User talk:Alien333

Welcome to Wikisource

Hello, Alien333, and welcome to Wikisource! Thank you for joining the project. I hope you like the place and decide to stay. Here are a few good links for newcomers:

• Help pages, especially for proofreading
• Help:Beginner's guide to Wikisource
• Style guide
• Inclusion policy
• Wikisource:For Wikipedians

You may be interested in participating in

• Proofread of the Month
• Monthly Challenge
• Requested texts

Add the code {{active projects}}, {{PotM}} or {{Collaboration/MC}} to your page for current Wikisource projects.

You can put a brief description of your interests on your user page and contributions to another Wikimedia project, such as Wikipedia and Commons.

Have questions? Then please ask them at either

• Wikisource:Scriptorium; or
• Wikisource:Scriptorium/Help

I hope you enjoy contributing to Wikisource, the library that is free for everyone to use! In discussions, please "sign" your comments using four tildes (~~~~); this will automatically produce your username if you're logged in (or IP address if you are not) and the date. If you need help, ask me on my talk page, or ask your question here (click edit) and place {{helpme}} before your question.

Again, welcome! Beeswaxcandle (talk) 09:33, 6 June 2023 (UTC)

For some reason I removed it two months later, without having read any of it, and it was certainly one of the worst decisions I've made. Readding it now, it's long overdue as a useful reminder to myself to actually pay attention. — Alien  3
3 3

Italics

Please note that italics do not carry across line breaks. You either have to stop and restart on the next line. or (better) remove the line breaks.

Regards -- Beardo (talk) 16:55, 24 August 2023 (UTC)

Yes, sorry, I know, it's just that it took me a little while to realize and that when, then, I tried to go back and correct myself, I missed a few. I'm pretty new at this and so I more or less learned by experience.

˜˜˜˜ Alien333 (talk) 17:44, 24 August 2023 (UTC)

No problem. I think most of us have learned that way. I recently learned from Wikisource:Scriptorium#De-linting.. that there is the page linked there which lists such errors. -- Beardo (talk) 18:38, 24 August 2023 (UTC)

Titles on subpages

Making this change in the header of the subpages will turn Poems (Nora May French) by Nora May French into the correct Poems by Nora May French EncycloPetey (talk) 19:59, 20 December 2023 (UTC)

Ok! Sorry, I'll correct it. Alien333 (talk) 19:21, 21 December 2023 (UTC)

Template:asc

You may find {{{asc}} useful, especially for A.M., B.C., and roman numerals that are printed in capital small caps. Yes, you could use {{sc}} with lower-case letters, but typically books do not use lower-case letters for these things, and putting lower-case into the text with small-caps will not preserve the case when someone grabs the text using copy-paste, such as for a quote in a school paper or for quoting in a Wikipedia article or on Wikiquote. The advantage of {{asc}} is that you can write the text in the correct case and still get it to display in reduced capitals. --EncycloPetey (talk) 20:54, 23 February 2024 (UTC)

OK! Alien333 (what I did and why I did it wrong) 11:17, 24 February 2024 (UTC)

Page:Poems Barrett.djvu/134

Check your module , >><<< seems to misbehave , by throwing a supurious closing SPAN tag? ShakespeareFan00 (talk) 17:24, 1 May 2024 (UTC)

If you are talking about the nearly-empty poem, there was a cleaner way to do it (and I corrected it), but if it's not that I don't see what you mean about that closing SPAN. As far as I can see, it only adds a </span> at the same time as adding a <span>. — Alien333 (what I did & why I did it wrong) 17:31, 1 May 2024 (UTC)

Reminder to vote now to select members of the first U4C

You can find this message translated into additional languages on Meta-wiki. Please help translate to your language

Dear Wikimedian,

You are receiving this message because you previously participated in the UCoC process.

This is a reminder that the voting period for the Universal Code of Conduct Coordinating Committee (U4C) ends on May 9, 2024. Read the information on the voting page on Meta-wiki to learn more about voting and voter eligibility.

The Universal Code of Conduct Coordinating Committee (U4C) is a global group dedicated to providing an equitable and consistent implementation of the UCoC. Community members were invited to submit their applications for the U4C. For more information and the responsibilities of the U4C, please review the U4C Charter.

Please share this message with members of your community so they can participate as well.

On behalf of the UCoC project team,

RamzyM (WMF) 23:10, 2 May 2024 (UTC)

Poetry collections

Thanks for completing so many small books of poetry by authors whose works we do not have, and which won't be found in most libraries.

Would you consider also doing Fiddler's Farewell (1926) by poet and violinist Leonora Speyer? (external scan) Her poetry won the Pulitzer in 1927, so it's a significant work, by a poet for whom we have no works at all. --EncycloPetey (talk) 02:58, 9 May 2024 (UTC)

I could, but I'd need you to get it, as I'm not a member at Hathi and it'd be a bother to download each of the 136 pages manually.

If you are more interested by the author than the specific collection, there are two scans of A canopic jar (external scan) (external scan) available on IA, which I prefer. — Alien333 (what I did & why I did it wrong) 06:29, 9 May 2024 (UTC)

Fiddler's Farewell is the Pulitzer winning work, so it's the one I'm interested in, but I cannot grab Hathi downloads either. --EncycloPetey (talk) 14:27, 9 May 2024 (UTC)

Poetry foundation says it's available in the Poetry magazine, which to my surprise we do not have but that is on jstor, more specifically in the issue of Jan. 1926, and the poem itself, p201-205 is there, which according to jstor is in public domain as © 1926 Poetry Foundation. I'll get at it some time soon, probably after finishing Index:Poems Shipton.djvu, but I think eventually I'll try to do the whole magazine. — Alien333 (what I did & why I did it wrong) 14:56, 9 May 2024 (UTC)

Note: did she get the prize for the poem or the book? Because there appears to be a collection of the same name (136 vs. 5 pages), that is the one at Hathi, and the poem after which it appears to have been named, that is what I found. EDIT: after just looking on WP it appears to have been for the book. Once more unto the breach, then... — Alien333 (what I did & why I did it wrong) 14:58, 9 May 2024 (UTC)

*sigh* I've given up, looks like it's Hathi or nothing. I've started taking the pages. EDIT: on top of all the rest, the preview images are scaled down. Well, 700*1000 will have to be enough, and I'm not going to go 136 times through their download dialog — Alien333 (what I did & why I did it wrong) 15:15, 9 May 2024 (UTC)

Well, there it is: Index:Fiddler's Farewell.djvu. The PNGs were acceptable at best, all pdf mergers I found (the three that let me upload 136 pages) made it terrible, for some reason the OCR on djvu conversion appeared not to work, and it has two watermarks, but it's there. As I said, will get at it at some point during next week. — Alien333 (what I did & why I did it wrong) 15:53, 9 May 2024 (UTC)

@EncycloPetey It's done. — Alien333 (what I did & why I did it wrong) 10:48, 11 May 2024 (UTC)

Thanks! --EncycloPetey (talk) 14:34, 11 May 2024 (UTC)

Transclusion in page order

Hi @Alien333,

As EncycloPetey said above, many thanks for completing so many poems. However, in the case of Poems (Baldwyn), I am inclined to believe the transclusion should be in page order, regardless of the ordering in the table of contents. I am in no way asking you to change it, although at some point, someone with greater concerns about the matter may add an (what I would consider "loud") template on the main page indicating it doesn't conform to Wikisource standards, unless things have changed since the last time I recall this happening. At the very least, information for the future. As an aside, if you are interested in having some of your work validated, especially more famous works (like the Fiddler's Farewell) mentioned above, we would be happy to have it included in the Monthly Challenge, if you are okay with that. Up to you though.

Regards, TeysaKarlov (talk) 20:42, 9 May 2024 (UTC)

Fiddler's Farewell is more or less an exception and at request, most of the time I just do random books called Poems. Feel free to include anything you want. I have collections from ~20 content pages to >400, so there's probably something of the right size.

On TOC's, I've already encountered the same problem with Poems (Cromwell), so if correction there is it would have to be done there too. I made that decision on the basis that it would be awkward to not be able to navigate in the sense of the TOC (and maybe also out of laziness of having to scroll through the TOC to find the right capitalization of the titles). — Alien333 (what I did & why I did it wrong) 20:57, 9 May 2024 (UTC)

Given how far you have already progressed with Fiddler's Farewell, I might just leave it as is. You are ever so efficient with the use of those ppoem templates.

I suspect that the previous/next sections of the header were to imply flipping forward and backward through the actual pages of the text (just like a real book!), but in terms of sensible, I don't see a great deal of difference. I guess just consider this a heads up then, unless someone else has graver concerns. Regards, TeysaKarlov (talk) 23:38, 10 May 2024 (UTC)

We'll probably be able to include Fiddler's Farewell quite soon.

(Honestly, regarding ppoem, most of the work of figuring when to put what end/start is done by a script of mine nowadays, alongside with indenting.) — Alien333 (what I did & why I did it wrong) 07:17, 11 May 2024 (UTC)

It's done, so you can include it. — Alien333 (what I did & why I did it wrong) 10:47, 11 May 2024 (UTC)

Disambiguation pages

Hi, I noticed that you've added new works to various disambiguation pages (not everyone does). The convention adopted with these appears to be:—

1. the list is alphabetised by author surname;
2. if there's more than one work of the same name by an author (usually poems), the first line is quoted; and
3. parts of books (e.g. a short story or essay, or individual poems) are given in double quotes, titles of whole works are in italics.

Chrisguise (talk) 07:51, 13 May 2024 (UTC)

Well, ok, by coincidence I've just made a post to ask for the conventions, and possibly officialize it, as everyone does not appear to be aware of there conventions, for example titles are often left plain. — Alien333 (what I did & why I did it wrong) 08:01, 13 May 2024 (UTC)

I made these up for myself as I went along based on what appeared to be most common practice, and most helpful. Chrisguise (talk) 08:05, 13 May 2024 (UTC)

Poems (Smith)

Hi, I noticed that you've used {{AuxTOC}} to create a table of contents for this work when it has one of its own (albeit in a different format to most books). I've just done one (The Canary) which has a (mostly) alphabetised ToC based on first lines rather than titles. For some reason the 'O' section is in reverse alphabetical order. Chrisguise (talk) 17:52, 18 May 2024 (UTC)

Up to a week ago (such as for Poems (Cromwell) and Poems (Baldwyn)), I'd used the original TOC in these cases when the TOC is not in order of apparition, until I was asked about a week ago by @TeysaKarlov to transclude instead in order of apparition, so I also put a second TOC that would match the order of transclusion because it would be awkward to navigate in a totally unrelated order. Usually, I also leave the original TOC after, with the links (like in Poems (Hazlett-Bevis)) but the one in Poems (Smith) was incomplete (did not show poems of the same name, only the first one) so I delinked it. What do you think I should do? — Alien333 (what I did & why I did it wrong) 18:26, 18 May 2024 (UTC)

@Alien333 Sorry if I caused confusion. I was not suggesting to modify the TOC in any way. All I meant was that all the previous/next links for each poem should follow in page order. That aside, what do you mean by "(did not show poems of the same name, only the first one)"? The table of contents seems to have many (if not all) of the poems in Poems (Smith), although I have not checked every one, to see if it is incomplete. However, if a table of contents is missing an entry, you can add an auxiliary line(s) to the original TOC (e.g.~Page:Eliot - Middlemarch, vol. I, 1871.djvu/9). Hope that helps, and thanks again for all your poetry efforts, TeysaKarlov (talk) 21:22, 18 May 2024 (UTC)

There were entries like "To Willie", of which there are two (Poems (Smith)/To Willie (Willie, may thy life abound) and Poems (Smith)/To Willie (Willie, may thy life be pure)), but the original TOC listed only one. Same for other poems that shared a title. Led to redlinks in the toc when @Duckmather linked it. Thanks for reminding me of the aux-toc lines, I'd forgotten they existed.

