C9 Lectures: Dr. Erik Meijer - Functional Programming Fundamentals, Chapter 3 of 13

I like the syntactic symmetry of inverse associativity for function declarations and function applications. 

f :: (Int -> (Int -> Int))         f :: Int -> Int -> Int
     ((f 1) 2)                             f 1 2

So the paranthesis' become denser at the left or the right.

The zip function is quite beautiful and metaphoric. Erik's definition

zip :: ([a], [a]) -> [(a, a)]

is super-symmetrical*. But then there is something nagging me (just a little bit) - here currying is suddenly not as pretty anymore because it breaks the symmetry, which is the motivation for Erik's own definition. 3 notions of structure: curried function (function to function to sequence), list (homogeneous unbounded sequence) and tuple (heterogeneous bounded sequence).

I assume curried functions can be treated as sequences as well - flipping and reversing arguments. And lazily (always here) so no time is wasted while jusing doing what I think Erik refers to as symbol pushing. Must see...

The book has not arrived yet, still waiting. That's not unusual, but hopefully it will arrive before the end of the lecture series.

* Sometimes (especially functional) programming language design looks like the search for a unified theory of everything - for computation, types and syntax. Slowly crawling closer and closer to nirvana. Haskell looks like the closest thing so far, although maybe still susceptible to decades of further axiomatization.