User talk:Benwing2

Archive

• 2012-2019
• 2020-2021
• 2022
• 2023
• 2024

Regarding less than signs for hypernymy and holonymy

Hi Benwing — thanks for your efforts. Sorry to trouble you by it. My development of using the less than signs (<) was a nice idea, something that it would be nice if any wordnet-capable type of dictionary (such as WordNet exports or Wiktionary) could helpfully do for its users, to show wherever invariable "levels" of hypernymy or holonymy exist (for example, that a neutron is a type of baryon and a baryon is a type of subatomic particle). It has good teaching and learning value in tying vocabulary to semantic concepts clearly, showing clearly how things relate. The idea is that it would only be done in the strict/true class of invariably hierarchical relation. So the one where I accidentally did it for orange < fruit < food was imperfect and would have needed correction because not all fruits are foods. But it's still a cool and useful idea when applied to the "strictly/truly" class, though (like the neutron example). However, if there is some issue of template performance being hurt by it, then unfortunately I guess I have no choice but to abandon it for Wiktionary's purposes, as a good idea whose implementation can't happen here. Alas, R.I.P. good idea. I tried using a character entity instead (<), but it doesn't work. If you or someone else who is good at editing templates ever were willing to build a capability for this feature into the templates, I hearily endorse it, but I'm probably not the one who can or will achieve it, alas. Maybe someday. Quercus solaris (talk) 15:22, 8 January 2025 (UTC)

@Quercus solaris I see. As a short-term fix maybe I can work around this syntax. In the longer term we'd need a better way of showing this. Benwing2 (talk) 15:27, 8 January 2025 (UTC)

@Quercus solaris I fixed things in a way I think is pretty good, which is to allow < to be a separator in its own parameter. See junk DNA or rapeseed oil for examples. We should redo the ones where I or someone else removed the < sign. Benwing2 (talk) 16:27, 8 January 2025 (UTC)

Wow, that's amazingly productive and rapid of you. My hat is off to you. Many thanks! Quercus solaris (talk) 16:37, 8 January 2025 (UTC)

@Quercus solaris You're welcome. Here is the list of places where a < sign occurred with {{hyper}} as of Jan 1: User:Benwing2/hyper-less-than-2025-01-01-dump. Benwing2 (talk) 16:50, 8 January 2025 (UTC)

Regarding your reverts on my Old English edits

Hello Benwing2,

I hope this message finds you well and that you've had a lovely holiday season.

I saw that you have reverted many of my edits and calling them questionable.

Well, I am happy to inform you that these edits are attested in the Anglo-Saxon gospels published by Walter Skeat. They are specifically from the M. S. V Hatton manuscript.

Have a nice day.

Leornendeealdenglisc (talk) 07:03, 9 January 2025 (UTC)

Can you give me an example? You need to indicate the source of alternative forms and such rather than just randomly sticking them in. Benwing2 (talk) 00:21, 10 January 2025 (UTC)

Templates like pl-adjective

Hello

I noticed this template like many others is incompatible with Dark Mode, due to having hardcoded colors. I can do the refactoring to make it at least accept foreground and background colors as constants; that'd be a start. Alternatively, default coloring could be reassumed.

Please advise

Regards Petros Adamopoulos (talk) 17:52, 9 January 2025 (UTC)

Column templates

I think that JS code is ready to be translated to Lua. I just tested it on a random selection of pages using col and colN templates (in Vector 2022) and it produces acceptable results on all of them except 頗為, where it makes the columns slightly wider than necessary.

Two comments:

• There's a bug in the JS when used on a list with <= 8 entries - it tries to read one past the end of the array - but I'm sure you would have noticed that anyway when translating to Lua!
• We have to be careful in the following cases:
  • 9 items in 4 columns results in a blank column on the right (at least in my browser):
    • a
    • b
    • c
    • d
    • e
    • f
    • g
    • h
    • i
  • Similar issues with 11, 12 and 16 items in 5 columns:
    • a
    • b
    • c
    • d
    • e
    • f
    • g
    • h
    • i
    • j
    • k
    • a
    • b
    • c
    • d
    • e
    • f
    • g
    • h
    • i
    • j
    • k
    • l
    • m
    • n
    • o
    • p
• So I think we should set CSS column-count to 3 when there are <= 9 items and 4 when there are <= 16 items, noting that when both column-count and column-width are specified, column-count enforces a maximum possible number of columns. (This assumes that no items break over multiple lines, which is the usual case.)

This, that and the other (talk) 07:08, 13 January 2025 (UTC)

Thanks! Will implement. Benwing2 (talk) 07:11, 13 January 2025 (UTC)

bot request for Sanskrit quotations

I'd like to do a very basic bot request (for WingerBot) and I apologise if this is not the right place. What should happen is that sequences of Vedic accent (either ॑ U0951 or ॒ U0952) + colon (U003A) should be changed to visarga (U0903) + the respective Vedic accent. This to correct wrongly formatted text that was copy-pasted from a site like wisdomlib.org. Exarchus (talk) 10:44, 15 January 2025 (UTC)

Can you give me an example or two where you've made the change? And does the visarga go before the accent or after? Benwing2 (talk) 10:54, 15 January 2025 (UTC)

Here and here are examples (I also added a space before the danda's but let's ignore that for now).

The visarga should go before the accent. Exarchus (talk) 11:30, 15 January 2025 (UTC)

@Exarchus Sorry for the delay. I made the changes offline but I'm not sure they are correct; can you tell me if the following diff is correct?

@@ -199,7 +199,7 @@
 # [[bright]], [[resplendent]]
 #* {{Q|sa||RV}}
 # [[clear]], [[pure]], [[spotless]]
-#* {{Q|sa||RV|4|51|9|quote=ता इन्न्वे॒३॒॑व स॑म॒ना स॑मा॒नीरमी॑तवर्णा उ॒षस॑श्चरन्ति।<br> गूह॑न्ती॒रभ्व॒मसि॑तं॒ रुश॑द्भिः '''शु॒क्रा'''स्त॒नूभि॒: शुच॑यो रुचा॒नाः॥|t=Thus they go forth with undiminished colours, these Mornings similar in self-same fashion, concealing the gigantic might of darkness with radiant bodies bright, '''pure''', and shining.}}
+#* {{Q|sa||RV|4|51|9|quote=ता इन्न्वे॒३॒॑व स॑म॒ना स॑मा॒नीरमी॑तवर्णा उ॒षस॑श्चरन्ति।<br> गूह॑न्ती॒रभ्व॒मसि॑तं॒ रुश॑द्भिः '''शु॒क्रा'''स्त॒नूभिः॒ शुच॑यो रुचा॒नाः॥|t=Thus they go forth with undiminished colours, these Mornings similar in self-same fashion, concealing the gigantic might of darkness with radiant bodies bright, '''pure''', and shining.}}
 # light-coloured, [[white]]

 ====Declension====

The line preceded by - is the existing line and the line preceded by + is what it will change to. The difference is in the third-to-last word in the Devanagari. Benwing2 (talk) 10:36, 21 January 2025 (UTC)

Yes, this is correct. Exarchus (talk) 10:48, 21 January 2025 (UTC)

Hi, I still find 64 results when looking for U0951 + colon. Maybe you only fixed U0952 + colon? Exarchus (talk) 09:04, 3 March 2025 (UTC)

I fixed both so I don't know what happened. It's a big pain working with Unicode combining chars so maybe I messed something up. I will try again doing the two combining chars separately. Benwing2 (talk) 09:45, 3 March 2025 (UTC)

Ahhh, I see now, I accidentally wrote 0591 instead of 0951 in the initial search. Benwing2 (talk) 09:46, 3 March 2025 (UTC)

I see you fixed it, thanks. Exarchus (talk) 10:16, 3 March 2025 (UTC)

Yes, it's done. Benwing2 (talk) 10:17, 3 March 2025 (UTC)

caps

Hi. Could you fix the caps of the last 4 letters at [Template:list:Greek script letters/aat]? I can't move pages w/o a rd, plus I don't want to get into an argument about 'vandalism' again. kwami (talk) 02:05, 20 January 2025 (UTC)

I'll respond to this and your other post tomorrow, going to bed now. Benwing2 (talk) 10:47, 20 January 2025 (UTC)

errors with WingerBot

FYI, you've used this with topics and not just glosses, so eg at ◌̯ we said,

(antonym(s) of “IPA”): ◌̩
(antonym(s) of “NAPA”): ◌̣

I changed those back to the 'sense' template. kwami (talk) 03:41, 20 January 2025 (UTC)

re phab:T384134

Are you absolutely sure this is a regression? One of the test cases for subpageText shows that the expected output for "Talk:Has/A/Subpage" should be "Subpage", not "A/Subpage". This matches the behaviour of MediaWiki core: {{SUBPAGENAME:Reconstruction:Test/one/two}} → two This, that and the other (talk) 01:19, 24 January 2025 (UTC)

@This, that and the other Yes. This used to function differently and has been recently changed. I think MediaWiki has a per-Wiki, per-namespace setting that indicates whether the above subpage behavior you're describing is enabled. It is clearly not enabled for Wiktionary mainspace since pages like 8514/A display correctly and are not truncated after the last slash, but somehow has gotten enabled for the Reconstruction namespace. Benwing2 (talk) 01:49, 24 January 2025 (UTC)

Probably a dumb question.

With regards to "Category:fr:Terms derived from Old French nominative", where do I go to properly create it (other than the grease pit)? Is there any button to press that takes you to an editing window to put in the JavaScript or Lua or whatever else to build it?

These questions betray both my ignorance of how coding & the category system on Wiktionary work.

If you believe that the category is not worth the trouble of creating, then I will work on something else. :) Pvanp7 (talk) 13:26, 1 February 2025 (UTC)

Hi so a couple of things. First, if we were to have the category the name should be 'Category:fr:Terms derived from the Old French nominative case' or similar. But in fact we used to have a series of such categories, one for each Romance language, and they were eventually removed and converted into an appendix Appendix:Survivals of the Latin nominative in Romance because in categories you're not able to include any discussion of the questionable cases or present the evidence for such a survival (and there are several cases where there are disagreements over the etymology of the word). Cf. @Nicodene who created the appendix. One thing we could do is link the appendix to the etymology section of each term mentioned in the appendix, perhaps via a template. As for how to add a category to the category system, once it's added there's an edit button that takes you to the place where it's defined, but there's no button to take you to the right place to define it if it doesn't already exist, because it's in general impossible for the system to determine that. I think the best we could do is link to Module:category tree/poscatboiler/data/documentation, but that documentation is pretty detailed and not super beginner-friendly. (And in fact we've set the permissions to autopatroller so that only people who are known to be good contributors can modify the underlying modules, because some modules kept getting messed up by people who didn't know what they were doing.) Benwing2 (talk) 20:28, 1 February 2025 (UTC)

So the category already sort of exists & no it's not really possible for me to create categories on my own.

Thanks so much for taking the time to write this!

Pvanp7 (talk) 22:23, 1 February 2025 (UTC)

Sure, let me know if you need something done. Benwing2 (talk) 22:36, 1 February 2025 (UTC)

cyrillic use of latin letters

I added Cyrillic use of k, l, w in phonetic transcription. An example of a source that has uvular [k] and dental [l] is Рамстедт Г.И., Сравнительная фонетика монгольского письменного языка и халхасско-ургинского говора, which is available online. kwami (talk) 03:48, 2 February 2025 (UTC)

Automated Baybayin entries

Hi @Benwing2, is it possible to automate Baybayin page creation (just one run for all right now) by simply copying the main entries etymology count (or no etymology if it's just one etymology) and part of speech then simply adding a {{tl-bay|<entry redirect>}} to them? That's usually the Baybayin pages right now anyway. The Baybayin page to be created is the one defined per entry.

example

Etymology 1

Noun

Benwing2 (Baybayin spelling ᜊᜒᜈ᜔ᜏᜒᜅ᜔2)

1. a Wiktionary admin

would be translated as

Etymology 1

Noun

ᜊᜒᜈ᜔ᜏᜒᜅ᜔2 • (binwing2)

1. Baybayin spelling of Benwing2

𝄽 ysrael214 (talk) 23:04, 2 February 2025 (UTC)

@Ysrael214 Hi, this is possible but it will take some work, so I can't give you a timeline as to when it will be done. Benwing2 (talk) 00:20, 3 February 2025 (UTC)

@Benwing2 No problem, thanks still! 𝄽 ysrael214 (talk) 01:12, 3 February 2025 (UTC)

Wiktionary:Todo/Lists/Entries with missing headword lines#Icelandic

I do hope you get the rest of the bugs worked out of the Icelandic templates, because right now no one else has a prayer of fixing problems like this. Chuck Entz (talk) 00:03, 3 February 2025 (UTC)

Thanks for the heads-up. The issue with those terms is a bit tricky to solve but I'll figure it out since it seems like Fia has been adding stuff that's triggering it. Benwing2 (talk) 00:12, 3 February 2025 (UTC)

@Chuck Entz all right, they should be fixed; for these words the derivatives don't have the obsolete alternatives that were added, in any case. Benwing2 (talk) 00:19, 3 February 2025 (UTC)

There are still 4 of these, though I suppose they might not be the same as the February ones. Chuck Entz (talk) 22:23, 22 June 2025 (UTC)

⁅ and ⁆

I'd like to create articles for these, but AFAICT they're only used in Swedish dictionaries, and it would be weird to present the Swedish usage as multilingual. Can I create a Swedish entry for them? kwami (talk) 19:05, 7 February 2025 (UTC)

@Kwamikagami IMO it's really not on my talk page but in the Beer Parlour where you should be posting this sort of thing. There isn't a clear consensus that I know of about how to handle such symbols and it would be good to get input from various people. Benwing2 (talk) 19:58, 9 February 2025 (UTC)

okay, i asked there kwami (talk) 00:07, 10 February 2025 (UTC)

Replacement of quotation templates (February 2025)

Hi, could you please do the following replacements when you have the time?

• {{RQ:Beckett Watt}} – 1st edition (1953) has been added to the template. If |page= specified, add |edition=US to the template.
• {{RQ:Elyot Governour}} – 1st edition (1531) has been added to the template. If |page= specified, add |year=1907 to the template.
• {{RQ:Hollinghurst Line}} – if |page= specified, add |edition=US to the template.
• {{RQ:Jonson Alchemist}} – 1st edition (1612) has been added to the template. If |page= specified, reduce the page number by 1 (for example, if |page=31 is specified, change it to |page=30). (Note to self: reverberate already refers to 1st edition and doesn't require changing.)
• {{RQ:Milton Divorce}} – 1st edition (1643) has been added to the template. If |page= specified, add |edition=2nd to the template. (Note to self: waver already refers to the 1st edition and doesn't require changing.}}

Thank you. — Sgconlaw (talk) 18:48, 9 February 2025 (UTC)

Just a reminder about this. — Sgconlaw (talk) 18:31, 8 March 2025 (UTC)

@Sgconlaw Apologies, I meant to get them earlier. I'll get to them today. Benwing2 (talk) 19:46, 8 March 2025 (UTC)

@Sgconlaw Should be done. Benwing2 (talk) 03:23, 25 March 2025 (UTC)

Thanks! — Sgconlaw (talk) 23:04, 27 March 2025 (UTC)

Doublets gone away

Hi, Category:Doublets by language is empty. I think it's because of last edition in Module:etymology/templates. Vriullop (talk) 13:52, 10 February 2025 (UTC)

Old Czech adjective headword

Hi, about the Module:cs-sk-headword, it doesn't make the superlative short forms in Old Czech adjectives automatically from the short comparative (e.g. the page mladý shows najmlazší by providing mlazší, but not najmlazí by providing mlazí). The previous template did that. Could you please repair that? You must have missed my message in December. Zhnka (talk) 13:05, 17 February 2025 (UTC)

Incorrect language header cats

Oddly enough, these were very useful: I went to the parent category and did a cmd-f search for "0 e" to see how many categories on the page were non-empty. Generally, there would be newly non-empty categories more than once a day on average, and a fair percentage were types of vandalism and other bad edits that weren't easy to find by other means.

While it's true that all but a few dozen will probably never have anything in them again, it would be nice not to have to wait until the relevand Todo lists run once a week or you do your WantedCategories runs even less frequently. Is there any way you could add a Wiktionary:Tracking link at the same time the tracking category is generated? Maybe Wiktionary:Tracking/incorrect language header? Chuck Entz (talk) 03:11, 23 February 2025 (UTC)

Apologies for deleting them, I figured most would never be non-empty. FWIW I do WantedCategories runs every 3 days; what happens infrequently is deleting categories. I will add the tracking link. Benwing2 (talk) 03:15, 23 February 2025 (UTC)

arrow rd's

I removed the rd from ➡ because it wasn't defined at the target, but I don't have a definition for it as its own article. It should probably just be deleted. I noticed the rd wasn't appropriate when someone removed a dab from a definition at the target. I rv'd them because we needed to support the rd, but then realized they were right, and the rd shouldn't have directed here. [To be consistent, if we kept the rd, all or at least most of the definitions would need the same dab, which seems undue.] Other arrows have similar rd's. kwami (talk) 07:13, 25 February 2025 (UTC)

Yeah hard redirects are generally not a good idea and it seems to me all Unicode chars should have their own pages unless there's a really good reason not to. Benwing2 (talk) 07:20, 25 February 2025 (UTC)

San Antonio

You may not have noticed, with the flood of module errors demanding asterisks in display forms, but this entry has been intermittently showing up in CAT:E with Lua timeout errors since your last edits to Module:place: it has over 200 transclusions of {{place}}. Chuck Entz (talk) 15:55, 26 February 2025 (UTC)

Thanks. In fact I didn't notice, partly due to those errors and partly because my edits are occasionally triggering other errors that I have been cleaning up. It's not there currently but let me see what I can do. Might be a good idea to consider using mw.loadData() on the place data modules; I didn't do it before because it actually increased the memory on some pages but if memory isn't the issue it should definitely help. But probably there is a lot to optimize even without that; a lot of unnecessary processing is happening in the data modules. Benwing2 (talk) 16:43, 26 February 2025 (UTC)

Укротимый

Hello, Benwing2 NickRzh (talk) 21:53, 5 March 2025 (UTC) You created a page for the word «укротимый», but it simply does not exist in the Russian language. The Russian National Corpus does not have any mentions of it. This is a type of word that is used only in negative form, неукротимый.

@NickRzh place an {{rfv|ru}} template on the page, save, and then click on the (+) sign, which will send the word to WT:RFVN, where you can challenge it. Benwing2 (talk) 22:25, 5 March 2025 (UTC)

handgag

I have requested a deletion to replace the document, but there are too many documents to be deleted, so I am contacting you. Category:Request templates YeBoy371 (talk) 10:03, 7 March 2025 (UTC)

@YeBoy371 Using the speedy deletion tag is a misuse of that tag. You should use {{rfd}} and post your rationale in WT:RFDE. Also please don't ping administrators with deletion requests, we're very busy. Someone will eventually get to it. Benwing2 (talk) 10:50, 7 March 2025 (UTC)

Please

How do I edit in wiktionary without getting banned? Please? 5.82.167.107 12:57, 7 March 2025 (UTC)

Who are you? Benwing2 (talk) 20:29, 7 March 2025 (UTC)

yucatec entries

FYI, I'm moving Yucatec entries substituted with curly apostrophies to the ejective/glottal stop letter, per the majority of entries. ASCII apostrophes I'm leaving as an alt form, though we don't do that in most cases. kwami (talk) 20:20, 7 March 2025 (UTC)

Can you give me an example? Benwing2 (talk) 20:28, 7 March 2025 (UTC)

Sure. I moved tseek’ to tseekʼ. That's what we do for the vast majority of entries, plus it's the Unicode assignment. We get duplicate articles and red links when the wrong character is used like this. kwami (talk) 20:31, 7 March 2025 (UTC)

I see, that's fine. Generally there is some disagreement when it comes to converting straight ASCII apostrophes to any sort of Unicode apostrophe-like character, but not from one such character to another. Benwing2 (talk) 20:34, 7 March 2025 (UTC)

Well, for ASCII we get incoherent results such as the plural of t'uut' being t'uut'oʼob, with a proper glottal stop in the suffix. But those I'm leaving as alt forms, to avoid duplicates and because they're likely search terms.

