Wiktionary:Beer parlour/2022/April

I do not habitually use diacritics in my writing, so when I wanted to look up German verdrt I spelled it, in the widely accepted alternative way, verdraengt. Wiktionary did not seem to recognise this spelling,which seems to me a limitation that could be avoided.

I'm sorry if this is the wrong place, or an over-Frequently Asked Question, but it's a bit hard to find one's way around the relevant areas. — This unsigned comment was added by 202.56.55.90 (talk).

The word "verdrät" doesn't exist and brings no hits in Wiktionary. If you want to search for a German word spelled with an Umlaut, you can type the word without diacritics, e.g. "Madchen" and linger, the word Mädchen will show at the top list. You can simply click on search and you will find similar results.

"Maedchen" (replacing "ä" with "ae" is a telegraph style) doesn't necessarily bring you the wanted result, unless someone makes a redirect page.--Anatoli T. ^{(обсудить}/^вклад) 05:52, 1 April 2022 (UTC)

By adding {{misspelling|Maedchen}} somewhere on the page for Mädchen, the latter will show up in a search for the former. (For example, search for naxalone.) For use for alternative spellings, the name of the template is a misnomer.  --Lambiam 19:30, 1 April 2022 (UTC)

@Lambiam: That spelling is conventional and not a misspelling by any means, just very unusual. It would be preferable if the digraphs were recognized as equivalent, but the equivalence will not generalize. Confining such substitution to German results is technicly not feasable as far as I can tell. I have no experience with Elastic search, so I could be wrong. ApisAzuli (talk) 07:55, 7 April 2022 (UTC)

I realised that you were searching for verdrängt (you missed "ng" in your example - "verdrt"). Same method will work. --Anatoli T. ^{(обсудить}/^вклад) 05:54, 1 April 2022 (UTC)

Should we list such attestable archaic spellings, which are even found with the ⟨æ⟩ ligature as {{alternative form of}}?  --Lambiam 19:18, 1 April 2022 (UTC)

We have such “telegraph” spellings for Esperanto (both x system and h system), why not for German? MuDavid 栘𩿠 (talk) 00:35, 6 April 2022 (UTC)

I'm pretty sure that options like adding alternative spellings in a bot run have been discussed before and have, apparently, not lead anywhere. ApisAzuli (talk) 07:59, 7 April 2022 (UTC)

A bot cannot verify that such a specific spelling is attested.  --Lambiam 10:41, 7 April 2022 (UTC)

Spellings do not need attestation. The difference between ä and ae in German is almost equivalent to a mere change in font. MuDavid 栘𩿠 (talk) 03:21, 11 April 2022 (UTC)

I don't see why not. Vininn126 (talk) 10:51, 7 April 2022 (UTC)

FYI: https://languagelog.ldc.upenn.edu/myl/rysyn2017-final.pdf —Justin (koavf)❤T☮C☺M☯ 02:32, 2 April 2022 (UTC)

Someone added a category at паляниця as a Ukrainian shibboleth. I removed it as we don't have this scheme and it wasn't prefixed by a language code, but I wonder if others in the community think it's worthwhile to add this to a module and have a scheme for them. I think it is but I don't feel strongly enough to start adding it to lollapalooza and other such entries. Thoughts? —Justin (koavf)❤T☮C☺M☯ 02:52, 3 April 2022 (UTC)

We've got Appendix:Shibboleths - I think that probably covers it? (especially as there aren't that many in the list). I've added a link to this on паляниця#See also. Voltaigne (talk) 15:55, 6 April 2022 (UTC)

The so-called “usage example” needs to go, though. That sentence is completely unilluminating. — Sgconlaw (talk) 11:32, 9 April 2022 (UTC)

https://lareviewofbooks.org/article/how-dictionaries-define-us-margaret-boyle-and-ilan-stavans-in-conversation/ —Justin (koavf)❤T☮C☺M☯ 21:36, 3 April 2022 (UTC)

Interesting, thanks for sharing. " only took a month from when it was first uttered to its appearance on the company’s website." –Jberkel 19:34, 14 April 2022 (UTC)

@Jberkel: Similarly, see here and here. —Justin (koavf)❤T☮C☺M☯ 03:16, 15 April 2022 (UTC)

Anyone have any objections if I add a language code for Proto-Cariban (sai-car-pro)? It’s popping up in a good deal of etymologies and seems fairly secure/reconstructible. — Vorziblix (talk · contribs) 16:24, 6 April 2022 (UTC)

Well, it’s done. Next work is to add some widely-recognized subfamilies within Cariban, following the scheme in the right-hand column at User:Vorziblix/sandbox/Carib and using various exceptional sai- codes. (I’ve tried to be relatively conservative in what groupings to accept.) I’ll keep on working unless/until anyone objects. — Vorziblix (talk · contribs) 02:45, 8 April 2022 (UTC)

It seems odd that despite both the labels giving "humorous" as the display output, we have Category:Jocular terms by language. I think it would be more consistent if it is changed to Category:Humorous terms by language to match the label; thoughts? —Svārtava (t/u) • 14:22, 7 April 2022 (UTC)

Support. I'd also oppose changing the label to jocular because it's pretty dated these days. Theknightwho (talk) 14:26, 7 April 2022 (UTC)

Support. I was actually confused a few days ago with the current category title. AG202 (talk) 18:03, 7 April 2022 (UTC)

Support. "Humorous" is better than "jocular" IMO. Equinox ◑ 18:14, 7 April 2022 (UTC)

I have no strong feelings one way or the other. I can't really imagine a language having a distinction (I could be wrong). I never felt like jocular was all that dated, but right now the split seems off, unless there's some language out there with this bizarre distinction. Vininn126 (talk) 19:09, 7 April 2022 (UTC)

Support. Binarystep (talk) 00:11, 8 April 2022 (UTC)

Support. User: The Ice Mage ^{talk to meh} 13:41, 8 April 2022 (UTC)

Support: I've often wondered why this is, too. — Sgconlaw (talk) 17:20, 8 April 2022 (UTC)

Support. --Daniel Carrero (talk) 10:10, 10 April 2022 (UTC)

Support. Imetsia (talk) 20:32, 10 April 2022 (UTC)

Support. Sartma (talk) 09:50, 11 April 2022 (UTC)

Support merging into "humorous". — This unsigned comment was added by Koavf (talk • contribs) at 03:16, 15 April 2022 (UTC).

