Wiktionary:Beer parlour/2021/May

Persian transliteration in etymologies
Some questions about the transliteration of Persian in etymologies:
 * In the etymology section of words borrowed from Persian in South Asian and Turkic languages, should the transliteration be given in modern Iranian or Classical Persian?
 * In the etymology section of words borrowed into Persian from other languages (especially Arabic), should the transliteration be given in modern Iranian or Classical Persian?

For example, the current etymology at 🇨🇬 would give a lay reader the impression that there were two shifts in the first syllable's vowel quality, when in reality there was none.--Tibidibi (talk) 12:39, 4 May 2021 (UTC)


 * For South Asian languages, the Classical Persian transliteration should be followed. Only few editors like AryamanA, Kutchkutch, or I use  as an etymon instead of  . For Turkic languages too, I think likewise (though ping Turkic editors like  for input). And the etymologies of our old entries need to be updated. - ⸘ -  dictātor · mundī  13:14, 4 May 2021 (UTC)


 * The vast majority of Persian loanwords in Ottoman Turkish were borrowed in the 16th century, or in any case not later than the 17th century. Does that make the donor “Classical Persian”? --Lambiam 11:56, 5 May 2021 (UTC)
 * , "Classical Persian" does not really exist as a discrete linguistic category per se, but Maggi and Orsatti ("From Old to New Persian", Oxford Handbook of Persian Linguistics) defines "Classical New Persian" as the relatively standardized literary language between Saadi in the thirteenth century to increasing European influence beginning in the late eighteenth and early nineteenth centuries.--Tibidibi (talk) 12:27, 5 May 2021 (UTC)


 * For South Asian languages, the classical transliteration should be used in most cases instead of the modern transliteration since it would provide a more historically accurate representation. As mentioned at User_talk:Inqilābī:


 * At a maximum wiki would be used for all pre-1800s borrowings, so wiki would almost always be wiki with the unwritten transliteration system discussed at User_talk:AryamanA/2019.


 * However, as AryamanA said there:
 * Of course, I have no deep knowledge of Persian (let alone Classical Persian) myself.
 * If there is a transliteration for a modern Persian term at Wiktionary or another dictionary, it's just easier to use that transliteration rather than trying to guess what the classical transliteration would be. As Inqilābi said there:
 * so it is high time we had a Transliteration page dedicated to Classical Persian, following Steingass's standard.
 * Judging by, it appears that the meaning of the word wiki in the term wiki (referring to ) is very loose. Kutchkutch (talk) 10:10, 7 May 2021 (UTC).

Thracian
Is this where should I address requests for creating templates to enable rendering entries for Thracian language terms and for the modification of the already existing templates so Thracian terms can be represented by the Latin alphabet in addition to the Greek alphabet already accepted by Wiktionary? Antiquistik (talk) 16:36, 4 May 2021 (UTC)
 * If this isn't the place to address the request, where should I request these changes instead? Antiquistik (talk) 14:45, 11 May 2021 (UTC)

Yiddish
Why is Hebrew or Biblical Hebrew not considered ancestor of Yiddish? EdwardAlexanderCrowley (talk) 03:32, 5 May 2021 (UTC)
 * Because the language itself is Germanic, even if you can trace individual terms back to Hebrew or Aramaic. Chuck Entz (talk) 05:19, 5 May 2021 (UTC)
 * I know Yiddish is Germanic, but according to the doc of Module:languages/data*, multiple ancestors can be listed for pidgins, creoles and mixed languages. EdwardAlexanderCrowley (talk) 09:01, 5 May 2021 (UTC)
 * But Yiddish is none of those things. It's a Germanic language with a lot of Hebrew loanwords, just as English is a Germanic language with a lot of French loanwords. —Mahāgaja · talk 10:45, 5 May 2021 (UTC)
 * That is correct. Yiddish is no more a Semitic language than English is a Romance language, despite having a ton of borrowings from French and Latin. Benwing2 (talk) 04:52, 8 May 2021 (UTC)

Exotic Page Numbers
When referencing a page of a book that using numerical page numbers that are neither Western Arabic nor Roman, is it better to use transliterated numbers or the raw numbers? For example, when referencing a book whose page numbers are marked using Burmese digits, is it better to write p171 or p၁၇၁? Now, if that were the 171st page of the referenced PDF of a scan, there might be an advantage to using the Arabic number, but if there are prefatory pages, one would have to back-transliterate "171" to Burmese script to know that one had found the right page number. RichardW57 (talk) 06:42, 5 May 2021 (UTC) Of course, if Arabic numbers and native page numbers are used for different sequences (e.g. Arabic for the prefatory pages and native for the main body, as I have seen), then one has to use the numbers as written. RichardW57 (talk) 06:42, 5 May 2021 (UTC)
 * Maybe "171 (၁၇၁)"? --Geographyinitiative (talk) 09:00, 5 May 2021 (UTC)


 * If we are likely to do any arithmetic with these numbers then they should be in a consistent format (which on computers generally means Arabic numbers). If they are just for display purposes then I suppose it doesn't matter much...? Equinox ◑ 19:56, 10 May 2021 (UTC)
 * We're not going to do any sensible arithmetic operation on page numbers "xxiii" (from the preface) and "15" (from the main body), let alone "vi (bis)" - I have seen prefaces have two runs of Roman page numbers! Geographyinitiative's suggestion seems good for users - not so much fun for editors.  (Keyboard switching can get temperamental.)  The only downside I can think of is that there might be some checking that the page number is a number, but that not a problem with the quote family of templates. --RichardW57 (talk)
 * Wait. How is switching keyboards temperamental? You just use Windows Key+Spacebar and you're done. If you're citing a digital source, you can copy-paste, and if you're working with a physical document that uses non-Arabic, non-Roman numeric, chances are you're well versed in and prepared for typing in that language, so you'd have that script stored in your quick-access keyboards. So I don't really see the problem on a fundamental level. 110521sgl (talk) 15:43, 20 May 2021 (UTC)
 * Clicks and key presses can go missing on a busy system - and windows can temporarily forget which keyboard they're using!. Moving to the 'next' keyboard doesn't work well in such a situation - you don't know whether the key press has gone missing or whether that keyboard is still being loaded.  That's rather frustrating when you've got about a dozen keyboards to cycle through.  (Pali is written in 10 scripts.)  I decided that the slower process of pointing to choose was less frustrating than the switching shortcuts. --RichardW57m (talk) 17:19, 2 June 2021 (UTC)
 * I think this is the correct option or possibly "p. ၁၇၁ (171)". The page numbers are the page numbers: just like how a book can have Roman numerals or appendices of the sort "p. C-17", we shouldn't really change the format or remove the page numbering entirely for our benefit. —Justin ( koavf ) ❤T☮C☺M☯ 21:52, 28 May 2021 (UTC)


 * I agree with Justin: the original information first, and then the 'translated' information.  But in general, follow the same pattern as used for the title of the work (assuming there is one).  In which case you could alternatively put the entire reference in the original language, followed by a translation of the entire reference.  —DIV (1.129.106.0 05:21, 17 June 2021 (UTC))


 * I repeatedly find that one must use such numbers, because it is e.g. an edition of a work written and numbered in Arabic (.–٩) with translation and commentary appended in German numbering (0-9, I like to call it like that for distinction, because you know how ambiguous “Arabic numbers” is, and “Western” would not distinguish from Roman numbers I–X).
 * Example books and templates are  and (of the same publisher, but that publisher is still popular). Currently however I cannot calculate the links to page scans having German numbers in the URLs using Arabic numbers, which I would have to, but I write ugly stuff under entries like 12 (١٢) as a mere Western number (“12”) would be already assigned and make the reference as a mere reference invalid.
 * And yes, indeed, in Orientalist literature one refers to pages with Arabic numbering, appearing usual enough that I did not take notes of it, but e.g. in that very reference, so it is not wrong to do it.
 * How can I convert an Arabic number to a German one? If it is not possible yet then Module:ugly hacks should 1. have a function whereby number systems can be converted, which should be easy for Arabic/Persian numbers since they are decimal, however I imagine that people could use it for Roman numbers (I–X), as some works have large forewords numbered by them, while URLs on websites have German numbers in any case. 2. apply the “match” to Arabic numbers too, which would be necessary even if one converts numbers before using them for URLs, or necessary for the very process of conversion, since page number parameters can contain ranges like ١١١–١٢٥ and other things 3. perhaps even detect “whether Arabic or German” so one does not need to have separate parameters page and arabicpage. –Fay Freak (talk) 15:29, 6 July 2021 (UTC)
 * Module:foreign numerals is relevant. --Vahag (talk) 17:53, 6 July 2021 (UTC)
 * For Roman numerals we have and . No doubt similar templates (and their supporting modules) could be developed for other major numbering systems. — SGconlaw (talk) 18:17, 6 July 2021 (UTC)
 * It would be pretty easy to add support for conversion between Arabic and western decimal numbers to Module:foreign numerals. The module is a bit overcomplicated though; because of the way Unicode is set up, conversion between decimal numeral systems is as simple as converting each digit to the code point, subtracting the zero code point of the current number system (, U+0660), and adding the resulting number to the zero code point of the target number system ( , U+0030), and converting back to a character. So there can be a single decimal number conversion function that uses a list of number systems with their zero code points, like this. — Eru·tuon 21:16, 6 July 2021 (UTC)

Enabling global sysops on this wiki
Hi, I propose allowing Global Sysops to work on this wiki. It is currently not enabled because the community has more than 10 admins/3 active sysops, but I strongly recommend that the community opt-in because they often help in combating spam and vandalism (eg GRP). As an en.wikibooks admin, I can attest to the work they do and have no issues with them at all. Thanks in advance, and please ping me if you need further input, since I don't watch this page.

P.S: Global sysops won't interfere with normal Wiktionary matters (for instance they do not have access to Special:UserRights) - their role is codified in the policy page and is more or less handling spam or vandalism. This wiki can enact a global rights policy if needed. They'll only help you. Leaderboard (talk) 09:44, 5 May 2021 (UTC)
 * Why would we want them? Global sysops don't know our customs and norms, nor the first thing about lexicography, and in some cases simply don't care. For example, a global sysop was tagging useful talk pages for speedy deletion; if he had been able to, he would have presumably just deleted them. That global sysop was acting in good faith, but good faith alone doesn't mean we should give them extra rights. —Μετάknowledge discuss/deeds 05:35, 8 May 2021 (UTC)
 * That seemed like an unfortunate error in the part of the global sysop (which was handled appropriately), and is a reason why a global rights policy should be enacted, as that will help the GS to decide what to, and not to do (for instance, enwiki's equivalent states that global rollbackers are not to use their rights except for anti-vandalism). It would be reasonable to mention that deleting talk pages should not be done unless it is for vandalism, and ask for flood if making tons of edits at a time. And answering your main question, global sysops end up being very useful during cross-wiki attacks, where they can effectively suppress vandalism/spam. This will also allow stewards to handle routine spam/vandalism, which they cannot do right now. Remember that global sysops are held to a high standard, and should they repeatedly use their rights to violate local policy, they can, and have been, removed. Leaderboard (talk) 07:49, 8 May 2021 (UTC)
 * , pings only go through if you add them in the same edit as your signature. As for your points, this is just one example of how outsiders may not be able to distinguish vandalism from useful edits. Most global sysops are unfamiliar with Wiktionary; one of them is outright hostile to the idea of improving Wiktionaries. I suppose we could limit them to fighting cross-wiki attacks, but it's not like we have any difficulty in handling those with our own admins. —Μετάknowledge discuss/deeds 22:49, 8 May 2021 (UTC)
 * That is true, but would that be a problem for sole vandalism or spam? For a GS that mainly focuses on cross-wiki patrolling, I think the average GS knows when to act and when not to (and they normally err on the side of caution from experience). The odd error may happen, but as you noticed above, GS will usually handle it appropriately. The spam/vandalism on Wiktionary should be no different from any other project, and that's the focus. Additionally, the global rights policy can be used too - ours at Wikibooks say that global sysops must stop if asked to by an admin. Leaderboard (talk) 07:20, 9 May 2021 (UTC)
 * I don't think we are worried about the "average" GS. We are worried about the energetic ones who don't understand Wiktionary, but think they do, and cause problems before they can be asked to stop. DCDuring (talk) 18:20, 14 May 2021 (UTC)
 * I understand your concern, but could that not be handled by a global rights policy? That way, you can inform GS what they cannot do, and this is something all GS have to adhere to (even the "energetic" ones). Also, if they make a mistake, they should revert it. As Metaknowledge showed, they are not fault-free. But they won't repeat a mistake if you tell them, and I think the GS in question didn't either. That is, GS are usually careful enough not to "cause problems before they can be asked to stop".
 * (also, for what's worth, I just encountered a cross-wiki attack which was halted only because I flagged a global sysop - it was fortunate that the wiki being attacked was a GS wiki otherwise I would have struggled for far longer. That wiki had 50 admins.) Leaderboard (talk) 13:30, 15 May 2021 (UTC)
 * something that I remembered just now with regards to the link you gave: the user was not even a global rollbacker (let alone global sysop) back when the first concern (and most likely the second) were noted - DannyS712 got global sysop late in 2020. Leaderboard (talk) 17:06, 15 May 2021 (UTC)


 * Sure, let's not be too dramatic about these things. We'll be delighted to have them. You don't have to know anything about lexicography to delete blatant vandalism. Indian subcontinent (talk) 22:06, 17 May 2021 (UTC)
 * Oppose, there is more than enough admins here to deal with vandalism/LTA and presumably, the sysops here are scattered across the globe. If it's an LTA, a steward can just step in by locking the account. --Minorax (talk) 14:39, 22 May 2021 (UTC)


 * If this means the kids who go around using "Huggle" and "Twinkle" tools then no, they do more harm than good. Equinox ◑ 17:00, 22 May 2021 (UTC)
 * No, it's not the case that any Huggle/Twinkle user gets the ability to delete pages everywhere... Leaderboard (talk) 08:25, 26 May 2021 (UTC)


 * I'll also weigh in as opposed. Our formatting and infrastructure is finicky, and easy to get disastrously wrong if you're a newbie to The Way of Wiktionary and have too much power and enthusiasm.  Many of us are here only part-time, and we still seem to manage vandalism well enough.  But if someone with the keys to the whole palace mistakenly started ripping things out without knowing what they are, it could take us a while to clean up the mess.  ‑‑ Eiríkr Útlendi │Tala við mig 00:13, 25 May 2021 (UTC)

