User talk:Xover

From Wikisource
Jump to navigation Jump to search

Vaguely remember ...[edit]

... that I saw a recent comment about "p" and "pp" and why not just have a string, and I cannot remember where. The reason that I put them that way was two-fold. 1) Due to transclusions of templates that nest "header" as they will have one or two parameter and it mirrors that function though kept the naming short though obvious; then following 2) this allows a pull into WD with the harvest template bot, and gets around issues of whether we are using hyphen or ndash, and we can change as we need to do so. Nothing stops anyone using just "p", except it doesn't change the label. — billinghurst sDrewth 00:05, 11 June 2021 (UTC)[reply]

Need help (two missing pages again)[edit]

Hello. There are two missing pages (159, 160) in Dickens vol. 16. Could you please add them from this file? Ratte (talk) 20:32, 14 June 2021 (UTC)[reply]

@Ratte: Sorry for the slow turnaround time. I should be able to look at it some time this weekend unless real life decides to be recalcitrant. Xover (talk) 13:32, 25 June 2021 (UTC)[reply]
@Ratte: Yes check.svg Done Xover (talk) 20:34, 27 June 2021 (UTC)[reply]
Thank you! Ratte (talk) 16:52, 29 June 2021 (UTC)[reply]

So...Vue is fun[edit] So much easier than gadgets!

That's even a WVUI button in the bottom left. Inductiveloadtalk/contribs 08:31, 24 June 2021 (UTC)[reply]

@Inductiveload: Neat! Grazing off the recent changes feed I presume? Is it just a WVUI testbed or do you have a specific goal in mind? (if it's patrolling qua patrolling I need to dig up some links) Xover (talk) 13:36, 25 June 2021 (UTC)[reply]
Oh, also, the settings dialog is borked on mobile if you're not aware. It gets mispositioned by the edits and suchlike. /me waves hands vaguely… Xover (talk) 13:37, 25 June 2021 (UTC)[reply]
Yes, it uses the RC event stream. It's more of a plaything for Vue/Vuex in general rather than WVUI, since they have a very limited widget selection (even checkboxes aren't released yet). Since it's mostly a toy, and it doesn't use BootstrapVue like (which is actually a functional tool), it's not really "expected" to work all that well until WVUI invents dialogs or panels or similar.
It's mostly for fun, but it does make slightly different sounds for proofreads vs mainspace pages and page creations and edits on the Scriptorium and {{New texts}} get different sounds too. But it's not really designed for practical use as such. Though I'll probably improve incrementally as I learn Vue and as WVUI learns widgets. Like make it able to change sounds and store state etc, etc.
Also the sounds go weird after a few thousand pings, no idea what that's about: it's probably leaking something. ¯\_(ツ)_/¯ Inductiveloadtalk/contribs 13:54, 25 June 2021 (UTC)[reply]

A History and Defence of Magna Charta[edit]

Could you please add pp. 100 and 101 to this file? (It also needs the space at the end of the file-name removed, but I’ll request that move later.) TE(æ)A,ea. (talk) 20:42, 27 June 2021 (UTC)[reply]

@TE(æ)A,ea.: Yes check.svg Done Xover (talk) 13:21, 28 June 2021 (UTC)[reply]
  • It turns out that pp. xxiv and xxv are also missing. They could be recovered from this scan, but should I just replace the current scan with that one, since they are evidently the current edition? TE(æ)A,ea. (talk) 14:56, 30 June 2021 (UTC)[reply]
    @TE(æ)A,ea.: I'm not familiar with the work, but if it's the same edition and otherwise complete the technical quality looks much better than the existing scan. In those circumstances I would definitely just upload over it (don't forget to update the file description page). Xover (talk) 17:46, 30 June 2021 (UTC)[reply]
    • There is, at least, one other edition, because the scan that you used to add pp. 100 and 101 has a different introduction, and did not have the correct pp. xxiv and xxv. I will check over the scan, to make absolutely sure, but it looks fairly similar. TE(æ)A,ea. (talk) 18:49, 30 June 2021 (UTC)[reply]
    • Update: they certainly look the same to me. The HathiTrust scan is, however, blocked from general download as a Google scan. Could you download it, and upload it as the new file? I can readjust the Commons metadata and index structure. TE(æ)A,ea. (talk) 18:59, 30 June 2021 (UTC)[reply]
      @TE(æ)A,ea.: Yes check.svg Done Xover (talk) 13:29, 1 July 2021 (UTC)[reply]
      • Thank you. The description on Commons has been updated, as has the pagelist &c. here. However, the pages I have created so far are offset by one page (up); could you please shift the pages. (Sorry for troubling you so much.) TE(æ)A,ea. (talk) 13:53, 1 July 2021 (UTC)[reply]
        @TE(æ)A,ea.: No trouble. Should be done now. Xover (talk) 14:28, 1 July 2021 (UTC)[reply]

<span><span>followed by</span></span>[edit]

Hi! I used a wikidata property call with a subst and got <span><span>followed by</span></span> (and for follows). I am thinking it was a fix for some other wiki that got pasted here when the Wikidata module was installed.

I was wondering if I could get permission to find and remove those spans, so, am I permitted to find and remove them? Perhaps they are for something here, but I doubt it.

I can link to things if you need it.--RaboKarbakian (talk) 16:19, 1 July 2021 (UTC)[reply]

