> A perfect language should be defined in terms of mathematical logic, of axioms and rules of inference. No reference to any computer or mechanism should be necessary to understand it. This is the basis of portability. Algol's designers saw this goal; but it is most difficult to achieve without sacrificing power of expression. Yet, any new language must be measured on the degree to which it comes close to this goal.
Um, no. Any new language must be measured on the degree to which it helps programmers to say what they need to say. Anything else is about something else than a language being a programming language.
Now, I'm sure that Wirth thinks that his definition implies helping programmers say what they want to say. He thinks one follows from the other. And, to some degree, he's right.
But I don't think he's totally right. Languages that are purely mathematical and logical can be rather clumsy when you're trying to do some things that aren't like mathematical logic. (Even SQL, which is based on the mathematics of relations, has various things in the syntax that are there for performance reasons, because the need was for something that would do the mathematics of relations in an acceptable amount of time. That "time" element is not present in any mathematical-logic-based language that I know of.)
Great teaching language. I wish I'd chosen the other door and done LISP as my introductory programming language back in 79. (It was literally a binary choice)
Never liked pascal systems, pcode, UCSD. I think it was a dead end but I know devs who said it worked well.
Um, no. Any new language must be measured on the degree to which it helps programmers to say what they need to say. Anything else is about something else than a language being a programming language.
Now, I'm sure that Wirth thinks that his definition implies helping programmers say what they want to say. He thinks one follows from the other. And, to some degree, he's right.
But I don't think he's totally right. Languages that are purely mathematical and logical can be rather clumsy when you're trying to do some things that aren't like mathematical logic. (Even SQL, which is based on the mathematics of relations, has various things in the syntax that are there for performance reasons, because the need was for something that would do the mathematics of relations in an acceptable amount of time. That "time" element is not present in any mathematical-logic-based language that I know of.)