Re: MES-1 deficient

From: Markus Kuhn (Markus.Kuhn@cl.cam.ac.uk)
Date: Thu Mar 09 2000 - 13:04:02 EST


"Brendan Murray/DUB/Lotus" wrote on 2000-03-09 16:44 UTC:
> The reality is that it's a lot cheaper, easier, and enhances compatibility
> enormously if one totally disregards this report and implements full
> Unicode support.

Sorry, but this is nonsense. It would cost enormous amounts of money if
European procurers would specify for each and every of their systems
that Hangul Jamo, Arabic, and Devanagari is supported. These scripts
(all part of "full Unicode") require quite sophisticated processing
mechanisms such as bidi, ligature substitution, etc. that will never be
needed in most European systems, because none of the people using them
can be expected to be familiar with the respective scripts. On the other
hand, there are subsets of Unicode that require not much more to be
changed then the switch from ISO 8859 to one of the UTFs and somewhat
larger fonts, and these things are indeed *very* easy to implement even
on existing legacy systems (we have just done that for X11's xterm for
instance). The 11 million NLG is political nonsense of course.

Remember that not all IT system run on GUI/OS combinations with OpenType
support. Remember embedded applications with LCDs and microcontrollers,
etc.

There is certainly a need for well-defined and easily quotable Unicode
subsets, and the MES was one attempt to generate a family of these. I am
not claiming that the result was particularly useful, especially
considering the huge amount of silly politics that played a great role
in this project.

Other national standards bodies have already defined their respective
Unicode subrepertoires, e.g. JIS X 0221-1995 for Japan and GB 13000.1-93
for China do exactly what MES does for Europe.

Markus

-- 
Markus G. Kuhn, Computer Laboratory, University of Cambridge, UK
Email: mkuhn at acm.org,  WWW: <http://www.cl.cam.ac.uk/~mgk25/>



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:59 EDT