Sigh, not seeing the support here and hence I will mark this as ❌, requesting a trusted user to close this in accordance with local policies. Leaderboard (talk) 08:26, 26 May 2021 (UTC)


 * , I'm a bit confused. At the meta:Global sysops page in the #Scope section, I see this:

 By default, global sysops may use this global user group's permissions on wikis that meet one or both of the following criteria:
 * fewer than ten administrators exist; or
 * fewer than three administrators have made a logged action within the past two months.
 * The English Wiktionary doesn't seem to meet either criteria. So why this thread?  (And I might suggest that that #Scope section be rewritten for better clarity.)  ‑‑ Eiríkr Útlendi │Tala við mig 18:05, 26 May 2021 (UTC)
 * Because the policy is not absolute - any wiki can opt-in (or out) of global sysops by obtaining consensus (which this wiki didn't for inclusion), and that is mentioned later on in the linked page. Leaderboard (talk) 18:10, 26 May 2021 (UTC)
 * That leads to my second point, implied by my parenthetical suggestion -- the #Scope section is unclear, as currently written, and may be cause for confusion. ‑‑ Eiríkr Útlendi │Tala við mig 20:46, 26 May 2021 (UTC)
 * I think I see what you mean - but this will require getting global consensus. I'll see what I can do. Leaderboard (talk) 21:05, 26 May 2021 (UTC)


 * As a super-trusted user, I hereby close this. Indian subcontinent (talk) 21:12, 26 May 2021 (UTC)

inquiry
Can i take wiktionary data for a website and to make a bot which searches for words and display, can i be rendered instruction
 * I'm not a lawyer, or an admin here. In fact, I don't even understand your question. But hey, go for it! Indian subcontinent (talk) 22:03, 17 May 2021 (UTC)
 * Wonderfool the lawyer. *shudders* -- 04:37, 18 May 2021 (UTC)

Mainspace sandbox
For testing of (new) templates which categorise too, I would like a mainspace sandbox because I've observed that categorisation by templates somehow doesn't work on WT:Sandbox and User:U/Sandbox. Any other possible solutions? 🔥 शब्दशोधक 🔥 02:51, 8 May 2021 (UTC)
 * Categorization is normally disabled in userspace, but in modules you can add a force_cat argument to defeat this. I'll add a similar argument to, and . Will that help? Benwing2 (talk) 04:51, 8 May 2021 (UTC)
 * I wanted to test out the new changed sa-root and its new parameters along with some other templates. A similar argument to cat, cln and top would mean that the categorisation would work only if an extra parameter is specified. Wouldn't it be better to have a mainspace sandbox? Also 🔥 शब्दशोधक 🔥 05:22, 8 May 2021 (UTC)
 * Let me instead see about making categories work in WT:Sandbox. Benwing2 (talk) 07:31, 8 May 2021 (UTC)
 * That would be really nice; after all WT:Sandbox is for that purpose! 🔥 शब्दशोधक 🔥 07:40, 8 May 2021 (UTC)
 * Let me know if it works now. Benwing2 (talk) 07:50, 8 May 2021 (UTC)
 * It's working, thanks a lot! But then it will categorise it into "English lemmas" all the time, it that okay? 🔥 शब्दशोधक 🔥 07:54, 8 May 2021 (UTC)
 * We'll have to live with that. Periodically people can erase the contents if it bothers them. Benwing2 (talk) 07:58, 8 May 2021 (UTC)
 * Ok, thanks. Though can you help me with sa-root? for tidiness here. 🔥 शब्दशोधक 🔥 08:00, 8 May 2021 (UTC)
 * Can you give me an example where a space is needed? Benwing2 (talk) 08:05, 8 May 2021 (UTC)
 * For example at तुष्टि : From . . should give "From . Synchronically analysable as..." instead of "From .Synchronically analysable as..." 🔥 शब्दशोधक 🔥 08:09, 8 May 2021 (UTC)
 * I mean some pages that use where a space is needed. Benwing2 (talk) 08:06, 8 May 2021 (UTC)
 * Should be fixed. I moved the call to to the end. Benwing2 (talk) 09:08, 8 May 2021 (UTC)

Image captions
I was surprised by the caption on the image in, which reads "Teledu clyfar". Only by going to did I discover that "clyfar" means "clever", and noticed that "teledu clyfar" is there as a red-link, but is glossed as "smart TV". This caption does not seem very helpful to me, but rather than editing it I went looking for a policy on image captions, and found none - though I did find some discussions on particular questions about captions).

My feeling is that the caption should read simply "Teledu" - it's not relevant that it's a smart TV. Alternatively, if "Teledu clyfar" is kept, then either "clyfar" should be wikilinked, or the whole phrase should be glossed in English. (Arguably, "teledu clyfar" should be listed as a derived term there too, but that's not about the caption).

I've brought this up here rather than at the Tearoom, because I'm startled not to find a policy about what image captions should contain. --ColinFine (talk) 14:55, 8 May 2021 (UTC)
 * If there's only one sense, and the image is simply illustrating/depicting that sense (as it should) there's no need for a caption at all. – Jberkel 19:50, 8 May 2021 (UTC)
 * That makes sense, . But I'm troubled that (as far as I can find) there is nothing anywhere that specifies how or why images should be used in an entry, never mind what should be in the caption. --ColinFine (talk) 21:43, 9 May 2021 (UTC)
 * Maybe it's time to document some of the current practices then. There were a few related discussion recently which could be used as a starting point (Information desk/2021/February, Tea room/2021/April). My general impression is that many editors aren't overly interested in them. – Jberkel 21:58, 9 May 2021 (UTC)

Removing Module:root and using Template:root without it
, mainly for the purpose of long-lang-codes-made-short (like  to  ). I created root/sandbox for this and tested it at the sandbox (and it seems to word). 🔥 शब्दशोधक 🔥 07:40, 9 May 2021 (UTC)
 * also 🔥 शब्दशोधक 🔥 07:41, 9 May 2021 (UTC)
 * If you want to make changes to, it's better to modify the module than re-create it in template space. But I would be opposed to adding a bunch of hacks to this module for the sole purpose of saving a few keystrokes in typing language codes. Furthermore, if we want to create shorter aliases for language codes, it should be done in places like Module:languages/data2 instead of in Module:root, so the shorter codes are available everywhere; but I'd be opposed to that, I don't see the benefit of doing this just to save a few keystrokes. Benwing2 (talk) 18:59, 9 May 2021 (UTC)
 * I also object to shortening the codes for proto-languages. The code  applies to the Indo-European language family; it's different from , which is the Proto-Indo-European language. —Mahāgaja · talk 19:41, 9 May 2021 (UTC)

Major revision to
I carried out a major revision to, the meta template that controls the formatting of the of reference templates, in the following ways for the reasons stated:


 * The date of publication has been shifted from after the author’s name to after the publisher’s name. There have been complaints about the position of the date on and off for years, which led to the creation of a separate meta template . The proposed edit will make that template redundant. Also, having the date in its current position makes the template coding quite complex, which has led to an extra space sometimes appearing after the date which we’ve not been able to eliminate.
 * The relocation of the date also makes the reference templates somewhat more similar to the quotation templates.
 * The name of a translator of a work, if stated, has been shifted to the front of the reference, as there have also been comments that the translator should be given more prominence as the true “author” of a translated text. — SGconlaw (talk) 10:48, 9 May 2021 (UTC)
 * Surely that comment applies to quotations, where the usage of words and their syntax is what matters, rather than citations for information, where it is the facts that matter. RichardW57 (talk) 11:55, 9 May 2021 (UTC)
 * There have been calls to update the quotation backend too, but since that has been converted to Lua I no longer handle that. The rationale applies in both cases, though. — SGconlaw (talk) 13:30, 9 May 2021 (UTC)
 * Who are these people making complaints and calls? I oppose both changes above. The two most import elements of a reference is the author's last name and the date, and as such, should come first. This is in keeping with most academic sourcing. -- 17:15, 9 May 2021 (UTC)
 * Several editors over a period; I don't recall exactly who. I hope that they will see this discussion and comment. — SGconlaw (talk) 18:04, 9 May 2021 (UTC)
 * Like quotation templates, the templates would also support 2ndauthor, title2, etc.
 * 2ndauthor is a non-standard parameter. cite should be made to use author2. What is title2 for? -- 17:15, 9 May 2021 (UTC)
 * author2 to author5 still exist; 2ndauthor, title2, etc., are for the purpose of specifying a second work (see the documentation of ). This might be used, for example, if the content of a website (for example, Dictionaries of the Scots Language) is from a book (A Dictionary of the Older Scottish Tongue). — SGconlaw (talk) 18:04, 9 May 2021 (UTC)
 * What I'm saying is cite should conform to cite-meta as 2ndauthor is a non-standard parameter format. Also, we really shouldn't be creating of alternative parameter names unless absolutely necessary. -- 22:43, 20 May 2021 (UTC)
 * 2ndauthor isn’t an alternative to author2; it’s used for a different purpose. See . — SGconlaw (talk) 04:34, 21 May 2021 (UTC)
 * I see now what it's supposed to do -- still a bad choice for a parameter name. More to the point though, I don't think it should exist in cite-meta because we can do that much better with  --  04:44, 21 May 2021 (UTC)
 * I’m not following – what do you mean? (Regarding the parameter name, it’s already used at so I don’t think it’s a good idea to introduce a different name.) — SGconlaw (talk) 08:51, 21 May 2021 (UTC)
 * That standard parameter format is nameN. 2ndname, 3rdname, etc. is really bad parameter naming. I only care about cite-meta though, so I'll let someone else die on that cross. -- 09:57, 21 May 2021 (UTC)

has said that I should have sought consensus for these changes, and on hindsight I agree. Please do provide your views. — SGconlaw (talk) 10:48, 9 May 2021 (UTC)
 * This discussion is not attracting much attention. I've reinstated the non-controversial changes to the template as they are causing errors in some reference templates. — SGconlaw (talk) 19:52, 7 July 2021 (UTC)

Functioning of
Several reference templates weren't functioning as expected after the revision, and this was discussed at Template_talk:R:CDIAL. User:SodhakSH had to redesign the templates to resolve the issues. Kutchkutch (talk) 11:19, 9 May 2021 (UTC)
 * yes, some reference templates would need minor updates after the changes, chiefly changing url to entryurl. — SGconlaw (talk) 13:30, 9 May 2021 (UTC)
 * If url is removed, there should be a script run to prevent this. -- 17:15, 9 May 2021 (UTC)
 * It's not removed entirely. It would just be realigned so it would work in the same way as url in the quotation templates, and a URL would be linked to a chapter or entry using chapterurl or entryurl (as in the quotation templates). This would avoid unnecessary differences in the operation of parameters between the two sets of templates. — SGconlaw (talk) 18:04, 9 May 2021 (UTC)

ubor
I believe that all unadapted borrowings should be automatically categorized as learned borrowings (lbor). Or else we would have to separately categorize instances of unadapted borrowings as learned loans. - ⸘ - dictātor · mundī  11:09, 10 May 2021 (UTC)
 * I am already confused as to what your learned borrowings are. I often have Thai television on in the background, and I frequently hear words of English included in the broadcast, especially in current affairs.  Now, if these are not code-switching, are they learned borrowings, or will that depend on how well they are adapted to Thai phonology?  To complicate matters, Thai phonology is taking a hammering from English, with not only final fricatives, but even final consonant clusters. RichardW57 (talk) 18:24, 10 May 2021 (UTC)


 * Is the type of borrowing of going to depend on whether the '19' is pronounced more-or-less as in the English word or as in the Thai word?  I've even heard a faint final /n/ in the first syllable, an effect of Thais learning English. RichardW57 (talk) 18:24, 10 May 2021 (UTC)
 * Such loans from English cannot be learned borrowings, for sure. Using bor is fine. - ⸘ - dictātor · mundī  10:33, 12 May 2021 (UTC)


 * I find it weird to think of as a learned borrowing from French.  --Lambiam 16:28, 11 May 2021 (UTC)
 * I, too, wonder if we could have some clearer guidance as to what constitutes a learned borrowing, as compared to an ordinary (unlearned) one. I would regard as a learned borrowing from Latin as it is unchanged from its source language and only ever used in scholarly discourse, but am unsure whether the label should be applied in other situations (e.g., ?). — SGconlaw (talk) 05:07, 12 May 2021 (UTC)
 * is both a learned borrowing and an unadapted borrowing. There are quite a few terms that can be both, and hence is my proposal to automatically categorize all instances of ubor as lbor. (Admittedly though, ubor is not so important, but still there’s nothing wrong with it also.) As for Frenchisms, I am not so sure, though bor seems fine. - ⸘ - dictātor · mundī  10:30, 12 May 2021 (UTC)
 * oh, your proposal is just to add an additional category? No objection, I guess. I thought you were thinking of eliminating altogether and just having everyone use . — SGconlaw (talk) 13:03, 12 May 2021 (UTC)

Narrunga language
has been creating entries for this without a headword, which is a very bad idea. We don't seem to have a language code for it and I don't know enough about the language to figure out whether it's covered by an existing language code or to create one if it isn't. Pinging. Chuck Entz (talk) 12:46, 10 May 2021 (UTC)
 * , Category:Narungga language.--Tibidibi (talk) 12:48, 10 May 2021 (UTC)
 * : Thank you! I was wondering why I couldn't find any information in the usual places. Chuck Entz (talk) 12:57, 10 May 2021 (UTC)

