28 Nov
2018
28 Nov
'18
6:27 a.m.
On Nov 27, 2018, at 9:23 PM, Fred Cisin via cctalk
<cctalk at classiccmp.org> wrote:
I learned it about 15 years ago (OpenAPL, running on a Solaris workstation with a modified
Xterm that handled the APL characters). Nice. It made a handy tool for some
cryptanalysis programs I needed to write.
I wonder if current APL implementations use the Unicode characters for APL, that would
make things easy.
> I have long wondered if there are computer
languages that aren't rooted
> in English / ASCII. I feel like it's rather pompous to assume that all
> programming languages are rooted in English / ASCII. I would hope that
> there are programming languages that are more specific to the region of
> the world they were developed in. As such, I would expect that they
> would be stored in something other than ASCII.
On Tue, 27 Nov 2018, William Donzelli via cctalk wrote:
APL.
APL requires adding additional characters. That was a major obstacle to acceptance, both
in terms of keyboard and type ball (my use preceded CRT), but also asking the
user/programmer to learn new characters. I loved APL! I love the use of an arrow for assignment. ...
One of the strangest programming languages I've used is POP-2, which we used in an AI
course (Expert Systems) at the University of Illinois, in 1976. Taught by a visiting prof
from the University of Edinborough, I think Donald Mickie but I may have the name
confused.
Like APL, POP-2 had the same associativity for all operators. Unlike APL, the designers
decided that the majority should win so assignment would be left-associative like
everything else -- rather than APL's rule that all the other operators are
right-associative like assignment. So you'd end up with statements like:
n + 1 -> n
More at https://en.wikipedia.org/wiki/POP-2
paul