There are also cases where a root has an ASCII apostrophe but its derivation has a glottal stop. kwami (talk) 20:39, 7 March 2025 (UTC)

Yeah that's a mess. Benwing2 (talk) 21:07, 7 March 2025 (UTC)

Okay, the only article that still has a curly apostrophe and is not tagged for deletion [after moving] is peek’. It has a page history, but before I could move it someone had created a duplicate article at peekʼ. I don't know if it's worth doing a history merge. kwami (talk) 23:37, 7 March 2025 (UTC)

links to reconstructions on French wiktionary

Not very urgent, but the search "insource:/fr:Annexe:/" in the Reconstruction namespace gives 93 results. "Annexe" should be changed to "Reconstruction". Exarchus (talk) 15:04, 8 March 2025 (UTC)

Nevermind, I've done it myself as I can now use JS Wiki Browser. Exarchus (talk) 15:24, 23 March 2025 (UTC)

Category:Latvian definite adjectives, Category:Latvian adverbial forms of adjectives

Hi Benwing! How are you? I hope you are well. Anyway, I would like to ask for your help so that I can clean up the Latvian categories. There are a ton of "useless" categories that were created more than 10 years ago that we can consider obsolete today, so I would like to ask for your help to delete this category and remove from the headword of all adjectives the function that applies this category to them with the help of a bot. There are several other categories with these same problems that I also want to delete, but these are the ones that I believe are most "urgent". Anatolijs LTV (talk) 23:14, 16 March 2025 (UTC)

The definite adjective forms need to be converted to non-lemma forms (by bot). As for Category:Latvian adverbial forms of adjectives, the definition of {{lv-adv}} should be changed to not generate it, but I don't know enough about Latvian grammar to know whether these should be classified as adverbs (i.e. lemmas) or as non-lemma adjective forms. In other languages, adverbs are lemmas, so all we would need to do is change the definition of {{lv-adv}}, but maybe Latvian is special. Benwing2 (talk) 23:23, 16 March 2025 (UTC)

I don't believe that Latvian works "differently enough" from other languages to need to be categorized. What really happened is that the creator of most (if not all) Latvian templates had a habit of putting automatic categories in all of them, although some of them are actually useful. All that needs to be done is that the definite forms are converted to non-lemma and that the adverbial forms are categorized as just adverbs like in any other language, there is no need for a separate category. Anatolijs LTV (talk) 23:29, 16 March 2025 (UTC)

@Anatolijs LTV I eliminated those categories (as well as the corresponding comparative categories) and did a big cleanup of Latvian non-lemma forms (which is still saving), making the lemma be the indefinite form instead of the definite form and combining the resulting {{infl of}} tags whenever possible. I have a question about superlatives, though. The superlatives are lemmatized at the definite form, e.g. visabstraktākais; is this correct? If so, I need to fix {{lv-decl-adj}}, since I made it always output both indefinite and definite forms but I need to make it conditionalize for these superlatives to only output the definite forms. Benwing2 (talk) 06:32, 17 March 2025 (UTC)

Thank you very much for your work! Both the comparative and superlative definite forms should be characterized as non-lemma, i.e. it is already good as you left it. But there is still something "bothering" me; is it necessary to specify "the comparative degree of" in non-lemma pages like in būtiskākais? Once again, thank you very much for your work 😄 Anatolijs LTV (talk) 13:32, 17 March 2025 (UTC)

And I would also like to eliminate that "adverbial form of" text in adverbs like būtiski, in case you don't disagree. Anatolijs LTV (talk) 13:33, 17 March 2025 (UTC)

Superlative adjectives are being treated as lemmas currently. It looks like the following needs to be done:

1. Definite participles need to be converted to participle forms and have their declension tables removed (in progress).
2. Superlative adjectives need to use just 'superlative adjective' as their headword POS (which makes them non-lemma forms); same for comparative adjectives.
3. The declension table for superlative adjectives needs to be definite-only.
4. Remove the {{infl of|...|adv|form}} and move it to the Etymology.
5. Participle cleanups (e.g. {{infl of|...|p|m}} -> {{infl of|...|m|p}}).

Benwing2 (talk) 23:25, 17 March 2025 (UTC)

BTW as for the |comp-of=, let me think about that. I added it as a convenience for the reader, so they don't have to click through to see the base form of a comparative. Certain other languages e.g. Gothic and Greek do the same; but not all. Benwing2 (talk) 23:27, 17 March 2025 (UTC)

Thank you very much for your work, Benwing 🙏 Anatolijs LTV (talk) 23:51, 18 March 2025 (UTC)

Indonesian headword templates

Hello Benwing, can I make a request with your bot? Specifically, removing |nolinkhead=, |tr=, and |pl=duplication from {{id-noun}}—it can automatically create plurals with full reduplication without any parameter, while the other parameters don't work or were included in error, and removing |pl= from {{id-adj}} and {{id-verb}}?—because there are no plural adjectives and verbs in Indonesian. They can be tracked here. Alfarizi M (talk) 01:01, 18 March 2025 (UTC)

@Alfarizi M I'm working on this. I downloaded all the Indonesian lemmas and I'm cleaning them up. They need a lot of work. Note that |nolinkhead= does have a legitimate function, which is to prevent linking in multiword terms where the individual components aren't Indonesian words. Examples are accent aigu, ad libitum, affaire de coeur, etc. But currently it seems to be misused so badly that it's probably best to remove it entirely and re-add it on a case by case basis. Benwing2 (talk) 05:49, 19 March 2025 (UTC)

@Alfarizi M I made those changes and a bunch of other formatting-related cleanups. I'm pushing the changes but it's gonna take several hours to complete. I notice you're also tracking uses of |pl=1 and |pl=-; are those supposed to be removed as well or is this tracking for some other reason? Also there are lots of pages that use {{head}} instead of one of the {{id-*}} templates, along with |head2=. You should make |head= a list parameter, and generally add parameter parsing using Module:parameters (maybe you did that in some of the changes I reverted ...). Benwing2 (talk) 07:31, 19 March 2025 (UTC)

Thank you very much! It's much better now than I expected. And yes, if it's possible, remove |pl=1 and |pl=- from {{id-noun}}—because first, |pl=1 doesn't work—and since it previously didnt show any "uncountable" label, |pl=- was often misused or misplaced even though the noun is actually countable. |- can be added if the noun is really uncountable. Alfarizi M (talk) 07:47, 19 March 2025 (UTC)

OK, I'll probably do that tomorrow as it will take awhile to push the current changes. (In 25 minutes it got through 1300 pages, and there are 24,400 of them, so it will take maybe 500 minutes = 9 hours to push all the changes, due to MediaWiki rate limiting when saving.) Benwing2 (talk) 07:52, 19 March 2025 (UTC)

BTW in the meantime please hold off on editing Indonesian pages until the bot has passed that section of the alphabet, as it's going alphabetically and just reached amplop; if you make a change to a page before the bot gets to it, the bot will skip it and I'll have to go back later and manually merge the change. Benwing2 (talk) 07:54, 19 March 2025 (UTC)

@Alfarizi M There were no occurrences of |pl=1 but I removed all the |pl=- occurrences. Benwing2 (talk) 22:43, 19 March 2025 (UTC)

Thank you very much, Benwing! Alfarizi M (talk) 01:25, 20 March 2025 (UTC)

Hey Benwing, can I make this request again? Specifically, removing these following parameters |pl=, |pl=word-word] (the plural form is inserted manually or has invalid value), |pl=+, and replacing |pl=0, |pl=-, |pl=a, |pl=* with bare |0, |-, |a, |* respectively in {{id-noun}}? Alfarizi M (talk) 18:48, 6 April 2025 (UTC)

@Alfarizi M Sure but I'm not quite sure what you mean in the first part of your request. What do |pl= and |pl=word-word] mean? Benwing2 (talk) 19:14, 6 April 2025 (UTC)

I mean remove those if |pl= has empty or invalid value, and |pl=word-word is when the plural form with full reduplication is inserted manually like {{id-noun|pl=serat-serat}} or {{id-noun|pl=sêrat-sêrat}}. Alfarizi M (talk) 19:41, 6 April 2025 (UTC)

@Alfarizi M There are no occurrences of |pl=- because I removed them in the previous round. Are you sure you want |pl=+ removed? (I guess this requests the default plural?) Also when removing |pl=word-word should it use the check against the pagename or the value of |head= or both? Also what does the default algorithm do with cases like asisten dosen? This has an explicit plural |pl=asisten-asisten dosen, should that be removed? Also there are a number of cases like dekan with plural |pl=dekan-dekan, [[para]] dekan or farmasis with plural |pl=para farmasis, farmasis-farmasis, what should happen there? Benwing2 (talk) 21:49, 6 April 2025 (UTC)

Yes, manually inserted plurals like |pl=asisten-asisten dosen should be removed too (the template can handle which part should be pluralized). I forgot to tell you that if the plural is like |pl=dekan, para dekan-dekan or |pl=para farmasis, para farmasis-farmasis which has para should be replaced with bare |a. Also, thank you for reminding me that you have removed |pl=-. Alfarizi M (talk) 21:55, 6 April 2025 (UTC)

What about cases like this:

{{head|id|noun|head=abdomên|head2=abdomen}}

Is this correct? Also there are cases like this which are certainly wrong:

{{head|id|noun|head2=aliasê}}

Should the |head2= be moved to |head= or should a value for |head= be inserted? Benwing2 (talk) 21:57, 6 April 2025 (UTC)

Similarly:

{{id-noun|head=detektor, détèktor|pl=detektor-detektor}}

Should that be converted to

{{id-noun|head=detektor|head2=détèktor}}

? Benwing2 (talk) 21:59, 6 April 2025 (UTC)

Also pemuda has

{{id-noun|pl=[[para#Indonesian|para]] pemuda, [[pemuda-pemudi]], [[pemuda-pemuda]]}}

Should this be converted to just

{{id-noun|a}}

? Benwing2 (talk) 22:02, 6 April 2025 (UTC)

• To make sure, I want |pl= to be deleted, not |pl2= or |pl3= because the other two can provide alternative plurals.
• If the double head parameter is used correctly like in the entry abdomen which the noun template has {{head|id|noun|head=abdomên|head2=abdomen}}, because the noun has two pronunciations.
• However, it's incorrect if only the second head exists as in {{head|id|noun|head2=abdomen}} or {{head|id|noun|head2=abdomên}}. To correct this, first if the word in the parameter isn't different with the entry title like {{head|id|noun|head2=abdomên}} then replace |head2= with |head=, or if the headword isn't different like {{head|id|noun|head2=abdomen}} then remove |head= completely (don't worry).
• If the |head= has two words like {{id-noun|head=detektor, détèktor}} then group them appropriately as {{id-noun|head=detektor|head2=détèktor}}
• Lastly, I'm not sure if the result would be correct for this (but I trust you), if the noun has more than two plurals or have a unique plural like in pemuda which the noun template has {{id-noun|pl=[[para#Indonesian|para]] pemuda, [[pemuda-pemudi]], [[pemuda-pemuda]]}}, then change it into something like this {{id-noun|a|pl2=pemuda-pemudi}} (if it has para) or {{id-noun|+|pl2=sayur-sayuran|pl3=sayur-mayur}} (if it doesn't have para). Alfarizi M (talk) 22:22, 6 April 2025 (UTC)

  @Alfarizi M I have done a bunch of cleanups to Indonesian lemmas and I'm in the process of pushing them all to Wiktionary (which will take several hours more). I also rewrote Module:id-headword to use Module:parameters and support arbitrary numbers of plurals and |nolinkhead= and such things, and I will push this after all the cleaned-up lemmas are pushed, since otherwise we'll get a bunch of module errors. Benwing2 (talk) 01:01, 7 April 2025 (UTC)

  Thank you very much, Benwing! The quality of Indonesian entries is getting better now than I expected. Alfarizi M (talk) 01:07, 7 April 2025 (UTC)

Pasadena categories

FYI, I consider your depopulation, deletion and creation-protection in error and have started an RFDO to have the categories restored. Purplebackpack89 12:33, 22 March 2025 (UTC)

Dude, why do you always have to be such an asshole? We don't normally have categories for cities under 1,000,000 people and Pasadena has < 200,000. On top of that, you created the categories manually so they don't fit into the {{auto cat}} system. Obviously you are from Pasadena and love your city, but it's only the 45th biggest in California per Wikipedia; we can't reasonably create categories for every random city in the US. Benwing2 (talk) 22:09, 22 March 2025 (UTC)

@BD2412 @DCDuring It is not appropriate for Benwing to call me the a-word... It's also not appropriate for him to make a claim about where I am from, that's OUTing. I would like his edit deleted and hidden. Purplebackpack89 15:31, 24 March 2025 (UTC)

I wholeheartedly agree that use of person-oriented pejoratives is bad practice. I would like it to be bad policy. I expect better of a bureaucrat. DCDuring (talk) 16:48, 24 March 2025 (UTC)

All right, I apologize about calling you names. As for where you are from, this is purely a guess based on your evident interest in this city. I don't see why this is "outing" and I will note that you state on your Wikipedia page that you're from California and have several references to the Southern California area. Benwing2 (talk) 19:56, 24 March 2025 (UTC)

Furthermore, discussion about whether or not a city need be a certain size to have a category is a perfectly legitimate point of discussion. So is the minimum number of entries needed to sustain a category. Purplebackpack89 15:17, 24 March 2025 (UTC)

Arabic verb generation bugs

Hi. I just wanted to inform you on several bugs in Module:ar-verb. No idea whether you will fix them or not, but just in case you decide it's worth it :)

1) verb VIII doesn't admit assimilation for ضاد. See page for verb اِضَّجَعَ, alternative form of اِضْطَجَعَ (found in Lane's lexicon)

Masdar, participles should change accordingly: مضّجع, اضّجاع. It's the only form VIII verb which assimilates ضاد (at least I haven't found any others) instead of just employing طاء.

2) form VII doesn't admit verbs with first root letter نون (which should assimilate). Try makeshift verb اِنَّصَرَ "be supported" with root ن ص ر (formally doesn't exist in other dictionaries but can be derived by an advanced speaker artificially, and btw, it's mentioned in preface to Lane's Arabic lexicon dictionary as an example: https://lexicon.quranic-research.net/T1.html ).

It will be rendered as اِنْصَرَّ in the conjugator, it assumes geminated root ص ر ر.

3) see verb استوأى of form X. It renders as استوءى in conjugator. The exception from hamza seat rules probably has something to with the case of very complicated root و ء ي with 2 week letters and hamza in the middle. Past tense and passive participle (renders as مستوءًى instead of مستوأًى) seem affected.

4) also about hamza seat. After short a. Not a mistake but an orthographic convention. See verbs like انفأى, اشتأى, ارتأى for examples. Imperatives/active participles render as اِرْتَأِ/مُرْتَأٍ, اِشْتَأِ/مُشْتَأٍ, اِنْفَأِ/مُنْفَأٍ. In modern orthography, we don't see hamza under alif except in word's beginning (like إرادة). Hence why we see cases like المبتدأِ.

But what we follow classical rules, hamza should be below alif. Maybe both variations can render at the same time? E.g. اِنْفَأِ/اِنْفَإِ and مُنْفَأِ/مُنْفَإِ for انفأى?

Also applies to case with geminated hamza like in form V verb ترأّى. Active participle applies renders as مُتَرَأٍّ (alt: مُتَرَإٍّ) and verbal noun as تَرَأٍّ (alt: تَرَإٍّ).

Thanks for reading this wall of text :) Fixmaster (talk) 19:58, 23 March 2025 (UTC)

@Fenakhay can you comment on these? These are getting into obscure corners of Arabic grammar that I'm not sufficiently familiar with. Benwing2 (talk) 20:02, 23 March 2025 (UTC)

1. We should add it. It is attestable but it should always be an alternative form of the nonassimilated form.

2. I thought you can override the inferred radicals with I/II/III. But it throws an error when you type: {{ar-verb|VII.I:ن}}:

3. Yeah, that should be fixed.

4. We could show both. I agree. — Fenakhay ^{(حيطي · مساهماتي)} 21:54, 23 March 2025 (UTC)

Small automation request

I would like the category Romanian terms suffixed with -i (infinitive) to be cleared and all its members transferred to the plain Romanian terms suffixed with -i. To this end, can you please remove any |id=infinitive parameters in the etymologies of the ~100 entries in question?

To explain why I want this measure: the ID granularity was a result of various homonymous suffixes being overzealously specified in useless etymologies (for instance, in entries for plural forms of nouns). With those occurrences cleared and the ‘infinitive’ suffix in question the only one of its homonyms left in etymology sections, there is no point in having the subcategory anymore. ―⁠K_(ə)tom (talk) 21:54, 25 March 2025 (UTC)

@Ktom Done. Benwing2 (talk) 23:15, 27 March 2025 (UTC)

ongoing RL issues

Hi Benwing. Re Special:Diff/84384797, I'm sorry to hear about your family emergency. Whatever it is, I hope it resolves itself without long-term issue. 0DF (talk) 09:02, 26 March 2025 (UTC)

Thanks! My wife needed surgery but she's doing fine now. Benwing2 (talk) 22:56, 27 March 2025 (UTC)

Oh, my! I hope it was not major surgery and that she is indeed, as you say, fine. 0DF (talk) 01:27, 28 March 2025 (UTC)

Wingerbot Issue

A couple weeks ago, you/your bot modified Taklamakan and replaced a link to the Uyghur language with one to Ukrainian. Since you used a bot for this, I figure I should let you know in case there's an underlying issue there with your code. https://en.wiktionary.org/w/index.php?title=Taklamakan&diff=prev&oldid=84218950

Eiszett (talk) 16:41, 26 March 2025 (UTC)

That's just me getting confused. All of these {{place}} and {{tcl}} link changes were done manually in a text editor; the only code involved is the script that pushed the results back to Wiktionary. Evidently I typed uk when I meant ug. Thanks for the heads up. Benwing2 (talk) 17:37, 26 March 2025 (UTC)

Another issue with module:ar-verb

Hi. Sorry for bothering again :) Just wanted to inform on a few more issues I found within module:ar-verb, whether you will fix or not, at least you should know as a maintainer of the template anyway :)

1) When dealing with form VIII verbs, the conjugator can only handle replacing تاء if your first root letter is either of 4: صاد, ضاد, طاء, ظاء.

I found 2 verbs with alternative forms, both have جيم as their first root letter:

اَجْدَمَعَ (participle مجدمع and verbal noun مجدمع), alternative form of اجتمع with root ج م ع

Mentioned in Lisan al-Arab: جمع : جَمَعَ الشَّيْءَ عَنْ تَفْرِقَةٍ يَجْمَعُهُ جَمْعًا وَجَمَّعَهُ وَأَجْمَعَهُ فَاجْتَمَعَ وَاجْدَمَعَ and also in Lane's lexicon for root ج م ع

اَجْدَزَّ (participle مجدزّ and verbal noun اجدزاز), alternative form of اجتزّ with root ج ز ز.

Μentioned in several books, e.g. in Abu Zakaria al-Farra's explanation of Quranic meanings:

فقلت لصاحبى لا تحبسانا * بنزع أصوله، واجتزَّ شيحا قال: ويروى: واجدزّ يريد: واجتز، قال: وأنشدنى أبو ثروان:

These 2 are probably the only ones (no idea, maybe I just couldn't find others), I created entries for them, the conjugator doesn't handle them, needs the ability to replace تـ with دـ if our root begins with ج.

2) Verb أْسْطَاعَ (pr. tense يُسْطِيعُ) is mentioned in Lane's lexicon, al-Razi's dictionary, etc.

It is the augmented alternative form of أطاع, form IV with extra سين before the first root letter, participles مسطيع/مسطاع, verbal noun إسطاعة. The conjugator can't render it.

3) in addition to اِسْطَاعَ, there's another short form of استطاع found in Lane's and elsewhere: اَسْتَاعَ/يَسْتِيعُ (participles مستيع مستاع, verbal noun: استاعة).

Instead of dropping تاء from form X prefix استـ like اسطاع does, استاع removes the first root letter طاء! Entry created. Thankfully, the conjugator handles it with with "reduced" argument.

Actually, the last one isn't about module:ar-verb code itself, it's about Wiktionary category "Arabic form-X reduced verbs":

When you open its page, the description says it's the category for the verbs dropping سـ from their prefix. The description should be edited to accomodate for the verbs where their dropped letter isn't from the prefix, but from their root itself.

I don't know how to edit this category description (it's created by autocat template) :)

Thanks for reading my whining. Sorry again for bothering. Fixmaster (talk) 19:53, 26 March 2025 (UTC)

@Fixmaster Thanks for the bug reports. Which ones (if any) are pressing? It may be awhile before I'm able to get to all of these. Benwing2 (talk) 20:08, 26 March 2025 (UTC)

title_plain

{{quote-journal}} needs an equivalent to the title_plain parameter that applies to the journal instead of the article. ―⁠K_(ə)tom (talk) 21:41, 28 March 2025 (UTC)

Apostes

what does this word mean? Your bot worked on it a few times. I was wondering if you yourself knew? K2520025 (talk) 17:43, 30 March 2025 (UTC)

"templatize langname categories for langcode=nl using {{cln}}"

While this kind of error is kind of funny, it could cause problems for bot work later on. I think I got them all, but it wouldn't hurt to check. I wonder if there are any other cases where language names that start with other language names have resulted in similar errors. If you use a query to look for these, you'd have to allow for cases like "German Low German" where both parts of the name are language names in their own right- Category:German Low German cardinal numbers and Category:Low German cardinal numbers are both valid category names. Chuck Entz (talk) 21:31, 31 March 2025 (UTC)

Perhaps Category:Egyptian Arabic cardinal numbers and Category:Arabic cardinal numbers would be a better example, since both categories exist. Chuck Entz (talk) 21:42, 31 March 2025 (UTC)

Oops. How did you find the instances of these? Benwing2 (talk) 22:02, 31 March 2025 (UTC)

I see, it must be this list: Wiktionary:Todo/Lists/Template language code does not match header (sorted by language). Benwing2 (talk) 22:10, 31 March 2025 (UTC)

There are several instances of {{cln|fr|French}} and similarly {{cln|id|Malay}}, {{cln|hsn|Chinese|language}} and {{cln|en|English}}, but no others of the sort you mention above (at least per the Mar 20 dump). @-sche and I have talked about renaming Category:French French to be Category:France French and similarly for Category:English English. Category:Indonesian Malay seems a contradiction in terms. Benwing2 (talk) 22:16, 31 March 2025 (UTC)

There seems to be a practice among some Malay or Indonesian editors (not sure if past or ongoing) to treat the two languages as varieties of each other, despite having different L2s. Do you know the history of our treatment of these lects? I know little, but from a cursory look it may be a bit like the Norwegian situation. This, that and the other (talk) 10:52, 1 April 2025 (UTC)

@This, that and the other: Indonesian was a new standard created out of Malay at the time of independance. We had a thorough discussion, and the editors in both languages came to a consensus that it was better to treat them as separate. My impression was that it was a somewhat arbitrary decision: we couldn't have it both ways, so they chose one. There's a separate issue of some Indonesian editors being rather sloppy with language codes in general- not just Indonesian and Malay. Chuck Entz (talk) 12:06, 1 April 2025 (UTC)

That said, I've been looking, but have yet to find any record of the discussion. Strange. Chuck Entz (talk) 13:03, 1 April 2025 (UTC)

There was Wiktionary:Votes/2012-12/Unified Malay, but that was more a proposal to merge that didn't get consensus (3-4) than a clear consensus for separation. (Wiktionary:Votes/2016-10/Unified Malay Revote never started, but suggests there were probably other discussions somewhere.) BTW, for France French, the option of "Metropolitan French" might be worth considering (cf. Peninsular Spanish). - -sche (discuss) 19:39, 1 April 2025 (UTC)

@-sche Is "Metropolitan" normally used to refer to the French spoken in France? It suggests to me specifically Parisian French rather than French of France as a whole, whereas "Peninsular Spanish" makes sense as referring to Spanish of the Iberian Peninsula. Benwing2 (talk) 20:09, 5 April 2025 (UTC)

