From Wikisource
Jump to navigation Jump to search
Warning Please do not post any new comments on this page.
This is a discussion archive first created in October 2021, although the comments contained were likely posted before and after this date.
See current discussion or the archives index.

ia-upload needs a kick[edit]

Slim Application Error

A website error has occurred. Sorry for the temporary inconvenience.

been this way for 6+ hours — billinghurst sDrewth 11:40, 3 October 2021 (UTC)[reply]

Tpt says that it is disk full. — billinghurst sDrewth 12:37, 3 October 2021 (UTC)[reply]
Cleaned; working. — billinghurst sDrewth 12:53, 3 October 2021 (UTC)[reply]
This section was archived on a request by: — billinghurst sDrewth 13:57, 3 October 2021 (UTC)[reply]

September Monthly Challenge summary[edit]

Wikisource laurier.svg

The September Monthly Challenge is now complete and the numbers are in: 2606 pages were processed (marked no text, proofread or validated), which is well over the target of 2000. We also had out longest streak ever of days with over 100 pages processed at the start of the month with 9 consecutive days! The following works were fully proofread:

And the following were validated:

Thank you to all contributors and I look forward to October's challenge, which starts today. There are many new works to the challenge, including (but certainly not limited to):

Now that the "nights are drawing in" as my grandmother would say (for those of us in the Northern Hemisphere), there's no need to be distracted by such fripperies as "going outside", "daylight" or "fresh air". Come and join in here! As always, please drop in as you wish and nominations are open for proposals and comment for future works. Inductiveloadtalk/contribs 06:37, 1 October 2021 (UTC)[reply]

Enoch Arden, etc[edit]

@Chrisguise: The title page uses a period after the word "etc", which leads me to believe perhaps Enoch Arden, etc. is the official title. Any objection to me moving the title pages of this work to Enoch Arden, etc.? PseudoSkull (talk) 18:23, 3 October 2021 (UTC)[reply]

Translation:Shulchan Aruch/Orach Chaim/507[edit]

I have just come across Translation:Shulchan Aruch/Orach Chaim/507. I cannot speak Hebrew, but Google translate suggests that there are completely different Hebrew and English texts. What can be done about it? Should it be deleted? --Jan Kameníček (talk) 08:01, 3 October 2021 (UTC)[reply]

Perrsonally I think we should just get rid of all Wikisource translations that aren't scan-backed, because they are impossible to verify in practice; and in particular they make it effectively impossible to verify random changes to them while patrolling recent changes. But so long as we permit this crud to accumulate there is little to be productively done about cases such as Translation:Shulchan Aruch/Orach Chaim/507 (or all of Translation:Shulchan Aruch for that matter: it is all equally unverifiable, and I regularly see arbitrary changes to the very same bits of it since I happen to have it watchlisted after some technical maintenance). I have essentially given up both patrolling and other improvements to non-scan-backed stuff in the Translation: namespace. Xover (talk) 08:09, 3 October 2021 (UTC)[reply]
Incidentally, see also Translation:Manshu, currently listed at WS:PD#Translation:Manshu (with predictable results), which claims to be a translation of some ancient work, but which is really mostly an original analytical/comparative/interpretative work by the contributor and in violation of multiple of our policies. But because it is in the Translation: namespace it seemingly gets a pass on any number of sins (policy violations). We should just go ahead and rename it "ThePolicyFreeZoneAndAnythingGoes:" and be done with it. Xover (talk) 08:15, 3 October 2021 (UTC)[reply]
Hm, that is strange… While a couple of hours ago Google translated the Hebrew text in a completely different way to what the page contains in English (i.e. as if it were a completely different text about something different), now I tried it again and now the meaning of the translations more or less coincides. Looks like Google makes fun of me. --Jan Kameníček (talk) 13:57, 3 October 2021 (UTC)[reply]

Let's talk about the Desktop Improvements[edit]

Annotated Wikipedia Vector interface (logged-out).png


Have you noticed that some wikis have a different desktop interface? Are you curious about the next steps? Maybe you have questions or ideas regarding the design or technical matters?

Join an online meeting with the team working on the Desktop Improvements! It will take place on October 12th, 16:00 UTC on Zoom. It will last an hour. Click here to join.


  • Update on the recent developments
  • Sticky header - presentation of the demo version
  • Questions and answers, discussion


The meeting will not be recorded or streamed. Notes will be taken in a Google Docs file. The presentation part (first two points in the agenda) will be given in English.

We can answer questions asked in English, French, Polish, and Spanish. If you would like to ask questions in advance, add them on the talk page or send them to

Olga Vasileva (the team manager) will be hosting this meeting.

Invitation link

We hope to see you! SGrabarczuk (WMF) 15:09, 4 October 2021 (UTC)[reply]

Tech News: 2021-40[edit]

16:31, 4 October 2021 (UTC)

Wikisource:WikiProject Data[edit]

I have started Wikisource:WikiProject Data to try to bring together the wider WS strategy (such that there is one) for Wikidata best practices.

If you are interested, you can sign up for {{ping project|Data}} notifications here. Inductiveloadtalk/contribs 18:55, 5 October 2021 (UTC)[reply]

Enable upload-by-URL for autoconfirmed users[edit]

Currently, no Wikisource can use the "upload by URL" feature that can be used at Commons. phab:T293205 would add this capability to all Wikisources from a short allowlist of acceptable sources (for now, the Internet Archive and a couple of Toolforge areas), but to actually use it, we also need to assign the upload_by_url user right to users. At Commons, this right is granted to all users. At English Wikisource, we grant the normal upload right to autoconfirmed users. Since there is no intrinsic extra "danger" to URL uploading (because if the user wanted to, they could download it and upload normally anyway), I suggest to grant auto-confirmed users this right, in addition to the upload right they already have.

