User talk:Benwing2/2020-2021

Archive

 * 2012-2019

Atmos
1. Plural of Atmo 2. (Sound production) clipping of ??? atmosphere

What does this mean: clipping of? Sound production? Ætherdog (talk) 12:39, 2 January 2020 (UTC)
 * It seems to mean that the term is used specifically in sound production. Benwing2 (talk) 15:18, 2 January 2020 (UTC)

Script to add translation tables
Hi. I've been working quite a bit with French idioms and their English translations lately, and I'm often adding translation tables to English entries, before. Do you think you could whip up a script that would allow me to do the first step (i.e. add an empty translation table) in one click? Canonicalization (talk) 10:32, 13 January 2020 (UTC)
 * I can definitely write a script that can be run to add empty translation tables to a list of French terms. The only thing is that I'd have to run the script because it requires a bot account and various things set up on my computer. The idea would be that you'd give me a list of pages and I'd run the script on those pages. We can repeat this as many times as is needed. If you're looking for something more interactive, you might want something written in JavaScript. For that, ask User:Erutuon, who is much better than I am at JavaScript. Benwing2 (talk) 10:55, 13 January 2020 (UTC)
 * He has asked me before; I probably should do it finally. I'll put it on my mental list. — Eru·tuon 09:44, 17 January 2020 (UTC)

Category:Valencia
Believe it or not, this along with its ca, es and pt children, has shown up with "out of time" module errors. If it helps any, none of these shows any categories, so it must be timing out before that part. The only thing I can think of is that there may be some kind of recursion or conflict due to Valencia being both a region/state in Spain and a city in California. Please have a look. Thanks! Chuck Entz (talk) 01:10, 27 January 2020 (UTC)
 * This was because I had "Valencia" as both an autonomous community of Spain and a city within that autonomous community. For now I've removed it from the city list. Benwing2 (talk) 01:19, 27 January 2020 (UTC)

Category:en:Washington, D.C.
Why did you recreate this? Did you see Category:en:Washington, D.C., USA? —Justin ( koavf ) ❤T☮C☺M☯ 06:08, 29 January 2020 (UTC)
 * Yes, I know about that and I'm planning on deleting it. I've added support to for major cities all over the world, and I've decided not to include the country name after the city. The logic is as follows: (a) nearly every existing city category does not include the country, and Category:en:Washington, D.C., USA is the only exception I know of; (b) including the country name causes problems e.g. for the city of São Paulo, where there is already Category:São Paulo, Brazil for the state of São Paulo (similarly for Rio de Janeiro). Benwing2 (talk) 06:13, 29 January 2020 (UTC)

re: "shiretown" parameter
Methinks you left out a step... Chuck Entz (talk) 00:42, 16 February 2020 (UTC)
 * Oops. I forgot to fix Module:place, will fix now. Benwing2 (talk) 00:49, 16 February 2020 (UTC)

Wingerbot is using gt/lt signs instead of curly brackets
For example. —Mahāgaja · talk 13:05, 21 February 2020 (UTC)
 * Oops. These changes were made manually by me in a text file, and I messed up here. Benwing2 (talk) 13:28, 21 February 2020 (UTC)

Delete request on شمالات
Can you take a look at this entry? Your bot made it and an IP requested it be deleted. Thanks! - TheDaveRoss  15:49, 25 February 2020 (UTC)
 * One of the plurals appears wrong but the other one is correct if rare, per Lane. I fixed the entry accordingly. Benwing2 (talk) 03:46, 27 February 2020 (UTC)

Bot request
I'm replacing a bunch of templates with new tone-requiring versions of them. For each, the former parameter  (or  ) is now going to be the first positional parameter in the new template, with the other positional parameters being incremented by 1 when present. I need some help with the ones I can't easily do by hand, and Erutuon told me you have a script to do this. The replacements are: Thank you! —Μετάknowledge discuss/deeds 17:43, 28 February 2020 (UTC)
 * to
 * to
 * to
 * to
 * Should be done. Benwing2 (talk) 06:02, 29 February 2020 (UTC)
 * Thank you! I've now made the /new templates into redirects — could you do a run to chop off /new from the transclusions? —Μετάknowledge discuss/deeds 07:21, 29 February 2020 (UTC)
 * Done. Benwing2 (talk) 07:41, 29 February 2020 (UTC)
 * And could you please also do the same (chop off the /new) for ? —Μετάknowledge discuss/deeds 07:51, 1 March 2020 (UTC)
 * Done. Benwing2 (talk) 11:36, 1 March 2020 (UTC)
 * I have another one, if you don't mind. I want to replace with : the former parameter   (or  ) is now going to be the first positional parameter in the new template (if neither of these parameters are present, the first parameter should just be the pagename), and the former   is going to be the second positional parameter (but it needs the first character stripped, so   becomes  ). Sorry for the extra trouble with this one. —Μετάknowledge discuss/deeds 18:19, 2 March 2020 (UTC)
 * Done. Benwing2 (talk) 03:43, 4 March 2020 (UTC)
 * Another one, just a simple replacement: to, please. Thank you! —Μετάknowledge discuss/deeds 19:04, 10 March 2020 (UTC)
 * Done. Benwing2 (talk) 01:29, 11 March 2020 (UTC)
 * And another: we can replace with, which I assume was the plan all along. —Μετάknowledge discuss/deeds 03:59, 23 March 2020 (UTC)
 * OK, it is running. As there are over 40,000 pages using the template, it will take a little while. I'm running 5 processes, each one handling 8,000 pages running at 1 per second, so it should be done in a little over 2 hours. Benwing2 (talk) 04:22, 23 March 2020 (UTC)
 * I have another request — please let me know if you want me to stop. This time, I'd like to replace with  in all Proto-Bantu entries. —Μετάknowledge discuss/deeds 02:45, 1 April 2020 (UTC)
 * Done. Don't worry, these requests are mostly easy to fulfill. Benwing2 (talk) 03:46, 1 April 2020 (UTC)

Block
Hi Ben. I need to be blocked, please. I've been spending too much time here again. --AcpoKrane (talk) 12:16, 1 March 2020 (UTC)

FYI
Chuck Entz (talk) 04:08, 4 March 2020 (UTC)
 * Sorry about that, I should have checked CAT:E after pushing those changes. I try to do that but occasionally forget. BTW, any time you see a sporadic error like this, it means I used the bot to push changes that I manually made to a text file, and the error is because I mistyped something. When I write a script to make the changes, typically you won't see any such errors, or if you do, you will see a lot :) Benwing2 (talk) 04:36, 4 March 2020 (UTC)

Bulgarian headwords and manual transliterations - bot request
(moved to Module talk:bg-nominal)

Mass removal of Russian manual transliterations
Hi,

Special:Contributions/86.134.66.200 has mass-removed Russian manual transliterations on "чн"/"шн". Is there an easy/quick way to undo this? --Anatoli T. (обсудить/вклад) 22:40, 5 March 2020 (UTC)
 * Ooof. Some bad abjad removals too. -- 23:00, 5 March 2020 (UTC)
 * It looks like there are some good changes mixed in with the bad ones. Can you identify the bad ones? If you can do that, I have a script I can use to undo those changes (even if there are subsequent changes from other users, although it looks like mostly there aren't). Ideally, save the HTML from the User contributions page and edit it; that will preserve the diff ID's. But that might be painful. If so, just select the lines from the User contributions page that contain the bad changes and paste into a text file, and I'll parse out the page names and undo the changes to those pages from this user. Note that there are some bad changes to pages that don't have чн in the page name, e.g. the change to тряпица. Benwing2 (talk) 01:36, 6 March 2020 (UTC)
 * Thanks. I will see what method is best, otherwise, I'll just clean up myself. --Anatoli T. (обсудить/вклад) 02:00, 6 March 2020 (UTC)
 * A script to revert any removals where the transcription differs from the automatic one would be nice. -- 02:11, 6 March 2020 (UTC)
 * No. The automated for is "staníčnik", the manual and the correct one is "staníčnik, staníšnik". --Anatoli T. (обсудить/вклад) 02:45, 6 March 2020 (UTC)
 * I think you misunderstand me. -- 03:41, 6 March 2020 (UTC)
 * Maybe, please clarify. --Anatoli T. (обсудить/вклад) 03:54, 6 March 2020 (UTC)
 * I think we got all of them reverted. Benwing2 (talk) 04:20, 6 March 2020 (UTC)
 * I'm not capable of evaluating the changes to Avestan or Old Persian; you'll have to look at them and roll them back as necessary. Benwing2 (talk) 04:21, 6 March 2020 (UTC)
 * You can't have the script look at the previous revision to see if it's in C:Terms with manual transliterations different from the automated ones? -- 04:27, 6 March 2020 (UTC)
 * I could write a script to do that, but (a) it would take longer than just manually reviewing the commits, (b) some of the changes (at least for Russian) were correct. Benwing2 (talk) 04:31, 6 March 2020 (UTC)
 * That's too bad. I'm mostly just going to revert all his edits on Iranian entries. -- 04:39, 6 March 2020 (UTC)
 * I only see about 20 of them, can't you just review them manually? Benwing2 (talk) 04:48, 6 March 2020 (UTC)
 * I don't mean just Proto-Iranian entries, I mean entries within the Iranian family. -- 05:05, 6 March 2020 (UTC)
 * Yes. There are the most recent 19 contribs, and about 30 more on March 1st, and that's it. Do you have rollback privileges? If not, I can give it to you, it will make your life easier. Benwing2 (talk) 05:11, 6 March 2020 (UTC)
 * I'm manually going through them now, but from what I see, it's more than that -- mostly Kurdish links. -- 05:14, 6 March 2020 (UTC)
 * @Benwing2: Yes, thank you. The rest of edits were OK. I wonder if manual (required) translits should have a hidden comment or something. I feel that I have to do it on a regular basis. I still feel that manual transliterations (even with commas) are important, like this one . . --Anatoli T. (обсудить/вклад) 04:25, 6 March 2020 (UTC)


 * These are also from the same person: Special:Contributions/5.81.100.151, Special:Contributions/86.145.59.25. -- 05:40, 6 March 2020 (UTC)
 * Damn, this guy is persistent. There are an awful lot of Iranian-related changes; if they need to be mass-reverted and can be done by a script, let me know how, and I'll see if I can write the script. Benwing2 (talk) 05:57, 6 March 2020 (UTC)
 * Yeah... annoying. I would actually even just settle for reverting any edits with  and  . Those are the worst edits. They don't seem to realize that some characters can represent multiple transcriptions, which is why manual ones were set. --  06:19, 6 March 2020 (UTC)
 * OK, how would the script go about figuring this out? Which templates are involved? Benwing2 (talk) 06:22, 6 March 2020 (UTC)
 * l and m, so reverting any edits (lines?) containing,  ,  ,  . --  06:27, 6 March 2020 (UTC)

T:R:Mackenzie
Benwing, mentioned you might have a bot script for moving links. If you have a moment, could you move all links to Template:R:Mackenzie over to Template:R:pal:Mackenzie:1971? I'd like to use T:R:Mackenzie for another reference template. Thanks for any help. -- 06:01, 8 March 2020 (UTC)
 * Done. Benwing2 (talk) 16:42, 8 March 2020 (UTC)
 * Thanks! -- 19:47, 8 March 2020 (UTC)

Can we simply combine the ‘Europe’ and ‘Asia’ parents?
I did this a while ago, but somebody told me that it wasn’t a good idea. Should I undo it? — (((Romanophile))) ♞ (contributions) 02:29, 16 March 2020 (UTC)
 * Yeah I don't think this is a good idea. Doing this makes things more obscure as you won't find e.g. Russia under either Europe or Asia when you'd expect it under both. BTW I don't think any country except Turkey and Russia should be listed as being in both Europe and Asia; the rest are only in Asia. Benwing2 (talk) 03:54, 16 March 2020 (UTC)
 * All right, sorry about that. Undone. —(((Romanophile))) ♞ (contributions) 04:12, 16 March 2020 (UTC)

Vowel length in Latin verno, vernus, vernalis
I was surprised to see these words marked with a long vowel, and looking through the edit history for verno, I noticed that you had added a macron to it in September 2019. I see that Lewis 1890 marks these words with a macron, and Bennett 1907's entry for vernus does also, simply stating "from vēr" (page 66).

However, vowels before a consonant cluster starting with a resonant are presumed to have been shortened in Latin at some point by a sound change, often called "Osthoff's Law" (a name that also applies to a similar sound change in Greek). Some exceptions are thought to be present in Classical Latin, but I don't know of any firm basis for supposing that the vern- words were such an exception. According to de Vaan 2008, the root of ver itself originally had a short vowel, and the long vowel found in Classical Latin vēr is secondary, resulting from compensatory lengthening when s was lost before n in the genitive: vesnos > ve:nos (with later replacement of ve:nos > ve:ros). Our entry in Wiktionary agrees with this account. De Vaan says that the vern- in vernus might come from either vesin- (in which case the e would have been short all along) or from ve:ri-n-. In the second case, Osthoff's shortening still seems like a possibility. "Osthoff’s Law in Latin", by Ollie Sayeed, assumes that vernus has a short vowel (page 157, in Indo-European Linguistics 5 (2017) 147–17).

So the etymological situation seems inconclusive, and as far as I know, there is no non-etymological evidence of a long e in the vern- words (e.g. in the form of either Latin-era inscriptions with apices, or distinctive Romance reflexes). Do you know more about this?--Urszag (talk) 05:04, 19 March 2020 (UTC)
 * I added the long vowel based on Alatius's web page, which quotes Bennett but contains corrections from several later authors. vērnus isn't corrected so I took it as correct. I have heard the arguments about Osthoff's Law but AFAIK it applied long before the Classical Latin period; at least that's how I learned this law worked. There are several exceptions like fōrma (as shown by Spanish horma not *huerma, French fourme), vēndō, quīnque, probably vāllum, ūllus, sūrsum, etc. So I am skeptical there was an Osthoff's Law that applied late enough to make a big difference in Classical Latin. De Vaan is of the Leiden school, which has its own peculiar ideas about Indo-European linguistics, so I wouldn't take everything he says at face value. I don't know about Ollie Sayeed but I see he's a PhD student. Benwing2 (talk) 05:33, 19 March 2020 (UTC)
 * Yes, there certainly were words in Classical Latin that for various reasons had long vowels before resonant-initial clusters, as I mentioned. The date of the shortening law would only matter if vernus was a late formation from vēr: if it was formed early on, then one possible scenario is the vowel being shortened by Osthoff's Law and remaining short after Osthoff's Law ceased being actively applied. Is there any evidence that it was a late formation?--Urszag (talk) 06:23, 19 March 2020 (UTC)
 * I don't have any evidence one way or other. But the assumption that Osthoff's Law continued to be applied well into the Classical period seems dubious to me given the large number of exceptions. It seems more logical to me that it applied very early on and then ceased to be active even before the Old Latin stage. Is there specific evidence that Osthoff's Law continued to apply into and past the Old Latin period? The only cases I know of where shortening before resonant + consonant seems to have occurred are before 'nt' and 'nd'.
 * Ultimately it seems that the best we can do is add a note indicating that there is some disagreement in the sources as to the length of 'vērnus/vĕrnus', maybe by writing it as 'vē̆rnus' with a note indicating which sources say it's long and which ones say it's short. Benwing2 (talk) 06:53, 19 March 2020 (UTC)
 * For many specific consonant clusters of the form RC, there aren't many examples of OL in Latin. Sayeed mentions shortening before -rn- in perna (>Spanish pierna), and before -mb- (from -ms- or -ns-) in membrum and in month names ending in -ember. Perna, membrum and -ember are supposed to be from PIE roots with long e (pages 156-157).--Urszag (talk) 07:25, 19 March 2020 (UTC)
 * The problem is that perna and membrum look to be very old constructions, meaning Osthoff's Law might have applied at the Proto-Italic stage or even earlier (in fact, the Wiktionary entry for membrum effectively dates Osthoff's Law to Proto-Italic or earlier by assuming shortening already at the Proto-Italic stage), and it's far from obvious how the -ember nouns evolved. In fact, Wiktionary's etymology for september assumes that the -em comes from the end of septem, not from the originally long ē of mēns. All of this is not to say that Osthoff's Law couldn't have applied later as well, but I feel we need better evidence. Benwing2 (talk) 04:10, 20 March 2020 (UTC)

Module:compound
The categories no longer work at all. DTLHS (talk) 16:01, 19 March 2020 (UTC)

Sri Jayawardenepura Kotte
I noticed a rather made by User:WingerBot so I fixed it, but you may want to look into it as a potentially larger problem. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 00:21, 27 March 2020 (UTC)
 * Thanks. This happened because I accidentally deleted the line that demarcated the division between two adjacent pages when manually editing the text. This is not indicative of a larger issue, just human error :) Benwing2 (talk) 00:54, 27 March 2020 (UTC)
 * Oops. Benwing2 (talk) 00:54, 27 March 2020 (UTC)
 * Thank you for your response, I am relieved. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 01:12, 27 March 2020 (UTC)

Bulgarian and Ukrainian pronunciation modules - minor fixes, hopefully
Hi,

When you have a chance (I know you're busy), could you please fix some most striking problems or issues:


 * 1) Handling of "я" and "ю" by Module:bg-pronunciation/testcases when they are not following consonants, should be [j], not [ʲ] like in Russian.
 * 2) There are no cases yet but Module:uk-pronunciation shouldn't make "и" as [e] in unstressed positions but always [ɪ]. I don't have a good reference on Ukrainian pronunciation. Ivan Štambuk must have based on some old book, which not so valid.

Pls let me know if you want anything on the Bulgarian noun inflection module. --Anatoli T. (обсудить/вклад) 03:56, 27 March 2020 (UTC)
 * I'll get to this in the next day or so. Benwing2 (talk) 03:57, 27 March 2020 (UTC)


 * Thanks for that, also please make Ukrainian "е" to always be [ɛ], also unstressed. This will make it a bit more phonemic but also more generic for most speakers.
 * There are few weird things as for the current pronunciation but it's hard to make it perfect, e.g. is [vɔu̯k] by most speakers, not [wɔu̯k]. No rush. Let's do Bulgarian first. --Anatoli T. (обсудить/вклад) 04:18, 27 March 2020 (UTC)
 * Hiya, I've changed my mind about the Ukrainian modules changes. It's better to follow some model, rather than mixing. Fixing Bulgarian module's errors are essential. --Anatoli T. (обсудить/вклад) 21:43, 29 March 2020 (UTC)

WingerBot creating errors
Hi, edits like are creating module errors in Bulgarian entries. —Mahāgaja · talk 06:24, 30 March 2020 (UTC)
 * Thanks. Fixed. --Anatoli T. (обсудить/вклад) 06:48, 30 March 2020 (UTC)

Белоруссия in
Hi,

Nah, it's not obsolete but is more respectful to Belarusians, even if  is obsolete. --Anatoli T. (обсудить/вклад)

Thesaurus:important person
I decide to be bold and renamed Thesaurus:big cheese to Thesaurus:important person. Could you do a bot run and replace occurrences of the former to the latter? Thanks. — SGconlaw (talk) 07:09, 12 April 2020 (UTC)
 * Done. Benwing2 (talk) 03:48, 16 April 2020 (UTC)
 * Thanks! — SGconlaw (talk) 05:01, 16 April 2020 (UTC)

New Bulgarian entries made by WingerBot
I noticed a mass-creation of Bulgarian entries by your bot. They look good - with the right senses, stresses and inflections! How?! It takes a long time to create entries manually. Did you extract all of them from a dictionary? Did it take you long?

I want to ask if it's possible to repeat this feat in the future (theoretically) with Ukrainian and Belarusian entries - The Ukrainian and Belarusian inflections exist on public sites, a smart bot would be able to load them. Russian might use some missing feminine forms, for example. Ivan Štambuk was able to mass generate Ukrainian entries with inflections, which is not easy to do if you have to do it manually. --Anatoli T. (обсудить/вклад) 09:11, 19 April 2020 (UTC)
 * I manually create a text file specifying the declensions, stresses, meanings, synonyms, antonyms, derived and related terms, and then use a script to generate the entries, and another script to push the entries using my bot. It takes a long time: the recent batch of 259 entries took 2 days to create. The time goes into looking up the entries, figuring out the inflection and meaning, and finding related terms. Running the actual scripts is fast. If I didn't have to worry about definitions or derived/related terms, it would go faster, for sure. The first few entries in the manually-created text file look like this:

n роб raw:From_West_Slavic,_from_,_from_._. <+и+ове[a]>|f=роби́ня|adj=ро́бски (also)(f)slave;prisoner;(f)servant;ux:Ваш_поко́рен_роб|your_obedient_servant der:ро́бство n ро́бство роб+-ство  slavery,servitude;syn:робу́ване;yoke;syn:и́го rel:роб:роби́ня n ро́ба - <> robe,dressing_gown n хълм inh:sla-pro:*xъlmъ <+ове+и[p]> hill der:хълми́ст n щрих de:Strich <+и>|adj=щри́хов (l)line,stroke der:щрихи́рам,щрихо́вам Benwing2 (talk) 09:25, 19 April 2020 (UTC)


 * Thanks for the hard work! --Anatoli T. (обсудить/вклад) 09:32, 19 April 2020 (UTC)
 * Hiya. How is going? Do you want me any help with Bulgarian - something I could do? The verbs are going to be a bit more complex but I hope there will be much more commonalities by types, for example, a missing verb has many verbs with similar or identical conjugations of type 173ti in https://rechnik.chitanka.info/type/173ti. We will need to check if a common type also includes stress patterns. --Anatoli T. (обсудить/вклад) 23:29, 21 April 2020 (UTC)
 * Hey. I've written the code for multiword nouns and adjectives and I'm testing it now. Afterwards comes the verbs. It would help if you could sort out the different possible stress patterns of verbs, as I have little idea of the possible variations. Maybe make a table similar to User:Benwing2/test-bg-ndecl that lists various verbs and certain forms. I think Bulgarian verbs don't have stress movement within a single tense, unlike Russian, so it may be enough to list the following forms:
 * the first-person singular present (future for perfective verbs0
 * the first-person singular aorist
 * the first-person singular imperfect
 * singular imperative
 * singular imperative
 * masculine indefinite singular present active participle
 * masculine indefinite singular past active aorist participle
 * indefinite plural past active aorist participle
 * masculine singular past active imperfect participle
 * plural past active imperfect participle
 * masculine indefinite singular past passive participle
 * indefinite plural past passive participle
 * adverbial participle
 * singular indefinite verbal noun
 * Apologies for requesting so many forms per verb. I think it may be possible to do without the plural variants of the participles, which would eliminate three forms (14 -> 11). But it looks from that Bulgarian verbs are just very complicated. Benwing2 (talk) 02:15, 22 April 2020 (UTC)
 * OK, thank you. I have just started User talk:Atitarev/test-bg-vconj User:Atitarev/test-bg-vconj. Just trying to make sense at the moment. --Anatoli T. (обсудить/вклад) 03:49, 22 April 2020 (UTC)
 * I have corrected the link. Please check if the number of columns is sufficient. --06:59, 22 April 2020 (UTC)

vēnus
This Latin noun is documented only in the dative and accusative. Is there a way to reflect this fact in the declension table? I could find no instructions in the template documentation for missing forms. --EncycloPetey (talk) 16:35, 22 April 2020 (UTC)

Incorrect bot edits
I notice that on 21 August, you made a lot of incorrect bot edits to forms of the verb compono/conpono, removing the length in -po- where it should have been kept. Would you mind correcting these errors? 2A02:2121:30A:5F8E:2CEB:77A0:3248:7871 08:18, 30 April 2020 (UTC)

auxiliandi
Could you explain why you deleted this? I can find it attested in Caesar, but I could be missing something. StudiesWorld (talk) 14:24, 4 May 2020 (UTC)

Belarusian and/or Ukrainian
Hi,

You have done an amazing piece of work for Bulgarian. Thank you! There's always room for improvement but it seems the infrastructure is now in a very good shape.

Please let me know if you're interested in improving Belarusian and/or Ukrainian contents as well. I am more familiar with these languages but I can't give much of a technical advise and I don't think we'll get much help from others. Resources are somewhat better for Ukrainian. Some work on conjugation templates is currently going on Russian and Ukrainian Wiktionaries. The inflections are more complex and there is more variety than Russian but even small improvements would be appreciated. I am OK to continue to add inflections manually but some templates don't even allow that, for example, Ukrainian templates can't handle reflexive verbs. --Anatoli T. (обсудить/вклад) 02:00, 13 May 2020 (UTC)
 * Sure, I can do some work on Ukrainian, esp. since the resources are better than for Belarusian. I will start by looking into the issue with reflexive verbs. If you could point me to any resources on Ukrainian declensions or conjugations, it would be helpful. Benwing2 (talk) 03:03, 13 May 2020 (UTC)
 * Thanks. You can start by analysing our currently used templates. For example, a transitive verb has a full paradigm. If you add any term in https://goroh.pp.ua/Словозміна, it will give you inflections, sometimes more than one for different senses. It's very good for nouns and adjectives. The verb conjugation is not 100% complete there. We will need to fill the rest - or leave unpopulated to be filled. E.g. https://goroh.pp.ua/говорити doesn't provide participles, and only one type of the future tense, e.g. говори́тиму - I will talk (infinitive + му ending for 1st pers. sg). The missing type is formed the same way as Russian, e.g. бу́ду говори́ти - I will talk. --Anatoli T. (обсудить/вклад) 03:29, 13 May 2020 (UTC)
 * BTW, many Ukrainian terms have audio files, so you can get a feel how Ukrainian sounds. You may wonder if unstressed "и" is represented correctly by the pronunciation module. E.g. говори́ти is [ɦɔu̯ɔˈrɪte] but you will hear [ɦɔu̯ɔˈrɪtɪ]. It is based on a more classical accent, which is fading away. --Anatoli T. (обсудить/вклад) 03:39, 13 May 2020 (UTC)
 * I have found another online resource for Ukrainian conjugations, the dictionary I have been using for a while - https://www.lingvolive.com/en-us/translate/uk-ru/давати - need to scroll down. It doesn't give stresses here but it gives MORE forms missing at https://goroh.pp.ua/Словозміна. (The dictionary needs some getting used to. It annoys you with requests to log on but t won't do it again after the first query - you can switch languages and look up other words. Clicking refresh is always better and adding the term to the URL). --Anatoli T. (обсудить/вклад) 08:02, 13 May 2020 (UTC)
 * It looks like Ukrainian grammar is pretty similar to Russian grammar, so as a first approximation maybe we can use Zaliznyak's system of notating nouns and verbs. Benwing2 (talk) 01:50, 14 May 2020 (UTC)
 * Yes, there are many similarities.
 * For verbs, there are more forms - alternative future, роби́тиму/бу́ду роби́ти ("I will do"). Ukrainian is unique in this among Slavic languages, forming is easy but care should be taken for reflexive verbs - -сь/-ся are also used, like in Russian but the rules are different.
 * Many verbs in -ти have alternative infinitives in -ть, the tables could use those: розмовля́ти/розмовля́ть but not verbs like нести́.
 * Pluperfect tense is sometimes described, e.g. "чита́в був" (I had read), no need to include in tables, IMO.
 * Present active participle is normally missing, it may be confused with a noun, e.g. даючий. So, I am not sure if it should be included. In any case, it's a bit hard to find for each verb and find the right stress. Perhaps as a parameter, rather than an automatic feature.
 * "past_pasv_part_impers", is currently not displayed, e.g. "ро́блено" in tables but it should. It's very typical and is also used with intransitive verbs.
 * All nouns have vocative forms, even if they are not used in the real life, that's what grammar books do.
 * Some challenges will be with alterations, such о/і, г/з, к/ц, х/с, л/в, у/в, which are not present in Russian.
 * You can request inflections for different terms and I will try to make entries or just inflection tables. If you improve the current conjugation tables, it will make it easier for me, so that I could use reflexive as well. I will be able to add multiple different verb conjugations on a page, so that you could build a module, if this is what you're planning. Just let me know. If you want to focus on nouns first, that's fine. They are easier. You'll get exposed to many sound changes, which are common for Ukrainian. Some things are still very unfamiliar to me and even current adjective templates include a number of archaic/very rare forms. It's definitely more complex than Russian.
 * You can also take a look at Ukrainian Wiktionary inflection templates, some are good but far from comprehensive in coverage. --Anatoli T. (обсудить/вклад)


 * Great job, Ben! Are you able to run a bot and add on all Uk, Be, Bg and Ru terms requiring inflections but missing them (and no )? --Anatoli T. (обсудить/вклад) 07:28, 27 May 2020 (UTC)
 * Again, great job, but I noticed that your bot added the conjugation table for касувати to . İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 13:00, 27 May 2020 (UTC)
 * Thanks for spotting. That has been fixed. --Anatoli T. (обсудить/вклад) 23:41, 27 May 2020 (UTC)
 * I added to all Uk, Be, and Bg nouns, verbs and adjectives needing inflections. Haven't yet done proper nouns or Russian. Benwing2 (talk) 02:06, 31 May 2020 (UTC)
 * Thank you for this and the fixes and the enhancements to Module:uk-headword. I saw you added comparatives, diminutives, etc. --Anatoli T. (обсудить/вклад) 03:31, 31 May 2020 (UTC)
 * : Speaking of Belarusian declensions, added a few some time ago, but generally without the stresses. See  or  for example. PUC – 10:05, 31 May 2020 (UTC)
 * Thanks, it would be good to label them for attention but maintenance categories for Belarusian are now quite large. I personally have no issue adding a stress mark at all. I am using SC Unipad, which decomposes diacritics and I just copy/paste the stress mark in the right place. At http://www.slounik.org/, you can see the underlined stress marks but you can't copy it. Belarusian headwords need a lot of rework too. --Anatoli T. (обсудить/вклад) 10:15, 31 May 2020 (UTC)

I am your student
Hello from el.wiktionary. I know nothing about Lua, computers, but I am fascinated by your work. And i learn a lot from your comments. I have been trying to unify templates for 'places'. Nothing complicated like your Module:place. I tried a /data page at el:Module:sarritest but I cannot make even one little link work, I cannot make them speak to each other. Is there a magic word for it. &#8209;&#8209;Sarri.greek &#9835; | 17:45, 23 May 2020 (UTC)
 * Hello! I can help you if you can specify exactly what you need to happen. Benwing2 (talk) 00:08, 24 May 2020 (UTC)
 * O! thank you. I do not wish to bother you. Here is a brief copy: el:Module:topos. I fell SO stupid... &#8209;&#8209;Sarri.greek &#9835; | 00:20, 24 May 2020 (UTC)
 * O! disaster: The data page el:Module:topos/where will not accept greek terms with spaces or with dialytics. e.g. el:Αρούμπα, el:Template:test I have tried various things. &#8209;&#8209;Sarri.greek &#9835; | 11:49, 6 June 2020 (UTC)
 * I'll take a look later today. Benwing2 (talk) 16:27, 6 June 2020 (UTC)
 * Can you explain exactly how it's not accepting terms with spaces or diacritics? I'm not quite sure how the module is being used and what error you're seeing. Benwing2 (talk) 16:44, 6 June 2020 (UTC)
 * OK, I see. Benwing2 (talk) 16:46, 6 June 2020 (UTC)
 * The issue appears to be line 21, where you have this:
 * local where = require("Module:sarritest/where")
 * if you change it as follows:
 * local where = require("Module:topos/where")
 * then el:Αρούμπα works fine. Benwing2 (talk) 21:39, 6 June 2020 (UTC)
 * Also everything on el:Template:test. Benwing2 (talk) 21:41, 6 June 2020 (UTC)
 * It was a silly bug?? I am so sorry to have bothered you. I thought it was some very difficult font issue. I have asked Lua support for small wikis at meta.wikimedia.org but noone answers. I thank you very much. I would never solve it.  &#8209;&#8209;Sarri.greek &#9835; | 21:48, 6 June 2020 (UTC)

Virile and nonvirile
Can you add the virile and nonvirile genders to Module:gender and number? The workarounds function, but they are not entirely accurate (e.g. utensylia, drzwi). İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 01:49, 26 May 2020 (UTC)
 * What do "virile" and "nonvirile" genders mean? We already have personal and non-personal, are they the same? Benwing2 (talk) 01:53, 26 May 2020 (UTC)
 * Definition of nonvirile from Appendix:Glossary: "In Slavic languages, a plural gender used for all groups that do not contain men, as well as plurals of masculine animate, masculine inanimate, feminine and neuter nouns. Contrast virile." Nonvirile and virile are typically abbreviated as "nv" and "vr" on templates and "nonvir" and "vir" in glossing (according to the Oxford Handbook) respectively by the way. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 01:59, 26 May 2020 (UTC)
 * I see. We currently use npers for non-personal in glosses, is it ok to use nvir for non-virile or would you prefer nonvir? Benwing2 (talk) 02:03, 26 May 2020 (UTC)
 * Sure, that works. Thank you. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 02:04, 26 May 2020 (UTC)
 * Done. If you want we can also add categories for virile and nonvirile nouns. Benwing2 (talk) 02:22, 26 May 2020 (UTC)
 * That would be great as well. My username is Ilawa-Kataka by the way. Thank you. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 02:28, 26 May 2020 (UTC)
 * Done. My apologies for getting your name wrong. Benwing2 (talk) 02:36, 26 May 2020 (UTC)
 * Thanks again and don't worry about it. My real name is a rather common English name, but I have been called everything from Mitchell to Julia. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 02:47, 26 May 2020 (UTC)
 * There are three problems with the current execution. Firstly, the category descriptions are misleading in that they sound like they are singular whereas virility is strictly plural. I would write for the virile and nonvirile categories respectively: "Polish nouns that refer to a group with at least one male human." and "Polish nouns that refer to a group without male humans." Also, I think both categories should be (also?) subcategories of Category:Polish pluralia tantum since these categories by their nature refer to plural-only nouns. Lastly, Module:pl-headword can neither handle virility categorisation nor display the virile and nonvirile genders. While it looks to be a straightforward edit adding a couple of lines between lines 190 and 191, I am not remotely confident in my moduling skills. Thank you in advance. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 16:58, 26 May 2020 (UTC)
 * Never mind, I did everything myself. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 17:50, 27 May 2020 (UTC)
 * Apologies for not getting to it faster. I'm not sure it's correct to put virile nouns under pluralia tantum, though; this is a particularity of Polish. I think it should be added by the Polish module. Benwing2 (talk) 01:01, 28 May 2020 (UTC)
 * Go ahead and change it, I am uncertain of any application of virility outside of Polish. Also, there is no need to be sorry, I was glad to get a bit of moduling experience. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 01:12, 28 May 2020 (UTC)

Ojibwe verb categorization
Hi there,

I saw your recent creation of the Ojibwe verb "gaanda'an". I've been trying to sort out the Ojibwe verb classes, and was aiming for a 4-way classification: VII-VAI-VTI-VTA. I'm no expert in wiktionary editing, but this 4-way classification matches the consensus in Ojibwe grammar. That said, your "gaanda'an" entry (and i assume others) doesn't follow that, separating the transitivity from the animacy, which i don't think works because those two elements can't be disassociated from one another. I don't want to assume my classification (following the Ojibwe People's Dictionary and others) is most appropriate for wiktionary. So, before continuing, i thought i would check with you.

