🚀 go-pugleaf

SDENARO@vm.poly.edu writes:
>>#  I will start him on Pascal in a year or two.
>The only options I have seen are Pascal and BASIC. [...]

I think Ken's point was that Logo itself is a perfectly good vehicle for
advanced study.  We Logo people tend to go berserk when someone talks
about what a great introduction Logo is, and then just assumes that the
goal is to move on to something else.  People who learn Pascal first
don't seem to feel that way about Pascal!

I want to make two different points about this:

1.  Aside from the particular languages involved, I think many kids have
the idea that learning computer science means learning a dozen programming
languages.  You can, indeed, get some valuable insights from studying how
and why various languages differ, but I think that for a beginner it's
better to get some depth in one language; beginners aren't ready for a
comparative languages study and instead often learn a lot of details
superficially.

2.  The same qualities that make Logo both easy and powerful for a beginner
also make it both easy and powerful for a more advanced learner.  These
qualities include automatic garbage collection, list processing, freedom
from type declarations and array sizing, a uniform evaluation model (everything
is a procedure), and dynamic scope.  As a pedagogic tool I wrote a Pascal
compiler in Logo (it's in the third volume of my _Computer Science Logo Style_)
and it was enormously easier than the reverse would be.  I'd hesitate to write
a Logo interpreter in Pascal under any circumstances, and I certainly wouldn't
expect to use the resulting program as an object of study for children.

As for "Pascal bashing," the point isn't really how good or bad Pascal is
on some absolute scale, but rather that Pascal represents a coherent point
of view about the nature of computer programming in general that's really
strongly opposed to that represented by Logo.  When you speak of moving
from Logo to Pascal as if that were a simple progression, rather than a
dramatic shift of perspective, it's disconcerting.

(Should I elaborate on that?  If you have access to _The Computing Teacher_
you can read my article "Symbolic Programming vs. the A.P. Curriculum" in
vol 18 no 5, Feb 1991.  Or I could post it or mail it if you like.)