Wiktionary:Beer parlour

Wiktionary > Discussion rooms > Beer parlour

Click here to start a new Beer parlour discussion.

Wiktionary discussion rooms (edit) see also: requests

Information desk start a new discussion | this month | archives Newcomers’ questions, minor problems, specific requests for information or assistance.

Tea room start a new discussion | this month | archives Questions and discussions about specific words.

Etymology scriptorium start a new discussion | this month | archives Questions and discussions about etymology—the historical development of words.

Beer parlour start a new discussion | this month | archives General policy discussions and proposals, requests for permissions and major announcements.

Grease pit start a new discussion | this month | archives Technical questions, requests and discussions.

All Wiktionary: namespace discussions 1 2 3 4 5 – All discussion pages 1 2 3 4 5

Welcome to the Beer Parlour! This is the place where many a historic decision has been made, and where important discussions are being held daily. If you have a question about fundamental aspects of Wiktionary—that is, about policies, proposals and other community-wide features—please place it at the bottom of the list below (click on Start a new discussion), and it will be considered. Please keep in mind the rules of discussion: remain civil, don’t make personal attacks, don’t change other people’s posts, and sign your comments with four tildes (~~~~), which produces your name with timestamp. Also keep in mind the purpose of this page and consider before posting here whether one of our other discussion rooms may be a more appropriate venue for your questions or concerns.

Sometimes discussions started here are moved to other pages for further development. In particular, changes to a major policy or guideline may be discussed on the corresponding talk page and “simple votes” (as opposed to drawn-out discussions) can be conducted on our votes page.

Questions and answers typically remain visible on this page for one to two months, but they can always be found in the appropriate monthly archive (based on the date discussion was initiated). While we make a point to preserve all discussions that were started here, talk that is clearly not appropriate for this page may be deleted. Enjoy the Beer parlour!

Beer parlour archives edit
2025 January February March April May June July August September October November December 2024 January February March April May June July August September October November December
Earlier years 2023 January February March April May June July August September October November December 2022 January February March April May June July August September October November December 2021 January February March April May June July August September October November December 2020 January February March April May June July August September October November December 2019 January February March April May June July August September October November December 2018 January February March April May June July August September October November December 2017 January February March April May June July August September October November December 2016 January February March April May June July August September October November December 2015 January February March April May June July August September October November December 2014 January February March April May June July August September October November December 2013 January February March April May June July August September October November December 2012 January February March April May June July August September October November December 2011 January February March April May June July August September October November December 2010 January February March April May June July August September October November December 2009 January February March April May June July August September October November December 2008 January February March April May June July August September October November December 2007 January February March April May June July August September October November December 2006 January February March April May June July August September October November December 2005 January–March April–June July–September October–December 2004 January–March April–June July–September October–December 2003 January–March April–June July–September October–December 2002 December

September 2025

Recent flood of protologisms

As anyone can tell from RFVE, we've had a lot of people lately addding terms they just made up. I don't think this is random chance. My hunch is that these people are using us to promote their creations by feeding them into AI. After all, AI is being aggressively promoted everywhere, and its combination of seeming omniscience on the front end and gullibility on the back end makes it a great tool for deception.

It seems to me that we need to address this somehow. Basically, we run the risk of violating NPOV by exaggerating the importance of these new coinages, and of distributing promotional material. Yes, everything will eventually be deleted via RFV, or at least changed to reflect actual usage- but who knows what damage will be done in the meanwhile?

However we respond, we'll need to avoid discouraging addition of legitimate new content, and to come up with workable tests for when to apply different rules. Off the top of my head, perhaps we need to look for references to a person or website that would benefit from the exposure.

Thoughts? Chuck Entz (talk) 20:45, 1 September 2025 (UTC)

@Chuck Entz - maybe you can speedy or just blank anything that you think is blatant slop, as a measure? If a legitimate word gets hit by this measure, then the author might just have to re-create it with citations, which is no problem if the word really exists. But there's no harm in wiping some words away, as if they really do exist they'll be back in no time, I think. Kiril kovachev (talk・contribs) 21:11, 1 September 2025 (UTC)

pinging editors who may be interested: @J3133 @Polomo @-sche @Surjection. Juwan (talk) 21:13, 1 September 2025 (UTC)

IIRC we used to have "Protologism: please see WT:CFI, use WT:LOP" as a deletion reason, back when we had WT:LOP. (We still have "Promotional material" as a deletion reason.) If I searched and found no results for something in Google Books / Scholar / Archive.org, and it looked rare and promotional/recent on e.g. Google, and I were seeing a large number of such entries, I would probably start deleting them with a manual "see CFI" summary (and it seems like others are already doing similar), but whether it's possible to formalize that approach into a set of non-gameable guidelines for speedying things or whether it requires admin discretion, I don't know. One idea for giving things community attention at RFV without having the entry getting "public" attention: start an RFV discussion for the entry, then immediately speedy delete it; even after an entry has been RFV-passed/failed/speedied/etc the discussion is supposed to stay on the central page for a week, which gives people time to notice and say "oh, actually I'm familiar with this one, here are some cites". - -sche (discuss) 21:22, 1 September 2025 (UTC)

I think there’s a couple cases to look at here:

1. Entry is clearly AI-generated — I don’t think that the “protologism” part even figures into this at all. The problem is that the entries have been created using AI, and, attested or not, we should delete AI-generated content, just like the folks at Wikipedia do. I believe they’ve been allowing admins to delete LLM content on sight lately.
2. Entry for a protologism that’s a stub but not AI slop — this is more my subjective opinion, but I don’t see any problem with straight up deleting either. We have that default deletion reason and even a template for it, {{uw-protologism}}. If a user really wants their entry in their dictionary, they can question the deletion in the talk page; if they’re not convinced then, they can go to RfD to request undeletion and be shot down there.
3. Entry for a protologism that’s reasonably well-formatted — I don’t see any issue with keeping the entry with an RfV tag for a month, though it probably wouldn’t be bad to delete it and ask the creator for cites either. The text displayed by the RfV tag really makes it clear that the entry is dubious, so we’re giving negative promotion if anything. Is that bad?

I know Surjection deletes protologisms pretty often, and I have too. Most recently, I deleted Hostile indifference for being AI-generated; before that, I deleted Cuckoldsmobile, Plushnik, and Promptosis due to a lack of results even on Twitter.

One thing to keep in mind is that RfVed entries often get forgotten and end up being kept for way more than a month. If we want to keep entries no longer than necessary, we could try and add a parameter to the {{rfv}} template, specifically for likely protologisms, that automatically tags the entry for speedy deletion after 1 month has passed since the tag was placed. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 22:31, 1 September 2025 (UTC)

Comment for the bullet 1, see WP:AIREMOVAL and WP:G15 on Wikipedia. Juwan (talk) 23:02, 1 September 2025 (UTC)

Just to be clear, I'm talking about people who add entries because they want AI to see it and incorporate it into AI's answers. For my comments, it doesn't matter whether they wrote it with or without the help of AI- it's basically the same sort of thing as spammers who add links to sites they want the search engines to see in order to increase page ranking. Using AI to create content isn't new: people have been using Google Translate for years. IMO we should treat AI content the same as anything from dubious sources, including mass-added translations from all the Wikipedias. Chuck Entz (talk) 23:28, 1 September 2025 (UTC)

Hm, I guess I must’ve misread. What makes you think that’s the intention? People have wanted to promote themselves using Wikimedia projects for ages, and that’s usually for other people, I believe? Anyway, the only connection I would readily draw is that self-promotion is much, much easier and tempting if you don’t have to do any work at all (i.e., if you use an LLM chatbot), and that’s why there’s an influx. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 00:36, 2 September 2025 (UTC)

I'm not saying the ease of production hasn't been a factor, and I'm not saying that people promoting their protologisms is entirely new- but it seems like there's been a change in the type of many of the protologisms recently: far more are primarily focused on promoting their names and/or websites as originators. Chuck Entz (talk) 01:52, 2 September 2025 (UTC)

@Chuck Entz: just curious—what makes you think the entries are AI-generated? — Sgconlaw (talk) 04:50, 2 September 2025 (UTC)

What makes you think I think that? Chuck Entz (talk) 05:00, 2 September 2025 (UTC)

@Chuck Entz: you said you had a hunch? I looked at one entry and to me it looks indistinguishable from other entries, so I was wondering what special insight you had… — Sgconlaw (talk) 13:51, 2 September 2025 (UTC)

@Sgconlaw: I said I had a hunch that people were taking advantage of AI's practice of picking up information from us by adding protologism entries in hopes that the AI would integrate them into its answers. It has nothing to do with with how they're created (OCRed off the back of a napkin?), but everything to do with why they're created.Chuck Entz (talk) 04:30, 3 September 2025 (UTC)

@Chuck Entz: oh, I see … — Sgconlaw (talk) 05:20, 3 September 2025 (UTC)

We should have more of a "shoot first and ask questions later" approach to dubious entries in general. In the case of doubtful verifiability, the deletion message can tell editors to populate the Citations page then go to the RFV forum to ask for undeletion. Vox Sciurorum (talk) 12:18, 2 September 2025 (UTC)

Agreed. I furthermore trust Chuck's (and generally other admins' judgement) that a clearly nonsense term is just self-promotion for some in-joke or other waste of time crap. No need to send everything to RFV, which is a bloated backlog already. —Justin (koavf)❤T☮C☺M☯ 13:54, 2 September 2025 (UTC)

I really like the idea of prompt deletion and placement on a list of these, somewhere other than RfD or RfV, so that they can be reviewed conveniently and, if warranted, be undeleted. The problem Chuck has suggested applies to new entries or, at least, L2 sections, but suspect definitions of relatively common words are also a problem. IMO, the draconian speedy deletion approach is not likely to work as well for suspect definitions as for suspect entries, because search for different senses of words is more complicated.

I don't think we have a special page or log that already makes it easy to find deleted pages of this kind. Is there a way to automagically identify such pages or to tag them in such a way to make them automagically identifiable?

Ultimately we will need AI-type tools to help with these matters. DCDuring (talk) 14:20, 2 September 2025 (UTC)

Perhaps we could consider not allowing unregistered users to create entries. I’ve often been tempted to propose this for the Reconstruction namespace, at least. Nicodene (talk) 14:48, 5 September 2025 (UTC)

This would require gathering some statistics on what proportion of IP-created entries are junk. 2A00:23C5:FE1C:3701:981B:1EF5:4A85:16FE 15:52, 5 September 2025 (UTC)

Another example just now: Zerune. I have asked the creator (on talk page) how they created the entry, and whether AI was used. Our best bet is probably to ask people when we see these, until we find out the truth. 2A00:23C5:FE1C:3701:B5BB:F9F5:7CF6:B0BB 08:27, 17 September 2025 (UTC)

And this is a re-creation of essentially the same entry from a couple of days ago. --Hiztegilari (talk) 08:32, 17 September 2025 (UTC)

Reviewing blocks by another administrator

Hi, Wiktionary says "you may appeal the block by requesting that another administrator review the block.". However, this rules isn't followed. For example take a look at the user Aliwonde who requested an unblock, but the only people who "reviewed" it were the ones who blocked them. And, also the administrator "Surjection" blocked them for editing their own talk page for no reason. Can other editors take a look at it? Cokedtacos (talk) 09:53, 2 September 2025 (UTC)

You have evaded your block numerous times to continue trolling and vandalism. Stop wasting our time. — SURJECTION ^{/ T / C / L /} 10:29, 2 September 2025 (UTC)

No, I evaded my block becaused it was an unfair one. I've never trolled or vandalized here apart from reverting (or trying to revert) Vininn126's edits who started an edit warring and then blocked me.

Maybe if you didn't block me every time I tried to participate in the discussion and didn't delete my replies, we would both be in a different place now. Actually, I'm quite shocked to see that you didn't delete this tread as well. I apreciate it, but unfortunatelly, you removed some other replies of mine, which only made things longer Onceopts (talk) 12:44, 11 September 2025 (UTC)

The fact that you still refuse to understand that block evasion is still not allowed further reinforces the view that -sche is entirely correct in referencing w:WP:IDHT. — SURJECTION ^{/ T / C / L /} 12:55, 11 September 2025 (UTC)

I would note that Chealer was recently globally banned for exactly the same behavior: telling veteran admins and contributors How Things Are Done Here, and not listening to any counterarguments. Chuck Entz (talk) 14:46, 11 September 2025 (UTC)

OK, as another administrator, I have looked into your many blocks. Because you resorted to extensively socking after your initial block and subsequent blocks, and that was all very recently, I do not think it would be appropriate to unblock you in the near term. Also, while I will defer to our Polish-editing editors regarding the nature of your Polish edits rather than try to evaluate those, your edits in relation to long live and the fact that you socked once blocked there too give the impression of someone who's convinced they're right and isn't willing to hear otherwise (what Wikipedia calls w:WP:IDHT), which doesn't make me think you're here to edit constructively. People can change, and a block appeal in (say) a year could be entertained, but right now, while you're still socking to evade your blocks, unblocking is a non-starter. - -sche (discuss) 21:30, 4 September 2025 (UTC)

Thanks, for your reply.

I evaded my initial block, because it was an unfair one. The details are explained in User talk:Aliwonde.

I strongly disagree that I'm not willing to hear others. Note that when the user J3133 asked me to see the Tea room discussion, I did so. I tried to take part in the discussion, but some of my replies were removed. I respect that the community consensus is more important than my perspective but I can't agree on being deprived of my right to participate in the discussion.

I guess that this might not be particularly strong argument here, but at least one of my IP addresses has a huge contribution to the Polish wiktionary and some good edits on the English Wikipedia. If you took a closer look at them, you would see that I've always edited constructively, been willing to hear other editors and haven't stuck to my viewpoint against the consensus or rules Onceopts (talk) 12:43, 11 September 2025 (UTC)

I started Wiktionary:Artificial intelligence

We should eventually have this page relatively good (written by AI?) Vealhurl (talk) 06:58, 3 September 2025 (UTC)

Delete Wonderfool trolling. 2A00:23C5:FE1C:3701:497E:4A03:99D9:D159 07:31, 3 September 2025 (UTC)

OK, maybe one day we'll have such a page! Vealhurl (talk) 12:13, 3 September 2025 (UTC)

Pie charts on various Turkic language categories

Some Turkic language categories, such as Category:Turkish language and Category:Azerbaijani language, contain etymological pie charts added by User:BurakD53. In my opinion, these are not that useful and a failed experiment at best. For one, they're based on fundamentally wrong methodology:

• "inherited" terms only include the inherited category, which means any derived native terms aren't included
• they ignore that an entry may be present in multiple "derived from" categories, e.g. Ottoman Turkish and Arabic at the same time,
• they assume "unstated" is the number of lemmas minus any entries in the aforementioned categories, which excludes all other languages and, again, gives incorrect results, since entries may be present in multiple categories and will be counted multiple times if so.

I suggest we simply get rid of them. — SURJECTION ^{/ T / C / L /} 20:47, 4 September 2025 (UTC)

I think these are interesting, but I agree that there are some severe methodological difficulties with calculating and labeling the inherited/unstated categories. Would it perhaps make sense though to include such a a pie chart at Category:Azerbaijani borrowed terms instead, limited to showing the relative proportions of source languages for borrowed terms?--Urszag (talk) 20:56, 4 September 2025 (UTC)

They're interesting—at least for my part I wouldn't mind if the user moved them to some page in their userspace (or maybe to the categories' talk pages?)—but I agree the methodological issues (and the fact that "unstated" so dwarfs all the other categories, and the way they seem to tease or invite the kind of puristic effort to bring the percentage of borrowings down that recurringly plagues our Turkic entries) makes them inappropriate to have on the category pages themselves. - -sche (discuss) 21:01, 4 September 2025 (UTC)

In addition, subcategories within category pages are also counted as entries. For example, you might say "Turkish derived terms," but this category categorizes itself by language. The template that indicates the number of entries in a category doesn't distinguish between entries and subcategories.

It's possible to remove this from the page, but I don't think it's right to remove it entirely instead of improving it and making it useful data. Of course, the pie chart uses the categorization system, because it's impossible to generate data by counting all those words in Turkish one by one. If it's [1], I can do it without it, I also did, since there are only 90 entries, but that's not possible for Turkish. BurakD53 (talk) 00:43, 5 September 2025 (UTC)

I should also point out that the derived category on the Turkish page is based on the categorization of suffix pages, so I didn't take the data from the derived or borrowed terms from X language category. There are still missing additions, though; I was just too lazy to fill them all in. BurakD53 (talk) 00:53, 5 September 2025 (UTC)

Furthermore, when you consider the actual availability of someone, turning this pie chart into a reliable table isn't that difficult:

1. You remove the same number of subcategories from the template using the subtraction template.
2. You group all suffixes under the derived category. For loanwords, use borrowed instead of derived to avoid confusion with other words. For inherited words, use the inherited category, which will also avoid confusion because the derivations will be in the suffix pages.

This solves the problem mentioned in both the first and second points. Even the third point, because if an item isn't counted twice, the problem is solved. So, improving the pie chart and ensuring it produces accurate results isn't that impossible, nor is it utopian. BurakD53 (talk) 01:07, 5 September 2025 (UTC)

The only problem here will be if either the suffixed term is inherited at the same time. In this case, we won't use both the inherited and derived/suffixed templates simultaneously on the pages. If derived/suffixed, it will be derived/suffixed, and if inherited, it will be derived/suffixed in the entry where it is inherited. BurakD53 (talk) 01:21, 5 September 2025 (UTC)

And my final speech, it is appropriate to move a userpage until it is developed. BurakD53 (talk) 01:31, 5 September 2025 (UTC)

I have moved them to User:BurakD53/Turkic pie charts. — SURJECTION ^{/ T / C / L /} 06:57, 5 September 2025 (UTC)

Thanks, eline sağlık.👍 BurakD53 (talk) 08:36, 5 September 2025 (UTC)

If you try to subtract all of the categories from each other, you'll quickly come across the expensive call limit. There is a limit on just how many times you can do {{PAGESINCATEGORY}} on a single page.

The whole thing is an interesting idea, but just not technically feasible like this. — SURJECTION ^{/ T / C / L /} 06:10, 5 September 2025 (UTC)

I see, and unfortunately it's not that basic. If you find it appropriate, it can stay for the Bulgar language, since I didn’t provide it with the template; here User:BurakD53/VolgaBulgarpiechart it is counted by origin. But still, it can remain here as well. – BurakD53 (talk) 08:39, 5 September 2025 (UTC)

• I like pie charts, we should use them more Jin and Tonik (talk) 15:55, 5 September 2025 (UTC)

Issues with WT:CLTR

there is a growing backlog of requests at WT:CLTR with little engagement from editors. admins don't close these threads even those that are not particularly controversial. previous discussions are not able to be archived, collapsed or otherwise hidden, so we have years of topics piled up, some closed, some not. Juwan (talk) 23:13, 4 September 2025 (UTC)

Unfortunately, this is also the case with WT:RFD and WT:RFV with open discussions that last for years. I am a little discouraged about closing some of these from my history of trying to close them before. I have also brought up this issue several times and I don't think there's an appetite for mass-closing several of these long-term discussions. —Justin (koavf)❤T☮C☺M☯ 23:39, 4 September 2025 (UTC)

@-sche thank you for archiving some of the discussions, may I ask however, where did they go...? the page Wiktionary:Category and label treatment requests/Archives supposedly for archives is empty. Juwan (talk) 12:10, 5 September 2025 (UTC)

Interesting, I didn't realize there was such a page so Fenakhay and I just archived them with aWa to relevant talk pages as is done for RFM etc, e.g. the discussion of Category:Urine is now on Category talk:Urine, as noted in the edit summaries. I suppose I can see benefits and drawbacks to either approach; with aWa relevant discussions are in places people will look for them (someone looking at Category:Urine to propose renaming or removing it will see the discussion that led to creating it on its talk page) but it's harder for someone to find all discussions topic-agnostically in one place as they could if things were moved to a central archive... but editing Template:archive-top to categorize them like it categorizes RFVs etc would solve that, I think. - -sche (discuss) 16:05, 5 September 2025 (UTC)

I think I'd prefer having things archived in one place, esp. as many of the CLTR discussions cover several categories or even whole classes of categories. When archiving such a discussion to talk pages it ends up semi-arbitrarily on the talk page of one of the categories, which makes it hard to find; in the past, for example, I've gone looking for archived discussions that I know exist and haven't been able to find them, and Wikimedia's internal search really sucks so it's often of no help. Maybe the solution is to archive them to a central place but put a message on all relevant talk pages pointing to the appropriate archive page. (This assumes the archive pages themselves don't get restructured; IMO we should put enough thought into the archive structure that it's unlikely we'll need to restructure later.) Maybe in fact this what you're referring to by saying editing Template:archive-top to categorize them like it categorizes RFVs etc would solve that? Benwing2 (talk) 22:35, 8 September 2025 (UTC)

Re Template:archive-top, what I mean is that for e.g.RFVs or RFDs, the template automatically categorizes them into categories like Category:RFV result, and we could presumably make it also categorize all CLTR discussions into a category. I don't know if this would make it adequately easy to search for old discussions or not. Meh. I don't object to archiving things to a central page instead, in which case I suppose we need to either undo the change that let aWa archive CLTR discussions (and rely on people to archive things manually), or (preferably?) make it so aWa archives CLTR discussions to the chosen central place. - -sche (discuss) 23:02, 8 September 2025 (UTC)

Hmm. I didn't know about Category:RFV result, but it has 21,000+ entries in it, and just lists the talk pages where the discussion were archived, not the actual title of the discussion, so I don't think that would make it easy enough to find old discussions. Probably we should fix aWa to auto-archive CLTR to central place; I don't know how aWa works but I assume that wouldn't be so hard. Benwing2 (talk) 23:38, 8 September 2025 (UTC)

for RFV results, decentralised archiving is mostly fine as these are typically for one word at a time. for threads regarding multiple entries, it would be good to have some note to link readers to the right place. Juwan (talk) 23:50, 8 September 2025 (UTC)

WT:CLutTeR... Chuck Entz (talk) 05:32, 9 September 2025 (UTC)

The problem is that changes to the category tree require thought and discussion if we're not just to have a random mess, which is what we were tending to before, when anyone with autopatroller (or sometimes just autoconfirmed) status could edit the category tree modules. I would strongly oppose getting rid of WT:CLTR but I will commit to spending a couple of hours tomorrow night going through and responding to as many CLTR topics as I can, and dealing with any that seem to be actionable. Just FYI, some of the topics concern ramifying the music categories; there's a long discussion in the Beer Parlour (or maybe the Grease Pit?) from near the beginning of this year that I started, proposing one such (very-detailed, 18-way-split) ramification. Juwan, it would be great if you could merge that discussion into WT:CLTR and merge in the existing music-related topics. Benwing2 (talk) 08:06, 9 September 2025 (UTC)

DCDuring and editing other people's comments

Apparently DCDuring thinks it's justified to alter headings of discussions that other people wrote without warning if he finds them objectionable. I don't agree (I believe at least some warning should be given before such an alteration), but what does the community at large think of this attitude? — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 15:13, 5 September 2025 (UTC)

Is your issue with my behavior or my attitude?

BTW, this heading is misleading.

