Wikidata:Property proposal/stroke count

stroke count

Originally proposed at Wikidata:Property proposal/Lexemes

   Done: stroke count (P5205) (Talk and documentation)

Description	number of stroke of CJKV character (Chinese, Japanese, Korean, Vietnamese)
Represents	stroke (Q1689024)
Data type	Quantity
Domain	CJKV character (Q item [instance of CJKV (Q53764732)] or Lexeme?)
Allowed values	positive integer
Example	一 (Q4025820)/"一" (Lexeme?) → 1 no item/"口" (Lexeme?) → 3 叶 (Q3594950)/"叶" (Lexeme?) → 5 雪 (Q3595029)/"雪" (Lexeme?) → 11
Source	"sn" parameter of wiktionary:Template:Han_char

Motivation

I'm not sure Chinese characters should be Q item or Lexeme, but this property is important to describe Chinese characters. Okkn (talk) 15:49, 19 May 2018 (UTC)

Discussion

•  Support but take a look at Wikidata:Property proposal/number of strokes. NMaia (talk) 10:55, 21 May 2018 (UTC)
  • I didn’t notice this former proposal. I think storoke count is one of the most basic and important characteristics of Chinese characters, so we should use an independent property rather than has part(s) of the class (P2670). --Okkn (talk) 12:14, 21 May 2018 (UTC)

 Comment I this the name is too ambiguous. It has nothing to do with two-stroke engine (Q152323) or four-stroke engine (Q191801). /ℇsquilo 08:51, 22 May 2018 (UTC)

• That’s a good point. But is the term “stroke count” commonly used in the context of an internal engine? I searched “stroke count” on Wikipedia (https://en.wikipedia.org/w/index.php?search=stroke+count&title=Special:Search) and found that “stroke count” is commonly used in the context of Chinese characters or kanji. --Okkn (talk) 09:26, 22 May 2018 (UTC)
•  Support Name is fine. I think this should be a property on lexemes, I assume on their forms, though I'm not sure how that works for Japanese & Chinese characters. ArthurPSmith (talk) 17:53, 22 May 2018 (UTC)
  • @ArthurPSmith: AFAIK, Asian language doesn't really have inflections, you can't conjuguate a character (it would be like to add a stroke to A to say two A). Cdlt, VIGNERON (talk) 09:38, 23 May 2018 (UTC)
•  Support --Was a bee (talk) 21:04, 22 May 2018 (UTC)
•  Support name is ok to me. « Chinese characters » should be used in a broad sense here for all CJKV characters (Chinese, Japanese, Korean, Vietnamese), I slightly changed the description to reflect that. I've also added the example of 口 (strangely no Q-item) made of 3 strokes, which is simple but can be counter-intuitive for peoples outside Asia. Finally, there should be a constraint to limit this property to "one character lexeme only". Cdlt, VIGNERON (talk) 09:38, 23 May 2018 (UTC)

  @VIGNERON: I have no definitive answer, but is a CJKV character really "one character lexeme"? Alphabetic characters are also lexemes? If so, what is the lexical category for these lexemes? It is true that Chinese characters themselves have sense, but I suppose Chinese characters should be distinguished from single character words. --Okkn (talk) 11:42, 24 May 2018 (UTC)

  @Okkn: oh sorry, I didn't explain my idea right, a CJKV character is not "one character lexeme" in my mind (more exactly they can be but it's an other and different matter) and yes, absolutely, « Chinese characters should be distinguished from single character words ». My point was just to expand this property and not to limit it to Chinese. Cdlt, VIGNERON (talk) 11:51, 24 May 2018 (UTC)

•  Comment the proposal indicate a source but it is a Wiktionary ;; wiktionarians loudly said that they don't want data imported from Wiktionary to Wikidata. So, is there an other source available and used as a reference? Cdlt, VIGNERON (talk) 11:51, 24 May 2018 (UTC)
  • In regard to Japanese kanji, w:en:List_of_kanji_by_stroke_count is available. --Okkn (talk) 12:07, 24 May 2018 (UTC)
•  Support David (talk) 12:36, 26 May 2018 (UTC)