If pages should always be transcluded in order of apparition, when the TOC is not in that order, a secondary, auxiliary TOC is I think useful for navigation. At any rate, it is for proofreading, because often in poetry titles on the pages of these poems are in all-caps, and the correct capitalization is only present in the TOC. This makes it for most poetry collections a headache to transclude without a TOC in order of apparition to find what is the exact name of the following/preceding poem.

Imagine someone wanted to read one of these collections. They could either a) fish for the smallest page number in the TOC, assuming it's correct, and take the "next" links, or b) start from another one, maybe the first in the TOC, and then land at some random point in the collection and then have to go through the "next" links and the "previous" links if they want to read the whole of it. Same goes if they were interrupted and want to re-start reading at a specific point in the book.

This inconvenience exists specifically and only when the order of tranclusion is different from the order of the TOC. That was why first I always transcluded in the order of the TOC, and after learning that transclusion has to be in page order, I add a second TOC that matches the order of transclusion to ease navigation. — Alien333 (what I did & why I did it wrong) 07:44, 19 May 2024 (UTC)

sib links

As long as the target subpage and target display name are the same, you can use this syntax, which is just as compact as the template but without requiring a template. --EncycloPetey (talk) 21:34, 20 May 2024 (UTC)

Fair enough. — Alien333 (what I did & why I did it wrong) 05:06, 21 May 2024 (UTC)

Hello, new reader here...

https://en.m.wikisource.org/wiki/User_talk:Alien333#/editor/0 ImaginarySusan (talk) 09:00, 2 June 2024 (UTC)

Feel free to ask if you have any questions. — Alien333 (what I did & why I did it wrong) 09:07, 2 June 2024 (UTC)

Merci beaucoup!

( High school French... from 40 years ago!)

🙏👩‍🎨🇲🇫💜 ImaginarySusan (talk) 10:12, 2 June 2024 (UTC)

oh I am fumbling. I wrote a bunch of information to you and then thought it might be in the incorrect place so I copied the message to my clipboard and just tried to send that text to you.

I can't seem to recover it now so I'll attempt to rewrite what I contacted you about.

Serendipitously, i crossed your path by researching wireframes on wiki as i am intrigued with learning to write code... but this was my first trip down the "rabbit hole"

I am a poetry enthusiast, also and while wandering around your contributed content appreciated your knowledge.

Also, Alien333 resonates with me for a variety of reasons..and it happened to be your username, which was my first encounter here. ImaginarySusan (talk) 09:12, 2 June 2024 (UTC)

What you sent was only the link to editing this page.

I left the usual welcome message on your page, it's useful.

If you are interested in poetry, I also suggest you take a look at Template:Ppoem, that is as of now more or less the best alternative for formatting poetry.

If you want to get started, here's an poetry index, picked at random: Index:The Poems of John Donne - 1896 - Volume 1.djvu.

Of course, feel free to do whatever you prefer.

If you can specify what sort of books you want to do, I might be able to fish a file in the Internet Archive.

Cheers, — Alien333 (what I did & why I did it wrong) 09:27, 2 June 2024 (UTC)

Oh my! Much appreciated that you responded so considerately! I didnt expect it and you have launched my enthusiasm to pursue this endeavor! I will pick this up upon waking as soon i will be going to sleep, but certainly hope to be in touch with you more if you will find it comfortable and worthwhile to mentor me for a bit!

My favorite poet is Walt Whitman...if I must choose from many I love.

My gratitude to you !

ImaginarySusan! ImaginarySusan (talk) 09:39, 2 June 2024 (UTC)

Here's one of Whitman's collections for you: Index:Drum-Taps.djvu.

(I myself only went down the rabbit hole a few months ago). — Alien333 (what I did & why I did it wrong) 09:53, 2 June 2024 (UTC)

oh thankyou! I will check that out before sleep... and yes, I notic3d your 1 yr. anniversary on wiki was just two weeks ago!

My how far you've come! What an inspiration! ImaginarySusan (talk) 09:56, 2 June 2024 (UTC)

damn auto correct!

Im obviously not proofreading my messages to you...

..as " spell check overnights" was supposed to be oversights*!

Lol. ImaginarySusan (talk) 09:53, 2 June 2024 (UTC)

Oh, and also: you might want to create your user page, with a bit of information about yourself.

User pages are also often used to keep things (such as links) close at hand, since you can go to your user page from anywhere. — Alien333 (what I did & why I did it wrong) 09:30, 2 June 2024 (UTC)

Yet another reply: I recommend you read Help:Proofread.

I'm assuming you want to contribute, of you don't that's fine and then Help:Reading would probably be more appropriate. — Alien333 (what I did & why I did it wrong) 09:34, 2 June 2024 (UTC)

oh yes I will read everything you suggested. It may take me some time..lol

"Back in the day" i started my graphic communication career in typesetting, copywriting and the REAL old fashioned skill of original proofreading! I was very good...and to this day I don't casually read a thing without noticing typos, grammatical errors, and spell check overnights! (Notice the Oxford comma!) Lol.

I am in the NW Pennsylvania area of the US..and an artist, writer and night owl... I see you are in UK?

The morning bird songs are beginning here, as it is almost dawn. I will let you know once I've started reading, and if i get stuck understanding anything.

My best to you! ImaginarySusan (talk) 09:51, 2 June 2024 (UTC)

No, I'm not in the UK, I'm French, so my english is always going to be somewhere between british and american english. — Alien333 (what I did & why I did it wrong) 09:54, 2 June 2024 (UTC)

Eggless recipe book for cakes . . .Index

I wondered the same thing. I suspect it's something in the scan file causing the issue. --EncycloPetey (talk) 20:27, 7 June 2024 (UTC)

Said it in an edit summary for a near-null edit, it's the table { width: 100%; } in the index CSS that naturally causes the info to expand to fill the whole width, as it's a table, and then it's wrapped and it ends up under the image. If we'd put something like td { background-color:red; }, it would also have applied. I would call index CSS applying to default mediawiki layout a problem, but we do need it to apply to pages transcluded, e. g. for the TOC. Maybe we should open a ticket about this. Left a comment at WS:S#Index CSS applying to mediawiki layout to see if others might know a bit more about that — Alien333 (what I did & why I did it wrong) 06:22, 8 June 2024 (UTC)

Baltimore

Yes, for clarity I DO in fact own the city of Baltimore!!! 50.75.166.42 19:21, 14 June 2024 (UTC)

Oh stop it, will you? (WP vandal coming over here) — Alien333 (what I did & why I did it wrong) 19:22, 14 June 2024 (UTC)

Rossetti. Poems

We may need a versions page for this. The original was published in 1890, but there was an expanded 1891 edition. I do not know yet whether the 1901 edition that you are editing follows the 1890 or the 1891, or is further expanded. --EncycloPetey (talk) 19:56, 20 June 2024 (UTC)

Do we have the other editions, to check the difference?

If it helps, the 1901 one says "new and enlarged" and "First complete edition printed November 1890, Reprinted December 1890, January 1891, August 1891, 1892, 1894, 1895, 1896, 1897, 1899, 1901". — Alien333 (what I did & why I did it wrong) 20:13, 20 June 2024 (UTC)

From a quick look at the TOCs of the 1890 one and the 1891 one, they all look the same.

The 1890 edition was in itself already marked "new and enlarged", so I think all three are of (nearly) the same text, already expanded from some earlier collection of poems, maybe this 1866 one, this 1872 one, or that 1888 one. More likely, each edition expanded from the last one, since they all share the same beginning and some poems are added progressively at the end. — Alien333 (what I did & why I did it wrong) 20:40, 20 June 2024 (UTC)

New texts

Indeed, the problem has been corrected now. --EncycloPetey (talk) 18:38, 22 June 2024 (UTC)

(Note: that was four hours before you reverted, I think it's just {{spl}} that got you confused.) — Alien333 (what I did & why I did it wrong) 18:40, 22 June 2024 (UTC)

Poems, 1909-1925

This collection of poetry by T. S. Eliot was first published in 1925; here is a link to the 1926 reprint on IA: (external scan). The collection includes editions of some poems we already have, but also some that we do not. It is about 100 pages. --EncycloPetey (talk) 23:44, 10 July 2024 (UTC)

Will do, after finishing Index:Poems Hornblower.djvu. I intent to overwrite Poems (Eliot) for this, as it's unsourced and its contents are included in this 1926 collection. — Alien333 (what I did & why I did it wrong) 07:33, 11 July 2024 (UTC)

That's a different collection, with completely different poems. T. S. Eliot published a series of Poems books, starting with that one. Subsequent volumes had a year range as part of the title, and the contents were different each time. It would probably be better to turn that into a versions page as a result of the differences between the many editions. --EncycloPetey (talk) 14:51, 11 July 2024 (UTC)

The TOC of the 1926 book you pointed me to contains the subsection "Poems (1920)", that contains exactly the same poems as the other one (compare this and that). — Alien333 (what I did & why I did it wrong) 16:16, 11 July 2024 (UTC)

Subsection, yes, but there are also additional poems not in the 1920 edition. So the two editions are different from each other, yet both are titles Poems. And the 1932 edition contains further poems not in the 1920 or 1926 edition, and we will want to host the 1932 edition as well. My point is that we will eventually have additional editions, and the page Poems (Eliot) is the logical place to disambiguate those editions. So, rather than put the 1926 edition at that location, convert it to a disambiguation page listing the 1920 and 1926 editions, and providing us a place to also list the 1932 edition in future. The alternative is to have to redo all of the internal and external links the next time an edition of his poetry is transcribed here. --EncycloPetey (talk) 17:25, 11 July 2024 (UTC)

Ok then, I'll put it at Poems (Eliot, 1926). Note: All of these editions will anyways (I think?) be listed at Poems, so I don't know if it's worth putting a separate dab page at Poems (Eliot). — Alien333 (what I did & why I did it wrong) 07:48, 12 July 2024 (UTC)

@EncycloPetey: Done (though I'm not sure the titles, quotations and poems are positioned the right way). — Alien333 (what I did & why I did it wrong) 13:30, 15 July 2024 (UTC)

Template:Sandbox/PleaseDontDelete.css

I'm curious, if you move this page to your userspace, does it keep the "sanitized CSS" content model, or automatically switch to unsanitized CSS? It does the former for me, but since I'm an admin, I have the ability to change a page's content model, so it occurs to me that you might see different behavior. —CalendulaAsteraceae (talk • contribs) 06:30, 24 July 2024 (UTC)

Yep, that works, it has the right content model. Thanks! — Alien333 (what I did & why I did it wrong) 07:24, 24 July 2024 (UTC)

Glad I could help! As you can see from the deletion log for Template:Sandbox/styles.css, this is a useful (if slightly silly) trick. —CalendulaAsteraceae (talk • contribs) 20:57, 24 July 2024 (UTC)

I'd seen it already, but I'd thought it was just for testing. — Alien333 (what I did & why I did it wrong) 05:32, 25 July 2024 (UTC)

Poems (Shore)

Hi, You might want to take a closer look at the transclusion of the works in Part III. There seem to be bad interactions between 'ppoem' (where used) and your personalised version of it. Also between the 'pseudoheading' templates and normal ones (e.g. small caps). It looks like you probably need to use entirely one or the other, not mix and match. Chrisguise (talk) 06:04, 1 August 2024 (UTC)

My template only spits out a ppoem with some lines that have an additional styling, it's completely compatible as I've seen in my 78 other books where I've used it. I already saw and fixed an issue like that yesterday at Poems (Shore)/Olga, it's just caused by a mismatch of ppoem start/end across a page break, namely stanza/follow. It only happened in part III, because that's where the longest poems are, so more chances to mess up start/end's. — Alien333 (what I did & why I did it wrong) 07:26, 1 August 2024 (UTC)

All fixed now. — Alien333 (what I did & why I did it wrong) 07:39, 1 August 2024 (UTC)

There are places where lines starting with a character name (formatted using 'small caps') are right aligned, not left, and instances where the character name formatted with 'pseudoheading' are overwritten by the following text (i.e. the following text seems to be left aligned). Chrisguise (talk) 21:21, 1 August 2024 (UTC)

I've fixed already, it's caused by a lint error with an unclosed div. Did I miss some? — Alien333 (what I did & why I did it wrong) 07:47, 2 August 2024 (UTC) EDIT: I've re-read all of part III, and I still haven't found any left. — Alien333 (what I did & why I did it wrong) 08:01, 2 August 2024 (UTC)

Belgic Confession

This is a complete work. The volume it was transcribed from contains a set of documents pertaining to the Reformed Dutch Church in America. This will need to be moved to be part of the containing volume, and that will take some investigation to be sure everything is organized correctly. I am working on that, but am also plagued by computer issues today, which is hampering my progress. --EncycloPetey (talk) 18:41, 14 August 2024 (UTC)

Yeah, I realized after reverting, sorry. — Alien333 (what I did & why I did it wrong) 07:37, 15 August 2024 (UTC)

TIF files from TE(æ)A,ea.