Support, but jocular gets to intent and humorous gets to result. I find many items on the category unfunny, though intended in jest. DCDuring (talk) 14:57, 15 April 2022 (UTC)

@DCDuring: Could you better illustrate your point with an appropriate example? —Svārtava (t/u) • 15:21, 15 April 2022 (UTC)

I have yet to find a category member that made me smile, let alone laugh. DCDuring (talk) 18:51, 15 April 2022 (UTC)

I think someone's just a grouch. Vininn126 (talk) 18:52, 15 April 2022 (UTC)

I never said that about anyone here. DCDuring (talk) 20:53, 15 April 2022 (UTC)

Oppose: Jocular sounds more formal than humorous, so the category name is fine. Further, jocular is probably a more comprehensive term, as we define it as ‘humorous, amusing or joking’. ·~ dictátor·mundꟾ 17:10, 15 April 2022 (UTC)

Support, though I think DCDuring correctly identifies the distinction between the words. Andrew Sheedy (talk) 07:00, 16 April 2022 (UTC)

•  Done. —Svārtava (t/u) • 16:32, 4 May 2022 (UTC)

See e.g. 'Obby 'Oss. Although Northern Ireland is part of the UK, this category is misleading because it now looks as though this traditional (from Cornwall, England) is somehow specific to Ireland! Equinox ◑ 20:18, 7 April 2022 (UTC)

It's because of Module:labels/data/regional#L-3828. 98.170.164.88 20:21, 7 April 2022 (UTC)

Thanks. So, we should remove that category, or is there disagreement here? Equinox ◑ 20:24, 7 April 2022 (UTC)

@Equinox, I've removed it for now. AG202 (talk) 02:16, 8 April 2022 (UTC)

@AG202 I added this when I was very new to Wiktionary. I do think Northern Irish English needs to be handled better (but I'm unsure of the best way to do it). "Britain" and "UK" are not the same thing, after all. Theknightwho (talk) 09:14, 8 April 2022 (UTC)

The labels “Britain” and “UK” should not be adding entries to categories of regional varieties of English. — Sgconlaw (talk) 11:36, 9 April 2022 (UTC)

You can find this message translated into additional languages on Meta-wiki.

More languages • Please help translate to your language

The Trust and Safety Policy team published the results of the Universal Code of Conduct Enforcement guidelines ratification vote. The vote ended 21 March 2022. See the results and read more on Meta-wiki.
--Mervat (WMF) (talk) 22:03, 7 April 2022 (UTC)

While editing the article Earth Overshoot Day, I found it linked to Category:en:Ecology. This category, however, seems to be populated with entries pertaining to the field of science and not nature conservation. I tried searching but could not find a more fitting place for the term. Does one exists? Should one be created? brittletheories (talk) 14:24, 9 April 2022 (UTC)

--Mervat (WMF) (talk) 21:18, 11 April 2022 (UTC)

Currently, a lot of Sanskrit entries are formatted like दन्त (danta), with abbreviated references to works that use the word included on each definition line. I think it would be more useful to readers to convert these to quotations. 70.172.194.25 13:37, 12 April 2022 (UTC)

That's because they're basically Monier-Williams dictionary entries converted to wikitext. As I've said before, this is really bad from a usability standpoint. The problem is that there's a lifetime of work that went into gathering all that data, and it will take another lifetime of work to find all of those quotes and copy them into the entries. So, yes, it would definitely be more useful to readers to convert them to quotations, just as it would be nice to end poverty, war and injustice- but how? Chuck Entz (talk) 14:10, 12 April 2022 (UTC)

Are we planning to abandon Wiktionary? If not, we can crowd-source the job, making sure that the product remains useful throughout. Also, {{sa-a}} could be changed to give the name of the work, probably ideally as a Wikipedia link (though for ill-guarded articles we may need to reference a local appendix instead, and we may need a guide as to the most useful sources for the text and translation), and we can also search for quotations. I'd also like to see dates of some sort, though I appreciate that India tends not to do dates. I would be inclined to keep the MW references even when expanded to a quotation so not to be thrown by disputed quotations. We may want to incorporate things like hymn numbers in the argument list of the template. --RichardW57m (talk) 14:58, 20 April 2022 (UTC)

It would probably be a good idea to have a template with an accompanying category so sanskritists with time on their hands know where to concentrate their efforts. Given the use of abbreviations that refer to an appendix, it might be possible to look for Sanskrit entries that link to that appendix. For that matter, we could even use the specific abbreviations to compile lists of the texts referenced, so that someone could go through, say, the Mahabarata, could find the entries that need quotes from it. Chuck Entz (talk) 15:09, 20 April 2022 (UTC)

I envisage a 650-case switch for the implementation of {{sa-a}}, amalgamating the 650 branch Template:temp:saAbbrName2No and Appendix:Sanskrit abbreviations, with the content of the latter being expanded. One can stick the categories into the text delivered, splitting the categories by book where appropriate. If the category is for managing the project, then we might add a parameter |found=1 to remove lemmas from the category as we add the quotations. --RichardW57m 12:31, 21 April 2022 (UTC)

Incidentally, I'm limbering up to use the Bhagavad Gita to provide quotations for the use of words in the Thai script. There are multiple out of copyright translations around. As ever, the logistics of demonstrating durable archival seems to be the main technical challenge, especially given the generally poor quality of Thai typesetting. I'll also use the Thai script quotation for the main, Devanagari lemma if it doesn't yet have a quotation, but I won't count that as 'found'. Or should I? --RichardW57m 12:31, 21 April 2022 (UTC)

This doesn't seem to be reflected on Wiktionary, for example ember doesn't show the fact that the two vowels are different. The exact pronunciation is different by dialect, but at least some indication should be present. There are minimal pairs which are only distinguished by this difference, for example "mentek" can have 3 different meanings depending on which e is open vs closed (and even the fourth combination can be understood although it's not in common use). — This unsigned comment was added by Gyorokpeter (talk • contribs) at 07:58, 14 April 2022 (UTC).

So how are you going to solve this? Fay Freak (talk) 18:46, 14 April 2022 (UTC)

One way to indicate the difference is to use "e" for open e and "ë" for closed e. This would be similar to how other languages have "dictionary-only" features like macrons for Latin or accent marks for Serbo-Croatian which are not used in normal writing (and so also not reflected in article titles).

However since this would require changing many pages, obviously many would remain unchanged so the page would need to indicate if it uses the distinction, as it also needs to extend to the suffixes in the inflection tables. Gyorokpeter (talk) 06:41, 15 April 2022 (UTC)

I've noticed that there are a lot of English surname entries having {{R:en:DAFN}}, but without any corresponding entry. Example: Julliard. This is perplexing, especially because ancestry.com's interface makes it trivial to check whether the entry exists or not (even if the formatting of the text itself sometimes leaves something to be desired). Is there a reason why this template has been used in this way? I don't know of any other reference template used so willy-nilly. Perhaps the argument is that it doesn't hurt, and can be useful in the cases where the entry does exist, but I think it can cause confusion. See e.g. Template talk:R:en:DAFN where an experienced user came across such an example, assumed there was just a paywall to access it, and made an account on ancestry.com just to find out that the reference wasn't there at all. 70.172.194.25 17:30, 14 April 2022 (UTC)

It does cause confusion. I have asked @Samubert96 not to use the template when DAFN has no corresponding entry. If the idea is linking ancestry.com which has interesting content in addition to DAFN, then a separate template should be made specifically for ancestry.com. Vahag (talk) 18:37, 14 April 2022 (UTC)

Or say explicitly in brackets or after a comma that no entry was discerned at the time of the accessdate. The links are also for comparison. (“Pree, we are better!”) Fay Freak (talk) 18:44, 14 April 2022 (UTC)

