* Set abstractions and operations including infinite, realizing the
impossibility of any complete implementation of such infinite sets, but
undaunted in the desire to still implement, for example, many natural
useful infinite sets, i.e. the set of integers, the set of X-collated
lists, etc.
* Use of set abstraction instead of more-rigid types/classes of most
languages, i.e. a function is declared or determined to accept odd
integers and return the octal string representations of integers 1..100.
This is to explore a concrete functional language implementation that is
more-easily susceptible to mathematical analysis, proof of safety, etc.
by declaring/capturing much more accurately even if inherently
incompletely, what can be determined about a function's domain,
codomain, range, etc. than traditional compiler type systems easily allow.
Bennu
[Well, there's always SETL. -John]
SETL's cool, but only for finite sets. For infinite sets, check out
CM Lisp for ideas. Though I know of no implementation, sadly.
Preston