Wikipedia talk:Manual of Style/Japan-related articles

WP:VG/GL#Non-English games

Archives of this discussion can be found at /VGGL and /VGGL2

Proposal regarding the format of Japanese personal names

According to WP:JTITLE, Japanese born prior to 1868 should be given family name first while those born later should be given family name last. In general, academic usage is to put family name first, while journalists tend to put family name last. As near as I can tell, the idea of using a specific year as a cutoff is unique to Wikipedia. I have created the following options to allow editors to express preference. For example, "BACD" indicates that "B" is the editor's first preference, "A" second, and so forth. Fernando Danger (talk) 02:50, 28 May 2015 (UTC)

Options

A. Default to surname first, with exceptions as required by WP:COMMONNAME.

B. Default to given name first, with exceptions as required by COMMONNAME.

C. Follow general common name principles with no default.

D. Retain current language in guideline.

Here is what some authorities on style have to say:

• "In Japanese usage, the family name precedes the given name. Japanese names are sometimes westernized, however, by authors writing in English or persons of Japanese origin living in the West. [Examples:] Tajima Yumiko; Tajima; Yoshida Shigeru; Yoshida; but Noriaki Kurosawa; Kurosawa." Chicago Manual of Style, "Japanese names" (§8.16). N.B. "Tajima Yumiko" is a "Jane Doe" name, not a real person. This example suggests that when in doubt Japanese name order should be used.
• "Use them in the customary Western fashion, with the surname after the given name." New York Times Style Guide, "Japanese names."
• "Although the Japanese put the family name first in their own language (Koizumi Junichiro), they generally reverse the order in western contexts. So Junichiro Koizumi, Heizo Takenaka, Shintaro Ishihara and so on." The Economist, "Japanese names"

CMOS was produced by the University of Chicago Press. It is recommended in WP:MOS in two separate places. The Economist and the New York Times are both news organizations.

I have produced ngrams for three hopefully representative post-war figures: Kishi Nobusuke (ngram and Britannica), Nakasone Yasuhiro[1][2], and Nakagami Kenji[3][4]. When these people were in the news, the westernized form of their names dominated. Later, the non-westernized form became prevalent.

The original rationale for the current guideline, which was adopted in 2006, was that Japanese schools taught students to put their given names first in English. However, this is no longer the case. Mass ping time: @Curly Turkey, Imaginatorium, Cckerberos, Margin1522, Msmarmalade, Giraffedata, and Nihonjoe: @Sturmgewehr88, Elinruby, and WhisperToMe: @Closeapple, BDD, and SMcCandlish:.

Comments

Please put comments and votes below

• I support A; I'll put up with B, with exceptions allowed for standardizing names within an individual article; and I oppose C and D.

  There are enough battles over the ordering of Japanese names that removing a default (C) would simply feed these battlegrounds. (D)'s lack of flexibility is problematic. With (A), we would still have redirects, and piping would remain an option in articles where Western name order were deemed more appropriate. Japanese name ordering is a problem that will not go away, and none of the solutions really solve the issue; (A) is the simplest.

  It should be emphasized that COMMONNAME exceptions should be truly exceptional—COMMONNAME should be narrowly defined as the name that is overwhelmingly used in English-language sources (Yoko Ono), and not merely the one that gets more Google hits (so not Shinzo Abe). Curly Turkey ¡gobble! 04:30, 28 May 2015 (UTC)

  Also, I'd ditch the Japan Probe link—it's not an RS, and half of it is a quote from the Big Daikon cesspool message board. I can confirm that FAMILY–GIVEN is indeed what is taught in junior high and high school textbooks these days (I just checked through five), but that source is just embarrassing. Also, the attempted ping didn't ping me, so I suspect it didn't reach the others, either. Curly Turkey ¡gobble! 04:32, 28 May 2015 (UTC)

  Following up on the textbook thing: I asked a couple of (Japanese) English teachers. The first was surprised to find out that all the textbooks use FAMILY–GIVEN, and said he always teaches GIVEN–FAMILY. The second said he knew about it, and that it was related to another initiative to give historical figures their actual names, rather than the names they've traditionally been known as in Japanese: for example, textbooks now give Ferdinand Magellan's name as マガリャンイス, rather than the traditional マゼラン. He says he just ignores all this and continues to teach GIVEN–FAMILY, and says all the teachers in that school's English department do the same. I'm sticking with my !vote above, though, and remain strongly opposed to a default of COMMONNAME, which is just trouble waiting to happen. Curly Turkey ¡gobble! 07:47, 28 May 2015 (UTC)

  An ngram might suggest putting the surname first, while a news archive search would suggest given name first. Since COMMONNAME doesn't provide any guidance on how to resolve such a contradiction, the Japan MOS needs to. Fernando Danger (talk) 08:11, 28 May 2015 (UTC)

• Strong support for B. We should follow English practice, not Japanese, and it is overwhelmingly "given name first". The same principle applies to Hungarian names, and follows the principle of least astonishment. No such user (talk) 07:17, 28 May 2015 (UTC)
• Support ACB. Anything is better than using an arbitrary and overly precise year as a cutoff. I favor strict Chicago Manual style to put Wiki in line with Britannica and other published encyclopedias. We should show the reader how to do it right, not follow others' mistakes, however "common" they might be. If it's a question of academic vs. media style, WP:MOS suggests that we should follow academic style. It recommends CMOS as well as several other academic style books. Neither the AP or NYT style books, the two main media-oriented guides, are even mentioned. Fernando Danger (talk) 14:54, 28 May 2015 (UTC)
• BCAD. And the year wasn't arbitrary, so stop with the attacks on that already. We already know your opinion about it. It was chosen because that's when Japan Westernized, and that cutoff was also very common in academic works. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 15:48, 28 May 2015 (UTC)
• D for the reasons given in the recent discussion, which would be tedious to repeat. – Margin1522 (talk) 04:35, 29 May 2015 (UTC)|
• Strong support for A, and agree that any that COMMONNAME exceptions should be truly exceptional; strongly oppose B. Do we really want "Yoritomo Minamoto" or "Shoin Yoshida", "Shikibu Murasaki", etc? --MChew (talk) 14:56, 29 May 2015 (UTC)
• Strong support for A – This is the only proper way to list a Japanese person's name, barring a more commonly used western-style form. RGloucester — ☎ 17:45, 29 May 2015 (UTC)
• Support A – This is not an easy decision to make. It is true that current journalistic discourse mostly gives family name last, while academic discourse now mostly gives family name first. But even then, film journalists are more frequently using surname first: [5], [6], [7], etc. Personal usage in Japan is also in transition (in the film world, I see artists, festivals, and distributors increasingly using Japanese name order (e.g., [8], [9], etc.)). But that's part of the problem: people in some worlds (academia, the arts) are doing it one way, while those in other worlds (some journalism, politics) are doing it another way. I thus am skeptical of any argument that claims one practice is a majority because it can be like comparing apples and oranges. This can be a problem with COMMONNAME because it offers little on how to adjudicate these differences. That's why I cannot support C, since it will only lead to endless and pointless arguments. If COMMONNAME can help us make a decision, it is in its stipulation that we look at what is common in RS, and given that the most reliable of sources are academic ones, it makes sense to go with A. I cannot support D because I never thought the 1868 birth year cutoff to be logical. Coming out of academia, I was never taught that and the textbooks I used, like John Whitney Hall's history, don't change name order when you get into the modern era. (If you have to insist on a change due to "modern" usage, then do it based on years the person was active, not on birth year.) If people think it's too soon to go to A, I can grudgingly accept B as a possibility, but only if, as with A, we settle on rules for how to adjudicate COMMONNAME exceptions. Michitaro (talk) 00:48, 1 June 2015 (UTC)
• ACBD – Like Michitaro, I'm inclined to support what I've observed to be academic usage over what is more common in journalistic sources (and academic sources only dealing tangentially with Japan). I think the near-universal use of macrons for long vowels is another reason to support the use of native Japanese order. --Cckerberos (talk) 11:14, 2 June 2015 (UTC)
• Strong A, then CB. I will focus on countering No such user's interesting mention of the principle of least astonishment, as Fernando Danger covered all other points I wished to make. I agree it should be kept in mind, but I don't think it applies here.