As a heads up, TE(æ)A,ea. does book scans, and the TIFs are raw page scans, uploaded here so that the files can be grabbed, cropped, processed, and the resulting images then uploaded here or at Commons. In general, raw scans that are TIF format are not suitable for use as is, but are uploaded here temporarily. --EncycloPetey (talk) 18:29, 17 August 2024 (UTC)

Addendum: And sometimes the images need to be researched separately. So, for example, The Vampire by Summers is in PD in both the US and UK, but some of the illustrations are not. The frontispiece is a painting by a Hungarian artist who died in 1961, so his paintings are not yet hosted at Commons, because they are still protected in the EU by copyright. Illustrations that are works of art and photos of that art, can have licensing that differs from the book in which the illustrations appear. --EncycloPetey (talk) 18:38, 17 August 2024 (UTC)

Ok, thanks for the explanation! — Alien333 (what I did & why I did it wrong) 19:10, 17 August 2024 (UTC)

Frances Ellen Watkins Harper

We have Index:Poems (IA poems00harp).pdf that has not been started, but seems well within your personal sphere of activity. She is severely underrepresented on Wikisource. --EncycloPetey (talk) 19:22, 1 September 2024 (UTC)E

I originally intended to stop at a 100 of them and then go transcribe something else, I've already done 90 (91 counting T. S. Eliot, but I usually do women authors for the gender gap, so that one is a bit apart) and I have ten more in stock, but I'm continuously finding new ones I want to do. Will do, thanks for the suggestion. — Alien333 ( what I did
why I did it wrong ) 22:40, 1 September 2024 (UTC)

@EncycloPetey: Done. — Alien333 ( what I did
why I did it wrong ) 22:33, 2 September 2024 (UTC)

QuickTranscribe and other proofreading software

Sorry for the late reply, MediaWiki doesn't seem to have a very robust way to mark messages unread. Well, it turns out the document you found, User:SnowyCinema/QT.py, was a place I jotted ideas down for the project a long time ago, a document I completely forgot about and doesn't have any relevance to the code right now.

I point you to User:SnowyCinema/QuickTranscribe, the main project page, if you're interested in details. It's not completely up to date, and there are a few more features not mentioned there. I even was toying with poetry collections and anthologies very recently with QT (Fox Footprints, poetry; Lords of the Housetops, anthology).

I am extremely impressed by your work here with poems and your ability to just churn these out! I would love to collaborate with you. I'll work to get my code documented and cleaned up for you soon, and also would love to have a lot of this work we both did centralized in one place, like a frontend application. I'm getting to a point where I think I'm ready to come back to the project, so I appreciate you for giving me some motivation also!

We'll be in touch about teaming up in our vision to populate Wikisource ridiculously quickly! :) SnowyCinema (talk) 03:02, 5 September 2024 (UTC)

On poetry, taking a look at Page:Fox Footprints (1923).pdf/58, I noticed there seems to be an issue with styling (assuming it's generated auto), as it really doesn't match the scan (details).

On my churning them: very much related a) to my efforts to get good OCR before starting, I feel like proofreading time is directly proportional to OCR quality and b) to my scripts and {{tpp}}.

On QT (I don't even know how I found that page (: ), a wild thought, as I haven't even read the codebase (I intend to do so soon), but maybe I could lend a hand? I consider myself a decent Python and JS programmer, for the better or for the worse. — Alien333 ( what I did
why I did it wrong ) 18:50, 5 September 2024 (UTC)

Your feedback on the CSS work in that collection is valid. You'll probably notice my styling is usually not exactly perfect to the original because for one I'm no CSS guru, and for two I focus on getting work out faster, with a focus on content over exact styling because this proofreading work is already horribly time-consuming as is, even with these "extra tools" I created, let alone without them. But, I try to keep everything in CSS classes so they can more easily be changed if needed. I know "the rules say" you have to get it perfect, but I hope you understand why I make this "is it readable vs. does it look perfect" compromise. Proofreading a novel and a film a day or whatever, with a few extra hours of admin maintenance and QT coding etc. sprinkled in, was completely consuming my entire life as it was.

(That's not whatsoever an exaggeration by the way—Wikisource was a serious personal addiction issue if I may open up a bit. I was having trouble living. I'm wagering I'm balanced and stable enough to be able to continue this by now, however.)

ANYWAY, yes, going to do some work on documentation at the very least. I want to make this a collaborateable project. The one thing I will say is that the code I have is intrinsically not fully automatic. There are always edge-cases every couple of works that require some manual intervention, but overall it makes a whole bunch of the process much smoother. So when a frontend or more UX-friendly build is made, we can design it so manual intervention is easy. SnowyCinema (talk) 19:31, 5 September 2024 (UTC)

(I'll probably improve poem code so that it takes all poem pages in larger blocks so modification is easier. I designed it with defaults that are generally correct to early 20th century styling.) SnowyCinema (talk) 19:46, 5 September 2024 (UTC)

To be fair to you, in my 91 collections, I've only seen the same type of styling once (and yet, that didn't have {{sc}} on first words). — Alien333 ( what I did
why I did it wrong ) 19:49, 5 September 2024 (UTC)

Totally agree on the time-consumption side, but eh, what am I going to do with my life instead? Some occasional drama/wikistress/mistakes makes me step back enough so that it doesn't eat up the whole of my life (and the rest is devoted to programming anyways, so...)

On styling, I get your point, and I also like keeping things in stylesheets, but to me that's exactly the point of them, that it takes what, a few minutes, to look at the file and set up the styles? It's not like other stuff like header & footers, . . . vs {{...}}, which are more minor and time-consuming (I still do them, but I haven't yet gone fully "speed first"). It happens to everyone to have not exactly the same styling (primarily because publishers are apparently puzzle maniacs), but I think stylesheets are rare enough (once a work) and small enough (usually only 2-4 rules, at least for me) to be worth doing manually. — Alien333 ( what I did
why I did it wrong ) 19:47, 5 September 2024 (UTC)

@SnowyCinema: what I meant with the placement & margins etc of the title is that instead of doing

{{cblock|TITLE
{{dhr}}
{{ppoem|
text
of
the poem}}}}

TITLE

 

textofthe poem

you can just do

{{ppoem|TITLE

text
of
the poem}}

which gives basically the same result, sparing a template (the break is slightly smaller, but in my experience most of the time it's the right one).

TITLE

textofthe poem

— Alien333 ( what I did
why I did it wrong ) 21:37, 5 September 2024 (UTC)

43000th edit!

Just in case you didn't know, as of me typing this you are on 42,999 edits. So next one shall be 43,000th. Congrats. ExclusiveEditor (talk) 05:39, 8 September 2024 (UTC)

Thanks. (Looked at my ec after seeing my name in that banner thing, I suppose?) — Alien333 ( what I did
why I did it wrong ) 08:45, 8 September 2024 (UTC)

Yes. ExclusiveEditor (talk) 19:50, 16 November 2024 (UTC)

Poetry collection requests

Would it be too forward of me to give you requests for certain poetry collections I find here or there? I think you're quite well suited to transcribe these. They're annoying for me and I'm not too interested in verse honestly, but lots of disambiguation pages need blue links. Is a requests page in your user space warranted, that I can add requests to? SnowyCinema (talk) 22:52, 30 October 2024 (UTC)

Will do whatever's given to me, as long as it's not old enough to have ſ.

(It's incredible how far specialization can go, now I can do most of the poem formatting on a page by typing four characters and pressing one shortcut.)

As to where, you can just drop them here, I don't mind. — Alien  3
3 3 09:32, 31 October 2024 (UTC)

Alright, I'll put some here for now.

• War Drums (1899), a poetry collection by Louis Edward Scharkie (external scan). He was Australian and this is almost certainly his Findagrave: https://www.findagrave.com/memorial/159763780/louis-edward-scharkie - for the disambiguation page War Drums
• Cofachiqui, and Other Poems (1884) by Castello Newton Holford (external scan) - just for his author page
• Pebbles and Shells (1895), by Clarence Hawkes, a bit on the longer side... (external scan)

This is something to start off with. I would highly recommend a request subpage, because I'll find myself throwing a ton here (if you want). SnowyCinema (talk) 13:39, 31 October 2024 (UTC)

There it is, if you really want it: User:Alien333/Poetry requests.

Also, a side note, though I won't die on these hills: I tend to prefer works

• without watermarks, because those are always a bore
• that don't have already-uploaded scans (to be able to redo the OCR myself.)
• available somewhere else than at hathi's (I don't have membership and it's really a pain to get each page individually). (for War Drums I'm going to take (external scan), at IA).

— Alien  3
3 3 13:47, 31 October 2024 (UTC)

Yo, you're an absolute hero! Thanks! Seems like the guy died young, just after he got his first collection of verse out. This could've been one of the quintessential poets in Australia. Wonder what disease it was. Well, now his voice can be heard again!

Hopefully also the NaN problem isn't causing you too much trouble. In an Index page, next to the transclusion status ("Fully transcluded") there's a button that lets you check and see if all the pages are transcluded. This might help you find out if errors happen in transclusion in the future! Wow, great, clean, quick work, impressed as usual! SnowyCinema (talk) 18:35, 1 November 2024 (UTC)

I did use the transclusion button, but it's not perfect and it counts <pages> tag errors as transclusions. (the NaN was caused by an OCR error, when I reused the page numbers in my code.) — Alien  3
3 3 18:37, 1 November 2024 (UTC)

side note: I've been here for a year and a half, actively for a year, so sometimes I want to protest that I'm not that clueless, but I often discover things I should have known, the latest being that the "Entered according to Act of Congress", &c is actually copyright note, and not something added by the LOC. — Alien  3
3 3 19:23, 1 November 2024 (UTC)

Would it be okay if you used title case for the poems of the works I request? It will make it easier for me to disambiguate. It's up to you—but it's my personal preference and I'm only requesting it. Some things can make the casing iffy, like for example the novel Resurrection Rock is listed in some places as "Resurrection rock" (in sentence case), which is an issue because the title of the novel (being "Resurrectio Rock") is named after the title of the fictional rock in the novel which is itself a proper noun ("Resurrection Rock").

