On Mon, 10 Jun 2002, John Hinsley wrote:
> The only "but" is that, in my pretty limited experience,  there can be issues
> when using the full range of Central European and Italian characters together.
> Anyone any experience (at any level) of doing something like this?

I have no experience with Swikis, but as far as HTML is concerned, you may
consider (i) named or numeric references to the non-ASCII characters in
the universal character set.  The UCS is the document character set (of
abstract characters allowed in conforming documents) required by HTML 4
and accordingly by XHTML 1.  Drawback:  some browsers may not handle all
relevant character references.  (A version of Emacs W3 even breaks on
numeric references to cyrillic characters.)  If Swikis are used to
substituting ampersands in user input by &s  there will be a problem.
        (ii) Use the Unicode UTF-8 character encoding (which contains
ASCII).  The HTTP server or at least the document would have to tell
clients about it.  (UTF-8 and UTF-16 are the default char encodings for
XML documents, and therefore also for XHTML delivered as text/xml.)
Drawback:  browser support?

rthappe