Prolog Day Two

Haskell, But Quickly

Heads Up!

I'm writing a book about building and deploying web applications with Haskell and Yesod. Want to know when it's released? Click here.

The second day of Prolog focuses on lists, maths, and recursion. We also dig deep into unification, which is a core element of doing things the Prolog way. An interesting and important idea in Prolog is the way lists decompose; you split a list into its head and tail, and recursively split this way to iterate over the whole collection.

  1. Things to find
  2. Fibonaccis & factorials
  3. A real-world community
  4. Things to do
  5. Reverse a list
  6. Find the smallest list element
  7. Thoughts

Things to find

Fibonaccis & factorials

Find some implementations of a Fibonacci series and factorials. How do they work?

The Rosetta Code project is a boon for comparing the syntax and idioms of programming languages. It’s hard to not be distracted by esoteric passages of code written in Befunge or Golfscript. Given the number of language entries for Fibonacci, it seems it’s the text-book example of a programming problem. Here’s the first RC example of a Prolog Fibonacci implementation:

fib(1, 1) :- !.
fib(0, 0) :- !.
fib(N, Value) :-
  A is N - 1, fib(A, A1),
  B is N - 2, fib(B, B1),
  Value is A1 + B1.

This snippet introduces a couple of new operators — the ! sign which “prevents backtracking of clauses”, and the is keyword which handles assignment. When you pass a number to the above rules, Prolog tries to unify the input with either of the first two rules. If the values can be unified (matched), then it returns immediately.

Magic happens in the third rule. If we pass a value of two or more, Prolog begins exploring the third rule. Two numbers are derived from the input value by subtracting 1 and 2 from it, and the derivatives are passed to two recursive fib calls. The recursion continues until the values of the derivatives are 1 and 0. Prolog adds the numbers to find our answer.

I’m not sure why the ! marks are necessary; I believe Prolog can unify without them. When I try it however, I overflow the stack sooner without the ! marks. Prolog computes an answer very quickly, but it hits a stack overflow with a miserably low input number — 27 on my machine. When I tried to inline the subtractions, I hit an immediate stack overflow. Yikes. Prolog is touchy.

Prolog can avoid a stack overflow with tail-call optimisation. Apparently, you would just need to position your recursive subgoal at the end of a recursive rule, but that’s another exercise for another day.

Let’s move on to factorials. From Rosetta Code: The factorial function of integer n is the product of sequence n, n-1, n-2 etc. For example, the factorial of 7 would be 7*6*5*4*3*2*1 = 5040.

fact(X, 1) :- X<2.
fact(X, F) :- Y is X-1, fact(Y,Z), F is Z*X.

This snippet is similar to the Fibonacci example. The first rule is there to prevent Prolog from counting down past 1. The second rule recursively calls itself with X-1 until Prolog can unify with the first rule.

A real-world community

Find a real-world community using Prolog. What problems are they solving with it today?

I found this article which proves that Prolog is pretty much everywhere in the world, and thanks to Boeing, even out of this world. Prolog is mostly used in academia and research, but I did find a long list of applications. Apparently, Prolog also heavily influenced Erlang.

Things to do

Reverse a list

Based on what I learned from the Fibonacci rules, I came up with this:

rev([], []).
rev([Head|Tail], Answer) :-
  rev(Tail, ReverseTail), append(ReverseTail, [Head], Answer).

This works by splitting a list into its head and tail, and then appending the head onto the reverse of the tail. We reverse the tail by recursively calling the rules with a gradually dimishing tail. Once the tail become an empty list, Prolog unifies with the first rule and the recursion stops. I wanted to put the recursive call at the end of the list of subgoals, but strangely enough I ended up with a longer stack trace.

Find the smallest list element

After spending a good deal of time thinking about this problem, I gave up and Googled an answer. The answer I found kind of makes sense to me, but I wouldn’t be able to explain it to you if you were six years old, which means I don’t understand it well enough.

smaller(X, Y, X) :- (X =< Y).
smaller(X, Y, Y) :- (Y < X).

smallest([Head|[]], Head).
smallest([Head|Tail], Answer) :-
  smallest(Tail, What),
  smaller(What, Head, Answer).

The final Prolog challenge for this chapter is to sort a list, but I’ve already found this language challenging enough so I’m skipping that before I give up on Prolog completely.


Prolog is easily the most confusing challenge I’ve faced in my programming career so far. I’m convinced it’s an incredibly powerful tool, but I can’t easily transition from imperative to declarative thinking. Any knowledge I have of more mainstream languages doesn’t mean shit in Prolog.

I feel that I’m not able to break problems into small steps or components. In something like JavaScript, you can step through the flow of logic with strategically placed console.log() calls and understand what the machine is doing every step of the way. Not only am I not sure how to achieve the same in Prolog, I don’t think you’re even supposed to think in this fashion because it’s imperative and not declarative.

I admit Prolog has defeated me, but I will finish what I can, and come back to it another day.