The matter is/was about revising the uninformative heading "Arbitary break" to be a bold non-heading (inserting a semi-colon) and inserting a new heading in its place that was the preceding heading with the word "(continued)". No words were deleted or altered. Only the heading was added. The main effect was to make more informative Watchlist and Recent changes displays. DCDuring (talk) 15:33, 5 September 2025 (UTC)

I posted here to get third-party answers on "what's the proper etiquette" that we are disputing, not to propose any disciplinary action. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 15:39, 5 September 2025 (UTC)

This is the diff. DCDuring (talk) 15:47, 5 September 2025 (UTC)

My only concern is with the semantics here. Please don't start a definition list/dl by using a semi-colon without then following it with a colon and the next part of the key–value pair. If you just want stuff to be bold, use three apostrophes. —Justin (koavf)❤T☮C☺M☯ 23:59, 5 September 2025 (UTC)

@User:Koavf I never heard of this. I use the semicolon, unpaired with anything, in taxonomic name entries often, though for sublists of references, never for definitions. I have been doing so for a long time, probably just imitating something I observed. I have no idea what the key-value pair might be when it's at home. Is this documented somewhere? DCDuring (talk) 00:18, 6 September 2025 (UTC)

Sure, thanks so much for asking. Definition lists in MediaWiki start with ;Some stuff to start the list and then :One or more of these entries. These used to be called "definition lists" (which is appropriate for here), but are now defined as "description lists", so it's not just a stylistic thing that makes things bold, but a semantically meaningful relationship between some information. And lest it seem like I'm being just a judgemental jerk (which, God knows, I have been before :/), I have used ";" to make stuff bold many, many times. —Justin (koavf)❤T☮C☺M☯ 00:25, 6 September 2025 (UTC)

I like saving five keystrokes to achieve bolding in a subheading. Does something bad happen as a result of using the semicolon this way? DCDuring (talk) 00:39, 6 September 2025 (UTC)

Bad semantics is bad for things like screen readers for the blind, so yes, it is a bad thing that happens, particularly to the most vulnerable. There are also second-order bad things like it makes the outline of the page wrong or makes indexing from search engines wrong, etc. Proper semantics is not a mere technical issues but a fundamental access one. —Justin (koavf)❤T☮C☺M☯ 01:10, 6 September 2025 (UTC)

The documentation that you pointed to does not suggest any rigidity to the use of the semicolon, just use to achieve bolding for headers and subheaders. The documentation doesn't mention lists of items starting with asterisks, though, which we have in great abundance. DCDuring (talk) 01:41, 6 September 2025 (UTC)

What? The MediaWiki page isn't about the importance of semantics, that's a broader issue that's true in literally all of the Web. And yes, you should not start a definition list and then proceed to have an unordered list: that is a perfect example of bad semantics. —Justin (koavf)❤T☮C☺M☯ 15:23, 6 September 2025 (UTC)

I see the semantics of free use of leading semicolon in multiple presentations of lists with bold headers that don't show up in tables of contents. If one didn't want bold in the headers, the semicolon could be omitted. Wikitext was designed to not be rigid "semantically". Whether the initial semicolon as I use it is vestigial is immaterial. DCDuring (talk) 15:33, 6 September 2025 (UTC)

It's not. Please don't use incorrect semantics. —Justin (koavf)❤T☮C☺M☯ 18:19, 6 September 2025 (UTC)

Unlike a comment that a person leaves in a discussion, which normally shouldn't be changed by others (though of course there are cases when it can and should be, like to remove spam, fix mismatched '''{{ that is affecting the rest of the page, add a missing signature, etc etc etc), a header is "shared" by the people in the discussion and more broadly on the page, and while I wouldn't change headers willy-nilly (and I personally am used to "arbitrary break" as a subsection header from its common use on Wikipedia), I have changed headers (e.g. someone used {{temp|lat|es}} in the headers which makes section-linking hard), and I don't see anything wrong with replacing "arbitrary break" with an objectively more informative header. Calling the other header "a waste of people's time" was unnecessarily rude; DCDuring, it would beho(o)ve you to have been polite in making the change. I would like to hope that if someone made such a change to a header in a polite way, no-one would try to make a mountain out of it. - -sche (discuss) 16:30, 5 September 2025 (UTC)

OK, but I wanted to say why the change, briefly. DCDuring (talk) 18:34, 5 September 2025 (UTC)

Changing header names is not a prototypical case of "editing other people's comments". I don't generally think of the header name as being attached to a particular participant in a conversation (they aren't uanbiguously attached to a single user's signature). From what I can tell it is correct that DCDuring did not change or remove the text of the header, but just demoted it from a header and added another header above it, and I think the reasons for this are understandable. I agree with -sche. (I'm also not sure why adding this header was necessary in the first place as opposed to something like Template:outdent.)--Urszag (talk) 19:37, 5 September 2025 (UTC)

The diff shows pretty conventional refactoring. This is misleading nonsense. —Justin (koavf)❤T☮C☺M☯ 00:00, 6 September 2025 (UTC)

• • This is the lamest discussion ever Vealhurl (talk) 06:29, 12 September 2025 (UTC)

I want people to leave my comments alone apart from technical fixes, but I welcome people editing headings if they're unclear as long as I like the end result. If not, I might edit them back. It's unfortunate to edit headings when they have appeared in edit summaries, but sometimes it's necessary. — Eru·tuon 16:17, 13 September 2025 (UTC)

Obscure ancient steppe languages

Hi everyone! I noticed some entries directly entered under the name of ancient obscure unclassified languages.

There is a recently entered "Xiongnu" entry. Are there any Xiongnu texts? I am not talking about the loans of titles into Chinese, they should be listed under Chinese, otherwise they would fail RFV. Also another thing is we currently give possible etymologies of Xiongnu etymologies like this:

• >? Xiongnu: a Chinese term borrowed from Xiongnu (like in Reconstruction:Proto-Iranian/hwatā́wā and its Yeniseian alternative Reconstruction:Proto-Yeniseian/atɬ)

Shouldn't they be given like this?

• >? Xiongnu: (unattested, a term cannot be provided unless it is a reconstruction)
  • → Chinese: term

On another note, same goes for Hunnic. The terms Xiongnu and Hunnic are confused (by a lot of sources too even though any historical connection is contentious) really often, like we can see "Xiongnu" descendants in Greek script, or we can see "Hunnic" terms in Chinese characters. Someone should clean this confusion up in reconstructions of P-Turkic P-Iranian and P-Yeniseian.

There is also one Rouran entry. I do not really know about Rouran, are there glossaries/texts in it? Otherwise it would also fail RFV.

A different case can be made for Tuoba, Jie (possibly Rouran as well if we follow Vovin's reading of the Bugut Inscription), and some other obscure ancient steppe languages mentioned in Chinese sources as they at least have a short passage/a short wordlist. I'm open to your inputs. Thanks in advance!

(Pinging the editors of 閼氐 and 去汾 @Saam-andar @AmaçsızBirKişi @Theknightwho) Bartanaqa (talk) 02:24, 6 September 2025 (UTC)

For Xiongnu, Quoting this study: "150 glosses, as well as possibly a very short text consisting of just two-lines in Chinese transcription." Also p. 9 of [2]

I was also confused by the "Xiongnu" terms written in Greek. I’ve put them as Ancient Greek, idk if they should be Category:Hunnic language instead. Saam-andar (talk) 09:35, 6 September 2025 (UTC)

There are no Xiōngnú texts, no. There are Chinese glosses and recordings of such, some of which I've compiled under User:AmaçsızBirKişi/mis-xnu. Ideally, one should dig up the Chinese text the Xiōngnú words were taken from, but that's hard, not for a lack of trying by my part. I guess providing references that tackle Xiōngnú terms is enough in the case of Xiōngnú.

AmaçsızBirKişi (talk) 11:31, 6 September 2025 (UTC)

@AmaçsızBirKişi Not entirely true - as pointed out above, this study by Vovin seems to contain a genuine Xiongnu inscription. The real question is which language it's written in. Theknightwho (talk) 12:44, 6 September 2025 (UTC)

That couplet is classified as the Jié language. It is recorded in the "language of the elite and the ruling class of the Later Zhào Dynasty", hence, Jié. It even has a language code, [mis-jie]. I also have a subpage dedicated to that particular "inscription", User:AmaçsızBirKişi/mis-jie, if by inscription you mean a recorded sentence. Xiōngnú did not leave behind any attested sentences, as far as we know.

AmaçsızBirKişi (talk) 13:22, 6 September 2025 (UTC)

Actually there is one, there is a recording with Chinese characters in Talât Tekin's book titled the language of Xiongnus. BurakD53 (talk) 13:36, 6 September 2025 (UTC)

@AmaçsızBirKişi Sorry - linked the wrong study: [3]. Theknightwho (talk) 13:44, 6 September 2025 (UTC)

Oh, I didn't know that. Thanks for linking that article.

AmaçsızBirKişi (talk) 14:03, 6 September 2025 (UTC)

Accepting that inscription is indeed in Xiongnu lanugage, are there also any inscriptions containing 閼氐, or should it be moved to be listed under Chinese? Bartanaqa (talk) 07:22, 7 September 2025 (UTC)

@Bartanaqa I may be wrong, but as far as I know the only directly-attested examples of Xiongnu are the two inscriptions mentioned by Vovin. I'm in two minds about what we should do with 閼氐: if we can only trace it back to Chinese sources, then that does suggest it should be recorded as Chinese, but I do think that the recent confirmation that the Xiongnu used hanzi to record their language means it could have been coined in that form by the Xiongnu themselves, which was then recorded by the Chinese (e.g. from diplomatic correspondence). Theknightwho (talk) 13:54, 7 September 2025 (UTC)

單于 definitely deserves a Xiongnu entry, though - it's clearly legible on the first inscription. Theknightwho (talk) 13:59, 7 September 2025 (UTC)

I'm moving it to Chinese for now, if someone founds another Xiongnu inscription containing it we can move it back to Xiongnu. Bartanaqa (talk) 11:21, 13 September 2025 (UTC)

There's also already a Chinese entry for it: 閼氏／阏氏 (yānzhī). Saam-andar (talk) 11:35, 13 September 2025 (UTC)

It seems a Xiongnu entry is unnecessary, I am moving the content of the page to Chinese. I do not anything about Chinese so I do not really know what to do with this page. Bartanaqa (talk) 11:43, 13 September 2025 (UTC)

@Bartanaqa Yes, there are genuine Rouran inscriptions. There are also some using the Brahmi script (with two extra letters, apparently coined by the Rouran themselves). Theknightwho (talk) 12:47, 6 September 2025 (UTC)

Some languages aren't recorded by their speakers, but by their neighbors, such as Cuman. I think the quality, rather than the quantity, is important. For example, if the word has truly entered Chinese and is used in Chinese sentences, let's give it in Chinese. But if it's recorded as the Xiongnu saying it, I think we shouldn't consider it a loanword. BurakD53 (talk) 13:32, 6 September 2025 (UTC)

This is true, I refer to Wiktionary:Etymology scriptorium/2025/April#ναρί. Fay Freak (talk) 13:58, 6 September 2025 (UTC)

My point was, are we going to move all "Chinese" words to Xiongnu? Also languages like Cuman and Karakhanid, although technically listed in dictionaries, have a lot of passages. There's a whole sermon recorded in Codex Cumanicus and lots of poems in DLT, so it is not really controversial to list them under a different name. But there is no whole text, not even a full sentence in Xiongnu, so it is undeniably controversial to list them as a different language, I just wanted to check the community's thoughts on this since no one would need a dictionary to read a "Xiongnu" inscription or text as there is none. Bartanaqa (talk) 07:16, 7 September 2025 (UTC)

@AmaçsızBirKişi, what is your source for the pronunciation of Xiongnu lemmas, in particular 閼氐? Dybo reconstructs the Xiongnu transcription as /ɣāt.tə̄j/, and according the Pulleyblank, the Middle Chinese of that period would be /ʔat.tɛj/. Where does /ʔen.tsye/ come from? --{{victar|talk}} 03:23, 13 September 2025 (UTC)

That's from Vovin (2003).

AmaçsızBirKişi (talk) 09:39, 13 September 2025 (UTC)

That page seems to mix up two variants of the word; 閼 has both "at" and "en" readings in Middle Chinese, while 氐 is not palatalised, but 氏 is (and has a k- initial in Old Chinese rather than t-). Xeroctic (talk) 14:45, 13 September 2025 (UTC)

If it's not attested, than it's a reconstruction. I've moved it to RC:Xiongnu/ɣāt-tə̄j. --{{victar|talk}} 18:45, 15 September 2025 (UTC)

Selecting one reconstruction out of the multiple proposed ones with no consensus seems problematic. I think we should delete the reconstruction and move the speculation to the Chinese page. Bartanaqa (talk) 20:48, 15 September 2025 (UTC)

I disagree. Multiple reconstructions for a single term is not unusual in linguistics, nor this project, and citing them together on a reconstruction entry provides the most comprehensive approach. Furthermore, the forms given by Bailey and Vovin appear to be transcriptions of the Middle Chinese borrowing rather than genuine reconstructions of Xiongnu. --{{victar|talk}} 21:43, 15 September 2025 (UTC)

• Merge *ɣāt-tə̄j to the Chinese entry. I do not have the confidence that we even have enough of an idea what the Xiongnu language even is to create any reconstruction pages for it. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 18:09, 19 September 2025 (UTC)

Accent labels in audios

Discussion moved to Wiktionary:Beer parlour/2025/August#Audio pronunciation qualifier formatting: should country names be included?.

Improving help and project pages

as Wiktionary currently has very poor maintenance of project pages, it is difficult for new users to get started (from experience, I basically just went ahead copying code from similar pages and hoped for the best). it is worse for those unfamiliar with wiki editing, who need to start from scratch. this also affects regular editors who need to figure out what is acceptable by osmosis when the community consensus is unwritten.

as such, in order to help both new editors get started and regular editors to more easily find information, I call upon other editors to help write and rewrite. these new pages should be categorised as think tank policies, as vote policies are particularly difficult to change (sometimes frustratingly so). over the course of time, I will add some pages that require attention to this topic or in a to-do.

for help pages (category) in particular, I suggest that (most of) these be incorporated into regular projectspace pages or soft-redirected to Wikipedia, Meta-Wiki and MediaWiki. this would be an improvement to readers (as they get better updated and more in-depth information) and for editors (who don't need to work on those pages anymore). Juwan (talk) 23:42, 8 September 2025 (UTC)

Which pages in particular do you want to change and/or improve? Some time ago I made extensive edits of Help:Audio pronunciations because I've been told that I can do this on that page without specifically asking for anyone's permission. --Ssvb (talk) 15:02, 10 September 2025 (UTC)

The problem is that experienced editors don't need the help pages, so never visit them and hence neglect to update them (if they even know of their existence in the first place). I'd support any updates, bearing in mind that the pages should be kept as simple as reasonably possible so as not to overwhelm newbies with excessive info, like Wikipedia has a habit of doing with its equivalent pages. (I had a bit of a go at Help:Category a while ago, and I acknowledge it gets into the long grass a little bit, but you could argue that adding a category to the tree is a task for at least intermediate editors, not total newbs.) I don't think I would support any dramatic cull or rearrangement of these pages, although the navigation experience could be improved a bit. This, that and the other (talk) 11:45, 11 September 2025 (UTC)

@This, that and the other thank you for putting it well. it is a shame because, even excluding pages for total newbies, editors neglect to update pages for useful topics that they wish would likely wish to read as well. so many a time have others said that they remember a discussion but no one wrote it down and so it was lost to the winds and into the depths of MediaWiki's poor search function. Juwan (talk) 00:10, 13 September 2025 (UTC)

• I go through the Help pages every few years, changing things like obsolete practices or tyops. Unrewarding stuff...Vealhurl (talk) 21:03, 12 September 2025 (UTC)

  it is the boring and unsexy work that keeps the show runningm. Juwan (talk) 21:13, 12 September 2025 (UTC)

Category:English rebuses mostly invalid?

A rebus consists of more than one element, e.g. honey-bee + "N" = "been". Most of the entries in this category are things like a picture of a key meaning (database) key. Not a rebus. 2A00:23C5:FE1C:3701:4DD7:133:70FE:E0AE 18:46, 10 September 2025 (UTC)

I noticed this category when I was browsing other entries — namely 🐐 — and decided to add it to 🔑 based on the rather questionable precedent of single-element single-image entries added into this category. I admit I was surprised but made the edit in a short moment of absentmindedness. From the definition given on the category page, I too believe single-element single-image entries — that is, exhaustively: 🌽, 🍇, 🐐, 🐜, 📠, 🔑, 🥷, 🧢, and 🪢 — do not meet the criteria to be considered rebuses. If anything, this should simply be clarified in the category page. B555c444 (talk) 19:31, 10 September 2025 (UTC)

I wonder if the entries have ended up in Cat:English rebuses because the more sensible location (Cat:English emoji or something) doesn't yet exist. This, that and the other (talk) 11:49, 11 September 2025 (UTC)

Where did you get the idea that a rebus consists of more than one element? Maybe in children's puzzles, but this is not the understanding in Sinology or Egyptology, for example, where a character is considered a "rebus" when a picture is borrowed to represent a word of similar sound. Hftf (talk) 12:55, 12 September 2025 (UTC)

Where did you get the idea they don't? Rebuses are puzzles which rely on an arrangement of multiple elements to spell something other than their intrinsic meaning, whether individual or collective.

Hence, the classification is quite simple:

• multiple elements, one or more of which are an image → rebus
• one element, one of which is an image → emoji, pictogram, ideogram, etc.

Ultimately, the choice whether the category should be depreciated by the inclusion of emoji forms of single words, which are in turn already marked as a wordplay form of a different word (at most), is arbitrary, but it needs to be decided by consensus.

As for Sinology (the only one I'm familiar with out of the mentioned), the example is right on the page of rebus. It's a phonetic hint that is paired with another sinogram. A "rebus" here is not a classification that a standalone sinogram can have. B555c444 (talk) 19:21, 12 September 2025 (UTC)

No, that's not correct whatsoever; then our English entry for rebus appears at the moment to be overly specific (wrong). The pairing of a meaning-bearing element (semantophore) with a phonetic element (i.e., picture rebus) is a typical later development in these writing systems that occurred once the ambiguity of whether a picture indicated what it literally depicted or a different word that sounded alike needed a new solution. Obviously, one of the six traditional classes of sinograms is the rebus character (phonetic loan, jiajiezi). Hftf (talk) 20:52, 12 September 2025 (UTC)

I admit I may have expressed myself unclearly. However, both of my points still stand. This is simply an example of a specialized sense derived from a figurative interpretation of an existing word. Unfortunately, the clarity of my second point is diminished by the fact that the Sinological meaning of "rebus" is a noticeable shift from the puzzle/composition sense to just meaning a part of such a composition.

Your own statement also supports that:

"Obviously, one of the six traditional classes of sinograms is the rebus character (phonetic loan, jiajiezi)."

Again, obviously, the jiajiezi in the derived phono-semantic compound is the "rebus", the jiajiezi itself is not.

Even if my supposition about the semantic shift (of the specialized sense of "rebus") is not true, it does not affect the sense of the "puzzle" meaning of rebus, which the category we are ultimately discussing refers to.

Also, just an additional example for my initial argument: "🌊🦁" to mean "sea lion" → rebus, "[theoretical emoji of a sea lion]" → pictogram. Furthermore, supposing "sea lion" were a wordplay form of another unrelated term, it still wouldn't be grounds for the classification of "[theoretical emoji of a sea lion]" as a rebus. As such, terms like "🔑" ("🔑" → keys → "kys") aren't rebuses, especially since the meaning of the word represented by a pictogram is standalone. Besides, emojis which have no significant disparity from their pictogrammic interpretation usually don't even have pages. B555c444 (talk) 14:20, 13 September 2025 (UTC)

This is just completely incorrect. Jiajiezi is a class of individual rebus characters, which has nothing to do with anything "in the derived phono-semantic compound". Please google "jiajiezi rebus" or "jiajie rebus" or "rebus characters" etc. You don't even need to invoke indirect wordplay like "🔑" → "kys" to find a single-character rebus; "🔑" used to mean a key on a computer keyboard or piano is a rebus because it is not referring literally to a metal tool whose purpose is to unlock a door.

As for what the category name "English rebuses" is supposed to mean, this is a dictionary that often uses technical linguistic terms in categorizing and labeling phenomena, and I would not have thought twice that the category clearly refers to the concept of a rebus in the technical linguistic sense and not in the children's puzzle sense. Hftf (talk) 14:43, 13 September 2025 (UTC)

Since it's hard to argue with someone who disagrees with clear-cut facts, I won't continue reproving the Sinology argument futilely.

However, I'd like to add that by "the 'puzzle' sense" I'm clarifying the manner of interpretation ("composition of multiple elements" vs the derived "the phonetic element of a phonophore-semantophore composition [of which it originally bore the meaning of]") of the word "rebus" of which "a children's puzzle" candidly mentioned by you is only a small subset of (analogously the first sense on the page rebus). The word "rebus" was most likely the one specialized because of its nuance of implied phonetic involvement, which supports the argument that the point is usually wordplay, not what you've described. This is further indicated by the subcategories listed on the category page. (Have you even read the category page?) There's no such thing as a "technical linguistic [rebus] sense" that refers neither to the actual linguistics phenomenon nor to the base meaning. So, no, the category in fact does not refer to the "technical linguistic sense". B555c444 (talk) 15:20, 13 September 2025 (UTC)

@Hftf: Chambers Dictionary agrees with me: "an enigmatical representation of a word or name by pictures punningly representing parts of the word, as in a puzzle or a coat of arms; such a puzzle". 2A00:23C5:FE1C:3701:B5BB:F9F5:7CF6:B0BB 13:10, 17 September 2025 (UTC)

Thanks, but why not instead of cherry-picking a general-purpose dictionary definition, try playing devil's advocate and look for actual linguistic contexts to corroborate what I'm saying about this term's use in linguistic contexts? Anyway, I see the online edition is using the plural much more ambiguously, just as Category:English rebuses. Hftf (talk) 15:56, 17 September 2025 (UTC)

@Hftf: "Cherry-picking"? As if I went around until I found one that agreed with me. It was the first dict on my shelf. I won't pursue debate with such a bad-faither. You have been proved wrong by another user already in this thread, so it's unnecessary too. Bye bye! 2A00:23C5:FE1C:3701:BD91:D983:D60:4265 18:41, 27 September 2025 (UTC)

There's nothing preventing a single character from being used as a rebus. It's not a rebus if it represents the word that refers to the object the character depicts (which would be the case for 🔑 = database key, since "key" in that sense is just a metaphorical extension of the same lexeme), but something like 🐜 = "anti" is a rebus since it's using an image of an unrelated word for its phonetic value.--Urszag (talk) 16:24, 17 September 2025 (UTC)

how we define "obsolete" and "archaic"

Two questions:

• We define "obsolete" as "No longer in use, and (of a term) no longer likely to be understood." Should we cut that to just "No longer in use"? (Or "That has not been in use for a long time" or a specified number of years or generations?) Consider an obsolete taxonomic term *"foobarimorph", and imagine there are texts, none more recent than 1784, using the term "non-foobarimorph": "non-foobarimorph" is just as "likely to be understood" now as in 1783 — it transparently means "not foobarimorph" — but if it hasn't been used in hundreds of years, it's obsolete. No? This point was also raised in 2024. We already removed "likely to be understood" as a factor when it comes to labelling spellings obsolete, since most obsolete spellings, e.g. the kinge of Frannce, are still understandable.
• We define "archaic" as "No longer in general use, but still found in some contemporary texts that aim for an antique style, like historical novels." Any novel set in a particular time period, whether that's the 1850s or the 1980s or the 2010s, may try to use that era's slang, to sound like it is of [or at least set in] that time; for example, almost every 1811 Dictionary of the Vulgar Tongue term wound up in a Georgette Heyer story set in that decade (so technically we need to relabel them all from "obsolete" to "archaic"). Given that, where is the cutoff between "archaic" ("I want this novel to sound like the 1810s") and "dated" ("I want this novel to sound like the 2010s")? Is it "if a significant number of people alive today were using it when it was en vogue, it's "dated", but if no-one alive today grew up using it normally, it's archaic"?

