[EM] General PR question (from Andy Jennings in 2011)

Kristofer Munsterhjelm km_elmet at t-online.de
Thu Oct 2 23:48:53 PDT 2014


On 10/03/2014 01:54 AM, Toby Pereira wrote:
> With the food example, I meant it so that each food type wasn't
> repeatable so that you wouldn't have to make arbitrary decisions about
> liking a food less after the first time (so pizza/pizza wouldn't be an
> option). And while it isn't exactly analogous to political voting, I
> think there are probably other situations where it could be relevant. So
> in this case, I would reject the more proportional curry/fry-up because
> it is Pareto dominated by pizza/curry or pizza/fry-up, which I wouldn't
> necessarily do in political elections.

Alright. Then, to be simple again:

If it were a political election, picking pizza/curry would give the 
curryists more power than the fryup-ists because the pizza candidate is 
even about who he supports and the curry candidate supports the 
curryists alone.

But in food, relative power is not an issue, so having one pizza makes 
everybody more well off.

That's really what I said last time, but I thought I should simplify it :-)

You could of course recast it so that electing the pizzaist would be 
preferable (e.g. if he's universally approved because of his high moral 
standing, and therefore would automatically shift his position to keep 
balance in the council). But the method has no way of knowing that. If 
it's proportional and elects BC, then it's just being cautious in favor 
of proportionality, in a sense.


More information about the Election-Methods mailing list