Re: status: EDUML v0.6

[Barret Dolph <white129@ms7.hinet.net>]

A question from an ignorant headmaster:

Would this work with non-european languages such as Chinese or Japanese.
Example: All Chinese "words" have one syllable. One then combines these
words into groups to make a meaning. Example: dong, east + shi, west =
thing. The reason why I say this is to show that Chinese works quite
different from languages we are used to. Furthermore, each word is a
picture. Formatting changes from Spanish to English are quite easy.
Formatting changes from English to Chinese are not. Example: An English
based Access Database may or may not work in Chinese because of formatting
difficulties.

If anyone is interested in how Chinese could be integrated in such a way
that it would not have formatting problems or require the operator to know
what English the translator is guessing I would be much obliged.

Cordially,
S. Barret Dolph
Headmaster
White Horse English
Development Center
Taipei Taiwan

PS How does this work with Russian. My Russian is too poor to think through
this problem.

Duane Morin wrote:

> > <locale>           // internationalization (optional)
> >  <school fr="ecole" es="escuela" eo="lernejo"/>
> ...
>
> I'm not sure my original comments on this subject made it out, or maybe
> I just got overruled, but I'll say it again in case.  If  you want to go
> this way, and assign the languages as attributes, then you lock yourself
> into a set of languages for good.  If in the above example somebody
> wanted to add "it" for "Italian", then the DTD would have to be recoded,
> as well as all the software that reads it.
>
> More flexible, though perhaps more work as well, would be to create a
> dictionary something like this:
>
>    <dictionary>
>     <word lang="en"> <!-- "language the base is coded in" -->
>      <base>name</base>
>      <alternate lang="fr">nom</alternate>
>      <alternate lang="es">nombre</alternate>
>      <alternate lang="eo">nomo</alternate>
>     </word>
>     <word>
>      <base>school</base>
>     </word>
>    ...
>    </dictionary>
>
> That way your dtd is language independent, and translating the tags is a
> relatively straightforward algorithm:
>
>    if (tagname IN getDictionary().getBaseWords()):
>      for a in getDictionary().getWord(tagName).getAlternates:
>          if (a.lang == desiredLanguage): return a.getText()
>
> or something like that.  It also encapsulates all the translation info
> in one spot, so if you don't want to internationalize, just don't add a
> dictionary element.
>
> Duane
>
> ______________________________________________________
> Get Your Private, Free Email at http://www.hotmail.com
begin:vcard 
n:Dolph;S. Barret 
tel;fax:022-937-6633
tel;work:022-937-6633
x-mozilla-html:TRUE
url:http://www.angelfire.com/on/WhiteHorse/index.html
org:White Horse English Development Center
version:2.1
email;internet:White129@ms7.hinet.net
title:Mr. S. Barret Dolph
adr;quoted-printable:;;#1 Bao Yi Road, Lane 26, #1=0D=0AWen Shan Area Mucha District;Taipei;;;Taiwan
fn:White Horse English Development Center
end:vcard