- -sche (discuss) 20:31, 11 September 2025 (UTC)

I argue that archaic is the wrong word. The right label for the definition is archaicizing. It's more about style. So really there are two other age labels, dated, meaning recognized but falling out of use, obsolete meaning not recognized. Vininn126 (talk) 20:33, 11 September 2025 (UTC)

• Wonderfool uses these tags often when dealing with Category:Webster 1913 entries. Generally, if there's been no use in 50 years, it's dated. 100 years, archaic. 200 years, obsolete. But those are kinda arbitrary guidelines. Perhaps we should bundle them up into {{lb|en|old}}... Vealhurl (talk) 06:26, 12 September 2025 (UTC)

Among taxonomic terms, there are large numbers of terms, especially about the rank of family, that are not up-to-date, but differ from terms in current or accepted use only by a prefix (eg, perhaps, eu-) or suffix (eg, -morpha). I take these as being archaic (even if in use as recently as 20 years ago). But most who come across the term would not know its placement and circumscription in much detail or precision, so it could be termed obsolete. I doubt that any argument can be made that literary use is any but an exceedingly rare consideration for taxonomic names. I haven't been using "dated", except in cases or recent, but no longer controversial, changes like Aves being redefined as a subclass of Reptilia. I'd like to make the labels for taxa as compatible as possible with use for real languages, but that might put unreasonable constraints on use for normal languages. Perhaps the specific meanings for taxonomic names could be put on WT:AMUL or a new WT:ATAX page, linked from WT:AMUL. DCDuring (talk) 19:50, 12 September 2025 (UTC)

I don't think this is always the best approach. Vininn126 (talk) 19:57, 12 September 2025 (UTC)

I have been trying to stick to a distinction that I believed (from reading previous discussions like this one) that many other Wiktionarians were following too: (1) "obsolete" denotes something that's not only old but also hard to "see through" for most readers: they can't even guess what it means (or, at most, their guess is achieved only by context alone, not by the morphology of the word); (2) "archaic" denotes something that's old and not in much use anymore but is still transparently or semitransparently "see through" for most readers: they can see pretty clearly what the meaning is, but the word sounds or looks old-fashioned, which is why it is, or could easily be, used by writers for archaicizing effect. Thus "the kinge of Frannce" is merely archaic, not obsolete, by these standards, because it's obvious what it means, and it looks old-fashioned. I use "dated" when a term is old but is still used and is not truly particularly good for any archaicizing effect. I am willing to change to another system if others agree on a new one. Quercus solaris (talk) 05:04, 13 September 2025 (UTC)

User:Agamemenon

The user carries out edit wars (see the history of อนามัย for example). The user has added and is trying to add large numbers of sum-of-parts to Wiktionary (as seen in 86405364, 86405724, 86565052, 86690673, etc), and when those SOPs were removed, he reverted the removals without stating any reason (such as in 86718416). He has been informed of Wiktionary's policy not to include SOPs. Moreover, it is a general consensus not to add periods to definitions of foreign terms (according to Andrew Sheedy). When the periods he added were removed, he instantly reverted the removals for no reason (such as in 86718416).

The user also adds external links to Wiktionary entries without reason. For example, he adds a link to a dictionary website to Wiktionary entries while stating that the terms in those entries are not on that website (as seen in 86582095, 86671244, 86638736, etc). Isn't this spamming or potential advertisement?

--Miwako Sato (talk) 04:58, 11 September 2025 (UTC)

Hi there.

"The user has added and is trying to add large numbers of sum-of-parts to Wiktionary (as seen in 86405364, 86405724, 86565052, 86690673, etc), and when those SOPs were removed, he reverted the removals without stating any reason (such as in 86718416). He has been informed of Wiktionary's policy not to include SOPs."

So all these terms come from the Thai dictionary maintained by SEALang, as I stated in 86714355, which you just-so-conveniently choose to omit in your report. I believe Thai lexicographers, whose input forms the basis of SEALang's Thai dictionary, are a reasonably credible source for "what counts as more than a SOP", given the extremely specific, idiomatic, and unpredictable nature of collocations and compound words. You then proceeded to completely disregard my explanation and wholesale re-reverted everything, and as I was on mobile, I didn't have the time to re-explain the issue when restoring the edit.

"Moreover, it is a general consensus not to add periods to definitions of foreign terms (according to Andrew Sheedy). When the periods he added were removed, he instantly reverted the removals for no reason (such as in 86718416)."

There is a vague rule of thumb on Wiktionary that one should try to respect the "local formatting standards" chosen by previous editors of a given language. I noticed that virtually all the Thai entries had periods added at the ends of definitions, and thus chose to leave most of them as-is. This includes the entry you cite, in which the user who originally added such periods was Wyang (40558766), as well as... you. I never added periods to this entry in particular, I just left what was already there.

"The user also adds external links to Wiktionary entries without reason. For example, he adds a link to a dictionary website to Wiktionary entries while stating that the terms in those entries are not on that website (as seen in 86582095, 86671244, 86638736, etc). Isn't this spamming or potential advertisement?"

This mainly comes from a desire to have a certain "marker" of having finished filling out a given entry, as well as maintaining a consistent source throughout the entries I edit. That said, if you take umbrage at the "potential advertisement" of a publically-funded dictionary repository in cases where it lacks a given entry, feel free to delete the "Further Reading" section and the link to SEALang. Thanks for understanding. Agamemenon (talk) 06:03, 11 September 2025 (UTC)

@Agamemenon It seems like you don't know Thai, but let me tell you that SEALang should be used with caution because it does contain errors and has not been updated for quite some time (as stated on the website, the last update on its Thai dictionary seems to have taken place in 2006, or about two decades ago). Some of the terms you took from SEALang do not deserve entries here, as they are erroneous or actually meaningless or are merely sum-of-parts. For example, the term "องค์การอนามัยแห่งโลก" you added to อนามัย does not exist as the correct term is "องค์การอนามัยโลก", and the term "เลขที่๕๖" you added to เลข just means "number 56" and has nothing special, which makes it a mere sum-of-part. That's why the terms you added have been removed by Thai editors. I hope you will do some careful check before adding entries, instead of just copying and pasting. Thank you. --YURi (talk) 14:16, 11 September 2025 (UTC)

Hi there, thanks for weighing in. A few clarifications before this is taken to Beer Parlour:

• On SEALang: It isn't flawless, but it remains one of the most comprehensive references for Thai. Notably, it has even been used by other editors for other languages — for example, the OP has consulted it for Khmer entries here. Dismissing it outright as "spam" or "meaningless" overlooks its value. Like any dictionary, errors exist, which is exactly why collaborative editing is essential. I’m still learning Thai, but my contributions are carefully sourced, and the collaborative editing model allows for refinement where needed. The goal is to build out entries systematically, not to claim perfect mastery.
• On SOPs and borderline entries: I haven’t systematically spot-checked all existing entries, but I’m happy to filter out obvious SOPs in future additions. That said, there’s rarely anything truly "obvious" — many compounds and collocations are borderline, and Wiktionary itself recognizes this. For example, both you and OP criticized my addition of องค์การอนามัยแห่งโลก to อนามัย as non-existent, noting that the correct term is องค์การอนามัยโลก. Detecting such nuanced or rarely attested forms is not realistic even for native speakers, and I would wager that plenty of people worldwide — including in Thailand — may not know what the WHO is or its local name. Red-linked placeholders exist to mark concepts for future expansion, and slight phrasing issues can be corrected collaboratively rather than discarding entries entirely.
• On Pali-Sanskrit connections: Many entries, such as อนามัย, trace back to Pali and Sanskrit roots and previously had red links further down the etymology chain until I filled them out. Completing these links builds a richer, cross-linguistic framework that benefits everyone, even if surface Thai editors do not immediately recognize the deeper etymology.
• On intent and net effect: My goal is to systematically build out entries, connect red links, and expand attested compounds. The result is more coverage, more blue links, and a stronger framework for others to refine — exactly what Wiktionary is designed to do.
• On specific examples: If a particular term is clearly erroneous or trivial (like เลขที่๕๖), I welcome correction or removal. This is less about personal judgment than about the collaborative model at work: entries can be refined over time, which is exactly what Wiktionary encourages.

Thanks for raising these examples — I’ll keep them in mind as I continue to edit. Agamemenon (talk) 18:35, 11 September 2025 (UTC)

Wiktionary is primarily built around finding three citations in trustworthy, durably archived sources (such as scans of published books) for each of the terms or their senses. See WT:CFI for more details. Yes, adding such citations is often skipped for common terms, which obviously exist in the language. But what is "common" and "obvious" can be only reliably assessed by native speakers. As you are a non-native speaker, you have no way to know when an external dictionary is inaccurate. If you copy-paste such inaccurate entries from that external dictionary and especially engage in edit wars to keep them in Wiktionary, then such activity happens to be harmful and disruptive without you even realizing what's going on. I would suggest you pay less attention to external third-party dictionaries and instead focus on finding and adding citations. These would be very valuable contributions. --Ssvb (talk) 14:38, 12 September 2025 (UTC)

Hi there, thanks for your input. I want to clarify that my edits in this case have focused on expanding derived terms sections, not creating standalone entries for obscure or neologistic words. The concern raised by the OP was about sum-of-parts or minor phrasing issues in derived terms, not citations for new entries.

While citations are valuable and my primary focus in most languages is adding etymologies with proper citations, derived terms sections are allowed to list forms without citations under Wiktionary policy. Relying solely on "native-speaker intuition" is also not fully reliable — many attested compounds and collocations are unknown to many native speakers, especially rare or specialized forms. My use of SEALang as a reference is for systematic expansion, and all entries are open to collaborative correction — which aligns with the "be bold" principle and does not constitute harmful edit-warring.

As an example of recent self-checking, I filtered out a clear sum-of-parts from TDP (กลัวผี "to be scared of ghosts") when expanding the derived terms list for both syllables.

I hope this clarifies the intent and scope of my edits so that discussion can remain focused and constructive. Agamemenon (talk) 04:38, 13 September 2025 (UTC)

This not the appropriate place for this, which is basically a combined content dispute and policy discussion- not something that admins need to know about immediately because of the imminent potential for serious damage. Please move this to a discussion at the Beer parlour. It definitely needs to be resolved, but not Right Now. Thanks! Chuck Entz (talk) 14:57, 11 September 2025 (UTC)

Discussion moved from WT:VIP. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 21:13, 11 September 2025 (UTC)

Million pages recently

Hey, sorry for asking here, but does anyone know about how the page count of our Wiktionary has gone up so much just recently? Just today I noticed it now reads 9.7 million entries, but from a Kiwix dump from not two weeks ago, it's only 8.6 million! I don't recall noticing such a higher number any of the previous days, only today did it make an impression on me. Further, I couldn't find in the contributions of any known bot any mass-created entries, and the statistics page doesn't yet reflect where the change came from, so I really would be curious to know if anyone is aware! I wondered perhaps if we aren't simply counting entries differently now, but I don't know anything about that. Kiril kovachev (talk・contribs) 23:09, 11 September 2025 (UTC)

Unzipping https://dumps.wikimedia.org/enwiktionary/20250801/enwiktionary-20250801-all-titles-in-ns0.gz from 1 August, I get a file with 8,548,101 lines (one line for every entry in Wiktionary's main namespace). Doing the same for https://dumps.wikimedia.org/enwiktionary/20250901/enwiktionary-20250901-all-titles-in-ns0.gz from 1 September, I get 8,584,455 lines. So there has been a modest increase of about 36,000 entries in a month.

I wonder where you saw that figure of 9.7 million entries? The total number of all pages (not just entries) on the wiki is over 10 million, so it can't be that figure. This, that and the other (talk) 00:49, 12 September 2025 (UTC)

The main page says Wiktionary has "9,655,629 entries with English definitions from over 4,500 languages". I wonder if this is related to changes (which I recall were being discussed on Phabricator somewhere) regarding the fact that not all pages which were declared to be "content" pages (e.g. reconstruction pages, IIRC) were being counted as "content" pages by certain Mediawiki or dump-generating functions; maybe they are counted now? Do we have a million Reconstruction: and Appendix: pages? Edit: ah, it's diff, but that just moves the question one stage, what is EntryCount is counting that NUMBEROFARTICLES wasn't? - -sche (discuss) 00:53, 12 September 2025 (UTC)

Okay, well Module:EntryCount counts the number of L2 sections on the whole wiki using the cat:Pages with entries hierarchy (so the number of pages in "Pages with n entries" is multiplied by n, then all those values are added up). That explains the higher count. Have we ever agreed as a community what "entry" means? Does it refer to a page or an L2? This, that and the other (talk) 02:55, 12 September 2025 (UTC)

@This, that and the other Aha, thanks a lot, that makes a lot of sense now. I think "entry" should refer to an L2, since if Wiktionary were treated as a collection of many different dictionaries put together (one for each language), each L2 would be one entry of that language's foreign-English dictionary. A whole page is just how we have to group them for wiki purposes, in my opinion, not necessarily a grouping that matters by itself. Also consider that if thousands of new "entries" were created by the old definition, but the author constrained themselves to only adding homographs of existing page names, then our total of entries wouldn't go up at all, despite a significant increase in content. This is approximately the reality today: I often create Bulgarian L2s that already have Macedonian, Serbo-Croatian or Russian L2s on the same page, so this new counting method would help to capture the true increase in unique content that we're creating. Kiril kovachev (talk・contribs) 12:42, 12 September 2025 (UTC)

Good point. In fact, the use of the word "entry" in WT:EL presupposes that "entry" = "L2" (which I had never noticed before).

Would anyone object to adding a definition for entry at one of our glossaries? Since the term is used on the Main Page and is therefore reader-facing, I think it should go in AP:Glossary.

entry

Each language section on a Wiktionary page is referred to as an entry. For example, the page titled "invention" contains two entries: one for English and one for French.

This, that and the other (talk) 02:06, 13 September 2025 (UTC)

I agree with that idea, that sounds good to have. Kiril kovachev (talk・contribs) 12:00, 13 September 2025 (UTC)

Some also use "entry" to refer to each individual etymology for homonyms. — SURJECTION ^{/ T / C / L /} 14:09, 13 September 2025 (UTC)

• Most of the million pages were Wonderfool's doings (under various usernames, of course). Been working very hard recently... WF really wants to hit the 10 millionth entry too. Hopefully with sth like poopyface...Vealhurl (talk) 06:20, 12 September 2025 (UTC)

  I must say, this was really funny... Kiril kovachev (talk・contribs) 00:24, 14 September 2025 (UTC)

  Tough that your thunder was already stolen on that word, though :) Kiril kovachev (talk・contribs) 00:24, 14 September 2025 (UTC)

adapted vs. unadapted borrowings (or is it calques vs. borrowings?)

-sche and I disagree about whether things like cosplay, dummy out, bodycon, galge, and kogal are adapted borrowings or calques.

My position is that they are more likely to be adapted borrowings, whereas a word like puroresu is an unadapted borrowing. The original focus was on borrowings from Japanese, and I noted that when English words are recognised, it's not uncommon for that to be reflected in the forms of words when they are borrowed (back) into English.

-sche, on the other hand, believes that this is an example of loan translation, saying that it appears that the morphemes are translated in cases like those.

What is the consensus here?

Pinging some relevant Wiktionarians: @-sche, @Eirikr, @Korn. Tharthan (talk) 21:59, 13 September 2025 (UTC)

When this first came up (in 2020), Wiktionary didn't have "adapted borrowing" as a thing (it was introduced in 2024 AFAICT) : it could provide an alternative way to classify these terms (kyabakura et al as unadapted borrowing, light novel et al as adapted borrowing), but I think recent discussion of that was trending in the direction of doing away with "adapted borrowing". My main contention is that terms like kyabakura (which was borrowed in its Japanese form, instead of calquing it [back] to cabaret club), and terms like light novel (which was calqued/translated to English words, instead of borrowing it as *raito noberu), seem like distinct phenomena, and it seems incorrect to call them both "borrowings" indiscriminately. - -sche (discuss) 22:40, 13 September 2025 (UTC)

Calling them calques seems wrong to me. If "adapted borrowing" is being phased out, then just calling them "borrowings" seems like the best option.--Urszag (talk) 04:20, 14 September 2025 (UTC)

"Adapted borrowing" is simply incorrect. The term is being phased out precisely because it is confusing and so many people understand it wrong. — SURJECTION ^{/ T / C / L /} 08:26, 14 September 2025 (UTC)

support for your analysis. most borrowings are adapted in some form (then we don't need to specify "adapted"). by contrast, unadapted borrowings are typically loaned without being respelled, typically formally spelled with italics (in Latin script), typically with a different pronunciation and grammar rules, typically more modern, etc.

as a Portuguese-language editor, this analysis is pretty simple and works well. for Japanese and English terms, it is a bit more complex. for recent katakana borrowings, the borrowed term is still perceived as English in some manner. in romanisations, borrowed terms sometimes are simply written following English orthography ("adapted" into English) rather than Hepburn ("unadapted" form). in my view, it makes more sense to call them borrowings twice rather than a calque. for galge for example, something like gal-game or girl-game (gal isn't used that much anymore?) would be expected if it was a calque. Juwan (talk) 15:08, 14 September 2025 (UTC)

Yes, "gal" has a dated feel to it. Benwing2 (talk) 20:29, 14 September 2025 (UTC)

To be clear, Japanese does use the word "gal" here. The language actually distinguishes between the English words "girl" and "gal." English "girl" is 「ガール」 ["gaaru"], with the long vowel marker serving for approximation of a rhotic vowel, whereas "gal" is 「ギャル」 ["gyaru"], with [ヤ] being an approximation of /æ/. In any case, I agree with Juwan that if the English word were, in fact, "girl game," then we would be dealing with a calque. But that's not the case. Tharthan (talk) 02:51, 15 September 2025 (UTC)

AFAIK the intention of "adapted borrowing" is/was a borrowing where a morpheme was added to the end of the source word in the borrowing language. So for example if Russian гипнотизи́ровать (gipnotizírovatʹ, “to hypnotize”) is considered a borrowing from English "hypnotize", it's "adapted" by the addition of the -ировать (-irovatʹ) ending. It was not intended for cases where the spelling in the borrowing language was somehow "adapted" to make the word closer to existing words. Depending on the case, we consider these as folk etymologies (cucaracha -> cockroach), calques, orthographic borrowings, etc. In this case I agree with @-sche that these are closer to calques than anything else. Benwing2 (talk) 20:28, 14 September 2025 (UTC)

I don't understand the debate. By our own definition a calque is "A word or phrase in a language formed by word-for-word or morpheme-by-morpheme translation of a word in another language." "Cos", "con" (in this meaning) and "ge" are not English words, therefor not translations, and that settles that for me. As for "dummy out", it depends. If "dummy" was not used this way in English before the Japanese used it this way, then the word (meaning) is borrowed, even if the "word" (spelling and pronunciation) existed beforehand. I would generally also consider merely katakana-ised versions of English words, e.g. ライトノベル, to be identical to the English ones. I.e. ライトノベル is essentially "light novel", not a new Japanese word raitonoberu. Thus "light novel" in English is not a translation, it's a re-borrowing of a specific meaning for English words. I would label it in total as a borrowing. I also find it funny that this discussion started while I was in Japan for the first time. Korn [kʰũæ̃n] (talk) 07:59, 27 September 2025 (UTC)

 

A thought: is "wanderlust" a borrowing or a calque? Now, obviously, the evidence indicates that it is a borrowing, but the pronunciation /ˈwɑndɚˌlʌst/ and the fact that it is universally understood by English speakers as wander + lust seems to put it dangerously close to words like beer garden, lightland, copper-nickel and brewmaster.

For that matter, would it even be possible to distinguish a borrowing of German Wanderlust from a calque of Wanderlust? Tharthan (talk) 02:51, 15 September 2025 (UTC)

@-sche, Metaknowledge, Benwing2: The reason why I bring this up is that I think that when dealing with loanwords, there are some instances in which the line between a borrowing and a calque can be fuzzy. I think that that is relevant to our present discussion. Tharthan (talk) 19:37, 16 September 2025 (UTC)

FYI Metaknowledge is no longer active. Wanderlust is an edge case where the distinction between borrowing and calque is indeed fuzzy, but I don't see how that's relevant. Note also that if the word were pronounced 'Vanderloost' in imitation of the German pronunciation, that would make it an unadapted borrowing (see {{ubor}}}); here, "adapted" is being used in the sense you were trying to use it in, which may be adding to the confusion. Benwing2 (talk) 19:46, 16 September 2025 (UTC)

The relevance is that the distinction between borrowings and calques can sometimes be fuzzy. But more to the point: when the word gal was borrowed into Japanese as 「ギャル」, it shifted in pronunciation from /ɡæl/ to /ɡʲa̠ɾɯ̟/, and—if written in the Latin alphabet—it went from gal to gyaru in the process of being loaned into Japanese. When 「力車」 ["rikisha"] was borrowed into English from Japanese, it was borrowed as rickshaw. When 「大君」["taikun"] was borrowed into English from Japanese, it was borrowed as tycoon. All of these could be called adapted borrowings. In contrast, there are many other borrowings from Japanese into English that have been left unadapted (and there are some borrowings from English into Japanese that are written out in the Latin alphabet rather than in katakana—ＡＩＤＳ for example.) Therefore, I would argue that the difference between kogal and kogyaru is no more than a difference between an adapted and an unadapted borrowing.

As we know, whether words tend to be borrowed unadapted or adapted differs from language to language (and, sometimes, between one national variety and another.) In the case of English, I'd say that it's mostly been over the course of the 21st century (outside of weeaboo lingo, at least) that unadapted borrowings into English from languages like Japanese have significantly picked up. Previously, many borrowed terms were adapted to some extent when borrowed into English. This is what I assert is going on with cosplay, dummy out, and bodycon. Tharthan (talk) 22:00, 17 September 2025 (UTC)

I think the pronunciation of wanderlust as /ˈwɒndəˌlʌst/ (as opposed to /'vændəˌlʊst/ or similar) makes it clear that it is a calque based on wander and lust. It helps that the combination of those words makes sense in English (= ‘lust for wandering’), whatever oddity there may be from the sexual connotations of lust. Nicodene (talk) 22:55, 27 September 2025 (UTC)

Category:Linear A language

Is this redundant to Category:Minoan language? Should it be deleted and its subcats reconfigured to be subcats of Minoan? —Justin (koavf)❤T☮C☺M☯ 02:38, 14 September 2025 (UTC)

@Koavf Could you move this discussion to WT:LTR? BTW it definitely sounds like "Minoan language" and "Linear A language" refer to the same thing but since Ethnologue/ISO 639-3 accepted codes for both, it would be enlightening to see the justification for creating whichever one was created later. Benwing2 (talk) 20:15, 14 September 2025 (UTC)

Moving. Keeping this for reference. Please post any new comments there. —Justin (koavf)❤T☮C☺M☯ 20:17, 14 September 2025 (UTC)

On better supporting definite article suffixes in Bulgarian and maybe creating a Category:Articles in general

I stumbled on the -ът (-ǎt) entry and decided to add more information on the article usage. The other article definite forms are missing and I will add them soon. @Kiril kovachev and @MoesianLion should give opinions on whether a small template is to be made to autofill the same information for all the entries or we should just copy paste.

As I was doing the changes, I was thinking "we should discuss adding categories for the definite articles in Bulgarian". However, I realized that no such generic category structure seems to exist for other languages either. While not all languages use an article (and some only a definite), and so this meta-category will not be applicable to all languages, I do think it might be valuable. @Benwing2, could you opine, too, please? Kaloan-koko (talk) 15:27, 14 September 2025 (UTC)

@Kaloan-koko I think there are only -ът, -а, -та, and -то for Bulgarian, so it would be fine to copy-paste the content, but also nothing wrong with making a template (might just be a lot of effort for not a lot of worth). The article category is under "Category:<Language> articles", such as Category:Bulgarian articles. However, currently there are no entries in this category... we should add the Bulgarian ones there. There is also a corresponding category Category:German articles which is a good example. Kiril kovachev (talk・contribs) 17:05, 14 September 2025 (UTC)

I just created the missing pages.