Similarly with Cofachiqui, and Other Poems/Grant county, it was named after a county in Wisconsin, which is (at least nowadays; I don't know if in 1884 this would have been valid) traditionally spelled "Grant County". When I make the Wikisource portal for that Wisconsin county, it would be nice if the work titles I list there match the casing of the portal, being Portal:Grant County, Wisconsin (in the future). Do you mind if I move at least that one to Cofachiqui, and Other Poems/Grant County? SnowyCinema (talk) 20:10, 2 November 2024 (UTC)

I generally stick to the case of the TOC, and when there is no TOC (or it's {{asc}}), as poem titles (and running headers for that matter) are most of the time all-caps, I don't have a way to determine the original title, so I choose to not make assumptions because title case is not applied consistently across the centuries and all over the world.

I make an exception for cases where I am sure that a word should be capitalized, mostly for proper nouns. Feel free to move Grant county, I wasn't aware of the custom of capitalising the word county in county names (not being american). — Alien  3
3 3 09:44, 3 November 2024 (UTC)

"He didn't write one poem titled "Pebbles and Shells", but fourteen (and not versions, all clearly distinct)"

Lmao. This is what I'm here for. SnowyCinema (talk) 19:18, 8 November 2024 (UTC)

Testing the DT API.

Testing the DT API. — Alien  3
3 3 15:34, 7 December 2024 (UTC)

Answering to myself (if it works)! — Alien  3
3 3 15:45, 7 December 2024 (UTC)

even that autoindents (moving up for testing, sorry of this pings). — Alien  3
3 3 15:53, 7 December 2024 (UTC)

Wikidata links

How are you figuring out which Portals need a link at Wikidata to Wikisource? I see you just linked seven of them. I had asked if someone could modify an existing bot to do just that. RAN (talk) 18:47, 22 November 2024 (UTC)

With the bit of code I linked to at WS:S#Qid. It is, in fact, a bot, these were only the test edits. — Alien  3
3 3 18:49, 22 November 2024 (UTC)

And to answer your question about the logic, I explained it in detail at the BRFA, there. — Alien  3
3 3 18:51, 22 November 2024 (UTC)

• Good stuff! It just found another 7. Does the bot check Category:Surnames for linking to Wikidata? The bot found individual news articles and portals, so far not surname categories. I only created the concept of surname categories a week or so ago, to link portals of people with the same surname. That way if you had two people with similar names, you could look at the category to work out who was the correct person. --RAN (talk) 17:37, 23 November 2024 (UTC)
• Can you fix an error I made. Wikidata item Q7344166 links to an article instead of Portal:Robert Ensko. The error needs to be corrected at Wikidata, I corrected it at this end. --RAN (talk) 14:38, 3 December 2024 (UTC)

  (I wasn't aware of the ... circumstance you mentioned at Mike Peel's talk.

  Now it makes more sense why you would ask for individual edits.

  Sorry, but I feel uncomfortable making possibly controversial edits for another user.

  This doesn't change anything as far as the bot is concerned, or sitelink corrections in general, as that is an uncontroversial task.)

  Question: what is supposed to be the point of surname categories? (I don't know any of the wider context around that story.) From what I can see, they just duplicate d:Special:WhatLinksHere/[id of name page]. — Alien  3
  3 3 13:50, 4 December 2024 (UTC)

The surname categories let you look to see if a person has an entry under a variation of the name. Someone might be "A.J. Smith" or "Allen J. Smith" or "Allan J. Smith" or "Allen James Smith" or "Allen James Smith I" or "Allen James Smith, Sr." or "Allen James Smith Sr."

• (Oh, and I did do that edit.) — Alien  3
  3 3 13:51, 4 December 2024 (UTC)

  Another question, related to your block, if you don't mind answering: people at WD consider that your creation of certain items was wrong (I don't know whether that was right or wrong and I have no intention to try and find out); couldn't you just comply, e.g. promise to not create any items that aren't immediately notable due to having a sitelink? Not being able to make any edits at WD, at all, is going to be a big obstacle to editing here (as most of our data is there, &c). — Alien  3
  3 3 14:07, 4 December 2024 (UTC)

• We already have a process for nominating entries for deletion at Wikidata. That process was skirted, I complained and asked for 5 deleted items to be restored, created by a third party. This all started earlier as a harassment campaign when I reversed an edit by someone with admin rights, they then retaliated by nominating the previous 400 images I loaded for deletion at Commons. When that was reversed they posted a message at Wikidata that someone should do the same there. I was permabanned out-of-process by an editor who had just one month of experience. We already have a 10 year old objective rule at Wikidata that the entries should have a "public and serious" sources which I abide by. If I agree to follow these new, vague, and subjective rules, they will just continue to harass me. The guy that nominated the 400 images will just delete whatever I add and ask that I be banned again under the new vague rules. Unfortunately there are just two bureaucrats and no Arbcom committee at Wikidata. Several people wrote me saying they were afraid of getting banned too if they supported my side. It is also crazy that the guy who created the entries that I asked be restored is still active. The whole project suffers since I would spend 8 hours each Friday adding Library of Congress images and create Wikidata entries for the people in the images. Same for the two local historical societies I belong to. I would scan and add the images and create a Wikidata entry for them. See for example: d:Q116700477 and d:Talk:Q106445178, that chart took me three months of research scouring historical papers in the archive. I stopped all three projects, and even if unbanned, will not start up again. Finding where I left off will be too difficult. --RAN (talk) 17:54, 7 December 2024 (UTC)
• Your bot would also be great at Wikiquote, there are also entries not linked to Wikidata. The problem there is that there is no backlink from Wikiquote to Wikidata for the bot to see. The site could also benefit from closer integration with Wikidata. I asked at Wikiquote:Wikiquote:Village_pump about adding the "authority control" and the "sister projects" template, but there are too few people contributing, no one responded. We could set up a test of the template with the backlink. --RAN (talk) 20:03, 9 December 2024 (UTC)

  The advantage of these specific cases of the header |wikidata= parameter is that we can be certain that the item matches the author (or at least it isn't our responsibility but that of the editor who added the parameter). For bot-volume editing, if the bot is going to make some decisions, I want to be sure that the error rate will be low. I will probably add the surname cats sometime soon, because there it is still pretty clear (there likely won't be two "instance of" "family name" with the exact same item name). But for other pages, e.g. authors, I'm not that sure, cf Elizabeth Gifford, there were plenty of WD items called "Elizabeth Gifford", but none the right one (born after, or died a while before, publication of work). If the bot went solely by item names, here it would have linked the author to one of the incorrect items. We can do a more complex algorithm, but it should be thought out carefully (the more steps there are, the greater the chance of error). — Alien  3
  3 3 20:32, 9 December 2024 (UTC)

Interest in being an adminstrator?

Hi Alien333, is adminship something you would be interested in? If so, I'm prepared to nominate you. Take some time to think about it and read up what it entails. Beeswaxcandle (talk) 06:56, 10 December 2024 (UTC)

I would be interested.

I think the main uses I would make of a mop, would be:

• The administrative backlog (e.g. edit requests tend to be dealt slowly, I'd keep an eye on that).
• I have also some technical knowledge (HTML, CSS, JS, and Lua), so I could help with that sort of stuff.
• (although minor) speedy delete myself the M2 pagemove redirects I leave behind instead of leaving more work to the admin team.

I am unsure of whether I am ready for the job, though I guess that'll be the community's role to decide, for these reasons:

1. I find myself clueless more often than I would have liked.
2. I have the impression I sometimes have some trouble communicating with other users.
3. Something that I should disclose, in all fairness, is that in about nine months' time I will start something IRL which will reduce my leisure time, so I won't be as active as I am today, though I won't go inactive.

On the other side, it could be said in my favor that:

1. In one year I couldn't know everything.
2. No one has ever mentioned that to me, so it's just an impression.
3. From reading around, admins have from time to time had periods of reduced activity and this was apparently not seen as too much of a problem.

I'd like your opinion on these three (possible) issues, or any other you have noticed. If, taking all of that into account, you think me ready, then I accept. (And btw thanks for welcoming me back in June of last year.) — Alien  3
3 3 09:29, 10 December 2024 (UTC)

1. Recognising that you're not sure is a strength in my view. Then asking before jumping into a solution is a pattern I see in you.
2. There are always some people with whom it is harder to communicate with—particularly as we're restricted to the written word.
3. Yes, there are times for all of us when RL gets in the way of doing what we really want to do. My own editing pattern has been very variable—partly depending on what works I'm focused on, but also what else is going on in my life. As long as the tools are being actively used and an admin is keeping up a minimum of 50 edits over 6 months, it's not a problem.

I'll go and do the nomination now. It is customary for candidates to confirm their acceptance of a nomination. Beeswaxcandle (talk) 07:46, 11 December 2024 (UTC)

HathiTrust scans

Hi @Alien333,

At some point, I thought you had figured out a way to get Hathi scans, but then randomly passing by your Poetry Requests page, it seems I might have been mistaken. Thus, I have uploaded File:From an Old Garden (Cloud).djvu and File:Travelling Standing Still (Taggard).djvu to Commons. If you would prefer the pdfs instead (to redo the OCR in some other fashion), I can also upload those. (P.S. Congrats on your admin nomination above).

Regards, TeysaKarlov (talk) 20:42, 15 December 2024 (UTC)

Well, until today, I hadn't. I hadn't tried, assuming that they'd somehow prevented just fetching the images (and I don't want to spend a single cent to fund and support these attempts to essentially paywall and privatise the public domain). But once you gave me the idea, I've been fiddling with it for a few hours and I managed to find a painless JS solution to do that (could also have clicked n times "save image as", but very time-consuming). Maybe I should write it or mention it somewhere, others must have asked themselves the same question. (On the admin nomination, it was very unexpected. I feel like some were passed over, that arguably are more knowledgeable than me but were never nominated, but eh, it's not my business, maybe they have issues I'm not aware of or they don't want to be admins.) — Alien  3
3 3 16:21, 16 December 2024 (UTC)

(For curiosity's sake: How did you download it? With the images I got, the end result is about twice as large as the ones you uploaded. If you have the "real" file, it maybe means that the displayed images are stretched, in which case I should zoom less before fetching.) — Alien  3
3 3 16:36, 16 December 2024 (UTC)

Hi @Alien333,

I downloaded them with institutional access, so no scripts necessary, just clicked download really (in pdf format). There only appear to be resolution options when downloading images, instead of a pdf, so maybe I should have done that and then converted to djvu. Not something I know a great deal about. If the OCR on my pdf to djvu downloads are still useful, feel free to ask for more in future, but if your JavaScript option is better, and simple enough, then maybe it is worth more of us using it (and describing somewhere).

Regards, TeysaKarlov (talk) 19:56, 16 December 2024 (UTC)

Thanks for the offer, but as it stands I think I'll do it on my own. (The OCR I use ([1]) has the advantage of recognising emdashes as emdashes & not hyphens. On the other hand, it only works for pdfs, but I haven't managed to get ocrodjvu working (python version issues).)

After a more detailed analysis of image quality: I've thought a bit more about it and I can bring it to the point where it actually downloads the best images available. These are not the pdf, but the individual images (with the full-res option on). For scale, my sketchy way of fetching the display images is about twice as large as downloading the pdf, and downloading the images individually (which is a tedious process but can in fact be automated much easier) is about 4/3 larger than my sketchy way. So, in the end, automating it the "right" way would be better than institution access. Will do tomorrow, and possibly in the coming weeks discuss this at WS:S.