Confusion on Japanese name order exists in En.Wikipedia articles already. That's why Template:Japanese name and Template:Okinawan name are available. Readers run into both orders of Japanese names here and elsewhere, so it's important to raise awareness of this situation, precisely to prevent the reader's confusion here and elsewhere, now and in the future—which the templates do quickly and with a link to more info.

But more than that, the benefit of learning in this case outweighs the small chance of astonishment impeding the reader. Surnames can look like given names and vice versa in any language with both. For example, I was classmates in 5th grade with Walker Riley and Railey Walker. However, the astonishment doesn't get in the way of adjusting. Most importantly, en.wiki can educate readers on the correct Japanese order of the name as an aside, which only helps when the reader runs into multiple versions of Japanese names elsewhere. They'll be aware of the possibilities and less likely to assume a certain way is correct.

(If we had an inline version to explain either order, similar to the ^? of Template:Nihongo, it would help in articles that merely mention a Japanese person who is not the center of the article. Maybe name it Template:Name order so it can be used with many languages, with text such as, "This family name is first/last." Just a brainstorm, because I've not seen one like that. I can't create this due to real-life limitations, but I might be able to give feedback if someone else does.) Got here via Signpost, --Geekdiva (talk) 10:39, 13 June 2015 (UTC)

I love the ^? suggestion. I hope it gets created with a human-readable template name. Perhaps two templates: {{Surname first}} and {{Surname last}}? Curly Turkey ¡gobble! 11:33, 13 June 2015 (UTC)

Yeah, that would be helpful.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:52, 25 June 2015 (UTC)

• Corrected approach to C, which amounts to a strong A, then D as a last no-consensus resort, but forget B. Prefer what modern high-quality not news style sources do, unless common usage dwarfs that, which is often the case for pop-culture figures, many of whom intentionally and preferentially have their given names first in Western media. This is not normal as the default for notable Japanese people who are not movie/tv/music stars. For businesspeople and academics, family name first is generally preferred. But even the fairly general given-name-first-in-the-West order favored by many Japanese pop culture figures is hardly universal. Utada Hikaru uses her native Japanese family-first order (see [official English-language website, and covers of all albums that use her full name), despite having grown up half in Japan and half in California; our present article title for her, Hikaru Utada is flat-out wrong; it needs to be WP:RMed, and please ping me if you list it. For option C, particular care must be (and to date has not been) taken to exclude all Western sources with an editorial policy of always using given-name-first out of jingoistic habit, against the practices/wishes of the subjects. That means most US magazines and newspapers. Utada Hikaru is actually a good litmus test - if they get her name backwards, they are not a reliable Western source for COMMONNAME purposes with regard to any Japanese people. As for options D and B, the semi-arbitrary date [yes, we know why it was chosen, but not everyone finds that compelling] in the current wording is kind of pointless, but it's preferable to always putting given name first just because lowest-common-denominator publishers like American newspapers do it that way. It is definitely the case that both orders are already being used across en.wp, so we need to always clarify which order is being used.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:52, 25 June 2015 (UTC)
  PS: C cannot really be avoided, since it's a policy. The trick is to correct our approach to it to stop engaging in utterly shameless WP:BIAS. We'll default back to D if this discussion can't come to a consensus, but that translates into a continuation of the incorrect approach to C, with a result that amounts to B.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:58, 25 June 2015 (UTC)
  PPS: Old sources are not reliable for this and amount to primary sources, especially if they date to the mid-20th century, a period of extreme bigotry toward the Japanese, including intentional disrespect toward cultural differences. It's pretty shameful that we're using Western name order for all the key figures in World War II Japan, a clear case of "history is written by the victors".  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:02, 25 June 2015 (UTC)
• While I don't want to entirely contradict your last point, I do think it would raise eyebrows with Japanese people. I think you'll find vanishingly few Japanese people would consider GIVEN-FAMILY to be a result of wartime bigotry (especially since it was established generations before the war). Curly Turkey ¡gobble! 11:12, 25 June 2015 (UTC)
  • That's not the argument I'm making. Western writers were already doing that, over native Japanese objections, yes. But ever since WWII (when Western defiance against Japanese cultural norms peaked), they've continued to do so despite our treatment of other cultures becoming notably more respectful. We even got Mao Tse Tung/Mao Zedong's, and Ho Chi Minh's name order right, despite Western distrust of and hatred toward Asian Communism. But oh no, never the Japanese. Not even when, as with Utada Hikaru, every single indication we have from east of our own newspapers is family-first order, including in Latin script. It's patent prejudice, even if it's sublimated.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:28, 25 June 2015 (UTC)
    • Western writers were already doing that, over native Japanese objections, yes.: I'd like to see a citation on that. I've lived in Japan for seventeen years and have a hard time believing it. As I've noted above, Japanese English teachers continue to teach GIVEN-FAMILY even nearly a decade after all the textbooks were changed to FAMILY-GIVEN. Curly Turkey ¡gobble! 11:45, 25 June 2015 (UTC)

Assessing consensus

The 30 day discussion period has past. Here is how I count it:

Family name first

1. (A) is the simplest. It should be emphasized that COMMONNAME exceptions should be truly exceptional Curly Turkey
2. I favor strict Chicago Manual style to put Wiki in line with Britannica and other published encyclopedias. We should show the reader how to do it right.Fernando Danger
3. Strong support for A, and agree that any that COMMONNAME exceptions should be truly exceptional. MChew
4. Strong support for A – This is the only proper way to list a Japanese person's name, barring a more commonly used western-style form. RGloucester
5. If COMMONNAME can help us make a decision, it is in its stipulation that we look at what is common in RS, and given that the most reliable of sources are academic ones, it makes sense to go with A. I cannot support D because I never thought the 1868 birth year cutoff to be logical. Michitaro
6. I'm inclined to support what I've observed to be academic usage over what is more common in journalistic sources. Cckerberos
7. The benefit of learning in this case outweighs the small chance of astonishment impeding the reader. Geekdiva
8. Prefer what modern high-quality not news style sources do, unless common usage dwarfs that, which is often the case for pop-culture figures. SMcCandlish ☺

   I must clarify (and I said this above) that we should prefer what the subject prefers when this can be determined from published material, even in the case of pop-culture figures (e.g. Utada Hikaru prefers that family-first order, even in Latin script, when not using the stagename "Utada"[10]). This point is more central to my view on the matter than my quoted point above about pop-culture figures, though I think my "[when] common usage dwarfs" caveat equals Curly Turkey's and MChew's "truly exceptional" caveat.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  16:56, 13 July 2015 (UTC)

Given name first

1. Strong support for B. We should follow English practice, not Japanese, and it is overwhelmingly "given name first". The same principle applies to Hungarian names, and follows the principle of least astonishment. No such user
2. D for the reasons given in the recent discussion, which would be tedious to repeat. – Margin1522
3. BCAD. [The year] was chosen because that's when Japan Westernized, and that cutoff was also very common in academic works. 日本穣^? · 投稿 · Talk to Nihonjoe Fernando Safety (talk) 13:15, 13 July 2015 (UTC)

Ruby RfC June 2015

The following text as shown in this revision of Wikipedia:Manual of Style/Japan-related articles#Ruby (minus the boilerplate) has become dated.

Do not use the <ruby> tag to further annotate the kanji, as most browsers cannot display it properly.

Therefore I am requesting that proposals to replace it be submitted and evaluated. Each proposal should have the user name in the header (such as produced by ===Proposal by ~~~===) with nested text and comments subsections placed just above the general comments header. – Allen4names (contributions) 01:27, 23 June 2015 (UTC)

Background

This RfC follows on previous extensive discussion at Wikipedia talk:WikiProject Japan#Revisiting the issue of ruby character usage on Wikipedia, itself a follow-up to ten previous discussions of the idea.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:21, 25 June 2015 (UTC)

Proposal by – Allen4names (contributions)

Last modified 21:34, 29 June 2015 (UTC)

The included notes may or may not be included in the final text.

