KW>3. That said, any proposal to encode another asterisk must
address the cost-benefit issues involved in further
disunification of the asterisk. Besides all the compatibility
forms and dingbats, there basically are two asterisks in
Unicode already: the normal, ambiguous one at U+002A, and the
mathematical operator at U+2217.
KW>If another asterisk were to be added, how would it be
distinguished from these two that already exist? How would a
user know which to enter for what circumstances? And how would
software deal with the functional overlap with usage that
already involves the existing characters?
I think Ken has hit on key issues here and asked the right
questions. The distinction, as I understand the proposal, is
that the new character REGULAR ASTERISK (I think REGULAR
EXPRESSION ASTERISK would be better), would be intended to
represent a specific mathematical operation, whereas U+2217 can
represent other mathematical operations. How would software
deal with this? How would users deal with this? A lot of users
would probably ignore it, but there are probably a small number
interesting in symbolic computation that really want a separate
character to mean REGULAR EXPRESSION ASTERISK. Most likely, the
only software that would know to do anything special with the
character would be software developed and used by that group of
people.
I can think of questions that might be asked at this point:
Q: "If there is a limited group of people working on specific
research areas with a special character need, but a character
with that function is not needed by a more general community,
could they not get together and assign a PUA character for that
need?"
Possible objection to the inferred answer: "The people who
would use this are not in a close working relationship whereby
they can cooperate on PUA allocations."
Objection to the objection: "If there isn't enough interaction
for such cooperation, then why do they need such a character to
be added to an international standard? Each can makes their own
PUA allocation if they need a character with this meaning."
Objection to the objection to the objection: "Oh, but they'd
want to use the same software. The group isn't really *that*
limited, by the way. And, after all, there is work underway to
add a bunch of math stuff; why not add this too."
I don't know where that dialogue ends up. I think it's clear,
though, that this character would be used by particular people
when they're doing particular work (and probably using
particular software), but otherwise ignored: most of use would
continue to write things like 0*1 using U+002A when talking
about regular expressions because it's right there on my
keyboard, and the only processing I expect to be done on that
string will be done by human brains that will figure out the
meaning from the context. (Don't take this to mean I'm opposed
to this; I'm of no mind one way or the other. I'm just trying
to help sift arguments.)
Peter
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:47 EDT