I merely meant my other implementation (lzp_rel) seemed (to me) more in keeping with the ethos of prolog: declaring the necessary relationships and expecting prolog to take care of the execution details for you. It also could not be written that way in Lisp, so it seemed more ‘prology’ in that respect also - a form of expression that prolog supports that most other languages don’t. Whereas recursion is something both languages utilize. I didn’t mean that recursion is somehow not prolog and I also understand that recursive definitions also express relationships.
(shades of this discussion I guess. I wouldn’t try to say recursion isn’t declarative, and I think z5h’s remark on the matter is a good one, but I can kind of see where emiruz is coming from)
Thanks for the link to the clpBNR documentation. This looks very useful - will read!