Text

Ruby (furigana) shall not be used to further annotate Japanese in text.

Notes

Comments (Allen4names)

• Per comments made by User:Curly Turkey below and concerns I have about old browsers I have made some minor changes to my proposal. – Allen4names (contributions) 17:34, 23 June 2015 (UTC)
• I have struck item 2.3 with some reluctance. I hope this can be re-added or a substitute found (Nihon-shiki perhaps) but until then I do not expect it to remain in the final draft. I have chosen a technical approach to limiting some of the use of ruby (list item 2 in the text section above) to the lead and infoboxes but this is by no means foolproof. I would like input as to what terms should be used ie. "should" or "must" and how the furi parameter should be set "inline" or "none" when the Nihongo template is used outside the lead. – Allen4names (contributions) 20:13, 25 June 2015 (UTC)
• Clean up of my initial proposal (preferred) and add an alternative (fallback). – Allen4names (contributions) 12:39, 26 June 2015 (UTC)
• Abandoning the original text. – Allen4names (contributions) 17:11, 29 June 2015 (UTC)
• Clean up and clarify. Only the substitution of "should not" for "shall not" is contemplated now. – Allen4names (contributions) 21:34, 29 June 2015 (UTC)

General comments

• The technical bits of the proposal are way above my head, but I strongly support introducing the use of ruby characters (furigana) in the manner prescribed by it. RGloucester — ☎ 04:46, 23 June 2015 (UTC)
• There should be something to make it clear that there ruby should never be used in place of a romanized gloss. The text must remain readable to those who don't read Japanese. Curly Turkey ¡gobble! 05:06, 23 June 2015 (UTC)

  That appears to have been addressed. See Wikipedia:Manual of Style/Japan-related articles#Japanese terms. – Allen4names (contributions) 05:35, 23 June 2015 (UTC)

  Assuming people don't try to interpret the above text as overriding that instruction (as it would be "redundant")—we should proactively discourage such disputes. Curly Turkey ¡gobble! 05:45, 23 June 2015 (UTC)

  In other words, I really want to support this, but I want to see explicit wording that prohibits the use of furigana in place of romanization. Curly Turkey ¡gobble! 05:12, 24 June 2015 (UTC)

  At bare minimum. I want to see it limited to a specific class of cases where it's thought to be genuinely necessary.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:26, 25 June 2015 (UTC)

• I too find it hard to follow exactly what is being proposed. But I don’t support anything that would encourage the use of ruby. Whenever Japanese appears in the body of an article it should be alongside the Romanisation, whether immediately and inline such as using {{nihongo}} or in a table or infobox Adding furigana is therefore redundant. For most readers it is meaningless clutter; even those who can read it can just read the Romanisation, probably more easily. It can get in the way of e.g. copying text. The one benefit is it helps readers determine the reading of a particular character. Or at least it does in theory: I notice that the examples given all fail to associate the kana with particular characters. But even if used properly few people will use this; wp is not a Japanese language dictionary. Anyone interested in how a particular character is read can use a proper dictionary, such as wiktionary.--JohnBlackburne^words_deeds 18:25, 23 June 2015 (UTC)
  • I should note that the above comes with obvious exceptions: articles discussing the use of ruby, such as Furigana, should obviously use it. In other cases where the ruby is relevant or notable, or just needs explaining such as when describing an image, then it could be given. But otherwise I do not see the need and see it as unnecessary.--JohnBlackburne^words_deeds 18:31, 23 June 2015 (UTC)
• For the oft-cited case of A Certain Scientific Railgun, the furigana is part of the title (とある科学の超電磁砲【レールガン】) and the fallback of putting "レールガン" in brackets after 超電磁砲 is not optimal since the reader doesn't know where レールガン begins if it's in brackets and not above. I would therefore recommend that if furigana is used as part of the title in the original work, then ruby should be used to reflect the official title. I understand this is a rare case, but we should examine these in a case-by-case basis and not ban the use of ruby in a broad stroke. On the other hand, ruby should not be overused since they don't add much to the English reader. _dk (talk) 19:45, 23 June 2015 (UTC)
  • I would support that usage. I also support removing the prohibition as pointless now since pretty much all browsers from the last few years support ruby. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 04:05, 24 June 2015 (UTC)
• I would be OK with lifting the prohibition, as long as it is clear that ruby will rarely be useful. The railgun title might be one exception, but in general 1) blocks of Japanese text in the English Wikipedia are rare, much less blocks with ruby in the original source 2) We don't need to flag unusual readings for our readers. They may be unusual to Japanese readers, but the vast majority of our readers are reading the romanization and they don't care if a reading is unusual. 3) Similarly, they don't care about possible ambiguity in romanized kana because they're not reading the kana, they're reading the romanization. 4) It would be nice if Japanese sources regularly provided furigana for personal and place names, but in practice they don't, so why should we?
  

What I'm afraid will happen if we relax this too much is a proliferation of kanji character names + ruby in video game and manga articles. We shouldn't be providing kanji and ruby just because they look cool. Also if someone wants to use ruby, it is that person's responsibility to set the line height to a value large enough to display them without messing up the page layout, and I doubt that editors would go to that much trouble, even if they could. Or even if we want them to. Are we going to have a different default line height for manga articles or other articles that might contain ruby? – Margin1522 (talk) 09:44, 24 June 2015 (UTC)

It would be nice if Japanese sources regularly provided furigana for personal and place names, but in practice they don't, so why should we?: because we don't want to be stuck in the same Web 1.0 purgatory that the majority of Japanese websites are? (Oops—did I say that out loud?) Curly Turkey ¡gobble! 11:24, 24 June 2015 (UTC)

WP:GREATWRONGS/WP:SOAPBOX/WP:ADVOCACY. It is not WP's job (much less en.wp's) to out-Japanese the Japanese. This is not an HTML technology demo site.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:31, 25 June 2015 (UTC)

(That was supposed to be a joke at the expense of japanese websites for those of us who get scarred by them daily) Curly Turkey ¡gobble! 10:48, 25 June 2015 (UTC)

Noted, but there's an undercurrent to previous discussions that leans in this direction. "The technology is finally here to do it right, so we certainly must!", absent any usability rationale for going on such a linguistic campaign. I call this the the "Robocop fallacy": Just because we can use technology to do something weird that someone thinks will solve a problem, like build a mind-wiped cyborg cop to suppress supposedly intractable urban crime, doesn't mean we should, absent any evidence that the proposal is actually the best solution, to a real problem.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:41, 25 June 2015 (UTC)

