Thanks for the swift reply,
see inline
On Wed, 27 Oct 2004 09:51:29 +0100, Graham Klyne
I think the first question you have to address is whether you really want to represent a *set* of reals or an *interval* of reals.
A set of intervals, I would assume... The reason for that is that i will probably end up with set theoretic operations like complement, and in that case the complement of an interval of reals will have a hole in it somewhere. That's why i represented them as lists of pairs (lower and upper bound of the sub-interval, so to speak) , but that will probably get ugly before long.
Then, some other questions follow: - possibly infinite sets within any given interval?
I was not explicitly thinking of infinite sets, i just wanted to keep the precision open for now. I would say, up to 4 decimal figures maximum.
- open or closed intervals?
closed intervals, but with holes. (see above)
and probably more.
#g --
cheers, stijn