[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
A small twist/objection to Ivan Moore's question about why traits are
centralized in a 'traits' object instead of accessed by a message to the
prototype: Asking a prototype for its traits is a reflective operation
(what if it doesn't have any traits? should that be revealed by its interface?).
A better way of doing it would be to access the traits through a mirror on
the prototype. At any rate, it shouldn't be done through the prototype
interface itself- this sort of thing is exactly why Smalltalk's programming
environment is so hard to disentange from programs written using it.