SteveGat (talk) 15:21, 28 May 2020 (UTC)
 * Hi. I guess you are referring to the categories that classify a verb separately by animacy and transitivity? I'm confused as to why this doesn't work. What is it about these two properties in Ojibwe that makes them so intimately bound? In many languages, nouns, verbs, etc. have multiple properties, and just because a verb is e.g. identified in the dictionary as "transitive inanimate" doesn't mean it needs to be categorized with both at once and not with the two separately. That said, I'm not opposed to creating a category like "Ojibwe transitive inanimate verbs" but in that case I think we should also categorize at least under "Ojibwe transitive verbs" for consistency with other languages. This can be accomplished by creating a special template that takes a parameter to specify the transitivity and animacy; I can create that template for you if you want. Benwing2 (talk) 01:37, 29 May 2020 (UTC)


 * You can see my overall response in the Beer Parlour. Specifically on animacy versus transitivity: intransitive verbs have different paradigms depending on the animacy of the subject, while transitive verbs (which by definition have animate subjects) have different paradigms based on the animacy of the object. In other words, animacy isn't a feature of the verb itself, so classifying a transitive verb as "animate" (VTA - animate object) has a completely different meaning than classifying an intransitive verb as "animate" (VAI - animate subject). I hope this clarifies the issue, and i appreciate your helping my think this through. SteveGat (talk) 17:11, 29 May 2020 (UTC)

Small bot request
Can you make changes like ? Ultimateria (talk) 17:31, 2 June 2020 (UTC)
 * Done. 88 replacements. Benwing2 (talk) 01:06, 3 June 2020 (UTC)
 * Thank you! Ultimateria (talk) 02:39, 3 June 2020 (UTC)
 * Here you go:
 * Hypoderma bovis larvae young.jpg


 * ;-) Chuck Entz (talk) 05:37, 3 June 2020 (UTC)
 * Oops, I just saw this. I don't think they survived my vacation 😨 Ultimateria (talk) 04:21, 26 June 2020 (UTC)

South American countries (bot edits)
WingerBot's templatisation of South American country entries has marked them with  instead of   causing improper categorisation. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 14:32, 22 June 2020 (UTC)
 * Thanks for letting me know; fixed. Benwing2 (talk) 02:30, 23 June 2020 (UTC)

Replacing uses of on English citations pages with
Hi, would it be possible to replace uses of on English "Citations:" pages with ? wishes to convert from a redirect to a general-purpose template, but this is currently distorting its use as a redirect. — SGconlaw (talk) 15:02, 23 June 2020 (UTC)
 * Sure, I'll do that this evening. Benwing2 (talk) 15:30, 23 June 2020 (UTC)
 * Thanks. Much obliged. — SGconlaw (talk) 15:52, 23 June 2020 (UTC)
 * Done. It converted about 12,000 pages and left about 250 alone as they were in other languages. Benwing2 (talk) 04:04, 24 June 2020 (UTC)
 * Thank you! — SGconlaw (talk) 05:10, 24 June 2020 (UTC)

Ukrainian
How well do you actually speak Ukrainian? --Nueva normalidad (talk) 00:39, 25 June 2020 (UTC)
 * I don't, really. Benwing2 (talk) 00:41, 25 June 2020 (UTC)
 * Oops. Benwing2 (talk) 00:42, 25 June 2020 (UTC)


 * : May I ask, why you are asking? --Anatoli T. (обсудить/вклад) 04:24, 26 June 2020 (UTC)
 * Because you've been spending most of 2020 writing ridiculously complex Ukrainian Modules. --Nueva normalidad (talk) 13:51, 26 June 2020 (UTC)
 * Hey. I don't know if you realised that it wasn't Benwing2 asking you the question. "You" must be referred to him.
 * The modules are complex but it's the grammar, variations and inconsistencies that are ridiculously complex in Ukrainian, especially nouns. I can attest that Benwing2 is bloody efficient at this and the modules are very powerful if used correctly. It's not only Ukrainian this year - Bulgarian modules for all inflections are also done this year. I think with the experience (Russian, Arabic, Latin modules) and the approach he has taken, many things are possible, you don't really have to speak the language well. That's why it's even more amazing. In any case, I can't complain about the development speed or accuracy, LOL. --Anatoli T. (обсудить/вклад) 02:03, 27 June 2020 (UTC)

Module:ang-verb
May I copy this module to Russian wiktionary? ПростаРечь (talk) 17:42, 9 July 2020 (UTC)
 * Sure, go ahead. Benwing2 (talk) 02:16, 10 July 2020 (UTC)

спусціцца
was a mistake on WingerBot's part right? İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 22:16, 11 July 2020 (UTC)
 * Not a mistake. The deleted verb doesn't belong on that page; it's already on its own page. Benwing2 (talk) 22:34, 11 July 2020 (UTC)
 * Thank you, I see that now. Sorry for bothering you. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 22:42, 11 July 2020 (UTC)
 * No worries, you're not bothering me by asking about my edits. Benwing2 (talk) 22:44, 11 July 2020 (UTC)

More non-native audios
Can you remove Portuguese audios by User:Adélaïde Calais WMFr? Example file File:LL-Q5146 (por)-Adélaïde Calais WMFr-abatido.wav. I'll add any more that I find to this section. Ultimateria (talk) 22:37, 14 July 2020 (UTC)
 * Done. Benwing2 (talk) 00:38, 15 July 2020 (UTC)

PAGENAME in audio file names
Hi, what is the benefit of ? If the file name were to be changed at Commons, would the bots that fix file names still recognize this as the same file? —Mahāgaja · talk 08:17, 19 July 2020 (UTC)
 * I did this because User:Atitarev has been consistently making changes of the same sort manually. I figured it makes it easier to copy text from one entry to another because you don't have to manually enter the pagename. However, I didn't think about file name changes. Under what circumstances does this happen, and who runs the bots to handle these changes? Can you point to an example diff, as I don't recall having seen such changes? If it ends up making these bots not work, potentially I could certainly do a run to undo all the changes although easier might be to fix the bots to handle this. Benwing2 (talk) 14:35, 19 July 2020 (UTC)
 * is an example where the bot CommonsDelinker changed the name of a file at Wiktionary because the name of the file was changed at Commons. That bot is run by c:User:Magnus Manske; I guess you'd have to ask him whether the bot would still recognize the name if it uses PAGENAME. Of course, files don't get their names changed very often anyway; it usually only happens when there's a typo in the old name. —Mahāgaja · talk 18:10, 19 July 2020 (UTC)
 * Thanks. All of these audio files have a very simple and consistent naming format, e.g. Ru-альтернатива.ogg so they've probably been put there by bot and are unlikely to change. However, I'd definitely like to hear from c:User:Magnus Manske. Benwing2 (talk) 18:54, 19 July 2020 (UTC)
 * There's also the chance of them being deleted, since removing links to deleted files is CommonsDelinker's other task. Again, it's unlikely to happen, but it is theoretically possible that some such file was later found to be a copyvio or something and deleted, and if CommonsDelinker didn't know we were using it, we'd be left with a red link. —Mahāgaja · talk 19:36, 19 July 2020 (UTC)
 * I, in turn adopted this method, when I saw User:PUC using it in Belarusian and later Ukrainian and Russian entries. It is a big time-saver when mass-creating entries and should be embraced by others, IMO. If audio or image files get renamed, deleted or replaced, it is a good practice to change all linked entries and I've seen conscientious editors do exactly that with edit summaries. If it doesn't happen, then well, we'll have to fix it ourselves. There's no full protection from this, whether we copy the page name or just replace it with PAGENAME. I'd like to continue doing it, if there are no objections and recommend doing the same for other languages where the same pattern works. If bots don't work with PAGENAME, then I guess, it needs to be trained to work with it. --Anatoli T. (обсудить/вклад) 02:51, 20 July 2020 (UTC)
 * But surely saving time when mass-creating entries is no reason to change existing file names so that they use PAGENAME instead of the actual name? Can't you subst it in when mass-creating the entries? —Mahāgaja · talk 07:51, 20 July 2020 (UTC)
 * PAGENAME serves as a template for new entries or entries to be enhanced. If they are all consistent, then all you need to do is copy that line from similar entries. You won't get many entries with wrong audio files copied from a different entry like of, which belongs to . --Anatoli T. (обсудить/вклад) 08:01, 20 July 2020 (UTC)
 * Fine, I guess it's not a big deal. —Mahāgaja · talk 08:06, 20 July 2020 (UTC)

Category standardisation
Why have you had WingerBot standardise topic categories as rather than ? For me, is slightly more efficient and has a logical progression into. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 12:21, 20 July 2020 (UTC)
 * Good question. I chose because it's fairly clear whereas  seems a bit obscure; on the other hand I didn't think about the parallels with . Benwing2 (talk) 04:15, 21 July 2020 (UTC)
 * For me, is less obscure than  since it links to a "c"ategory. Do you think this warrants a Grease pit discussion? İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 15:38, 21 July 2020 (UTC)
 * If you wish. Not sure it will get anywhere, though; we had a discussion about a similar topic awhile ago and the upshot was there are differing opinions on which template name is preferred. Benwing2 (talk) 08:51, 22 July 2020 (UTC)
 * I think I remember that one, so never mind. But anyways when I categorise a page it is with and I will assume that is not a problem unless I hear otherwise. In an unrelated matter, I left a template edit request at the Grease pit about the Kyrgyz declension template—could you add singulare and plurale tantum support when you have some extra time? İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 19:27, 22 July 2020 (UTC)
 * Feel free to use ; no objections. I will take a look at the Kyrgyz declensions. Benwing2 (talk) 03:45, 23 July 2020 (UTC)

problem
At Santa Catarina (and other Brazilian state entries), producing "the the", as well as linking to a Portuguese entry when it should be linking to an English one. —Μετάknowledge discuss/deeds 08:46, 22 July 2020 (UTC)
 * Thanks. There's some old hacks involving 'macroregion' that don't work well; I need to remove them. Converting them to regular 'r/' = region fixes them. Benwing2 (talk) 08:49, 22 July 2020 (UTC)
 * Same problem at Jerusalem, Batman, Heligoland, and surely more. —Μετάknowledge discuss/deeds 08:59, 22 July 2020 (UTC)

Belarusian months
When WingerBot cleaned up the Belarusian month entries, it made them countable (which I recall Atitarev says is incorrect). İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 13:21, 26 July 2020 (UTC)
 * I thought User:Atitarev said they're countable. They are countable in other languages (e.g. "it happened two Septembers ago"). If this is incorrect, I will fix them. Benwing2 (talk) 18:40, 26 July 2020 (UTC)
 * Belarusian months are countable, even if plurals are uncommon, so we have to infer or search for correct forms. --Anatoli T. (обсудить/вклад) 22:44, 26 July 2020 (UTC)
 * Sorry about that, far too late a night I guess. Thanks for correcting me. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 23:36, 26 July 2020 (UTC)
 * No problem. The latest discussion was in Talk:травень. --Anatoli T. (обсудить/вклад) 00:58, 27 July 2020 (UTC)
 * That is the one to which I was referring, though my memory failed me this morning. I admire your and Benwing's recent work on the Slavic languages and have been following it closely, though I do not know enough about those languages to really contribute. If you get to Polish I would be more active, but its infrastructure is in a much better state than several other Slavic languages so it's not a priority. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 01:25, 27 July 2020 (UTC)

New requests
Hi,

Thank you for all the work - Bulgarian, Ukrainian and Belarusian inflections and headwords are so much better this year!

Are you able to run a bot to replace grave accents to acute accents for all Bulgarian templates?

Would you be interested in much smaller improvements for Macedonian (headword) and Hindi (inflections are simple)? --Anatoli T. (обсудить/вклад)
 * I can run a bot to replace the grave accents, and I'll work on the other tasks afterwards. Benwing2 (talk) 02:33, 5 August 2020 (UTC)
 * Thank you for that. I'd like to learn a little from you on module development too. --Anatoli T. (обсудить/вклад) 02:37, 5 August 2020 (UTC)
 * How much programming experience do you have? If you know how to program, writing modules isn't that hard, although you have to learn Lua (which is similar in many ways to Javascript and somewhat similar to Python). You can refer to Scribunto to get started, and copy an existing module. Benwing2 (talk) 05:45, 5 August 2020 (UTC)
 * I ran a bot to check for grave accents in Bulgarian headword templates but it found only one instance, азе. What other templates are you thinking of? I also added Module:mk-headword. I'm looking into Hindi nouns; do any nouns require manual translit? It looks like it's enough in most cases to specify the gender of the noun, and the rest can be inferred automatically. Benwing2 (talk) 02:27, 6 August 2020 (UTC)
 * Thanks. I come across grave accents all over the place. E.g. User:Benwing2/sla-pro-bg-redlinks shows a few like "бадàкам". I meant, , , , , , etc. Many sorts of templates.
 * Thanks for Module:mk-headword. I see that older features are preserved too. What functionality has been added? I see dim, adj, m2/f2, etc. are working now. Great!
 * As for Hindi transliterations, adding them for inflections seems cumbersome. I think you can use the automated transliteration in 99% of cases. The only problem is the shwa-dropping and dropping on the right syllable, which is the work of the translit module. In cases where the automated correct transliteration is impossible, an invisible virama symbol could be used to manipulate or phonetic respelling. Say, I want  to be displayed and declined in all forms as "alārm klŏk" (dropping "a"), it could be phonetically described as  (with a virama at the end of the first word). Perhaps the phonetic respellings could be used as an additional optional parameter. E.g.  can be spelled in a number of ways, with or without nuqta, with  or with  but the phon could force the desired reading. So, I think respellings (exposed or non-exposed, to be discussed?) is better than manual transliterations. I am not sure yet if we want the respelling to be shown to the user, e.g  for
 * Template Needs a categorisation by language, please, similar to Category:Russian spellings with е instead of ё, I think. --Anatoli T. (обсудить/вклад) 03:00, 6 August 2020 (UTC)
 * I see. I think I can trick the translit module into tracking cases where grave accents occur. As for Hindi, the manual transliterations would be needed only where the translit module gets it wrong, similar to Russian transliterations, and only on the lemma, not on the inflections. The declension module will take care of adding the translits to the inflections. So it looks like the choice is between manual translits and respellings. It's the same amount of work in the module in either case. As a first effort I will probably not worry about this, as the existing declension templates don't support it; I can add it afterwards. BTW I don't see the virama at all that you added, either in the normal display or when editing. What do you want the nuqtaless-form-of category to be called? Something like Category:Hindi nuqtaless forms? Benwing2 (talk) 03:14, 6 August 2020 (UTC)
 * OK, just automating the inflections is a good start.
 * Virama is a little diacritic at the end of the of अलार्म् (at the bottom) - letter by letter: अ ल ा र ् म ् . How about . Can you see here? I am using Chrome. You can download SC UniPad, which is very good for decomposing conjuncts, diacritics, accents, etc. You can copy/paste most of the symbols you need (good for vocalising Arabic sentences, for example).
 * Yes, Category:Hindi nuqtaless forms is good. --Anatoli T. (обсудить/вклад) 03:39, 6 August 2020 (UTC)
 * For the example you give, I see viramas after the fourth and fifth letters and under the dotted circle. But it looks like you're putting them under spaces instead of letters, which is probably why I see them. I am also using Chrome, on a Mac Book Pro. Benwing2 (talk) 03:41, 6 August 2020 (UTC)
 * BTW All the Macedonian templates should have documentation describing their params. Benwing2 (talk) 03:42, 6 August 2020 (UTC)
 * I have only put virama under spaces in this example अ ल ा र ् म ्, after र (r) and after म. Normally is spelled with one virama after r. You can't see it, as it's all merged into a conjuct र्म, with र appearing as a little hook on the top. Without a virama - रम look as separate too letters and it would be pronounced as "ram" (with the final inherent "a" dropped). अलार्म्  with two virama after r and m is a (possible) respelling to force the final inherent "a" to also drop, i.e. "alārama" becomes "alārm". Virama is also called a vowel/shwa killer, I think.
 * Please download "SC UniPad", it's light and useful. --Anatoli T. (обсудить/вклад) 03:52, 6 August 2020 (UTC)
 * Unfortunately per the docs it only works on Windows. Benwing2 (talk) 03:54, 6 August 2020 (UTC)
 * That's a shame. Perhaps you're missing some fonts on your Mac? Do you see all symbols in Module:hi-translit, in particular under the line which says "virama"? --Anatoli T. (обсудить/вклад) 03:59, 6 August 2020 (UTC)
 * BTW, in Australia, “it’s a shame” = “it’s a pity”, not shaming you but regretting you can’t access the tool :) --Anatoli T. (обсудить/вклад) 08:52, 6 August 2020 (UTC)


 * Just FYI. I just found a good example where a respelling or manual transliteration would be required:, spelled +  +  +  +  + , should be pronounced as "mehnatī", not "mehantī". The way to tell, which shwa (a) should be dropped could be  (not the actual spelling), just applying a virama after the consonant  where the inherent vowel "a" should be silent - ha -> h.
 * Hindi Devanagari is very phonetic. The few issues with spellings is that shwa-dropping rules (not complicated for humans) sometimes don't work with automated transliterations or syllabification can't be predicted, like in this case or in loanwords.
 * Another issue is that Hindi speakers (writers) typically fail to write nuqta (dot), not unlike Russians change ё to е or don't always use a couple of other diacritics: ॊ = ǒ, 'ॆ = 'ě', ॉ = 'ŏ', creating cases where pronunciations differ from spellings. At Wiktionary we make terms with nuqta and with strict spellings the main forms, just like we do with Arabic (hamza-less alif, etc.) or Russian - less formal native spellings. --Anatoli T. (обсудить/вклад) 01:11, 7 August 2020 (UTC)
 * I am working on this. I will implement phonetic spelling for forcing manual translit. Benwing2 (talk) 03:58, 8 August 2020 (UTC)
 * Thanks (and fixing a typo in my post). --Anatoli T. (обсудить/вклад) 23:16, 8 August 2020 (UTC)
 * Hi, @Benwing2 and . Thank you very much for improving Hindi modules and templates! The complexity actually exceeded my original knowledge and I can't access good grammar resources, so I wasn't able to contribute much at the end, especially with verbs and adjectives. I am sure I will get back to Hindi work later. I am currently busy learning Korean. --Anatoli T. (обсудить/вклад) 07:36, 30 September 2020 (UTC)

Kyrgyz
Two more things with : it currently cannot handle ь (used in loanwords such as Тайвань), which is either removed or not removed (more commonly removed) in declension but does not affect the suffixation otherwise; and it also does not handle possessive suffixes, which I think are best explained at https://www.researchgate.net/publication/313774172_Kyrgyz_Orthography_and_Morphotactics_with_Implementation_in_NUVE at page 7. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 19:02, 6 August 2020 (UTC)
 * Hi. I'll get to these soon. The issue with ь can maybe be handled by an argument, something like keep_soft_sign for those cases where the soft sign should not be dropped. Benwing2 (talk) 01:43, 7 August 2020 (UTC)
 * How about b(oth)/y(es)/n(o) or just soft? For Тайвань I found both in use (with Тайвандын getting a few thousand results and Тайваньдын a few hundred—both used in professional contexts). İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 00:13, 8 August 2020 (UTC)
 * Sounds good to me. Benwing2 (talk) 03:59, 8 August 2020 (UTC)

Source for formal grammar for templates
I was wondering if, for example, you know a good formal source of the declension rules for Russian verbs? I see you're the author of the rather impressive ru-conj table, and was wondering if that work was based on some formal set of rules codified somewhere?

I'd like to have a crack at rewriting the conjugation system for my own learning, and was wondering if there's an authorative source, rather than interpreting your code.

Thanks
 * The authoritative reference is called Грамматический Словарь Русского Языка by A.A. Zaliznyak. It's in Russian so you have to be able to read it at least somewhat, or make liberal use of Google Translate. Benwing2 (talk) 00:48, 22 August 2020 (UTC)

пат
Etymology 2 at Macedonian needs to display uncountable. Ultimateria (talk) 20:40, 27 August 2020 (UTC)

de-IPA question
Hello Benwing2!

I was wondering if it would be possible to write a module that creates automatic German IPA pronunciations, just like Module:fr-pron does for French words. German pronunciations seem to follow certain fixed rules, at least for Hochdeutsch, but certainly it would require quite some effort to write it. LinguisticMystic (talk) 11:06, 28 August 2020 (UTC)
 * But you wouldn’t know before pleongraphs whether the vowel is the long or the short one. And it can’t know the stress in prefixed verbs where the prefix can be either separable or inseparable, like . is lower than in French but with these things there is this more left unexpressed in the script. Fay Freak (talk) 11:23, 28 August 2020 (UTC)
 * Even languages with the "shallow orthography" often require stress marks, additional symbols and respellings, such as Erde as "Ehrde". --Anatoli T. (обсудить/вклад) 12:47, 28 August 2020 (UTC)
 * Which means, that such a module would be very hard to implement, or nigh impossible, I guess. LinguisticMystic (talk) 23:36, 28 August 2020 (UTC)
 * Not impossible. I implemented a similar module for Old English, which has many of the same concerns as German. As Anatoli notes, you would need respellings to accommodate unpredictable elements. This would include long vowels where short vowels are expected (indicated by adding a macron over the vowel or an h afterwards), short vowels where long vowels are expected (indicated by adding a breve over the vowel or doubling the following consonant), unpredictable stress due e.g. to prefixes like um- or loanwords (indicated by an acute accent over the appropriate vowel), secondary stress (indicated by a grave accent and/or by adding a hyphen to divide components of compound words), glottal stops in the middle of words, etc. In Old English respelling, for example, I used < to separate an unstressed prefix that can't be predicted, > to separate an unstressed suffix that can't be predicted, + to prevent prefix separation that would normally occur, and - to separate components of compound words that receive individual stresses. More respellings are required than in French, for sure, but it's not impossible. Benwing2 (talk) 01:55, 29 August 2020 (UTC)
 * I see the main problem being education about the need for respellings. There have been far too many bad edits consisting of replacing hardcoded IPA with empty language-specific IPA templates. Module errors are one option, but the worse offenders are mass-editors (often IPs) who don't preview and don't look at the entry after they click "Publish changes". There's also using the technique by which taxlink alerts editors that there's already an entry for the taxonomic name, but there again, it only shows in preview. The question then becomes whether we're creating an attractive nuisance, like an unprotected pit that pedestrians fall into if they're not paying attention. Chuck Entz (talk) 02:26, 29 August 2020 (UTC)
 * The situation differs per language, depends on how many conscientious and thorough editors exist per language. Manual IPA is no guarantee for errors, either. We do correct such bad edits when we see them. Modules can trap a few errors or slack editing, like headwords requiring accents, the same could be done for pronunciation modules. German is not very different from many languages already automated and it has a high level of predictable pronunciation. Like Slavic languages, it may require stress marks, a list of prefixes, which may determine the stress, respellings, marks for morpheme or component word boundaries.
 * Module:de-IPA/testcases already uses orig. (German is known to retain partial or full pronunciation from source languages.)
 * There hasn't been much support from native speakers, the obstacle being variations in pronunciations, even for Hochdeutsch. I personally think it would be good to choose a variety/standard and stick to it for the purpose of module development. Regionalism/versions could be added later, if they are really needed. --Anatoli T. (обсудить/вклад) 02:53, 29 August 2020 (UTC)
 * Oh, I've just realized that you already started to work on this as an experiment three years ago ... LinguisticMystic (talk) 13:24, 29 August 2020 (UTC)

Category:Coinages by language
A lot of people forget to use the  parameter for people who have only ever made one coinage, causing a bunch of bot-created categories that will only ever have one entry (in which case they shouldn't exist). Conversely, it is possible that people may use  in error, not knowing that somebody does in fact have multiple coinages to their name. Would it be possible for you to do regular bot runs to add or subtract this parameter as necessary, and delete the empty categories that will result? —Μετάknowledge discuss/deeds 02:39, 30 August 2020 (UTC)
 * Apologies, I'll add it to the list of tasks to do, I have a bit of a backlog now :) Benwing2 (talk) 02:42, 30 August 2020 (UTC)
 * Oh, don't apologise — it's due to human error, not bot error. But it would be good for this to be something that gets run on a regular basis. —Μετάknowledge discuss/deeds 02:43, 30 August 2020 (UTC)
 * I was just reminded of this. No rush, as before, but a reminder in case you forgot too. —Μετάknowledge discuss/deeds 03:55, 1 October 2020 (UTC)
 * Your regular reminder. —Μετάknowledge discuss/deeds 20:56, 21 November 2020 (UTC)
 * This time I'll actually get to it :) Benwing2 (talk) 21:17, 21 November 2020 (UTC)
 * This is running. It output some warnings that you should probably look at:


 * Page 159 star-crossed: WARNING: Lang en, coiner William Shakespeare has 6 total words coined but has nocat=1:
 * Page 256 brave new world: WARNING: Lang en, coiner William Shakespeare has 6 total words coined but has nocat=1:
 * Page 258 commonty: WARNING: Lang en, coiner William Shakespeare has 6 total words coined but has nocat=1:
 * Page 376 primrose path: WARNING: Lang en, coiner William Shakespeare has 6 total words coined but has nocat=1:
 * Page 531 cold fish: WARNING: Lang en, coiner William Shakespeare has 6 total words coined but has nocat=1:
 * Page 647 all the world's a stage: WARNING: Lang en, coiner William Shakespeare has 6 total words coined but has nocat=1:
 * Page 684 Ailurus: WARNING: Lang mul, coiner Frédéric Cuvier has 2 total words coined but has nocat=1:
 * Page 721 Ailurus fulgens: WARNING: Lang mul, coiner Frédéric Cuvier has 2 total words coined but has nocat=1:
 * Page 101 ambivalence: WARNING: Lang en, coiner Eugen Bleuler has 2 total words coined but has nocat=1:
 * Page 215 autism: WARNING: Lang en, coiner Eugen Bleuler has 2 total words coined but has nocat=1:
 * Benwing2 (talk) 05:48, 22 November 2020 (UTC)
 * Thank you! I think it would be helpful to both add and subtract the nocat parameter in an automated way; these should all have it removed, and I don't think a human needs to look over this kind of report in general. —Μετάknowledge discuss/deeds 06:16, 22 November 2020 (UTC)


 * Relatedly, can we get the coinage cats to sort by last name instead of first name? It looks sort of amateurish this way, e.g. at Category:English coinages. —Μετάknowledge discuss/deeds 04:44, 20 December 2020 (UTC)
 * Done. I purged all the subcategories of 'English coinages' so they get re-sorted by last name, but it will take a little while for subcategories of other languages to get sorted by last name. Benwing2 (talk) 04:56, 20 December 2020 (UTC)
 * Thank you, looks good. —Μετάknowledge discuss/deeds 04:59, 20 December 2020 (UTC)

Request for creating new parameter
Hello! Sometimes we need to quote only the title of a work to illustrate words, and in such cases, it is desirable, while using the or the  template, to hide the colon that appears at the end of bibliographical informations, as we are not giving the passage in such cases. So, could you possibly create a parameter something like nocolon to that effect? Thank you. —inqilābī [ inqilāb zindabād ]  10:58, 31 August 2020 (UTC)
 * My two cents: (1) try not to use just a title as an example wherever possible; (2) just repeat the title as the text to be quoted using the passage parameter, and make the cited word boldface there. That way, there won't be unusual boldface in the title of the work. — SGconlaw (talk) 12:10, 31 August 2020 (UTC)
 * Thanks for your suggestions. I myself have used a title as an example only once, but I have seen quite a few ones that other users have added earlier (such as this); and I have seen that they did not use any templates: so, would it not be better to not use templates while quoting titles, as a solution to the problem? But I do not like the idea of repeating the title as the passage, because that would give the impression that the author themself wrote that way. —inqilābī [ inqilāb zindabād ]  11:13, 1 September 2020 (UTC)
 * My main point is the first one – titles are not really good illustrations of terms as they are usually very short and don’t provide enough context of the terms. Sometimes they do no more than repeat the term, so I don’t see much value in them. — SGconlaw (talk) 11:38, 1 September 2020 (UTC)

, do you endorse the idea of having a new parameter nocolon for this purpose (as stated in the first paragraph)? —inqilābī [ inqilāb zindabād ]  11:13, 1 September 2020 (UTC)
 * I also wanted such a thing, albeit not because of quoting titles but to give other editions/loci containing the same text, although it would also be useful for the aforedescribed purpose because I often refrain to type off a whole passage and link to a scan instead (transcribing the editions is extra work that begins at even deciding where to begin the quote if it is a medieval text not using punctuation, or worse if it is a tablet in a fragmentary state); for the purpose of multiple loci the template should also not cause a line break. Fay Freak (talk) 20:21, 1 September 2020 (UTC)
 * So d'you want this to be ta'en to BP? —inqilābī [ inqilāb zindabād ]  11:44, 2 September 2020 (UTC)
 * For colon and line break options? Why, this is an uncontroversial functionality extension. It just enables to do the things we do anyway more orderly. Fay Freak (talk) 11:50, 2 September 2020 (UTC)
 * I meant the parameter to hide the colon: let us see if the community supports this proposal... But I did not get what you mean by "line break" & "multiple loci". —inqilābī [ inqilāb zindabād ]  11:59, 2 September 2020 (UTC)
 * If you use two times for the same quote – because it is found in various editions –, then the first breaks the line. See for example, or the hack at . Fay Freak (talk) 12:08, 2 September 2020 (UTC)
 * What is the need to provide multiple editions? Never seen such a thing before. Do not take so much trouble: just give the earliest or the major edition. —inqilābī [ inqilāb zindabād ]  15:43, 2 September 2020 (UTC)
 * Readings, variants? Not everyone has the same edition? It is good for SEO? Wrong question. It is the superobligatory. There is no need, but we like to do things better or more than is required. It’s free! also often adds bibliographic information to the quotation templates which are “not needed” but he does it because he can. Fay Freak (talk) 16:05, 2 September 2020 (UTC)
 * Providing multiple editions and providing full bibliographical informations are wholly otherly things. The latter is meedful, the former is frowned upon. —inqilābī [ inqilāb zindabād ]  16:27, 2 September 2020 (UTC)
 * if there's a need for some reason to state more than one edition (for example, a reprint of a first edition), you can just use the additional parameters ending with a " 2 " (title2, location2, publisher2, etc.) as described at ; it isn't necessary to use a new quotation template. But note that "Wiktionary:Quotations" states: "The year should be that of the earliest edition known to use the word. Where feasible, the page number should be taken from the first edition, but if a later edition is used (found in a library or digitised by Google Books], etc), then the publication date should be added in parentheses after the publisher’s name. In these cases, publication details should reflect the work actually cited: do not give the name, location etc. of the publisher of the first edition if you are not citing it directly." — SGconlaw (talk) 17:43, 2 September 2020 (UTC)
 * This is all aimed at English quotes. There technically isn’t a first edition if the text circulated in manuscript, no reprint of works written before the age of printing (which is for Arabic-script lands before 1800). The year in year should of course be that of the composition – which is often approximated –, not the publication year of the earliest edition as claimed by Quotations, that is also year_published (if the first edition is given). And usually modern Arabic editions of medieval texts are better than the 19th editions of Orientalists who had few manuscripts, which are however nonetheless quoted, but the editions stand side by side. Fay Freak (talk) 18:06, 2 September 2020 (UTC)
 * gotcha, in which case you can still use the “ 2 ” parameters as I mentioned. That would avoid the problem of the extra colon and line break in your scenario. — SGconlaw (talk) 18:47, 2 September 2020 (UTC)

