Wiktionary:Beer parlour/2020/May

discussion rooms: Tea room • Etym. scr. • Info desk • Beer parlour • Grease pit ← April 2020 · May 2020 · June 2020 → · (current)

---

AWB bot request

I made a newer version of the templates {{tr-conj}} and {{tr-conj-v}} that is easier to use, is more correct, and most importantly supports irregular verbs. If you want to check it out, a test page with comparisons is in here and the template itself is here. I'm making this request to replace the current template with the new one, and then using a bot to change the template syntax to use the new one. (the bot acc will be user:tr-conj-bot) --betseg|g 07:45, 2 May 2020 (UTC)

Are you going to use the bot to make conjugated forms of the Turkish verbs? That would be awesome. --Elvinrust (talk) 22:18, 3 May 2020 (UTC)

This will allow the conjugation of affetmek and siblings to be generated, and “bakıyorlarmış” is a necessary correction. And the new “-lermiş” forms are more common than the old ”-mişler”. But I think “bakmaz mımış?” instead of “bakmaz mıymış?” in both the old and the new version is a typo. I hope you will also supply better documentation then we have now. Which aorist vowels are “regular”? Those that are different from the |2= vowel? It took me some experimentation to figure out that the current {{tr-conj-v|oyn|u|a|oynar|a}} should be replaced by {{User:Betseg/tr-conj-template|oyn|u|d|vowel=1}}. By the way, {{User:Betseg/tr-conj-template|oyn|o|d|vowel=1}} works just as well; is it true that the |2= vowel can always be taken equal to the last vowel of the stem?  --Lambiam 11:57, 4 May 2020 (UTC)

• "But I think “bakmaz mımış?” instead of “bakmaz mıymış?” [...] is a typo."

  Yep, Fixed

• documentation

  Will do.  Done

• "Which aorist vowels are “regular”?"

  It's not that the vowels are irregular, it's that a couple verbs take non standard vowels. Going by the rules, it should be "alarım" but it's "alırım", in irregular verbs' aorist forms the vowel is close instead of open. (now that i think about it, i can add a named parameter to it and use {{tr-v2-v4}} instead of |aorist= (update:  Done))

• "is it true that the |2= vowel can always be taken equal to the last vowel of the stem?"

  Yes, for all Turkic-originated verbs and most of the borrowed ones (the irregular ones can be done with a named parameter), but I couldn't find a way to do it on MediaWiki. The 3rd parameter also can be removed by checking if the last consonant of the stem (if the last letter is a consonant) is voiced.

-- betseg|g 17:04, 4 May 2020 (UTC)

• Support. Great improvement.  --Lambiam 16:55, 5 May 2020 (UTC)
• Any more comments or opposes? Or is this not the right place to ask for AWB requests? --betseg|g 15:02, 11 May 2020 (UTC)
  • @Betseg, Tr-conj-bot: OK, I added you to WT:AWB, you should now be able to use AWB. - -sche (discuss) 21:02, 11 May 2020 (UTC)

Italian pronunciations

Are there any plans to add the pronunciation to words like pianga that don't have it yet? Please also look at https://en.wikipedia.org/wiki/Talk:Italian_language#%E2%9F%A8ng%E2%9F%A9 --Espoo (talk) 08:12, 5 May 2020 (UTC)

Well, one can always add {{it-IPA}} (which I've just done), but some weird cases might not be handled correctly. I think /ˈpjan.ɡa/ is correct here though. – Jberkel 09:01, 5 May 2020 (UTC)

I was interested in a project to mass-add Italian IPA to entries a while ago, but it required someone with technical expertise. @Jberkel, let me know if you're interested in working on it (and absolutely no pressure if you're not). —Μετάknowledge^{discuss/deeds} 17:24, 6 May 2020 (UTC)

I can help with that, I assume we could automatically add it to entries where we know the generated IPA will be correct. {{ca-IPA}} has a nice feature where it warns about ambiguous input–something similar could be added to {{it-IPA}} as well. – Jberkel 22:44, 6 May 2020 (UTC)

Planned maintenance operation on May 7 @ 05:00 AM UTC

Hi, There's a planned maintenance operation that will be performed on Thursday 7th May at 05:00 AM UTC. CentralAuth-based services (rename account, change password, etc.) may not work. See also: phab:T251157. NB: This wiki will not go read-only during this operation. --Kaartic (talk) 11:01, 6 May 2020 (UTC)

Headword line - abbreviated case names

Is there a standard regarding the format of case names in headword lines? I couldn't find anything written on this, I just assumed the case names should always be spelled out and not be abbreviated. We are planning to add more items to the Hungarian headword line for nouns (the current contains only the plural form). For example:

makroobjektív (accusative makroobjektívet, plural makroobjektívek, third-person singular possessive makroobjektíve or makroobjektívje, superessive makroobjektíven)

The abbreviated version:

makroobjektív (acc makroobjektívet, pl makroobjektívek, pos makroobjektíve/-je, sup makroobjektíven)

I want to make sure the additions conform to our standards but the spelled out version is probably too long, and the short version might be confusing. Thanks. Panda10 (talk) 16:59, 6 May 2020 (UTC)

Are those forms predictable to Hungarian learners? If so, I don't think they really belong on the headword line. If they aren't, and you want to abbreviate, you should use tooltips with the full names. —Μετάknowledge^{discuss/deeds} 17:22, 6 May 2020 (UTC)

(edit conflict) We don't usually use abbreviations in headword lines unless there's a {{tooltip}} so that the full form is revealed by mouseover. But your example looks like way too many forms in the headword line anyway. Shouldn' they be in an inflection table in an ===Inflection=== or ===Declension=== section.? —Mahāgaja · talk 17:25, 6 May 2020 (UTC)

@Mahagaja They are meant to provide the minimal core information that makes it possible for intermediate or advanced language learners to create all the other forms correctly. The full declension table is given below but many of its forms are superfluous with respect to this minimal information. @Metaknowledge Certainly, tooltips would be supplied without doubt. Adam78 (talk) 17:45, 6 May 2020 (UTC)

Cambridge Grammar of the English Language

I think editors will benefit from this true pdf <REDACTED> --Backinstadiums (talk) 19:14, 10 May 2020 (UTC)

Unless you can show permission for you to distribute this copyrighted work has been granted by the copyright owner(s), linking to a download of it is strictly against Wikimedia policy. Chuck Entz (talk) 05:00, 11 May 2020 (UTC)

If somebody needs a copy, PM me. --Backinstadiums (talk) 10:42, 11 May 2020 (UTC)

{{R:Webster 1996}}

Why does this template link to dictionary.com? What is it supposed to link to? An anonymous user brought it up at Talk:naze. Ultimateria (talk) 20:51, 10 May 2020 (UTC)

Per Wikipedia, Webster's Revised Unabridged Dictionary was published in 1913 and after the expiration of its copyright was digitized by MICRA in 1996. It was one of the source materials for dictionary.reference.com when that template was created back in 2006. The site seems to have removed the 1913 Webster's content from its entry pages not long after ("naze" in particular lost it sometime between December 2008 and June 2010). {{R:Webster 1913}} links to the same material on a dedicated website. --Quesotiotyo (talk) 20:00, 20 May 2020 (UTC)

"male or of unspecified sex"

We have almost 500 entries with this phrase on the definition line (see Zuwanderer for example). Can we safely remove it? Not only is this label needlessly wordy (especially with a gloss), but I think it explains a broad feature of an entire language that doesn't really belong at the entry for every agent noun. The headword line already shows the masculine gender, and these largely German and Italian entries use parameters for female equivalents, so the issue of gender isn't unclear. Ultimateria (talk) 20:16, 11 May 2020 (UTC)

The gender in the headword line is grammatical, while the gender in the definition is natural. So one does not necessarily imply the other. —Rua (mew) 09:01, 12 May 2020 (UTC)

@Rua: I agree with User:Ultimateria. It's actually a feature of German that almost if not all agent nouns have genders. The masculine form is the default when the gender is unspecified. It is similar to other languages but not the same. I think the usage is closer to French but I don't think it's the same as in e.g. Russian or Dutch. (Russian is an odd one and agent gender usage causes issues for other Slavic speakers). We need a native German speaker. --Anatoli T. ^{(обсудить}/^вклад) 11:41, 12 May 2020 (UTC)

I had to make longer usage notes at президе́нт (prezidént), до́ктор (dóktor) and врач (vrač) (more words fall into the same category) because some people have assumed that these are male occupations but they are used for both genders), unlike учи́тель (učítelʹ), which s a male teacher or "unspecified gender". It's not perfect. Do we need that for German agent words? --Anatoli T. ^{(обсудить}/^вклад) 11:48, 12 May 2020 (UTC)

@Atitarev: but agent nouns have diminutives, and those are all neuter, despite having a referent with a natural gender. Natural gender cannot automatically be derived from grammatical gender in German. —Rua (mew) 11:57, 12 May 2020 (UTC)

@Rua: You're right about the neuter diminutives, which may be confusing. Not just agent nouns, common words like Mädchen, Fräulein could have a note about the gender. The former one does. --Anatoli T. ^{(обсудить}/^вклад) 22:38, 12 May 2020 (UTC)

We could potentially have a usage note at any Indo-European agent noun explaining the connection between grammatical gender and natural gender, but unless they're about the specific use of that word (like the notes at Fräulein), these notes are clutter that doesn't belong at thousands of dictionary entries. Take the usage note {{es-note-noun-mf}}: it describes a grammatical feature, not a lexical one. It's a text bubble from a level 1 Spanish textbook. If it belongs at all, it belongs in an appendix on Spanish nouns.

Let me rephrase my original question: does anyone object to removing "male or of unspecified sex" from definition lines? Ultimateria (talk) 00:08, 14 May 2020 (UTC)

@Ultimateria, Rua: I agree it looks clattering. We have to come up with a policy how to both inform the users (maybe just in one place) and remove it from definitions when it's a feature of the language for a very large group of words.

Just to clarify again, I felt the need to put notes in президе́нт (prezidént), до́ктор (dóktor) and врач (vrač), which will apply to many loanwords as well, ending in -ик, -лог, such as фи́зик (fízik), био́лог (biólog) - no grammatical feminine exists for these. However, a large number of other words, such певе́ц (pevéc), учени́к (učeník), преподава́тель (prepodavátelʹ) have feminine equivalents (grammatical and natural) including loanwords ending in -ист, -aт such социали́ст (socialíst), демокра́т (demokrát). The usage for those is similar to German, no need to say {{m|ru|учи́тель} is "male or of unspecified sex" for each of these. So, in short, in Russian, we have, at least two groups of agent nouns.

1. German Lehrer, Lehrerin = Russian учи́тель (učítelʹ), учи́тельница (učítelʹnica) - a typical example, the grammatical masculine forms are "male or of unspecified sex", no notes required in the entry, IMO.
2. German Präsident, Präsidentin <> Russian президе́нт (prezidént). The Russian word is used for both genders, the form президе́нтша (prezidéntša) is too colloquial and not respectful.
3. German Dichter, Dichterin <> Russian поэ́т (poét), поэте́сса (poetɛ́ssa). Partial match. The Russian term for "poetess" is somwhat dispreferred by females.
4. German Mädchen = Czech děvče - grammatical neuter but natural female. A generic usage note is in order.

I would only use special notes when it's atypical, uncommon, like in cases I described above.--Anatoli T. ^{(обсудить}/^вклад) 01:35, 14 May 2020 (UTC)

There's also the case of Irish, which has no neuter gender, but where dimunitive suffixes likewise dictate grammatical gender even if it's different from natural gender. So cailín (“girl”) is masculine, gasóg (“boy scout”) is feminine. —Rua (mew) 09:11, 14 May 2020 (UTC)

why are the unicode Arabic Pedagogical symbols blacklisted?

why are the Unicode Arabic Pedagogical symbols U+FBB2 to U+FBC1 blacklisted?

they seem to be part of a large range blacklisted? I got these messages:

The title "﮻" has been banned from creation. It matches the following blacklist entry: .*[\x{FB50}-\x{FDC7}\x{FE70}-\x{FEFC}].*
The title "Wiktionary:﮲" has been banned from creation. It matches the following blacklist entry: .*[\x{FB50}-\x{FDC7}\x{FE70}-\x{FEFC}].*

A lot of that set are for compatibility with old systems that encoded positional forms عـ ـعـ ـع differently (the set i am looking for are a subset of "Arabic Presentation Forms-A" U+FB50 - U+FDFF) but these are different. This set are designed for demonstrating features of the letters, and a linguistically relevant dictionary entry could be written about them. Unless the positional forms which get included with the current code-point for each letter e.g. س includes all forms. The positional forms get included with the current code-point for each letter e.g. س includes all forms. But including these with each letter they relate to doesn't make much sense, unless they link back to a central entry for the mark. They each relate to multiple letters.