Since Appendix:Bulgarian articles is informative for the usage, I just altered it slightly and referred the other pages to it.

I also added the entries to Category:Bulgarian articles. Kaloan-koko (talk) 05:17, 15 September 2025 (UTC)

@Kiril kovachev Could you also add audio for кратък член (kratǎk člen)? Kaloan-koko (talk) 05:42, 15 September 2025 (UTC)

No problem, I'll go do it shortly. Kiril kovachev (talk・contribs) 09:44, 15 September 2025 (UTC)

@Kaloan-koko We do have Category:Articles by language. -ът is classified as a suffix instead of an article because it is not a separate word. I would actually suggest that the other article forms should just refer the reader to -ът instead of either copy-pasting or creating a template. This is consistent with how we handle suffix forms and more generally non-lemma forms elsewhere. Benwing2 (talk) 19:56, 14 September 2025 (UTC)

O'odham spelling conventions

Hey all, I've hit a bit of an impasse and could use some opinions from more seasoned contributors regarding an "official" Wiktionary spelling convention for the O'odham language. The core issue is that there are two major competing spelling standards, both official in different jurisdictions, both of which present some issues for Wiktionary entries.

For those unfamiliar, O'odham is an endangered Uto-Aztecan language spoken in central/southern Arizona and northern Sonora. The variety spoken around the Phoenix area (Maricopa and Pinal Counties), in the Gila River Indian Community (GRIC), Salt River Pima-Maricopa Indian Community (SRPMIC), and Ak-Chin Indian Community (ACIC) is also known as "Akimel O'odham" or "Pima"; the variety spoken around Tucson (Pima County), in the Tohono O'odham Nation (TON), and northern Sonora is also known as "Tohono O'odham" or formerly "Papago" (the term now considered pejorative). These varieties are mutually intelligible so Wiktionary entries are under a single "O'odham" header.

There are two spelling standards for O'odham: Alvarez-Hale and Saxton with use as follows:

• Alvarez-Hale: Official orthography in TON and SRPMIC (variant used in the latter). Used in one of two available dictionaries: Mathiot's ‎‎Tohono 'O'odham-English Dictionary, published in 1973 and revised around 2013. Used in the only published grammar, Zepeda's A Tohono O'odham Grammar, published in 1983. Alvarez-Hale also tends to be the written standard in modern academic publications.

  Example: Ṣ hab kaij g cu꞉wĭ, "Pi g 'am ñ-hu꞉g n-apt pi hemho 'o wa mu꞉. Baptp ñ-elid mañ ḍ cu꞉wĭ ñ 'eḍa ḍ ge ko'owi t g ñ-ko'okdag 'o m-mea."

• Saxton: Official orthography in GRIC. Used in one of two available dictionaries: Saxton's Dictionary: Tohono O'odham/Pima to English, English to Tohono O'odham/Pima, published in 1983.

  Example: Sh hab kaij g chuhwi, "Pi g am ni-huhg n-apt pi hemho o wa muh. Baptp ni-elith mani d chuhwi ni eda d ge ko'owi t g ni-ko'okthag o m-mea."

Alvarez-Hale has wider use and so would seem the natural choice, but presents multiple problems for Wiktionary entry creation (and text input in general).:

• Word-initial apostrophes: In Alvarez-Hale, words starting with a vowel (in practice, a glottal stop then vowel) start with a glottal stop apostrophe. This causes issues with sorting (all apostrophe-initial words are sorted under ' ) and entry creation (when creating a vowel-initial entry the user needs to create a corresponding apostrophe-less entry for redirects). e.g. Alvarez-Hale "child" is 'ali and "person" 'o'odham. In Saxton these are ali and o'othham.
• Colon character: Long vowels are marked by a colon, which is a character that causes issues for entries, so entries have to use substitute characters like modifier letter colon ꞉ or triangular IPA colon ː, neither of which are easily searchable. Again a user would need to create the corresponding links elsewhere, e.g. for ha꞉l (not ha:l), a link under "See also" on the page hal. Saxton uses an H after long vowels, so the corresponding Saxton spelling is hahl.
• Special characters: Alvarez-Hale sticks to a one-sound-one-letter rule, sometimes with special characters, whereas Saxton tends towards using English digraphs. Again, creating an entry with diacritics necessitates creating a corresponding entry without diacritics for redirects. Demonstrative examples: "boys" cecoj vs chechoj; "nose" da:k vs thahk; "chair" daikuḍ vs thaikud; "shoes" ṣu:ṣk vs shuhshk.

In summary:

Alvarez-Hale

+ More widely used. Basis for academic publications and teaching materials by TON and SRPMIC.

− Punctuation colon must be substituted with a special character.

− Special characters necessitate creating redirects to new entries, e.g. susk redirecting to ṣu꞉ṣk.

Saxton

+ Easier to search for entries.

− Less widely used. Official in GRIC but very little teaching material available.

− Omits certain distinctions made by Alvarez-Hale, e.g. short vowel ĭ (omitted or written as i) or ñ (written n or ni).

There doesn't appear to be any plan to unify these standards among the O'odham-speaking jurisdictions even despite continued language loss, so it puts Wiktionary in the weird position of needing to pick a side in a discussion that no one is having. I'd appreciate any suggestions y'all might have about how to approach this, internally among Wiktionary editors or (hopefully) involving native speakers and those jurisdictions that have a stake in language preservation.

-Sumiaz (talk) 17:26, 14 September 2025 (UTC)

@Sumiaz (FWIW O'odham is the local indigenous language where I grew up (in Tucson) and I studied it somewhat in linguistics classes at the University of Arizona.) Some comments:

1. There are lots of languages that have diacritics like ṣ and ḍ in them e.g. Serbo-Croatian with š č ž đ ć etc. In general we don't create redirects from diacritic-less spellings to spellings with diacritics unless the diacritic-less spellings are commonly found in written sources. MediaWiki in fact will auto-redirect from diacritic-less spellings to spellings with the proper diacritics if the diacritic-less word doesn't otherwise exist. So I don't see these as compelling arguments against using Alvarez-Hale.
2. Since Alvarez-Hale is the dominant spelling system we should definitely use it as the canonical spelling of the terms. However, we can treat the Saxton spellings as alternative spellings and add soft redirects from the Saxton spellings to the Alvarez-Hale spellings. This is similar to what we do in many languages with multiple scripts or spelling systems.

Benwing2 (talk) 20:11, 14 September 2025 (UTC)

@Benwing2 Oh that's awesome! I'm based there now so I'm hoping to get some more first-hand exposure outside of the literature but I'm stuck with paper sources and recordings for now.

1. I should clarify, I think "redirect" wasn't the right term. It was more about connecting the words with diacritics to the diacritic-less form as a "See also" at the top of the page, less so about creating a redirect page. But you're right, that shouldn't be a barrier to using Alvarez-Hale.
2. Definitely! I'm all for using Alvarez-Hale except that the long-vowel colon is causing problems... Maybe that's the core issue here? The character that someone typing O'odham would use is the punctuation colon : (U+003A), but this character causes problems for Wiki links (I think?) as in the example above where "ha:l" (with punctuation colon) links to the Hausa language entry for "L". In order to get around this, I had been creating entries using the modifier letter colon ꞉ (U+A789), but then these terms aren't searchable if typing the punctuation colon. I was looking for the example ke꞉k (letter colon) but searching ke:k (punctuation colon) won't bring up the entry in a search.

- Would the convention for unsupported titles in this case be to use the letter colon (U+A789) in the title and the punctuation colon (U+003A) in all other cases (i.e. the body of the entry)? (I guess the issue here is that if there are no usage examples then only the letter colon is searchable on the page).

- OR should we use the punctuation colon (U+003A) that is actually used in practice, and create a page with an unsupported title when necessary? I suppose it would be possible to create an unsupported title page Unsupported titles/ha:l but I want to make sure this is permitted/expected so someone else doesn't come along and change them all to the letter colon after the fact (for example as on the page 'a:li).

I've been using letter colon in this setting only by necessity but no one would actually seek out that character in regular typing, so I'd rather prefer to avoid it if possible. The use of punctuation in the middle of an unabbreviated word maybe seems odd to the outsider but I think it's necessary in order to allow Alvarez-Hale entries to be searchable. Thoughts?

-Sumiaz (talk) 15:12, 15 September 2025 (UTC)

@Theknightwho Can you comment on this? How annoying will it be for editors if we use the unsupported-title mechanism to handle words with colons in them that would otherwise be interpreted as references to other-language wiktionaries, and will it solve the searching problem? BTW the U+A789 just shows up as a box in my browser (Chrome on MacOS 13.7), but that might be fixable with appropriate font settings in MediaWiki:Gadget-LanguagesAndScripts.css. Benwing2 (talk) 21:31, 16 September 2025 (UTC)

Should North Picene exist as a language?

I recently created two entries for the North Picene language. However, I have since learned that the entire existence of the North Picene language might be a hoax. Currently, the only attestation for the North Picene language is a set of inscriptions referred to as the Novilara Stelae. The provenance of these artifacts remains obscure as the original excavation was poorly documented. Moreover, a recent study concluded that the inscriptions were likely forged. Unfortunately, I cannot access this paper and thus cannot judge the strength of its arguments. It does seem like this study goes against the preexisting communis opinio, which affirmed the veracity of the inscriptions (at least if Trismegistos is to be believed). I could also find several other philologists who all discussed the inscription extensively as if it were completely genuine: Harkness 2011 and Blažek 2009. Though, all of these studies were written before the publishing of the paper that concluded it was a hoax.

I'm not entirely sure if North Picene should even be a language on Wiktionary, considering that it might be completely fictitious Graearms (talk) 17:41, 15 September 2025 (UTC)

I support the removal of the language from Wiktionary modules. Whether fictitious or not, it currently has no purpose, as no entries can be made and the language wouldn't appear in other entries' etymology sections. The increasing academic scepticism on their authenticity makes it even more of a preferrable choice. Catonif (talk) 16:43, 16 September 2025 (UTC)

I agree with @Catonif here. Since the language is untranslated and likely untranslatable, there's no way to make any entries in it, so there's no particular point in having a code for it esp. if there's reason to believe it may be a hoax. Benwing2 (talk) 19:54, 16 September 2025 (UTC)

If it's a hoax, let's move the entries to an appendix like Appendix:Grunge speak hoax. Ioaxxere (talk) 11:16, 19 September 2025 (UTC)

We could if there was scholarly consensus around what the words (even only some of them) should mean, I'd be surprised if there is. Catonif (talk) 11:32, 19 September 2025 (UTC)

@Catonif There are only two words for which I can find a proposed meaning or etymology in English-speaking sources: mimniś and rotnem. The former is defined by Harkness as "monument," a theory which apparently was also proposed by Poultney, although I cannot access the original source by Poultney. This term has also been proposed to be cognate with Oscan memnim. There is also rotnem, although Harkness seems be the only English-spekaing source that actually provides a definition. Graearms (talk) 18:26, 19 September 2025 (UTC)

@Graearms I saw those and assumed they would only crop up in those two papers. Searched for a bit and sure enough Agostiniani, Luciano (2003), Le iscrizioni di Novilara, p. 603 in ΑΙΩΝ 25 (Scritti scelti, vol. 2) does mention the connection with Italic langauges for mimniś and rotem and connects śoter/śotriś/šoteri, polem, isperion, vilatas, teletaú, partenúś to Greek and trút to Etruscan. One could make an appendix though it seems an unwise spending of our time.

Very scientific translation of the Novilara stele by two people from Pesaro I found and translated in English

MIMNIS. ERÚT. CAARES. TARES

Suddenly (violently) the Carians and Tarians assaulted

ROTNEM. ÚVLIN. PARTEN. ÚS

the once partially non-fortified which is

POLEM. ISARION. TET

the city of Pesaro. This

SÚT. TRATNESI. KRÚVI

assault terrified us hidden

TENAC. TRÚT. IPIEM. ROTNEM

with tenacity. They assaulted the above non-fortified

LÚTUIS. THALÚ. ISPERION. VÚL

with mournings [= killings]. He filled untill late with pleasure

TES. ROTEM. TEÚ. AITEN. TASÚR

for them the chariot. He hence kept the loot.

SOTER. MERPON. KALATNE

The saviour Merpon shouted

NIS. VILATOS. PATEN. ARN

to our farmers to get to the shore

ÚIS. BALES TENAC. ANDS. ET

whence they would tenaciously throw the swords, and

SÚT. LAKÚT. TRETEN. TELETAÚ

they assaulted. They shouted to strike fear. He freed

NEM. POLEM. TISÚ. SOTRIS. EÚS

our city. To the mercy of its saviour.

Catonif (talk) 19:23, 19 September 2025 (UTC)

Diacritics for Old North and South Arabian

The current transliteration schemes for both Old North Arabian and Old South Arabian use numbering in the form of s¹, s², and s³ for the fricatives 𐪊 𐪆 and 𐪏 in Old North Arabian and the fricatives 𐩪 ,𐩦, and 𐩯 in Old South Arabian.

I am wondering whether using diacritics such as ś, š and s might be preferable to make the transliteration less confusing for readers.

I am not necessarily requesting that this change be made, but it's certainly worth discussing. Antiquistik (talk) 13:38, 16 September 2025 (UTC)

It depends on (a) what the recent scholarly conventions are, and (b) whether there's universal agreement on how the three sibilants were pronounced. The reason the numbering scheme was originally adopted was precisely because there wasn't any consensus on how they were pronounced, and if that's still the case, then using notations like ś, š and s that imply specific pronunciations will only make things worse. Benwing2 (talk) 19:49, 16 September 2025 (UTC)

Why would it be cOnFuSiNg specifically?

Both are current scholarly conventions. They are already independent of our insights about pronunciation, similarly to the names of the Slavic palatalizations, which aren't chronological according to current knowledge. Fay Freak (talk) 20:15, 16 September 2025 (UTC)

It's confusing because the use of ś, š and s would imply pronunciations something like /ɬ/, /ʃ/ and /s/ respectively, and if it turns out that š is /s/ and s is /ʃ/, it will be hella confusing. As for the Slavic palatalizations, (a) the first did precede the second, and (b) the third is now usually called the "progressive palatalization" rather than "third palatalization" precisely because its order isn't clear. Benwing2 (talk) 21:26, 16 September 2025 (UTC)

@Fay Freak, Benwing2 Those are fair objections. I suppose it's better to keep the present scheme. Antiquistik (talk) 11:07, 21 September 2025 (UTC)

Adding an label parameter to usex and collocations

The quotation modules have a parameter |lb=, while {{ux}} and {{co}} seem to only have |q=. Could we add it? Catonif (talk) 14:17, 16 September 2025 (UTC)

Yes, please! Vininn126 (talk) 14:23, 16 September 2025 (UTC)

What is such a parameter for? — Sgconlaw (talk) 16:15, 16 September 2025 (UTC)

I often end up having dialectal usex's that aren't always best for a quote, etc., Vininn126 (talk) 16:18, 16 September 2025 (UTC)

Some examples: енӄоӈ, boenë, kaz. Catonif (talk) 16:26, 16 September 2025 (UTC)

@Catonif: thanks. — Sgconlaw (talk) 19:04, 16 September 2025 (UTC)

@Catonif: Are we sure those are created by our own users (especially the Ket one)? Because if not, they shouldn't be usexes, but rather quotes (or just removed). Thadh (talk) 16:45, 16 September 2025 (UTC)

They aren't user-made, and I disagree with what you say. But it's besides the discussion as you would need the parameter anyways for user-made usexes. Catonif (talk) 16:49, 16 September 2025 (UTC)

Yes, this is an oversight, I will add it. Benwing2 (talk) 18:08, 16 September 2025 (UTC)

How should we transcribe syllabic consonants in English?

I wonder how we should phonemically transcribe the pronunciation of English words that contains syllabic vowels. I made a few edits that do this by using /l̩/ because I saw that the entry "table" also do this, but I'm not sure if it was a good idea. I think that's a very clear way to transcribe it, but /l̩/ doesn't create a minimal pair, so you can argue that it should only appear phonetically i.e. [l̩].

For comparison: OLAD just uses /l/, but Cambridge transcribes /(ə)l/.

Your thoughts? 5.172.255.98 23:29, 16 September 2025 (UTC)

Unless current practice changes drastically, English pronunciation entries aren't going to be consistent among each other in practice. John Wells has argued that syllabic consonants are phonemically all /ə/ + consonant, which seems good enough to use as a standard phonemic transcription to me, despite the fact that their usage of syllabic consonants is variable across accents and you might be able to find some speakers who make distinctions on lexical rather than fully phonological grounds. Notation like [(ə)ɫ]] is potentially ambiguous in phonetic transcriptions when it precedes a vowel (e.g. it's not clear whether [ˈtʰeɪb(ə)ɫɪŋ] is meant to stand for all three of [ˈtʰeɪbəɫɪŋ], [ˈtʰeɪbɫ̩ɪŋ] [ˈtʰeɪbɫɪŋ]) although this is not really a significant issue.--Urszag (talk) 00:35, 17 September 2025 (UTC)

LiquidThreads will be removed from all wikis

Hello community,

As mentioned earlier, LiquidThreads will be retired from all wikis.

This extension is not maintained and create a lot of issues on the maintenance of the wikis that use it. It is also a blocker to the deployment of Temporary Accounts to your wiki. It is used on a few pages at your wiki.

To prepare for this:

1. Move LiquidThreads pages to subpages (see below)
2. LiquidThreads pages will be converted to Flow boards. While Flow will also be decommissioned, this conversion step is a needed phase to prepare the next steps.
3. The original LiquidThreads pages will be replaced with redirects to the corresponding Flow discussions.
4. We will then freeze LiquidThreads so no new topics can be created.
5. Then Flow boards will be set to read-only as well, as they are on other wikis. Farther in the future, we plan to convert all Flow boards to plain wikitext across wikis.

Once we freeze LiquidThreads, we’ll be ready to deploy Temporary Accounts. We expect to start this un-deployment on Thursday, 18 September 2025.

Please move your pages to subpages by the end of Thursday, 18 September 2025. Wikimedia Foundation staff will manually move some pages if needed.

We are sorry for the late notice. Thank you for your understanding and cooperation.

Trizek (WMF) (talk) 09:06, 17 September 2025 (UTC)

@Trizek (WMF) it says "see below" next to step 1, but I'm not sure exactly what we're supposed to see.

To be honest, I think it's best to let the relevant WMF team handle all steps of this process, including step 1 - too many cooks spoil the broth and all that. There only appear to be 41 pages (plus WT:LiquidThreads testing) that need moving, so it won't take the staff member very long, and they will know the correct actions to take in each case. This, that and the other (talk) 11:52, 17 September 2025 (UTC)

@This, that and the other, the "see below" was for my second sentence in bold. Sorry for the lack of clarity.

The number of cooks depends on how specialized the cooks are. The idea there is to let your community make the sauce (the manual page moves) the way you prefer before we, who aren't sauciers, make it.

We will start the page moves soon. Trizek (WMF) (talk) 10:04, 18 September 2025 (UTC)

@Trizek (WMF) I see your motivation and it's always good to check in with communities in this way. Luckily for you, all the pages concerned (with the possible exception of User talk:Rua) are practically-zero-traffic talk pages, so I doubt anyone here is greatly bothered about how the moves are carried out - so long as we can still read the talk archives somehow or other.

Incidentally, I noticed a message at User talk:Commander Keane#Archives. I assume his archives aren't lost forever - can you bring them back? This, that and the other (talk) 11:15, 18 September 2025 (UTC)

Thank you for noticing the Commander's message, we will have a look at it shortly. Trizek (WMF) (talk) 13:51, 18 September 2025 (UTC)

@Trizek (WMF) @ESanders (WMF) I see this has been done with the exception of User talk:Rua. I have gone ahead and moved this one myself. This, that and the other (talk) 07:46, 19 September 2025 (UTC)

Thanks, as that board was recently active we were giving them an extra day to update. ESanders (WMF) (talk) 11:03, 19 September 2025 (UTC)

Cleanup notes:

Could some admin please use Special:MergeHistory to histmerge all but the last edit of User talk:Rising Sun/LQT Archive to User talk:Rising Sun, since that talk page was effectively c&p moved.

Will enwiktionary want me to run mw:User:Flow cleanup bot to export the content of these previous LQT boards to wikitext with history, once the WMF is done with their processed?

Pppery (alt) (talk) 15:50, 19 September 2025 (UTC)

I've done the history merge. I have no idea about the bot, but if it's the best way to keep the discussion history, I guess so. I don't know what the alternatives are, though. — SURJECTION ^{/ T / C / L /} 18:02, 19 September 2025 (UTC)

@Pppery (alt) ultimately the end game should be to have the discussions preserved in wikitext form. How that happens is not really of great consequence, so long as the tool used does not make a mess. Trizek's comment implies that the WMF has or will have their own solution to this problem, but if your bot is known to do a good job, I don't see any reason why we shouldn't take you up on your generous offer. This, that and the other (talk) 14:02, 20 September 2025 (UTC)

(Same person as Pppery (alt), now editing from my main account) My guess is that the WMF will either run You can see on MediaWiki.org what the bot did and come to your own conclusions about whether it is known to do a good job - I would say so (albeit there are certain things I wish I had done differently on MediaWiki.org) but I'm biased of course, and I will need to make minor fixes to the script to better handle the Flow-as-a-temporary-state angle. My guess is that if I don't do anything, many many months down the line the WMF will either enshrine my bot as the one way of converting Flow to wikitext, or run some trivial variation of convertToText.php - the main difference between the two is that my bot also produces a revision history showing an approximation of how the page would have looked if it had been using wikitext from the start, which means edits to previously-LQT pages will show up in users' contribution histories among other things.

There has been a kind of ridiculous series of unfortunate events associated with running bitrotted code so the LQT convert isn't done yet, but once it's all done:

The Wiktionary community will need to create equivalents for the templates mw:Template:LQT-enabled, mw:Template:Flow summary, mw:Template:LQT imported revision (feel free to give them different names, that's what I called them on MediaWiki.org) - the script also uses mw:Template:Archive top and mw:Template:Archive bottom but I can update it to use {{archive-top}} and {{archive-bottom}} on this wiki and I'm not even sure if that case survives the LQT->Flow import or if it's Flow exclusive.

My bot account will need to be granted at least XML importer rights (which can only be done by a Steward), and preferably also admin rights. I assume that would need a vote.

I assume you would want to delete the temporary "/LQT Archive" Flow board and replace it with the wikitext export, and also delete the "/LQT Archive/LQT Archive 1" (yes, really) page leftover from the multiple moves, leaving the ultimate state just being one archived page, not three. Is that correct?

* Pppery * _{it has begun...} 22:31, 25 September 2025 (UTC)

Tracked in Phabricator
Task T405080

@Pppery thanks for the info. We'll await completion of the LQT-to-Flow migration for now. I can assist with the bot-related steps later. (By the way, mw:Template:LQT imported revision doesn't appear to have ever existed. Did you mean something else?) This, that and the other (talk) 01:40, 26 September 2025 (UTC)

Sorry, bad link. I meant mw:Template:LiquidThreads imported revision. * Pppery * _{it has begun...} 01:47, 26 September 2025 (UTC)

Possible problem with en-verb and complicated 'heads'?

See bowler hat out. The verb template is {{en-verb|[[bowler hat]]<> out}} but it does not link to bowler hat as one might expect, but to the two words separately. 2A00:23C5:FE1C:3701:B5BB:F9F5:7CF6:B0BB 13:08, 17 September 2025 (UTC)

Fixed it! The template needed a separate "head" parameter so it's now {{en-verb|bowler hat<> out|head=[[bowler hat]] [[out]]}} and appears to work.

