The inexact->timespec and timespec->inexact procedures should probably
forbid inexact integers, just to state the obvious. If there are missing
digits in the integer part, it's not obvious how to interpret that.
What is the canonical term to talk about inexact floats - do we call
them floats, real numbers, or something else?
An inexact integer is just an inexact number that happens to have an integral value (like 12.0), so there's nothing wrong with that. In principle, inexact numbers don't have to be floats (they could be, say decimal numbers with 10 integral digits and 3 fractional digits), but the whole reason inexact numbers exist in Scheme is efficiency (exact arithmetic can get very slow if you do a lot of divisions). So since floats are provided in all modern hardware, Scheme models that.
John Cowan
http://vrici.lojban.org/~cowan xxxxxx@ccil.orgRather than making ill-conceived suggestions for improvement based on
uninformed guesses about established conventions in a field of study with
which familiarity is limited, it is sometimes better to stick to merely
observing the usage and listening to the explanations offered, inserting
only questions as needed to fill in gaps in understanding. --Peter Constable