Re: Universal Translation Language

This message is in MIME format.

-------18881-----Avana-Web-Mail------
Content-Type: text/plain
Content-Transfer-Encoding: 8bit

Marcos skribis:

> On Mon, 31 May 1999 14:25:03 +0100, "Raymond A. Brown"
> <raybrown@...> skribis:
>
> >>It has not being devised to author or to edit texts directly on it.
> >
> >So I should hope - that'd would IMO only add a further _complication_ to
> >the whole translation process.
<snip>

> The point of the entire UTL project, Ray, is that one makes the input
> directly on UTL, so one gives the computer an easy tractable text
> which it can translate well to NLs. This is very different of the
> traditional approach of inserting an interlingua between source and
> target natural language.
So in other words, a human manually translates from the source language to UTL,
and software translates that into various target languages?

That might work fairly well a lot of the time, though the thread about how
desirable it is to eliminate ambiguity has made good points & I reckon it would
be more useful for business/legal translation than other kinds.
But I don't see how this is really:

> very different of the
> traditional approach of inserting an interlingua between source and
> target natural language.
My understanding of the "traditional approach" is:

software translates source to interlanguage, then interlanguage to target language

Yours seems to be:

human translates source language to UTL, software translates UTL to target language





-------18881-----Avana-Web-Mail--------