Click this to show or hide table.

&#x (code#) ;

﮲

﮳

﮴

﮵

﮶

﮷

﮸

﮹

﮺

﮻

﮼

﮽

﮾

﮿

﯀

﯁

[[:wiktionary: (character) ]]

wiktionary: ﮲

wiktionary: ﮳

wiktionary: ﮴

wiktionary: ﮵

wiktionary: ﮶

wiktionary: ﮷

wiktionary: ﮸

wiktionary: ﮹

wiktionary: ﮺

wiktionary: ﮻

wiktionary: ﮼

wiktionary: ﮽

wiktionary: ﮾

wiktionary: ﮿

wiktionary: ﯀

wiktionary: ﯁

[[:wiktionary: &#x (code#) ; ]]

wiktionary: ﮲

wiktionary: ﮳

wiktionary: ﮴

wiktionary: ﮵

wiktionary: ﮶

wiktionary: ﮷

wiktionary: ﮸

wiktionary: ﮹

wiktionary: ﮺

wiktionary: ﮻

wiktionary: ﮼

wiktionary: ﮽

wiktionary: ﮾

wiktionary: ﮿

wiktionary: ﯀

wiktionary: ﯁

[[ &#x (code#) ; ]]

﮲

﮳

﮴

﮵

﮶

﮷

﮸

﮹

﮺

﮻

﮼

﮽

﮾

﮿

﯀

﯁

unicode

U+FBB2

U+FBB3

U+FBB4

U+FBB5

U+FBB6

U+FBB7

U+FBB8

U+FBB9

U+FBBA

U+FBBB

U+FBBC

U+FBBD

U+FBBE

U+FBBF

U+FBC0

U+FBC1

Example Letters

ض ف غ خ ن ذ ظ ز ڬ ڧ ڶ

ج ب ڊ

ت ق ڗ ڱ

ي ڲ

پ ش ث چ ڜ څ ڠ ڮ ؼ ڸ ڽ ۋ ؿ

ٿ ڐ ڙ ڦ

ڇ ڀ

۽ ۾

ڂ ٺ

ڳ ې ٻ

ټ ډ ړ ڼ ګ ؠ

ٹ ڈ ڑ

though linking typical letters doesn't work this way, so maybe i'm doing something wrong, i was trying to link here from a wikipedia article i was editing. — This unsigned comment was added by Irtapil (talk • contribs). sorry, forgot to sign Irtapil (talk) 02:23, 13 May 2020 (UTC) updated for clarity: Irtapil (talk) 02:37, 13 May 2020 (UTC)

They were blacklisted by @Erutuon. DTLHS (talk) 00:58, 13 May 2020 (UTC)

Yea, after my suggestion, User talk:Erutuon/Arabic presentation forms. Barring the mentioned codepoints which have no visual equivalents all uses of these characters are erroneous. I had to work through a long list of pages containing these characters to replace them with the normal forms. And to create pages with these characters in the title there is even less reason. Many people have been saved from using them without knowing that they do not use the normal characters, one of them has expressed himself at Wiktionary:Grease pit/2020/March § Permission error trying to create page "أغروم". The name “pedagogical characters” OP is giving to them is funny, pedagogy is not their purpose; if you want to display positional forms you can use the kashida or bidirectional control characters (which are under-used and should be used more often legitimately to outperform the spammers).

The purpose of these characters is display on terminals which do not support complex text rendering; web browsers of this kind can’t exist in 2020, and for terminals this is to be avoided too. These signs are not specific for displaying linguistic content anyway so they do not belong into a dictionary. Fay Freak (talk) 01:32, 13 May 2020 (UTC)

@Fay Freak: yes i know how to use kashida that's how i made the عـ ـعـ ـع forms in the question. These are something different. Irtapil (talk)

@Fay Freak, Erutuon: "Pedagogical" U+FBB2 to U+FBC1 is a subset of the "Presentation Forms-A" U+FB50 - U+FDFF, which also contains the positional block (Unfortunately i can't show the documentation for this because the Unicode technical site is still offline, and the mirrors don't include this level of detail). If you were trying to exclude the positional forms i think thus set was included in error?

I understand the reason for blacklisting the positional forms, but there are no "normal" forms for most of these characters in U+FBB2 to U+FBC1. A few have corresponding identical-looking combining diacritics such as ــؕـ for ﯀ but i don't think these actually are the same character linguistically. One of them looks the same as something described as a "vowel diacritic", which is definitely different linguistically. This set U+FBB2 to U+FBC1 are not used to write words, they are used to describe sets of consonants, e.g. ٹ ڈ ڑ all having the same modifier U+FBC0. But there are meaningful and useful entries for other things not used to write words, such as ۞ "RUB EL HIZB۔".

They are definitely still in use, there is a pending proposal to add an additional one of this type of character corresponding to the mark added to Alif ٱ in some classical Arabic.

Irtapil (talk) 02:22, 13 May 2020 (UTC)

The blacklist entry is intended to exclude positional forms of Arabic letters, as mentioned by User:Fay Freak, so I've removed U+FBB2-U+FBC1 from the regex. They are categorized as symbols rather than letters. It's not clear to me that there should be entry names containing them, but on the other hand I don't see a clear reason to prohibit them at the moment. — Eru·tuon 03:04, 13 May 2020 (UTC)

Thank, you. @Erutuon: Should i attempt to write entries? Using other symbol entries as a template? i'm not sure how literally to take "It's not clear to me that there should be entry names containing them". Irtapil (talk) 18:43, 13 May 2020 (UTC)

@Irtapil: I mean, maybe there could be entries for the characters themselves, if they pass WT:CFI, but maybe no other entry names should contain them. — Eru·tuon 21:59, 13 May 2020 (UTC)

@Erutuon: yeah, they shouldn't appear within another name just ike most symbols & ۝ ۞ @ ¶ ½ € % would not be part of a word. Theoretically maybe you could use them for writing a foreign word in a font that doesn't have it, but it would be atypical, like using double dot diacritic instead of ä, and i'm not even sure if they work in that way. Irtapil (talk) 01:15, 14 May 2020 (UTC)

As far as i can gather they're mostly for talking about the letters, without needing to add images to a document, because as soon as you add images thing get more complicated and/or expensive. Irtapil (talk) 03:44, 14 May 2020 (UTC)

On a related note, i'm seeing quite a few Unicode character pages that contain dead links to the positional forms on the forbidden list, e.g.

ہ    ۀ    ۆ

How can i help fix them? They're done with a template so i'm not sure how to remove the dead links without removing the forms themselves? Showing the positional forms is useful, but the next symbol and previous symbol links being all dead is a bit messy.

* Previous letter: {{l|az|ی}} * Next letter: {{l|az|ێ}}

that one is prticularly confusing, because they don't look mich like the forms i'm seeing shown:

ﯚ ﯙ

There are also some pages still exist for forms on the forbidden list, e.g.  ﻫ  How do i help tidy these up if i spot them?

(Some leave out the forms completely, e.g. ۂ   ۄ    ۇ but position forms might not exist for them, they seem to be obscure characters or from less common languages which might have only got into Unicode after the fixed positional forms became obsolete.)

Irtapil (talk) 03:44, 14 May 2020 (UTC)

Redlinks in {{character info}} could be unlinked, but they don't bother me personally. If more people would prefer them to be unlinked, I can do it. As for ﻫ, an admin reckoned it should be created. Admins can override the title blacklist if they want. You can nominate it for deletion (WT:RFD) if you think it shouldn't exist. — Eru·tuon 04:01, 14 May 2020 (UTC)

That entry was also created back in 2006, which appears to predate MediaWiki:Titleblacklist, heh. (It also seems to specify a rationale, the different place it comes in the abjad, although that could be handled with a usage note in the "main" entry.) Anyway, IMO each of these characters, as an individual character, should have an entry even if for the "positional" characters that entry is just a hard redirect to the "main" form. - -sche (discuss) 17:00, 14 May 2020 (UTC)

@-sche: An automated redirect could be confusing? if people search for the obsolete character and end up at the new one? the person searching won't know what they started with? a good strategy would be a short standard entry "obsolete Unicode character: see (new character)". It would be more informative than a blacklist? Irtapil (talk) 07:37, 15 May 2020 (UTC)

fr-IPA

Can this template be programmed to produce /é/ instead of /è/ as the default pronunciation of the spelling ai in the first syllable of words with two syllables? I fixed aimer and baiser manually, but there are many more: aider, aigu, aîné... --Espoo (talk) 17:26, 13 May 2020 (UTC)

@Espoo: WT:GP is probably a more appropriate place. I also recommend to post on Module_talk:fr-pron. User:Benwing is the main contributor to the module. --Anatoli T. ^{(обсудить}/^вклад) 01:48, 18 May 2020 (UTC)

@Espoo This can potentially be done. Maybe it should list both pronunciations? I assume that the pronunciation with /è/ is still possible, just more formal. What about three-syllable words like aiderions, four-syllable words like aimablement, etc.?

@Espoo Benwing2 (talk) 04:21, 22 May 2020 (UTC)

Esperanto : imperative vs. volitive

Hi, I recently changed the label of -u column of the {{eo-conj}} table from "imperative" to "volitive" based on a recent change on the French Wiktionary. This change was based on the book "Parlons espéranto" by Jacques Joguin. Metaknowledge (talk • contribs) reverted me because I made the change without any discussion with the community. So here I am!

I know that imperative is sometimes used to describe this mood, but volitive is a more general mood. I found it in Parlons espéranto by Jacques Joguin for French litterature, and also in Esperanto: Language, Literature, and Community by Pierre Janton or The Grammar of Esperanto by Christopher Gledhill.

What do you think about updating the table label to "volitive"?

Please, feel free to correct my message if it contains any mispellings or errors. Lepticed7 (talk) 18:34, 13 May 2020 (UTC)

I don't see any value here. Google Ngrams puts imperative at 1,000 times more common in English than volitive, so there's a distinct cost in readability and clarity by the change.

It also doesn't seem more accurate: w:Volitive modality says the Esperanto sense is more of a deontic mood, and Adventures in Esperanto says:

I want to talk about the “U-mood” of verbs in Esperanto. Most English books seem to call it the “imperative” mood. But on page 67 of “Being Colloquial in Esperanto”, David Jordan points out that its functions include things that could be considered “imperative”, “volitive” or “subjunctive” when comparing to how these moods are used in some other languages.

To quote from the Esperanto Teacher:

To give an order or command, or to express will, desire, purpose, etc., the verb must end in u, as Donu al mi panon, Give (to) me bread; Iru for, Go away; Estu feliĉa, May you be happy! Vivu la reĝo! (Long) live the king!

