Omicron was required because it was distinguished from O in at least one of the
source standards for Unicode. Thus there was never any discussion of whether it
would otherwise have been unified or not.
I was never part of the Cyrillic discussions, so if someone would summarized the
reasons that Q and W cannot be correctly processed if used as Kurdish, I would
appreciate it. The only reason off hand that I can think of that someone might
claim that a mixture of Kurdish and (say) English could not be correctly sorted.
But we know that mixtures of any two arbitrary languages, e.g. French and Swedish,
cannot necessarily be correctly sorted without tagging.
Roozbeh Pournader wrote:
> I somehow agree with Michael in this case. The reason for the need is
> exactly the same as the reason behind a greek letter Omicron.
> On Fri, 21 Jan 2000, Michael Everson wrote:
> > >Only if you are a strict scriptist. Otherwise, those are encoded in Latin1,
> > >right?
> > They are NOT encoded in the Universal Character Set, Mark, and we have gone
> > over this again and again with the same arguments. Multiscript unilingual
> > data in Kurdish cannot be correctly processed making Latin Q and W do
> > double duty in two different scripts, the unification was bogus, and we
> > need to add the two characters to the UCS.
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:58 EDT