Question

In Clojure you can write:

(cond (= 1 2) 1 
      :else 5)

In Scheme the equivalent would be:

(cond ((= 1 2) 1)
       (else 5))

The :else 5 syntax is not as consistent as the (else 5). What is the reason the else syntax is implemented in this seemingly inconsistent way in Clojure?

Answer 1

I think (else 5) is less consistent. (cond ...) arguments are stated as condition – value pairs. :else value is consistent because :else is just a convention – it works because :else is just an expression that’s always true. There’s no special rules for :else at all.