I think Dyalog's recent work shows that symbols can
be used quite well in Unicode. Here's a cut/paste
from Firefox to Evolution from a recent complang.apl posting on
Benkard's Distributed Round function.
red←{⍺{⍵{⍺+(×⍺)×(⍳⍴⍺)∊(|⍵)⍴⍒|⍺}⍺-+/⍵}⍵+⌊0.5+⍵×(⍺-+/⍵)÷+/⍵}
Note that we're coming up on two decades of the
early design decisions on J, and it's only now
that Unicode is starting to be usable in practice, so
I don't think it's fair to pick on Ken, Roger, or
any of the rest of us who pushed the adoption of
ASCII as a way forward.
If you really like APL (or other) symbols, it should be
a relatively simple matter to write a "skin" for J that will
map between your favorite symbols and J symbols.
Now, if I could just find the correct keys on the keyboard...
Bob
ps: I, too, find APL easier to read than J, but I'm biased.
On Sat, 2007-12-15 at 21:25 +0000, James C Field wrote:
> I deeply mourn the passing (well almost) of the APL character set. Such
> a well-thought-out set of symbols coupled [tripled?] with an equally
> well-thought-out language and award-winning brilliant mainframe
> timesharing implementations: IBM, Burroughs, et al. should have some
> kind of future.
>
> It seems the beautiful APL symbols (and consequently the language) lost
> out to a short-lived technical limitation. The language-hard-core guys
> seemed to accept the failure of the character set and go for a
> 'quick-fix' which limited itself to the ancient ASCII character set.
> ASCII is(was) limited, ambiguous [ANSI, ?] and offers no room for expansion.
>
> Specialist keyboards and mappings can be bought for a song these days.
>
> For all of J's powerful syntactic innovations and improved
> programability IT IS NOT PRETTY! J looks like a dog's breakfast.
>
> Jim Field
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