Well, jokes' on me: downloading the high-res images gives a result the same size as the institutional pdf, except it's more blurred (may or may not be due to making one more conversion (jpg → pdf & pdf → djvu)). In the end, looks like the way I did it at first is better (I still have a suspicion of streched images, but that isn't much of a problem.) — Alien  3
3 3 20:59, 16 December 2024 (UTC)

Easy LST

Hi, just letting you know that "Easy LST" is turned on by default for new users and most of our editors have no idea that there is an alternative. Personally, I think it should never have been implemented, but I was a lone voice at the time. Beeswaxcandle (talk) 07:24, 20 December 2024 (UTC)

I know it's default (as I arrived after it was set up, so at first I had it too. It got me very, very confused when I transcluded my first work, after which I just turned it off when I found out that was possible). 100% agree with you on it being a bad idea.

The main issues I can see with the current way are:

a) new users tend to not know what it means, and think it just gives a limit that can both be used for start and end; so it's not easier to understand

b) it requires putting ## ... ## twice for a single section break; the manual way does the same, but I'm saying that Easy LST doesn't help type much less either

c) users even have to type more in cases where the begin is not right after the end (e.g. there's a separator that shouldn't be transcluded in either)

However, I get where it's coming from, as typing the <section ...="..."/> can be bothersome. I think there is probably something better to be salvaged from this. To me, the "right" way of doing it would have been just making a ##x#y## shortcut, where it expands to something in the lines of (in pseudo-code)

result = ""

if !x.match(/^\s*$/) (meaning is not of the form ## # ... ##)

then result += '<section end="'+trim(x)+'"/>'

if !y.match(/^\s*$/) (meaning is not of the form ## ... # ##)

then result += '<section begin="'+trim(y)+'"/>'

return result

While we're on it, I'd like to ask you a question about section titles. Personally, I think that individual labels that do not follow an easy pattern only take more time. In my first work, I labeled chapter beginning/ends with c[chapter num], and it was a nightmare to keep track of it when transcluding. Ever since, I always call the sections, a, b, c, d, &c in that order (so end=a, begin=b, end=b, begin=c, &c, and reset to end=a on every page). If this gets consensus, the above proposal could be even better, such as ### to put a end & begin, #### for only end, and ##### for only begin. It would need no more work.

What do you think of that? — Alien  3
3 3 11:27, 20 December 2024 (UTC)

I can't comment on your suggestion of the "right" way it should have been done. In re the choice of labels, I think it depends on the type of work being dealt with. If there are lots of small sections on a page, then the a, b, c, d, … works well. But if we're dealing with sections that go over multiple pages, then I tend to label them in accordance with their name. This means that when I'm transcluding, I don't have to think about I called the sections as they matches with the title of the subpage. In the end, I see it as an individual thing and would prefer not to dictate how to label. That's not to say that the Help: page can't have a suggested "ideal." Beeswaxcandle (talk) 00:48, 27 December 2024 (UTC)

Litanies

Hello! You helped me format litanies with ppoem. The litanies span several pages, so now I am trying to get them to format correctly. I followed the instructions at Help:Table#Spanning_Pages and Help:Page breaks and put the table coding in the footers, but now in the trancluded namespace, it places the tables (pages) next to each other, rather than one above the other. Will you please take a look and tell me what I am doing wrong? Blessed be God/Devotions To The Holy Name Before I put the table codes in the headers and footers, the transcluded pages had pages 2 and 3 at the left margin instead of block centered. Laura1822 (talk) 15:59, 26 December 2024 (UTC)

In a table, you have to delimit the rows, as it can't guess what row each cell should be on. To mark a new row, use |-, on its own line. See w:Help:Table for more details.

There was also another issue, that was my fault. In a table, the cells appear one above the other, but they are separared. With ppoem, this means that the poems are not actually joined, so the start=follow's, that expect a ppoem right before, caused this alignment issue you mentioned. However, the table itself is centered (margin:0 auto), so we don't have to join the ppoems. We can therefore remove the starts and ends. But then, they will all be centred in the same column, but they're not the same width, so the beginning of lines wouldn't ve aligned. Luckily, the table provides us with a way to align them together: stripping them of their native centering, which makes them all go to the left of the column, so they're aligned.

There's a last problem which needed to be taken care of. By default, there will be some space between rows of a table, and we don't want that. It can be suppressed by adding border-collapse:collapse to the table.

Should be good now, I hope the explanation was clear enough. — Alien  3
3 3 18:17, 26 December 2024 (UTC)

Wow, I cannot thank you enough. I don't think I would ever have figured that out by myself! I think I understand each of the things you describe. We'll see if I can duplicate it for the next one! I appreciate this so much. Merry Christmas! Laura1822 (talk) 20:15, 26 December 2024 (UTC)

Note: is is better to size the braces with "em"s, a unit that sizes in function of the text, so that with a smaller browser font size or things like that, it still looks right. The default line-height on Mediawiki is about 1.6em (why I don't know), so e.g. when you need a braces 18 lines high, you can write: {{brace3|calc(18 * 1.6em)|r}}. calc() is a CSS function, that permits using arithmetic operations in the code. This leaves cleaner code, in this case when you see calc(18 * 1.6em) you know it's 18 lines, whereas if it was just 28.8em it'd be much less clear. The line height is in fact closer to 1.5714, and when multiplying by large numbers you may want to use the more precise value, as at that scale the offset becomes visible, e.g. 100*(1.6-1.5714) = 100 * 0.286 = 28.6em, which is not negligible.

Also, something important: do not use curly quotes (” “ ’ ‘) in code. In the text, you may use whatever you want (as long as it's consistent in a work), but curly quotes are not recognized in code (this includes HTML and CSS). Any piece of code in which you use curly quotes instead of straight ones will either silently do nothing or send an error, depending on the language. — Alien  3
3 3 17:48, 30 December 2024 (UTC)

Thank you for both the reminder about avoiding curly quotation marks in code and for how to use ems for vertical space. I was wondering about what would happen with different sized fonts, and if dhr would do it correctly when, for example, an ereader offers choices to the reader about line height vs. how a browser or Mediawiki does it. Will including the calculation in the code this way work with ereaders too? Why doesn't dhr include the calculation? Never mind, I don't really need to understand these things at that level. Laura1822 (talk) 14:04, 31 December 2024 (UTC)

Well, in the end, {{brace2}} is preferable to {{brace3}} (because of the output markup), but you can give it basically the same argument (just remove the "em", it only takes numbers). dhr's units are in function of line height, therefore they should adapt. The calculation should work with ereaders too, normally (on export, many things are precalculated, e.g. in the PDF/Epub/etc each word has its placement precisely given).

I think it's good to ask questions, else you can never know. — Alien  3
3 3 14:11, 31 December 2024 (UTC)

I think I see what you mean. I had disregarded brace2 because I thought it was only for showing equations. So you're saying instead of brace3 with the calculation, just use brace2 with the number of lines and that will be better? It sure looks easier! Laura1822 (talk) 14:22, 31 December 2024 (UTC)

{{brace2}} should only be preferred to {{brace3}} inside a ppoem (this is necessary because {{brace3}} is a block element, which can't fit inside the ppoem lines, which are inline elements, see H:DIVSPAN). On a closer inspection, I'm afraid that {{brace2}}'s arguments don't correspond to anything that I can find out. Inside {{ppoem}}, you should use it and just test values until it fits. Also, brace2 doesn't accept CSS, so no calc(). — Alien  3
3 3 14:30, 31 December 2024 (UTC)

Okay, yes, I just tried brace2 in my sandbox, and 18 was too short. 23 looks right--but maybe that's only with my browser font preferences?

Also, the ppoem code end=follow was commented out, I assume in relation to the issue discussed below. So leave that out? Laura1822 (talk) 14:36, 31 December 2024 (UTC)

end=follow? where? — Alien  3
3 3 14:38, 31 December 2024 (UTC)

In my sandbox, before the brace code.

And to make sure, I should still use dhr to control vertical placement of the brace, correct? Laura1822 (talk) 14:43, 31 December 2024 (UTC)

On the end=follow, it's because of the ppoems not actually been joined. You shouldn't put the ends and starts whenever the ppoem is in a table (because ppoems in different tables cells are separated.).

Forget all I just said today about braces, I just found a way to make {{brace3}} compatible with ppoem. — Alien  3
3 3 14:45, 31 December 2024 (UTC)

LOL!!! I am so grateful for your time. Glad I could help (in my very small way) to niggle your brain to the best solution!

No ends and starts within tables, got it. Thank you! Laura1822 (talk) 14:54, 31 December 2024 (UTC)

Honestly, using the number lines needed (in the brace2 template argument) rather than defining the brace length by px (as I was trying to do it before) is MUCH easier for me. Can I do that now with brace3? Is that what you meant? Laura1822 (talk) 15:01, 31 December 2024 (UTC)

brace2 does not work in term of lines; or it only does at low scale. See e.g. [2] for ten lines. The calc way with brace3 is the way to go. (and, in general, avoid sizing things with px as often as you can.) — Alien  3
3 3 15:12, 31 December 2024 (UTC)

Thank you for the clarification! I thought you meant "don't bother with the calc thing" when you said "forget what I said about braces today." But now I see that you told me about the calc thing yesterday!

So, to sum up: 1. Use calc with brace3 to get the length of the brace, within dhr to get vertical placement. 2. No ppoem ends and starts within tables. 3. Don't forget to put a pipe-dash at the beginning of a table. 4. Don't forget that curly quotation marks foul up codes.

I just need to make sure I've got all the codes correct in my sandbox so that I can copy/paste. Laura1822 (talk) 15:33, 31 December 2024 (UTC)

Think that's it, yes. — Alien  3
3 3 15:43, 31 December 2024 (UTC)

One more thing! Above, you said, "There's a last problem which needed to be taken care of. By default, there will be some space between rows of a table, and we don't want that. It can be suppressed by adding border-collapse:collapse to the table." Where do I put that? At the beginning of the table? Does it need to be on its own line, or with its own pipe? Does it need to be in the subsequent page headers?

ALSO, THANK YOU for noticing and fixing the disambiguation problem! Laura1822 (talk) 16:29, 31 December 2024 (UTC)

It's in the table style, at the beginning, next to margin:0 auto, separated by a ; — Alien  3
3 3 16:37, 31 December 2024 (UTC)

Thank you! I have just (re)discovered {{brace table parameters}}. When I add it, it tightens everything up considerably, to I think 1em. So I don't have to use the calc parameter, can just put, e.g., "12em" --before adding this template I could not seem to elminate the space between pages, and I am pretty sure it was the braces that were forcing too much space.

So my next question is, when using that template, do I still need the border-collapse:collapse in the table style, or any table style parameter at all? Or the style=margin:0 parameter within ppoem? Please check my work at Blessed be God (Callan)/Devotions To The Holy Name and let me know if it looks right to you.

Edited to add: I just noticed that it now is flushleft instead of block centered, so maybe the {{brace table parameters}} has a conflict? Laura1822 (talk) 21:36, 31 December 2024 (UTC)

What is? — Alien  3
3 3 07:34, 1 January 2025 (UTC)

Blessed be God (Callan)/Devotions To The Holy Name when using the {{brace table parameters}}, as I mentioned just above ("Thank you! I have just (re)discovered {{brace table parameters}}. . ."). Laura1822 (talk) 15:13, 1 January 2025 (UTC)

Indeed, conflict with {{brace table parameters}}. That template, though, is I think useful only for {{brace}}, and useless for {{brace2}} and {{brace3}}. Indeed, its purpose is for images to butt together seamlessly, and the other two templates use only one image for the entire brace. — Alien  3
3 3 15:17, 1 January 2025 (UTC)

I added the template, and it "worked" with brace3, in that it removed the significant vertical space between pages, reduced the amount of space between each line (from 1.6em to 1em, it looks like), which made it much easier to figure out the length of the brace (in lines/ems) (though it was still not perfect over many lines). It allowed the ends of the braces to abut together very closely. Before adding the template, the calc and dhr were getting very complicated as I tried and tried to eliminate the gap by adjusting the length of the braces, then the vertical space between/above with dhr to get them into the right place. The only issue I see as a user (not a coder) is that the table is no longer block centered.

