Kenneth Whistler wrote:
>
> Brendan suggested:
>
> > Antoine Leca <Antoine.Leca@renault.fr> wrote:
> > > What about the pairs
> > > 0041;LATIN CAPITAL LETTER A
> > <snip>
> > > (which will be the one chosen by almost any software for this use).
> >
> > This is a little too simplistic: these characters have specific numeric
> > properties and behaviors separate from those of A and B.
Good point.
> > The hexadecimal-ness of A through F is not encoded in Unicode, but these
> > characters are commonly used as hex digits *and* in Latin-based scripts.
> > However, these extra two characters are clearly not interchangeable with A
> > and B, so I don't believe that using these is a complete solution.
I fail to see why we could not do a parallel.
> From the point of view of character encoding, I don't think it would
> be correct to substitute out either the A/B usage (from hexadecimal
> implementations)
Just for the record: although the most widely known usage is definitively
hexadecimal notation using A through F, the letters A to Z (or a to z,
depending of variying factors) for the "digits" above 10 in base 11 to 36
are in wide use in the 80's programming languages, including in the
ISO standards for Ada (the base#...# notation) or C (the strtol function,
and the non-Standard itoa/ultoa functions).
> or the Roman numeral characters, although it is
> easy to see how one could implement dozenal numerical formatting
> with either (or for that matter, using "t" for ten and "e" for
> eleven: 1 2 3 4 5 6 7 8 9 t e 10 11 ... 19 1t 1e 20 ... etc. or
> any other arbitrary choice of characters.
Agreed, except that the last two "digits" should be "d" for 10,
and either "u" or "o" for 11!
> But if the Dozenal Society of Great Britain feels that its particular
> usage of characters is important, then they (or anyone wishing to
> sponsor their usage) could submit the appropriate Summary Proposal
> Form for encoding characters to UTC and WG2 for consideration.
If it is the way to go, I believe digits 12 to 15 are to be included
as well, due to their usage at least as wide as 10 and 11.
Antoine
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:05 EDT