Given that the -u translates the English imperative, Esperanto doesn't have two distinct conjugations for imperative and volitive, and since -u seems to be the way to express a command that is not a desire, it is questionably more accurate, and definitely less clear to our users.--Prosfilaes (talk) 19:36, 13 May 2020 (UTC)

Hi, it’s not because a word appears 1,000 times more than another one that you chose it. You choose a word because it conveys the meaning you wanted. If most english references describing esperanto use "Imperative", it is okay. I’ve a last question: is May you be happy! considered as imperative in english? I think in french, it would be considered as subjunctive, but I don’t know for english. Lepticed7 (talk) 06:10, 14 May 2020 (UTC)

If you choose a word because it conveys the meaning you wanted, it is important to understand whether it does convey the meaning you want, and exceedingly rare words tend not to convey meaning well. I'm not sure what the technical term for "may you be happy!", but it would not be imperative. Though I would say that "Estu feliĉa" could be translated "Be happy", which would be imperative, though a rather dickish thing to say.--Prosfilaes (talk) 20:15, 14 May 2020 (UTC)

So the fact that "May you be happy!" is not imperative, but "estu" is presented as imperative does not bother you? It looks like an error to me, but I surely don't understand some subtleties of the English language . Lepticed7 (talk) 11:34, 15 May 2020 (UTC)

Does the fact your first sentence is grammatically a question, but does not truly involve a request for information, bother you? There's a certain imprecision involved in these names, and as I pointed out above, "volitive" isn't any better. "Donu al mi panon" would not normally express a desire; it's a straightforward command. There is no word that's going to exactly cover what the -u suffix in Esperanto does, and imperative is reasonably well understood and accurate.--Prosfilaes (talk) 19:46, 15 May 2020 (UTC)

How about just labeling the mood "imperative/volitive"? Andrew Sheedy (talk) 20:13, 15 May 2020 (UTC)

Or make a new word, impolitive. According to The Official Wiktionary Rules, we are allowed to coin words and spread them into daily use. --Undurbjáni (talk) 00:30, 16 May 2020 (UTC)

It's a slightly pointless discussion because the meaning of any of these terms will generally vary depending on language anyway. "May you", "may he", etc. certainly is expressed by the imperative in some languages. Ancient Greek has a third-person imperative, for example. Biblical Hebrew uses the jussive, imperative, and another one called the cohortative. In English the category of such expressions is sometimes called optative but is grammatically expressed with a modal verb (as in "may he ..." itself) or, more rarely, with the subjunctive ("perish the thought"). Jussive or volitive seems to be a usual term for Esperanto. Just pick a term and stick with it; explain the choice in an appendix if necessary. —Nizolan ^(talk) 03:58, 16 May 2020 (UTC)

Copying examples from other dictionaries

What is the etiquette of copying usage examples from other dictionaries? For example, @Nizolan found some examples of a queried sense of "about" in the OED [1], but can I just copy these and use them in our article, or do we have to find our own examples using our own research? Mihia (talk) 17:50, 14 May 2020 (UTC)

If quotations are provided in a dictionary that is out of copyright, like the early editions of the NED/OED are, then you can clearly use them, AFAIK: if you can find the original works, it's best to cite them directly (sometimes quoting the full sentence if the dictionary has abbreviated it), otherwise I have occasionally seen things formatted like "original date, original author, original work, quoted in such-and-such dictionary". If usexes are present in a dictionary is in copyright, and alert you to search for a particular collocation and thus find works using a particular word/sense (even the same works the other dictionary found), I think it's OK to quote those works you found (directly). In both cases it would probably be good to add the dictionary to the ===References=== or ===Further reading===. - -sche (discuss) 19:38, 14 May 2020 (UTC)

Often the exact wording of citations found in other dictionaries cannot be found in the available online versions. My suspicious mind worries that the other dictionaries have set traps for us. I try to provide direct links to online editions. I also believe that we should encourage users to compare other dictionaries' sets of definitions of a term with ours by including {{R:OneLook}} in References. DCDuring (talk) 01:55, 15 May 2020 (UTC)

Yes, I would only add a citation if we can find the original (and add its wording), or if the dictionary whose wording we're using is out-of-copyright. - -sche (discuss) 06:45, 15 May 2020 (UTC)

This is the policy I've been following for the Medieval Latin entries I've been adding recently, since the dictionary I'm referencing (Niermeyer) is still in copyright. I try to look up different examples online and, failing that, at least to verify the quotation myself and prepare a translation from there. —Nizolan ^(talk) 04:02, 16 May 2020 (UTC)

Speaking of traps...If other dictionaries do it, we should too. I'll start including some random traps in my future editing, and encourage you all to do likewise. --Undurbjáni (talk) 00:02, 17 May 2020 (UTC)

If some content (e.g. a definition, or research on citations, etymology etc.) was first published in a dictionary edition that is now out of copyright, but has been perpetuated through to later editions, even current editions, that are in copyright, would that content still be under copyright or not? Mihia (talk) 21:37, 17 May 2020 (UTC)

If a work is out of copyright, further reprintings or new editions of that work will not bring the original work back under copyright; otherwise, nothing major would ever leave copyright.--Prosfilaes (talk) 23:03, 17 May 2020 (UTC)

Citation templates (again)

A recent citation I added read as follows:

1. • 1646, Thomas Browne, Pseudodoxia Epidemica, III.28:

     Much wonder is made of the borametz, that strange plant-animal or vegetable lamb of Tartary, which wolves delight to feed on, which hath the shape of a lamb, affordeth a bloody juice upon breaking, and liveth while the plants be consumed about it […].

This was changed to use the {{RQ:Browne Pseudodoxia Epidemica}} template, so that it now reads:

1. • 1650, Thomas Browne, “Of Some Others”, in Pseudodoxia Epidemica: […], 2nd edition, London: […] A[braham] Miller, for Edw[ard] Dod and Nath[aniel] Ekins, […], →OCLC, 3rd book, page 152:

     Much wonder is made of the Boramez, that ſtrange plant-animall or vegetable Lamb of Tartary, which Wolves delight to feed on, which hath the ſhape of a Lamb, affordeth a bloudy juice upon breaking, and liveth while the plants be conſumed about it; and yet if all this be no more then the ſhape of a Lamb in the flower or ſeed, upon the top of the ſtalk, as we meet with the formes of Bees, Flies and Dogs in ſome others, he hath ſeen nothing that ſhall much wonder at it.

(Please don't change the template during this discussion!) This is worse, I maintain, in just about every way. Note that the full title, with subtitle, is given twice in full! It is, I'm sure, a lot of fun making these templates more and more detailed so as to represent every datum of an early-modern title page, but this is not the point of them and it is not helpful for their purpose. In many cases, the book details are now much longer than the actual citation, running across three, four or more lines. Our citation templates go against every citation "house style" that I know of. I had thought the reasons for citation conventions were fairly self-evident, but, that apparently not being the case, I would propose the following principles which, if there's some consensus, I'll make into a formal vote to add to our guidelines.

1. Cite book titles without their subtitles.

   Persuasion, not Persuasion: A Novel

   Psudodoxia Epidemica, not Pseudodoxia Epidemica: Or, Enquiries into Very Many Received Tenents, and Commonly Presumed Truths

2. If a book title has a well-known conventional short form, use it.

   Henry IV, part 1, not The First Part of Henry the Fourth, with the Life and Death of Henry surnamed Hot-Spvrre

   Tom Jones, not The History of Tom Jones, A Foundling

3. Cite chapters but not chapter titles.

   "Partition II, Section II, chapter 3", not "‘Ayre Rectified. With a Digression of the Ayre.’, Partition II, Section II, chapter 3"

4. Cite the publication details only of the edition actually being cited.

   i.e. do not record a separate title, publisher etc. of the first edition unless you're quoting from it.

5. Printing location, publisher and publication date should be given, but not their addresses or connecting prose such as "printed for".

   "London: J. Johnson 1791", not "Printed for Joseph Johnson, 72 St Paul's Churchyard"

6. Date of cited edition should be given, but not other descriptive details.

   There is no need to say that something is a "Newly enlarged and corrected edition".

In some cases, there may be lexicographic reasons to break some of these guidelines, which is fine, but as a general rule we should keep this information as concise and clean as possible, IMO. What do others think? Ƿidsiþ 06:39, 15 May 2020 (UTC)

I agree on all points except possibly point 2 (I would be inclined to use whatever title the book actually has, that someone searching for [that edition of] it would find). Perhaps the excess data can be made into "tooltips". Or perhaps it can be hidden by default but displayed for users who opt in! - -sche (discuss) 07:25, 15 May 2020 (UTC)

Do not agree with any of them; and 4 is even unclear (does that mean that people shall not give the original publication date? But the date of something said is amongst the most important; the edition should in fact often be avoided because popular works have many editions and people generally won’t use the same one as the editor). And even if I agreed on any formatting, I would not agree on a vote for a guideline on such. Long live anarchy!

I would agree if there were a method and apparatus whereby one could a) see one short form, as in the first example b) expand it into long styles even of multiple editions, one of which is the second example; sometimes indeed in philology one gives references for multiple editions. Fay Freak (talk) 09:10, 15 May 2020 (UTC)

4 does not mean that people shouldn't give the original publication date. It means they should not give a book title twice, or give the name of a publisher of an edition that is not being cited. If there was a way to have a short form, expanded by mouseover or something, I guess that could be a solution. Ƿidsiþ 09:53, 15 May 2020 (UTC)

I enthusiastically agree with each and every point, with the sole provisos that the items in 3, 4, and 5 are optional. Also, it has always seemed important to me that the author should be the translator, if any, (eg, Eleanor Marx, not Gustave Flaubert) and the date should be the date of the earliest edition in which the English term can be found. DCDuring (talk) 14:40, 15 May 2020 (UTC)

I agree about the translation stuff, but I think that's already usual practice. Ƿidsiþ 06:17, 16 May 2020 (UTC)

It may be more common to attribute translated text to the translator, but the famous FL original author usually gets more prominent placement and many translators are not acknowledged at all. It often requires considerable sleuthing to find the translator and the date of the translation, which is probably why the remaining false attributions remain. DCDuring (talk) 13:51, 19 May 2020 (UTC)

If there is consensus that the translator, if known, should be listed first, the quotation template can be updated to reflect this. — SGconlaw (talk) 21:29, 20 May 2020 (UTC)

Also agree with all of this, especially point 4 on not citing publisher details for an edition that's not actually being referenced. The preference should be to provide whatever is necessary to look up a quotation, not to tabulate every piece of publishing info. —Nizolan ^(talk) 16:05, 15 May 2020 (UTC)

It seems like an easy solution would be to have an expanded form of the template on the documentation page (which can be to linked from entries using the template). DTLHS (talk) 16:06, 15 May 2020 (UTC)

+1 for simpler bibliographic info. – Jberkel 16:37, 15 May 2020 (UTC)

I disagree with point 2 but agree with most of the others. For point 6 there are cases where further details of a cited edition could matter, say, if there were two editions issued in the same year, but apart from such exceptional cases I agree. Subtitles would be nice to have under a tooltip or expandable text, and, really, that might be a good solution for a lot of this. — Vorziblix (talk · contribs) 17:37, 18 May 2020 (UTC)

Definitely support this. I should be able to see the author and title at a glance, without having to wade through ridiculously long subtitles and publication info. I think there are valid reasons for exceptions to all the points you propose, but it would be nice if people were less eager to document every possible piece of information about the book. Andrew Sheedy (talk) 23:45, 18 May 2020 (UTC)

This looks like @Sgconlaw's style? I agree that simpler is better. —Μετάknowledge^{discuss/deeds} 20:39, 20 May 2020 (UTC)

My responses:

1. No real objection to placing the subtitle within a tooltip using {{nb...}}. I already abbreviate most long subtitles, but if the subtitle is short and provides information about the work I usually just leave it.
2. Uncertain and tending towards disagreeing with this; there may be disagreements on when an alternative title is well-known enough.
3. Disagree. Why shouldn't the chapter name be given if it is known?
4. Agree. I think I did this because I discovered that some quotations were dated to the first edition, but I could not actually find any online version of that edition and so was unable to verify whether the quotation given was actually stated in that edition. In order to preserve the original date I added the bibliographic details of the first edition but then also provided details of the closest available edition. Happy to change this to the date of the earliest edition available online. (I encourage everyone to check quotations against actual scans of works at Google Books, the Internet Archive, or the HathiTrust Digital Library. On several occasions I have found quotations only found in subsequent editions, but dated to the first editions of works.)
5. I have already been putting non-essential information within a tooltip using {{nb...}}. I disagree that there can be a hard-and-fast rule about about omitting "connecting information". Perhaps we can compromise on omitting the information if the name stated is clearly that of a publisher, but if it is someone else (for example, a printer where no publisher is stated) then clarifying words can be added.
6. Just following standard bibliographic practice, but no objection to limiting the edition to an ordinal number, "new", or "revised".

