Wiktionary:Grease pit/2024/October

Module:number_list/data/la has now been extended beyond one million, which caused the Roman numerals automatically generated by Module:foreign numerals to get out of hand. Wikipedia represents 1,000,000,000 in Roman numerals as M with two lines over it; I don't know where it comes from, but that is at least more likely to have ever been used than repeating M̅ a thousand times. Urszag (talk) 22:19, 1 October 2024 (UTC)

I don't generally use QQ for searching, since I am often searching books.google for specific date ranges, looking for earliest and latest cites, or looking for uses of rare senses, eg currently the noun sense of scarper, which I was unaware of until I read it in the Guardian, following which I had to look through about 130 bg quotes to find two more uses. (I couldn't use QQ because, even when it hasn't reached what seems to be its daily quota, it seems to stop after around 20 hits, and anyway, there is sometimes insufficient context in the snippet it shows to know whether the cite is using the word in the sense I'm searching for.) I'm not expecting that to change anytime soon -- I believe I'm using sufficiently fuzzy skills that it will take a while yet for AI to catch up with me, and at the same time, I read slowly enough that Google knows I'm not a machine, so they don't shut me down.

BUT, having found my cites, I then spend ages filling out the details into the template -- often longer than it took me to do the search. Would it be possible to modify QQ (or some other gadget) to take an input of a particular bg page, rather than searching for it itself, and format the cite from that? Hopefully, that would be easy, since it must do something very similar at present, after users click on the cite they want to use. --Enginear 01:54, 3 October 2024 (UTC)

phab:T376361, itself caused by a bungled fix to phab:T375876...

What even is integration testing? Ioaxxere (talk) 02:39, 3 October 2024 (UTC)

This is a {{ja-see}} issue; 辨へ is the old kana form of 辨え, the kyūjitai form of 弁え, which is an alternate form of わきまえ (which has the old kana form わきまへ). All of the other forms work, but not 辨へ. TE(æ)A,ea. (talk) 02:42, 4 October 2024 (UTC)

@TE(æ)A,ea.: Seems the template lacks a default way of handling alternative kyūjitai forms. Thankfully, it accepts custom fields. Binarystep (talk) 05:31, 24 October 2024 (UTC)

bodkin isn't showing up in the Category:English terms suffixed with -kin (diminutive). Can someone explain why it's happening and/or show me how to fix it? Thanks! Katya0133 (talk) 04:26, 4 October 2024 (UTC)

@Katya0133 It was using {{m}} instead of {{af}}. Anyways, that part should ideally be covered under Middle English. Catonif (talk) 12:21, 4 October 2024 (UTC)

Please add the additional attributes to m at Module:languages/data/3/h:

m = {
    "Hanunoo",
    35435,
    "phi",
    "Hano, Latn",
    translit = {Hano = "hnn-translit"},
    override_translit = true,
    entry_name = {Latn = {remove_diacritics = c.grave .. c.acute .. c.circ}},
    standardChars = {
        Latn = "AaBbKkDdEeGgHhIiLlMmNnOoPpRrSsTtUuWwYy",
        c.punc
    },
    sort_key = {
        Latn = "tl-sortkey",
    },
}

Thanks! 𝄽 ysrael214 (talk) 11:06, 4 October 2024 (UTC)

Tested the translit module and  Done. Catonif (talk) 12:46, 4 October 2024 (UTC)

@Catonif Hi can you also add "hnn" to data.hyphen_not_multiword_sep at Module:headword/data? 𝄽 ysrael214 (talk) 14:56, 4 October 2024 (UTC)

Following this discussion. There is currently an issue with the display of hieroglyphs not supported by WikiHiero, as {{egy-h}} has to treat them separately (traditionally this was done manually via egy-glyph), which leads the final result to go to a new line when those glyphs are encountered if the window is too small. For example, check the quotation under hrm and decrease your window size (or access it from your cellphone, where the width is insufficient already) and notice the unwanted line breaks. This can be solved via a JS snippet, currently at User:Catonif/hierotables.js (I know modifying raw HTML with regexes is usually considered forbidden practice, but handling this with JQuery seemed to get too lengthy and bug-prone), which also solves another issue, this time intrinsic to WikiHiero, about even more unwanted wrapping in complex hieroglyphs groups.

Given that we already have a JS script running on Common.js for WikiHiero, MediaWiki:WikiHieroTempFix.js (this makes WikiHiero results display inline with the text rather than displaying block), launched whenever WikiHiero calls are detected, I believe the content of hierotables.js snippet could be simply incorporated in it.

