> As for your aside, how so? CLOS multimethods do not "belong" to an object or even to a class declaration. Particular implementations of generic methods are d

> As for your aside, how so? CLOS multimethods do not "belong" to an object or even to a class declaration. Particular implementations of generic methods are declared globally, just like Haskell type class instances. Although, as Peaker noted, type classes can dispatch on any part of the type signature. It is impossible to make a CLOS multimethod with signature: (SomeClass a b) => String -> (a, b) > Perhaps a discussion we can have over email if you're interested. Sorry, I never check email. But I am almost always on Freenode. My nick is pyon. > You sound smart. Not really. The regulars in #haskell - now they are frigging smart.