If this proposal is supported by consensus, I can prepare and submit a configuration change and schedule it for deployment.

Note: please do not start suggesting other import URLs to allow in this discussion: start a new thread if you know of a specific domain we would want to import files from (that are not suitable for Commons). That would be a completely separate configuration change. Inductiveloadtalk/contribs 15:13, 13 October 2021 (UTC)[reply]

  • Symbol support vote.svg Support There's issue with it on large general purpose projects like enwp, where uncritical import from Flickr and similar creates a patrolling headache, but that doesn't apply to use; and when even Commons (which has similar issues to enwp) has it enabled for all users it should be fine for us too. We can always tighten the permission if it should turn out to be a problem. Xover (talk) 08:17, 24 October 2021 (UTC)[reply]
This has now been requested: phab:T294447 Inductiveloadtalk/contribs 16:14, 27 October 2021 (UTC)[reply]
This was deployed and is now working. The same rules apply to URL uploads as normal uploads: only upload here if they are 1) public domain in the US and 2) Commons will not accept them because they are copyrighted in the country of origin. Inductiveloadtalk/contribs 15:01, 1 November 2021 (UTC)[reply]
This section was archived on a request by: Inductiveloadtalk/contribs 15:01, 1 November 2021 (UTC)[reply]

Orphaned pages[edit]

Hi. We should work out a way to avoid these pages to appear in Orphaned pages: Page:Hesiod, The Homeric Hymns, and Homerica.djvu/100.Mpaa (talk) Mpaa (talk) 22:18, 3 October 2021 (UTC)[reply]

@Mpaa: Transwiki it to either elWS or mulWS and then delete it. — billinghurst sDrewth 22:20, 3 October 2021 (UTC)[reply]
I tried to understand the process but I guess I do not have rights for Special:Import. Mpaa (talk) 20:08, 5 October 2021 (UTC)[reply]

Adding Alternative Image[edit]

I've noticed that on many books there is the image from the book and often a second, higher-quality version of the same image. For example, a black-and-white painting and then a full color, modern scan of the same image. This sometimes leads to debates over which one to use. I was wondering if it would be possible to add an alternative image (maybe to floating image) and then toggle it in the same way that long s is dealt with. The alternative parameter would be optional. In this way, each user can decide whether to see the original or the modern scans.

{{img float | file = original file | alternative = modern scan … }}

Languageseeker (talk) 22:36, 8 October 2021 (UTC)[reply]

Wikisource:WikiProject Women in Red[edit]

In preparation for Ada Lovelace Day this coming Tuesday I have started this WikiProject to tie in with it. The purpose of the wider Women in Red project is to deal with the gender bias across the Wikimedia sites and in particular to turn the red-linked women's names blue. The main focus of Ada Lovelace Day is on Women in STEM and most participants will be creating biographies on WP and items on WD, but I am hopeful that there will be some interest in coming here to create Authors and maybe do a bit of proofreading.

Going on from there, when and as you come across the names of women authors in your proofreading, and you can't work out who they are, please add them to the project page in an appropriate section. Adding any information you glean, including the title of a work, will help others when they come to create the Author: page. Beeswaxcandle (talk) 05:32, 10 October 2021 (UTC)[reply]

Why I write (George Orwell)[edit]

Hi all, sorry new to wikisource and not sure if this is the right place for this, but wanted to query about "Why I Write" by George Orwell.

I can't find a copy on here for the text, and wanted to see if this was a copyright issue at all? I think the work is public domain now (published in 1946). Jamzze (talk) 18:27, 6 October 2021 (UTC)[reply]

Hello. Unfortunately, this work is in the public domain in the UK, but for English Wikisource it is necessary it was in the public domain in the United States, which will not happen until 95 years after publication. For details see Wikisource:Copyright policy and Help:Public domain. --Jan Kameníček (talk) 18:45, 6 October 2021 (UTC)[reply]
Thanks - I thought it might be something to do with copyright! Jamzze (talk) 19:13, 6 October 2021 (UTC)[reply]
here it is at Internet Archive, [5] in collected essays, not digital controlled lending. here is File:George Orwell - 'Politics & the English Language' (1946) (IA orwell-politics-the-english-language-1946).pdf, but you are at the mercy of the uneven URAA enforcement, given the legal memo suggesting actual evidence of a copyright violation. --Slowking4Farmbrough's revenge 13:50, 7 October 2021 (UTC)[reply]
This has nothing to do with the URAA; under US law it's a US work that dotted all its i's and crossed all its t's.--Prosfilaes (talk) 09:18, 9 October 2021 (UTC)[reply]
good, then upload the local version of the essays right now. --Slowking4Farmbrough's revenge 22:45, 9 October 2021 (UTC)[reply]
The point is that it has nothing to do with the URAA or the status in the UK. Why I Write is copyrighted in the United States because it was published in 1953 with renewal (Renewal: RE121351). Politics and the English language was published in 1950 and renewed (Renewal: RE014467) MarkLSteadman (talk) 23:15, 9 October 2021 (UTC)[reply]
thanks, we are left to guess, when broad claims of copyright are made, and works exist on commons, a deletionist place; and "Politics and the English language" may be public domain, since there is enough of a gap between the serial and book, but i leave that to your deletion discussion. Slowking4Farmbrough's revenge 15:16, 10 October 2021 (UTC)[reply]

Tech News: 2021-41[edit]

15:30, 11 October 2021 (UTC)

Template:Default layout[edit]

Why does Template:Default layout not work in user pages? --PastelKos (talk) 13:07, 10 October 2021 (UTC)[reply]

Because it's designed for Mainspace: only. Beeswaxcandle (talk) 03:57, 12 October 2021 (UTC)[reply]

Maintenance Project Proposal: Wikisorcify Wikipedia[edit]

On Wikipedia, many of the texts available on Wikisource are either unlisted or are at the bottom of the page making it difficult for users to know that the text. However, most of the articles on Wikipedia about text use the template {{Infobox book}} that has an optional parameter | wikisource = name_of_text_on_Wikisource. This optional parameter adds a link to the Wikisource inside of the info box making it extremely visible. For example, see Jane Eyre. Adding this will bring two major benefits

  1. It will improve Wikipedia by making it easier for users to find the original texts described in the article.
  2. It will benefit Wikisource by raising awareness of this project attracting more users to Wikisource. Some of these users will then contribute to Wikisource.