The second issue is that this WikiHieroTempFix.js I just mentioned is not called on MediaWiki:Mobile.js, but only on MediaWiki:Common.js, which means that if I access pages such as ⲑⲱⲟⲩⲧ from my cellphone (weirdly enough, not if I access it in mobile mode from the PC however), the hieroglyphs in the etymology rather messily do not display inline with the text. I don't know whether we're planning to merge Common.js and Mobile.js, as @Ioaxxere recently finally did for their .css counterparts, which would get rid of this issue, but as long as that's not the case I call for WikiHieroTempFix.js to be run also by Mobile.js.

CC: @Vorziblix. Catonif (talk) 12:18, 4 October 2024 (UTC)

If no one has any objections, I support putting all of User:Catonif’s suggestions into operation. This JS snippet resolves some very longstanding issues with WikiHiero and should make mobile browsing of relevant entries much less messy. — Vorziblix (talk · contribs) 23:41, 7 October 2024 (UTC)

Not sure when this went bad, but attempting to process table entries thorugh the linkify_entry function no longer correctly creates links to reconstructed entries. Instead, an input like *jptwj creates a link to Reconstruction:/Unsupported titles/Reconstruction:/jptwj; see the links at {{Egyptian demonstratives}} for a number of examples. This only happens when the asterisk comes at the start of the input; inputs like jptw, *jptwj, where multiple entries are listed and the reconstructed one doesn't come at the beginning, are processed fine. Anyone know how to fix it? — Vorziblix (talk · contribs) 14:32, 4 October 2024 (UTC)

@Vorziblix This is caused by {{lang}}, not by Module:table tools, and is due to something in embedded_language_links in Module:links. I've asked User:Theknightwho to look into it. Benwing2 (talk) 00:08, 5 October 2024 (UTC)

also: JPEG artefact

• JPEG artifact: compression-induced image distortion

188.4.61.73 21:35, 4 October 2024 (UTC)

It's not clear what you are saying. — Sgconlaw (talk) 21:48, 4 October 2024 (UTC)

@Sgconlaw context-free posts to the GP can often be elucidated by checking the user's abuse log. In this case I think the abuse filter is not set up properly, as the user did provide a definition, even though they failed to remove the rfdef template. This, that and the other (talk) 04:15, 5 October 2024 (UTC)

I've been going through these, and I'm finding a few common types of false positives:

1. Zero affixes like ∅- and -∅ have entries in various L2s and are included in etymologies.
2. Combining diacritics with a dummy character as a holder, such as ◌́, ◌̂, ◌̈, ◌̀, etc.
3. Links to Wikipedias and other Wiktionaries. Wikimedia sites can have entry names in any language, so links to their entries should not be treated as text in a specific language. For instance, all the Cyrillic characters have entries on English Wikipedia with Cyrillic titles, but the links to those entries in {{wikipedia}} or {{pedia}} are not English terms in Cyrillic. Likewise, a Wikipedia in a non-Latin script will have articles on organisms and their taxa under their taxonomic names, but links to those are not terms in the language of the Wikipedia in Latin script.
4. Languages with no script specified. What's the point of putting the entire language in an entry-maintenance category?
5. Obsolete or rare orthographies like Belarusian Łacinka, or Arabic script for various languages in the Philippines, Indonesia, South Asia, Eastern Europe, etc. I realize that these have to be handled on a case-by-case basis, but I thought I would mention them, to be thorough.
6. Multilingual symbols such as mathematical or scientific notation, signs for gods, astrology and alchemy terms. Here again, this is probably case-by-case.

Some common mistakes. I'm not sure there's a technical fix for these:

1. Transcriptions copied from comparative literature by people who don't know or care that they're in the wrong script.
2. Bad language codes
3. Wrong templates: people using {{l}} instead of {{lb}} for labels is probably the most common (there was even someone who used {{l|lb}}, which made them into Limburgish).
4. Errors in template syntax: for instance, not putting the right number of pipes between positional parameters, so a translation or transliteration is treated as a display form.

One special case worth mentioning: Hindi and Urdu headword templates have links to the equivalent term in the other language. If someone uses {{head}} instead, they try to reproduce this by having fields for "Urdu spelling" or "Hindi spelling", but these are tagged as the language of the headword rather than as another language. There may be a way to work around this, but I haven't found it yet.

