At 17:01 -0700 6/30/1999, Tony Harminc wrote:
>On 30 Jun 99, at 12:35, Asmus Freytag wrote:
>
>> In the case of ASTERISK, the analysis that is needed, and that, as far as
>> I have seen, has not been made, is to present evidence that cases exist
>> (or are easily conceivable) where *both* the ASCII asterisk and yet
>> another asterisk are needed in the same text, and with consistent
>> distinction in use or formatting.
>
>My first thought was APL. I was then surprised to discover that
>there doesn't seem to be an APL asterisk, although there are
>composite APL characters such as "Apl Functional Symbol Circle Star",
>which oddly enough are not listed as compositions. (Oddly, at least
>because historically these characters were keyed by the user
>explicitly typing one base character, a backspace, and the other
>base.)
Many APLers felt that the APL star should have been given its own code
point, but there wasn't a formal character code standard for APL when
Unicode was taking shape, so it didn't meet the requirement for a
preexisting standard requiring guaranteed round-trip conversion. The
current APL standard defines a minimum set of characters but not their
encoding. IBM is still the biggest APL vendor, and still uses an APL
overlay on EBCDIC for the 94 base printing characters in APL, while other
APL vendors use APL overlays on ASCII. (The overlay idea originally meant
that you overload the codes and change your Selectric typeball, which holds
88 printing glyphs.) Besides these 94 base characters, there are numerous
APL characters composed using overstrikes, and no two APL vendors agree on
which composed characters should be included, or how they should be encoded.
I wrote an article in APL News about the Unicode-APL mappings in the
context of IBM's announcement of its own particular version of Unicode
support in APL2, including a new mapping between the APL2 character set and
Unicode. It had been agreed in the APL standard committees (ANSI, ISO, and
others) that some APL characters should be unified either with ASCII or
math characters of similar appearance, but nobody seemed to know which
ones. My article included a table of characters that had more than one
possible mapping, a table of APL characters in page 22 of Unicode, and
other data .
If anyone wants to see the article, I can dig out a hard copy, but I won't
attempt to post an electronic version. I probably still have the file, but
I don't have software that can read it. :-[
>On 30 Jun 99, at 10:56, Kenneth Whistler wrote
>
>> And the function is obviously quite variable in plain text:
>> Multiplication: 2*4=8 (this is the one addressed by U+2217,
>> clearly)
>> Exponential: 2**10 (varies with 2^10, and shown with formal
>> superscripting
>> in properly typeset mathematics)
>
>And in APL, 2*4=16, i.e. the * indicates exponentiation. But surely
>the * in 2*4=8 was not known to conventional mathematical notation,
>but was introduced in the 1950s by FORTRAN in the absence of a small
>x or suitable dot that could be distinguished from a letter or
>decimal separator respectively.
and meant multiplication, for which APL used a multiplication sign.
Exponentiation was ** in FORTRAN, IIRC.
>So back to the top: I think quoting APL expressions, particularly in
>the context of more conventional mathematical notation, may be a case
>for multiple asterisks. But I know little of the history of
>incorporation of APL characters into Unicode, so I may be way off
>base.
>
>Tony H.
-- Edward Cherlin President Coalition Against Unsolicited Commercial E-mail Help outlaw Spam. <http://www.cauce.org/> Talk to us at <news:comp.org.cauce>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:47 EDT