From: Jim Allan (jallan@smrtytrek.com)
Date: Sat Jan 03 2004 - 12:30:13 EST
D. Starner posted:
> Not safe for what? I've come across six characters that weren't in
> Unicode at all. Does this mean that Unicode isn't safe to use?
Not safe to *assume* that because a character appears in the pan-Turkic
alphabet and is not known to you (or me) from other earlier
orthographies, that it was first invented for the pan-Turkic alphabet.
Similarly for any other orthographical system. If you *know* when that
character was devised, then you are not assuming.
> True. But this is not a closed system, where the only characters usable
> will be those in the panel. If we ever get texts with the letters, we
> can deal with it.
Then I'm puzzled as to the purpose of this proposed subset.
Books before 1923, especially scholarly books concerned with language or
mathematics and logic, might contain almost any character currently
coded in Unicode as well as characters not currently coded in Unicode
including idiosyncratic characters that will never be encoded in
Unicode. They are also likely to contain characters from non-Latin
scripts and many symbols.
It would take an enormous amount of research to *prove* for many
characters that they (and characters with which it would be reasonable
to unify them) were *never* used before 1923 in *any* published work,
attempting to prove a negative.
Why prescribe a closed subset?
Jim Allan
This archive was generated by hypermail 2.1.5 : Sat Jan 03 2004 - 13:05:00 EST