Definition from Wiktionary, the free dictionary
Jump to navigation Jump to search



The following discussion has been moved from Wiktionary:Requests for moves, mergers and splits.

This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.

Why is this entry's pagetitle not CD? The romaji entries use regular Latin letters like C and D... Likewise CM, DVD, JR, SMS should be moved to CM, etc. - -sche (discuss) 22:58, 12 January 2013 (UTC)

What are those strange letters for? Are they part of Japanese somehow? —CodeCat 22:59, 12 January 2013 (UTC)
They are variants created to occupy two terminal columns. Why the hell did Unicode include them? — Ungoliant (Falai) 23:02, 12 January 2013 (UTC)
(e/c) They're fullwidth letters. They display better than regular letters, if embedded into the middle of a Japanese text and viewed by someone with enough font support to display them, but not enough font support to display regular letters nicely. I see zero reason we use them in cases like this. - -sche (discuss) 23:02, 12 January 2013 (UTC)
Also all the other fullwidth Latin-letter entries in Category:Japanese abbreviations, acronyms and initialisms, and any other similar entries anywhere else. - -sche (discuss) 23:02, 12 January 2013 (UTC)

Sorry to jump in. The Japanese use only the full-width letters when they write Latin, because otherwise they would look extremely out of place in vertical writing. Therefore, these should be kept. -- Liliana 23:17, 12 January 2013 (UTC)