Module:ru-noun
Your recent edit seems to have broken некого and нечего (sounds like material for a -style routine- нечего исправить, perhaps?), or at least exposed a problem that was already there. Chuck Entz (talk) 13:51, 15 September 2020 (UTC)
 * Yeah, you are right, it exposed an existing problem. I fixed this issue and also the issue of the lemma showing up as — in the title of the table. Benwing2 (talk) 02:10, 16 September 2020 (UTC)

Yet another fix for Module:la-verb
An adjustment is needed for the particular case of. The present header,, is an oversimplification; the supine  does exist, and the verb is also optionally semi-deponent—ōsus sum was used for  (e.g. Plautus, Amphitryo 900). Therefore, the header should be. This, however, creates problems: disappeares and  is shown to be in the same tense as. I honestly don’t know how it should be displayed in the end (whether the present infinitive should still be displayed or instead of it and ). Thorny stuff. --Biolongvistul (talk) 14:34, 15 September 2020 (UTC)

Sanskrit reverse transliteration
Hello! I was trying to find out how to combine these two modules, Module:sa-utilities/translit/IAST-to-SLP1, and Module:sa-utilities/translit/SLP1-to-Deva into one and through a template get a reverse transliteration of a Latinized Sanskrit word, so far unsuccessfully. Could you help me out? It would be quite useful to have it, as I have a lot of data in IAST, which could be used here in their Devanagari forms. LinguisticMystic (talk) 12:17, 24 September 2020 (UTC)
 * I actually found it out by myself. Thank you. LinguisticMystic (talk) 22:26, 24 September 2020 (UTC)

Your bot changed plural to s
Revision. J3133 (talk) 06:39, 26 September 2020 (UTC)
 * That was intentional, but the module had an error handling s and es when there is a qualifier. Fixed. Benwing2 (talk) 07:01, 26 September 2020 (UTC)

Special:Contributions/213.166.146.171
Hey, you know some Arabic. Is this IP actually trying to contribute or just messing with stuff? It looks like the latter to me, but I don't know what I'm talking about. —Globins (yo) 05:09, 2 October 2020 (UTC)
 * Most of the changes by this IP look fine, but not the change to . I don't know why the IP keeps edit-warring over this. The vocalization ending in -awn is not normal in Arabic, for sure. Benwing2 (talk) 05:18, 2 October 2020 (UTC)
 * I protected the page to prevent further edit warring. Benwing2 (talk) 05:21, 2 October 2020 (UTC)
 * @Benwing2, : The IP may be right. The reading is definitely supported by H. Wehr dictionary. Just try putting صهيون in the search window on on this site. Applies to the derivation, for which @Fay Freak wants protection as well. --Anatoli T. (обсудить/вклад) 06:15, 2 October 2020 (UTC)
 * Interesting, thanks for pointing this out. Pinging User:Fenakhay. Benwing2 (talk) 06:24, 2 October 2020 (UTC)
 * Thanks. User:Fenakhay also reverted the IP's edits on, although the dictionary suggests the reading must be valid (maybe rare or dated but valid). --Anatoli T. (обсудить/вклад) 06:55, 2 October 2020 (UTC)
 * Thanks for the ping. That vocalisation is certainly not common at all, and I haven't encountered before. I'll readd it. فين أخاي (talk) 13:56, 3 October 2020 (UTC)
 * Since one does not learn it systematically, and I have only become cognizant of it because of adding many of these words, I note that there is a pattern where Standard Arabic allegedly has KiLMawN or KiLLawM, and the classical dictionaries list only this, but the most real pronunciation, which the Saudi IP despises and which you virtually always hear when someone on TV speaks Modern Standard Arabic, is KaLMūN or KaLLūM. Other examples include, , , ,  (the last word is not found often so I cannot speak of its real pronunciation, but of Zionists one speaks often); with a remarkable difference . You may see that there is no word with such a pattern which is not to be deemed a borrowing. In some other words like , , ,  for some reason that KiLMawN or KiLLawM pattern is somehow never mentioned. On other occasions, and more often, the alleged norm is KuLMūN or KuLLūM as opposed to KaLMūN or KaLLūM, e.g. . Which works analogously with , as in  instead of  which latter the Saudi IP hates much, , . Due to borrowings in other languages and the source forms it is also certain that the allegedly colloquial forms have always been of greatest use. I wonder in which contexts one could actually and unironically hear the alleged norm forms. Fay Freak (talk) 12:44, 8 October 2020 (UTC)

Female equivalents
While you're on these I have a request. Can you remove m parameters from the headword line on these pages? I think the redundancy at e.g. mercière is pointless and distracting. Ultimateria (talk) 06:37, 3 October 2020 (UTC)
 * Sure, I'll do that afterwards. Benwing2 (talk) 06:38, 3 October 2020 (UTC)
 * Thanks. Also, I cleaned up the ~120 Portuguese female equivalents that didn't have, but the accelerated entry script will keep creating them. Can you take a look at it? Ultimateria (talk) 08:17, 4 October 2020 (UTC)
 * This should be fixed. I added tocadora as the feminine of tocador (not sure this is correct though), and when you click on it you get the headword with and the definition with . Benwing2 (talk) 03:01, 5 October 2020 (UTC)
 * BTW I still have to fix the feminine plural accelerator. Benwing2 (talk) 03:02, 5 October 2020 (UTC)
 * Feminine plural accelerator should be fixed; it now generates instead of . Benwing2 (talk) 03:19, 5 October 2020 (UTC)
 * Excellent, thank you! Ultimateria (talk) 16:33, 5 October 2020 (UTC)

Can you replace instances of in noun sections? . Ultimateria (talk) 21:04, 8 October 2020 (UTC)
 * Done. Benwing2 (talk) 02:28, 9 October 2020 (UTC)

Do you still plan on removing m parameters? To be clear I'd only like them removed from sections whose only definition line includes. Ultimateria (talk) 18:24, 16 November 2020 (UTC)

Renaming of Kurdish entries
You've renamed Kurdish entries to Northern Kurdish and by doing so you broke some links. See for example mereq. The link in "Please also see mereq in the Northern Kurdish Wiktionary" is wrong because it links to kmr.wiktionary.org. The Kurmanji Kurdish Wikipedia and Wiktionary use the ku code, not kmr and since it is just called Wîkîferhenga kurdî (translation: Kurdish Wiktionary. Just "Kurdish" is used in many other places referencing the Wikipedia or Wiktionary too) the "Northern Kurdish Wiktionary" is partly wrong. I know that the dialects differ a lot and perhaps specifying the dialect is good but the name "Northern Kurdish" is pretty uncommon (I know that it's not made up by you, it is used in other templates too). -- Guherto (talk) 19:14, 24 October 2020 (UTC)
 * I see, you're referring to Tbot links. I can fix them to use ku as the Wiktionary code. It will still be referred to in the entries as the "Northern Kurdish Wiktionary", which isn't completely wrong since that's what it really is. Meanwhile, if you think we should rename "Northern Kurdish" -> Kurmanji and/or "Central Kurdish" -> Sorani, feel free to bring that up in the beer parlour. Benwing2 (talk) 19:41, 24 October 2020 (UTC)
 * Okay, thanks. I will think about requesting it since almost all speakers call it kurmancî (Kurmanji). -- Guherto (talk) 19:57, 24 October 2020 (UTC)

Hindi
Just wanted to say again, thanks for all your work on declension and conjugation! Looking back now, it is seriously impressive that you managed to learn the whole noun and verb paradigms so quickly :) Really a huge improvement on what we had before. —AryamanA (मुझसे बात करें • योगदान) 03:52, 27 October 2020 (UTC)
 * You're welcome! I still need to add docs to, and . I've actually written up more docs privately than are on those pages, just need to copy the text basically. Benwing2 (talk) 03:54, 27 October 2020 (UTC)

Why prefer LINK to for non-english definitions?
Hi there Benwing2! I noticed that your bot changed some links from the format to the LINK format. I thought that when specifically the English definition of a word is being referenced as a definition to a non-English word, it would've been preferable to directly link to the English section of the entry, which is achieved by (or also LINK, I suppose). I was just wondering if you had any particular reasoning why it might be preferable to link to the top of the article instead? – Guitarmankev1 (talk) 12:20, 30 October 2020 (UTC)
 * Hi, this is a good question. The general practice I've seen across Wiktionary is to use bare links in English definitions except when the foreign lemma is spelled identically to the English definition, as e.g. on Welsh (in which case the bare link will display in bold and not be clickable). AFAIK the reason for this is that it's much easier to type a bare link, and it's visually less intrusive in the source code, and the English definition is already at the top of the page so it usually doesn't matter very much whether you link to the top of the page or to the English definition. I think there may have been votes on this but I'm not sure. I know that some editors don't even like using  links in synonyms/antonyms/derived terms/related terms sections of English entries, but here the practice of using  seems more established. Benwing2 (talk) 14:15, 30 October 2020 (UTC)

Request for mechanical fixing of several dozen misworded Jeju entries
Currently a lot of Jeju entries have "Sino-Jeju word from " in their etymology sections. This is pretty bad because first, "Sino-Jeju" is a term nobody uses either in Western academia or in Korea itself, and second, because Jeju does not have its own character readings that correspond to e.g Sino-Korean, Sino-Vietnamese, Sino-Japanese.

The correct way to handle these should be, because these words derive from Sino-Korean readings.

Could you set up a bot so the Wikitext sequence "Sino-Jeju word from " is automatically changed to "  "?

Thank you so much! :) --Karaeng Matoaya (talk) 18:32, 1 November 2020 (UTC)
 * Sure, I'll see if I can get to this today. Benwing2 (talk) 18:35, 1 November 2020 (UTC)
 * Apologies, it took me several days to get to this. I fixed things up, and also replaced Sino-Jeju -> Sino-Korean in some synonym sections. Can you review the pages with these latter changes? They are as follows:

Also, the following pages mention Sino-Korean in the Etymology section, which can maybe be cleaned up: Other pages whose Etymology sections need cleanup: Benwing2 (talk) 05:40, 9 November 2020 (UTC)
 * ᄋᆢᄃᆞᆸ
 * 넷
 * 다ᄉᆞᆺ
 * 두게
 * 둘
 * 마은
 * 설룬
 * 셋
 * 쉰
 * 시믈
 * 아은
 * 열
 * 예쉰
 * 일곱
 * 일흔
 * 궨당
 * 전상
 * ᄀᆞᆸ
 * ᄎᆞᆷ
 * ᆢ
 * ㅐ
 * ㅢ
 * ㆍ
 * 가사
 * 가차이
 * 골히
 * 드르
 * 복송
 * 비낌
 * 새베
 * 새위
 * 서월
 * 시다
 * 심방
 * 아시예
 * 얄룹다
 * 일홈
 * 장게들다
 * 줴께
 * Benwing2 (talk) 05:41, 9 November 2020 (UTC)


 * Having looked at those Jeju entries and tried to fix some of them (mostly by removing the false "from Middle Korean" etymology and just linking to Korean cognates instead), I am now of the strong opinion that Jeju should be merged back into Korean:


 * The etymology section for Jeju is unmanageable separate from Korean etymology sections, because 1) Jeju is only attested in the twentieth century but 2) Jeju is not directly descended from fifteenth-century Middle Korean, so all that every etymology section can really say "Cognate to Middle Korean blah blah; see Korean blah blah".
 * It is aesthetically very unpleasing to have Jeju above Korean in entries like . The definition "A Hangul syllabic block made up of ㄴ and ㅏ" should absolutely come at the top of the page.
 * The significant majority of Jeju lemmas are also attested in some form of Early Modern or modern dialectal Korean. So far the Jeju-Korean distinction works because 99.5% of our Korean lemmas are only from Contemporary Standard Seoul, but our coverage of the dialects and of the historical forms has been improving lately and eventually there will come a point when most Jeju entries have a Korean entry below them with identical pronunciation and identical semantics. This seems very unhelpful.
 * With just a single tweak for the Jeju vowel, this will allow us to use Module:ko-translit for Jeju forms instead of transliterations being manually inputted.
 * People in Jeju Island today speak a Jeju-influenced Korean, but a lot of these Korean words seem to have been added as Jeju entries (presumably because a lot of online resources fail to make this distinction). Merging Jeju into Korean solves this issue.


 * What do you think? Should I take this to Beer Parlor?--Karaeng Matoaya (talk) 09:52, 9 November 2020 (UTC)
 * Yes, definitely. Similar considerations led me to suggest merging Scots and English. Nearly all the distinctive Scots vocabulary is also attested in dialectal northern England English, and since we do have coverage of those dialectal terms we end up with a huge amount of duplication, which just adds a lot of unnecessary load onto editors. On top of this there aren't really any Scots editors to maintain the Scots terms in any case. In this case, someone else created a vote which was shot down; there was a lot of huffing and puffing about how Scots was obviously a separate language and to suggest merging them was to deny the linguistic identity of Scots, which in reality wasn't and isn't anyone's intention. But for Jeju vs. Korean you might get more traction since (a) it's not so familiar as with English vs. Scots so less likely to elicit emotional reactions, (b) there's already the example of the merged Chinese varieties. Benwing2 (talk) 15:03, 9 November 2020 (UTC)

Removing macrons from Latin words built on signum (and other words containing -ign-)
I started out trying to do this manually, but there are too many for me to be able to do it well. Could you please feed WingerBot a list of all the Latin words containing -sign- that are now marked with a macron (-sīgn-) to have it remove the macrons? It added them in August 2019, but the idea that these words had a long vowel is not well supported, and the general current consensus is that they didn't. Please also remove the macrons from any words built on other roots that we transcribe with -īgn-, e.g. from dignus.

Per W. Sidney Allen: "The change of e to i indicates a short vowel for an early period in ignis, dignus, lignum, signum, ilignus (cf. p. 23) and Romance evidence points to a short vowel at a later period in dignus, pignus, pugnus, lignum, signum (e.g. Italian degno, French poing)." (Vox Latina page 72)

Likewise, Carl Darling Buck writes in "The Quantity of Vowels before gn" that "For the cultivated language, which is what we aim to represent in our pronunciation and spelling a long vowel before gn is to be recognized only where it is long in origin, as, for example, in rēgnum." (page 314, The Classical Review, Vol. 15, No. 6 (Jul., 1901)).

--Urszag (talk) 18:45, 8 November 2020 (UTC)
 * Yeah I'm not sure why I added those macrons originally. My primary source has been here, and under section 38 they mention Buck's views, saying "Buck’s argument is a very strong one, and his conclusions deserve at least provisional acceptance. It should be noted, however, that three words, rēgnum, stāgnum, abiēgnus, being derived from stems with a long vowel, were legitimately entitled to their long quantity and always retained it." Are the macrons only on -sign- words and not on any other words with -ign-? If so, I was probably influenced by this evidence: sIgnum, CIL. vi. 10234; seignvm, xiv. 4270; sIgnificabo, vi. 16664;. But I will fix this. It would help if you could make a list of all the lemmas in question, or at least the classes of lemmas (e.g. "words in -sign-", "words in -dign-", etc.), so that I can look for them. Here is what appears to be the complete list of lemmas in -sign-:

adsignatio adsignifico adsigno antesignanus assignatio assignifico assigno bisignatus consignate consignatio consigno designatio designator designo in hoc signo vinces insigne insignificabilis insignio insignis insigniter insignitor insigno obsignator obsigno resigno signaculum signale signanter signarius signate signatim signator signifer signifex significatio significo signo signum subsigno
 * Benwing2 (talk) 20:35, 8 November 2020 (UTC)

Thanks. I found some words built on ignis that also need to be fixed: ignesco igneus ignifer ignio ignipes ignitus ignivagus --Urszag (talk) 21:24, 8 November 2020 (UTC)
 * This is running. Benwing2 (talk) 04:01, 9 November 2020 (UTC)

Primitive datapages
Sir, the primitive auto cat trials are going well at el.wikt. Thank you for your help, so much! If you could help me with splitting the data pages? I tried but failed as explained at :el:Module talk:yy/alldata. I do not understand why :( &#8209;&#8209;Sarri.greek &#9835; | 20:13, 8 November 2020 (UTC)
 * Sure, I'll take a look. Benwing2 (talk) 20:24, 8 November 2020 (UTC)
 * I looked at your code. I'm not exactly sure what you're trying to do, but I think what you want is this:

-- --                      GRAMMAR Γραμματική                             -- -- -- βλ. και Παραρτήματα γραμματικής = Παραρτήματα, Αμάρτυροι τύπoι = Βικιλεξικό local label = { -- Αμάρτυροι τύποι = βλ. Ετυμολογία ['Αντωνυμίες'] = { key = 'αντωνυμιεσ', parent = 'Γραμματικές κατηγορίες' }, ['Άρθρα'] = { key = 'αρθρα', parent = 'Γραμματικές κατηγορίες' }, ... }

return label

Benwing2 (talk) 03:51, 9 November 2020 (UTC)
 * Thank you . &#8209;&#8209;Sarri.greek &#9835; | 03:55, 9 November 2020 (UTC)

Could do you a bot run and replace occurrences of as follows?



seems to be an unnecessary duplication of. Thanks. — SGconlaw (talk) 13:02, 11 November 2020 (UTC)
 * Done. Benwing2 (talk) 02:13, 14 November 2020 (UTC)
 * Thanks! — SGconlaw (talk) 05:03, 14 November 2020 (UTC)

bring someone to book > bring to book
Hello Benwing. I don't oppose the redirect per se, but bring to book isn't the "standard" form: afaict, there is absolutely no consistency in the way we lemmatize such entries.

A proposal was made about a related issue, but it hasn't got anywhere yet.

In the meantime, I wouldn't delete the original page, because with a red link, someone is bound to create a duplicate entry, not knowing that we already have one at bring to book. PUC – 17:39, 12 November 2020 (UTC)
 * I made the change because when I looked at lots of English multiword verbal expressions, it became clear that the cases without someone are much more common than the cases with someone. Feel free to recreate the someone variant as a hard (or soft) redirect if you prefer. Benwing2 (talk) 05:11, 13 November 2020 (UTC)

and
Could you please do the following replacements?


 * And where man ended, the continued vest, Spread on his back, the houss and trappings of a beast.
 * And where man ended, the continued vest, Spread on his back, the houss and trappings of a beast.

→




 * When last I saw thy young blue eyes, they smiled.
 * When last I saw thy young blue eyes, they smiled.

→



Thanks! — SGconlaw (talk) 16:54, 15 November 2020 (UTC)
 * I did this as well as I could. There are a few usages still needing to be fixed manually. If you could make those changes, I'll delete the old templates. Benwing2 (talk) 19:55, 15 November 2020 (UTC)
 * Thanks, I did the manual fixes and deleted the redirects. — SGconlaw (talk) 19:07, 16 November 2020 (UTC)

Something wrong with Module:category tree
See Category:Xiang nouns. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 18:45, 15 November 2020 (UTC)
 * Actually it is in every category now. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 18:47, 15 November 2020 (UTC)
 * Thanks for letting me know; fixed. Benwing2 (talk) 18:57, 15 November 2020 (UTC)

Module:tl-pron/sandbox
Have created a sandbox version of Module:tl-pron, but there appears to be a bug when I'm trying to test it to rewrite the existing one.--TagaSanPedroAko (talk) 04:29, 16 November 2020 (UTC)

Hi, another one:


 * We are both love's captives, but with fates so cross, / One must be happy by the other's loss.
 * We are both love's captives, but with fates so cross, / One must be happy by the other's loss.

→



Thank you! — SGconlaw (talk) 18:34, 16 November 2020 (UTC)
 * Done. Benwing2 (talk) 04:11, 17 November 2020 (UTC)
 * Much obliged! — SGconlaw (talk) 16:11, 17 November 2020 (UTC)

Another one, when you have the time:


 * of which great worth and worship may be won
 * of which great worth and worship may be won

→



Thank you! — SGconlaw (talk) 16:11, 17 November 2020 (UTC)
 * Done. Benwing2 (talk) 05:41, 18 November 2020 (UTC)
 * Thanks again! — SGconlaw (talk) 08:40, 18 November 2020 (UTC)

Another duplicate template:


 *  Policying of cities.
 *  Policying of cities.

→



I'm not sure if all the quotations have been italicized using wikitext markup. — SGconlaw (talk) 16:37, 21 November 2020 (UTC)
 * Done. Benwing2 (talk) 20:05, 21 November 2020 (UTC)
 * Thank you! — SGconlaw (talk) 04:11, 22 November 2020 (UTC)

Greek categories
There is another problem with Module:category tree, this time only affecting Greek categories like Category:Greek verb conjugation group '-νω'. İʟᴀᴡᴀ–Kᴀᴛᴀᴋᴀ (talk) (edits) 00:43, 22 November 2020 (UTC)
 * Thanks; fixed. Benwing2 (talk) 01:08, 22 November 2020 (UTC)

Could you please replace these uses of ?


 * They discerned a body of five cornets of horse very full, standing in very good order to receive them.
 * They discerned a body of five cornets of horse very full, standing in very good order to receive them.

→



That will eliminate an unnecessary redirect. Thank you. — SGconlaw (talk) 22:03, 22 November 2020 (UTC)
 * Done. Benwing2 (talk) 23:39, 22 November 2020 (UTC)
 * Thanks! — SGconlaw (talk) 03:47, 23 November 2020 (UTC)

Categorization of coinages
Hello Benwing2, I noticed that you recently removed the code  from an instance of Template:coinage at autism and some other pages. Most of the removals look good and I thank you for making them. For autism and ambivalence, though, my understanding is that they are borrowings of German coinages and were not coined natively in English. Because of that I think keeping  is appropriate on the pages. I do think the equivalent German terms, Autismus and Ambivalenz should be categorized as coinages. Let me know what you think and I hope you the best. &mdash;The Editor's Apprentice (talk) 03:57, 24 November 2020 (UTC)
 * What you should do when you find these is to remove altogether; it's only to be used for coinages, not borrowings thereof. —Μετάknowledge discuss/deeds 05:00, 24 November 2020 (UTC)
 * Besides which, the only reason to use the template is for the categories. If you just want to display the text, type the text in directly- you get the same results without having to type 1, 1 and 1. Chuck Entz (talk) 07:46, 24 November 2020 (UTC)
 * I agree, and I had a similar realization a recently. I can't think of or find any examples at the moment, but I believe I have been doing so. &mdash;The Editor's Apprentice (talk) 18:14, 24 November 2020 (UTC)
 * Fair enough, I'll go ahead and do so. &mdash;The Editor's Apprentice (talk) 18:14, 24 November 2020 (UTC)

,, and
When you have time:


 * Preventive physic [...] preventeth sickness in the healthy, or the recourse thereof in the valetudinary.
 * Preventive physic [...] preventeth sickness in the healthy, or the recourse thereof in the valetudinary.

→




 * 1) * or
 * [passage]
 * [passage]

→




 * The doors of plank were; their close exquisite.
 * The doors of plank were; their close exquisite.

→



Then the unnecessary templates, , and can be deleted. Thanks. — SGconlaw (talk) 16:03, 5 December 2020 (UTC)
 * Done. Benwing2 (talk) 17:57, 5 December 2020 (UTC)
 * Thank you! — SGconlaw (talk) 13:44, 6 December 2020 (UTC)

Template:mk-phrase
Hello! Macedonian was not listed in Category:Phrase_templates, so I followed the Template:meta-phrase and changed the code in Template:mk-phrase. Now Macedonian is listed there, but the example in Template:mk-phrase/documentation is not shown correctly. Do you know how can we fix this? Thanks! --Горец (talk) 16:59, 4 December 2020 (UTC)
 * Hi. Why did you change it? It was working fine before and the previous implementation supported more features (multiple heads, id=). If the issue is placing it in Category:Phrase templates, that is easy to do without using, which is IMO not very useful (if you look at the history, it used prior to August, when I changed it to use Module:mk-headword). Benwing2 (talk) 18:01, 5 December 2020 (UTC)

Another replacement of a redundant quotation template:


 * Childish, unworthy dares / Are not enough to part our powers.
 * Childish, unworthy dares / Are not enough to part our powers.

→



Much obliged. — SGconlaw (talk) 18:59, 5 December 2020 (UTC)
 * Done. Benwing2 (talk) 20:44, 5 December 2020 (UTC)
 * BTW this change has caused a bunch of errors, e.g. on inflame, tappish, etc.: A big red "Unexpected < operator". Presumably this is because the volume or something isn't explicitly given. Could you look into this? Benwing2 (talk) 20:49, 5 December 2020 (UTC)
 * Yup, I noticed and fixed this. Now it displays a maintenance notice. — SGconlaw (talk) 20:50, 5 December 2020 (UTC)


 * The cion overruleth the stock quite.
 * The cion overruleth the stock quite.

→



The unnecessary template can then be deleted. Thanks. — SGconlaw (talk) 13:44, 6 December 2020 (UTC)
 * Done. There are a couple of cases that didn't get fixed up that need to be changed manually. Benwing2 (talk) 23:32, 6 December 2020 (UTC)
 * BTW same goes for some of the other orphaned templates I did previously; there are a few stragglers that need manual fixing. Benwing2 (talk) 23:38, 6 December 2020 (UTC)
 * OK, thanks. — SGconlaw (talk) 18:14, 7 December 2020 (UTC)

and
Could you please replace and  with ? Thank you. — SGconlaw (talk) 18:14, 7 December 2020 (UTC)
 * Done. Benwing2 (talk) 03:44, 8 December 2020 (UTC)
 * Thanks! — SGconlaw (talk) 03:35, 9 December 2020 (UTC)

and
Hello, could you please replace and  with ? Thank you. — SGconlaw (talk) 19:53, 8 December 2020 (UTC)
 * Done. Benwing2 (talk) 01:27, 9 December 2020 (UTC)
 * Thanks! — SGconlaw (talk) 03:35, 9 December 2020 (UTC)
 * I didn't realize they got deleted, lol. Feel free to replace them all again when I've finished up adding a whole bunch of Milton quotes, if you want. Oxlade2000 (talk) 19:56, 13 February 2021 (UTC)
 * Now, for example, I've finished Milton Oxlade2000 (talk) 11:09, 15 February 2021 (UTC)

, et al.
following from above, could you please carry out the following replacements?



Thank you! — SGconlaw (talk) 21:38, 17 February 2021 (UTC)
 * Done. Apologies, missed your previous request. Benwing2 (talk) 20:16, 21 February 2021 (UTC)
 * No worries. Much obliged! — SGconlaw (talk) 21:17, 21 February 2021 (UTC)

and
Hi, kindly replace:


 * with ; and
 * with

when you have the time. Thank you. — SGconlaw (talk) 17:47, 10 December 2020 (UTC)
 * Done. Benwing2 (talk) 02:50, 11 December 2020 (UTC)
 * Thanks! — SGconlaw (talk) 05:03, 11 December 2020 (UTC)
 * we now have 26 pages in Category:ParserFunction errors that weren't there before. I hope you have plans to fix that. Chuck Entz (talk) 04:31, 12 December 2020 (UTC)
 * I fixed them, we're now down to 1 page in that category. Benwing2 (talk) 05:24, 12 December 2020 (UTC)
 * Thanks! Chuck Entz (talk) 06:12, 12 December 2020 (UTC)
 * Yes, thanks, Benwing2! whoops, it looks like I updated one part of the template and forgot to update another part lower down. — SGconlaw (talk) 09:32, 12 December 2020 (UTC)

Hello, when you have time, please replace with. Thanks! — SGconlaw (talk) 16:20, 11 December 2020 (UTC)
 * Done. Benwing2 (talk) 05:29, 12 December 2020 (UTC)
 * Thank you! — SGconlaw (talk) 09:35, 12 December 2020 (UTC)

Hi, when you have time, please replace with. Thanks! — SGconlaw (talk) 13:46, 12 December 2020 (UTC)
 * Done. Benwing2 (talk) 19:53, 12 December 2020 (UTC)
 * Thank you! — SGconlaw (talk) 07:03, 13 December 2020 (UTC)

Category:Arabic redlinks
Hey! Is there a reason why this category is empty? — فين أخاي ( تكلم معاي · ما ساهمت ) 06:53, 13 December 2020 (UTC)
 * Yes. It's controlled by redlink category and language codes have to be added to a list- which I've now done. Chuck Entz (talk) 08:17, 13 December 2020 (UTC)
 * Thanks! Can you add ary and mt as well? — فين أخاي ( تكلم معاي · ما ساهمت ) 09:40, 13 December 2020 (UTC)
 * : You can also find lots of red-links (correct and incorrect translations) at User:Matthias_Buchmeier (under each English letter). Some SoP translations may need a split. --Anatoli T. (обсудить/вклад) 05:50, 15 December 2020 (UTC)

Maintenance category?
Hi,

Thanks for all the work. How difficult would it be to generate a maintenance category of entries that are simultaneously using ko-etym-sino and ko-IPA? The majority of such entries are missing a parameter in their etymology sections and need to be manually fixed.--Karaeng Matoaya (talk) 09:01, 14 December 2020 (UTC)
 * Never mind, I managed to do it :)--Karaeng Matoaya (talk) 13:29, 14 December 2020 (UTC)

Module:ko-etym
Hi,

Sorry to bother you again, but when you have the time, could you add a new parameter ("hangul=y") to Module:ko-etym so that ko-etym-native produces "In the Hangul script, first attested in..." instead of just "First attested in..."? There are a fair number of Korean words which are attested before Hangul, but the earliest Hangul form is always still important to note, typically because it's the first phonologically transparent orthography.

Additionally, could you have the module put the language as "Late Old Korean" instead of "Middle Korean" if the year is before 1300?

Thank you so much!--Karaeng Matoaya (talk) 00:56, 22 December 2020 (UTC)
 * Apologies for taking so long! I did the first part. For the second part, we should probably add an etymology-only language "Late Old Korean" that has Old Korean as its parent language. However, per Wikipedia, there is disagreement about whether to categorize the Korean period from 900-1300 as Old Korean or Middle Korean. I'm not familiar enough with Korean to know what to do here. Anyone else have comments? Benwing2 (talk) 22:29, 26 December 2020 (UTC)
 * Thanks so much!!
 * For the second part, I actually wrote the Wikipedia article on Old Korean, and while I tried to be neutral WRT the periodization question there, there really is a growing consensus in South Korean academia that the periodization up to 1300 is correct/more useful. The previous 900 periodization still holds weight IMHO mainly because of academic inertia, but it was formulated at a time when we knew much less about Old Korean than we do now—almost nothing, really. I've actually already added many thirteenth-century words as Old Korean entries, and explained the reasons for doing so on About Korean/Historical forms.--Karaeng Matoaya (talk) 00:28, 27 December 2020 (UTC)
 * OK, I went ahead and created Category:Late Old Korean with the code  and fixed Module:ko-etym appropriately. Let me know if you see any issues. Benwing2 (talk) 01:01, 27 December 2020 (UTC)

Template:categoryboiler
Hi, you deleted categoryboiler for being obsolete, but it appears on every empty category when one starts to create it. We should either restore the template or arrange it (somehow) that categories in the process of being created display something else. —Mahāgaja · talk 21:32, 26 December 2020 (UTC)
 * Hmm, didn't realize that. The template is becoming less and less useful; we should probably just display a message to use . Do you know where the Wikicode is that includes ? Benwing2 (talk) 21:34, 26 December 2020 (UTC)
 * No, I don't. And auto cat still doesn't work in every single case. —Mahāgaja · talk 21:35, 26 December 2020 (UTC)
 * You are right, but only mentioned five templates and all of them are subsumable under . Benwing2 (talk) 21:36, 26 December 2020 (UTC)
 * Let me see if I can find it by searching through the dump file. Benwing2 (talk) 21:37, 26 December 2020 (UTC)
 * MediaWiki:Newarticletext; will fix. Benwing2 (talk) 21:47, 26 December 2020 (UTC)
 * Fixed. Benwing2 (talk) 21:54, 26 December 2020 (UTC)

"Headword tables"
Re, maybe "headword lines" would be better. Headwords don't really have tables. —Mahāgaja · talk 12:39, 31 December 2020 (UTC)
 * Hmm, yeah I can fix that. I said "headword tables" by analogy with existing categories like Category:Latin terms with red links in their inflection tables. I didn't like categories like "Missing Welsh plurals" because it wasn't obvious whether "missing" meant "plural is a red link" or "plural is unspecified". Benwing2 (talk) 16:50, 31 December 2020 (UTC)
 * Why not CAT:Welsh terms with red links in their headword lines (rather than just nouns) and expand it to include other parts of speech? (I always disliked "missing plurals" too.) —Mahāgaja · talk 16:53, 31 December 2020 (UTC)
 * In the current setup, CAT:Welsh nouns with red links in their headword tables has CAT:Welsh terms with red links in their headword tables (not yet created) as its parent. I kind of like having different categories for different parts of speech, although I can definitely expand it to cover other parts of speech. Benwing2 (talk) 16:57, 31 December 2020 (UTC)
 * OK, having the nouns category as a daughter of the terms category is an even better idea. —Mahāgaja · talk 17:00, 31 December 2020 (UTC)

et al.
Hello, when you have time, kindly replace:


 * with ;
 * with ;
 * with ; and
 * with.

