>As before, not being picky or whatever here, but I'm stll not understanding your use of 'recursively defined'. As a Algol/Algol68/Pascal/C/C++ (in order with some missed out), I would say that a recursive definition was
>
>#define TRUE (! FALSE)
>#define FALSE (! TRUE)
>
>'recursively defined' has some implication that it is defined in terms of itself.

I'd disagree here (not being picky< g >). In this context, it means being defined terms of something that preceeded it, not it terms of itself. This is also a definition of "recursive". As I mentioned, this is also one of the premises of OOP.

>(As well as having a thing about correct use of data types (TRUE/FALSE are ONLY boolean), I have a thing about unnecessary use of recursion too. C++ classes are prone to some very nasty hidden recursion, especially in destuctors - a real 0x627567676572 to track down)

In this context, absolutely. I'm as quick to use recursion as the anybody, if, and only if, the situation warrents it. This, of course, can only be determined via quantitative analysis of the problem.