So sorry for my mistakes. I'm using physical dictionaries for all of my entries. Will stop for the moment and try to get on top of the coding issues. Thanks for your help. (Peterbruce01 (talk) 04:57, 12 May 2021 (UTC))
 * , I fixed some of the entries you created (such as, , , , , ), and other editors also did likewise; but there are many more that need a cleanup. Could you please fix them yourself by seeing how your bad edits were cleaned up in the specified instances? Thanks. - ⸘ - dictātor · mundī  15:42, 10 May 2021 (UTC)
 * Some sysop please temporarily block this editor!— they are going on adding substandard entries. These entries will remain in their bad shape for years, and they are not heeding any request to clean up after themself. - ⸘ -  dictātor · mundī  09:14, 11 May 2021 (UTC)
 * Thank you for your response! Seems like you saw your notifications very late. Sorry if my behaviour was a bit harsh, but please ensure that you clean up all the entries you created in bad shape, sooner or later, based on how other editors fixed some of them (using the head template, using headers, etc.). Feel free to ping me in the edit summary (by User:Inqilābī ) for any problems with the formatting details. Thanks again! - ⸘ - dictātor · mundī  05:26, 12 May 2021 (UTC)
 * I need to delete a false entry ("mudlhakanti", a misspelling). How do I do this, please? Peterbruce01 (talk) 04:12, 13 May 2021 (UTC)
 * Apply . Recent misspelling by the author is generally accepted as grounds for a 'speedy' deletion. RichardW57 (talk) 07:43, 13 May 2021 (UTC)
 * . I went ahead and deleted it. For future reference: when you're the only one who's made any substantial edits and no one is likely to object, just tag it with delete or its alias d. If the entry is okay aside from the spelling, you can always move it to the correct spelling (provided there isn't already an entry there). That will leave a redirect behind, but you can tag it for deletion if necessary. Chuck Entz (talk) 04:27, 13 May 2021 (UTC)

Ngadjuri
We don't seem to have the either, although it has its own ISO code (jui). Am I, too, missing something? Thadh (talk) 14:32, 11 May 2021 (UTC)
 * We simply never got around to assessing whether it was different enough to merit inclusion when the ISO code got added. Given that it's extinct and poorly documented, I would lean toward including just to be conservative. —Μετάknowledge discuss/deeds 16:53, 11 May 2021 (UTC)
 * Yes, let's add it. Or what would we consider it to be a variety of if we didn't add its own code? Looking at the history of the ISO code, added in 2012, we may have just not noticed when it was initially added and then (when I did note its absence on WT:T:LT) didn't get around to assessing it, as Meta says. - -sche (discuss) 02:15, 12 May 2021 (UTC)

I'm using the dictionary in "Ngadjuri, Aboriginal People of the Mid North Region of South Australia" (2005). I tried the code jui but it obviously didn't work. Thank you for your circumspection.(Peterbruce01 (talk) 04:57, 12 May 2021 (UTC))
 * has been added with the canonical name "Ngadjuri", so it should work now. —Mahāgaja · talk 09:21, 12 May 2021 (UTC)
 * Brilliant, thank you. Will get on to all the Ngadjuri entries today. Peterbruce01 (talk) 23:46, 12 May 2021 (UTC)

Hi again. I'm keen to transcribe the (short) Nukunu dictionary by Louise Hercus onto Wiktionary. Do I need to reference this dictionary with every entry? Such referencing doesn't seem to be the case across Wiktionary, although the editor of the Barngala dictionary, Ghilaad Zuckermann, is asking me to do this when I get around to transcribing his work. Peterbruce01 (talk)
 * Please do. Note that most dictionaries and even individual dictionary entries are copyrighted. Do you have some explicit proof that this is freely licensed? Once you can show that (e.g. by having someone publish this notice on a web page), then it would be easy to create a template that you can add to every new entry--it's not a lot of work, just adding something like at the end of every entry. —Justin ( koavf ) ❤T☮C☺M☯ 00:40, 24 June 2021 (UTC)
 * I'm pretty sure we don't need the dictionary to be out of copyright to use it as reference, it's rather we can't copy-paste its data ("Note that copyright law governs the creative expression of ideas, not the ideas or information themselves. Therefore, it is perfectly legal to read a dictionary entry, or other work, reformulate it in your own words, and submit it to Wiktionary...")[1] So feel free to create R:nnv:Hercus:1992 (compare also other reference templates, like R:aa:Parker:1985), referencing shows that the data presented is verifiable and that it's attested. Cheers! Thadh (talk) 07:48, 24 June 2021 (UTC)
 * Thadh, of course it's legal to cite or publish a reference to a copyrighted work. Peterbruce01 was asking if he could transcribe that entire dictionary here, entry-for-entry, which would not be acceptable unless that dictionary was freely licensed or in the public domain. —Justin ( koavf ) ❤T☮C☺M☯ 07:59, 25 June 2021 (UTC)

Why do we include rhymes?
Out of interest, what is our rationale for including rhymes? People constantly complain about the inclusion of anagrams, saying they are only useful for word games etc., but whom are rhymes useful for? Poets? Why are poets more deserving than word-game players? (I do not want to remove rhymes but I feel there is a double standard at play.) Equinox ◑ 20:02, 10 May 2021 (UTC)
 * Sometimes, a rhyme line is the only line in a pronunciation section and thus gives pronunciation information alone. Then, it might help to recognize words that one has misheard or lead to the correct word like, or help resolve manuscript corruptions like . Fay Freak (talk) 14:56, 11 May 2021 (UTC)
 * You could also argue that this has future historical value. Knowing which words were considered to have rhymed, at different points in the past, could be valuable to linguists interested in reconstructing spoken language from written language. Khromegnome (talk) 10:10, 12 May 2021 (UTC)
 * Any "future historical value" would only occur if the reader can date the rhymes. Presumably the default is to look at the article history to know when the rhyme was added, but that restricts the history to start from only a decade or so ago.  What about words that rhymed in the year 1066:  should they be included in current articles (with appropriate annotation)?  What about ten (or more) years from now when several words that currently rhyme no longer do:  are they to be deleted?
 * —DIV (1.129.106.0 05:27, 17 June 2021 (UTC))


 * Would you agree that rhymes can be distinguished from anagrams by not being a pure "function" of the word? That is, it's clear how you would write some code to populate an anagrams section (and that code even exists). I assume it'd be a lot more difficult with rhymes. Khromegnome (talk) 10:03, 12 May 2021 (UTC)


 * Yes, definitely: I think I've made the function/transformation point myself in the past. This makes them less "lexically interesting" than rhymes, but isn't itself an argument to omit them. Equinox ◑ 16:20, 12 May 2021 (UTC)
 * I think it does put them into a large class of such “functions”, though, and it’s hard to justify including this whole class, or singling anagrams out for inclusion. I think this might be why the anagram sections seem jarring to me. I think things like “This list could be generated on the fly (with a cache if you want), so why bother storing it on the page?” and “If anagrams are here, why not, say, all words formed by extending this word?”. Khromegnome (talk) 09:02, 18 May 2021 (UTC)


 * A minor point, but I think you're mischaracterizing the main objections to the inclusion of anagrams. Being useful for word games is presented as an argument for keeping the lists, not removing them. Khromegnome (talk) 10:08, 12 May 2021 (UTC)


 * Well, I said that people complain they are only useful for word games, i.e. that word games are not enough. But it's natural to resort to a dictionary when doing crosswords etc. Equinox ◑ 16:20, 12 May 2021 (UTC)


 * FWIW, I use Wiktionary all the time to look for rhymes, when writing poetry. If you think it's not worth having, there might be more people who write rhyming poetry than you think (mostly students, probably). I've also used the namespace for trivia purposes, like to verify common claims along the lines of "nothing rhymes with _____".
 * As far as rhymes in pronunciation sections go, I think it's most useful for making people aware of the main rhyme pages. I wouldn't have found out about them otherwise. I don't think it would be super useful to mention rhymes in entries if the Rhymes namespace was more visible. Andrew Sheedy (talk) 19:06, 18 May 2021 (UTC)


 * I'm not sure why we include rhymes or, for that matter, why we include synonyms because we suck terribly at both. I use Word Hippo for synonyms, and for rhymes I usually just google because there are plenty of good rhyming dictionaries out there like Rhymer, Rhyme Finder, What Rhymes With, and even one that's a wiki. But yes, I do occasionally have the need, when I'm flexing my creative juices I guess (can't recall exactly what for).
 * Wiktionary is absolutely superior at defining new terms, better than Urban Dictionary, and at translations. How many dictionaries do you know that translate more than one other language? This website is a hidden treasure that not nearly enough people are aware of, and yet tons of people are familiar with Urban Dictionary somehow. Improving the purposes here that are lacking would strongly suggest and not just hint at a single stop for every need. It would allow for domination of the space, much in the way that Wikipedia dominates its space.
 * There's not nearly enough thought put into this, as it's been established very firmly through inheritance of the wiki framework, and through our layout as a result, what we should be, with little wiggle room. Better we should have not just a rhyming dictionary, but a rhyming dictionary across languages. Likewise for an acronym finder. Neither of these would be specific to English or any particular language.
 * We should have a way of specifically identifying each sense, easily accessible to any editor, and then building synonym lists automatically off of that. We should have tabs for each of these, and not for Talk. But at the moment, we do not excel at these, we underwhelm. DAVilla 04:52, 19 June 2021 (UTC)


 * Re: "I'm not sure why we include rhymes or, for that matter, why we include synonyms because we suck terribly at both." You realise this means we should just pack up and go home, because OED is better at definitions. Equinox ◑ 22:12, 6 July 2021 (UTC)
 * I'd say we're doing pretty good on alternative forms, better than most. --Geographyinitiative (talk) 00:16, 7 July 2021 (UTC)


 * On what basis? If I look up an everyday word in the OED like dog then the headword will probably contain a million forms like dog, dogge, doge, dag,... whatever, but they have dates attached to them. We just have an "alt forms" header and sometimes there is no clue that these are even obsolete forms! NNES learner would see some of our shitty entries and think "oh, so catte is totally acceptable 21st-century form of cat". It is that bad. Equinox ◑ 00:23, 7 July 2021 (UTC)

Add a note on pages with Lua memory errors (2)
I created a discussion at the Beer parlour in January (and the errors will likely not go away soon), however, no one replied. As readers constantly ask us whether we know about the errors (and likely many who do not ask are confused), if no one opposes, I will proceed with adding notes. We should not have unexplained errors in our entries. J3133 (talk) 16:51, 12 May 2021 (UTC)
 * No objection. — SGconlaw (talk) 08:12, 13 May 2021 (UTC)
 * Would “Regarding the Lua memory errors on this page, see Lua memory errors.” at the top of the page be sufficient? J3133 (talk) 17:43, 15 May 2021 (UTC)
 * wondering if you should make it into a template using . The text seems fine. — SGconlaw (talk) 17:48, 15 May 2021 (UTC)
 * Template:Lua memory errors. J3133 (talk) 18:00, 15 May 2021 (UTC)
 * Wouldn't it be better to have the memory error message itself link to the page? —Mahāgaja · talk 18:03, 15 May 2021 (UTC)
 * Sorry, I don't I understand. — SGconlaw (talk) 18:08, 15 May 2021 (UTC)
 * At the moment, there's just a red message saying "Lua error: not enough memory". Couldn't it say "Lua error: not enough memory. See Lua memory errors for more info"? —Mahāgaja · talk 18:28, 15 May 2021 (UTC)
 * oh, you mean edit the red error message itself. I have no idea how that can be done though, or whether it can be done at Wiktionary (a Phabricator ticket might have to be filed). And if the message resides outside Wiktionary, it would not make sense for it to refer to a Wiktionary page. — SGconlaw (talk) 18:39, 15 May 2021 (UTC)
 * see https://en.wiktionary.org/wiki/Special:AllMessages?prefix=scribunto&filter=all&lang=en&limit=50, or specifically: MediaWiki:Scribunto-lua-error. I suppose one might include logic that added the desired wording depending on the value of $2, but I don't know if that's available to the code one would use in that context. You should probably bring it up at the Grease Pit, where it will be seen by people who are familiar with handling interface-message syntax. Chuck Entz (talk) 19:56, 15 May 2021 (UTC)
 * Cool, so it can be done on our end. — SGconlaw (talk) 20:10, 15 May 2021 (UTC)

Apparently it hasn’t been done yet...? — MuDavid 栘𩿠 (talk) 00:59, 28 May 2021 (UTC)
 * As some users wanted to change the message instead, I refrained from adding the template, however, it seems no one is changing it. J3133 (talk) 01:22, 28 May 2021 (UTC)
 * You’ll need to ask someone who knows how to edit modules; I don’t. — SGconlaw (talk) 04:20, 28 May 2021 (UTC)
 * As it is not my proposal, I am not sure why I need to ask, unless you are referring to MuDavid who wants to see it done. J3133 (talk) 04:28, 28 May 2021 (UTC)
 * This has very little to do with the workings of modules. It's a matter of the workings of the Wikimedia system messages. Judging by what was done with the categorization of module errors it would seem that some of the same syntax that's used in templates may be available. The problem here is that this syntax would need to be able to tell an "out of memory" error from other types of lua errors. I have no idea what variables are available to this particular system message, or what might go wrong if there's an error in an error message. Chuck Entz (talk) 04:53, 28 May 2021 (UTC)
 * is this something you can help with? — SGconlaw (talk) 10:27, 28 May 2021 (UTC)
 * Thanks to whoever made this change. It should both reduce the number and increse the average quality of BP inquiries about entries with such errors. DCDuring (talk) 17:59, 28 May 2021 (UTC)
 * Tried checking the text of the message supplied to MediaWiki:Scribunto-lua-error and it seems to work. Now the pages with memory errors have lots of links to Lua memory errors. To change the out-of-memory message, after this change, non-interface administrators can edit Template:Scribunto-lua-memory-error. — Eru·tuon 18:04, 28 May 2021 (UTC)
 * great, thanks! — SGconlaw (talk) 19:11, 28 May 2021 (UTC)
 * Thanks, we’re making progress. ☺ Now let’s find a way to get rid of those errors permanently. MuDavid 栘𩿠 (talk) 01:47, 29 May 2021 (UTC)

Glossing as both "idiom" and "slang"
WF often adds "idiom" glosses to things already glossed "slang". Isn't slang always idiomatic, thus this is redundant? (I think he only does it trying to get on my nerves.) Equinox ◑ 10:59, 17 May 2021 (UTC)
 * Getting on Equinox's nerves is always a satisfying byproduct of Wonderfool's editing, but on these occasions it wasn't the main reason for her adding the idiom tag. You can't have one-word idioms like conky or trendwhore. Indian subcontinent (talk)

Is not such practice, a blatant attempt at vandalism?