@RaboKarbakian: Not all templates work with subst: (it often requires coding specifically for it). If you give me some links I'll take a look. Xover (talk) 16:28, 1 July 2021 (UTC)[reply]
The Allies' Fairy Book/Frost is where SF removed one. d:P155 and d:P156 are the properties. And the Module:Wikidata I think is the one that I used. It was a really, really simple call, not like Module:WikidataIB. The documentation at wikidata promised simplicity and it was, except the html was embarrassing.--RaboKarbakian (talk) 16:39, 1 July 2021 (UTC)[reply]
The subst is how I can tell what is being really rendered! I used both modules for User:RaboKarbakian/WD version, but not with P155 and P156.--RaboKarbakian (talk) 16:43, 1 July 2021 (UTC) (It wasn't a template, it was a simple call to the property....)[reply]
@RaboKarbakian: What was the subst: code you used? Xover (talk) 17:33, 1 July 2021 (UTC)[reply]
I just added a "published by" to The Allies' Fairy Book/Frost using P123. That is what I did with the follows and followed by properties, only this without the subst. Like I said, I used the module directly. There is a syntax for using it when the page is not installed as a wikidata item. Here: The Allies' Fairy Book/Frost follows The Tongue-cut Sparrow. --RaboKarbakian (talk) 17:43, 1 July 2021 (UTC)[reply]
@RaboKarbakian: If you want just the value you will have to call getRawValue, otherwise the module will attempt to format it for you (hence the spans). However, going by way of Wikidata directly in wikitext makes no sense: it's much easier to just copy and paste the text you need and the end result is exactly the same. Unsubsted Wikidata calls should not be used in normal wikitext. Pulling information from Wikidata should happen from inside a template in a controlled scenario, for example the way the {{author}} template does. Xover (talk) 17:49, 1 July 2021 (UTC)[reply]
That is not what the how to at wikidata said. I used subst because the information is static. If source ever makes a portal for Heinmann, then that will be linked where I put it. I think those spans are for another wiki and should not be included. But, let me know what you think, maybe, after some more familiarity with it. I used it and thought about it for a long while.--RaboKarbakian (talk) 17:57, 1 July 2021 (UTC)[reply]
@RaboKarbakian: Which howto and what did it say? Xover (talk) 18:32, 1 July 2021 (UTC)[reply]
It's like I am telling you how to be my boss! :) And really, I used it and thought about this for a while, so let's talk about basic module behavior. A module like Wikidata, when asked to get an item and provides the link (if it is linked) -- shouldn't that provide just an <a href="linkto">Wikidata property value</a>? If whatever template needs it, the spans should be included at that level, the template level, not from the module. It is sloppy html. Would you write html like that? I wouldn't and I don't think that the module should be either. I would be more than happy to check a number of these and find whatever bad html there is in there so that it can be gotten rid of before the module gets used too often. That you hadn't seen this before, makes me think that P155 and P156 are not being used here much, yet.
That how to wasn't exactly hidden at wikidata....--RaboKarbakian (talk) 19:05, 1 July 2021 (UTC)[reply]
@RaboKarbakian: It may not be hidden, but there are many howtos there so knowing which one you meant is kinda hard. And if a howto on Wikidata is giving instructions for what to do on Wikisource I'm interested in checking that it doesn't lead people astray.
Regarding modules to access Wikidata, the situation is not comparable to HTML. Templates and modules were not developed with an overarching architecture and separation of concerns in mind: they are an organic development happening at very different points of the technological development curve and trying to solve the same problem. For example, a lot of templates are now simple wrappers around a module that does all the heavy lifting: templates provide a slightly simpler invocation syntax and nothing else.
In addition, some modules are designed to be used in specific contexts where emitting pre-formatted results is the most appropriate interface. WikidataIB, for example, is designed to be used for implementing generic Wikidata-integrated infoboxes on English Wikipedia, where a goal is to let wrapping templates delegate all the Wikidata fields to the module and only contain code for the non-Wikidata functionality.
Wikisource doesn't develop its own Wikidata access modules (we don't have resources to sustain that) but import our modules from elsewhere. We explicitly do not want to make local changes to these modules because those changes have to be re-integrated every time we resync the module from the source wiki, and there is no good way to even detect that local changes have been made, leading to mysterious failures after each update.
So while, yes, in some design contexts it would be most natural for a module to return unformatted values, that is by no means universally true; and, no, we will not change the module's interface locally. If there's a particular Wikidata access module out there somewhere that works like you want we can consider importing it here (with the same caveats), but barring that you'll have to make do with the ones we have.
I have a todo for looking at options for Wikidata access, because Module:Wikidata is deprecated and I find Module:WikidataIB irksome in several ways, so I am by no means suggesting the status quo is perfect. But it is complicated and not an area to be waded into lightly. Xover (talk) 18:02, 2 July 2021 (UTC)[reply]

I just wrote a script that recursively gave a style suggestion:"leftt", and the table I was generating, the last four values on the original are wrong, a 1921 "off by .10 error", I guess. And got a "no" from you about my request, so the feeling here is more terse than usual. Warning, some unloading to follow. I do not want to allienate the local DJVU expert, so tough up a bit for the next bits....
So, a summary. But first, some background. I got more involved with my computer than the average due to the developer/enthusiast group I first encountered. A collaborative group who spoke confidentally of experiences and were open-minded about application. I miss them. I get a lot of authoritative mumbo-jumbo from first time viewers now, interspersed with escalation bot bs. You seem to be quite an expert at djvu, which is mysterious and mystical to me. It would be great and fine by me if you explained that. For example, I am really good at some things, but other things are (like I said) "magical and mystical" to me; it would be good to be not alone in that. (longing for home)
If you agree or disagree with my opinion that span span unspan unspan is crappy html is not the issue. What is the issue is that the module is pasted here from some other place and local changes to it will make repasting updates too complicated to consider. (Summary)
If I want to get the spans out of the module, I will need to make my appeal at Which is so weird, since we are in an environment sitting upon versioning software. But at this time, some software is moved from "mountain top" to "hill top" as the tunnels are all tied up with minutia? Summary
It is okay not to be great at everything, in fact, it is perhaps a reality. If there are questions about how to use wikidata, you can send them my way until a more experienced expert shows up. (longing for home)
I read the same warning at the wikidata module as you did. And I read more in other places. I think that warning is itself a deprecated opinion and came with the paste. IB is really good for some things and the data module for others, seriously, look at the source of my version template, both module have their purpose/use. There is a problem with one of my templates, a damn space!!! And I have forgotten how to mark an input as optional so the damn things need a second pipe whether it uses it or not. But perfectly formated versions with only two pastes is just freakin great! (confident speaking about experiences)
It should be okay to say that you don't know something, shouldn't it? Confidence about experiences and open-minded about application -- it's a nice and honest way to interact with others. Please re-read the kudos for being a djvu guru and forgive my tersity (longing for home)
So, two summary, a lot of longing for home, a kindly request to be real and one confident voice.--RaboKarbakian (talk) 00:05, 3 July 2021 (UTC)[reply]
@RaboKarbakian: I don't actually understand what it is you are criticising me for. Please be specific and concrete. You can also stop worrying about offending me: constructive criticism politely presented does not offend me (even if I should disagree with it).
Perhaps it would be relevant for you to know that I am a bit frustrated about both the state of available Wikidata modules (not to mention API access from JavaScript) and the limitations of MediaWiki for things related to code reuse and version control?
The current Wikidata access modules are too specialised for specific uses and make assumptions related to those uses. When those assumptions do not hold true the way the modules behave are often either unnecessarily hard to use or too inflexible. It doesn't help that they are unusually hard to understand by being too closely tied to Wikidata and its data model rather than general computing concepts and common API designs. I wish we had a better and more flexible general-purpose Wikidata access module, and I am actively looking for one we could use (on and off, it's not my highest priority by any means).
It is also quite frustrating, to me, that MediaWiki doesn't have better facilities for sharing modules between projects, and for maintaining local modifications and adaptions. The "version control" built in is very specifically designed for prose text in Wikipedia articles, and lacks nearly every single property that is desirable for a code control system. There is also far too little collaboration between the projects, and consequently there is too little affordance made for other project's needs to make local modifications without having to take on maintenance of the module entire.
In other words, I wish I could point you at better facilities for doing whatever it is you are trying to achieve—I certainly don't mean to be dismissive—but I'm afraid we're mostly limited by the cold hard realities of our platform here. If there is something specific you're having trouble with I would be happy to help, if I can, but then you'll have to articulate your need in a way I can understand. Xover (talk) 09:50, 4 July 2021 (UTC)[reply]

Lint errrors[edit]

Well, I'll stick to 'content' type pages for now..

If you wanted to resolve some of the other remaining lint-errors outside of these:-

A lot of the simpler cases I'd already dealt with, having to a large extent cleared out a fair proportion of the Main and Page namespace misnesting , over the last few months. ( I will note I also cleaned out some simpler cases over on Wikiversity and Wikispecies.) ShakespeareFan00 (talk) 11:12, 7 July 2021 (UTC)[reply]

@ShakespeareFan00: Thanks. I may try to deal with some of these eventually, but I also don't think most lint errors are very critical so it has pretty low priority. Xover (talk) 06:22, 8 July 2021 (UTC)[reply]

Bot flag active[edit]

Hi. Did you mean to still have your bot flag active? Its been on since 9 May 2021 and its only because I include bot changes in my watchlist that I was able to see some of your edits - you've participated in some discussions while still having the flag on, so some editors might have missed your comments. --DannyS712 (talk) 00:26, 8 July 2021 (UTC)[reply]