The Japanese clearly don't always use these Unicode characters, though; "CDです" gets plenty of Usenet hits. It also gets BGC hits, though it's probably impossible to know if the books themselves were originally typed up with fullwidth Unicode characters, or with Latin letters displayed as wide as the Japanese characters around them. We are not, IMO, bound to carry over the aesthetic practice into our pagetitles. Books in many languages use curly apostrophes, but we use straight ones. - -sche (discuss) 23:28, 12 January 2013 (UTC)
Would someone in Japan come to en.Wiktionary and not be able to type Latin script and instead use full-width characters? How does our search engine handle the search? DCDuring TALK 23:45, 12 January 2013 (UTC)
Does Unicode have any recommendation for or against their use? This seems like a typical case of characters that were included for compatibility, not because they should be used. —CodeCat 23:58, 12 January 2013 (UTC)
Actually, though it depends what fonts you have installed I guess, this test shows that CD written in fullwidth with Jpan as the script actually looks horrible, in fact it looks like a ligature of O and D rather than 'CD'. Almost unreadable, or perhaps actually unreadable. Mglovesfun (talk) 00:55, 13 January 2013 (UTC)
As far as I can tell, the WT search considers CD and CD as different, but similar. CD will get me to the full-width and CD will get me to the regular, but if I mix the two (CD or CD), it won't take me directly to either, but the list of hits following contains both the full-width and regular-character entries. On my Mac, I'm given a choice of 4 keyboards for Japanese (Romaji, Hiragana, Katakana, and Full-width Romaji, as well as a "Kana palate" that has separate tabs for Romaji, Hiragana, and Katakana, with a checkbox under the Romaji tab for using Full-width instead of regular Romaji. From this I would guess that Full-width Latin characters are something one would have to choose specifically, and not something that would be a barrier for most Japanese visitors. The main reason we would want to have this entry is for those who copypaste it into the search box- but even without it, it looks like the search would get them close enough. Chuck Entz (talk) 02:38, 13 January 2013 (UTC)
I'm actually OK with hard redirects from CD to CD; redirects are cheap. Conversely, if we keep the content at CD (which I oppose), we should have 'redirects' (which will need to be soft, i.e. {{alternative form of}}) from CD. - -sche (discuss) 02:55, 13 January 2013 (UTC)
I think these should be kept even though only trolls and Japanese people use them, because they are the standard way to write the Latin script in Japan, and presumably the search problem can be negated by means of {{also}} . —Metaknowledgediscuss/deeds 03:04, 13 January 2013 (UTC)
(also @Liliana): Do you think our romaji entries should be moved to use fullwidth characters? What about our pinyin entries? - -sche (discuss) 03:11, 13 January 2013 (UTC)
No, the romaji and pinyin entries are fine where they are. It's only Japanese terms incorporating Latin letters (like this abbreviation) that need fullwidth letters. -- Liliana 10:18, 13 January 2013 (UTC)
I'm not convinced these Unicode characters are standard. As Mg's sandbox test shows, regular Latin letters have appropriately full widths in good Japanese fonts (while fullwidth Unicode characters are, ironically, illegible), so it's hard to know whether books were typed up using the fullwidth Unicode characters or normal Latin letters. And both "CD" and "CD" are attested on Usenet. I note that ja.Wikt deleted w:ja:CD and uses w:ja:CD (and "80nm" and other regular Latin characters) in its entry on CDs. Even the Japanese Wiktionary itself redirects CD to CD, probably the most damning argument against making CD a lemma. - -sche (discuss) 03:23, 13 January 2013 (UTC)
OK, I concede. Redirect.Μετάknowledgediscuss/deeds 03:26, 13 January 2013 (UTC)
Actually, you are incorrect: in Japanese fonts, regular Latin letters are exactly half width. This means they look completely out of place within Japanese text. -- Liliana 10:18, 13 January 2013 (UTC)
As far as I know, Japanese keyboards input the fullwidth Latin characters by default (when set to input Latin characters, obviously), with the option to switch to regular width. --WikiTiki89 03:55, 13 January 2013 (UTC)
Halfwidth Roman letters are now standard in Japanese Wikipedia but Japanese printed books use fullwidth. Japanese news articles use both fullwidth and halfwidth. Forums usually use fullwidth. Japanese keyboard, when switching to numbers produce fullwidth numbers: 0123456789. Japanese editors here, including User:TAKASUGI Shinji are against fullwidth as names of articles but happy to use them as redirects or headers. As the creator of the above entries, I'm OK for them to be moved/merged with halfwidth but I'd prefer the headers to display fullwidth (e.g. head=CD). "CD" combines much better with the Japanese script than "CD". Halfwidth characters look really out of place with kanji but it's becoming common on the web. --Anatoli (обсудить/вклад) 12:51, 13 January 2013 (UTC)
See also this PDF: File:Japanese writing.pdf. The pages shows a Japanese font with halfwidth letters, a standard proportional font, and a Japanese font with fullwidth letters, respectively. The first page shows horizontal, the second page vertical orientation. See what difference it makes? -- Liliana 13:23, 13 January 2013 (UTC)
Yes, vertical writing is the main reason why the fullwidth characters will continue to be used. Even if halfwidth characters were positioned properly, they never align vertically with the Japanese kana and kanji. --Anatoli (обсудить/вклад) 22:11, 13 January 2013 (UTC)
  • Redirect all of these to their normal Latin character equivalents. The distinction between the fullwidth characters and their normal equivalents may be important for typesetting Japanese, but it's not important to human beings reading Wiktionary. By all means keep the redirects so people can find them (and won't re-create them), but don't keep the info on separate pages as if the fullwidth characters comprised a separate alphabet in any meaningful sense. —Angr 16:00, 13 January 2013 (UTC)
{{form of|fullwidth form}} would work for me. Mglovesfun (talk) 00:03, 14 January 2013 (UTC)
I'd vote for hard redirects. --WikiTiki89 00:13, 14 January 2013 (UTC)
I agree, a hard redirect should be enough. The distinction between the two types of letters is really a matter of font alignment and not so much of the appearance of the letters. So C really is the same character as C, it just has a different Unicode representation and is treated a little differently in fonts and software. But it really is a computer-only distinction. On the other hand, the same could be said for Cyrillic А vs Latin A vs Greek Α... —CodeCat 00:35, 14 January 2013 (UTC)
Cyrillic is used in more than one language, so you have the risk of language A having a lemma where language B has only a redirect. Full-width characters are specific to Japanese, so can be redirected without worry. Of course, someone could always use them in artificial ways like User:Doremítzwr used specialized IPA characters as superscripts, but I wouldn't lose any sleep over precluding that kind of entries. Chuck Entz (talk) 02:15, 14 January 2013 (UTC)
Hard redirect all such IMO.​—msh210 (talk) 04:05, 18 January 2013 (UTC)
Hard redirect per Angr. — Ungoliant (Falai) 06:03, 19 January 2013 (UTC)
I redirected all the entries I could find, but I may have missed some because the computer I'm on displays all fullwidth characters as boxes. - -sche (discuss) 19:39, 21 February 2013 (UTC)