> Hmmm. I accept Marco's statement that while it's a single symbol, it
> doesn't necessarily have to be a single character. What is the rule
> for deciding whether a single symbol needs one or more characters?
I don't know if Unicode's UTC has a rule or decides case by case.
Applying common sense, I would say that an important criterion should be the
appearance of the symbol (that's why I asked you for a picture).
Although Unicode does not encode glyphs, if the glyph is visually distinct,
then it's hard to say that a vaguely look-alike sequence is appropriate.
Another criterion could be semantics and character properties. E.g.:
- Should that symbol be usable in an file name or resource locator? If yes,
composing it with ":" (a punctuation character, often forbidden in these
cases) could be inappropriate.
- Should that symbol be recognized as a Norwegian word with a specific
pronunciation? If yes, a sequence that can be confused with something else
can be inappropriate for, e.g., a screen reader application.
> What happens if I find a font that has this as a single character, for
> example?
This is a circular argument: fonts don't contain characters, they contain
glyphs. And each glyph can be mapped to one character or to a sequence or
characters, and this mapping can even be subject to contextual rules.
_ Marco
This archive was generated by hypermail 2.1.2 : Tue Apr 30 2002 - 10:50:38 EDT