I hope I haven't overloaded this topic. I was afraid that I would never get around to posting this if I had to decide how many individual posts to break it into. Thanks! Chuck Entz (talk) 17:31, 5 October 2024 (UTC)

I tried to solve false positive type 3 in Special:Diff/82308974 and 4 in Special:Diff/82309298, but it might take a while to filter through. As for 6 (multilingual symbols), we shouldn't be writing {{syn|en|❤️}} or whatever. I guess it should be {{syn|en|mul:❤️}}. This, that and the other (talk) 11:41, 7 October 2024 (UTC)

@Theknightwho reverted the second change, pointing out that this is a useful way of keeping track of which languages need script codes added to their language data. TKW might also be able to help fix false positive types 1 and 2 by making the relevant code ignore the ∅ and (presumably) ◌ characters. This, that and the other (talk) 23:37, 7 October 2024 (UTC)

In five#Translations it seems usages of Template:tt are broken and only show "true" or "true, true" Xie1995 (talk) 19:04, 5 October 2024 (UTC)

This looks to be fixed. Benwing2 (talk) 21:03, 5 October 2024 (UTC)

Is there a particular reason for it to appear always in italic when it's linked? Sérgio R R Santos (talk) 11:37, 7 October 2024 (UTC)

Hi, here on enwiktionary, {{bn-noun}} gives বই • (boi) (see বই#Noun). On bnwiktionary, we don't need the translit part for Bengali word (in this example, the (boi) part). So, i comment out translit part from languages/data module. But the template still says "Transliteration required". How can we remove it? (We still want "Transliteration required" for other languages but not for Bengali language.) আফতাবুজ্জামান (talk) 17:28, 7 October 2024 (UTC)

@আফতাবুজ্জামান Hey it looks like you should undo your change and instead add "bn" to the list here: Module:headword/data#L-434. Benwing2 (talk) 21:55, 8 October 2024 (UTC)

Is it possible to update {{IPAchar}} so that it does not allow a line break after a slash (/), to avoid a result like what is simulated below?

In German schon is pronounced /

ʃoːn/.

I tried inserting a zero-width no-break code after the slash when using the template, but this causes it to throw a "unrecognized character" error. — Sgconlaw (talk) 16:44, 8 October 2024 (UTC)

We could add word joiners after opening and before closing at Module:IPA#L-538. It looks as though this would solve the problem. (Using CSS would too, but then IPA pronunciations would never wrap, which is not correct behaviour.) Would the word joiners cause trouble on any known system, or with copy-pasting? This, that and the other (talk) 09:09, 9 October 2024 (UTC)

Change the link m:Help:Page name#Restrictions to mw:Manual:Page naming#Page title restrictions. Also while you're at it you may want to consider adding a similar message to MediaWiki:Title-invalid-characters which is more likely to be seen. * Pppery * _{it has begun...} 23:35, 8 October 2024 (UTC)

@Pppery:  Done — would it make sense for MediaWiki:Title-invalid-characters to be an exact copy? Ioaxxere (talk) 01:40, 9 October 2024 (UTC)

Suggested code:

<div class="plainlinks" style="background-color:#FFE7DD;border: 1px dashed #884444;width:90%;margin:auto;text-align:left;text-align:center;margin-bottom: 10px">
Sorry, titles containing the character $1 are ]. It is possible that the entry you are looking for is below; if not, please ] it or  yourself.
</div>
{{Appendix:Unsupported titles}}
{{Appendix:List of protologisms/Long words}}

Almost an exact copy, but I modified the first sentence a little to include the specific invalid character. Some more advanced logic could in theory be added but that already exists in MediaWiki:Gadget-UnsupportedTitles.js so I wouldn't bother. * Pppery * _{it has begun...} 01:42, 9 October 2024 (UTC)

Done, I also tweaked the CSS a little. Ioaxxere (talk) 02:28, 9 October 2024 (UTC)

https://en.wiktionary.orghttps://en.wiktionary.org/w/index.php?title=Special:Search&limit=500&offset=0&ns0=1&search=insource%3A%22NavFrame%22 Ioaxxere (talk) 03:20, 9 October 2024 (UTC)

Your post is wildly obscure, so I have to tease out what you're trying to tell us, but it seems like, e.g. ორომს has a declension table that was inserted as HTML rather than MediaWiki syntax. Looking at {{lzz-conj-table}}, it is very different from the conjugation table in this entry, so I suppose that User:Gubazes inserted it by hand deliberately. Maybe he can tell us more about what happened there? And maybe you can elaborate on what you're trying to tell us with this thread or if you have a proposal for what to do? —Justin (koavf)❤T☮C☺M☯ 05:48, 9 October 2024 (UTC)

