Template talk:mul-script

Purpose and scope

The alphabet display boxes are very nice, but I question the need for this root template. I think we should just have the child templates for displaying the alphabet boxes. First, the name of {{mul-script}} is too vague, viewing the wikitext I have no idea what this template does. Relatedly, we should have the boxes for Armenian, Hiragana, and Katakana but those aren't Translingual (so not mul). What is the purpose of the root template? Why do the script auto-detecting (which isn't scalable)? If you're on ⠷ why not just put {{mul-script/Brai}}? --Bequw → ¢ • τ 20:56, 12 October 2009 (UTC)

This template comprises a considerable number of automatical choices. Please see the entry ć, that contains a {{mul-script}} without any parameters, yet it shows a display box of Latin script, with a section "Variations of letter C" and a section "Letters using acute accent or double acute accent". Hypothetically, we could achieve this effect manually through the introduction of parameters, like {{mul-script|Latn|c|acute}} (We could even include every information through parameters, such as possibly {{mul-script|Latn|section1_title=Variations of letter C|section1_contents=ĆćĈĉČčĊċC̄c̄ÇçḈḉƇƈᴄＣｃ}}, which would be highly undesirable in my opinion). I chose to keep any technical details in the templates, not in the entries, because this system facililitates both entry editing and template editing: Entries are edited more easily because you just add a {{mul-script}} as theoretically every information is fulfilled automatically, and templates are edited more easily because the characters are all together.

I'm planning to make a documentation soon (perhaps after I finish my work on small katakana syllables and eight-dot braille symbols), but I think the idea is simple: mul-script analyses the character and calls a subtemplate related to its script, such as mul-script/Brai or mul-script/Latn. Then, the subtemplate shows a box with the characters in that script.

As for your third question, if you're on ⠷, you might use {{mul-script/Brai}}, but you probably wouldn't use {{mul-script/Brai}} in A, B, 愛 or п, so a manual distinction is not necessary. When a distinction is necessary, you may edit this list.

I consider Armenian, hiragana and katakana as Translingual, therefore code mul. One reason is, the katakana script is used in at least three languages at Wiktionary (Japanese, Miyako and Ainu). --Daniel. 06:40, 13 October 2009 (UTC)

I think the autodetection in the child templates is wonderful (like what {{mul-script/Latn}} does on ć), it's just the root template that I find unpalatable. We will want to have alphabet-box templates for all the alphabets but many of them will not have Translingual entries as many scripts are only used to write a single language. It's confusing to have a smattering of language specific alphabet templates (e.g. {{Tibt-script}} or {{hy-script}}) plus a semi-generic {{mul-script}}. Additionally, it is untenable to try and pack all the alphabet symbols into a generic template (there's a lot at w:List of alphabets). The root template is even more troublesome if community attitudes of what is "Translingual" changes or no consensus is reached on particular cases. This is evident already as you categorized certain alphabets as translingual that I do not. My comment above about Armenian, Katakan and Hiragana was motivated by what appeared to be an emerging consensus at WT:BP#Translingualness of letters. If you disagree with the criteria there, please contribute and discuss. As for the how those criteria apply to these alphabets here:

• Armenian is only used to write the different temporal periods of Armenian. Additionally, we unify them all here on en.wikt. Why would you categorize this as Translingual?
• You're right that Katakana is used to write other languages such as Palauan, so I agree that this can be thought of as Translingual.
• Hiragana is only used for Japanese and some of the Ryukyuan languages. There has been no serious discussion that I can find (only this) about whether to consider these on enwikt as dialects of Japanese or as separate languages. I think the jury is out on this one. Hopefully something we be discussed on WT:AJA.

--Bequw → ¢ • τ 15:41, 14 October 2009 (UTC)

A categorization scheme which considers û as Translingual and ű as non-Translingual would be slightly confusing. Anyway, if the Translingual section is simply representing the fact that a character is used in two or more languages, we don't need such section anyway, just the individual language sections. --Daniel. 04:51, 17 October 2009 (UTC)

You would classify a term as Translingual just because it shares an etymology (and shape) with another term that is Translingual? If you would like to put forward your own criteria, please do. --Bequw → ¢ • τ 01:48, 19 October 2009 (UTC)

From what I see, classifying "a term as Translingual just because it shares an etymology (and shape) with another term that is Translingual" would be, for example, this rough hypothetical Translingual definition of É:

1. The letter E with an acute accent, used in Spanish, Hungarian, Portuguese, French, Danish, Irish, Kashubian, Vietnamese and other languages, usually representing /eː/, /ɛː/ or ɛ/, also used in Chinese pinyin as high-rising tone and in English in words of French origin.

