(iso15924.85) "Native" derivations for codes

[Manfred Kiefer <eurmxk@drv.cbc.com>]

> 
> Ar 12:25 +0100 1998-01-09, scr=EDobh Manfred Kiefer:
> >
> >I tend to see a stronger interest in the latter part
> >of the definition ("expression of scripts in coded form").
> >These codes might soon be used for identifying scripts
> >in machine-(aided) translation, machine-(aided) transliteration
> >and the like. It is true, in this context it is not of great
> >importance whether numerical codes or alphabetic ones (either
> >"English-related" or "native") are used - I still tend to prefer
> >"English-related" ones - they are most likely easier to
> >memorize for most people, hence they are more useful
> >as far as the first part of the definition is concerned.
> 
> N=EDl s=E9 sin ceart ar chor ar bith. Is g=E1 script agus a hainm f=E9in a a=
> ithint
> chun c=F3d a =FAs=E1id l=E9i. N=ED g=E1 a hainm B=E9arla a aithint leis!
> 
> Das ist aber nicht richtig. Man mu=DF eine Schrift und ihren Name kennen um
> Kod f=FCr sie zu benutzen. Man soll nicht den Name der Schrift in Englisch
> auch kennen m=FCssen!
No, you do not have to know the codes, first of all.
When you look the codes up in a manual or what, you will likely have an index
and a description in your native language. This is like checking the code
for a chemical element. Pb is "lead" or "Blei" or "plomb" or whatever language
you use. Who minds the the names of the codes are based on Latin?
The more often you will use them, the easier you keep them in mind, and in this
case it's just practical to use English-based ones.

> 
> >This sounds
> >like a bias towards English but it really shouldn't. I would regard
> >them as pure codes (which happen to be based on English language terms
> >for practical purposes).
> 
> Basing the codes on ISO 639-2/T will do everything you want and will cause
> you or anyone else no great hardship.
> 
> >There might even be more variants of others, e.g. Coptic as a variant
> >of Greek or (with more right even) Gothic as a variant of Greek also.
> 
> Actually the disunification of Coptic from Greek is proceeding these days.
> I have a proposal for doing this at
> http://www.dkuug.dk/jtc1/sc2/wg2/docs/n1658.htm
It makes sense depending on how you define a script. If there are just variances
in shape the scripts should be unified. In the case of Coptic (with additional
letters taken from a different script, i.e. demotic) it makes sense to disunify
it.

Regards,
Manfred Kiefer 

--
          
=====================================================================
Manfred Kiefer                     SAS Institute GmbH
e-mail : eurmxk@eur.sas.com        European Headquarters
Phone  : (49) 6221 416 0           Neuenheimer Landstr. 28-30
Fax    : (49) 6221 47 48 50        D-69120 Heidelberg, Germany
=====================================================================