Final thoughts on FP (Why should I care?)

The influence of functional languages

In Programming Languages: Concepts and Constructs, Ravi Sethi quotes J.H. Morris, who gives the following description of the historical role of functional programming:

Functional programming as a minority discipline in the field of programming languages bears a certain resemblance to socialism in its relation to conventional, capitalist economic doctrine. Their proponents are often brilliant intellectuals perceived to be radical and rather unrealistic by the mainstream, but little-by-little changes are made in conventional languages and economies to incorporate features of the radical proposals.

-- Morris, "Real programming in functional languages"

Polymorphic static types, garbage collection, metaprogramming, and lexically scoped first-class functions all originated in the functional languages community. Today these features are all found in different widely used languages (from Python to C++ to the next major revision of Java). Understanding "academic" languages prepares you for learning future languages, and gives you a framework for more deeply understanding present languages.

Functional style

Functional style can be very powerful and useful even in non-functional languages. Many people believe that functional style leads to clearer, more concise, and more flexible programs. Elements of functional style:

• Avoid side effects: instead, compute and return values. Prefer return values to output parameters.

• First class functions: consider passing functions instead of, or in addition to, data. Parameterize algorithms with functions for flexibility. In languages without first-class functions, use function pointers, "function objects", or interfaces.
• Polymorphic types: C++ templates and other language mechanisms for "generic" or "parameterized" types are essentially forms of polymorphism. Think operations as polymorphic functions that work with multiple types; it leads to more reusable code.
• Decompose functions: functional programs are usually constructed by composing many small functions rather than a few large procedures, because functional languages provide powerful and simple ways to compose functions. Think of ways to decompose systems into small pieces, often with a recursive structure, so that your code modules each become small and manageable.
• Lazy evaluation: This is something we haven't studied this quarter, but if you're interested then you you could play with Miranda or other "lazy languages". In lazy evaluation, you never compute a value until you need it. "Pull" values through a system rather than "pushing" them. While not always appropriate, lazy evaluation can sometimes be a useful technique.