On 2/9/25 6:32 AM, Charlie Gibbs wrote:
On 2025-02-09, rbowman <bowman@montana.com> wrote:
On Sat, 8 Feb 2025 23:36:12 -0500, WokieSux282@ud0s4.net wrote:
>
Maybe the best cross-over was Professor Wirth with Pascal/Modula.
Alas it was a shorter fence to straddle back then ....
>
Before it was adapted for the real world Pascal was known as a language
that excelled at telling itself secrets. I/O was an afterthought. Lisp
needed some improvement before it was useful too.
That was true of all the Wirthian languages, wasn't it?
I've heard them referred to as "bondage and discipline" languages.
Early ALGOL didn't even have I/O per-se. It was
an educational demo, not real-world. The last
versions were much more usable.
Pascal/Modula ... kinda "improved ALGOL".
As for "bondage" ... not sure where you heard
that. Pascal - esp the later Borland-derived
versions - are a great all-purpose environment
and you can actually READ yer code a year later.
Modula-2/3 were a little more 'stiff', but still
usable. Can't find a native M3 compiler for Linux
that will install correctly alas ...
If you want B&D, look up little Miss ADA :-)
I fell in love with assembly language at first sight. It was so nice
to have a machine that would just do what I wanted in a couple of
instructions, rather than jumping through all the hoops you had to
do to coax a high-level language to do it. (Yes, you could also
shoot yourself in the foot, but that's part of the learning process.)
Did plenty of ASM early on. Now compilers can be
fairly cheap and cross to almost everything. Didn't
used to be that way. For micro-controllers you were
kinda stuck with ASM.
Didactic computer languages tend to emphasize concepts over utility. I
spent a winter a long time ago working my way through the Wizard book
which used Scheme. I found the concepts interesting but the back of my
mind kept saying 'Why would you ever do it this way?'
Exactly ...
There were (kinda still are) a lot of 'concept' languages
with esoteric syntax and ways of looking at things. Can
be interesting to think about, but you don't wanna use
them for anything real-world.
Maybe they can re-do the entire IRS database in BrainFuck ?
THAT might frustrate Vlad and Xi :-)
I once had a term project reviewed by the department head, who happened
to be one of the Algol 68 development teams. Every so often he would
pause and, in pain, say, "Why did you do it in assembly language?"
Ten years later, along came C, which was a godsend.
I remember 'C' when it was new ... still have my K&R
book ... and you could go almost (now more) tight
with 'C' than ASM. It made everything better.
But for larger stuff, I still trend towards Pascal
when possible. I mean you can DO sockets in Pascal
but it's a lot easier/smaller in 'C'.
Someone in the groups was complaining the other day
about a prob even with 'C' these days - they kinda
changed the rules/defs/syntax over the years. This
made large volumes of his old code into a gigantic
pain. There ARE compiler switches and such to kinda
bring things back to K&R assumptions but IMHO that
should be the default all along. Don't remember if
he was using GCC ... likely ... however there are a
number of good 'C' compilers for Linux now so maybe
he'd have more luck with them. Sometimes stuff that
spits out many errors in GCC compiles fine with
Clang/LLVM. JetBrains has a fair 'C' compiler+IDE
now as well but I don't think it's entirely free.
There's always TCC as well if you don't need
esoteric libs. Visual ... um ... pref to stay away
from M$ .......
Keep promising to learn 'D' ... never quite get
around to it :-)