Thanks! — SGconlaw (talk) 11:12, 16 January 2021 (UTC)
 * Done. Benwing2 (talk) 00:39, 17 January 2021 (UTC)
 * Thank you! — SGconlaw (talk) 06:26, 17 January 2021 (UTC)

Module:th-pron
The mistake in this Module is that at one point it attempts to wrap a TD with a DIV, whereas it should be the reverse. As you are more experienced in writing Lua code, you might want to figure out how to reverse this behaviour, as doing so will considerably reduce the number of LintError's being generated.

In a specific example :-

in ลำไส้ใหญ่

what was seeminlgy generated for one row is :-

The class should be applied to the TR surely?, or the relevant class should be applied to the TR or TD not the DIV, and the DIV tag removed. I don't understand how this is rendered in the Lua, so perhaps you can make more sense of it than I can? ShakespeareFan00 (talk) 23:12, 17 January 2021 (UTC)

Request for using User:WingerBot
Hello. Recently, root-sa has been created to categorise Sanskrit terms by root (see CAT:Sanskrit terms by root). I was just asking that can you use your bot to add root-sa below the etymology sections of Sanskrit entries for words which are derived from roots (which have something like  in their etymologies)? Thanks, 🔥 𑀰𑀩𑁆𑀤𑀰𑁄𑀥𑀓 🔥 05:45, 18 January 2021 (UTC)
 * is completely unnecessary (and badly named, it ought to be if it were to need to exist). I fixed  so you can use it instead, e.g. instead of  just write . Benwing2 (talk) 06:18, 18 January 2021 (UTC)

Intended correct treatment of Latin gerunds
The pages for some gerund forms (e.g. addicendo) list the form under the "Verb" heading and treat it as an inflection of the verb lemma (e.g. addico), with the basic template to link back to this lemma. Others (e.g amando) list the form under the "Gerund" heading, treat it as a form of the gerund head form (in the accusative, e.g. amandum), and use the template to link them back to this gerund head form; in turn, the page for the gerund head form (again, e.g. amandum) sometimes uses the  template to mark this as the gerund head form, and the  template to generate a gerund declension table. Which of these is the intended correct treatment of Latin gerunds? Wewebber (talk) 06:04, 18 January 2021 (UTC)
 * I'm not really sure, but I think probably the former treatment is better because it's not obvious which case of the gerund should be basic. It feels strange to me to say that amandum is a "gerund" and the others are "gerund forms"; in reality, all of them are on equal footing. Benwing2 (talk) 04:49, 23 January 2021 (UTC)

Module italics and recognising escaped brackets?
Would the logic in Module:italics/sandbox be something like what was needed, or am I fumbling in the dark again? ShakespeareFan00 (talk) 20:00, 22 January 2021 (UTC)
 * Apologies, can you explain again exactly what the issue is? I never completely understood what the problem is you're trying to solve. Benwing2 (talk) 04:34, 23 January 2021 (UTC)


 * The 'problem' I was trying to solve was that of the italic brackets issue. In comments made by others in respect of  It was indicated that there was an issue with Module:italics failing to recognise that &amp;#91; and &amp;#93; were equivalent to [ ], when used as a replacemet. The code I attempted to add in Module:Italics was my attempt to add a recogniser for that situation.  However, this is only my third or fourth to modify Lua code, and thus I'd like a second view on the suitability of the code concerned.

ShakespeareFan00 (talk) 09:18, 23 January 2021 (UTC)


 * Apart from the Linter errors which are apparently being created (but which don't appear to lead to any visual effects), we've identified the following:
 * The current version of the template which uses square brackets creates a clash in quotation templates if used in combination with an external link, for example, if is used in chapter and chapterurl is used to apply an external link to the chapter.
 * This problem is solved if the brackets in are replaced with   and  . However, the quotation templates don't recognize these codes and so are unable to unitalicize the brackets and ellipses in titles. (I think this is handled by Module:italics.)
 * Typing " " within, for example , breaks . However, a workaround is to type "  instead. If we're going to implement the workaround, then we'll need your help to replace occurrences of " " within  and.
 * (Pinging for information.) — SGconlaw (talk) 10:45, 23 January 2021 (UTC)
 * (I did not receive the ping (signature and ping have to be added in the same edit); however, I do not have more information; see also Template:nb.../testcases.) J3133 (talk) 17:49, 23 January 2021 (UTC)
 * ah, I didn’t know that. Thanks. — SGconlaw (talk) 18:33, 23 January 2021 (UTC)

Module:form of/cats
Given that Module:form of/cats is restricted to template editors, and you're the one who worked on it the most recently, could you make some fixes to it? isn't assigning the proper categories to Middle English verb forms (see, , ); I think I've identified what needs to be changed in Module:form of/cats: Hazarasp (parlement · werkis) 06:32, 24 January 2021 (UTC)
 * At line 304, lua should be replaced with lua
 * Lines 306-317 need to be entirely rewritten; replace with: lua lua lua lua
 * I went ahead and made the changes. Let me know if you want to be a template editor; if so I can give you the permissions. Benwing2 (talk) 21:23, 24 January 2021 (UTC)
 * Middle English verb categories still aren't being assigned properly; I could tell you what I think's wrong, but it'd be easier if you gave me template editor permissions. Hazarasp (parlement · werkis) 06:43, 25 January 2021 (UTC)
 * Done. Benwing2 (talk) 07:33, 25 January 2021 (UTC)

Hello, kindly replace with. Thank you! — SGconlaw (talk) 17:27, 27 January 2021 (UTC)
 * Done. Benwing2 (talk) 05:52, 28 January 2021 (UTC)
 * Thanks! — SGconlaw (talk) 05:54, 28 January 2021 (UTC)

Hi, kindly replace with. Thanks! — SGconlaw (talk) 17:18, 30 January 2021 (UTC)

Actually, I realized there is already a more complete request at "Template talk:RQ:Buk Baibel" – please see that page. — SGconlaw (talk) 17:48, 30 January 2021 (UTC)

Hello, please replace with  when you are free. Thanks! — SGconlaw (talk) 16:22, 1 February 2021 (UTC)
 * Done. Benwing2 (talk) 02:45, 2 February 2021 (UTC)
 * Thank you! — SGconlaw (talk) 04:36, 2 February 2021 (UTC)

etyl
Hi, please don't use a bot to eliminate etyl. Bots don't know whether a term is borrowed or inherited or calqued or what. Thanks! —Mahāgaja · talk 00:18, 2 February 2021 (UTC)
 * Don't worry. All uses of were eliminated by hand; I reviewed every case and decided manually whether to substitute, , , , etc. The only purpose of the bot was to push the changes (that's what "manually assisted" means in the commit message). Benwing2 (talk) 02:43, 2 February 2021 (UTC)
 * Great, thanks! ☺ —Mahāgaja · talk 09:57, 2 February 2021 (UTC)

Module:form of and Module:form of/templates
seems to be putting things in incorrect categories; for instance it places in Category:Middle English first-person singular forms and Category:Middle English second-person singular forms, even though it isn't a first- or second-person (indicative) form. The problem seems to be that lua processes all tags inputted as one large lump when it should be processing each tag set individually. Can you look into this; I would've seen what I could do, but I'm otherwise occupied and I fear my Lua knowledge isn't quite up to scratch. Hazarasp (parlement · werkis) 09:06, 9 February 2021 (UTC)
 * I'm confused. is given as 1st-person singular indicative as well as 1st/2nd/3rd-person singular subjunctive, so it's correct to place it in both Category:Middle English first-person singular forms and Category:Middle English second-person singular forms. Can you explain how you intend for it to work? Benwing2 (talk) 01:59, 10 February 2021 (UTC)
 * Sorry, I probably wasn't clear enough; Category:Middle English first-person singular forms/Category:Middle English second-person singular forms/Category:Middle English third-person singular forms are only for the respective indicative forms; see the category descriptions (their names are misleading). All are defined in Module:form of/cats as requiring the tag, so   should not trigger them. Hazarasp (parlement · werkis) 11:27, 10 February 2021 (UTC)
 * I see. I'll try to look into this today or tomorrow. Benwing2 (talk) 04:07, 11 February 2021 (UTC)
 * I implemented this but it needs testing, which I'll probably get to tomorrow. Benwing2 (talk) 06:05, 12 February 2021 (UTC)
 * I pushed the code to fix this. However, there are various existing issues with Middle English verb forms, so you won't necessarily see what you expect. For one thing, the category spec for Category:Middle English first-person singular forms and such is specified as, so it will trigger on subjunctive forms as well as indicative forms. If you want it to trigger only on indicative forms, you should probably add the tag   to the spec; but it looks like not all existing verb forms include this tag (e.g. fole does but wones doesn't). An alternative, then, is to use  , but that won't trigger if there are additional tags like   present. You could use something like this:

{"or", {"tags=", {"1", "s", "pres"}}, {"hasall", {"1", "s", "pres", "ind"}}, "first-person singular forms" }
 * but IMO you're better off just fixing the verb forms to consistently include the  tag.
 * Another thing is the naming of the categories themselves. If you want the category to only include first-person singular present indicative forms, IMO it should be named Category:Middle English first-person singular present indicative forms, so that the name correctly reflects what's in the category.
 * A third thing is this spec:
 * This should now be changed back to
 * as you had it before; the latter now works, and the former doesn't.
 * If you want, I can make some of these changes by bot. Benwing2 (talk) 02:41, 15 February 2021 (UTC)
 * Thanks for doing this; to solve the continuing miscategorisation, I've chose to implement a solution entirely different from anything you've suggested (but of course which was made possible by your work!), so nothing should have to be recategorised (as far as I know); see my edits to Module:form of/cats. Hazarasp (parlement · werkis) 03:49, 15 February 2021 (UTC)
 * Yup, it looks like what you did should work. I would still recommend renaming the categories, though, to reflect their actual purpose. Benwing2 (talk) 03:58, 15 February 2021 (UTC)
 * Thanks for doing this; to solve the continuing miscategorisation, I've chose to implement a solution entirely different from anything you've suggested (but of course which was made possible by your work!), so nothing should have to be recategorised (as far as I know); see my edits to Module:form of/cats. Hazarasp (parlement · werkis) 03:49, 15 February 2021 (UTC)
 * Yup, it looks like what you did should work. I would still recommend renaming the categories, though, to reflect their actual purpose. Benwing2 (talk) 03:58, 15 February 2021 (UTC)

Hello, kindly replace with. Thanks! — SGconlaw (talk) 09:26, 9 February 2021 (UTC)
 * Done. Benwing2 (talk) 02:17, 10 February 2021 (UTC)
 * Thank you! — SGconlaw (talk) 13:55, 11 February 2021 (UTC)
 * And you can do it again when I've finished tagging all of them Oxlade2000 (talk) 11:09, 15 February 2021 (UTC)

Possible bot task
Most or all instances of "#*: {{quote" should read "#* {{quote". Changes like Special:Diff/61794231 could be automated. Vox Sciurorum (talk) 10:41, 10 February 2021 (UTC)
 * Actually, "#*: {{quote-" because {{tl|quote}} alone does need the indent marker. Vox Sciurorum (talk) 14:09, 11 February 2021 (UTC)

Module:en-headword
Have you seen the message I wrote on the talk page on 5 February? J3133 (talk) 06:20, 12 February 2021 (UTC)
 * Sorry, I got that ping but accidentally overlooked it. Fixed. Benwing2 (talk) 06:33, 12 February 2021 (UTC)

, et al.
Hello, please replace the following: Thanks! — SGconlaw (talk) 16:01, 14 February 2021 (UTC)
 * Done. Benwing2 (talk) 19:51, 14 February 2021 (UTC)
 * Thank you! — SGconlaw (talk) 19:53, 14 February 2021 (UTC)

"form of" module errors
It may not have been obvious due to Chinese terms flooding the category (whatever error it was has been cleared- now it's just the queue), but there are a half-dozen+ Welsh and Norwegian entries that have module errors in the inflection of template that seem to be due to your edits on "form of" modules. It's not a huge issue- I'm only bringing it up because I wasn't sure if you were aware of it. Chuck Entz (talk) 02:18, 15 February 2021 (UTC)
 * Thanks, I just saw them and pushed a partial fix; the remaining fix will come shortly. Benwing2 (talk) 02:22, 15 February 2021 (UTC)
 * Special:WhatLinksHere/Template:mn-verb form of Chuck Entz (talk) 05:38, 16 February 2021 (UTC)

. J3133 (talk) 03:36, 15 February 2021 (UTC)

Have you seen (2 February)? J3133 (talk) 05:30, 15 February 2021 (UTC)

Etymologies of German prefixed verbs
Hi Benwing2, your/your bot’s distinct etymologies for the separable and inseparable forms of German verb seem a bit dubious. Some possible sources for the entry expressly do not make this distinction (while others mention no etymology at all). Are there any sources supporting your distinction +  vs. + ? On the face of it (cf. referring to  for use in compounds) the two would appear to be the same. ―BlaueBlüte (talk) 06:25, 17 February 2021 (UTC)
 * This also affects, for example, recent edits to durchzogen and durchwachsen. ―BlaueBlüte (talk) 06:33, 17 February 2021 (UTC)
 * Most verbs that are both separable and inseparable already had this system of etymology. I simply systematized it across all such verbs. To me, this makes sense, because separable prefixes are fundamentally distinct words that just happen to be written attached to the verb in some circumstances, whereas inseparable prefixes are fundamentally prefixes that don't exist as distinct words. This is clearer on the one hand with prefixes like ver- and ent- and on the other hand with collocations like and  (formerly written  and ). It's true that for some pairs like ich durchziehe vs. ich ziehe durch the meanings can overlap, but in many they don't at all, as in ich umfahre vs. ich fahre um. Benwing2 (talk) 07:02, 17 February 2021 (UTC)
 * Without having come to a final conclusion on the issue, in my eyes the distinction between durch and durch- is artificial, but if we make it, we should be consistent and I agree with Benwing that the separable prefixes could be written as separate words, so Benwing's approach seems reasonable to me. I now almost regret that I've created the entries -fähig and -gleich, and am considering if it wouldn't be better to treat the words with these and similar elements as compounds with fähig and gleich etc. Same with durch- et al. --Akletos (talk) 07:29, 17 February 2021 (UTC)
 * +1 re: artificial and consistency. Although, consistency probably shouldn’t stop at how we analyze such verbs in terms of etymology, but also look at the entries for those prefixes. For example, the ‘prefix’ entry for claims to pertain both to the separable and inseparable case, and the ‘distinct word’ entry  pretty much mentions the meanings for both ich umfahre and ich fahre um (even more clearly so for ich umgehe (‘go around, circumvent’) vs. ich gehe um (‘go about, haunt’)). It would seem that if such a distinction in etymologies is to be made on the lemma level, further consistency issues will arise. I’m not sure I’m clear on the ‘written as a separate word’ part yet. Would the test for prefix vs. compound be whether the verb can be rewritten with an adverb/a preposition that is ‘more separate’ than a separable prefix? Ex.: ich umfahre das Hindernis ⟶ ich fahre um das Hindernis [herum] ich fahre den Baum um ⟶ *ich fahre um den Baum (except maybe poetic) (a similar test could be devised for  and : widerstandsfähig ⟶ zum Widerstand[e] fähig) At any rate, maybe the etymology for separable and inseparable forms could be explained distinctly in an overall less involved way if the etymology, while referring to the same lemmas as the compound elements, were more specific about the distinct (albeit perhaps artificially so) mechanism of compound formation. This used to be the case for (see erstwhile source; I can’t reproduce what the rendered page looked like at the time with templates expanded). Also I wonder if this should perhaps be discussed more widely—or maybe that has happened already? ―BlaueBlüte (talk) 08:38, 17 February 2021 (UTC)

Wingerbot's Latin macron edits from September 2020
Hi, Benwing2. I see that some people have discussed a couple issues with the Wingerbot edits to Latin macrons above (regarding vērnus/vĕrnus and -ign- words). I recently noticed was probably erroneously given a macron... Is *ḱorh₂- > *ko:r- a possible vowel compensation? Spanish has cuervo < cŏrvus at least, and the Alatius page with the discussion by Bennett also has cornīx without a macron. Is there also a way to get a list of the words WingerBot changed at the time (did it include all short -rn- words?)? I would like to try examining them... The discussion on vērnus/vĕrnus seems interesting.--Ser be être 是 talk/stalk 14:19, 21 February 2021 (UTC)
 * I think cōrnīx was an error on my part, I can't find any evidence for the long ō. The full list is here: User:Benwing2/latin-macrons It's a concatenation of 7 files that served as the direct input to my macron-frobbing script and you have to edit the source in order to make sense of it. Benwing2 (talk) 16:46, 21 February 2021 (UTC)
 * Ah, thanks! It's a lot of words. Looking quickly at the list I now see why you'd do it though, to improve the coverage of less obvious macrons. Some of these really aren't obvious, like the ū of rūrsum/rūrsus... or the ē of comprehendō > comprēndō. Of course they're vastly correct, and it's only cōrnīx and maybe a few others that annoyingly got through the cracks. Hm.--Ser be être 是 talk/stalk 00:08, 22 February 2021 (UTC)

Request for simple botting
Hello, when you have the time, could you run your bot to change all instances of "deriving transitive verbs" to "deriving active verbs" and "deriving intransitive verbs" to "deriving passive verbs" in Korean etymologies (there should be around fifty of both)? Thanks, and hopefully this isn't too much of a bother!--Tibidibi (talk) 14:47, 22 February 2021 (UTC)
 * Apologies, I overlooked this. I'm looking now and it seems you've already made all the changes? I only see one term with "deriving transitive verbs" in it and none with "deriving intransitive verbs". Benwing2 (talk) 23:33, 7 March 2021 (UTC)

Weirdness in CAT:E
Today 16 Georgian entries showed up there, and I can't figure out why. None of them has been edited recently, nor has any of the main forms pointed to by the first argument in the template. I went through the entire transclusion list of one of those entries: aside from one unrelated edit to Module:scripts/data on the 16th, absolutely nothing has been edited since your edits to the "form of" family of modules.

It just so happens that the list of those Georgian entries exactly matches the list of mainspace entries in Special:WhatLinksHere/Special:WhatLinksHere/Template:ka-form_of, Special:WhatLinksHere/Module:ka-form_of and Special:WhatLinksHere/Module:ka-form_of, so I can only guess that either someone did a null edit on one of those three, or the  edits in the queue from those February 14th edits finally hit one of those three. I've spent years patrolling CAT:E and checked the revision histories of hundreds of entries for patterns, but this is the first time I've seen such an obvious pattern so completely without anything in the revision histories to explain it. Chuck Entz (talk) 06:29, 24 February 2021 (UTC)

de-verb/conj
Just wanted to say thanks for your great work on this. Adding new German verbs is a now a piece of cake. – Jberkel 18:52, 25 February 2021 (UTC)

Hi, kindly replace with. Thank you. — SGconlaw (talk) 20:03, 2 March 2021 (UTC)
 * Done. Benwing2 (talk) 05:04, 3 March 2021 (UTC)
 * Thanks! — SGconlaw (talk) 05:15, 3 March 2021 (UTC)

Hi, kindly replace with. Thank you. — SGconlaw (talk) 15:08, 3 March 2021 (UTC)
 * Done. Benwing2 (talk) 06:34, 4 March 2021 (UTC)
 * Thanks! — SGconlaw (talk) 06:54, 4 March 2021 (UTC)

, et al.
Hello, kindly replace the following:


 * and →.
 * and →.

Thank you. — SGconlaw (talk) 19:22, 4 March 2021 (UTC)
 * Done. Benwing2 (talk) 20:55, 6 March 2021 (UTC)
 * Thanks! — SGconlaw (talk) 21:00, 6 March 2021 (UTC)

Kurdish cats
Hi! Hope you are doing well! Just came here to say maybe it's time to move "ku" to a family. Your bot keeps creating a lot cats like Category:Kurdish terms borrowed from Turkish even when there is only a single page, so we can get rid of this problem. Most of the pages in cat:Kurdish language right now are there since we don't know from which Kurdish language a word originates. Once we make a "Kurdish languages" code, we will be able to tag those pages appropriately. Thank you! --Balyozxane (talk) 04:04, 7 March 2021 (UTC)
 * I have done this. It may flush out some errors; let's be on the lookout over the next day for this. Most of the issues should already have been fixed, as I have done various bot runs to convert uses of 'ku' to either 'kmr' (Northern Kurdish) if the term is in Latin script or 'ckb' (Central Kurdish) if the term is in Arabic script. Any remaining issues will be things that were added using 'ku' since I did the last bot run. Benwing2 (talk) 18:48, 7 March 2021 (UTC)
 * OK, most of the remaining errors are in translation tables (which should use 'kmr' or 'ckb' as above) or in descendants tables. Since 'ku' is no longer a language, it can't be used in, even in a section header. I have changed them to write out 'Kurdish:' as raw text. The alternative is to avoid grouping the different Kurdish languages and put each one alphabetized appropriately, e.g. Northern Kurdish under N. Benwing2 (talk) 23:20, 7 March 2021 (UTC)
 * Note to Benwing and, since I suspect at least the former is not even acquainted with the possibility and the latter followed lead, that one also puts the borrowing arrow as raw text, or with , or because one has done it so often there is . Such as in front of “Hindustani” and ”Aramaic”. Now you could argue that unlike with Hindustani you put the arrow in front of the individual Kurdishes as there is not one borrowing but separate borrowings as there are multiple languages, but one can say alike that one just puts it once in front of the brackets like in mathematics (what is the equivalent idiom for ? In probably none due to the structure of common law, but else?) and it does not even depend on whether something is really one or multiple languages but it should just look nicer, on one level with the other mentioned languages in the same table and uniform. Sometimes one also puts a gloss at the top, if the meaning is the like. Fay Freak (talk) 13:48, 8 March 2021 (UTC)
 * I have a script to clear these errors that moves the 1 down to the individual languages, on the theory that they're individual languages and potentially each one could have borrowed it separately. If it's agreed to do it the other way, I'll write another script to undo the change. Benwing2 (talk) 16:26, 8 March 2021 (UTC)
 * Most of the errors are gone now, I'll tackle the remainder this eve. Benwing2 (talk) 16:31, 8 March 2021 (UTC)

, et al.
Hi, please help to carry out the following replacement when you are free:


 * → (remove the pageref parameter).
 * → (add an explicit passage parameter).

Thank you. — SGconlaw (talk) 14:02, 7 March 2021 (UTC)
 * Done. Benwing2 (talk) 22:49, 7 March 2021 (UTC)
 * Thanks! — SGconlaw (talk) 03:58, 8 March 2021 (UTC)
 * Allow me to add that Category:ParserFunction errors went from 39 to 1 because of your combined work (the one remaining is due to out-of-memory errors). Chuck Entz (talk) 03:55, 11 March 2021 (UTC)
 * Great to hear. Benwing2 (talk) 06:42, 11 March 2021 (UTC)

Spanish lemmas
Heads up, I am in the process of pushing a bunch of changes I made to Spanish lemmas. As these were mostly done manually there may be some errors. It is running overnight (UTC-6); if there are any errors I'll try to fix them in the morning if they are still around. Benwing2 (talk) 06:16, 8 March 2021 (UTC)
 * BTW the bug in acribillar appears to be an underlying bug in Module:compound, but I don't have time to fix it right now, have to go to sleep. Will fix tomorrow. Also, there are continuing to appear errors due to 'ku' appearing in ; I have a script to fix these as they come up but it will take a little while for them all to appear. Benwing2 (talk)
 * Why did you remove the ending period in Special:Diff/61966688? The definition looks like a pseudo-sentence, not a word, to me.  Vox Sciurorum (talk) 10:18, 8 March 2021 (UTC)
 * On the general theory that there shouldn't be periods in foreign-language defns. I agree it doesn't look perfect as-is; either we can put the periods back or add an argument to lowercase the initial a. Benwing2 (talk) 16:28, 8 March 2021 (UTC)
 * Another observation: When linking to a different page, en => is better than a => a because it skips over a potentially long table of contents.  On the same page the   form is better.  Vox Sciurorum (talk) 16:46, 8 March 2021 (UTC)

Hello, is it feasible to do the following bot edit?


 * The enemies of our happy establishment seem at present to copy out the piety of this seditious prophet, and to have recourse to his laudable method of club-law, when they find all other means of enforcing the absurdity of their opinions to be ineffectual.
 * The enemies of our happy establishment seem at present to copy out the piety of this seditious prophet, and to have recourse to his laudable method of club-law, when they find all other means of enforcing the absurdity of their opinions to be ineffectual.

→



In particular, all dates after 1 January should have the year 1716 instead of 1715; WF indicated the year incorrectly. (Addison published The Free-holder between December 1715 and June 1716.) Thanks. — SGconlaw (talk) 18:18, 8 March 2021 (UTC)

Kurdish language code
Benwing, where's the discussion to move the Kurdish language code  to a family code? -- 10:42, 9 March 2021 (UTC)
 * We had several discussions a few months back, in November, I think, and all the Kurdish editors were in agreement. See the Beer Parlour. I made almost all the changes at the time to convert 'ku' to 'kmr' or 'ckb', but never finished it, and a Kurdish editor (see just above) asked me now to finish the job, so I did. Benwing2 (talk) 15:27, 9 March 2021 (UTC)
 * I remember there being talk about moving pages from the Kurdish header, but I don't recall any discussion about making  a family code. If we make   a family code, we should have a   to replace all the instances of ku.  isn't a good solution to the problem. --  19:57, 9 March 2021 (UTC)
 * Hi wouldn't that imply those words are borrowed/derived in "proto-kurdish" rather than the individual kurdish langs? --Balyozxane (talk) 20:07, 9 March 2021 (UTC)
 * Yes, that would be implied. In the case exampled in the link, they certainly weren't borrowed from Aramaic multiple times. Using  would be the best one-to-one replacement for all such examples. --  20:53, 9 March 2021 (UTC)
 * Oof, I see WingerBot already replaced ku with  everywhere. Again, I don't recall any discussion on that choice. --  20:58, 9 March 2021 (UTC)
 * I recall a lot in the direction of removing the  code completely., . Fay Freak (talk) 21:13, 9 March 2021 (UTC)
 * I can certainly do a bot run to replace  with something else if needed, but I strongly disagree with a blanket replacement to  . Declaring that a borrowing happened in Proto-Kurdish is only appropriate if the term was borrowed 1000 or more years ago (or whatever the age of Proto-Kurdish is). This may be appropriate for some Aramaic borrowings but hardly for the majority of borrowings into Kurdish. It's much more likely IMO that terms were borrowed into a single Kurdish language and then diffused through the others, or borrowed simultaneously into multiple Kurdish languages around the time the originating term was created. In both cases a noncommittal "Kurdish:" header is much better than "Proto-Kurdish". I have no problem with using the arrow template, as Fay Freak suggested, before "Kurdish:" rather than separately adding 1 to each language; either way works fine for me. Benwing2 (talk) 02:42, 10 March 2021 (UTC)
 * It was likely borrowed in the Middle Iranian period from Middle Aramaic, so yes, Proto-Kurdish is probably appropriate. The scenario you're giving sound more like a much later Arabic borrowing. Just because we have a proto language code, doesn't mean it should be used for entries, i.e. Proto-Armenian. That said, if someone wanted to come along and start creating reconstructed entries using the proper sources and methodology, I wouldn't object. -- 06:02, 10 March 2021 (UTC)
 * For now Benwing's solution seems to do the job but I agree with victar, a "ku-pro" code might come handy. --Balyozxane (talk) 07:50, 10 March 2021 (UTC)
 * "For now" isn't a solution. If it's level in the descendants list, it needs a language code. -- 06:34, 11 March 2021 (UTC)
 * It does not. We already have several examples of this, e.g. "Hindustani" and "Punjabi" on مسجد and many other pages, "Northwest Iranian" and "Central Iranian" on Reconstruction:Proto-Iranian/káHmah, "Lurish" and "Caspian" on Reconstruction:Proto-Iranian/HyúHā, etc. etc. Benwing2 (talk) 06:40, 11 March 2021 (UTC)
 * Those are examples to be remedied, not exemplified. I haven't gotten around to cleaning up the branches of SWIr, and we haven't decided on a name for the Caspian branch. Also, some families are areal, not genetic, like Southeastern Iranian, and I agree that those should be limited to family codes. You reply to my last comment, but you haven't said why a Proto-Kurdish code shouldn't exist. -- 07:16, 11 March 2021 (UTC)
 * I never said a code for Proto-Kurdish shouldn't exist. Feel free to create ku-pro if you wish. What I said was I object strongly to a blanket replacement of Kurdish: in Descendants tables with Proto-Kurdish. This should *only* happen if the term was actually borrowed into Proto-Kurdish, not if it was borrowed at a later period and diffused through the various Kurdish languages. In other words, you have to go case by case deciding whether to use ku-pro. For Iranian terms that were inherited, of course it's ok to use Proto-Kurdish, but that's not the majority of the cases. I also disagree with the idea that every line in a Descendants table needs to use a language code, that seems just an arbitrary assertion on your part. Areal families, for one, should not have any associated code; nor should "Hindustani" even though it's a convenient grouping of what is essentially a single language. Benwing2 (talk) 07:41, 12 March 2021 (UTC)
 * If you have no objections to a  language code, I'll create it. As I said, areal codes are one thing, but there are cases where Proto-Kurdish did borrow terms into it, which surely applies to those borrowed from Middle Aramaic. --  04:50, 15 March 2021 (UTC)
 * Semi-related: is the format at بئر considered ideal? Personally, I'm not fond of how "Kurdish" is missing an arrow (making it look like an inheritance), and then there is an arrow before each Kurdish language (making them look like borrowings from "Kurdish"). It's not a problem in the sense that humans can easily intuit what we actually mean, but it still feels like we could be doing a better job of presenting this. —Μετάknowledge discuss/deeds 21:36, 16 March 2021 (UTC)
 * I see your point. I had my bot make the changes in this fashion but User:Fay Freak thinks it's better to put the arrow before the word "Kurdish:" (even though it's not a language) rather than before each language. If you agree, I can do a bot run to change things in this fashion. Benwing2 (talk) 03:41, 17 March 2021 (UTC)
 * I agree, but I think some kind of template is better than just typing in an arrow, especially for automated programs trying to read our etymologies. I also feel it might be useful to get more community engagement on this, rather than just running with my and FF's preference. —Μετάknowledge discuss/deeds 03:45, 17 March 2021 (UTC)

acribillar
This is an odd one: the problem is caused by the "LL." in the af lang2 parameter- if I change it to "la", the problem goes away. If etymology-only languages aren't allowed in such cases, the code should detect for it and give a real error message instead of Lua error in Module:compound at line 164: attempt to call method 'makeEntryName' (a nil value). Chuck Entz (talk) 15:57, 9 March 2021 (UTC)
 * Thanks for figuring out that the issue was the etymology language. I fixed the code so it can handle them correctly. Benwing2 (talk) 02:11, 11 March 2021 (UTC)

Hi, kindly replace:


 * His jaw was underhung, and when he laughed, two white buckteeth protruded themselves and glistened savagely in the midst of the grin.
 * His jaw was underhung, and when he laughed, two white buckteeth protruded themselves and glistened savagely in the midst of the grin.

