>
>
> If you unload P3 you would get P2.
> If you then unload P2 you would get P1.
>
> Or,
>
> If you unload P2 first you would remain at P3.
>
> But,
>
> If you then unload P3, you would get P2, which is a bug.  P2 
> doesn't/can't "unload" the shadowed version (which only exists in the 
> .changes file) and so that stays in the stack.

I see in VW3.0 there were hidden namespace so that at unload of p3 you 
would get p2.
Can I summarize that when you load overriding method you lose the 
conflict or methods attached to each package?
I guess that this is right. I try to see the implication of keeping 
them somewhere attached to the package but not shown to the user. This 
can be a bit complex when you start to edit P1>>C>>m for example. With 
your model, the method just changes its host and this is simple which 
is good.

Thanks for the discussion it was good and it helps me to clarify 
points. I like this list...ok back too hacking packages representation 
for our wonderful reengineering environment :)

Stef