RE: XHTML validator doesn't completely support Unicode

 
> Are you sure it's the end of line characters that give the problem?
> 
> I'd guess it's the BOM ("U+FEFF") that's the culprit. It's 
> not very common to use a BOM in UTF-8 files. Some even say 
> it's not allowed in UTF-8. It's certainly not necessary to 
> use a BOM in UTF-8.

Actually, I do find it necessary, and it exists for a very good reason.

Without it at the beginning of a file, an editor has no sure way of
determining what format the text file is. In practical testing with SC
UniPad, I have found that the use of either the BOM or EOL character is
the only sure way for the editor to open the file as UTF-8. Their
absence causes the editor to assume plain ASCII-UCN, which causes real
problems if you don't notice and correct it.