— SGconlaw (talk) 21:21, 20 May 2020 (UTC)

OK, do we need a formal vote on this (in which case: would someone like to draft one?) or is consensus above clear enough to trim a lot of the excess out of these templates? - -sche (discuss) 23:27, 21 May 2020 (UTC)

I can draft a vote based on this, I think, as soon as I have some free time to sit down and do it (hopefully in the next week or so). Ƿidsiþ 13:42, 28 May 2020 (UTC)

Less is more, there should be enough information given that a person who needs additional details can get them, and no more. This is a dictionary, the purpose of the citation is to demonstrate usage over time, details about publishing houses and font size are very rarely going to be of any value, if ever. - TheDaveRoss 14:14, 22 May 2020 (UTC)

Restore to the Widsith original. And restore status quo ante until Sgconlaw demonstrates consensus for their practices. --Dan Polansky (talk) 11:08, 19 June 2020 (UTC)

OED Antedatings on Twitter

Is anyone here following the #oedantedatings hashtag on Twitter? People are giving evidence of word-usages predating the first-cited in OED - but of course the examples can be used here also; and many of the images are eligible to be uploaded to Wikimedia Commons. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:11, 15 May 2020 (UTC)

Hm, neat. We should peruse it to see if any of the quotations provided require our own entries' dates of attestation to be updated. - -sche (discuss) 21:42, 16 May 2020 (UTC)

Interesting. Today I looked up emo in the OED and their earliest "citation" was in a two-word song title (it was "Emo Overdose" or something). Pushing it a bit I thought! Equinox ◑ 18:59, 21 May 2020 (UTC)

Planned maintenance operation (read-only time) on May 19 @ 5:00 A.M. UTC

Hi, There's a planned maintenance operation in the upcoming week. It will happen on Tuesday 19th May at 05:00 AM UTC, for 15 minutes. This wiki will go read-only during this operation. During this operation, Wikidata would also be in read-only so services targeting Wikidata may not work during the meantime. See also: phab:T251981. --Kaartic (talk) 15:11, 16 May 2020 (UTC)

Planned maintenance operation (read-only time) on May 19 @ 5:00 A.M. UTC

Hi,

This is a reminder about the planned maintenance operation that is will happen happen on Tuesday 19th May at 05:00 AM UTC, for 15 minutes.

This wiki will go read-only during this operation. During this operation, Wikidata would also be in read-only so services targeting Wikidata may not work during the meantime.

See also: phab:T251981

--Kaartic (talk) 15:11, 16 May 2020 (UTC)

Evidence for Less-Documented Languages

The Criteria for Inclusion says of less documented languages, "For all other spoken languages that are living, only one use or mention is adequate, subject to the following requirements:

the community of editors for that language should maintain a list of materials deemed appropriate as the only sources for entries based on a single mention,...".

For a use, do we need to generate a citation? There may be a problem with it's being subject to copyright restrictions. Some sources may have a significant problem with permanence - and even being carved in stone is no protection against vandals!

Are there any examples of such lists?

What is the 'community of editors for that language'? Does one need to explicitly form one to create such lists, and if so how does one go about forming such a community? It's particularly important if one is to guard against errors in word lists - apparently impossible forms do show up in lists occasionally. RichardW57 (talk) 12:45, 19 May 2020 (UTC)

"[D]o we need to generate a citation?" Yes, all Wiktionary entries can be challenged, and should have a quotation or reference that can be added to them. It has long since been established that this is fair use.

"Some sources may have a significant problem with permanence" Only durably archived media count, as explained elsewhere in the CFI. We have traditionally counted carved stones, and even if the original is destroyed, faithful representations like drawings or photographs suffice (cf. the Golden Horns of Gallehus). By the way, these are uses, not mentions, which you seem a bit confused about.

"Are there any examples of such lists?" They're not usually real lists in practice. See WT:ARAP for an example.

"What is the 'community of editors for that language'?" It could just be one person. This is just to say that people who edit a given language are the ones who should decide. —Μετάknowledge^{discuss/deeds} 20:37, 20 May 2020 (UTC)

By asserting fair use, you are saying that Wiktionary is protected, even if the author isn't. Apparently English law doesn't know the concept of 'fair use'. RichardW57 (talk) 21:47, 20 May 2020 (UTC)

I was thinking of proposing that a mention in certain dictionaries plus a quotation should suffice. I thought copyright was an issue with using quotations as citations, but it seems not as far as Wiktionary itself is concerned. However, durability may be an issue with the available quotations. A reference to a good dictionary (with a page number in some cases!) should be verifiable (but some of them may have very limited print runs); an impermanent quotation adds protection against dictionary errors or traps. Perhaps the impermanent quotations should go on the talk page, ready for any challenges! RichardW57 (talk) 21:47, 20 May 2020 (UTC)

A reference to a good dictionary will always suffice for LDLs. If you have a tweet or blogpost that confirms usage, putting that on the talkpage is certainly not a bad idea, but isn't necessary (think of how rare an error in a good dictionary is). —Μετάknowledge^{discuss/deeds} 03:14, 21 May 2020 (UTC)

I've clearly been too generous with the word 'good' - some of the dictionaries I have in mind have errors or dubious pronunciations that even I can spot. Possibly in single figures (if you ignore inaccurate typography), but still there. And that's overlooking the inconsistent alphabeticisation, which is why I mentioned page numbers. --RichardW57 (talk) 05:44, 21 May 2020 (UTC)

@RichardW57 in many other countries there is the w:Right to quote. In the US the right to quote is covered by fair use instead. Alexis Jazz (talk) 10:24, 25 June 2020 (UTC)

I've been thinking about proposing a separate CFI for endangered languages, but I haven't completely thought it through. The idea would be to have fluent speakers accredited by us as references, and arrangements made to have the information they provide archived. This would require:

1. A gatekeeper, preferably a linguist who works with the language and knows who the fluent speakers are, as well as who can be trusted to collect information from those who aren't able to contribute directly.
2. Fluent speakers who are the sources
3. Volunteers, preferably from the community where the language is spoken. They would gather data and add it to the repository, as well as entering it into the dictionary.
4. A repository. Some place where a permanent record of the data gathered from the fluent speakers can be saved for posterity
5. A Wiktionary liaison/sponsor. A regular here who can help with the process of converting the raw data into dictionary entries, either as an advisor or an active worker.

I can see how more than one of the roles might be covered by a single person: for instance a fluent speaker might be their own gatekeeper and perform some of the volunteer roles- or even learn how to edit entries and do everything themselves. Or the volunteer roles might be split into a number of sub-roles.

Let's say, for instance, that a school in one of these communities has a literacy program. The teacher could be the gatekeeper and recruit speakers from the elders in the community, the repository could be set up at the school or a community center and the students could be at least part of the volunteer force- they would be doing some of the same work anyway as class assignments.

The only way this would work on our end would be to certify and mark in some way the accounts that are the conduits for authentic data so that we can recognize edits by vandals and good-faith editors who aren't qualified. This would be especially important after the people directly involved move on.

The whole thing would be a lot of work and require skill in getting all the pieces to fit, but if we do it right we could play an important role in preserving languages that might otherwise simply vanish as the last speakers die. It would be good for the dictionary and provide good PR for Wikimedia. Chuck Entz (talk) 03:49, 21 May 2020 (UTC)

Requiring all that would be dangerously close to banning many LDLs. The best is the enemy of the good. --RichardW57 (talk) 05:44, 21 May 2020 (UTC)

I don't think you understand Chuck's proposal... —Μετάknowledge^{discuss/deeds} 05:51, 21 May 2020 (UTC)

To be clear, my idea would be in addition to what we already have, not a replacement. If the language has coverage in reference works, no need to go through the hoops I described. It's just that there are too many languages where the only complete source for the language is the memory of a few elders who may not be around much longer. Anything that can be done to bring recognition to the importance of these elders and of their community can only help- simply put, they are the language, and our role should be to record and preserve what they have to offer. Chuck Entz (talk) 07:21, 21 May 2020 (UTC)

It's a good idea, if anyone wants to do it. It seems like one hurdle is that we don't have many/any linguists asking to use the site like that, so it might require outreach from us. (As for marking "accounts that are the conduits for authentic data": once the gatekeeper gave us the list of approved accounts, we could list them all on an admin-protected page as is done to approve AWB users. Changes by users not on the approved list could then be identified and rolled back.) On a smaller scale, I've thought (ever since we had a user who was working with the Wauja) that if a linguist or native school etc finds, for whatever reason, that it's easier to collect and store data on our site than on their own blog, there's no reason they can't use their userspace for that...that would preserve the data somewhere accessible, while requiring less/no effort from us to validate it since it wouldn't be part of the dictionary (but who added what would still be recorded in the edit history, if the linguist wanted to go back and validate it themselves before publishing it somewhere that we could then cite). - -sche (discuss) 03:06, 25 May 2020 (UTC)

Ukrainian extended animacy features

We've got Category:Animate nouns by language and Category:Animate nouns by language and Category:Inanimate nouns by language.

Ukrainian features humans, creatures (all animals and some mythological creatures) and objects (everything else, all other nouns).

The difference between the three only show in accusative plural.

1. Humans - accusative plural coincides with genitive plural

   люди́на (ljudýna, “human”) - accusative plural люде́й (ljudéj) = genitive plural

2. Creatures - accusative plural coincides with genitive plural and nominative plural (two possible forms)

   вівця́ (vivcjá, “sheep”) - accusative plural ві́вці (vívci) = nominative plural or ове́ць (ovécʹ) = genitive plural

3. Objects - accusative plural coincides with nominative plural.

   мі́сто (místo, “city”) - accusative plural міста́ (mistá) = nominative plural

Module:category tree/poscatboiler/data/lemmas has these animacies: "animate nouns" = "nouns that refer to humans or animals" "inanimate nouns" = "nouns that refer to inanimate objects (not humans or animals)"

As you can see, these descriptions are imperfect for Ukrainian nouns.

If Module:category tree/poscatboiler/data/lemmas is the right module to maintain genders and animacy, can we please define additional animacies for Ukrainian?

What should they be called? "human nouns" (excluding animals) and "animal/creature nouns" (excluding humans)?

The feature is probably unique to Ukrainian (Not quite unique, see Polish below). Russian and Belarusian definitely don't have this feature, neither has Polish, as it seems or any other Slavic or any other language.

Interesting enough a noun may have a different declension type dependent on the sense. If inflections are ever automated, then we definitely need to have distinction.

1. лев (lev, “lion”) - (creature) accusative plural ле́ви, ле́вів
2. лев (lev, “lev, Bulgarian currency”) - (object) accusative plural ле́ви
3. жук (žuk, “beetle”) - (creature) accusative plural жуки́, жукі́в
4. жук (žuk, “rogue, crook”) - (human) accusative plural жукі́в

@Benwing2, Rua. --Anatoli T. ^{(обсудить}/^вклад) 02:10, 22 May 2020 (UTC)

@Atitarev I added code 'anml' = animal to Module:gender and number, which is where these distinctions are maintained. I also added the appropriate categories to Module:category tree/poscatboiler/data/lemmas. See змій (zmij) and жук (žuk) for examples. Benwing2 (talk) 04:16, 22 May 2020 (UTC)

@Benwing2: Excellent, thank you! --Anatoli T. ^{(обсудить}/^вклад) 04:23, 22 May 2020 (UTC)

@Benwing2: Can terms with more than one gender/animacy please be added to Category:Ukrainian nouns with multiple genders: execting these змій (zmij), жук (žuk) and лев (lev)? --Anatoli T. ^{(обсудить}/^вклад) 04:29, 22 May 2020 (UTC)

@Atitarev It's not very easy to do this automatically because it requires looking at the contents of the page to find the invocations of {{uk-noun}}, but it's possible; Module:descendants tree does something like this. I'll look into it. Benwing2 (talk) 04:52, 22 May 2020 (UTC)

(edit conflict) @Benwing2: Is this right? A term like че́шка (čéška) belongs to the category, although there are multiple animacies, not multiple genders? --Anatoli T. ^{(обсудить}/^вклад) 04:56, 22 May 2020 (UTC)

After the "ec". I see. Maybe it needs a cleanup. The Russian че́шка (čéška) doesn't belong to multiple gender cat. --Anatoli T. ^{(обсудить}/^вклад) 04:56, 22 May 2020 (UTC)

@Atitarev It's easy to add a noun like че́шка (čéška) to Category:Ukrainian nouns with multiple genders because both inanimate and animate genders are specified in the same template call. It's harder to do this with nouns like змій (zmij), жук (žuk) and лев (lev) where the different genders/animacies are in different template calls. Benwing2 (talk) 04:59, 22 May 2020 (UTC)

@Benwing2: I understand. My question is, is this categorisation correct? Ukr. че́шка (čéška) belongs to only one gender (feminine) but two types of animacy. The Russian че́шка (čéška) doesn't do that. So, maybe the same template call shouldn't say "multiple genders" when the gender is the same, even if there are different animacy types? --Anatoli T. ^{(обсудить}/^вклад) 05:07, 22 May 2020 (UTC)

@Atitarev I can fix that so it says Category:Ukrainian nouns with multiple animacies for nouns with different animacy categories. Benwing2 (talk) 05:57, 22 May 2020 (UTC)

@Benwing2: That would be great. Also applicable to other languages, Russian, Belarusian, etc. BTW, I am shocked to see that Polish nouns, which must be similar to Ukrainian don't categorise by animacy:

1. człowiek (“human”) - accusative plural ludzi = genitive plural, identical to Russian or Ukrainian

   niedźwiedź (“bear”) - accusative plural niedźwiedzie = nominative plural, like one of the variants of Ukrainian but, like both Russian and Ukrainian have accusative singular niedźwiedzia = equal genitive singular

   dom (“house”) - accusative plural domy = nominative plural, identical to Russian or Ukrainian.

