Chapter 10 considered functional programming with data involving names and name abstractions. The new language features introduced in that chapter were motivated by the theory of nominal sets introduced in Part One of the book. We encouraged the reader to think of the type and function expressions of the FreshML functional programming language as describing nominal sets and finitely supported functions between them. However, just as for conventional functional programming, the sets-and-functions viewpoint is too naive, because of the facilities these languages provide for making recursive definitions. Giving a compositional semantics for such language features requires solving fixed-point equations both at the level of types (for recursively defined data types) and the level of expressions of some type (for recursively defined functions). As is well known, solutions to these fixed-point equations cannot always be found within the world of sets and totally defined functions; and this led the founders of denotational semantics to construct mathematical models of partially defined objects, functions, functionals, etc., based on a fascinating mixture of partial order, topology and computation theory that has come to be known as domain theory. For an introduction to domain theory we refer the reader to Abramsky and Jung (1994).

In this chapter we consider merging domain theory with the concepts from nominal sets – names, permutations, support and freshness. As a result we gain new forms of domain, in particular domains of name abstractions.

A very useful feature of functional programming languages such as OCaml (http://caml.inria.fr/ocaml) or Haskell (http://www.haskell.org) is the facility for programmers to declare their own algebraic data types and to specify functions on that data using pattern-matching. This makes them especially useful for metaprogramming, that is, writing programs that manipulate programs, or more generally, expressions in formal languages. In this context the functional programming language is often called the meta-level language, while the language whose expressions appear as data in the functional programs is called the object-level language. We already noted at the beginning of Chapter 8 that object-level languages often involve name binding operations. In this case we may well want meta-level programs to operate not on object-level parse trees, but on their α-equivalence classes. OCaml or Haskell programmers have to deal with this issue on a case-by-case basis, according to the nature of the object-level language being implemented, using a selfimposed discipline. For example, they might work out some ‘nameless’ representation of α-equivalence classes for their object-level language, in the style of de Bruijn (1972). When designing extensions of OCaml or Haskell that deal more systematically with this issue, three desirable properties come to mind:

1. • Expressivity. Informal algorithms for manipulating syntactic data very often make explicit use of the names of bound entities; when representing α-equivalence classes of object-level expressions as meta-level data, one would still like programmers to have access to object-level bound names.

2. […]

A personal perspective

This book has its origins in my interest in semantics and logics for locality in programming languages. By locality, I mean the various mechanisms that exist for making local declarations, restricting a resource to a specific scope, or hiding information from the environment. Although mathematics and logic are involved in understanding these things, this is a distinctively computer science topic. I was introduced to it by Matthew Hennessy and Alley Stoughton when we all arrived at the University of Sussex in the second half of the 1980s. At the time I was interested in applying category theory and logic to computer science and they were interested in the properties of the mixture of local mutable state and higher-order functions that occurs in the ML family of languages (Milner et al., 1997).

Around that time Moggi introduced the use of category-theoretic monads to structure different notions of computational effect (Moggi, 1991). That is now an important technique in denotational semantics; and thanks to the work of Wadler (1992) and others, monads are the accepted way of ‘tackling the awkward squad’ (Peyton Jones, 2001) of side-effects within functional programming. One of Moggi's monads models the computational effect of dynamically allocating fresh names. It is less well known than some of the other monads he uses, because it needs categories of functors and is only mentioned in (Moggi, 1989), rather than (Moggi, 1991).