But it would be unfortunate if we ever thought that modeling IS inference,
or that any model or set of models could EXHAUST or entirely capture the
rules or meaning of inference, or that we fully understand the rules and
criteria for the game of MODELING itself.
Still, I can list two reasons for pursuing rules and criteria for the game
of modeling itself. The first is historically longstanding, the second is
new:
1) Settling contentious public issues involving competing models (the legal
reason)
2) Enlisting the aid of a machine in performing the evaluation (the primary
UAI reason)
Clark Carrington