The Problem with unification and logical variables would be that they have to be dynamic in scope and work with a call-by-name evaluation strategy, at least in Prolog.
Anyway, you could build a Prolog-Like DSL with pat-m and amb
While logic Programming in Arc can make problem-solving easier, i think it introduces problems for libraries:
- How should you call functional functions from logical functions?
- How should you call logical functions from functional functions without violating referential transparency?
- Do closures over logical variables make any sense?
- Should goals be limited in scope? Lexical or dynamic? If I call 2 logical functions from different modules I may or may not want the first to be retried when the second fails, depending on the context.
