Re: Transliteration from Jonathan Rosenne on 1997-11-20 (www-international@w3.org from October to December 1997)

From: Jonathan Rosenne <rosenne@NetVision.net.il>
Date: Thu, 20 Nov 1997 07:12:56 +0200
To: "Martin J. Dürst" <mduerst@ifi.unizh.ch>
Cc: Carrasco Benitez Manuel <manuel.carrasco@emea.eudra.org>, www-international@w3.org
Message-Id: <3.0.1.32.19971120071256.006ab088@mail.netvision.net.il>

At 21:25 19/11/97 +0100, Martin J. Dürst wrote:
>- Will they be produced on the server side, or on the browser side?
>
>- Will they be short-lived (produced on the fly) or long-term?

I don't believe that in the general case transliteration can be produced by
machines, except by means of a huge table lookup or very complex analysis.

This is true for English and Hebrew.

It is especially true when one considers that one of the prime targets for
transliteration is names. Such as Gloucester.

Jonathan

Received on Thursday, 20 November 1997 00:13:13 UTC