Right, let's get to the point. The first, tables are different because it represents intransitive and transitive verbs, whereas ორომს (oroms) is bitransitive (they are conjugated for both the subject and the object). It's irregular, that's why it's been insterted by hand. I tried to create a Laz conjugation table for the template so people could see what the conjugation looks like, but I failed. I couldn't figure out how to represent tenses, which to include and which to not, how to determine root, preverbs and so on. I think the template page needs to be deleted until someone with more experience with Wiktionary templates can create a better script. Gubazes (talk) 13:34, 16 October 2024 (UTC)

The current alphabetical order on Pannonian Rusyn (in Category:Pannonian Rusyn lemmas and so on) is wrong. I've copied the correct one into T:list:Cyrillic script letters/rsk, and I can also confirm it's correct because it's the one used by the 2010 dictionary (T:R:rsk:RSS) which I use as my main reference, and publications from the 70s are listed on the Wikipedia page which state the standardized orthography. Can someone more technologically savvy than myself fix the order? Thanks. Insaneguy1083 (talk) 09:57, 9 October 2024 (UTC)

@Insaneguy1083: Done: diff. It might take a while for everything to update itself to the new order. — Vorziblix (talk · contribs) 21:29, 9 October 2024 (UTC)

Thanks! Insaneguy1083 (talk) 21:32, 9 October 2024 (UTC)

@Vorziblix: actually, I think you did it wrong? Г and Ґ are treated as separate letters, as are И and Ї and Е and Є. They're not meant to be collated together. Insaneguy1083 (talk) 21:35, 9 October 2024 (UTC)

@Insaneguy1083: They are not collated as if they are the same letter—Ґ will always come after Г, and so forth. The sorting itself should be correct. Unfortunately the headings in the category pages will treat them as the same letter, but there’s no way around that—it’s the same situation we have in Ukrainian, etc. — Vorziblix (talk · contribs) 21:42, 9 October 2024 (UTC)

@Vorziblix: Hmm. Curious. Well, thanks a lot for updating the sorting anyway. If there's any updates on being able to separate Г and Ґ etc. in Ukrainian/Carpathian Rusyn, do let me know. Insaneguy1083 (talk) 21:49, 9 October 2024 (UTC)

There is a way around that these days - we have a gadget for this very purpose. — SURJECTION ^{/ T / C / L /} 19:23, 13 October 2024 (UTC)

This changed just recently. I assume it's a wiktionary thing, because control-x, c and v for 'cut', 'copy' and 'paste' still work properly on this page, in the url window, and in edit windows on WP.

Currently, in the edit window of a wikt article, if I hit control-x I get bold formatting, with control-c I get italic, and with control-v I get superscript. Presumably this has something to do with me using a dvorak keyboard (dvorak x and c correspond to qwerty b and i), but other commands are not affected. E.g. control-z and -y are still 'undo' and 'redo', despite corresponding to the qwerty keys for t and /. Also, as noted, this page is not affected: All keys work properly right now in this edit window.

Can I do something with my css to override this behaviour? kwami (talk) 02:12, 10 October 2024 (UTC)

Which skin are you using? —Justin (koavf)❤T☮C☺M☯ 05:42, 10 October 2024 (UTC)

Vector 2022. I just switched to Vector legacy, and the behaviour is the same. kwami (talk) 06:26, 10 October 2024 (UTC)

The same thing is now happening on WP, so I'll ask there as well.

BTW, it occurs here if I edit the thread, but not if I 'respond' to a comment, opening a new window. kwami (talk) 00:13, 14 October 2024 (UTC)

Update: On WP they said that the patch that broke this has been reverted, and will be going out on thursday. Meanwhile, for anyone reading, the syntax highlighter breaks this function and can be used as a temp fix. kwami (talk) 01:28, 14 October 2024 (UTC)

Hi all! I've been adding photos to plant family name definitions, and I'd love to be able to use the photo montage template from Wikipedia; it'd be ideal to show more of the taxon's range of diversity, rather than just the type species, and this would be much faster than creating and uploading an individual collage image for each family.

Is there any equivalent template on Wiktionary, something that'll take a few Commons images and arrange them in the space of one thumbnail? If that doesn't already exist here, is there any reason I can't or shouldn't just copy and paste the module code from Wiki -- citing the source, of course? Can templates be ported that easily? ("That's not how any of this works" is an extremely valid answer, especially seeing as I've never written a Mediawiki template before! I'm willing to learn, though.)