The only restriction that I am proposing is that the text should be scan-backed. I know that there is quite a lot of debate about scan-backed vs non-scan-backed, but I strongly believe that scan-backed works are the true strength of Wikisource and the best way to advertise this project. Languageseeker (talk) 01:29, 12 October 2021 (UTC)[reply]

I don't like the boxes, I don't see them unless I look for them. I do look in the list of links, "Further reading" where they list off-site transcriptions, if they exist. The template to put it in that list is {{cite wikisource} or {{wikisource cite} or something like that. While looking for that template the last time, the term they use (instead of infobox, which has a lot more information) is "poster", making it, for me, all the less likable.--RaboKarbakian (talk) 01:44, 12 October 2021 (UTC)[reply]
@Languageseeker: I imagine a very large amount of this can be automated: w:Template talk:Infobox_book#Automatic_Wikisource_link_via_Wikidata_sitelink. Inductiveloadtalk/contribs 13:42, 13 October 2021 (UTC)[reply]
@Inductiveload: I should have figured that you would find a much better and more elegant way to implement one of my ideas. :) That would greatly simplify thing. If this gets implemented, it might make sense to also make a scan for Indexes that are transcluded, but not listed on Wikipedia to figure out what needs to be done manually. Languageseeker (talk) 16:52, 13 October 2021 (UTC)[reply]
@Languageseeker Indexes that are transcluded, but not listed on Wikipedia: I'm not quite sure what you mean by this.
The biggest issue will be dealing with the (common) case where WP links to the work, but the Wikisource page links to an edition. This happens whenever we do not have a versions page for a edition, usually because there is only a single edition at WS. This may be better to solve with a dedicated template at Wikipedia to follow has edition or translation (P747). Inductiveloadtalk/contribs 17:13, 13 October 2021 (UTC)[reply]
@Inductiveload: I meant that not every text that is transcluded will be automatically matched to Wikipedia. Some books do not have articles or the algorithm could fail. Therefore, it would be nice to be able to easily distinguish between the books that are listed or Wikipedia and those that are not. Perhaps, it might be possible to even add a field in the Index ns Listed on Wikipedia that would be automatically set to true if the data-match happens or can be manually toggled. Then, it would be possible to find out which books are not listed. Languageseeker (talk) 01:03, 14 October 2021 (UTC)[reply]

Help with formatting a transcription[edit]

Can someone help me with my transcription, I want to match the format and font more closely. See: Commons:File:War Department letter to Eloise Lindauer II (1856-1935) concerning the reburial of Louis Julius Freudenberg I (1894-1918).png --RAN (talk) 06:39, 9 October 2021 (UTC)[reply]

@Richard Arthur Norton (1958- ): Something like this? Xover (talk) 08:30, 9 October 2021 (UTC)[reply]
  • Perfect, thank you! I will use it as a template for other War Department letters I am transcribing. --RAN (talk) 15:08, 9 October 2021 (UTC)[reply]
@Xover: More questions: Where would I add Portal:Eloise Lindauer and how would I change the font to Courier, to see if it more closely matched the original font. I am sorry to bother with such basic questions. --RAN (talk) 23:44, 9 October 2021 (UTC)[reply]
@Richard Arthur Norton (1958- ): Specific fonts are challenging here, but you might look into <code> and its variations or even use html <pre>. As for the "Portal": here, people are typically found in the Author namespace (like Author:Edgar Allan Poe or Author:Lewis Carroll) and groups of people or Subjects are found in the Portal namespace (Like Portal:Brothers Grimm or Portal:United States Office of Education or Portal:Biology). Perhaps your Portal should be moved to Author.--RaboKarbakian (talk) 18:35, 10 October 2021 (UTC)[reply]
@RAN: Portals are added through a parameter to the {{header}} template, like this.
Regarding the font, we do not add specific fonts (vs. a general type of font), both because we cannot guarantee all users have it available and because we want users to be able to choose their font for a given style. The only exceptions are {{blackletter}} and some specialised fonts for non-English scripts (Hebrew etc.). If you just want to experiment you can manually do it with raw HTML (<div style="font-family: Courier;"> page content </div>) and preview, but please don't actually use that in any non-sandbox page. Xover (talk) 17:51, 11 October 2021 (UTC)[reply]

wikipedia link for work via wikidata[edit]

A wikipedia article on a work here is what I expect and usually see being linked in the header, one-to-one for a text if it exists, however, I notice that the link will correspond with the 'main subject' of a work at its own wikidata item. I assume that must be unintentional and can be corrected in our header's coding for sister links. Cygnis insignis (talk) 15:28, 11 October 2021 (UTC)[reply]

In the early days of linking with WD there was some automations that got it wrong. The way to fix is to repoint over at Wikidata. Beeswaxcandle (talk) 03:59, 12 October 2021 (UTC)[reply]
The links to Wikipedia are graduated. 1) Direct interwikink; then 2) Through the WD edition of; then 3) main subject. There is no need to do direct WP links, it is better if the links are managed through WD so the most relevant link is used, AND it better corresponds with the target page at WP wherever it may be. If there is nothing at WD then no link; all works well. — billinghurst sDrewth 07:05, 13 October 2021 (UTC)[reply]
Think we documented these at Template talk:plain sister. I could be mistaken that we had a brief conversation here, I forget. — billinghurst sDrewth 07:08, 13 October 2021 (UTC)[reply]

I have a related question about this, and to mention that I think the wikidata interaction described above is very cool. My question is: Does wikipedia sift through the versions to find the wikisource version, and if not when?--RaboKarbakian (talk) 13:36, 13 October 2021 (UTC)[reply]

Voting for the MCDC Election[edit]

Voting for the election of the members for the Movement Charter drafting committee is now open. In total, 70 Wikimedians from around the world are running for 7 seats in this election.

