Misplaced Pages

Revision as of 07:52, 13 September 2005

Template:Babel-4 I am a translator of Gaim, in a certain two-member translation team; a (rather lazy) translator/developer of Koha; and a (also rather lazy) member of linux-community (a GNOME translation team).

This should be enough information for you to figure out my name :D

About my name

My user name on wikipedia is "Gniw", which is "Wing" written right-to-left. I guess this means you can pronounce my wiki name as either GNU (since /i/ and /j/ are allophones, and <w> is a "double u") or as Wing.

About the name "Wing"

"Wing" is a word frequently used in Chinese names. It is the most significant word in my real name for the purpose of forming diminutives. When forming diminutives or other nicknames of Chinese names, the last word in a given name is usually the part that is retained — the rest of the given name would be replaced by the prefixes 小 (little), 阿 (Ah), or suffix 仔 (little one), etc.).

(For obvious reasons, if the last word in the name is known to be not actually the “most significant” word, diminutives are never formed and the full given name is always used.)

"Wing" is a transliterated Cantonese pronunciation of more than one possible word. In the "most significant" position (i.e., as the last word in the given name), it is most often either 詠 (song), 穎 (excellence), or 榮 (glory). 詠 and 穎 are usually used in female names while 榮 is often used in male names. In Cantonese, the words are pronounced differently (a difference in tones); in English, however, this difference is lost and it is not possible to discern whether someone known simply as "Wing" is male or female.

In Mandarin, the different possible Chinese words corresponding to Cantonese "Wing" have completely different pronunciations (pinyin yǔng, yǐng, and róng, respectively) and would not cause confusion in English.

The Mandarin "r" sound

This brings us to the Mandarin "r" sound. Some people keep wondering why foreign words are not usually transliterated with "r" words, and some (e.g., the name of Switzerland) who is transliterated with an "r" word do not have an "r" in the original.

The previous discussion about the word "Wing" gives a clue: the Mandarin "r" sound was not historically an "r"; in fact, it does not quite sound like an "r" when correctly pronounced.

Recall that the three possible words transcribed as "Wing" are pronounced yǔng, yǐng, and róng, in Mandarin. Is there a possibility that the "r" sound actually is a kind of "y" sound? The answer is yes; in fact the way many native Mandarin speakers say the "r" sounds more like a "y" than what one expects from an "r" (unless you speak Gaelic, perhaps); this is reflected in the decidedly "not r" spelling of the "r" sound in Wade-Giles.

The _jyt ˈjɐm ˉdziŋ _duk _dzi _wɐi (ISBN 962-948-509-5, and yes, you are reading it right, the title of the book is in IPA — it also has a Chinese title, though), pp.413–414, states that

In Middle Chinese, retroflex sounds probably did not exist. However, after the Yuan dynasty, .

In Mandarin classes, Cantonese speakers are often reminded that Mandarin "r" sounds are not rolled; one simply need to make the back of the tip of the tongue touch the top of the mouth.

In contrast to English names

The wikipedia article on Chinese names mentions that Chinese names are different than English names in that Chinese names can be made up of any words, while English names are usually drawn from a limited set of names.

This has not always been so. Old English names were constructed just like Chinese names.

The writing direction of Chinese (or, "Why Gniw?")

Chinese was a right-to-left language. As recent as about 30 years ago, there were still some old people in Hong Kong who simply could not read left-to-right Chinese. Store signs were frequently written right-to-left, as were newspaper headlines.

Nowadays, Chinese is usually written left-to-right when written horizontally. This writing direction has been standardized in the People's Republic of China.

Right-to-left Chinese is still sometimes seen, for example in historic monuments, and for writing the Chinese New Year greeting "出入平安" (peace in going outs and coming ins) which is always written horizontally. In areas where traditional Chinese is used, it is also sometimes seen on store signs (but usage is decreasing), newspaper headlings (usage also decreasing, probably attributable to lack of software support), etc.

When written vertically, Chinese is still written from right to left. However, there has been a recent trend for newspapers and magazines (the first time I've seen being from a few years ago in a Korean newspaper) to write even vertical Chinese from left to right. This has alarmingly been becoming more widespread in Chinese media; while done to harmonize the writing direction of headlines and body text, this practice in fact makes the headlines difficult to read because it is not the expected writing direction of vertical text.

Most software is unable to process horizontally-written Chinese as a right-to-left language; this lack of support might explain the accelerated adoption of a left-to-right direction. Recently (around 2005), the Canadian edition of the World Journal has switched the writing direction of its news headlines from right-to-left to left-to-right; as it happens, World Journal uses a typesetting system from Founder (a spinoff of the University of Peking), which does not support a right-to-left writing direction.

The disuse of the right-to-left direction might also be attributed to an apparent lack of standardization in how English and other left-to-right text should be incorporated into right-to-left Chinese.

About my language teams

My language team is traditional Chinese, normally indicated by the language code "zh-TW". (IANA has, however, assigned a new code, "zh-hant", to represent "traditional Chinese". As far as I can tell, this new code is not actually in use "in the wild".)