Thanks for your time. --Photosynthetic430 (talk) 18:56, 11 October 2024 (UTC)

@Photosynthetic430 you're more than welcome to give it a go. As this template is Lua-based, you'd need to copy w:Module:photo montage too. That module does not appear to have any dependencies, so it should copy across cleanly. This, that and the other (talk) 03:20, 14 October 2024 (UTC)

On it, then! I'll report back if it works, haha. -- Photosynthetic430 (talk) 20:52, 14 October 2024 (UTC)

Looks like it works! Now to finish removing the 'pedia-specific bits of the documentation. -- Photosynthetic430 (talk) 21:22, 14 October 2024 (UTC)

In e.g. Category:Old English lemmas, the TOC is "A B C... Z", "aa... ba... ca... ... zz", with no "þ" AFAICT, so it's not obvious how to navigate to where the entries starting with thorn are. In fact, trying to manually change the URL to reach them, I notice that they aren't sorted under either thorn or th. I eventually tracked down that they, and eth, are sorted after tz; whether this makes sense I don't know, but it'd be useful to add to the TOC a thorn anchor pointing to where they are. (I have forgotten offhand where the TOC is generated from.) - -sche (discuss) 00:50, 13 October 2024 (UTC)

@-sche I think you have to make {{ang-categoryTOC/full}}, noting that {{ang-categoryTOC}} exists for categories with between 200 and 2500 entries. We could also do with {{is-categoryTOC/full}}. This, that and the other (talk) 04:37, 17 October 2024 (UTC)

Is there a working French lookup extension for French Firefox? Ineuw (talk) 04:33, 13 October 2024 (UTC)

@Ineuw: You click onto the loupe in your search bar when on the French Wiktionary mainpage, click its icon for “adding French Wiktionary”, and Bob’s your uncle. Ctrl + K focuses the search bar. Fay Freak (talk) 09:01, 15 October 2024 (UTC)

I've listed myself as mzs-1 on my user page, but the message for that (i.e. "This user has basic knowledge" blah blah blah) is in English. Where can I go to write these messages for Macanese? Insaneguy1083 (talk) 10:07, 13 October 2024 (UTC)

They are known as infoboxes or userboxes. Look at my collection of infoboxes and if you have questions, post back here.Ineuw (talk) 18:28, 13 October 2024 (UTC)

Insert code like:

{{#babel:en|pt-3|mzs-1|de-1}}

etc. Let me know if that's confusing. —Justin (koavf)❤T☮C☺M☯ 18:33, 13 October 2024 (UTC)

I already did that. That's not the issue - the issue is that the message is in English, because no one's written a personalized one for Macanese yet, and I was wondering where I could do that. Insaneguy1083 (talk) 18:35, 13 October 2024 (UTC)

Oh, I see. translatewiki: is where these messages are translated. —Justin (koavf)❤T☮C☺M☯ 18:36, 13 October 2024 (UTC)

While going through WT:Todo/Lists/Broken links to Wikipedia, I noticed a number under Aragonese. These are governed by Module:labels/data/lang/an, and I saw that the bad links were only from those labels that had { } tables in their Wikipedia parameter. From the documentation, I would have expected these to generate a link to the first of the values in the list:

labels = {

Wikipedia = {"es:Aragonés belsetano", "Bielsa"},

regional_categories = true,

should link to w:es:Aragonés belsetano. Instead it links to w:Belsetán- a behavior the documentation says should happen when Wikipedia = true

Am I missing something, or is this an undocumented bug/feature? (I should add that the linking on the category pages such as Category:Belsetán Aragonese seems to work as described, so something would be lost by simplifying it to :Wikipedia = "es:Aragonés belsetano"},). Chuck Entz (talk) 21:41, 14 October 2024 (UTC)

This sounds like a bug. I'll take a look. Benwing2 (talk) 22:17, 14 October 2024 (UTC)

Fresh out of QQ (but with underlining added), a cite often looks like:

• 2016 April 28, Andrew Picard, Myk Habets, Theology and the Experience of Disability: Interdisciplinary Perspectives from Voices Down Under, Routledge, →ISBN, page 51:

  ... the saltland for its dwelling place ? It scorns the tumult of the city ; it does not hear the shouts of the driver . It ranges the mountains as its pasture , and it searches after every green thing . ( Job 39 : 5-8 ) Yahweh assigns 