I'd be against it. For consistency, Translingual entries would exist for all phonetic and ideographic symbols, regardless of how many languages use them. These Translingual sections would comprise generic definitions, links to probably wide groups of related entries (for example, Translingual letter ű is evidently related to Translingual letter û, but Hungarian ű is not directly related to û) and comparison between scripts. Language-specific sections would include pronunciation, etymology, own alphabetical order when possible, usage notes and symbol names. --Daniel. 03:09, 19 October 2009 (UTC)

As my statement was only about classifying characters (which L2), it was not dealing with definitions.

You appear to want all characters to have a Translingual entry for purposes or organizing information rather than for any linguistic reason (that's originally why I held your point of view as well). Single-characters have many similar types of information, as you point out, that should be recorded. But there's a better way to organize the information than stretching the meaning of "Translingual" to encompass, for instance, an Ugaritic character, which defeats the meaning of the word an is confusing to users. Some of the information is separate from language concerns; two languages could define and use the colon differently but each is using ":" U+003A. This type of information then is best shown then with {{character info}}. Some of the information is common to the script such as links to other related characters (transliterations are more tricky as they are often specific to source and target languages and therefore better dealt w/ in language sections and appendices). If a character is used in several languages, then a script display box in the Translingual entry will show all the links (and subsequent language sections can show different alphabetical orderings). If a character is used in only one language, that language's entry will have the script display box. Either way though, the links to similar entries is still only display once (in the top entry). So having a Translingual entry just because an entry is a single character doesn't even make sense from the a standpoint of organizing information. (Sorry if this was long but I wanted to describe how I went from your view to mine:) --Bequw → ¢ • τ 21:21, 19 October 2009 (UTC)

I typed that definition as an example for comparison between Translingual sections and other L2 language sections, and how to distribute information between them. There are many characters in Latin script which are used in only one language, therefore from your arguments, these characters would merit entries without a Translingual section. Then, I'd like to know if you have any criteria for the opposite situation: Latin characters such as A, B, C, a, b, á and ô, which appear in two or more languages. Most importantly, why we need Translingual sections for them and what information to include in such sections. You don't need a Translingual section to express the fact that the colon sign is ":" U+003A, because the Unicode box is at section 0 anyway. --Daniel. 09:39, 21 October 2009 (UTC)

Right, you'd list at the Translingual section for ":" the identical meanings (uses) of the colon (e.g. "used to separate numbers in a ratio"). I steered clear of definitions because, as far as I know, there have been no thorough discussion about whether a definition listed in a Translingual section should apply universally to all languages that use the term, or just a subset. And even then, how do you list the languages to which a definition can apply. But these are concerns universal to Translingual entries, not specific to single-characters at all. Additionally some characters such as Ⓐ afford an independent definition. Letter senses of characters are much less interesting. A reasonable definition at the Translingual section of D (which has one since it is used in several languages) could be "the capital letter dee in the Latin alphabet, usually proceeded by C and followed E". Related links would then also be in this section. --Bequw → ¢ • τ 18:56, 21 October 2009 (UTC)

Encoding information

The encoding information (Unicode codepoint) is best left outside the language sections. I've been showing it with the generic {{character info}} or one of the easier shortcut templates. I've done this for all the Braille entries so I'm going to remove the Unicode info from {{mul-script/Brai}}. The same should be done for the Katakana and Hiragana. --Bequw → ¢ • τ 16:28, 15 October 2009 (UTC)

Width

With the width=70% on the tables they were overflowing the RHS elements. I removed the width clause from the Braille entries. I think it looks better, but I'm worried the other tables (Hiragana, Katakana, etc.) won't look as good since the cells aren't fixed width. Ideas?

Your removal of width=70% looks good on Braille display box, but I'm also worried about other scripts (in addition to the non-fixed width, the quantity of characters is also of major concern). I'll undo your edit and place the template below See also, this may fix the current issue. Anyway, you may use {{mul-script|width=50%}}, for example, if necessary on individual pages. --Daniel. 12:46, 19 October 2009 (UTC)

I switched it to be just a bulleted list of characters. It now takes up less spaces, doesn't attract extra attention, doesn't overlap other items, and appears consistent with other elements in the See also section. Once the Latin letter occurrences of this template are moved down to the See also section, the same should be done for those display templates. --Bequw → ¢ • τ 18:59, 21 October 2009 (UTC)

Use in non-Translingual sections

I think the top-line output of this template needs to be rethought for letters used in only one language. For instance, ű is only used in Hungarian, so the top-line display of this template should show the Hungarian alphabet not the basic Latin one (the bottom two lines about "u" and "double accents" variants are fine). As we will want a Hungarian alphabet display template anyways (to show their ordering) is there a way we can just swap out the top display for a different one. Also, as this only affects a few entries, maybe a special template is easier than modifying this one. --Bequw → ¢ • τ 19:07, 21 October 2009 (UTC)

Please add Y-ogonek to the ogonek selection

kwami (talk) 13:44, 27 May 2023 (UTC)