. this thread on enWP. Some information has gone out on the technical channels and you should expect a MassMessage notification to the village pumps (i.e. here) soonish. Xover (talk) 06:10, 19 April 2023 (UTC)
The first half of the meeting will be focused on non-technical updates and conversations like events, conferences, proofread-a-thons and collaborations. The second half will be focused on technical updates and conversations, such as talking about major challenges faced by Wikisource communities, similar to the ones conducted in previous Community meetings.
If you are interested in joining the meeting, kindly leave a message on sgill@wikimedia.org and we will add you to the calendar invite.
Meanwhile, feel free to check out the page on Meta-wiki and suggest any other topics for the agenda.
Do you want to host an in-person or virtual session at Wikimania 2023? Maybe a hands-on workshop, a lively discussion, a fun performance, a catchy poster, or a memorable lightning talk? Submissions are open until March 28. The event will have dedicated hybrid blocks, so virtual submissions and pre-recorded content are also welcome. If you have any questions, please join us at an upcoming conversation on March 12 or 19, or reach out by email at wikimania@wikimedia.org or on Telegram. More information on-wiki.
All respondents to the vote had the opportunity to provide comments regarding the contents of the revised Enforcement Guidelines draft document. A total of 369 participants left comments in 18 languages; compared to 657 commenters in 27 languages in 2022. The Trust and Safety Policy team completed an analysis of these results, categorizing comments to identify major themes and areas of focus within the comments. The report is available in translated versions on Meta-wiki here. Please help translate into your language.
Again, we are thankful to all who participated in the vote and discussions. More information about the Universal Code of Conduct and its Enforcement Guidelines can be found on Meta-wiki.
This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. Yann (talk) 16:08, 4 April 2023 (UTC)
Tech News: 2023-14
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
The system for automatically creating categories for the Babel extension has had several important changes and fixes. One of them allows you to insert templates for automatic category descriptions on creation, allowing you to categorize the new categories. [1][2][3][4][5]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 4 April. It will be on non-Wikipedia wikis and some Wikipedias from 5 April. It will be on all wikis from 6 April (calendar).
Some older Web browsers will stop being able to use JavaScript on Wikimedia wikis from this week. This mainly affects users of Internet Explorer 11. If you have an old web browser on your computer you can try to upgrade to a newer version. [6]
The deprecated jquery.hoverIntent module has been removed. This module could be used by gadgets and user scripts, to create an artificial delay in how JavaScript responds to a hover event. Gadgets and user scripts should now use jQuery hover() or on() instead. Examples can be found in the migration guide. [7]
Some of the links in Special:SpecialPages will be re-arranged. There will be a clearer separation between links that relate to all users, and links related to your own user account. [8]
You will be able to hide the Reply button in archived discussion pages with a new __ARCHIVEDTALK__ magic word. There will also be a new .mw-archivedtalk CSS class for hiding the Reply button in individual sections on a page. [9][10][11]
Future changes
The Vega software that creates data visualizations in pages, such as graphs, will be upgraded to the newest version in the future. Graphs that still use the very old version 1.5 syntax may stop working properly. Most existing uses have been found and updated, but you can help to check, and to update any local documentation. Examples of how to find and fix these graphs are available.
re __ARCHIVEDTALK__, I have added that magic work to Template:archive header which should cover the majority of our community talk pages and many user talk pages (list at Special:WhatLinksHere/Template:Archive_header). If people believe that we should not be imposing that on user talk pages, then we can do some further tweaking to allow people to exclude pages in user ns. I feel that the application as a default is advisable. — billinghurstsDrewth22:05, 5 April 2023 (UTC)
Would this be in scope for Wikisource? Is the license acceptable here? This was deleted on Commons for being a plain text document. If the license is OK and the transcription is accepted, then it can be undeleted on Commons. Source mentioned from [12] and [13]. Thanks, Yann (talk) 14:01, 8 April 2023 (UTC)
There has been discussion about the OGL and the general trend of opinion is against it being an acceptable license.--Prosfilaes (talk) 14:18, 8 April 2023 (UTC)
Legislation Versions as Synthesis of Works
Not sure if this should have been posted on the Versions talk page, but since that's been inactive since 2013, thought I'd post it here.
I do understand that we don't do syntheses of works for a reason, but I also can't think of any other situation in which it would be as useful as with legislation.
Each version of an Act is quite literally a synthesis of works. It is the act as enacted, but with some paragraphs removed/replaced/added based on other works, the amendment acts. An amendment act will literally say "Remove section 7 and replace with: "
Let's say you've got "The Example Act 2022", some 500 pages long. We transcribe the whole beast of an act, and then they release "The Example Act Amendment Act 2023", where they've added "section 7A" in between sections 7 and 8. Then they publish "The Example Act 2023 (Version 2)", which itself is a synthesis of The Example Act 2022 and The Example Act Amendment Act 2023.
It doesn't make much sense for us to duplicate all that text with a slight adjustment in version 2. That leaves room for mistakes, it creates way more work, and isn't a very efficient use of our time. It also means you have to have confidence in the transcribing of two documents. Confidence that one hasn't been incorrectly edited. And if you're correcting one version that has a mistake that's been in all versions, you have to go correct them all.
It makes far more sense to me to have Version 2 be a synthesis of works. Transcribing all the content of the act as enacted, except for the sections that have been amended, which will instead be transcribed from the Amendment Act (Or entered as plaintext if the full amendment act hasn't been transcribed).
Then, if something needs correcting due to poor transcription, amending it in the act as enacted or the correct amendment act would correct it in every relevant version of the act. This is "Single Source of Truth" thinking.
I do get there isn't much in the way of multiple versions of legislation, but I've been trying to think of ways to make it work better, and I see the Criminal Code Act 1995 discussion above touched on it.
Although I am tempted to agree, I can also see one problem which I consider important: I am afraid that such syntheses can be in some (many?) cases quite complicated and there is a big danger of misinterpretation. Also checking the work done would not be as easy as simple checking of text against a source text. I do have doubts that people would make such complicating checks, and with the extremely limited number of people who do patrolling here I am quite afraid that the amount of bad syntheses would be piling up quickly. Besides that, even now, when this approach is not allowed, I meet poor syntheses which have escaped patrolling. For this reason I am more inclined to vote against it. If such a change were to be done despite these objections, I would suggest founding a new namespace for such syntheses. --Jan Kameníček (talk) 23:12, 2 April 2023 (UTC)
The solution to misinterpretation and checking against source text is simple: There is still a source text. Every version of legislation still gets a new source text published, and that should be attached to the synthesised version. In terms of checking/verifying, you can check the synthesised text against the physical publication. No room for interpretation, our rule about matching the source would still be in place.
The synthesis is purely a timesaving and accuracy measure. The source text is a synthesis, the wikisource version would do exactly the same. ElDubs (talk) 23:50, 2 April 2023 (UTC)
I think you're thinking of Wikisource primarily as a place to host copies of legislation. From that perspective what you're proposing makes some sense. But Wikisource isn't about legislation, it's about published works. And in the majority of cases what has been published is not a complete amended version of an act, but rather an act to amend. What was published doesn't read "§4. Flurbs are henceforth prohibited.", it reads "§4. of the Prohibitions FTW Act shall read ‘Flurbs are henceforth prohibited’". If we are to read those instructions, interpret them, and modify the base act (the actually published version) accordingly we will be creating an entirely new work that did not exist before. And at what point the competent legislative body chooses to publish a consolidated act, or not, or replace it with an entirely new version, is also relevant information.In other words, in order to do what you propose here, as an exception just for legislation, we would need dedicated tooling to handle all these concerns (and we can barely get the base Proofread Page extension maintained). Meanwhile, there are lots of websites out there that do just that: show synthetic consolidated versions of legislation, with traceability and annotations to indicated the amendment history of any given section. At best we'd spend a whole bunch of limited resources on becoming a low-rent copy of these websites. Xover (talk) 07:35, 3 April 2023 (UTC)
At the recent Wellington WikiCon I was discussing this very issue with someone with extensive experience in drafting legislation and then shepherding it through the various processes towards enactment. A distilled version of their response is, "Why? It's all available at nzlii.org and legislation.govt.nz". There are similar legal websites in most other jurisdictions and where a WP article needs to link to an Act that's where they will make the link, rather than to here—particularly as there are hundreds of acts enacted at each parliamentary session and we cannot possibly keep up with such. Where there is particular purpose for us to host an act (such as those on Portal:Royal Society of New Zealand), then by all means—but still only at that point in time. Beeswaxcandle (talk) 18:52, 3 April 2023 (UTC)
@Beeswaxcandle: nzlii.org is run by one person (met her, she's lovely). It provides no transcription for historical legislation. Legislation.govt.nz is a fantastically designed website, but they have no plans to add historical legislation (which is why nzlii holds it in the first place, it was a government money saving decision). This means that there can be instances where each of these websites may not be appropriate for a user's use case.
Wikisource, being built and driven by users, can have any published work uploaded and transcribed. I consider it a far better solution than either of those websites for this reason. Certainly, we can never "keep up" with the amount of legislation that comes out, so if a user determines a particular act is useful to them to transcribe, they can do so.
Which is why for acts in my area of interest, and how those acts have changed over time, I'd love to be able to use timesaving measures like largely transcribing previous re-prints of acts via synthesis with amendment acts to fast-track the transcription of new re-prints. ElDubs (talk) 22:53, 3 April 2023 (UTC)
@Xover: To clarify, the "synthesis" I'm suggesting is definitely of actual published works. So we won't be creating an entirely new work that did not exist before. We'll be duplicating the method used to create a work that does exist.
The Act as enacted is a published work, the amendment act is a published work. Both of these we agree on. However there is also the re-printed act.
The re-print is a published work which is very simply a synthesis of the act as enacted and the amendment act. At the very least in NZ, every single amendment to an act gets a re-print of the Act published. So what was published will indeed read "§4. Flurbs are henceforth prohibited.".
I completely agree we shouldn't create new works here on Wikisource, and I wouldn't suggest it. What I suggest is we upload the reprint (The officially published synthesis), and then instead of transcribing it, we just transcribe the act as enacted, but edit with the amendment act. The result will be a perfect match of the published re-print. From a reader perspective there will be absolutely zero difference. The reason for the synthesis is purely to save time, and for the other reasons I mentioned above. At no point would we be creating a new work.
With this in mind, I don't anticipate any impact on the base Proofread Page extension or any need for dedicated tooling. It's simply a recognition that in this instance, the synthesis of works is not creating a new work, and should therefore be allowed. ElDubs (talk) 22:47, 3 April 2023 (UTC)
Comment Our primary purpose is the creation of versions, pure and clean. If someone wishes to see a version of a time, that is our role.