It refers to the French spoken in the part of France that's in Europe, "metropolitan France". Indeed, while I had only been suggesting it as a more euphonious name, I now realize it differs from "France French" that way: "France French" would be a higher-level category including both European France's French and the French of non-European parts of France, like Guianese, Martinican or Reunionese French (like British English including Bermudian English). - -sche (discuss) 09:05, 6 April 2025 (UTC)

Three Categories

Hey Benwing2, I see you are doing some geography stuff. I wonder if you would be willing to look at the possibility of creating three new categories. The categories are collected in nascent form on the entries in the "See also" sections at Haishenwai, Taihoku, and Yalu. I do not have the perspective to understand if any of them might fit into the larger Wiktionary scheme somewhere as actual categories. They might need to be "rebranded" or similar. Just a project for you if you are interested. --Geographyinitiative (talk) 12:40, 5 April 2025 (UTC)

Hi. I think categories of this nature would have to be manually added as there's no functionality in {{place}} currently for specifying the ethnic origin of a given term. Benwing2 (talk) 20:08, 5 April 2025 (UTC)

Okay, thanks for your feedback! I think I will wait and see if the entries can get a little stronger (more entries, more citations for entries, etc.) and then I may try something; if you want to try something, feel free. --Geographyinitiative (talk) 20:49, 5 April 2025 (UTC)

Hey, thanks again for working on these geography things. I see that you are adding the q numbers to some articles- I tried to use Template:tcl to add id=Q1195273 on Huxi ("Lua error in Module:transclude at line 386: Couldn't find the template {{senseid|en|Q1195273}} within entry Huxi. "), but it's showing an error when I try to add it. I'm going to try to learn how that works, please show me if you want to. Also, it looks like this category has a problem: Category:en:Provinces of China, I'm not sure what to do to correct this. --Geographyinitiative (talk) 11:56, 26 April 2025 (UTC)

@Geographyinitiative Hi! {{tcl}} is meant to be used on a foreign translation of a geographical term to copy the definition from the corresponding English term. You put {{sid|en|Q1195273}} on the line in Huxi that defines it (at the beginning of the line, just before {{place}} or any preceding labels), and then e.g. on the Chinese equivalent of Huxi you'd write {{tcl|zh|Huxi|id=Q1195273}} and it will copy the definition. As for provinces of China, I consolidated provinces and autonomous regions into Category:en:Provinces and autonomous regions of China (and did similarly for Canada, Australia, India and Pakistan, which each have two main types of top-level administrative divisions). If you define something as a province or autonomous region of China it automatically goes into the combined category; the remaining ones in Category:en:Provinces of China are manually categorizing and need cleanup. Benwing2 (talk) 19:23, 26 April 2025 (UTC)

BTW many of the entries in Category:en:Provinces of China are not provinces but former provinces, which are properly categorized differently (under Category:en:Former political divisions, usually). Benwing2 (talk) 19:30, 26 April 2025 (UTC)

@Geographyinitiative Please make sure that former counties, prefectures and the like are preceded by the word 'former'. For example, under 化州 there are three definitions, two of which refer to former counties and prefectures but which aren't correctly tagged using 'former', which means they end up categorized as if they're current. By using 'former' we segregate the former political divisions; if there are enough of them, we can create a category specifically for them (cf. Category:Former municipalities of the Netherlands). Benwing2 (talk) 23:22, 27 April 2025 (UTC)

Esperanto 'interfixes'

Theknightwho has changed the POS of all Esperanto suffix roots, such as -ism-, to 'interfix'. Discussion is pretty much in consensus that they're simply suffixes [e.g. that Esperanto -ism-o is equivalent to Italian -ism-o], and even Theknightwho admits that they're not interfixes. Would it be okay for me to change them back? kwami (talk) 20:19, 7 April 2025 (UTC)

@Theknightwho any comment? Benwing2 (talk) 20:32, 7 April 2025 (UTC)

@Benwing2 Kwami is being dishonest by saying that I "admitted that they're not interfixes", as I said they were infixes. Theknightwho (talk) 22:16, 7 April 2025 (UTC)

They can't be infixes, which are inserted in the middle of morphemes rather than between morphemes. Benwing2 (talk) 22:24, 7 April 2025 (UTC)

When I said 'There are no interfixes in Eo,' you said, and I quote, 'Right, but that makes them infixes instead.' Summarizing that as you admitting they are not interfixes is both honest and accurate. You then continued to change 'suffix' to admittedly incorrect 'interfix' in multiple articles. kwami (talk) 22:31, 7 April 2025 (UTC)

I don't know what an interfix is, so I looked it up here. No, these are just multiple suffixes. There's a suffix -ism- then there's the noun suffix -o. That's how Esperanto is built up: multiple suffixes. The -o in hundo isn't an integral part of the word, it's a suffix indicating the word is a noun. --Hiztegilari (talk) 22:39, 7 April 2025 (UTC)

To be fair, in some traditions a prefix or suffix that cannot occur at the edge of a word -- as in Bantu languages -- is called an 'infix', and sometimes even the same affix may be variably called either a prefix/suffix or an infix depending on where it occurs in any particular word or inflection. Wiktionary follows this tradition for some Bantu languages, presumably because that's the terminology of our sources, but this usage is deprecated for general linguistics and is it not used in descriptions of Esperanto. The only distinction I'm aware of in Eo is between 'true' pre/suffixes and 'pseudo' pre/sufffixes, depending on whether the affix can be treated as a root that follows normal Esperanto word-combining rules [where AB = B of A], or if it is fixed in position despite violating the order expected from semantics. kwami (talk) 23:44, 7 April 2025 (UTC)

I'm a bit confused why most of these exist at all. It seems that -ism- only occurs with nouns so why isn't it just -ismo? In other languages, we don't normally break down suffixes that contain a part-of-speech suffix within them into separate affix-like entities. I would only expect this to occur with affixes like -âc- that can occur with different parts of speech. Benwing2 (talk) 23:56, 7 April 2025 (UTC)

Some suffixes don't have a single inherent part of speech. I think for those that do, it makes sense to include the vowel. That seems to be what Wikipedia and Wikiversity do. But in the case of -ism-, it looks like it's essentially a soft redirect, so I guess having it doesn't really do any harm. By the way, I agree that even affixes like -aĉ- should be categorized as suffixes, not as interfixes or infixes.--Urszag (talk) 00:37, 8 April 2025 (UTC)

Agreed, it's a little weird, but Esperanto grammars often give suffixes in their root form, so that would be a likely way for people to look them up, esp if they don't know what the inherent POS is. kwami (talk) 04:46, 8 April 2025 (UTC)

@Kwamikagami Simply saying I "admitted that they're not interfixes" is deeply dishonest, because it completely misrepresents my view. This is an absolutely classic example of the kinds of manipulative tricks you play on Wikipedia all the time, and that kind of thing will not be tolerated here. Theknightwho (talk) 03:35, 8 April 2025 (UTC)

How can it misrepresent your view when it's what you yourself said? If you didn't admit they're not interfixes when you agreed that they're not interfixes, and then repeated yourself here, then I have no idea what you intended to say. It's hardly dishonest to take you at your word. You've accused me of being dishonest before when I've done that, but all I have to go on to understand your POV is what you say here. kwami (talk) 04:43, 8 April 2025 (UTC)

Being able to deduce something from what someone said is not the same as their "admitting" it. For one thing, there's the implication that they were saying something that they would rather not have had to say. Then there's the matter of putting words in someone's mouth that they didn't actually say. Using your methodology, I could claim that you have admitted that you're not speaking Swahili, and that you're not illiterate. Chuck Entz (talk) 06:21, 8 April 2025 (UTC)

If I were claiming to be speaking Swahili, and you said I wasn't, and after arguing a bit I agreed with you, then indeed I would be admitting that I wasn't speaking Swahili, and you wouldn't be putting words in my mouth if you noted that fact. Theknightwho claimed these were interfixes, I said they weren't, and after some back-and-forth they agreed. Substitute 'agreed' for 'admitted' above if you like; the point is that even after they agreed that these were not interfixes, they continued to change the subheadings to 'interfix', and I was asking if it would be okay for me to revert. kwami (talk) 07:03, 8 April 2025 (UTC)

@Kwamikagami You completely misrepresented what my view was, and now you're doubling-down, because apparently you know my own opinion better than I do. Are you this arrogant in real life, too? Theknightwho (talk) 15:05, 8 April 2025 (UTC)

I apologize for offending you, as I apparently have, but what was your view? I really don't understand. I said your edits were wrong, and at first you said they were right, but then you agreed they were wrong, and yet you continued to make the same edits after that.

Can you really speak to my motivation for a natural choice of wording? I wouldn't object if our roles were reversed and you characterized my agreement with you as an admission that you were right, so I truly don't understand what the issue is.

I suppose I should have immediately asked what you thought was dishonest, or to summarize what your view was when you said I was dishonest, but frankly it's annoying to be accused of dishonesty when I'm just trying to reverse an error, one that you acknowledged on your talk page, and again in this thread, is an error. kwami (talk) 17:03, 8 April 2025 (UTC)

Sometimes your line of argumentation is sort of like "2+2=did you know he one the Nobel Peace Prize?" Chuck Entz (talk) 04:45, 9 April 2025 (UTC)

I really don't see how using the plain reading of a text can be controversial, let alone 'deeply dishonest'. It would be different if Theknightwho said they were being sarcastic and that I should've understood they didn't really mean it -- it's easy to miss sarcasm in text, -- but instead they repeated their assertion here and were clearly not being ironic when they did so.

On their talk page, when I said, 'There are no interfixes in Eo,'

they answered, 'Right, but that makes them infixes instead.'

I understand 'right' to be a mark of agreement. I don't think that is controversial.

The plain reading of that response is that [a] they agree that these are not interfixes, and [b] they now assert that they're infixes.

But if they agreed that these were not interfixes, why did they continue to change the POS to 'interfix' in dozens more articles? I would've understood if they'd starting changing the POS to 'infix', but not to something they conceded was wrong. That's why I came here. kwami (talk) 14:49, 9 April 2025 (UTC)

Superlative adjectives

Yes, I'd like it if {{id-adj}} could show adjective inflections, like in User:Alfarizi M/Sandbox. Is this okay? Maybe there are suggestions you would like to share. Alfarizi M (talk) 06:54, 8 April 2025 (UTC)

Hi. I'm a bit confused by your suggested template arguments. Here's what I'd suggest:

{{id-adj|+}} or {{id-adj|peri}} (where peri means "periphrastic") generates:

terpanah (comparative lebih terpanah, superlative paling terpanah)

{{id-adj|+|eq=+}} or {{id-adj|peri|eq=se}} generates:

kaya (comparative lebih kaya, superlative paling kaya, equative sekaya)

{{id-adj|+|+|sup2=ter|eq=+}} or {{id-adj|peri|peri|sup2=ter|eq=se}} generates:

cepat (comparative lebih cepat, superlative paling cepat or tercepat, equative secepat)

The logic here is that:

1. |2= specifies the comparative, and additional comparatives are specified using |comp2=, |comp3=, etc.
2. |3= specifies the superlative, and additional superlatives are specified using |sup2=, |sup3=, etc.
3. |eq= specifies the equative, and additional equatives are specified using |eq2=, |eq3=, etc.
4. A value of + or peri for the comparative generates a periphrastic comparative using lebih before the adjective.
5. A value of + or peri for the superlative generates a periphrastic superlative using paling before the adjective.
6. A value of ter for the superlative generates a superlative using the prefix ter- before the adjective.
7. A value of + or se for the equative generates an equative using the prefix se- before the adjective.
8. If the comparative is given as + or peri, the superlative by default uses the same value and is also generated. You can override this by specifying specific superlative values, or using - to indicate that there is no superlative.

This is just based on your examples. If there are other ways of generating the comparative, superlative or equative, let me know. Benwing2 (talk) 07:08, 8 April 2025 (UTC)

I see. I think this is enough for handling Indonesian adjectives. Are you going to code it? I'm curious. Alfarizi M (talk) 07:32, 8 April 2025 (UTC)

Yeah I can code it. Benwing2 (talk) 07:34, 8 April 2025 (UTC)

Icelandic cleanup

Hi.

Would you happen to have time to take a look at User talk:Jöttur? We have a user claiming to be a native speaker of Icelandic who has been adding blatantly incorrect entries (some under ungrammatical forms), shockingly ungrammatical or unrepresentative usage examples, incorrect IPA, and uploading audio samples despite barely knowing the language.

Even when this user's entries are somewhat correct, they still manage to include errors, e.g. slightly incorrect usage notes.

I cannot think of any way to fix this apart from deleting all of this user's contributions to Icelandic. 130.208.182.103 10:02, 9 April 2025 (UTC)

I am fine with removing all this user's Icelandic content. Unfortunately it's all mixed in with other content, any ideas whether that content (e.g. in Yiddish) is equally bad? Can you help me compile a list of all the entries the user changed? I can probably write a bot script to undo them all. Benwing2 (talk) 18:17, 9 April 2025 (UTC)

Thank you, I see you've already deleted those thousands of entries. Certainly an interesting case. 130.208.182.103 12:55, 11 April 2025 (UTC)

There are still a few hundred edits where he changed existing entries that aren't all done. I'm in the process of writing a script to help with this. Benwing2 (talk) 16:57, 11 April 2025 (UTC)

kalla ekki allt ömmu sína

Hi, i thing your bot made a mistake with this edit[1] since it changed the correct formatting for a sense:id Sérgio Santos (talk) 13:15, 11 April 2025 (UTC)

Thanks, I'll fix the underlying script to account for this and see if there are any other such instances. Benwing2 (talk) 16:58, 11 April 2025 (UTC)

Replacement of reference templates (April 2025)

Hi, please carry out the replacement of the following reference templates when you are free:

• {{R:Canadian Oxford 2004}} → {{R:Canadian Oxford}} (template has been expanded to include multiple editions).
• {{RQ:Herbert Travaile}} → {{RQ:T. Herbert Travaile}} (to distinguish from works by George Herbert).
• {{R:Websters}} and {{R:Webster NCD 1967}} → {{R:Webster's Collegiate Dictionary|year=1967}}.
• {{R:Webster NCD 1974}} → {{R:Webster's Collegiate Dictionary|year=1974}}.
• {{RQ:Wood New Englands Prospect}} → {{RQ:W. Wood New Englands Prospect}} (to distinguish from works by Anthony Wood).

Thank you. — Sgconlaw (talk) 20:10, 13 April 2025 (UTC)

Hope you're able to get to this. — Sgconlaw (talk) 23:03, 21 May 2025 (UTC)

@Sgconlaw Should be done. Benwing2 (talk) 23:58, 21 May 2025 (UTC)

Thanks! — Sgconlaw (talk) 14:53, 22 May 2025 (UTC)

bad audios

Hey! I see you fiddled with {{audio}}. Can you add a bit/tell me where to catch bad audios like at casino? My suggestion is Category:English entries with shitty audios (probably Wonderfool) Vilipender (talk) 10:26, 14 April 2025 (UTC)

Use |bad=probably Wonderfool or similar. Benwing2 (talk) 22:12, 14 April 2025 (UTC)

/w̝/

This sound in es-pr should only appear in narrow transcription, not in broad, also, why hua results in /w̝a/, while wa is /wa/?. Rodrigo5260 (talk) 02:26, 19 April 2025 (UTC)

Kiowa letters

[B [b, [D [d and [G [g are the common form of the Kiowa barred letters, found in informal publications like the tribal newsletter. Would it be okay if I created these specifically for Kiowa? I don't know of any translingual use. (The barred letters are used in dictionaries etc but there's been some resistance to switching over daily usage) kwami (talk) 21:40, 24 April 2025 (UTC)

Also, how would I fix the 'see also' section at ∶, where the 1st link is an unsupported title?

I suspect that the square brackets might interact in strange ways with wiki markup and with parsing in some templates and modules. It may take some tinkering to get things set up right. Chuck Entz (talk) 04:20, 25 April 2025 (UTC)

Okay, I've created articles for [G and [g, and they seem to work. The letters display correctly at the top of the page and I've checked the links work. I've gone ahead and created the rest - [B, [b, [D, [d, [G, [g. kwami (talk) 23:49, 1 May 2025 (UTC)

WingerBot mismatched angle brackets

Look at the recent edit history of Chittagong. WingerBot made changes to two lines adding double angle brackets, but the second of them went slightly wrong at the end. The conditions seem to be identical. 2A0A:EF40:110E:E201:844:432C:1D95:2FA5 15:29, 1 May 2025 (UTC)

This was done manually in a text file (hence the "manually assisted" note at the end), and is just a case of me mistyping the output. Benwing2 (talk) 19:22, 1 May 2025 (UTC)

Obscure Arabic characters

There are a number of Arabic Unicode characters whose entry consists only of a translingual section with a 'definition needed' tag. I was thinking of requesting deletion of the articles, but found usage for some of them. These are mostly obscure characters used for a single language, so I replaced the empty translingual section with one for that language. If you think the translingual section should be retained, I'll restore it.

Also, the 3 tone diacritics of Rohingya are covered by 6 Unicode characters, despite the proposer saying ideally there would only be 3. They even look the same in my default font, so these seemed an appropriate use of redirects. kwami (talk) 23:07, 1 May 2025 (UTC)

@Kwamikagami I still think there needs to be a Beer parlour discussion to iron out when we use "Translingual" and when we use a specific-language entry, and I'm not the arbiter of this. Can you make such a posting? Benwing2 (talk) 23:09, 1 May 2025 (UTC)

Sure. I'm writing it now. kwami (talk) 23:22, 1 May 2025 (UTC)

A request for a bot task (and maybe general help)

I am updating the {{pt-pre-reform}} template to include a small little-known European reform from 1920. The code I’m going to paste into the template, later is at Template:User:Polomo47/sandbox and User:Polomo47/sandbox/internal. My request is two-fold:

• The way I set it up, we would need a bot to change all pages with the sequences on the left to the ones on the right:
  • pt=1 > pt=11
  • pt=2 > pt=45
  • pt=3 > pt=73
  • pt=4 > pt=90
  • br=1 > br=43
  • br=2 > br=71
  • br=3 > br=90

This is because the new "=1920" reform would have to be "=2", so I'd rather get rid of the numbers entirely than refactor the numbering scheme. Would need a bot either way.

• If you're willing to, I also invite you to try and improve the template’s code in whatever ways you can. In this diff, I was trying to make it declare errors, but then it wasn’t working and I gave up for now.

Always grateful, and hoping I make myself clear — I've failed terribly at it quite a few times... By the way, did you see my ping at Module:pt-verb? I don’t think I treat you as highly as I should, Mr. ’Wing, but I hope to improve!

Polomo47 (talk) 01:06, 2 May 2025 (UTC)

@Polomo47 Hi, you say "change all pages with the sequences on the right to the ones on the left" do you mean the other way around? I can get to this in the next day or two. As for your ping on Module:pt-verb, yes I saw it but I wan't sure what you were pinging me about ... did you just want me to check out whether it's correct (it looked fine to me)? Benwing2 (talk) 01:13, 2 May 2025 (UTC)

Heh, yes, that’s what I meant. It’s easier to mix up "right" and "left" when typing, it seems (not the first time I did it). Thanks!

As for why, I pinged you in Module:pt-verb, it was because, in spite of my efforts, {{pt-conj}} wasn't displaying correctly at coar. I can only assume the first declaration (for all verbs in -oar) is overriding the second, narrower, one. I was indeed not clear. Polomo47 (talk) 01:15, 2 May 2025 (UTC)

@Polomo47 I see. Yes, the specs are processed in order from top to bottom, and so the "oar" spec is processed first and overrides the "^coar" spec. You should move the latter up before the former. Benwing2 (talk) 01:20, 2 May 2025 (UTC)

Oh, alright. I think I considered that — and it really is pretty obvious — but for some reason decided against it. Heh. On it. Polomo47 (talk) 01:23, 2 May 2025 (UTC)

@Polomo47 Should be done. There were relatively few pages needing fixing; only about 40 or so. I also fixed a few cases where there was a clear mistake, e.g. apnéico said it was the pre-reform form of desoxirribonucleico. As for the template, I would support an |id= parameter instead of hardcoding things like {{pt-pre-reform|colar#Portuguese: necklace|pt=11|br=43}}. If you pass the |id= value directly to {{l}}, it will automatically generate the correct anchor, which protects you against a possible future format change in the anchors. Benwing2 (talk) 06:36, 3 May 2025 (UTC)

Thank you so much, mister! Right, about IDs, that was one thing I wanted to implement but that slipped my mind, just now. I also remembered this old |obs=1 parameter that I need to handle. I’ll implement IDs in the template code, and then ask you to run another bot task, alright?

Though, I could try running AWB myself, like I already plan to do for something else today... Dunno why I didn't think of that last time. Polomo47 (talk) 16:28, 3 May 2025 (UTC)

Showing Respect

Hello,

Given my young age and lower status, would it be polite to refer to you as Mr. Wing?

Thank you Flame, not lame (Don't talk to me.) 21:18, 9 May 2025 (UTC)

@Flame, not lame No need! Just use my first name. Benwing2 (talk) 03:16, 10 May 2025 (UTC)

Understood. Flame, not lame (Don't talk to me.) 12:21, 10 May 2025 (UTC)

A very strange new editor

Jh38 (talk • contribs • global account info • deleted contribs • nuke • abuse filter log • page moves • block • block log • active blocks): adding dative and accusative "inflection of" senses to Spanish entries, creating a Bengali entry for কুকি (kuki, “cookie”) adding entries in Arabic, Hungarian, and other Wiktionaries. Either we're dealing with a randomly impulsive and clueless human who knows lots of random stuff, or a LLM. What do you think? Chuck Entz (talk) 06:21, 14 May 2025 (UTC)

@Chuck Entz The user seems to be a person who knows some Spanish and may well be using an LLM to generate Spanish entries (which they then correct by hand) and translate random terms into random languages; or just entering the terms into Google Translate. IMO definitely someone impulsive and clueless, of the sort who thinks they know a lot more than they do. Benwing2 (talk) 06:40, 14 May 2025 (UTC)

The possibility of adding a footnote about the second person singular imperative form of -er verbs before en and y (e.g. vas-y !, achètes-en pour moi !) to the French conjugation table

This form with a euphonic 's' is mentioned on the conjugation table description for aller but I don't know if it is on any others. I think a footnote should be added to all -er verb tables (edit)=to clarify this point. I am still undecided if we should add this usage as a definition to the non-lemma forms through the inflection of template on all applicable entries, like on the French wiktionary. Thank you. Pvanp7 (talk) 06:19, 15 May 2025 (UTC)

@Pvanp7: @Nicodene, PUC I'll let you two respond to this. I suspect achètes-en pour moi is more informal (or even nonstandard) compared with vas-y. I presume when you say "add this usage as a definition to the non-lemma forms" you mean adding the 2nd sing imperative as an additional definition for achètes and such. My instinct is not to do that but I'll again defer to the French experts. Benwing2 (talk) 06:24, 15 May 2025 (UTC)

@Benwing2: Yes, that was my intended meaning. :) Pvanp7 (talk) 06:34, 15 May 2025 (UTC)

I believe a euphonic [z] is obligatory given an immediately following y or en, though indeed informal to the extent that using a second-person singular verb in general is.