Idiomatic usage of the language code "zh-TW"

The code "zh-TW" comes from the ISO 639 code "zh", meaning "Chinese" (pinyin zhōng wén (中文), meaning "the Chinese script") and the ISO 3166 code "TW", meaning "Republic of China" (but which is obviously an abbreviation for "Taiwan"). Contrary to popular belief, however, "zh-TW" is idiomatically used to generically refer to "traditional Chinese"; it almost never means "Chinese (Taiwan)".

This idiomatic usage seems to be little understood. As a result, even though "zh-HK" should fall back to "zh-TW" (because Hong Kong also uses "traditional Chinese"), this is often not done. This could very possibly account for the low usage of the "zh-HK" locale.

Lack of foresight in language tagging on the Web (1)

On the World Wide Web, "zh-TW" is associated with the Big5 character set, while "zh-CN" ("CN" being the ISO 3166 code for " China") is associated with the GB2312 (GB) character set. While this usually gives the intended results, it also introduces ambiguity into the markup.

A consequence of the above is that text marked as belonging to "lang=zh-TW" is displayed in a Big5 font while text marked as belonging to "lang=zh-CN" is displayed in a GB font. Because of the way fallback handling of language codes works, any unknown "zh" language codes are displayed in a GB font by default.

This leads to the question: How can one properly mark up pinyin transliterations? According to RFC 1766, this should be a code consisting of a primary tag "zh", indicating "Chinese", plus a subtag of "pinyin" or similar. As it happens, because of the assumptions described above, pinyin text would be displayed using a GB font, which is not what is expected; it is possible to correct this using cascaded style sheets (CSS), but the results would look wrong when CSS is turned off.

Similarly, how does one mark up Japanese romanized words (or transliterated Hebrew, Russian, etc.) in an English document? A language code similar to "ja-romaji" would be appropriate according to RFC 1766; this would also give the wrong results.

(Note about the new IANA codes "zh-yue", "zh-guoyu", "zh-hans", "zh-hant", etc. These codes also suffer from all the above-mentioned problems.)

Lack of foresight in language tagging on the Web (2)

Another problem arises when one discusses Mainland Chinese idioms in a piece of text written in a traditional Chinese–using region. For example, if we are to write (in traditional Chinese)

The word "quality" (質素) is written "mass" (質量) in Mainland China. This introduces an ambiguity because in scientific usage "mass" always means "mass" , not "quality" .

The second word rendered as "mass" in the above example is actually the Mainland Chinese way to write the word "quality"; it should therefore be marked as zh-CN. However, marking it as zh-CN would cause the browser to display it using a simplified Chinese font, would would not work.

The opposite is also true. For example, if we are to write in simplified Chinese

In Taiwan they call it the "internetwork network" (网际网路), and in Hong Kong they call it the "interconnected network" (互联网); we just call it by their English (transliterated) name, the "Internet" (因特网)

even though everything should be in simplified characters, there is a Taiwan term which should be marked zh-TW and a Hong Kong term which should be marked zh-HK. This would not work, as the browser will try to display simplified characters using a Big5 font, causing display problems.

In other words, when one actually tries to use the codes "zh-TW", "zh-HK", and "zh-CN" properly—to describe dialects—, the browsers' assumption regarding the association between dialects and scripts would fail, causing undesirable results; as a result, these codes are in practice never used to refer to actual dialects.

This problem, unfortunately, does not only affect CJK languages. For example, the following, if written in English, would also likely not work if languages are properly tagged:

The name of the country China, when written in ideograms, are identical in Chinese and Japanese, but pronounced differently: in Mandarin Chinese, Zhōngguó , in Cantonese /tsuŋ¹ kwɔk³/ , and in Japanese, chūkoku /tʃuːkɔkə/. In many other languages the name might be allegedly derived from the word Qin /tsin/: China /tʃainə/ in English, chine /ʃin/ in French, Kiina /kiːna/ in Finnish, siin /siːn/ in Hebrew, etc.; if we assume that the modern Mandarin and Cantonese pronunciations reflects the old pronunciation, the /ts/-like sound have become variously /tʃ/, /k/, /s/, and /ʃ/…

In such an example, only the words tagged as French and Finnish (i.e., those assumed to be written in the same character set as English) would display correctly.

About me

My interests

In alphabetical order, my interests include: CJK computing, support for creation science, free software, integrated library systems (ILS's), languages and (to a lesser extent) linguistics, translation, and typography

Perhaps I should steal a line from another wikipedian and add: editing various Wiki pages in the English Misplaced Pages, so as to correct the wrong/misleading information and fill the missing information that I think is important

This turns out to be more difficult than expected, because some people don't believe you.

My wikipedia wishes

The pages on typography in wikipedia are biased towards western typography. I would like to, if possible, make the articles refer more to CJK typography.

Misplaced Pages's articles on scientific creationism cannot be described as neutral at all. The article for the Noah's Ark is relatively close to neutral, though there is still much unnecessary bias against creationism; the bias in a lot of the other articles are much, much worse. Since the wikipedia guidelines says articles should be neutral, all such biases should be removed; this is, however, a daunting task because there is so much bias.

I agree with another wikipedian that surnames should be in all caps, especially if the surname is not in the usual English position. This is not even a CJK issue; the French does the same thing too.