News for hackers. Stuff that matters.
Re:Statically Typed Lisp by tempodox on Thursday November 13, 2014 @03:35AM (#8600507)
This is, of course, correct and I do use that technique in SBCL myself, but it still doesn't make SBCL a statically typed language (even if SBCL has the strictest type checks and the best native code generation I've seen in an open-source Lisp to date). The static typing in OCaml, Haskell - heck, even C is still another thing.
Re:Statically Typed Lisp by xenophonf on Thursday November 13, 2014 @11:21AM (#8602322)
I don't know OCaml or Haskell, but I think you're mistaken about C. Given the following C code:

  char *foo(char *bar) {
    return bar;
  }
  
  int main(int argc, char *argv[]) {
    foo(1);
    return 0;
  }
I get compile-time warnings from clang substantially similar to what SBCL emits:

  $ cc test.c 
  test.c:6:7: warning: incompatible integer to pointer conversion passing 'int' to parameter of type 'char *' [-Wint-conversion]
    foo(1);
        ^
  test.c:1:17: note: passing argument to parameter 'bar' here
  char *foo(char *bar) {
                  ^
  1 warning generated.
I'm sure that with suitable flags set, both compilers would throw type errors at compile-time instead of warnings.