RE: Transliteration

> 
> After the discussion on this list a few months ago,
> the consensus was the the following is needed
> 
>  - Source language
>  - Target language
>  - Transformation method
> 

My thought would be that the source and desitination might
be something more like what the Text Encoding Initiative
referred to as writing systems. (I don't have their
pages in front of me so I'm not sure if it is an exact
match).

(I'm a little bothered by terminology here too; if it was just
language to language, it's translation, not transliteration).

Don't lose the distinction between languages, scripts, and 
character encodings. (After the work done to make HTML
support unicode character references, I hope we can keep
"charset" as orthogonal as possible.)

I think of cases like Hindi and Urdu which are almost the
same languages, written in different scripts. Or to pick
an example closer to home, I've seen Japanese text that
was originally written in romanized form but with JIS-encoded
fixed-width characters instead of ASCII. I could convert
it to kanji & kana or ASCII without changing the language.

--
    Albert Lunde                      Albert-Lunde@nwu.edu

Received on Friday, 16 October 1998 10:48:09 UTC