> So what is the accepted way to add character references to

> the output? Would I have to run some kind of find-and-replace script > 
after the XSLT transformation? What do other people do?



You shouldn't care whether characters in the output are represented as
themselves or as character references. No process that consumes the output
XML or HTML is going to make a distinction, so it doesn't matter. Leave it
to the serializer to decide. If you want to look at the output in a text
editor, choose an encoding that your text editor can handle, and the
serializer will do the rest.

Michael Kay
http://www.saxonica.com/

If that were true, it would be fine with me. I am not trying to output 
character references just for the sake of doing it. I was under the 
impression that I should try to do it because some browsers would have a 
problem with UFT-8 encoded data. Specifically, articles such as this one at 
A List Apart:



http://www.alistapart.com/articles/emen/



claim that "many more browsers choke on UTF-8 characters than do on named 
entities." Yet you claim that there is no distinction?



Is this true or isn't it?