I am not working with Polish but I think it's a big oversight. So, the situation with Ukrainian is not unique but has its specifics. Note that lables but no categories associated with labels. Advising @BigDom, Shumkichi, editors I found working with Polish. --Anatoli T. ^{(обсудить}/^вклад) 08:06, 22 May 2020 (UTC)

@Atitarev I'm a bit confused. The examples you cite show that Polish uses anim = animals, pers = people and inan = objects, exactly as I originally did before introducing a separate gender/number category for animals. The only thing they don't do is create categories like Category:Polish animate nouns, Category:Polish personal nouns, Category:Polish inanimate nouns, which is easy enough to set up. Benwing2 (talk) 22:02, 24 May 2020 (UTC)

@Benwing2: Yes, it's confusing but I didn't mean to confuse. To me, Ukrainian (and Polish) personal and animate nouns are all animate, so splitting animates by animals and persons seems the right thing to do. Persons are also animate, aren't they? That's why I suggested new categories. Polish nouns may follow different from Ukrainian grammatical conventions, I don't know, I didn't research. I called two Polish editors in hope of clarifications. I agree it should be consistent (e.g. both Polish and Ukrainian) and no need to create unnecessary categories. If there are no objections, maybe we should add new Polish categories, which will be the same as Ukrainian? --Anatoli T. ^{(обсудить}/^вклад) 22:52, 24 May 2020 (UTC)

Appendix to mention phenomena we don't include ("fasque", "I hates it", "selahs", etc)

I think we should have an Appendix,* either one per language or just one overall, to record general phenomena in various languages that we decided not to give entries or sense-lines to even though someone unfamiliar with a language might look it up. (This could also house or be the same thing as the "WikiGrammar" page Equinox has suggested.) For example, for Latin, we decided that since any word can be suffixed with -que we don't include -que forms even though they look like single words and someone unfamiliar with Latin might look them up; for Tzotzil, we don't include -e forms; for English, we don't give separate sense lines for the likewise lexicon-wide use in some dialects of e.g. the third-person singular as the first-person singular or vice versa ("I hates it"), and we've deleted some entries that were just "plural" or even "singular" forms of (non-noun) words meaning "occurrence(s), in a text, of [that word]", even though, again, they look like single words.
*(Or perhaps a Wiktionary-space page, perhaps even our former "About English"-type pages, but that one is now more clearly named as a Wiktionary-internal, inward-facing page on "English entry guidelines", and in general I think Wiktionary-space pages are for Wiktionary-internal things as opposed to the Appendix-space which is theoretically part of the outward-facing dictionary.)
Thoughts as to what such an appendix should be named, and whether it should be separate from our existing "About English"/"English entry guidelines"-type pages? (Edit to add for clarity: the appendix would not list all Latin words suffixed with -que, for example, but have a general notice that for Latin no -que forms are included, etc.) - -sche (discuss) 19:48, 22 May 2020 (UTC) edited: - -sche (discuss) 00:08, 23 May 2020 (UTC)

There is Wiktionary:Criteria_for_inclusion#Exclusions. Equinox ◑ 20:00, 22 May 2020 (UTC)

Some of these exclusions are worth drawing to the attention of noobs trying to understand a foreign text, and may well belong in public-facing notes on the language. -- RichardW57 (talk) 22:36, 22 May 2020 (UTC)

I've always thought we should have separate "About" pages for editors and readers. Right now the main emphasis is on laying the ground rules for editors so their edits are consistent with community consensus for a language. That's important. but there's also need for the equivalent of the introduction in a print dictionary. Things like the basic structure of the language, and how the morphological and syntactic features are represented in our entries, the main varieties of the language, etc. The idea is to help readers with where to look for things, how to interpret the shorthand we use in the entries, and some of the unwritten assumptions that tend to show up only in rfd, rfm and rfv discussions. Where do we explain the reason that we don't cover English possessives in "-'s", or have adjective sections in English for attributive nouns? Or that what is called the present tense isn't like the present tenses in other languages?

I would envision a reader's guide and an editor's guide for each language, with the current "About" pages being changed to editors' guides to start with, and reader's guides being added as people have the time. Although there would be considerable overlap with the Wikipedia coverage of the languages, I would contend that we need a presentation that focuses specifically on what would help our readers get the most out of our dictionary. Chuck Entz (talk) 23:53, 22 May 2020 (UTC)

Thanks, Equinox, for reminding me of that section. :) It links to our inward-facing "Wiktionary:English entry guidelines". I would suggest (in line with Chuck) that we have a separate, public-facing page like "Appendix:About English" (or something). Then, WT:CFI#Exclusions could link to (let's say) the category of all such appendices and (representatively) the appendix for English (instead of linking as it does now to the inward-facing WT:AEN which never actually documents many of the things we exclude for English). For languages where we already use the mid-dot to link to the pages where we explain our transliteration schemes, we could add some "see also" links to the bottoms of those pages to promote the visibility of the corresponding "exclusions" pages. I also think we wouldn't need that much overlap with Wikipedia; looking at the lead of English language, only a few sentences seem like ones we'd mirror here. - -sche (discuss) 00:31, 23 May 2020 (UTC)

Language Log on 'rona-related hot words

How many of these look durable and sufficiently tested enough? https://languagelog.ldc.upenn.edu/nll/?p=47179 —Justin (koavf)❤T☮C☺M☯ 01:45, 23 May 2020 (UTC)

None of them. Wait a year and if any of us are still alive they can be added. DTLHS (talk) 02:12, 23 May 2020 (UTC)

The article is a bit sensationalistic. It curiously does not include some more likely candidates for Wiktionary, including a verb sense for Zoom, which predated the pandemic in oral use (but finding written use is difficult). —Μετάknowledge^{discuss/deeds} 02:18, 23 May 2020 (UTC)

We have got covidiot (after some initial wrangling; see talk). Equinox ◑ 14:19, 23 May 2020 (UTC)

When does the hot word clock start

Sometimes someone adds a "hot word" which only become "hot" enough within the last year to enter durable media, but which existed before then in non-durable media. For example, here are tweets mentioning murder hornets from 2013 (with link to news article clarifying that the Asian giant hornet is meant and not, as in some tweets, any hornet the speaker is dysphemizing), 2015 (with photo), etc, but it's tagged as a hot word from 2020, when it became common in durable media. (Another example, where I later found older durable cites and could just untag the word, was gender-critical.) I'd think we'd start the "hot word" clock from the first use even if it's not durable, because we're aiming to allow words which simply can't meet the "spanning a year" criterion because they didn't exist before, but which meet all the other criteria and seem sure to stay in use, like words for newly discovered species and chemical elements. (OTOH, I understand the desire to include words that people are currently trying to look up, like murder hornet, and which probably will continue to see use in durable media now that they've "broken through".) What to do? If it's not attestable yet, we could always {{no entry}}-redirect murder hornet to Asian giant hornet so people looking it up still reach content. - -sche (discuss) 19:19, 23 May 2020 (UTC)

I think we should aim to include the earliest uses of a word even if they didn't occur in durable media. While such usages may not count towards attestation, I think their existence would certainly justify "hot word" status even if a word hadn't made it into durably archived print media yet (especially since it usually takes a bit of time for that to happen after a word appears). I also think something we should provisionally include popular words that are not really "hot", but are common on social media and Internet forums, even if they are not durably archived, since it might take a while before a novel or something includes them. But that's another conversation. Andrew Sheedy (talk) 19:28, 23 May 2020 (UTC)

In theory I agree with this because if we later want to find out how a term originated we won't care "when was it first posted to Usenet" (lol) we will probably care about Twitter or something. The OED uses Twitter! The problem isn't that, really, the problem is that without durability (i.e. having a meaningful "resource identifier" whether it's URI or ISBN) it's easy for vandals to make shit up. "It hasn't been a big problem yet!" you cry. That's because we have CFI. Wait until psychosplattermatic comes up for RFV in 2025 and all the sources are from some shit like Instagram that disappeared a decade earlier. Equinox ◑ 21:51, 23 May 2020 (UTC)

It is easier to make things up by referring to books, especially now when libraries are closed. With ephemeral websites, at least things were accessible for some time to check. Therefore such citing could make a request for verification pass: Like it went through the process once and editors confirmed, so no need to later complain that the sources are gone – Wiktionary’s archives then guarantee the legitimacy. Of course one would require interaction, like a number of choice editors putting their checkmark. Fay Freak (talk) 22:11, 23 May 2020 (UTC)

If we are genuinely concerned about this, we can have a validation system like at s:. —Justin (koavf)❤T☮C☺M☯ 22:41, 23 May 2020 (UTC)

To ask my question in a more concrete/specific way: should "murder hornet" be deleted as a "hot word older than a year" (in that it has existed for more than a year on e.g. Twitter, and durable uses of it fail the "spanning a year" criterion), or should it be kept as a "hot word newer than a year" because the first durable uses weren't until this year? - -sche (discuss) 06:48, 24 May 2020 (UTC)

Lack of durable uses could have led to its deletion to begin with, though I could understand why no one would bother. I suppose that only durable uses start the clock. DCDuring (talk) 17:20, 24 May 2020 (UTC)

I think this not a question of logic, but a question of policy. Giving hot words a relatively long time to get attestation doesn't bother me. A hot word that can't be attested in durable media after a full year seems well worth an RfV, but not an automatic deletion. DCDuring (talk) 17:27, 24 May 2020 (UTC)

"idiomatic" label

The label {{lb|en|idiomatic}} creates a link to Appendix:Glossary#idiomatic, where the word is defined as meaning "Pertaining or conforming to the mode of expression characteristic of a language". Yes, this is one definition, but is it the one relevant to this label? Shouldn't this label apply only to "idiomatic" in the sense "not (easily) understandable from the individual words/parts"? Mihia (talk) 19:36, 23 May 2020 (UTC)

Personally, I think many (if not all) uses of the label are inappropriate, and am prone to remove it when I come across it, on the basis that if a term wasn't idiomatic we wouldn't (normally) include it, though I concede that's debatable. - -sche (discuss) 02:42, 25 May 2020 (UTC)

You may be right that it is often a bit pointless. It certainly isn't consistently applied. But if/where it is used, do you think it has, or should have, the sense "not (easily) understandable from the individual words/parts" rather than "pertaining or conforming to the mode of expression characteristic of a language"? Mihia (talk) 17:50, 28 May 2020 (UTC)

Decimal integers in Thai citations

The Thai example template {{th-usex}} arranges three lines for a citation: the text in Thai script, a transcription, and a translation. The text and transcription are processed as a single string. The template attempts to convert each word but the lemma itself into a reference to a Wiktionary entry. Like English, Thai has a few digit strings that are Thai rather than translingual, such as 555 (555) or allegedly 69 (69) (currently merely a red link).