We have had that sort of discussion about commentary/interpretation/comparison of works that we have and it has been a clear decision that commentary/interpretation/comparison was possible outside of the clear version. The problem has always then been how to do it cleanly. Unfortunately the system DIFF comparison tool doesn't work with comparing works with transcluded pages, though it may do so okay at a more granular Page: to Page: level, though that is problematic as the works are less likely to line up at that granular level. There is the possibility that someone could fork and fix the mw:extension:DoubleWiki tool, though that is a wish, not a certainty. [As a note here I am suggesting that we could do an internal wiki comparison rather than an intrawiki language comparison]. I would almost suggest that it is either that sort of extension, or someone creates a tool in wmflabs that allows side-by-side comparison of pages. — billinghurstsDrewth22:37, 5 April 2023 (UTC)
@Billinghurst: In this instance I'm not raising the issue of comparing versions, I'm discussing creating versions, which as you say, is our primary purpose. I'd like to streamline substantially similar versions by having future versions of legislation just be a synthesis of works as described above. This would not require significant effort to change any rules, simply acknowledging that in the case of legislation, it makes sense for these reasons:
Versions of legislation are "officially published" syntheses of works. They take the act as enacted, they adjust based on amendment acts, and then they publish that as a reprint. The reprint has no new information, it's just the original, with edits taken verbatim from the amendment act. If the official one is a synthesis, then why can't ours be so long as it perfectly matches?
If I want to put 50 versions of one act here, being allowed to make 49 of them a synthesis of works would simplify that process significantly.
It follows the principle of "single source of truth", since that information is in fact exactly the same, then on Wikisource, it will always just be transcribed from the original source, the act as enacted.
We reproduce works, and that this is in their entirety, what it seems you are talking about is an excerpted version as a diff. That is a change to how we are doing our works. I am not seeing the benefit, nor how we could proofread it from a source. You are talking a whole new scope. You could just as easily have the first version, then put all the latter changes on the respective talk pages in a chronological order. We have enough issues with people treating these as dynamic works, and all I see is that we are starting to fork a small section of work that makes it near impossible to properly manage. Legislation is just problematic. — billinghurstsDrewth07:16, 7 April 2023 (UTC)
I'm talking about reproducing a works in their entirety. No extra work, no difficulty managing because it's connected to the original. I'll give it a go and come back with an example. ElDubs (talk) 04:09, 9 April 2023 (UTC)
I think that there are some crossed thinkings here. The technique by which you create the Pooh Bah Act (1899) as amended by the Pooh Bah Amendment Act (1905) and the Amendments to Several Acts Act (1978) is not important—as long as there is a published document for the amended act in the form in which we are hosting it. Preferably that published document is in the Index: and Page: namespaces and transcluded. The question I have is about the cost/benefit ratio. Do the benefits of us hosting multiple versions of an Act outweigh the costs to the Wikisource community to take the versions through the proofread/validate process? Or future scan-backing? While the Proofread process is "simplified" a little by you doing the bulk of the work, each version will need to have second eyes for validation. If you go down this path further, then I recommend waiting for the validation process before creating versions using the previously proofread text. Beeswaxcandle (talk) 05:32, 6 April 2023 (UTC)
@Beeswaxcandle:, I do think there's some crossed wires here. Because what I describe would not need the proofread/validate process for multiple versions. I'll try and do a workflow to show why.
I upload and transcribe the Pooh Bah Act (1899), it's 100 pages long and has 300 sections. It is the "act as enacted", as in it's the original act passed in 1899. The wikisource community proofreads/validates that.
That act then gets transcluded to the main space as "Pooh Bah Act (1899)".
I upload and transcribe the Pooh Bah Amendment Act (1900), it's 3 pages long. It is an amendment act, section 3 of this act states: "This act repeals section 9 of the Booh Bah Act (1899)". A very small document, the wikisource community proofreads/validates that.
That amendment act then gets transcluded to the main space as "Pooh Bah Act (1899)"
I upload the Pooh Bah Act (1899) (Version 2), this is a re-print of the Pooh Bah Act (1899), but with the Amendmend Act incorporated. It is 100 pages, and apart from one crossed out paragraph in section 9, it is the same document as the original. But it is its own published document.
I then again transclude a combination of the Pooh Bah Act (1899) and the Pooh Bah Amendment Act (1900). I'm not at all transcribing or transcluding the (Version 2) document. No one is proofreading or validating it because they have already proofread and validated the documents that make it up. We allow this because as I said, (Version 2) itself is officially a synthesis of those other two documents. So we're duplicating that process here.
By proofreading just the act as enacted and the small amendment acts, you automatically proofread all the versions because they transclude those documents. In this way, you don't need to wait for the validation process, if the validation process finds an error, then that error only needs to be fixed once on that original version. And because all versions are transcluded from the original Index/Page, they will all automatically update.
In terms of cost/benefit, by doing this we can eliminate proofreading of "versions" and just proofread acts passed. That I think is very desirable.
I'm not keen to just pursue something that doesn't have support, but I am keen to discuss any perceived issues to see if I can get that support. ElDubs (talk) 07:45, 6 April 2023 (UTC)
If you are talking AN ADDITIONAL compiled presentation on top of the pure existing publications versions, then my complaints are lessening, I do just wonder whether this is not something that would be sitting better WMFLABS outside of Wikisource, though utilising the data. Do your internal section builds to a doc and you could probably pull and present sections. The issue taht I see is that in the form you suggest that you are missing the integral part of changes and from whence the legislative change occurs, and the bill from which is occurs, and NEVER as well as can be done with sites that are dedicated to this design type and designed that way. For example, this Act's presentation https://content.legislation.vic.gov.au/sites/default/files/2022-09/06-16aa026%20authorised.pdf and then a site like http://austlii.edu.au/ — billinghurstsDrewth07:25, 7 April 2023 (UTC)
@Billinghurst: It definitely includes the changes from where the changes occur.
As I mentioned, the Pooh Bah Act (1899) and Pooh Bah Amendment Act (1900) are both uploaded, transcribed and added as their own published documents. The Amendment Act is where the change occurs.
We'll then upload Pooh Bah Act (1899) (Version 2), which is a document the government published that incorporates the amendment into the act.
But we'll never transcribe/transclude it. Instead, the page for version 2 will show this:
This is how Version 2 would be constructed. It takes all of the original act, but it inserts the part of the amendment act that gets amended. It'll look exactly like the document the government publishes because of this. No need for anything new at WMFLABS, as the tools to do this are already in place here.
You're right that we won't do it as well as the government legislation website, but NZ legislation is only "complete" back to around 2007 on the government website. We will however do it far better than NZLII, which holds all pre-2007 legislation, but has done nothing with it other than upload and OCR with no verification.
With this, we could add the legislative history of an act with relatively little effort. Transcribe amendments, and then just use this method to "generate" the re-prints. But requiring that this can only be done if re-prints have been officially published by the government, because we don't create new works. ElDubs (talk) 11:53, 7 April 2023 (UTC)
I think to some extent someone else had already used this approach with certain South African Legislation. It took a while figuring it out, when repairing Linter concerns.ShakespeareFan00 (talk) 14:38, 8 April 2023 (UTC)
thanks @Billinghurst:. In our project artist and art object are linket to Wikidata in property name (P2561) qualifier point of time P585, reference: stated in P248, page P304, volume P478. You can see an example here. Susanna Giaccai (talk) 09:15, 9 April 2023 (UTC)
Will this document associated with General Douglas MacArthur be acceptable in Wikisource?
Can anyone advise me whether the following unpublished document will be accepted/deleted if entered in Wikisource?
I note, under "How to add a work to Wikisource", there is:
"In almost all cases, works on Wikisource must have been professionally published. Wikisource does not host original or vanity press works; everything in our library must have been previously published. The only exception to this is for source documents of historical importance; in this case Wikipedia-style notability requirements apply."
The document is an application for extension of an alien bond made by General Douglas MacArthur on November 13, 1939 in respect of Loh Chui, the amah who cared for his four-year-old son and who accompanied the son, Mrs MacArthur and the general during their perilous escape from the Philippines after the Japanese invasion. She has been described by the MacArthur Memorial as being in effect part of the MacArthur family, continuing to be a companion to Mrs MacArthur until her death in the 1970s.
As such, I believe the document passes the Wikipedia requirement for notability. However, it's possible others may not agree.
The document was initiated by the Department of Labor, Government of the Philippines. Section 176 of the Intellectual Property Code of the Philippines, Works of the Government, states "No copyright shall subsist in any work of the Government of the Philippines." It is therefore in the public domain.
The scanned copy was provided by the MacArthur Memorial but there is no intention currently to publish it. Cheers, Simon – SCHolar48 🇦🇺💬 at 06:30, 9 April 2023 (UTC)
This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. SCHolar44 (talk) 10:04, 4 April 2023 (UTC)
Tech News: 2023-15
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
There was a problem with all types of CentralNotice banners still being shown to logged-in users even if they had turned off specific banner types. This has now been fixed. [16]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 11 April. It will be on non-Wikipedia wikis and some Wikipedias from 12 April. It will be on all wikis from 13 April (calendar).
