Side-effect free containers for E

[Jonathan S. Shapiro]

> >Weak (sensory) capabilities don't seem to lend themselves to a static
type
> >characterization at all.
>
> I think your last examples illustrate my point.  Yes, there are types, but
> not in a language sense

Actually, these *are* types in a language sense -- just not static types.

The problem with sensory that makes a static type description impossible is
that the output of a key fetched through a sensory key depends on the input.
In consequence, the whole transitive type space needs to be described, and
this is prohibitive in a statically typed system.

In a dynamic type system, this is both normal and okay.

shap