How should translingual digit strings be handled? I have added notes on current implementation because I can foresee process being made only incrementally.

Firstly, should they be transcribed as pronounced, or simple written as a string of European digits? The second question arises if they should be transcribed as pronounced.

Secondly, most multi-digit numerals are, if I understand correctly, a sequence of words when written out by hand. I see four cases:

1) If Wiktionary has, or should contain, the whole number as a translingual entry, I believe that entry should be referenced. Is it acceptable for the transcription to suggest that the number is a whole word in Thai even though it isn't, or should it show the breakdown into words? @Octahedron80, Bitterschoko The template's Lua currently needs some modification to show it as multiple words, and there may be good reason for deliberately making that difficult. For example, {{th-x|14{สิบ-สี่}|fourteen}} yields

14

sìp-sìi

fourteen

but {{th-x|14{สิบ สี่}|fourteen}} currently raises an exception.

2) If the number does not merit an entry, and digits do not share words, it at first sight seems reasonable to make each digit in the text into a link to the digit. We may ultimately have to fix the template to prevent line breaks occurring within the decimal number. @Wyang

3) Zeros with no corresponding sound can simply be written as a non-linking zero with no transcription. The template needs modifying to accommodate this. @Atitarev

4) The number might not merit an entry, but digits share words. I think we don't have to worry about this one because it doesn't occur. For example, for 20,000,000, I envisage that being split up into "20" with transcription ยี่-สิบ-ล้าน and then six non-lining zeros. An example that needs tweaking with as yet undefined flags is {{th-x|20{ยี่-สิบ-ล้าน},000,000 คน|20 million people}} or {{th-x|๒๐{ยี่-สิบ-ล้าน},๐ ๐ ๐,๐ ๐ ๐ คน|20 million people}}, which yield:

20,000,000 คน

yîi-sìp-láan , 0 0 0 , 0 0 0 · kon

20 million people

๒๐,๐๐๐,๐๐๐ คน

yîi-sìp-láan , 0 0 0 , 0 0 0 · kon

20 million people

--RichardW57 (talk) 23:17, 24 May 2020 (UTC)

@RichardW57: Hi. I didn't get your ping. Sorry I am not module-savvy and Wyang has left Wiktionary, so maybe someone else can help you. The Chinese module Module:zh-usex can handle what you require, so it is possible, e.g. 14  ―  shísì  ―  14. So the corresponding Thai module Module:th (the usex part) needs to be able to handle a new parameter @ to suppress the linking. --Anatoli T. ^{(обсудить}/^вклад) 10:14, 26 May 2020 (UTC)

Thanks. You've answered the technical question I had for you - you've identified input conventions that seem to be workable. Lua doesn't scare me. My question was what we should aim to do, which is why I raised this in the Beer Parlour rather than the Greasepit. I'm a bit surprised that no-one has chimed in that they hate quotes being converted to a series of links. --RichardW57 (talk) 11:28, 26 May 2020 (UTC)

'wrestling' label vs. 'professional wrestling'.

A number of terms (face, heel, kayfabe) have senses labeled using {{lb|en|professional wrestling}} (which doesn't add a category) or {{lb|en|professional|_|wrestling}} (which adds it to Category:en:Wrestling). There's some overlap here, but I think the concepts of wrestling and professional wrestling are different enough that's it's worth adding 'professional wrestling' to Module:labels/data/topical and using the former label-type to point to it. I wanted to get a bit of feedback first, as I haven't played with these before. Any thoughts? grendel|khan 02:49, 26 May 2020 (UTC)

Agreed, professional wrestling is a very different thing than the proper grappling sport and in particular is very different in its terminology. —Justin (koavf)❤T☮C☺M☯ 02:54, 26 May 2020 (UTC)

I agree. DCDuring (talk) 13:45, 26 May 2020 (UTC)

I have gone thru Category:en:Wrestling to take out some terms that I know are related to pro/entertainment wrestling and not Greco-Roman/sports wrestling but I'm not sure of all of them. Some of the moves may overlap both but I'm being conservative about taking things out of the category and putting them into the new one. —Justin (koavf)❤T☮C☺M☯ 23:17, 26 May 2020 (UTC)

Thanks for the feedback! See Category:en:Professional wrestling, which looks lovely. grendel|khan 23:50, 26 May 2020 (UTC)

Should CAT:Professional wrestling be a subcategory of CAT:Wrestling? —Mahāgaja · talk 04:58, 27 May 2020 (UTC)

I had considered this myself and I lean toward no, as professional wrestling is really not a type of Greco-Roman–style wrestling. It's descended from it but has diverged so far that they are fairly unrelated. I'm open to someone arguing me out of this position, tho. —Justin (koavf)❤T☮C☺M☯ 10:51, 27 May 2020 (UTC)

Well, wrestling doesn't just mean Greco-Roman and professional, right? The general category CAT:Wrestling could/should include terms relating to sumo, Turkish oil wrestling, mud wrestling and so forth. —Mahāgaja · talk 11:16, 27 May 2020 (UTC)

That's true. The others are all actual grappling martial arts but professional wrestling is theater and sports entertainment. Still on the fence, I suppose but there is a distinction between the athletic events and the pre-scripted reality TV. And for that matter, backyard wrestling... —Justin (koavf)❤T☮C☺M☯ 11:49, 27 May 2020 (UTC)

Shouldn't it also be a subcategory of Theater? DCDuring (talk) 15:41, 27 May 2020 (UTC)

Theater, performance art, something. —Justin (koavf)❤T☮C☺M☯ 18:27, 27 May 2020 (UTC)

Names and the criteria for inclusion

See Wiktionary:Requests for deletion/English#Dog.

I don't even necessarily oppose removal of that sense, but there should be no reason to rely on reasoning like that's common sense!, ridiculous, include only "proper" names and last but not least: stupid entry.

Common sense should be an absolute last resort for rare edge cases. Included in policy or not, we should be able to explain why Dog as a name shall not be included without resorting to name-calling. If we are unable to, that's an invitation for making arbitrary decisions.

Are names to be excluded if the name is (a form of) the noun that identifies the subject? People who are named "Human", "Female", "Male", "Man", "Woman", "Boy", "Child". Cars called "Car" or "Automobile", practitioners of magic called "Wizard" or "Magician" and we should probably unlink "Disk Operating System" on DOS.

That would exclude Dog, Doggie, Chihuahua and Hondje as dog names, but doesn't exclude them as names for other animals. (which can also be attested at least for Dog)

So perhaps we'd say that a name must be restricted to a certain kind of group which shouldn't span multiple species. But now we must discuss Spot which could be a name for anything from an elephant to a bunny all the way to a turtle.

I also suspect we want to exclude names like Desire, Dangerous and Demonic regardless of whether those names can be attested, but correct me if I'm wrong.

I'm guessing (fair warning, just a guess) that the actual reason several people want to exclude Dog (and likely Dangerous and Demonic) is that the writers/name givers are just messing with people. I don't think that'll make for a good exclusion basis though because we can't (and imho shouldn't) try to judge what is serious and what isn't.

We may or may not append the policy, but I suggest we try to get down to exactly why we should(n't) exclude Dog, Cat, Human, Desire, Dangerous, etc. to avoid making decisions arbitrarily. If we remove Dog as a name that's fine, but not because it's a "stupid entry". Alexis Jazz (talk) 07:01, 26 May 2020 (UTC)

I suspect that it's because English speakers (some at least) feel that names are what is grammatically an albeit large closed category; one can't just take anything pronounceable and make it a name. By that criterion, names are part of the language, where arbitrary inventions don't take off - a word is usually useless if no-one else understands it - and will fail the CFI. Now in Thai, reportedly any word may be used as a name, and accordingly the Thai-editing community here has decided to completely exclude personal names. The alternative they face is that eventually every Thai page will hold a proper noun, and such proper noun entries will be largely useless. It would also be a way for brand names to creep in - we know a Thai lady named 'Fanta', after the drink. For English, capitalisation would usually make them separate pages. --RichardW57 (talk) 09:37, 26 May 2020 (UTC)

Perhaps we should say that there is a grammatical process that can make any noun (certainly concrete, and probably also abstract), and perhaps any adjective of the right sort, into a proper noun, and in general refuse to document them on that basis. That wouldn't help with non-standard spellings. --RichardW57 (talk) 09:37, 26 May 2020 (UTC)

If my memory serves me right, one can't just take three independent English language gazeteers, and dump their common contents into Wiktionary - geographic names have to be for places of some notability. We're looking for a rule of that sort - though what we have in that case is more of a guideline. That would allow use to keep proper nouns like Frank. --RichardW57 (talk) 09:37, 26 May 2020 (UTC)

We wouldn't want to exclude Faith, Hope, and Charity. Desire was used as a given name by Puritans. Desirée/Desiree is not rare. DCDuring (talk) 14:39, 26 May 2020 (UTC)

Hondjee would be pretty reasonable as an English name for a dog, as would Beast. Having allowed names to be included, we will inevitably have names some of us will find frivolous or otherwise inappropriate. I don't think we actually have any notability standard for inclusion. DCDuring (talk) 14:48, 26 May 2020 (UTC)

There's an attempt in that direction at Wiktionary:CFI#Place_names. I am not claiming it's succeeded, but it's an idea of a principle to strive to implement. --RichardW57 (talk) 16:56, 26 May 2020 (UTC)

@DCDuring: I just thought of something. Desire wouldn't be hard to attest even for current use. You don't want to Google these at work, but several if not all are probably durably archived on DVD: Desire Ware, Desire Mandrile, Desire Moore, Desire Collins, Desire Sevilla, Desire Delgoto and Desire Taylor. I just stumbled upon the reason why they don't meet the criteria for inclusion: With respect to names of persons or places from fictional universes, they shall not be included unless they are used out of context in an attributive sense. If only someone had cited that instead of calling names on Wiktionary:Requests for deletion/English#Dog.. We might also consider adding a requirement for names to be attested in non-fictional media, but that may not even be needed right now. Alexis Jazz (talk) 13:01, 27 May 2020 (UTC)

CFI Place names does not have any implications for exclusion of place names. It provides a safe harbor for some classes of names that may have entries. CFI Fictional universes does not have much bearing of names of real people. I don't see any reason whatsoever to want to exclude any given names that are in use. What do you think we should try to do with regards to given names? Which names seem to you not worthy of inclusion? Why? DCDuring (talk) 14:25, 27 May 2020 (UTC)

We could enact some rule that required any given name definition to be attested in durable media before being entered. That is it could be deleted on sight if without attestation. DCDuring (talk) 14:28, 27 May 2020 (UTC)

"Which names seem to you not worthy of inclusion? Why?"

@DCDuring: I'd say any names (be it for people, things, organizations, countries, planets, etc) that can't be attested in non-fictional works or in direct reference to the non-fictional work the name came from. So any mention of C-3PO in a fictional work doesn't count. "C-3PO is a cute robot" in a review (non-fiction) of a Star Wars movie is a direct reference, so also doesn't count. But "The performances were about as stiff as C-3PO" in a review of a Broadway musical would count, as would "Lisa Simpson is essentially the C-3PO of the Simpsons" in a review of The Simpsons. (I'm just making stuff up, okay?) I also support your suggestion to require attestation before being entered. Alexis Jazz (talk) 16:30, 27 May 2020 (UTC)

But isn't the treatment of words from fictional universes a settled matter? The specific names mentioned in the initial comment under this topic are likely attestable in the real world. That they are also used in fictional universes is immaterial. DCDuring (talk) 21:23, 27 May 2020 (UTC)

"Common sense should be an absolute last resort for rare edge cases." I think calling a dog (or even more so another animal) "Dog" as its actual name is a rare edge case, and agree too much time has been wasted on it. Equinox ◑ 14:33, 27 May 2020 (UTC)

Now let's talk about Bunny, Kitty (wait.. why is that sense there?), Doggie and Babe. I don't consider this an edge case and a lot of wasted time could have been avoided if someone had referenced that other CFI bit sooner. And at any rate, those who have just been calling names were not contributing to a sensible reasoning for exclusion, only to making the environment more hostile. Alexis Jazz (talk) 16:30, 27 May 2020 (UTC)

@Alexis Jazz: I gather that you are unhappy with my RFD comments, which you have liberally quoted above. Let me explain to you that I am dismissive of this entry because it seems like a joke -- and if not originally meant as a joke then appearing as one to the average reader. "Dog = a name given to a dog" is surely well worthy of Wiktionary:Bad jokes and other deleted nonsense. I find it rather absurd that serious analysis and consideration should be given to it, that we should be searching out citations of people who call their dog "Dog", or that this general kind of titting around should be required in order to get rid of such patent nonsense. Mihia (talk) 22:14, 28 May 2020 (UTC)

Should non-lemma forms only attested in certain expressions be included?