Voting is open from October 12 to October 24, 2021.

The Movement Charter committee will consist of 15 members in total: The online communities will vote for 7 members, 6 members will be selected by the Wikimedia affiliates through a parallel process, and 2 members will be appointed by the Wikimedia Foundation. The plan is to assemble the committee by November 1, 2021.

You can learn more about each candidate to inform your vote here

You can also learn more about the Drafting Committee here

We are piloting a voting advice application for this election. Click through the tool and you will see which candidate is closest to you! To try out this tool, visit: App

Go vote at SecurePoll: Vote

Read the full announcement: here

Best, --Civvi (WMF) (talk) 15:13, 14 October 2021 (UTC)[reply]

Technical issues with lilypond markup[edit]

See [6] and compare with the previous version. I've tried a few of the commands in the documentation, but none appear to work. I don't know if this is something caused by the new method lilypond is implemented. @ShakespeareFan00: FYI, you seem to have edited some of those recently. RandomCanadian (talk) 02:08, 12 October 2021 (UTC)[reply]

Yes, some of the markup commands no longer work in the safe implementation. The A&N Hymnal is a low priority for me at present, so I haven't taken the time to look into what's going on. By the way, we're now on version 2.22. Beeswaxcandle (talk) 03:56, 12 October 2021 (UTC)[reply]
@Beeswaxcandle: Would it be a better idea, then, to do something similar to this template I've created (adapting it for the specific formatting of the A&H hymnal, naturally)? RandomCanadian (talk) 18:02, 15 October 2021 (UTC)[reply]
No. And I would not recommend using such in the English Hymnal either. The reasons for including the text as part of the Lilypond score is that when a user grabs a copy of the score, then they will get that header text with it. If it's in a separate template, then the data will be disconnected and they will just have a score with no indication of what it is and where it came from. This is why I was gradually removing the hymn header template (or whatever it's called) out of the A&H hymnal pages. Beeswaxcandle (talk) 19:55, 15 October 2021 (UTC)[reply]
@Beeswaxcandle: I've just tested this out but the new implementation does not appear to yield much if any data. The MP3 has strictly nothing useful, the midi only has the title (see below).
Auto-generated MIDI source code

MThd � � ��€MTrk ˆ ÿ�7Behold the Bridegroom cometh in the middle of the night ÿ� creator: ÿ��GNU LilyPond 2.22.0 ÿX����� ÿQ��¢^‚Ü ÿX�� ��° ÿ/ MTrk �û ÀI ÀI ÿ��flute ÿY�� @> ;>Œ @ ; C> ;>† C ; B> ;>† B ; @> ;>† @ ; @> 9>† @ 9 @> ;>† @ ?>ƒ ; ;>ƒ ? ; @> ;>† @ ; C> ;>† C ; B> ;>† B ; @> 7>ƒ 7 9>ƒ @ 9 @> ;>† @ ; ?> ;>† ? ; @> ;>Œ @ ; † C> @>† C @ H> @>† H @ G> @>† G @ E> @>† E @ E> @>† E @ E> @>† E @ D> @>† D @ E> @>† E @ H> @>† H @ G> @>† G @ E> <>ƒ < >>ƒ E > E> @>† E @ D> @>† D @ E> @>Œ E @ † B> >>† B > G> >>† G > E> >>† E > >>† > C> <>† C < C> >>† C > B> >>† B > C> >>† C > G> >>† G > E> >>† E > C> ;>ƒ ; <>ƒ C < C> >>† C > B> >>† B > C> >>Œ C > † C> ;>† C ; >> ;>† > ; @> ;>† @ ; B> >>† B > E> >>† E > C> ;>† C ; B> ;>ƒ B @>ƒ ; @ ?> ;>† ? ; @> ;>† @ ; C> 7>† C 7 B> >>† > ;>ƒ B @>ƒ @ @>Œ ; @ ?> ;>† ? ; @> ;>Œ @ ; ÿ/ MTrk �– ÁI ÁI ÿ��flute ÿY�� ‘4>Œ ‘4 ‘4>† ‘4 ‘6> ‘2>† ‘6 ‘2 ‘7> ‘4>† ‘7 ‘4 ‘4> ‘0>† ‘4 ‘0 ‘6> ‘/>‰ ‘6 ‘/ ‘6> ‘/>ƒ ‘6 ‘/ ‘4>† ‘4 ‘4>† ‘4 ‘2> ‘/>† ‘2 ‘/ ‘4> ‘0>† ‘4 ‘0 ‘6> ‘/>† ‘6 ‘/ ‘6> ‘/>† ‘6 ‘/ ‘4>Œ ‘4 † ‘4>† ‘4 ‘9> ‘->† ‘9 ‘- ‘;> ‘,>† ‘; ‘, ‘<> ‘->† ‘< ‘- ‘9> ‘->† ‘9 ‘- ‘;> ‘4>† ‘; ‘4 ‘;> ‘4>† ‘; ‘4 ‘9> ‘->† ‘9 ‘- ‘9> ‘->† ‘9 ‘- ‘7> ‘4>† ‘7 ‘4 ‘9> ‘5>† ‘9 ‘5 ‘;> ‘4>† ‘; ‘4 ‘;> ‘4>† ‘; ‘4 ‘9> ‘->Œ ‘9 ‘- † ‘2>† ‘2 ‘7> ‘+>† ‘7 ‘+ ‘9> ‘2>† ‘9 ‘2 ‘;> ‘7>† ‘; ‘7 ‘7> ‘4>† ‘7 ‘4 ‘9> ‘2>† ‘9 ‘2 ‘9> ‘2>† ‘9 ‘2 ‘7> ‘+>† ‘7 ‘+ ‘7> ‘+>† ‘7 ‘+ ‘6> ‘2>† ‘6 ‘2 ‘7> ‘4>† ‘7 ‘4 ‘9> ‘2>† ‘9 ‘2 ‘9> ‘2>† ‘9 ‘2 ‘7> ‘+>Œ ‘7 ‘+ † ‘7> ‘4>† ‘7 ‘4 ‘6> ‘/>† ‘6 ‘/ ‘7> ‘4>† ‘7 ‘4 ‘9> ‘2>† ‘9 ‘2 ‘6> ‘2>† ‘6 ‘2 ‘7> ‘4>† ‘7 ‘4 ‘7> ‘4>† ‘7 ‘4 ‘6> ‘/>† ‘6 ‘/ ‘4>† ‘4 ‘4> ‘;>† ‘4 ‘2>ƒ ‘; ‘9>ƒ ‘2 ‘9 ‘4> ‘7>† ‘7 ‘4 ‘4>† ‘4 ‘6> ‘/>Œ ‘6 ‘/ ‘4>Œ ‘4 ÿ/