→



Thank you. — SGconlaw (talk) 17:23, 10 March 2021 (UTC)
 * I fixed this issue and the previous Addison Freeholder issue except that it's too hard to change the date in the Addison Freeholder usages, because the date format is so varied. I would suggest you make the changes by hand; there are only about 30 usages. Benwing2 (talk) 06:42, 11 March 2021 (UTC)
 * OK, thanks! — SGconlaw (talk) 08:12, 11 March 2021 (UTC)

, et al.
Hello, kindly carry out the following replacements:


 * XYZ
 * XYZ

→




 * XYZ
 * XYZ

→



→


 * XYZ
 * XYZ

→




 * XYZ
 * XYZ

→



For, the number in the 2 position should be deleted as it is now meaningless. Thank you! — SGconlaw (talk) 16:36, 11 March 2021 (UTC)


 * Apologies, this got lost somehow. I'll get to this tomorrow probably. Benwing2 (talk) 04:58, 9 April 2021 (UTC)
 * Nudge. — SGconlaw (talk) 18:28, 20 May 2021 (UTC)

Detection of script by text
Hi, Benwing. For Sanskrit terms in other scripts (see Template talk:sa-sc), do you know any template (can you create one if not?), which simply detects and returns the script name/code of a particular text? I am looking for a template like देवनागरी which returns "Devanagari" or "Deva". Even a module would do, if you can tell me which of its parameter to invoke. I think this is possible because:
 * sa returns
 * which implies that "some module" has the ability to detect script by text. I tried really hard to find something like this, but I could not. Sorry for bothering you. 🔥 शब्दशोधक 🔥 04:43, 12 March 2021 (UTC)
 * If you know the language of the text in question, you can use  in Module:scripts/templates; you can invoke this from template code using  . This should return the script code  . If you don't know the language of the text, you have to call the module function   in Module:scripts from another module. I could create a template interface for this in Module:scripts/templates if you need it. Benwing2 (talk) 07:47, 12 March 2021 (UTC)
 * Thanks! But can you please tell me why isn't T:sa-noun categorising entries into CAT:Sanskrit nouns in Devanagari script? 🔥 शब्दशोधक 🔥 05:42, 13 March 2021 (UTC)
 * Relatedly, I was wondering about this for Ladino. As you can see in Category:Ladino headword-line templates, they currently require the script to be supplied, but they could be a lot smarter. —Μετάknowledge discuss/deeds 05:46, 13 March 2021 (UTC)
 * In order to get a category 'LANG POSes in FOO script' you need to use the cat sc parameter, e.g. adding Hebr in will add Ladino nouns to Category:Ladino nouns in Hebrew script. I'll add support for auto to autodetect the script of the headword. Benwing2 (talk) 08:03, 13 March 2021 (UTC)
 * OK, it looks like cat sc has a bunch of problems, so don't use it for now. I'll fix it shortly. Benwing2 (talk) 08:20, 13 March 2021 (UTC)
 * OK, it looks like cat sc has a bunch of problems, so don't use it for now. I'll fix it shortly. Benwing2 (talk) 08:20, 13 March 2021 (UTC)


 * ✅ If you can do something to make these (sa-noun, sa-adj, sa-adv, sa-pron, sa-noun form, sa-proper noun, sa-adj form) templates neater, please do so. Thanks. 🔥 शब्दशोधक 🔥 15:20, 13 March 2021 (UTC)
 * I implemented 1 in ; this automatically adds the correct 'LANG POSes in FOO script' category. I fixed the above Sanskrit headword templates to use this param. Metaknowledge, if you want the same thing done to the Ladino headword templates, I can do it too. Benwing2 (talk) 18:50, 13 March 2021 (UTC)
 * Wow, thanks a lot, it works perfectly now! 🔥 शब्दशोधक 🔥 04:32, 14 March 2021 (UTC)
 * Yes please! And if you're willing, it would be great if a bot could remove the parameter supplying the script once the templates can supply it automatically. —Μετάknowledge discuss/deeds 19:31, 13 March 2021 (UTC)
 * Done, see MOD:lad-headword/documentation for the parameters of the templates. Benwing2 (talk) 21:42, 13 March 2021 (UTC)
 * Looks excellent. Thank you! —Μετάknowledge discuss/deeds 21:46, 13 March 2021 (UTC)

Bot operation to change sa to sa-noun and others
Hi, will this be possible? I've seen many (espicially the older ones) Sanskrit entries use sa, which doesn't categorise by scripts, which is why this should be fixed. If this will be too tedious to do, can you do something like this to the template head: ? Thanks. 🔥 शब्दशोधक 🔥 12:15, 31 March 2021 (UTC)
 * Hi. This is easy to do by bot. However, what about the parameters that need to go into ? They can't be filled in very easily automatically. I can certainly provide you the list of verbs that are missing parameters to, but you'll have to either fill them in or provide me the correct parameters and I can fill them in. Benwing2 (talk) 01:35, 1 April 2021 (UTC)
 * See User:Benwing2/head-sa-verb. This is a list of all the cases that use . There are 56 entries. Each of them is in the form . Here,   is the current text of the headword, and   is currently a copy of this same text but you should change it to the appropriate call to  while leaving the   text alone. If you can do this, I will use my bot to push all these changes to the appropriate entries. This should be much easier for you to do than having to go through and edit and save each page by hand. Benwing2 (talk) 04:47, 1 April 2021 (UTC)
 * I did the same for nouns; see User:Benwing2/head-sa-noun. There are 334 entries here, but I've already converted to  on the   side and made a few other fixups, e.g. removing the   param if present, so all you need to do is review the existing entries and make any changes, e.g. adding missing genders and translits. Thanks! Benwing2 (talk) 05:30, 1 April 2021 (UTC)

Help with subcategorizing Latin adjectives of the third declension
Hello, thanks for all your work on the modules and templates for Latin. I would find it helpful to have categories for Latin third declension adjectives that have one ending, two endings, or three endings based on gender. I think the auto-categorization of Latin words based on declension type is currently handled by Module:la-nominal, but I don't understand the code there well enough yet to know how it could be changed to add subcategories "Category:Latin third declension adjectives of one termination", "Category:Latin third declension adjectives of two terminations" and "Category:Latin third declension adjectives of three terminations". Could you possibly help me with this?--Urszag (talk) 06:52, 14 March 2021 (UTC)

Pali สะวากขาตะ
The transliteration of this word isn't the same as the Latin script form, and pi-adj does not support explicit transliteration, so I had to resort to head. RichardW57 (talk) 17:36, 14 March 2021 (UTC)
 * How can the transliteration be different from the Latin script form? That doesn't make a lot of sense to me, although I'm not very familiar with Pali. Benwing2 (talk) 18:19, 14 March 2021 (UTC)
 * This form without implicit vowels is a spelling error for, but the misspelling is commoner than the correct form without implicit vowels. I'm not sure if the spelling difference implies a difference in tone for the second phonetic syllable (this form , but the correct spelling ) - perhaps  can advise on that issue.  I have found a Thai blog complaining about the misspelling.  The misspelling very occasionally turns up in the Roman script, but I don't think at a frequency often enough to record.  Besides, I have no durable quotation for the Roman script misspelling. Contrariwise, I don't recall finding a durable quotation for the correct Thai script spelling without implicit vowels. RichardW57 (talk) 20:03, 14 March 2021 (UTC)
 * Sounds like a very weird edge case. Instead of resorting to and a transliteration that just makes things even more mysterious, why don't you leave it using  and add a usage note explaining exactly what's going on, including the Latin translit and why exactly it differs from the Latin spelling?  doesn't generate a translit so it isn't wrong. Benwing2 (talk) 20:09, 14 March 2021 (UTC)
 * A usage note would be an odd place to put the transliteration. Doesn't an explanation of the misspelling belong in the etymology?  The explanation is that the epenthetic (I think 'svarabhakti' would be more precise) vowel in the Thai pronunciation of the onset (ancient /sv/) has been wrongly treated as a full vowel in the misspelling. RichardW57 (talk) 20:29, 14 March 2021 (UTC)
 * The problem is that no other Pali word even has a transliteration so it's far from clear what "transliteration" even means with Pali if not for the Latin spelling. I personally have no idea what the distinction is. So including a "transliteration" that's different from the Latin spelling is IMO worse than not including it; it's just confusing. Better to not include it, and include an explanation of what's going on (a usage note is generally the best place for such explanations, as a synchronic explanation is not an etymology; but it can go in the etymology if you prefer). Benwing2 (talk) 22:05, 14 March 2021 (UTC)
 * I've put forward a Pali transliteration module, but I've had no progress in getting it adopted (or sent back for improvement). There are interface issues for the Thai and Lao scripts - I occasionally need to know whether I'm dealing with an abugida or an 'alphabet'.  The glosses for non-Latin entries almost always say "xxx script form of blah", so it seems redundant to have the transliteration in the header, and without automatic translation, it is an invitation to error.  There is also a specific transliteration issue with Lao.  When it uses only the character set for Lao, d, dh, ḍ and ḍh are all represented by the same letter, ທ.  Do I transliterate these all as 'd'?  So far I've ducked the issue. RichardW57 (talk) 22:51, 14 March 2021 (UTC)
 * The point here is that the word "savākkhāta" does not exist in Latin script Pali. It is a Thai script error, induced by Thai phonology.  The correct form of the word is "svākkhāta".  (Remember that the West has gone 'native', and accordingly uses Pali in its own script, i.e. that of the West.  The East seems to have largely adopted European punctuation, though the Indians might be doing their own thing.)  I repeated my Google frequency determinations.  It turns out that, if I actually chase down the Google hits, correct and incorrect spellings get roughly equal numbers of hits in both systems, namely with and without implicit vowels.  Without implicit vowels, the correct spelling gets about 7 times as many raw hits as the wrong one; the preponderance seems to be due to multiple copies of the same text. RichardW57 (talk) 22:51, 14 March 2021 (UTC)
 * I have updated the entry to use pi-adj. RichardW57 (talk) 22:51, 14 March 2021 (UTC)

สะวากขาตะ is misspelling. As in สฺวากฺขาต or สวากขาตะ, สว is a cluster sv. You just can't add -a (or -ะ) inside it. Anyway, please create the entries สฺวากฺขาต and สวากขาตะ too.--Octahedron80 (talk) 23:57, 14 March 2021 (UTC)
 * The entry for already exists - I had a durably archived quotation in my bookcase.  I don't have a durably archived quotation for  or a good reason to bend the rules to add it, and without it, I wouldn't.  Fortunately, I've found it on p12 of a book, but I'm having trouble working out how to cite it. Fortunately, from the bar code 9789744974020 I was able to find it in a catalogue entry at http://www.thammasapa.com/product/459119/%E0%B8%A1%E0%B8%AB%E0%B8%B2%E0%B8%A1%E0%B8%87%E0%B8%84%E0%B8%A5%E0%B8%84%E0%B8%B2%E0%B8%96%E0%B8%B2-(%E0%B8%98%E0%B8%A3%E0%B8%A3%E0%B8%A1%E0%B8%AA%E0%B8%A0%E0%B8%B2).html, but I'm having trouble working out how to turn that into a citation.  Unfortunately, the sentence has an error in it - it starts "สวากขาโต ภะคะวาตา"  It's not the first time I've seen that spelling error in the second word.  Drilling down, I find 124 hits for the phrase, out of 1210 raw Google hits.  I don't think this misspelling merits documentation. RichardW57 (talk) 08:54, 15 March 2021 (UTC)
 * I've found a verifiable quote for yet another spelling of the word, "ส๎วากขาโต ภะคะวะตา ธัมโม". I hadn't seen yamakkan in alphabetic Pali before!  It's in the Wat Concord chanting book. RichardW57 (talk) 00:38, 20 March 2021 (UTC)
 * I've finally found and used a verifiable quotation for . RichardW57 (talk) 20:15, 20 March 2021 (UTC)
 * Octahedron80, you may have missed the ping. RichardW57, I still don't think the "transliteration" in the headword belongs there, whether you write it manually or use, since no other Pali entries have transliteration in the headword. Benwing2 (talk) 03:47, 17 March 2021 (UTC)
 * So how do you stop them from automatically appearing once a transliteration module has been accepted? Are you implying that there is a policy against having a transliteration module for Pali?  If so, I suggest you reply at my request on the Information Desk for Module:pi-translit to be adopted. RichardW57 (talk) 08:07, 17 March 2021 (UTC)
 * For an example of what I would expect to see, see the Azerbaijani word مطبخ. RichardW57 (talk) 08:32, 17 March 2021 (UTC)
 * For another case of discordance, see . It's possible that this Thai form is also a misspelling - .  RichardW57 (talk) 08:57, 17 March 2021 (UTC)

The +preo template
I mentioned in Grease pit/2020/October that the template +preo is marked experimental and not to be widely deployed. This did not get much response so I thought the template was, or should be made obsolete. But when I tried to replace the template with what I thought was more appropriate format, WingerBot undid the change, see the change history for denken. My main problems with the +preo template are that it is rather cryptic and that it puts grammatical information in a gloss when it would normally go in the label. Plus, as mentioned, it's marked experimental and it's undocumented so I don't know how to fix it if it comes out wrong.

There really should be some standard way to deal with prepositional verbs, but German, the language I'm mainly working on, has a number of variations on the idea and I not sure how a template can deal with them all rationally; is the preposition required or optional, are different prepositions possible, and in which case is the prepositional phrase if the preposition allows more than one? Plus I'm still not sure how to make the whole thing readable to someone who may not be terribly familiar with German grammar and doesn't want to have to figure out a string of cryptic symbols. I'd like to see some kind of consensus on the format first, then if possible have it implemented with templates. At least document the template, so someone other than the person who created it can use it correctly, before deploying it. --RDBury (talk) 02:26, 15 March 2021 (UTC)

Lang. code
Hi, again. I just thought you might be able to change  from an etymology-only language code to full-fledged language code, so that any Prakrit entries from now on will be "Prakrit" instead of "Maharastri Prakrit". The bot implementation can definitely wait until AryamanA has the time to do so. The ancestor of this new "Prakrit" should be Ashokan Prakrit -> Sanskrit -> PIA -> PII -> PIE. I'm not thinking of altering the already existing codes like, etc. into etymology-only-codes (it'll be done later). If this is done, I'll be able to make some of its basic templates. Thanks. 🔥 शब्दशोधक 🔥 03:23, 15 March 2021 (UTC)
 * No, there shouldn't be a general Prakrit language. -- 04:44, 15 March 2021 (UTC)
 * AFAIK, there was a discussion among the South Asian contributors who agreed to group all the Prakrits similarly to how the Chinese languages are currently grouped. This would imply creating an overarching Prakrit language, I think. I did not take part in this discussion and I forget where it occurred; User:SodhakSH can you link it? Benwing2 (talk) 06:01, 15 March 2021 (UTC)
 * Oof, something that big should go to vote. -- 06:07, 15 March 2021 (UTC)
 * Are you referring to Category talk:Prakrit languages? It actually started on my talk page (see this archived conversation), then was continued on the category talk page of Prakrit languages. 🔥 शब्दशोधक 🔥 06:16, 15 March 2021 (UTC)
 * Ordinarily I would agree with you but I think a lot of the Prakrit language codes were recently created so there isn't really any longstanding consensus on how to handle them. Benwing2 (talk) 06:33, 15 March 2021 (UTC)
 * Some codes should probably be done away with, but a Prakrit header? No, that shouldn't be the case IMO. -- 06:37, 15 March 2021 (UTC)
 * I think this should be up to the editors who actually contribute in this area. Benwing2 (talk) 06:38, 15 March 2021 (UTC)
 * While I think the idea of merging some of the Prakrits is warranted, it does seem like something that deserves a vote, especially since some Prakrits will be merged and some will not. Regardless, the current language code distribution is way too messy and seems to be hindering progress for MIA. —*i̯óh₁n̥C[5] 09:37, 15 March 2021 (UTC)

The vote has been created. 🔥 शब्दशोधक 🔥 02:19, 16 March 2021 (UTC)

aï and aü
Greetings! Can you do something to Module:typing-aids/data/sa so that aï changes into अइ on using sa? We all are currently working on Prakrit alternative forms in Devanagari script. I'll tell you why this is needed- see 𑀭𑀇 - on automatic Devanagari conversion, the spelling is given "रï" instead of the correct one "रइ". sa and sa already give अइ and अउ, so can you make it so that sa and sa are also able to? Thanks. 🔥 शब्दशोधक 🔥 02:48, 30 April 2021 (UTC)


 * Still not okay. sa now gives रि (ri) instead of the desired रइ (raï). 🔥 शब्दशोधक 🔥 03:00, 30 April 2021 (UTC)

Replacing old language codes with new ones
Some (hopefully small) tasks for WingerBot here:
 * 1) Wherever, in etymologies, there is , replace it with.
 * 2) Replace the following codes in etymology sections of entries:
 * by
 * by
 * by
 * by
 * by
 * by
 * Thanks, 🔥 शब्दशोधक 🔥 07:45, 6 May 2021 (UTC)
 * I can help but why are you suggesting replacing the language codes? If you are trying to convert full languages into etym-only languages, it's not necessary or correct to change the codes. You should keep the same codes; just fix the cases that won't work when they are etym languages (I can help you find those cases) and then directly make them etym languages and fix any errors generated. Benwing2 (talk) 03:02, 7 May 2021 (UTC)
 * Now Kutchkutch and Bhagadatta have already made the new codes. These older codes have a lot of lemmas so can't straightaway change them into etym-only. Once the lemmas of the old codes are cleared, we can simply remove them. 🔥 शब्दशोधक 🔥 03:05, 7 May 2021 (UTC)
 * IMO it's still not correct to change the codes. (Furthermore, the new codes are not ideal. It's preferred to use the first three letters of each language name in the code whenever possible, rather than a random subset of letters.) For example, when I changed the code 'ku' for Kurdish from a language to a family I did not need to change the code. Instead I used my bot to move all the lemmas to Northern Kurdish or Central Kurdish (depending mostly on the script). In this case, a bot could potentially move the lemmas to be under Prakrit. (I don't know if this is realistic, it depends on how much reformatting is needed and whether this can be done automatically.) Benwing2 (talk) 03:10, 7 May 2021 (UTC)
 * Still, we've started using these codes and I think these are just ok. Before too, such codes were there : Maharastri Prakrit = pmh, Magadhi (Indic) = inc-mgd, etc. 🔥 शब्दशोधक 🔥 03:17, 7 May 2021 (UTC)
 * That's not a good reason. See . I will wait to see what Kutchkutch and Bhagadatta say. Benwing2 (talk) 03:23, 7 May 2021 (UTC)
 * I know that moving the codes psu, pmh etc to the etymology only module was the simpler solution, but doing so without first changing all instances of pmh and psu to inc-pra etc would cause errors. I'd have suggested waiting for these entries to be converted to Prakrit first, but by then the update of the descendants and etymology sections of different languages were already underway, which made these etym only codes necessary. Now I would suggest first replacing all pmh and psu entries with inc-pra so that pmh psu etc have zero lemmas and THEN replacing the etym only codes inc-pse, inc-pmh etc with psu, pmh respectively. A lot fewer entries have inc-pmh instead of pmh so converting inc-pmh to pmh will be easier than converting pmh to inc-pmh. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 03:28, 7 May 2021 (UTC)
 * @Benwing2: Yes, the duplicate language codes are unnecessary in the long run. However, until a bot can operate on the entries affected by the merger, the duplicate codes are/were needed as a temporary measure to prevent errors while editors edit Prakrit-related terms with the merged Prakrit language. Prakrit editors could have waited until the bot operations until creating the merged Prakrit language, but waiting would lead to a loss of valuable time that could be used to improve Prakrit coverage.


 * The following codes are defined by :


 * wiki: Ardhamagadhi
 * wiki: Maharastri
 * wiki: Sauraseni


 * Thus, if is to be followed, then these codes would remain the same after being changed to etymology-only. However, as discussed at:
 * Wiktionary_talk:Hindi_transliteration
 * Beer_parlour/2021/January
 * many of the standards created by tend to arbitrary for the Indo-Aryan languages and are not indicative of actual usage.


 * If wiki and the initial wiki is disregarded, then the codes for the lects in would be:


 * wiki: Ardhamagadhi
 * wiki: Maharastri
 * wiki: Sauraseni


 * The codes for the lects not in would be:


 * wiki: Elu/Helu
 * wiki: Khasa
 * wiki: Magadhi
 * wiki: Paisaci


 * However, since wiki for Sauraseni for could be confused with Category:Sauraseni Apabhramsa language, perhaps the the initial wiki is needed. Kutchkutch (talk) 08:54, 7 May 2021 (UTC)
 * If you need to switch the codes of these lects from the original ones used as full languages, I'd strongly recommend using the codes suggested just above by @Kutchkutch rather than the "new existing" codes like . We can come up with a different scheme for Apabhramsa variants, maybe   or just  . I can help rename any uses of the "new existing" codes to these regularized codes. Benwing2 (talk) 05:12, 8 May 2021 (UTC)
 * Fine by me. First codes like pmh, psu, pka, inc-mgd etc in etymologies and descendants section will have to be replaced by inc-pra. Then, etym-codes like inc-pse, inc-pmh etc need be replaced by the new proposed codes like inc-sau, inc-mah and so on. Right? -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 05:22, 8 May 2021 (UTC)
 * I think you can replace things like 'inc-pmh' with 'inc-mah' now. Benwing2 (talk) 05:55, 8 May 2021 (UTC)
 * An alternative naming scheme is 'pra-ard' (Ardhamagadhi), 'pra-elu' (Elu/Helu), etc. Benwing2 (talk) 07:24, 8 May 2021 (UTC)


 * This one seems the best: pra-ard, pra-elu, pra-sau, etc. 🔥 शब्दशोधक 🔥 07:26, 8 May 2021 (UTC)
 * If wiki can be used instead of wiki, that seems better. Kutchkutch (talk) 09:01, 8 May 2021 (UTC)

Template:es-adj-inv
Hi, template guy. As you're currently working on a new Template:es-adj, I was wondering if you could work Template:es-adj-inv into it. That template is very simple, and only used for invariable Spanish adjectives. Ideally, the entries would have instead with the same output. Oxlade2000 (talk) 11:30, 20 March 2021 (UTC)
 * Hey WF. This is already implemented in and I will convert the uses of  shortly. Benwing2 (talk) 11:38, 20 March 2021 (UTC)

Template:es-noun
Also, Template:es-noun is no longer as functional as before. The template used to recognise that the plural of golpe de timón would be golpes de timón, but no longer does. My apologies if you weren't the one to break it! Oxlade2000 (talk) 11:39, 20 March 2021 (UTC)
 * I did break this, but I added new functionality to replace it. Use the value  for the plural to pluralize only the first word, e.g., and the value   to pluralize the first and last word, e.g.  for abanico aluvial plural abanicos aluviales. There's also   to pluralize only the second word but I'm not sure how useful this is. I added it originally for adjectives for expressions like más lento que el caballo del malo (plural más lentos que el caballo del malo). In adjectives you can write e.g.  to have it automatically generate the plurals and feminine based on only the second word, for example. I will document this all shortly. Benwing2 (talk) 11:48, 20 March 2021 (UTC)
 * IMHO, that's not an improvement to the template! Oxlade2000 (talk) 12:16, 20 March 2021 (UTC)
 * The problem is that the old algorithm was very fragile and didn't work well in a lot of cases, and the code was very hard to work with. It worked OK if there was a 'de' or 'a' in the word but in two-word expressions it was wrong much of the time and it was hard to predict how it would function. I prefer making it a bit more explicit which words need to get pluralized. Benwing2 (talk) 12:24, 20 March 2021 (UTC)
 * I restored the old plural handling of words with 'de(l), a(l)' in them, now also 'con' and 'por'. You should still use,   or   for multiword expressions without a preposition in them. Benwing2 (talk) 15:10, 20 March 2021 (UTC)
 * Also 'en' and 'para'. Benwing2 (talk) 15:21, 20 March 2021 (UTC)
 * Good job, Ben. You know modules much better than me, I'll leave them to you. Oxlade2000 (talk) 19:44, 20 March 2021 (UTC)

In light of your automated standardisations…
Hello. Since you have lately been doing bot edits to standardise the Etymology section of some language entries by inserting the terms “Inherited from …” and suchlike stuff, do you not think it would be worthwhile editing the templets, so that the full sentence appears: “Inherited / Borrowed from …”, just as other templets as , , etc. function? If that be acceptable, then the parameters nocap, notext could be revisited. Thanks! -⸘- inqilābī  ‹inqilāb·zinda·bād› 21:17, 7 April 2021 (UTC) P.S. Though, notext is now not really needed, given the said standardisation. -⸘- inqilābī  ‹inqilāb·zinda·bād› 21:23, 7 April 2021 (UTC)
 * At one point generated the text "Borrowing from ..." before the term. It was changed to remove the text to make it consistent with  and  and because people preferred to write "Borrowed from" instead of "Borrowing from". If we were to add the text back to  and  it would need to be discussed at the Beer Parlour to make sure people are on board with it. Benwing2 (talk) 03:23, 8 April 2021 (UTC)
 * Benwing, could you please see my revised proposal in the BP? And also could you address the concerns raised by Erutuon & -sche? Thank you. - ⸘ - dictātor · mundī  07:33, 19 April 2021 (UTC)

change to Schnaps
In [//en.wiktionary.org/w/index.php?title=Schnaps&diff=next&oldid=60898128 diff], English (with the right language name but wrong code) was changed to 🇨🇬 when it should've been 🇨🇬. I don't know if you have a way of telling whether the bot changed any other cases where a language name and code didn't match...? (Ideally there wouldn't have been [m]any such mismatches, but if there were, removing the language name will have made the potentially incorrect code harder to detect.) - -sche (discuss) 03:15, 8 April 2021 (UTC)
 * Thanks for pointing that out. That change was made by hand by me (in a text file containing all the German lemmas); the only thing the bot did was push the change to Wiktionary. (That's what "(manually assisted)" means in the changelog.) So it's unlikely there are very many more like this. I did changes like that fairly quickly, and I imagine I overlooked the context since "schnapps" looks like a German word, and just assumed the preceding text was wrong rather than the language code. Benwing2 (talk) 03:27, 8 April 2021 (UTC)

Minor issue with Accelerated Generation in Welsh
Hi Benwing! Hope everything is going well for you. Not sure if you're the right one to bring this up to, but maybe you can be some help.

I was on the page tlodaidd, and was about to created an accelerated entry for the softly mutated form "dlodaidd". However, when I went to do so, the auto-generated page incorrectly included that "dlodaidd" was also the equative degree. I believe this is because another accelerated link for "dlodaidd" in the page for tlodaidd, where it mentions the equitive degree as "cyn dlodaidd". However, "dlodaidd" alone is not an equative form, and is just the softly mutated form in the "cyn + _" equative construction. Compare with cyfforddus / cyn gyfforddus, and contrast with cyfoethog / cyfoethoced.

Also note that I haven't created dlodaidd yet, so you can try it out an see for yourself... Hopefully nobody rushes in to generated it before that happens!

This isn't a showstopper or anything since I can easily remove the lines stating that "dlodaidd" is the standalone equative, but I thought I ought to raise that it's an issue, in case it's a simple fix with a template somewhere.

Thanks! – Guitarmankev1 (talk) 14:04, 8 April 2021 (UTC)
 * Should be fixed now. Arguably the accelerator module itself should be smart enough to skip multiword forms when generating accelerators if only part of the form is linked, but I fixed it in Module:cy-headword. Benwing2 (talk) 03:23, 9 April 2021 (UTC)


 * Thank you Benwing! – Guitarmankev1 (talk) 12:57, 9 April 2021 (UTC)

Hi, when you have time, kindly replace:


 * with (the website has been renamed as it is no longer formally affiliated with the OED).

Thank you! — SGconlaw (talk) 20:22, 10 April 2021 (UTC)
 * Done. Benwing2 (talk) 04:44, 11 April 2021 (UTC)
 * Great, thanks! — SGconlaw (talk) 04:50, 11 April 2021 (UTC)

Category:Spanish verbs by consonant alternation et al
Thanks for these categories. I'm not sure what the previous category name for these were, but this looks pretty useful and user-friendly for Spanish learners. Yellow is the colour (talk) 22:01, 10 April 2021 (UTC)

Deleting etymology-only code wiki
The etymology-only wiki has been upgraded to Category:Old Punjabi language as per Talk:ਕਤੇਬ. Before deleting the etymology-only code wiki from Module:etymology languages/data, how do I check if it is still being used somewhere? Kutchkutch (talk) 08:04, 12 April 2021 (UTC)
 * You have to search through a recent dumpfile. See User:Benwing2/pa-old-usages. Alternatively you just delete the code and see where the errors pop up :) ... Benwing2 (talk) 00:39, 13 April 2021 (UTC)
 * There's a third option, but I'm a bit fuzzy on the details. You can use Special:Search with the  keyword. The tricky part is escaping any characters that have a function in the search syntax so you don't get false results. The idea is to search for the string "|pa-old" in the wikitext in various namespaces. You would have to use some other search syntax for the Template and Module namespaces, but I'm sure there wouldn't be much use of hard-coding for an obscure code like that. I would suspect that most of the usage would be in mainspace and the Talk, User Talk, Reconstruction, Reconstruction Talk and Wiktionary namespaces. See mw:Help:CirrusSearch for documentation on the search syntax. Chuck Entz (talk) 03:25, 13 April 2021 (UTC)
 * Yes, the code is so obscure that I'm the one who added almost all instances of it. I tried looking for documentation on the search syntax, but wasn't aware of mw:Help:CirrusSearch.
 * @Benwing2: Thanks for the dumpfile.
 * In any case, it seems safe to delete wiki now. Kutchkutch (talk) 09:11, 13 April 2021 (UTC)

Spanish templates
Template:es-prefix and Template:es-suffix probably don't need to exist either. Yellow is the colour (talk) 23:08, 12 April 2021 (UTC)
 * Same with Template:es-verb form of/adverbial. It's super-old, so might interest you Yellow is the colour (talk) 22:28, 13 April 2021 (UTC)
 * And Template:Spanish conjugation boiler and Template:Spanish conjugation type look like failed experiments, probably deletable too. Yellow is the colour (talk) 22:32, 13 April 2021 (UTC)
 * I can't delete Template:es-verb form of/adverbial as it's used on > 5000 pages. Template:es-suffix can eventually be deleted (just use, etc. and it will automatically categorize correctly) but before then we need to clean up the existing uses. Benwing2 (talk) 02:10, 14 April 2021 (UTC)

Declensions of Sanskrit consonant-stem terms
Hi Benwing. Can you do something so that sa-decl-noun-m, sa-decl-noun-f, and sa-decl-noun-n are able to decline a few more Sanskrit consonant-stem terms like those ending in -t (-at, -it, -ut, -ṛt, etc.) -d, -dh, -bh, -j, -c and more? If it is possible, I can help with how these words are declined. Thank you. 🔥 शब्दशोधक 🔥 04:35, 25 April 2021 (UTC)
 * Yes, I can look into this. I will respond to your pings tomorrow morning (it's late night where I am). BTW I hope you are staying safe, the news out of India now is pretty terrifying. Benwing2 (talk) 06:09, 25 April 2021 (UTC)

Can you look into this today? 🔥 शब्दशोधक 🔥 03:19, 2 May 2021 (UTC)
 * I will need tables showing how to decline all these nouns, including where the Vedic accents go. Benwing2 (talk) 03:47, 2 May 2021 (UTC)
 * Yes, I'll tell you. For now, though, see शिच् - declension is pretty simple for those ending in -c. Decl of -ic will be -ik -icau -icaḥ, -uc will be -uk -ucau -ucaḥ and same for -[any other vowel]c. Masculine is just like feminine in these. Neuter is a little different. See below neuter endings (apply for all -ac, -ic, -uc, -ṛc, etc.):
 * k cī ñci (nom) [Deva=क् ची ञ्चि]
 * k cī ñci (voc)
 * k cī ñci (acc)
 * (rest - dat-loc is like masculine/feminine)
 * I want the table which appears on using sa-decl-noun-m to appear like it does in any normal a-stem declension.
 * Pitch accent is usually on the same vowel throughout except the vocative case, where it shifts to the first vowel only. 🔥 शब्दशोधक 🔥 04:21, 2 May 2021 (UTC)
 * I notice that the vocative singular or neuter nouns is different from the nominative singular, e.g. कमल. Is that correct? I always thought all IE languages had neuter vocative and nominative singular (and dual and plural) the same. Benwing2 (talk) 05:31, 2 May 2021 (UTC)
 * Yes, that's true for some while not true for some others. Here it's different. 🔥 शब्दशोधक 🔥 05:39, 2 May 2021 (UTC)
 * Sorry, I'm confused, are you saying the existing declension of e.g. कमल is wrong? If so I'll fix it. Benwing2 (talk) 20:11, 2 May 2021 (UTC)
 * No it is correct. The nominative singular is 'kamalam' while the vocative is 'kamala' (similar to the actual word). For some other words like manas, both the nominative and vocative are same - manas (contracted form : manaḥ). 🔥 शब्दशोधक 🔥 02:46, 3 May 2021 (UTC)
 * I aded support for -c nouns but for the others I need exact declension tables along with several examples so I can properly test on them. Benwing2 (talk) 20:46, 2 May 2021 (UTC)

Error
I wanted declension for those terms ending in त्. There is this problem:. See my edits on Mod:sa-decl/data. I want to know how (and where) is the module for those ending in -स्. 🔥 शब्दशोधक 🔥 10:51, 30 July 2021 (UTC)


 * eg., etc. which decline differently. How are they dealt with? Same for -t. 🔥 शब्दशोधक 🔥 10:55, 30 July 2021 (UTC)