Starting on Wednesday, a new set of Wikipedias will get "Add a link" (Moroccan Arabic Wikipedia, Danish Wikipedia, Dinka Wikipedia, Lower Sorbian Wikipedia, Ewe Wikipedia, Greek Wikipedia, Emiliano-Romagnolo Wikipedia, Esperanto Wikipedia, Estonian Wikipedia, Basque Wikipedia, Extremaduran Wikipedia, Tumbuka Wikipedia, Fulah Wikipedia, Finnish Wikipedia, Võro Wikipedia, Fijian Wikipedia, Faroese Wikipedia, Arpitan Wikipedia, Northern Frisian Wikipedia, Friulian Wikipedia, Irish Wikipedia, Guianan Creole Wikipedia, Scottish Gaelic Wikipedia, Galician Wikipedia, Gilaki Wikipedia, Guarani Wikipedia, Goan Konkani Wikipedia, Gothic Wikipedia, Gujarati Wikipedia, Manx Wikipedia). This is part of the progressive deployment of this tool to more Wikipedias. The communities can configure how this feature works locally. [17][18]
Hi, I am trying to complete editions of Tolstoy. The Complete Works of Lyof N. Tolstoï is half complete, and the list of volumes is wrong. [19] shows at least 14 volumes, and volume 14 is "The Kingdom of God is Within You". Any idea how to find the missing volumes? Thanks, Yann (talk) 17:52, 9 April 2023 (UTC)
There are at least two sets of volumes floating around here with variances between djvu and pdf. Someone with the patience and inclination to sort out one complete set under one file type (preferably djvu) is needed. Then the appropriate page moves can be botted and the other set(s) disposed of. There are various conversations about the sets in the archives. Beeswaxcandle (talk) 20:03, 9 April 2023 (UTC)
See my comment on the discussion page. The edition that was started had each physical volume containing two "volumes" of the series (e.g. physical volume 4 contains what would be Anna Karenina Volume I and Anna Karenin volume II which are in the single volume sets volumes 7 and 8). I posted a link to the missing two double volumes in google scan versions if you want to complete the set. Open to ideas about how to make this more clear to readers, the volumes were released that way with a combined title page so it isn't just afterwards combining two separate books. MarkLSteadman (talk) 20:35, 9 April 2023 (UTC)
To emphasize the potential confusion with "Volume": you have one particular section of Anna Karenina labeled as Volume II on the title page within the text, referred to as volume viii in the index and is in physically the fourth book on the shelf (e.g. in the call number it is labeled v4). MarkLSteadman (talk) 20:40, 9 April 2023 (UTC)
@Beeswaxcandle: These sets are by another translator, Leo Wiener, publisher by Estes, called The Complete Works of Count Tolstoy, while this one is by Nathan Haskell Dole, published by Crowell, with a different title. And there is a third complete collection translated by Aylmer Maude published by C. Scribner's Sons. Yann (talk) 21:10, 9 April 2023 (UTC)
So, which one should be completed first? Which is the most important/significant? Decide on one, sort out the scans and bibliographic mess, get it fully proofread. Then we can work out which of the other sets needs to be dealt with after that. Beeswaxcandle (talk) 22:54, 9 April 2023 (UTC)
From what I can tell, the Maude complete translation is often considered "definitive" but it started with OUP in 1928 and finished in 1937 so likely is covered by URAA restoration, and with the Garnett translations is the more commonly used today, the Wiener is considered more scholarly but less literary (he was not a native English speaker), while the Hapgood/Dole has issues with fidelity due to censorship (even if Hapgood was "authorized" and was friends with Tolstoy). MarkLSteadman (talk) 02:13, 10 April 2023 (UTC)
The Maude translations have been discussed on WS:CV several times and been found to still be copyrighted. And while Tolstoy definitely isn't my field, my understanding based on those discussions concurs with yours: the Maude translations are considered the "best" ones. Xover (talk) 05:04, 10 April 2023 (UTC)
This is not really true: Maude's translation was published in 1902, so it is in the public domain. Now the Tolstoy Centenary Edition contain a new introduction in each volume (see [20]), so this is still be under a copyright. Otherwise, it is only a republication of a public domain work. Yann (talk) 09:03, 10 April 2023 (UTC)
My apologies, I phrased myself poorly. Not my field, as mentioned, so I pointed at the Maud translation based on what Mark wrote. What I do know is that (texts from) one of the main translations have been up for discussion on multiple occasions and found to be still in copyright. Which one of them that was I cannot recall. Xover (talk) 09:21, 10 April 2023 (UTC)
Many of Maude's translations date from the 1910s to early 1920s as well. While much of the OUP editions may be reissues someone needs to actually sort through what is new and what is reissues. It is also possible we should just start by proofreading Maude's works separately and start hunting down the scans for say Kreutzer, AK and W&P. MarkLSteadman (talk) 13:40, 10 April 2023 (UTC)
Just for reference, there's a deletion discussion for one of these at WS:PD#The complete works of Count Tolstoy that is also stalled due to confusion over what's what. If someone with an interest in Tolstoy comes up with the "definitive" list of what scans we should prefer for the different major Complete editions I am happy to take care of page moves etc. Right now it's complete chaos, and whoever has sufficient interest to do the works gets to decide. @Uzume, @R. J. Mathar: you have both expressed interest in this in the past so here's a courtesy "FYI" ping. --Xover (talk) 05:00, 10 April 2023 (UTC)
Yes, I have asked there one and half month ago, and didn't get any answer. AFAICT, the PDF files and indexes should be discarded. Yann (talk) 07:17, 10 April 2023 (UTC)
I'm not sure I understand what you're asking here. Are you looking for advice on how to get a definitive list of volumes in this edition? Or how to find scans for them? Something else?At the risk of telling you what you already know: For scans it's trawling through the terrible metadata at the Internet Archive, then whatever HathiTrust chooses to make public, and then the crapshoot that is Google Books. There are of course archives and libraries that offer scans through their own services, but these are a bit case-by-case and I rarely go looking for them. Some bigger institutions also offer to scan things for you on request (for a fee, of course). Xover (talk) 09:27, 10 April 2023 (UTC)
Now based on this discussion and examination of all available scans, I would drop this edition (The Complete Works of Lyof N. Tolstoï): the quality of the scans is not equal, and I can't find a ToC of the whole, so we are not even sure it is complete. The Novels and Other Works of Lyof N. Tolstoï contains a complete set of 22 volumes translated by Maude, so it should be the preferred edition. Yann (talk) 09:47, 10 April 2023 (UTC)
My understanding of the publication history is roughly this.
1880s: Hapgood and Dole produce individually published translations for Thomas Crowell and Company. (e.g. AK in 1886 [21] and W & P in 1889 [22]), Life in 1888 [23]). If you look in the last one you will see advertising for many of these in the beginning.
1890s L and A Maude start their effort at translation. A. Maude includes some of the more political works. E.g. What is Art? [24] (Crowell), L. Maude translates Resurrection [25]
1898 to 1902: Thomas Crowell apparently decides to commission Dole to edit and provide new introductions etc, for the earlier translations, makes new plates, adds illustrations etc. to make a uniform set. This is published in 12 volumes by Crowell as The Complete Works of Lyof N. Tolstoï and in 22 volumes by Scribner as Novels and Others Works of Lyof N. Tolstoï. This contains some of Maude's work as well,
1902 or so, Grant Richard starts publishing Maude's: Essays and Letters [26], Plays [27], Tolstoy and His Problems [28], etc. These are labeled as Revised Edition of the Works of Leo Tolstoy. These come out in Funk and Wagnalls in the US slightly later.
The Dole and Hapgood are reissued with what appears to be variations in the volumes (e.g. from 22 to 24)
Maudes continue translating solo works (e.g. the Kreutzer Sonata in 1924, [29], AK in 1918, W&P in 1922) until in 1928 OUP kicks off the centenary edition.
In parallel there is the Leo Wiener complete translation (which is called The Complete Works of Count Tolstoy) in 24 to 28 volumes (I haven't dug into detail on this version).This was uploaded to WS in both pdf and djvu and is the subject of the deletion proposal and sorting out which version to keep, moving pages between the two etc. MarkLSteadman (talk) 13:23, 10 April 2023 (UTC)
@Yann: I am sure you noticed but I recently cut through the pagelists of {{The Novels and Other Works of Lyof N. Tolstoï}}. In so doing I have noticed a few things I thought I would bring to your attention (since you seem to be the commons uploader). Volume 7 appears to have no title page so I suspect it might be missing pages. I noticed you uploaded a new version of volume 8 at 749569023 so I fixed the pagelist at 13133109. Volumes 12 and 18 appear to have frontispiece captions without frontispiece images so I suspect they might be missing pages. Also it would be nice if you did not make edits like 13132826 which kill the page numbers (even though such might make navigation at the index pages slightly more interesting/useful). —Uzume (talk) 14:59, 12 April 2023 (UTC)
@Uzume: Thanks for your message. I reconstructed the whole set from several sources. OK, I will check if pages are missing. Mentioned the ToC in the pagelist is useful. I don't find mentioning the covers useful, as they are not corrected and not transcluded. Yann (talk) 15:41, 12 April 2023 (UTC)
@Yann: I agree the ToC in the pagelist is useful on the index pages but the page numbers are also useful when the pages are later transcluded (not to mention that can be used when constructing the headers too). This is why I don't think killing the page numbers to get ToC labels on the index pages is the right thing. We should only provide our own "meaningful" page numbers when there are no actual page numbers from the original work (be it table of contents or covers, etc.). If you really want to label the ToC pages on the index pages, you can do that by breaking the pagelist into multiple parts with labels without changing the actual page numbers. —Uzume (talk) 15:53, 12 April 2023 (UTC)
@Yann: I do not see how that is the same result at all. Help:Index pages#pagelist specifies examples like: Index:A Dictionary of Music and Musicians vol 4.djvu and Index:Tracts for the Times Vol 1.djvu. The former shows (among others) how to separate pages into a group named "Index". So it is possible to retain (the roman numerated) page numbering while also labeling them in sections (e.g., entitled "Contents" or similar, etc.). I admit it might be a tad cumbersome to construct but it is possible. Then the pages will have their numbering when transcluded, during header creation, etc. and will also be easy to see as part of the table of contents on the index page as well. —Uzume (talk) 22:01, 12 April 2023 (UTC)
@Uzume: Headers are not transcluded, so it has zero incidence of the final document. I primarily do what is important for the readability of the work, and then for the facility of the proofreading. What you suggest require extract work without giving any advantages. You can do that for the works you proofread yourself. Regards, Yann (talk) 22:18, 12 April 2023 (UTC)
@Yann: Exactly! Help:Index pages states: "Whenever in doubt, unique labels are always preferred over the re-use of previously assigned labels." If you change all the ToC pages so they have the same page number "ToC" then they will all be confusingly labelled the same when transcluded via <pages ... />. If you must change the page numbers at least make them unique like ToCv, ToCvi, ToCvii, etc. or ToC1, ToC2, ToC3, etc. —Uzume (talk) 22:27, 12 April 2023 (UTC)
Several works need to be moved to make place for translations:
I am noticing that there is a right-hand gutter appearing throughout the Mainspace now: A blank buffer zone on the right-hand side of the page. This does not include the header, but does include the text and content of the page.