@DannyS712: Grmpl. No, I just forgot to turn it back off. Thanks for the headsup! That would have just gotten increasingly embarrassing as time went on. :) Xover (talk) 06:37, 8 July 2021 (UTC)[reply]
Maybe in the future have it automatically expire after some time (a week?) in case you forget? Just an idea. Thanks, --DannyS712 (talk) 06:38, 8 July 2021 (UTC)[reply]
@DannyS712: Yeah, the thing is, I normally don't do that because I only need it for a couple of minutes or an hour or so, so any timeout would leave it on too long. Certain sayings spring to mind. I've been meaning to write a user script that gives me a visual reminder that the flood flag is on, and maybe also toggle it without having to go to the Special: page. But, you know, when priorities and laziness conspire… I should probably just start setting it to expire like a normal person would. :) Xover (talk) 06:47, 8 July 2021 (UTC)[reply]
Happened again, you're still flagged from 11 July. If you want me to make a script that will give you a visual reminder, I can try - just let me know what that visual reminder should be and where on the screen to show it --DannyS712 (talk) 21:12, 15 July 2021 (UTC)[reply]
@DannyS712: Aaaargh! This is now officially embarrassing embarrassing. For my sins I have performed the following penance: User:Xover/seawall.js (critique always welcome, but caveat quick-personal-hack!) @Inductiveload: You may be interested in this. Once baked a bit it might be useful as a admin-only gadget too.
BTW, does MW / Vector actually have any "API" (area of the page) that would actually be intended for displaying this kind of indicator? The mw-indicator stuff seems to be for… other types of indicators. I think we're probably in "will always be a hack" territory, but if anybody happens to know of a "proper" way… Xover (talk) 07:31, 16 July 2021 (UTC)[reply]

Eis et al. running here too[edit]

OK: the monoblock, itwikisource-inspired tool runs here too: see User:Alex brollo/common.js, User:Alex brollo/common.css and User:Alex brollo/PersonalTools.js. Obviously I've to test scripts a little bit deeper, but luckily critical points turned out to be few and simple to fix.

I realize that a wide doc would be absolutely needed... the scripts summarize years of painful tries and a very personal editing style, focused on editing speed of wikicode. --Alex brollo (talk) 12:54, 29 July 2021 (UTC)[reply]

@Alex brollo: Awesome! I'll be sure to take a look as soon as time allows. Xover (talk) 12:56, 29 July 2021 (UTC)[reply]
Years ago, I contributed at en.wikisource, how I can fastly find a list of simple abandonated transcriptions, just to test my script here? --Alex brollo (talk) 05:17, 15 August 2021 (UTC)[reply]
@Alex brollo: I'm not sure we have any list specifically of abandoned transcriptions (we have way too many of them), but we have the Monthly Challenge that lists works for community collaboration. You should be able to find something in there. There is also Category:Incomplete texts which lists those transcriptions that have been explicitly tagged as being incomplete using {{incomplete}}. Xover (talk) 07:59, 15 August 2021 (UTC)[reply]
Thanks, I'm going to test tools on Index:A daughter of the rich, by M. E. Waller.djvu. Obviously the script needs a little bit of debugging, and I can't fix issues but testing it while editing. The only clues to scripts edits are a memoRegex section into Index talk page and a new eis level1 into the edit comment. --Alex brollo (talk) 09:10, 15 August 2021 (UTC)[reply]

non-emoji entities[edit]

When I placed them there, I knew they would render into creepy candy colored things on my mobile device, but I did not know there was another option. (I have such sadness about the emoji-fication of the Unicode Consortium)

Where did you find those things?

What are those things?

Where can I get my own collection of those things?--RaboKarbakian (talk) 14:14, 30 July 2021 (UTC)[reply]

@RaboKarbakian: Welll, I'm on macOS which happens to have a pretty decent "Emoji and Symbols" helper that, despite the name, gives you access to most of Unicode. But you can usually google your way to this stuff: there's tons of web pages with Unicode tables, including Wikipedia. Xover (talk) 16:53, 30 July 2021 (UTC)[reply]
I used Gucharmap which is the gtk+ unicode do-hickie. Mine came from the "Unicode Dingbat" block. I don't have the emoji unicode data installed. How did I get emoji and you get what gucharmap was supposed to deliver? (the app is using Red Hat's Liberation font, btw).--RaboKarbakian (talk) 18:42, 30 July 2021 (UTC)[reply]
Also, emoji are unicode: see --RaboKarbakian (talk) 18:50, 30 July 2021 (UTC)[reply]
Sorry, I am not done here. I was just fooling around with the fonts in my browser and with this page: ("Miscellaneous symbols" is at the 2600 block) When I use Dejavu font, I get the textural glyphs, when I use Liberation, I get the emoji. Mozilla installed an emoji font set so I am assuming that if the font selected does not contain the character, Mozilla puts an emoji there. So, while I did not build this Gucharmap, I have built it before and know what goes into it. I am going to assume that I used unicode characters. Whatever you pasted there made a "diff" so it is not unicode (because that would have not shown a change). Since you got your app out of a box, I think I will never know. Unicode was a very respectable group of developers before this emoji thing. Maybe it is like that HGTG story about the planet with Group A, Group B, and Group C, where the Group B people picked the colors of things. Any day now, maybe they will be sent off....--RaboKarbakian (talk) 19:15, 30 July 2021 (UTC)[reply]
I seem to need to express myself in between researches. I just answered my own question. See ☘ vs 🍀. {{Unicode|&#x2618;}} vs {{Unicode|&#x1F340;}}. I will be happier not asking the question "Why did my gucharmap paste give me the new one and yours didn't?" but I will always kind of wonder of it.--RaboKarbakian (talk) 19:31, 30 July 2021 (UTC)[reply]
@Xover: can you point your browser at User:RaboKarbakian/Symbols and tell me if there is anything in the text column that looks like an emoji and if there are any text in the emoji column? 1) it all looks like text to me (the text column has been x-largered) and 2) there is no special emoji for zodiac so I am wondering if calling by the unicode name is what makes emoji appear in your browser). 20 years ago, it was javascript for IE or for Netscape. *shrugs*, *pats ascii* --RaboKarbakian (talk) 01:06, 2 August 2021 (UTC)[reply]
@RaboKarbakian: My apologies, I appear to have missed this question. In the tables on that page, the vine leaf ones in the first table appear unsupported here (haven't checked why); the "New Moon" up to "Sun with face" are Emoji-styled, as are "Aries" through "Pisces"; otherwise they appear as textual forms. Xover (talk) 17:54, 15 August 2021 (UTC)[reply]
@Xover: Both sides? meaning the left side is using the xml number and the right side is using the unicode number? I was expecting the xml to display as text and the unicode to display as emoji styled. There are not special unicode identities that point to something other than what the xml points to {like the shamrock). If both sides of my table display as emoji-like, then I have no idea what it was that you pasted or how to for certain access it. And the vines don't display here either. I put them there so that one day they might, and maybe they too will be familiar printers marks like the floral leaves are.--RaboKarbakian (talk) 19:24, 15 August 2021 (UTC)[reply]
@RaboKarbakian: It's ♈︎ (U+2648 ARIES + U+FE0E VARIATION SELECTOR-15) vs. ♈️ (U+2648 ARIES + U+FE0E VARIATION SELECTOR-16). VS-15 asks for a text presentation and VS-16 requests Emoji presentation. The astrological symbols are Emoji by default. This only works with Unicode code points that explicitly support variations. Xover (talk) 06:09, 16 August 2021 (UTC)[reply]
@Xover: My eyes don't exactly cross, but something in my brain does the same thing when I see unicodes. I blame hex, but that might be wrong. Anyway, I just wanted to drop this off, here: Along with that &#FE0E being the same as &#65038;. There is at least one browser, chrome on someones mobile device that does not draw them correctly. I am checking an ipad later.--RaboKarbakian (talk) 17:01, 4 September 2021 (UTC)[reply]

