Hacker News Reader

Hacker News

Comment by toufka | Hacker News Reader

Parent Full thread

toufka

It's curious, because he's actually (unusually) explicit about what this referendum would be for.[1]

1) 'Equal Right to Vote' - "...automatic registration, and shift election day to a national holiday."

2) 'Equal Representation' - "ranked choice voting" & degerrymandering

3) 'Citizen Funded Elections' - to align, [money = citizen] rather than [money = moneyed citizen]

[1] https://lessigforpresident.com/the-act/

View on HN

Unfortunately, the second seems to be implemented via a site that promotes Instant Runoff Voting (IRV), which is one of the only voting systems that manages the impressive feat of being worse than first-past-the-post voting (our current system). Not least of which because IRV is one of the few systems where voting for someone can make them lose. IRV encourages people to list their full preferences, but then completely ignores all rankings except the first until the first loses. So if you vote A>B>C, your vote does absolutely nothing to help B win unless A loses first.

There are several different systems that are quite a bit better, including approval voting (simpler) and Condorcet (closest to ideal, still easy to explain until someone asks what happens with a tie, which rarely happens anyway).

As for #3, sure, I'd love to see candidates' election campaigns funded primarily by citizens, at the option of those citizens. That doesn't mean I want to see them funded by mandatory taxes. Where can I cash in my voucher for an "all of these candidates suck" refund, for instance?

#1, on the other hand, seems like a great idea. Could go hand-in-hand with making sure it's a severe crime to deprive anyone or any group of their ability to vote (such as the various stunts that have occurred in past elections where certain districts "mysteriously" had malfunctions).

I assume that in a voucher-funded campaign system, if you don't give your voucher to any particular campaign, it would expire and go back into the general fund.

I really like the voucher idea because it solves the issue of "I don't want my tax money being spent on candidates I don't agree with".

> I assume that in a voucher-funded campaign system, if you don't give your voucher to any particular campaign, it would expire and go back into the general fund.

One would hope, though that's still wasteful. And without seeing a particular proposal, I don't know whether it would think to include such a provision or whether it would simply decide it knows better how to allocate those funds towards election campaigns based on the people who do use their "vouchers".

> I really like the voucher idea because it solves the issue of "I don't want my tax money being spent on candidates I don't agree with".

There's a much easier way to solve that issue: don't give tax money to candidates.

I don't think IRV is worse than first-past-the-post (FPTP) at all. IRV eliminates vote splitting, reduces strategic voting, and results in fewer wasted votes. Also, the fact that IRV ignores all rankings except the first until the first is eliminated means that you won't hurt your most preferred candidate by changing the order of your other preferences. If you're voting A>B>C, then if A wins, then you wouldn't really care whether you helped B win or not.

Approval voting has the problem of electing a candidate that's acceptable instead of one who the majority actually likes.

Also, IRV has a greater tendency to elect the Condorcet winner when compared to FPTP.

Every voting system has its problems but IRV is still a lot better than FPTP.

> I don't think IRV is worse than first-past-the-post (FPTP) at all. IRV eliminates vote splitting, reduces strategic voting, and results in fewer wasted votes. Also, the fact that IRV ignores all rankings except the first until the first is eliminated means that you won't hurt your most preferred candidate by changing the order of your other preferences. If you're voting A>B>C, then if A wins, then you wouldn't really care whether you helped B win or not.

IRV doesn't eliminate strategic voting. Yes, if A is going to win, then A>B>C is great. However, voting A>B>C rather than B>A>C can cause C to win instead of B. It doesn't "eliminate vote splitting" except in the case where a third-party candidate has no chance; in the case where the third-party candidate actually has a chance, IRV can break horribly.

See http://minguo.info/election_methods/irv and http://minguo.info/election_methods/evaluation .

> Approval voting has the problem of electing a candidate that's acceptable instead of one who the majority actually likes.

And that's a bug? Approval tends to find satisficing solutions, yes. Condorcet does better, though; it's just harder to deploy (but no harder than IRV).

Condorcet is better?

http://ScoreVoting.net/AppCW.html

According to the world's most robust Bayesian regret calculations, IRV is actually significantly better than Plurality Voting. Although it's roughly equivalent in quality to the top-two runoff (TTR) system.

Condorcet methods are typically even better, but NOT "closest to ideal". The best system (if we exclude "exotic" varieties that are too complex to be practical) is Score Voting aka Range Voting.

http://ScoreVoting.net/BayRegsFig.html

It's also plausible that Score Voting and Approval Voting are, in practice, better Condorcet methods than real Condorcet methods.

http://ScoreVoting.net/AppCW.html

Clay Shentrup Co-founder, The Center for Election Science

# 3 Sounds great

# 2 Sounds like a move to a Parliamentary System which I'm for.

#1 Well, Unfortunately I'd be more in favor of ensuring voters actually know who and what they're about to cast a vote for. Our Nation was Founded as and is a Republic, The Average American Citizen is not informed enough for us to live in a true Democracy

Re #1, thankfully we live in a republic, so your fears about the ignorant masses won't come true.

Unfortunately the ignorant masses, continue to reelect the problem at a 95% rate

I don't think #1 is as bad as you think. It works if you make it easier for people to vote, but don't force them to vote. Additionally, emphasizing that citizens have a right to vote and giving them more time to do so can lead more a more engaged population and voters that are more informed.

We'll have to agree to disagree and use for example cities and rural areas where literacy rates hover around 50%, see Detroit. Do you honestly believe these people have an accurate grasp on the civics of the nation, or the logic and understanding necessary to come to an informed decision?

Do you honestly believe we should restrict their ability to vote based on your preconceptions about their understanding of the world?

Edit: I do agree that lots of people don't posses the quality of information to make good voting choices. I just think that isn't a good justification for disenfranchising people.

No I would not do it on my preconception of their understanding, that is what a civics test would confirm or deny. If its required to gain citizenship why should it not be a requirement of voting? LowInformation voters have not been a benefit to this country.

Reply on news.ycombinator.com