> That may make more sense. What about allowing for a means to provide > bidirectional bindings? Any Thoughts? Yes, I think this is something we should support, when the expression represents an lvalue (in other words, something that can be assigned to). G