[idn] usability of Unicode

["Adam M. Costello" <idn.amc+0@nicemice.net.RemoveThisWord>]

Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp> wrote:

> Unicode is not usable in international context.

I think you mean that Unicode doesn't work very well for displaying
free-form text, unless you have more meta-information about the text,
and you may be right.

But for text used as identifiers, Unicode is extremely useful, for
testing whether two identifiers are the same.

Suppose there are 100 charsets in use around the world.  Many of them
have characters in common, so two identifiers that use different
charsets might still contain the same characters in the same order.
To test for that, do we write 10,000 comparison algorithms to handle
all possible combinations?  Of course not.  We write 100 transcoding
algorithms to convert each charset to Unicode, plus one normalization
algorithm for Unicode.

AMC