🚀 go-pugleaf

Hello, Kyan.

> (1) natural language <--> natural language translation is a
> difficult problem for machines because, basically, machines
> cannot understand the meaning of texts, at least not with
> current technology.  my idea is to remove the "semantic"
> part of this problem so that the remaining task for the
> computer is reduced to syntax / transliteration aspects,
> things that computers do well.  strange as it may sound,
> the point is to make _humans_ adapt to computers.
I'd like to refer you to at least Routledge Encyclopedia of Translation
Studies (1998, $169 + shipping from Amazon.com ;((), section on Machine
Translation. You'll find answers to your questions ;) And I think Mona
Baker (ed.) had also included some selected bibliorgaphy for you. What
you're looking for is called a limited or restricted input. In Canada
computer translates loads of weather forecasts every day from E into F
(or vice versa, don't remember) and noone notices.

+: if you're looking for an interlingua, it's a different approach.

Yakov Karchev

> Talking to children, foreigners, and retarded adults...


--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---