--- In qalam@yahoogroups.com, "Richard Wordingham"
<richard.wordingham@...> wrote:
> Doesn't look hopeful. Google distinguishes 'role' and 'rĂ´le' in
> English, and Unicode-compliant applications ought to treat composed
> characters and equivalent sequences identically.

When I google in French I don't keyboard the accents and I still get
search results. I naively deduced from that that diacritics could be
discounted in google. Then I was disappointed to find searching the
Cree dictionary so difficult.

My distaste for accents date back to when I lost marks in translation
exercises in classical Greek when I was 14. I don't mind reading
Polytonic Greek but I usually write Greek using only the breathing
marks and the subscript iota.

> In a bespoke application of course, the issue is little different
> from case-insensitive sorting. For DIY applications, dropping the
> pre-aspiration is the most difficult aspect. You might get more of
> an issue for the long-vowel versus diphthong alternation.
> You might like to try contacting Bill Jancewicz or Marguerite
> MacKenzie if you haven't already. From their article 'Applied
> Computer Technology in Cree and Naskapi Language Programs' at
> http://llt.msu.edu/vol6num2/pdf/vol6num2.pdf , it sounds as though
> there's a lot of activity devoted to converting between encoding
> systems.
> Richard.