"james...@alumni.caltech.edu" <jameskuyper@alumni.caltech.edu> writes:

> On Wednesday, April 5, 2023 at 2:21:46?AM UTC-4, Tim Rentsch wrote:
> ...
>
>> not due to inherent limitations in the language. It would be
>> harder, for example, to think and write procedurally in APL.
>
> I'm curious about your basis for making that statement. APL was my
> second computer language, after Fortran I (sic), in the mid 1970's. I
> was hired for my first full-time job in the early 80's because of my
> knowledge of APL (though I made negligible use of that knowledge,
> since I could never extract from my boss a sufficiently detailed
> explanation of what the APL project was supposed to do - I ended up
> spending most of my time there on non-APL projects).
> I haven't used it much since then, but contrary to APL's reputation as
> a write-only computer language, I can still read with understanding
> programs that I wrote at that time. And what I remember about how
> APL works was basically procedural. I'm curious - has APL changed
> significantly from what I learned, or did I not even notice it's non-
> procedural features?

APL is definitely an imperative language.

On the other hand, APL is not very good for writing programs
in the way that they would be written in, for example, C.
Imperative is not the same as procedural. Assembly language,
for example, is certainly an imperative language, but not
really suited to procedural programming.

APL does have procedures, but they are not up to even the
level of procedural programming offered by Pascal. It isn't
that it's impossible to do procedural programming in APL,
but the impedance mismatch makes it feel like swimming in
molasses.

Further explanation in my response to Ben Bacarisse in this
thread.