Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bluecaribou
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
3 ms
·
1.
▲
by
bluecaribou
7y ago
> We just need _better_ right now, we need momentum. Agreed. But we shouldn't be relying on local governments to provide that momentum. I think the people in the best position to legitimize improved voting systems are the pollsters.
2.
▲
by
bluecaribou
10y ago
As I explained in my other comment, that doesn't have to be a problem. Just let everyone vote with a traditional "single choice" ballot. But you also allow the voter to optionally enhance/refine their preference with a s
3.
▲
by
bluecaribou
10y ago
So first of all, I think that only three evaluation methods should be considered - ballot conversion/reduction to traditional "single choice", "sum of ratings", and Schulze. You might need to start off with conversi
4.
▲
by
bluecaribou
10y ago
Recounting my comment from a previous article[1], comparing all these voting systems is not really the right way to think about it. First, you need to separate the mechanism used to express voter preference (i.e the ballot design), from the
5.
▲
by
bluecaribou
10y ago
Just to clarify, when I say "relaxed rank voting is just a restricted version of score voting", I'm talking about the ballots, not the method for evaluating the winner. As I said, those two things should be separated. I agree
6.
▲
by
bluecaribou
10y ago
The point is that "approval" voting and "score" (aka "range", aka "rating") voting are not fundamentally different systems. Approval voting is just score voting where the voter is restricted to giving
7.
▲
by
bluecaribou
10y ago
Pitting "ranked choice" versus "approval" (versus "range") voting is not really the way to think about it. Consider a "relaxed" version of ranked choice where voters are allowed to give multiple candi