Reply to thread

Message: <blockquote data-quote="fusangite" data-source="post: 2062430" data-attributes="member: 7240">What I want you to do is explain these conditions. While Umbran and I can actually explain what we are advocating on this thread and how it works, all we are getting from you is links to web sites advertizing your favourite system. If you cannot actually defend your system in your own words, this indicates a problem to me. Besides, if you actually had to explain the exercise that you are referencing in the context of the ENNies, you would have to construct a model where this highly rare scenario took place with five rather than three candidates -- a far more difficult thing to do because it represents a far more improbable outcome.Again, what you seem to be doing is favouring a system under all circumstances, regardless of what it is being used to count and the social/political environment in which it is being used. Finally, I note that you have yet to counter any of my arguments for what happens under your system when people vote strategically. If you are indeed leaving these completely uncontested, I think I have conclusively shown that even if IRV breaks under certain real world conditions, approval voting breaks under far more.I wasn't disputing that it was called this on the group's web site. I was disputing that this name was actually descriptive. In order to call it this, you have to falsely describe second, third, fourth, etc. preference votes as essentially not votes or as inferior votes; they are not, as in Borda systems, inferior votes. They are conditional votes under IRV. What I was objecting to was the inaccurate and rhetorical naming of a popular and widely used system.Why would I rank a product that was terrible? I wouldn't. You seem to assume that people will cast votes with no regard to the strategic outcome of doing so. IRV votes are transferable votes; why would I establish any conditions whatsoever under which my vote would transfer to a bad product?Not if people vote rationally. You argue that I for some unknown reason I would choose to use my ballot to describe a product that I thought should lose under all conditions a0s my second choice. Why on earth would I do such a thing? If there were only one product on the list about which I had a positive opinion, the rational thing for me to do would be to put a "1" next to it and leave it at that. How is that different from your model? If there are unknowns, why would I give them votes? I would do exactly the same thing I would do under IRV; I would refrain from casting a vote in favour of a product about which I knew nothing. So, just as I would not cast an approval vote for a product I knew was bad, I could not, in good conscience, cast one for a product about which I knew nothing and therefore might be even worse. So, just as under IRV, voting under your system the product would also never ever get my vote. So what's the difference?This doesn't answer my question. You have somehow established conditions under which a product about which few people know anything can get more votes than a product that many people like. My point is that designing such a system is a foolish thing to do. Fortunately, you have yet to convince me that people will use approval voting in such a way that your desired outcome will even happen.Well, if such a system were unacceptable, you would have to reject Conaill's too. There are only two conditions in his system: voting for something and NOT voting for something. Abstaining in approval voting is, just as under IRV, identical to voting against the product.</blockquote>

Verification