Comparing Functional Languages

I wrote this blog post about a year ago but finally got a chance to publish it.

How many ::’s does it take to screw in a light-bulb? A car’s worth.

I’m currently in a class where we are using Scheme extensively. This has made me very used to the idea of structuring my algorithms using car and cdr for any sort of data manipulation.

car returns the head of a List while cdr returns the tail of List.

Here is a simple implementation of a sum function to understand how car and cdr works.

(define (sum some-list)
    (if (null? some-list) 0
        (+ (car some-list) (sum (cdr some-list)))
    )
)

This is a structurally recursive approach in which we add the first element of each recursive step up until we have no more elements in the list, in which case we return a concrete 0.

Having been doing Scheme for about 4 months, this is a very common pattern I see. It’s the construct that allows for the general idea of “iterating through a list” to function.

What I wanted to do was compare this with another functional language, Scala, and see what the differences were.

Scala is the best language I have ever written in. (Not true anymore :D, now it's Go) I have been using it for about a year and a half now and absolutely love it. Since both of the languages are considered functional, they share many similarities. However, I feel like Scala has many more abstractions that make things a lot easier.

To mirror the above code, here is the same idea applied in Scala.

def sum(someList: List[Int]): Int = {
    someList match {
      case Nil => 0
      case x :: xs => x + sum(xs)
    }
}

There is a very clear distinction between these two. Scala uses a concept known as pattern matching to get the head and tail of a List whereas Scheme uses predefined functions to do the same thing. What is interesting, though, is how both of these concepts are trying to do the same two things — wrangle with the structure of lists.

:wq