New Python implementation

[dn]

On 16/02/2021 17.57, Dan Stromberg wrote:
> On Mon, Feb 15, 2021 at 8:52 PM Igor Korot <ikorot01 at gmail.com> wrote:
> 
>> Hi, guys,
>> Let me try to throw in another one - PL/1.
>> This guys used to be very popular with the accounting community...
>>
> 
> Actually PL/I is basically proprietary Pascal - from IBM.  My Intro Comp
> Sci classes at the University of Cincinnati were in PL/I, because IBM had
> an office not far from the University, and they liked to hire interns from
> UCinci.  I was told that otherwise the classes would've been in Pascal.


Might a coincidence of location have been conflated with language
development?


PL/I (note the Roman "one") was developed out of the IBM System/360
effort, ie one computer architecture for all types of users (cf
different computers for 'commercial' and 'scientific' clients - which
you can see again today in AWS' range of compute options, for example).
It was to be 'one language to rule them all'.

Initially, it was based on FORTRAN. We joked that it took all of the bad
parts of FORTRAN and COBOL and mashed them into a single ... language.
However, like many such efforts, it attempted to take 'the best' from
many sources (including ALGOL).

Sadly, (maybe) it, like Ada (another attempt to be 'all things to all
men') only ever really existed in a niche market.


Pascal, named after the mathematician and philosopher the French credit
with 'inventing the computer', was developed in the European world -
which even back-then had marked differences in thinking to the American
approach/domination of computing.

Prof Niklaus Wirth published a seminal book "Algorithms + Data
Structures = Programs" (which I seem to have lost, somewhere in the
mists of time). I don't think it gave examples in Pascal, but it did use
a pseudo-code/-language which illuminated various ideas and ways of
developing programmatic solutions. As I recall, it (and Pascal) was very
ALGOL like, and thus heavily influenced by "stack architecture" (cf
IBM's collections of "registers" in an ALU). Certainly the language was
clearly based upon ALGOL-60. Clearly it led to the refinement and/or
development of Pascal.

Pascal's value as a teaching language was that it embodied many aspects
of structured programming, and like Python, consisted of a limited range
of items which could be learned very quickly (in contrast to PL/I's many
'bells and whistles'). Thus, once absorbed, learning attention could be
directed to ComSc/algorithms + data!

Pascal was also very efficient, to compile and to execute, which made it
a competent fore-runner of Micro-Python (etc) in the micro and
single-board computer arenae/arenas.

Without Python, I think I'd prefer to use (an updated) UCSD-Pascal or
Borland Turbo-Pascal to this very day! PL/I, not so much - even on a
mainframe project!
-- 
Regards,
=dn