But the template is for an actual existing book printed in 2003. If the entry is not in it, the accessdate to a website later does not matter. Vahag (talk) 18:49, 14 April 2022 (UTC)

Ah, ok, then it would be also confusing to mention the accessdate as such, and one may do it otherwise. (Sometimes it is intransparent if a web resource based on a dead-tree book gets updated.) Fay Freak (talk) 19:06, 14 April 2022 (UTC)

I personally use DAFN also to access the profiles of historical people with a specific surname, because it can be helpful in finding the origin of a particular surname that has changed over time. For example, if you scroll down the DAFN page related to Julliard you can find a list of persons with this surname and discover that the surname is of French origin, even if there is no specific entry about the surname itself. If there are no entries and no historical profiles, then of course it's useless to insert the DAFN template Samubert96 (talk) 19:33, 14 April 2022 (UTC)

That content is not from DAFN, but from Ancestry.com which has licensed DAFN and presents DAFN content in addition to its own content. You can create a separate {{R:Ancestry.com}} template for Ancestry.com, using {{cite-web}}. Vahag (talk) 19:46, 14 April 2022 (UTC)

It would be possible to generate a list of DAFN links that don't actually exist in DAFN. Once completed, we could either remove the template from all those pages, replace it with {{R:Ancestry.com}} on those pages, or make some kind of hack such that the DAFN template changes its output on those pages? (Would only require basic Lua, but would save us from having to make potentially thousands of bot edits to entries. But the semantic template would still be wrong.) 70.172.194.25 19:34, 17 April 2022 (UTC)

Hello all,

Something I've noticed is that many languages have duplicate entries in different scripts where the same definition lines are repeated. Some examples:

• Ladino לשון and Ladino lashon
• Inuktitut aiviq and Inuktitut ᐁᕕᖃ (efiqa)
• Serbo-Croatian го̀нити and Serbo-Croatian gòniti

I understand that for Serbo-Croatian in particular there could be political reasons at play (Croatian uses only Latin, Serbian uses both), and that there is a policy at Wiktionary:About Serbo-Croatian § The headword. But that doesn't apply to many other languages that do this. I'm wondering if other people see this as sub-optimal or whether it's just me.

It requires manually keeping multiple pages in sync, which does not always happen. Compare the Ladino entries above: one has the etymology and IPA, the other has more definitions, synonyms, and a derived term. It feels like it would be better to have all of this in one centralized place and direct everyone there. On the other hand, the benefit of this approach is that the information is presented to the user in one click instead of two, and it doesn't run into the problem discussed at Wiktionary:Beer parlour/2022/February § Non-lemma form entries are confusing to casual readers. These advantages cannot be dismissed entirely. But by the same token you could argue that we should do away with {{alt form}} and instead manually copy the text of an entry to every attested spelling variation, as is done for these script variations, which seems a bit absurd.

The way e.g. Sanskrit handles this seems reasonable to me. For example, the Khmer-script Sanskrit entry ភូមិ (bhūmi) gives a short gloss of the main definition and a quotation of the script-specific form, and then links to the Devanagari Sanskrit entry where you can find much more detailed information. Copying all of that would increase complexity significantly, especially if it ever got out of sync and you needed to do a merge.

A third way would be some kind of technical wizardry to automatically transclude the definitions from one page to the other, but it seems like there are a lot of ways that could go wrong if the implementation isn't perfect. Making one form canonical and linking is simpler, even if it forces one to pick favorites.

Does anyone have thoughts on this matter? 70.172.194.25 08:11, 17 April 2022 (UTC)

I'm not sure that I have the perfect answer here, but we may want a solution like #3, which is basically what happens on Chinese entries between simplified and traditional script, but I think it would be hard to choose which script is the canonical one. Otherwise, a bot? Or hosted lexemes at Wikidata? —Justin (koavf)❤T☮C☺M☯ 15:08, 17 April 2022 (UTC)

I can't work out what you mean by '#3'. Three distinct methods are being used or are considered:

1. Manual duplication. Hard for two writing systems, infeasible when there are many. Reportedly the Serbo-Croat solution.
2. Chinese/Pali/Sanskrit (incipient)/English mechanism, whereby one form is chosen as the Wiktionary main lemma and others are treated as alternative forms. There are simple rules for choosing the Wiktionary main lemma for Chinese, Pali and Sanskrit, while English is more arbitrary. Incidentally, the quotation is used to verify the spelling of the alternative forms for Pali (it isn't always obvious) and because some of the Sanskrit community want evidence of use for non-Devanagari forms.
3. Automated duplication. This needs wizardry, and the data may need writing-system-specific tweaking. Possibly too difficult for most users, which for example currently makes Pali inflection tables for irregular words a pain. --RichardW57m (talk) 10:30, 22 April 2022 (UTC)

My preferred practical solution is the Chinese/Pali/Sanskrit solution. If there are attestation problems for the main form, then select the most prominent of the others, and note in its entry that the 'main form' is not attested. Where relevant, I think inflection tables, where appropriate for the language, should be provided for all forms. (Sanskrit inflection templates, last time I looked, suffered from being implemented via Romanisation, which only works if one can make the Romanisation work reversibly.) --RichardW57m (talk) 10:30, 22 April 2022 (UTC)

Entries often consist only of grammatical information, even when that doesn’t help with meaning. As one example, the entry for Spanish despacito says merely that it’s a diminutive of despacio. Well, for anyone even remotely familiar with Spanish—admittedly far from every Wiktionary user—the -ito ending makes the diminutiveness obvious. But what is not obvious is what, exactly, a diminutive of despacio might mean. Is it “somewhat slowly”? “Extremely slowly”? Some third thing?

Off the top of my head I can recall no other example of such entries in which some defining would be useful, but the resulting feeling of disappointment and frustration from encountering them feels pretty familiar. This lexicographic practice isn’t too different from providing an etymology and nothing else. Is this an affirmative Wiktionary practice? If so, then why? And if not, then what should a user do? Squawk in the tea room?—PaulTanenbaum (talk) 13:24, 17 April 2022 (UTC)

Don't forget to look at the quotation(s) if any, it's translated as "A little bit slowly" there. I might have gone with "nice and slow" General Vicinity (talk) 13:40, 17 April 2022 (UTC)

I've always thought diminutives should be handled in the main POS template, much like gendered nouns and Japanese kyūjitai spellings. Binarystep (talk) 00:40, 18 April 2022 (UTC)

As the template used in this example, {{diminutive of}}, is 'not meant' to be used in etymology sections, I can only include that the meaning of diminutives is meant to be part of the grammar of Spanish, just like the meaning of verb forms. That then raises the question of whether Wiktionary should document the default meaning of Spanish diminutives. --RichardW57 (talk) 08:02, 22 April 2022 (UTC)

As I am going through some of these entries, I understand one reason why I wanted adjectives so badly - I mostly want relational adjectives, which are the Latin/Slavic equivalent of attributive nouns. Seeing as we are going to allow nouns and verbs, what if there was a change for relational adjectives? This would not include other kinds. I realize this is awkward timing, but it didn't occur to me until I started implementing the change. Vininn126 (talk) 15:32, 17 April 2022 (UTC)

e.g. this Vininn126 (talk) 16:26, 17 April 2022 (UTC)

@Vininn126 Just go ahead and add collocations to relational adjective entries, I reckon. If anyone objects we can revisit it. But it seems so eminently within the spirit of the concept that I can't see an issue with it. This, that and the other (talk) 05:24, 19 April 2022 (UTC)

--Mervat (WMF) (talk) 12:01, 18 April 2022 (UTC)

In editing thousands of etymology sections, there is a specific issue I have not infrequently encountered. For background, {{der}} is a very overloaded template; it is primarily intended for indirect derivations that do not qualify as inherited ({{inh}}), directly borrowed ({{bor}}), calqued ({{cal}}), or so forth. In practice, it is further used as a "catch-all" for cases where we want to say that a term ultimately derives from some source, but the exact pathway is unclear to the editor. (For example, is a Spanish term inherited from Latin, or was it a later semi-learned borrowing? Was an English word borrowed from Latin directly, or borrowed via some stage of French?)

