[YG Conlang Archives] > [katanda group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

RE: [katanda] machine translation



> No.  It's intended to represent all possible meanings, WITHIN REASON.
> Language specific subtleties will either not be translated or will
> require periphrasis.  Also, the way that a meaning is represented in
> Katanda may differ from the source language.  For example, Katanda might
> use a prefix while another language will use a word or phrase, or
> vice-versa.  Katanda is quite flexible, but it's not a "shape-changer"
> that's capable of being anything at all.

So how much "noise" would show up in translations into katanda and back into
the original language? Have you done tests with this for english? Have you
tried translation with languages other than english?

> Of course, but then this is true of all languages as long as the
> vocabulary is enhanced as needed to represent new concepts.  However,
> unlike natural languages, the meanings of Katanda words, once created,
> will not change.
>
>
> Regards,
>
> Rick Morneau
> http://www.srv.net/~ram
> http://www.eskimo.com/~ram
>

there are a couple big problems with real languages though, which i'm sure
you already know:

1. they change over time
2. they are dependent on a particular culture's way of thought
3. idioms/slang

the problem with ever using a natural language for a timeless translation is
that because of (1), (2) and (3) offer big problems for foreign people
translating that language. You might as well leave it in the original
language if you have to deal with these problems.
--
Garrett Jones
http://www.alkaline.org