• I'm generally (with exceptions) opposed to this for the various reasons mentioned in this and previous discussion of the idea. This is en.wp, and use of furigana does not help our readers. If anything it's just confusing to many of them. English Wikipedia's purpose does not encompass aiding learners of the Japanese language in working out unusual readings of Japanese. Whether "The ... text as shown in this revision ... has become dated" as the RfC asserts (without evidence) is essentially irrelevant to the question of whether to permit ruby markup of furigana in general Japanese representations in en.wp. I don't think anyone objects to it in special cases, such as the Furigana article itself, and in rare direct quotations of Japanese script that included it in the original. IIRC, Onna Nabe uses this, and a case has been made for doing so at Gosei (meditation). But it simply doesn't do anything useful for en.wp's target readership, English speakers; there's evidence it is still problematic in some browsers that are still in fairly popular use; and it presents display problems like excessive spacing between lines in all browsers when used in running prose. Some kind of narrow compromise could probably be worked out, e.g. including it in the first line of an article, only, when it's presence is especially meaningful, e.g. in helping to distinguish between two different uses of the same Japanese words; and also using it, again where deemed crucial, in quotations; and probably some other encyclopedic usage case I'm forgetting [reading other comments above, I see that Underbar_dk has proposed one]. As a more general approach to the presentation of Japanese in en.wp, it's unlikely that anyone will derive anything at all from it other than native and very advanced-learner readers of Japanese. The main desired case for it seems to be pop-culture trivia like teenagers' pronunciation of video game titles, about which I honestly don't think en.wp gives a damn.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:17, 25 June 2015 (UTC)
  PS: I agree that if furigana has been included, that placing it in brackets after the main Japanese text isn't helpful. But this is as much an argument for eliminating the furigana as for doing it in ruby. How does it help me or any other English speaker, however it is marked up? That's the key question.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:33, 25 June 2015 (UTC)
  • (edit conflict)I have no idea what "teenagers' pronunciation of video game titles" is supposed to refer to. Something that happens in Japanese literature is to give the kanji for a particular meaning and to gloss it with a word that has that meaning but whose pronunciation cannot be gleaned from the kanji: for example, in Yukio Mishima's The Sound of Waves, the kanji 巻揚機 "wind-up lifting machine" is given the furigana gloss ウヰンチ uinchi "winch", a reading impossible to get from the kanji as uinchi is a foreign loanword. The kanji provides the meaning of the word while the furigana provides the intended actual reading—almost as if the kanji were the gloss, even though it is the kanji that appears in the regular running text. One could not quote this text properly without the furigana (this is pretty much what's happening with the "railgun" example above). It's in places like this that the restriction on ruby should be lifted. Curly Turkey ¡gobble! 10:47, 25 June 2015 (UTC)
  • Having the ruby appear in brackets is the standard W3C recommended fallback for the ruby tags—in Japanese the furigana is sometimes necessary and not just "helpful", so if can't be displayed properly it at least gets displayed. Curly Turkey ¡gobble! 10:51, 25 June 2015 (UTC)
    • I linked to previous discussions for a reason. The fact that people mostly want to inject furigana into en.wp articles for video games, anime/manga titles, and other pop-culture stuff is no accident. It does not help English speakers in any way to give a pronunciation guide for kanji here in obscure Japanese furigana characters, since this is not ja.wp; just give Latin-script pronunciation guides like we would for Russian Cyrillic or any other non-Latin writing system. It's pure trivia that uinchi is a loanword from winch. (On the W3C side point: I know the bracketed display is a standardized fallback. That doesn't change the fact that it's not helpful; we've already been over two reasons why it's not helpful: One is the same, that furigana is generally not helpful here at en.wp at all, while the other is that when put after the kanji, it's unclear exactly what in the kanji the furigana was referring to. This has already been addressed in prior discussions.) I'm done here other than adding that if an editorial consensus agrees that The Sound of Waves actually is one of the cases where furigana would be helpful for some reason, then fine. I'm already allowing for that. I just personally think that pronunciation and etymological trivia regarding pop-culture titles is perhaps the least compelling kind of case. My take that some exceptions should be made, while keeping a general rule against furigana (a usability analysis), is severable from my opinion of what is a good case to qualify for the exception (an encyclopedic relevance analysis).  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:19, 25 June 2015 (UTC)
      • It's pure trivia that uinchi is a loanword from winch.: the point went entirely over your head—it is not a case of highlighting that a word is of foreign origin—the same technique is used with native Japanese words that are not normally associated with the kanji displayed. It's a technique that makes use of furigana as part of the text, and not as a mere pronunciation guide. Curly Turkey ¡gobble! 11:39, 25 June 2015 (UTC)
    • re in Yukio Mishima's The Sound of Waves, the kanji 巻揚機 "wind-up lifting machine" is given the furigana gloss ウヰンチ uinchi "winch": that’s a Japanese text, the gloss is to help Japanese readers (or people with near native Japanese reading skills). It is of no help to the vast majority English readers; even those who have picked up some Japanese and can perhaps read the kana will gain nothing from having it presented in addition to the Romanization, which should always be included.--JohnBlackburne^words_deeds 14:42, 25 June 2015 (UTC)
      • No, you too have missed the point that this use of furigana is part of the text and not a mere gloss. Curly Turkey ¡gobble! 20:56, 25 June 2015 (UTC)
• So far the only proposal submitted is mine. If you insist on an alternative from me you can simply replace the text with "Furigana shall/should not be used." @Curly Turkey: I am not going to add romanization must be used to my proposal as it would just be bloat. If that means no furigana, so be it. – Allen4names (contributions) 19:39, 25 June 2015 (UTC)
  • Then I'll have to oppose. Curly Turkey ¡gobble! 20:56, 25 June 2015 (UTC)
    • As I am now convinced that Japanese has been overused I now oppose all use of ruby to annotate Japanese. – Allen4names (contributions) 17:16, 29 June 2015 (UTC)
• I also oppose using ruby. Thank you. Alec Station (talk) 01:24, 2 July 2015 (UTC)

Proposal 2

The use of <ruby> to add Furigana to Kanji is generally discouraged, for the following reasons

• Romanisation should always be present, making the kana unnecessary
• It cannot be read by most readers
• The extra line height can disrupt layout
• It can get it the way of copying and pasting text
• Browser support, although improving, is not universal

Comments on proposal 2

Here’s a new proposal, which I think addresses the various concerns above. It softens the prohibition slightly while giving the main reasons not to use <ruby>, I don't think it needs noting that there will be exceptions, that is explicit in all MOS guidelines.--JohnBlackburne^words_deeds 18:14, 29 June 2015 (UTC)

• I would support that, including the implication that there will be exceptions. As Curly pointed out, authors sometimes do make creative use of their freedom to assign readings to kanji. The railgun case is a good example and I can see using the ruby there. In the Mishima example, almost always the only thing that needs to be said is that he mentioned a winch, which can be translated as "winch". If someone wanted to comment on his use of ruby in an analysis of his writing style that would be another exception, but generally I don't think we need to do that. – Margin1522 (talk) 07:32, 2 July 2015 (UTC)
  • Of course, such a usage should be limited to cases where such a direct quote were deemed important—unlikely in the actual example I gave, but you don't have to think too hard to imagine where such a (rare) case could come up. We should allow for such (rare) usages, while discouraging the idea that furigana should be used as a gloss in running text, as the gloss should always be in rōmaji, making the furigana redundant in all but corner cases (as well as somewhat disruptive with lineheight issues etc). I'd like to see the total ban lifted, yet furigana should still be discouraged. I don't like the proposed wording above, though, as it really much to wordy. Maybe a wording along the lines of "furigana should only be included when its absence would do disservice to the article". Curly Turkey ¡gobble! 08:54, 2 July 2015 (UTC)
    • The {{style-guideline}} at the top of the page already makes it clear that style guidelines are only that, guidelines with occasional exceptions, not hard rules. As such I don't think adding a note to any particular guideline is necessary. It is not that this should be handled any differently to any other cases of IAR, in which case editors can refer to/invoke that.--JohnBlackburne^words_deeds 18:06, 3 July 2015 (UTC)
      • Of course, but at WP:JAPAN we have a unique problem: the otaku. Curly Turkey ¡gobble! 22:43, 3 July 2015 (UTC)
        • Of course, every subject has its otaku. The way I see it, the MOS guidelines should be followed unless there is a really good reason for not following them. They should be the de facto standard, in order to establish and maintain a common appearance and usage across related articles (meaning "all Japan-related articles"). That's the whole point of the MOS. Exceptions should be exceptional and rare. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:26, 30 July 2015 (UTC)
          • Every subject has its otaku, but not every subject is overwhelmed by them. Curly Turkey ¡gobble! 00:34, 31 July 2015 (UTC)
            • True, but Japan-related topics are hardly the only subject "overwhelmed" by otaku. Pretty much every ArbCom case deals with otaku of one sort or another. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 06:10, 31 July 2015 (UTC)

Proposal 3

The use of <ruby> to add furigana to kanji is generally discouraged. It is generally only acceptable when giving titles where the pronunciation may be otherwise non-obvious (e.g., A Certain Scientific Railgun (とある科学の超電磁砲【レールガン】) Toaru Kagaku no Rērugan). If used, it should generally only be used once in the lead of the article (with rare exceptions), and Romanization should always be present.

Other usage of ruby is strongly discouraged as the extra line height can disrupt layout, it can interfere with copying and pasting text, and it is not useful to the majority of users of the English Wikipedia.