In cleaning up uses of {{etyl}}, a lot of work has been done in properly sorting derivations into their correct types. But even after {{etyl}} is gone, this issue will remain. For this reason, I think it may be a good idea to create new templates, say {{der?}}, {{bor?}}, {{inh?}}, which would produce the same textual output as their regular counterparts, but also add some sort of categorization or tracking link to enable editors to find derivation templates that may need to be corrected. It would also be possible to just use {{rfe}} or {{attention}} in such cases, but I personally think that being able to find this more specific type of etymology question would be helpful; and {{rfe}} is a little too "loud" for such cases.

Anyway, I'm just floating the idea. There may be people who dislike it, and if so, the proposal can be discarded. 70.172.194.25 19:30, 19 April 2022 (UTC)

How would this interact with {{bor+}} and the like? Vininn126 (talk) 22:06, 19 April 2022 (UTC)

See {{der?}} and {{der+?}} for how I would imagine it working. The category scheme I added is just an example, inspired by Category:etyl cleanup/en/fr, but there should be some way of tracking it. It might also be possible to add a parameter to the templates like {{der|unsorted=1}} instead of making new templates, but I like the der? actually because it's simple.

I guess I'm mainly trying to bring up the fact this issue of unsorted derivations exists already, and people may want to add them in the future, and here is one potential way we could go about tagging them. But I'm not wedded to the particular solution. 70.172.194.25 22:20, 19 April 2022 (UTC)

I could see this being useful, however if people aren't too keen on a whole new category or something, having it categorize to the appropriate rfe or attention would be a compromise. That is, using a ? template would add the word to one of those two categories (probably rfe?). Vininn126 (talk) 22:26, 19 April 2022 (UTC)

What about something like {{etycheck}} (placed above the etymology and categorising the page) instead? I take it that the distinction in inh/bor/der isn't really important here, it's more about creating some central category for uncertain etymologies. Thadh (talk) 22:58, 19 April 2022 (UTC)

I like that better, as I was leaning towards sending them to one category anyway. Vininn126 (talk) 23:02, 19 April 2022 (UTC)

1st, while I salute your interest in precision, outside of documented scientific terminology and the like, we simply do not know the exactly precise paths that each word comes to us from. If Newton calques some word from a Latin text while knowing that the French are already using something similar, does his knowledge of the French matter? If someone who only knew French used the same word in their book 3 years earlier, does it matter if all subsequent English usage derived from Newton's? It's actually counterproductive to pretend that our templates make any provided etymology unimpeachably correct. Surely, this is just like the 10s of 1000s of entries that use the prefix and suffix templates for things that look like internally constructed English words but actually derive from Middle English forms borrowed from French and Latin: the wrong etymology sits there until someone goes to a more authoritative source, fixes it, and cites their work. Having even more templates for people to memorize isn't going to change that process at all and risks making the wrong ones look more authoritative from formatting.

2nd, what would this even look like? All the current {{der}} templates display "ultimately from ..." and the {{der platinum}} templates would just function normally? — LlywelynII 12:20, 22 April 2022 (UTC)

I agree that there are inherently grey areas. The scientific coinage example you suggested is one. In other cases it's just uncertain. I'm fine with leaving {{der}} if it's the best we can do.

But there are entries where I know it's not the best we can do, but I have no good way of tagging this fact directly for further review. The thing is that it's often not entirely obvious to me which more specific template is appropriate, so I refrain from categorizing more specifically when I see this. (E.g., I might guess that a Spanish term derived from Latin is inherited if it has significant sound changes, and borrowed if it seems uncommon and/or nearly identical to the Latin, but I'd rather leave this to people like User:Nicodene who know what they're talking about.)

In the hypothetical I suggested, the displayed text would be the same, but it would add a hidden maintenance category to allow people to go through and clean up the etymology. And all the other etymology templates would stay the same as they are currently. I imagine that the alternative {{etyl check}} idea would work similarly in terms of categorization, but it would be on its own line. 70.172.194.25 19:28, 24 April 2022 (UTC)

Why not retain the already existing etyl for cases where neither bor, nor inh, nor der is the desired description for certain language-specific situations? The explanation might be different for each language and could be discussed at the "About:Xxxx language". E.g. We have such a problem for Modern Greek words from Ancient Greek (but not inh), as {{lbor}}, but not external lbor, but internal, which is a vast special Category, not available in en.wikt. Thank you. ‑‑Sarri.greek ^♫ I 12:25, 22 April 2022 (UTC)

For one thing, the order of parameters to {{etyl}} is inconsistent with all the other derivation templates, and it makes you use {{m}} directly afterwards. 70.172.194.25 19:28, 24 April 2022 (UTC)

Do changes to Module:scripts/data need to be co-ordinated? It seems to me that changes would trash most of the Wiktionary page cache. I have one correction and a trio of tweaks to make to the Eastern Nagari 'scripts', Beng and as-Beng. Firstly, U+09B2 ল LA needs to be added to as-Beng. Secondly, the letters ড় U+09DC RRA, ঢ় U+09DD RHA and য় U+09DF YYA may as well be removed from both Beng and as-Beng, because the characters cannot appear in NFC text such as Wiktionary pages. Or should I just go ahead and make the changes? --RichardW57m (talk) 12:46, 20 April 2022 (UTC)

Since no one bothers to reply to these technical matters that you post, you can go ahead and do whatever changes you think should be made. I for one don’t understand these intricate things, and unless you screw things up, no one would even care what you have done. ·~ dictátor·mundꟾ 14:40, 22 April 2022 (UTC)

Would en.wiktionary consider the title "Other forms" instead of "Alternative forms"? I find that the title 'alternative' gives an immediate signal "you may also use this form instead". Which is not true in many cases. Because most of the time, the context for the use is different. Thank you. ‑‑Sarri.greek ^♫ I 12:05, 22 April 2022 (UTC)

Other forms would be a much clearer signal that "you may use this form instead", so it's completely counterproductive for your purposes. You just need to label them with { { q | } } glosses to tell people what's obsolete, archaic, &c. regardless.

