People in racket are using typep racket, clojure has annotation, in lisp you can declare types and sbcl infers them. Haskell also is fond of types. Perhaps in e

People in racket are using typep racket, clojure has annotation, in lisp you can declare types and sbcl infers them. Haskell also is fond of types. Perhaps in erlang types are not so important because you try to decompose the problem in small parts and hence is not so important to declare types?