Re: reinventing ASCII?

wildhalcyon <wild.halcyon@[EMAIL PROTECTED]
> writes:

> Hypothetically, if you could reinvent ASCII in a modern context, what
> would you remove or add to the specification?
>
> Unicode looks like its set to take ASCII's place somewhere down the
> line, so this is mostly a thought problem, but I'm curious about what
> changes others might implement. I think the first logical change would
> be implementing it as 8-bit, rather than 7-bit, to take advantage of a
> full byte. After that, maybe remove most of the non-printable
> characters. I don't know any modern programs that take advantage of
> most of them. What to fill the additional spaces with though, I don't
> know.

I agree that most of the non-printable character should go, but even
more im****tant would be to standardise how newlines work.  Different
systems use different combibations of CR and LF characters to denote
newlines, and this is a mess.

I would add diacritical marks (accents, umlauts, cedille, ...)  as
prefix characters that would combine with the following character, so,
for example, å would be repersented as °a.  If the following character
is a space, the diacritical mark would appear as itself.  Underline
could be a diacritical mark, so you can have underlined text.

This does play havoc with lengths of strings, as the number of shown
characters is not the same as the number of bytes in the string, but
you do get a lot of characters without having to have separate ASCII
codes for all possible combinations of diacritical marks and letters.

On my old mechanical typewriter, diacritical marks did not advance the
carriage, and bitmapped screens and printers can easily combine two
bitmapped characters, so it would not be a problem for such systems.
More advanced systems could have special characters for common
combinations, so accents could be placed differently on different
letters.

        Torben