Comments on proposal 3

Thoughts? One thing to note: ruby doesn't seem to work in the {{nihongo}} template. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:38, 30 July 2015 (UTC)

I think the requirement that ruby should only be used at the beginning of articles would be detrimental to lists where Japanese script is included. Maybe a mention that the requirement is for prose only? _dk (talk) 03:03, 31 July 2015 (UTC)

I tweaked the wording a little bit. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 06:07, 31 July 2015 (UTC)

I think it'd be better if the wording were more along the lines of: "The use of ruby can be problematic, and should be restricted to cases where not using it would be detrimental to the article." In the Railgun example the pronunciation is not "confusing", it's deliberately non-standard—the furigana is not a mere pronunciation guide. Curly Turkey ¡gobble! 03:30, 31 July 2015 (UTC)

It is, therefore, confusing, since it is deliberately not pronounced in a standard way. This seems to be a "po-tay-toe" "po-tah-toe" thing. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 06:07, 31 July 2015 (UTC)

You can't be serious—this is nothing remotely like "po-tay-toe" "po-tah-toe". This is more like Raymond Luxury Yacht". Curly Turkey ¡gobble! 07:05, 31 July 2015 (UTC)

You saying it's not confusing and me saying it is confusing is absolutely a "po-tay-toe" "po-tah-toe" situation. You say it's one way, I say it's the other. Don't get so bent out of shape. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 07:12, 31 July 2015 (UTC)

Sorry, miscommunication: by "the pronunciation is not "confusing"", I meant "the pronunciation is not merely "confusing"". "Confusing" is not an acceptable reason for including ruby. Curly Turkey ¡gobble! 07:21, 31 July 2015 (UTC)

Changed to "non-obvious". ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:25, 2 August 2015 (UTC)

To clarify: the pronunciation of 下手 is "confusing", but we don't provide furigana for it, but a Roman gloss. This is not what's happening with the Railgun example. Curly Turkey ¡gobble! 03:34, 31 July 2015 (UTC)

_dk pointed me here because I got rid of ruby on one page because the katakana in ruby is so tiny. It's so hard to read. You can barely even see that it says "rerugan" here, particularly with the dakuten on the ga being so tiny http://puu.sh/jjXwq/cfb81c534b.png --2601:140:8200:23E5:C461:55B8:F4AE:CD77 (talk) 10:12, 31 July 2015 (UTC)

It's no smaller than in printed Japanese books. Again, it's not for everyone, but there are a few cases where it would be fine to include it. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 23:55, 31 July 2015 (UTC)

Additionally, anyone with any experience reading Japanese books which use ruby will know that's a dakuten since there are no handakuten on "ga". Japanese relies a lot on context and simply knowing things. It's the way it is. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:25, 2 August 2015 (UTC)

But I can't even see the dakuten. It looks like "ka" instead of "ga". It's just a single pixel bump. --2601:140:8200:23E5:CD58:C288:A5B7:C6E9 (talk) 06:19, 3 August 2015 (UTC)

And that indicates your unfamiliarity with printed Japanese. Those familiar with it would interpret that "bump" as the dakuten. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 17:13, 5 August 2015 (UTC)

I don't like it. The pronunciation is given immediately after in the romanisation, and should always be there or somewhere obvious. So in this case there is no need for ruby, for English speakers. Those with some Japanese may be able to read the kana but that, again, just matches the Romanisation. Only those with a fairly advanced understanding of Japanese, who can read the characters and compare their normal reading to the one given, will benefit from such ruby. But they make up probably a tiny fraction of the readers of en.wp, who are very unlikely to be reading this encyclopaedia’s coverage of Japanese topics, given that jp.wp’s coverage is probably much better. So per my earlier comments it should not be used unless it is the subject of the article, e.g. Furigana, or where the ruby is discussed directly.--JohnBlackburne^words_deeds 00:33, 5 August 2015 (UTC)

• Your objection ignores the Railgun example. Curly Turkey ¡gobble! 00:38, 5 August 2015 (UTC)
  • How did I ignore it? That example is in the proposal, and it is precisely that, or other cases like that, that I disagree with. It is of almost no benefit to our readers to add ruby; either they can’t read it, or if they can it just repeats what’s in the Romanisation, or the very few that can read Japanese well enough to make sense of it will probably read jp.wp in preference for its much better coverage of such topics.--JohnBlackburne^words_deeds 01:58, 5 August 2015 (UTC)

As a test I tried looking at this page with another browser, Chrome on Android. And it does something horrible to this:

(とある科学の超電磁砲【レールガン】)

it minifies the part of the text with ruby. I.e. it makes both the kanji and the furigana smaller so it is no taller than the characters before them. They end up between 3/4 and 1/2 the size they are meant to be. It varies between these two extremes as the screen rotates. I have uploaded a screenshot of the more extreme case. This is an up to date version of Chrome on Android 4.4.2.--JohnBlackburne^words_deeds 02:40, 5 August 2015 (UTC)

• • • You ignored it in that the romaization in no way handles what's happening in the Railgun example. If your objection is that the technology still doesn't handle it, then make that your objection rather than that romanization will handle what romanization cannot handle. Curly Turkey ¡gobble! 02:48, 5 August 2015 (UTC)
      • The ruby just matches the Romanisation. I don’t see how giving both is helpful to the vast majority of readers. And I just did a test with {{Ruby-ja}} and that has the same problem in Chrome for Android.--JohnBlackburne^words_deeds 02:54, 5 August 2015 (UTC)
        • So (a) the problem is technical; and (b) you don't understand what's happening with the furigana in Railgun. Curly Turkey ¡gobble! 02:59, 5 August 2015 (UTC)
          • (b): yes and no. I do understand it because it has been discussed here but I would not if I saw it in an article as although I can read kana I know very few kanji, my study of written Japanese not getting so far. And even my very limited Japanese is exceptional; almost all readers of this encyclopaedia know none at all.--JohnBlackburne^words_deeds 03:19, 5 August 2015 (UTC)

Proposal 4

There is a really very simple principle which answers almost all of the issues discussed here:

• English WP is written in English, using the character set and typographical conventions of English.

The only exception to this is:

• But English WP can and should quote other languages, and when it does, it can and should use the character set and typographical conventions of those languages.

Quoting should be judicious, and may occur typically in the lead, or it may occur throughout an article on, for example, the typographical conventions of another language. This is all that is needed to guide the use of "ruby" in quotes from Japanese. Imaginatorium (talk) 14:09, 1 August 2015 (UTC)

I must say this is the proposal I like best. Ruby does not serve the average English reader, so it being a pronunciation guide is not the issue. The issue is, as Imaginatorium rightly points out, providing the original title as it is presented in the original language (like the Railgun example). _dk (talk) 03:25, 5 August 2015 (UTC)

Some policy analysis on Japanese names

I don't want in any way to hold up closure of the above discussion on Japanese individuals' names (two major threads up). I think some things are worth additional (policy-esoteric) discussion on the side, without affecting the outcome of the above, but reflecting instead on where it's clearly going and how we deal with policy questions about it when it gets there.

Four (or 4.5) important things to me are detailed below. The nutshell version: 1) Where usage varies widely between two options in RS, either is "consistent" with RS. 2a) "Prevalence" is not a count of G-hits – reliability and relevance of the sources is important; 2b) relatedly, "naturalness" doesn't logically apply to forced Westernization of Asian names. 3) Policy tells us (but it's not been happening) that COMMONNAME has to be balanced against other criteria (like recognizability, which is also increasingly questionable for forced Westernization), and all of AT balanced against other WP concerns, like MOS clearly telling us to pay attention to subject preference (both individually and collectively), and common sense questioning the application of an arbitrary rule the basis of which is unsourced OR. 4) ABOUTSELF policy has largely been ignored in this debate, while the OFFICIALNAME essay has been mistakenly treated as if it trumps all other concerns. I conclude with an observation that COMMONNAME is not junk, but excessive devotion to it without balance has led to an NPOV and BIAS problem. I also suggest a post-RfC test case.

