Re: Vallian (was: How to minimize "words")

On 2/26/07, Henrik Theiling <theiling@...> wrote:
> Or Philip had thought of the canonical ordering algorithm that assigns
> to each diacritic a value by which the diacritics must be sorted to
> make a Unicode sequence canonical.
>
> However, these numbers are for diacritic 'attachment points' that do
> *not* interfere.  Those that attach at the same point have the same
> sort value and the order is significant (for stacking) and must not be
> changed by the sorting algorithm.
That's the one I was half-remembering. Thanks for the correction.

(It did seem odd to me that it was "not possible" to distinguish
between a-diaeresis-tilde and a-tilde-diaeresis. That it's not
possible to distinguish [since the two sequences are equivalent when
normalised] between a-acute-dotbelow and a-dotbelow-acute, on the
other hand, is not a problem.)

Cheers,
--
Philip Newton <philip.newton@...>