2016-09-08 14:11 GMT-04:00 C.Benham <cbenham@adam.com.au>: > On 9/8/2016 7:59 AM, Jameson Quinn wrote: > > It is not more complicated than MTA; it's the same. > > >> Its extra complexity over >> those methods doesn't seem to buy much. >> >> if you include in U/P 's explanation all the palaver about the default >> ratings it becomes quite long and complex. >> > > So don't. > > > C:How default ratings are handled is an important part of the method's > definition. Default ratings should be handled in a > way that is simple, clear and consistent. Normally the default rating > should be bottom-most, but if say a score ballot is > used that includes positive and negative numbers it might be at least > understandable and plausible to use zero as the default > score. > For me, the bottom line with default ratings is to have clear, unambiguous rules that approximate the will of the voter as closely as possible. Simplicity is a concern, but secondary to intent. I understand you disagree on these axioms. I think that now we've both stated our axioms, further debate won't go anywhere. > Also U/P uses a more complex ballot than either MTA or 3-slot IBIFA. It > allows voters to give a candidate more than one mark. > Um. I am assuming that the ballot is a piece of paper. Paper is a technology that admits an unlimited number of marks. The fact that I explicitly stated what I would do when such marks were made, while you did not in describing IBIFA, is neither here nor there. > > Taking the full descriptions of the 3 methods into account, U/P is much > more complex than MTA and has a longer and not less > complex description than 3-slot IBIFA's. > Again, if you believe this, let's do the person-on-the-street test. Or if you'd like to propose some other test, I may agree that it's fair. Each side asserting things is not a test. > > BTW in all the examples (at least the ones where the voters confine > themselves to using no more than one of the 3 marks/ratings for > each candidate), U/P gives the same result as MTA. > > Can we perhaps see an example where you think U/P gives a better result > than MTA? > 40: C 35: A>B 24: B>A 1: strategic ego with good information about others' likely ballots (to within an error lower than the relevant margin). Honest preferences are B>A. Strategic options are B or B>A. In MTA, truncation strategy would work for this voter, so they vote B; in U/P, it wouldn't, so they vote B>A honestly. > > The provision you've stuck on to U/P about hobbling "majority rejected" > candidates in the following election is (it bears repeating) > unfair, undemocratic, rigid, vindictive and absurd. > More naked assertions. Yet again: I've consistently said that IBIFA is a great system. You're claiming that U/P, which would give the same result in a large majority of realistic scenarios, is all kinds of bad things. Which one of us is being absurd? > If you're looking for something "NOTA"-like then you could give the voters > the possibility of giving the winner a shorter-than-usual > term. So they could mark as many candidates as they like with L or S, and > if for the winner more than half the voters gave hir an S, > then that winner would serve a short term. > Um. I'm really speechless. ... It seems to me that your suggestion is similar enough to mine that I cannot fathom why you put a moral gulf between them. I can imagine arguments that yours is better or vice versa, but those would be in terms of marginal differences, not a night-and-day gap.