W3C home > Mailing lists > Public > www-international@w3.org > January to March 1997

Re: Web browsers and the new language code for Hebrew

From: Gadi Doron <Gadid@accentsoft.com>
Date: Tue, 11 Feb 1997 09:47:10 +0200
To: Chris.Lilley@sophia.inria.fr, koen@win.tue.nl
CC: misha.wolf@reuters.com, www-international@w3.org, unicode@unicode.org
Message-ID: <83799D7A62@pmail.accentsoft.com>
> Date:          Mon, 10 Feb 1997 21:27:34 +0100 (MET)
> From:          Chris Lilley <Chris.Lilley@sophia.inria.fr>
> To:            koen@win.tue.nl (Koen Holtman),
>                Chris.Lilley@sophia.inria.fr (Chris Lilley)
> Subject:       Re: Web browsers and the new language code for Hebrew
> Cc:            misha.wolf@reuters.com, www-international@w3.org, unicode@unicode.org

> On Feb 10,  9:05pm, Koen Holtman wrote:
> > Why does these browsers need to know the language to render the script?
> >
> > (Martin J. Duerst suggested text-to-speech conversion as one option.  Does
> > that apply here?)
> Assuming the browsers in question are visually based, no. There are cases
> where knowing the the language is essential for visual rendering, for
> example for selecting an appropriate font (eg CJK), or desirable, eg
> selecting a hyphenation dictionary, etc.
> So in this instance I suspect that the language information is not
> essential. That does not, of course, help in deciding what value to
> put in the Content-Language header when returning the document.

The language marks are also useful for translation engines. For
example, the Hebrew character set can be used for either Hebrew or
Yiddish text (and Arabic script can be used for Arabic or Farsi).
Therefore, in order to translate text marked as Hebrew charset,
you'll need first to know it's language.

Gadi Doron
Accent Software International
Received on Tuesday, 11 February 1997 02:49:06 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 21 September 2016 22:37:16 UTC