Help creating templates for Akkadian entries
Hi! I've found your User name in the history of. I've started to revise Akkadian, Sumerian and cuneiform entries these days, but I'm not familiar at all with templates or any programming language. I'd really like to clean up Akkadian and Sumerian entries and give them a better structure so that they might become useful to, if not scholars, at least people who are studying those languages. I've been trying to see what I can do with existing templates, but being quite a newbie, I'm between "I think I got it now" and "Please god kill me". Would you have time and energy to help/guide me on this endeavour? Thank you! Sartma (talk) 18:08, 25 April 2021 (UTC)
 * Hi. I moved your message to the bottom so things stay in chronological order. I can help you if you have specific requests. Benwing2 (talk) 18:21, 25 April 2021 (UTC)

Deleting crap
Hey. Can you delete the empty categories in Category:Requests for date by source? It started getting filled up again Yellow is the colour (talk) 22:30, 28 April 2021 (UTC)

Boring formatty crap
Also, if you want to change boring stuff like this at distream, you can. I did a bunch of edits without the asterisk, coz I'm sloppy bitch. Yellow is the colour (talk) 22:36, 28 April 2021 (UTC)

Italian IPA project
Hey Ben. You've been working on IPA templates for various European languages, so I was curious whether you wanted to take on a project with Italian that I've been thinking about for a while (but which no bot-runner had the time and interest to take on). The problem is not with, which is good, but with the fact that it is so underused. We can, in theory, bot-add with the appropriate parameters to all Italian entries that lack IPA, have, lack , and lack spaces. We can also bot-add it to entries that lack, but do have where stress is marked (so where one of the arguments to  contains a vowel with a grave accent). Finally, we can bot-add it to any disyllabic Italian entries that lack both and, but lack diacritics in the pagetitle and have any of as their first vowel — but these would have to be scanned by a human first to catch borrowings with unchanged orthography, mostly from English. That would cover a lot of entries with minimal human effort. —Μετάknowledge discuss/deeds 21:50, 7 May 2021 (UTC)
 * I will look into it. I have also thought of writing a bot to add to all Spanish pages without it, except for those that are likely or possibly unadapted English words. I used a complicated regex to look for words that don't fit normal patterns of spelling in Spanish:

This pulls out 3,561 words that might have a nonstandard pronunciation (including some false positives), and 52,049 words that probably have standard pronunciation (including a few false negatives that have nonstandard pronunciations, which we might be able to pick out by hand). A sample, from booster to brownie, most of which I would not trust to have standard pronunciation: booster bootear BOPE bórax borraj bortsch bosón de Higgs bosón W botnet botox bótox Botswana botswanesa Boucher bouchri boulangerie bouldering bouleuterión boulevar bourbon bourguibismo boutade boy bpd bracket braford brahman brahmán bráhmana bralette brandy brasserie bratwurst Brazzaville break breakdance breakdancing brechtiano Brest Brexit brexit Brézhnev Briansk BRIC Bridgetown brie brik briozoo briquet british britpop broker Broussard browniano brownie Benwing2 (talk) 05:05, 8 May 2021 (UTC)
 * The regex method is really smart. Writing a version of that for Italian would help save a lot of work. —Μετάknowledge discuss/deeds 05:12, 8 May 2021 (UTC)
 * I looked into this. First of all, I do think changes should be made to . For one, I don't really like how it defaults to penultimate stress, closed [e] and [o], and unvoiced [t͡s]. I would prefer it to throw an error rather than defaulting to any of these, forcing the user to specify explicitly whether they want penultimate or antepenultimate stress, closed or open  and , and unvoiced or voiced . This is similar to how operates (w.r.t. stress) and  operates (w.r.t. open or closed vowels and  except in certain positions). Otherwise, it's too easy for lazy editors to just add  to a page without looking into whether the output is wrong. What I'd prefer to do is run a bot to add the missing pronunciations whenever  is given without an argument, and then make specifying these features mandatory. I'd also prefer to use respelling to indicate whether  is voiced or unvoiced (e.g. using  for unvoiced,  for voiced), in case for example there is a combination of voiced and unvoiced  in a word, or multiple words with different pronunciations of  in them. Also, it looks like  doesn't yet handle multiple words. These are all things I can fix.
 * As for automatically adding, I think this is a good idea. I generated some counts:
 * Total number of Italian lemmas: 117,079
 * Number of current lemma pages with : 8,363
 * Number of current lemma pages with : 7,848
 * Number of current lemma pages with either or : 16,193
 * Number of current lemma pages with /: 9,619
 * Number of current lemma pages with / with stress marked: 9,293
 * Number of current lemma pages with /: 10,014
 * Number of current lemma pages with either / or /: 14,516
 * Number of current lemma pages with either / with stress marked, or with /: 14,270
 * Number of current lemma pages with either / with stress marked, or with /, and without : 11,362
 * Number of current lemma pages with either / with stress marked, or with /, and without either or : 10,706
 * So we have around 11,000 pages where the pronunciation can probably be autogenerated. I haven't yet looked into including disyllabic words with stressed or excluding pages with  or pages that look like English words. I also think we should exclude pages with 'gli' + consonant (cf. aglina with /ʎ/, but aglifo with /gl/), and with 'sci' + vowel (cf. sciame with /ʃa/, but sciare with /ʃi.a/). I also think it's fairly safe to include words with certain well-known endings, e.g. verbs ending in '-are' or '-ire', nouns ending in '-mento', adverbs ending in '-mente', etc. I think the Catalan Wiktionary module for Italian pronunciation has a long list of these things (although it's not able to look into the contents of the page to see what the part of speech is, which my bot can do). Benwing2 (talk) 19:30, 9 May 2021 (UTC)
 * Agreed on all counts. I don't think the defaults in are a big issue, but the logic behind changing them is sound. The only other thing I have to add is that this is a fantastic resource, if you don't know about it already. —Μετάknowledge discuss/deeds 19:49, 9 May 2021 (UTC)
 * Thanks for taking this on. I like how handles this, it detects ambiguous patterns and errors automatically if no input is provided, and in most cases you can specify simply one vowel without spelling out the whole word again. – Jberkel 21:09, 9 May 2021 (UTC)
 * That is a good idea but needs to be modified for Italian, because there is no default stress (it can either be penultimate or antepenultimate). I'm thinking you can specify a single vowel, but you'll get an error if the same vowel occurs in the penultimate and antepenultimate syllable (or if there are multiple words). Benwing2 (talk) 21:51, 9 May 2021 (UTC)
 * What endings should the module automatically handle? Currently only -izzare, automatically respelled -iddzàre. I'm thinking -mento (respelled -ménto), -mente (respelled -ménte), -ezza (respelled -éttsa), -zione (respelled -tsióne), -tore (respelled -tóre), -trice (respelled -trìce). Other possibilities: -are (either verb or adjective, but in both cases -àre), -ale, -ire, -oso (respelled -óso), -ello (is this always -èllo?), -ella (likewise), maybe -etto, -etta, -evole, -ibile. Maybe this is less necessary if we support the single-vowel notation and/or a partial-word notation like [àre] for words in -are. Benwing2 (talk) 22:06, 9 May 2021 (UTC)
 * I think it might be overkill to stop treating penultimate stress as default. More respelling than it's worth, in the end. But I will accede to those actually doing the work, of course. —Μετάknowledge discuss/deeds 22:17, 9 May 2021 (UTC)
 * Maybe; although definitely the è ò vs. é ó needs to be specified explicitly, defaulting to é ó is just asking for trouble. Benwing2 (talk) 22:22, 9 May 2021 (UTC)
 * Erutuon, can you speak to the section in Module:it-pronunciation that handles z? The logic is rather complex in deciding whether to default to voiceless or voiced, and whether to generate a single z as double. When looking through the existing manually specified pronunciations using, there is no consistency as to whether a single z between vowels is specified as single or double. It actually appears that most voiced z between vowels are given as single, although some are double. Cf. alcazar given as /alˈkad.dzar/ or /al.kadˈdzar/, but apozema given as /aˈpɔ.d͡ze.ma/|[äˈpɔːd̪͡z̪emä], including with vowel lengthening before single z. There are even weirdnesses like azoto given as /adˈd͡zɔ.to/|[äˈd̪͡z̪ːɔːt̪o], with double pronounced zz in the phonemic notation but single pronounced z in the phonetic notation. Who are the active editors here who are native speakers? Perhaps they can speak to what's going on? Benwing2 (talk) 01:32, 11 May 2021 (UTC)
 * Not a native speaker, but my understanding is that there is no normative distinction in pronunciation between singleton and geminate z in Italian (whether voiced or voiceless), and different speakers have different intuitions about whether they are neutralized to long or short consonants. The same goes for [ʃ], [ɲ], [ʎ]; this set of consonants is sometimes referred to as "inherent" geminates. (According to Martin Maiden, some northern speakers who don't have robust consonant length distinctions in their native accent may make a spelling-pronunciation distinction in consonant length between words like spazi and spazzi, but this is not a feature of the established standard description.) Since there is no distinction, Wiktionary could transcribe the phonemes either way, but should be consistent (for each phoneme, and probably, in the treatment of the set of length-neutralized/"inherent geminate" phonemes). The correct phonetic length transcription is a tricky question since phonetic length is not dichotomous.--Urszag (talk) 02:07, 11 May 2021 (UTC)
 * I think I based the z logic on the generalizations given at Italian orthography. The code certainly is hard to understand because I was trying to be clever and do it all in one run. — Eru·tuon 02:31, 11 May 2021 (UTC)
 * BTW I'm pretty sure I'm going to discard the current logic in Module:it-pronunciation to decide whether to default z to voiced or voiceless, and require that all z be explicitly respelled either ts or dz, except in certain recognized endings (probably -izzare, -zione, -izzazione, -ezza, -izia). Benwing2 (talk) 01:40, 11 May 2021 (UTC)
 * Heads up, I am planning on making changes to . In particular:
 * I'm going to fix the module to correct many of the existing bugs (e.g. lack of multiword support, lack of support for falling diphthongs, incorrect handling of monosyllables including lack of recognition for which monosyllables are stressed vs. unstressed).
 * I'm going to discard the current complex logic that defaults  to either /t͡s/ or /d͡z/, and also discard the voiced param. Instead, you'll have to respell  as either  or , and respell <zz> as either or, except in certain recognized suffixes (see below). If you don't, you'll get an error. Note, the current logic that automatically converts single <z> between vowels (respelled <ts> or <dz>) to double /t.t͡s/ or /d.d͡z/will be kept.
 * I'm going to require that, in cases with 3 or more syllables, the stress is marked using an acute or grave accent (in the process specifying the quality of <e> and <o>), except in a fairly large set of recognized suffixes (see below). If you don't mark it, you'll get an error.
 * I'm going to require that, if the stress falls on <e> or <o>, the quality is marked with <é> <è> <ó> <ò>, even if otherwise the stress wouldn't need to be marked (e.g. in monosyllables and disyllables). Again, exceptions for a large set of recognized suffixes (see below). If you don't mark it, you'll get an error.
 * I'm going to add the ability to specify only the stressed vowel, as long as the stress falls on the penultimate or antepenultimate and both vowels aren't the same. E.g. in anfiteatro you can just write, which is the same as writing . This won't work in e.g. angioedema because the penultimate and antepenultimate vowels are both <e>; if you write , you'll get an error. This works for glides as well, e.g. for adenopatia you can just write.
 * The idea behind the changes above is to reduce the errors that are likely to appear when one of two fairly arbitrary choices is defaulted (penultimate vs. antepenultimate, é vs. è, ó vs. ò, /t͡s/ vs. /d͡z/), since lazy editors, esp. those less familiar with Wiktionary and with templates like, are likely to write without params and without properly checking the output.
 * The current set of suffixes that will be recognized is as follows (they are given in their automatically respelled form, but will be recognized without any accents and with <z> rather than <ts> or <dz>): -ménte, -ménto, -ènte, -ènto, -iddzàre/-iddzàrsi, -àre/-àrsi, -ìre/-ìrsi, -iddzatóre, -sóre/-tóre, -iddzatrìce, -trìce, -iddzatsióne, -tsióne, -óne, -àcchio, -àccia/-àccio, -àggine/-ìggine/-ùggine, -àglia/-àglio, -ìglia/-ìglio, -àia/-àio, -àntsa/-èntsa, -àrio, -sòrio/-tòrio, -àstra/-àstro, -èlla/-èllo, -étta, -éttsa, -fìcio, -ièra/-ièro, -ìfero, -ìsmo, -ìsta, -ìzia/-ìzio, -logìa, -tùdine, -ùra, -ùro when not directly following a vowel (as in e.g. centauro), -iddzànte, -ànte, -iddzàndo, -àndo/-èndo, -àbile/-ìbile, -ànico/-ènico/-ìnico/-ònico/-ùnico, -àstica/-àstico/-ìstica/-ìstico, -àto/-àta, -àtica/-àtico/-ètica/-ètico, -ènse, -ésca/-ésco, -évole, -iàna/-iàno, -ìva/-ìvo, -òide, -óso. I'm specifically excluding <-etto>, <-otto>, <-osa> as having too much ambiguity in their pronunciation. I'm also focusing here on lemmas specifically; my assumption is that nonlemma pronunciations will likely be generated by bot, where the defaulted suffix rules are less of an issue. For this reason, I currently include <-iano>, which is frequently unstressed or <-ìano> in third person plural verb forms. Maybe I should exclude it instead.
 * The plan for how to implement this is first to use a bot to explicitly specify the pronunciation (including accent mark and respelled <z>) for all current instances of, then change the module according to the plans above, and then do another bot run to convert instances to the default form or short vowel-only form as appropriate.
 * Benwing2 (talk) 02:22, 12 May 2021 (UTC)
 * I'm pleased with the idea of of using more straightforward morpheme-based rules for z. It was a cute idea to rely on generalizations, but because they were complicated and not always correct, it was hard to ensure the output was correct in all cases. About using dz, ts, I'm wondering if Italian ever has contrasting with  or  contrasting with . I guess probably not, but they could be written t.s and d.z anyway. — Eru·tuon 18:49, 17 May 2021 (UTC)
 * I found an example that has in it, which is outsider, and yes for this case it works to spell it t.s. Benwing2 (talk) 05:31, 18 May 2021 (UTC)

Template:bnt-desc
This one is truly great! Can we not have the same thing for all language names in the Descendants section? (I had proposed this earlier, though.) - ⸘ - dictātor · mundī  18:42, 10 May 2021 (UTC)
 * Personally I am in favor of this, but as you saw earlier, some people seem to be strongly opposed. Benwing2 (talk) 01:21, 11 May 2021 (UTC)
 * Then how come we have that for Bantu languages, just curious? - ⸘ - dictātor · mundī  07:35, 11 May 2021 (UTC)
 * User:Metaknowledge implemented that. Apparently they are in favor of linking languages in Descendants sections. Benwing2 (talk) 02:24, 12 May 2021 (UTC)
 * I am, but as should be obvious, the main point of that template is to display Guthrie codes. Even if were to link to Wikipedia, it would still need to exist. —Μετάknowledge discuss/deeds 02:54, 12 May 2021 (UTC)

fixed Italian IPAs
All correct, thanks for the pings. Ultimateria (talk) 15:44, 14 May 2021 (UTC)

Endless Loop in Module:it-pronunciation
Until I removed the secondary stress, ˌai sènsi at ai sensi was running out of execution time. I tried ˌai ˌdai sènsi  ˌdai in preview, with the same result. There was no error if I gave it the primary stress. Chuck Entz (talk) 19:58, 14 May 2021 (UTC)

Different Etymologies Ignored by Wingerbot
In this edit: two sets of definitions with different etymologies were added as if they all sprang from the same etymology. I have corrected the page, but I thought this might be relevant info about the operation of WingerBot. --Geographyinitiative (talk) 20:25, 15 May 2021 (UTC)
 * Thanks, this is my own doing, not my bot's. The text "manually assisted" in the commit message means that the only use of the bot was to push changes I manually made in a text file into the entry. I don't remember this exact word but I imagine I looked up in Wikipedia and added definitions accordingly, not correctly accounting for the different characters in the different uses. Benwing2 (talk) 19:05, 16 May 2021 (UTC)

pt-IPA: Pronunciation of
Cheers! I was making some edits and I noticed the result of Template:pt-IPA for. First of all, I don't know about the Brazilian pronunciation (yours says it's /baˈtõ/, while the page states /baˈtõw̃/; I have no idea which one should be correct). However it was the European pronunciation that raised my curiosity. The pronunciation should be /baˈtõ/ with an open "a" (see Infopédia and Priberam). I've edited the page accordingly, but I refrained from using the template.

The main reason I'm writing this comment is because I'm wondering if it's a rule to have an open vowel in the penultimate syllable, preceding the "-om". Alternatively, could it just be an artifact from French? Another examples are (I personally read all of them with an open "a", I also read the /õ/ as /ɐ̃w̃/, but I think that's unrelated):
 * (Priberam says it's an open "a", but Infopédia a close one)
 * (Infopédia says open, Priberam doesn't state anything)
 * (Infopédia says open, Priberam doesn't state anything)

However, words like and  are pronounced with ɐ.

I can't think of examples with other vowels, but at least I get the impression that the "om" opens the vowels in the previous syllables. At least that's I would pronounce them. I'm sorry I can't be more specific or accurate describing the issue, but I hope this helps you somehow. - Sarilho1 (talk) 15:22, 18 May 2021 (UTC)

(New) Sanskrit (neologisms)
Hi. The categories CAT:New Sanskrit and CAT:Sanskrit neologisms are essentially the same. One can say that a newly coined Sanskrit term to be a neologism or a part of New Sanskrit vocabulary. There should be one proper category of such Sanskrit terms which contains all of these - not that half in one and the other half in the other cat. I tried to do something, but then all the neologism labels crashed. Either of these 2 can be done:
 * 1) Hard-redirecting CAT:Sanskrit neologisms to CAT:New Sanskrit
 * 2) Bot operation to change all sas to sa
 * 3) Altering the label module for turning sa to an alias of sa, as I had tried (preferably without causing the failure of the other LANGs neologisms)
 * 4) Making CAT:Sanskrit neologisms a sub-cat of CAT:New Sanskrit manually and having no entries in the latter
 * 5) Bot operation to change all sas to sa
 * 6) Altering the label module for turning sa to an alias of sa
 * Which one is your preference and would this be possible? 🔥 शब्दशोधक 🔥 14:45, 19 May 2021 (UTC)
 * If there's no essential difference between "New Sanskrit" and "Sanskrit neologisms", I'd prefer to go with the latter as we have "Foo neologisms" for various languages. I think it should be possible to change the label 'New Sanskrit' to be an alias of 'neologism' for Sanskrit. If we do that, it seems to me there's no reason to a category Category:New Sanskrit to exist. Benwing2 (talk) 00:55, 20 May 2021 (UTC)
 * wiki is an etymology-only language code (and also pretty unnecessary IMO) for New Sanskrit. I don't think it is always necessary to give in an etymology that the source term is newly-coined or a thousand-year-old. So perhaps wiki could be removed. All "sa-neo"s in etymologies would have to be changed to "sa"s and then CAT:Terms derived from New Sanskrit would be empty. If New Sanskrit should be deleted, then even a label "New Sanskrit" should not exist. Would you be able to manage these 2 bot operations - etym. sa-neo to sa and label change and  to  ? 🔥 शब्दशोधक 🔥 03:55, 20 May 2021 (UTC)
 * Hmm. I thought about this and I'm not sure "New Sanskrit" and "neologism" are the same. By analogy with Latin, for example, "New Latin" (also known as "Modern Latin") includes any Latin used after 1500 or so. A neologism, on the other hand (per Wikipedia), is a term coined in the last 20 years or so that hasn't gained wide acceptance. In the case of Latin, there are tons of New Latin terms (e.g. the names of almost all chemical elements, terms of plants and animals, etc.) that aren't neologisms in the sense that they're widely accepted, frequently borrowed into other languages, etc. An example is . On the other hand, terms like or  might well be neologisms (although the latter term was used by Pope Francis in a Vatican encyclical). The fact that a term like  has three synonyms, ,  suggests to me that all are neologisms. Wiktionary isn't always good at marking the distinction but it clearly exists. I don't know enough about Sanskrit to say whether the same applies. Benwing2 (talk) 06:01, 20 May 2021 (UTC)
 * Conceptually, there could be distinction between "New Sanskrit terms" and "Sanskrit neologisms". "Sanskrit neologisms" could refer terms coined within the last "generation", while "New Sanskrit terms" could refer to terms coined after a certain point in time but before the current "generation". However, maintaining such a distinction may be difficult. Kutchkutch (talk) 09:24, 21 May 2021 (UTC)
 * Yes, maintaining such a distinction would be difficult. And "New Sanskrit" and "neologism" are used interchangeably here. I am still for merging these 2. शब्दशोधक—شَبْدَشودَھک—śabdaśodhak 10:05, 21 May 2021 (UTC)
 * There is Sanskrit literature from 300 or 400 years ago which is different from the Sanskrit used in school textbooks which has words for modern appliances, etc. Benwing is right. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 02:10, 22 May 2021 (UTC)


 * every neologism would come under new Sanskrit, so I'll still make it a subcat. 🔥 ಶಬ್ದಶೋಧಕ 🔥 05:41, 23 May 2021 (UTC)
 * Not necessarily. Sanskrit neologisms come under ‘Contemporary Sanskrit’ rather than ‘New Sanskrit’; compare vs. . By tradition these ought to be classified differently, so a subcat. is not a good idea.  If need be, we can even delete the neologism categories for classical languages and instead use the label “Contemporary LANG” for them. Would that be okay? - ⸘ -  dictātor · mundī  15:19, 27 May 2021 (UTC)

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



Thank you. — SGconlaw (talk) 16:03, 24 May 2021 (UTC)
 * Done. My apologies about the other request of yours, I have been meaning to get to it for awhile. Template renamings are easy, while other sorts of formatting may require more work because they need a custom script. Benwing2 (talk) 03:08, 27 May 2021 (UTC)
 * Great, thanks! — SGconlaw (talk) 04:28, 27 May 2021 (UTC)

Ongoing vote
Hey, you said in the BP discussion that you were in favour of creating the two templates inh+ & bor+. So would you like to cast your vote in the crucial vote?— given that you cannot implement something when there’s no consensus for it. Thank you. - ⸘ - dictātor · mundī  00:30, 27 May 2021 (UTC)
 * I voted, however I disagree with the general premise of the vote, if everyone needed a vote to do add any new feature, nothing would get done. Benwing2 (talk) 03:06, 27 May 2021 (UTC)
 * Benwing, I agree with you, but you forgot about the (in)famous Victarian opposition! And, some more users were opposed to the proposal as well. I really wish we could ignore their oppositions— is that achievable? - ⸘ - dictātor · mundī  15:37, 27 May 2021 (UTC)
 * Since the vote has failed, I am thinking of an alternative way to make the etymological text appear: can you please add a new parameter withtext for inh & bor? That would be very nice, and we can preferably have a shorter one, wt. There’s even no need to ask approval for this as only a new parameter is to be added. - ⸘ - dictātor · mundī  02:04, 29 May 2021 (UTC)
 * I believe some votes were against the text itself (i.e., regardless of new templates). J3133 (talk) 02:08, 29 May 2021 (UTC)
 * , do you like the alternative idea to use a parameter to display the text for inh & bor? - ⸘ - dictātor · mundī  00:39, 30 May 2021 (UTC)
 * That seems to be ignoring the result of the vote (i.e., against no consensus to add the text; the vote was for the templates, however, the point was to add text in etymology sections). J3133 (talk) 00:50, 30 May 2021 (UTC)
 * I am not aware of any user who was entirely against having the text. Only one or two, did say (even then they did not put much emphasis on it) that no wording was necessary for inheritances (inh only). On the whole, a formal vote on this was needless (as Benwing & others have stressed), and most people support the proposal anyway. Lastly, as a side note I would welcome comments from people who took part in the vote (folk from all sides, not just the supporters) rather than take your honourable advice (you did not even take part in the BP discussion(s)!), so I bid farewell to your bureaucratic self. - ⸘ - dictātor · mundī  01:12, 30 May 2021 (UTC)
 * Apparently I am bureaucratic when you needlessly created a vote only to do what you wanted after it ended. Lastly, I will quote Mahagaja (who voted and therefore should be “welcomed”) instead of being hushed after your farewell made in bad faith to ignore opposers: “These are unnecessary. If you want to say "Borrowed from" or "Inherited from", just write it in.” This seems to be against your idea, are we bidding farewell to Mahagaja? J3133 (talk) 01:29, 30 May 2021 (UTC)
 * User:Mahagaja was against the templets, not the etymological text. I am really worried about your comprehension skill. ‘These are unnecessary’ = ‘These [templates] are unnecessary’. - ⸘ - dictātor · mundī  03:09, 30 May 2021 (UTC)
 * As you could only read the first words, I believe yours is what we should worry about. “If you want to say "Borrowed from" or "Inherited from", just write it in.” See . J3133 (talk) 03:13, 30 May 2021 (UTC)
 * Not sure what you really want to tell or prove. The opposers were against what MK called a ‘templatisation creep’ or what Jberkel called ‘redundant/overlapping templates’. I have no time to explain the same thing over and over again; why not ask those opposers themselves if you have comprehension disability? would be a good person to start with. The lost vote has no implication of any prohibition to introduce a single, new parameter. I am just waiting for Benwing’s response, no one had invited you for a debate in his talk page. Lastly, to analyze your bearing in this post, it is very natural you would behave thus: having neither taken part in the BP discussion(s) nor cast your vote, you have become all-knowing, right? - ⸘ -  dictātor · mundī  03:58, 30 May 2021 (UTC)
 * Next time your disability prevents you from understanding etymologies perhaps do not suggest more ‘templatisation creep’ and immaturely shoving your opinions on everyone who opposed (as can be seen above). If you were all-knowing, you would understand if your templates were opposed (and caused you to hurriedly create a vote before thinking), your parameters, also, are not immune, rather than being salty and mindlessly blathering. J3133 (talk) 04:30, 30 May 2021 (UTC)


 * Benwing, could you please add the parameter withtext for inh & bor as I suggested erewhile? Thanks! ·~   dictátor · mundꟾ  11:59, 27 June 2021 (UTC)
 * withtext is too long to type, so please choose 1 or 1. As I said, link to the glossary is very important, so still I advocate a bot operation. I hope you have not forgotten the old discussion on this — "Inherited" is not as obvious and self-explanatory as you think it to be since I couldn't understand what it was. 🔥 शब्दशोधक 🔥 13:55, 27 June 2021 (UTC)
 * Such headstrongness is unwholesome, I have said sundry times that I am against a bot operation, and I have stated the reason clearly. By the way, putting the parameter should produce the link automatically, of course. ·~   dictátor · mundꟾ  14:09, 27 June 2021 (UTC)

Replacement requests
Hello, could you please carry out the following replacements?



Thanks. — SGconlaw (talk) 14:17, 28 May 2021 (UTC)
 * Done. Benwing2 (talk) 07:42, 3 June 2021 (UTC)
 * Thank you! — SGconlaw (talk) 09:55, 3 June 2021 (UTC)

: learned borrowing
Module:etymology/templates (line 452) should have “learned borrowing” not “learned borrowed”. J3133 (talk) 21:03, 29 May 2021 (UTC)
 * I went ahead and fixed it. Chuck Entz (talk) 22:59, 29 May 2021 (UTC)

Pronunciation questions
Thanks, Ultimateria (talk) 17:36, 2 June 2021 (UTC)
 * 1) Is pt-IPA fully ready to use? If so it needs documentation.
 * 2) It it-IPA fully ready to use? If so, will you be removing manual parameters soon? Do we need to add them in the meantime?
 * pt-IPA is close to being ready. I need to make some changes to the handling of hiatuses. Please don't use it until then. As I make those changes I'll document it.
 * it-IPA is ready and in use. What do you mean by manual parameters?
 * Benwing2 (talk) 07:41, 3 June 2021 (UTC)
 * I'm referring to edits like ; having giocàre produces the same thing as raw it-IPA. I assumed you were reviewing/getting pronunciations uniform before removing these parameters when they're the expected output from the pagename alone. I guess it's fine to keep any redundant parameters, but my question is whether there's a reason to add them now in cases where raw it-IPA is correct. Ultimateria (talk) 16:51, 3 June 2021 (UTC)
 * I see. Yeah, originally I had to make all the parameters explicit before changing the handling of the module. At this point it's fine to use by itself when it works. I've been habitually adding the explicit pronunciation but I'll probably stop doing this, and do a bot run to remove the redundant params. Benwing2 (talk) 03:38, 4 June 2021 (UTC)
 * I see. Yeah, originally I had to make all the parameters explicit before changing the handling of the module. At this point it's fine to use by itself when it works. I've been habitually adding the explicit pronunciation but I'll probably stop doing this, and do a bot run to remove the redundant params. Benwing2 (talk) 03:38, 4 June 2021 (UTC)

Bot's Italian mistakes
Hi,

I believe the past participles listed at and  are wrong (venuto and avvenuto instead of venito and avvenito), as confirmed by the conjugation tables. Kritixilithos (talk) 10:02, 3 June 2021 (UTC)
 * Didn't look carefully at the edit, I mean s/Bot// Kritixilithos (talk) 10:21, 3 June 2021 (UTC)
 * Okay fixed those two individual entries. I don't understand the grave vs acute accent distinction, so have used the same grave accent for the corrected participles. Kritixilithos (talk) 11:07, 3 June 2021 (UTC)
 * Thanks. You are correct about venuto and avvenuto. I also fixed convenire, prevenire and several others. The grave vs. acute accent distinction applies to e and o and indicates the quality of the vowel: é =, è = , ó = , ò = . Benwing2 (talk) 03:21, 4 June 2021 (UTC)

Category:Esperanto quotation templates
I saw you deleted this category saying it is empty. It isn’t; it contains one entry Template:RQ:Kabe Faraono, but for some reason it doesn’t show up. MuDavid 栘𩿠 (talk) 01:06, 4 June 2021 (UTC) I did a null save on Template:RQ:Kabe Faraono to make it appear in the category, and re-created the category. Benwing2 (talk) 03:23, 4 June 2021 (UTC)
 * Thanks! MuDavid 栘𩿠 (talk) 03:29, 4 June 2021 (UTC)

Quotation template replacements
Hello, kindly carry out the following replacements when you have time:



Thank you. — SGconlaw (talk) 19:33, 6 June 2021 (UTC)
 * Done. Benwing2 (talk) 02:04, 7 June 2021 (UTC)
 * Thanks! — SGconlaw (talk) 04:10, 7 June 2021 (UTC)

Quote module
Hi Benwing2, could you please address my ping at Grease pit/2021/May? I pinged you because you are the main editor of Module:quote. This, that and the other (talk) 09:49, 11 June 2021 (UTC)

Quotation template replacements
Hi, when you have time could you please carry out the following replacements?