On the other hand, it is shorter and much easier Germanic wording that would be clearer to ESL users. Weird it took until now for someone to suggest it. — LlywelynII 12:09, 22 April 2022 (UTC)

@LlywelynII, Sarri.greek Using Germanic wording generally makes things obscurer for European non-native speakers. My gut interpretation tends towards that of Sarri.greek, but perhaps there is no common agreement on the nuances. However, the term 'other forms' could also be taken to include inflected forms, which are generally not 'alternative' forms. (This, of course, assumes that the head form can be used in the language in question.) --RichardW57 (talk) 00:21, 24 April 2022 (UTC)

@LlywelynII, RichardW57 Thank you for your comments. Sorry that i asked a question without understanding fully the words 'atlernative' and 'other'. Not being an anglophone, I took alternative as in 'there is also an alter choice / It's ok if you also alternate with these forms', and other, like 'something else / see also these forms'.

Of course there is the template {{q}} after each form, which describes its characteristics. The problem of 'alternative' as a prompt 'you may also use these forms' arises sometimes e.g. in Modern Greek for etymological spellings, proscribed spellings, or popular misspellings. PS I did not know what Germanic wording means, but I can guess? ‑‑Sarri.greek ^♫ I 04:30, 24 April 2022 (UTC)

@Sarri.greek Alternative is from Latin, other is from Proto-Germanic. In English, on average latinate words tend to be more precise (=to the point) but require a more extensive vocabulary (=need you to know more words). Germanic words are easier to understand and more basic, but there's the worry that they cover broader senses and can be taken more than one way (=introduce ambiguity). I don't think it's really a concern here, but Richard does. — LlywelynII 12:14, 24 April 2022 (UTC)

I will make a comment here on how I understand the concept of alternative forms. To me, the concepts of 'alternative forms' and 'alternative spellings' are very similar. What makes a pair of words "alternative" of each other is that on some very core level, the two forms are really more "like one word" than they are "two different words", and hence they are "alternative forms of the same word". For example: Taiwan Straits and Taiwan Strait are alt forms and not synonyms because they are one letter apart and essentially exactly the same thing with almost the exact same word, and Pei-ching and Beijing are alt forms because the sound in the language of origin is supposed to be literally identical, and there's merely a difference in letters mechanically generated by the transliteration scheme involved. Now whether these words are "other" forms of each other, I don't know- I just assumed that "alternative forms" is the default academic term and I never considered other nomenclature. Guangzhou and Canton are not alternative forms of the same word- they are different words-- totally different sound and mode of derivation, really just synonyms. There is a gray zone between alternative forms and synonyms where I go more narrow and Fish bowl goes more broad-- see Talk:Ryukyu.
As to the issue that "Using Germanic wording generally makes things obscurer for European non-native speakers.", this would be totally irrelevant in any determination of whether "other forms" or "alternative forms" were to be more appropriate- the core analysis has to be English language scholarship on English and what terminology is used in those fields- has nothing to do with the continentals and interaction between their languages and English (unless that's what the academics are doing I guess). --Geographyinitiative (talk) 19:40, 24 April 2022 (UTC)

Ok, given the content and the interested userbase, I completely understand keeping the thing under lock and key. Short of applying for modhood, though, what right/rite/privilege/blood oath/sacrifice/procedure/&c. do I need to receive or undergo to get the ability to tweak articles like this for obvious failings? (Some phrasing and formatting could be tightened but there should definitely be some extra categories, weeb is derived term and not a synonym, &c.)

I've been around for over ten years and apparently have a 98+% retention rate on a couple myriad of edits, so I double pinky swear I'm not going to piss it away for the lulz. — LlywelynII 12:09, 22 April 2022 (UTC)

We could ask @Equinox if he might consider dropping the protection back to autoconfirmed. As far as I can tell, much (if not all) of the vandalism before the page was protected in 2017 was from IPs, so autoconfirmed protection ought to be sufficient. (Perhaps there is something going on that I can't see, however.) This, that and the other (talk) 13:24, 22 April 2022 (UTC)

Maybe the new autopatroller level protection would work better? — SURJECTION ^{/ T / C / L /} 13:29, 22 April 2022 (UTC)

@LlywelynII, you can now edit it. This, that and the other (talk) 14:16, 22 April 2022 (UTC)

Thank you kindly.

To the others, I really do understand why that article would need special protection overall. It's a topic of particular interest to the l337 hacker 4chan and G-d knows how ornery xe can get. — LlywelynII 17:21, 22 April 2022 (UTC)

@LlywelynII: Actually I can't remember why I protected it so "deep". Usually it's sufficient to protect from unregistered non-whitelisted users only. Anyway, we relaxed the edit rule as you requested. Equinox ◑ 11:05, 1 May 2022 (UTC)

I saw. Thanks. — LlywelynII 17:10, 1 May 2022 (UTC)

forvo.com have many pernunciation for all languages please use in wiktionary worlds thanks Amirh123 (talk)

I think those are under a different lisence, so they're not usable. Vininn126 (talk) 13:50, 22 April 2022 (UTC)

It's been two years since I've made the Tuvan pronunciation appendix, and yet individual entries' pronunciation key links (example) still link to an inexistent Wikipedia page. I understand the appendix could be wrong, but this seems to have led some users to change the IPA on some pages that I've edited without seeing the basis for it first. Making it more visible could incentivise users to change the incorrect appendix page before individual pages, which in the occasion of error are more cumbersome to correct later. @LibCae

Alves9 (talk) 20:15, 22 April 2022 (UTC)

Because, the way the IPA modules currently work, you have to manually add tyv to the variable langs_with_infopages in Module:IPA/data. The module could be changed to instead check whether the "Appendix:$LANGUAGE pronunciation" page exists, but there are several pronunciation appendices like Appendix:Slovak pronunciation that are merely soft redirects to the Wikipedia article, and in such cases we would prefer to link directly to Wikipedia instead of making people go to a page that then redirects them there. 70.172.194.25 19:03, 24 April 2022 (UTC)

I've gone ahead and added tyv to the module. Thadh (talk) 19:18, 24 April 2022 (UTC)

As of now, we have to manually categorize terms that are compound nouns and compound adjectives. For this reason, CAT:English compound nouns contains a measly 40 entries! It would be better if the categorization could be automated, for example, by using a parameter |type= that will take the values "noun" or "adj". ·~ dictátor·mundꟾ 17:06, 23 April 2022 (UTC)

Adverbs too can be compound, like after +‎ -ward. |pos= may indicate the purpose of the parameter more clearly. There are case where more than one PoS applies; e.g., anti- +‎ racist can be an adjective as well as a noun. |pos=adj,noun?  --Lambiam 18:35, 23 April 2022 (UTC)

As of now we don’t treat affixed terms as compounds. Maybe we should, or else categories like CAT:Words by prefix by language could as well be made a subcategory of CAT:Compound words by language. However, terms like afterward are actually univerbations, not compounds. Most univerbations are presently wrongly categorized as compounds, but I’ve been fixing such entries. And, we already use the parameter |posN= to signify the constituent elements’ PoS, whereas we already use |type= to classify compounds. ·~ dictátor·mundꟾ 19:04, 23 April 2022 (UTC)

