[EM] apportionment, morality & goals

[Warren Smith]

I am highly dubious of both Ossipoff's "moral argument" and of my own.
But anyway, I have now put them both at
  http://rangevoting.org/NewAppo.html
in the sections titled
 Which one is the most "morally right"? Smith: Alternative method #2.
and
 Which one is "morally best"? Ossipoff: the original method is most moral (as well as
 being the simplest)

I think really, any such arguments have to be backed up by some considerably deeper
arguments about maximizing summed utility in some models,
and at their present level of development are little more than intuitive ranting.

Concering Olson's questions about morality, yes, those are good questions and your
goals happen to correspond quite well with known global optimality theorems
discussed in
   http://www.rangevoting.org/Apportion.html
section titled "Global optimality properties."

Ossipoff said of Olson's list:
>Of the standards that you listed, I prefer the one that minimizes the
>greatest under-representation.

--that standard leads exactly to Adam's method.

However... I do not think Ossipoff really thinks Adams is best!
That is an example of the troubles we get into trying to say "X is morally best"
but without having any clear theory of what morality is.
I believe it ought to be possible to build a theory of morality based on maximizing
summed utility in some models (or at least get closer).

Warren D Smith
http://rangevoting.org