<div><br></div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, May 26, 2024 at 10:20 Chris Benham <<a href="mailto:cbenhamau@yahoo.com.au">cbenhamau@yahoo.com.au</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">In a way Approval isn't a voting method, it's more like a voting <br>
"anti-method".</blockquote><div dir="auto"><br></div><div dir="auto">???</div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
Voting methods have some algorithm.</blockquote><div dir="auto"><br></div><div dir="auto">Approval has an algorithm: “Sum the approvals”. </div><div dir="auto"><br></div><div dir="auto">Is it much simpler, & more easily & securely computed algorithm? Absolutely!!</div><div dir="auto"><br></div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"> They try to reduce the need for <br>
strategy, hopefully make some attempt to resist strategy, and (normally <br>
and hopefully) give the voter the option of expressing their full <br>
strict ranking of the candidates.</blockquote><div dir="auto"><br></div><div dir="auto">You’re referring to the rank-methods. Yes, as I said, their goal is to allow you to merely state your sincere preference-ordering, & then do it all for you, sheltering & isolating you from the choice that you’d have to make without that complicated, computer-dependent & humungously computation-intensive automatic-machine that is a rank-method.</div><div dir="auto"><br></div><div dir="auto">I’ve agreed that wv Condorcet would be desirable for public political elections were it not for the problems that I’ve specified.</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
Approval has no algorithm. </blockquote><div dir="auto"><br></div><div dir="auto">See above.</div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto">It invites "strategy" and more-or-less relies <br>
on it to give a reasonable result. </blockquote><div dir="auto"><br></div><div dir="auto">No, some ways of choosing whom to approve don’t involve strategy, & are completely sincere. I discussed them in my long posts about how to vote in Approval.</div><div dir="auto"><br></div><div dir="auto">Approve everyone that you like.</div><div dir="auto"><br></div><div dir="auto">Approve all the acceptable candidates & none of the unacceptable ones.</div><div dir="auto"><br></div><div dir="auto">That latter way of approving is how I’d approve in a public political Approval election. No strategy. Just sincere expression regarding acceptability.</div><div dir="auto"><br></div><div dir="auto">But yes, some of the other ways of approving are strategic. Admittedly, Approval doesn’t do it all for you.</div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto">Rather than give the voter a way of <br>
expressing their full sincere rankings, it confines the voter to <br>
choosing a set of candidates whose members they wish to uniformly and <br>
indiscriminately help versus the non-members.</blockquote><div dir="auto"><br></div><div dir="auto">…an alternative wording of Approval’s best descriptive definition:</div><div dir="auto"><br></div><div dir="auto">“Approval let’s each voter rate each candidate as approved or not approved.”</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
It ensures "honest voting" by severely curtailing voter expression.</blockquote><div dir="auto"><br></div><div dir="auto">Approval doesn’t curtail any currently-available voter-expression. It improves voter-expression to allow & count preference & merit differences among a set of more than 2 candidates. </div><div dir="auto"><br></div><div dir="auto">Does it provide for expression of all of your pairwise preferences? No. It isn’t a rank-method. It doesn’t take all of your preferences, but it lets you choose which ones are the more important ones. </div><div dir="auto"><br></div><div dir="auto">The added expressivity of rank-methods comes at too high a price (…here. I can’t speak for Australia or Norway.)</div><div dir="auto"><br></div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
It's based on the assumption that voters only care about the effect of <br>
their votes on who wins…</blockquote><div dir="auto"><br></div><div dir="auto">…as can be said just as much for any method. But with any method, there’s voting intended for expression or pure principle.</div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto">and enjoy compromise-compression strategising. </blockquote><div dir="auto"><br></div><div dir="auto">I don’t know what compromise-compression is. Enjoy? Approval voting is easy, as I’ve described. It can but needn’t be strategic. </div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
Obviously a huge improvement on the FPP "anti-method" but nonetheless <br>
understandably difficult for some people to get enthusiastic about.</blockquote><div dir="auto"><br></div><div dir="auto">…especially people who have been led to expect rankings. That’s why it’s good that RP(wv) & MinMax(wv) are available as alternative proposals for people who insist on rankings.</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
There would still be some incentive for dishonest polls, trying to scare <br>
voters into approving unneeded Compromise candidates to guard against <br>
some Greater Evil. But it is a great benefit that a rational voter can <br>
at least never be conned into not approving their sincere favourite <br>
(along with any other candidates they prefer to their Compromise candidate).</blockquote><div dir="auto"><br></div><div dir="auto">Yes—a big improvement l.</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
So "over the course of a few elections" Approval would strongly tend to <br>