Could QQ be made slightly smarter, to remove the extraneous spaces? (This has been suggested before.) - -sche (discuss) 23:13, 15 October 2024 (UTC)

These terms, such as татуируя (tatuiruja), should be in Cat:Russian non-lemma forms but are not for some reason. I note that Russian present active/passive participles do appear to be properly categorised. Can anyone advise? This, that and the other (talk) 06:02, 16 October 2024 (UTC)

Is it really as simple as this? I searched module space for the singular form (mod:"present passive participle") but I guess I should have looked for the plural. This, that and the other (talk) 06:35, 16 October 2024 (UTC)

Yeah that should be all you have to do. Benwing2 (talk) 05:45, 17 October 2024 (UTC)

I wish to request new templates for creating links to thesaurus and appendices more easily. similarly to the templates such as {{temp}} (creates a link to templates) and {{w}} (to Wikipedia), these should exist so that we do not need to tediously type the whole link. examples below:

{{thes|en|good}} gives ]

{{snowclone|en|X with a capital Y}} gives ]

there are likely other usecases that I haven't thought about, please feel free to reply with some. Juwan (talk) 08:27, 19 October 2024 (UTC)

Is there a reason why these need language parameters? I don't know that we have a lot of multilingual appendices or thesaurus entries... —Justin (koavf)❤T☮C☺M☯ 12:01, 19 October 2024 (UTC)

{{thes}} exists now. —Justin (koavf)❤T☮C☺M☯ 12:07, 19 October 2024 (UTC)

thank you very much! I would prefer them (at least the thesaurus one and other specialised templates) to have a language parameter, because 1. it is useful for the few entries that are multilingual, 2. it is already required that all thesaurus pages must have a language header, so why not make use of it, and 3. these can bring some help for accessbility for screen readers reading the title (if that implemented in the future). the last one is a whole can of worms to itself but I digress. Juwan (talk) 12:42, 19 October 2024 (UTC)

@JnpoJuwan why do you want this? When inlining synonyms (as we should be doing) using the {{syn}} template, you can write {{syn|en|...|Thesaurus:happy}} to get

Synonyms: ...; see also Thesaurus:happy

This, that and the other (talk) 21:54, 19 October 2024 (UTC)

@This, that and the other it's used outside of {{syn}}! a big example is under the see also header in the thesaurus namespace or when mentioning it in talks. Juwan (talk) 23:23, 19 October 2024 (UTC)

If anything, a more generic Appendix link would maybe be useful for typing, like {{ap}}. —Justin (koavf)❤T☮C☺M☯ 12:10, 19 October 2024 (UTC)

that's true! but why not both? that was just one example appendix among many that I could have chosen. Juwan (talk) 12:43, 19 October 2024 (UTC)

I thought these were errors that would need to be fixed one at a time, until I noticed that there are "/testcases/" pages in CAT:Pages with module errors/hidden. Would it be possible to make all documentation pages of hidden pages also hidden? Pinging @Theknightwho. Chuck Entz (talk) 21:31, 19 October 2024 (UTC)

I request to change the entry of Peranakan Indonesian as follow.

m = { "Peranakan Indonesian", 653415, "poz-mly", "Latn", ancestors = "ms", } Xbypass (talk) 00:20, 20 October 2024 (UTC)

Trying out Vector 2022 dark mode, I notice that QQ still has a light background; could it be given a dark-mode-compatible version? - -sche (discuss) 19:26, 21 October 2024 (UTC)

@-sche:  Done Ioaxxere (talk) 03:15, 26 October 2024 (UTC)

Hello everyone, I previously wrote on the 27th September to advise that the Wikidata item sitelink will change places in the sidebar menu, moving from the General section into the In Other Projects section. The scheduled rollout date of 04.10.2024 was delayed due to a necessary request for Mobile/MinervaNeue skin. I am happy to inform that the global rollout can now proceed and will occur later today, 22.10.2024 at 15:00 UTC-2. Please let us know if you notice any problems or bugs after this change. There should be no need for null-edits or purging cache for the changes to occur. Kind regards, -Danny Benjafield (WMDE) 11:28, 22 October 2024 (UTC)

I wish to request that the following list of templates be deprecated and replaced with the template {{head}} in all appendix entries. toki pona, as an isolating language, does not have any type of inflection, making these templates simply redundant.

