Wednesday, November 14, 2012

Functional Programming in Scala

Here I go, congratulating myself for finishing Functional Programming Principles in Scala, an online class taught by Martin Odersky, creator of the Scala programming language, professor at EPFL in Lausanne, Switzerland and co-founder of Typesafe.

Scala is a well engineered modern successor to Java. It's a statically-typed functional/OO language with curly-brace syntax, pattern matching and type inference that runs on the JVM. Scala's libraries are well-thought-out and consistent. And, it's type system is some serious rocket-science.

The course took some material and examples from the classic Structure and Interpretation of Computer Programs (SICP), which is a great way to learn functional concepts in a new language. Topics included:

  • higher-order functions
  • pattern matching
  • immutability and immutable collections
  • laziness
  • touched on inductive proofs of correctness

Boatload of Scala resources

Class staff and students compiled a great list of resources for learning Scala.

While we're on the topic, Twitter has invested big in Scala, open-sourcing some of the results. The presentation Systems Programming at Twitter gives a peek into how Twitter uses Scala for "programming the datacenter". Kestrel is a distributed message queue.


Principles for Good Design

The class was a great introduction to the language and some ideas that will come in handy in any language. I had a lot of fun with it. Odersky wrapped up the course with some principles for good program design.

  • Name everything you can
  • Put operations into natural scopes
  • Keep degrees of freedom for future refinements


  1. Hey Chris, did you see that the "Statement of Accomplishment" is available as of today ? Just printed mine.

    1. I put in exactly the right ratio of effort to slacking. I was over the "with distinction" boundary by .1%