I think that this is skin dependent as I don't see it when logged in. But it is there when not. IIRC, the latest Vector skin has space for "stuff" in the right hand side of pages. Beeswaxcandle (talk) 09:39, 12 April 2023 (UTC)
Hmm. Vector 2022 does indeed squish all content into a narrow column in the middle, as well as do some very weird things with the tools sidebar and tables of contents (the built-in MW one). However, Vector 2022 is not (yet) the default on Wikisource so unless Petey selected it explicitly it shouldn't be what's affecting this. Much more likely to be the dynamic layouts (although those have not changed recently) or some other Gadget or user script, or even a local browser issue. Xover (talk) 11:34, 12 April 2023 (UTC)
I am indeed not using Vector 2022, and I have checked pages with and without applied layouts. I am seeing this effect only where a Layout has not been applied; pages that I've looked at look fine that have layouts. I have also not updated a browser recently, or changed anything else at my end. I am not sure when this phenomenon started for me, since I haven't been doing much editing here for several months. I did notice this behavior last week. --EncycloPetey (talk) 14:00, 12 April 2023 (UTC)
@EncycloPetey: are you able to identify the named element style of the gutter through the browser console? That may enable us to explore what may have happened in through the mediawiki development space. — billinghurstsDrewth23:36, 12 April 2023 (UTC)
I don't know how to do that. That is, I can pull up the code for an affected page, but don't know what I'm looking for. --EncycloPetey (talk) 23:48, 12 April 2023 (UTC)
@EncycloPetey: Depending on your browser you may be able to right-click inside the gutter and then choose "Inspect element" from the context menu. With any luck that'll open the browser's web development tools to a view of the web page's element structure (it's not the actual source of the page, but a rendered view of it). If we're lucky you'll have an element highlighted (a <div> most likely) that has an id= or class= attribute that we can use to identify it, and thus try to figure out where it comes from. But if you're not somewhat acquainted with digging through the DOM tree already I don't think you should expend too much energy pursuing this path. It's a pretty high geek factor kind of thing.BTW, two things you should always try (and report the results of) in these kinds of situations because it often helps narrow down what's causing the issue:1. use different web browsers (Chrome, Firefox, Safari, Opera, etc.) and check whether the problem is visible in all of them. Testing in both Firefox and Chrome is especially useful as the two do not share a rendering engine (Internet Explorer, Opera, and a number of other browsers use Chrome's rendering engine).2. Test it while logged out of Wikisource. When you are logged in your Special:Preferences affect a lot of things, such as the skin, and your common.js/common.css and other personal scripts and stylesheets get loaded. When you are logged out you get all default everything, so we can eliminate your personal stuff as a cause.(Oh, and the third thing you'll always be asked for is a specific page where you see the problem and your friendly neighbourhood tech geek can try to reproduce it, but since you already provided that I won't repeat it.) Xover (talk) 05:17, 13 April 2023 (UTC)
Checking Chrome, I did not see the issue, whether logged in or out, but then double-checking in Firefox (my usual browser), the issue is no longer present today. So I have no idea what the problem was, but it seems to have disappeared (at least for now). --EncycloPetey (talk) 15:08, 13 April 2023 (UTC)
I think that it is of some value in doing a first level CONSOLE inspection FAQ as I have found it helpful in working through CSS issues for Index:….css understanding. If someone can identify the issue that way, it does help in digging quicker IMNSHO. Not hi-tech, something quick and dirty. — billinghurstsDrewth07:37, 13 April 2023 (UTC)
As a comment, you can quickly move between respective views by appending to the url the text ?useskin=skinname, eg. ?useskin=monobook?useskin=vector?useskin=vector-2022?useskin=timeless — billinghurstsDrewth07:43, 13 April 2023 (UTC)
Poll regarding April 2023 Wikisource Community meeting
Hello fellow Wikisource enthusiasts!
We will be organizing this month’s Wikisource Community meeting in the last week of April and we need your help to decide on a time and date that works best for the most number of people. Kindly share your availabilities at the wudele link below:
Why are the possible times always the same? It's always 0700-1000 UTC, which is extremely unfriendly to the Americas. Can't you switch it up month-to-month? — Dcsohl(talk) (contribs)17:28, 14 April 2023 (UTC)
Not very practical for Europeans either. I would only be available in the evening, i.e. 1600 to 2100 UTC. Yann (talk) 18:29, 14 April 2023 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
At Wikimedia Commons, some thumbnails have not been getting replaced correctly after a new version of the image is uploaded. This should be fixed later this week. [33][34]
For the last few weeks, some external tools had inconsistent problems with logging-in with OAuth. This has now been fixed. [35]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 18 April. It will be on non-Wikipedia wikis and some Wikipedias from 19 April. It will be on all wikis from 20 April (calendar).
At Wikimedia Commons, some thumbnails have not been getting replaced correctly after a new version of the image is uploaded. This should be fixed later this week.This will fix some particularly annoying problems we've had with page images not updating after a PDF or DjVu has been reuploaded. There are still bugs lurking with thumbnail generation that can bite us, but the fix mentioned there should eliminate a big chunk of them. If all goes to plan it should roll out for us some time this afternoon UTC. After the fix has gone live, using the "Purge" function on the File: description page on Commons should more consistently force generation of new "thumbnails" (page images). Xover (talk) 06:18, 19 April 2023 (UTC)
Ideally, you need an image, and to upload that image to wherever the scan of the book is housed. When the cover is part of the published work, and is going to be transcluded in the work, then it is part of the work, and counts for validation. However, if the "cover" is a library binding, is blank, or is a repeat of minimal information repeated on the title page (or half-title) and is thus marked as "Not transcluded", then it isn't part of the work. In this case, the cover indicates the series of the book, and contains an image, and isn't marked "not transcluded", so it is part of the work and counts toward/against completion of validation. --EncycloPetey (talk) 19:37, 17 April 2023 (UTC)
OK, I added the cover, but after thinking about it, this is probably still under a copyright on the cover. This book was published in 1987, and this design is clearly above threshold of originality for simple logos, even in USA, not even talking about UK. Idem for Index:Agatha Christie-The Murder on the Links.djvu. Yann (talk) 19:07, 19 April 2023 (UTC)
It depends on whether the cover design was copyrighted. It needs to be placed under a copyright, just as the text does, and for Murder on the Links, the only copyright date is 1950, and that is only for the text. A similar situation came up for The Case of Charles Dexter Ward, where no copyright for the cover was claimed, so the cover is in PD because of the mistake. You may need to do some research if you think either Christie cover is under copyright. If they are copyrighted, then we would need to redact the cover from the scan. --EncycloPetey (talk) 19:16, 19 April 2023 (UTC)
This is much older. A copyright notice was needed at this time. Post 1977 publications do not need a copyright notice. Yann (talk) 19:22, 19 April 2023 (UTC)
I don't have first-hand knowledge, but I believe the last time I saw it discussed the Index sort key was said to be non-functional. Xover (talk) 05:13, 20 April 2023 (UTC)
@Yann: Probably a good idea to drop something at the talk page of Module:Proofreadpage index template, which is where the sort stuff seems to now hang. Once it disappeared into that code, it falls out of my ability to fix. If we cannot figure it out there, then a message at m:Tech can sometimes elicit some knowledgeable people. — billinghurstsDrewth05:46, 20 April 2023 (UTC)
The module already outputs the requisite magic word, so it's presumably ProofreadPage or MediaWiki that's not picking it up. And—again based on vague recollection of others' discussions—this was a problem long before the Lua-ification of the template. That being said I'll try to do some digging and see whether I can figure out what's going on. Xover (talk) 09:54, 20 April 2023 (UTC)
Ah, yes, this is T219015. Known broken, probably since the dawn of time, and unlikely to be fixed any time soon since the backlog of stuff needed in PRP exceeds available developer hours by a factor rapidly approaching the heat death of the universe. --Xover (talk) 10:09, 20 April 2023 (UTC)
Milestone.
After some effort, the number of entries on Special:LintErrors of a 'structural' nature has been reduced in content namespaces. Does anyone want to take on the bulk of the remaining ones (mostly mismatched bold and italic formatting?) ShakespeareFan00 (talk) 21:48, 17 April 2023 (UTC)
The Wikimedia Foundation tests the switch between its first and secondary data centers. This will make sure that Wikipedia and the other Wikimedia wikis can stay online even after a disaster. To make sure everything is working, the Wikimedia Technology department needs to do a planned test. This test will show if they can reliably switch from one data centre to the other. It requires many teams to prepare for the test and to be available to fix any unexpected problems.
All traffic will switch on 26 April. The test will start at 14:00 UTC.
Unfortunately, because of some limitations in MediaWiki, all editing must stop while the switch is made. We apologize for this disruption, and we are working to minimize it in the future.
You will be able to read, but not edit, all wikis for a short period of time.
You will not be able to edit for up to an hour on Wednesday 26 April 2023.
If you try to edit or save during these times, you will see an error message. We hope that no edits will be lost during these minutes, but we can't guarantee it. If you see the error message, then please wait until everything is back to normal. Then you should be able to save your edit. But, we recommend that you make a copy of your changes first, just in case.
Other effects:
Background jobs will be slower and some may be dropped. Red links might not be updated as quickly as normal. If you create an article that is already linked somewhere else, the link will stay red longer than usual. Some long-running scripts will have to be stopped.
We expect the code deployments to happen as any other week. However, some case-by-case code freezes could punctually happen if the operation require them afterwards.
This project may be postponed if necessary. You can read the schedule at wikitech.wikimedia.org. Any changes will be announced in the schedule. There will be more notifications about this. A banner will be displayed on all wikis 30 minutes before this operation happens. Please share this information with your community.
The first half of the meeting will be focused on non-technical updates and conversations like events, conferences, proofread-a-thons and collaborations. The second half will be focused on technical updates and conversations, such as talking about major challenges faced by Wikisource communities, similar to the ones conducted in previous Community meetings.
There are going to be updates about Transkribus and we will be sharing more information during the meeting.
If you are interested in joining the meeting, kindly leave a message on swilson@wikimedia.org' and we will add you to the calendar invite.
Meanwhile, feel free to check out the page on Meta-wiki and suggest any other topics for the agenda.
Sorry, not possible for me at this time. During the week, I am only available in the evening (1600 to 2100 UTC). Yann (talk) 14:06, 24 April 2023 (UTC)
Certainly not in that format. We wouldn't list the author's name on every work, because it's on the Author page for that person. We put the date after the title of the work, and wouldn't abbreviate all the journal names. It's also incorrect to link the title of a work off-site. That's at a minimum. --EncycloPetey (talk) 15:23, 25 April 2023 (UTC)
I know mostly the German Wikimedia projects, therefore I am not really sure what is usual here, sorry. In the German projects, the Wikipedia articles give only the most important works of a person and the less central publications like smaller articles are mentioned, if anywhere, only at Wikisource (at least if they are in the public Domain and therefore a possible future addition to Wikisource). I added the exhaustive list to Wikisource because I deleted most of the entries in the bibliography of the article in the German wikiped But if that doesn't fit the rules in the English Wikisource, you can delete it if you want to. Best regards, DerMaxdorfer (talk) 16:09, 25 April 2023 (UTC)ia.
