How much of your Perl code deals with non-European scripts, i.e. beyond the ASCII range? Do you handle such code differently, or are you programming for UTF8 compatibility with everything you do, even if you do not foresee a special need for UTF8 with it?