At 02:01 24/10/1999 -0700, Otfried Cheong wrote:
>
>
> > At 00:22 24/10/1999 -0700, Otfried Cheong wrote:
> > >Long-s versus ordinary s is exactly the same distinction as KAF versus
> > >FINAL KAF, TSADI versus FINAL TSADI, etc. in Hebrew, yet those do not
> > >have a compatibility decomposition. Why?
> >
> > 1. Compatibility with existing standards, such as 8859-8.
> >
> > 2. Compatibility with a huge installed base.
> >
> > 3. It is not exactly the same, since in Hebrew there are many
exceptions to
> > the rule and it is quite complicated or even not possible to decide
> > algorithmically which letter to use.
>
>I don't see how that differs from the distinction between round-s and
>long-s in German. You would have to do a pretty clever morphological
>analysis to decide algorithmically.
Well, in Hebrew it would require a dictionary and the dictionary would have to
be maintained.
>Is there a minimal pair in Hebrew that shows that KAF/FINAL KAF are
>different letters?
What do you mean? No one denies that they are different shapes of the same
letter, but we say that you have to encode them at source because the shape
cannot be determined algorithmically in a practical way.
>
>Otfried
>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:54 EDT