Thank you. — SGconlaw (talk) 12:59, 11 June 2021 (UTC)
 * Done. Benwing2 (talk) 02:35, 12 June 2021 (UTC)
 * Great, thanks! — SGconlaw (talk) 04:51, 12 June 2021 (UTC)

position of qualifiers in nyms
Hello,

while I don't mind having the qualifiers before the referred-to items, I think you need to be aware that it necessitates hundreds of changes to be made manually in cases when the qualifier referred to more than one term and now it gives wrong information, changes like this one. Could you help me find such instances in Hungarian with some smart search? Adam78 (talk) 06:08, 13 June 2021 (UTC)
 * I can go through the latest dump (of 2021-06-01) looking for uses of and  in Hungarian lemmas (maybe also non-lemma entries) that have a qualifier attached to the last entry. Would that work? BTW my instinct for handling cases like this is to duplicate the qualifier on each entry needing a qualifier, but your solution seems to work fine too. Benwing2 (talk) 06:12, 13 June 2021 (UTC)
 * This happens in English entries too (or at least I've used it in that way before). I'm a little uncertain whether it's more common for the qualifier to be in front of rather than after the term, though. In "Derived terms", "Related terms" and "Translations" sections, we seem to place it after the term. — SGconlaw (talk) 06:33, 13 June 2021 (UTC)
 * How have you used it, similarly to what User:Adam78 has done in saying "the last two are colloquial" (or whatever)? If you just say "colloquial", it becomes ambiguous whether you're referring to the preceding entry only or the preceding two/three/etc. entries. My thought for putting it before is that (a) most (all?) templates that take a q or qual param put it before (the templates you mention above don't have such a param, it may be just convention to put a separate qualifier after, and in those sections the ambiguity doesn't appear since the terms usually stand by themselves), (b) if you have a list of several entries and the last one needs a qualifier, putting it before makes it ambiguous. The idea is that it's more likely the things at the end of the list need qualifiers than the things at the beginning of the list. But I could reconsider if several people think it's better to go after. Benwing2 (talk) 06:47, 13 June 2021 (UTC)
 * yes, I've sometimes used expressions like "the last two obsolete" or "all obsolete", if I recall correctly. I don't really know whether the qualifier should be before or after the term, but I suppose we ought to be consistent one way or another. One possible difficulty of having the qualifier in front is how it would look in combination with the use of – you'd have the sense in parentheses, followed by another set of parentheses with the qualifier which might be a bit odd. Putting the qualifier in front on rhymes pages might also look a bit odd. (On a related point, perhaps  should have a qualifier parameter built in so that it isn't necessary to separately add .) — SGconlaw (talk) 07:06, 13 June 2021 (UTC)

I'm not sure it only concerns those where there is a qualifier referring to the last term; I think it would be more expedient to find strings expressing English numerals (including "both") in any qualifier field within nym invocations. But maybe we'd better postpone it to early July so that cases from the first half of June can be included, and also we can see by that time if it's really the final solution or perhaps reverted. Adam78 (talk) 12:45, 13 June 2021 (UTC)
 * I'll go ahead and pull out all cases of where there's a qualifier and see what's going on. Also, the dump is generated twice a month (on the 1st and the 20th), so we don't have to wait till July. Benwing2 (talk) 17:01, 13 June 2021 (UTC)

Quotation template replacements
Hello, when you are free kindly carry out the following quotation template replacements:


 * and →

Thank you. — SGconlaw (talk) 17:52, 13 June 2021 (UTC)
 * Done. Benwing2 (talk) 23:56, 13 June 2021 (UTC)
 * Thanks! — SGconlaw (talk) 07:29, 14 June 2021 (UTC)

Category:nl:Neighborhoods in Gelderland, Netherlands
This has had a module error since it was created over a week ago. I'm definitely not up to speed on the backend for the place categories, but I'm guessing that some combination of "Neighborhoods" vs. "Neighbourhoods" and "Netherlands" vs. "the Netherlands" is messing up the automated recognition for this category. I would note that Category:Places in Gelderland, Netherlands exists, and so does Category:Neighbourhoods in the Netherlands, but the module seems to be looking for Category:Neighborhoods in the Netherlands. I tried the obvious quick fix of moving Category:nl:Neighborhoods in Gelderland, Netherlands to Category:nl:Neighbourhoods in Gelderland, Netherlands, but that isn't even recognized as a valid label. To make things even more confusing, Kralingen has " " and Het Loo has " ", but the first is in Category:nl:Neighbourhoods in the Netherlands, and the second is in Category:nl:Neighborhoods in Gelderland, Netherlands. Is there something obvious I'm missing? Chuck Entz (talk) 02:47, 14 June 2021 (UTC)
 * Yes, the issue is somewhere in the spelling of "neighborhoods" vs. "neighbourhoods". I made all European countries and all former British colonies use the -our- spelling and everywhere else use the -or- spelling. Let me look into this and see why it's generating the -or- spelling. Benwing2 (talk) 02:49, 14 June 2021 (UTC)
 * BTW, the issue with Kralingen is that Module:place/shared-data lists the province as "South Holland" rather than as "Zuid-Holland". I will add an alias for the latter. Benwing2 (talk) 02:54, 14 June 2021 (UTC)

et al.
Hi, when you have time, please help to carry out the following quotation template replacements:



Thank you. — SGconlaw (talk) 13:57, 18 June 2021 (UTC)
 * Done. Benwing2 (talk) 07:19, 19 June 2021 (UTC)
 * Thanks! — SGconlaw (talk) 08:54, 19 June 2021 (UTC)
 * Just curious to know, why are you going about augmenting the text in quotation templets? ·~   dictátor · mundꟾ  16:15, 19 June 2021 (UTC)
 * mmm, why not? I’ve been creating quotation templates which link to first or early editions of works which are available online for use in our entries for some time now. When I come across a template that has not been linked, I see if an online version of the actual work is available, and if so I link the work to the template. — SGconlaw (talk) 16:51, 19 June 2021 (UTC)
 * No no, I was talking about changes like Bronte Wuthering → Emily Bronte Wuthering Heights, i.e., those changes of the names of the templets (sorry my wording was unclear) that you are asking Wingerbot to make. ·~   dictátor · mundꟾ  17:03, 19 June 2021 (UTC)
 * oh I see. In many cases there was already a pre-existing template like . However, the editor Wonderfool (who goes by a changing series of usernames), who has been helping to locate quotations in entries tagged with, has a tendency to create new templates or redirects which aren't really necessary. When I come across these, I ask Benwing2 for help to replace the unnecessary uses with the primary quotation template. (For example, I recently found that had the redirects  and  which are difficult to interpret, so I shortened the primary template to  and asked for replacement of all uses with the primary template so the redirects can be deleted. — SGconlaw (talk) 17:57, 19 June 2021 (UTC)

Removing ‘Old Galician’
Hello, could you please change all occurrences of ‘from Old Galician and ’ to simply ‘from ’ in entries? Old Galician is but an alias of Old Portuguese, and I think there’s some user who’s wrought this redundancy. Thanks. ·~  dictátor · mundꟾ  16:15, 19 June 2021 (UTC)
 * That might not be the best solution. From a Galician point of view, "Old Portuguese" is but an alias of "Old Galician". It might be better to substitute "from Old Galician/". At any rate, I think the Galician editors should have a say in this- or at least be informed about it before it's done. Chuck Entz (talk) 16:36, 19 June 2021 (UTC)
 * Fala is also a descendant of Old Portuguese, so I am not sure how listing two aliases together would work. We should stick to one name. ·~   dictátor · mundꟾ  17:25, 19 June 2021 (UTC)
 * @Inqilābī, be aware that this issue is politicized in modern Galicia. There are two camps, one which thinks of Galician as a separate language from Portuguese and spells it using Spanish conventions (sanctioned by the government of Spain) and another which thinks of Galician as a dialect of Portuguese and spells it using Portuguese conventions. Both grew from the common Old Galician-Portuguese language, which Wikipedia calls and Wiktionary calls "Old Portuguese". The tricky thing here is that Old Galician-Portuguese was physically spoken in the area of modern Galicia up until about 1100 or so, at which point it expanded south during the Reconquista. Since Wiktionary uses the term "Old Portuguese", it's arguable that the etymology sections should just read the same, but the Galician editors understandably don't seem to like that much. Benwing2 (talk) 19:00, 19 June 2021 (UTC)
 * We're talking about Galician entries here, not Fala. Besides which, nobody talks about "Old Falan" in English. My main point, though is that you were advocating something that might very well be seen by Galician editors as a sneaky attempt to make a political point- sort of like taking "Myanmar" out of the template at Category:Rohingya language. Right or wrong, such things should only be changed by consensus. Chuck Entz (talk) 19:28, 19 June 2021 (UTC)
 * Well… then I guess it would be noncontroversial to change the name of the ancestor language to Old Galician-Portuguese, or maybe even Old Galician since it originated from Galicia. It just looks so odd to have ‘Old Galician’ written manually beside the templatised ‘’. Let us finalise what name to choose, and then perhaps I would suggest that in the BP. (The name Galician-Portuguese is not a goodun, as it sounds like a modern dialect continuum rather than an ancestor language.) ·~   dictátor · mundꟾ  22:22, 19 June 2021 (UTC)

Lua local environment
What's your local environment for running Lua? My local environment involves many hacks. Thanks. EdwardAlexanderCrowley (talk) 04:33, 25 June 2021 (UTC)
 * I don't run Lua locally on my laptop. All my scripting is done in Python. Benwing2 (talk) 04:36, 25 June 2021 (UTC)
 * Try asking User:Erutuon, who AFAIK does run Lua locally. Benwing2 (talk) 04:42, 25 June 2021 (UTC)

Remove all mnw: link
eg. မၞိဟ်တြုံ I usually use bzgrep at zhwikt, but I don't download enwikt dump, can you help removing the links? EdwardAlexanderCrowley (talk) 13:13, 29 June 2021 (UTC)


 * All 33 pages done. EdwardAlexanderCrowley (talk) 04:38, 2 July 2021 (UTC)

RfD
Hello, could you please delete this term brét in Romagnol, that I've created by mistake? The plural is brèt. Thank you.--BandiniRaffaele2 (talk) 00:10, 30 June 2021 (UTC)

Howewer in Italian is razzi the second-person present conjugation of razziare, not razzii, that would be pronunced as /rat͡s.ˈt͡siː/ (razzìi), but not /ˈrat͡s.t͡si/ (ràzzi) as the verb; you can see here, in Italian razzii is the plural of razzio. I counsel you to use this site for the conjugations, because it is reliable.--BandiniRaffaele2 (talk) 00:52, 30 June 2021 (UTC)
 * Treccani and Hoepli  both say razzìi. They are the most reliable dictionaries out there.  Native speaker input? Benwing2 (talk) 01:00, 30 June 2021 (UTC)
 * The site you quote is not so reliable, and doesn't even give the position of the stress or the quality of written e and o. DiPi agrees with the two dictionaries I cited. Benwing2 (talk) 01:01, 30 June 2021 (UTC)
 * razzii is correct. The Zingarelli and Devoto-Oli agree too. Imetsia (talk) 14:43, 30 June 2021 (UTC)

Wrong auxiliary verb
Hi. On January 8th, 2021, Wingerbot did this edit:

converting to.

However, in this case (aussäen), the auxiliary verb "sein" is wrong. In German language, there are two auxiliary verbs which you can combine with the past participle. These two auxiliary verbs are "haben" and "sein". The former one is the prevalent form whereas the latter one is more seldom used. So, you can't do this converting job by using a bot. Instead, in each single case you have to check first which one of the two auxiliary verbs is the correct form.--2003:CF:3F3F:6CD:9908:7B97:F9FA:9CD2 10:44, 4 July 2021 (UTC)
 * Hi. Not sure if I can ping an IP, you should create a user account. It may be not obvious but in this case the former template had an  in its params indicating that the auxiliary should be sein. The conversion merely preserved the auxiliary that was already there. If there was an error to begin with, it got propagated; my bot didn't make any new mistakes. Benwing2 (talk) 13:00, 4 July 2021 (UTC)
 * Hi Benwing2. Well, I'm glad to hear that this error wasn't caused by your bot which implies there is no further action needed. It was a single and isolated error made by the creator of this entry. For me, it wasn't possible to find out what caused this error bcs the previous template had already been deleted. Thanks for your quick reply.--2003:CF:3F3F:65E:DD8A:3F93:D3F6:C3D9 15:56, 5 July 2021 (UTC)

New CJK Pages
By creating these pages from scratch, you've taken them off of everybody's watchlists. They should have been created by the method used for the monthly pages. I'll have to get myself back up to speed on what I did at the beginning of the year to see if we can fix this. I believe it would involved something along the lines of temporarily deleting the CJK pages, moving the N pages on top of them, then moving the N pages back with redirects left behind, then restoring the CJK pages over the redirects. Chuck Entz (talk) 13:09, 6 July 2021 (UTC)
 * I went ahead and did it for the rfv page. I forgot to delete the redirect before restoring the other edits and had to redelete and selectively restore, but it should be back to normal. I see I was mistaken about your having already created the rfd page. I went ahead and moved rfdn to rfdcjk, then moved it back. Rfdcjk is now a redirect that's on everybody's watchlists. Whenever we get around to creating the new page, it will simply be a matter of replacing the redirect. Chuck Entz (talk) 13:57, 6 July 2021 (UTC)
 * Thanks. Maybe you can create an SOP (standard operating procedure) aka "runbook" page documenting how to do these things? Benwing2 (talk) 05:18, 7 July 2021 (UTC)
 * Everything I know I pieced together from examining how used to do it. At the beginning of the year it looked like no one was going to create the new monthly pages, so I looked at the edit histories from years past and figured it out just well enough to get it all done- with a bit of trial and error. Apparently, when you move a page the system updates all of the watchlists with the new page name, but doesn't remove the old pagename. That means you can add a new pagename to everybody's watchlist by moving a page that's already on their watchlists to the new pagename, then moving it back over the redirect. That leaves you with the original page still in the same place and on all the same watchlists, but now you have a redirect page that's also on the same watchlists- in effect, you've cloned the watchlist membership of the first page onto the redirect page, which you then replace with the content for the new page. If you need more pages, you can clone the clone, then the clone of the clone, etc. I believe all the monthly pages we've ever had were created using this method. Chuck Entz (talk) 15:17, 7 July 2021 (UTC)
 * Didn't I once post the source code of the script that does this every year? —Rua (mew) 15:38, 7 July 2021 (UTC)

et al
Hi, please carry out the following quotation template replacements when you're free.



Thanks. — SGconlaw (talk) 16:44, 8 July 2021 (UTC)
 * And Template:RQ:Mlry MrtDrthr too - dth t vwls, rght???! Roger the Rodger (talk) 16:47, 8 July 2021 (UTC)
 * can't be replaced by bot and redirected to as they are different templates with different parameters. The intention is to eventually replace all uses of the former with the latter, but it has to be done manually. — SGconlaw (talk) 18:45, 8 July 2021 (UTC)
 * actually, no need to work on . For some reason I thought there were a lot of uses (maybe I was looking at some other template), but now it seems there are only a handful, in which case I might as well do it manually. — SGconlaw (talk) 19:03, 8 July 2021 (UTC)
 * Done. Benwing2 (talk) 07:47, 9 July 2021 (UTC)
 * Thank you! — SGconlaw (talk) 10:36, 9 July 2021 (UTC)

Bot operation
Hello, Benwing. Would you be able to do a bot operation related to sa-sc? Basically, this has to be done: change SCRIPTCODE, sa-sc to TERM (as now the script can be auto-detected, thanks to your help) and remove any tr parameters. Thanks! 🔥 शब्दशोधक 🔥 07:09, 9 July 2021 (UTC)


 * Would it be possible? 🔥 शब्दशोधक 🔥 08:30, 9 July 2021 (UTC)
 * This is possible but are you sure you want all script code and translit params gone? What about non-standard transliterations and cases where the script recognition doesn't work? Benwing2 (talk) 05:06, 10 July 2021 (UTC)


 * The transliteration is once already shown at the head level, so showing it again is redundant. Your point about the scripts is correct, I'll fix it with an sc parameter. 🔥 शब्दशोधक 🔥 05:18, 10 July 2021 (UTC)
 * I'm confused now because the first param of isn't a script code, it's the Devanagari equivalent. You'll need a param for this to override it; maybe your intention is to add a param other than 1 to specify the Devanagari and move all params down by 1? Benwing2 (talk) 05:40, 10 July 2021 (UTC)


 * I changed that. Earlier 1st parameter was the sc code, hardly needed. 🔥 शब्दशोधक 🔥 05:50, 10 July 2021 (UTC)


 * Earlier = Special:Permalink/62293253 🔥 शब्दशोधक 🔥 05:53, 10 July 2021 (UTC)
 * OK but you've made a mess of things by hand-converting some of the usages to the new format but not all of them. How can my bot know which ones have been hand-converted, so it won't try to further convert them? I would recommend either you make a list of all the entries you already converted, so I can ignore them, or undo those changes. Benwing2 (talk) 06:12, 10 July 2021 (UTC)


 * The sc code is in Latin script, so perhaps best is if you can do: if parameter 1 contains Latin script, move to sc. If this is not possible, I'm thinking of another way. 🔥 शब्दशोधक 🔥 06:19, 10 July 2021 (UTC)


 * Just to add, the script can be removed all together, except for those in Bengali, and Assamese script. 🔥 शब्दशोधक 🔥 06:30, 10 July 2021 (UTC)
 * Everything converted. If the Devanagari isn't needed in most cases, it should probably be made a named param and the remaining params moved down by one. Benwing2 (talk) 18:46, 10 July 2021 (UTC)


 * Thanks. I'll try to do that. 🔥 शब्दशोधक 🔥 05:26, 11 July 2021 (UTC)


 * Please make a list of uses for this template like User:Benwing2/head-sa-noun. Still some uses are not correct, which I'll manually check. Also for pra-sc. Hope it's not a very tedious task. If you can, then I'll check them and then it can be replaced by bot. 🔥 शब्दशोधक 🔥 09:47, 12 July 2021 (UTC)
 * 🔥 शब्दशोधक 🔥 15:20, 13 July 2021 (UTC)
 * Done; see User:Benwing2/sa-sc, User:Benwing2/pra-sc. I included the whole line, for reference; you can modify any part of the line after the, and my push-changes script will pick it up properly. Benwing2 (talk) 00:34, 14 July 2021 (UTC)


 * , How difficult/easy is it for you to make lists like User:Benwing2/sa-sc, etc.? 🔥 शब्दशोधक 🔥 05:38, 17 July 2021 (UTC)
 * This is not hard, I have a script to do it. Benwing2 (talk) 00:30, 18 July 2021 (UTC)


 * If it is not hard, will you be able to this for some more templates I want to fix? There are just optional fixes, which would make the templates better if made. But if it is very time-taking for you, we can do without these changes. 🔥 शब्दशोधक 🔥 04:59, 18 July 2021 (UTC)
 * Sure, let me know which templates. Benwing2 (talk) 12:52, 18 July 2021 (UTC)


 * Well, here is a task with the bot, now from the list like one. Template: R:hi:Dasa
 * Convert R:hi:Dasa and R:hi:Dasa and R:hi:Dasa to 000
 * TERM and TERM and TERM and TERM to TERM
 * This is because of . Now the template can make the first parameter the page (if number) and headword (if string). If it is string, the second parameter is the page.
 * If, this is tough for you, please undo my change at the template. Thanks. 🔥 शब्दशोधक 🔥 15:54, 18 July 2021 (UTC)
 * I can do this. However, it is a bit strange to overload arguments the way you've done so that the page number goes in either 1 or 2; I would prefer that the page number goes in a consistent argument. If you put the page number in 2=, there's only one extra character (a vertical bar) to type compared with putting it in 1. What do you think? There are only about 33 current uses of page numbeers in . Also, what you call disp is standardly called alt. There are no current uses of disp so I'll go ahead and change it to alt. Benwing2 (talk) 16:30, 18 July 2021 (UTC)
 * I went ahead and changed disp to alt in, and ; this param wasn't actually used in any of these templates in any case. Benwing2 (talk) 16:35, 18 July 2021 (UTC)


 * Yes, page in 2 seems a better option. I'll change that. 🔥 शब्दशोधक 🔥 06:36, 19 July 2021 (UTC)


 * ✅. Now, can you please replace the pg and page in R:hi:Dasa with 2? 🔥 शब्दशोधक 🔥 07:02, 19 July 2021 (UTC)
 * Done. Benwing2 (talk) 07:21, 19 July 2021 (UTC)


 * I've changed R:inc-hnd:Platts and R:hi:Bahri. Please remove page and pg and move it to 2. Also please change all R:UR-HI-EN to R:inc-hnd:Platts since the template already has already has the shortcut R:Platts. Thanks. 🔥 शब्दशोधक 🔥 15:54, 21 July 2021 (UTC)

Template:id-IPA
Hi. Would you help me to create Indonesian IPA template? ―Rex Aurōrum｢Disputātiō｣ 04:38, 10 July 2021 (UTC)
 * I don't know anything about Indonesian. You'd need to specify in detail all of the pronunciation rules and give a bunch of examples. Benwing2 (talk) 04:40, 10 July 2021 (UTC)


 * Okay. Here the basic. There ara 26 letter in the alphabet. Each letter represent one sound. Same sounds as IPA: a, b, d, f, h, k, l, m, n, p, r, s, t, v, w, z. Different: c /t͡ʃ/, g /ɡ/, j /d͡ʒ/, q /k/, y /j/. Clusters: ny /ɲ/, ng /ŋ/, sy /ʃ/, kh /x/. ―Rex Aurōrum｢Disputātiō｣ 05:00, 10 July 2021 (UTC)
 * I need a lot more info than that, including a native speaker who also has a linguistics background. I don't really have time to devote a lot of energy to this right now. Benwing2 (talk) 05:05, 10 July 2021 (UTC)

I don't know Indonesian native speaker with linguistics background. But i made more comprehensive note. Can i contact you via email or Discord? ―Rex Aurōrum｢Disputātiō｣ 03:40, 11 July 2021 (UTC)

List of entries using specific headword parameters
Kutchkutch (talk) 12:00, 16 July 2021 (UTC)

Adding sub-modules to la-pronunc
Hi Benwing2,

Do you know how one would go about adding an additional pronunciation to the Latin module? For instance there is already a Classical one, an Ecclesiastical one, and a 'Vulgar' one. Let's say I want to add an additional one named 'X', with its own phonetic rules and such: how would I do this? I have tried to do so but ran into some trouble when the phonetic output of 'X' would ignore phonetic_rules_X no matter what I tried.

Regards,

The Nicodene (talk) 00:23, 17 July 2021 (UTC)

Substitution
Looking at, unfortunately the substitution seems not to have been successful.--Tibidibi (talk) 04:26, 19 July 2021 (UTC)
 * My apologies, I should have clarified, the change I made to Module:languages/data2 is just the first step. I want to make sure it doesn't cause any additional out of memory errors, and if so I will implement the remainder tomorrow, which involves some changes to Module:script utilities. Benwing2 (talk) 07:08, 19 July 2021 (UTC)
 * Oh I see, thank you again. Looking at Category:Pages with module errors it doesn't seem to have caused any noticeable issues, though you would know better than me.--Tibidibi (talk) 07:34, 19 July 2021 (UTC)
 * Really sorry, any updates on this?--Tibidibi (talk) 02:25, 23 July 2021 (UTC)
 * My apologies. I have not forgotten about this. It's just that I need to find 3 hours or so to sit down and implement and debug this, and my RL work keeps getting in the way ... Benwing2 (talk) 06:53, 23 July 2021 (UTC)

replacement
Hi, see Special:Diff/61964501. The bot replaced  with. The former was certainly an incorrect way to use the template, but nevertheless rendered sensibly unlike the latter. It would be good to check the content of the template argument and guard against that. –mwgamera (talk) 22:32, 20 July 2021 (UTC)
 * Hi. Yes, I discovered this issue awhile ago and fixed it in my script. I tried to correct the mistakes of this sort that the bot made, but I seem to have missed some. Benwing2 (talk) 04:58, 21 July 2021 (UTC)

et al.
Hi, kindly carry out the following replacements when you're free:


 * → (the number at parameter 2 is now redundant and can be deleted).
 * → (the number at parameter 2 is now redundant and can be deleted).
 * → (the number at parameter 2 is now redundant and can be deleted).
 * → (the number at parameter 2 is now redundant and can be deleted).

Thank you. — SGconlaw (talk) 18:30, 21 July 2021 (UTC)
 * Are you sure about removing the page number from ? The documentation says it's mandatory in some cases to make the link work, and the template code definitely makes use of the page number. Also if we rename 1 to chapter and rename or delete 2, we should also rename 3 to text so it doesn't end up stranded. Benwing2 (talk) 02:42, 22 July 2021 (UTC)
 * Yes, that is not a page number but is an artefact left over from an old version of the template which used to link to some other website. You can certainly rename any 3 to text or passage if you find any. In fact, if you find:


 * [The quoted text.]
 * [The quoted text.]


 * you can change it to:




 * However, if this is troublesome to program the bot to do, then don't worry about that part – I guess it can be done manually as and when someone comes across it. — SGconlaw (talk) 05:49, 22 July 2021 (UTC)
 * OK, it is running. I will look into incorporating the manually given quoted text into the template tomorrow. Benwing2 (talk) 06:59, 22 July 2021 (UTC)

et al.
Hi, kindly carry out the following replacements when you are free:



Thank you. — SGconlaw (talk) 20:45, 29 July 2021 (UTC)

A few more, please:



Thank you. — SGconlaw (talk) 14:38, 30 July 2021 (UTC)
 * Done. Benwing2 (talk) 02:37, 31 July 2021 (UTC)
 * Thanks! — SGconlaw (talk) 17:23, 3 August 2021 (UTC)

Index:Korean and subpages
I've came to the conclusion that Index:Korean, Index:Korean/Hanja and their subpages should be moved to the Appendix namespace, just like the corresponding Chinese indexes. Though I've barely touched Wiktionary's Korean content, I feel obligated to note that they appear to contain useful content that isn't found elsewhere on WT; their deletion would thus be a detriment. Alternatively, the content of the indexes could be transferred to a template/category system (this is probably the eventual desideratum); all that matters to me is the preservation of the content. Do you have any thoughts? Hazarasp (parlement · werkis) 10:02, 2 August 2021 (UTC)
 * I can see how Index:Korean/Hanja and Index:Korean themselves are useful, but not subpages like Index:Korean/ㄹ. These subpages buy you nothing over Category:Korean lemmas (which also has a table of contents indexed by character, see for example ), and they haven't been updated since 2008 so they are woefully out of date. Benwing2 (talk) 05:53, 3 August 2021 (UTC)
 * Notice the carefully noncommittal wording I used: "they appear to be useful". I'm not well-versed in Korean or WT's Korean content and only gave the index a quick look after moving Index:Korean/Hanja/KS X 1001 hangyol code and Index:Korean/Hanja/KPS 9566 hangyol code, so I was unable the full details of the situation.
 * By the way, the content of Index:Korean/Hanja seems to mostly be covered in Index:Korean, so preservation of Index:Korean/Hanja seems hard to justify. Hazarasp (parlement · werkis) 12:20, 3 August 2021 (UTC)
 * Makes sense, thanks. Benwing2 (talk) 02:21, 5 August 2021 (UTC)

Hello, when possible, please make the following substitution:



Thanks in advance! (as a sidenote, thank you for your work at Module:es-pronunc, it has been very useful for creating the Asturian and Basque pronunciation modules) --Santi2222 (talk) 15:17, 3 August 2021 (UTC)
 * Done; glad you were able to make sure of the existing module. Benwing2 (talk) 04:25, 4 August 2021 (UTC)
 * Thanks! --Santi2222 (talk) 11:52, 4 August 2021 (UTC)

et al
Hi, when you're free please do the following replacements:



Thank you. — SGconlaw (talk) 17:20, 3 August 2021 (UTC)
 * Done. Benwing2 (talk) 04:25, 4 August 2021 (UTC)
 * BTW I went through all the templates in Category:English quotation templates and compiled a list of all of them that have redirects to them, in User:Benwing2/english-quotation-templates-redirects. In this list, the template on the left is the canonical name, and the one on the right is a redirect pointing to the canonical name. If you go through the list and decide what should be the canonical name and which ones you want renamed, I can rename them all in one fell swoop. Benwing2 (talk) 04:58, 4 August 2021 (UTC)
 * Thanks! OK, let me have a look. — SGconlaw (talk) 05:15, 4 August 2021 (UTC)

ODBL reference templet
Hello. Please do the following replacement:

R:bn:ODBL → R:ODBL

The thing is, this reference is used not only for Bengali but also for Middle Bengali, Old Bengali, other Eastern Indo-Aryan languages like Oriya and Chittagonian, Magadhi Prakrit, Ashokan Prakrit, etc. The language code  is just needless. Thank you. ·~  dictátor · mundꟾ  09:41, 13 August 2021 (UTC)

Incorrect stressed form on Italian
Hi! Just a heads up that this WingerBot edit was incorrect. It should be citòfono and not *citofóno. See here. Andrew talk  (afc0703) 21:33, 26 August 2021 (UTC)

Module:hyphenation recent change incompat
See. Though this can be done by a bot who only takes care of pages before 2021-07, I strongly recommend checking the hyphenations with this Module by concatenating all the syllables. For example, zh:ako is in zh:Category:與詞彙不一致的斷字（hyphenation different from word）, so I found  in both zh:ako and ako. EdwardAlexanderCrowley (talk) 03:34, 27 August 2021 (UTC)

Category:English irregular plurals ending in "-ra"
I restored this because there are plenty of English irregular plural categories, I don't see a linked deletion discussion or post to the talk, and this category is still generated from the template. —Justin ( koavf ) ❤T☮C☺M☯ 18:07, 28 August 2021 (UTC)

Replacer
Hi, replace-guy! Could you replace all Aeropagitica with Areopagitica? Surprisingly, it was, and not Wonderfool, who made these errors. TVdinnerless (talk) 14:41, 2 September 2021 (UTC)


 * , pages including Citations:enchiridion were done by CrowleyBot. ps. I don't run CrowleyBot frequently, though. Crowley666 (talk) 16:04, 28 September 2021 (UTC)

Noun plural forms for Mirandese and Sardinian
Why did you remove these? I recreated one but want to get an explanation before I made more. Is there some conversation about this? If so, linking it in your edit summary would have been helpful. Your only deletion rationale was that it was empty but you emptied it without no explanation. —Justin ( koavf ) ❤T☮C☺M☯ 17:19, 6 September 2021 (UTC)
 * ??? —Justin ( koavf ) ❤T☮C☺M☯ 06:33, 1 October 2021 (UTC)
 * My apologies, I have not been intentionally ghosting you. RL has been interfering for several months and taking me away from Wiktionary. I think I posted to the Beer Parlour before removing noun plural forms of Romance languages, and got some agreement on this (and no disagreement). The basic reason I removed them is that 'noun plural forms' are entirely redundant in languages (as in most Romance languages, and almost certainly including Mirandese), where plurals are the only possible noun inflection. (Diminutives and augmentatives do not count as inflections. They are derivational forms that are their own lemmas, rather than non-lemma forms.) In other words, in these languages all 'noun plural forms' are 'noun forms' and vice versa so there's no point in having 'noun plural forms' as a category distinct from 'noun forms'. I did not remove 'noun plural forms' for any Romance languages where 'noun forms' existed that were not plurals. The same logic was used before me (by User:Rua) for removing 'noun plural forms' as a category in English, and maybe in other languages as well. Apologies for not including a better explanation in the commit message. Benwing2 (talk) 06:09, 7 November 2021 (UTC)
 * Gracias/merci/obrigado. If you say there was a consensus, I'll buy it. —Justin ( koavf ) ❤T☮C☺M☯ 14:39, 7 November 2021 (UTC)

avoir or être
Salut! On sortir and other pages (Special:WhatLinksHere/avoir_or_être) the table links to avoir or être instead of avoir and être separately LinguisticMystic (talk) 17:27, 25 September 2021 (UTC)

WingerBot adds false etymologies?
Please see my contribution to the tea room: Tea_room/2021/October PJTraill (talk) 12:13, 24 October 2021 (UTC)
 * Responded in the Tea Room. Benwing2 (talk) 06:38, 7 November 2021 (UTC)
 * Oops. Benwing2 (talk) 06:39, 7 November 2021 (UTC)

WingerBot made a redundant second “Etymology” for е́дучи.
See https://en.wiktionary.org/w/index.php?title=едучи&diff=next&oldid=52619564#Russian. Perhaps the “dated” tag in the first section confused it. Both sections basically say



but the first also has.
 * I think it actually got confused by the ==Verb== header instead of ==Participle==. You can see in the changelog that it says . The  tag shouldn't make a difference. I will see about fixing the code to handle this situation. Benwing2 (talk) 03:50, 12 November 2021 (UTC)

Deletion of Category:en:First Nations
I believe that your deletion of this yesterday was the wrong call. Instead of deletion, it needed to be integrated into the category tree (which, BTW, is waaaaay too complicated). Deleting it now means that Category:en:First Nations tribes is uncategorized. Please undelete. <b style="font-family:Verdana"><b style="color:#3A003A">Pur</b><b style="color:#800080">ple</b><b style="color:#991C99">back</b><b style="color:#C3C">pack</b><b style="color:#FB0">89</b></b> 15:33, 13 November 2021 (UTC)


 * My bot deleted it because it was empty, no other reason. But IMO we don't need Category:en:First Nations or Category:en:First Nations tribes; Category:en:Native American tribes is enough (and wrongly listed under "Canada" instead of just "North America"). "First Nations tribe" is just the Canadian equivalent of "Native American tribe", and in any case the two tribes you put under this (Assiniboine and Salish) exist in both Canada and the US. Benwing2 (talk) 19:39, 13 November 2021 (UTC)


 * I just happened to see this conversation and wondered, per this last point, if First Nations might also include tribes from South America? Which Native American doesn't cover, according to the government definition. At least, it wouldn't include Australia etc. DAVilla 11:46, 29 November 2021 (UTC)
 * If we are going by government definitions I think "First Nations" would only cover Canadian tribes. Wikipedia refers to and maybe that should be the category name. Benwing2 (talk) 04:02, 2 December 2021 (UTC)

weorþan broken
The conjugation for broke, as its Verner alternations no longer are applied when they are supposed to, generating things like *wurþon instead of. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 23:10, 16 November 2021 (UTC) Fixed. Benwing2 (talk) 01:41, 17 November 2021 (UTC)

aconteciese
Your comment was "delete bad Spanish verb form". I don't know Spanish too well, but this word is linked in the conjugation table and appears in other dictionaries, so I've restored it. DAVilla 11:41, 29 November 2021 (UTC)
 * Yes, my mistake. I was trying to delete all non-third-person forms and I seem to have added this by mistake to the list. Benwing2 (talk) 03:55, 2 December 2021 (UTC)