 * No, because your edits did not comply with Entry layout, and introduced meaningless content into the entry. — SGconlaw (talk) 21:22, 17 May 2021 (UTC)
 * You introduced into the entry honorificabilitudinitatibus the following:
 * “Exceptional extraordinarily long outstanding abilities to achieve the state of being of such honorificabilitudinitatibus.
 * [ The Honorific Loop:
 * Exceptional extraordinarily long outstanding abilities to achieve the state of being of such exceptional extraordinarily long outstanding abilities to achieve the state of being of such ...
 * - Loop to Honorificabilitudinitatibus ]”
 * That seems meaningless to me, and to the other editors who reverted your edits. You also made changes to the format of a quotation to one which we do not use, either in entries or here on this discussion page. Finally, you added the plural form ”$DEITY”s (including the quotation marks) to the entry $DEITY. I don’t see any evidence for such a strange plural form. — SGconlaw (talk) 04:37, 18 May 2021 (UTC)

Arrgh, this tiresome incident: on these covert vandals!

 * Translation: anyone who doesn't let me do whatever I want is a vandal, and getting rid of the ugly argumentative graffiti I put at the top of the page in a dictionary is "impeding the consensus reaching progress".
 * As for notifying functionaries, a pattern of:
 * violating community rules and practices
 * edit-warring against the admins that the community has chosen to enforce those rules and practices
 * starting out by calling everyone vandals before discussing anything, and not discussing content.
 * using a content page for discussion that belongs on a talk page
 * is going to get you blocked on any wiki, and the functionaries are just as likely to step in to stop you and clean up the damage as anyone else.
 * As for your content: this is a descriptive dictionary. We describe the language as it is used, as clearly and succinctly as possible. "Exceptional extraordinarily long outstanding abilities to achieve the state of being of such honorificabilitudinitatibus" is extremely verbose, ugly, and hard to read, as well as being circular. It has all the marks of someone trying very very hard to show how clever they are.
 * This is a wiki, "the dictionary anyone can edit". That means you can add content, but that also means others can remove it. You have yet to explain why your content makes the entry any better. I'm not saying that you have to do that before you add it, but when challenged, you should at least try.
 * Also, you keep invoking WT:NPOV, which is completely irrelevant here. As long as the entry isn't slanted one way or another, we can and should have opinions about what content should be included and about matters of style and presentation.
 * So far, all I've seen from you is bluster, accusations and unreadable walls of horribly-formatted text. If your goal is to demonstrate why you should never be allowed to edit anything visible to the public, you're doing a marvelous job. Chuck Entz (talk) 22:48, 30 May 2021 (UTC)

Disagreement over the Classical Latin pronunciation of non-Classical lemmas
I'm having a disagreement with user over this and previous edits: Special:Diff/62545149. He keeps incosiderately reverting my edits, claiming that the word being Late Latin and "vulgar" means that the usual Classical pronunciation shouldn't be included.
 * Firstly, to my knowledge no such policy exists, and it would be untenable, as by that logic many post-Classical words shouldn't have a Classical pronunciation. Yet this is the standard pronunciation today and anyone wishing to use these words may want to know how to pronounce them (concerns regarding eclecticism and "drawing on the Latin of all eras" notwithstanding).
 * Secondly it's based on an arbitrary assumption that register is tied to pronunciation, which I don't share. If we suppose that 6th century CE Romans had more than one pronunciation, I would not support arbitrarily dividing Latin words into ones they would supposedly be more likely to pronounce in one or another pronunciation. I think such a situation is sufficiently covered by including the Classical pronunciation, and some appropriately chosen putative local variant (in this case Roman because the cited inscription is from Rome).
 * I believe the Ecclesiastical pronunciation is best excluded because I believe it would be totally ahistorical to try and pronounce substandard Late Latin in a pronunciation stemming from the turn of the 20th century, and these type of words don't come up in liturgical Latin, hymns etc where this pronunciation has found its home.
 * Finally, I'd like to point attention to the belligerent attitude, lack of humility or respect, and a general aversion to civil discussion shown by comments like "Keep reverting if you like, you will not change anything", as well as the whole situation of aggressively revering an informed user's edits without any attempts to contact them and reach consensus first, unfortunate traits that have resulted in me coming to blows with this user in other internet venues. Brutal Russian (talk) 03:07, 18 May 2021 (UTC)


 * The form capus (alteration of the original Classical caput 'head') is attested in the sixth century AD, not any earlier, and yet you are attempting to give it a reconstructed pronunciation from the first century b.c. By the sixth century the Classical system of phonemic vowel length was well and truly dead. That is not merely my opinion, that is very much the communis opinio of scholarship in Romance Linguistics.


 * What I have added is not meant to be an Ecclesiastical pronunciation, rather it is meant to represent the popular pronunciation of Late Latin as reconstructed by comparative Romance Linguistics. Lengthening of stressed vowels in open syllables in the aftermath of the collapse of the Classical system of phonemic vowel length is a well-established fact.


 * Incidentally, it was you who first came to the page titled capus and undid the pronunciation that I had put there, without bothering to "contact [me] and reach consensus first".


 * Finally, pardon me if I find it difficult to be polite to someone who has so consistently been toxic and condescending to me, and incidentally several other people as well, on the 'other online venues' that you allude to. Perhaps you have really have changed in the last few months, but I have no a priori reason to believe it. The Nicodene (talk) 03:22, 18 May 2021 (UTC)
 * I don't understand why you feel the need to explain to me when the word was attested and things like open syllable lengthening. You understand perfectly well that I already possess this information, and I've taken it into consideration when forming my opinion, which I've laid out above. I would like you to re-read my actual arguments and address them.
 * I removed the pronunciation as part of the general conversion of the entry from a confused mix of Classical and "Vulgar Latin" into a normal Latin entry. To repeat what I say above and in the edit summary: all Latin words theoretically deserve to have some Late Latin pronunciations. The trouble is that there wasn't a standard Late Latin pronunciation, as I hope you realise, and so the choice of such a pronunciation becomes a difficult one. Is being attested at Rome a good enough reason not to give all the other possible Late Latin pronunciations of the word that has passed into every single Romance language? I don't believe so; my belief is that Late Latin pronunciations should be either automatic or exceptional. Even overlooking this problem, I don't believe the proposed pronunciation was correct, specifically the presence of the final -s; I doubt that the word in question had a subject case at all, when the current thinking in Romanistics is that only animate nouns had it - the S might simply be hypercorrect. Even details like the open syllable lengthening are not as secure as you seem to be implying. Have you read Loporcaro 2015? He's the person who propounds OSL as a pan-Romance phenomenon, and in this he disagrees with numerous previous studies, which makes the issue far from settled. Accepting Loporcaro's arguments still doesn't settle the precise dating of OSL in different regions, making it a matter of guessing. I belief that guessing is best left to the readers.
 * All in all, as I mentioned in previous discussions, I think proto-Romance reconstruction should be kept out of Latin as far as possible in absence of a strong consensus and firm references. 6th century Latin at any rate is firmly in the realm of "anybody's guess" - even the morphology and syntax of that variety is in the air! As my final edit shows, I'm fine with having a rough approximate of this particular word simply as a compromise to avoid further disagreement, but this should firmly remain an exception as long as we don't have a systematic, automated transcription. Otherwise all the entries will have different transcriptions into the Bulgarian Latin of the author's choosing (recall the mess that the Vulgar Latin la-IPA used to be).
 * I will finish by asking you again to engage with the arguments in my original comment and explain specifically why the Classical pronunciation should be excluded, and from what other words/layers of vocabulary you believe it should also be excluded. Brutal Russian (talk) 04:07, 18 May 2021 (UTC)


 * A Classical pronunciation should not be given for a word that is clearly not a Classical word. To do so is quite clearly anachronistic.


 * You point out that there wasn't a "standard Late Latin pronunciation" as if precisely the same is not true about the Classical pronunciation. Clearly we have to do the best with the material that we have.


 * If you think that /s/ in word-final position had already fallen in Central Italy by this period, I would encourage you to read the statistical analysis of Politzer (1953: 34–35), in which he establishes that its regular loss rather took place in the eighth century, sporadic loss or assimilation in earlier periods notwithstanding.


 * Your claim that the "current thinking" in Romance Linguistics is that inanimate nouns lacked a subject case entirely by the sixth century AD is dubious. A look at the grammar of either Old French or Old Occitan—ideally both, and for bonus points also the Sursilvan dialect of modern Romansh—would quickly disabuse you of that notion.


 * Loporcaro disagrees with older scholarship on the issue of OSL, yes, but as far as I am aware there is no modern scholar who denies the overall phenomenon, whatever disagreements they may have on certain details of it. (For instance, perhaps the view that in Iberia, or at least the future Castile, all stressed syllables must have lengthened, because later diphthongization operates without regard for syllable structure, may yet have some adherents. That is, however, entirely irrelevant to the issue at hand, namely the pronunciation of capus.)


 * As it happens I too have a distaste for the term 'Vulgar Latin' and would prefer that it not exist. There was simply Latin, tout court; that there were differences of register or style is only what one would expect of any language, particularly one with a long literary or poetic tradition. When Romanists reconstructed the pronunciation of so-called 'Proto-Romance', what they had really done, without realizing it, was discover the pronunciation of Late Latin, as Roger Wright has pointed out (2002: 199).


 * Capus will remain an exception until I get around to programming a system for outputting reconstructed Late Latin (~'Proto-Romance') pronunciations.