(Based on this). It's possible to have both the custom-formatted header and this (limited) information, though. RandomCanadian (talk) 21:17, 15 October 2021 (UTC)[reply]
I couldn't give a toss about the midi as I never use it. It's the score image itself that is taken and printed. If I need the score of a hymn (or any other piece), then I don't want the Wikimedia fluff around it, I just want the music. Then six months later, when I'm sorting through a pile of printed music, if I don't have at least the tune name, metre, composer and author, then it's useless. These need to be part of the Lilypond score, otherwise we're not producing a hymnal of any practical use to church musicians. Beeswaxcandle (talk) 21:37, 15 October 2021 (UTC)[reply]
Well there's multiple solutions to that problem (copy-pasting the header and the score image to Word or some other software before printing also allows you to keep the information; and this isn't much different or more complex than just saving the score image...; you can also just take a screenshot (which will include the header, however it is formatted behind-the-scenes), ...). And having one easily useable template for all the 700-odd hymns (give or take a few, depending on whether you want to count the gregorian examples, since these can't work with Lilypond in safe mode, see for ex. Page:The English hymnal (1906).djvu/26 - not caused by a lilypond error AFAICS); is certainly quite more useful for both editors in terms of code readability and readers in terms of final output than having to deal with a disabled lilypond feature. RandomCanadian (talk) 22:00, 15 October 2021 (UTC)[reply]

Author:T. S. Eliot‎ vs Author:Thomas Stearns Eliot[edit]

I was doing some routine maintenance stuff yesterday, and I cleaned up a bunch of authors with initials in their page titles. I noticed, however, that despite our practice of moving Author pages with initials wherever possible, nevertheless Author:T. S. Eliot‎ remains as the only prominent holdout. Is there community consensus to keep Author:T. S. Eliot‎ at its current location with its initials, or should we redirect it to Author:Thomas Stearns Eliot like we do with everyone else? —Beleg Tâl (talk) 14:47, 14 October 2021 (UTC)[reply]

@Billinghurst, @Cygnis insignis: pinging interested parties —Beleg Tâl (talk) 14:49, 14 October 2021 (UTC)[reply]
There is no exception to the guidance. Just an individual's commentary that there is no consensus for such a move and we that are to follow the author's preference for that form. I have the conversation over and over with the individual, and asked them to bring that conversation here, and that has not happened. I thank you for raising the issue. I believe that the standard/guidance should stand. — billinghurst sDrewth 06:11, 15 October 2021 (UTC)[reply]
  • @Beleg Tâl: why did you redirect Shakespeare to Bill Shaxberd? Cygnis insignis (talk) 12:41, 15 October 2021 (UTC)[reply]
    • Updating links after a page move, that's all. If you think that it should point somewhere else, like Author:William Shakespeare, or even a separate disambig page for anyone surnamed Shakepseare ... well I personally don't think that surname-disambig is something we should do, but I won't object to it either. I don't think the name "Shakespeare" by itself can really be expected to refer to any Shakespeare but the one it currently redirects to, but I would not be unhappy to be proven to be mistaken. —Beleg Tâl (talk) 13:04, 15 October 2021 (UTC)[reply]
@Beleg Tâl: There is no such consensus that I am aware of, and I am frankly surprised the point needs to be argued. Do we really need to hold a vote and make it a policy? Don't get me wrong, I like having stuff like this enshrined clearly in policy, but it seems quite a waste of the community's time on this particular point. Whether you like the "full name" page naming scheme or not, it is the long-standing and essentially universal practice on the project and I see no reasonable argument why Eliot should be an exception. Xover (talk) 17:18, 15 October 2021 (UTC)[reply]

Fixing the Cover Image of Strange Case of Dr Jekyll and Mr Hyde[edit]