It's the formatting that is the main problem. Look, for example at Author:Henry David Thoreau or Author:Howard Phillips Lovecraft. Wikisource does not use the citation format of the Wikipedias, because all the listed works are by that author, so naming the author over and over for every work is pointless. Author pages exist primarily to list (and link) the works Wikisource hosts, with clearly marked {{ext scan link}} for links to external scans at the end of the listing. External links should never appear on the work's title. --EncycloPetey (talk) 16:23, 25 April 2023 (UTC)
OK, thanks for this tool, it is useful. There is no indication how to add this to one's JS subpage.
IMO line breaks do not matter, as they have no impact on the transclusion. Sometimes keeping the line breaks makes proofreading easier (not the case here). Yann (talk) 11:35, 26 April 2023 (UTC)
As for the necessity of removing the line breaks: it is usually quite strongly demanded that they are removed before a work is marked as proofread (or even validated). Others may give more reasons, I can only say that line breaks make problems when they interfere with various templates or with links. This problem may arise especially when somebody wants to add such templates or links to the work later, e. g. using a bot. They also cause lint errors when they interfere e.g. with italics. --Jan Kameníček (talk) 13:29, 26 April 2023 (UTC)
Yes, I know all that. In most texts that I validated, line breaks were kept, so I didn't remove them. But I am OK with removing them by default. Yann (talk) 18:04, 26 April 2023 (UTC)
All I can say is that there are circumstances where line breaks do cause issues in mediawiki, though usually in conjunction with other code; it all depends on the MW interpretation. Sometimes I have seen the issue appear several pages later in the transclusion due to an earlier piece of code. Please ensure that hyphenation is removed. — billinghurstsDrewth00:31, 27 April 2023 (UTC)
I would do it that way, or at least capitalize the first non-article word (first word that isn't a, an, or the), but sentence case is an acceptable pagename format on Wikisource. See Wikisource:Style guide, under Page titles, #1. Most websites and databases use sentence case for journal article titles, and many libraries do the same for books. --EncycloPetey (talk) 21:35, 26 April 2023 (UTC)
Depends on whose style manual you are following. Here we typically follow the contributor's wishes as both are accepted. Modern cataloguers typically do sentence case, where cataloguers of the 19th century did that more title case. To us it really doesn't matter which, we would simple create redirects from one form to the other as references in works, and listings on author pages could be either. — billinghurstsDrewth00:21, 27 April 2023 (UTC)
These are deprecated, not obsolete. They are not and will never be problematic. Browsers will interpret these appropriately until the day the internet freezes over. These should not be anybody's priorities. — billinghurstsDrewth11:49, 30 April 2023 (UTC)
Looking at Harper's, it is a bit of a mess. All the subpages are under a number of different hierarchies, and we should either amalgamate them to one, or set up root pages to manage each hierarchy separately. I am equivocal on both choices as I don't know (nor want to) much about the publication itself.
The current page itself, could probably sit as a portal page [[Portal:Harper's]] and take the superfluous information that has been gathered, plus list available scans and point to the one or multiple root pages for the work.
a) Is anyone (@Languageseeker, @Tylopous) officially or unofficially running the monthly challenge any more (just curious, no pressure)?
b) Does someone know why the MC page counter seems to be broken?
c) Is the monthly challenge category still supposed to be added to works, when rolling over the MC (thanks @TE(æ)A,ea. for adding some of these, although not for the current month?)
d) Many people still seem to be making MC nominations, but I am not sure if someone is planning to respond to/action any of them. I mean, can't any user update the monthly challenge module?
e) Although I realise it is not directly monthly challenge related, I sometimes notice people remove "(transcription project)" links from author pages. Is there a policy on when removing this is preferred (e.g. post proofread+transclusion vs. post validation)? Personally, I think the earlier the better, for the sake of cleaner author pages, but I am often hesistant to remove them if they are also to incentivize new users joining/validating texts.
To answer e), we remove the transcription links from Author pages once transcription is complete, which typically means fully Proofread at least once. Those links on an Author page are always meant to be temporary. However, removal of such links from the Portal namespace is a matter for personal discretion, as the Portal namespace is newer and meant to be more flexible. --EncycloPetey (talk) 02:40, 15 April 2023 (UTC)
I feel like some of these questions are owing to the MC’s lack of a clear power structure. But to respond:
a) I don’t think that anyone was running it officially, although Languageseeker started it; it’s really just who chooses from among the nominated works.
b) The page counter is run by InductiveBot, which has been broken for a few days. I asked Inductiveload, but he hasn’t been around for a while, eiher.
c) The MC categories have not been automatically updated since last October or November; I have been adding the categories on and off for a while, and will probably get to some more recent months soon.
d) The practice has been, so far as I know, if there are not a substantial number of works, to include all works to which objections were not made.
e) Different people have different practices. The most common practice is to remove it after validation and transclusion, although some people choose to remove it at an earlier date. Even then, they can be left for a while because of forgetfulness or laziness.
Regarding e) I would keep these until a work is validated, at least. I don't feel removing them makes the author pages "cleaner". Some pages of prolific authors are a mess even without them (e.g. Leo Tolstoy). Yann (talk) 16:43, 16 April 2023 (UTC)
If validation went faster / sooner, I might agree. But most works that get proofread will not get validated for years, if ever. The exception being works in the PotM or MC. The scan can be accessed using the "Source" tab, so once it's proofread, we might as well remove the scan link. They aren't drawing editors to validate, and limiting scan links to just works that are not fully transcribed is a more useful visual cue for people looking for items. That is, seeing a scan is a visual cue that the work is incomplete, which saves the reader from having to go into the work to determine that fact. --EncycloPetey (talk) 17:41, 16 April 2023 (UTC)
Personally, I remove the link as soon as I've transcluded the preliminary pages and the link to the work goes blue. The "source" tab is sufficient, IMO. Beeswaxcandle (talk) 05:23, 17 April 2023 (UTC)
@Beeswaxcandle, @EncycloPetey, @Yann e) Given there doesn't seem to be a clearly defined policy, I will lean towards removing post proofread+transclude.
Hi, first of all, I'm glad that there is still lots of activity in the Monthly Challenge. I once said I'd be happy to help running the challenge if I have the time; during the last few months, I didn't have much time.
To points (b) and (c): My understanding always was that the MC categories were necessary to make the page counter work. However, starting about the end of last year, the page counter seemed to be working even without the categories. For this reason I stopped participating in the maintenance of these categories.
As for point (d), yes, in principle any user can update the Monthly Challenge module. Nevertheless I would encourage every MC participant to take part in discussions on the nominations page.
(e) I usually remove the transcription project link on an author page after a work has appeared in the New Texts box.--Tylopous (talk) 12:28, 20 April 2023 (UTC)
@Tylopous All good about helping with the MC, I was just curious what was happening. Also happy not to add categories if they aren't necessary. Thanks, TeysaKarlov (talk) 20:20, 23 April 2023 (UTC)
Tech News: 2023-17
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Due to security issues with the Graph extension, graphs have been disabled in all Wikimedia projects. Wikimedia Foundation teams are working to respond to these vulnerabilities. [37]
For a few days, it was not possible to save some kinds of edits on the mobile version of a wiki. This has been fixed. [38][39][40]
Changes later this week
All wikis will be read-only for a few minutes on April 26. This is planned for 14:00 UTC. [41]
The new version of MediaWiki will be on test wikis and MediaWiki.org from 25 April. It will be on non-Wikipedia wikis and some Wikipedias from 26 April. It will be on all wikis from 27 April (calendar).
There's only about 2100 of them, which means that if 21 dedicated contributors did a 25 of them a day it could be cleared in a week or so. 25 minor edits is not a big ask I think. ShakespeareFan00 (talk) 16:46, 25 April 2023 (UTC)
How do we determine what kind of "LintError" is present or identify its location? Some of these pages have a lot of formatting. --EncycloPetey (talk) 16:59, 25 April 2023 (UTC)
Most of the time, it's a case of carefully comparing the italicisation or emboldening against the Page scan ( as most entries on that list have scans.)
I am going to try to revive Wikisource:Featured text candidates which has been inactive for a couple of years. Any help appreciated. Especially checks of nominated pages are needed, as well as new nominations of validated quality works. My idea is that nominated works should be not only well proofread, but should also be of special value in some way. -- Jan Kameníček (talk) 20:15, 25 April 2023 (UTC)
Almost any work listed at Portal:American literature has some special value. However, more than a few are not scan-backed. And of those that have a scan, many are not fully proofread. Of of those that have been proofread, few have been validated. But it does at least provide a list of potential works from American literature to target for backing by a scan, proofreading, and validating, to get nominations.
I'd like to see similarly selective listings appear for other national literatures, especially Irish, Scottish, Welsh, Canada, Australia, New Zealand, and South Africa. A selective literature portal is useful precisely because it is selective, and not exhaustive, presenting the works typically included in literature survey courses rather than everything. --EncycloPetey (talk) 20:37, 25 April 2023 (UTC)
Seeking volunteers for the next step in the Universal Code of Conduct process
As follow-up to the message about the Universal Code of Conduct Enforcement Guidelines by Wikimedia Foundation Board of Trustees Vice Chair, Shani Evenstein Sigalov, I am reaching out about the next steps. I want to bring your attention to the next stage of the Universal Code of Conduct process, which is forming a building committee for the Universal Code of Conduct Coordinating Committee (U4C). I invite community members with experience and deep interest in community health and governance to nominate themselves to be part of the U4C building committee, which needs people who are:
Community members in good standing
Knowledgeable about movement community processes, such as, but not limited to, policy drafting, participatory decision making, and application of existing rules and policies on Wikimedia projects
Aware and appreciative of the diversity of the movement, such as, but not limited to, languages spoken, identity, geography, and project type
Committed to participate for the entire U4C Building Committee period from mid-May - December 2023
Comfortable with engaging in difficult, but productive conversations
Confidently able to communicate in English
The Building Committee shall consist of volunteer community members, affiliate board or staff, and Wikimedia Foundation staff.
The Universal Code of Conduct has been a process strengthened by the skills and knowledge of the community and I look forward to what the U4C Building Committee creates. If you are interested in joining the Building Committee, please either sign up on the Meta-Wiki page, or contact ucocprojectwikimedia.org by May 12, 2023. Read more on Meta-Wiki.
I would like to suggest to our community members that people's user pages, and somewhat their user talk pages, are having other users too busily editing those pages for no clear reason. If a user has an error,[1] a red link, or whatever on their user page which has no flow on impact, that is usually only the business of that user. If one is concerned about that error/red link/whatever, then the best cause of action is to mention it the person on their user talk page. If they fix it, great; if they don't, so be it, move on. Our presentation pages are those that are within the default for the search engines as defined by content in the respective namespace.