 * Politzer, Robert L. 1953. Romance trends in 7th and 8th century Latin documents. Chapel hill: University of North Carolina Press.
 * Wright, Roger. 2002. A sociophilological study of Late Latin. Utrecht: Brepols.
 * The Nicodene (talk) 05:16, 18 May 2021 (UTC)
 * But anachronism is real, and some “truly dead systems” are popular, in spite of being dead. If we now know how the Classical pronunciation was it follows that the 6th-century Romans could also know it, as we only know it from sources existing then already, which have become less rather than more over time, counterpoising the circumstance that the methods of learning have improved. And if they were cognizant of it they also realized it. Potuit, decuit, ergo fecit. Fay Freak (talk) 21:17, 18 May 2021 (UTC)
 * What we now call the Classical pronunciation is a very recent (in historical terms) reconstruction by modern linguists. It does not at all follow that a typical Roman, even an educated one, living in the 6th century AD would be any more aware of the Roman pronunciation of the 1st century b.c. than a typical person today would be aware of the pronunciation of English in the fifteenth century. The Nicodene (talk) 21:28, 18 May 2021 (UTC)
 * But why do you assume that there was no such person beyond the wits of today’s lackwit? Why focus on typicity? We also describe the atypical, if it’s kind of normative. So that it is more commendable to give a Classical pronunciation of a New Latin word than the typical English-accented one. The norm is now rarely fulfilled, yet the bare frequency of pronunciations is exactly not of utmost importance. Fay Freak (talk) 21:43, 18 May 2021 (UTC)
 * If by some miracle there was a precocious chap in ancient Rome who had a thorough understanding of diachronic linguistics on-par with that of modern specialists, he has unfortunately left us no record whatsoever of his existence or his ideas. Such a person, furthermore, would hardly have written capus for caput. The Nicodene (talk) 21:58, 18 May 2021 (UTC)
 * Yet you dodged the answer on what to do with Modern Latin words. It is not aloof to give Classical pronunciations for them, as the judgment of those who understand rather than those who do not master the language may be decisive. The more so that, as you say, it is “a very recent reconstruction” by the moderns. We want that, and we want it indiscriminately of the period a word is attested it—distinction would only make it complicated and it is natural from this understanding of today to pronounce any Latin word, in general, with this same pronunciation, and it is not wrong but rather it is odd to try to switch chronolects while pronouncing based on attestation periods and assumed pronunciations of each. So we just answer our readers the question how they are supposed to pronounce it if they want to adhere to Classical pronunciation rules, in place of pronouncing every word differently which is unrealistic, anachronistic, just because of the attempt to adhere to realities varying by age. Your concept of “reconstruction” thus comes up short as the whole picture is Frankenstein's monster. Reconstruction of Classical Latin pronunciation is not interpreted as “the here given pronunciation of this particular word definitely happened” (to be honest with many Latin words from scientific Latin no pronunciation ever happened, yet we assume one). The pronunciation line is an indication, not an allegation. Fay Freak (talk) 23:09, 18 May 2021 (UTC)
 * What was there to dodge? You hadn't even asked a question about 'what to do with Modern Latin words' in the first place.
 * The modern reconstruction of a 1st century b.c. pronunciation has added another way of pronouncing Latin to what was already a somewhat confused mess (although previously the traditional Latin pronunciations were at least fairly consistent within each country, even if not across different ones). I don't think any of them would be wrong for a modern neologism.
 * That said, all such pronunciations would be wrong for a sixth-century 'vulgar' form like capus, as well as others like it, none of which are 'Modern Latin' in any sense of the term. Let's not lose sight of the point of this thread. The Nicodene (talk) 23:37, 18 May 2021 (UTC)
 * Do you believe the Classical pronunciation should be removed from all post-Classical (including Medieval and New Latin) entries?
 * I believe it's possible to pintpoint one exact pronunciation that can be called Classical, the urban speech of Rome around the time of Vergil, which is what we have been aiming to transcribe under that moniker. There existed other concurrent pronunciations that can be called by different names other than Classical. I don't think something comparable can be found for Late Latin.
 * I do not agree with your comparison of the situation Late Latin and modern English. I believe the situation in Late Latin was much closer to that of diglossia, more comparable to modern Arabic, or at the very least the German-speaking world. Lack of vowel lengths in these spoken varieties don't preclude the existence of vowel length when speaking Standard Arabic or Standard German. This situation is not dependent on the existence of modern linguistics - the pronunciation of Classical Sanskrit has been preserved for over two thousand years by oral tradition, that of Homeric Greek for at least a thousand. Wright's suggestions are stimulating but not conclusive.
 * I would object to outputting phonetic Late Latin pronunciation in Latin entries in the absence of a comprehensive reference, also because their number could be arbitrarily large. Brutal Russian (talk) 22:23, 18 May 2021 (UTC)
 * The Classical pronunciation is entirely inappropriate for terms that were first used in Medieval or Early Modern Latin, yes.
 * Why would you think that doing precisely the same isn't possible for Late Latin?
 * The idea of an early Latin diglossia, even before the fall of the Empire, is an unfortunate relic of a more primitive era in linguistics that is dying a slow, but steady, death in modern scholarship. There is nothing to substantiate its existence in that era. (Comments from Cicero et al. about a sermo uulgaris simply indicate differences in register, not language.) Incidentally the concept of 'Vulgar Latin', which you apparently reject, is the main pillar holding up the diglossic model, which you apparently believe in.
 * For comprehensive references, see the articles I wrote on Proto-Romance (defined there essentially as Late Latin prior to major regional splits) and the sound changes that produced it.
 * The Nicodene (talk) 23:17, 18 May 2021 (UTC)
 * This is clearly against the currently adopted policy and I strongly doubt you will find anyone agreeing with you on this. Why, by this logic, would we list the Roman Ecclesiastical pronunciation for terms coined before the publication of the first Liber Usualis where it was codified (1934 or so)?
 * Also replying to you replying to below, I think you're fundamentally misinterpreting what Latin is and isn't. Unlike Middle Dutch, Latin is not exinct and is a language in contemporary use. The Classical pronunciation is the de-facto idealized standard pronunciation for the contemporary use of Latin. Therefore not giving a Classical pronunciation for later-coined words would be precisely the same as not giving modern Dutch (or English, or Russian) pronunciation to words coined before that pronunciation appeared. This is the reverse of the Roman Ecclesiastical pronunciation. Or you can equate it completely with the Roman Ecclesiastical and claim as you already have that it's "a very recent  reconstruction by modern linguists" on par with the any other post-Renaissance pronunciation.
 * Uhm, because there seems to have existed no such supra-regional pronunciation, unless it be precisely the scholastic, MSA/Standard German-like one, tied to the literary tradition. Which is what Classical already transcribes. If you're aware of some description of a supra-regional Late Latin pronunciation different from Classical, I'll be glad if you point me towards it.
 * I really have no idea what Cicero has to do with this at all. How can you appeal to the sociolinguistic situation in the 1st century BCE as an argument for the linguistic situation in 6th century CE?? I don't reject the concept of Vulgar Latin, I reject the use of this term which lacks any definable underlying concept. Most of 20th century scholarship uses it to refer to the exact same entity you're referring to as Proto-Romance (Grandgent and the whole of Vulgar Latin).
 * Proto-Romance, as adopted in most of the materials you seem to be citing in your article (such as DERom), is an entirely abstract reconstruction, using the comparative method and nothing but it, that does not profess to reconstruct a language variety that ever existed. It's been purposefully divorced from Latin and makes not using any written Latin evidence as one of its founding principles. This approach is a diglossic approach at its very core, but it jettisons the entirety of linguistics that the comparative method is unable to reconstruct, and the resulting entity is not a natural language, but a set of results of comparative reconstruction, a language in the most abstract sense of a set of lexical items and grammatical rules not used as such by anybody. This contrasts sharply with Late Latin as a physically attested, undeniably extant linguistic entity.
 * The belief that Proto-Romance is Latin is incompatible with using no attested Latin to reconstruct... Latin? These two entites are as conceptually divorced as they can possibly be. Proto-Romance is a theoretical linguistic demonstration without any claims regarding sociolinguistics or anything else it doesn't barter in. I can't bring myself to re-read that Do Romanists Need... article I linked you, but I believe that's essentially what it says as well.
 * What reference attempts to reconstruct phonetic shapes of either Late Latin or Proto-Romance? I don't believe this has been the goal of any Proto-Romance reconstruction.
 * Finally, I believe you're confusing the concepts of written, attested Latin, with Late Latin as its temporally subsequent variant, and Proto-Romance; and I firmly believe that Late Latin phonetic transcriptions belong to the reconstructed mainspace and not to mainspace Latin entries.
 * Brutal Russian (talk) 02:36, 19 May 2021 (UTC)
 * Simple: because Classical words continued to exist after Classical times, so giving them later pronunciations in addition to earlier ones makes sense. You cannot take what I'm saying, reverse the arrow of time, and pretend like I should agree with the resulting statement.
 * I think you are greatly overestimating the prevalence of the reconstructed Classical pronunciation among modern speakers of Latin; a trip to Italy, France, or Spain would suffice to show you that it is far from being the 'de-facto idealized standard'. Anyway, giving a Classical pronunciation for a neologism that actually exists in Modern Latin is reasonable enough. Note however that the form capus, or other Late Latin 'vulgar' forms like it, do not exist in Modern Latin, so this is essentially irrelevant to the topic at hand.
 * How would you argue for the existence of a "supra-regional pronunciation" for the Classical period in a way that cannot be used to argue the same for the Late Latin period? You (this is a general 'you') have arbitrarily picked a period (1st c. b.c.) and a place (Rome) and worked with what evidence exists to reconstruct a sketch of the pronunciation found in that approximate time and place. Exactly the same can be (and has been) done for another approximate period and place, so long as a suitable corpus of evidence exists.
 * The fact that older scholars like Grandgent labored undered that misconception does not in any way affect the validity of the evidence they have adduced regarding ongoing sound changes.
 * The part that you are missing is that my disclaimer that you have read on that Wiki page also applies to your reconstructed Classical pronunciation. It is a "simplified abstraction", and it is most certainly "should not be taken as an exact equivalent" to the pronunciation of Latin on a specific date, in a specific place, or in the mouth of a specific person. All we can do is work with what data we have to make informed guesses.
 * As mentioned above, the validity of the work of older scholarship on, and examination of evidence for, post-classical phonetic changes in the spoken language does not depend in any way on what label they chose to use for the latter. It is not 'diglossic at its core': it just so happens that many of its early pioneers were, for lack of a better term, diglossicists (which is only to be expected, as the diglossic model had yet to be seriously challenged at the time). Incidentally, the DERom itself essentially sees its reconstruction of Proto-Romance as a sort of reconstructed sketch of Late Latin (and interestingly applies many of its conclusions, such as an affricate realization of intervocalic /-b-/, even to Classical Latin after examining data from studies on Proto-Italic).
 * What you are not seeing, under the opaque mess of terminology, is that 'Proto-Romance' fundamentally is Late Latin.
 * The Nicodene (talk) 03:39, 19 May 2021 (UTC)
 * Let me stop you before you inevitably say 'No, Late Latin refers only to the literary language of late antiquity' [1]. Late Latin can and does have more than one meaning, and incidentally this general medley of amorphous and shifting definitions (Late Latin, Vulgar Latin, Proto-Romance...[2]) is one of the many unfortunate side-effects of the diglossicist model. (A model which, incidentally, both Dworkin and Eva Buchi reject. Do re-read that .pdf if you find the time.) It would be best, in all likelihood, to simply speak of Classical Latin and Late Latin, to which an additional, and optional, qualifier can be added if the register needs to be specified.
 * [1] No doubt followed by some claim along the lines of 'you are simply confused' or 'you are misunderstanding [insert elementary concept]'. I should like to point out here that you have already betrayed your ignorance of the general topic at hand by making the outlandish claim that 'the current thinking in Romanistics is that only animate nouns had [a subject case by the sixth century]'. No, it quite simply is not. One might add that 'Romanistics' isn't even the name of the field.
 * [2] You said something earlier that I would like to address now. No, Proto-Romance is not purely reconstructed from comparative Romance data, pace Robert Hall, whose own reconstruction of it was essentially a slightly-modified version of Classical Latin. (He went so far as to 'reconstruct' all six CL cases for Proto-Romance.) In reality, any serious overview of Proto-Romance makes extensive use of evidence from the Late Latin period. The Nicodene (talk) 09:57, 19 May 2021 (UTC)
 * I will take a bit of time to lay out some choice quotes from Dworkin's article, just for the record.
 * In the very first paragraph on the first page, he says:
 * “…specialists in Romance historical linguistics are blessed (or are they cursed?) by having at their disposal an abundant documentation of various written registers of the source language, namely Latin […] almost all Romanists realize that the relatively uniform written Latin of these texts offers a fuzzy and distorted image of the diverse regional and social linguistic realities of the Spoken Latin of the Roman Empire […] just as written modern standard French or standard English do not faithfully represent the multifaceted realities of the spoken languages.”
 * On page five we find some more illuminating quotes:
 * 1) “… the Proto-Romance bases [in the DÉRom] appear in phonemic transcription, since what is being reconstructed is a linguistic form of the spoken language, of which the written Latin form is often a very imperfect representation.”
 * 2) “[Written] Latin and Proto-Romance (in essence, spoken Latin as reconstructed through the comparative method) are in reality two different registers of the same linguistic system.”
 * On page six, continuing into page seven:
 * “Perhaps the written Latin material should not be included in the Romance data set used in the first round of reconstruction. However, the written Latin forms can then be used to modify and nuance as needed the results of the purely Romance-based reconstruction […] The comparative method is not an absolute formula set in stone with immutable procedures, and was not originally designed for languages that had the good fortune to have a well-documented register of the source language. Romance linguists should take advantage of all the data at their disposal. In my view, Romance linguists are blessed rather than cursed by having available written Latin."
 * On page 13:
 * “This protolanguage in its narrowest sense (“protoroman sensu stricto”, to use Buchi’s terminology) is the language spoken throughout the extensive Roman Empire prior to the beginning of what Romance scholars view as the beginning of the political, with the consequent linguistic, fragmentation, of the Empire…”
 * It should be abundantly clear by this point that, first of all, neither Dworkin nor Buchi subscribe to the diglossic model, and second that the DERom's Proto-Romance is not a purely abstract system, divorced from any real-life existence, and that it is rather, to quote Dworkin yet again, ‘spoken Latin as reconstructed through the comparative method’ that has been informed by a thorough consideration of real evidence from written (or carved) Latin.
 * As a disclaimer, no, I do not agree with everything that Dworkin or Buchi say. On this much, however, I do.The Nicodene (talk) 11:07, 19 May 2021 (UTC)
 * In response to your first bullet point:
 * Not all words continued to exist after the Classical period. But how would we know which ones did and which ones didn't? If attested, yeah, later realizations should be listed. If not, we should probably not list them (in the same way we shouldn't give later realizations of Ancient Greek words if we're sure they fell out of use.)
 * I just thought of something. If both caput and capus stayed in use throughout "Late Latin" and "Proto-Romance" and whatnot (and specifically doesn't originate from a set sound change that took place widespread,) we should consider it an alternative form of the word, and thus should treat it as if it had been such throughout history, listing it for all eras. If, however, it either naturally evolved from caput or gradually took over caput in many or all subsequent languages, we should treat it as a separate word without giving the Classical pronounciation. The key here is that there is already an equivalent Classical form: caput. For words that don't have a Classical equivalent, the non-Classical pronounciation would be disfavored in modern times, instead choosing a Classical interpretation. So capus, from caput, would not need the Classical pronounciation, while formaticum would.
 * 110521sgl (talk) 15:33, 20 May 2021 (UTC)
 * I fully agree.
 * Capus, and its other inflections like accusative *capu(m), does in fact appear to be the ancestor of all the Romance descendants. Supporting evidence:
 * No Romance descendant contains /t/ or a consonant derived from /t/. One might have expected e.g. Spanish *cade, from a hypothetical Old Spanish *cabde, from Latin capite, ablative of caput. Cf. codo < cobdo < Latin cubitum, and also nombre < nomne < Latin nomine, originally the ablative of the neuter noun nomen 'name'.
 * All descendants are phonetically consistent with a derivation from capus (or acc. *capu).
 * No Romance descendant reflects a feminine derived from the Latin plural capita 'heads' (the plurals of original Latin neuters are often reinterpreted as feminine nouns).
 * The Old French nominative singular chiés 'head' still retains the final /s/ of capus.
 * The Old Occitan caps (see §100 here) also retains the /s/ of capus. (Inaccurate: see below.) The Nicodene (talk) 19:27, 9 June 2021 (UTC)
 * The Nicodene (talk) 18:15, 20 May 2021 (UTC)
 * As a nitpicky aside, wouldn't the expected Spanish result be caude, not cade? Compare < . Hazarasp (parlement · werkis) 09:17, 29 May 2021 (UTC)
 * @Hazarasp Ah good point. The stressed vowel in cobdo must have simply absorbed the /b/ (after vocalization to [w]) due to both being labial, and this wouldn't be the case for *cabde. The Nicodene (talk) 17:59, 31 May 2021 (UTC)
 * What etymological dictionaries did you consult when reaching these conclusions? Brutal Russian (talk) 12:52, 9 June 2021 (UTC)
 * @Brutal Russian Looking at the REW entry, there do seem to exist derivatives of caput and capita in Sardinian and Romanian. I was not aware of that. The Nicodene (talk) 19:17, 9 June 2021 (UTC)
 * I agree with Brutal Russian here. Just like in, for example, Middle Dutch, where, no matter where in the period between Middle Dutch and contemporary Dutch a word first appeared, the standard spelling and pronounciation of the oldest form of Middle Dutch is given, to my knowledge. With Latin, and its over a millennium of usage, I could see it reasonable to give an approximation of the changes in pronounciation over time, like with Ancient Greek. But, since the "Latin" we're talking about regarding the header of the word, with its ISO code and its position in time and space indeed being Romae anno Vergilio, refers more to specifically that variant, any words should also be given with the pronounciation tied to that variant. We consider - everywhere except for in linguistics - all non-ecclesiastical Latin to have been pronounced the same way (even though most people mispronounce it due to maternal tongue or ecclesiastical influence), regardless of its progress in evolution. Late Latin is a variant of Classical Latin in terms of classification on Wiktionary, so I propose giving the Classical Latin version first and then - like with the variation present in the different forms of English - also including the assumed Late Latin version of the relevant region, with a note like "(Late Latin, Rome, 6th century AD)" to clear things up, adhering both to the idea of Latin being a single language code requiring a single base phonology, and the realization of the word in that it was pronounced with variant phonetics considering its place in geography and history. I'll be happy to hear your thoughts. 110521sgl (talk) 22:57, 18 May 2021 (UTC)
 * What he is proposing to do would be the equivalent of giving an eleventh-century pronunciation to a Dutch word first attested in the late 1600's.
 * Incidentally I rather like what wiktionary has done with Greek words and would simply like to implement the same system for Latin, though probably on a more limited scale. First would come the Classical (1st c. b.c.), if the word actually existed in Classical times, or soon after. Next would come the Late transcription, reflecting pronunciation around the year 400 or so. Perhaps even a third stage could be added for the period around 700, but by this point Latin (~Romance) has already broken into dialects, so the pronunciation will have to be split on a very approximate regional basis. (Intervocalic voicing of /p t k/ for France, for instance.) And this transcription would only be appropriate for words that were actually coined in that period, for instance formaticum 'cheese' ~*[for'mad(e)go].The Nicodene (talk) 23:48, 18 May 2021 (UTC)
 * Yeah, I agree with your reasoning. If a word was only coined around that 700 AD mark, I wouldn't include the 400 AD hypothetical precedent pronounciation. But, especially in scholarly environments, the current accepted classroom pronounciation of words - regardless of the first language of the scholar - mimics that of the Classical era. And since these words still constitute the Latin language, in its current form, people would probably want to know how to pronounce its words in practice. If you would want to talk about cheese in Latin today, you'd use the hypothetical Classical pronounciation, even though the word didn't exist back then. The way I see it, we should give the progression of the pronounciation starting from the first coinage, like with Ancient Greek, but also, if the Classical era predates its first use, the equivalent Classical pronounciation, since that is the current "living" form we would attest in a speaker today. 110521sgl (talk) 14:40, 20 May 2021 (UTC)
 * It may not be correct, but people often want to know the "Classical" pronunciation for most Latin words, though maybe not so much for an intermediate form that is morphologically unlike its ancestors or descendants. Chuck Entz (talk) 15:10, 20 May 2021 (UTC)
 * To provide some background info, the (Classical) Latin word for cheese is caseus, and it does survive quite widely in Romance languages, ranging from Spanish queso to Romanian caș. The competitor formaticum or formaticus appears to have arisen at a fairly late period in France (first attested in the year 795 according to the TLFi) and it displaced caseus as the word for 'cheese' throughout Gallo-Romance. That caseus had existed before that period in France, as elsewhere in the Latin-speaking world, is supported by 'relic' forms such as casier 'cheesemaker'.
 * Anyway, if a speaker of Modern Latin wants to refer to cheese, he or she will naturally reach for the word caseus. I at least would be very surprised to hear someone say formaticum, as only a small subset of Latinists (those informed about the diachronic evolution of Latin into French or Occitan) could understand it. The Nicodene (talk) 18:29, 20 May 2021 (UTC)