Perhaps we could put an auto-generated note on the relevant conjugation tables. Nicodene (talk) 07:40, 15 May 2025 (UTC)

Yes, that is exactly what I was hoping for. On a related note the verb laisser has an obligatory 's' added, even when the adverbial pronoun in question is agreeing with a different verb. This will require its own note. See this page on the OQLF for more information. Pvanp7 (talk) 08:09, 15 May 2025 (UTC)

Ah, I hadn't thought about what happens if the y/en happens to be followed by an infinitive. That's worth mentioning, yes. Nicodene (talk) 08:32, 15 May 2025 (UTC)

Editnotices

Hope you can explain this to me. How does one create an editnotice like the one displayed at "Wiktionary:Word of the day/May 22"? I see that an editnotice can be created in the form "MediaWiki:Editnotice-1", but how does one then associate it with a particular type of page? Also, how does the numbering system for our editnotices work? Thanks. — Sgconlaw (talk) 23:01, 21 May 2025 (UTC)

@Sgconlaw Apologies, I will get to your template rename request right now. As for the notice displayed at Wiktionary:Word of the day/May 22, if you're referring to the informational notice that begins with From 2020, Word of the Day pages are in the format "Wiktionary:Word of the day/[year]/[month] [day]", that is not an edit notice but is displayed due to the template {{WOTD}}, which calls the {{notice}} template to display it. As for the edit notices themselves, I'm not quite sure how they work. I had assumed the mapping between pages and edit notices is built into MediaWiki itself or into some Wiktionary configuration file that you would need a Phabricator ticket to change, but maybe it's under our control. I think @Erutuon might know more about this. It also might depend on which page you're trying to display an edit notice for. Benwing2 (talk) 23:15, 21 May 2025 (UTC)

@Sgconlaw: Edit notices start with MediaWiki:Editnotice- and can be seen at Special:PrefixIndex/MediaWiki:Editnotice-. What type of page are you thinking of? At least you can make an editnotice display on all pages with a certain base page name: for instance, MediaWiki:Editnotice-828-languages, which displays on Module:languages/data/2, Module:languages/data/3/a, etc. — Eru·tuon 02:42, 22 May 2025 (UTC)

@Sgconlaw: The part after the "Editnotice-" part is the namespace number. See Wiktionary:Namespace for a table of all the namespaces numbers. Namespace 1 is the Talk namespace, so Mediawiki:Editnotice-1 displays when you edit a page in the Talk namespace. From Erutuon's examples, I would assume that the part after the next dash is a prefix that matches all pages in the specified namespace that start with that sequence. I'm guessing that "MediaWiki:Editnotice-3-Sgconlaw" would display when your talk page or any of its subpages were edited. Chuck Entz (talk) 03:05, 22 May 2025 (UTC)

@Benwing2, Chuck Entz, Erutuon: ah, I see. Thanks for all your explanations. I was thinking of adding an “editnotice” for FWOTD fallbacks, but it seems I shouldn’t create an actual editnotice for this but just edit {{FWOTD}}. Nonetheless it’s good to have some insight into how editnotices work. They’ve always been a bit of a mystery to me. — Sgconlaw (talk) 06:34, 22 May 2025 (UTC)

Problems with Template:tcl conversions

If you look through the entries in Category:English terms in nonstandard scripts, you'll find quite a few that are there because last month WingerBot converted synonyms, main forms, etc. to tcl parameters without specifying their language, so they defaulted to English (see diff for an example) I'm sure this is only the tip of the iceberg, since terms in Latin-script languages with the same problem wouldn't show up there. I believe I've fixed other problems with those conversions, but I didn't think to keep notes so I don't have anything useful to say about that. Chuck Entz (talk) 20:44, 23 May 2025 (UTC)

Hi Chuck. Recently I cleaned up the logic for deciding when a given extra-info form should be interpreted as English vs the language of the tcl call. In the diff you showed, that term is supposed to be being interpreted as Georgian. I may have messed up the logic when cleaning it up. The tcl calls themselves are correct. Let me take a look when I’m on my laptop. Benwing2 (talk) 21:09, 23 May 2025 (UTC)

@Chuck Entz Fixed. Benwing2 (talk) 23:21, 23 May 2025 (UTC)

After purging, the category went from 249 to 158. As for other problems, there's always Wiktionary:Todo/Lists/Template language code does not match header#{{tcl}}, but most of those are just people with no clue what the first parameter is for. There are things like diff, but they seem to be just data entry errors in your batch files- nothing systemic. Chuck Entz (talk) 00:38, 24 May 2025 (UTC)

@Chuck Entz Yeah, sometimes for the language code I type in QQQ and then run a script that corrects the QQQ to the right code based on what language the entry is for, but sometimes I enter the language code manually, and in the latter case, occasionally I make mistakes. I should get in the habit of using the QQQ method more often; it's an extra step but it pretty much ensures these errors won't occur. Benwing2 (talk) 00:44, 24 May 2025 (UTC)

WingerBot should not "add missing final period (full stop)" when term contains one

I reverted this edit. J3133 (talk) 06:21, 25 May 2025 (UTC)

@J3133 OK thanks, I'll fix up that script to account for this. Benwing2 (talk) 06:24, 25 May 2025 (UTC)

obs-std display

In 賊, {{lb|zh|obs-std} shows obsolete on its own in Standard Chinese, I feel it's a little long, can we make it more concise. Kethyga (talk) 22:22, 26 May 2025 (UTC)

@Kethyga We had a long discussion about this. You need to (a) propose an alternative, (b) get consensus among the various Chinese editors to use that alternative. Benwing2 (talk) 22:34, 26 May 2025 (UTC)

some bullets to columns

Hi, since you worked on column templates (also pinging @This, that and the other) I was wondering if you'd be interested in converting certain bulleted lists to column templates. The target I have in mind is Derived terms and Related terms sections with 6 or more lines in a row beginning with asterisks. I've seen others express that 5 is their preferred max height for bulleted lists, and I agree. I'd exclude See also's because they tend to be less uniform in their contents. Ultimateria (talk) 22:43, 29 May 2025 (UTC)

@Ultimateria I checked and there are 13,494 ==Derived terms==/==Related terms== sections in 13,206 pages that begin with 6 or more bulleted lines. My preceding script converted bulleted lines between {{topN}} and {{bottom}}, so it shouldn't be very hard to adapt for this purpose. The only tricky thing is figuring out whether auto-sorting should be disabled. When I converted the previous stuff I went over all the conversions manually to see which ones needed to have sorting turned off -- a tedious and mind-numbing exercise. I'm thinking I should be able to check to see if the entries are sorted or almost-sorted, and only go manually over the ones that aren't. Benwing2 (talk) 08:07, 2 June 2025 (UTC)

My imagination is failing me; when does sorting need to be turned off in these sections? The only ordered lists coming to mind (planets, taxa) would be See also's or Coordinate terms. Ultimateria (talk) 20:41, 2 June 2025 (UTC)

@Ultimateria So for example the word "day" might have derived terms "Monday, Tuesday, ..." that should be ordered. Looking through the derived term examples I see X that has Roman numerals as derived terms:

* [[IX]]
* [[XI]]
* [[XII]]
* [[XX]]
* [[XXX]]
* [[XXXX]]
* [[XL]]
* [[LX]]
* [[LXX]]
* [[LXXX]]
* [[LXXXX]]
* [[XC]]

But these are fairly rare, and it's not the end of the world if we fail to correctly turn off sorting; the underlying order is still in the wikitext so it's easy to fix if need be. Benwing2 (talk) 20:50, 2 June 2025 (UTC)

@Ultimateria I fixed up the script and ran it and it would affect 11,532 pages. I need to do more verification of the output before running it and probably post to the BP about doing this, but it's looking good. Benwing2 (talk) 03:04, 3 June 2025 (UTC)

Removing capitalization and trailing punctuation

Hello, it seems like the WingerBot is making edits that it shouldn't be (at least, throughout the Telugu entries). Here is an example. I'm not sure why it's doing this, but please remove this. Tstcikhthys1 (talk) 04:49, 2 June 2025 (UTC)

@Tstcikhthys1 It is standard practice for non-English glosses to be lowercase and without final period. This is used throughout Wiktionary. You seem to be a new user, and so it's likely you don't yet understand the standard practices at Wiktionary. Please learn them before complaining about the way something is done. Benwing2 (talk) 04:56, 2 June 2025 (UTC)

@Benwing2 Where is this practice documented? Tstcikhthys1 (talk) 07:44, 2 June 2025 (UTC)

Wiktionary:Style_guide#Definitions. Examine several languages and you will see this in practice. Benwing2 (talk) 07:51, 2 June 2025 (UTC)

Thank you for pointing this out. I'll keep this in mind for my future edits. Tstcikhthys1 (talk) 06:07, 3 June 2025 (UTC)

Could you readd VDCs to Module:place/placetypes?

When I added village development committees to the module, an error occurred, and the first line was deleted. This led to User:Fenakhay reverting the entire edit and protecting the module page. I left a request on his talk page to return my additions to the module and explained what happened, but he isn't responding. [ɪˈɫa.wə kəˈta.kə] (talk) (edits) 17:59, 3 June 2025 (UTC)

@Ilawa-Kataka I can add this but I looked up w:village development committee (Nepal) and it appears it's no longer used; in place are gaunpalikas aka rural municipalities, and the latter already exists as a placetype (and IMO should be used in preference to the native term gaunpalika as it's more understandable). Are you planning on adding former VDC's as toponyms? Benwing2 (talk) 02:33, 4 June 2025 (UTC)

Generally multiple VDCs were merged into a single municipality, so I think it would be useful in some cases to qualify towns or villages as former VDCs. [ɪˈɫa.wə kəˈta.kə] (talk) (edits) 02:47, 4 June 2025 (UTC)

@Ilawa-Kataka OK I added it and marked it as "inherently former" so it will be treated as former even if you forget to add the "former" qualifier (but IMO you should always include this qualifier). Benwing2 (talk) 02:55, 4 June 2025 (UTC)

Don't be alarmed

I'm going to briefly delete your new request page so I can do the usual watchlist merge operation. Chuck Entz (talk) 05:36, 8 June 2025 (UTC)

@Chuck Entz OK sounds good. Benwing2 (talk) 05:38, 8 June 2025 (UTC)

Done. I deleted the new page, moved RFDO over it, moved RFDO back, then undeleted the new page. It should now be on all the same watchlists as RFDO. Chuck Entz (talk) 05:45, 8 June 2025 (UTC)

@Chuck Entz Great, thanks! The underlying specifics of how watchlists work are still a bit of a mystery to me. Benwing2 (talk) 05:48, 8 June 2025 (UTC)

bot error report

Reporting an error, not sure if it might also appear in other entries. In 密西西比 (Mìxīxībǐ) (see Special:Diff/82933381/84108671), the bot filled in the wrong id. Kethyga (talk) 07:00, 16 June 2025 (UTC)

@Kethyga Thanks for the report. This edit was done by me manually editing a text file and pushing the results in a batch; it was probably just a one-off problem, a brain fart, so to speak. There wasn't any script involved so there probably aren't any other such errors but I will check. Benwing2 (talk) 08:18, 16 June 2025 (UTC)

OK, I checked and I don't see any other such issues. It may have been a search-and-replace issue where I forgot to change Minnesota to Mississippi in a particular search-and-replace that only affected Chinese. Benwing2 (talk) 08:24, 16 June 2025 (UTC)

Replacement of quotation and other templates (June 2025)

Hello, could you please carry out the following template replacement?

• {{RQ:Eddison Worm|[page number in 1st parameter position]|[passage in 2nd parameter position]}} → {{RQ:Eddison Worm|page=[page number]|passage=[passage]}}

Thank you. — Sgconlaw (talk) 18:41, 19 June 2025 (UTC)

Thanks for getting to this so quickly! Thought I would add more requests here:

• {{asdfg}} → {{uw-protologism}} (moved following a discussion at RFM).
• {{RQ:Conrad Secret Agent}}: if |page= or |pages= used, then add |edition=copyright.
• {{RQ:Harvey Pierces Supererogation}}: if |page= or |pages= used, then add |year=1870 as 1st edition has been added to the template. (Note to self: snudge already refers to the 1st edition.)
• {{RQ:Langley Rhetoric|[page number in 1st parameter position]|[passage in 2nd parameter position]}} → {{RQ:Langley Rhetoric|page=[page number]|passage=[passage]}}

Thank you. — Sgconlaw (talk) 19:05, 23 June 2025 (UTC)

@Sgconlaw Sorry for the delay! All should be done. Benwing2 (talk) 23:26, 23 July 2025 (UTC)

Do you have any suggestions?

So, you know how Portuguese entries have the label {{lb|pt|European Portuguese spelling}} {{lb|pt|Brazilian Portuguese spelling}}, with aliases {{lb|pt|European Portuguese form}} {{lb|pt|Brazilian Portuguese form}}? I really wanted to split that alias into an independent label, so we could use it, e.g., in ===Alternative forms=== sections (instead of “Portugal”, “Brazil”) and more appropriately distinguish between what’s a form and what’s a spelling. I know how to make a new label in the labels module, but idk how I would go about replacing the entries that currently use the alias "European/Brazilian Portuguese form" with "...spelling".

My idea: using the dump that's coming in the next few days, make a list of pages with the sequences European Portuguese form, Brazilian Portuguese form, so I could... manually search, I guess, and change whatever needs to be changed? I don't expect there to be a lot of pages — I hardly ever see this alias used.

Do you pick up what I’m putting down, mister? Have you better ideas? And if not, could you scrape the database for me, pretty please?

CC: @Sérgio R R Santos

Grateful as always,

— Polomo47 (talk) 16:17, 22 June 2025 (UTC)

@Polomo47: It won't create a list, but if you enter insource:/lb\|pt\|European\ Portuguese\ form/ in the box at Special:Search, it will return every page with the precise string of characters "lb|pt|European Portuguese form" in the wikitext. If you want to allow for things like the label not being the first argument or different wording, it would be more complicated than I want to bother with now- but you can do all of that with help from the documentation. If you just want to find things, the search utility can probably do it without bothering with dumps, etc. It also is a lot more up to date- it doesn't update instantly, but I've never had to wait more than 20 minutes for edits to be reflected in the searches. That's a lot better than twice a month. Chuck Entz (talk) 20:15, 22 June 2025 (UTC)

Hm, yeah, but it’s not just in the {{lb}} template. It’s in {{tlb}}, {{standard spelling of}}, {{standard form of}}, and maybe even {{alt}}. Thought searching for the sequence European Portuguese form in a dump would yield better results. I don’t need super up-to-date; not many will have been made between the start and end of the dumping. Polomo47 (talk) 20:35, 22 June 2025 (UTC)

@Polomo47 Sure. The latest dump is as of two days ago so it should be fine. Just searching for European Portuguese form and Brazilian Portuguese form should find everything without many false positives. There are a few aliases of this like pt-br form but I will take care of them. Benwing2 (talk) 21:08, 22 June 2025 (UTC)

@Polomo47 I scanned the dump. There are a lot of hits (2900) and hardly any are false positives; there are a few from templates and modules that are arguably false positives, but I left them in. I created two files. User:Benwing2/brazilian-european-portuguese-form-spelling-orthography-2025-06-20-dump-readable is in case you want to edit each page manually; the displayed form of the file is human-readable, including links for each of the pages in question. User:Benwing2/brazilian-european-portuguese-form-spelling-orthography-2025-06-20-dump-begin-end is in case you'd rather do offline editing. In that case, copy the latter file to another page, and edit that page directly, making any changes to any lines but making sure not to delete any line (you can put indications of any sort outside the <begin> ... <end> portions on each line, and they will be ignored, so you can use this to put notes or done markers or whatever). When you are done, I can do a bot run to push all the changes to Wiktionary in a batch. Benwing2 (talk) 05:23, 24 June 2025 (UTC)

BTW on a different matter, I've been going through and fixing all the language-independent form-of templates so they display without any final period, and display with initial caps only for English. This is consistent with the general trend for non-English definitions to be formatted as lowercase-initial phrases without final period, whereas English definitions are formatted as full sentences. There are still a bunch of lang-specific form-of templates that automatically have initial caps and final period, and I'm planning on changing them, but I want to make sure that's OK with you for things like {{pt-pre-reform}}. So for example instead of

# Pre-reform spelling (used until 1943 in Brazil and 1911 in Portugal) of hipotenusa.

it would look like

# pre-reform spelling (used until 1943 in Brazil and 1911 in Portugal) of hipotenusa

and instead of

# Pre-reform spelling (used until 1943 in Brazil and 1990 in Portugal) of ação. Still used in countries where the agreement hasn't come into effect; may occur as a sporadic misspelling.

it would be something like

# pre-reform spelling (used until 1943 in Brazil and 1990 in Portugal) of ação; still used in countries where the agreement hasn't come into effect, and may occur as a sporadic misspelling

Benwing2 (talk) 05:33, 24 June 2025 (UTC)

For the files, thank you. 2900 is not a lot actually — it’s all a winter break away! (heh...)

Regarding the change to capitalization, I am generally favorable to it, yes. It’s worth noting that our policy on capitalization in definitions does have more nuance than “English is capped; other languages are not”: notably, the sentence A full definition should start with a capital letter. motivated me to get into trouble with Mr. Fenakhay just recently (see history). I did browse through the 2024 Beer Parlour, but I’m not sure if this was mentioned.

So, I do consider that definition lines with {{alt sp of}} and its coordinates are glossy in nature, and I agree with making them lowercase. Say, we’ve had {{plural of}} lowercase for a while, and no one complained, did they? I believe it’s a matter of getting used to, for anyone who was taken aback and attributed it to anything inherently negative.

Now, when it comes to a template like {{pt-pre-reform}}, with three whole sentences — the latter of which are nonoptimally written and thus need a semicolon to boot —, it’s hard to consider it a “gloss”... Nevertheless, I guess I’d rather it be lowercase for consistency.

So, all I want to say is, yes, but probably drop the comma before “and”. Lol. Polomo47 (talk) 18:59, 24 June 2025 (UTC)

Replacing

Hey! Can you quickly change all these guys so they link to lloviznar instead of lloviznarse. It's not a reflexive verb. Worm spail (talk) 18:08, 24 June 2025 (UTC)

Boxball

Could you add a separate Wikipedia article of Box Ball just wondering if this is possible? EliG233 (talk) 03:51, 29 June 2025 (UTC)

It is not four square EliG233 (talk) 03:52, 29 June 2025 (UTC)

Huh??? Benwing2 (talk) 04:05, 29 June 2025 (UTC)

Box Ball has four square https://en.wikipedia.org/wiki/Four_square this isn’t Boxball this is a separate sport all together played with Boxes EliG233 (talk) 04:12, 29 June 2025 (UTC)

If you can create a Wikipedia article of Boxball that would be good EliG233 (talk) 04:14, 29 June 2025 (UTC)

Or at least create a page called Hand Football on Wikipedia EliG233 (talk) 04:23, 29 June 2025 (UTC)

Can you give me an explanation of why you deleted it EliG233 (talk) 04:37, 29 June 2025 (UTC)

@EliG233: please read our Criteria for inclusion. Wiktionary is a descriptive dictionary based on usage- no usage, no entry. Period. Searching Google for "Hand Football" turns up a lot of people named Hand associated with football, touch football, giant foam hands people bring to football games, foosball, and various random bits of text where "hand" and "football" somehow end up next to each other. Nothing about the particular game that you apparently made up yourself. You've already been banned from Wikipedia for trying to promote things you made up. If you keep trying that here, you'll end up banned from Wiktionary as well. You also shouldn't misuse Wiktionary talk pages to attempt recruiting people to help you evade your block by doing the very things you got banned for. You're already on very thin ice. Don't make it worse. Chuck Entz (talk) 05:38, 29 June 2025 (UTC)

Kyrgyz declension

Could you make a change to your Kyrgyz declension module? Root-final <к> should become <г> between vowels. For example, the first-person singular possessive form of укук should be укугум, not укукум. That applies to all the possessive forms, of course. Thanks, Dylanvt (talk) 16:40, 30 June 2025 (UTC)

@Benwing2 Dylanvt (talk) 02:53, 9 July 2025 (UTC)

Thanks for the reminder, let me take a look. Benwing2 (talk) 02:58, 9 July 2025 (UTC)

@Dylanvt Does this apply only to root-final к, or to all unvoiced consonants? Do you have a good reference for Kyrgyz declension? Benwing2 (talk) 03:09, 9 July 2025 (UTC)

Oh whoops, п does it too. This grammar gives a decent brief overview. On page 7 it mentions the voicing of root-final п and к. Dylanvt (talk) 03:24, 9 July 2025 (UTC)

There's also a mistake in the verb conjugation template that was made by a now-blocked user. Past habitual vowel harmony works differently, as the ending vowel is always rounded (i.e. always у or ү). So e.g. for табуу, the past habitual should be тапчумун, not *тапчымын. Along with a few other oddities, but I'm not 100% sure about those, so I won't point them out for now. Is that template still fixable with its creator now blocked? Dylanvt (talk) 03:34, 9 July 2025 (UTC)

Yes, it might be best to rewrite it in Lua but it's definitely fixable. Benwing2 (talk) 03:52, 9 July 2025 (UTC)

@Dylanvt I spent some time analyzing the endings in the code to try to compress them. Some questions:

1. The code only gives 1s, 2s informal and 2s formal single and multiple possession forms. Are there third singular and 1/2/3 plural possessive forms?
2. The dative singular uses a-e-o-oe-type vowel harmony when all other singulars use y-e-u-oe-type harmony. Is this a mistake? Note in particular the similarity between the endings of the dative and locative singular, but the locative singular uses y-e-u-oe-type harmony.