• {{tok-adj}} (adjectives)
• {{tok-adv}} (adverbs)
• {{tok-cardinal}} (cardinal numbers)
• {{tok-conj}} (conjunctions)
• {{tok-noun}} (nouns)
• {{tok-part}} (particles)
• {{tok-prep}} (prepositions)
• {{tok-pron}} (pronouns)
• {{tok-verb}} (verbs)

hope I didn't miss any, that'd be silly. Juwan (talk) 18:51, 22 October 2024 (UTC)

This is normally dealt with at WT:RFDO, but I agree that these hand-coded headword templates need to be done away with. This, that and the other (talk) 01:29, 23 October 2024 (UTC)

@This, that and the other I didn't know, sorry! moving discussion there. Juwan (talk) 08:14, 23 October 2024 (UTC)

in my opinion these could possibly be replaced by content words and particles, though i don’t know БудетЛучше (talk) 11:19, 10 November 2024 (UTC)

Per MediaWiki talk:Gadget-CodeLinks.js, I'm going to undefault CodeLinks.js and save our readers some bandwidth. Anyone who still wants the Wiktionary-specific features can reenable the gadget in their preferences. Ioaxxere (talk) 04:46, 24 October 2024 (UTC)

While going through Category:Greek terms in nonstandard scripts, I found a number of entries where the second positional parameter (originally the third) was a transliteration rather than the expected plural. This is because of changes (including a move and a change to a redirect target) starting a bit over a decade ago that converted things so the behavior of that parameter was completely different, but without updating a number of the entries.

There seem to be dozens of these, so I'm bringing it here to see how best to remedy this. There are well over a thousand entries in Special:WhatLinksHere/Template:el-noun-proper, so simply changing the parameter to match a minority of old uses doesn't look like a good idea- besides which, all the ones I've checked so far have a transliteration that's redundant to the automated one.

As I see it, there are two options:

1. Have a bot go through and remove all the transliterations from the second (or third?) parameter
2. Add a named parameter for the transliteration, then have a bot go through and convert the positional parameters with transliterations to the named parameter.

Either way, if the bot detects a transliteration by the presence of Latin script, there are a couple of Latin-script keywords that are used instead of the Greek-script plural for indeclinable and uncountable proper nouns- so those will have to be allowed for.

Pinging Greek editors for their opinions and insights- (Notifying Saltmarsh, Sarri.greek, Rossyxan, FocalPoint): Chuck Entz (talk) 14:59, 24 October 2024 (UTC)

@Chuck Entz hello. Could you please indicate two-three examples? This will help me in understanding the issue. FocalPoint (talk) 16:31, 24 October 2024 (UTC)

M @Chuck Entz, if you are in a hurry to clean up, either way is fine. Cat:Greek proper nouns was at my 'to do' list (1899 reviews, most of them have a plural to be added, +ipa), after I finish the pending corrections (approx.1000) at WT:LTR#plan for Medieval Greek. PS May I ask, about Categories. Sometimes Subcategory members are fully indexed at a supracategory, sometimes they are not (they are different, extra members). Is this indicated in some way? Thank you ‑‑Sarri.greek ^♫ I 16:45, 24 October 2024 (UTC)

On second check, @Chuck Entz, I need to do Category:Greek terms in nonstandard scripts by hand. The English ones are ok, The Greek are less than 180, and probably need more attention. ‑‑Sarri.greek ^♫ I 19:52, 24 October 2024 (UTC)

I wanted to add customer-side to customer#Derived_terms, but this was seen as harmful. 81.217.68.78 14:36, 25 October 2024 (UTC)

Unfortunately this is a global abuse filter, managed on Meta-Wiki, and we can't do anything about it here. The best way to avoid this type of issue is to simply create an account. This, that and the other (talk) 08:17, 27 October 2024 (UTC)

Hello! I was attempting to add a detailed, factual description for the African beaked snake (Rhamphiophis oxyrhynchus) but was blocked by an automated filter. My goal was to contribute additional information on the species' appearance, habitat, behavior, diet, reproduction, and interactions with humans, presented in an informative, neutral tone. I believe this content would enhance understanding of the species for readers.

If there’s a specific part that triggered the filter or needs revision, please let me know, and I’d be happy to adjust the submission accordingly. Thank you for reviewing this! PrajinExplores (talk) 17:32, 27 October 2024 (UTC)

Wiktionary is a dictionary. "dditional information on the species' appearance, habitat, behavior, diet, reproduction, and interactions with humans" does not belong in a dictionary. — SURJECTION ^{/ T / C / L /} 17:43, 27 October 2024 (UTC)

@PrajinExplores: You tried to create a long article that mostly didn't match Wiktionary formatting, and it was detected by a filter that was created to prevent people from posting long articles about themselves. The fact that you were doing it without an account probably didn't help. Really, the first few lines were all that would make sense for a dictionary entry, plus a single-sentence definition line.

