On Thu, 11 Jan 2001, Mark Davis wrote:
> By the way, I am still missing the following. If anyone can supply them, I'd
> appreciate it.
>
> [BOPOMOFO]
[snip]
>[MONGOLIAN]
[snip]
> See http://www.macchiato.com/unicode/Unicode_transcriptions.html for
> details.
It's still not very clear to me what this is supposed to be a list of.
The title says "Transcriptions of Unicode", and a note at the bottom says
"For non-Latin scripts the goal is to match the English pronunciation --
not spelling."
Some of the entries (leftmost column of the table) are names of languages,
while others are names of scripts. e.g., "Russian" and "Japanese" are
names of languages, with examples given in Cyrillic and Katakana,
respectively. For some scripts, there is basically only one language that
uses it, such as Katakana (used by Japanese) or Hangul (used by Korean),
while other scripts are used by many languages. It this supposed to
suggest that Russian is the representative language to give a Cyrillic
example in, and say, not Mongolian?
In some cases, it seems the example is not necessarily a transcription of
the English pronunciation, but a translation into another language,
most likely a loanword, with attendant sound changes. e.g., Japanese
"yunikoodo". I notice the lack of a request for an example using the
Hiragana script (which is also used by Japanese), which suggests that the
Japanese example is not a transcription of the English pronunciation into
Katakana, but a Japanese word (albeit a loanword). Otherwise, it would be
possible to provide a Hiragana example, however nonsenical or non-existant
it may be in reality. There is also the particular case of the Chinese
entries, written in CJK "ideographs", which *are* translations using the
calque strategy.
It seems to me that this list is intended to showcase a variety of ways to
write "Unicode", be they transcriptions, transliterations, or
translations--whatever maximizes the number of scripts that one can show
off, apparently.
This raises some questions of what an example showcasing the Bopomofo
script should look like. Basically, it is used only for Chinese,
primarily Mandarin (zh-guoyu). It is also primarily an auxiliary script
for ruby annotation of Chinese text written in CJK "ideographs", although
it may stand alone. So, if it is a transcription of English
pronunciation, then it will have to go through the language filter of
Mandarin Chinese, and this form may or may not be attested in
reality--perhaps as a "best-fit" colloquial attempt to say a foreign
(English) word. And this version would have the script standing alone.
Alternatively, it could be a transcription according to Mandarin Chinese
pronunication of the already existing Chinese translations written in CJK
"ideographs". In this case, it could either stand alone, or be attached
as ruby annotation to the CJK "ideograph" version (in Chinese).
Implemenation-wise, it would be problematic seeing the Bopomofo at the
size it would be in for ruby annotation of text in a 96x24 bitmap (as
requested on the page. Also, Bopomofo does have an inclination to be used
with Chinese text written top-to-bottom, so the horizontal shape of the
96x24 bitmap is problematic--more generally, vertically written scripts
such as the traditional Mongolian script (also requested) cannot be
demonstrated within this framework.
Thomas Chan
tc31@cornell.edu
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:17 EDT