double-in says it specs a 64-bit floating point number https://clojuredocs.org/clojure.spec.alpha/double-in what did s/expain return?

Clojure doesn't really support floats in general (other than for interop). you can of course make your own spec for floats, or cast into a double etc.

my brain just tripped over the fact a 64 floating point number isn't a float. err, or i guess it is, but not the 64 bit variety. as an aside, are the other numbers just not buoyant?