> ü seems likely to be the UTF-8 value for □.
> Maybe it's being converted to UTF-8 twice (I've had that), or maybe
> the browser is set to use a specific character set.

This can happen if you are using one of the third party converters that
spits out utf-8, but you have specified a different encoding such as
iso-8859-1. Or perhaps you need to explicitly tell it that it's utf-8,
although I think this seems unlikely.

One problem with passing html entities through is that you can
no longer search for words containing them... eg:
xxx□zzz vs xxxüzzz
Users will type the first into the search box...

Having said that, if you are sure you don't want to convert HTML
entities such as é etc, edit perllib/plugins/HTMLPlug.pm
and comment out (by inserting a '#' at the start of the line)
the line near the bottom that says:
$$textref =~ s/&([^;]+);/&ghtml::getcharequiv($1,1)/gseo;

Incidentally, this function converts entities to their utf-8
code.

Hope this helps
John McPherson