>  Your new scheme don't solve the problem, I think
Oh I didn't intend that as a new scheme, like I said its more what I
expected than anything else.
I read the article on compatibility it is interesting but I don't think they
are solving much, I'll have
to reread it (I scanned the end part, it is late here) but I don't see how
they implement their
properties without defining new messages which raises both up and downward
compatibility
issues since their property metaclass do not parrallel the class hierarchy.
I'll also try to apply this thing to ruby.
Benoit
PS:
I guess I've been constrained enough by the java class model to ignore those
problem, in java
there is but one metaclass, the class Class and it is singleton.
PPS:
I'll have to reread the ruby book