Hong Kong Case Law Files[edit]

I've seen your Hong Kong case law file uploads like File:HKSAR v. Xu Shengqi (CACC 463-2010).djvu from Hong Kong's Legal Reference System. I'd therefore like to ask how can one download the original DJVU (or PDF) file from the system, nice and neatly? After all, the print function on that website doesn't give such quality as those judgments uploaded by you. Many thanks.廣九直通車 (talk) 10:07, 1 August 2021 (UTC)[reply]

@廣九直通車: I believe I downloaded the Microsoft Word (.doc) version, printed that to PDF, extracted the page images, and used a custom script to generate a new DjVu file with an OCR text layer. If you need more of these just let me know and I can do the same. It's a bit fiddly so no guarantees on how quickly I can get them done. Xover (talk) 10:19, 1 August 2021 (UTC)[reply]
Many thanks! I've now managed to upload some legal documents from that website, thanks!廣九直通車 (talk) 03:52, 4 August 2021 (UTC)[reply]

Transitory Provisions[edit]

Could you move/copy this to my user-space, please? TE(æ)A,ea. (talk) 14:53, 3 August 2021 (UTC)[reply]

@TE(æ)A,ea.: Yes check.svg Done . User:TE(æ)A,ea./Transitory Provisions of the Constitution of Ireland. Xover (talk) 16:41, 3 August 2021 (UTC)[reply]


What exactly changed, because it no longer shows the link to the actuall ticket, just an Open, stalled etc line? ShakespeareFan00 (talk) 13:28, 5 August 2021 (UTC)[reply]

@ShakespeareFan00: I can't reproduce. Template:Tracked/testcases looks as expected to me. Where are you seeing it fail? Xover (talk) 17:08, 5 August 2021 (UTC)[reply]
@ShakespeareFan00: Hmm. But I am seeing the Toolforge tool that provides the API to Phabricator is currently down, so it's likely the difference in symptoms is due to browser differences. I'll follow up the Toolforge thing and see if we can get the right people to administer a kick where one is needed. Xover (talk) 17:12, 5 August 2021 (UTC)[reply]
Previously , the second Test cases would have shown the NUMBER of the ticket, with the Status being the second parameter. In the updated version the nominal Phabricator URL is seemingly being converted to a status indication.

ShakespeareFan00 (talk) 09:04, 6 August 2021 (UTC)[reply]

@ShakespeareFan00: This should be fixed now. Tracked tasks should now show up with either the task number (an exception, cf. later) or the task description, followed by the status. Note that the backend service giving the status data is flaky and sometimes returns errors when queried. When that happens the template will just show the task number (instead of the description) and no status. Xover (talk) 07:31, 13 August 2021 (UTC)[reply]

More deletions[edit]

In some recent deletion discussions, some individual parts seemed to have been forgotten.

Per this discussion (A glossary of words used in the neighbourhood of Sheffield), please delete these pages:

Per this discussion (Introduction (Molesworth to Hobbes, X) and Homer's Iliads in English), please delete this page:

Per this discussion (Leaves of Grass (Gutenberg edition)), please delete these pages:

Per this discussion (Ponniyin Selvan), please delete these pages:

Per this discussion (Sartor Resartus (Project Gutenberg edition)), please delete these pages:

Per this discussion (The Early Christian Attitude to War), please delete these pages:

Per this discussion (Variants and Analogues of some of the Tales in the Supplemental Nights), please delete this page:

Per this discussion (Landmark Education suffers humiliating legal defeat in New Jersey Federal Court), please delete these pages:

Redirection pages are painful, it seems; thank you for your work in deleting most of the pages. TE(æ)A,ea. (talk) 20:09, 6 August 2021 (UTC)[reply]

@TE(æ)A,ea.: Wow, that's an impressive haul. Thank you!
The problem is that the mass delete tool doesn't show you any extant redirects; they only show up when you manually delete pages one by one. So for these we have to rely on various other methods to catch them. Thanks for the above list which saved us having to hunt them down! Xover (talk) 09:47, 7 August 2021 (UTC)[reply]

my ego vs. seeming to complain[edit]

That was a very nice note that you dropped on my talk page. I was letting my ego express itself yesterday. Sometimes that expression seems like a complaint though. I remember feeling dismayed more than eh, "complainatory". I complain (or did complain) when I felt harassed or asked to lower my skill level (which I discovered yesterday was not the site-wide requirement it had been presented as).

Not complaining about volunteers and the current skill set comes with the territory. The opposite of complaining would be gushing, I guess.

Gushes: There is no better place for a djvu jockey than here! Proofers should be fluffing your pillows and such. I took a look at the tools in debian and omg! My script for making a djvu file failed miserably (commons finally deleted it about 4 or 5 years ago) and making one is nothing compared to fixing one! Shuffling cards to be in the right order would be difficult, shuffling a 'certain type of deck' even more so. So, I am very glad you are here.

Inductiveload, seems to have started life at the wikis as an SVG artist. Putting style class manipulation skills to good use is a plus.

Sometimes I deeply regret not learning how to script for wiki, but a lot of people who know "a little of this and a little of that" should always benefit from those who know a lot about one thing (as well as a little of this and a little of that).

There is a fortune file that says "I am sorry for the length of this, but I did not have time to make it shorter". <--That!--RaboKarbakian (talk) 18:32, 7 August 2021 (UTC)[reply]

@RaboKarbakian: I live in eternal hope of identifying the actual source of that quote, but like many of its kind it is variously attributed. It is, however, a perfect summation of most of my contributions here: "I would have written shorter, but I didn't have the time." Xover (talk) 20:10, 7 August 2021 (UTC)[reply]

partial repairs?[edit]

I have two books, with still red links, that are off by one. The first parts are done(ish yellow not green). Can parts or even end parts of books have their text realigned? And, can you pdf-- wait, I just checked and the one is repaired.... The other is Index:The Construction of the Wonderful Canon of Logarithms.djvu--RaboKarbakian (talk) 22:10, 9 August 2021 (UTC)[reply]

