On Mon, 19 Jul 2004 13:25:38 +0100 Timwi
<timwi(a)gmx.net> wrote:
Andre Engels wrote:
In the Pywikipediabot, it is now specified to use
alphabetical order
by language name
How do you alphabetically sort 中文, 日本語, etc.?
Using the most common transliteration into the Latin alphabet.
However, I know that on nl: exactly this problem was the reason to use
alphabetic by code instead. The same might hold for de:, I don't know
about them.
To be more correct it should be "romanization" instead of
"transliteration". It has to do with how a set of non-alphabetical
symbols in a language are represented with the Latin alphabet. Pinyin
representation may be close to the pronunciation of Chinese in some
European languages, but that is not an essential element of
romanization. It should be valid for use in contexts of all languages.
That would make everyone's life much simpler. The difficulty may be in
attempting to determine what is "most common". At one times the
Wade-Giles system prevailed for the romanization into English, but that
has rapidly become obsolete. I imagine that there is a similar
situation for other languages. Japanese romanji may be in a similar
situation.
Ec