Benwing2 (talk) 06:15, 9 July 2025 (UTC)

1. There are additional possession forms. 3s and 3p are identical: -I (i.e. и ы у ү) after a consonant, and -сI after a vowel. 1p is -бIз after vowels, -IбIз after consonants (but it's often not used, they'll just use the bare form). 2p informal is -ңAр (i.e. ңар, ңор, ңер, ңөр) after vowels, -IңAр after consonants (also sometimes not used). 2p formal is -ңIздAр after vowels, -IңIздAр after consonants (again, not always used).
2. The vowel harmony for the dative singular should be the same as for all other cases. I'm not sure what you mean by a-e-o-oe-type and y-e-u-oe-type, and don't understand the documentation enough to know what to look for. But the harmony type should be the same. Dylanvt (talk) 18:18, 9 July 2025 (UTC)

   Oh I found it and I understand the issue. It's that after <у> you don't get <о>, but <а>. So if I'm reading the documentation right, it's like that to get correct forms like укукка instead of the expected *укукко. Same with locative. But accusative has a front vowel in the ending, so you get the expected укукту. I think that's why there's the separate types.Dylanvt (talk) 18:25, 9 July 2025 (UTC)

   So the issue I was bringing up was actually due to a mistake I made. It seems you consistently have a-e-o-oe vowel harmony with low-vowel suffixes and y-e-u-oe vowel harmony with high-vowel suffixes; the difference is only in the у and ю vowels, which get grouped with аыя in the a-e-o-oe harmony but grouped with оё in the y-e-u-oe harmony. Essentially, the y-e-u-oe is a simple front/back, rounded/rounded harmony but the a-e-o-oe groups low+high back vowels against mid back vowels. There's a different issue I noticed, though, which is that sometimes you have а/о/е/ө variation in the suffixes, and sometimes you have а/а/е/ө variation. You can see for example that the locative singular has да/де/до/дө after a vowel, but locative first-singular singular possessive has мда/мде/мда/мдө after a vowel instead of мда/мде/мдо/мдө like you might expect; see lines 219 and 291 of the code. Is this intentional? Benwing2 (talk) 21:20, 9 July 2025 (UTC)

   Yes, that would be because the 1s possessive introduces a high vowel, after which you'll never get <о>. So e.g. you get доско and досто for dative and locative of 'friend', but досума and досумда for dative & locative of 'my friend'. The high /u/ blocks the rounding harmony from spreading to all subsequent affixes. Dylanvt (talk) 00:11, 10 July 2025 (UTC)

   Also note with the 2p informal that that /A/, like the /A/ in the dative and locative, should be [a] after /u/, not [o]. E.g. чокуңар, not *чокуңор. Dylanvt (talk) 00:42, 10 July 2025 (UTC)

   OK, a couple more questions:

   1. Does the blocking effect of у apply across a stem/suffix boundary? E.g. if you have a stem ending in -у, do you get plural -лор or -лар?
   2. I notice that the dative is sometimes -гА/-кА but sometimes just -А. For example the dat 1s singular possessive is -име/-ума etc. not #-имге/-умга as I would expect, and similarly with the 2s informal -уңа/-иңе, but the 2s formal uses -уңузга/-иңизге, where the -г- reappears. What's the rule here?

   Benwing2 (talk) 03:23, 10 July 2025 (UTC)

   1. After a stem with -у, you get plural -лар. So чокуда (loc sg), чокуңар (possessive), чокулар (plural), чокуларыңарда (all three).
   2. Dative takes г/к except in three cases: after 1s and 2s inf. possessive forms, where it's just the vowel; and after 3s/p possessive forms, where you get -нA: доско, досума, досуңа, досуңузга, досуна, досубузга, досуңарга, досуңуздарга, досуна.
      1. Locative also has an [n] appear after 3s/p possessive: чокуда, чокумда, чокуңда, чокуңузда, чокусунда... Dylanvt (talk) 19:11, 10 July 2025 (UTC)

      The ablative is -дан, very similar to locative -да. Does it also change to -ндан after the 3s/p possessive? Benwing2 (talk) 03:50, 12 July 2025 (UTC)

      @Dylanvt Benwing2 (talk) 03:51, 12 July 2025 (UTC)

      No, ablative after 3rd poss just takes the usual -н that it always gets after vowels (e.g. чокусунан). Dylanvt (talk) 17:27, 12 July 2025 (UTC)

      @Dylanvt Please see and review User:Benwing2/test-ky-decl-noun and feel free to add more nouns for testing. Benwing2 (talk) 21:48, 12 July 2025 (UTC)

      Please note, I don't know the terms for third-person singular/plural, first-person plural, etc. so I just copied сиздин from the second-person formal singular. Benwing2 (talk) 21:55, 12 July 2025 (UTC)

      Ah-ha, that would be анын (3s), биздин (1p), силердин (2p inf), сиздердин (2p f), алардын (3p). Yes, third person has different possessive pronouns but the same ending on the possessed noun. Dylanvt (talk) 22:02, 12 July 2025 (UTC)

      Oh also, бала has an irregular plural балдар lol. I think it's the only noun with an irregular plural, but I'm not completely certain. Dylanvt (talk) 22:04, 12 July 2025 (UTC)

      @Dylanvt I added a pl= arg to override the plural and fixed the remaining possessive pronouns. Let me know if you see any errors or other issues; if not I'll push it live. Benwing2 (talk) 01:00, 13 July 2025 (UTC)

      As far as I can tell everything's good. Thanks a lot! Dylanvt (talk) 16:59, 13 July 2025 (UTC)

      @Dylanvt I pushed the code. I'm working on the verb module; can you tell me all of the issues you currently know about? I also have a question about verbs whose stem ends in a vowel like окуу. The current table has the bare form оку given for 4 slots: 2nd singular informal imperative, 3rd person singular and plural future intentional, and 3rd person plural past simple. The imperative is almost certainly correct but I wonder about the others. Benwing2 (talk) 03:58, 14 July 2025 (UTC)

      Okay wow, that's quite the mess.

      1. past simple 3p should be окушту
      2. past indirect: This tense has two forms, more common seems to be with the suffix –тур- in all forms: окуптурмун, окуптурсуң, окуптурсуз, окуптур, окуптурбуз, окуптурсуңар, окуптурсуздар, окушуптур. But also acceptable is the current paradigm.
      3. present tense vowel harmony is wrong, all the 'ы's should be 'у's.
      4. окуур not *окур is normative for all future presumptive forms. You get long vowel when the root ends in a vowel, short when it doesn't (барар, келер vs. байлаар, жээр)
      5. I have no idea what "future intentional" is. Googling forms like окупун and барпын gives no Kyrgyz results. I asked a linguist who does Kyrgyz and he also had no idea what that's meant to be.
      6. 1s imp should be окуюн. Ending is -(A)jIn: барайын, келейин, байлаыйн, коёюн, сүйөйүн etc.
      7. 1p imp: the vowel harmony is wrong, should be окуйлу(к). Because also, the /к/ of that suffix is very frequently dropped. Both variants are widely attested.
      8. optative 2p and 3p are wrong. Should be: окугуңар келет, окугуңуздар келет, and окушкусу келет.
      9. conditional vowel harmony is partially wrong. Should be: окусаңыз, окусаңыздар Dylanvt (talk) 16:50, 14 July 2025 (UTC)

      I'll get back to you with more notes later today. Dylanvt (talk) 18:21, 14 July 2025 (UTC)

      A couple more notes from looking at баруу:

      1. Vowel harmony in past habitual is wrong: should be барчумун etc.
      2. I now realize what future intentional is. For баруу it's correct (except that 1s should be бармакмын, 2p should be бармаксыңар and бармаксыздар, and 3p should be барышмак). For some reason future intentional is completely messed up on окуу. Should likewise be окумакмын, окумаксың, etc. Future intentional is broadly speaking the "would" meaning, I'm not sure what (if any) source calls it "intentional", but whatever.

      There's also lots of missing forms, e.g.

      1. all the negatives.
      2. participles like барып/окуп (perfective), бара/окуй (imperfective), and ofc the negative participles барбай/окубай.
      3. remote past барган элем, барган элең, барган эле etc.
      4. remote past habitual барчу элем etc.
      5. unaccomplished past бара элекмин, бара элексиң, бара элек etc. (I haven't yet...)
      6. the two conditionals барар элем/окуур элең etc. and барат элем/окуйт элең etc. These are of identical meaning. The -мак conditional is more counterfactual. The -са form that's currently called conditional is the "if/when" form.

      Dylanvt (talk) 23:08, 14 July 2025 (UTC)

      @Dylanvt I can add all of these but it would help if I had a good reference. I was looking for a PDF copy of the "100 Kyrgyz Verbs Fully Conjugated in All Tenses" book but couldn't find one. If you have that book and can send copies of a few pages that would be great. What you could also do is write out the missing forms for a few verbs of different sorts, e.g. иштөө is a good one since it's vowel-initial with final vowel -е which rounds in certain circumstances that I'm not sure of. баруу might be a good consonant-initial verb. Then I will produce an attempt (the module is largely written already but need some of the core logic updated) and we can iterate. Benwing2 (talk) 23:17, 14 July 2025 (UTC)

      I unfortunately only have PDFs given to me by professors, idk if there's anywhere to access something like it online.

      Here's all the forms, excluding those with auxiliary verbs, to which end I think we should also exclude the optative. There's 3 groupings of endings, which I'll list first:

More information

P1 (past/conditional)

• -[I]м
• -[I]ң
• -[I]ңIз
• —
• -[I]к
• -[I]ңAр
• -[I]ңIздAр
• —

P2 (copula)

• -мIн
• -сIң
• -сIз
• —
• -BIз
• -сIңAр
• -сIздAр
• —

P3 (present)

• -мIн / -м
• -сIң
• -сIз
• -т
• -BIз
• -сIңAр
• -сIздAр
• -т

nonfinite

• inf: X+Uː
• past part (ptp): X+GAн
• future part (fp): X+Aр (after C) / X+ːр (after V) / X+р (after и ы)
• neg future part (nfp): X+BAs
• trans past part: X+(A)тIр (*NB I've never seen this form and am struggling to find many citations of it)
• prs prf part (prp): X+(I)п
• prs impf part (pip): X+A (after C) / X+й (after V)
• neg prs part (npp): X+BAй

Note that for all third person plural forms, there's two variants: either the same as 3sg, or the same as 3sg but with an -(I)ш attached to the root (i.e. before suffixes). You can see this in the иштөө examples below. For each tense I have positive above and negative below. non-past / aorist

• pip+P3
• npp+P3

recent past

• X+DI+P1
• ptp жок+P2 / X+BA+DI+P1

remote/general past

• ptp+P2 (1sg also: X+GAм)
• ptp эмес+P2

distant remote/general past

• ptp эле+P1
• ptp эмес эле+P1

indirect/unwitnessed past

• prp+тIр+P2 / prp+P2 *but in 3rd person -тIр is required
• X+BA+п+тIр+P2 / X+BA+п+P2 *but in 3rd person -тIр is required

past habitual

• X+чU+P2
• X+чU эмес+P2

remote past habitual

• X+чU эле+P1
• X+чU эле эмес+P2

past unaccomplished

• pip элек+P2

uncertain future

• fp+P2
• nfp+P2

“would” conditional 1

• fp эле+P1
• nfp эле+P1

“would” conditional 2

• X+A/й+т эле+P1 (/A/ after C, й after vowel) (this is just the 3s aorist form + эле)
• X+BA+й+т эле+P1

counterfactual conditional

• X+мAк+P2
• X+мAк эмес+P2

imperative/hortative (here positive is first, negative second)

• 1s: X+(A)йIн, X+BA+йIн
• 2sinf: X, X+BA / X+GIн, X+BA+гIн
• 2sfor: X+(A)ңIз, X+BA+ңIз
• 3s: X+сIн, X+BA+сIн
• 1p: X+A/й+лI, X+BA+й+лI / X+A/й+лIк, X+BA+й+лIк (/A/ after C, й after vowel)
• 2pinf: X+GIлA, X+BA+гIлA
• 2pfor: X+(A)ңIздAр, X+BA+ңIздAр
• 3p: X+сIн, X+BA+сIн / X+ш+сIн, X+ш+пA+сIн

“if” conditional

• X+сA+P1
• X+BA+сA+P1

Here's all the forms of иштөө as an example. nonfinite

• inf: иштөө
• past part: иштеген
• future part: иштээр
• neg future part: иштебес
• trans past part: иштетир (*NB I've never seen this form and am struggling to find many citations of it)
• prs prf part: иштеп
• prs impf part: иштей
• neg prs part (npp): иштебей

non-past / aorist

• иштейм/иштеймин, иштейсиң, иштейсиз, иштейт, иштейбиз, иштейсиңер, иштейсиздер, иштейт/иштешет
• иштебейм/иштебеймин, иштебейсиң, иштебейсиз, иштебейт, иштебейбиз, иштебейсиңер, иштебейсиздер, иштебейт/иштешпейт

recent past

• иштедим, иштедиң, иштедиңиз, иштеди, иштедик, иштедиңер, иштедиңиздер, иштеди/иштешти
• иштеген жокмун, иштеген жоксуң, иштеген жоксуз, иштеген жок, иштеген жокпуз, иштеген жоксуңар, иштеген жоксуздар, иштеген жок/иштешкен жок
  • иштебедим, иштебедиң, иштебедиңиз, иштебеди, иштебедик, иштебедиңер, иштебедиңиздер, иштебеди/иштешпеди

remote/general past

• иштегенмин/иштегем, иштегенсиң, иштегенсиз, иштеген, иштегенбиз, иштегенсиңер, иштегенсиздер, иштеген/иштешкен
• иштеген эмесмин, иштеген эмессиң, иштеген эмессиз, иштеген эмес, иштеген эмеспиз, иштеген эмессиңер, иштеген эмессиздер, иштеген эмес/иштешкен эмес

distant remote/general past

• иштеген элем, иштеген элең, иштеген элеңиз, иштеген эле, иштеген элек, иштеген элеңер, иштеген элеңиздер, иштеген эле/иштешкен эле
• иштеген эмес элем...

indirect/unwitnessed past

• иштептирмин, иштептирсиң, иштептирсиз, иштептир, иштептирбиз, иштептирсиңер, иштептирсиздер, иштептир/иштешиптир
  • иштепмин, иштепсиң, иштепсиз, иштептир, иштеппиз, иштепсиңер, иштепсиздер, иштептир/иштешиптир
• иштебептирмин... иштебептир/иштешпептир
  • иштебепмин... иштебептир/иштешпептир

past habitual

• иштечүмүн, иштечүсүң, иштечүсүз, иштечү, иштечүбүз, иштечүсүңөр, иштечүсүздөр, иштечү/иштешчү
• иштечү эмесмин... иштечү эмес/иштешчү эмес

remote past habitual

• иштечү элем... иштечү эле/иштешчү эле
• иштечү эле эмесмин... иштечү эле эмес/иштешчү эле эмес

past unaccomplished

• иштей элекмин, иштей элексиң, иштей элексиз, иштей элек, иштей элекпиз, иштей элексиңер, иштей элексиздер, иштей элек/иштеше элек

uncertain future

• иштээрмин, иштээрсиң, иштээрсиз, иштээр, иштээрбиз, иштээрсиңер, иштээрсиздер, иштээр/иштешер
• иштебесмин, иштебессиң, иштебессиз, иштебес, иштебеспиз, иштебессиңер, иштебессиздер, иштебес/иштешпес

“would” conditional 1

• иштээр элем... иштээр эле/иштешер эле
• иштебес элем... иштебес эле/иштешпес эле

“would” conditional 2

• иштейт элем... иштейт эле/иштешет эле
• иштебейт элем... иштебейт эле/иштешпейт эле

counterfactual conditional

• иштемекмин, иштемексиң, иштемексиз, иштемек, иштемекпиз, иштемексиңер, иштемексиздер, иштемек/иштешмек
• иштемек эмесмин... иштемек эмес/иштешмек эмес

imperative/hortative

• 1s: иштейин, иштебейин
• 2sinf: иште, иштебе / иштегин, иштебегин
• 2sfor: иштеңиз, иштебеңиз
• 3s: иштесин, иштебесин
• 1p: иштейли, иштебейли / иштейлик, иштебейлик
• 2pinf: иштегиле, иштебегиле
• 2pfor: иштеңиздер, иштебеңиздер
• 3p: иштесин, иштебесин / иштешсин, иштешпесин

“if” conditional

• иштесем, иштесең, иштесеңиз, иштесе, иштесек, иштесеңер, иштесеңиздер, иштесе/иштешсе
• иштебесем... иштебесе/иштешпесе

I think the only forms you might not be able to predict for e.g. баруу would be:

• fp = барар
• pip = бара
• 1s imp = барайын (cf. коюу = коёюн, сезүү = сезейин)
• 1p imp = баралы(к)

Dylanvt (talk) 02:39, 15 July 2025 (UTC)

@Dylanvt I took out the indentation in the box so it worked. I'm going through trying to get the existing forms of баруу and олуу correct and then I'll tackle the list you just posted. I have a question: why is the vowel harmony in the present past habitual of баруу барчумун etc.? What is the rule here that causes a у after an а? Is there a rounded-only type of harmony here? Benwing2 (talk) 02:47, 15 July 2025 (UTC)

BTW (1) thank you for all the info! (2) some of this discussion may be easier to carry out on Discord. Are you on Discord? If not see WT:DISCORD. Benwing2 (talk) 02:51, 15 July 2025 (UTC)

Oh duh! I am on the Discord, just not super active. I didn't think to send stuff over there. Oh well, next time (or if there's anything else needed for this). Dylanvt (talk) 02:53, 15 July 2025 (UTC)

Thanks for fixing the box.

Yes, that suffix has a vowel that's "underlyingly specified" as rounded –– just like the infinitive vowel, which is why we get баруу instead of барыы (well, also because there is no long и or ы in the language, but you get the point). There's also the always-round adjective-forming suffix -LUU (e.g. ылайыктуу Dylanvt (talk) 02:52, 15 July 2025 (UTC)

@Dylanvt Please see User:Benwing2/test-ky-conj. I haven't yet implemented round vowel harmony so there are mistakes there and probably lots of others; I will review your table and comments above and fix them. I also haven't added any of the forms not found in the existing tables; that comes next. In the meantime please add more verbs for testing. Note that the |pagename= parameter defaults to the current pagename and won't be necessary when used on mainspace pages, but the |stem= parameter will probably be needed for all vowel-final verbs (maybe I can reduce it to just the final vowel instead of specifying the full stem). Benwing2 (talk) 03:21, 15 July 2025 (UTC)

@Dylanvt Please see the latest User:Benwing2/test-ky-conj. I'm not sure about барчо- in the past habitual. It seems like maybe it should be барчу- but if so I don't understand the rounding harmony rule. I'm also not sure about the thicker black lines; maybe instead we should change the background color of all the negatives. Also you didn't give any negative for past unaccomplished so I assume it doesn't exist; if this is not true, please let me know. Benwing2 (talk) 01:46, 16 July 2025 (UTC)

Oh, I forgot to include the remaining participles in the table. Benwing2 (talk) 01:47, 16 July 2025 (UTC)

You're right about барчу-.

I think changing the colors could be neat, since the negative label is already a different color from the positive? Dylanvt (talk) 02:13, 16 July 2025 (UTC)

And yeah, the past unaccomplished is already negative semantically, so there's no "negative" of it. Dylanvt (talk) 02:14, 16 July 2025 (UTC)

@Dylanvt I haven't heard from you on Discord for a few days so I went ahead and pushed to production the sandbox version with footnotes and support for overriding individual participles. Let me know if you see any mistakes. Benwing2 (talk) 05:35, 21 July 2025 (UTC)

Sorry, I've been busy past few days. I did check and didn't see any mistakes. My only thought was maybe the footnotes should be more consistent in their terminology; the descriptors you included are just the ad hoc words that came to my mind when I first explained them. When I have time I'll head over to Discord and we can come up with better descriptors. Dylanvt (talk) 15:19, 21 July 2025 (UTC)

Template:syn

Hi, here on enwiktionary, {{syn|bn|পরিকল্পনা}} gives Synonym: পরিকল্পনা (porikolpona). On bnwiktionary, we don't need the translit part for Bengali word (in this example, the (porikolpona) part). We still want "Transliteration" for other languages but not for Bengali language. How can we remove it? Could you please help? This module is very complex, and we don't have anyone who understands Lua code. আফতাবুজ্জামান (talk) 22:55, 30 June 2025 (UTC)

@আফতাবুজ্জামান This sort of change should not be done in Module:nyms but in the general language code. You need to find the module corresponding to Module:languages/data/2 and remove the translit entry from the section on Bengali (code bn). Benwing2 (talk) 23:04, 30 June 2025 (UTC)

When I search for 'bn-translit', I only see three results.[2] I'm not sure where it's coming from. আফতাবুজ্জামান (talk) 23:14, 30 June 2025 (UTC)

Yes, it's the third search result. Benwing2 (talk) 23:23, 30 June 2025 (UTC)

Thanks, that worked. However, a new problem has come up, e.g. see bn:বিশ্ব#উচ্চারণ. It says "Lua error in module:bn-IPA at line 91: The term 'বিশ্ব' could not be transliterated." I think we need to directly specify 'Module:bn-translit' to 'Module:bn-IPA', but I'm not sure how or in which line. Could you help? আফতাবুজ্জামান (talk) 00:15, 1 July 2025 (UTC)

@আফতাবুজ্জামান You need to change line 78 of Module:bn-IPA so it reads like this:

return (require("Module:bn-translit").tr(text))

Benwing2 (talk) 21:41, 2 July 2025 (UTC)

{{ja-readings}} referencing feature

Hi Ben, I don't recall in what level of detail we talked about this idea before, but I recall that I was supposed to help to remove some spurious kanji readings that we had imported a long time ago from Unihan. If I recall correctly, you said something about perhaps implementing something on the side of {{ja-readings}} that would show which readings were questionable, and basically provide a way to specify references for readings inline inside the list.