@RaboKarbakian: Yes check.svg Done And as a bonus you got about 3x the scan resolution which should improve ad hoc OCR if needed.
PDF files are mostly not possible to manipulate in this fashion, but in dire cases we can convert them to DjVu and migrate any existing Index: and Page: pages over to use the new file.
DjVu files can be manipulated in many ways, but often (as in this case) it is easier to just regenerate them from the source scans. The problem with offset OCR text is almost always due to a bug that's partially in MediaWiki's DjVu handling code and partly in the reference implementation of DjVu (DjVuLibre). What happens is that certain pathological pages end up generating malformed text data (often the frontispiece or other non-text content that OCR engines get confused by) which DjVuLibre accepts as input, but then fails to later extract from the same file. When this happens MediaWiki fails to detect the error, doesn't notice the missing page, and just assumes that the next page text that it gets belongs to the current page. The net result is that the text layer for, say, page 5 gets added to the scanned page at page index 4. And the effect compounds if there are more such malformed pages, so you can get page offsets of one, two, five, or more pages. My hacked up DjVu toolchain avoids this problem at the source by validating the text layer before inserting it in the DjVu file thereby never actually triggering this bug. Not perfect but it's caught all the cases I've dealt with so far.
The one big challenge with manipulating DjVu files is that when you insert or remove a page all the subsequent pages shift around accordingly. If we have meanwhile created those Page: pages they will now be out of sync with the underlying DjVu file. To fix it we have to move those Page: pages to the new location, and there are no tools to make that process easy. If there are a lot of pages that needs to be shifted by a fixed amount we can use a bot to do it somewhat sanely; but if there are multiple page offsets (some pages added, some removed, spread throughout the work) it becomes a huge pain and must be done manually. That's why we really really want Index: pages and pagelists to be checked thoroughly before starting proofreading.
But, in any case, so long as there actually is a decent scan available we can usually fix it one way or another. Inductiveload even created Wikisource:Scan Lab for just this purpose, so don't hesitate to ask when needed. Xover (talk) 07:30, 10 August 2021 (UTC)[reply]
"validating the text layer" would be the formal language that hides the actual magic! I guess I have previous experiences gumming up my works in a similar fashion; a "just deal with it" for old and now solved problems. Ever see an "animated jpeg"? I have. The jpeg plug-in in earlier versions of gimp-1.2 had such a problem that needed images to be flattened before saving (it was called saving back then). Truly, I flattened before saving and then flattened before exporting for the next ten years (or through gimp-2.6) even long after the jpeg plug-in became the supreme best jpeg maker ever (finding the original quality settings). Surely your hacked tools is a similar invention. There are three qualities I have noticed through out the years that people who have my sign sun all possess: wide feet (with the exception of one who has "narrow heels"), an extra striving for perfection (or nit pickers) and the strong tendency to error on the side of caution. Some of those other people have annoyed me to an extreme due to the differences in "caution". That second quality has caused me to avoid Inductiveloads projects (like the Scan Lab) as the mere mention of the name and I get a slurry of errors that software would probably not detect. That third quality has prevented me from seeking help for things like this ocr stuff. I would not have been able to fix the ocr, but on the other hand, finding solutions for other problems myself has been one of the delights of my lifetime, so some balance should be found. The first, the wide feet, the internet makes no requirement of shoes (\pats the internet).--RaboKarbakian (talk) 14:27, 10 August 2021 (UTC)[reply]

Members of Congress and copyright[edit]

I had mentioning the Murtha case, and said it 'wasn't really relevant' because of being about tort law. That's true, in the main....Murtha, while giving an interview to the press, as a ranking member of one of the Defense committees, defamed a number of Marines by calling them murderers, and it used a explicitly different definition of 'employee'.

What is relevant, though, is that the press interview, and everything he said in it, was part of his 'official duties' because, by saying they committed murder, he would make the Secretary of Defense look bad, and that would help him advance his political agenda. The leeway for what is 'legislative' seems to be, not that it gets tested a lot, very very long as it has arguably some connection to some law the critter is trying to get passed, some agenda they are pushing, or his 'job' as a sitting critter to interact with his constituents and at least pretend to listen to their concerns. Jarnsax (talk) 04:39, 11 August 2021 (UTC)[reply]

@Jarnsax: That accords with the practice we have established for dealing with works by congresscritters. But it all hinges on congresscritters falling within the scope of the {{PD-USGov}} exemption in 17 USC §105, as amended by 17 USC §101. If they are excluded, either by the definition of "employee or officer" or by the Emoluments Clause, from 17 USC §105 then we are left with only the edicts of government doctrine. And while Georgia v. Public.Resource.Org, Inc. significantly expanded the scope of what this covers, it is still a far more limited exemption than that provided by 17 USC §105 in that whatever the work in question is, it must be in some fashion informative upon a law to bear upon an edict of government rather than merely being within the scope of their duties as an "employee or officer of …".
In other words, I think we really need to settle whether 17 USC §105 applies to congressmen. Xover (talk) 06:20, 11 August 2021 (UTC)[reply]
I agree... I mean, I have certainly expressed opinions, lol, but I've tried to be clear between what I feel certain about (like, Members of Congress can't be Officers, no other way of reading it) and things where I think the case law is vague.... specific points that, as far as I have dug up, haven't been ajudicated.
It does seem that the scope of "legislative activity" has come up in terms of the Speech and Debate Clause, though. Similar enough that the reasoning in some of the relevant cases might apply... there is some discussion here [1]. I haven't actually looked up those cases yet, as I'm reading a CRS report on the same thing [2]. I'm hoping it might shed more light on where the line would fall...our problem, really, is we're not really looking for something that's stated in statute law, but instead for the common law hole that edicts fall through.
I will say, though, that I don't think PRO really 'expanded the scope', other than making it explicit.... like I said, people have been calling BS on Georgia for years. :) Jarnsax (talk) 07:00, 11 August 2021 (UTC)[reply]
@Jarnsax: Edicts of government determinations used to be based on whether the work in question carried some form of force of law. Georgia vs. PRO expanded this to also consider authorship as a factor. For US works the distinction was often academic due to {{PD-USGov}}, but for works from countries that do not have PD-USGov-like exemptions in their copyright legislation the difference is quite significant. Anything authored by the jurisdiction's competent legislative body is now at least eligible for consideration in light of the edicts of government doctrine in US copyright practice. And even in the US, everyone calling bullshit on Georgia does not mean we, here, can ignore it until case law overrules them.
Which is somewhat the point made at WS:PD somewhere: if the Compendium of Copyright Practices flat out states something that it requires interpreting the law to contradict, then we are not competent to do so and cannot ignore them even if they are wrong. In Georgia vs. PRO both the 11th Circuit and the Supreme Court gave us that interpretation, but for the application of 17 USC §105 to congresscritters we're currently walking very close to the line of legal innovation.
I'm starting to wonder whether this is an issue it would be worth bringing to c:COM:VP/C, since it affects them too and the community there is larger. If no conclusion is possible to obtain there we may have to punt it up to WMF Legal (it may just be general enough that they'd be willing to address it). Xover (talk) 07:57, 11 August 2021 (UTC)[reply]
My issue with using the term 'expanded' is most easily explained by quoting the PRO decision... "A century of cases have rooted the government edicts doctrine in the word “author,”..." That doesn't mean that I think it would have been appropriate for any of the WMF projects to just ignore their claims, though (don't get me started on Commons and the URAA) so it's fair to say it expanded the scope of acceptable content.
The House Judiciary Committee report on the 1976 Copyright Act is here [3], and discusses section 105 starting on page 58. The main thing I get from that is that "works for hire" and "works of the United States Government" are meant to be construed the same way.
Regarding the Compendium, I'd agree that we can't just say it's wrong, but I think it's easily misread. If legislators "cannot be the authors" of works they create in the course of their official duties, then the US Government could not (if 17USC105 didn't exist) become the owner of "works for hire" created by them, because there was never a copyright to be transferred. The language in 313.6(C)(1) talk about works 'created by' the President, Congress, etc... and then "or any other officer or employee".... the words "any other" imply the entire sentence is talking about officers and employees.
I do think taking this over to the Commons is probably a good idea, but at the same time... points at comment about URAA above. There is an established group over there who has (IMO) a history of ignoring WMF legal. Jarnsax (talk) 10:05, 11 August 2021 (UTC)[reply]


FYI, my little ramble about Hague IV etc. was essentially just meant to summarize what I had learned while looking into the situation of who had actual legal authority... like my other comments over there, I'm just mostly trying to give more info to hopefully help you guys with closing stuff.

This actually got me off on a nice tangent of reading way too much about the legal situation when the USSR dissolved, and how it was legally built. Anyhow, I think zapping it is the right decision... what I have basically learned is that people were are technically 'citizens' of whichever SSR, as well as Soviet citizens, but borders changed repeatedly, people moved around, and the citizenship law for the SSRs was "intentionally kept vague" and was basically irrelevant.