What I saw in the filter log wouldn't be bad as a Wikipedia article, without the first few lines (they would only make sense in a Wiktionary entry) and with the addition of references and the standard Wikipedia templates. You might want to create a user page at Wikipedia, and submit the article as a draft. Chuck Entz (talk) 23:29, 27 October 2024 (UTC)

The second parameter should not default to {{{1}}}. This should be fixed; apparently, this is not the first time this has happened. -BRAINULATOR9 (TALK) 23:20, 27 October 2024 (UTC)

@Babr: who edited it fairly recently. —Justin (koavf)❤T☮C☺M☯ 00:56, 28 October 2024 (UTC)

It looks like @Almanbet Janışev is the one who edited the relevant part of the template a month ago, and they have a history of breaking things. Chuck Entz (talk) 02:48, 28 October 2024 (UTC)

MediaWiki message delivery 20:56, 28 October 2024 (UTC)

Hi, is there an Android app for Wiktionary? How does it work? I have been advised that there is no infrastructure for push notifications for Android apps for sister wikis and I would be interested to know more. Related: phab:T378545. Thanks! Gryllida 23:13, 29 October 2024 (UTC)

@Gryllida Others may be able to comment more, but I don't know of any mobile app for Wiktionary on any platform. Instead there is just the mobile website, which leaves a lot to be desired (but for which there are active efforts by editors like @Ioaxxere, @Surjection and others to improve its appearance). Benwing2 (talk) 05:49, 1 November 2024 (UTC)

There was a Wiktionary Android app on the Google Play store back in the day; I had it on my old phone. I think it's the one mentioned here: mw:Wiktionary Mobile. It was good enough for basic browsing but the Android UI of the app felt outdated even 10 years ago, so it's no surprise it was retired. This, that and the other (talk) 09:20, 2 November 2024 (UTC)

@Gryllida: The app English Dictionary - Offline uses Wiktionary for its data but unfortunately only covers English (the same company has versions in a couple other langauges as well). I wished for dedicated Wiktionary app as well but if no one steps up I guess I'll have to create it myself... Ioaxxere (talk) 02:46, 2 November 2024 (UTC)

I wonder if this would be a useful template for anyone else. for non-English languages, there should be a template where you may pass parameters and it would output a definition line regarding the pronoun, similarly to templates such as {{place}} and {{demonym-noun}}. this would be good for standardisation and harmonisation between entries and possibly helpful for categorisation.

{{pronoun|pt|1|sg}}

First-person, singular pronoun; I, me

{{pronoun|es|2|pl|f|inf}}

Informal, feminine, second-person, plural pronoun; you, you all

{{pronoun|pt|3|obj|clitic}}

Oblique, third-person, clitic pronoun; him, her, it, them

Juwan (talk) 14:56, 31 October 2024 (UTC)

@JnpoJuwan IMO this is a good idea, although I'm not sure how easy it is to automate the generation of the English-equivalent glosses. The way I imagine it, any arbitrary {{infl of}} tags could be supplied to {{pronoun}}, which would make it hard to come up with a generic algorithm to convert these tags into English-equivalent pronouns. This would be especially problematic, for example, with second-person pronouns, where English has only "you" and various dialectal plural forms (you all, you guys, y'all, yinz, etc.). For example, given the wealth of ways of saying "you" in European Portuguese (tu; archaic vós; você, vocês; o senhor, a senhora, os senhores, as senhoras; plus object variants te, ti, vos), any auto-generated English-equivalent gloss would be highly misleading. Instead you'd need a custom non-gloss definition following the output of {{pronoun}}. Japanese, among others, would be similar. Benwing2 (talk) 05:46, 1 November 2024 (UTC)

@Benwing2 I believe that the English glosses being "lossy" in detail is not a critical issue. these to me serve as a way for others to have a basic idea of what pronoun is roughly equivalent in English. all the variants of "you" in Portuguese glossed as "you" is okay, as the rest of the non-gloss serves to disambiguate what is different. this is how glosses already work in some way right now. the template having a dedicate gloss template to override the automatic gloss would be a good compromise to me. Juwan (talk) 18:50, 1 November 2024 (UTC)