The detailed version, numbered for later reference:

1. WP:AT is often being read with "selective blindness". A key AT policy phrase is "consistent with usage in reliable English-language sources". Where usage is widely mixed (as it almost always is with Japanese names in English today), then either is in fact "consistent". WP:COMMONNAME has a particular rationale in it where it addresses "official" names: "the name that is most commonly used (as determined by its prevalence in reliable English-language sources) as such names will be the most recognizable and the most natural." This has two obvious breaking points when it comes to a case like this.
2. 1. "Prevalence" is not synonymous with "frequency", but strongly implies a evaluative judgement tied to influence/authority and continued longevity; this is reinforced by the reference to "reliable ... sources". That is the issue at the heart of the above arguments about high-quality vs. low-quality sources. By way of analogy, a strict headcount of American citizens will show that a majority of them believe in the Biblical creation story, not human evolution, but this view is not actually prevalent; it is not prevailing – not intellectually, where it is considered quaint, nor over time numerically, where it is steadily losing ground. The jingoistic practice of reversing Asian name order is suffering exactly the same two kinds of lack of prevalance.

      While not always put in such stark terms, this issue is to be found in plenty of other WP:AT and WP:RM discussions. As one ongoing example, it's at the heart of the application of WP:MEDRS expectations with regard to the name of the Electronic cigarette aerosol RM debate, which is divided sharply along along "do what popular culture does" and "do what quality sources on biochemistry do", and guess which side is losing ground? The subject-matter relevance of the source has a lot to do with how reliable it is in case like this, where we are deal with a desire to rename pre-existing names to fit a pattern preferred by others, against the original usage. (This is not true of all style matters; see the WP:Specialized-style fallacy, in which expert sources on technical topics are treated as if they are authorities on English-language usage, in attempts by some editors to force all of Wikipedia to use specialized jargonistic practices from their narrow field's internal publications, in ways that violate basic English-language practice.)

   2. No "naturalness" argument can plausibly be made for inverting people's name orders, in a world in which even American, English, etc., school children today know that Japanese (among many other Asian) name orders are naturally family-name first, especially given the enormous surge in popularity of Japanese media in particular, which of course often gets the Japanese name order correct. Wikipedia is actually on the "slow" end of the curve here. Before much longer, its forcible rewriting of Asian names to seem more Western is going to look as ignorant as people still referring to Asians as "Orientals".
3. The COMMONNAME criterion explicitly says to balance it against all the others: "Editors should also consider the criteria outlined above." But many just don't "see" that part. When we don't do what most quality sources do (or even what mid-quality but more specialist ones do, e.g. magazines devoted to Asian popular culture), but only what everyday journalistic ones (also mid-quality) do, this actually calls into question the recognizability criterion as well as the naturalness one already addressed. "Recognizable" to whom? Are most people looking for information on Japanese public figures 45-year-olds reading newspapers? Recognizability has to account for the expectations of the majority demographic of the WP article in question, and of WP as a whole, which is not just Americans and Brits reading The Wall Street Journal and The Financial Times.

   Even aside from COMMONNAME's relation to the rest of AT, the entire AT policy exists in balance with all the rest of our principles and best practices. One of these is obviously MOS:IDENTITY, which directly addresses this: "When there is a discrepancy between the term most commonly used by reliable sources for a person or group and the term that person or group uses for themselves, use the term that is most commonly used by reliable sources; if it isn't clear which is most used, use the term that the person or group uses." On what rational basis would this apply to group names, but not individual ones? And on what basis would Japanese group preference for family-first order be ignored? It defies common sense that this could only ever apply to demonyms, but that's how it's been treated in relation to this issue. The idea that it's so limited is a fantasy, as is the idea that "MOS is only about content". MOS:IDENTITY is regularly used not just to change in-article wording, but to move articles, including on individuals, e.g. transgendered persons who have undergone a personal name change. AT policy and its NC guidelines directly defer to MOS on all style matters in at least a dozen places, and identity has been determined to be a style matter, just as has the use of diacritics, and various other such name-affecting matters. Not much could be more central to identity that a personal name.

   Another principle AT must be balanced against is WP:COMMONSENSE, which is defied by a "rewrite all Japanese names of people born after a particular date, even though that date has nothing to do with name order" doctrine. Yes, we know it wasn't actually "arbitrarily" chosen, but its relevance has never been demonstrated. There's a principle that "extraordinary claims require extraordinary sources", and we have none for this. 1868 was a big year for Japan, and brought a lot of change, but flipping name order doesn't seem to have been among them (nothing at Meiji Restoration and related articles suggests any such change). It's blatant original research. The evidence before our very eyes is that Japanese people never stopped using family-first order, even if sometimes using it with Westerners, and they continue to use their natural order to this very day, no matter how much of a Westernizing influence they've been subjected to since the Victorian era and again post-WWII (even if some of them also continue to use Western order with Westerners, sometimes).

4. The words "secondary source" are not mentioned anywhere in WP:AT, so subjects' own publications being primary sources is of precisely zero concern for article title sourcing purposes. No source for what someone's name reall is can be more reliable that the person to whom the name belongs, absent something weird going on, like a criminal using aliases. AT policy pulls directly from WP:V, which does not require secondary sources for everything; rather, it says "Base articles on reliable, third-party, published sources...", and the WP:RS guideline and WP:NOR policy elaborate on this, while notably also allowing for primary sources. To the extent there actually are "rules" about secondary sources, they're about article content, which article titles are not. (There actually is one real rule in this regard, in NOR: "Articles may make an analytic, evaluative, interpretive, or synthetic claim only if that has been published by a reliable secondary source." And it has nothing to do with this sort of article titling matter.) For AT purposes, the salient part of V policy is WP:ABOUTSELF (which actually even applies to some content, too), and it very obviously applies to how we title articles about people, living or otherwise. This is also the policy basis for the changes to MOS:IDENTITY regarding transgendered persons, as well as the shift toward properly using diacritics in Latin-script names, so we're not even testing any new ground here.

   People also fetishize the WP:OFFICIALNAME essay and imagine that it means we avoid using names preferred by the subject, but it says nothing of the sort (to the extent that what it says matters), only that we're not required to do so; in actual practice, we almost always do so. The "official name" stuff in COMMONNAME (i.e. the only place in AT policy that addresses subject preference at all) was inserted principally to get at cases like "The Artist Formerly Known as Prince", and "the State of Rhode Island and Providence Plantations".

   In short, any case of a Japanese person clearly preferring family-first order in their Latin-script name is a strong counter to any COMMONNAME expectations. Contrariwise, any such person clearly preferring Western given-first order in their Westernized name also arguably meets the "truly exceptional" caveat that Curly Turkey and MChew have in mind; a combination of subject preference and common usage surely trumps any wikiproject preference for family-first as a default (thus Yoshitaka Amano is at the correct title per the English-language editions of his art books). Same goes for people known in English almost exclusive by given-first order (e.g. Ken Watanabe), where we have no direct evidence of subject preference. For many cases, this will be uncertain, and usage will be mixed, with ten of thousands or more cases of both name orders showing up in search results (e.g. Toshiro Mifune vs. Mifune Toshiro), skewed toward given-first order by low-quality sources. That skewing, combined with the fact that forced Westernization isn't really "natural", is why I favor family-first as a basic default.

I think it would be interesting to take the disambiguation page Amano as a test case, BTW, and start working down that list to see which names need their order flipped. I suspect it will be the case for more of them than COMMONNAME fans would like, but possibly fewer than CT and MChew would like. :-) In my view, COMMONNAME is hardly stupid or obsolete, and in many cases, especially public figures with a Western following, Western name order is clearly not just more common, but expected, and should probably be used here even in cases where we can't determine their own personal preferences. E.g. the credits in the English-language version of an anime film or TV show are going to reflect the decisions of the producers/translators, not necessarily those whose names are being presented in Latin script. Anime illustrators and such are generally not the subject of many academic sources, so "do what the academic sources do" does not give us an answer to what to do with such subjects.

Given that most of these people are semi-obscure in the West, a pure COMMONNAME approach fails COMMONSENSE. The presumption that everyday news sources can get the names right of Japanese people they know almost nothing about is closely akin to the idea that the same news sources get medical and technical details right when writing about specialist fields in which the journalists have no direct experience. News sources are not magically extra-reliable all of a sudden just because the topic has changed from astrophysics or oncology to Japanese people.

