"Suzanne M. Topping" wrote:
> In hunting around for negative opinions about Unicode, ...

Let me add one complaint to your list:

    Thai is not stored/used in logical order in Unicode.

I know that many here have fought hard to get every script encoded and used in logical 
order so that processing is fast. They lost.

This comes up in our (ICU) project right now because we are reimplementing our 
collation code, and we have to hardcode Thai reordering there to in effect transform 
the text on the fly to logical order.

In general, I would probably list every deviation from the principles: precomposed 
characters including Hangul, compatibility characters, etc. They all need some special 
code, like normalization, that would otherwise not be necessary.

markus

Reply via email to