John Cowan <cowan at ccil.org> writes:
> Currently, (eqv? 1+2i 3+4i) is defined to be #f as a consequence of
> the rule about "yield[ing] different results (in the sense of eqv?)
> when passed as arguments to any other procedure". This not only
> appears to be recursive (eqv? is defined in terms of eqv?) but
> the work it does can be covered by a rule such as this:
I don't see the recursion as a problem in this case, as the base cases
are covered by the axiomatic specification.
> Obj1 and obj2 are numbers such that = returns #f, at least one
> of obj1 and obj2 is non-real, and both the real and the imaginary
> parts of obj1 and obj2 are rational numbers.
Why would this be an improvement? It seems pretty hairy compared to
the intuitive spec we currently have.
--
Cheers =8-} Mike
Friede, V?lkerverst?ndigung und ?berhaupt blabla
Received on Thu Jun 14 2007 - 03:39:28 UTC