If you look at the original cover image for Page:Stevenson - Strange case of Dr. Jekyll and Mr. Hyde (1886).djvu/1, it's obvious that the cover was printed in color. However, the image used on the page is black and white, can somebody redo the cover image for this text so that we can have a proper cover image for this month's featured text. Also, could an administrator add the cover image to the header on Strange Case of Dr Jekyll and Mr Hyde so that it can be exported? Languageseeker (talk) 01:12, 18 October 2021 (UTC)[reply]

File:Strange Case of Dr Jekyll and Mr Hyde-1896-0001.jpg. If you get a better one, don't worry about my feelings and such; a better one could be made.--RaboKarbakian (talk) 02:47, 18 October 2021 (UTC)[reply]

Adding Category:Children's books to Beatrix Potter's Books[edit]

The works of Author:Beatrix Potter should all be tagged with [[Category:Children's books]], but most are not. I'm unable to do so because the books are protected so that only an Administrator can modify them. Can someone add this category to all her books? Languageseeker (talk) 05:07, 18 October 2021 (UTC)[reply]

Yes check.svg Done I have also removed all the WP and Commons links. It would be worthwhile checking that each of the works has all the edition linking so they collect the right linking. Thanks. — billinghurst sDrewth 13:29, 18 October 2021 (UTC)[reply]

Tech News: 2021-42[edit]

20:53, 18 October 2021 (UTC)


i want to upload this, but I can't. 18:14, 19 October 2021 (UTC)[reply]

Hi This should be uploaded as the complete scan file and then Help:proofread from that. I will set it up for you. In the meantime, if you create a user account, you will be able to contribute by proofreading the text when it arrives (to be clear, you can edit while not logged in, but you will not be able to set pages to Proofread. Inductiveloadtalk/contribs 20:08, 19 October 2021 (UTC)[reply]
File is now uploaded at Index:New York State Birth Index - 1893.djvu. You can proofread the text now by visiting the individual "pages" (e.g. Page:New York State Birth Index - 1893.djvu/2 and entering the correct text. I have set up a template {{NYS birth entry}} for you. Inductiveloadtalk/contribs 21:36, 19 October 2021 (UTC)[reply]

Learn how Movement Strategy Implementation Grants can support your Movement Strategy plans[edit]

Movement Strategy Implementation grants now provide more than $2,000 USD to put Movement Strategy plans into action. Find out more about Movement Strategy Implementation grants, the criteria, and how to apply here.

Also, the Movement Charter Drafting Committee election is still ongoing. It would be great to increase community participation. If you haven't voted now is the time. Please vote here before October 24. Regards, --Civvi (WMF) (talk) 12:42, 21 October 2021 (UTC)[reply]

Talk to the Community Tech[edit]

Magic Wand Icon 229981 Color Flipped.svg

Read this message in another language


We, the team working on the Community Wishlist Survey, would like to invite you to an online meeting with us. It will begin on 27 October (Wednesday) at 14:30 UTC on Zoom, and will last an hour. Click here to join.


  • Become a Community Wishlist Survey Ambassador. Help us spread the word about the CWS in your community.
  • Update on the disambiguation and the real-time preview wishes
  • Questions and answers


The meeting will not be recorded or streamed. Notes without attribution will be taken and published on Meta-Wiki. The presentation (all points in the agenda except for the questions and answers) will be given in English.

We can answer questions asked in English, French, Polish, Spanish, German, and Italian. If you would like to ask questions in advance, add them on the Community Wishlist Survey talk page or send to

Natalia Rodriguez (the Community Tech manager) will be hosting this meeting.

Invitation link

We hope to see you! SGrabarczuk (WMF) (talk) 23:00, 22 October 2021 (UTC)[reply]

Version Page for Common Sense[edit]

There are currently two editions for Common sense: Common Sense and Common Sense Addressed to the Inhabitants of America. Can an Administrator create a version page and move Common Sense to Common Sense (unsourced edition) and Common Sense Addressed to the Inhabitants of America to Common Sense (Sixth Edition). I would also support deleting the unsourced edition from 2004 and moving the proofread text to Common Sense. Languageseeker (talk) 20:26, 20 October 2021 (UTC)[reply]

There are no subpages to Common Sense, so it doesn't require an admin to do the move. I don't see that we would need the other identified version to something tagging as 6th edition, create a redirect if you think that it is needed/ — billinghurst sDrewth 13:22, 21 October 2021 (UTC)[reply]
@Languageseeker: it appears that Common Sense is an edition that Paine had published by W. and T. Bradford before his main publisher Bell even released the second "official" edition. Apparently there were two dozen editions published in 1776? So perhaps a more robust disambig scheme is needed, e.g. Common Sense (Bradford, 1776), Common Sense (Carter, 1776)? —Beleg Tâl (talk) 21:01, 25 October 2021 (UTC)[reply]
(see for very detailed publication history of this work) —Beleg Tâl (talk) 21:05, 25 October 2021 (UTC)[reply]
Oh and here's a scan for the Bradford editionBeleg Tâl (talk) 21:08, 25 October 2021 (UTC)[reply]

Year Zero[edit]

There is a template {{deaths by year}} used in categories like "1890 deaths". However, if you visit e. g. Category:1 BCE deaths, the box on the right offers a red link for "Category:0 BCE deaths", which is a non-sense, as there was no year zero in history. The same applies for {{births by year}}. Could the templates be fixed not to offer such a year? The red link can only attract jokesters to try to found the category, as has recently happened. --Jan Kameníček (talk) 18:57, 24 October 2021 (UTC)[reply]

I have protected the 0 year categories, but it would be nice if the red links disappeared anyway. --Jan Kameníček (talk) 19:18, 24 October 2021 (UTC)[reply]
Should be fixed now. The template actually had a measure in place to remove the year zero, it just wasn't working properly. —Beleg Tâl (talk) 20:35, 25 October 2021 (UTC)[reply]

Tech News: 2021-43[edit]

20:08, 25 October 2021 (UTC)

Geopolitics, carbon-neutrality and uploads[edit]

So here's a fun economic second-order effect. Fuel prices are way up this winter for us Northerly folks for various excitingly adult reasons. I am told that I need to heat my house. Some of my heating is electric. Computers are also electric. A computer is in the house. Converting things to scan files produces a bit of heat and a useful side effect: tasty, tasty scans! And the power is free, because it would be used in a radiator anyway! Therefore, the high fuel prices encourage scan imports.