Do you remember whether you did implement such a thing? Don't worry if not, but I'm just asking so I know whether I should put a proposal out for how it could look if it's not already been done somewhere. I've been working on readings a lot recently, and it's just resurfaced to me how useful this could be. Kiril kovachev (talk・contribs) 20:31, 6 July 2025 (UTC)

@Kiril kovachev Almost positive I didn't implement this, and I don't see any evidence in Module:ja-kanji-readings of my having done this. It would be great if you could dig up the original discussion as I barely remember it now. But if you create a concrete proposal for doing this and need implementation help, I can probably help. Benwing2 (talk) 21:39, 6 July 2025 (UTC)

@Benwing2 the discussion was here, way back in September 2023: Wiktionary:Beer parlour/2023/September#really long single-kanji readings. I ought to apologize again for having not gotten round to doing what I'd said I would then, and I still can't promise anything for a short while. But, as a first step, I'm definitely interested in getting this part done, since it means there can be some accountability at least for all these obscure readings.

Thanks for offering to help. I may take you up on that if you happen to have some time, but I don't wish to trouble you or anything. I'd seen that the module hadn't seemingly been edited, but I asked just in case you had some private changes on GitHub or somewhere perhaps. Anyway, maybe tomorrow (jinxed!), I'll have a go at putting up a proposal for how it could look. Kiril kovachev (talk・contribs) 21:54, 6 July 2025 (UTC)

Thanks for the link. I had totally forgotten about that, but it looks like I did create a script called find_bad_japanese_readings.py which is probably still in the github repository at https://github.com/benwing2/WingerBot (sorry, I need to clean up the repository as it's enormous and most files are at top level still). I checked and I don't have any private changes for Module:ja-kanji-readings. Ping me when you have a proposal ready. Benwing2 (talk) 22:03, 6 July 2025 (UTC)

Russian short forms

Please stop removing Russian short forms (e.g. https://en.wiktionary.org/w/index.php?title=адъективный&diff=prev&oldid=85488078). Your effort is very misguided.

First of all, the fact that short forms don't exist in a corpus is not evidence of the short form not existing: if for some rare noun with a very obvious declension pattern the e.g. instrumental case happens to be missing from a corpus, surely you will not mark that noun as lacking an instrumental case? Because every native speaker would be able to produce that form readily if the need ever arises.

Second, the corpus which you are using to validate short forms is insufficient. It took me 15 seconds of googling to find a PDF containing "адъективен": https://rhema-journal.com/Rema-2-2017.pdf - and I am sure that for many of the other adjectives from which you removed short forms examples could be found too. Please ask native speakers for advice before embarking upon a massive content removal! Tetromino (talk) 20:41, 7 July 2025 (UTC)

Please supply the quote. Poetic truncated forms don't count. Benwing2 (talk) 20:43, 7 July 2025 (UTC)

Short forms are not like instrumental forms because they depend on the semantics of a given word. адъективный is of the wrong semantics to have a short form. You may be a native speaker but that doesn't mean your grammar is perfect. Benwing2 (talk) 20:45, 7 July 2025 (UTC)

Please elaborate your argument why "адъективный is of the wrong semantics". Are you trying to argue something like "адъективный is a relational adjective and Russian relational adjectives don't have short forms"? Because that is a simplistic rule that is asserted by schoolbooks but which in reality has thousands of exceptions. There are some specific kinds of relational adjectives whose semantics really don't allow short forms (for example, I cannot think of any relational adjectives which relate to the name of a place which have a short form); but адъективный is not one of them. Tetromino (talk) 21:00, 7 July 2025 (UTC)

Pure relational adjectives tend not to have short forms. Many relational adjectives have non-relational metaphorical meanings that do have short forms, but for non-obvious cases you need to find actual examples, and IMO truncated poetic forms don't count. We are a descriptive dictionary and so "I'm a native speaker and I assert such-and-such exists" doesn't cut it in non-obvious cases: you need to supply citations. That's what I've asked you to do, so please don't revert any change I make without such a citation. I found your example of адъективен and I agree it's a good example but it's the only one I could find, which means it's very rare (and there's an indication of this, rare-a*). As an example, I searched Google Books for адюльтерен and адюльтерны and found no examples, which surprises me, but I take it there are no short forms for this adjective as you'd expect they would be easy to find in the meaning "adulterous". Benwing2 (talk) 21:08, 7 July 2025 (UTC)

"We are a descriptive dictionary and so "I'm a native speaker and I assert such-and-such exists" doesn't cut it in non-obvious cases: you need to supply citations." - the whole problem is that адъективный is an obvious case! Perhaps non-native speakers don't have an instinct for what is obvious and what is not. Re адюльтерный - that is a rare, very bookish word; and there are few occasions when one would want to use it. When describing a case of adultery, you would not say адюльтерные отношения or similar - you would say супружеская измена. When describing an adulterous person, you would not say that person is адюльтерный - you would say that person изменяет жене/мужу. The noun адюльтер already smells of something from a dusty 19th century book on international family law, and the adjective derived from it is something I personally would never use even in formal writing. So I am not surprised that the short form of it doesn't occur in google books. Tetromino (talk) 21:39, 7 July 2025 (UTC)

Well none of this concerning адюльтерный is indicated in the definition, and I found lots of examples of адюльтерный as a long form adjective in Google Books, including in novels that aren't especially formal (see the quote I provided). So it surprises me you say it's rare and very bookish. But if this is true it should definitely be indicated. As for адъективный, I don't see at all why you think this is an obvious case, and if it were, there would be lots of examples, not only one. "Obvious cases" are words with semantics like "active", "confusing", "strange", etc. that are easy to form a comparative of or modify with a degree adverb like "very", "somewhat" or "so". I don't go looking for short forms of such adjectives; instead I look for cases like афферентный (I found examples), апикальный (I found one example, which also supplied an example for дентальный), анкетный (I found a single example and was surprised to find any), антигенный (I found examples) and алкогольный and аллювиальный (I couldn't find examples of either). Benwing2 (talk) 21:53, 7 July 2025 (UTC)

Алкогольный and аллювиальный are obvious cases (short forms are easy to produce and have a reason to exist). But I am having some doubts about адюльтерный. Tetromino (talk) 22:20, 7 July 2025 (UTC)

... and to prove what is obvious, here is a PDF with "алкоголен" found in just a few seconds: https://www.uanalyticon.ru/junior2023/georgy_cherkasov.pdf. Tetromino (talk) 22:23, 7 July 2025 (UTC)

@Tetromino Where is the quote for алкоголен in that paper? It appears on its own not in a sentence, which doesn't count. And what about аллювиален? I couldn't find examples even outside of Google Books. Benwing2 (talk) 00:22, 8 July 2025 (UTC)

ru-pron syllable count

Hi, how do you add the "Russian X-syllable words" category to each word in the ru-pron module? I'm trying to add that to the rsk-IPA module. Thanks. Insaneguy1083 (talk) 06:48, 9 July 2025 (UTC)

AFAIK this is handled in Module:IPA/data. You add to langs_to_generate_syllable_count_categories, and if there are any diphthongs that are written with vowel signs, you have to tell Module:IPA/data about it. Benwing2 (talk) 19:41, 10 July 2025 (UTC)

formatting fraktur

hi,

here, under usage notes, i formatted some text as fraktur, but the only model i have to work from links the word, which isn't appropriate in this case. which template should i use here? kwami (talk) 04:24, 14 July 2025 (UTC)

Use {{lang}}. Benwing2 (talk) 04:29, 14 July 2025 (UTC)

thanks kwami (talk) 04:32, 14 July 2025 (UTC)

Czech conjugation module

Hello, I would like to ask whether you still plan on making a conjugation module for Czech. I see that you started it two years ago. If not, I would make it myself based on the Old Czech verb module, that is originally based on your adjective module. It works really great and, like the original adjective module, it isn't too long and complicated. It would finally allow Czech to have verb class categories etc. Thank you in advance for your answer. Zhnka (talk) 11:41, 14 July 2025 (UTC)

There are some complications I found with modern Czech verbs and since I was close to finishing, I think it's best if I finish what I have. Give me a couple of weeks and it should be in a good state. Benwing2 (talk) 21:24, 14 July 2025 (UTC)

Template:R:ru:Vasmer

I just cleaned up an invisible module error at лох from an empty |1=. The url contained the string "%3ELua+error%3A+bad+argument+%231+to+%26%23039%3Bgsub%26%23039%3B+%28string+expected%2C+got+nil". Also, the link to starling.ru.net doesn't work on my browser (Firefox for Mac). It apparently tries and fails to make an https out of the http somewhere along the way. I'm not sure if you can do anything about the second issue, though. Chuck Entz (talk) 15:08, 16 July 2025 (UTC)

@Chuck Entz I fixed the first issue. Starling was working yesterday for me and it's not today (same error as you) so something has gone wrong in the meantime. Benwing2 (talk) 17:57, 16 July 2025 (UTC)

Sources for Proto-West Germanic inflection Templates

Hi, I found your contribution to the template for some Proto-West Germanic inflection templates, and I would like to find sources from which thy are written. Thanks! Themistokl (talk) 11:59, 18 July 2025 (UTC)

@Themistokl Which contribution? I have barely contributed to Proto-West Germanic. It looks like the modules were mostly written by @Rua, who's no longer active, with some more recent contributions by @Victar, who AFAIK contributed many or most of the PWG lemmas and who you might ask. Also try asking @Mellohi!, who does a lot of Indo-European stuff. I assume one of the primary sources is Ringe, possibly specifically this book by Ringe and Taylor [3]; beyond that I dunno. I assume some of the entries list sources, you might check for that as well. Benwing2 (talk) 19:04, 18 July 2025 (UTC)

Benwing is correct, {{R:ang:DOE}} would certainly be the best source amongst the few we have. --{{victar|talk}} 20:34, 18 July 2025 (UTC)

Thank you, I'll check. Themistokl (talk) 21:57, 18 July 2025 (UTC)

A quick message for you.

Why is your main account named User:Benwing always less editing and protects many pages? Rage Ragest (talk) 09:42, 20 July 2025 (UTC)

Error with your bot

It nonsensically turned "short for" into "blend", which broke some entries, e.g. [4]. 2A00:23C5:FE1C:3701:6D67:7D19:9A43:6A92 17:31, 20 July 2025 (UTC)

This. I've just fixed:

• MOAL (Diff ~85377690)
• гідроелектростанція (Diff ~85377576)
• dewas (Diff ~85377353)
• Chunnel (Diff ~85376223)

— Alexis Jazz (talk) 02:05, 31 July 2025 (UTC)

@Alexis Jazz Sorry, those were done manually offline and I got confused about {{blend of}} vs. {{blend}}. Thanks for fixing. Benwing2 (talk) 02:59, 31 July 2025 (UTC)

Hello Ben! Regarding edit of Appuy

Hello, Ben! I saw ur bot worked on the word “appuy” and I was wondering if u could help me out I am new to Wiktionary. I am willing to learn and scope out for new words to specify. I just had a few issues with putting citations verifying the word exists. Would be greatly appreciated. Santiago, José Noé (talk) 15:02, 24 July 2025 (UTC)

another non-English definition template to uncapitalize?

{{la-praenomen}}, as seen on e.g. Quīntus. (Why do we have a template for praenomina but not e.g. cognomina, I wonder?) - -sche (discuss) 17:33, 3 August 2025 (UTC)

Problems with spanish modules

Hi, I will be brief but I have to leave you two remarks about Spanish words.

1. the module for mounts does not process final hiatos, look what I had to do at saín.
2. The alternation for culiar is wrong, is not í but e. It is said "Yo culeo" instead "Yo culío".

Leave up to you to fix these things. Cheers. Tmagc (talk) 14:09, 16 August 2025 (UTC)

For #1 I take it you are referring to plurals. I will see about fixing that. For #2, I would need confirmation from other Spanish speakers. AFAIK yo culío is correct. Benwing2 (talk) 17:46, 16 August 2025 (UTC)

@Tmagc: We treat culiar as a variant of culear. That means all of its forms should use the "i", while all the forms of culear should use the "e". The variant with the "i" seems to be far less common and may even be nonstandard, but the Diccionario de americanismos has an entry for it. As a descriptive dictionary, we should, too. Chuck Entz (talk) 18:17, 16 August 2025 (UTC)

@Chuck Entz You are right but I didn't said that. Of course is nonstandard, however the shift of e to i occurs only when is non stressed. The stressed forms keep the e. So, it is correct as it is in Spanish Wiktionary, keeping the e on stressed forms. And not necessary all forms use the i, vowel alterations exist on stressed forms of many verbs, and so your module is missing the i->e alteration, culío does not exist. Tmagc (talk) 18:20, 16 August 2025 (UTC)

Edit request

Hi, can you add Tamil to the "langs_with_modules" in Module:category tree/lang? Thanks. Pixelpito (talk) 17:11, 20 August 2025 (UTC)

@Pixelpito Hi. I took a look at the module you created, Module:category tree/lang/ta. Normally we don't categorize subtypes of verb forms or noun forms like this; the decision was specifically made awhile ago to eliminate things like noun plural forms, so I think there is no need for a category auxiliary verb forms. Benwing2 (talk) 03:20, 22 August 2025 (UTC)

Some minor bot errors

See LWIR, SWIR and USCT. I cleaned up MWIR and (possibly related) Mercedes. JeffDoozan (talk) 00:00, 31 August 2025 (UTC)

Thanks. As always, these are due to editing mistakes/typos I made while manually editing an offline file. Usually there are only a few such errors but I'll check the offline file to make sure there aren't any others. Benwing2 (talk) 00:05, 31 August 2025 (UTC)

@JeffDoozan OK, there were 8 of them, some already fixed; I fixed the remainder. Note that commas between different possible abbrevs in something like MWIR should not have a space after them; without a space they are treated as separate items and separated with or or and, as appropriate. Benwing2 (talk) 00:33, 31 August 2025 (UTC)

Found an Inconsistency?

Hey, I appreciate your work on the geography terms, and I wanted to point out an inconsistency I'm seeing: I think Category:en:Places in Beijing should be "Category:en:Places in Beijing, China" because it's province-level. Same for Tianjin, Chongqing and Shanghai; let me know if any confusion; if you don't agree then nevermind. Thanks again for making the whole Wiktionary space better. Also, I have managed to mess up the Guangxi category so bad that it doesn't even appear in the list at Category:en:Places in China, I would appreciate it if we could restore that back to that category, apologies! Geographyinitiative (talk) 22:27, 14 September 2025 (UTC)

@Geographyinitiative Yeah I struggled with that. Ultimately I concluded that Beijing, Shanghai etc. are more city-like than province-like so should be treated as cities rather than higher-level administrative divisions. In some cases it may make sense to distinguish both; hence we have both Category:Delhi (the city) and Category:Delhi, India (the union territory surrounding the city). China in particular is especially thorny to handle due to the dominant system of "prefecture-level cities", which aren't really cities so much as prefectures and where the urban portion of the prefecture-level city (what we normally think of as the city proper) is usually called a district and has a different name. In fact I wrote a long comment in Module:place/locations about the special way these are handled:

	-- In China, a "prefecture-level city" is not a city in any real sense. It is rather a prefecture, which is an
	-- administrative unit smaller than a province but bigger than a county, which is administratively controlled by
	-- the chief city of the prefecture (which bears the same name as the prefecture), in a unified government. Prior
	-- to the mid-1980's, in fact, prefecture-level cities *were* prefectures, and a few of them (especially in the
	-- western portion of China) have not yet been converted. Generally a given province is entirely tiled by
	-- prefecture-level cities, another indication that they should be treated as prefectures and not cities per se.
	-- Yet another indication is that prefecture-level cities can contain counties and county-level cities (which, much
	-- like prefecture-level cities, are effectively counties surrounding a chief city of the county, again which bears
	-- the same name as the county-level city).
	--
	-- For this reason, we treat prefecture-level cities as non-city political divisions, and separately enumerate the
	-- most populous so we can separately categorize districts and counties under them instead of lumping them at the
	-- province level.
	--
	-- Note also that China separately distinguishes "urban area" from "metro area". Sometimes the two figures are
	-- identical but sometimes the metro area is larger (and very occasionally smaller, which I assume is an error). I'm
	-- guessing that the "urban area" is the contiguous urban area over a certain density while the metro area includes
	-- all urban areas above a certain density; when the latter is greater, it's because of satellite cities in the
	-- metro area separated by suburban/exurban or rural land.

	-- At first I chose all prefecture/province-level cities with a total prefecture/province-level population of at
	-- least 6,000,000 per the 2020 census with data taken from https://www.citypopulation.de/en/china/admin/ (a total
	-- of 67, including the four direct-administered municipalities), and also chose all prefecture/province-level
	-- cities whose "urban population" was at least 2,000,000 per the 2020 census with data taken from Wikipedia
	-- [[w:List of cities in China by population#Cities and towns by population]] (a total of 61 cities; if we cut off
	-- at 1.5 million we'd have 84 cities, and if we cut off at 1 million we'd have 105 cities). Merging them produces
	-- 87 cities. Note that this leaves off a few well-known cities (Guilin, Qiqihar, Kashgar, Lhasa, ...) but includes
	-- a lot of obscure cities.
	--
	-- At a later date I added all cities from citypopulation.de whose "urban" population per the 2020 China census was
	-- >= 1 million, and then finally added all urban agglomerations from citypopulation.de whose 2025-01-01 estimate
	-- was >= 1 million. These are sorted below by the urban agglomeration value (which is generally of the "adm-urb" =
	-- "administrative area (urban population)" type) and sometimes groups nearby cities into a single agglomeration
	-- (most notably in the case of the Pearl River Delta, grouped under Guangzhou with an agglomeration population of
	-- 72,700,000 but including a large number of nearby large cities in the agglomeration (although for some reason not
	-- Hong Kong, maybe due to the administrative issues involved). In addition, citypopulation.de includes divisions
	-- under a prefecture-level city if they are city-like and have an agglomeration population of at least 1 million;
	-- this includes several county-level cities, one county and one district (Wanzhou, a "district" of Chongqing
	-- despite being 142 miles away). None of the county-level cities or counties have districts under them, only
	-- subdistricts, towns and townships.

As for Guangxi, I'll take a look and see what's up. Benwing2 (talk) 00:04, 15 September 2025 (UTC)

Hmmm, I don't see anything wrong with either Guangxi or Category:Guangxi, China. Could you be more specific as to what is wrong? Guangxi is correctly categorized into Category:en:Places in China and Category:en:Provinces and autonomous regions of China, and Category:Guangxi, China is also correctly categorized. Benwing2 (talk) 00:08, 15 September 2025 (UTC)

I made a mistake and dis not specify the category I was referring to concerning Guangxi. Category:en:Places in China should include Category:en:Places in Guangxi, China (distinct from Category:en:Guangxi and the like) in the "Subcategories" list for that category, along with the places in the other province level divisions, but it does not appear. Also, at Guilin, you can see that the name of Category:en:Places in Guangxi, China is italicized. This was due to my error somehow. Thank you for reviewing the other issues, I have no meaningful opinion on the issues. That's fine if Beijing etc do not mention China, I was just confused. Thanks again fir your great work! --Geographyinitiative (talk) 10:07, 15 September 2025 (UTC)

mod:ms-pron

Hello Benwing2!

Can you tell me which part causes an error as I didn't see anything popping out in Category:Error. Thanks Sponge2490 (talk) 04:45, 15 September 2025 (UTC)

@Sponge2490 It's Category:Pages with module errors. You need to edit the broken revision and preview some of the pages still there like Butterworth, Hungary or putera to see the error. You won't see the errors now since I've already reverted the bad commits. Benwing2 (talk) 04:51, 15 September 2025 (UTC)

I fix the bug and there's no more error for all the input in here. But how can I make sure there's no more error before changing the main module? Sponge2490 (talk) 05:35, 15 September 2025 (UTC)

@Sponge2490 The best you can do is have a bunch of testcases, like you do, and after deploying, check CAT:Pages with module errors every so often for awhile afterwards (e.g. an hour or two) to see if any errors appear. Pages are regenerated gradually after you make a module change so you won't necessarily see any errors immediately. Benwing2 (talk) 06:08, 15 September 2025 (UTC)

Southern Altai Verb Inflection

I made some templates about Southern Altai Verb, but they aren't operating, how I fix these templates?

If people who can know Turkic languages help me, It would be appreciated.