I don't think that we have to have policy or written direction about this, it is simply the courtesy to let people have and manage their pages in those aligned namespace.
↑in user ns, and only impacting user: ns, no flow on to the community
I agree with the general principle of leaving user pages alone, and that for active users the first stop is a message on their talk page, both as a courtesy. However, I think that for straightforward technical fixes (the kind you'd do with a bot, say) we shouldn't fuss so much about User: namespace pages being holy. For example, the syntaxhighlight extension switched from the tag name "source" to "syntaxhighlight" several years ago, and is now adding tracking categories for uses of the old name prior to dropping support for it. Running a bot through to replace the tag name should be entirely uncontroversial: this is a public and collaborative wiki, not your social media profile, so your User: space is only personal to the extent that it serves the project (cf. why you can't put just anything there: copyvios, advertising, attack pages, etc. are no more permitted in User: space than anywhere else). It's a matter of not interfering with a user's User: pages, more than not ever editing them.And for most of the maintenance and tracking categories it is necessary to remove the backlogs of trivial stuff so that real issues can be detected and fixed, and so that newly created issues can be detected. Cf. the CommonsDelinker issue and Category:Pages with missing files. The LintErrors fixes are kind of borderland in that there are so many of them and they usually have no visible effect currently, and are there mostly to prepare the ground for changes to MediaWiki's parser in the future. But as with most things I think we should not get religious on either side of this issue: User: namespace pages should be approached with respect and courtesy for the contributor, but are in no way sacrosanct and exempt from necessary maintenance; and maintenance and tracking cats should be processed and worked on, but not at any cost and not all issues are equally important.Similarly with /Archive pages: in general these should not be edited, and if they must be edited then the change should change the historic page as little as possible. But if whatever maintenance issue it is is non-trivial then, yes, we do need to edit even /Archive pages. Its original state and the change made to it will be visible in the revision history so it's not that big of a deal. "Avoid it when possible, but do it if you must." Xover (talk) 07:08, 27 April 2023 (UTC)
If we are looking to run uncontroversial clean-ups, then we should have that as a general open point of reference/conversation/background. This then also has a diff that can be part of any summary. Noting that I did focus on user pages, and I did qualify it for being no flow-on to the community. And yes it is a collaborative wiki, however, user ns is less collaborative, it is the one space that should have a really good reason to edit; and that is not what I am addressing. There are enough public facing and important issues that need fixing, so keeping out of user ns is desirable where possible. — billinghurstsDrewth07:31, 27 April 2023 (UTC)
Endorse proposal - It would be nice to focus contributor attention on actual "content". Is there a config option so that LintErrors isn't crowded by non-"content" pages? ShakespeareFan00 (talk) 09:39, 27 April 2023 (UTC)
In respect of one very specfic use case, do we have a user script that could put an additional column on Special:LintErrors pages when Page: namespace pages are listed that shows thier status as Proofread/validated etc? ShakespeareFan00 (talk) 06:52, 4 May 2023 (UTC)
Do we want CommonsDelinker bot on enWS?
CommonsDelinker is a bot (running without bot flag and without bot approval, incidentally, which we should fix either way) that checks file deletion logs on Commons and then removes now-broken references to that file on other projects. The idea is to prevent red links and broken images visible to readers. On Wikipedias this is a great idea. However, on Wikisource this frequently happens in Page: namespace pages, like this, where they will usually go unnoticed. Our page editing pattern is such that any given Page: namespace page will only have one, or at most two, watchers—whereas a Wikipedia article is likely to have many—and in a large number of cases those contributors are not even active any more.
When this happens we will likely never discover the problem. Whereas if the image reference gets broken (points at a non-existent file) it will show up in Category:Pages with missing files where we can track them and fix them.
I'd therefore like to solicit some input on whether it makes sense to run CommonsDelinker here, or if we should disable it.
@Xover: We have an AF that checks and reports every edit by CDL. I then check every edit it makes, and follow-up at Commons as required; and I have a wide range of responses and actions that I undertake. The bot is not acting as a bot with rights, and it is behaving exactly as we want it to do and the visibility that we require. The fact is that we want to see what Commons is doing to our files, and the CDL is the best way to see, as there is no ready way to see files nominated for deletion, so we are always acting reactively. Can I say that it is far worse where Commons deletes the pdf/djvu file as CDL cannot remove the Index: or Page: files, so they happen quietly and more lethally. And I am not the bot operator and have never been, though as part of my Commons admin role, I can occasionally manage deletions there which CDL may do its cleanup. — billinghurstsDrewth07:20, 27 April 2023 (UTC)
Sure, but using AF and contribs requires someone to actually watch contribs all the time; and when you get busy IRL they go unnoticed again. It becomes a "bus factor" issue: "how many people have to get hit by a bus before the system breaks down?" Right now I suspect that factor is one. Keeping the breakage visible in the maintenance categories allows a much wider set of contributors to monitor for such issues and work the backlog.(Which reminds me, I should get off my behind and make a list of tracking cats we should try to monitor more actively to fix problems as they occur)Xover (talk) 07:33, 27 April 2023 (UTC)
@Xover: Sure, though missing files is just going to show that they are not held locally, not that they were uploaded and deleted at Commons. There is also no real time ability to note their deletion, and relies on the whim of someone needing to go to the category, then having to know what to do from first principles, then going to Commons, and knowing how to get files undeleted, and the long argument from a position of weakness with an admin. My plan works way better, though yes it does rely on me, so it fails the bus on person test, whereas yours passes the bus test but you have no buses in the neighbourhood on a quiet rural lane, no map, no keys and a long walk to the station.
Also noting that the CDL edits from the abusefilter all have a tag on them to enable review, so you can see them through the tags filter in Special:RC. So what I have done is added some text to identify that it needs review. What we should be doing is creating target pages within Wikisource:Maintenance (or alternatively Wikisource:Maintenance of the Month) that we point to, and describe the process for resolution. — billinghurstsDrewth08:09, 27 April 2023 (UTC)
Oh, yes, indeed: irrespective of this particular issue I whole-heartedly agree we should be better at facilitating collaborative approaches to maintenance such as through Wikisource:Maintenance. Teaching lots of people to fish is bound to be more important than any kind of technical nitty-gritty of how we track the fishing spots. Xover (talk) 08:16, 27 April 2023 (UTC)
Excellent example: here CommonsDelinker removed the image and nobody noticed, so it's been sitting broken for over a year. If we didn't use CommonsDelinker the image would have sat as a redlink (broken image reference) and would have shown up in Category:Pages with missing files. Even if nobody was watching that category at the time, the fact it was broken would have stayed visible even after the edit itself scrolls off of Special:RecentChanges and Special:AbuseLog. Right now we have no idea how many images have been removed by CommonsDelinker that nobody noticed. Xover (talk) 13:23, 28 April 2023 (UTC)
@Xover: We do know what is happening with CDL edits ... Special:Contributions/CommonsDelinker everything is clear. You can also see what has been reverted and what has not by the Tags => Tags: Reverted CommonsDelinker. I will recover and migrate those identified missing images in the next couple of days. Moving them over to here is now such a painful task unfortunately. — billinghurstsDrewth12:00, 30 April 2023 (UTC)
Right. It's not that we can't find out what it's done, it's that the logs-based approach is best suited to 1) realtime monitoring of what's happening now and 2) research into what happened historically. If nobody's watching when the edits happen they'll scroll off the horizon and someone needs to go actively looking for them, unlike a tracking category that we can show in relevant places as Category:Pages with missing files (7). For tracking something that should not get lost but can not necessarily be fixed immediately, a tracking category where items stay permanently until the problem is fixed is less fragile and easier to handle. We can't base stuff like this on a single contributor never getting sick (or fed up, or busy IRL, or ...), and we have enough problems getting patrolling coverage of the critical stuff at Special:RecentChanges without adding more such real-time tasks. I think we should seriously consider whether this particular CommonsDelinker task is the best way to do it, or whether letting image references break and automatically end up in the maintenance category would not be a better approach. Xover (talk) 09:48, 1 May 2023 (UTC)
It is really not helped when it is one of our own community members nominates these for deletion prior to them being repatriated
Sidenote: you can detect what the bot changes by following its contributions. Maybe it's more fiddly than looking at broken images list: pick which is best for you. -- grin✐18:19, 3 May 2023 (UTC)
I will accept whatever you people decide (preferably with voting or by other wide-coverage method); the bot can skip enwikisource, and you'll have broken image links which can be handled individually. --grin✐18:16, 3 May 2023 (UTC)
Keep Yes we want CDL here. No, we don't want to have rename files that are moved. We have been pretty well managing the deletions that have occurred. Come up with a better solution prior to making more work and more holes. We have a solution that is working, yours is not a solution, just a bigger evident problem. — billinghurstsDrewth21:31, 3 May 2023 (UTC)
@Billinghurst: To be clear, when I'm suggesting we stop using CommonsDelinker for deleted images that does not necessarily mean dropping it for moved files. I'm sure grin could disable just the one task for enWS but leave the other one running. Xover (talk) 05:30, 5 May 2023 (UTC)
Fair enough. You're the one putting in the work here, so your voice has weight on this. But let's just say it's not like you're likely to run out of tasks any time soon, and making it practical for others to take some of that load would have been a nice bonus. Xover (talk) 10:13, 5 May 2023 (UTC)
1928 PD (2023) scan project
Looking back at the past few years of Public Domain Day, I think it would be a good idea to have a few scans in store for the change of year. This is especially so because many of the books just entering the public domain won’t have scans on the major scan-hosters because of the up-to-then existing copyright restrictions. I have access through my library to a lot of the books on WorldCat, and would be willing to scan them in before the new year. (It’s a little early now, I know, but I thought it would be better to ask now rather than to push it off until quite late.) I’m thinking of items listed at Wikisource:Requested texts/1928, but I would be especially interested in works where editors promise to proofread the works once they are ready. I would create scans in the coming months, and hold on to them until January 1, 2024, when I would upload them to Wikimedia Commons (or here, if there are other copyrights), so work can begin right away. TE(æ)A,ea. (talk) 02:10, 29 April 2023 (UTC)
I am interested by Satyagraha in South Africa by M. K. Gandhi, and The Mystery of the Blue Train by Agatha Christie. And there are quite a number of short stories by Christie which we don't have yet, but are already in the PD in USA. Yann (talk) 11:35, 29 April 2023 (UTC)
Yann: Satyagraha in South Africa (OCLC 7965175) and The mystery of the blue train (OCLC 315963986, reserve OCLC 1075695763) have been called upon. I’ll also provide updates as the process of getting and scanning the books continues. TE(æ)A,ea. (talk) 12:05, 29 April 2023 (UTC)
Yann: Satyagraha in South Africa has just come in (and The mystery of the blue train should be here soon). It is a revised second edition (from the 1950s), which itself carries no copyright notice (apart from the original, 1928 notice). Is that acceptable? TE(æ)A,ea. (talk) 20:42, 10 May 2023 (UTC)
@TE(æ)A,ea.: It is in the public domain in India. I guest the copyright status in USA depends on the amount of changes from the first edition. Difficult to say without having both. Yann (talk) 20:49, 10 May 2023 (UTC)
@TE(æ)A,ea.: If possible, I would be very interested in books "A forest story" by Josef Kožíšek (here) and "The Magic Flutes" by the same author (here and here). They were published in 1929 but their copyright was not renewed. What do you think, would it be possible to get them? --
Jan Kameníček: A forest story (OCLC 1686335) and The magic flutes (OCLC 3079241) have been asked for. In addition, I would like you to know that I am close to finishing scanning The Net of Faith, and should be able to upload it within twenty-four hours. TE(æ)A,ea. (talk) 14:58, 2 May 2023 (UTC)
Jan Kameníček: A forest story and The magic flutes both came in to-day. I have started scanning A forest story and should have it and The magic flutes scanned in to-morrow. It might take a little longer for me to upload them, however. TE(æ)A,ea. (talk) 20:42, 10 May 2023 (UTC)
I've noticed two issues with {{Block right}} and {{Block right/s}} which I believe should be addressed.