/ʀ/ in pt-IPA
The Brazilian phonetic transcription shows [h] in the template but the more common realisation is [ɦ]. Please look into it
 * You should use  to sign your comments. As for [h], I think this is in fact correct. I have spent time in Brazil and I definitely heard [h] in words like carro; if it had been [ɦ] I would have noticed it. [ɦ] does occur but primarily before voiced consonants as in carne, pardo. Benwing2 (talk) 03:59, 2 December 2021 (UTC)

Oh, sorry, I left that message from my phone and thought there was an automatic sign. RonnieSingh (talk)`

Oh, sorry, I left that message from my phone and thought there was an automatic sign. RonnieSingh (talk) 07:34, 2 December 2021 (UTC)

Your user pages show up in maintenance categories
Hey. See Category:Romanian_term_requests. Is there a way to fix this? Fytcha (talk) 12:53, 1 December 2021 (UTC)
 * This is because Module:links is inserting categories directly instead of using Module:utilities to format the categories, which will not put userspace pages and such in categories. This should be changed although carefully so it doesn't use more memory, i.e. it should respect the list of high-memory pages in Module:links/data. Benwing2 (talk) 03:53, 2 December 2021 (UTC)

Changing Prakrit's script
Hello. Recently the Prakrit editors have decided to change the main script of Prakrit from Brahmi to Devanagari. Is it possible for you to run a bot to change all instances of, say, to ? Similar change has to be done with such occurrences of some other templates also, but first let me know if you can do it. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 03:15, 2 December 2021 (UTC)
 * Yes, I can probably do something like that although it may take a little while to implement. Benwing2 (talk) 03:42, 2 December 2021 (UTC)
 * How much ever time it takes isn't really a problem. If you can do it, please replace  in these templates: l, link, m, mention, cog, cognate, nc, ncog, noncog, noncognate, desc, descendant. Same for alt, alter but if there are multiple continuous parameters like inc-pra then it'll have to be  . In etymology templates (der, inh, inh+, bor, bor+), for example,  would have to be changed to  . —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 04:38, 2 December 2021 (UTC)
 * Here are some additional considerations regarding this change:
 * Would the format at پیوݨ and descendants trees using sclb be affected by your suggested implementation scheme, or would they remain as is?
 * Do you have an opinion regarding the Latin script used by European scholars for Prakrit similar to Pali? There is a resource that advocates using the Latin script since a distinction could possibly be made between ṃ and nasalisation ̃ when the anusvara is used in a Brahmic script. Despite the potential convenience, it was decided not have Latin script for Sanskrit. Would the same still hold true for Prakrit?
 * I haven't attempted to assess what the script distribution is for Prakrit encountered by English speakers, or indeed by L1 English speakers. For Pali, a quick survey of the totality of Pali on the Internet put the Roman script in third place, behind the main two Thai writing systems.  (I've even seen some surprising evidence that the Sinhalese. but more precisely, those who could afford to buy, preferred the Roman script to the Sinhalese script for Pali!  Maybe the Roman script printing was more legible.)  I wouldn't be surprised if the Roman script were commoner than the Kannada script, though.  IAST already allows bindu and candrabindu to be distinguished.  We could also just have the Roman script as a 'soft redirect', but with transliterations themselves made into links for looking up inflections. --RichardW57 (talk) 22:47, 2 December 2021 (UTC)
 * Although early European scholars thought Maharastri and Jain Maharastri were two distinct lects, it was later found that they constitute a single lect. So would you consider adding Kannada script for Jain Maharastri so that ಗುಜ್ಜರತ್ತಾ appears on the headword line of the main entry? Kutchkutch (talk) 18:27, 2 December 2021 (UTC)
 * This is just a headword question, isn't it? We already have Kannada as one of the scripts in Module:languages/datax. --RichardW57 (talk) 22:47, 2 December 2021 (UTC)
 * @, : After Template talk:pra-noun, designed headword templates to show the Kannada script only when the lect is Maharastri since Bhagadatta wanted to be cautious about which lects are attested in the Kannada script and which lects are not. However, see Talk:𑀓𑀬: text. Therefore, at a minimum the Kannada script should display in the headword line for both Maharastri and Jain Maharastri lects. Should this cautious approach be disgarded, and should the Kannada script be used for all lects? Kutchkutch (talk) 03:13, 4 December 2021 (UTC)
 * I would favour the cautious approach. I'm confused by the documentation for the headword template. If a word is attested in some form of Maharashtri Prakrit, is that to be explicitly recorded?  Do we need some way of saying, 'All dialects, so also in Kannada script'?  There might be merit in allowing + to mean to give it in Kannada form, using automatic transliteration. --RichardW57m (talk) 11:19, 16 December 2021 (UTC)

For now, the Kannada script is displayed for both Maharastri and Jain Maharastri. I think it's okay to continue with the cautious approach and let the Kannada script form be there for (Jain/) Maharastri for now at least. The headword-line templates is coded to generate the Kannada form, if the values  or   are entered in any or it's parameters, to specify if the term is attested in those lects. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 15:31, 16 December 2021 (UTC)
 * FYI I already have code that knows about most of the templates out there with foreign-script text in them and which parameters need to be changed; I originally wrote it several years ago for vocalizing Arabic text based on the transliteration and have used it for various purposes since. As for پیوݨ and descendants trees with sclb, I may have to add special-case code for this; it would help if you could list as many examples as possible that have this sort of situation in them. Benwing2 (talk) 01:44, 3 December 2021 (UTC)

Categories for gendered nouns by language
Would it be appropriate to integrate categories of the following type into Category:Fundamental?
 * Category:Urdu nouns with gendered forms
 * Category:Punjabi nouns with gendered forms
 * Category:Marathi nouns with gendered forms
 * Category:Hindi nouns with gendered forms
 * Category:Gojri nouns with gendered forms
 * Category:Bengali nouns with gendered forms

Kutchkutch (talk) 18:56, 2 December 2021 (UTC)
 * Sure, although I'd prefer nouns with gendered equivalents or nouns with gender equivalents or maybe nouns with other-gender equivalents. "Forms" normally refers to non-lemma forms, and that's what I thought of when I first saw these category names. Note that we already have a category female equivalent nouns such as Category:Italian female equivalent nouns, which is generated by the template. Benwing2 (talk) 01:36, 3 December 2021 (UTC)
 * Thanks for the feedback. Although it's a bit wordy, nouns with other-gender equivalents seems like the most accurate of the three. I've never added or edited a category before, so could you explain how this would be implemented? Would it be added to Module:category tree/poscatboiler/data/lemmas? Kutchkutch (talk) 03:57, 4 December 2021 (UTC)
 * See . Benwing2 (talk) 23:40, 6 December 2021 (UTC)
 * Thanks for the implementation! Once all the hidden categories above have corresponding categories in Category:Fundamental, they would be deleted. For English, the equivalency of actor vs. actress could be considered more morphological than king vs. queen. Perhaps it should be left to the editors to decide the exact nature of equivalency. Kutchkutch (talk) 01:09, 7 December 2021 (UTC)
 * actor/actress and king/queen are just examples of nouns that may go into this category. It's common in descriptions of categories to list examples to help understand the nature of the category, and they're usually in English because it's not easily possible to make the category description be customized per language. I think it's reasonable to include an example of both morphologically-related pairs and morphologically-unrelated pairs because both sorts will typically end up in the category; but if you really don't like them we can remove them. Benwing2 (talk) 03:21, 7 December 2021 (UTC)
 * Those examples are fine since it's true that wiki. I've used cln in entries, but I wasn't aware that it ensures that userspace and talk pages don't end up in the categories. I'm not proficient at coding, but Module:hi-pa-headword is certainly very interesting. Kutchkutch (talk) 05:45, 7 December 2021 (UTC)
 * I did something similar in Module:uk-be-headword, since as with Hindi and Punjabi these two languages are very similar. BTW can you take a look at खत्रप? I tried to fix all the bad params in Punjabi terms but this one has the gender specified as neuter and I don't know what the proper gender is. Benwing2 (talk) 05:59, 7 December 2021 (UTC)
 * खत्रप with the meaning manure must be an error for this Kutchkutch (talk) 08:07, 7 December 2021 (UTC)
 * Perhaps due to the changes at Module:hi-pa-headword & Module:hi-headword, or in Template:hi-adj is now causing module errors such as at दसवाँ and सातवाँ: Special:PermanentLink/64051118. Kutchkutch (talk) 08:20, 7 December 2021 (UTC)
 * What is the purpose of this param? I intentionally removed it because it had no effect, and you've added it back without effect. Is it to mark ordinal numbers? If so we should create or similar. BTW the current state of Hindi numerals is messed up; e.g. we have both  and  doing the same thing, and they are being used for both cardinal and ordinal numbers even though they explicitly label the term as cardinal. Benwing2 (talk) 15:56, 7 December 2021 (UTC)
 * It does have an effect, it generates . —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 16:53, 7 December 2021 (UTC)
 * How? I don't see it in the code anywhere. In any case I've added, fixed and , and cleaned up all the Hindi numerals. Benwing2 (talk) 18:17, 7 December 2021 (UTC)
 * Never mind, I see it was in the code. In general this isn't the right way to add categories; they should be added to the module code. Benwing2 (talk) 18:25, 7 December 2021 (UTC)
 * OK, I see why you did this; in any case, we now have, which should be used instead. Benwing2 (talk) 19:30, 7 December 2021 (UTC)

Do you think it's right to change the "adjective" to "numeral"? Even English entries like seventh use "adjective" POS header. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 02:52, 8 December 2021 (UTC)
 * You are right; I fixed them. Benwing2 (talk) 03:54, 8 December 2021 (UTC)

Incorrect coding of |footer=
Hi, I made a mistake in the coding of the footer parameter that I added to various quotation templates. Could you do the following replace?



Thanks very much. — SGconlaw (talk) 19:22, 2 December 2021 (UTC)
 * Do you still need this fixed? I see you've already fixed some of the templates. Benwing2 (talk) 01:23, 3 December 2021 (UTC)
 * Yes, please. I fixed some of them but I’m not sure if I corrected them all. — SGconlaw (talk) 03:43, 3 December 2021 (UTC)
 * OK. I notice you didn't add footer to a lot of templates, is that intentional? Benwing2 (talk) 04:00, 3 December 2021 (UTC)
 * Fixed; there were 67 templates needing fixing. If you need other templates to include this, I can do that. In the longer run I should write some Lua code to facilitate creating quotation templates of this sort; this will make it so you don't have to forward all the parameters but will only need to specify those that differ from template to template. Benwing2 (talk) 04:13, 3 December 2021 (UTC)
 * Great, thank you. In the past I used to include footer on a case-by-case basis, but I think it’s probably better to just include it in all quotation templates.
 * Speaking about updates to Module:quote, other possible changes are (1) to check that all external links use https:// where possible (I noticed the WorldCat one is still using http://); and (2) in location and publisher (and location2, etc), to automatically link (or “S.l.”) and  to the glossary. I think I may have mentioned some of these things at Module talk:quote. — SGconlaw (talk) 05:25, 3 December 2021 (UTC)
 * Linking s.l. and s.n. can be done in the module itself. http:// occurs in the following templates:
 * Template:RQ:Doyle Land of Mist
 * Template:RQ:Glanvill The Vanity of Dogmatizing
 * Template:RQ:Lawrence Lady Chatterley's Lover
 * Template:RQ:Orwell Animal Farm
 * Template:RQ:Orwell Burmese Days
 * Template:RQ:Orwell Homage
 * Template:RQ:Wells William Clissold
 * Benwing2 (talk) 05:50, 3 December 2021 (UTC)
 * Thanks. I actually meant the URLs that are inside the module, like those that link to the Library of Congress, Open Library, and WorldCat. For URLs in quotation template, those linking to Google Books, HathiTrust, and the Internet Archive should use https://, but for some reason Project Gutenberg Australia doesn’t, which explains most of the templates you found above. — SGconlaw (talk) 06:20, 3 December 2021 (UTC)
 * BTW the following garbagey templates don't even include passage; you might want to fix them:
 * Template:RQ:Barrow Math — now . — SGconlaw (talk) 21:14, 4 December 2021 (UTC)
 * Template:RQ:Boyle Cold — updated. — SGconlaw (talk) 14:51, 6 December 2021 (UTC)
 * Template:RQ:Burke Noble Lord — updated. — SGconlaw (talk) 14:51, 6 December 2021 (UTC)
 * Template:RQ:zhx-teo:Fielde Dictionary
 * Template:RQ:Hale The History of the Common Law of England — now . — SGconlaw (talk) 21:13, 10 December 2021 (UTC)
 * Template:RQ:Irving A History of the Life and Voyages of Christopher Columbus — now at (and the two entries using this template were not from this work – sheesh). — SGconlaw (talk) 17:14, 3 December 2021 (UTC)
 * Template:RQ:Joseph Hall Hard Measure — now . — SGconlaw (talk) 18:33, 11 December 2021 (UTC)
 * Template:RQ:zhx-teo:Lim Hiong Seng
 * Template:RQ:Peake Titus Groan — SGconlaw (talk) 21:14, 4 December 2021 (UTC)
 * Template:RQ:Stevenson Wrong Box — replaced by (which already existed). — SGconlaw (talk) 17:14, 3 December 2021 (UTC)
 * Template:RQ:Taylor The Rule and Exercises of Holy Living — now . — SGconlaw (talk) 18:33, 11 December 2021 (UTC)
 * Benwing2 (talk) 05:56, 3 December 2021 (UTC)
 * Not created by me, obviously. OK, will have a look at them. — SGconlaw (talk) 06:20, 3 December 2021 (UTC)
 * Doing a run now to add footer almost everywhere, as well as to ensure that templates support both text and passage (and I standardized them to prefer text over passage, hope that is OK). Benwing2 (talk) 06:12, 3 December 2021 (UTC)
 * Guess that’s OK. I’m rather used to using passage, personally. — SGconlaw (talk) 06:20, 3 December 2021 (UTC)
 * Shouldn't matter as long as they aren't both used. Benwing2 (talk) 06:37, 3 December 2021 (UTC)
 * Speaking about problems with quote templates, there are about 65 pages in CAT:PFE that need fixing. You changed the first positional parameter for RQ:Milton Paradise Regained from being the book name to being the page number, so all of the entries that don't use the book parameter now have errors. Do you think maybe "book=" should be inserted in front of all first positional parameters that aren't numeric? Chuck Entz (talk) 06:33, 3 December 2021 (UTC)
 * Looks like this message from Chuck is directed to you. Benwing2 (talk) 06:36, 3 December 2021 (UTC)
 * Yes. You got to it before I could fix it. Thanks! Chuck Entz (talk) 06:38, 3 December 2021 (UTC)
 * oh, yes, could you please add book to those entries as Chuck suggests? Thanks. — SGconlaw (talk) 10:25, 3 December 2021 (UTC)
 * : never mind, I figured out a fairly quick way to do it myself. Chuck Entz (talk) 03:06, 4 December 2021 (UTC)
 * ah, great! How? — SGconlaw (talk) 04:57, 4 December 2021 (UTC)
 * Nothing high-tech: just assembly-line techniques with tabbed browsing and text searching to find where to paste without scrolling. It took me 27 minutes, but it was worth it to clear everything up. If I was really in a hurry, I could have done it faster. It does take a lot longer than it would with a bot, but I can just do it in less time than it takes to get a bot to start on it. Null edits are a lot quicker- I've cleared hundreds of entries from CAT:E in less time. Chuck Entz (talk) 06:55, 4 December 2021 (UTC)
 * I see! Thanks. — SGconlaw (talk) 09:36, 4 December 2021 (UTC)


 * You mentioned "garbagey" quotation templates - there are a load of them, probably all created by Wonderfool. I remember someone once searched for all RQ: pages without a "passage=" bit or anything else - there were over 1000. It's probably useful to have on a long-term to-do list the de-garbagifaction of suche templates Notusbutthem (talk) 16:22, 4 December 2021 (UTC)

Discord
Hey, would you like to join Discord server? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 12:16, 5 December 2021 (UTC)
 * Sure, probably a good idea to do so. Benwing2 (talk) 15:47, 5 December 2021 (UTC)
 * It indeed is nice to be there and would be good to have you. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 15:54, 5 December 2021 (UTC)
 * You said that it was a good idea to do so, but didn't actually join. Could you join today? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 06:13, 6 December 2021 (UTC)
 * You will find me there. Benwing2 (talk) 23:50, 6 December 2021 (UTC)
 * I DMed you regarding something important on Discord, please see it and reply there. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 05:58, 8 December 2021 (UTC)

Manlius
Hi Benwing,

In 2019, WingerBot made the a long, "per Bennett (with corrections by Allen and Michelson)." Do you have a link for that ref? It's not supported by Lewis and Short, who show all the vowels as short. (Please ping me if you answer.) kwami (talk) 21:58, 8 December 2021 (UTC)
 * Hi Kwami. Haven't seen you much around Wiktionary. The ref is here: I don't trust Lewis and Short at all for hidden quantities; their info is from the late 1800's and quite out of date in this respect. Benwing2 (talk) 22:52, 8 December 2021 (UTC)

Thanks! Great ref.

Yeah, I got a bit burned out after the fight about not being able to use IPA for Chinese. kwami (talk) 23:17, 8 December 2021 (UTC)

Marathi
Hello! Could you take a look at User:Kutchkutch/mr-decl, and let me know if this could be done eventually? I started this in 2017 using templates, but AryamanA pointed out that this has to be done by a module. Although there's no rush, it would be nice to at least be able to show the most common declension classes on entries. Kutchkutch (talk) 03:15, 10 December 2021 (UTC)
 * This should be done by converting the Hindi declension module Module:hi-noun. This will take awhile; Marathi looks quite similar to Hindi in its declensions but there are always subtleties. Benwing2 (talk) 03:43, 10 December 2021 (UTC)


 * Thanks for taking a look at it. Module:hi-noun is certainly very long, and I understand that it may take time to adapt it for Marathi. I just wanted to bring it to your attention.


 * AryamanA started Module:mr-decl based on an earlier version of the outline of declension paradigms at User talk:AryamanA/sandbox. However, as AryamanA was working on it, I realised that the outline had to account for subtleties that were not apparent at first. This is why I created a newer version based on your work for Hindi.


 * A year ago AryamanA started Module:mr-verb based on Module:hi-verb, after I created User:Kutchkutch/mr-conj, but he never had time to finish it. Although verbs are also important, the number of nouns would be expected to be greater than the number verbs. I believe AryamanA intended finish both declension & conjugation, but it seems that he’s been too busy in real life for the past year. Kutchkutch (talk) 15:13, 11 December 2021 (UTC)
 * I created User:Benwing2/todo to help me keep track of requests. Feel free to modify it or add to it. Benwing2 (talk) 18:40, 11 December 2021 (UTC)
 * Thanks for accepting the request and adding it to your todo list. I intend to continue editing the outline to try to make it as clear/comprehensive/accurate as possible. Feel free to ask any questions. Hopefully, AryamanA will be back to work on it as well. Kutchkutch (talk) 18:32, 13 December 2021 (UTC)

Replacement of unnecessary redirects
Hi, a while back you created a list of redirects to quotation templates at User:Benwing2/english-quotation-templates-redirects and suggested that I go through them and let you know which ones should be replaced. There are too many for me to handle at once, but if you don't mind I'll try to clear them in batches. Here is the first batch:


 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →
 * ,, (not sure if entries still use that) →

Also, if it is convenient, please (1) carry out the following replacement when you encounter it (probably best to do it only for the above templates and future batches that I post here rather than for every quotation template, as not all such templates may have the passage or text parameter yet):


 * This is the passage quoted.
 * This is the passage quoted.

→



and (2) on documentation subpages, please add:


 * footer – a comment on the passage quoted.

after the text or passage line if it does not already exist. (In some cases I believe I typed "a comment about the passage quoted" – just wanted to point that out so that the bot doesn't insert another statement about footer in those cases. I'm trying to remember to standardize the wording!)

Thanks! — SGconlaw (talk) 18:23, 11 December 2021 (UTC)
 * Done. Benwing2 (talk) 19:56, 11 December 2021 (UTC)
 * Well, except for modifying the doc pages, I'll do that a bit later. Benwing2 (talk) 19:57, 11 December 2021 (UTC)
 * Thanks! No rush. — SGconlaw (talk) 20:30, 11 December 2021 (UTC)

Replacement of unnecessary redirects (part 2)
Here is the next batch:



Thank you. — SGconlaw (talk) 14:17, 12 December 2021 (UTC)
 * Done. Benwing2 (talk) 19:56, 12 December 2021 (UTC)
 * Thanks! — SGconlaw (talk) 21:32, 12 December 2021 (UTC)

Proposal for improvements
Thank you for the incredible rewrite of and, it was a huge undertaking and a big improvement over what we had before. Using the foundation you built for Module:es-verb, do you think it would be reasonable to extend it to ? I'm envisioning something like a single definition with that would replace the four very complicated  definitions on hable. This would be much easier for humans to use, it would ensure that page shows all of the correct verb forms, and it would enable future editors to improve the presentation without needing to edit editing 600,000 pages. JeffDoozan (talk) 12:51, 13 December 2021 (UTC)
 * This is doable although the specification in would in some cases have to be more complex than just the verb infinitive; essentially you'd need to specify the same params as are passed to . It would work by generating the full conjugation and then searching through it to find forms that are the same as the page name. I've added it to User:Benwing2/todo, so it won't get lost. Benwing2 (talk) 02:59, 14 December 2021 (UTC)

Replacement of unnecessary redirects (part 3)
Here is the next batch:



Thank you. — SGconlaw (talk) 20:29, 13 December 2021 (UTC)
 * Done. Benwing2 (talk) 03:01, 14 December 2021 (UTC)
 * Thanks! — SGconlaw (talk) 09:43, 14 December 2021 (UTC)

Replacement of unnecessary redirects (part 4)
Here is the next batch:



Thank you. — SGconlaw (talk) 17:17, 14 December 2021 (UTC)
 * Done. Benwing2 (talk) 05:49, 15 December 2021 (UTC)
 * Thanks! — SGconlaw (talk) 10:11, 15 December 2021 (UTC)

Declensions and conjugation tables for Akkadian
Hi! I've been working on Akkadian for a while now and I'd like to add declension and conjugation tables to Akkadian Noun and Verb entries. I've seen you've done an amazing job with Arabic and I was wondering if you could help me. Unfortunately I have zero knowledge of programming, so I'm quite useless in that regard, but I do have the linguistic knowledge. Could I interest you in a collaboration? Thank you! Sartma (talk) 13:47, 15 December 2021 (UTC)
 * I can definitely help you although I may not have much time right now. I would have concerns about implementing noun or verb inflection modules using cuneiform due to the inconsistency of cuneiform spellings, but I see that Wiktionary lemmatizes Akkadian using Latin spellings, which avoids this issue. Benwing2 (talk) 04:23, 17 December 2021 (UTC)
 * @Benwing2 That's ok, I won't have much time until January myself. We can discuss about the cuneiform. Personally, I'm happy not to use it on inflection tables (I feel it makes everything heavier), but I know that other users here have stronger feelings about it and would like to keep it as well. If we limit ourselves to the basic cuneiform syllabary plus a couple common triliteral signs and stick to the Old Babylonian syllabic writing style, it shouldn't be impossible. Are you on Discord? It might be easier to communicate there. Shall I get in touch again in a month's time or so (mid January?)? Thank's again! Sartma (talk) 21:19, 17 December 2021 (UTC)

Spanish autoplurals
The templates, modules, or secret societies governing Spanish nouns are way above my pay grade ... can you fix something to make ciudadano de a pie have the correct plural? Br00pVain (talk) 21:47, 16 December 2021 (UTC)

Russian feminitive bot request
Hi,

I wonder if you could write a bot to create missing Russian feminitives (noun entries) from noun headwords where f exists? Nouns like -ница, -ица, -ка, иня, -ха, -ша and adjectival nouns in -ая?

They should be all stress pattern "a", gender=f-an. Examples:
 * 1)  (all of these -ка nouns have reducible stems if preceded by consonants)
 * , (the stress can be on the ending as with adjectives)
 * , (the stress can be on the ending as with adjectives)
 * , (the stress can be on the ending as with adjectives)
 * , (the stress can be on the ending as with adjectives)

I am also somewhat interested in generating feminine forms of surnames but I don't know how hard it is and it's of lower priority. There's no rush but maybe you could add it to your to-do list. Thanks! --Anatoli T. (обсудить/вклад) 00:55, 20 December 2021 (UTC)
 * The definition line could simply be something like
 * Sure. I already did something like this awhile ago for Arabic and can probably repurpose some of the code. The definition should probably use . As for generating feminine forms of surnames, you mean generating an entry for e.g. and ? This is hardly any more difficult than doing it for common nouns. The hardest part of either, I think, is handling the insertion of the actual entry: i.e. do nothing if the entry already exists, create a page if no page exists, create a Russian-language entry if an entry exists for another language but not Russian, add a new etymology section if necessary, etc. Pretty much the same thing that my script to create non-lemma forms does, and the logic will be the same whether I'm creating a common or proper noun. The only thing I'm uncertain about is what a feminine surname entry should look like. Should it be a lemma or non-lemma form? What should the definition look like (e.g. should it use )? Should it have a declension table, and if so, what should the table look like? Should it be the same as the masculine table but missing the masculine column? Benwing2 (talk) 01:23, 20 December 2021 (UTC)
 * Yes, sounds good for both. Yes, inserting into existing entries will be harder, sorry. The matter is further complicated somewhat by feminitives being rare, slangy, (low) colloquial or even rude, e.g. докторша, канцлерша, извергиня, президентша, врачиха but masculine forms may not even have f for those. I just need to keep checking those entries.
 * I'm OK to refer to the main (masculine) form for declension of female surnames, similar to because I don't think we have a specific declension template for them and this will be a duplication. What do you think would be best? They are sort of on the border between lemmata and non-lemmata. --Anatoli T. (обсудить/вклад) 01:41, 20 December 2021 (UTC)
 * I have just generated from green links (semi-automated). It's probably fine. Let me think about it. --Anatoli T. (обсудить/вклад) 01:46, 20 December 2021 (UTC)

New parameter for RFV and RFD templates
I was recently thinking of a new parameter for T:rfv and T:rfd, which would put a page into Cat:Candidates for speedy deletion. The purpose of this parameter would be to allow non-admins to close RFV/RFD's which involve page deletion. You have said earlier that adding features without changing the existing features of a template doesn't require consensus beforehand, so for the template, could you possibly tone down the protection to autopatroller from template editor so that I can edit it accordingly? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 06:23, 20 December 2021 (UTC)
 * Let me know when you're around and I'll downgrade the permissions temporarily so you can edit them. Benwing2 (talk) 01:56, 21 December 2021 (UTC)


 * Ok I'm here. Just downgrade the protection for maybe 2 hours, that should be enough. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 03:49, 21 December 2021 (UTC)
 * Should be working. Benwing2 (talk) 04:00, 21 December 2021 (UTC)
 * Thanks, I've edited rfv. An example of the new parameter is here. Do you agree or object to this change of adding ? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 04:43, 21 December 2021 (UTC)
 * I think this is OK with me. I wouldn't like a param name  (too short) but   seems fine. Benwing2 (talk) 04:49, 21 December 2021 (UTC)
 * Okay, thanks for the feedback. Do I have to take it to BP to get community approval or just community knowing about this? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 05:15, 21 December 2021 (UTC)
 * Dunno? Probably not? If someone objects then you'll presumably find out. Benwing2 (talk) 05:32, 21 December 2021 (UTC)
 * I've added the parameter to rfd also. It was a bit tricky choosing the wording; I finally ended up on this one: "The voting and discussion is closed now, and this page is awaiting speedy-deletion by an administrator. If you think this page should not be deleted, please start an undeletion discussion for the same." This is supposed to make it clear that after this even admins can't vote since the rfd is closed, they just have to delete this. Any suggestions or objections? Btw I think since rfd is a wiktionary process template, it deserves a template-editor level protection. Before you changed it to "autopatroller", it was just "autoconfirmed" and anyone with user account could edit it. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 10:14, 21 December 2021 (UTC)
 * Your changes seem fine to me and I upgraded the protection to template-editor level. Benwing2 (talk) 02:57, 22 December 2021 (UTC)

Replacement of unnecessary redirects (part 5)
Here is the next batch:



Thank you. — SGconlaw (talk) 18:37, 22 December 2021 (UTC)
 * Done. Benwing2 (talk) 03:44, 23 December 2021 (UTC)
 * Thanks! — SGconlaw (talk) 15:13, 23 December 2021 (UTC)

No vowels in RQ template
Can you get a list of all RQ templates without vowels? Partly to save SG pestering you every 5 minutes with redirecting. One to do is the ridiculously named Template:RQ:SWymn ChpngBrgh to Template:RQ:Weyman Chippinge Borough. Br00pVain (talk) 10:52, 23 December 2021 (UTC)

inh+, bor+ in Russian
Hey, did you discuss this with other Russian editors? Russian's relations with other languages aren't ambiguous, so using these templates doesn't seem helpful. Thadh (talk) 23:02, 26 December 2021 (UTC)
 * The main other editor is User:Atitarev, who I seriously doubt will object; if so I'll undo the change. The text on many or most pages already read "Borrowed from ..." or "Inherited from ..." so this is mostly just templatizing the text. Benwing2 (talk) 23:06, 26 December 2021 (UTC)
 * User:Tetromino may also be interested in this. Anyway, I think it's just generally a good idea to ask first, act later. Of course if nobody objects, you can freely ignore my opinion. Thadh (talk) 23:13, 26 December 2021 (UTC)
 * Hi. Which edits are you concerned about? Yes, what @Benwing2 says is right. Borrowed terms use, inherited and derived  (also preferred if borrowed via another language). --Anatoli T. (обсудить/вклад) 00:29, 27 December 2021 (UTC)
 * I understand your question regarding the templates with (+) plus and I support it, since it's an improvement - less typing is required to achieve the same thing. --Anatoli T. (обсудить/вклад) 05:31, 27 December 2021 (UTC)
 * from my point of view, the long debate and the series of votes about the + templates was pointless bikeshedding. If active editors in the Russian section started using the + templates, I will also use the + templates. Tetromino (talk) 17:40, 27 December 2021 (UTC)
 * Sure, in that case all's good. Wanted to make sure this wouldn't be a surprise to you all. Thadh (talk) 17:54, 27 December 2021 (UTC)

inh+, bor+ in Indo-Aryan
Relatedly, what do you think about these 2 templates in Indo-Aryan languages? In most of the cases, it's already "Inherited/Borrowed from" typed out, so adding the + templates to those pages would mostly be just templatisation. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 14:08, 28 December 2021 (UTC)
 * No objections from me. If other Indo-Aryan editors do not object, I have no issue adding them by bot. Benwing2 (talk) 03:59, 29 December 2021 (UTC)
 * @Benwing2: The other Indo-Aryan editors can be pinged, of course, but I think the votes + related discussions, etc. made it clear enough that we all are in support of that (well, with the exception of who is neutral but always types out the whole text). OTOH there are many more in support and some have already started using them: Taimoorahmed11, Rishabhbhat, Bhagadatta, Inqilabi, Imranqazi90, me, etc. —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 04:27, 29 December 2021 (UTC)
 * Which are the main languages needing conversion? Benwing2 (talk) 07:48, 29 December 2021 (UTC)
 * There're quite a few of them, but I'd say the "main" are hi, ur, pa, gu, mr, bn, as, ne, pi, inc-pra. Is it too difficult to do? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 07:52, 29 December 2021 (UTC)
 * Not too hard. I am saving the Assamese changes currently. These changes are done semi-automatically and I made a bunch of additional fixups to Urdu and esp. Bengali lemmas; many of the latter were really messy. But these additional fixups take significant time so I will probably skip them on the other languages. I included your username in the changes; hopefully you won't get pinged a zillion times ... Benwing2 (talk) 05:21, 30 December 2021 (UTC)
 * @Benwing2: Thanks a million for the amazing work!! :) But now that these 2 templates have been standardised in these languages, any template-to-text replacement would have a much much greater negative impact and disruption. Per Imetsia's essay, “… it is not time to rest on our laurels just because attacks on the templates have become less frequent. If they wanted to, any one template-opponent could effectively ban the templates under the aegis of Mahagaja’s holding. Even more troubling is how to deal with that potentiality …”, so at this point, where Victar's or any other opposer's template-replacement would be really devastating, I think a policy/rule might be needed to forbid such replacements before it's too late. Your thoughts? —<u style="color:#E21F1F; font-variant:small-caps; font-size:120%;">Svārtava <sup style="font-size:80%;">[t•c•u•r] 12:20, 30 December 2021 (UTC)
 * What sort of policy/rule are you thinking? My concern is that bringing this up at this point would just stir up more trouble than it's worth. Maybe better to wait until it actually becomes an issue? Benwing2 (talk) 01:47, 1 January 2022 (UTC)