For example:

• Module:alt-verb
• Template:alt-v2
• Template:alt-v3
• Template:alt-v4
• Template:alt-v5

118.216.30.67 13:29, 16 September 2025 (UTC)

On IPA license

Hello Benwing2. First of all, I am really impressed by your work. Amount of effort to develop and validate all these things is just enormous. My respect. Moreover, this all is published under free licenses, can't thank you enough. So yes, my question is about the license of produced IPA transcriptions. I know that everything on Wiktionary is CC-BY-SA. Still, structured data on Wikimedia Commons and Wikidata is CC0. Hence copying your IPA from Wiktionary to Wikidata is not possible. Do you consider this an issue? This Wikidata page (gato) has an example of how I would like to use IPA on Wikidata. Is releasing your IPA data under CC0 acceptable for you? This is not a request but a question. Dunkaist (talk) 04:23, 3 October 2025 (UTC)

Several comments:

1. Please sign your comments with four tildes, like this: ~~~~. That ensures that the comment is timestamped. Also please place new topics at the bottom rather than the top; I moved your comment to the bottom accordingly.
2. The short answer about whether you have my permission to use this data is, yes, go ahead from my perspective.
3. How are you planning on entering the data into Wikidata? Are you planning on doing a bulk import (using a bot) of all the Portuguese IPA pronunciations? There are thousands of them, so it's not feasible to do by hand.
4. I have some general concerns about the entire Wikidata approach to things like this, because the IPA will be entered static, and any later fixes to the module won't get reflected in the Wikidata. This is a general problem with Wikidata and is one of the reasons that Wiktionary doesn't make much use of Wikidata.
5. From a legal perspective, I don't actually know whether it's disallowed to copy factual programmatic outputs from a more restrictive to a less restrictive license, because the data is factual; it could be argued that it fails the creativity test required for copyrightable work. However, I Am Not A Lawyer so I really have no idea if this is correct. We do have a lawyer, User:BD2412, who knows a lot about this stuff and occasionally still edits Wiktionary entries and they may be able to comment more definitively. Benwing2 (talk) 04:51, 3 October 2025 (UTC)

@Dunkaist Adding a ping so you'll get notified of this response. Benwing2 (talk) 04:52, 3 October 2025 (UTC)

If I am correctly understanding the gravamen of the question, my legal opinion would be that there nothing copyrightable specifically in an IPA pronunciation of a dictionary word or phrase. bd2412 T 18:34, 3 October 2025 (UTC)

@BD2412 Thanks! Yes that was the question I was asking. Benwing2 (talk) 18:37, 3 October 2025 (UTC)

👍 bd2412 T 19:02, 3 October 2025 (UTC)

He also adds “pronunciation varieties” to Wikidata, evidently from our labels attached to the IPA, which are but peculiar in some intentionally effortful hall-of-fame edge cases, like قيطون, but even that is no creative work and rather mechanical description after a bunch of available sources, as is required for the threshold of originality, in my legally qualified opinion, in spite of my own interest in being the author of that page. Fay Freak (talk) 19:42, 3 October 2025 (UTC)

@Fay Freak Well, distinguishing IPA transcriptions of European and Brazilian dialects is not hard even for a beginner like me. As a rule, I understand it by IPA itself without double-checking against Wiktionary. Anyway, if I can get the IPA strings from the bot directly, I can get the dialect from it too. Dunkaist (talk) 23:35, 3 October 2025 (UTC)

1. Sorry for the inconvenience of posting at the top and the short signature. I have done a number of edits here and there across Wikimedia projects, still I am new to discussion pages. Thank you for pointing me to these.
2. Lovely, I really appreciate. I have started learning European Portuguese recently and I find IPA transcriptions extremely useful to check and correct my pronunciation. If I understand correctly, a more experienced learner of Portuguese knows the rules and doesn't need transcription. I am far from that level at the moment though.
   1. I use Anki flashcards to extend my vocabulary. While there are quite a few publicly shared decks for Portuguese, many of them target the Brazilian dialect, or lack the IPA, examples, etc. Most decks just lack any license information and look like a copypaste of unknown source. It is hard to get in touch with authors to fix mistakes. As a mathematician, I happen to be pedantic about these things. Therefore I decided to create yet another standard deck without the mentioned drawbacks. Yes, creating flashcards can be boring and slow, yet I consider it a part of learning. Moreover, it helps me to systematize and structure knowledge, not just memorize stuff. It became obvious soon that the best flashcard deck is the one that you can generate for your needs, e.g. with or without pictures, with or without IPA, with or without pronunciation audio, or maybe you prefer a male or female voice, or translation into a non-default language, or eventually just a definition.
   2. So, the project can be logically split into two parts: the script that generates flashcards and the database with all the language things. I am a software engineer and can write and maintain the script myself. Wikidata with its structured data looks perfect to accumulate language data. Well, some aspects of natural languages are hard to formalize, I agree. That said, for the limited purpose of vocabulary flashcards Wikidata offers far more than enough. First I plan to spend more time on the script that generates flashcards and update Wikidata pages manually. I need to get familiar with the data model, add some examples of each part of speech, etc. Then I can implement a mechanism to track or maybe even update changed IPA automatically.
3. So yes, after playing with things a while I will definitely need to do a bulk import. As I said, I am a software engineer. Also, I have some experience with massive changes in OpenStreetMap. Therefore this sounds not easy but doable to me.
4. As for maintenance, you mentioned the weak point yourself: static data. Still, I believe, it is possible to do as follows:
   1. annotate each IPA generated by your bot with something like Pxxxxx(generated by)=Qyyyyyy(your bot);
   2. run another script weekly, or monthly, or by request that for each IPA annotated as above updates the IPA if changed. That is it.
5. Since you have agreed to release the IPA data under CC0, this is more of a theoretical discussion now. Anyway, here is my understanding. There is a classical example of a telephone book where each phone number is a known fact but this doesn't mean anybody can just copypaste the whole book under their name. I believe the same principle applies to your IPA data too. That is why I decided to ask you. Dunkaist (talk) 23:29, 3 October 2025 (UTC)

   This all sounds good. I created flashcards (the paper kind) when I was learning Mandarin Chinese, in order to better learn the characters, so I understand the value of them. BTW if you're looking for European Portuguese IPA, you can find them in Infopedia, e.g. https://www.infopedia.pt/dicionarios/lingua-portuguesa/racional for racional. Unfortunately I don't know of any source for Brazilian Portuguese IPA online; none of the Brazilian dictionaries I've encountered have IPA in them. In response to your message I would just add:

   1. Writing a script to do bulk import isn't all that hard. I think the biggest issue is figuring out the data model of Wikidata and constructing the correct API calls. You can use pywikibot to do the actual interfacing with Wikidata and Wiktionary, or just call the REST API directly.
   2. I assume if you want to do a bulk import of IPA to Wikidata, you should get consensus from the Wikidata community (maybe the admins, or the Portuguese Wikidata community if there is such a thing). I have no idea though how the Wikidata community works.
   3. Periodically rerunning the script to check for changes is a reasonable approach, except that there isn't really an easy way of finding the pronunciations that have changed in Wiktionary in the last month, so you might end up having to upload all of them each time (which isn't a huge deal since there are only around 100k of them).

   Benwing2 (talk) 03:09, 4 October 2025 (UTC)

1. Infopedia has that famous 'Todos os direitos reservados' line at the bottom of each page, which I interpret as 'We are not sure about our license but we prohibit everything we do not like'. Maybe I am wrong here, but I feel really uncomfortable about copying their data as CC0 therefore.
2. Pywikibot looks good, thank you. I am definitely more fluent in Python than in Lua.
3. I am aware of a pair of Wikidata user chats, will reach folks there when/if I get that far.
4. My plan wasn't to mirror Wiktionary to Wikidata. I think I can cycle over Wikidata lexemes, generate an IPA and check if it differs from what is already in Wikidata. So, no dependency on Wiktionary at all. I can miss something, of course, but it sounds possible.
5. May I ask two things?
   1. How can I run your code locally to convert e.g. gato to ˈɡa.tu?
   2. How do you validate such amount of data?

Dunkaist (talk) 04:26, 4 October 2025 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Dunkaist I understand your concerns about Infopedia; just pointing out that it is a good reference for EP pronunciations. The pronunciations themselves, as factual data, can't be copyrighted, which is why I have not felt any compunctions about using their pronunciations as references for the pronunciations we generate. However, there might be issues with mass copying; I dunno. As for running the code, it should be possible to set up a Scribunto Lua interpreter locally and stick all the needed modules in; I know for example that @Erutuon has done this and maybe also @This, that and the other. However, I have never done that; instead I make a call to Wiktionary, telling it to run the module. This can be done like this in Pywikibot:

site.expand_text("{{#invoke:pt-pronunc|IPA_json|SPELLING|style=STYLE}}")

where site is the Pywikibot connection to the site (in this case, English Wiktionary), SPELLING is the respelling and STYLE for you might be "pt". The only tricky thing is you need the respelling, which means you need to use Pywikibot to pull down the page, find the occurrences of {{pt-IPA}} and extract the respellings. You can use mwparserfromhell to parse MediaWiki syntax. I would do it like this:

def parse_text(text):
  return mwparserfromhell.parser.Parser().parse(text, skip_style_tags=True)

and then after pulling down the text of the page, you can do something like this:

parsed = parse_text(pagetext)
for t in parsed.filter_templates():
  if str(t.name).strip() == "pt-IPA":
    ...

The tricky thing is what you do if there are multiple possible pronunciations; you'll have to decide that.

As for validating the data, that is not easy to do and I don't really have an answer for it. I suppose you could download the pronunciations from Infopedia or the Cambridge dictionary you mentioned (thank you for that), and compare them against Wiktionary pronunciations to see if there are discrepancies; but that assumes those sites are accurate, which is not always the case, and in practice there are a zillion complications that arise.

As you can see from this, you do have to take a dependency on Wiktionary, because the spelling of Portuguese words does not in general provide enough information to correctly generate the IPA. For example, the pronunciation of the mid vowels e and o is not generally indicated in the spelling (cf. forma "form" with /ɔ/ vs. forma "baking mold" with /o/); nor are the myriad cases where unstressed vowels are unreduced in European Portuguese. (Compare, for example, pregar "to nail" [pɾɨˈɣaɾ] in EP vs. pregar "to preach" [pɾɛˈɣaɾ] in EP, whereas both are pronounced the same in BP.) And these are just two such issues; there are at least a dozen more. Benwing2 (talk) 05:01, 4 October 2025 (UTC)

Oh, I forgot to mention the Brazilian IPA online: Cambridge dictionary, an older rule-based generator. Dunkaist (talk) 04:38, 4 October 2025 (UTC)

@Dunkaist: The content of a telephone book is not a case of copyright either but of database right unless there are presentations according to rational principles more intricate than the alphabet making it a database work. As was definitely decided in the century when they were widespread: BGH, Urt. v. 06.05.1999 - I ZR 199/96 = GRUR 1999, 923. Feist Publications, Inc. v. Rural Telephone Service Co. (1991).

We can hardly agree, given that we aren't everyone, who releases here, but given this uncoordinated effort all is ex lege not protected = CC0, as described. Fay Freak (talk) 00:39, 4 October 2025 (UTC)

Replacement of quotation templates (October 2025)

Hi, could you please carry out the following quotation template replacements?

• If |page= or |pages= used, {{RQ:Milton Of Reformation}} → {{RQ:Milton Of Reformation|year=1916}}
• If |page= or |pages= used, {{RQ:Montgomery Anne of Green Gables}} → {{RQ:Montgomery Anne of Green Gables|year=1909}}
• If |page= or |pages= used, {{RQ:Nabokov Ada}} → {{RQ:Nabokov Ada|year=1970}} (note to self: wick already using the 1970 version)

Thank you. — Sgconlaw (talk) 21:01, 4 October 2025 (UTC)

@Sgconlaw Should be done. There were only about ~10-15 pages per template needing fixing. Benwing2 (talk) 22:13, 4 October 2025 (UTC)

Thanks! — Sgconlaw (talk) 22:14, 4 October 2025 (UTC)

It's still October, so I thought I would list a few more requests here:

• {{RQ:Hawthorne House Seven Gables}} → {{RQ:Hawthorne Seven Gables}}.
• {{RQ:Joseph Hall The Balm of Gilead}} → {{RQ:Hall Balme of Gilead}}.
• {{RQ:Joseph Hall Satires}} → {{RQ:Hall Virgidemiarum}} (the same work, and not by Ayliffe).
• If |page= or |pages= used, {{RQ:Nashe Strange Newes}} → {{RQ:Nashe Strange Newes|year=1867}} (note to self: entreat and intreat already using the 1st edition).
• If |page= or |pages= used, {{RQ:Virgil Stanyhurst Aeneid}} → {{RQ:Virgil Stanyhurst Aeneid|year=1836}} (note to self: revive already using the 1st edition).

Thank you. — Sgconlaw (talk) 21:28, 17 October 2025 (UTC)

@Sgconlaw Should be done. Note that the bot skipped potluck when adding year=1867 to {{RQ:Nashe Strange Newes}} even though it has a page number, because it uses 2= for the page number; you might want to fix this by hand. Benwing2 (talk) 20:13, 23 October 2025 (UTC)

Thanks! — Sgconlaw (talk) 22:26, 23 October 2025 (UTC)

Some Basque-related bot requests

Hello, I have a few bot requests involving Basque templates. They're simple template substitutions:

• {{eu-decl-anim}} to {{eu-ndecl|an}}, {{eu-decl-both}} to {{eu-ndecl|both}}, {{eu-decl-inanim}} to {{eu-ndecl|in}}
• {{eu-decl-proper-anim}} to {{eu-ndecl|an.pr}}, {{eu-decl-proper-both}} to {{eu-ndecl|both.pr}}, {{eu-decl-proper-inanim}} to {{eu-ndecl|in.pr}}
• {{eu-decl-adj}} to {{eu-adecl}}, {{eu-decl-num}} to {{eu-numdecl}}
• {{head|eu|verb form}} to {{eu-verb form}}
• {{R:EH}} to {{R:eu:EH}}, {{R:OEH}} to {{R:eu:OEH}}, {{R:EDB}} to {{R:eu:EDB}}

I've already handled the cases with parameters manually. Thanks in advance! Santi2222 (talk) 23:39, 4 October 2025 (UTC)

@Santi2222 Should be all done. I think we should delete the old templates; if you're OK with that I'll go ahead and do that. Benwing2 (talk) 06:51, 5 October 2025 (UTC)

Perfect, many thanks! I agree that the old templates should be deleted, no issues on my part. Santi2222 (talk) 09:51, 5 October 2025 (UTC)

Old templates deleted. Benwing2 (talk) 19:53, 5 October 2025 (UTC)

Template:be-ndecl

Just leaving a note that I neutralized the module error here so you don't have to deal with it right away. My best guess is that it was caused by putting a nonexistent spec as the default in Line 2332 of Module:be-noun]. Chuck Entz (talk) 03:43, 11 October 2025 (UTC)

@Chuck Entz Thanks. Your guess was exactly right; when cleaning up the code, I accidentally copied the Ukrainian example over the old Belarusian example. The indicator io is Ukrainian-specific (and handles the і/о alternation in that language, which doesn't exist in Belarusian). Benwing2 (talk) 03:51, 11 October 2025 (UTC)

WingerBot error

For some reason WingerBot removed a curly bracket here: https://en.wiktionary.org/w/index.php?title=superchat&diff=87429423&oldid=87417030 BirchTainer (talk) 10:24, 13 October 2025 (UTC)

Thanks. Most of this run was done by me manually typing offline in a text file, so it's very possible I just forgot to type the second right brace. Benwing2 (talk) 18:05, 13 October 2025 (UTC)

Nomination

Hi. You've been around a bit, and save WT's asses on a daily basis. I'd like to nominate you as The Official Poobah, Supremely High Internet Titleholder. Please accept hastily Vealhurl (talk) 20:22, 14 October 2025 (UTC)

About the Persian page adab ادب

The word adab on the Persian page suggests that the Arabic word adab is a borrowing from the Persian dab, ultimately from Sumerian. However, there is no evidence that the Arabic word adab comes from the Persian word dab, which has no proven existence. John daar (talk) 21:09, 15 October 2025 (UTC)

@John daar I have no idea why you're posting this on my talk page. It sounds like you should post it to WT:Etymology scriptorium. Benwing2 (talk) 05:27, 17 October 2025 (UTC)

Category:Spanish combined forms

Hi Ben! I've pared Category:Spanish combined forms down to just 417 typos, invalid forms, or unverified obsolete forms all added by WF many years ago, who seems to have lost interest in them. I've verified that none of the pages include other L2s or contain any other useful information so I think it's probably safe to mass-delete everything in there.

On a separate note, Category:Pages_using_deprecated_templates lists nothing in the mainspace. Is that good enough to remove the #if:{{{lang wrapper on templates like {{inflection of}}, {{audio}}, {{IPA}}, {{plural of}}, {{hyphenation}}, {{rhymes}}? JeffDoozan (talk) 18:46, 21 October 2025 (UTC)

@JeffDoozan Thank you for doing that. I'll delete everything in CAT:Spanish combined forms. As for the #if:{{{lang stuff, it's for making the page history somewhat viewable, not for current pages. That's why for example we have {{infl}} (an older version of {{head}}), which isn't used anywhere. Benwing2 (talk) 18:53, 21 October 2025 (UTC)

@JeffDoozan All deleted, after making sure there were no entries from other languages on those pages. I also deleted {{es-compound of}} and Category:Spanish combined forms and moved Category:Spanish verbs with lexical clitics out of the former category. Benwing2 (talk) 01:14, 23 October 2025 (UTC)

Oops, you already checked for other L2's :) ... Benwing2 (talk) 01:14, 23 October 2025 (UTC)

Excellent, thank you. The handful of mainspace pages in Category:es-verb_form_of_with_old_params can be purged, too, which should complete the conversion of everything to the new style {{es-verb form of}}. JeffDoozan (talk) 13:26, 23 October 2025 (UTC)

@JeffDoozan should be all done; let me know if I missed anything. Benwing2 (talk) 19:57, 23 October 2025 (UTC)

• Yes, thanks. It was a big boring job. Vealhurl (talk) 21:20, 29 October 2025 (UTC)

Serbo-Croatian demonyms

Hi there. My mother tongue is Serbo-Croatian. I've noticed that demonyms (male and female) in general are designated as Nouns, and that the Serbo-Croatian demonyms written in Latin are designated as Nouns while the same ones written in Cyrillic are designated as Proper nouns. A few months ago I added a few Serbo-Croatian demonyms and I designated them all as Proper nouns, and now I've noticed that you changed the ones written in Latin to Nouns. Can you tell me why is that? Why did you make a change to the words in Latin and not to the ones in Cyrillic, and why did you make a change at all? I understand that demonyms can have a plural form, but I am pretty sure that in Serbo-Croatian every noun that begins with a capital letter is a proper noun. Maybe demonyms should be written in lowercase letters, like in other languages (including Slavic ones, like Russian), but until that happens, they are regarded as proper nouns. So, how should I designate them here? Galicijac (talk) 14:32, 29 October 2025 (UTC)

@Galicijac It is good to have some more native Serbo-Croatian speakers around; we don't have too many. As for proper nouns, generally they are nouns that refer to unique or mostly unique entities. This includes toponyms (place names), given names, surnames and language names, but not demonyms, because they refer to an entire group of people, just like professions like "teacher" or "biologist" or nouns of characteristic like "gambler" or "miser". Some languages capitalize demonyms, some don't; some languages (e.g. French) capitalize demonyms when they are nouns but not as adjectives. But that should not change how they're classified, because "proper noun" is a semantic concept, not an orthographic concept. If you take a look at French demonyms, for example, you'll see that Américain and Français are indicated as common nouns even though they're capitalized. It is a misconception that all nouns that are capitalized are proper nouns; English, for example, capitalizes a lot of common nouns (e.g. Monday, September, American), and German capitalizes all nouns, but that does not affect their status as proper nouns (otherwise every noun in German would be a proper noun, for example). The reason the change was not propagated to the Cyrillic entries is that it's been too much work for me as I've been cleaning up Serbo-Croatian entries to keep the Cyrillic ones in sync. My plan in the slightly longer term is to implement a template {{sh-tcl}} that copies a Latin entry to a Cyrillic one or vice-versa, transliterating the Serbo-Croatian terms but keeping all other terms as-is (and probably leaving quotations alone as well, since the quotation was written in a particular script and we should respect that). Your message has actually prompted me to think more seriously about what is needed to implement this and I'll probably do it soon, but until then the Cyrillic entries will be out of sync.

BTW I am mostly done creating a module that declines Serbo-Croatian adjectives properly (i.e. including the tones); I have been fortunate to have the input of a couple of very knowledgeable Serbo-Croatian speakers on Discord to help me with this as well as to help me sort out what are the best resources for this kind of information (and which resources are reliable). I have been working on other things recently but hope to get back to finishing the module soon. Benwing2 (talk) 21:41, 29 October 2025 (UTC)

wp|la: issue: Links for Latin wikipedia should be macron-stripped, but aren't

E.g. the link at sedecim should go to https://la.wikipedia.org/wiki/Sedecim, but currently goes to the nonexistent https://la.wikipedia.org/wiki/S%C4%93decim. Urszag (talk) 18:48, 1 November 2025 (UTC)

@Urszag I don't think this ever worked, but it should be fixable. Benwing2 (talk) 19:30, 1 November 2025 (UTC)

Module:cs-sk-headword

Hello, I want to apologise for not writing my reasons into the edit summary on the page Template:zlw-ocs-adj. I did not realise it would cause trouble later finding out why I did not like your module. Alas, you do me wrong by assuming I did not tell you, for I have tried to contact you multiple times on Wiktionary, but you never replied. I had also tried previously to repair the part of the module that does not make short superlative forms automatically from the input short comparative forms, yet I had been unsuccessful.

On the other hand, as I had been active on Wiktionary for a couple of years, I would have expected you to contact me while reverting my edits or at least give me some notification that something had changed. I do not deserve to be treated like a vandal.

Thank you in advance for a quick upgrade to your module that will make Old Czech short superlatives automatically from the input short comparatives. Zhnka (talk) 07:43, 2 November 2025 (UTC)

@Zhnka In truth you did not ping me when you replaced the module call with hacky Wikicode; I only found out months later. That sort of replacement is never the right thing to do.

In any event can you let me know what you need done? I no longer have the context on what the issue is. Benwing2 (talk) 07:53, 2 November 2025 (UTC)

I have told you everything right now. You see the page mladý. If you input the comparative mlazší, it automatically generates najmlazší as a superlative. If you input the short comparative mlazí, the short superlative is not generated. Zhnka (talk) 07:57, 2 November 2025 (UTC)

Is that the only issue? That you need a short superlative auto-generated using naj- for Old Czech? What about Old Slovak, does it have short comparatives and if so does it need the same treatment? Benwing2 (talk) 08:02, 2 November 2025 (UTC)

I do not know much about Old Slovak. Overall, Old Czech is here defined as anything Czech until around 1500 with many attestations and big literature works, whereas Old Slovak is here the pre-standard period of Slovak (anything until the 18th century). I do not know whether the short forms are attested in Old Slovak, it would make sense for the earliest language, but I have not found many attestations of its earliest form in the dictionary {R:sk:HSSJ} and overall the earliest attestations are Slovakized Czech. Zhnka (talk) 08:45, 2 November 2025 (UTC)

Unfixed bot error

Currently, up shit's creek without a paddle might arguably belong in Category:English autological terms. If I fix the missing curly bracket (which probably explains why you missed it). that exposes a module error- but I don't know the new syntax well enough to fix it easily. Thanks! Chuck Entz (talk) 23:14, 2 November 2025 (UTC)

@Chuck Entz Thanks. There are currently issues with 's in the middle of the left side of a replacement pattern, which is why the module error occurs. I haven't figured out the best way to fix this yet, so I just replaced it with the old syntax. Benwing2 (talk) 23:26, 2 November 2025 (UTC)

Request

Hey, would you mind checking this topic? Thanks, Yacàwotçã (talk) Yacàwotçã (talk) 15:47, 3 November 2025 (UTC)

optional {wikipedia} syntax

All: I am a little surprised by (for example) a recent Wingerbot edit:
{{wikipedia|metalorganics}} → {{wp|+s}}
(Only a minority of the series of edits makes this specific kind of mod.)

