I'm gonna take a look at Wadler's paper—but does anyone want to take a shot at a high-level description of how an algebra would be applicable to the problem of pretty printing?

Any more general comments on applying algebras to solving programming problems would also be appreciated.

My rough understanding of the benefits/rationale atm:

1) the algebra is a nice/minimal factoring of the problem

2) being able to formally classify it means we know a bunch of its properties, which tells us certain approaches are definitely (im)possible (and these properties are drawn from a fairly small set: e.g. associativity, distributivity, existence of inverses)

3) If the algebra is explicitly represented in code (maybe in a library), we can use that general structure to verify (and generate?) certain aspects of our particular problem solution.

Edit: The paper mentions "Over the years, Richard Bird and others have developed the algebra of programming to a fine art" —and I found Bird's book "Algebra of Programming". Seems like what I'm looking for maybe, but would be grateful to hear of any alternative resources as well.