A long time ago I was able to write on my HP67 a complete program for calculating the compression, tension and bending forces of a continuous beam on supports (

A long time ago I was able to write on my HP67 a complete program for calculating the compression, tension and bending forces of a continuous beam on supports (for example a railway rail on sleepers), applying the equations I learned in my engineering school. In fact it was the first time I understood something about this equation. Until I translated it into code I only knew the music, not the details. That's why, when I explore a code, here the evaluation of an RPN expression, I try to keep the foundations to a minimum. The minimum minimorum is the lambda-calculus. Many code examples use the subtleties of this or that language, with the drawback of hiding the fundamental structure of the algorithm. For these reasons I prefer the "list" version to the "array" version. Basically nothing else than the three fundamental "cons, car, cdr". I am interested in your opinion.