- Sumiaz (talk) 15:29, 17 September 2025 (UTC)

User:DanielParoliere creating loads of illegal redirects

Please help, I can't keep up reverting them, lol. 2A00:23C5:FE1C:3701:B5BB:F9F5:7CF6:B0BB 14:58, 17 September 2025 (UTC)

• Get an account! Vealhurl (talk) 20:38, 17 September 2025 (UTC)

  You have more than enough accounts for both of us, WF. 2A00:23C5:FE1C:3701:F964:38DE:F1EC:9F85 09:24, 18 September 2025 (UTC)

User:NaomiAmethyst for template editor

I'd like to get template editor to help out with Word of the Day (per User talk:NaomiAmethyst#Helping with WOTDs), as well as to help out with other templates that are protected behind the group. I've made a few modules (Module:Wiktionary:Word of the day/Status and Module:was wotd). I'm also an admin on enwiki. — Naomi Amethyst 00:43, 18 September 2025 (UTC)

• Support. — Sgconlaw (talk) 03:17, 18 September 2025 (UTC)

I would recommend you start with requesting autopatroller rights; this is the level below template editor. I see from your user page that you have a lot of programming experience and experience with Wikipedia modules, but template editors are expected to know the Wiktionary module framework well, which is very different from Wikipedia's module framework and in several ways more complex and well-developed. Template editors can end up doing significant damage if they're not familiar with the functionality of and interactions between core modules, which can be subtle. We've had issues in the past with people importing Wikipedia modules into Wiktionary that either duplicated existing functionality or weren't properly rewritten to follow Wiktionary standards and interface in normal ways with other Wiktionary modules (instead people have tended to import all the needed Wikipedia dependencies directly, leading to a messy parallel hierarchy). As a result we tend to tread carefully when giving template editor rights. One thing that could be done esp. if you're mostly interested in working on particular modules and templates (e.g. the WOTD-related ones) is for you to ask for autopatroller rights, and if granted, we can lower the protection on these particular modules and templates to autopatroller. Benwing2 (talk) 04:01, 18 September 2025 (UTC)

I certainly wouldn't be looking to import Wikipedia modules to Wiktionary—I've been rather impressed with what is here on Wiktionary as far as foundational libraries and general consistency. But restricting template editor doesn't restrict people from importing Wikipedia modules, because you don't need to have the group to create new modules and templates, just improve the existing ones.

Autopatroller may be sufficient for specific areas, but it also requires lowering the protection level of each of the pages in areas I want to work on. — Naomi Amethyst 05:51, 18 September 2025 (UTC)

Yes you are right that anyone can still create imported Wikipedia modules; I brought that up just as one thing that could go wrong (and without template editor status it's not possible to hook an imported Wikipedia module into an existing core module, which would make it that much harder to disentangle). We had a certain admin here who caused a bunch of havoc a couple of years ago editing core modules without proper testing and without a full understanding of the connections between the modules and the need to work carefully when changing them, so I am leery of granting template editor permissions to someone whose work I don't know. (Although template editor status is not identical to admin status, it's tantamount to the same thing when it comes to modules because nearly all protected modules are protected at the template editor level or below, rather than the admin level.) It's not that I don't trust you to do good work, it's just that I would like to tread lightly to avoid any possibility of something going wrong.

Currently, the number of existing template editors is low (~ 20), and all of them first went through autopatroller status (the only ones without autopatroller status are a few bots, like my own). What I would propose is that we grant you autopatroller status and lower the protection of the area(s) you want to work on first, and then once we've verified the quality of your work, grant you template editor status.

BTW what are the modules you are thinking of working on besides WOTD-related stuff?

Also I'd like to hear from some of the other people out there who work on core modules. Benwing2 (talk) 02:55, 19 September 2025 (UTC)

Support. good work! Juwan (talk) 09:03, 18 September 2025 (UTC)

Support 0DF (talk) 12:20, 18 September 2025 (UTC)

Support Rodrigo5260 (talk) 14:29, 18 September 2025 (UTC)

Temporal and locative adverbs categorisation

Notice CAT:English locatives, CAT:Ojibwe locative adverbs, CAT:Ottawa locative adverbs and CAT:Ojibwe temporal adverbs, CAT:Ottawa temporal adverbs. I would say this could be properly implemented and generalised into the categorisation modules. Catonif (talk) 16:21, 18 September 2025 (UTC)

@Catonif There's already CAT:English time adverbs which is in the category tree, along with three subcats for duration, frequency and point-in-time. We could add locative adverbs but the temporal adverb categories should use the existing ones. Benwing2 (talk) 17:23, 18 September 2025 (UTC)

For that matter, CAT:English location adverbs is also in the tree. Benwing2 (talk) 17:24, 18 September 2025 (UTC)

@Benwing2 Thanks very much, :) I'll move the Ojibwe and Ottawa entries to their right categories. I'll leave the English thing for further discussion. Catonif (talk) 18:02, 18 September 2025 (UTC)

@Catonif You're welcome! It looks like CAT:English locatives is a mixture of adverbs and prepositions, and CAT:English location adverbs only has adverbs pertaining to static position. Maybe we could create subcats of 'location adverbs' for static position, movement towards and movement away, or something like that, and either create a parallel hierarchy for prepositions or just de-categorize the prepositions, since (a) most prepositions indicate some sort of location, (b) prepositions often can be used with either time or location interchangeably. Benwing2 (talk) 18:09, 18 September 2025 (UTC)

I like the idea of subcategorisation for the location adverbs. Doing the same for pre- and postpositions sounds trickier (cross-linguistically) as each often has a very large number use cases, though still possible I guess. Catonif (talk) 18:52, 18 September 2025 (UTC)

@Catonif Sounds good. Do you like the idea of splitting by static position, movement towards and movement away? If so I'll go ahead and add that. Benwing2 (talk) 18:53, 18 September 2025 (UTC)

@Benwing2 I'd be fine with it. Perhaps only one movement category, since I can't imagine there being many of those, the vast majority being static. However we are only two people, I assume some may want to have a say. Catonif (talk) 19:10, 18 September 2025 (UTC)

Sounds good. I'll go ahead and make a two-way split and if anyone else disagrees later, we can always change it. Benwing2 (talk) 19:14, 18 September 2025 (UTC)

@Benwing2, Catonif: Just one question: Why does the category structure call these "location" and "time" adverbs, rather than "locative" and "temporal" adverbs? 0DF (talk) 23:45, 18 September 2025 (UTC)

Good point, although the current syntax may be simpler, the more common and professional way to do it (and FWIW, what I looked for first) is with the adjective, Google Ngram links for temporal and locative adverb, although I'm fine with any name as long as the category is there. Catonif (talk) 11:37, 19 September 2025 (UTC)

@Catonif: That was my sense of the usage, too. The other alternative I've seen is "of time", although I associate that with grammatical cases, rather than parts of speech. @Benwing2: Unless there's a good reason for using the nouns in apposition, how about changing the category structure to call these "locative" and "temporal" adverbs? 0DF (talk) 13:01, 19 September 2025 (UTC)

Pronunciation templates' changes

I changed the phonemes in Module:gmh-pronunciation according to The Oxford Guide to MHG, but they are being reverted with no rationale given. There are also some unaddressed suggestions at Module talk:ru-pron. Qbli2mHd (talk) 15:40, 19 September 2025 (UTC)

I addressed your posts about Module:ru-pron. I don't know why your changes to Module:gmh-pronunciation by @TheFells are getting reverted (or even who that user is), but you should discuss with them why they are reverting. Benwing2 (talk) 03:55, 20 September 2025 (UTC)

Madness: zh-pron where all Chinese languages are listed together

I believe the Chinese languages should be broken up into separate language headers rather than combined into "Chinese", which is not a language but a language group. Middle Chinese and Old Chinese are included in Chinese, along with separate languages like Mandarin and Cantonese.

Here is all the pronunciation data about the word "one" in "Chinese" as found in zh-pron on the 一 entry (see below).

It would make more sense, and there would be less need of a drop-down option in zh-pron, if each one of these languages had their own language header section on the 一 entry, and pronunciation section under that header. zh-pron could be used under each language header.

Pronunciation

Pronunciation

   Mandarin

       (Standard)

           (Pinyin): yī (yi1)
           (Zhuyin): ㄧ
           Duration: 1 second.0:01

       (Chengdu, Sichuanese Pinyin): yi2
       (Xi'an, Guanzhong Pinyin): yǐ
       (Nanjing, Nanjing Pinyin): yi̊q
       (Dungan, Cyrillic and Wiktionary): йи (yi, I)

   Cantonese

       (Guangzhou–Hong Kong, Jyutping): jat1
       (Dongguan, Jyutping++): jak7
       (Taishan, Wiktionary): yit2
       (Yangjiang, Jyutping++): jat3

   Gan (Wiktionary): it6
   Hakka

       (Sixian, PFS): yit
       (Hailu, HRS): rhid
       (Meixian, Guangdong): yid5
       (Changting, Changting Pinyin): yi2

   Jin (Wiktionary): ieh4
   Northern Min (KCR): ĭ
   Eastern Min (BUC): ék

           Duration: 1 second.0:01

   Southern Min

       (Hokkien, POJ): it / et
       (Teochew, Peng'im): ig4 / êg4
       (Leizhou, Leizhou Pinyin): ieg4

   Southern Pinghua (Nanning, Jyutping++): at2
   Wu (Wugniu)
       (Northern): 7iq
       (Jinhua): 7iq
   Xiang

       (Changsha, Wiktionary): i6
       (Loudi, Wiktionary): i2
       (Hengyang, Wiktionary): i6

   Mandarin
       (Standard Chinese)+
           Hanyu Pinyin: yī
           Zhuyin: ㄧ
           Tongyong Pinyin: yi
           Wade–Giles: i1
           Yale: yī
           Gwoyeu Romatzyh: i
           Palladius: и (i)
           Sinological IPA (key): /i⁵⁵⁻⁵¹/

Note:

   Subject to retrograde tone sandhi (see Wikipedia for details):
       before fourth tone, pronounced with second tone;
       before any other tone, pronounced with fourth tone;
       when final (or at the end of a multi-syllable word), pronounced with the original tone (first tone);
       when used in a series of numbers, as a part of a number, in dates, or in addresses, pronounced with the original tone (first tone).
   When used between two reduplicated words, it may be pronounced with neutral tone.

       (Chengdu)
           Sichuanese Pinyin: yi2
           Scuanxua Ladinxua Xin Wenz: i
           Sinological IPA (key): /i²¹/
       (Xi'an)
           Guanzhong Pinyin: yǐ
           Sinological IPA (key): /i²¹/
       (Nanjing)
           Nanjing Pinyin: yi̊q
           Nanjing Pinyin (numbered): yiq5
           Sinological IPA (key): /iʔ⁵/
       (Dungan)
           Cyrillic and Wiktionary: йи (yi, I)
           Sinological IPA (key): /i²⁴/

           (Note: Dungan pronunciation is currently experimental and may be inaccurate.)

   Cantonese
       (Standard Cantonese, Guangzhou–Hong Kong–Macau)
           Jyutping: jat1
           Yale: yāt
           Cantonese Pinyin: jat7
           Guangdong Romanization: yed1
           Sinological IPA (key): /jɐt̚⁵/
       (Dongguan, Guancheng)
           Jyutping++: jak7
           Sinological IPA (key): /zɐk̚⁴⁴/
       (Taishanese, Taicheng)
           Wiktionary: yit2
           Sinological IPA (key): /jit̚⁵⁵/
       (Yangjiang Yue, Jiangcheng)
           Jyutping++: jat3
           Sinological IPA (key): /jɐt̚²⁴/
   Gan
       (Nanchang)
           Wiktionary: it6
           Sinological IPA (key): /it̚⁵/
   Hakka
       (Northern Sixian, incl. Miaoli)
           Pha̍k-fa-sṳ: yit
           Hakka Romanization System: idˋ
           Hagfa Pinyim: yid5
           Sinological IPA: /it̚²/
       (Southern Sixian, incl. Neipu)
           Pha̍k-fa-sṳ: yit
           Hakka Romanization System: (r)idˋ
           Hagfa Pinyim: yid5
           Sinological IPA: /(j)it̚²/
       (Hailu, incl. Zhudong)
           Hakka Romanization System: rhid
           Sinological IPA: /ʒit⁵/
       (Meixian)
           Guangdong: yid5
           Sinological IPA: /it̚¹/
       (Changting)
           Changting Pinyin: yi2
           Sinological IPA: /i²⁴/
   Jin
       (Taiyuan)+
           Wiktionary: ieh4
           Sinological IPA (old-style): /iəʔ²/
   Northern Min
       (Jian'ou)
           Kienning Colloquial Romanized: ĭ
           Sinological IPA (key): /i²⁴/
   Eastern Min
       (Fuzhou)
           Bàng-uâ-cê: ék
           Sinological IPA (key): /ɛiʔ²⁴/
   Southern Min
       (Hokkien: Xiamen, Quanzhou, Zhangzhou, Jinjiang, Nan'an, Zhangpu, Changtai, Longyan, General Taiwanese, Penang, Singapore, Philippines)
           Pe̍h-ōe-jī: it
           Tâi-lô: it
           Phofsit Daibuun: id
           IPA (Quanzhou, Jinjiang, Nan'an, Longyan, Philippines): /it̚⁵/
           IPA (Xiamen, Zhangzhou, Zhangpu, Changtai, Taipei, Kaohsiung, Singapore): /it̚³²/
           IPA (Penang): /it̚³/
       (Hokkien: Hui'an)
           Pe̍h-ōe-jī: et
           Tâi-lô: et
           Phofsit Daibuun: ed
           IPA (Hui'an): /et̚⁴/
       (Teochew)
           Peng'im: ig4 / êg4
           Pe̍h-ōe-jī-like: ik / ek
           Sinological IPA (key): /ik̚²/, /ek̚²/
       (Leizhou)
           Leizhou Pinyin: ieg4
           Sinological IPA: /iek̚⁵/
   Southern Pinghua
       (Nanning Pinghua, Tingzi)
           Jyutping++: at2
           Sinological IPA (key): /ət̚³³/
   Wu
       (Northern: Shanghai, Jiading, Songjiang, Chongming, Chuansha, Suzhou, Kunshan, Yixing, Changzhou, Jingjiang, Jiaxing, Tongxiang, Haining, Haiyan, Deqing, Hangzhou, Xiaoshan, Shaoxing, Cixi, Ningbo, Zhoushan)
           Wugniu: 7iq
           MiniDict: ih入
           MiniDict: ieh入
           Wiktionary Romanisation (Shanghai): 4iq
           Sinological IPA (Shanghai): /iɪʔ⁵⁵/
           Sinological IPA (Jiading): /iɪʔ⁵⁵/
           Sinological IPA (Songjiang): /iɪʔ⁴⁴/
           Sinological IPA (Chongming): /ʔiəʔ⁵/
           Sinological IPA (Chuansha): /iɪʔ⁵⁵/
           Sinological IPA (Suzhou): /iɪʔ⁴³/
           Sinological IPA (Kunshan): /iəʔ⁵⁵/
           Sinological IPA (Yixing): /iəʔ⁵⁵/
           Sinological IPA (Changzhou): /iɪʔ⁵⁵/
           Sinological IPA (Jingjiang): /ieʔ⁵⁵/
           Sinological IPA (Jiaxing): /iɪʔ⁵⁵/
           Sinological IPA (Tongxiang): /iəʔ⁵⁵/
           Sinological IPA (Haining): /iəʔ⁵⁵/
           Sinological IPA (Haiyan): /iəʔ⁵⁵/
           Sinological IPA (Deqing): /iəʔ⁵⁵/
           Sinological IPA (Hangzhou): /iəʔ⁵⁵/
           Sinological IPA (Xiaoshan): /iəʔ⁵⁵/
           Sinological IPA (Shaoxing): /ieʔ⁴⁵/
           Sinological IPA (Cixi): /ieʔ⁴⁵/
           Sinological IPA (Ningbo): /iɪʔ⁵⁵/
           Sinological IPA (Zhoushan): /iəʔ⁵⁵/
       (Jinhua)
           Wugniu: 7iq
           Sinological IPA (Jinhua): /iəʔ⁴⁴/
   Xiang
       (Changsha)
           Wiktionary: i6
           Sinological IPA (key): /i²⁴/
       (Loudi)
           Wiktionary: i2
           Sinological IPA (key): /i¹³/
       (Hengyang)
           Wiktionary: i6
           Sinological IPA (key): /i²²/

   Dialectal data

Variety Location edit 一 Mandarin Beijing /i⁵⁵/ Harbin /i⁴⁴/ Tianjin Jinan /i²¹³/ Qingdao /i⁵⁵/ Zhengzhou /i²⁴/ Xi'an /i²¹/ Xining /i⁴⁴/ Yinchuan /i¹³/ Lanzhou /i¹³/ Ürümqi /i²¹³/ Wuhan /i²¹³/ Chengdu /i²¹/ Guiyang /i²¹/ Kunming /i⁴²/ Nanjing /iʔ⁵/ Hefei /ieʔ⁴/ Jin Taiyuan /ieʔ²/ Pingyao Hohhot /iəʔ³⁴/ Wu Shanghai /iɪʔ⁵/ Suzhou /iɪʔ⁴/ Hangzhou /iɪʔ⁵/ Wenzhou /iai³²³/ 白 /i³²³/ 文 Hui Shexian /i³³/ Tunxi /iʔ⁵⁵/ Xiang Changsha /i²⁴/ Xiangtan /i²⁴/ Gan Nanchang /it̚⁵/ Hakka Meixian /it̚¹/ Taoyuan Cantonese Guangzhou /jɐt̚⁵/ Nanning /iət̚⁵/ Hong Kong /iɐt̚⁵/ Min Xiamen (Hokkien) /it̚³²/ /tsit̚⁵/ 俗 Fuzhou (Eastern Min) /eiʔ²³/ /suɔʔ⁴/ 俗 Jian'ou (Northern Min) /i²⁴/ /tsi⁴²/ 俗 Shantou (Teochew) /ik̚²/ Haikou (Hainanese)

   Middle Chinese: 'jit

Rime Character 一 Reading # 1/1 Initial (聲) 影 (34) Final (韻) 質 (48) Tone (調) Checked (Ø) Openness (開合) Open Division (等) III Fanqie 於悉切 Baxter 'jit Reconstructions Zhengzhang Shangfang /ʔiɪt̚/ Pan Wuyun /ʔit̚/ Shao Rongfen /ʔjet̚/ Edwin Pulleyblank /ʔit̚/ Li Rong /ʔiĕt̚/ Wang Li /ĭĕt̚/ Bernhard Karlgren /ʔi̯ĕt̚/ Expected Mandarin Reflex yi Expected Cantonese Reflex jat1

   Old Chinese

       (Baxter–Sagart): /*ʔi[t]/
       (Zhengzhang): /*qliɡ/

Baxter–Sagart system 1.1 (2014) Character 一 Reading # 1/1 Modern Beijing (Pinyin) yī Middle Chinese ‹ ʔjit › Old Chinese /*ʔi[t]/ English one

Notes for Old Chinese notations in the Baxter–Sagart system:

Zhengzhang system (2003) Character 一 Reading # 1/1 No. 14833 Phonetic component 一 Rime group 質 Rime subdivision 2 Corresponding MC rime 一 Old Chinese /*qliɡ/

Geographyinitiative (talk) 05:47, 20 September 2025 (UTC)

This seems to be essentially a rehashing of the argument for and against unified Chinese, just coming at it from the pronunciation angle. The sheer amount of work involved, combined with no clear vision for how everything would work (what exactly should entries look like, how would templates and modules operate, etc.), has been identified as a serious blocker in past discussions.

Do you have any concrete proposals for splitting all of the Chinese entries? Do we even have ISO codes for all varieties? I see what looks like a lot of geographically bound dialectal pronunciation information in the more developed of these entries, and I don't have any clear idea how this would be implemented otherwise.

When it comes to referencing and linking terms, we have {{ltc-l}} and {{och-l}}, which display the pronunciations for the specified varieties of Chinese -- Middle Chinese for {{ltc-l}}, and Old Chinese for {{och-l}}. It looks like {{m}} may already do the same for other varieties, such that {{m|yue|一}} gives us 一 (jat¹) with the expected Cantonese pronunciation.

Side note: @Geographyinitiative, your copy-paste seems to have gone a little funny, as apparently there was whitespace at the start of the lines, causing the Wikicode parser to interpret that all as monospace code. ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:21, 24 September 2025 (UTC)

Abolish WT:ATTEST

Here's a big lie on Wiktionary, right on the front page: "Welcome to the English-language Wiktionary, a collaborative project to produce a free-content multilingual dictionary. It aims to describe all words of all languages using definitions and descriptions in English." [4]

I submit that protologisms and hapax legomenon are part of the set of all words of all languages, i.e. that WT:ATTEST is directly preventing the achievement of the purpose of Wiktionary.

Hence I believe it should be abolished. --Geographyinitiative (talk) 11:44, 20 September 2025 (UTC)

No.

1) Protologisms are just that, protologisms.

2) Hapax legomenon are already allowed for LDL's.