 * I agree with Chuck and 110521sgl's comments from the 20th. IMO, we should include the "Classical" pronunciation (where it is unambiguous what it would be) because it's what (enough of) the sort of people who still pronounce Latin would use and/or be looking to learn from our entries. I'd be find with some kind of footnote indicating that the word didn't exist until after the end of the period when the Classical pronunciation was widely used by native Latin speakers. - -sche (discuss) 02:01, 25 May 2021 (UTC)
 * @-sche The problem is that the 'vulgar' formaticus did not exist in Classical Latin, nor does it exist in Modern Latin, where a classical pronunciation might be used. Giving that word a classical pronunuciation is anachronistic no matter how you look at it.
 * If you object to giving the word an approximation of the actual pronunciation that the late eighth-century scribes who first wrote it would have given it, then perhaps it's best to simply give no pronunciation at all. The Nicodene (talk) 17:23, 31 May 2021 (UTC)
 * It's worth considering the sociolinguistic situation in the eighth century. No Romance spelling system existed yet, so no matter what one wanted to write, one had to write it "in Latin", even if the word in question never existed in Roman times. Hence a scribe trying to spell a newly invented word such as */forˈmadʒ(ə~o)s/ 'cheese' would have had no choice but to write as something like ⟨formaticus⟩ (probably the 'best' spelling, as it accurately reflects the etymology of each morpheme) or ⟨formadeus⟩, etc. The Nicodene (talk) 18:14, 31 May 2021 (UTC)
 * As Brutal Russian has tried to tell you many times already, Latin and Old French have, had, to be distinguished. “You are substituting definitions“ in Tea room/2021/June In France itself they had to have a way to pronounce Latin that is not French. It is worth considering copyist practice: How did they copy manuscripts? One bloke read aloud a manuscript to a squad of others. This caused -tio and -cio endings to be not distinguished while copying, but I cannot discern that formaticus was considered to have the pronunciation, though the spelling formadeus existed. Nor did one guy read in Old French and the others write it down in Latin with all endings, as such copyists often not even understood the language they copy, and such translation would not even provide for the same results (sufficiently), so would not even be copying. Spellings do reflect pronunciations of words used in speaking Romance, but once they are out in Latin they do not needs get pronounced again as in the source language. What would they have pronounced it in far Germany or Poland anyway, if they got hold of the manuscript? Such a reader likely hasn’t even heard French spoken, the same way a 16th-century German has heard more Latin than Italian. There were some standard pronunciations back then: though they have been untrue to Classical Latin, they weren’t exactly Romance either. And now the pronunciation one uses when one gets hold of such a manuscript tends to be a Classical Latin one.
 * You should explain what the purpose of giving “modern Egyptological” pronunciations for Egyptian words is, which we have even when ancient ones are given. You see here how your argument that word X is “not used” in modern Latin does not work, because “Modern Egyptian” does not exist. Hint: These modern pronunciations are owing to how a spelling may be transmitted by speech across the learned. Fay Freak (talk) 02:49, 3 June 2021 (UTC)
 * It is not necessarily true that two highly distinct systems of pronunciation, the 'vulgar' and 'cultivated', always coexisted. Wright (1982) examines the evidence—including e.g. rhymes, spelling mistakes, comments by grammarians, and such— and finds no evidence for such a division in pronunciation in Romània prior to the Carolingian Renaissance. In fact the general idea had occurred even to Grandgent over a century ago (1907: §218), although the cited comment focuses specifically on the pronunciation of vowels. A bit later Mildred Pope (1934: §§646–647) expanded on the very same idea.
 * That said, the Carolingian Renaissance occurred toward the end of the eighth century, and the first attestations of formaticus also happened to occur at about that time, so one might ask whether it could have been given a reformed (not classical) pronunciation based on spelling, which would have simply amounted to /for'matikus/. The problem with this though is that the Carolingian reforms aimed at a wholesale 'classicization' of the literary language, and that would have included replacing 'vulgar' terms such as formaticus with the 'correct' term caseus. That said, it is conceivable that someone could have, nevertheless, used the 'incorrect' vulgar word but still given it the reformed pronunciation. Note however that, although we keep referring to formaticus, various other spellings are found, such as, indeed, ⟨formadius⟩ or also ⟨formacius⟩, which give an idea of how the word was really pronounced. Scribes were not always able to think of the correct latinized spelling of popular terms that had not existed in Classical Latin and hence did not have a traditional spelling that they could use for reference. Cf. ⟨anoget⟩, ⟨iornalis⟩, ⟨sorcerus⟩, and ⟨impruntatum⟩ in the Reichenau Glossary, terms which 'should' have been spelled, according to their etymologies, as *inodiat, *diurnalis, *sortiarius, *impromutuatum.
 * I suppose a German scribe who had never been to France, if he found himself needing to read aloud the word formaticus in some document from there, would have been forced to give it a spelling-pronunciation, and assuming they got the stress right, that would indeed be /for'matikus/. Still, we do not include German pronunciations of English words, for instance, on Wiktionary entries, since non-native pronunciations are not considered relevant.
 * We've allowed Mon pronunciations of Pali, even though Pali has long lacked native speakers, and according to some, never had any. (I don't like having them, but that's because it opens the door to a flood of local pronunciations.)  More locally, what then are church Latin pronunciations doing on Wiktionary? --RichardW57 (talk) 06:10, 4 June 2021 (UTC)
 * Latin ceased to be a native language, eventually, and so Church Latin pronunciations do make sense for Latin entries in general.
 * What is being discussed here is an edge case: a non-standard local term attested for a couple of centuries in France that never existed in Classical Latin and never made it into any 'proper' sort of Latin. It basically amounts to a latinized spelling of a local Gallo-Romance word (one of many), hence my reference to native speakers. A German who had never been to France would not have known the 'popular' i.e. native pronunciation of ⟨formaticus~formadius~formacius⟩. The Nicodene (talk) 07:54, 4 June 2021 (UTC)
 * Since there is not any evidence of a separation between Latin and Romance prior to the ninth century (and even long afterward the distinction remains murky), for our purposes the French scribes writing formaticus in the eighth century were native speakers.
 * Egyptian is not my area of expertise, but as I understand Egyptological pronunciations are used by modern scholars for convenience when discussing ancient history. There is no similar convention in Romance Linguistics for pronouncing Latin terms in general. For instance Martin Maiden pronounced genus alternans in a recent seminar as /gɛnəs æltɜːnɑns/ (that is approximately how I remember it), meanwhile an Italian colleague might have rendered it as /dʒɛnus altɛrnans/, etc. Incidentally I have yet to hear a classical pronunciation used in such a discussion. The Nicodene (talk) 21:21, 3 June 2021 (UTC)
 * I think we're going to have to do a vote on this, or at least split it in multiple different parts. We've got past issues that have accumulated coming together here, edge cases from a troublesome era causing problems, a huge discussion in the BP of last month that hasn't given us an answer, and many differing and opposing views on the matter, all at once. Either we need to take a step back and look at the original issue and decide on a course of action, or we need to settle the underlying conflicts concerning Latin in the first millennium AD. Regardless of whichever one takes preference, this subject clearly needs a resolution. 110521sgl (talk) 15:15, 14 June 2021 (UTC)
 * I think we're going to have to do a vote on this, or at least split it in multiple different parts. We've got past issues that have accumulated coming together here, edge cases from a troublesome era causing problems, a huge discussion in the BP of last month that hasn't given us an answer, and many differing and opposing views on the matter, all at once. Either we need to take a step back and look at the original issue and decide on a course of action, or we need to settle the underlying conflicts concerning Latin in the first millennium AD. Regardless of whichever one takes preference, this subject clearly needs a resolution. 110521sgl (talk) 15:15, 14 June 2021 (UTC)

Is attestation required for comparative and superlative?
See PG-13. The comparative and superlative were removed, but they are linguistically correct and do exist in use (these examples are mostly not "permanently recorded"): I know MTV targets a younger, more PG-13 audience these days, Can Josh Jackson be a little more PG-13?, It seems to lend itself to being more PG-13 because it’s family-oriented, I’ve been trying to be more PG-13, less rated R, AOL's considerably more PG-13, Just had the most PG-13 campaign ever with Bretonia, Oof those three things are the most pg-13 parts of the film compared to the rest lol, This could be the most PG-13 story I’ve ever had on @MorningEdition, What is the most PG-13 thing done in the Pokemon anime?, One day, one or two seasons from now, they’re going to fall in love and have the most PG-13 of sex scenes and it will be highly satisfying for audiences everywhere, with Milo and Atticus fending off an entire Roman battalion as the city cheers on at the local arena, the two muscle-bound warriors cutting their way through many a throat in the most PG-13 way imaginable.

I thought we only needed attestation for the base word. Otherwise, will we also remove conjugations that fail attestation? Alexis Jazz (talk) 11:19, 19 May 2021 (UTC)
 * I don't if we have a policy or even an established custom in cases like this, but in my opinion, these examples are sufficient to restore the comparative and superlative to the headword line. —Mahāgaja · talk 13:30, 19 May 2021 (UTC)


 * We don't require attestations for every single inflected form if there are citations enough to demonstrate that a word inflects, and there just happen to not be examples of the third-person singular feminine past habitual form (but we know what it must be). But if no comparative / superlative is attested at all, that'd be what ("not comparable") is for. (And in the case of Latin, we've removed inflected forms if they're not attested and it's not clear what they'd be, like when it's clear a word inflects but it's unclear which of the various possible genitive endings it takes.) Assuming, for a moment, that there were a word where there were no durably-archived cites of the comparative/superlative, but there were internet cites like the ones above, I'd be inclined to leave it alone but IMO it would/should fail if someone wanted to be a stickler and RFV it. In this case, though,  finds enough books that I don't see what the issue here is...? (I also see a use of "the tight silvery top was the most PG-13 thing she owned".) - -sche (discuss) 01:53, 25 May 2021 (UTC)
 * , I had missed that, thanks! Alexis Jazz (talk) 02:27, 25 May 2021 (UTC)
 * Did we establish a procedure for challenging alleged inflections? I've a feeling  and  have strong opinions on that.  And do we have method of establishing the general forms of the 3sf past habitual when the textbooks disagree? --RichardW57 (talk) 23:33, 2 June 2021 (UTC)
 * The inflection class needs to be clear. We should somehow know, or have reason to tendentially assume, that what we give is morphologically sound. This should be enough to list the inflections – inflection presentation cannot depend on whether a certain form has been written down, it would be abhorrend, contrary to lexicography tradition, and incompatible with the fact that inflections are partially unwritten (as with most endings in Arabic verbs and nouns). It is also licit that for a word we only have inflected forms attested and the lemma form is assumed, but this does not make it a term to be starred because this would imply the whole thing is unattested (often in Old Armenian entries here; the attestation situation sometimes demands more than a star for clarification). A very easy view, meseems (man opposed Votes/pl-2020-04/Attestation of comparatives and superlatives because it tried to make things more problematic than they are).
 * It is only that in case of comparatives and superlatives, they should not be automatically created in so far as some may be inverisimile. This is similar to our recent discussion on including Classical Latin pronunciations for Latin words not attested for classical Latin. Classical Latin pronunciation is the now standard pronunciation and dictionary users want it. So theoretical comparatives and superlatives may be given unless they are unthinkable because of being contrary to the sense of the adjective, so in Latin where the language has some formal peculiarities how they are formed and a learner wouldn’t necessarily guess them because he wouldn’t know the rules. (Not so Arabic elatives, as they are not as widely used and detached or separate lexical items from adjectives they relate to, so they aren’t automatically part of adjective inflection tables.)
 * This is to say that the thinkability is enough reason to list them but not to star them, although knowledge of our not being cognizant of their actual use is reason for preventing their creation by bot. Here we distinguish the positive from the possible and the never ever: a question of prognosis arises in so far as degrees are not positively demonstrated, by reason that their absence cannot be empirically disproven so as to claim incomparability and because starring inflected forms is confusing because readers do not know what is meant with it: if we had particular Wiktionary rules about it, it would only be private language readers unfamiliar with these rules would have difficulties with.
 * Starring and attestation attempts have limited reliability. To reformulate in short again: Dan Polansky forgot how the works here: If a degree is not attested this does not mean conversely that the adjective is “incomparable”. You cannot derive an inability from inactuality though you can conclude an ability from its actuality. And you have this problem to solve as always: How you deal in presentation of the details that you leave open, while already having sufficient information for an entry. Saying nothing would be harmless, as distinguished from claiming incomparability which is not empirically or apriorically proven, but then the reader asks what’s the craic about the unshown forms, so technically we always have to say something about them, the question is how frank the editor can be in an adjective entry and how he can be frank, about what he has witnessed and what he has but assumed—for which purpose no formal language exists yet. As sweepingly assuming “incomparability” based on not seeing attestation would make entries mendacious, as I have demonstrated, while starring inflected forms is ambiguous, as well as not starring and omitting. Fay Freak (talk) 01:05, 3 June 2021 (UTC)
 * I agree with the part of Fay Freak's post that I'm able to comprehend (yes this is a wish for a more straightfoward syntax and a less arcane vocabulary :P). A full range of inflection should be assumed unless given evidence to the contrary. Likewise, the most productive type of inflection or derivation should be assumed. This is basically Fay Freak's 'thinkability', the technical term being 'grammaticality'. We've fixed some Latin entries that have very clear-cut and well-known syntactic limitations, namely that the words are highly idiomatic and participate in a limited number of constructions ; this has nothing to do with some sort of taboo, ambiguity etc. most of the time, although there's a known phenomenon where languages avoid forms that strain or violate phonotactics (euphony), or where there is uncertainty about the correct formation: Russian has numerous., I don't recall seeing Latin forms removed due to uncertainty over their formation. Could you perhaps give a couple of examples? —On the topic of which, I have a big issue with this template: la-epithet. It basically limits the syntactic usage of random common nouns in Latin based it seems on their non-occurrence in English! This makes zero sense to me and I know no precedents of a language where taxonomy is indeclinable, normally or otherwise, or is syntactically limited in any shape or form. Brutal Russian (talk) 01:31, 3 June 2021 (UTC)
 * On Talk:synaeresis it was argued that some of the inflected forms were not certain, that they could be the forms given in the table but could also be other things depending on what declension category / set of endings the word used (which was argued to be not known with certainty). It seems they have been re-added to the table, hopefully based on evidence that they are correct or must have existed. The locative plural was removed from Talk:humus on the grounds that it wouldn't exist. In general, if a Latin word foobarus were only attested in the nominative and accusative and it was thus not clear whether the genitive would be foobarī or foobarūs (second or fourth declension), I hope we wouldn't just blindly guess! - -sche (discuss) 19:27, 3 June 2021 (UTC)
 * Ah, this issue is actually bigger than just this one word: Latin made use of Greek in unadapted and adapted form. The unadapted use included using Greek declension as well as pronunciation (word accent) - regardless of the alphabet used, although naturally Greek was the default in this case. The Latin use involved Latin declension and accent placement. This is explicitly stated by writers on grammar, and so the current treatment is a confusion not simply of different paradigms, but in truth of different languages. A separate Attic or mildly Latinised pronunciation should probably also be supplied to go with the Greek declensions. Though this would make Latin have the phonetic pronunciation that the Ancient Greek entries themselves currently lack, heh. As for the declension class problem you mention, I believe it's only theoretical - I've never come across it in practice; in case I do, there are objective criteria to decide this, such as that the 4th declension was generally recessive and limited to certain semantic fields (tree names, verbal nouns); also, etymology (expected formation) and of course other dictionaries can help decide this. Brutal Russian (talk) 23:14, 6 June 2021 (UTC)

