[Ecls-list] ECL's view of floats

[Lars Brinkhoff]

ECL likes to think of its two float types as short-float and
long-float (which are properly equivalent to single-float and
double-float, respectively), while the common wisdom seems to be that
if an implementation provides two float types, they should primarily
be considered to be single-float and double-float.

CLHS says:

  * Two internal representations can be arranged in either of the
    following ways:
    [...]
    * Two types are provided: single-float and double-float. An object
      is simultaneously of types single-float and short-float, or
      double-float and long-float.

which I guess kinda supports the conventional view.

I'm working on a patch to convert ECL to the single/double-float
style, and also to add a third immediate short-float type.

Why immediate short-floats?  For fun.  Any objections?