@Svartava: Could you add noun and adj to Module:compound as compound type parameters? Thanks in advance! ·~ dictátor·mundꟾ 14:20, 28 April 2022 (UTC)

@Inqilābī: Hi, don't we have |pos= already for this purpose? {{com|en|foo|bar|pos=noun}} => Category:English compound nouns. |posN= is used only for the components, not the compound itself. For something like |pos=noun,adjective some module expert (Surjection, Erutuon, Benwing2, Octahedron80) could help us out since I don't know enough Lua to do that. —Svārtava (t/u) • 15:06, 28 April 2022 (UTC)

@Svartava: Oh, you are right! I didn’t know that using |pos= already does the needed categorization. Thanks. ·~ dictátor·mundꟾ 15:19, 28 April 2022 (UTC)

@Svartava: |pos= doesn’t work for {{com+}}, could you edit its module? ·~ dictátor·mundꟾ 15:25, 28 April 2022 (UTC)

 Done —Svārtava (t/u) • 15:29, 28 April 2022 (UTC)

@Equinox, Justinrleung I need some dispute resolution here. Should Wiktionary show the Wade-Giles full forms with tones in the etymologies for Wade-Giles derived words? I say yes, LlywelynII says no. There is no room for compromise between the two positions; I am okay with either method but I think showing the Wade-Giles full form is better. See discussion here: concerning Chung-nan-hai. Thanks for any guidance. I will continue adding these unless you all decide otherwise. --Geographyinitiative (talk) 19:22, 23 April 2022 (UTC) (modified)

I think it is better to repeat it the way Geographyinitiative has done for SEO; plus as far as I see the Wade-Giles romanization with tone marks is not even found in the linked Chinese entry, which makes the “q. v.” a false promise. Fay Freak (talk) 21:06, 23 April 2022 (UTC)

I think Geographyinitiative's way is relatively standard for these entries (since it's mostly Geographyinitiative who is laying the groundwork, but it has been discussed and agreed upon with discussion). LlywelynII's way is kind of going against the consensus, and I don't see it as an improvement. It makes it harder for users to know how the spelling can be derived based only on the pinyin. — justin(r)leung _{{ (t...) | c=› }} 23:19, 23 April 2022 (UTC)

@Fay Freak, Justinrleung His perogative, but a bit knobbish to not ping me in and not post the germane points and mislink the discussion. This may not change your mind but you should at least be aware of where I'm coming from. The original version for Chung-nan-hai:

Chung-nan-hai

From Mandarin 中南海 (Zhōngnánhǎi) Wade-Giles romanization: Chung¹-nan²-hai³.

The edited version:

Chung-nan-hai

The atonal Wade-Giles romanization of the Mandarin pronunciation of Chinese 中南海 (Zhōngnánhǎi), q.v.

The original discussion with GI:

I absolutely believe it is an improvement. (a) The new etymology is terser without losing any information. That's an improvement. (b) Almost no one, including Wade and Giles themselves, actually appended the tone marks to anything. The previous version was well intentioned but hypercorrect. (c) The tones for those who are curious are clearly marked on the pinyin transcription. (d) Even if it were to be included (which again it shouldn't be), you put it in the wrong place in the etym. It should follow the word romanization or transcription before you get to the Chinese. It shouldn't go back and forth from Wade-Giles to Chinese back to Wade-Giles. (e) You should be marking the English forms as "the ... romanization" or "transcription" instead of "from ..." and (f) should note that these are the atonal forms, again making the superscripts needless, even though we want to see the tones on the tr= for the Chinese template, whichever one we use.

Yes, I absolutely believe that the new version was better in every way, with apologies for the work you put into formatting the Wade-Giles so carefully. Totally understand if you want to throw it up to a group discussion at the Village Pump or wherever, but the other problems noted would still need to be addressed even if we include the needless repetetition of the Wade-Giles form of a Wade-Giles term... For what it's worth, I am completely in support of making a Wade-Giles format (with tone numbers!) show up automatically on Chinese entries' pronunciation templates. We should have that.

We should have the tonal pinyin with any use of Chinese characters in etymologies. We don't need to duplicate that less well to discuss Wade-Giles that is toneless in application anyway.

@Fay Freak The q.v. is to look for further information on the meaning and origin of Zhongnanhai, not the tones. The tones are already marked on the pinyin. To the extent that we want toned Wade-Giles, it should just be made part of the default Chinese pronunciation templates. God knows it's more important to English users than the Zhuyin is.

@Justinrleung In what way does it make it harder to understand how the spelling is derived? The entry literally is exactly the spelling, hence why it's needless to repeat it. Or did you mean something else? — LlywelynII 00:43, 24 April 2022 (UTC)

Surely we shan’t append “q.v.” formulaically or habitually, otherwise it loses its suggestive signification to tell the reader to see an entry. When we say “which see” then we really mean it, meaning that there is peculiarly more information to be derived from the linked entry than usually expected.

Also it is kind of contradictory to claim that almost no one appended the tone marks in that romanization but then noting that the transcription is atonal, which according to your claim is the usual case. “romanization of … Mandarin pronunciation” is also wordy. I don’t think you made it terser. Geographyinitiative’s version was already as terse as possible plus a repetition of the Wade–Giles romanization with tones which can be excused as SEO and clarification that yes this is the Wade–Giles romanization—otherwise of course, repeated without tone marks, it would seem a stupid repetition, so the hypercorrectness is functional. Fay Freak (talk) 01:02, 24 April 2022 (UTC)

The q.v. is meant. It just didn't refer to what you thought it did. If people clicked through, they would see more information on the word itself and its origin. That said, I'm fine with removing it since the link itself is essentially a q.v. and it seems to have confused at least two people here.

It isn't. GI wants to include it because it should by theory be included; it's not "contradictory" to observe that it was almost never actually employed in practice and it's hypercorrect to bother with it at the entry here. The information about the tones (necessary) is already provided by the pinyin (necessary) and adding it to the Wade-Giles is repetitive, artificial, and unnecessary. — LlywelynII 12:28, 24 April 2022 (UTC)

Why does "atonal romanization" take front and center? Do we do anything similar with other languages, besides Korean (not even Japanese)? English doesn't have tone, what do you expect?

Why is writing something as wordy as "the Mandarin pronunciation of Chinese" necessary? Is Mandarin a mere pronunciation? Setting aside that we have a "Chinese" header here on Wiktionary, which we have established is out of convenience, rather than because we reject the identity of 方言 as languages, cf. the continued existence of "Mandarin" vs "Cantonese" in Translations sections.

q.v. what?

The new text is simultaneously longer and inferior. —Fish bowl (talk) 03:50, 24 April 2022 (UTC)

If that was a serious question, because it is precisely the pinyin or WG romanization without the tones, which are otherwise completely required. It's wrong to say that it is the pinyin romanization; it's misleading to say it's from the pinyin romanization; it is the atonal pinyin romanization. Neither Korean nor Japanese have tones, so I'm not sure what your point was there.