The first part of the edit comment is: "convert {{wikipedia}} to new syntax; ...":
Why would anyone start a mass-conversion to an optional feature?
Did anyone call for this? Did anyone discuss it? (If so, where is the link?) Did anyone approve the operation?

The second part of the edit comment is: "... convert {wikipedia} to {wp}":
Why would anyone mass-replace a template with one that redirects to it?
Why would anyone mass-replace a spelled-out template to a terse, cryptic redirect? To me it seems wrong. pointless. obfuscatory. without benefit. with drawback. a step backward. bloating storage with hundreds of edits.
Did anyone call for this, discuss it, or approve it? Does anyone appreciate it? (Maybe there's a fetish here for terse, cryptic abbrevs.) Is anyone else appalled by it? -A876 (talk) 19:00, 3 November 2025 (UTC)

(This specific Wikipedia link needed more work, some of which could have been automated.

The final "s" is not typical (Wikipedia has mostly singular titles), and not necessary ("Metalorganic" and "Metalorganics" both redirect to "Metal-organic compound").

Wikipedia also has a redirect at "Organometallic compound" (mentioned in the entry) to "Organometallic chemistry". (A possibly undesirable split or duplication exists here and there that suggests a need for links to both.)

So I changed this one to {{wikipedia|metal-organic compound,organometallic compound}}.) -A876 (talk) 19:00, 3 November 2025 (UTC)

@A876 Since you are basically a new editor with only 146 edits, and don't seem to know how Wiktionary works, I'm surprised you're coming in guns blazing making strong (and often misguided) criticisms of my work. Please see Wiktionary:Beer_parlour/2025/October#redoing_syntax_of_Template:wikipedia_and_Template:slim-wikipedia where this issue was discussed in detail and given the thumbs up before I did anything. FYI, the Beer Parlour is the standard place to discuss policy changes and major syntax changes. Benwing2 (talk) 20:33, 3 November 2025 (UTC)

Double-checking after "manually assisted" bot runs

I spent what seemed like an hour or two today changing "sclb" to "sclang" and "nolb" to "nolang" in entries where your bot had changed some, but not all of the occurrances in each entry. Then I did a simple search for insource:"sclb" and found 7 more. A similar search for "nolb" turned up a couple, but its use in other templates made it hard to tell whether I got them all. There were even a few instances of "lb" in {{desc}} that you missed (perhaps because they're under a "Derived terms" header instead of "Descendants"). Given the problems with the manual modification of text files, it might be a good idea to come up with more ways to check after bot runs. The insource:"sclb" method requires waiting at least 20 minutes after a bot run for the search engine's info to update, but it's very effective when you're replacing all of something

Then there are the occurrences of {{desc|xx|*foo ~ bar}} that are still popping up as Reconstruction links missing the "*"- I remember fixing one of them early on by changing it to {{desc|xx|*foo ~ *bar}}, but that doesn't work anymore. I could also do without the errors from "m, f" vs. "m,f", but it's not that big of a deal. Thanks for reading my grumbles all the way through... Chuck Entz (talk) 00:23, 9 November 2025 (UTC)

Sorry Chuck. I didn't realize you had fixed all these errors. I would fix them if I saw them but you seem to be getting to them too quick. I can fix them relatively quickly using my offline editing tricks. The reason for not all the sclb/nolb things getting fixed is I based my search off of the Nov 1 dump, and some people have added stuff since then. I didn't think about the insource: trick, I'll remember that for next time. The occurrences of {{desc|xx|*foo ~ bar}} should be fixable by adding a * before the bar; if that isn't working can you give me an example? As for the "m, f" stuff I suppose I could be more lenient about this, it's more that everywhere else the comma is not supposed to be followed by a space and allowing a space here makes it impossible in the future to have items with embedded commas followed by spaces (not that they're likely to ever occur ...). Are there any errors you are still seeing? Benwing2 (talk) 00:31, 9 November 2025 (UTC)

@Chuck Entz I am looking through the dump for places that still might have missing * after a tilde, and I found this: {{desc|mkh-kat-pro|*braak ~ raak|alt=*braak ~ *raak}}. Any idea what this means? Benwing2 (talk) 00:43, 9 November 2025 (UTC)

FYI sometimes the ~ seems to be part of the lemma; in that case, put brackets around the lemma, like this: {{desctree|jpx-ryu-pro|*[[kura ~ kuCa]]|t=child}}. Benwing2 (talk) 00:58, 9 November 2025 (UTC)

I tried to catch all the remaining cases of {{desc|xx|*foo ~ bar}} and m, f etc. (at least, all the cases present on Oct 31). Benwing2 (talk) 01:17, 9 November 2025 (UTC)

There's Reconstruction:Proto-Eastern Malayo-Polynesian/pat, where the current revision is me giving up and making two parameters out of it. Looking at it again, it seems that the problem there is that this is {{desctree}}, and it has to link to Reconstruction:Proto-Halmahera-Cenderawasih/pat ~ pati. The bracket method works in preview: both {{desctree|poz-hce-pro|[[*pat ~ pati]]}} and {{desctree|poz-hce-pro|*[[pat ~ pati]]}}. I've now fixed it. Thanks! Chuck Entz (talk) 01:51, 9 November 2025 (UTC)

The parameters |sclang= and |nolang= are both less intuitive and longer to type than |sclb= and |nolb=. I think there should have been a discussion started prior to implementing this change, involving users, such as myself, who rely on these parameters on a near daily basis. --{{victar|talk}} 18:06, 10 November 2025 (UTC)

Replacement of quotation templates (November 2025)

Hi, some quotation template replacements for when you have time:

• {{RQ:en:Permanent Record}} → {{RQ:Snowden Permanent Record}}.
• If |page= or |pages= used, {{RQ:Stoker Dracula}} → {{RQ:Stoker Dracula|year=1897}}.
• If |page= or |pages= used, {{RQ:Tillotson Works}} → {{RQ:Tillotson Works|edition=8th}} (note to self: interlard already using the 1st edition).

Thank you. — Sgconlaw (talk) 16:26, 9 November 2025 (UTC)

@Sgconlaw Apologies for the delay, should be done. You will have to change interlard manually. Benwing2 (talk) 03:32, 18 November 2025 (UTC)

No problem. Thanks! — Sgconlaw (talk) 10:18, 18 November 2025 (UTC)

Links to "Unsupported titles/'lowbar' in Wikipedia templates

Wiktionary:Todo/Lists/Broken links to Wikipedia has quite a few of these, which weren't there a couple of weeks ago. The interwiki linking code obviously shouldn't be doing that kind of thing. Thanks! Chuck Entz (talk) 06:14, 10 November 2025 (UTC)

Grrrr, should be fixed. Module:links is a real piece of junk and not well documented; needs rewriting. Benwing2 (talk) 06:24, 10 November 2025 (UTC)

Sillabyfication in Module:es-pronunc

hi, the current transcription for adyacente is [a.ð̞ʝaˈsẽn̪.t̪e], where the correct would be [að̞.ʝaˈsẽn̪.t̪e]. Afetr looking the module, I think is smarter to sillabify all the word first and then convert to IPA, and not viceversa; but that would require redeveloping almost all from scratch. A quick patch (which I also recommend anyway) would be removing the syllable dividers (except stress mark) in transcriptions, in es.wikt we did it by consensus. Leave it up to you, regards. Tmagc (talk) 15:58, 16 November 2025 (UTC)

I like having syllable divisions shown; this shouldn't be too hard to fix if it's only this case. Benwing2 (talk) 20:34, 16 November 2025 (UTC)

Caught an old mistake

This edit by your bot managed to break a template for almost the whole year. Oops. -Brainulator9 (TALK) 03:10, 18 November 2025 (UTC)

See my reply on the talk page of that template, it was an offline editing mistake rather than a bot mistake. I'm surprised someone didn't catch it earlier ... Benwing2 (talk) Benwing2 (talk) 03:11, 18 November 2025 (UTC)

Yeah, the system didn't tell me that my edit pinged you, which made me think it didn't happen. Sorry about that. -Brainulator9 (TALK) 03:12, 18 November 2025 (UTC)

Template:RQ:apc:Fleisch:1974

It looks like Module:quote and/or Module:parse utilities is counting the same prefix in one of the parameters twice. It's associated with w:, so it's probably a side-effect of your recent fix. Chuck Entz (talk) 04:35, 19 November 2025 (UTC)

Shit. Thanks for the heads-up. Yes, this is related to my recent fix; I'll have to be a bit more clever in how I do it. Benwing2 (talk) 04:47, 19 November 2025 (UTC)

Septem and Semitic origin hypothesis

I would appreciate it if you could attach the Semitic hypothesis regarding the origin of the Indo-European word septem. I urge you to consider this matter. Yeedan 12 (talk) 00:40, 21 November 2025 (UTC)

I have no idea what you're talking about, and it's unlikely the PIE word for seven originates in Semitic. Benwing2 (talk) 00:44, 21 November 2025 (UTC)

Haven't you heard about borrowing? Yeedan 12 (talk) 00:48, 21 November 2025 (UTC)

I mean this word septḿ̥ Yeedan 12 (talk) 00:50, 21 November 2025 (UTC)

I believe they're talking about this YouTube video. Perhaps unsurprisingly I can't find any academic sources corroborating it. viiii (talk) 02:43, 23 November 2025 (UTC)

WP templates and punctuation in titles

Lately I've been finding and fixing (diff, diff) entries where the template was filtering out question marks, even though the target Wikipedia page has that character in the title. It's true that the URL itself has it escaped as %3F, but it's still part of the title.

Oddly enough, the template has no problem linking to Live ?!*@ Like a Suicide in spite of the %3F in the URL. Apparently, Wikipedia has articles for all the punctuation marks, with the wikisyntax ones redirected to another spelling {{w|/}} works, though: /, and {{?}}/{{w|%3F}} go to the same place: ? / %3F. I suspect the module is applying Wiktionary assumptions about entry names to the Wikipedia article names.

Given that there aren't many of these and they're not hard to fix, I'm just bringing it to your attention rather than asking you to "¡Fix It Right Now!". Thanks! Chuck Entz (talk) 03:19, 22 November 2025 (UTC)

@Chuck Entz Hi Chuck. I think your comment about "the module is applying Wiktionary assumptions about entry names to the Wikipedia article names" is spot-on. Originally it didn't do anything like this, but then @Urszag asked me to do this explicitly, so that diacritics could be supplied in Latin Wikipedia links and it would link to the diacritic-less entries. I had some vague misgivings about this but went ahead and implemented it, and I'm guessing what you're seeing is a side effect. You can see the diff for this in Module:interproject on Nov 1, and I bet if you use the "view page with this template" feature and view a page with a ? in it using the commit directly before it, it will not have this issue. I'm not quite sure how to fix this; it may be a case of trying to have your cake and eat it too, although if it's only happening with question marks, I may be able to work around that by replacing a ? with %3F before calling makeEntryName() (which does the diacritic stripping). Benwing2 (talk) 03:27, 22 November 2025 (UTC)

Indeed, I just checked and this is correct. I will see if I can fix it by converting ? to %3F first. Benwing2 (talk) 03:29, 22 November 2025 (UTC)

OK, I made that change and it seems to work. Benwing2 (talk) 03:33, 22 November 2025 (UTC)

I won't especially mind if you end up needing to un-implement this: it's not a feature I regularly use, I just noticed that the link at sedecim had become broken. Rather than stripping diacritics, it would be simple enough to just use alternative formatting like {{wp|la:[[sedecim]]}} instead.--Urszag (talk) 21:44, 22 November 2025 (UTC)

@Urszag So far it looks like I was able to fix the issue without having to remove the diacritic stripping. We'll see if any more issues come up. Benwing2 (talk) 21:50, 22 November 2025 (UTC)

Found and fixed one link to a WP entry with a final exclamation mark/%21 (dun dun dun), but that's it so far. Chuck Entz (talk) 23:27, 22 November 2025 (UTC)

Fixed. Hopefully this won't end up being a game of whack-a-mole. Benwing2 (talk) 23:32, 22 November 2025 (UTC)

Character in mathematics

In the article for character, I attempted to make the definition more mathematically precise, changing it from

An assignment of complex numbers to each element of a group, in particular a finite abelian group.

into

A group homomorphism into the group of units of a field.

You reverted this revision, may I ask why? viiii (talk) 02:38, 23 November 2025 (UTC)

There were two reasons: (1) you introduced an error, putting the word "math" in the middle of the param "passage", but (2) the definition needs to make sense to someone who is familiar with math in general but not deeply versed in advanced algebra or group theory. I am such a person, and your revised definition is incomprehensible to me. I have no idea if it's mathematically correct but in any case it's far too concise and abstract for a general-purpose dictionary such as Wiktionary. Benwing2 (talk) 02:42, 23 November 2025 (UTC)

A character is not some random function $G\to \mathbb {C}$, it's specifically a homomorphism of groups. And the codomain is not even necessarily the complex numbers. My definition should be understandable to anyone who's taken an introductory course in modern algebra, which seems fine to me considering that group characters are a relatively advanced topic within that same branch of mathematics. If Wiktionary is not the place for these knife-sharp technical definitions, then I understand. viiii (talk) 02:48, 23 November 2025 (UTC)

Like I said, it needs to be comprehensible to the educated layman. You cannot assume the reader has taken an introductory course in modern algebra; very few people have done that compared with the readership of Wiktionary. You need at the very least to add context to your definition. If you look at the Wikipedia definition, they mention that it usually involves a field of complex numbers, and that it forms an abelian group under pointwise multiplication. You took out both of these facts, which makes the definition much less accessible. You say "the codomain is not even necessarily the complex numbers" but evidently it usually is, so this should be mentioned. And since very few people know what a homomorphism of groups is, you should put a parenthetical definition indicating that it is a type of function from a group to a field. You have to think not from your own perspective of someone who is intimately familiar with the subject matter, but someone who has only a vague understanding of anything but basic mathematical concepts but is nonetheless smart enough to figure out what's going on if given some context. Benwing2 (talk) 02:59, 23 November 2025 (UTC)

Thanks for the feedback. Perhaps the solution is to have the lay definition, followed by the technical definition? "A character is X. More precisely, a character is Y." viiii (talk) 03:01, 23 November 2025 (UTC)

Something like that will work, yes. Benwing2 (talk) 03:02, 23 November 2025 (UTC)

¡ay, caramba!- Here's another one.

The only way I could make this work in the {{pedia}} template was adding an &nbsp; in front to keep the inverted exclamation mark from being the first character. Replacing the inverted exclamation mark with %A1 causes the module to interpret the parameter as something else and throw an error when whatever it is doesn't work. Chuck Entz (talk) 02:15, 24 November 2025 (UTC)

OK, I'll take a look; maybe there is a more principled way of handling this. Benwing2 (talk) 02:33, 24 November 2025 (UTC)

@Chuck Entz the entry-name-conversion code actually deletes a lot of different special characters and does various transformations besides just stripping diacritics, so I suspect we'll have to turn it off. In any case we already have {{lw}} which is similar to {{w}} but takes its parameters like {{l}} and explicitly does entry-name conversion. We could just tell users to use that if they want diacritic stripping (and maybe harmonize its syntax with that of {{w}}). Benwing2 (talk) 23:25, 24 November 2025 (UTC)

requests

Hi! I could get no response on other fora so I take my pleas here:

In Module:Quotations, is it possible to add multiple rlFormat per works? if so, how? if not, how? Asking for plays for which I'd like to have both linkable lines per scenes and linkable lines from first verse to last one of the play.

Could you add the prepositive label to Module:labels/data (as in ready-(made)) with link to a new corresponding entry in the glossary, we already have {{lb||postpositive}}.

Thanks! Saumache (talk) 08:37, 25 November 2025 (UTC)

Hi, I'll add the prepositive label. Module:Quotations is sort of a no-man's land, no one understands it and the consensus is it needs to be rewritten. Benwing2 (talk) 08:48, 25 November 2025 (UTC)

I see, but in the meantime, no coder could take a look at it? I'm not sure how to format the Wikisource texts I'd want to be finished because of it. Saumache (talk) 09:01, 25 November 2025 (UTC)

I'll take a look but I can't promise anything because the code is rather a mess. Benwing2 (talk) 23:44, 27 November 2025 (UTC)

Thanks! I'll be so bold as to add some more desiderata, I'd like there to be labels support in {{la-verb}} (as we have in {{en-verb|<,,got,got[UK]:gotten[US]>}}) and notes support for Latin inflection templates (as in Italian essere). See sorbeo for an example of its future usefulness. Saumache (talk) 13:42, 28 November 2025 (UTC)

Category:Pages using the JsonConfig extension

Hi. Your bot recreated this category, but it is deprecated and thus shouldn't be. Could you please perform the changes necessary to prevent the bot from doing so again? Thanks. NguoiDungKhongDinhDanh (talk) 09:54, 27 November 2025 (UTC)

@NguoiDungKhongDinhDanh Can you explain why this page is deprecated and what to do about the remaining 13 pages in the category? Benwing2 (talk) 23:48, 27 November 2025 (UTC)

See T378352 for more information. The pages are actually no longer in that category; they need to be null-edited so that MediaWiki updates them accordingly. NguoiDungKhongDinhDanh (talk) 23:52, 27 November 2025 (UTC)

OK thanks, I null-edited those pages and deleted the category. Benwing2 (talk) 00:11, 28 November 2025 (UTC)

Bug with la-IPA

Hello, I noticed that the la-IPA template is now not generating the "Ecclesiastical" pronunciation label when the classical=0 parameter is set. An example: ieiūnus. Here, the in-line references only pertain to the Classical Latin pronunciation, so the pronunciation below it is meant to be the Ecclesiastical pronunciation: the bug is that both pronunciations are now labeled as "(Classical Latin)". Is it possible this was caused by something in your recent edit to make the module be able to indent pronunciations that have an annotation? I didn't notice it before then, but I'm not sure what's causing it; I'd appreciate it if you could take a look. Thanks! Urszag (talk) 01:30, 5 December 2025 (UTC)

Yes, this is likely due to my recent edit; I will take a look. Benwing2 (talk) 01:33, 5 December 2025 (UTC)

Module:cs-sk-headword – verb templates

Hello again,

I have a proposal for Module:cs-sk-headword, which I could implement myself. I have wondered for a long time what to do with those verb pairs like nést/nosit/nosívat, běžet/běhat/běhávat, jít/chodit/chodívat, být/bývat. In most places they are listed as abstract/concrete verbs. This site explains the modern classification of Czech verbs (http://nase-rec.ujc.cas.cz/archiv.php?art=8053) and its similar to what I see in the Module:zlw-lch-headword. Whether we call it concrete/abstract or determinate/indeterminate verbs, doesn't matter, so I would go with determinate/indeterminate, as that seems to be the prevalent classification in the West Slavic languages. And then, of course, the frequentative verbs. So we should add basicaly everything that the Module:zlw-lch-headword has (aspects "impf-det", "impf-indet", "impf-freq" and specs {"det", "imperfective determinate"}, {"indet", "indeterminate"}, {"freq", "frequentative"}).

Above that, there seem to be something like pairs of determinate/indeterminate perfective verbs donést/donosit (like nést/nosit) and the above mentioned page also talks about something like frequentative perfective verbs zamknout/pozamykat, but these are more of semantic differences and overall irregular. I think the aspects from the Module:zlw-lch-headword are enough. Do you agree? Should I add it or will you? Zhnka (talk) 08:38, 7 December 2025 (UTC)

@Benwing2, there's really not something not to agree with. It works the same way as in Polish and I have included a source confirming it. I just wanted to let you know before I add it. Zhnka (talk) 15:58, 8 December 2025 (UTC)

OK, go ahead. There was a recent discussion about the terms concrete/abstract and such that concluded we should probably use "unidirectional/multidirectional" (it's in the Beer Parlour I think). Benwing2 (talk) 19:14, 8 December 2025 (UTC)

@Benwing, thank you for your answer. I am sorry, I don't know how to add it correctly. I tried to add the whole verb section, all needed modules are there, but it still gives errors that the val argument is expected as a table, but is given as a string. The only place where the two modules differ is, I think, the main entry point, but I don't want to mess with that. And overall, it would be best if the "a=impf" was replaced with "impf" in the headwords as it already is in the Polish verb headwords (similarly as has been done with nouns which don't use "g=" anymore), but that needs to be done by bot. So, I'm leaving it up to you. Zhnka (talk) 21:04, 8 December 2025 (UTC)

OK, no problem, I will get to this within 2 days or so. Benwing2 (talk) 21:07, 8 December 2025 (UTC)

Entries with colons

Just to let you know that I fixed a module error at Swedish mms:ad caused by the module mistaking the last parameter of {{past participle of|sv|mms:a}} for an inline link with "mms" as a language code. The only reason there weren't more is that most of the Swedish abbreviations start with things that are also valid language codes, like {{past participle of|sv|sms:a}}, which displays as "past participle of Skolt Sami a", or are too long to be language codes. See Category:Swedish terms spelled with : ... Chuck Entz (talk) 09:16, 8 December 2025 (UTC)

@Chuck Entz Thanks. Not really sure how to handle that other than putting brackets around all such terms or preceding them with a colon. Benwing2 (talk) 19:16, 8 December 2025 (UTC)

The other possibility is to disable the language prefix functionality for Swedish in particular. This could make sense because, depending on the spelling convention chosen, we might have colons in the middle of O'odham words as well (see the discussion in the October? Beer Parlour about this). Benwing2 (talk) 20:06, 8 December 2025 (UTC)