Right now, wine claims to use DUCET data for lingustic sorting, but by
section 1.9.2 of that document (as of version 6.0.0), uses it in a wrong
way. The result of it are bugs such as #10767 and #9583.
A possible way around it would beby using ICU to get language specific
tailoring and applying some of wine-specific for the parts addressed by
#10767.
I will not go into the part of the discussion that has to do with
whether ICU will or will not resolve the issues. I will point out
that Wine used to have a soft dependency on ICU (introduced, for my
sins, by me) for the sake of BiDi processing. Everyone involved,
myself included, were all too happy to see it go. ICU is impossible
to dynamically link with, and it's size is quite huge if statically
linked.