The kind of details about the author's life that would be needed for a formal copyright clearance are probably, to us, unknowable. Jarnsax (talk) 20:24, 26 August 2021 (UTC)[reply]

@Jarnsax: Yeah, that was what I figured (if I thought you really disagreed I would have held it open to try to tease out a consensus). All contributions to the discussions at WS:CV and WS:PD are very much appreciated: copyright research, since it requires a bit of specialised knowledge and often a lot of effort, is especiially appreciated of course; but even just expressing an opinion on issues that are not clear cut helps tremendously. Even on WS:CV we have a lot of issues that boil down to what each participant's standard of evidence, risk tolerance, or general stance on issues without a bright-line regulation in our policy or US law is. On WS:PD most issues are decided based on the participants' individual opinions because we have little bright-line policy for non-copyright issues.
When there are few participants, and they represent opposite outcomes, the issues deadlock and never get resolved. Especially since I really don't like letting mere majority voting settle the outcome: if it's, e.g., 2 vs. 3 on the outcome and both sides feel strongly about the issue, that's effectively a deadlock. And having 3 people total participate is probably above average.
What I'm trying to say, in a roundabout way, is that I very much appreciate your participation in these processes, both in terms of just generally contributing to consensus-building and in terms of actual legal and bibliographic research. Thank you! Xover (talk) 06:17, 27 August 2021 (UTC)[reply]
It's nice to hear that the admin mainly closing them knows how a wiki is supposed to work. :) I try to make it as least fairly clear when I'm "reasoning" vs citing factual information or actual cases (you've probably notice I try to cite stuff).... people are of course free to disagree with my reasoning, and I still tend to think in terms of Commons copyright rules anyhow (the 'and at home' thing). Jarnsax (talk) 18:16, 27 August 2021 (UTC)[reply]
@Jarnsax: Yeah, it took me forever and a day to train the lump of lead I use for a brain to think in terms of US-only. It doesn't help that I actually agree that Commons' policy is the right balance on this and that enWS should adopt it. Oh, well. Anyways, thanks for the compliment: I try! And when I inevitably mess up please do let me know. I may be an ornery, cantankerous, grouch, but I try not to bite the heads off people with valid criticisms. Too much. Xover (talk) 18:40, 27 August 2021 (UTC)[reply]
Right, ours is 'simultaneous worldwide publication'. Jarnsax (talk) 18:45, 27 August 2021 (UTC)[reply]

table style: template to module transition left docs in the dust (they were already very dirty)[edit]

The docs for {{table style}} have seemingly always been out-of-date. Now even more so. I just tried to use {{ts|w5}} and got style="w5;" output. Similar for {{ts|w10}}. They used to work. Never saw "w0" before, nor the "two thirds" longhands. Oh, they were there! {{ts|sc}} was never documented, even though used in examples in the doc!

Should items documented in the current Template:Table_style/doc be checked against Module:Table_style/data and forced into /data if missing ?

In addition, should items found in Template:Table style/parse be checked against Module:Table_style/data and forced into /data if missing ?

Should items not present in Module:Table_style/data be dropped from Template:Table_style/doc ?

And of course the proposed "additional shorthand"s aren't present. Can we drop everything in the "additional shorthand" column?

And how (or should) we document the new aliases like "all small capitals" ? Oh, look, at one time they were called "natural language"!? Huh?

And then someone changed column "code inserted" to "natural language code"?? It's CSS! 'Tain't nothing natural about it!

Can you think of others that would be sensitive to documentation changes?

Arrgh! What are your wishes my master? Shenme (talk) 04:10, 30 August 2021 (UTC)[reply]

@Shenme: Things that were only in the doc page never worked, so I've not bothered with those. The codes in /parse worked with the new module, but additional codes were added after I copied them but before I migrated the template to the Lua module. I've resynced those so they should now work too. As for the documentation table, it is now generated automatically from the data in Module:Table style/data so it should now always be up to date (modulo caching etc.). Happy? :) Xover (talk) 08:34, 30 August 2021 (UTC)[reply]
Use table to generate doc... automatic synchronization... I can't fault that! Thanks.
I've updated Template:Table style/doc a little. Link CSS so people know to blame someone else for problems/quirks. Add a note that styling sometimes doesn't work, e.g. table rows.
Would it be okay for me to change the column title
:tag('th'):wikitext('Output CSS Style')
There are so many different mentions of 'shorthand's, 'shortcut's, 'aliases', 'style's, etc. that being very clear what the output actually *is* is important. Again, if they have problems, get them to blame CSS, not us! Shenme (talk) 20:22, 30 August 2021 (UTC)[reply]

@Shenme: Go right ahead. :) Xover (talk) 21:13, 30 August 2021 (UTC)[reply]

The Net of Faith[edit]

Hello. Although I still believe that pre-1978 theses without copyright notice are PD nowadays, I would also like to try to get more information from about their publication of The Net of Faith and ask Tom Lock about it. Can you please help me to word the questions that we need to be answered? Thanks! --Jan Kameníček (talk) 15:05, 4 September 2021 (UTC)[reply]

@Jan.Kamenicek: Well, the high-falutin' legal discussion is really orthogonal here. The discussion is really about whether PD can be assumed in the absence of an assessment of evidence specific to the case under consideration. As I mentioned at WS:CV, even if the assumption must be that they are protected by copyright the circumstances for a given dissertation or thesis paper may, on examination, lead to the public domain through failure to comply with the formalities or other means. And there are definitely many plausible ways that could be the case if we can just find the evidence.
So… if we're going to do the leg work, the first step is finding out what's source for the text is. Did they just grab it through inter-library loan and scan it or transcribe it? Was it submitted to them by the author? A third party? Were they given a "permission to use", and if so under what terms? Were they provided an actual license and if so what was that license?
My immediate guess would be that they found it in some uni. repo or library somewhere and just grabbed it. In that case their part in the chain of custody doesn't help us much, and we will have to track further back. Was the translated text actually a part of the thesis paper, or did the author just happen to translate it while working on that thesis? If it wasn't part of the thesis, what path did it take to end up at Was it published anywhere prior to going online there, and if so where and when and under whose authority? If it was part of the thesis paper, which educational institution was it submitted to and what were their instructions to students and policies for such papers? Did they archive physical copies, or microfilm? Were these distributed anywhere? Were they made available through inter-library loan, and if so under what restrictions in terms of who could get access and what they could do with them? Was it ever commercially sold, for example to other articles chives and libraries through the microfilm distributor? And, not least of all, does the original—wherever it was first published—contain a copyright notice? It is in my experience rare for this kind of text to do so, but it is absolutely not unheard of, not even in the relevant time period.
Judging by Worldcat, Molnar published commercially (well, depending on what you consider "commercial", but…) as late as 1980, so I hold it not at all unlikely that the translation in question was actually published commercially as well. If so it was definitely subject to "general publication" and the normal rules (notice etc.) apply. Xover (talk) 15:36, 4 September 2021 (UTC)[reply]


It requires autopatrol level which I don't have here. I have it at commons where I have compiled a list of properties.

I was thinking of something like a citation template that creates a data item for the scan, which I can do from the commons because, well, they like me there I guess and have given me more rights.

Then I was thinking, maybe @Xover: can make this here before I can make it there. And so I am here, with the link and also, with the collected properties.