Finally the Japanese language, and names within it, are verifiable facts of the real world; choosing to side with those who campaign to reverse them to suit Western expectations is a serious WP:NPOV and WP:BIAS problem that cannot be escaped by worshipping COMMONNAME as a holy commandment instead of one factor among many to consider. We already had this very same debate with regard to diacritics, and it ended conclusively in favor of not bowdlerizing the names of foreigners to make it allegedly easier for English-speaking school children and retail clerks to remember them, and concern that is silly to begin with, given our WP:REDIRECT system.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  18:20, 13 July 2015 (UTC)

• I don't want to undermine my own position (which like SMcCandlish is to go with SURNAME-GIVEN), but I really have to object once again to the suggestion that GIVEN-SURNAME for Japanese is somehow a racist imposition by the West, especially given that countries like China, which underwent brutal, violent colonization, did not have the same thing "imposed" on their names, and I've seen no documentary evidence that such an "imposition" ever took place—and it's certainly not the feeling I get from flesh-and-blood Japanese people I interact with every day. Curly Turkey ¡gobble! 23:19, 13 July 2015 (UTC)

• My position is simple:
  1. Use the name order most commonly used in reliable sources.
  2. If the above can not be established, use the order preferred by the subject (for instance, on their website, in official publications or releases, etc.)
  3. If the above can not be established, use Japanese order (SURNAME GIVENNAME).
• I think that covers every possible situation. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:22, 30 July 2015 (UTC)
  • My problem with this is that it brings us back to my original point—it will inevitably result in articles referring to related people with different name orders, e.g. "Suzuki Hanako is the mother of Tarō Suzuki." (as in Kanae Yamamoto (artist)) Curly Turkey ¡gobble! 00:33, 31 July 2015 (UTC)
    • I'm fine with making exceptions in an article to keep all names in the same order within a specific article. That's what WP:IAR is for. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 06:08, 31 July 2015 (UTC)

Historical vs modern

Only one of the eleven participants in the recent RFC expressed support for the "historical figure" vs "modern figure" distinction, so I think we have a mandate to consolidate these two sections. Fernando Safety (talk) 03:14, 18 July 2015 (UTC)

I assume that "only one of the eleven" refers to me. Actually I don't think that's an accurate count, and it doesn't count any of the other recent discussions on this topic. I didn't comment about it in this one because it gets to be tedious following all of the forum shopping and making the same arguments every time, but I don't accept the argument that CMOS or Britannica mandate family name first (those assertions have both been questioned), and I don't accept the priviliging of academic sources over mainstream sources. I definitely don't think it's Wikipedia's place to correct the usage of journalists and the Japanese people in question and teach readers how to do it the right way.

I am about to go offline for a while (medical reasons) but when I get back I will be happy to discuss this further. In the meantime let's please consider this question open (which means IMO that your recent edits on the name order question should probably be reverted.) – Margin1522 (talk) 02:55, 20 July 2015 (UTC)

The rationale for “privileging” academic style over media style is to avoid a deadlock when the two styles conflict. WP:COMMONNAME endorses a similar tie breaker: "Other encyclopedias are among the sources that may be helpful in deciding what titles are in an encyclopedic register, as well as what names are most frequently used." I already noted your opposition above. I guess we will see if others share it. Fernando Safety (talk) 14:59, 20 July 2015 (UTC)

Sorry for the delay in this reply. I think there has been quite a lot of tendentious, out-of-context quoting going on in the family-name-first arguments. The above quote from WP:COMMONNAME endorses names from encyclopedias in preference to unusual "official, scientific, birth, original, or trademarked names". There is no suggestion that names from encyclopedias should take precedence over names in major English-language media sources like the New York Times. To say something like that in WP:COMMONNAME would be nonsense.

Look, the rule we have had for the past decade has always been a compromise. There are two competing systems, equally valid, and the Wikipedia method for dealing with situations like that has always been to allow both as far as possible and when appropriate. Differentiating between modern and historical figures is one way of allowing both. It agrees with the actual practice of the majority of sources cited in Wikipedia, and IMO is the solution that is most typical of and compatible with Wikipedia.

I myself would prefer the family-name-first system. It would certainly make my day job easier. The problem is that none of my Japanese clients, friends, or colleagues want that. They all prefer the other way. I wish people would remember that. The given-name-first order isn't something that was imposed by a cabal of WP editors. It exists in the world and all we are doing is recognizing it.

Also, if I could add this, there must be tens of thousands of modern Japanese names in Wikipedia. The MOS currently says that they are in the wrong order. What are we going to do about that? Fixing it is going to be an enormous task. Do we have any volunteers to get started on it? If not, then wasn't it jumping the gun to change the MOS only? – Margin1522 (talk) 06:15, 4 August 2015 (UTC)

You have multiple positions that are outside the range of opinion of everyone else who participated in the RFC. Yet somehow in your mind your view is the “compromise”! No one else thought that including an arbitrary cutoff year in the MOS was a good idea. And no one else based their argument on New York Times style. If NYT style is the basis for the cutoff year, then the cutoff should be a lot later than 1868. COMMONNAME does suggest using Google News, but that goes back only about five years or so. Otherwise, I assume editors are expected to follow academic/encyclopedia style. In any case, I’ve always been more interested in getting rid of the cutoff date than in which order the names are in. — Preceding unsigned comment added by Fernando Safety (talk • contribs) 21:24, 4 August 2015 (UTC)

On the compromise, see this archive from 2004, which is when the Meiji divide was adopted. Compromise is how it was described at that time, including by the late editor Fg2, who originally suggested it.

Complaints about the arbitrariness of the cutoff date have been made since 2004. The advantage of having a single date is that it prevents tedious disputes about article titles, a chronic problem on Wikipedia since we have so many editors who would rather argue about the title of an article than contribute to it. But it that's the problem, then we could adopt one of the proposals to make the cutoff date fuzzier. For example, instead of using their birth date, we could say that historical figures belong to the period in which they were active -- when they fought their battles or published their books or whatever. That would allow the flexibility that some editors want to deal with borderline cases, at the cost of increasing the noise level. Whatever, maybe it's worth it. I wouldn't oppose such a proposal.

About the NYT, I'm not basing my argument on one newspaper or on any appeal to authority. I'm basing it on the prevalent usage in the majority of English-language sources actually cited in Wikipedia. IMO if you read the sources, the course that we should take is pretty clear. – Margin1522 (talk) 01:05, 5 August 2015 (UTC)

Uh, so you are now pro-"compromise" or anti-"compromise"? I guess it doesn't really matter. In the RFC, eight editors voted for more or less the current guideline, three for various other solutions. Nihonjoe’s recent post suggests that he has revised his thinking. The third editor just said we should “follow English usage,” which doesn’t engage these arguments at all. So whatever your "pretty clear" position is, you are on your own. That the article should follow the usage in its sources is good thinking. I note that MOS:FOREIGN already contains a provision of this kind. Fernando Safety (talk) 22:17, 5 August 2015 (UTC)

You keep trying to make this about me. I simply pointed out that the long-standing policy was a compromise, which you would have known if you had taken the advice last time to read the archives.

I've already said that I disagree with your vote count. Now let me say that I think the design of the RfC itself is tendentious. Option B is a strawman. Has anyone ever suggested that all Japanese names, including historical ones, should default to given name first? Of course not. So why is that an option? And why are options A, B, and C about name order, while option D is about the "current language" of the MOS instead of name order and the principle of distinguishing historical and modern names. Since your main aim is apparently to abolish that distinction, you'd think the RfC could at least mention it.

