Re: Inequivalence of models

[Tim Gwinn <***>]

Steve,

SJ:
> So here are my questions:
>
> 1) Are these two models of the car(mechanical and
> electric) really one model? Why or why not? What is
> the criteria that allow us to say that they are two
> different models.
>
> 2) If they are different models, what is the largest
> model that subsumes them? How would one go about
> constructing it?

TG:
1) I would say they are two separate models: each formal representation
maintains synonymies with very different aspects of the car.

2) I don't think it would be hard to overlay an electrical diagram onto a
mechanical diagram. (This obviously has to happen at some point during the
engineering, so that mechanically, they know where the battery will fit,
each wiring harness will pass, the bulbs will be located, etc.) The main
point, I think, is that the abstract symbols for bulbs, battery, wiring
connections, etc. can map onto the mechanical diagram, (or that both can map
onto a third diagram) -- there is a commensurability.  With inequivalent
models, there would be a lack of commensurability: there would be no way to
map from one to the other. So, for example, impredicative relational models
are incommensurable with state-based models.

Regards,
Tim

P.S. - Go New England Patriots!! (Reference to the NFL playoff game today
for those of you that don't follow American football.)