Quickstatements howto: Recently compiled property list: commons:User:RaboKarbakian/WD create book although, it might get renamed to "create scan"....--RaboKarbakian (talk) 17:02, 16 September 2021 (UTC)[reply]

Months ago, you asked me for the link at wikidata. To refresh your memory. one of the modules here claims to be "less expensive" than using the regular wikidata module. I am quite certain that the word "expensive" is being used to describe how much the computer is being used. I piffed at that statement, and still do, but I am willing to be corrected if I am wrong. The software chain is user, intermediate module, wikidata module. Sending a simple "#statements" to one module, who sends it to the wikidata module is the same as sending a simple "#statements" to the wikidata module or "less taxing". I think that the "expensive" part comes from sorting and qualifier finding. So, on my templates, if it was a simple thing, I just used the simple. If it really was complicated, I invoked one of the intermediate modules. Here is the that link I thought you should find yourself: (cryptically named, as you can see) I don't mind if you take my logic apart. I would prefer to be actually correct instead of just sounding correct and thinking that I am.
Also, you can probably win my challenge without even trying or even writing anything because I have no experience with Wikibase. I will need something like that to be able to get the right author id, location id, etc. The structured data interface at the commons does it really well. Have you seen that?--RaboKarbakian (talk) 19:36, 16 September 2021 (UTC)[reply]
Eek! Javascript! This is AC/DC . It is slick. It will gather all of the images in a category, or work with a list. It will find a property at wd, given enough to search with (copyright status shows up with "c o p" for instance, pub date with "p"). It also locates Q's when given to it. It makes structured data (for commons files) at commons, not uploading to wikidata though. Some mix of this and quickstatemnts. I am done with the links and bothering you.--RaboKarbakian (talk) 19:57, 16 September 2021 (UTC)[reply]
@RaboKarbakian: I'm not following you in the above. What exactly is it you are asking me to do?
If it's making a template or module that can directly create an item at Wikidata then that's not really doable (for all sorts of reasons). If it's user interface like the Structured Data for Commons here on enWS then that's going to be a pretty tall order, and is likely better solved by getting the WMF to add in-core support for it. If it's a utility javascript á la ACDC it's something we can look into, but since Wikidata doesn't really offer any sane JS API (that I've found) it's not a very inviting prospect (I've looked at existing scripts' code and it ain't pretty). Xover (talk) 06:43, 17 September 2021 (UTC)[reply]
My initial idea was for something like a citation template that would create the data entry. Templates here all get not put. Quickstatements does that, all that is needed is for the information to be correctly formated (for quickstatements, pipe using syntax, iirc) and then reformated to send via http to quickstatements. So, that will all work from here.
My idea faltered when I realized it needed to be able to locate and use the correct "other information" from wd, like author and place of publication. That is when I found that javascript for ACDC, which does all of that.
My question about what you wrote is "Is Sanity a quality that Javascript API is even able to possess?" I have translated Scheme to Python and Perl to Python but it was all for a base software that I understood. I pasted the ACDC javascript into a text editor, saw that it was all on one line, my vision blurred, I closed the editor before I began seeing dolphins or birds or "something" in the Magic Eye-like image that my text editor had turned into. Came here and wrote positive things, not mentioning the text editor experience. No nightmares, last night, as I thought maybe. That is all I know about ACDC, other than it searches wd for properties and items.
I built my idea here. Sorry for the mess. Quickstatements will create wd entries from here, via url commands. I thought that ACDC uses wikibase to search wd, but is that wrong?--RaboKarbakian (talk) 12:23, 17 September 2021 (UTC)[reply]
@RaboKarbakian: I'm not sure what ACDC uses.
Hmm. Ok, so you're looking for some way to get to a pre-filled QuickStatements batch from a wikipage? What sort of wikipage (Index:, File:, any random page, etc.)? And what information do you envision manually entering into that page or are assuming is already on that page, and what information do you envision being looked up before being fed to QuickStatements? And should the QuickStatements batch create a completely new item, and if so of what kind? Or should it add attributes to an existing item? Xover (talk) 14:27, 17 September 2021 (UTC)[reply]
@Xover:From when I still thought it could be a template commons:User:RaboKarbakian/WD create book. The properties for creating a wikidata item from the scan. If you interested in seeing the book template at work for the images used here, File:Midsummer-Nights Dream-Rackham-081.jpg is a good example I have handy. If not, avoid!! (I have been "GET" info at wd for that, this is the same only "PUT".)--RaboKarbakian (talk) 14:36, 17 September 2021 (UTC)[reply]
@RaboKarbakian: The book template doesn't create a Wikidata item, it pulls information from an already existing Wikidata item. And the Book template already exists, so I am assuming you are trying to achieve something different from what it does. Hence the questions in my previous message: I need to understand what it is you are trying to achieve before I can figure out how that might be achieved. Xover (talk) 15:16, 17 September 2021 (UTC)[reply]
@Xover: Others have suggested that there should be a way to glean the information (from a template was my idea) and put it at wikidata. The book template, it will add the scan (P996) to an existing wd item. Probably I should not have pasted an example of that. At commons, that template uses the art module, I cannot imagine working on that. But a different template say, something like the hypothetical {{WD create book scan}} that puts rather than gets.--RaboKarbakian (talk) 15:28, 17 September 2021 (UTC)[reply]
@RaboKarbakian: A template cannot itself modify Wikidata. What it can do, and which several different templates already do, is to construct a link to the external QuickStatements tool that will pre-fill the details of a batch when you click on it. If you want a template where you can fill in all the parameters listed on c:User:RaboKarbakian/WD create book and, on saving, get such a QuickStatements link, then that should be possible.
But unless I understand how and where you intend to use it, and what effects it should ultimately have at Wikidata, I am unable to comment on the specifics. Xover (talk) 07:53, 18 September 2021 (UTC)[reply]
@Xover: iaupload might be a good start.
I started remembering.... The book template at commons does search and create if no wikidata id has been given it. Seems like it did not "put" all of the information at wd when it created it though. Also, finding it once created was a challenge, but I am going to look for it this time in my contributions (which I hadn't before). That info was at quickstatements, but with url activation, maybe not so helpful. I'm gonna find one of those pesky pdfs and see what all of that is doing lately.--RaboKarbakian (talk) 14:52, 18 September 2021 (UTC)[reply]

No good deed... {{table style}}[edit]

Consulting {{Table style}} for the Nth time this week and just happened to notice...

When calling up the page to consult the documentation, there's a stray line at top


Dropping the documentation doesn't fix that. Dropping the module invocation

{{#invoke:table style|main}}

does make the stray text disappear.

I noticed that Module:Table style

function p._main(args)

has these lines which would run whether there are arguments or not:

local str = 'style="'
for ....
str = str .. '"'
return str

Perhaps instead the order something like:

local str = ''
for ....
if str is non-empty
  str = 'style="' .. str .. '"'
return str

or whatever the 'correct' lua would be. Shenme (talk) 00:32, 29 September 2021 (UTC)[reply]

@Shenme: Done. Xover (talk) 13:43, 29 September 2021 (UTC)[reply]

Version Page for Jane Austen.[edit]

I see that you removed the notes about other editions from Jane Austen. Would you mind creating version pages for Sense and Sensibility, Mansfield Park, and Pride and Prejudice? Right now the text of Pride and Prejudice are from the 1817 edition while the scan links on Jane Austen are from the 1811 edition. I'm also planning on adding the ssl to the other texts, but they will be mismatched as well. Languageseeker (talk) 12:29, 3 October 2021 (UTC)[reply]

