Message-Id: <6.2.1.2.2.20050515024721.03bdfdd0@mail.jefsey.com>
Date: Sun, 15 May 2005 03:07:14 +0200
To: "Chris Jacobs" <chris.jacobs@freeler.nl>
From: "JFC (Jefsey) Morfin" <jefsey@jefsey.com>
Subject: Re: what is Latn?
Cc: unicode@unicode.org
In-Reply-To: <002a01c558b7$bc7b8c00$5d1f79c3@u9t1f6>
References: <a06210203beab9d00690c@[83.250.195.133]>
 <6.2.1.2.2.20050514151551.05846590@mail.jefsey.com>
 <002a01c558b7$bc7b8c00$5d1f79c3@u9t1f6>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: unicode-bounce@unicode.org
Errors-To: unicode-bounce@unicode.org
Precedence: bulk

On 21:02 14/05/2005, Chris Jacobs said:
> > Could someone tell me where to find the list of the characters
> > belonging  to ISO 15924 "Latn" script?
>
>When I first read this question I mistook ISO 15924 for the name of the
>"Latn" script, but Google enlightened me.
>
>http://www.google.nl/search?hl=nl&q=iso+15924&btnG=Google+zoeken&meta=
>http://www.evertype.com/standards/iso15924/
>http://www.unicode.org/iso15924/iso15924-codes.html
>
> > I wish to know if they support all
> > the French characters and variants?
>
>Since Latn is a script, not a charset, you cannot expect an authoritative
>list of what is in it, but there is a list of which unicode chars are Latin,
>where Latin is Latn or Latf (Fraktur variant) or Latg (Gaelic variant).
>
>http://www.unicode.org/reports/tr24/tr24-4.html
>http://www.unicode.org/Public/UNIDATA/Scripts.txt

Dear Chris,
you are right. But the combination of ISO 639, ISO 3166 and ISO 15924 is 
the matter I consider. I see no big problem in piling more or less precise 
or erroneous lists: this hurts no one and permits the main vendors to 
proceed to their world market repartition. Now, the problem is when one 
discusses the possibilty of a default script associated to a language: does 
the script exist? ISO 15924 is a list of script names. I cannot associate a 
name to a language name without some kind of correspondancy rule I can 
check. I think reasonable to say that I can associate the name of a script 
to the name of a language if they share identical charsets when using (all) 
the encoding scheme(s) accepted by the programmers community?
jfc