So, we have two academics who voted for the academic order. Fine, that's their right. I myself, as someone who has spent more hours of my life than I like to remember trying to track down the roman spelling of Japanese names, would like to spare our readers the trouble. I would like the default name order to be the same as the one used by the Japanese government (Cabinet Office), corporations (Toyota), and major English-language media sources. I would like names in WP to agree with the names that show up in Google searches. Our readers will be happier. – Margin1522 (talk) 05:17, 6 August 2015 (UTC)

Frankly, you are becoming less and less coherent. I explained the cutoff date right at the top of the proposal. Two people voted for option B. They didn't think it was a strawman. Fernando Safety (talk) 22:12, 6 August 2015 (UTC)

Presumably the voters for option B wanted to endorse the existing practice of Western order for modern names, and would have if that option had been available. That didn't prevent MChew from noting that option B covers historical figures as well, and mocking it as "Yoritomo Minamoto" and "Shikibu Murasaki". Was that your intention?

I note that you still haven't indicated your plan for fixing the problem I mentioned earlier – that most modern names in Wikipedia are now in the wrong order, according to MOS-JA. As the person who rewrote the MOS, what do you think we should do about this? Does it need to be fixed, and if not why not? I think you owe us an answer on that.

Let me note one other problem. As it stands, the MOS says to give precedence to the form used by the person himself, or on the person's behalf, in English-language materials. Almost always this will be in Western order. For names that can't be confirmed in English-language materials, we are supposed to use the opposite order. What this means in practice is that articles will have modern names in both orders. Those confirmed in English will be given-name-first. Those confirmed in Japanese will be family-name-first. In other words, the problem of inconsistent name order that Curly was complaining about will be found in many more articles from throughout modern history, right up to today, instead of in a few articles about the Meiji period. This is an improvement?

If we are going to say that the default name order is family-name-first, then I don't see any way around this difficulty except the academic approach -- ignore the person's own preferences in English-language materials, and always use family-name-first. I wonder what the people who voted for option A think about this. Are they willing to be consistent and always use the style favored by academics who are literate in Japanese? – Margin1522 (talk) 00:21, 7 August 2015 (UTC)

As an editor, I'd say yes, it'd do away with so many headaches. Whether that best serves readers? I'm on the fence about that. Having surname first doesn't seem to raise issues with regards to Chinese and Korean names ... Curly Turkey ¡gobble! 01:46, 7 August 2015 (UTC)

• My first instinct is to support family name first across the board, aside from a few exceptions of the Yoko Ono type. That works for China, Korea, and Vietnam, all of which have nomenclature traditions similar to Japan's. It does occur to me that there are some hard cases to deal with. For a singer or an author, it may make sense for the article title correspond to the name under which the albums or books are sold. The name order used in the article's references can be a consideration as well. I do not support treating the subject's website or other promotional material as an indication of his "personal preference." Typically, we talking about Japanese-language sites. The designers of such sites are unlikely to have put much thought into the name order issue. Fernando Safety (talk) 12:05, 8 August 2015 (UTC)

I think the Yoko Ono case has little to do with "personal preference". It's what the audience knows and expects. Same with Ichiro Suzuki. Who knows what Ichiro prefers? For persons who are not well known, I think we might as well just go ahead and assume a personal preference for given-name-first, since that is almost always true, and then decide that we are going to ignore personal preferences and ignore the order used in the sources. It's just too confusing to have different orders in different articles, or even within the same article. We have to choose one or the other and enforce it. And since we – the small minority who are literate in Japanese – find surname-first the most natural then surname-first it is. That is, assuming we go this way, in defiance of normal WP conventions.

Note that this is different from honoring personal preferences in spelling – Goto, Gotoh, Gotou, Gotō, or whatever. Readers can figure that out. What they can't figure out is which part of Keita Gotō is the surname.– Margin1522 (talk) 03:30, 9 August 2015 (UTC)

I feel there has been a misunderstanding. No, I do not think the subject's personal preference should be a factor in titling. I certainly do not think we should assume a personal preference, which strikes me as a contradiction in terms. If a singer has put out a series of albums and his name appears the same way on the cover of each, that's probably his personal preferences. But in general, there is no way to tell what a celeb's personal preference might be. There are surveys which suggest that ordinary Japanese are split down the middle on this issue. So you can't assume anything. Fernando Safety (talk) 04:36, 9 August 2015 (UTC)

Japanese words spelled with the full stop

This follow a post at Wikipedia_talk:Article_titles#Getting_tied_in_knots_by_a_Japanese_fad.

Having looked at issues caused by spellings that include a terminal period ("."), and a little information such as at Japanese_punctuation#Words_containing_full_stops, it seems to me that the Japanese "fad" of including a terminal period in the spelling of some names should be considered a Japanese stylisation that should not be used as normal English, in either text or titles.

Examples that have come up include Melody. (Japanese singer) and Gangsta.. The use of of the period in the spelling in running text is disturbing to reading of the text, and the use in the title leads to it frequent use in reference to the subject in other articles. For example:

"The Japanese lady Melody. is an American-born fashion designer."

"Kohske illustrates the Gangsta. comic book series."

In Japanese, does the full stop convey meaning that justifies its use in English translation? In Japanese, does the full stop interupt sentence flow as it does in English?

Would it be appropriate to introduce a recommendation in Wikipedia:Manual of Style/Japan-related articles (WP:MOS-JA) that Japanese words or names containing a terminal full stop should not include the terminal full stop when translated? --SmokeyJoe (talk) 04:41, 27 July 2015 (UTC)

• The Japanese full stop is 。. I have no idea what the period in these cases residents represents, but I suspect it is purely decorative. Curly Turkey ¡gobble! 07:14, 27 July 2015 (UTC)
• I think it's used merely for decoration in these two cases. It would be fine to use it once at the beginning of an article, but I wouldn't use it elsewhere as it would likely cause confusion among editors as well as grammar bots. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:16, 30 July 2015 (UTC)

Proposal

There is a really very simple principle which answers almost all of the issues discussed here:

• English WP is written in English, using the character set and typographical conventions of English.

The only exception to this is:

• But English WP can and should quote other languages, and when it does, it can and should use the character set and typographical conventions of those languages.

The writing of what looks like an English full stop at the end of a name in Japanese is a typographical convention from Japanese. In fact it is simply "decoration", and all sorts of bits of cruft are much beloved by designers of bits Japanese written in roman letters. Therefore, such names should be written with the closest approach in English ("Melody" etc), but of course it is appropriate to say something like "styled in the Japanese as "Melody."", perhaps adding an explanatory comment that the "." here is not punctuation. There is a similar issue on the title of One-punch man, where the Japanese styling looks a bit like "ONEPUNCH-MAN", but in fact the mark resembling a hyphen is surely a nakaguro, mid-point, representing more separation between 'PUNCH' and 'MAN' than between 'ONE' and 'PUNCH'. As far as possible, translated titles should provide the same semantic overtones as the original, not the closest graphical similarity. (You know, like translating shaku as "R".) Imaginatorium (talk) 14:16, 1 August 2015 (UTC)

Quoting should be judicious, and may occur typically in the lead, or it may occur throughout an article on, for example, the typographical conventions of another language. This is all that is needed to guide the use of "ruby" in quotes from Japanese. Imaginatorium (talk) 14:09, 1 August 2015 (UTC)

Let's keep all the discussion above (in the other section) so people don't get confused at discussing the same thing twice on the same page. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 16:08, 1 August 2015 (UTC)

But this is a separate discussion -- about Japanese names written with decorative pseudo-punctuation. My point is that one principle pretty much answers all of these discussions, which eliminates the need for anguishing over exactly how and where to allow/encourage/discourage ruby, for example. Imaginatorium (talk) 15:22, 2 August 2015 (UTC)

COPYVIO at Kishi Nobusuke

Great big heaping chunks of the Kishi Nobusuke article have been lifted verbatim from "Kishi and Corruption: An Anatomy of the 1955 System" by Richard J. Samuels. I don't have the time myself to deal with it—could sometime spare the time to clean this up? Curly Turkey ¡gobble! 00:41, 5 August 2015 (UTC)

I removed everything I could find. Let me know if there is any I missed. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 17:25, 5 August 2015 (UTC)

Just so more people can see it, in the future I recommend posting a notice to WT:JA. More people watch that page. ···日本穣^? · 投稿 · Talk to Nihonjoe · Join WP Japan! 21:17, 5 August 2015 (UTC)