pyotr filipivich wrote:
"Bill in Co" surly_curmudgeon@earthlink on Sat, 16 Mar 2019 17:12:50
-0600 typed in alt.windows7.general the following:
pyotr filipivich wrote:
"Bill in Co" surly_curmudgeon@earthlink on Sat, 16 Mar 2019 14:01:35
-0600 typed in alt.windows7.general the following:

As we also said about C, it wasn't so much "released" as
"escaped". Same was said about RPG (Report Program Generator) from
IBM.

By the way, and just for thr record, Turbo Pascal does indeed still
compile just fine on a 32 bit Windows XP computer using the DOS cmd
shell. Interestingly enough, apparently some versions of Turbo Pascal
have been releassed into the public domain, and can be found on the
Internet. So maybe the OP can port his source code over to a newer
version of Pascal, although Turbo Pascal took some liberties in its own
version of Pascal that may need to be addressed

One difference I caught "back in the day": Pascal on the
Honeywell, exited a loop and left the value in the counter. In Turbo
Pascal, that didn't happen, which made the next set of calculations
really, really wrong. Orders of magnitude wrong.

I can only imagine! Sometimes I'd forget about uninitialized variables
too, and boy did that bite me, on occasion.. I think with Basic they were
normally initially zeroed out.

I think this was along the lines of at the end of the procedure
Variable A had a value N, which it retained when procedure exited on
the Honeywell, and was used in the next calculation. But in Turbo. In
Turbo, Variable A after the return had Value (Large "Random" number).
Subsequent calcs were way out of line.

I thought if you passed a var parameter as the argument in a procedure call,
it was retained, but not a value parameter (in the procedure calls). But
this is taxing my memory from two decades ago.