Question: Does it make sense to be able to name characters by their
officiel Unicode name, e.g. #<mathematical left angle bracket> instead
of #\u27e8?

I really would like to be able to name characters---and the best thing is
that the names are well documented in 'UnicodeData.txt'---but on the other
hand storing the names is a burdening the implementations.

Anyone knows how much the name mapping can be compressed?

Sebastian.