Otherwise this opens the door for anything and everything without any sense of verifiability. WT:ATTEST is literally what gives the project credibility. This is beyond a terrible idea. Vininn126 (talk) 11:48, 20 September 2025 (UTC)

Strong oppose per Vininn. Juwan (talk) 12:14, 20 September 2025 (UTC)

Strong oppose. Makes sense. — Fenakhay ^{(حيطي · مساهماتي)} 12:31, 20 September 2025 (UTC)

I’m not sure what you mean? The quoted message is a Finnish cake recipe. It just happens to use protologisms that are homographs for English words. Nicodene (talk) 12:40, 20 September 2025 (UTC)

@Nicodene WTF? Where are the Finnish cakes? Vealhurl (talk) 21:54, 20 September 2025 (UTC)

Well, you have to make them. Nicodene (talk) 05:18, 21 September 2025 (UTC)

While I don't believe WT:ATTEST should be deleted entirely, I believe it was a mistake to remove the "well-known work" exemption that we used to have. The thousands (millions?) of schoolchildren reading Shakespeare cannot use Wiktionary as a resource to fully understand his works because we lack his protologisms. It would be better to "one use in a widely-read work of literature agreed upon by Wiktionarians and listed at the relevant WT:AXX page". In the WT:AEN list you would put Shakespeare, Austen and such people (but not Joyce). Elitism, some might say - but I'm a pragmatist and I'd love it if we could be a bit more pragmatic about such things as this. This, that and the other (talk) 14:11, 20 September 2025 (UTC)

I have set certain lects as LDL's within Polish, namely Middle Polish and dialects. It has not caused major issues. Vininn126 (talk) 14:14, 20 September 2025 (UTC)

• Oppose DCDuring (talk) 14:31, 20 September 2025 (UTC)

WT:ATTEST has multiple parts. Deleting all of it is going way too far. I agree that "at least three independent instances spanning at least a year" requirement is overly strict in certain cases. E.g. it's proven difficult to find 3 independent cites for Etymology 2 of vagitate (as used by Samuel Beckett), but I think we should have an entry for it nevertheless. But I do not support removing the requirement without replacing it with some other criterion. We should have rules that do not allow people to make a post with one use of a protologism on their personal blog or social media account, and use this to prevent deletion of an entry for the "word" or sense that they just invented. Some examples of alternative criteria that I think could be applied: an exception for uses before a certain date, or before a certain number of years ago (if we don't allow hapaxes/protologisms within the last 100 years, that would be pretty much sufficient to prevent gaming of the system in this manner, while still allowing documentation of historically significant hapaxes before that cutoff) or exceptions for uses in significant works (which is a little harder to apply as a criterion, but can be done as This, that and the other mentioned).--Urszag (talk) 17:43, 20 September 2025 (UTC)

• Samuel Beckett is known to make up words, particularly borrowing between Romance and English, and this uses to be described in the plentiful literature on him. E.g. fr:trasciner from Italian trascinare, which I stumbled upon when reading that exact work, but I did not feel compelled to add it because it was ascertained to be restricted to that place and I was not prone to delve into literary science. Dogmatically, there lacks evidence for an -eme level if there is but one instance. You may still link an apparatus via Template:no entry. Fay Freak (talk) 18:18, 20 September 2025 (UTC)

• Weak oppose I can see in principle where you're going with this, but it's a very extreme measure to just abolish it. Let's try revising so it's better. That would be a more reasonable proposal. —Justin (koavf)❤T☮C☺M☯ 17:52, 20 September 2025 (UTC)

• Oppose “Language” there is exlusive of idiolects. Fay Freak (talk) 18:18, 20 September 2025 (UTC)

• What's wrong with having "a big lie […] on the front page" anyway? We also claim "Community portal […] [is] a page containing everything you wanted to know about Wiktionary" and that's subjective, and it claims "you can edit it [Wiktionary]", which isn't always true - I got blocked 1000 times, then I couldn't edit..... Anyway, joking aside, I suspect Geoggy is either suffering from what I've coined mission statementitis, taking it too literally, or wants this policy out as an excuse to add a Chinese pond no one in English has bothered to write about yet. Regarding the millions of schoolchildren reading Shakespeare [who] cannot use Wiktionary as a resource to fully understand his work, too bad. We can't be everything. Vealhurl (talk) 21:52, 20 September 2025 (UTC)

Oppose this comes off as such a "throw the baby out with the bathwater" type of suggestion — BABR・kurwa? 22:38, 20 September 2025 (UTC)

Strong oppose per Babr and Vininn. Davi6596 (talk) 10:38, 22 September 2025 (UTC)

Strong oppose Wiktionary’s presentation can be quite cluttered already. If anyone can coin a word and add an article about it right away, this would make finding information on pages even harder. Хтосьці (talk) 13:59, 23 September 2025 (UTC)

Funny trolling. Better to revise the slogan. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 20:39, 23 September 2025 (UTC)

Strong oppose No to the protologisms we are flooded with, though we should have a way of handling single instances found in the likes of Shakespeare and Spenser, as any other dictionary does. But not at the expense of letting the pedantic floodgates open: "Well if you include Spenser you also have to include X." --Hiztegilari (talk) 21:41, 23 September 2025 (UTC)

in the same spirit

All¹ words² in³ all⁴ languages⁵

The ordinary-word meaning of this slogan is somewhat misleading. The following notes explain the qualifications:

¹Not every word is included at all, let alone in a meaningful way. Obviously we haven't gotten around to all of them. Attestation requirements exclude many. Due to the narrowness of our contributor base many languages are unrepresented and many specialized contexts are unrepresented, even in English.

²"Word" can include letters, numbers, symbols, abbreviations, proverbs, idiomatic expressions, some non-idiomatic expressions, clitics, affixes.

³Some "words²" could fall between languages. A multi-word expression borrowed from a foreign language could be non-idiomatic in its original language and thereby not includable in that language. It may also only be found in italics or quotation marks in running text in other languages, indicating that authors and editors don't think it has entered the lexicon in that language.

⁴See Vote on Serbo-Croatian.

⁵Translingual is not a language. Many non-words are better characterized as things. Things that are not words are not part of languages.

from my user page. DCDuring (talk) 22:31, 20 September 2025 (UTC)

"Things that are not words are not part of languages", erm... but what about punctuation? 2007GabrielT (talk) 04:34, 22 September 2025 (UTC)

Punctuation is part of a writing system to represent a language like English graphically. I wouldn't say it's part of the English language per se, any more than printer ink is part of a delicious recipe for ragù alla bolognese. —Caoimhin ceallach (talk) 10:34, 22 September 2025 (UTC)

• Well, at least we're talking about something useful here (recipes for Finnish cookies and ragu, not Geoggy's stuff) Vealhurl (talk) 21:30, 23 September 2025 (UTC)

I honestly feel that the alleged creation of chaos by destroying this policy will not be so terrible as you may suppose. I think Wiktionary would thrive in the chaos of bringing truth to the words on the front page. To me, it's like the argument between feudalism and democracy. --Geographyinitiative (talk) 23:49, 23 September 2025 (UTC)

Ok. Count the votes. Vininn126 (talk) 23:54, 23 September 2025 (UTC)

All words are created equal. There is not an elite class of words with special status as the real words because they have a certain arbitrary pedigree of usage examples. --Geographyinitiative (talk) 23:58, 23 September 2025 (UTC)

This is true, but the words that matter to most people will be buried under heaps of ones that don't if any word qualifies without policing their entry. Sometimes, I'm with you, the criteria are a pain, and some people prefer to delete more than others, which can be tough if your work is being deleted. However, if it was deleted, that's only because it was something so obscure as to no longer qualify as relevant or important at all... after all, we aren't here to literally just add every word - we're here to make a dictionary that everyone can get use out of. "All words" is just our definition for completeness - the standard we wish we could meet in the long run. Our job will never end if everything ever said or written could qualify, I think. Kiril kovachev (talk・contribs) 00:30, 24 September 2025 (UTC)

• Err, our job will never end anyway. Thaat's kinda the point Vealhurl (talk) 05:43, 24 September 2025 (UTC)

It's kind of insane to call the requirement of usage examples an "arbitrary pedigree". Good scientific research is based on what has evidence supporting it. Something could theoretically could be a word, and to prove it, you need some type of evidence. Trying to get rid of evidence would kill any credibility the dictionary has. I'm not going to engage in this with you further. Vininn126 (talk) 09:27, 24 September 2025 (UTC)

I understand your views, and I appreciate your feedback. I'm just saying that whenever you all are ready to open the floodgates, whenever Wiktionary has the self-confidence to truly include all words, I'm on board with that. Quotations and citations are important, but I feel WT:ATTEST elevates them over the words themselves. I want to start to try to push to open your minds to like REALLY including EVERYTHING, actually all of it. I know so many words that have like only a few mentions, not even real use that I can find. Or words used only by one author, but repeatedly and authoritatively. Maybe I misstated my case above. Anyway, these other words outside WT:ATTEST are still words, and I feel sad to miss out on including everything. --Geographyinitiative (talk) 10:17, 24 September 2025 (UTC)

Just saying, IMO it's very sad and contributes negative value to the dictionary that some editors feel the most important words to add are the most obscure ones. That's why we have silly entries like Tejostadt and Tajostadt when we're missing a ton of commonly used, important German words. Benwing2 (talk) 19:48, 24 September 2025 (UTC)

• Failed the overambitious proposal. But please keep the Chinese places coming, Joggy - I want all villages in all China. Vealhurl (talk) 21:18, 24 September 2025 (UTC)

That's just the problem, I literally can't do all villages in all China. The words exist, but WT:ATTEST is like '3 cItEs oR iT dIdN't hApPeN bRo'. There is a vast array of abstruse terminology out there just waiting for Wiktionary to reform. Geographyinitiative (talk) 02:04, 25 September 2025 (UTC)

Another avenue is to have a separate attestation policy specifically for toponyms. --Ssvb (talk) 03:21, 25 September 2025 (UTC)

Effectively we already do; really obscure toponyms that satisfy the criteria listed under WT:CFI#Place names don't normally come with any attestations. I think @Geographyinitiative is the only editor who actually tries to add citations for place names. Geographyinitiative seems to take the 3-attestation criteria very literally and creates citations even for large cities and such that clearly would pass CFI; IMO their time would be better spent finding citations for more obscure places. I think if we actually enforced the attestation criteria for place names, a lot of unincorporated communities that have a very nebulous existence and are included only because they are found in Wikipedia would disappear, partly satisfying the complaints of people like @Fish bowl. Benwing2 (talk) 03:31, 25 September 2025 (UTC)

(to be clear, i'm neutral on (for example) the inclusion of American unincorporated communities under English, and extremely skeptical of the inclusion of American unincorporated communities under Japanese, etc.: look at the list at Washington#Proper noun and tell me with a straight face that this these can all be attested 3 times in Japanese; it all comes down to attestation —Fish bowl (talk) 03:51, 25 September 2025 (UTC))

Yes. I bet a surprising number of them can in fact be attested sufficiently in Japanese, but not all. For example, all 31 counties in the US named Washington are listed under the Japanese Wikipedia disambiguation article ワシントン郡 and each has an article on it, each with several footnotes. Granted, the footnotes are in English, but I would not be surprised to find that there exist gazetteers and such in Japanese that include blurbs on all sorts of towns, and random mentions of a lot of these places in newspaper articles about the US. Benwing2 (talk) 04:03, 25 September 2025 (UTC)

@Benwing2: The wording of WT:CFI#Place names could be improved to ensure that there's no possible misunderstanding about what "if they fulfill attestation requirements" means in that context. Does it refer to the "3 quotations" requirement mentioned elsewhere and then imposes additional restrictions listed below? Or does the list below fully represent the new "attestation requirements" that are specific to toponyms and supersede the usual attestation requirements? --Ssvb (talk) 10:46, 25 September 2025 (UTC)

• OR Geoggy, find a different webbsite for your damn villages! Vealhurl (talk) 10:58, 25 September 2025 (UTC)

  I'm not Geoggy. The wording of the policy could be improved nevertheless for better clarity.

  But you are right, there are dedicated catalogues and databases for toponyms elsewhere. For Belarus and the Belarusian language, more or less complete list of toponyms can be found in {{R:be:NNP}}. I surely have no intention to replicate all of its entries in Wiktionary and see no purpose of doing that. And of course, Wiktionary still needs toponyms that are frequently mentioned in the literary works or news. --Ssvb (talk) 11:14, 25 September 2025 (UTC)

  All¹ Chinese villages² in³ all⁴ languages⁵ Jberkel 11:25, 25 September 2025 (UTC)

Catalogues and databases do not reach the level of detail that we can reach here. GEONet Names Server is definitely big af, but the details for each variant of each word and pronunciations (bad and good) etc are not as high quality as here; I find errors in the GEONet Names Server when I go there. The wheel needs to be reinvented or at least reimagined. --Geographyinitiative (talk) 11:31, 25 September 2025 (UTC)

On a more serious note, wouldn't Wikidata be a potential candidate for an all-Chinese-villages-encompassing-database? Jberkel 11:46, 25 September 2025 (UTC)

What "new attestation requirements"? WT:PLACE gives a list of place names that should be included if they fulfill "attestation requirements" (of course meaning WT:ATTEST); I am not sure how you would read this list as anything about attestation, since it does not mention anything related to attestation. Later, the policy says All place names not listed above shall be included if they have three citations of figurative use that fulfill attestation requirements., which is accurately worded as an additional restriction on the usual WT:ATTEST (again referred to as "attestation requirements"). What "attestation requirements" means does not change between one mention and the other, as intended. What is the ambiguity here and how do you propose it be resolved? — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 14:41, 25 September 2025 (UTC)

@Polomo: There are at least two possible interpretations. One is precisely as you described. But another one is that the colon at the end of "... if they fulfill attestation requirements:" means that the list of these mentioned attestation requirements follows below. And, for example, Human settlements: cities, towns, villages, etc. as an attestation requirement means that any human settlement is eligible for Wiktionary. Instead of 3 citations, a proof that something is really a human settlement can be a photo of a road sign with that settlement name. Or some official document or bill with that settlement name in the address field, etc.

Please check the Benwing2's comment, I'm interpreting it as a statement that Geographyinitiative is unreasonably adding citations, contrary to what is supposedly needed per WT:PLACE and/or contrary to what the other editors are doing. So it seems to me that there are differences between your and Benwing2's understanding of the policy text. Hence I think that the policy text is ambiguous in its current form. --Ssvb (talk) 16:22, 25 September 2025 (UTC)

I think, rather, you’re misreading Benwing’s comment. It’s not that these terms do not need attestations, but rather that no one tends to bring them to RfV. We do not require three attestations for every new page creation; it’s just a necessity if the term is brought to RfV. I do not see the ambiguity you speak of, seeing as “being a human settlement” is indubitably not an attestation of any kind (since it is not a confirmation of anything, just a reported characteristic), especially not sense 4. Nevertheless, the policy wording can be changed, if it’s not for the worse, if you have any ideas. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 17:34, 25 September 2025 (UTC)

@Polomo is correct. WT:PLACE effectively imposes additional requirements on things not in the list of "allowed toponym types" (e.g. roads, dams, and the like), specifically that they be used figuratively three times, similarly to WT:BRAND. The list of allowed toponym types is there to exclude those types of toponyms from the figurative requirement. For example, you can surely find three uses of Country Club Rd. in Tucson, AZ (or nearly any similar street in any US city) in the local newspaper and other local sources, but that doesn't mean we should include this road in Wiktionary unless it's used figuratively in three separate instances (which seems unlikely as there's nothing special about this road, it's just one of an innumerable number of arterial roads in a mid-sized US city). Benwing2 (talk) 19:05, 25 September 2025 (UTC)

Pashto transliteration proposal

(Notifying Vahagn Petrosyan, Dijan, AryamanA, Saam-andar):
Our transliteration of Pashto seems all over the place. Not to mention, some entries use seemingly extremely dialect specific transcriptions. I am proposing the following for our Pashto transliteration (which is based on more conservative/dialectally broad pronunciation):

Proposal

• The same as Wiktionary:Persian transliteration/Classical with the following changes:
• All forms of yāʔ (ي ى) remain transcribed as they are already. , but ـي = ī not i seems preference is to use no vowel length in transcription
• The sibilant fricative series is transcribed as s, š, ṣ̌ for س، ش، ښ; and z, ž, ẓ̌ for ز، ژ، ږ
• the vowel ◌ٙ /ə/ is always written ə not ë (which is also more common academically)
• The remaining retroflexes are transcribed as in Wiktionary:Urdu transliteration and are the same as in Urdu and Punjabi: ṛ, ṭ, ḍ, ṇ (which is already the case)
• The dialectal nasal vowel ں is transcribed as in Urdu, ـاں = ā̃
• The affricate series ج، ځ، چ، څ is c, č, j, ǰ or ć, č, ź, j whatever has more support (the first is used by Iranica and Johnny Cheung, but the latter is more cohesive with other Iranian languages).

— BABR・kurwa? 17:46, 20 September 2025 (UTC)

there was no Pashto ping group (so I pinged the ira group), pinging others who may be interested @Samiollah1357, @Rizozoda34, @OblivionKhorasan. — BABR・kurwa? 17:48, 20 September 2025 (UTC)

Standard Pashto orthography does not distinguish length at all, always writing vowels as long even in cases where etymologically the modern vowel goes back to a short one. The orthophony is to always pronounce them as short. I think it would be somewhat misleading to transcribe e.g. اوښ as ūṣ̌ while it is pronounced /uʃ/ (not /uːʃ/) and equivalent to the first element of Persian اشتر (uštur).

Also, what leads you to prefer ṣ̌ to the current standard of x̌ for transliterating ښ? Rizozoda34 (talk) 21:17, 20 September 2025 (UTC)

ṣ̌ is more common academically + the letter ښ was added specifically to represent that sound but some Pashto dialects lost it. I think it's better to represent pronunciation mergers in the pronunciation section e.g.

Pronunciation

• (Southern) IPA^(key): /pəʂt̪o/
• (Northern) IPA^(key): /pəxt̪o/ (as if spelt پختو)
  

I would be fine with removing vowel length altogether and transcribing short and long vowels the same way. Some claim that vowel length exists in Perso-Arabic loans, but that could also be considered codeswitching. — BABR・kurwa? 22:12, 20 September 2025 (UTC)

That seems reasonable to me. Rizozoda34 (talk) 00:10, 21 September 2025 (UTC)

I prefer â for /ɑ/ because it's not a length distinction, but I guess most sources use ā.

Support the latter on the affricates.

Is stress going to get marked? Saam-andar (talk) 12:32, 21 September 2025 (UTC)

If we are not marking orthographic length, I'd rather use â as well. It's weird to have a single "long" vowel. As for stress, I'm not sure, that feels like something for the pronunciation section but its been common practice to put it in the header so 🤷 — BABR・kurwa? 23:07, 21 September 2025 (UTC)

for the record, I am still a bit conflicted regarding how to transcribe vowels. According to MacKenzie, Pashto used to have vowel length but nearly every dialect today has lost it. So I'm unsure if vowels should be marked as long or short, if we do, this system would do so purely based on the written form. — BABR・kurwa? 17:52, 20 September 2025 (UTC)

Proposal to Reevaluate Frankish’s Status as Etymology-Only Variant

Hello everyone, I’ve been following the ongoing discussions about the classification of Frankish on Wiktionary, particularly the decision from the 2020 vote to treat it as an etymology-only variant of Proto-West Germanic. Given the historical linguistic evidence supporting Frankish as a distinct West Germanic language, I believe this policy deserves reevaluation. Treating Frankish merely as a variant obscures its unique features and complicates etymological work involving Frankish-origin words. I propose starting a new formal discussion or vote to consider restoring Frankish as a full language entry with its own namespace. This would better reflect scholarly consensus and benefit Wiktionary’s accuracy and usability. I welcome thoughts from everyone — especially those with expertise in Germanic linguistics or experience in Wiktionary’s language classification policies. Thanks for considering this. — An interested contributor 38.43.34.225 04:09, 22 September 2025 (UTC)

By the way, these talks are in the etymology scriptorium under bring back Frankish. 38.43.34.225 04:15, 22 September 2025 (UTC)

As with many things, if this is something you would like to see changed, the best approach would be presenting a case with citations and specific examples.

ETA: preferably without AI-generated rubbish this time. Nicodene (talk) 08:15, 24 September 2025 (UTC)

Theknightwho's idea for affix entries

Since from the last discussions it seems most English editors support allowing and keeping all suffix inflection entries (I disagree with this but have to cope), I'm bringing up @Theknightwho's idea of treating affixes as the parts of speech of the words they form in terms of entry layout.

He said on the Discord server, "Tbh they [affixes] should be subdivided by part of speech. It's a little tricky, because we bundle affixes in as parts of speech for Wiktionary's system, but distinctions like prefix/suffix/infix etc. are all purely morphological (i.e. they're distinguished by the position of the affix), whereas parts of speech are distinguished by semantic function. In theory, a language could have prefixes that inflect like a verb.
In terms of categorisation, they should go under 'noun-forming suffixes', or something like that, or maybe "nominal suffixes" since it's a bit weird to say 'noun-forming suffix' when it's from another noun. Actually, hmm - it's a little more complicated. There are two separate concepts here:

1. Affix 'systems', where a single affix is the representative lemma (e.g. -sis and -ses).
2. Specific affixes, such as the individual inflections for a Latin verb.

Both are useful, and I suspect many 'non-lemma' affixes could have quite interesting etymology sections."

I don't oppose this, but my alternative idea is to keep the Suffix subheading but use noun templates for nominal suffixes, verb templates for verbal suffixes, adjective templates for adjectival suffixes, etc. like in -ial and Portuguese entries, for example.
Anyway, I'd like to hear y'all's thoughts on these suggestions. Davi6596 (talk) 11:00, 22 September 2025 (UTC)

@Davi6596 This feature already works in exactly this fashion for Spanish, Portuguese and several other languages. For example, -inho is defined using {{pt-noun}} not anything like {{pt-suffix}} or similar. Benwing2 (talk) 03:18, 23 September 2025 (UTC)

@Benwing2 Yes, I know this. But it isn't standard in English entries: the only one I know it uses the feature is -ial. Davi6596 (talk) 10:11, 23 September 2025 (UTC)

@Davi6596 I see, you're saying the support is there but not in regular use. That's probably because I went through and converted all the Portuguese, Spanish and Italian entries to use this format and deleted {{pt-suffix}} etc., but this hasn't been done for English. Benwing2 (talk) 02:09, 24 September 2025 (UTC)

@Benwing2 Exactly. I think you can do it for English since there's majority support for English suffix inflection entries to exist or remain (or, in an alt analysis, little opposition to it) and the conversion wouldn't put Theknightwho's proposal in jeopardy. Davi6596 (talk) 11:33, 24 September 2025 (UTC)

PIE "Unsorted formations"

Throughout the entries for reconstructed PIE terms, there currently appear to be several different (and seemingly uncoordinated) styles in which the sub-heading “Unsorted formations” has been formatted for inclusion:

1) Attached to the end of the bullet-point list of Derived terms, with colon [ • Unsorted formations: ], cf. https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/geh%E2%82%82w-

2) Detached beneath the list of Derived terms, in normal type, with colon [ Unsorted formations: ], cf. https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/%28s%29keng-

3) Detached beneath the list of Derived terms, in bold type, without colon [ Unsorted formations ], cf. https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/d%CA%B0eg%CA%B7%CA%B0-

Would reformatting the heading “Unsorted formations” cause any problems within other entries or templates (desctrees and the like)? Would there be any objection to standardising the formatting? And would there be a preference as to how it should be formatted? Key of Now (talk) 21:11, 22 September 2025 (UTC)

@Key of Now I don't work on PIE reconstructed terms but I think the answer to all three of your questions is no. Benwing2 (talk) 18:04, 24 September 2025 (UTC)

questionable modules and templates from User:Bytekast

1. This user created {{nrp}} which displays "non-rhotic pronunciations" but is supposed to be used for "non-rhotic pronunciation spellings" which is something very different. This is created as a general language-agnostic template which seems wrong; I am going to RFD this and propose using a simple category addition to track such cases in English. On top of this, {{nrp}} is a really bad shortcut; too short and opaque.
2. This user created Module:nickname with template {{nickname of individual}} and shortcuts {{nickname}}, {{nfi}} and {{noi}}. These last two are equally bad just like {{nrp}}. On top of this, I'm not sure this belongs at all; generally individuals don't qualify per WT:CFI, and the nicknames cited for use with this template (e.g. "Wizard of Menlo Park") remind of me of the "Dont Tread on Me" and "Hix Nix Stix Pix" discussions, about similar things that are more encyclopedic than dictionary material and may not pass CFI either.

@Bytekast, I appreciate your youthful energy but you need to learn what is appropriate for Wiktionary and what's not, and how things are normally done around here, before trying to make basic changes to the system. It would have been better, for example, for you to have proposed your new templates and modules *BEFORE* implementing them, to get feedback concerning (a) are they needed at all, (b) what's the best way to implement them. Benwing2 (talk) 07:48, 23 September 2025 (UTC)

I just found another bad module/template and redirect: {{tus}} -> {{template-usage}}. This is absolutely not needed; we already have {{temp demo}} and {{demo}} for precisely this purpose. I am going to RFD this as redundant. Benwing2 (talk) 07:50, 23 September 2025 (UTC)

There's also {{fq}} -> {{forum-quote}}; {{snowclone}} with non-working (deleted?) shortcuts; {{sqb}} -> {{brackets}}; etc. Benwing2 (talk) 07:54, 23 September 2025 (UTC)

• {{forum-quote}} is for formatting mentions of replies taken out of forums, for quotations, as a way to avoid writing block-quote tags and "<name> says" (or whatever, depending on how each forum fornats mentions). See fairy nuff, Fairy Nuff and fairy 'nuff for some good practical examples of its usage. Same as for {{non-rhotic}}: for convenience, standard formatting, and abiding to the DRY principle.
• {{snowclone}} is because I saw other templates being used for Appendix: entries, but the snowclones page lacked one and I thought it'd be useful. For the deleted shortcuts, see again this and this.
• {{sqb}}: That's for conveniently putting text in square brackets inside wikilinks, among other use cases (see examples in the doc). Completely opposite to the case of {{template-usage}} (see below), I deem that template to be actually superior to a similar, existing one, {{sqbrace}}, of which I had foreknowledge (unlike {{temp demo}} and {{demo}}): it is thorougher, has more functionality and more use cases, and the module looks and works great—it is not a mess like Module:templusa. So I would say: delete it, and keep sqb, then improve upon it even more.
  