As per title.

The particular problem in this case is that in Finnish, country names usually don't have plurals (they do exist grammatically, but make little semantic sense); despite this, the plural is used in the expression tehdä X:t, which roughly means something like "do a/an X". Islannit (literally “Icelands”) is an example that is currently in both RFD and RFV; tehdä Islannit would mean "do an Iceland", whatever the exact meaning is depending on some context. This expression is not exclusive to Iceland, and any other country name or even any other name can be substituted (which I suppose makes it a snowclone?)

Should the plurals of country names be included in such a case, and further, should they be listed under the inflection table? If they should, shouldn't there be some kind of warning (perhaps something like "chiefly uncountable" or "usually uncountable")? — surjection ⟨??⟩ 21:33, 26 May 2020 (UTC)

I think {{only used in}} could be used, which then points to the idiom. —Rua (mew) 07:22, 27 May 2020 (UTC)

Possibly, but should those forms be listed in the inflection tables then? — surjection ⟨??⟩ 12:02, 27 May 2020 (UTC)

Do the plurals of country names occur in any case other than the nominative/accusative? It would be silly to list all 15 cases of a plural country name if 14 of them are never attested. Can the inflection table be told to show only the nom./acc. and show em dashes for all other cases? —Mahāgaja · talk 12:18, 27 May 2020 (UTC)

Per the verb expression, they can show up as genitive when a noun is formed with -minen (Islantien tekeminen; whether this is attestable is another matter entirely), or as partitive when the verb is in negative (ei tee Islanteja). No other cases in plural would be used. — surjection ⟨??⟩ 12:22, 27 May 2020 (UTC)

No Finnish speaker ever talks about comparing two Finlands, eg., old Finland with new Finland? 15:45, 27 May 2020 (UTC)DCDuring (talk)

You can, but that's fairly rare and probably not attestable for most countries. (It seems English entries don't have plurals for country names either, despite the usage also existing.) — surjection ⟨??⟩ 16:10, 27 May 2020 (UTC)

I suppose you can speak of the two Germanys and the two Koreas in Finnish? —Mahāgaja · talk 16:15, 27 May 2020 (UTC)

Those are two of the obvious exceptions, yes. — surjection ⟨??⟩ 16:32, 27 May 2020 (UTC)

Well, if the inflection table doesn't already have such control, and it seems not to, it is always possible to add it. The Pali verb templates have such controls because the middle voice can be poorly attested, and the language seems to have a lot of irregularities. --RichardW57 (talk) 18:50, 27 May 2020 (UTC)

One idea, for countries only attested in the plural a few expressions and grammatical case forms, if having a declension table that has a whole column for "plural" forms with just a "nominative plural" field that only occurs in one(?) set phrase is considered too misleading: make the declension table have a little footnote at the bottom saying something like "a plural form x (nominative), y (genitive), z )partitive) occurs in certain set phrases" (or add that to entries as a usage note, but keeping it in the declension table is probably easier from a technical standpoint and possibly also more sensible). For cases where other forms are attested, i.e. the word is generally pluralizable, like "Korea", the table could list plural forms "as usual" i.e. in their own column etc like for other nouns. Re "It seems English entries don't have plurals for country names": this is only due to oversight; I've added attested plurals whenever I've come across such entries, also to language names and personal names. - -sche (discuss) 18:32, 27 May 2020 (UTC)

As to the general question: yes, if an inflected form is attested, we should have an entry so someone looking it up can figure out what's going on, and if it's only attested in < 3 phrases, then {{only used in}} seems like exactly the thing to use, probably in conjunction with (and on the same line as) some kind of "plural of X" template (in this case), and wikilinks to break the linked-to phrase up into its constituent words if it's SOP the way "do a X" is. In turn, mentioning the form somewhere on the lemma entry also seems helpful. - -sche (discuss) 18:37, 27 May 2020 (UTC)

That is similar to what I had in mind, but considering some of the arguments above concerning some other cases where plurals might exist, it's probably going to be a better idea to list all plural forms as normal but have some kind of a note about how the (proper) noun is usually not used in plural. Does any language have something like that right now? — surjection ⟨??⟩ 19:47, 27 May 2020 (UTC)

English entries mark words like rain and greenery "usually uncountable". —Mahāgaja · talk 05:04, 28 May 2020 (UTC)

Mandarin romanisation

(Notifying Tooironic, Suzukaze-c, Justinrleung, Mar vin kaiser, Geographyinitiative): : User:Geographyinitiative has created a bunch of Chinese geographical names in English where in etymologies he uses |tr= with Wade-Giles (WG) transliterations. For example, Hsi-ning in this revision, derived from from 西寧／西宁 (Xīníng), romanised as "Hsi¹-ning²" in WG.

I find this inconsistency unacceptable. Users only expect Hanyu Pinyin (pinyin) as the only transliteration of Chinese Mandarin. We never mix them, this is counter-intuitive and confusing. Different romanisations have very different reading rules. users wouldn't know how to read "hsi", "chi", etc. or would make wrong assumptions. I suggested to display WG in quotes ("Hsi¹-ning²") but got into edit-war with the user. While WG is helpful to understand the etymology of the English spelling, we shouldn't replace the current standard of may years and create alternative transliterations. It's very messy. I don't think Geographyinitiative has any understanding of consistency and never tries to follow conventions. So that I am not accused of bullying, please decide what to do with this kind of entries. --Anatoli T. ^{(обсудить}/^вклад) 08:50, 28 May 2020 (UTC)

I agree that transliterations should be consistent, which means using only Pinyin for Mandarin. —Rua (mew) 09:00, 28 May 2020 (UTC)

The etymologies in question would be more helpful with Wade-Giles since the spelling is directly derived from the Wade-Giles romanization rather than Hanyu Pinyin. I'm okay with Geographyinitiative's previous edits, but I also think Pinyin is probably useful and should be there for consistency across the dictionary. We need to have a way of putting both Wade-Giles and Pinyin for these particular entries. The current layout at K’un-ming is not ideal since quotation marks are meant for glosses. — justin(r)leung _{{ (t...) | c=› }} 09:27, 28 May 2020 (UTC)

The |tr= parameter isn't the place to put alternative romanizations. Period. There's nothing wrong with giving the Wade-Giles form separately, but having some entries transliterated differently because someone feels a need to make believe they're being revolutionary would be silly if it weren't damaging to the usability of the dictionary. It would be like driving on the left on streets named after places in England and on the right in streets named after places in the US- you just end up with confusion, gridlock, or head-on collisions. Chuck Entz (talk) 09:37, 28 May 2020 (UTC)

@Chuck Entz: Your analogy with traffic makes sense, especially if you consider a situation in many Taiwanese cities where a simple syllable like "xi", for example as in 西 (xī, “west”) can appear in street signs as "xi", "hsi", "shi", "si", "see", "shee" plus possible readings and transliteratiuon from Hokkien. So, a poor tourist who has no idea about the fight over standard transliterations in Taiwan is clueless whether they are in the right place. The street signs may be different even on the same street in different places. (You can Google about it)--Anatoli T. ^{(обсудить}/^вклад) 00:38, 29 May 2020 (UTC)

I agree that the etymology section of Hsi-ning should use the pinyin transliteration of 西寧. But I notice that at 西寧 itself, {{zh-pron}} provides several transliterations, but Wade-Giles isn't among them. We're given Pinyin, Zhuyin, Gwoyeu Romatzyh, and Tongyong Pinyin but not Wade-Giles. I'd support adding Wade-Giles to the transliterations that {{zh-pron}} supplies. —Mahāgaja · talk 10:09, 28 May 2020 (UTC)

I don't edit Chinese entries, but that seems like a good idea since from my understanding Wade-Giles is a major (if dated) transliteration scheme. — SGconlaw (talk) 12:05, 28 May 2020 (UTC)

@Mahagaja, Sgconlaw: Wade-Giles is very important, no doubt. The reason why WG is not part of Chinese entries, is that nobody has added it yet. Nobody added WG not because they don't want to but for technical reasons. Not easy to map from Hanyu Pinyin, there's no one-to-one correspondence. I am personally keen to have it in entries. Pls see my question and User:Justinrleung's answer at User_talk:Justinrleung/Archive_17#Wade–Giles. --Anatoli T. ^{(обсудить}/^вклад) 00:38, 29 May 2020 (UTC)

These romanization schemes created words which form a part of the English language. We can't read our books without knowing what these words mean. If the English language dictionaries ignore these words, we literally cannot read. As long as the full Wade-Giles form is included on the page somehow in a way that helps explain the etymology of the word (that's the important point), that's all that's really needed. --Geographyinitiative (talk) 10:27, 28 May 2020 (UTC)

@Geographyinitiative: We had never an issue regarding what should be the standard transliteration for Mandarin. Wiktionary:About Chinese/Mandarin mentions pinyin and all examples in Wiktionary:About_Chinese#Translations_into_Chinese_lects, etc. always use Hanyu Pinyin. It has been the practice and default for many-many years and nobody questioned the policy and conventions that language code "cmn" and |tr= should be always followed by pinyin. It's also automatically generated by Chinese specific templates, like {{zh-l}}, {{zh-x}}, etc.

You keep breaking the rules about mixing transliterations (also from different Chinese varieties), so I call for a standardisation and formally enforcing what should be the Wiktionary policy on Mandarin transliteration. Every language at Wiktionary has a standardised transliteration, e.g. Wiktionary:About_Korean#Han-geul,_Romanization,_and_English_translation, WT:RU TR, so should have Mandarin. If people vote for your mess, so be it. I don't think anyone will support all kinds of transliterations mixed. If an alternative transliteration is used, it should be clearly marked so, which you failed to do in your edits. --Anatoli T. ^{(обсудить}/^вклад) 00:38, 29 May 2020 (UTC)

The full-form Wade-Giles has to be shown as part of the etymology to explain the origin of the English language term, and it doesn't make sense to put it in scare quotes. I would say try to fix the Hsi-ning page so that works within those boundaries. There's no other transliteration scheme that is put in scare quotes. Also, the ordering "Chinese Mandarin" is beyond bizarre- that will have to be changed too. --Geographyinitiative (talk) 00:47, 29 May 2020 (UTC)

@Geographyinitiative: How about: From Mandarin Chinese 西寧／西宁 (Xīníng), Wade-Giles romanization: Hsi¹-ning²? --Anatoli T. ^{(обсудить}/^вклад) 01:01, 29 May 2020 (UTC)

@Atitarev Okay, that looks great to me. I will try to use it on a page I am about to create: Pa-tung. --Geographyinitiative (talk) 01:05, 29 May 2020 (UTC)

@Geographyinitiative: You can make or request template, so that Wade-Giles romanization: Hsi¹-ning² is displayed with WG romanisation as the input. Justinrleung has removed "Chinese" from the etymology, just use {{bor|en|cmn|-}}. --Anatoli T. ^{(обсудить}/^вклад) 01:22, 29 May 2020 (UTC)

• Alternative proposal -- put the details about the etymology of the English terms in the English entries. Problem solved without any changes to any Chinese entries at all. ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:50, 29 May 2020 (UTC)

• @Eirikr: Yes, we are talking about English entries, such as Hsi-ning. --Anatoli T. ^{(обсудить}/^вклад) 01:01, 29 May 2020 (UTC)