Who are the Election Volunteers in your community?


Would you like to get more people taking part in the Wikimedia Foundation’s Board of Trustees election?

The Wikimedia Foundation Board of Trustees announced the plan for the 2021 Board elections. That plan includes outreach and communication support for the Board elections. The Board election facilitators will:


 * Inform communities of the trustee selection process
 * Invite communities to engage in voting
 * Encourage people representing emerging Wikimedia communities to run as candidates

Voter turnout in prior elections was about 10% globally. It was better in communities with volunteer election support. Some of those communities reached over 20% voter turnout. We know we can get more voters to help assess and promote the best candidates, but to do that, we need your help.

We are looking for volunteers to serve as Election Volunteers. Election Volunteers should have a good understanding of their communities. The facilitation team sees Election Volunteers as doing the following:


 * Promote the election in their communities’ channels
 * Organize discussions about the election in their communities
 * Translate messages for their communities

'''Who are the Election Volunteers to connect your community with this movement effort? Is it you? Or someone you know? Check out more details about Election Volunteers''' and add your name next to the community you will support in this table or get in contact with a facilitator. We aim to have at least one Election Volunteer for Wiki Projects in the top 30 for eligible voters. Even better if there are two or more sharing the work.

Best, JKoerner (WMF) (talk) 18:12, 21 May 2021 (UTC)

definite/indefinite articles in gloss(-like) definitions
There seems to be lack of agreement whether these should be used or not, and even individual editors can't seem to decide on it. Special:diff/62455375 removes them from the Latin definitions while supplying them in cog traslations. It looks especially wierd in extended glosses and with relative clauses such as "part of the cheek that..."; they seem to be desirable in cases where their removal may change the meaning, such as "a mouth/head" vs "the mouth/head", "flight" (fleeing) vs "a flight" (on a plane). English entries seem to vary, but for ex. uses it in those confused capital-glosses:, and quite expectedly not in  , a difference that exists for a reason. Admittedly, this creates some questions with glossing languages that have articles, since "the mouth" is  in Italian but  in Latin, "a bird" is Pt. vs just - but again, En. is currently "A weapon"... Brutal Russian (talk) 13:08, 23 May 2021 (UTC)
 * Can’t comment about other languages, but for English I think the use of indefinite and definite articles helps to show whether a term is countable or uncountable, for one thing. — SGconlaw (talk) 09:29, 25 May 2021 (UTC)
 * If we follow the practice of starting our English definitions with a hypernym, then an indefinite article (or sometimes any) is appropriate for all definitions of countable nouns. I'd prefer that no English definition of a common noun used initial the. Eg, I would reword the first two definitions at arm. DCDuring (talk) 07:58, 26 May 2021 (UTC)

Ancient Greek noun declension tables
Does anyone else think that having articles and transliteration for every single form clutters things up unnecessarily? Dngweh2s (talk) 00:06, 24 May 2021 (UTC)
 * I believe it's convention to include transliterations within the declension tables of languages that use non-Latin scripts; compare the Hittite and Sanskrit declension tables. Hazarasp (parlement · werkis) 00:37, 24 May 2021 (UTC)

Niya Prakrit
I'd like to propose a new code for Niya Prakrit, closely related to but distinct from Gandhari. It was first proposed in 2018 by DerekWinters. Although gandhari.org treats Gandhari, Niya Prakrit and the language of the Ashokan edicts at Shahbaz Garhi and Mansehra as a single language, T Burrow and others distinguish between them. It's also evident with some of the forms viz. Niya mṛda vs Gandhari muda (both < OIA mṛtá). I notified User:AryamanA on Discord and he agreed. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 02:19, 25 May 2021 (UTC)
 * Why not have a label pgd and a Category:Niya Gandhari? 🔥 ಶಬ್ದಶೋಧಕ 🔥 09:31, 25 May 2021 (UTC)
 * Although the resources about Niya Prakrit are scattered, there should be a code for Niya Prakrit. Apparently, AryamanA didn't create a code for Niya Prakrit after DerekWinters proposed it at  User_talk:AryamanA/2018.
 * According to R:CDIAL and R:inc:CGMIA, Niya Prakrit appears to have a distinct identity, and it is associated with the Tarim Basin in China. Kutchkutch (talk) 09:44, 25 May 2021 (UTC)
 * How does Turner abbreviate Niya Prakrit? 🔥 ಶಬ್ದಶೋಧಕ 🔥 09:54, 25 May 2021 (UTC)
 * NiDoc Kutchkutch (talk) 10:01, 25 May 2021 (UTC)
 * Oh nice, "Language of `Kharoṣṭhī Inscriptions discovered by Sir Aurel Stein in Chinese Turkestan' edited by A. M. Boyer, E. J. Rapson, and E. Senart". I didn't find that as I was searching for "Niya". 🔥 ಶಬ್ದಶೋಧಕ 🔥 10:04, 25 May 2021 (UTC)
 * OK, thanks. I'll put together whatever material I can find and make a code for Niya - by next week or so. I won't be editing much for the next few days so I won't be able to find the time to do this now. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴) 10:53, 25 May 2021 (UTC)
 * I'll make the code over the weekend. Does  sound good? It will be a descendant of (North Western) Ashokan Prakrit, just like Gandhari. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴)  01:48, 4 June 2021 (UTC)
 * According to User:Benwing2's suggested naming scheme at User_talk:Benwing2, wouldn't the code be something like  or  ? Kutchkutch (talk) 09:31, 4 June 2021 (UTC)
 * Please use . I am planning on renaming all the remaining Prakrit etym codes to follow the same convention. Benwing2 (talk) 05:55, 5 June 2021 (UTC)
 * OK,  it is. I'll get around to it. -- 𝓑𝓱𝓪𝓰𝓪𝓭𝓪𝓽𝓽𝓪(𝓽𝓪𝓵𝓴)  06:32, 5 June 2021 (UTC)

FYI: New lexicon of ancient Greek to English
https://www.theguardian.com/commentisfree/2021/may/28/the-guardian-view-on-the-art-of-lexicography-ancient-greek-lives-on —Justin ( koavf ) ❤T☮C☺M☯ 21:48, 28 May 2021 (UTC)
 * Thanks for the heads-up. 212.224.226.12 23:16, 29 May 2021 (UTC)
 * Good to know, but this is apparently a scrap of a dictionary, for widely read texts and not rare words or sourced, so likely it will not find use here. They talk about Liddell-Scott-Jones but fail to mention (though one comment there does, but who should read comments?) the Brill Dictionary of Ancient Greek, which by its number of cites (three, by now) man uses too infrequently. Fay Freak (talk) 00:24, 30 May 2021 (UTC)

Why occasional new dictionaries are required for dead languages
I liked the suggestion made here: |here that, apart from occasions when further research suggests more-accurate definitions, the living language being translated into also moves on, both in changes of word meaning and in the acceptability of using English words (rather than Latin/Ancient Greek which will not arouse the servants) to describe taboo concepts. That seems so obvious in hindsight but, partly perhaps because I don't normally edit words in dead languages, I'd never thought about it. So it interested me, and I mention it here in case anyone else is interested. To precis it:

The author mentions in passing that pudenda literally means shameful things, so I'm glad that isn't a word I use! (Thinks: It may possibly explain some oddities in Vulgate-influenced bible translations.)

As an example of the need to issue new dictionaries of dead languages, it cites an Ancient Greek word λαικαζεῖν, which was defined in a dictionary edited in the 1930s as to wench, with classicists then generally considering it a synonym of βινεῖν, defined at the time (in an Ancient Greek/English dictionary) as coire of illicit intercourse, coire being itself defined by a 1982 Latin/English dictionary as to have sexual intercourse, or more succinctly by the latest Cambridge dictionary, as to fuck.

Even between academics in 1930, such things were spoken of in Latin, with A E Housman writing to the lexicographer Henry Stuart Jones that he didn't think “there is any place where λαικαζειν means wench”; it meant “fellare”, though certainly not “cunnum lingere”. Although that was ignored at the time, in 1980, a Professor H D Jocelyn apparently wrote 54 pages on the subject, and the latest Cambridge dictionary defines λαικαζεῖν as to perform fellatio -- though what is wrong with to fellate, I don't know. Perhaps there was some white space to fill. --Eng in ear 00:44, 29 May 2021 (UTC)
 * More likely because is a back-formation from  and might therefore have been considered too neologistic. I notice my browser's spell-checker knows the word fellatio but not the word fellate. —Mahāgaja · talk 07:05, 29 May 2021 (UTC)
 * Yes, good point! I was being a little flippant -- it's not good for a dictionary to describe a word using other words that the average user will have to look up as well. I think I prefer fellate because it is a nicer-looking word (to English eyes) with a more-English sound to it -- Having said which, I doubt I've used either until writing these posts. --Eng in ear 06:28, 30 May 2021 (UTC)
 * I don’t think this is actually any weighty reason for such a new dictionary, this is only the marketing to be published by progressive newspapers, especially considering that it is a student edition and in the classroom one avoids the topics anyway, and on the other hand the students do understand these circumstantial wordings, as they have a penchant for old language and practically one continues to encounter them. Fay Freak (talk) 00:46, 30 May 2021 (UTC)
 * Yes, good points -- I had failed to consider that English people studying Ancient Greek can be assumed to be interested in archaic language! And I agree that the journalist was probably responding to a well-aimed press release.  I thought the more-interesting point was about English mutating -- eg a 1930 dictionary might well translate λογιστής as computer using the sense current at the time, to some confusion nowadays (if the word doesn't mean that, my apologies -- I've shamelessly taken it from the translations section of computer to Greek and confirming that its etymology to Ancient Greek implied the same).  And amongst people not attracted to archaic language, I remember my daughter, then aged 11 and egged on by an older cousin, coming up to me giggling and saying "Dad, did you know that ass giggle, giggle, can mean donkey!  And a vicar telling me, while agreeing that the King James bible sounded wonderful when read out loud, that a woman he knew had reached her 40s before it became clear that she had misunderstood the meaning of the "woman with an issue of blood" as if she was a "woman who had an issue about blood"
 * And there are other taboos which might well be discussed in class with late-teenagers, such as racist tropes. For example, Twain's Huckleberry Finn, where the hero is a runaway slave referred to throughout as a nigger, who gives himself up to save the life of his young white friend, and with appropriate karma, finds that slavery has been abolished while he was hiding.  Or the fact that Dickens originally wrote Oliver Twist (serialised) with Fagin disparaged as a stereotyped Jew, until a Jewish friend of his said how upsetting that was for him -- an outcome Dickens had never even imagined -- at which point, he immediately changed the language in the later installments and then rewrote the early parts before publication as a book.  (I accept these examples have nothing to do with dead languages, but I am sure there are others that do.  And anyway, studying those books in a non-English speaking country would be more difficult if the dictionaries were Bowdlerised -- a case where modern dictionaries might be less useful than 1930s ones.) --Eng in ear 06:28, 30 May 2021 (UTC)

An article about French Wiktionary semantic domains
Dear colleague,

During the past year, French Wiktionary made a large change about the indications of semantic domains (like biology, architecture, etc.). Before that, it was a rough list of 400+ templates with plenty opaque aliases. Now is one template and a module with a controlled vocabulary containing explanations used when hovering the label and for pages of categories listing words from a domain. It deals with super-categories automatically. It may be a first step in the direction of a template similar as the Template:label in use here. Our list of domains is here: fr:Module:lexique/data.

