Talk:でち

RFD discussion: February–August 2016

The following information has failed Wiktionary's deletion process (permalink).

It should not be re-entered without careful consideration.

---

でち

Is there any POS defineable for this entry? This reading isn't used. If the reading were used, you could define an affix for it, employing an "only used in" if it's rare. Categorisation handles the index for readings. Nibiko (talk) 17:40, 13 February 2016 (UTC)

• I think this is a whole lot of bogosity, brought to us by poorly-vetted Unihan entries. I can't find any solid evidence of this purported reading.

Delete as bogus. ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:12, 15 February 2016 (UTC)

• I clicked on a few of the kanji in the list, and the ones I tried go to Japanese entries for that character, where the でち reading is repeated. So, if the reading is actually wrong (or the characters are not used in Japanese at all) then these should all be deleted too, I guess. I'm not clear if this is the case though, or if the deletions at でち are proposed because that reading can only be used in compounds. By the way, the でち readings for a few of the characters that I tried are also listed at http://www.romajidesu.com website, and one of them also at wwwjdic.com, but I'm not saying this is definitive. 109.151.61.197 15:00, 20 February 2016 (UTC)

• Many of the less-visited kanji entries on the Japanese Wiktionary appear to be little more than a reformatting of information from the Unihan database, so inclusion there is not necessarily an indication of validity.

http://www.romajidesu.com clearly states on its About page that most of its data comes from WWWJDIC and KANJIDIC.

In turn, the KANJIDIC readings data appears to be partially sourced from the JIS X 0208 standard (search this page for the text "In April 1996 the readings of all the kanji"). That said, I cannot find the purported dechi kanji in the standard list provided at http://www.asahi-net.or.jp/~AX2S-KMTN/ref/jisx0208.html, leaving me uncertain where these kanji came from. I cannot find them in my monolingual resources to hand, furthering my impression that these are vanishingly rare in Japanese.

Poking around more, I see that KANJIDIC was also based on a second dataset, where “the Japanese ON and KUN readings are mostly from the file of Unicode pronunciations (Pronunciations.text) prepared by the Taligent company.” This suggests that the similarity between the KANJIDIC listings and the Unihan database listings is because both derive from the same dataset. I don't know who this Taligent company is (possibly the Taligent of Apple's history?), but given the difficulties I'm having in verifying that these dechi characters are even used in Japanese, I suspect that they were shooting more for complete CJK character coverage, rather than lexicographically useful information.

Ultimately, though, we must hew to Wiktionary's descriptivist approach, by describing what is in use. And, so far, I cannot find any instance of these characters in use with the purported dechi reading. ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:47, 20 February 2016 (UTC)

There's this thing with Sino-Japanese dictionaries where they recontruct on'yomi, so certain characters that never had a goon or kan'on get their goon or kan'on reconstructed anyway. Kan'on is the more modern reading, so if a rare character were to be used, it usually takes that - for example, searching "泆" "いつ" on the web shows use of this character with a gloss of its kan'on reading, but "泆" "いち" doesn't bring up any use. Unihan takes this a step further and gives heaps of weird readings that don't even show up in Sino-Japanese dictionaries. I've sorted out the on'yomi of all the characters that were listed as having a でち reading, but for the record, sometimes on'yomi are actually used, sometimes on'yomi are reconstructions, and sometimes on'yomi are weirdness from the Unihan database. Nibiko (talk) 08:11, 21 February 2016 (UTC)

Could you comment on whether you think the characters in this list are used at all (with any reading) in Japanese? 109.151.61.197 18:40, 21 February 2016 (UTC)

I think that Eirikr put it best when he said that these are vanishingly rare. 挃 appears to be used in a metaphor (Daijirin/Daijisen). 昳 appears to be used in the name of an hour, 日昳, the eighth hour, representing the time period from 13:00-15:00 (w:ja:十二時辰/w:de:Japanische Zeitrechnung#Tageseinteilung). 澈 appears to be used in some names (Japanese wikipedia). 瓞 appears to be used in a name (kotobank) and in 瓜瓞 (wikimatome). 絰 appears to be used in some obscure terms like 墨絰 (kotobank) and 衰絰 (wikimatome, jigen). 荎 appears to be used in 荎草園, which is the name of a temple (Google). 泆 appears to be used in 淫泆 (Google). The usage of all these is extremely rare and limited at best, and these are just the characters that I could find significant usage for. The characters that are encoded in JIS X 0213 are the same as the ones that I mentioned with the exception of 荎. Nibiko (talk) 10:18, 22 February 2016 (UTC)

I see, thanks. 109.152.149.220 03:19, 28 February 2016 (UTC)

Deleted. bd2412 T 02:07, 10 August 2016 (UTC)