So, if anyone has a massive pile of Hathi or IA documents that they want to be chunked though, all they need to do is hand me a nice chunky XLSX as described at User:Inductiveload/Requests/Batch uploads before, say, March, and I will keep my little toes warm on the djvm exhaust and the conversion will be more-or-less carbon neutral. 🌲 Inductiveloadtalk/contribs 20:46, 25 October 2021 (UTC)[reply]

A Handbook of Indian Art[edit]

I have noticed that the copy of A Handbook of Indian Art is missing the Frontispiece and Plates 3a and 3b. I have found another copy of it on the Internet Archive here which contains the missing images. I have not checked either copy for any additional issues, apart from I noticed that the alternate copy also contains a title page followed by an advert page. Is there a process to repair or replace the existing copy? If this can be done, I'll continue working on it. Thanks Sp1nd01 (talk) 23:02, 25 October 2021 (UTC)[reply]

You can post a request in Wikisource:Scan Lab Languageseeker (talk) 23:15, 25 October 2021 (UTC)[reply]
@Sp1nd01: -- Done. Hrishikes (talk) 04:08, 26 October 2021 (UTC)[reply]
Thank you both, I wasn't aware of the Scan Lab, good to know if I come across other issues. Sp1nd01 (talk) 08:28, 26 October 2021 (UTC)[reply]
@Hrishikes: Sorry to report that there are still some page misalignment.
I thought that I had better have it rechecked before I continue working on it.
i.e. At the start of the book, I created the first 2 pages, then the front matter appears out of synch.
In chapter 1 Page 14 text is alongside an image of page 20 with various other misalignments following etc.
Can this message be moved to the Scan Lab for detailed investigation please? Thanks Sp1nd01 (talk) 09:23, 26 October 2021 (UTC)[reply]
@Sp1nd01: -- It is not misalignment. It is cache. On page 14, right click on the image button and open in a new tab, then you'll be able to see the scan of page 14. Same with the title page. You proofread it from the cache, after my page move. Hrishikes (talk) 12:41, 26 October 2021 (UTC)[reply]
@Hrishikes: Thanks you, I didn't realize about the cache. I have done as you suggested and it now appears as it should. Sorry to have bothered you. Sp1nd01 (talk) 13:13, 26 October 2021 (UTC)[reply]

Please request me to transcribe PD Maine-related works in 2022[edit]

I am planning to do an editing spree for works related to the US state of Maine starting next year. I'm going to fill up Portal:Maine and make lots of subportals in 2022. Please drop any and all Maine works you'd like transcribed at this page, in the form of bullet points in the correct categories: User:PseudoSkull/Maine. Anyone can edit this userspace page. Thanks. PseudoSkull (talk) 12:55, 26 October 2021 (UTC)[reply]

RfC: Constitutions => versions or disambiguations[edit]

Where we have constitutions of a country, do we think that we should be using {{versions}}/{{other version}} or {{disambiguation}}/{{similar}} as the lead/seealso template pairs. I am favouring "disambiguation" as they can have a complete rewrite so are clearly their own separate entities, even if one is transactionally changed. Trying to manage that way I see as too confusing, though happy to hear other opinions on what we should do. I will hazard a guess that such a resolution will flow through to other works, so it would become guidance on Help:Disambiguation. — billinghurst sDrewth 03:22, 28 October 2021 (UTC)[reply]

I think it depends entirely on whether they are actually different constitutions. Some countries' constitutions have a whole pile of Amendments (looking at you, USA) which I would definitely NOT consider to produce a different work; so a Versions page would be appropriate. However, other countries are known to replace one constitution with another one rewritten from the top, in which case a Disambiguation page would be more appropriate. I'm sure the line between the two scenarios is fuzzy, e.g. where a constitution is substantially rewritten but is also substantially the same, but we have generally used Versions pages for such scenarios in other types of texts. —Beleg Tâl (talk) 13:14, 28 October 2021 (UTC)[reply]
see also Looking at two little-known versions of our Constitution & How Different Are the Early Versions of the United States Constitution? An Examination.Slowking4Farmbrough's revenge 00:54, 29 October 2021 (UTC)[reply]

Questions about our OCR tool[edit]

Is the purpose of this OCR tool to rescan the djvu/pdf image and improve on the OCR layer included in the original?

The OCR tool activated in gadgets is a recipient of the improvements discussed in Wikiproject OCR?— Ineuw (talk) 11:04, 27 October 2021 (UTC)[reply]