Similarly, yes, it's the pronunciation of those Chinese words in the Mandarin dialect as opposed to other dialects. It also makes sure the word is covered in both our Mandarin and Chinese categories. Yes, there's a long philosophical/philological/political debate to be had about whether Chinese dialects count as separate languages. Y'all seem to be of two minds on the subject with, e.g., our entry on Standard Mandarin calling it (the artifical construction) a "language" but the actually spoken Beijingese it's based on a "dialect". In any case, 北京 isn't Mandarin. It's Chinese, and Beijing is its pronunciation in the Mandarin whichever. Yes, we should distinguish that; it's precisely why people get confused that Peking and Beijing are forms of the exact same word rather than completely separate words.

Highly disagree, and your questions seem to belie a complete unfamiliarity with the subject that really should remind you to let other people figure this out. GI and Justin eg might also completely disagree, but they would have good reasons for doing so. — LlywelynII 12:24, 24 April 2022 (UTC)

Whatever one might say about @Fish bowl, assuming "a complete unfamiliarity with the subject" itself shows "a complete unfamiliarity". They do a prodigous amount of front-line patrolling of new edits in all three of the CJK languages, and, judging by the comments of our subject-matter experts, with pretty good accuracy. Also, while Japanese doesn't have tones, it does have pitch accents, though I'm not sure if that's what they're referring to here. Chuck Entz (talk) 14:37, 24 April 2022 (UTC)

I prefer the way it was before because it provides more information. You can still see the tones in the Pinyin, but to translate between the numbers and the diacritics you need to be familiar with both transliteration schemes. Better to just include the full Wade-Giles form. This should probably also be included in the {{zh-pron}} box, but I don't see why not to include it in the etymology when it is relevant. I'm not sure why it's not already in the zh-pron box, since e.g. {{#invoke:cmn-pron|py_wg|Zhōngnánhǎi}} seems to produce the correct output. The only thing that doesn't sit completely right with me about the original format given is the lack of punctuation or separating words between "Mandarin 中南海 (Zhōngnánhǎi)" and "Wade-Giles romanization: Chung¹-nan²-hai³". It feels like there should be a comma there or something. But since this is the way it's consistently done on a lot of pages, I don't mind much. 70.172.194.25 18:36, 24 April 2022 (UTC)

Currently 4 vs 0. —Fish bowl (talk) 23:51, 30 April 2022 (UTC)

Cute. See my talk page, where the OP has subsequently agreed with my edits.

TLDR. User:Geographyinitiative, please confirm. —Fish bowl (talk) 00:01, 1 May 2022 (UTC)

Beyond which, Chuck just mentioned you have some (undemonstrated) familiarity with the subject, not any form of agreement here. That would make it 2-2 or 2-1, if this mattered beyond OP having already realized he was mistaken. — LlywelynII 23:55, 30 April 2022 (UTC)

"your questions seem to belie a complete unfamiliarity" is from you talking about me, not Chuck talking about me lmao —Fish bowl (talk) 00:01, 1 May 2022 (UTC)

I'm not sure how poor your critical reading skills are. You seem to be unable to skim but that might just be disinterest (so why bother continuing here?), but you should be able to follow that yes I was precisely saying Chuck was asserting that you have some (undemonstrated) familiarity with the subject.

In any case, if you're still hanging around here and have such great interest in Wade-Giles transcriptions, GI and I could absolutely use your support for finally including Wade-Giles in the {{zh-pron}} template. No idea why it was passed over before but it's incredibly simple to code over from pinyin and the tone numbers could be included there. — LlywelynII 00:26, 1 May 2022 (UTC)

@Chuck Entz, Equinox, Justinrleung Hey guys, I am requesting a clarification on these issues; I don't want two good editors to get hurt in this issue; please provide guidance so we can all move on here, if at all possible. Thanks for any help. I am making an extraordinary fifth comment on this general topic (which is against my personal policy) because of the problems. Help the editors if at all possible. Example: Please don't use discipline if possible. --Geographyinitiative (talk) 00:07, 1 May 2022 (UTC)

Eh? What are the problems? If you really did support restoring all the tone marks, that's fine. Whatshisface would be right that I'm completely outvoted. I assumed you'd changed your mind, since you'd said as much on my talk page.

@Geographyinitiative What did I misunderstand? or what still needs to be resolved? — LlywelynII 00:27, 1 May 2022 (UTC)

Sorry, I'm not qualified here because it's based on languages I don't speak at all. Equinox ◑ 11:03, 1 May 2022 (UTC)

You can find this message translated into additional languages on Meta-wiki.

More languages • Please help translate to your language

The Board of Trustees seeks candidates for the 2022 Board of Trustees election. Read more on Meta-wiki.

The 2022 Board of Trustees election is here! Please consider submitting your candidacy to serve on the Board of Trustees.

The Wikimedia Foundation Board of Trustees oversees the Wikimedia Foundation's operations. Community-and-affiliate selected trustees and Board-appointed trustees make up the Board of Trustees. Each trustee serves a three year term. The Wikimedia community has the opportunity to vote for community-and-affiliate selected trustees.

The Wikimedia community will vote to fill two seats on the Board in 2022. This is an opportunity to improve the representation, diversity, and expertise of the Board as a team.

Who are potential candidates? Are you a potential candidate? Find out more on the Apply to be a Candidate page.

Thank you for your support,

Movement Strategy and Governance on behalf of the Elections Committee and the Board of Trustees

--Mervat (WMF) (talk) 21:24, 25 April 2022 (UTC)

Hello!

Have you noticed that some wikis have a different desktop interface? Are you curious about the next steps? Maybe you have questions or ideas regarding the design or technical matters?

Join an online meeting with the team working on the Desktop Improvements! It will take place on 29 April 2022 at 13:00 UTC and 18:00 UTC on Zoom. Click here to join. Meeting ID: 88045453898. Dial by your location.

Agenda

• Update on the recent developments
• Questions and answers, discussion

Format

The meeting will not be recorded or streamed. Notes will be taken in a Google Docs file. Olga Vasileva (the Product Manager) will be hosting this meeting. The presentation part will be given in English.

We can answer questions asked in English, French, Italian, and Polish. If you would like to ask questions in advance, add them on the talk page or send them to [email protected].

At this meeting, both Friendly space policy and the Code of Conduct for Wikimedia technical spaces apply. Zoom is not subject to the WMF Privacy Policy.

We hope to see you! SGrabarczuk (WMF) (talk) 00:35, 26 April 2022 (UTC)

Personal opinion: I disagree that these are "improvements". Whenever I end up on fr.wikipedia.org I groan, mainly because I find the placement and interface of the language-switching tool far less convenient than the old-fashioned sidebar. I'm also not sure what is even the point of making the sidebar text collapsible if the panel still takes up the same amount of screen real estate, but that doesn't bother me too much. It just seems like a more complicated UI for no reason. But oh well, I suppose the powers that be have made up their minds and I should make my peace with it. 70.172.194.25 02:15, 26 April 2022 (UTC)

Hello, thanks for your opinion. First, I'll address the concern about "are these improvements, really?" and "the powers that be have made up their minds".