• Anatoli, I have no particular opinion on including Wade-Giles in Chinese entries, but as suggested above by this thread, Geographyinitiative's current approach appears to require a massive reworking of Chinese entries -- as details for English entries. That seems backwards to me -- the pertinent details belong on the English entries. ‑‑ Eiríkr Útlendi │^{Tala við mig} 05:22, 29 May 2020 (UTC)

• @Eirikr: Adding Wade-Giles transliterations is good to have and they are important. They will explain how historical European names for Chinese places came about. It's a separate issue, though. Even if this transliteration (WG) is of value, it's not to replace the current standard. Like we don't use IPA instead of the proper transliteration. Everything must have its use. As for the English entries, displaying them separately with a label is always possible. A label and a cursive seems to be acceptable now but not with |tr=, this is reserved for the standard. Some work is required to standardise the convention. Since no-one challenged conventions before Geographyinitiative (don't forget, per regular rants, that we are all out trying to diminish Chinese dialects, alternative transliterations, everything to hurt what is so dear to him, LOL), now the convention should become a written standard. --Anatoli T. ^{(обсудить}/^вклад) 05:36, 29 May 2020 (UTC)

Ojibwe (Anishinaabemowin) verb categorization

I'm looking for advice on how to categorize verbs in Ojibwe (Anishinaabemowin, but likely also other Algonquian) languages. The litterature on Ojibwe grammar uses a 4-way categorization for verbs, as follows:

• VII = inanimate intransitive verbs = intransitive verbs that take an inanimate SUBJECT
• VAI = animate intransitive verbs = intransitive verbs that take an animate SUBJECT
• VTI = transitive inanimate verbs = transitive verbs that take an inanimate OBJECT
• VTA = transitive animate verbs = transitive verbs that take an animate OBJECT

This 4-way categorization is critical for understanding Ojibwe syntax, morphology, and derivation, and there are systematic (quasi-morphological??) relationships between verbs belonging to different categories. So, i don't think the words can simply be categorized under "verbs," which ignores the richness of the language and misses an important aspect of how speakers percieve and use the language.

The Appendix:Ojibwe_verbs page describes this categorization well, though in greater detail than is necessary for wiktionary, i think. The Category:Ojibwe_verbs page, and the entries themselves, are a mess of different and sometimes parallel approaches. Luckily, there are few enough entries (under 500), that a solid fix now would be relatively easy to apply.

Thanks for your help.

SteveGat (talk) 18:20, 28 May 2020 (UTC)

I have a hard time seeing why this situation is so unique. Compare for example, Russian nouns, which can be categorized as masculine, feminine or neuter as well as animate and inanimate. There are interactions between gender and animacy but we have no problem categorizing them all as nouns and then also categorizing them separately as e.g. "Russian masculine nouns" and "Russian inanimate nouns". Benwing2 (talk) 06:06, 29 May 2020 (UTC)

I don't know if it is unique or not. I am unfamiliar with a similar classification outside of the Algonquian languages. That said, after some reflexion, i think the best categorization might be the following:

• Level 1 - verbs
  • Level 2 - intransitive verbs
    • Level 3 - inanimate intransitive (VII) (subject is inanimate)
    • Level 3 - animate instransitive (VAI) (subject is animate)
  • Level 2 - transitive verbs
    • Level 3 - transitive inanimate (VTI) (subject is animate, object is inanimate)
    • Level 3 - transitive animate (VTA) (subject is animate, object is animate)

This would maintain the crosslinguistic transitive vs intransitive categories, and respect the standard analysis of Ojibwe verb classification. As for separate categorization, the issue is that the animacy feature does not apply to the verb itself, but rather to subject and object it takes, so we can't say a verb is animate or inanimate independently of whether it is transitive or intransitive. SteveGat (talk) 17:01, 29 May 2020 (UTC)

First of all, I agree with what both of you have said. One related question I have is which verb form should be used as the headword entry? For instance, for a transitive animate verb (which uses different peripheral endings to signify different participants in the action), is there really a basic lemma to use as a dictionary form?

This then begs a larger question about how to add Algonquian entries. There are a few problems we will need to solve before we can really begin adding full verb entries; Since Ojibwe and other Algonquian languages are highly polysynthetic, it is difficult to organize words neatly based on lemmas. There are many possible forms which can be derived from a single root. If a prenoun or other affix is added, where should these words be included in the categorization? Additionally, many words contain morphemes which do not overtly appear in spelling.

E.g., Lenape (nulamhìtao) I believe him Analysis:(n- first person subject, -wëlamhìtaw- 'believe (an.)', -a- direct (transitivizer), -w third person object)

I know that Diné bizaad (Navajo)verb entries are organized by root. The root forms are not actually used, however they are probably the best way to organize such languages. Look at this example, -TŁIZH. Maybe we should consider this kind of solution to the long-term issue of organizing verb entries. That way all entries containing a given root could be linked to a single page, regardless of grammatical category, affixes etc...

Some dictionaries use specific forms since the root can easily be derived from them. For example, John O'meara's Delaware-English dictionary uses the third person singular entry. Some languages on wiktionary do likewise. Latin, for example, uses the first person singular present indicative active.Hk5183 (talk) 20:42, 5 June 2020 (UTC)

Thanks for this reply. I'm going to go ahead and reorganize the few verbs that are already there in accordance with the categorization i suggested above.

And because i don't know other Algonguian languages, so may miss the language family nuances, a few thoughts on the points you make:

• The citation conventions in Ojibwe seem pretty well established to me. For intransitive verbs (VII and VAI), the independent order 3rd person singular form, and for transitive verbs (VTI and VTA), the imperative order for the 2nd person singular with a 3rd person object. All the texts i use take this approach, but especially John Nichols/Maude Kegg and Rand Valentine (of the Ojibwe People's Dictionary).
• Once we have a few more verbs in, i think i would ask someone to write a template for an impersonal VII (say, gimiwan). The conjugation is the simplest for those verbs, i just don't know how to write a conjugation bot.
• Presumably, the other non-lemma forms would show up in the conjugations and can be added as non-lemma forms, eventually...
• I'm not sure whether we should add as non-lemma forms words that are derived by relatively stable rules (such as preverbs).
• For various spellings of moprhemes (eg waaboo vs aaboo), i'm not sure. I'm tempted to say go with the underlying form (usually with a w) even though they appear less often in the final spelling, but only because that is what Valentine does in the OPD.
• I have other questions/suggestions (eg initials, medials and finals), but i wonder if one of the more experienced editors could suggest a more appropriate place to start those conversations).

SteveGat (talk) 13:07, 8 June 2020 (UTC)

Experimental inline images template

For your consideration, {{inline images}}, and an example of its use on the this revision of the page "corona". I find this far superior to having images floating around randomly in the page text, when they often refer to specific senses. These images could be hidden in the same way that quotations are hidden by default. DTLHS (talk) 18:22, 28 May 2020 (UTC)

I can appreciate the benefits of structurally linking an image to its sense. Visually though it looks quite disruptive to me, having biggish pictures interrupting vertically between text definitions. Collapsibility could solve this, as you state, but then I suppose we would have no sense-images showing on an entry by default, which seems like a loss. Equinox ◑ 18:39, 28 May 2020 (UTC)

An issue is captions. Alt text is possible, but then it's impossible to use links. I don't know how to add captions to inline images, but maybe there's some CSS trick. DTLHS (talk) 20:16, 28 May 2020 (UTC)

Pending vote

Wiktionary:Votes/2020-04/Use_of_"pronunciation_spelling"_and_"eye_dialect"_labels has had its review period without comment. I sense that there is not much interest in this issue, but I wonder whether an Administrator could now quickly check that the vote is in order and all makes sense, and put it live if so, or contact me if not. Mihia (talk) 21:11, 28 May 2020 (UTC)

I wasn't even aware of this vote. For some reason it isn't on Wiktionary:Votes/Active transcluded onto our watchlists. —Mahāgaja · talk 19:58, 29 May 2020 (UTC)

I wasn't aware of it either. Added to Wiktionary:Votes/Active. I updated the timestamps too. The vote will run from June 5 to July 4. PUC – 20:11, 29 May 2020 (UTC)

@Mihia: Or it can start straight away, if you want? PUC – 20:14, 29 May 2020 (UTC)

@PUC: Thanks, June 5 to July 4 is fine by me. There is no hurry for this, just so long as at least someone has given a quick check that it does make sense. In the end I started mixing up my "standard" and "nonstandard", and getting confused myself. Mihia (talk) 19:18, 2 June 2020 (UTC)

short for, ellipsis of, or write out the definition

Many definitions are shorter forms of phrases with the last words eliminated. To pick the last example I looked at, transistor is a short form of transistor radio. There are at least three ways such definitions are written:

• Short for transistor radio.
• Ellipsis of transistor radio.
• A transistor radio

Is one of these styles preferred and if not should one be declared to be the preferred form? Does the more specific {{ellipsis of}} take precedence over {{short for}} when both are correct? Vox Sciurorum (talk) 13:57, 31 May 2020 (UTC)

I can only speak for myself: my preference is usually "short for X" because it makes it clear that this is a sort of "shortcut" to another main entry where you can find all the exciting stuff like translations. I don't like "ellipsis" because that is a very technical word that many dictionary users may not know. I don't object to the spell-it-out form ("a transistor radio") but I think that's more likely to make people add content like translations, derived terms, etc. because it's not so clear that the other form is what we chose as the "main" one. (How to choose the "main" one is another issue.) Equinox ◑ 14:38, 31 May 2020 (UTC)

Relatedly (?) one thing I have often seen that I dislike is the use of synonym of X where the synonym has some shade of usage that the base does not. For example (in my opinion) it's bad to define feminazi as synonym of feminist even if you put the derogatory gloss on it because "synonym" suggests a technical equivalence, whereas feminazi is very speifically a derogatory term. Just "(derogatory) a feminist" works better. Equinox ◑ 14:42, 31 May 2020 (UTC)

Should we even be using {{synonym of}}? I'd rather spell out the senses at each entry if they're not alternative forms. (I treat female equivalents differently, but that's another discussion.) Still it's better than # Variant of [[. *shudder*

On topic: I agree with Equinox and say "short for" since it's easily understood. Ultimateria (talk) 02:13, 2 June 2020 (UTC)

I think it's OK to use {{synonym of}} where appropriate, but editors have to be aware that if the term has more than one sense, the relevant sense(s) need to be specified using the |t= parameter. — SGconlaw (talk) 12:14, 2 June 2020 (UTC)

I'm not a big fan of the loose way that we seem to use "synonym" in definitions or lists. To me, if X is a "synonym" of Y, it should properly mean that any time I use the word X (in a particular stated sense) I could use the word Y instead with exactly the same meaning, which is not the case with many of our "synonyms". Mihia (talk) 21:54, 3 June 2020 (UTC)

I don't have a problem with "ellipsis", and I prefer it when it's the right term. PUC – 11:02, 2 June 2020 (UTC)

Same here, but it's true it's a bit technical. An option would be to keep the ellipsis template (which will do the categorization) and change the output to match {{short for}}. OTOH we already link to the Appendix, but it's another click away. Ideally the link preview / pop-up should handle this, but it's empty at the moment. – Jberkel 11:14, 2 June 2020 (UTC)

I didn't even know {{ellipsis of}} existed. My preference would be for the use of less technical language unless this would be inaccurate. I notice that Lexico now uses labels like "mass noun" (rather than "uncountable") and "with object" (rather than "transitive"), though the main OED retains the more technical terms. I'm not saying we should definitely go down that route, but it may be worth having a discussion about it at some stage. — SGconlaw (talk) 12:14, 2 June 2020 (UTC)

I learned transitive and intransitive in or before high school. I never met "ellipsis of" until a few weeks ago. Vox Sciurorum (talk) 23:08, 2 June 2020 (UTC)

• I'll add to the anecdotes for "transitive" not being all that technical. Meanwhile, I don't think I'd ever heard of "mass noun" until here at Wiktionary and it required a good bit of context for me to understand it, and "ellipsis of" is similarly opaque. ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:15, 2 June 2020 (UTC)

I prefer "short for", "transitive"/"intransitive" and "countable"/"uncountable". Mihia (talk) 10:48, 3 June 2020 (UTC)

I agree. — SGconlaw (talk) 13:02, 3 June 2020 (UTC)