Yes, if the original OCR layer is bad or even missing, the OCR tool should provide new OCR text. There is no direct connection with the Wikisource:WikiProject OCR, it was created after the previous OCR tool had been defunct for a long time and as a result of a 2020 community wish.
Unfortunately, the community’s wish was only partly fulfilled as the Tech Team decided not to reflect the call for a fully integrated tool and created another external tool which is quite good for basic OCR, but very impractical when a contributor tries to use additional features like setting the language or work with columns (the necessity to leave the OCRed page again and again for such features slows the work so much, that it is often better not to use them at all). --Jan Kameníček (talk) 11:24, 27 October 2021 (UTC)[reply]
They have in the advanced options, a croptool, for columns. Once at the advanced options site (it takes you to a different site with a lot of options) there is a url loader, which I have used successfully on wrongly oriented tables (after using the commons croptool to rotate them). So, I think it would be useful to add image rotating to that advanced options site. Further, I have bookmarked the advanced options site, and have skipped having to enable the edit toolbar. The url has been provided below, so, with some acrobatics, tool using and pasting image urls from commons, a lot can be ocred....--RaboKarbakian (talk) 17:58, 27 October 2021 (UTC)[reply]
Just brief description of how it slows the work down:
1) I often proofread English texts containing some foreign expressions, but I cannot add the other language directly. I have to open the tool at and only there I can add the other language, make OCR, copy it to clipboard and manually insert to Wikisource. And this has to be done with every individual page again and again, because the tool does not remember it. So it is usually much easier to give the language setting up and correct the non-English expressions manually.
2) If the page contains columns, I need to open the external tool, transcribe every column individually and individually copy it to clipboard and manually insert into Wikisource. However, the wish asked for an integral tool. The tool would be very useful and the work less frustrating if all this could be done directly in the page namespace without any repetitive manual copy-pasting from an external page to our page namespace.
I feel really disappointed by the result which we were waiting so long for :-( --Jan Kameníček (talk) 19:31, 27 October 2021 (UTC)[reply]
@Jan.Kamenicek indeed it's frustrating. Luckily, they did add the crop into the API, which I asked for deliberately so we could do it client-side ourselves because I knew it wouldn't possibly get shipped in the UI. I'm planning to add the regions thing to the on-wiki UI once phab:T283917 is reviewed. But that does rather seem to have stalled out. Inductiveloadtalk/contribs 22:32, 27 October 2021 (UTC)[reply]
@Jan.Kamenicek: Much thanks for your quick reply. Am I correct to assume that, our OCR tool is a separate piece of software, using Tesseract from a source unknown to me?— Ineuw (talk) 11:34, 27 October 2021 (UTC)[reply]
Yes, it is a separate piece of software. For more details see meta:Community Tech/OCR Improvements. --Jan Kameníček (talk) 11:39, 27 October 2021 (UTC)[reply]
@Ineuw yes. There is a web front end here:, and the source code is here Wikisource's UI can interact with the tool's backend it directly over its API. I am planning to add a region selector to the WS editor UI, but I am still road-blocked on phab:T283917, which has been stalled in code review for months now.
The Tesseract is, as far as I know, the one shipped with Debian Buster. Inductiveloadtalk/contribs 12:55, 27 October 2021 (UTC)[reply]
@Inductiveload It was that one, but then we manually upgraded to Tesseract 5 because the that's what the Internet Archive now uses and they said it was better (and stable enough for our use, despite not being officially released yet). See task T282150. —Sam Wilson 01:36, 29 October 2021 (UTC)[reply]
"decided not to reflect the call for a fully integrated tool" yeah, this is the default response. we are awash in tools, that are created, and maintained for a time, and break, and shut down when only one user can make them work. uptake based on word of mouth, very little UX design or integrated solutions to task flows, or tool life cycle management. we are lucky they are responding to our wishes, but tools are a major pain point across all projects. (don't ask about commons uploaders) Slowking4Farmbrough's revenge 00:49, 28 October 2021 (UTC)[reply]
My thanks to all for this info. It cuts through the fog and haze. @Inductiveload: Thanks for the new OCR link. Will try it out and compare it to our toolbar OCR, which also does a good job (for me at least).— Ineuw (talk) 08:58, 29 October 2021 (UTC)[reply]
@Ineuw if you mean the "Transcribe text" tool in the top right of the toolbar, it's the same thing. The "old" gadgets (the ones that read "OCR") are indeed separate. The Google one is probably the same result, but the black-text one is, I think, using a much older Tesseract, so I imagine the results will be different (probably less accurate in general). Inductiveloadtalk/contribs 09:15, 29 October 2021 (UTC)[reply]
I was referring to the old "OCR". On a rare occasion I use the Google gadget as well, to see which generates columns better.
I tried the,_or_Memoirs_of_a_Fugitive.djvu/123 and it was rejected on It requires an "upload.wikimedia" url. — Ineuw (talk) 09:27, 29 October 2021 (UTC)[reply]
@Ineuw the image URL is the actual image, not the page title: in your case: Using the "advanced options" button in the "Transcriber text" tool will take you to the correct page. Inductiveloadtalk/contribs 10:10, 29 October 2021 (UTC)[reply]
Thank you!!! I knew that a page existed, I just didn't remember how to get to it.— Ineuw (talk) 10:14, 29 October 2021 (UTC)[reply]
@Ineuw you hid the new tool in your user CSS, so that is probably not helping your user experience. Inductiveloadtalk/contribs 10:24, 29 October 2021 (UTC)[reply]
@Inductiveload: Phetools is using the stock Tesseract 4.1.1 from Debian Buster. So far as I know Tesseract 5 (which is just a tag in VCS right now) is mainly about internal refactoring, so quality improvements should be mostly incidental (bug fixes, mostly). They may have landed something more interesting (to us) after last time I checked of course, but at least for now I don't think 4 vs. 5 matters. If need be we can build a private Tesseract and have Phetools use that (I have access to the tool account now). Xover (talk) 12:24, 29 October 2021 (UTC)[reply]

@Inductiveload: Thanks for pointing it out. I forgot about it. I tried it out on six pages, and compared them in a text editor. So far they are nearly identical, the new version identified a couple of words whereas the old could not differentiate. Otherwise, they identically see and generate the same gibberish wherever it occurs. So the new Tesseract does seem to be only marginally better. — Ineuw (talk) 11:36, 29 October 2021 (UTC)[reply]