Andrew Lentvorski wrote:
In addition, you can lose optimization opportunities when faced with
weak typing. If I know that 100% of the time I have integers as
arguments to a function, I can throw away the time wasted to check
that I have integers rather than doubles. Weak typing can't do that.
Actually, with weak typing, you don't have to care about what the actual
type of the objects is so you can optimize away, but I am guessing you
were simply referring to dynamic typing.
Anyway, you'd be wrong to assume that you can't do those kind of
optimizations. Check out StrongTalk for a specific example of how it can
be done. It just requires a more clever runtime.
Languages exist in which typing is dynamic and strong. Common Lisp is
further toward the strong side and can use optimization directives to
take advantage of that. I believe that Haskell and OCaML are also
dynamic and strong in terms of typing.
Haskell and OCaml both use static typing, but they are both strongly typed.
I think your terminology is not consistent with standard terminology.
Can you explain what you mean by strong vs. weak and static vs. dynamic?
--Chris
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