I did NOT test removing the other table style parameters from the beginning/headers of each page after adding the {{brace table parameters}} after them on each page.

This has probably reached the limits of what is possible. I will do it in whatever way to tell me is the best practice. I am grateful for your skills and your patience! Laura1822 (talk) 15:43, 1 January 2025 (UTC)

Also, did you intentionally put calc(18 * 1em) on that page (the first one)? As a reminder, line height is 1.6em.

Please do tell me if a calc(18 * 1.57em) brace is not 18 lines high for you, as this means that I was wrong in my assumptions (namely the one that line-height is 1.6em on all platforms).

Note: this may be due to the fact that {{brace table parameters}}, intended for situations where there is one {{brace}} parameter for every image segment this high: , supposed that there would be one row per line, and so reduced line spacing. — Alien  3
3 3 15:34, 1 January 2025 (UTC)

Sorry, I did not see this before my previous reply. Yes, the template reduced the line height from 1.6em to 1em so it simplifies the calc. But then I could eliminate the calc because I could just put the number of ems (lines) instead of the calc. (Before we started this I didn't know I could do that at all, that's why they were specified in px, as he original creator of some of the pages did them.) Laura1822 (talk) 15:46, 1 January 2025 (UTC)

The calc(18 * 1em) is 18 lines but now there is an extra line between the pages when it is transcluded. Laura1822 (talk) 15:49, 1 January 2025 (UTC)

Actually, it looks like the brace is longer than 18 lines, which appears to be causing the extra space before the next page. But I tried tweaking these lengths on each page (then having to adjust their placement vertically) and got into decimals that were far more complicated than is useful. Laura1822 (talk) 15:53, 1 January 2025 (UTC)

It looks different on each page edited individually vs. the transcluded three pages together. Laura1822 (talk) 15:55, 1 January 2025 (UTC)

(That's just because {{brace table parameters}} was still in the second & third pages' headers.) — Alien  3
3 3 15:56, 1 January 2025 (UTC)

(And because of something else: actually, vertical-align:top needs to be on the row separators, not the table opening; this was probably one of the things that made weird things.)

I'm going to do the three pages the way I would've. Can you tell me if it makes sense to you? — Alien  3
3 3 16:01, 1 January 2025 (UTC)

Have done how I would've these pages, would like your opinion/feedback on that. Is it clear and easy enough?

1.6 is a bit too much, and the value is closer to 1.5714, but if you find yourself in a situation where your brace is slightly too long, just use a smaller value, e.g. 1.55 (I used that on the second page.) — Alien  3
3 3 16:09, 1 January 2025 (UTC)

Thank you very much. I have not looked at the code yet. Here is what I see on the transcluded pages:

First page: Bracket looks one line too long, so there is a one-line gap between the first and second pages.

Second page: First bracket looks one line too long, so takes in the first line that should have no bracket (i.e., "Be merciful, spare us, O Jesus."). The next bracket starts one line too low, and ends two lines too low, so that there is an even bigger space between pages 2 and 3.

Third page: Bracket looks one line too long, so takes in the first line that should have no bracket.

Also, I was just looking at {{dhr}} and it says that it uses 1.4em, not 1.6 ("n the default Vector theme"). So that is probably one reason why the dhrs and brackets and lines do not line up. I was using parameters for dhr with decimals to get it to work, and it was very hard to get just the right placement. I was wondering (hoping) if there is another template that will allow vertical spacing by lines/ems instead, but there doesn't seem to be. Laura1822 (talk) 16:21, 1 January 2025 (UTC)

(I don't know of another vertical space template, though you could use {{vrl|{{em|...}}}}, where ... is the number of ems.)

Well, sorry, I think we've reached the limits of available technology and/or my knowledge. With web styling, sometimes we can't win. Maybe someone else could've helped you better.

I suppose this is where it ends. This is probably a browser issue. — Alien  3
3 3 17:00, 1 January 2025 (UTC)

You've helped me TONS, and I am very grateful for your time, expertise, and patience! I have learned a LOT from this experience, all of which will remain valuable to me.

If I fix the lengths and spacing so that it looks right to me, will you tell me if it looks mis-aligned/too short to you?

BTW I think "vertical-align:top needs to be on the row separators, not the table opening" is indeed fixing some of the mis-alignment for which I was trying to compensate.

Happy New Year! Laura1822 (talk) 20:47, 1 January 2025 (UTC)

You're probably tired of hearing from me, but I wanted to let you know that I experimented with skins and the skin I was using, Monobook, was apparently causing the display discrepancies. It looks much better in either of the two Vector skins. (The newer Vector won't work for me because of a visual impairment I have which requires black backgrounds and lighter text, and none of the buttons show up.)

So anyway, now on the transcluded pages I see much smaller (and therefore tolerable) gaps, and the only big issue (which I did not mention before) is the vertical text which is misaligned. But if we're done tweaking the brackets, I can fix that now.

I have learned a lot from this experience (which has led me to revisit a lot of formatting templates, and finding many helpful ones) and I thank you. I'm only sorry that my questions apparently came at a bad time for you. Have a great day! Laura1822 (talk) 14:42, 2 January 2025 (UTC)

If you use monobook, that explains a lot of stuff. In general, Vector 22 (often V22 for short) is now (after a long story) the default skin here, so content should work under that. In most conditions, what works under V22 also works in V10 (the older skin), which is not always true the other way round.

For black background and lighter text, there are multiple dark modes. I don't recommend the V22 so-called "feature-level" dark mode (the one that pops up when you add the skin), as it makes a lot of stuff hard to see. What I use is the invert-style dark mode, where the brightness of everything is just flipped through CSS. I encourage you to try it with V22. The steps would be: 1) select V22 as skin; 2) uncheck Preferences > Appearance > Skin preferences > "enable limited width mode" 3) add @import "https://en.wikipedia.org/w/index.php?title=MediaWiki:Gadget-dark-mode.css&action=raw&ctype=text/css"; to your common.css or your global.css. You may also want to add a.new:visited { color:#A55858; } table { /*remove light gray background */ background:transparent !important } html .mw-file-description img[src*="svg"] { /* rm black bg by import */ background:none }, to fix a few things that import does, though it's a question of personal choice.

Regarding a bad time for me, I don't know if you could call it that. I did a lot of stupid stuff. It happens (not that it's not my fault; merely saying I should have been on the watch for that, and prevented it). Now I've got to get back up, learn something out of that, pay more attention, maybe step away from those things for some time, proofread, read up more, and try to do more good than bad.

There is never a wrong time for questions. If we don't learn, we can't progress.

Anyway, nice to have helped. (I don't think I told you, but with poems with braces and vertical text across multiple pages, you didn't choose yourself a piece of cake.) Good day to you too. — Alien  3
3 3 15:45, 2 January 2025 (UTC)

Thank you for the code! I copied it into my common.css but does any of it need hard line breaks? (Feel free to edit it.) (I don't really understand css.)

Part of the problem is that my browser has a dark mode and it takes some colors, etc., from my Windows setup, which is an inverted white on black high contrast theme.

So far the main thing that has changed is that the buttons are all outlined now, but they are empty at the top of the editing page. (At the bottom of the editing page, I do see publish, cancel, etc., but I do NOT see colors or popups for the Page Status radio buttons.) Also at the top of every page for things like Preferences and Notices. I do see text for some (but not all) of the navigation links at the top (e.g., History). (One reason I was using monobook was that it was more text-based with fewer button icons. Is there a way to tell Vector22 to use text instead of buttons?)

Then there are all the problems with page status colors, such as on Index pages. I have sort of solved this problem by switching to a secondary theme in Windows which has a few more colors. Some programs--not just Wikisource!-- simply will not work with standard high-contrast inverted colors.

And AHA! I just discovered that if I switch to my secondary Windows theme, the buttons all show up (properly inverted). So something in my browser (or Windows) must be overriding them. BUT now the insert-markup box has a bright background, and the buttons below it above the editing box are blank. I will play with the themes and see if there's a simple fix.

So you see, everything is a trade-off. Several years ago someone wrote me a lot of css to make things workable for me in WS, but with changes over time to how Windows does colors, how browsers do colors, and WM skins, most things eventually looked better without the css. So fiddling with it is probably more trouble than it's worth. It'll just break with the next update.

re: the other stuff, I can recommend to you a couple of prayers for humility. There's even a litany!  :) Laura1822 (talk) 19:14, 2 January 2025 (UTC)

I changed a browser setting and I can now see all the buttons! Yay! But there are still a couple of things with light backgrounds: the Wikimarkup box on the editing page, and behind the page status at the top of a page. When I switch back from my secondary Windows theme to my preferred theme, those bright backgrounds are fixed but the buttons disappear again. So nothing's perfect! C'est la vie! Laura1822 (talk) 19:34, 2 January 2025 (UTC)

Line breaks in the code would help for readability, I didn't put them in because DiscussionTools has an issue with multiple-line stuff. (No one can edit other's CSS.)

You should close the comment at the end of what is currently line 24.

I suppose textarea { background: #FFF; color: #000 } could help for the editing box. Can you tell me which other thing has a light background (if possible, inspect the html and give me a distinguishing feature, such as an ID or the classes.) — Alien  3
3 3 08:18, 3 January 2025 (UTC)

I am sorry to report that suddenly all my wikimedia pages were bright white, and after much testing (in common.css, Windows, and browser, including a full reboot), I finally reinstated the browser setting I have long used, which is "Force websites to use a dark theme." Now all works as before. The css code you provided is I suspect redundant (or being overridden by the browser setting). I am now using the older Vector skin and that seems to be the best compromise for me, since the spacing is now correct (for the brackets and such) and I can see more buttons than I used to, plus I actually prefer the layout with more text links instead of buttons, and the the sidebar down the left which I like. Thank you for your efforts! I think it was worth it. Laura1822 (talk) 16:24, 3 January 2025 (UTC)

Well, whatever works is good. — Alien  3
3 3 16:59, 3 January 2025 (UTC)

Congratulations, you are now an administrator on English Wikisource

May you make great use of the tools. Cheers! BD2412 T 00:25, 27 December 2024 (UTC)

Adding my own congratulations. If you have some language abilities beyond English, please add them to your line in the Table on WS:ADMINS. Beeswaxcandle (talk) 00:40, 27 December 2024 (UTC)

(Wanted to answer, forgot) Thanks for the congrats! noted the language abilities there. — Alien  3
3 3 18:25, 28 December 2024 (UTC)

I also do congratulate you very much. We have really needed such reinforcement :-) --Jan Kameníček (talk) 23:20, 28 December 2024 (UTC)

Is is correct?

I added the references to the text of 1844:

https://es.wikisource.org/wiki/Discusión:Un_puesto_de_chía_en_Semana_Santa

Then how the text is added to translate namespace?

--Rauzoi (talk) 21:31, 28 December 2024 (UTC)

Sorry, forgot to answer. As far as sourcing is concerned, this should be enough (though if/when WS:T gets adopted as a proposal, which could be pretty soon, it will require scan-backing.)