Bytekast (talk) 14:29, 23 September 2025 (UTC)

Oh! I just saw that template. I obviously did not know of its existence at the time! (Had I, it wouldn't have been created.) So, yes, I wound up wasting much time, with something worse, by not consulting the community beforehand. I hate that module I made; I was going to improve its code in the near future, but yes, I also think it's better if we delete it and its template altogether: it's hopeless and that existing template is so much simpler and superior. But it seems to lack usage! Let's substitute it for my template wherever I have snuck it in (to wit, in the documentations of {{para}}, {{temp}}, {{sqb}}, etc.). Bytekast (talk) 13:50, 23 September 2025 (UTC)

All right. Allow me to address each issue individually and per reply.

{{nrp}} ({{non-rhotic}})

This was supposed to be a shortcut for the "Non-rhotic pronunciation of <x>" that I had seen repeated on a few etymologies (such as brudda's; see more here). Upon seeing them, I thought, "Well, templates are good to avoid needless repetition, and we could make one with a category, out of these." It helps "unify" those etymologies under a single format, so that they all display the same instead of "nonrhotic pronunciation" in one and "non-rhotic pronunciation spelling" in another, for example.

I understood, in my limited experience as a Wiktionarian, such to be the raisons d'être of templates: convenience, decreasing repetition and redundancy, cross-entry standardization. (But of course, we can't create or have any under that pretense: it must be useful and have a good motivation.)

The nrp shortcut, in my view, abbreviates well the "non-rhotic pronunciation" text the template produces. However, surely, you could change the text to something more detailed and wordbookly, like "Non-rhotic pronunciation spelling of" or "Represents a non-rhotic pronunciation of"; no need to delete the whole template for that!

{{nickname of individual}} and its aliases and module.

This was born out of this idea (which see), by @Juwan. I found it interesting, agreed with it, and thought, "Well, why not‽ There is a category with many entries, and a perceived need for maintenance." Should we get rid of those entries too? And about "Wizard of Menlo Park": it is just documented as a hypothetical example, clearly not needing (or asking for) such an entry to be created (the community would change it to a better example one day if deemed necessary). Alongside it are real-entry examples.

@Benwing2, I appreciate your feedback. Indeed I see now how it would have been better to suggest and propose, rather than wasting my time writing something that might turn out to be uncalled-for/improper, or even unoptimized/badly coded. I did propose a template once, but only once unfortunately; I shall do more of that in the future, rather than simply and immediately creating new ones as I see fit.

Bytekast (talk) 13:36, 23 September 2025 (UTC)

@Bytekast Thanks for your courteous reply and apologies if my initial post was a bit impolite. Here are my thoughts about the templates you've created:

1. Keep in mind we already have {{tq}} = {{talk quote inline}} and {{tqb}} = {{talk quote block}} for quoting from other postings; given that, do we need {{forum-quote}}?
2. As for {{non-rhotic}}, IMO it needs some rethinking. The category it generates is Category:English non-rhotic pronunciations, whereas these are actually pronunciation spellings, which is different. Also, this is more or less an English-only phenomenon, so I'm not sure we need a general etymology template for this. Potentially this could be handled by introducing a non-rhotic spelling label, which displays as non-rhotic when used in conjunction with {{pronunciation spelling of}}, so we get non-rhotic pronunciation spelling of in the definition, with appropriate linking and categorization handled by the label.
3. As for {{nickname}}, I do see its point now, but I would suggest calling it simply {{nickname}} and deleting all the other aliases.
4. As for {{sqb}}, I think this should stay and replace {{sqbrace}} (which is misnamed since they are brackets not braces).
5. As for {{snowclone}}, I'm not sure it's going to be used enough for there to be a need for a template, although I suppose it doesn't hurt.

Benwing2 (talk) 02:46, 24 September 2025 (UTC)

Don't mention it! I didn't find your topic impolite; I saw it rather as a big misunderstanding, plus a little wariness of my being relatively new to the wikicommunity… and to (adult) life in general.

Anyway, find below my replies to your five thoughts.

1. Not the same purpose:

1.1. {{forum-quote}} is for use in the Main namespace, in example quotations for senses, to easily insert a mention as seen in a forum reply;

1.2. {{tq}} and {{tqb}} belong in the Talk namespace and are for quoting replies and topics, not out-of-wiki forum replies.

2. I seriously agree with you on this.

2.1. I have also felt that when editing some pages to include {{non-rhotic}}. I would think, "Um, this should be on the definition line, not here." The template's output seems to be more fitting for that.

2.2. I disagree that it be English-only whatsoever: other languages have such pronunciations spellings too (see the case of Brazilian Portuguese I wrote about here); English has more of them simply because it is a most freesome tongue, more tolerant of internal variation than other tongues.

2.3. I really like that idea related to {{pronunciation spelling of}}, except for the label part. I'd rather have it as a parameter of that template instead of {{lb}}. This way, it will be displayed directly on the definition line rather than within the labels' parentheses. If we implement that, then yes, we might as well get rid of the {{non-rhotic}} template (after editing the entries it is used in, deleting their etymology sections [or replacing those with the likes of See [[<standard form>]]., e.g. See [[brother#Etymology|brother]]. for brudda]).

3. Good one again! The name {{nickname of individual}} never made sense, besides being too long, because the template covers also companies and inventions. Let us leave only {{nickname}} — same name as the module, for consistency —, and {{nick}} as the sole shortcut.

4. Perfect.

5. Well, have you seen the many redlinks in Appendix:Snowclones? My template can aid with easy linking between themselves, and between them and regular entries. No need to type Appendix:Snowclones/ every time! As long as it is available whenever needed, it won't matter if it will be much less used than other Appendix templates (like {{glossary}}).

(By the way, what will we do with its shortcuts? What about this? How can we wait for community consensus if no one else has discussed there since then? Can't we just cancel and archive that topic of mine, and let the shortcuts simply exist? It has been almost two months.

Bytekast (talk) 18:32, 26 September 2025 (UTC)

Template:etymon and categories

@Saph, Ioaxxere: Lately (as a newly added feature?) Template:etymon has been adding words to categories based on the suffixes they were derived with. I'm not inherently opposed to this, but there's a problem because of the conflict between the naming philosophy of Template:etymon (with its requirement that every entry have its own required "etymology ID", separate from the entry name itself) and the naming philosophy of existing Wiktionary etymology categories, which only use additional clarifying text when actually needed to disambiguate. So for example, this has created categories such as Category:Latin terms suffixed with -ius (suffix forming adjectives), which is redundant to Category:Latin terms suffixed with -ius. I agree with the position that the id should not be required in etymon (I find it often unintuitive and so a pain to check what terms/affixes have been assigned which ids), but if it is required, there needs to be some way of dealing with this correctly so it doesn't disrupt existing categorization systems. Urszag (talk) 01:45, 24 September 2025 (UTC)

@Urszag {{etymon}} has recently been changed to not require an ID when there's a single Etymology section, something I heartily agree with; the documentation hasn't caught up yet. @Fenakhay did a bunch of refactoring of the {{etymon}} support. @Fenakhay has changed things so that {{etymon}} categorizes a lot less based on the ID than it used to, but it still categorizes in certain circumstances (that I'm not completely sure of). I personally think {{etymon}} should never categorize based on the ID, because the ID's are often meaningless, garbagey or inconsistent with other ID's, producing things like Category:Sanskrit terms suffixed with -ल (suffix). But maybe there's a way to put the ID-based categories generated by {{etymon}} in a cleanup category or something so that (theoretically at least) someone will clean up the bad ones. Benwing2 (talk) 02:26, 24 September 2025 (UTC)

if {{etymon}} is never going to categorize suffixes by ID, then it can never be used as a proper {{af}} replacement, because there are legitimate uses for suffix IDs and individual categories.

As a sidenote, I'd like to again express my indignation at editors who force {{etymon}} down everyone's throats by converting existing etymologies to use it no matter the language, despite there being no community consensus to do so. Even the core features of {{etymon}} keep changing regularly, so it's clearly not ready for prime time. It's completely unreasonable to expect everyone else to learn to use the template (and update their tooling, if any, to support it) when the very basics of its functionality keep changing. — SURJECTION ^{/ T / C / L /} 11:46, 24 September 2025 (UTC)

I've never seen the point of {{etymon}} at all. It doesn't do anything that handwritten etymology sections don't already do better. And last time I tried, I couldn't get it to assign terms to the correct sense IDs of PIE roots (e.g. CAT:Latin terms derived from the Proto-Indo-European root *men- (stand out) vs. CAT:Latin terms derived from the Proto-Indo-European root *men- (stay) vs. CAT:Latin terms derived from the Proto-Indo-European root *men- (think)). Maybe that works now with the most recent change, but I'm too sick of fighting against {{etymon}} every time I encounter it to attempt to find out. —Mahāgaja · talk 08:17, 25 September 2025 (UTC)

@Urszag: just so you know, {{etymon}} has been comprehensively overhauled and rewritten by User:Fenakhay (as mentioned above) and at this point I am not sure how things work, although Fenakhay has been working on the docs. For what it's worth I support your position: add an ID to the category name only if necessary to disambiguate. @Surjection I'm not sure where your comment comes from since this is the first time that a "core feature" has been changed, although admittedly the first-ever version (which existed entirely within my user page) used : as the separator, rather than >, and in fact now it seems like we've gone back to that. @Mahagaja the basic purpose of {{etymon}} is to keep different entries in sync in terms of etymology, so it's impossible for C to say A -> B -> C while B says D -> B. The reason this is beneficial is that if an editor modifies a fundamental PIE etymology, the change can *immediately* take effect across every single descendant (even if you think etymology trees are stupid, I'm sure you feel positively about X terms derived from Y categories at least). Ioaxxere (talk) 08:39, 25 September 2025 (UTC)

Another look at Category:Requests for unblock

This was deleted in 2023 after failing an RfD. The arguments presented there were that the category did not serve a purpose, seeing as no one monitored it and it was filled with already resolved requests (which @J3133 cleaned up during the discussion). I, on the other hand, believe it certainly could be useful, since, as @This, that and the other pointed out then, this project doesn't really have good methods for getting the attention of admins when this is required. It would be much easier for uninvolved admins to give their opinion if they didn't have to be summoned there by the blocking admin, who I believe would usually be the first to see an unblocking request. Compare #Reviewing blocks by another administrator.

I propose the categorization parameter be added back to {{unblock}}, and that the category be recreated and cleaned up for future use; or perhaps we could rename it something more generic, like [[Category:Requests for admin attention]], to include other cases, though a category with that name could well be misused. I wonder if there’s a way to get notifications about additions to a category via a web feed.

Ping other users in the RfD: @Metaknowledge, Fish bowl, Vealhurl, SnowyCinema, Fytcha, Chuck Entz, -sche, Ultimateria — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 23:09, 25 September 2025 (UTC)

Support reestablishing Cat:Requests for unblock for now. We can also think about a process for allowing users to request edits to protected pages, but a general "Requests for admin attention" category does seem like it would attract abuse. As for monitoring the unblock category, I'm sure a Discord bot could be set up (not that I'm on Discord). This, that and the other (talk) 23:49, 25 September 2025 (UTC)

Stripping nasal diacritic from Albanian page titles

There are currently 56 entries containing the circumflex to indicate vowel nasalisation, of which 29 were made by @Chihunglu83 on 28 February 2025, 8 by @FierakuiVërtet, 4 by @Torvalu4, 2 by me and the remaining by now inactive editors. My proposal is to get rid of the circumflex in page titles and only use it in {{head|head=}} and links, from which it would be stripped automatically by the modules, as is already done for the acute accent. Although it isn't hard to find the circumflex used in older books, I can't think of places where one would use it now in running text. For entries which would need to contain both, we could duplicate the headword, e.g.

{{sq-noun|m|gjunjë|gjuri|gjunjët}}
{{tlb|sq|Gheg}} {{sq-noun|m|head=gjû|gjûj|gjûni|gjûjt}}

gju m (plural gjunjë, definite gjuri, definite plural gjunjët) (Gheg) gjû m (plural gjûj, definite gjûni, definite plural gjûjt)

If the headwords do not coincide, e.g. zë and zâ each headword would be separated kept under its respective page. CC: @Mondiad2, SairiRM, and anyone else I'm sorry I forgot. Catonif (talk) 12:50, 26 September 2025 (UTC)

I am fine with that. Chihunglu83 (talk) 14:29, 26 September 2025 (UTC)

I also don't see any problem with that. A side note, what do you think about omitting the circumflex when the vowel is followed by a nasal consonant, for example gjuni, zani, kamba instead og gjûni, zâni, kâmba, letting the consonant indicate the nasalisation of the vowe?. I have read this proposal in Martin Camaj's grammar, but I don't know which was the more common standard for Gheg orthography. FierakuiVërtet (talk) 18:43, 27 September 2025 (UTC)

Thank you for making me notice that! :) I was hesitant to propose this as I wondered whether there existed cases where a non-nasal vowel could be followed by a nasal consonant. Traditional Gheg orthography I've seen has a tendency to place the circumflex whenever the vowel is nasal, but it's rarely consistent. The Franciscan printing house of Shkodër, which printed Kanuni i Lekë Dukagjinit (1933), Lahuta e Malcís (1937), has jânë, dhânë, zêmër, etc. but it spells kanû and kanuni as such almost regularly, meanwhile trim is seemingly always diacriticless, and dhima ~ dhîma (= standard dhemba) oscillates. The periodical Agimi (1919–1922), also in Shkodër, on the other hand, indeed seems to put the circumflex only when no nasal consonant follows, although exceptions can still be found (e.g. fêmnat). If there is no phonetic reason behind the inconsistency as Camaj leads to believe then I wholeheartedly agree to follow his approach to avoid giving others the confusion I had. Catonif (talk) 09:01, 28 September 2025 (UTC)

Organization of FWOTD nominations

Currently, WT:FWOTDN is sorted by continent. This gets a little tricky with languages that span different continents. I propose we sort nominations based on language family (or some kind of grouping based on language families). A couple of editors have voiced support for something like and have suggested the language family groupings be organized somewhat similarly to how the Discord channels are organized (e.g., major Indo-European branches are split up). If there aren't strong reasons, I will eventually reorganize the nominations to this proposed new format. (@Fenakhay, Polomo47, Svartava, Sgconlaw) — justin(r)leung _{{ (t...) | c=› }} 17:58, 26 September 2025 (UTC)

(Oops, @Polomo.) — justin(r)leung _{{ (t...) | c=› }} 17:59, 26 September 2025 (UTC)

Despite not being that involved, I want to express Support for the option that will help y'all. Vininn126 (talk) 18:00, 26 September 2025 (UTC)

I haven’t contributed to FWotD in a while, so I’m behind whatever you guys think will be better (like Vininn, lol). This seems like it could be a good change. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 22:58, 26 September 2025 (UTC)

I don't use Discord. Let's split FWOTD based on major language families or along the lines of the RFV forums, putting the nominations for each category on subpages. Vox Sciurorum (talk) 18:44, 27 September 2025 (UTC)

Support, makes sense. – Svārtava (tɕ) 17:56, 2 October 2025 (UTC)

Criticism of Wiktionary by Teflpedia

I just found this relatively obscure wiki called Teflpedia that's about teaching English as a foreign language. Apparently they have a page about us with some criticism. Any thoughts? HyperAnd (talk) 02:32, 27 September 2025 (UTC)

That page isn't even readable if you don't have an account; it just redirects you back to the main page. That should indicate the general quality of the site. Benwing2 (talk) 04:17, 27 September 2025 (UTC)

It doesn't for me? HyperAnd (talk) 04:19, 27 September 2025 (UTC)

Did you create an account? Benwing2 (talk) 04:23, 27 September 2025 (UTC)

No HyperAnd (talk) 04:25, 27 September 2025 (UTC)

Well, you will have to paste in whatever it says, because I can't access it. Benwing2 (talk) 04:27, 27 September 2025 (UTC)

Okay, here's the full text (under CC BY-SA 3.0 by various contributors in edit history of https://teflpedia.com/Wiktionary, though it's only User:Duncan as of writing this):

"Wiktionary (/wɪkʃənəri/) is a wiki-based dictionary, a sister project to Wikipedia, run by the Wikimedia Foundation.[5]

It is available in English and other languages, but each version also accepts both L1 and foreign entries.

The English entry has about 6 million content pages as of 2019[6], which is more than the English Wikipedia. Note that this is total different spellings in all their forms, rather than headwords.

Unfortunately, Wiktionary suffers from poorly structured data, because it conflates homographic spellings together into single entries, sometimes across multiple languages. For example, the entry on love[7] contains a main entry for English, plus additional entries for Czech, Danish, Dutch, French, Friulian, Inari Sami, Middle Dutch, Middle English, Norwegian Bokmål, Norwegian Nynorsk, Romani and Serbo-Croatian(!). Meanwhile, the entry for “like”[8] has no less than 9 entries in English, covering its use as a noun, 2 different verbs, an adjective, an adverb, a conjunction, and a preposition or particle(!)

The hall of fame for the most stupid Wiktionary entries can be found at Special:LongPages; highlights include a, go, line and set.

A better structure would see the non‑English words split off into subdictionaries, and then have a separate entry for each English headword. Ambiguous spellings should be treated as ambiguities are treated in Wikipedia (and indeed Teflpedia) and actively disambiguated.

Nevertheless, in the absence of anything better that’s licensed under a compatible Creative Commons, Teflpedia does link to Wiktionary.

Technically speaking, Wiktionary classifies words according to traditional grammar rather than modern scientific grammar.

They also fail to identify (1) which verbs can undergo dative shift and those which can't and (2) reporting verbs.

It also contains errors, such as classifying same as a pronoun (it can take a determiner, so it’s not a pronoun).[9]"

I don't know much about the quality of this site though, considering I only just found out about this site from voy:Teaching English. HyperAnd (talk) 04:59, 27 September 2025 (UTC)

Thank you. Most of this is IMO opinion rather than factual criticisms. Basically, they prefer a different structure, that is all. Some people do indeed argue that we should have separate pages per-language, but I have never seen up until now the idea that separate etymologies and/or parts of speech for a given word should be placed on different pages (which would make things very awkward IMO). Comparing Wiktionary to Wikipedia and using Wikipedia as a reference to be emulated is IMO missing the point, as dictionaries are very different beasts from encyclopedias. The only substantive criticisms I see are that we mostly fail to identify the argument structure of English verbs (which is true and is something I've talked about addressing, but it's complex to do so) and that it contains some errors (which is necessarily true of a crowd-sourced dictionary, but the one "error" they identify concerns determiners vs. pronouns, which is something that is frequently argued about, so I would not necessarily consider it a substantive criticism). Benwing2 (talk) 06:04, 27 September 2025 (UTC)

Some of the criticism could be valid but it's mostly bullshit. Here are the main points I could pick out:

• Unfortunately, Wiktionary suffers from poorly structured data […] A better structure would see the non‑English words split off into subdictionaries, and then have a separate entry for each English headword.

  Dividing entries by language seems like a good idea, and would fix Lua errors on a, but then further dividing up English specifically seems bizarre (this has been proposed before but there are major technical and UX challenges).

• Wiktionary classifies words according to traditional grammar rather than modern scientific grammar.

  I don't know what "modern scientific grammar" is but it doesn't look like any other dictionary (including Teflpedia) is using it...?

• They also fail to identify (1) which verbs can undergo dative shift and those which can't and (2) reporting verbs.

  I believe these are covered by Category:English ditransitive verbs and Category:English reporting verbs respectively.

• It also contains errors, such as classifying same as a pronoun (it can take a determiner, so it’s not a pronoun).

  "The parts of speech exist for categorization purposes, to make it easier to find. When it doesn't fit exactly, or when it bleeds—as long as the definition is there, you're well served."

Teflpedia itself is hilariously awful, like their page for idea which describes it only as something "important in philosophy", or their entry for understand which doesn't even bother with a definition and just leaves it at "English verb". Ioaxxere (talk) 07:43, 27 September 2025 (UTC)

$5 says it was founded by a disgruntled ex-Wiktionarian :) 2A00:23C5:FE1C:3701:8076:6C67:B80A:B146 07:46, 27 September 2025 (UTC)

Everything is either objectively mistaken (such as not seeing our categories), and the rest is personal opinion on how they think a dictionary looks prettier. I, for one, much rather have homographs in the same page, although that leads to some half a dozen pages sometimes breaking down.

The same user has another page about Wiktionary on his userspace, which says this:

Wiktionary is basically a huge bug:
[Potentially epic]
The data structure of Wiktionary is a 🥵 hot mess of 💩, and I'm going to tell you why this is, and how it can be solved. ✅
The ideal data structure is 1 idea, 1 page. This is not followed. Instead, if we have a look at e.g. the entry on love in the English Wiktionary: https://en.wiktionary.org/wiki/love it has an English entry, followed by entries for Czech, Danish, Dutch, French, Friulian, Inari Sami, Middle Dutch, Middle English, Norwegian Bokmål, Norwegian Nynorsk, Romani and Serbo-Croatian. Yikes. 😬
This is the first problem. As an English-speaking reader, I want a definition of an English word. I don’t really care very much about other languages. If I want to look up the definition of a German word, then I'd start looking specifically for a German→English dictionary, since my German is not good enough to read a monolingual German dictionary. It's very rare that I can’t specify the language I want to search for.
So, the first solution is to separate out the non-English words from the English words. Starting from the base, https://en.wiktionary.org is English Wiktionary. Let's put the English in the root directory, and prefix all non-English content with that language's code. So, for example, all French→English content (i.e. French words with English analysis) would be at https://en.wiktionary.org/fr/...etc.
A well-programmed bot can sort this out.
Immediately, that will solve a lot of the problems, but cause others (those will be addressed below).
The next problem is that for languages like English, we have to deal with two types of entities. These are (1) spellings and (2) lemmata (also known as headwords).
A spelling that may refer to multiple possible lemmata is called a homograph. An example in English is the word like, which can (1) be a preposition (What's it like?) or (2) a verb (I like cake) which also has secondary use as a noun (the page has 30 likes).
Then, each lemma can have multiple spellings. For example, the English verb know has know, knows, to know, knew, known and knowing as the spellings of its forms. Other words have
So. Imagine two columns here, one of spellings (on the left), and one of lemmata on the right. We're going to pick some example words and see how they relate to each other.
Let's choose…
Spellings
lemm
A standard print dictionary is usually a list of headwords and the dictionary user is expected to understand that (say) raisins is the plural of raisin, and look up raisin to obtain information about raisins.

And another in the project namespace:

Wiktionary is an online dictionary run by the Wikimedia Foundation (who also run Wikipedia).
Teflpedia has a separate entry on Wiktionary in the category:index. This Template:tp internal project page is about how and why to link to Wiktionary.
In Teflpedia, use Wiktionary to link to dictionary definitions where such definitions are needed.
The 👍good things about it are:

• It’s available online
• It’s licensed under a Creative Commons licence, as Teflpedia is, so there is idealogical compatibility.
• It uses traditional grammar categories, which are likely to also be used in teaching materials; e.g. yesterday is listed as an noun/adverb.

However, Wiktionary has major 👎problems, including:

• Conflating lemmas together into single entries based entirely on the spelling of a word (i.e. homographs are treated together, even when this is completely illogical).
• In the English version, failure to focus on English, instead it attempts to cover every language that has ever existed — and fails.
• It does not appear to be driven by corpus data, which is the modern approach to creating a dictionary.
• It uses traditional grammar categories; e.g. yesterday is listed as an adverb/noun, but in modern grammar analysis demonstrates it’s actually a pronoun. Traditional grammar categories often make little to no sense.