@Languageseeker: So far as I can tell we only host one version of each of those texts. Am I missing something?
Also, what's listed on author pages are works, not editions. Works don't have publication dates, they have dates of first publication. That the text we current have is a 1817 edition is entirely compatible with the author page showing the work to have been first published in 1811. Xover (talk) 13:08, 3 October 2021 (UTC)[reply]
I think the question becomes where do we put the scan links for the other editions. Right now, the following is confusing because the Pride and Prejudice refers to the 1817 version, (1813) refers to the first publication date, and (transcription volumes: 1, 2, 3) refers to the unproofread scans of the 1813 edition. How do we clear up the confusion? Languageseeker (talk) 15:35, 3 October 2021 (UTC)[reply]
The answer is Author:Jane Austen. We probably do not need ESL for all the editions, though if you have SSL then you can list them there. Don't go creating hot links to editions of works where we don't have a scan. — billinghurst sDrewth 15:43, 3 October 2021 (UTC)[reply]
@Billinghurst: Is something like this acceptable? Languageseeker (talk) 15:45, 3 October 2021 (UTC)[reply]
@Languageseeker: Since we have Billinghurst's attention I'll generally defer to them on this issue, but… I'd say that if there is work actually planned (and not just amassing links) for additional editions of these works, then add the ssl entries as sub-items (***) under the work. Just not as mainspace links for the additional editions: those go on a versions page, and only when the additional editions have actually been proofread. --Xover (talk) 16:12, 3 October 2021 (UTC)[reply]
I'm not trying to amass links for the sake of having more links, but to try and define what needs to be proofread. All the major authors have hundreds of editions of there works, but only a few of them are either those that the author contributed to or have some independent merit. Also, I've noticed that adding scans actually increases the likelihood that a text will be proofread. Languageseeker (talk) 01:37, 7 October 2021 (UTC)[reply]

Unused files[edit]

  • I have added some more Files for speedy deletion: I hope to do more work on them in the near future. TE(æ)A,ea. (talk) 19:36, 4 October 2021 (UTC)[reply]
    @TE(æ)A,ea.: Yes check.svg Done And thanks! I'll try to expedite these when you add them, but IRL is keeping my wikitime (and capacity for attention) fairly unpredictable so they may sometimes linger a bit before I get to them. Xover (talk) 06:57, 5 October 2021 (UTC)[reply]
    • You’re welcome! Don’t worry about expediting—I don’t think anyone is complaining too loudly about them at the moment—because I may also not get to them too much. I noticed recently that it had been almost a year since I worked on them, so I thought to get back to them now—with some small steps, at first. TE(æ)A,ea. (talk) 13:15, 5 October 2021 (UTC)[reply]

File:Appeal of the Council of People's Commissars to the Moslems of Russia and the East.pdf[edit]

I have created this file to scan-back Appeal of the Council of People's Commissars to the Moslems of Russia and the East; please trim the second page, please, and match-and-split. TE(æ)A,ea. (talk) 20:56, 5 October 2021 (UTC)[reply]

@TE(æ)A,ea.: File:Appeal of the Council of People's Commissars to the Moslems of Russia and the East.djvu, Index:Appeal of the Council of People's Commissars to the Moslems of Russia and the East.djvu.
As you can see, image (OCR) quality is really poor due to generational loss: scan/photo → PDF → JPEG → JPEG → DjVu. It's generally going to be better if we can work on the original files: scan/photo → DjVu. Xover (talk) 07:12, 6 October 2021 (UTC)[reply]
  • Yes, the DJVU does look to be of poor quality; in the future, I’ll export to TIFF instead of to PDF. Could you please delete the PDF scan? TE(æ)A,ea. (talk) 12:50, 6 October 2021 (UTC)[reply]
Awesome! If you have pictures of the TOC it might be useful to have those for reference in case someone wants to see about some particular document in the future (even if we just have the image in commons). MarkLSteadman (talk) 15:23, 6 October 2021 (UTC)[reply]
  • MarkLSteadman: I only took pictures of those three pages, and I have now returned the book, so that is not possible—sorry. There were several dozen documents in that volume alone, not counting the other two (I think) volumes, so that would have been helpful. TE(æ)A,ea. (talk) 17:08, 6 October 2021 (UTC)[reply]
No worries. Awesome that you were able to get it. MarkLSteadman (talk) 18:06, 6 October 2021 (UTC)[reply]

Ppoem pokes its head out of its burrow[edit]

...but will it see its shadow? I have removed the dire warnings of doom from {{ppoem}} (leaving in its place a note along the lines of "be gentle").

Do you think a general announcement is in order, or wait and see if/how the keen-eyed early adopters manage to make it explode and shower us all with tiny bits of shattered poetry for a while? Inductiveloadtalk/contribs 08:04, 8 October 2021 (UTC)[reply]

@Inductiveload: I'd say announce it, and let the announcement reflect that while the template itself is well-tested and stable the use of it is still experimental. We have a theory that it will be way way better than the alternatives, but only large-scale use by diverse contributors will tell us that for sure. Xover (talk) 06:04, 9 October 2021 (UTC)[reply]

File for redaction[edit]

It is Hotaling v. Church. Please redact the following:

  • On p. 199 (/1), the key symbol below “AFFIRMED” and the paragraph at the bottom of the right column;
  • On p. 200 (/2), everything except for the running header;
  • On p. 201 (/3), the paragraphs above the rule in the left column; and
  • On p. 205 (/7), the key symbol in the right column and the two paragraphs at the bottom of the right column.

Thanks! TE(æ)A,ea. (talk) 14:18, 9 October 2021 (UTC)[reply]

@TE(æ)A,ea.: Done and deleted. DjVu at File:Hotaling et al. v. Church of Jesus Christ of Latter-Day Saints.djvu. But are you sure the redacted bits are a copyvio? What's the source / who is the author? If it's just the syllabus and written by the reporter we host about a gazillion of those already (nearly every USSC case has it) and it's covered by the same license as the order and dissent. Xover (talk) 15:09, 9 October 2021 (UTC)[reply]
  • This opinion, as you can see, was published in the Federal reporter, Third series, which was published by West (now a wholly-owned subsidiary of Thomson Reuters). The syllabi and headnotes were prepared by West staff, and are thus copyright West. This is the reason newer volumes of the Federal reporter and regional and state reporters cannot be hosted: the extraneous material was not produced by someone with the ability to interpret law (or whatever the PRO phrasing is), but by a private individual or agency. I’m thinking of getting some more of these copyright-related opinions; is there a portal for (U.S.?) copyright law where they could be listed? TE(æ)A,ea. (talk) 15:45, 9 October 2021 (UTC)[reply]
    Ah, I see. There's no portal for copyright opinions that I'm aware of. Might be useful to have, both in Portal: and in Wikisource:. I keep having to google for things like PRO, and often struggle to recall the right google keywords. Xover (talk) 16:09, 9 October 2021 (UTC)[reply]
  • Xover: I have two more cases for you to censor, please: the district and circuit court opinions in this case. For the first one, please remove the first page (which relates to the whole volume), and censor the following: the West Key logos (on pp. 1350 and 1361); the syllabus and headnotes to the case (from “background” on p. 1350 to just above the rule at the top of the second column on p. 1352); and the syllabus and headnote to the following case on p. 1361. For the second one, please remove the first two pages (once again, relating to the volume) and censor the following: the West Key logos (on pp. 1229 and 1255); the syllabus and headnotes to the case (from “background” on p. 1229 to just above the rule at the bottom of the first column on p. 1231); and the syllabus to the following case (on p. 1255). TE(æ)A,ea. (talk) 21:26, 12 October 2021 (UTC)[reply]

multi-layered tiff[edit]

Just for your notes, GIMP 2.10 (maybe others also) handles them much in the same way it handles pdf.--RaboKarbakian (talk) 15:02, 9 October 2021 (UTC)[reply]


Your comment on Author talk:Miss Morrison has prompted me to create a tracking category for such pages. Observe: {{honorifics}} and Category:Authors with honorifics in title. —Beleg Tâl (talk) 14:14, 13 October 2021 (UTC)[reply]