First, on {{Block right}}, I believe the handling of "gutter" is erroneous. For all the other parameters, the relevant CSS is added only if that parameter is present, but for "gutter", the CSS is added only if "offset" is present. It looks to me like a simply copy-and-paste error, but it produces unexpected results. Fixing should be low impact, as it would only affect pages with "gutter" and without "offset", which are currently not rendering as intended anyway.
Second, I think {{Block right/s}} should support all the same parameters as {{Block right}}, but it does not, even though they share the same doc page which implies that they are supported. Fixing again should be low impact, as it is strictly additive and should not change any existing uses of this template (unless they use these parameters in which case I would argue they are not rendering as intended).
I'd like to propose these two issue be fixed (and if this isn't quite the right sort of topic for a "Proposal", my apologies and I can move this down to general discussions). — Dcsohl(talk) (contribs)17:55, 28 April 2023 (UTC)
This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. — Dcsohl(talk) (contribs)16:48, 17 May 2023 (UTC)
@Yann: If the copyright is uncertain, then upload locally, and put notes on the Index_talk: page for things to be checked and moved once certainty is established. Probably can put notes on the portal page too. Easier to upload here and migrate to Commons, than the reverse. — billinghurstsDrewth23:32, 17 May 2023 (UTC)
So I added {{free media}} to a number of license templates to make the files show up in Category:All free media. But it seems that only a few files will show up there (perhaps 1-2k).
So I wondered if there are really thousands of 'illegal' files on wikisource?
Then I noticed Category:Raw page scans for missing images with 16,215 files. I checked a number of random files and they
did not have a license. But they are an extract from files on Commons where there should be a license.
My question is what to do. Should the files be moved to Commons or are they needed locally? Is there an easy way to add the missing license? --MGA73 (talk) 18:29, 29 April 2023 (UTC)
MGA73: I don’t see that requirement here, but I’m not too knowledgeable of the exact rules. Those files have proper licenses through the files from which they are respectively derived. The files in that category were created as part of a process which now no longer creates local images. The files in that category have been gradually replaced with ones uploaded either here or on Wikimedia Commons, as appropriate. After that occurs, the person who created the new image should nominate the one here for deletion; however, that has not happened in a number of cases. I have gone through the category in the past—it used to have over 20,000 files—and will probably go through it again this summer, when I have more time. Thank you for reminding me of the category. As for your last paragraph, besides the review, no work in particular is needed. TE(æ)A,ea. (talk) 18:50, 29 April 2023 (UTC)
The {{free media}} categorization implies that the work is out of copyright worldwide, and for some files and for some licensing templates that is not true. I have reverted your addition of the category to several templates, and other folks here better acquainted with the copyright policies may revert more. Some files housed here in Wikisource should not be moved to Commons because they do not satisfy the requirements for hosting on Commons. The criteria for Commons are more stringent than those applied on Wikisource. In future, if you have a question, it is better to ask before changing a whole array of widely-used templates. --EncycloPetey (talk) 19:11, 29 April 2023 (UTC)
@MGA73: Long story short: these are a non-issue in terms of licensing, and they're a manual backlog we'll finish eventually.The longer version is that these are high-res versions of individual pages that contain illustrations and plates from scanned books. They were bot-uploaded to serve as placeholders until someone got around to properly extract the plate (it's used automatically by {{raw image}}/{{missing image}}). For various reasons we don't use these any more (new ones are not being uploaded), and most of them will show up on Special:UnusedFiles (well, eventually; it only caches 5k at a pop), but a previous discussion determined that we have to evaluate each of them individually (because reasons, I guess). So I chip away at them whenever I have a spare moment, and TE(æ)A,ea. has been checking larger batches, so we're now down from ~25k to ~15k.We host everything on Commons unless it is 1) out of scope for Commons (i.e. local screenshots and stuff) or 2) incompatible with Commons' licensing policy. enWS licensing policy allows works that are PD only in the US, while Commons requires US and the country of origin, so we have a significant number of local files for that reason. The majority have proper license tags, but trying to identify and fix the ones that don't is fairly hopeless until we clear the above backlog. I'm pretty sure we don't have a significant problem with unlicensed local file uploads; it's just an issue of the usual housekeeping. Xover (talk) 19:17, 29 April 2023 (UTC)
After working through a large number of files we are now down to ~4k in Special:UnusedFiles so the 5k cache limit should no longer be a concern. MarkLSteadman (talk) 22:30, 12 May 2023 (UTC)
@MarkLSteadman: Wow! Between you and TE(æ)A,ea. an intractably large job has now become something we can expect to finish in some reasonable time frame. Thank you!The bad news is that we still have 15k+ files sitting in Category:Raw page scans for missing images that are of dubious value, but still require manual intervention. Xover (talk) 10:09, 13 May 2023 (UTC)
TE(æ)A,ea. According to Wikisource:What is Wikisource? then "Wikisource – The Free Library – is a Wikimedia Foundation project". As a Wikimedia Foundation project Wikisource must follow the rules set out out by WMF and that include the wmf:Resolution:Licensing_policy. So it is a requirement that the files have a license. Files without a license should be deleted. On Wikipedia it happens after 7 days. Same should happen here.
It takes a long time to check manually so that is why it was made a global requirement that files have a machine readable license and why the two categories above was created.
User:EncycloPetey I know that not all files can be moved to Commons. But because you reverted my edits I can't easily make a list of files without a license. Wikipedia have w:Template:Free in US media but that does not exist here yet ({{Free in US media}}). Perhaps create that?
You can certainly ask for help identifying problematic files. Someone may be able to assist you with accumulating data. As I say, starting by changing Templates to cause a large number of pages to be re-rendered across the whole of Wikisource is not the best approach. Your edits will cause thousands of pages to be changed, and not just the files you're interested in, but every files on this site with one of those licenses. And as you aren't familiar with the licensing requirements here on Wikisource, how do you propose to help? --EncycloPetey (talk) 19:49, 29 April 2023 (UTC)
@MGA73: I'm sure we have a bunch of files that are not properly tagged, and a subset of those that are incompatibly licensed. Just as we have texts sitting in mainspace that lack license tags and a subset of those that are not compatibly licensed. We are aware of the problem areas and the scope of the problems (nowhere near critical), and we are dealing with them deliberately and in the way we have determined is best for our project. Our approach is very different to enWP (or Commons for that matter), so not employing a 7-day hard automatic delete for lack of license tagging is a deliberate choice (we find it user-hostile and counter-productive). So while I would certainly prefer we got through our backlog faster, we're steadily processing this backlog as we are all our other backlogs and with a priority commensurate with its importance. We're always happy to have more help, but then I might suggest starting with scan-backing some of our ~200k naked texts, or fixing the ones in Category:Index - File to fix. There's also any number of projects in Category:Index Not-Proofread where more help would be very welcome. Xover (talk) 20:17, 29 April 2023 (UTC)
Xover I think that all wikis have more work than they have hands. But copyright should be taken very serious because they represent a legal risk to the project. All files that does not meet the requirements MUST BE DELETED! Check this old diff by Jimbo. He even said that if things did not improve dramatically they would close for local uploads unless for users that have earned the right. So if wikisource host unlicensed text it should be deleted even if you do not like to do so. I think perhaps somewikis keep files for 14 days so I guess that will be okay too. But I do not think that Wikisource can decide to keep known unlicensed text for longer than that. --MGA73 (talk) 20:46, 29 April 2023 (UTC)
I'm sorry, but do you realise you are now implying that we do not take copyright seriously, screaming in all caps, appealing to Jimbo, and making bald assertions about what we "must" do?English Wikisource has a licensing policy that is fully in accordance with the terms of service, and we actively and effectively enforce that policy. We delete blatant copyvio immediately, and generally delete incompatibly licensed matter as soon as a proper determination of copyright status has been made.What we do not do is mechanically delete anything just because some poor newbie hasn't figured out how to add all the right technical gobledygook within some arbitrarily chosen deadline. Or because the file happened to have been uploaded back in the dawn of time before the standards were raised to their current level and we haven't gotten around to fixing it yet. We don't do those things because they are user-hostile and counter-productive, and because our experience shows us that very rarely are these actual copyvios they just lack the right tagging.So while we appreciate anyone wanting to help, this is an area we can manage just fine by ourselves (unless you actually want to roll up your sleeves and help out with the not-so-glamorous task of manually checking everything on Special:UnusedFiles and tagging anything we can safely delete). User:MGA73/Sandbox looks possibly useful as a manageable subset we can check and maybe improve sooner than we otherwise would have gotten to them, but I also have to note that the first ~50 entries (~10%) there I happen to personally know exactly what they are and why they're there (half are newbie uploads that should be on Commons, the rest are temporary files that will be deleted). I'll be very surprised if there is a significant number of actual licensing problems in there. Xover (talk) 22:31, 29 April 2023 (UTC)
MGA73: I agree entirely with Xover, except as to your usefulness. (A quick look at your sandbox page notices many hundreds of works marked with license templates, for example.) I may mention also that engaging in legal threats is most unbecoming of a user, and likely violates one of those “resolutions” to which you like citing; this is especially in the case of a new, inexperienced user such as yourself, who has also recently undertaken the modification of a number of highly-used templates without community consensus. We here are perfectly able to enforce our own policy, and do not need our policy to be redirected by someone from another site. TE(æ)A,ea. (talk) 23:26, 29 April 2023 (UTC)
Xover it usually only takes 1 minute to find files without a license provided the license templates have {{free media}} etc. So perhaps create {{Free in US media}} and add that instead? Then I could make a list to see if there are any files without a license besides those in Category:Raw page scans for missing images.