You still need to add a license to the Spanish page, to make sure that legally we can host it. It should be one of this list. — Alien  3
3 3 16:11, 29 December 2024 (UTC)

Congatulations, Broke 600 pages wiki on you did..

https://en.wikisource.org/wiki/Special:LintErrors/html5-misnesting -Mostly it's due to trying to put DIV based tags insisde P tags, which is bad HTML. ShakespeareFan00 (talk) 18:29, 30 December 2024 (UTC)

Reverted back to previous version. Now to figure out what went wrong.. ShakespeareFan00 (talk) 19:21, 30 December 2024 (UTC)

When we look at what you reverted, it is clear that this version of {{ppoem}}, (just like the old one), never adds divs inside stanzas. These divs can come only from the user input. Now, still just looking at the code, we can see that all user input that is interpreted as html is put inside the lines, which are themselves spans (and already were before). Thus, all these pages already were putting divs inside spans, and the problem is these pages, which were already lint errors, and misusing the template. Therefore, this change to the module was not the cause of the lint errors, so I will re-revert. The thing to do, here, would be to fix these 600 pages, which are probably (from experience) using block templates inside ppoem. Leaving it live is moreover useful, if not necessary, to know what the issues actually were.

I would like to remind you that lint errors are not breaking issues, so I would appreciate it if you'd be a little less ton-o-brickey. Thank you.

I must say that it comes across as rather:

• dismissive, to blame someone for not testing enough without trying to see what tests were done. I have tested this code, on three computers, on five desktop browsers, on mobile, and in exports (which is all written at WS:S#Poem formatting). So I think I have done my duty for testing.
• trigger happy, to revert apparently without looking at what you're reverting.

I am sure that this was not your intention, merely stating that this is what it appears as. No offence meant, and I hope none taken.

Regards, — Alien  3
3 3 08:31, 31 December 2024 (UTC)

Taking a look, suspicions confirmed: {{FI}}s, other {{ppoem}}s, {{rule}}s (all block elements), &c can indeed be found inside ppoems in these pages. — Alien  3
3 3 08:58, 31 December 2024 (UTC)

Ah, found what makes the issue slightly more proeminent: the p tags autoclose when the parser arrives to their invalid (because block) child nodes, whereas the div didn't. The pages that use block elements inside are still the problem. — Alien  3
3 3 09:31, 31 December 2024 (UTC)

(Now sorting through them, appreciate if you and/or @Zinnober9 could give a hand to solve these invalid invocations. Useful info for this task: {{rule|xem}}s can be replaced by <>{{bar|x}}). — Alien  3
3 3 10:00, 31 December 2024 (UTC)

What would be even more impressive is changing ppoem so

<|:x> <||:x> and <*:y> work as single line prefixs to add rules with needing to break out of ppoem.

Or make {{rule}} 'compatible'. .. The revert was because of the sheer number of pages broken. It's never personal, but the types of interactions concerned could have been uncovered during testing (albiet for them specifcally.) ShakespeareFan00 (talk) 10:40, 31 December 2024 (UTC)

Can you explain <|:x> <||:x> and <*:y> I don't see what you mean. As in native rule lines? Maybe I could, going to think about it. Note: {{***}} is already compatible, as it uses spans.

As-was, {{rule}} already had a bad interaction with {{ppoem}}, because the hanging indent caused rule to go 4ems to the right. So making {{rule}} compatible would be complicated.

Even if we leave the rules aside, there are plenty of simply wrong usage in these 600 (most notably the FIs (all I've met so far are in fact useless, putting {{FI|filename.ext|size}} when <>[[File:filename.ext|size]] would have sufficed), the nested ppoems, and using {{hi}} instead of {{hin}}). These should get rid of in any case. I've done some 60 pages, I'd appreciate if you could help. — Alien  3
3 3 10:47, 31 December 2024 (UTC)

Yes, I was asking for native rule lines in {{ppoem}}, Also ppoem recognising a line is in fact a block insertion (like FI and nested poems) would be reasonable.

The rule interaction, wasn't a problem when using DIV based stanzas, HR in P isn't allowed as we've identified. ppoem in ref inside ppoem, would need Mediawiki to actually support block based footnotes (something that's been a long standing issue for at least a DECADE). ShakespeareFan00 (talk) 10:57, 31 December 2024 (UTC)

Ppoem inside ref inside ppoem perfectly works. What doesn't is ppoem inside ppoem directly. — Alien  3
3 3 11:00, 31 December 2024 (UTC)

Prior to your template changes, the interactions were masked. You are welcome to ask for assistance, but it needs a clear change of what to migrate.

Annoyingly the test cases I added recently, aren't necessarily showing the same 'bad interactions' as lints for some reason. ShakespeareFan00 (talk) 10:57, 31 December 2024 (UTC)

The issue is stanzas autoclosing before the rules. If the rule was mid-stanza, this problem is more visible than if it was end-of-stanza. Personally, I'm seeing the same things in the lints and on the testcases.

On what to migrate: simple {{rule}}s to {{bar}}s, {{FI}}s to normal images, {{hi}} to {{hin}}. Nested ppoems, when used to mark a differently-aligned stanza, should replaced by their content, and that content indented to match the scan. These cases are already at least 80% of it. — Alien  3
3 3 11:04, 31 December 2024 (UTC)

For me, the lints aren't being reported on the testcases I added. But doe show up when I use Special:ExpandTemplates (Sigh. If adding testcases doesn't actually report a limt that DOES showup eselwhere.)

As I said when Stanza's were DIV based , these interactions were masked. ShakespeareFan00 (talk) 11:14, 31 December 2024 (UTC)

Some, yes, but not the 4em one. I've seen that one for a year and a half (doesn't always happen). — Alien  3
3 3 11:15, 31 December 2024 (UTC)

For some of the interactions you've identified, there isn't a clear migration, so I'm going to go back to fixing unclosed italics rather than assist in migrating {{ppoem}} due to changes in a template that wasn't broken pre change as far as I can tell. I strongly suggest you testcase the interactions you've identfied, to resolve the migration path. Are you also checking each usage of ppoem on Pages for custom styles that relied upon the div based vs P based behaviour? ShakespeareFan00 (talk) 11:30, 31 December 2024 (UTC)

Already did that, no one used div.ws-poem-stanza. — Alien  3
3 3 11:30, 31 December 2024 (UTC)

@ShakespeareFan00: Well, no html5-misnesting in contentspace anymore. that was faster than I thought it would be (partly because, as I discovered, 2/3 of it were me last year, doing specific mistakes over and over again that could be fixed with a regex). A question, since you're the linter expert: I know it's not instant, and some trickled in for a good two hours while I was fixing it, but how much do you think is yet to appear? At my (uninformed) first glance, it looks like it's over. — Alien  3
3 3 12:54, 31 December 2024 (UTC)

Template:PING - Should have cleared most of them. Not sure how many will trickle through in the next round. Also manged to fix some splt-table references, by eliminating the splits. Any chance you could take a look at the remaining 10 or stripped tag lints, Most of the remaining missing tags are essentialy mismatched formatting (and mostly on un-proofread pages.). If you can also take a look at some of the High prority lints, I am not able to edit as I don't have admin powers, much appreciatedShakespeareFan00 (talk)

(Note: I added a warning at {{ppoem/doc}} to not use block elements inside, hopefully will help.)

None of the high-priority lints appear in contentspace, apart from Duplicate IDs. That one is complicated. Has plenty of causes, the largest of which seem to be:

• Something, I think MediaWiki:Gadget-PageNumbers-core.js, which gives, as IDs, the page names as given in the index pagelist. This is what allows links of the form [[Work title#page number]]. To solve this, we could either add something to the IDs to keep them unique, but then they'd lose their purpose, or remove duplicate IDs altogether. Need to be careful around this.
• Templates that add an IDs to the result, depending on the argument or not.
• Hand-chosen duplicate IDs.

Tbh, I don't see how I can help you on this. If you have ideas, I'm open to suggestions. — Alien  3
3 3 15:02, 31 December 2024 (UTC)

Broken redirects

Shouldn't these have gone along with the targets:

• What can I upload to Wikimedia Commons
• 2004 Wikipedia Press Release

? -- Beardo (talk) 13:49, 31 December 2024 (UTC)

Indeed, thanks for noticing. Is there a way to find them, so I can keep an eye on it? I often do mass deletes when closing WS:PD, and during such I'm likely to miss redirects. — Alien  3
3 3 14:08, 31 December 2024 (UTC)

Old (and some newer) formatter codes - possible for short codes in ppoem...

code	Function
{bb} or (bl}	Embolden
{it} or {il}	Italcize
{ni}	Not italic
{nb}	Not bold
{bu} or {ul}	Underline
{ds}	Set doublespace (line spacing}
{cs}	Set 'compacted' (line spacing}
{tt}{mono}	Monospace font.
{ansi}	ANSI/VT type font ( intended for program output screens if those get transcibed at some point.) Monospce, Green on black, fixed 80 character width)
{cc}{888}	200%, high contrast - (intended for film dialouge,transcription. Page 888 used to be the UK teletext page for subtitles.)
{ls1}..{ls5}.. {lssquish}.. {lswide}	Letter spacing in ems , and commonly used variants.
{lv}{sp}{dhr}{nil}{blank}	Insert blank lines.
{sc}	Small caps.
{grc}	Ancient greek ( for single line/stanza - ppoem has lang attrib for entire passages)
{he}	Hebrew (for single line/stanza - ppoem has lang attrib for entire passages}
{ang}	Old English
"{ex}chequer" {court}{abbey}	"Court" hand (essentialy a formatter code to use Junicode font (via ULS if needed}} to accomodate 'recordtype' glyphs for scribal contractions, see examples used in Statutes of the Realm and related works..
{hl1} (hl2}	Heading levels- 1to9 - (Which a user will have to specify in Index styles for a work).

There may be more, (And if implementing I strongly suggest having a seperate /data module/stylesheet from the LUA scripts!/

I'm not sure if the ppoem formatter can be made to an 8-bit error hexdump type format though.. maybe thats what Syntaxhighlighting is for..:) ShakespeareFan00 (talk) 17:23, 31 December 2024 (UTC)

/* I think it's not a very good idea to have 2 two-letter aliases for the same class (risk of confusion). */
.ws-poem-bb /* to add as alias to .ws-poem-bold, {bl} duplicate */
/* {it} is already there, {il} duplicate */
/* {ni} would correspond to {{fsn}} and {nb} to {{fwn}}, so: */
.ws-poem-noitalics, .ws-poem-ni {
  font-style:normal;
}
.ws-poem-nobold, .ws-poem-nb {
  font-weight:normal;
}
/* might as well add {{fvn}} */
.ws-poem-font-variant-normal, .ws-poem-fvn {
  font-variant:normal;
}
.ws-poem-ul {
  text-decoration:underline;
}
.ws-poem-doublespacing, .ws-poem-ds {
  line-height:200%;
}
.ws-poem-compact, .ws-poem-cs {
  line-height:95%;
}
.ws-poem-monospace, .ws-poem-mono, .ws-poem-tt {
  font-family:monospace;
}
/* Don't think that {ansi} is going to be useful
 * for {cc} and {888}, not all film use poems, and the film templates already have styles
 * how much would fall under "commonly used variants" of sp? if there are a lot, might be better to use {{lsp}}
 * (gave me the idea to add a {{lspn}} equivalent:) */
.ws-poem-letter-spacing-none, .ws-poem-lspn {
  letter-spacing:normal;
}
/* a single blank line is a stanza break, and multiple should be a larger stanza break. Here are classes (to be applied to a stanza) to have larger stanza breaks: */
.ws-poem-mb2 {
  margin-bottom:2em;
}
.ws-poem-mb3 {
  margin-bottom:3em;
} /* Here, mb''x'' is for margin-bottom-x. Should avoid having a lot of these classes, users can define beyond 3. If this other spacing is used consistently in the whole work, just add .ws-poem-stanza:not(:last-child) { margin-bottom:''whatever'' } to index CSS.
 * sc already there
 * I'm afraid that with simple CSS we can't change lang attribute (would need a bit of hardcoding)
 * Is anyone really going to use nine levels of headers? Anyhow, anyone can already use whatever classes they want (the set is not restricted, if you want you can add {xkcd} to a line and define that.) */

— Alien  3
3 3 18:17, 31 December 2024 (UTC)

Apologies for being an idiot

@ShakespeareFan00, @RaboKarbakian:

I am the worst of idiots. I was completely, totally wrong. I owe everyone involved apologies, for causing unnecessary conflict through my stupidity, and for acting like an moron. (To SF00, more specifically: sorry for recruiting you into this mess.) I ought to go hide in a hole of shame and never come out.

