> Can anyone see any issues with using a "zero wdith no-break space" 
> as a way of preventing tag minimization:

some, if the element takes eleemnt content then putting a character ther
emakes it invalid (though that may not matter) if it does take mixed
content then of course it depends if the application recognises #65279;
for special behaviour, or if the font being used has an invisible glyph
for this.


> (#8203) doesn't seem to prevent the minimisation...
that would be a bug.

> but I would expect that if the browser can display XHTML then it must
> support unicode?

The only requirement is not to fall over while parsing text, there are a
lot of characters in unicode, and typically on most real systems most of
those characters render as a box or ? or some other missing glyph
marker.

This is often an issue in mathml (taking a random example) where you
sometimes face the choice of doing the "right thing" and marking up 2x
as 2 invisible-times x or just abitting the 2 and the x. If the
invisible times character (#x2062) renders as a missing glyph, there is
a temptation to miss it out...

David





________________________________________________________________________
The Numerical Algorithms Group Ltd is a company registered in England
and Wales with company number 1249803. The registered office is:
Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.

This e-mail has been scanned for all viruses by Star. The service is
powered by MessageLabs. 
________________________________________________________________________