Everyone. Since the files in Category:Raw page scans for missing images will be deleted and they link to Commons where there is a license then I agree that it is not a big problem that they do not have a formal license template. But unless you are sure that all other files have a valid license I suggest that you add one of the templates above to the license templates. --MGA73 (talk) 19:43, 29 April 2023 (UTC)
TE(æ)A,ea. okay. I do not need those changes if the project can find and fix unlicensed files without my assistance. As written above copyright should be taken very serious. --MGA73 (talk) 20:46, 29 April 2023 (UTC)
I made a list at User:MGA73/Sandbox of files that possible does not have a license. The list may not be as good as it could because I excludes a wide number of categories. Some files are there because the license template does not add the files to a category like "PD in xxx". If you fix the template the files will be removed if I update the list. If you ping me I can make an update. --MGA73 (talk) 21:07, 29 April 2023 (UTC)
@Xover, @TE(æ)A,ea.: I decided to jump to the bottom because I think it makes it easier to read and because it's a follow up on more of the discussions above.
To begin with I would like to clarify that I came here to help out and I do that on many wikis. Usually that works out fine but for some reason things went of track here. I'm sorry for my part of that happening.
You may think I'm a newbie because I do not have many edits here but I have millions of edits related to images on this account and my 2 bots so I do think I know something about copyright.
English is not my native language so that may be a part of the reason things went wrong.
First I read the comment "I don’t see that requirement here, but I’m not too knowledgeable of the exact rules." as a "I'm not sure Wikisource does not have to follow the WMF:Resolution:Licensing_policy". Thats why I tried to explain that Wikisource must follow it because it's a Wikimedia project.
Second I read the comment about not deleting unlicensed text/files and the "nowhere near critical" as "We know we have a lot of copyright issues but we do not think its a big problem". Thats why I linked to the old comment from Jimbo (I did not ping him or call for his help).
However, I do worry a bit when I read "our own policy, and do not need our policy to be redirected by someone from another site". If it is to be read as "We do not care what WMF have written in their policy" then I think we have a major problem. If its just frustration that I point to a problem areas in a non diplomatic way then I can just repeat it was not intented to be that way.
As for the value of the list on User:MGA73/Sandbox I would like to repeat that "Some files are there because the license template does not add the files to a category like "PD in xxx"." So I know some files have a license and as I wrote it can be fixed if you edit the template and make it add the files to a category like all the other license templates. (Perhaps on {{Legislation-SGGov}} add {{PD-EdictGov}}).
Xover, I do not mind helping out and I do that a lot on many Wikis. One of the latest tasks was jawiki where I helped fix 80k files. But when I'm told that "We here are perfectly able to enforce our own policy..." that does not really motivate me to help out. But I have made the list so its now easy to go through the files and enforce the policy as stated.
Last thing. If not allready done you might want to try to make a similar list of unlicensed text in main space etc. --MGA73 (talk) 10:22, 30 April 2023 (UTC)
@MGA73: No worries, and I'm sure we share the blame for not keeping this from going off into the weeds.I'm trying here to concisely address your concerns: 1) we are very much aware of the scope for local autonomy on licensing issues that the WMF affords us, and are operating within those boundaries; 2) our licensing policy, community practices, and enforcement of these does not differ markedly from other projects, and our enforcement is effective; 3) we are confident we do not have any significant problem with copyvio lurking in our backlogs, because we are aware of the issue and have spent quite a bit of effort on cleaning up things that are not properly tagged etc. and the proportion we find that have actual licensing problems (i.e. are copyvios) is miniscule; 4) because we are confident we do not have any "lurking unnoticed copyvio" problem worth speaking of, we do not want to summarily delete things that have deficient tagging as the vast majority of what we deleted would be perfectly valid, in-scope, and properly licensed content that just failed to observe all formalities possibly a decade before those formalities were made explicit.So when we do not consider lack of machine-readable licenses on files as a critical problem it is not because we do not care about appropriate licensing, but because we have ample reason to believe that it is purely a technicality (with possibly a trivial amount of actually problematic files). We do work on cleaning it up, but it competes for priority with all our other maintenance backlogs of the same type. And most of the low-hanging fruit has been picked, so, for example, what lacks tagging in mainspace (which we do track through various means) consists of a lot of stuff that needs detailed research to identify actual copyright status. I've personally spent countless hours on such issues and the vast majority I end up tagging with a compatible license rather than bring to WS:CV. Of those brought to WS:CV a large proportion are saved through the diligent research of other contributors. The content I've found here that is blatant enough that it can be speedied is a trivial amount, and even these require at least some research to verify.So while we do aim to get to an end state where every file and every text we host has an appropriate license tag, we are going about that with slow deliberation (quality over quantity). We also don't feel it likely that we have any significant blind spots here, so while we always welcome more hands to help we don't feel we need any external help pointing up stuff that just lacks tagging or similar (we already know and are tracking it).If you want to help the most effective thing you could do is start going through Special:UnusedFiles, checking that the raw page scan has been replaced with a properly extracted image (cropped, color-corrected, etc.), and then tag it for {{sdelete|A1 transwikied}}. That's our biggest backlog that is generating noise for other cleanup efforts (including getting proper license tagging in place), and it's going slow because it requires manual (human) verification and lots of clicking around on slow-loading pages (MediaWiki's multi-page file support is pretty crappy). Xover (talk) 11:05, 30 April 2023 (UTC)
The reason I wanted to make a list is because in many cases a wiki do not have an easy way to find the files without a license (if they did they would probably have fixed it long ago). If added to a list or put in a category it is very easy to work on the files.
As for the 16k files my thought was that if the files were needed I could perhaps move them all to Commons with FileImporter and fix the missing license etc. with my bot.
It doesn't add a cateory that is a subcategory of Category:Works by license like the other license / government templates, e.g. it should categorize into a subcat of Category:PD-EdictGov. MarkLSteadman (talk) 22:25, 12 May 2023 (UTC)
Done. I see we have quite a bit of cleanup to do in this area, not least is making the license cats hidden, consistent naming, and so forth. But for now it should get the SGGov works out of MGA73's list. Xover (talk) 08:21, 13 May 2023 (UTC)
Updated list again. It removed one third of the files. And yes cleaning up license templates is a nice little project. But I have seen much worse. Some wikis do not have Category:License templates so there it is hard to find the license templates. --MGA73 (talk) 18:21, 13 May 2023 (UTC)
@Liuxinyu970226: Ah. You're arguing that we should permit non-free licenses and fair use content? There have been some community pushes for that occasionally, but mostly for very limited subsets. Given the nature of the project and our content it'll be very rare that we have valid fair use rationales, so even if we permitted it it'd have very limited effect. Personally I think that'd be a very bad idea, for any number of reasons, but fair enough.The reason we have local uploads enabled is that a small but significant fraction of our applicable files cannot be hosted on Commons due to their additional requirement that files must be public domain in the country of origin in addition to being public domain under US law. This typically (but not exclusively) affects books first published in the UK more than 95 years ago, but whose authors died less than 70 years ago. Due to the difference in how the US and most other countries calculate copyright terms there's a gap where a book is public domain under US law but is still in copyright in the source country. In these cases we simply adhere to US law (since that's where the WMF is legally registered), but Commons has adopted the additional requirement that files' copyright must have expired in the country of origin too. There are also some few media files that are out of scope for Commons (project-internal stuff, not mainspace content) and must be hosted locally for that reason. Xover (talk) 09:37, 4 June 2023 (UTC)
As for my original question/comment I think it have been closed with the creation and update of User:MGA73/Sandbox so that it shows the files that do not have a formal license template. I think this section could be archived. --MGA73 (talk) 12:28, 9 July 2023 (UTC)
A maintenance thread
Hi to all. As mentioned in the previous section there has been questions about where are we with maintenance, and with its coordination. We have
which were our attempts to get us on track years ago, and probably have fallen away since.
I know that I have my own maintenance schedule, scripts, and the like that I potter through--on what seems like one monotonous journey of (re)painting a bridge--to the point that my actual transcription editing is few and far between these days. I also know that doing new works and new information provision is always more fun than maintenance.
So
Are there people interested in undertaking maintenance tasks?
What maintenance tasks do you wish to see?
What maintenance tasks would you have an interest in undertaking?
What information and guidance would you expect to see for your tasks?
Do you sort of want to own your task? Or do you just wish to lucky dip tasks that are there?
New one-offs? Or a repeated, regular.
Does anyone have an interest in coordinating/overseeing?
Have good vision on how maintenance has been done elsewhere that would give us a good starting model?
Mulling this one over still (thanks for taking the initiative!), but… Regarding Maintenance of the Month: it'd be a nice thing to revive, but it's also one of those things that would require someone to run the process (light, or more extensive, but someone would need to run it either way), so I think maybe reviving that is not where we first start. Keeping it in mind as a tool in our toolbox for organising efforts would be smart though, especially if someone volunteers to be the one to manage the process. Xover (talk) 10:19, 5 May 2023 (UTC)
This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. Yann (talk) 16:08, 4 April 2023 (UTC)
Done —Beleg Tâl (talk) 15:17, 4 April 2023 (UTC)
This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. Yann (talk) 16:08, 4 April 2023 (UTC) 
The new version of MediaWiki will be on test wikis and MediaWiki.org from 4 April. It will be on non-Wikipedia wikis and some Wikipedias from 5 April. It will be on all wikis from 6 April (calendar).
The deprecated 
Comment Our primary purpose is the creation of versions, pure and clean. If someone wishes to see a version of a time, that is our role.
In the visual editor, it is now possible to edit captions of images in galleries without opening the gallery dialog. This feature request was voted #61 in the 2023 Community Wishlist Survey. [14]
w:WP:SNOWBALL — billinghurst sDrewth 11:44, 30 April 2023 (UTC)
and 
is another possibility. Yann (talk) 21:28, 16 April 2023 (UTC)
wikimedia.org by May 12, 2023. Read more on Meta-Wiki.
Keep Yes we want CDL here. No, we don't want to have rename files that are moved. We have been pretty well managing the deletions that have occurred. Come up with a better solution prior to making more work and more holes. We have a solution that is working, yours is not a solution, just a bigger evident problem. — billinghurst sDrewth 21:31, 3 May 2023 (UTC)
