Patrick Andries asked:
> The Unicode Standard 3.0 (page 150) says that "U+2011 NON-BREAKING HYPHEN is
> present for compatibility with existing standards" as if it shouldn't really
> be encoded. But isn't its relation to U+2010, the same as the one that
> opposes SPACE to NO-BREAK SPACE, i.e. a semantic (behavioural) one ?
It went in, initially at least, for compatibility with XCCS (Xerox Character
Code Standard):
Unicode XCCS
U+002D HYPHEN-MINUS 000/055 Neutral dash
U+00AD SOFT HYPHEN 357/043 Discretionary hyphen
U+2010 HYPHEN 041/076 Hyphen
U+2011 NON-BREAKING HYPHEN 357/042 Nonbreaking hyphen
"Compatibility" in this sense doesn't necessarily mean "shouldn't have
been encoded".
In fact, in this particular case, if I recall, the distinctions were
probably considered to be good practice, and not something to be mapped
away. XCCS was often a *model* for early Unicode, rather than a character
encoding that forced the grudging inclusion of many icky "characters"
that we would have preferred didn't have to be there.
Keep in mind that U+00A0 NO-BREAK SPACE is *also* a compatibility
character -- for compatibility with ISO 8859-1, among other character
sets.
--Ken
>
> Patrick Andries
> Saint-Hubert (Québec)