This project was made by four people: an intern with a training in lexicography to list and describe semantic domains; a contributor accustomed to deal with massive corrections, Sebleouf; a contributor with skills in development, Lepticed7 and myself to supervise the internship and to guide the mapping. It was a decent job and we are quite proud of the result.

We considered it was a new kind of lexicographic task and it may be of some interest for academic lexicographers, so we wrote an article for a conference. Sadly, a reviewer wrote: "Also, the English in the paper is so un-English that it is sometimes difficult to understand what the authors even meant." Not so much of a surprise, considering Academic English is almost a separated language from English and we are not fluent writing it.

So, if this topic seems interesting for you, or if you want to help some fellow wiktionarians to get this article published, we invite you to improve our draft. The deadline is very soon as we had trouble dealing with it before, so every comment or direct correct are welcome. Feel free to say when our English is broken and to modify the draft directly. If you have any question, we'll be happy to discuss this theme here with you!  Noé 08:45, 29 May 2021 (UTC)


 * I think you should be proud of this article too! I speak as someone with a good knowledge of UK English, but with little exposure to academic gobbledygook, though I do have to read a number of academic papers relating to engineering every year. I found it much clearer than most academic papers, and hopefully it will not become formalised to the point where it becomes less accessible to non-academics!


 * I was very surprised that you received such a negative review, to the point where I wonder if there was some academic rivalry involved -- when I was at school, in the '60s, we were told that the Académie Française used to object to every new word they hadn't invented themselves (exaggerating only slightly), in which case, they would certainly object to (particularly Anglophone) wiki-speak. And so might other lexicographers who were wiki-virgins, and believed that only experts could produce excellence.


 * There were a fair number of words rarely used outside of wiki's, eg transcluded, which could make it less readable to an academic who was unused to the field. I suggest asking a lexicographer unused to wiki's to review the text and mark such words, so that you can add an appendix defining the words/concepts that confused them.


 * I've made some minor stylistic suggestions but, as I said, the standard was already better than that of most academic papers I have read (though I suppose it's valid to hold lexicographers to a higher standard than the engineers whose papers I usually read)! But honestly, it is well-translated.  Don't let a reviewer with a grudge persuade you otherwise! --Eng in ear 04:53, 30 May 2021 (UTC)
 * Thank you very much for your precious advice and suggestions. We had a nice bonus conversation about the word shambolic and some connotation it have now in UK. Very interesting. I just sent the paper and I think it will be accepted for publication soon. When the online article will be available (under CC BY-SA 4.0), I'll update this post with the link! It was quite exhausting, but I am happy we shared our explanation as the process was time-consuming as well. On the matter of semantic domains, French Wiktionary was worst than English Wiktionary and it may be equal or even better now, thanks to the glosses that define each domains, used in the controlled vocabulary (the /data page) and in categories pages Face-smile.svg Noé 22:24, 31 May 2021 (UTC)

Hi Enginear and others colleagues, the article just got published online: https://elex.link/elex2021/wp-content/uploads/2021/08/eLex_2021_15_pp259-268.pdf on this page: https://elex.link/elex2021/proceedings-download/

Thanks again for your help  Noé 15:00, 31 August 2021 (UTC)

Verb or adjective?
I am conscious that parts of speech have been mashed around since I was at school, and I never really got the hang of gerundives, gerunds and participles anyway. Nor am I certain what our policy is, but IIRC, do we class gerundives as adjectives and gerunds as nouns, while participles remain verbs?

Anyway, what prompted me to check was: Looking at the quotations for lampshade currently #3 "To adorn with one or more lampshades", don't they all belong with a (missing) adjectival, rather than a verbal sense? (In any case, the 3rd quote is rather wonderful -- each time I read it, my head swam more -- perhaps if I had some of whatever the author was on, it would make more sense!) --Eng in ear 00:54, 30 May 2021 (UTC)


 * The 1913 citation says that the room "had...been...flowered, cushioned and lampshaded". It's saying that this has been done to the room, and is (presumably) still the current state of the room, so I think that must be a verb. The other two citations are "the lampshaded glow" and "a fallen lampshaded being", which look more adjectival. Equinox ◑ 01:06, 30 May 2021 (UTC)


 * Thanks; moved. --Eng in ear 21:13, 30 May 2021 (UTC)

"Everyone besides Larry was invited"
"Everyone besides Larry was invited." Was Larry invited, or was he not invited? If you speak English, and you are reading this, please give your answer. Equinox ◑ 10:01, 30 May 2021 (UTC)
 * It's difficult to tell - "besides" can mean "in addition to" as well as the more usual "except for". SemperBlotto (talk) 10:04, 30 May 2021 (UTC)


 * Yeah but "in addition to everybody" is logical nonsense. You can't have everybody plus a person, because everybody already includes the person. Blotto you okay? Equinox ◑ 10:09, 30 May 2021 (UTC)
 * I think that "everyone" can mean all the members of a group. Yes, I'm still alive. SemperBlotto (talk) 10:13, 30 May 2021 (UTC)
 * Page 593 of the CambridgeGEL reads

[62] i) a. Kim too resigned. b. Only Kim resigned. ii) a. “Kim resigned” b. “Kim resigned” iii) a. “Someone besides Kim resigned” b. “No one except Kim resigned.” Both [ia] and [ib] entail that Kim resigned: the obvious difference between additive too and restrictive only is shown in [iii]. But there is also a difference with respect to the status of the component propositions given in [ii–iii]. We saw that with only the main assertion is [iiib], with [iib] being backgrounded. With too, however, it is [iia] that is the main assertion, and [iiia] that is backgrounded. And in fact [iiia] (unlike [iib]) is not an entailment, not a truth condition, but merely a conventional implicature. The only scenario in which [ia] can be false is one where Kim didn’t resign. To see more easily that this is so, consider a situation in the future. You say Pat will sign the cheque and I respond Kim too will sign it. And suppose that in fact Pat does not sign, and only Kim does so: it is clear that the prediction I made will be judged to have turned out to be true, not false. Correlating with this is a difference with respect to negation. We have seen that negating [62ib] affects [iiib]: Not only Kim resigned says that there was someone else besides Kim who resigned. But we can’t negate [ia] so as to cancel [iiia], while leaving [iia] intact. ∗Not Kim too resigned is ungrammatical, and Kim too didn’t resign has too outside the scope of negation, so that the two components are “Kim didn’t resign” and “Someone besides Kim didn’t resign”.

Also, "everybody" is a universal quantifier: http://www.glottopedia.org/index.php/Negative_polarity_item --Backinstadiums (talk) 11:45, 30 May 2021 (UTC) --Backinstadiums (talk) 11:43, 30 May 2021 (UTC)

Context: Preposition 3. Other than, else than: in negative and interrogative (formerly sometimes in affirmative) sentences, capable of being rendered by ‘except, excluding.’ https://oed.com/oed2/00020842 (Cf. beside https://oed.com/oed2/00020841, except : Preposition †2. Leaving out of account; hence, in addition to, besides, as well as. Obs. rare. https://oed.com/oed2/00079534 )


 * regarding your sentence "the stuff that you autistically, boringly copy from textbooks ", please do not use derogatory language --Backinstadiums (talk) 10:11, 30 May 2021 (UTC)


 * You're still wrong even if I used language that doesn't personally appeal to you. Stop tone-policing. Equinox ◑ 10:12, 30 May 2021 (UTC)


 * That doesn't seem like tone-policing, it sounds to me like a polite reminder that you are breaching Wiktionary's rules, which clearly state:
 * "Please keep in mind the rules of discussion: remain civil, don’t make personal attacks, ...."
 * There is (subjective) judgement required on what is merely tone and what is crossing the line, but let me chime in as an impartial observer that the above example is an infringement of the rules.
 * —DIV (1.129.106.0 05:54, 17 June 2021 (UTC))


 * In the absence of context forcing another interpretation, I would take "everyone besides Larry was invited" to mean Larry wasn't invited. Semper is right that it could be ambiguous, and the other interpretation can be forced e.g. if someone says they wanted you to invite Larry to a meeting to discuss something (privately) and you point out that you invited him to the meeting tomorrow, "so what's the problem? Larry was invited to a meeting." "The problem is everyone besides Larry was invited!" But I wonder if that's better viewed as still using "everyone besides Larry" to mean "the set of all people, aside from Larry", and then just conveying by context that both that set of people and also the set which consists of Larry were invited. This requires more thinking about, especially as it's not immediately obvious to me what change to our entries is being proposed based on one or the other interpretation(?). - -sche (discuss) 14:20, 30 May 2021 (UTC)


 * In my interpretation, Larry was not invited. Imetsia (talk) 15:31, 30 May 2021 (UTC)
 * Only definition 2 of seems normal in my contemporary English idiolect. (I am not sure about the "instead of" element of that definition.) Definition 3 would give us "Everyone next to Larry was invited", which seems quite implausible, but so does "Everyone in addition to Larry was invited." In my idiolect's dictionary, which would be prescriptive, "in addition to" would be marked obsolete and "beside" would be archaic or erroneous. DCDuring (talk) 15:59, 30 May 2021 (UTC)
 * To me, "besides" means "aside from", implying "every element of the set in question, exept for the elements that also constitute the following". I think there are two ways of using it.
 * "Everyone besides Larry was invited." vs. "Everyone was invited.": focuses on the fact that Larry wasn't invited (because saying specifically that "at least this set of people - that doesn't include Larry - was invited," implies that Larry isn't part of the complete set of people who were invited.)
 * "Someone besides Kim was invited." (1) vs. "Someone was invited." (2) vs. "Kim was invited too." or "Kim too was invited." (3): statement 3 focuses on the fact that Kim was invited, while also asserting that statement 1 is true. Here, our brains parse statement 1 as a substatement of statement 3. Why? Because if Kim wasn't invited, we'd say "Kim wasn't the one who was invited." The word "someone" can only refer to one person. So we're saying that "at least the set, which is only one person large, that contains a person who is not Kim was invited." We parse this as Kim being invited too because the alternative, "Someone was invited and it wasn't Kim," would already be implied in the simpler senctence "Kim wasn't invited," or "Kim wasn't the one who was invited."
 * In short, there's a distinction between "[a set] besides [a subset of that set]" and "[a neighboring, compatible set] besides [a set]." The both mean "at least the entirety of the first set, not including the members of the second set," but the first statement can only be parsed as "the first set, but not the second set, with focus on the second," while the second statement only as "both the first set and the second set, with focus on the first." If we would like to imply that Larry was invited too, we would say "everyone besides Larry was invited too." If we wanted to say that Kim didn't resign (or get invited), we'd say "Someone other than Kim resigned instead." So it all depends on whether the two sets are compatible or not.
 * So I think a good rephrasing of the second definition is "(when the antecedent is part of the current subject) not including, except for" while the first definition would then be "(when the antecedent can coexist with the current subject) next to, in addition to."
 * As a side note, the second quotation of the first definition has "besides" used as an adverb, not as preposition. The definition "in addition" doesn't work for the preposition.
 * 110521sgl (talk) 14:03, 3 June 2021 (UTC)
 * You raise a good example. I would note that context could still make "Someone besides Kim was invited" be interpreted as "Someone other than Kim, and not Kim, was invited", e.g. if there's only one invitation being extended to a particular set of people and it's expected to go to Kim, but then in a surprise move "someone besides Kim was invited". (This interpretation becomes the more salient/default one with a verb like "Someone besides Kim won", where context would need to establish that multiple winners were possible in order for "Kim won but so did someone else" to seem likely/possible.) - -sche (discuss) 21:07, 3 June 2021 (UTC)


 * "Everyone besides Larry was invited" is a bit equivocal, but I'd be pretty sure he wasn't invited; "Everyone, besides Larry, was invited" and "Everyone was invited, besides Larry" unambiguously mean he wasn't invited. Also unambiguous: "Besides Larry, everyone (else) was invited." (See Talk page at besides for further comment.) --Kent Dominic (talk) 10:56, 2 June 2021 (UTC)


 * By far the most natural interpretation is that almost "everyone" was invited, with the exception being that Larry was not invited.
 * As others have mentioned above, the alternative interpretation might be able to be forced by the context. Consider the conversation. "Everything's ready, I have even invited Larry."  "But what about everybody in the other group?"  "Don't worry about everybody in the other group:  everybody besides Larry was invited."  But even with that context the wording seems unnatural:  "Besides Larry, everybody (in the other group) was invited (too)." seems better.
 * The spatial interpretation doesn't work for me. It would have to be, "Everyone beside Larry was invited".
 * —DIV (1.129.106.0 05:47, 17 June 2021 (UTC))

Merge history, how does it work?
I'm a bit confused about Special:MergeHistory. I tried to see what would happen if I merged take advantage of into take advantage, but it says "no mergeable revisions". Doing it the other way, there's only one mergeable revision from 2007. How is this supposed to work? I'd like to preserve both page histories when merging the two entries, but apparently that's not possible. Benwing2 (talk) 18:36, 31 May 2021 (UTC)
 * I didn't know this page existed. Also keen to learn how it works. — SGconlaw (talk) 19:51, 31 May 2021 (UTC)
 * if any kind bureaucrat can give me temporary adminship, I can do some experimenting and write a report about for you guys Indian subcontinent (talk) 20:26, 31 May 2021 (UTC)
 * 😂 - -sche (discuss) 23:49, 31 May 2021 (UTC)


 * Wikipedia has some information at w:Wikipedia:Administrators'_guide/Fixing_cut-and-paste_moves; the issue may have to do with the fact that both pages are approximately the same (very old) age. - -sche (discuss) 23:49, 31 May 2021 (UTC)
 * I've been merging a lot of Proto-Germanic pages into Proto-West-Germanic by request of User:Victar when there are no non-West-Germanic descendants. (I feel like even mentioning it is inviting another Reddit thread ranting against Wiktionary, like the ones about Frankish, but whatever.) So basically, you can only merge the revisions in the merged-from page that are older than the oldest revision of the merged-to page. Fairly often I've found no mergeable revisions, when the Proto-Germanic page was created after the Proto-West Germanic page. Both take advantage and take advantage of were created in December 2007, so whichever way you merge, most of the revisions are doomed. :-( There's no clever Git scripting type feature (not even sure if that exists in Git) where we can interlace the revisions that were made around the same time. — Eru·tuon 05:45, 1 June 2021 (UTC)
 * In that case, I'd recommend making a hard redirect without merging the histories so that each history is preserved. —Mahāgaja · talk 09:48, 1 June 2021 (UTC)