There’s also a project page on Wikipedia by the same user (surprise), which briefly mentions Wiktionary. The user is really afraid of being bold. Maybe he’d rather be italics:

Wikipedia is a wiki-based encyclopedia run by the Wikimedia Foundation. It’s one of the world’s largest websites.
In some ways, Teflpedia is like Wikipedia, but in other ways Teflpedia is not Wikipedia.
== Category:Teaching English as a foreign language ==
The coverage of articles within https://en.wikipedia.org/wiki/Category:English_as_a_second_or_foreign_language is … interesting.
== Errors in Wikipedia ==

1. As of August 2019, The Wikipedia article lexical set [10] sets only considers phonetic sets, and excludes all other types of lexical sets, particularly semantic sets which are very important in MFL.
2. As of August 2019, The Wikipedia article on subtext [11] considers only texts that happen to be written narrative, excluding all other sorts of texts, i.e. all other genres of written texts and ALL spoken texts.
3. According to the Wikipedia article on quantifiers, [12] quantifiers cannot be numerals, which at the very least seems disputable.
   ==Missing articles ==
   The following articles do not exist:
4. As of August 2019, Wikipedia lacks an article on place name/toponym,[13] though it does have one on toponymy[14]. The latter is an academic field that studies the former. This is like having an article on linguistics, but not one on language, or one on second language acquisition but not one on second language acquisition studies!
5. Wikipedia lacks an article on grammatical unit [15]
6. Wikipedia has an article on discourse analysis [16], but lacks an article on discourse studies ([17]). This is akin to conflating language analysis with linguistics.
7. Wikipedia lacks various ELT-related biographies, including those for Lionel Billows [18]., Christopher Brumfit [19], E.V. Gatenby [20], Vere Redman [21] and probably others.

== Missing content ==
The article on Probability [22] is mostly about mathematical probability. Etymology is discussed, but a glaring omission is how languages (particularly English) talk about probability.
== Duplicate articles ==
# Wikipedia has articles on both adverbial[23] and adverbial phrase [24] even though given single-word phrases, they are the same.
== Badly-written articles ==
The article on reading education (reading education) redirects to Reading education in the United States, which for once is as bad as the tags on the page, and is an awful mess.
== Completely misnamed ==
The Wikipedia entry on free response item [25] is not about free response items, but extended response items. Most extended response items are free response items, but multiple choice items can also be free response.
== Wiktionary ==

1. The Wiktionary entry for wish [26] does not contain usage notes.
2. If [27] does not link to its antonym unless [28]

— Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 17:09, 27 September 2025 (UTC)

One point we could take from this is that some users are really fixated on how the data looks rather than what the data is. In the same way that an appropriately structured HTML page can be made to look totally different by changing the CSS ("presentation layer"), and e-mail messages can be displayed in various ways depending on the reader's client: perhaps we should offer a sort of raw data server that could be displayed by users in the way of their choice. Unfortunately it's hard to know what technology to use, since general "walled gardens" and anti-desktop, pro-Web approaches have destroyed most consistent UI of the past. 2A00:23C5:FE1C:3701:BD91:D983:D60:4265 18:32, 27 September 2025 (UTC)

I’m almost tempted to make the entry Yeah, well, that’s just, like, your opinion, man but this doesn’t meet the CFI. Nicodene (talk) 21:58, 27 September 2025 (UTC)

"surface analysis" -> "surface etymology"

It occurs to me that "surface etymology" may be clearer than "surface analysis". "Surface analysis" suggests something fancy, when surface etymologies are not fancy; they're based on what the etymology looks like on the surface. So I propose, instead of

By surface analysis, ...

we say

The surface etymology is ...

(or something similar) and link "surface etymology" to the glossary. The code can be made smart enough to insert an indefinite article when there's preceding text, so that e.g. {{surf|en|vibrate|-tion}} displays as

The surface etymology is vibrate +‎ -tion.

But {{surf|+cal|es|en|red-eye}} displays as

The surface etymology is a calque of English red-eye.

(or similar)

Benwing2 (talk) 04:14, 27 September 2025 (UTC)

This seems good. I wonder if we could even do "on the surface" (like you used just now to explain it). Like,

On the surface, vibrate +‎ -tion.

Ioaxxere (talk) 07:47, 27 September 2025 (UTC)

There are many debates as to what the wording of this concept should be. This seems fine to me, or "Synchronously". As to the wording, I don't mind By "X" or an adverb. Vininn126 (talk) 17:22, 27 September 2025 (UTC)

"Etymology" may be better than "analysis". However, I'm concerned that the word "surface" is jargony and does not have an obvious meaning in this context to non-linguists no matter what word follows it (I remember seeing a complaint on the talk page of w:Talk:Old English phonology from a user who found "The inventory of consonant surface sounds" an opaque way of discussing Old English consonant phonemes). When appropriate, I feel like saying something like "Equivalent to vibrate + -tion" is more straightforward.--Urszag (talk) 18:47, 27 September 2025 (UTC)

Synchronous etymology? Vininn126 (talk) 18:49, 27 September 2025 (UTC)

My issue with "equivalent to" is that it's too vague. "Equivalent" in what way or sense? At least "surface" indicates that this is derived from looking at the structure of the word as it appears directly on the surface (i.e. synchronic, but this is even more jargony). "Equivalent" comes across to me as very hand-wavy. Benwing2 (talk) 18:53, 27 September 2025 (UTC)

BTW I'm starting to think that if "surface" is unacceptable we should bite the bullet and just say "synchronic etymology" or "synchronically", because that has a precise definition that we can link to and can be Googled. Benwing2 (talk) 18:58, 27 September 2025 (UTC)

My thoughts exactly. It's jargon but Googable jargon. Vininn126 (talk) 19:02, 27 September 2025 (UTC)

I support ‘synchronically’, with a glossary link for any needed clarification. Concise, unambiguous (there being no plausible alternative reading of ‘synchronically A + B’), and in line with common usage in linguistics. Nicodene (talk) 21:53, 27 September 2025 (UTC)

I'm in favour of "synchronically" with a link to the glossary. —Caoimhin ceallach (talk) 21:32, 1 October 2025 (UTC)

There's always "superficial etymology", since "surface" and "superficial" are basically the same idea in isolation, minus the negative connotations in non-technical usage of the later. Chuck Entz (talk)

(Re Urszag) In fairness, I also find "The inventory of consonant surface sounds" confusing (and would find e.g. "The inventory of consonant synchronous sounds" confusing), in a way I don't find "surface etymology" confusing; "The inventory of consonant (surface|synchronous) sounds" is just a weird phrase, IMO; I could formulate some guesses as to things it could mean, but wouldn't feel certain of any of them. - -sche (discuss) 03:07, 28 September 2025 (UTC)

My response to all of this is that, of the options I've seen and have been presented here ("by surface analysis"/"equivalent to"/"analyzable as"/"corresponding to"/"synchronically"/"the surface etymology is"/etc.), I find myself in favor of "equivalent to". The use of "surface" in any sense, while bringing to mind "this is what it seems like to modern speakers, but it's not actually the case", comes off to me less as "because the affixation happened in a previous stage of the word's life" and more "because it's a false/folk etymology" (e.g. interpreting adultery as adult +‎ -ery when it's really more like ad- +‎ alter +‎ -y or cookie as cook +‎ -ie when it's really more like cake +‎ -kin). "Synchronically", while less ambiguous, is less common in the wild and runs the risk of making my head sad. As for "equivalent to", what I get from it is "if you redo the affixation of the etymon with the etymon's components' descendants, you'll get something like this". I suppose that's poorly-worded, though, because there are instances where equivalence is not readily apparent on a surface analysis (‎ad- + ‎rest → ‎arrest, ‎fly + ‎-th → ‎flight, ‎may + ‎-th → ‎might, ‎wer (“man”) + ‎eld (“age”) → ‎world, plenty of others I know I'm forgetting). I still think these non-obvious ones are worth noting, though. -BRAINULATOR9 (TALK) 02:03, 28 September 2025 (UTC)

I am as fine with "surface etymology" as with "surface analysis"; no strong preference between them. My main concern with "synchronically" is that it's opaque, although I guess if someone doesn't know what it means, and they just ignore it or try to guess from context, they will still understand the gist of the etymology. - -sche (discuss) 03:07, 28 September 2025 (UTC)

Just FYI, I would be strongly opposed to "equivalent to" by itself, but I would accept "synchronically equivalent to". Yes, "synchronically" is a technical term, but if you just ignore it you would still get the gist, and it will be linked to an explanation in the glossary. (I would explain it as, if this term were formed today, it would be formed in this fashion.) Benwing2 (talk) 21:23, 28 September 2025 (UTC)

If ‘synchronically equivalent to’ stands a better chance of meeting consensus, it has my support. I would much rather sacrifice conciseness than unambiguousness or correlation with common academic usage. Nicodene (talk) 00:33, 1 October 2025 (UTC)

Every option (for this wording) has someone who considers it more opaque than the alternatives. But the only objective thing we can say about it is that whichever wording is chosen, it should not fail to be hyperlinked to a glossary entry for an explanatory gloss that can be easily obtained by clicking or tapping. This remains true even if people cannot agree on the optimal wording of the explanation; whichever wording of the explanation currently exists (at any given time) is better than no link to any help at all. I suspect that I am preaching to the choir, but if not, then it was worth stating explicitly. Quercus solaris (talk) 23:39, 30 September 2025 (UTC)

If one just uses "equivalent to", as many us do, you don't need a hyperlink to a glossary as it's opaque enough to not need it. --{{victar|talk}} 19:42, 2 October 2025 (UTC)

Rename Template:commonscat (commons cat / "common scat")

Hey guys, I apologize, I'd just like to ask if someone could rename Template:commonscat. A few years ago I noticed that it not only spells "commons cat", but it also spells "common scat" ('everyday feces'). I was too embarrassed to bring it up, but I think it'd be nice to just go ahead and name it like, Template:commons-cat or Template:catcommons or other variation. If someone doesn't fix this, I'm gonna be stuck with this shit forever. --Geographyinitiative (talk) 15:22, 28 September 2025 (UTC)

Changing this seems unnecessary.--Urszag (talk) 21:28, 28 September 2025 (UTC)

Completely agreed. @Juwan has vaguely suggested adding spaces in certain templates for consistency, but it doesn't seem necessary for that reason and definitely not for this reason. Benwing2 (talk) 21:35, 28 September 2025 (UTC)

I agree that such a change seems overly sensitive and thus unnecessary. In any case, we should encourage the use of {{R:commons}} instead. — Sgconlaw (talk) 16:12, 29 September 2025 (UTC)

"Verbal combining form" as a POS

There are a few Ukrainian entries (-дягати (-djahaty), -дягнути (-djahnuty) and -дягти (-djahty)) that have been sitting in todo lists for a while because @Eilaiyas decided to use this POS in the {{head}} template. It's not recognized as a valid POS by our headword modules, so it isn't in either Category:Ukrainian lemmas or Category:Ukrainian non-lemma forms. There are also others in Category:Ukrainian verbal combining forms where this has been "fixed" by hard-coding the categories via {{cln|uk|suffixes|lemmas}}. Given that combining forms are a feature of many inflected languages, we need to think this through and either make it an accepted POS and convert everything possible to it, or convert it to whatever is already being used in other such entries and have Category:Ukrainian verbal combining forms as an additional category (easily done via the |cat2= parameter in {{head}} or via one of several definition-line templates). We need to be consistent. I might as well bring our Ukrainian workgroup in on this: (Notifying Atitarev, Benwing2, Underfell Flowey, Voltaigne, ZomBear): Chuck Entz (talk) 00:13, 29 September 2025 (UTC)

@AshFox needs to update his user name in the {{wgping}} template. Chuck Entz (talk) 00:17, 29 September 2025 (UTC)

Thanks for raising this. We should standardize our treatment of Ukrainian verbal combining forms and Russian verbal combining forms. The latter are handled by {{ru-verb-cform}} which categorizes them as verb lemmas as well as verbal combining forms (which is a subcategory of suffixes). This makes sense in that these are often 'ghost' verbs descending from standalone verbs but only surviving in the modern language in combination with prefixes. {{suffixsee}} still works on them. Voltaigne (talk) 09:19, 29 September 2025 (UTC)

In some cases I wonder if it wouldn't be better to analyze it differently, e.g. with -силати or -микати. Funtionally, they are part of the pattern where -ати imperfective derivatives of verbs with root-initial consonant clusters (from a now-disappeared hard yer) lead to the insertion of a -и- in its place. Since this pattern is (to my knowledge) fairly consistent, could we not describe it as an effect of the suffix, rather than as standalone forms? That's how we analyze Polish -ać, which can lead to equivalent changes (e.g. wymykać). Though that only applies to a couple entries in the category, most others do appear to descend from originally standalone verbs.

Only Russian and Serbocroatian seem to have categories for (verbal) combining forms out of the Slavic languages (from what I could find), and both seem to follow the strategy of "use the POS anyway and manually categorize". If that's what we're basing it on, I guess we might as well officially recognize it? But I'm surprised most other Slavic languages seem to be doing without any such category at all. Did the standalone verbs all incidentally survive long enough that there could be an entry for them, marked obsolete if needed? PhoenicianLetters (talk) 13:02, 1 October 2025 (UTC)

October 2025

Is the text= parameter of etymon meant to be used yet?

(this originated from User talk:Fenakhay but it developed into a more general discussion so I'm reposting it here)

I'm not sure if you saw my edit at eusebes; I noticed you added text=++ here and also have used this parameter at some other entries such as žirafa. I don't necessarily think this is a bad thing but I think there is existing consensus not to use this parameter to replace text etymologies, as per Wiktionary:Votes/2024-04/Allowing etymology trees on entries, and I'm confused because I haven't noticed any more recent discussions that changed that. Can you clarify if I'm missing something, or if you think this parameter is ready to be implemented, and community discussion is not needed before proceeding with it? Thanks. Urszag (talk) 21:26, 28 September 2025 (UTC)

@Benwing2, This, that and the other: as admins, would you be able to weigh in on the technical question of whether the result of Wiktionary:Votes/2024-04/Allowing etymology trees on entries is currently an active constraint on the use of text= as a replacement for other etymology text and templates, as in this edit to azucre, etc.? While I think the vote says not to make edits like this, I want to figure out whether I am misinterpreting it or missing more recent discussion before I start a Beer Parlour discussion about whether this use of etymon|text= should now be allowed.--Urszag (talk) 19:31, 29 September 2025 (UTC)

It was experimental eons ago and is ready now, and works reliably. The earlier consensus reflected caution while it was still in testing, but the current functionality is stable. It is used in many languages as you can see Category:Entries with etymology texts by language. — Fenakhay ^{(حيطي · مساهماتي)} 19:37, 29 September 2025 (UTC)

Thank you for the response. I think it would be nice to establish greater clarity. The documentation at Template:etymon/documentation has never been updated from calling text= "[EXPERIMENTAL]". It may be a fait accompli, but the number of entries using it is one aspect of determining if there is a practical consensus for its use: if those entries are mostly the work of only a small number of editors (such as yourself and @Trooper57), that means something different from if they are the work of many different editors. This also seems like something where the relevant consensus could plausibly differ by language.--Urszag (talk) 19:50, 29 September 2025 (UTC)

@Urszag IMO we should proceed very cautiously with replacing textual etymologies with {{etymon|text=+}}. For one thing, the syntax is likely to be difficult to handle and opaque to most editors, compared with the similar raw text. Back in Dec 2024 in Discord before all the recent work on {{etymon}}, I proposed a replacement for {{etymon}} called {{ety}}, which I'll quote here:

1. Hey I was going through trying to replace invalid uses of {{der}} in Spanish that should be {{bor}} or {{inh}} and I was thinking we might need some new etymology templates to reduce the need for {{der}}. I am thinking the following:
   1. {{rinh}} = "restructured inheritance"; an inherited term that was restructured in some irregular way e.g. through metathesis or other irregular sound change.,
   2. {{iinh}} = "indirect inheritance" and {{ibor}} = "indirect borrowing"; these replace {{der}} uses farther up the etymological chain, indicating that the derivation in question is either a borrowing or inheritance. Hence: From {{bor|es|LANG1|...}}/{{inh|es|LANG1|...}}, from {{ibor|es|LANG2|...}}, from {{ibor|es|LANG3|...}}, from {{iinh|es|LANG4|...}}, ....,
   3. If we adopt {{rinh}}, then logically we might want {{irinh}} = "indirect restructured inheritance".,I'm not sure whether these need to categorize differently but they could display differently; e.g. {{iinh}} might display as "inherited from LANGUAGE FOO" and {{ibor}} as "borrowed from LANGUAGE FOO" (and {{irinh}} as "inherited with restructuring from LANGUAGE FOO"; in that case the above chain would read From {{bor|es|LANG1|...}}/{{inh|es|LANG1|...}}, {{ibor|es|LANG2|...}}, {{ibor|es|LANG3|...}}, {{iinh|es|LANG4|...}}, ... without the need for "from" before each one. An alternative to this is a single {{ety}} template that has param per link in the chain: {{ety|es|b/fr:TERM1|i/la:TERM2|b/grc:TERM3|b/he:TERM4|...}}. Here the code b = borrowing, i = inheritance, ri = restructured inheritance, u = unknown if borrowing or inheritance, unspecified = added to a maintenance category, etc. and we don't need the "indirect" variants. This has multiple advantages, e.g. it's easy to add new codes, the formatting can be changed if someone e.g. wants a list, etc.
      
      

There was no immediate response, but when I brought this up again a month later, the feedback from @AG202 and @Hftf was that this syntax was difficult and opaque, and the same feedback seems to apply to {{etymon}}; I took a look at the uses on žirafa and they are somewhat opaque even to me. In response to that feedback, I proposed a free-form version of {{ety}} that used <<...>> blocks to indicate inheritances, borrowings and the like, and everything else is output as-is. This is much like the "new" (i.e free-form) syntax in {{place}} and {{lb}}, and I think it's very important to have something of this sort in {{etymon}} before we start using it more widely to replace the current text of Etymology sections. (NOTE: This is orthogonal to the issue I brought up in the above post concerning additional variants for indirect borrowings, inheritance with restructuring and the like; those should be handled in a separate discussion and IMO after we figure out when/whether {{etymon|text=+}} is allowed.) I disagree with Fenakhay that the stipulations on using |text=+ in that vote should be disregarded; I understand that they were based on it being experimental, but (a) I think that even if it currently works it's still essentially experimental because it's still under development and the syntax has not been finalized; and (b) more generally, the absence of non-consent is not consent, so we need an explicit, separate consensus discussion concerning the use of {{etymon|text=+}} (which could indeed end up concluding that it needs to be decided per language community). Benwing2 (talk) 22:47, 29 September 2025 (UTC)

@Benwing2: I am pretty surprised to hear that the uses of the template on žirafa were difficult to understand. When I look at something like {{etymon|cs|id=giraffe|:bor|it:giraffa<id:giraffe>|tree=1|text=+}} I can practically read it left to right like an English sentence: "The etymology of this Czech entry with ID `giraffe` is a borrowing of Italian giraffa (ID: `giraffe`); generate a tree and text." Is there something that can be improved in the documentation? As an aside, some of the new complex features, which aren't present on žirafa (reference names and groups, recursive inline etymologies) actually are obscure to me so I would encourage @Fenakhay to fully document them and add examples.

As for |text=, I did notice that over the past few days/weeks Fenakhay has been mass-adding etymology texts across dozens of languages. I think it is a bit silly to be concerned about edits where the output stays letter-for-letter identical, but we should just create a Beer Parlour thread where we can officially authorize |text= if editors feel that to be necessary. I am often tempted to use it simply because it saves keystrokes.

I do think you have an interesting concept for {{ety}} but I'm not sure whether a complex mechanism for free-form input is really necessary. Something like 95% of etymologies are something bog-standard like "From A + B" or "Borrowed from X" which an algorithm can easily write for me. A "nonstandard" etymology that comes to mind is go ham, where I'd say that a fancy template with free-form input is hardly an improvement over writing plain English in wikitext. Ioaxxere (talk) 06:11, 30 September 2025 (UTC)

Since I was summoned, my two cents: I'm broadly in agreement with Benwing here. The 2024 vote says "This feature is currently experimental and should not be used in mainspace" - presumably that text was put there to address the worries of certain users, and surely some voted to support the vote on that basis. We should at least have a discussion about the |text= parameter; if it has its detractors, they can explain themselves there, and supporters can come along to rebut their objections. This, that and the other (talk) 09:13, 30 September 2025 (UTC)

Benwing2 (talk) 02:23, 1 October 2025 (UTC)

Placement of {{&lit}}

In 2021, User:Fytcha added to the template’s documentation that it should be used as the last sense below the idiomatic and figurative ones. I looked for their contributions and could not find a record of any related discussions prior to the change. I believe this stipulation goes against common usage of the template, at least nowadays.

Many idiomatic terms whose literal uses are more common than their figurative ones place {{&lit}} before the figurative senses, as can be seen in, e.g., car crash, big boy, and come across (I lack the Special:Search skills to come up with more examples). Indeed, I believe this is consistent with WT:EL#Definition sequence, which says it is important that the most common senses of a term be placed first.

An exception specifically for {{&lit}} seems unnecessary and confusing, as it suggests an idiomatic sense is more common than the literal one when this is sometimes not the case; of course, even when this is the case, the stipulation is redundant with what is already at WT:EL. I propose that the sentence in question at Template:&lit/documentation be removed to reflect what appears to be pretty common practice. — Polomo ⟨⁠ ⁠oi!⁠ ⁠⟩ · 04:20, 2 October 2025 (UTC)

Obviating the need to pipe links for typographical apostrophes

Following a change to the Dutch perpetuüm mobile by WingerBot which, inter alia, changed the second plural given in its headword line from perpetuüm mobile’s to perpetuüm mobile's (note that they link to the same page and differ only in the display of the apostrophe), Mnemosientje and I disagreed about its restoration, with Mnemosientje finding it “unnecessary, messy and inconsistent with established practice https://en.wiktionary.org/w/index.php?go=Go&search=intitle%3A%2F%27%2F&title=Special%3ASearch&ns0=1 ”. Like correct typography, correct spelling and grammar in our entries is not strictly necessary as long as they're good enough to be understood, but it mars our credibility if we fail to meet professional standards of presentation in these respects. The charge of inconsistency with established practice is, I think, misplaced, since our disagreement concerns what to display, not the {{PAGENAME}} to which to link. (There is good reason for using U+0027 APOSTROPHE in page titles, namely search-accessibility, which I am not contesting.) However, I have to agree that the code that accomplishes this is indeed messy, but I believe there is a better way to achieve this. In the specific case of Dutch, what is needed, unless I'm mistaken, is to add:

entry_name = {

Latn = {

from = {"’"},

to = {"'"},

},

},

between the current lines 1671:

ancestors = "dum",

and 1672:

sort_key = {

of Module:languages/data/2. This would work like the automatic diacritic stripping that allows, e.g., {{m|la|amō}} to link to [[amo#Latin]] whilst diplaying amō, without the need for link-piping. In the case of perpetuüm mobile, that would mean the code for the second plural form could be just perpetuüm mobile’s instead of [[perpetuüm mobile's|perpetuüm mobile’s]].

Module:languages/data/2 is currently protected against being edited by those not either template editors or administrators. Therefore, I would be grateful if someone in either of those groups would make the necessary change I laid out above, provided that doing so wouldn't have the effect of breaking anything. More broadly, this is probably something worth adding to the data for very many languages that use the Cyrillic, Greek, and Latin scripts. So many, in fact, that it would be good to find a way to apply it at a more fundamental level, thus precluding all the undesirable redundancy that would exist otherwise. 0DF (talk) 12:17, 2 October 2025 (UTC)