"zero in" on the Condorcet winner. That is great, but the Condorcet <br>
supporter might well ask: "Why should we have to wait that long?".</blockquote><div dir="auto"><br></div><div dir="auto">Of course. For improvement, the sooner the better.</div><div dir="auto"><br></div><div dir="auto">But, for one thing, enacted sooner, Approval could bring better results before Condorcet does…because Condorcet can’t do much till it’s enacted. Two U.S. cities have enacted Approval. Zero have enacted Condorcet.</div><div dir="auto"><br></div><div dir="auto">For another thing, with count-fraud, it might not do any good to have theoretically better method.</div><div dir="auto"><br></div><div dir="auto">It seems to me that, in every one of our EM polls, Approval has chosen the CW. …without any waiting.</div><div dir="auto"><br></div><div dir="auto">Better to trust the voters to use Approval well, than to trust every count-participant to not accept a few bucks from someone in return for doing count-fraud.</div><div dir="auto"><br></div><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
Approval is a huge bang-for-buck contender, mainly by being almost no <br>
"buck".</blockquote><div dir="auto"><br></div><div dir="auto">Don’t underestimate the bang. It would bring a huge difference.</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex" dir="auto"><br>
<br>
Chris B.<br>
<br>
<br>
On 26/05/2024 7:20 pm, Kristofer Munsterhjelm wrote:<br>
> On 2024-05-16 08:04, Michael Ossipoff wrote:<br>
><br>
>> I’m glad you brought up the desire for the method to do it all for us <br>
>> (as Condorcet does)…taking our sincere rankings as input, & <br>
>> outputting the legalistic right choice. …because I’ve been meaning to <br>
>> address that matter.<br>
>><br>
>> …& that’s not even counting the much less expensive implementation <br>
>> (could be zero cost, without even needing new count-software), easier <br>
>> less expensive administration, & easier simpler explanation with <br>
>> consequent easier enactment.<br>
>><br>
>> Approval isn’t as difficult to vote as it opponents claim. There are <br>
>> many ways to choose what or whom to approve. That variety is a good <br>
>> thing, because you can choose how you like.<br>
>><br>
>> Condorcet is legalistic, but we don’t have to be legalistic! Approval <br>
>> guarantees election of the candidate who maximizes the number of <br>
>> voters for whom the outcome is in their preferred of the two <br>
>> merit-subsets, however the voter designates them.<br>
>><br>
>> e.g. If people approve what they like, Approval maximizes the number <br>
>> of people who like the outcome.<br>
>><br>
>> If people approve what’s acceptable, then Approval maximizes the <br>
>> number of people for whom the outcome is acceptable.<br>
>><br>
>> If people approve above expectation, then Approval maximizes the <br>
>> number of people for whom the outcome is above expectation…maximizes <br>
>> the number of people pleasantly surprised.<br>
>><br>
>> You don’t know the objectively-optimal vote? Neither does anyone <br>
>> else, so don’t worry about it!<br>
><br>
> This makes it very hard to reason about or define any relative <br>
> properties about Approval, though, because it kicks the can down the <br>
> road to the voter instead.<br>
><br>
> This is perhaps a bit exaggerated, but to show a point: does Approval <br>
> really pass clone independence? Suppose more candidates of a wing show <br>
> up, which embolden people to be more discerning than just approving <br>
> what's acceptable. Then the winner changes.<br>
><br>
> Is "Approval"[1] *really* FBC, or monotone? If it converges to the <br>
> Condorcet winner, then nope, because you can't have FBC and Condorcet. <br>
> If it converges to the essential set, then nope, because the essential <br>
> set is nonmonotone.<br>
><br>
> By saying "don't sweat it, people will do what they will do, there's <br>
> no objective meaning to the ballot except what the voters like", voter <br>
> behavior becomes so much more intermixed with the method. And then, <br>
> for criteria to compare apples to apples, they should take this <br>
> intermixing into account. Not being legalistic shouldn't let Approval <br>
> off the hook.<br>
><br>
> But how?<br>
><br>
> What I want is to have each method equally on the hook for what its <br>
> behavior implies. To paraphrase a reply I made to rb-j: if the winner <br>
> of an election can change when non-winners drop out, we don't really <br>
> care if the method passes IIA as such. It's a nice thing to have in a <br>
> box-ticking way, but that's it.<br>
><br>
> If a method wants to open up the can of worms to say "nope, this <br>
> method isn't meant to be used in isolation, it's supposed to be used <br>
> with /whatever/ internal way of going from utilities to the types of <br>
> expression the ballot accepts", then okay, let's take it at face <br>
> value. Let's ask the questions that matter. Can this make someone lose <br>
> if some of the honest voters like the candidate more? Can it make a <br>
> non-winner change who the winner is?<br>
><br>
> For ranking, this is easy, courtesy of the "legalism". For Approval, <br>
> not so much.<br>
><br>
> -km<br>
><br>
> [1] or rather, the method induced by iterating Approval or by <br>
> combining Approval with one or more thresholding strategies implicitly <br>
> chosen by a voter.<br>
> ----<br>
> Election-Methods mailing list - see <a href="https://electorama.com/em" rel="noreferrer" target="_blank">https://electorama.com/em</a> for list <br>
> info<br>
</blockquote></div></div>