My edit was the problem, because the line spans are not inline. They have display:block. Which is why the lint errors said "span". And why SF00 first mentioned div-in-p errors, as a blocked span is essentially a div. This display:block was also written in the style sheet, and therefore on ppoem in the browser console. I looked at both at least twenty times each in the last month. I do not know how I managed to not notice this whole thing as it was damn staring into my face. I found out because InductiveLoad mentioned that spans were blocked on his talk page in 2021. (On realizing, kinda wanted to ask, why'd no one correct me? But you did try, both of you, although you didn't know, or didn't manage to make me understand (wouldn't be your fault, rather mine), the exact reason why I was wrong, and so I wasn't convinced. I guess this is the problem with being too confident. Anyhow, the responsibility is all mine.) I suppose it only signaled on 600 pages (as opposed to tens of thousands) because a span-display-block that has only span children is considered in some step of the linting process to be an inline element, and therefore to not be an issue.

Now, for consequences of my actions. Of these 600-odd edits (counts approximate, from memory):

1. 200 were harmlessly moving a pipe after instead of before a custom rule. They're pointless, but harmless.
2. 50 were removing templates, and replacing them by something not worse ({{hi}} to {{hin}}), or a bit better ({{FI}} to simple images, as FI has no use in ppoem).
3. 300 were replacing rules by bars. They could be characterized as either changes slightly for the worse (semantic-wise), or slightly for the better (because of the 4em shift issue).
4. 50 of them were changes definitely for the worse (such as splitting a {{ppoem}} where something that couldn't be included was used, e.g. a double rule).

I think that reverting:

• the first and part of the second would be as pointless, and harmless, as these edits themselves.
• the rest of the second would be slightly for the worse, but not a great lot.
• the third group will be done, because one important issue with it is that, though it was not the intention, it did end up enforcing one side (as the pages already using {{bar}} did not appear in the list), for no valid reason.
• the fourth group will be necessary, as it's a worse result for no valid reason either.

I will reread all of these 600 edits and revert those that need to be.

As the proverb goes: sow the wind, reap the whirlwind. — Alien  3
3 3 20:35, 1 January 2025 (UTC)

This wasn't addressed to me, but may I jump in here and say:

. . . dimitte nobis debita nostra, sicut et nos dimittimus debitoribus nostris. Laura1822 (talk) 21:17, 1 January 2025 (UTC)

Done. For information, I had to revert 94 of them (The first group was a lot larger, and SF00's were already in great part reverted). — Alien  3
3 3 08:54, 2 January 2025 (UTC)

Line based ppoem formatting..

Page:A History of Domestic Manners and Sentiments in England During the Middle Ages.djvu/425 ShakespeareFan00 (talk) 11:14, 4 January 2025 (UTC)

Work around for non italics in other wise italicised poem :).. If there's an even quicker way , LMK. (Like perhaps marking the non-italics only with italic markup. (like the approach taken with Italic Block and other templates :) ShakespeareFan00 (talk) 11:17, 4 January 2025 (UTC)

(It hurts my head a bit to see '' in {fsn} in {it}.)

That page had an issue because an apostrophe line 4 was replaced by a ''.

What I'd do, tbh, is just use the inline {{fsn}} template. I did that at that page. What do you think? — Alien  3
3 3 11:24, 4 January 2025 (UTC)

Works for me, Thanks. ShakespeareFan00 (talk) 11:46, 4 January 2025 (UTC)

(To clarify, just to be sure, italics in {fsn} in {it} does work, it didn't because of the unclosed italics above.) — Alien  3
3 3 11:49, 4 January 2025 (UTC)

Author:Tilman Bayer

I note that you deleted the only work by the author following the decision to do so. Do you think that the author's page should remain ? -- Beardo (talk) 04:20, 5 January 2025 (UTC)

No, sorry, I missed it. I should pay more attention. (I intend to write up something one of these days to automate finding a list of pages to delete (pending human approval, ofc)). — Alien  3
3 3 06:49, 5 January 2025 (UTC)

Thanks. -- Beardo (talk) 02:16, 6 January 2025 (UTC)

Rapid transcription tools - are they available to anyone, and if so, how?

Hi, judging by the number of works you add to the home page, you clearly possess the means of rapidly generating and transcluding output from texts. While I do a lot of old texts, for which OCR produces at best only middling results, I also do some more modern stuff with cleaner typefaces, which seem to be what you focus on. I'd like to have a go with whatever these tools are, and was wondering if they're available, and if there's a tutorial covering their use? Regards, Chrisguise (talk) 15:35, 9 January 2025 (UTC)

It's probably much less automated then you think, and it's nearly all available. Details of my workflow, if you'll bear with me (I'm afraid most of it won't interest you):

I consider OCR quality to be key. Bad OCR is a lot of time lost. Therefore I always strive to get as good OCR as I can. My current mix for that is:

• Getting the JP2s from IA (I nearly only work with IA). I keep them around till I'm done, for illustrations. When I need some of those, I get JPGs from the JP2s, do whatever file manipulation I want to do with the JPGs, and then upload to commons (I chose a fixed format to save time: [Index name without extension] p[pagenum].jpg).
• Converting them to PDF with img2pdf. This intermediate PDF conversion between JP2 and DJVU is probably the weakest link right now, as it entails a slight loss in quality, but it is needed for the next step.
• Which is OCR itself, using ocrmypdf (using tesseract). After a lot of testing, I found that this gave much better result than other methods. It has an issue tesseract always has of often badly misunderstanding quotes, but for letters and other punctuation, it's top notch, as far as I've seen. I would like to get a same-quality equivalent for djvu, but haven't found yet ): (never managed to get ocrodjvu up and running). A property I like a lot with ocrmypdf, is that when it fails (which is rarely, apart from quotes), it either fails loudly (quotes also fail loudly, I'm merely saying that outside of them there are few errors), or fail in way that are made loud by some of the below tools. When it doesn't understand a word, it often spits out gibberish, instead of an easily-confusable incorrect version of that word.
• Conversion to DJVU using pdf2djvu.
• At this step, there might be misaligned OCR. I do djvused [filename].djvu -e "output-all" > test.dsed (djvused is from djvulibre), and watch if a text hierarchy error comes up. If it does, it means that some page returned invalid OCR, and PRP is going to have trouble with that, and the OCR will be shifted, which is huge pain. To solve this, go to test.dsed, look what's the last pagenum. If it's after the work's end (in the no text pages at the end), as it often happens, you can ignore it. Else, do, with djvused [filename.djvu] -e "select [last pagenum in the djvused+1];remove-txt;save", and then rinse and repeat till it's fine.
• Finally, then upload to the relevant place often with the filename [mainspace work name].djvu (sometimes remove the parentheses in the name, because I did a lot of works called Poems (author name), and so there were always parentheses. In the past, my code assumed the title was always a form of that. most of this, but not all, has been cleared.)

For proofreading itself: I have made plenty of scripts (all here online on WS) to assist with various steps. I have tried to provide doc so that others can use, feel free to ask if it's unclear. They are:

• User:Alien333/common.js: not much, and this one can't really be used by everyone, it's mostly temporary stuff. The one important thing (maybe should be moved out?) is near the end, it prevents saving a page where there are invalid italic/bold (a bit simplistic, might have false positives.)
• User:Alien333/cuts.js: provides access to various functions of the below, as well as navigation (shifting Page:s, shifting sibling through the {{header}}s, and so on, with for each the option to open in this tab or a new one) through key combinations (mostly ctrl-meta-something, with a bit of ctrl-something). see User:Alien333/cuts
• User:Alien333/clean.js: applies some regexes to clean OCR and do some basic formatting. This is invoked by many of my scripts; you may want to provide a dummy clean() function (that returns its output) in your user JS if you don't like it.
• User:Alien333/poemise.js: as the name says, to ease formatting poetry, being able to do the formatting with 3-4 characters and a keypress on most pages so far. It also brings an imperfect way to fix the di-ppoem interaction (premature wrapping), and functions to change ppoem ends and starts fluidly. see User:Alien333/poemise
• User:Alien333/nobr.js: I rarely use it (as I mostly do poetry), but it's the manyth version of a simple unwrapping script.
• User:Alien333/rhalt.js: A fork of the rh gadget, specialized for poetry. It does not replace it; it is complementary. I like having the two at hand. see User:Alien333/rhalt
• User:Alien333/addtpp.js: a simple script to correct typos in the name of {{tpp}}, and replaces {{ppoem}} by {{tpp}} when tpp features are used.
• User:Alien333/pagenum.js: very simplistic script that adds, in the page header for Page:s, their pagenum as defined in the pagelist
• User:Alien333/cmbb.js: a WIP wikicode editor, somehow similar to CodeMirror, for thorougher highlighting that includes the headers/footers and other features I wanted. Notably, this allows CSS styling of common scannos to identify them easier. see User:Alien333/cmbb

I also use Firefox's spell checker (with an extendable dictionary, which now comprises about 9000 items for old words that are valid) to find scannos (it takes no action).

I at this point probably should talk about {{tpp}}. It's a thing I made, which uses ppoem, to do some stuff I ended up doing often. This template could be debated. The reason I have not tried to add these features to ppoem is that ppoem has the characteristic of being clean; these are not, and in some cases cannot be, clean. Features (see the doc for details):

• Adding a title as first parameter, centered and targetable with CSS; that allows marvels, and permits the elimination of a great lot of repetitive formatting
• Wrapping in most cases the first words with a classed span (for small-caps first words); this has saved me a great lot of time
• Relative indent; essentially the same as typing {{phantom|previous line}} this line, but shorter
• Reverse indent; does exactly what it says on the tin; to be rewritten to be cleaner
• Separating of different parts of a poem not aligned together (in effect making multiple ppoems); it has the advantage of 1) being shorter and 2) taking better care of the height of the break between the poems, being the same as that of a regular stanza break

That's about as far as it goes for transcription. I have optimised as much as I could so that I could easily find most scannos, and that hard-to-find scannos are very rare. That is probably one of the things that speed the most up. (I stay on the lookout for the scannos I have trouble finding, and when I find one once in a book I wait until I'm finished and then I reread everything, checking specifically for that, as if there's one of these I may have missed more.) I have given a lot of though on that, and I believe that I do not go too fast for it to diminish the quality of my works; if that is not the case, please point me to where I messed up, and I will gladly reread and correct that or these books, and adapt my method.

For transclusion, I use User:Alien333/transclude.js, which indeed has allowed me to speed the process up tremendously. I intend to give this a proper (OOUI) interface one of these days, to make it more usable. see User:Alien333/transclude

For disambiguation, I maintain a local list of most of the poems I have transcribed, a piece of OCaml code takes care of finding matches and adding to the list. User:Alien333/dab.js does the actual on-wiki work (it is one of the things which rely on the title being Poems, which is why I do not disambiguate the other works I do).

There you go! I hope I've answered your question, please tell me if not.

I must say I admire you and thank you for your relentless proofreading efforts. You do more than me on a monthly basis, and without my tools.

(and sorry for the wall of text) — Alien  3
3 3 19:06, 9 January 2025 (UTC)