🚀 go-pugleaf

I'd appreciate some comments on something which may be at
the boundaries of the MOP.

I have been doing some work which involves taking over the
slot storage for metaclass instances.  The purpose in doing
so is to experiment with a fully versioned environment, where
the values of slots, as well as the slots hemselves are 
versioned.  Behaviors get version too, for that
matter. 

I have been using the July 92 PCL as the basis for this, and of 
course things aren't quite working - versioning of slot 
*values* is fine, but something isn't quite right in my
take over of the management of the slot storage for the metaclass 
instances.

I'm quite sure that I can get this to go (things like this 
worked, allowing for names, in the May 90 PCL), but I am 
curious to hear what the opionions of skillful MOP'ers are.

Am I beyond the pale here?  This question has also made me 
curious about exactly where the limits of the MOP design are, 
and how those limits might be described.  (Something better
than: "Whatever you can do, by changing the behaviors of
the GF's in the AMOP book"
  
If anyone has one, I would also appreciate any working example
of taking over slot storage for metaclass instances.

   ...Tom M