• Different people have different needs, and they consider different things as improvements. We are focusing on the improvements related to viewing (reading).
• Our interface is configurable, as any other MediaWiki interface. Editors can modify it, create gadgets, etc.
• Some editors may have so specific needs that even with gadgets and adjustments, they will be not satisfied. This is why we give an easy option to stay with the legacy Vector.
• But in general, by measuring (via API) and talking to people, we check if our changes are for the better. At the beginning of the work on the Desktop Improvements, we decided that if a given change is not positive, we will resign from it. See what we did when we were not satisfied with the first version of the language button.
• If you are curious how we work on the changes (and how little room there is for the powers to make arbitrary and bubble-determined decisions), see the newest Signpost.

Regarding the interface of the language-switching tool, I understand your point. The tool will be improved further. This is part of another team's task. While we are working on the interface overall, they are responsible for languages-related stuff. So what you see is only the first attempt.

Could you write more about what you mean by "if the panel still takes up the same amount of screen real estate"?

I invite you to our meeting, too! SGrabarczuk (WMF) (talk) 14:28, 26 April 2022 (UTC)

Re "screen estate", it just hides the sidebar, and doesn't use the space for anything else, it does feel a bit gimmicky. Perhaps the intention was a "distraction-free reading mode" ? btw, the signpost link is broken.– Jberkel 15:40, 26 April 2022 (UTC)

Yes @Jberkel, the intention was a "distraction-free reading mode". But that's just part of the answer. Our hypothesis was that making it collapsed by default for logged-out users would result in other links (such as Talk, History, and Edit) becoming more noticeable. We keep the sidebar uncollapsed by default for logged-in users, but it's also persistent (meaning, if you hide it and open another page, it will stay collapsed). Now, when the sidebar is collapsed, the TOC is displayed there. Moving page-related links from the sidebar to a dedicated menu on the right side of the screen (as visible in the prototype) would make sidebar more "integral". BTW, everyone who hasn't shared their opinions on the prototype - I invite you to do that! There's a pre-filled form already. SGrabarczuk (WMF) (talk) 16:30, 26 April 2022 (UTC)

Ah, I just tried it on fr.wikipedia.org and the TOC isn't displayed there, it's just blank. Maybe it's not the most recent version. But probably a good call to deprioritize the sidebar, it's crowded with many rarely used links. – Jberkel 17:01, 26 April 2022 (UTC)

I also prefer the narrower reading pane - it can be a bit laborious to read stuff on a wide screen. Wikisource has had the option of doing something similar for ages, for that exact reason. Theknightwho (talk) 19:18, 26 April 2022 (UTC)

Unless I'm missing something, the sidebar option doesn't make the article body any narrower or wider than it is already, though? It keeps the same dimensions but toggles the visibility of the links in the sidebar. If the goal is to shift attention away from sidebar links that are not needed by most readers, fair enough, but I never personally found that problematic. What is problematic to me is the new language-switcher, since it now takes multiple clicks to change between different language versions, which slows down a common use case for me. 70.172.194.25 22:53, 26 April 2022 (UTC)

Can't one just narrow the browser window no achieve all but the most extreme degrees of narrowness? DCDuring (talk) 23:17, 26 April 2022 (UTC)

• I resent calling this arbitrary change "Desktop Improvements". What's the evidence that it is an improvement for any population of users? At best it might be "Proposed Desktop Alternative". DCDuring (talk) 17:22, 26 April 2022 (UTC)

  @DCDuring, we do collect data. Each time we introduce a change, we perform A/B tests. One of the tasks of our Product Manager is to surface the data and write explanation to raw statistics, because these may be a bit difficult to find now. Also, please see the newest Signpost to learn more on, as I've wrote above, "how little room there is for the powers to make arbitrary and bubble-determined decisions". SGrabarczuk (WMF) (talk) 16:01, 27 April 2022 (UTC)

  The testing is only on logged-in users, who have the knowledge to opt out and are probably not a good proxy for users who are not logged in. Do you have any way of A/B testing non-logged-in users or do our policies prevent us from doing such tests? DCDuring (talk) 20:30, 27 April 2022 (UTC)

  @DCDuring, for logged-out, we use "before & after" (quantitative) and interviews (qualitative). We have no tool to make A/B tests with them. See our page about the new TOC as an example. SGrabarczuk (WMF) (talk) 22:45, 27 April 2022 (UTC)

  Is any of this tested with content other than Wikipedia. In particular, is it tested with Wiktionary-type content? DCDuring (talk) 22:48, 27 April 2022 (UTC)

  Yes, here's the list of our early adopter (aka pilot or test) wikis. We test our changes on French Wiktionary. (I recommend taking a look at our project pages, you'll have more answers to possible questions. If you have any suggestions for me on how to improve the pages, please share your recommendations :D) SGrabarczuk (WMF) (talk) 00:08, 28 April 2022 (UTC)

@SGrabarczuk (WMF) Looks cool, thanks for sharing. - TheDaveRoss 18:40, 26 April 2022 (UTC)

https://www.justice.gov/usao-ma/pr/california-man-arrested-and-charged-making-threats-against-lgbtq-community —Justin (koavf)❤T☮C☺M☯ 20:19, 26 April 2022 (UTC)

I think the categories Category:Compound adjectives by language and Category:Parasynthetic adjectives by language should be merged. ·~ dictátor·mundꟾ 14:42, 28 April 2022 (UTC)

@PUC, Lambiam, Fay Freak, any thoughts? ·~ dictátor·mundꟾ 23:42, 1 May 2022 (UTC)

Support merging the latter with the more standard "compound adjectives". —Svārtava (t/u) • 03:09, 2 May 2022 (UTC)

@Inqilābī: Should this be moved to Wiktionary:Requests for moves, mergers and splits? J3133 (talk) 06:23, 2 May 2022 (UTC)

@J3133: Technically yes, but I intentionally posted it here for more visibility. In the past, proposals for merging labels such as colloquial/informal at WT:RFM led to great contentions. ·~ dictátor·mundꟾ 18:09, 2 May 2022 (UTC)

The category description (“These are also known as compound adjectives”) is wrong. The adjectives that are parasynthetic compounds are a proper subset of the compound adjectives. In English, a simple compound adjective is pitch-black, black as pitch. It is not a parasynthetic adjective. An example of the latter is palefaced = (pale +‎ face) + -ed, that is, having a pale face. Parenthesizing makes the more complicated tree structure explicit. I have no opinion on whether having this more refined categorization is useful.  --Lambiam 06:54, 2 May 2022 (UTC)

A user is requesting to "global de-admin" (sorry I really don't know how likely if this could be happened) @Jusjih which in requestor's claim, has many disruptive and abusive behaviors in several Chinese-language projects and Meta-Wiki, the requestor is also mentioned en.wiktionary here where Jusjih is also an administrator, but didn't say anything else on their en.wiktionary's behaviors. Do any of our adminships that may or may not familiar Jusjih's works know how to resolve it? --Liuxinyu970226 (talk) 05:27, 29 April 2022 (UTC)

-- Johanna Strodt (WMDE) 11:13, 29 April 2022 (UTC)