6 ms·

Debian uses the Schulze method (if things haven't changed recently), so I do not see how that would have an impact on the result.

contravariant 5y ago
As far as I can tell the Schulze method does not satisfy the independence of irrelevant alternatives condition, and so adding an extra option can have an effect on the end result. However the upside is that you can actually recalculate what the result would have been if you remove one of the options, since the voting results are summarized as the number of people who prefer A to B, which shouldn't change if you remove one of the options.
- nulbyte 5y ago
  If you remove option A, it would necessarily be a different outcome for someone who would have voted for A over B.
  - YetAnotherNick 5y ago
    I think what GP meant that if you have two identical option A and B, we could remove A after voting and it would have the same result as not having option in the first place.
    - sjmm1989 5y ago
      I think you got their intent in sharing their belief correct; but their belief is wrong. If you remove A as a choice, or B, people might go with their next preferred option instead of just not voting at all, etc. This means you can't know for sure that results would equal the same with or without that option, because you cannot actually assume all actors will vote the same every time. To even begin to think such a thing as true is just asinine when the world proves to us on a daily basis that people are willing to compromise on their beliefs all the time. (No offense.)
      - winkeltripel 5y ago
        But its a ranked ballot. The sort should be stable.
- sjmm1989 5y ago
  The problem with this is that it is thinking in absolutes during a situation where people are often very willing to compromise their own beliefs just to be able to provide an answer. In normal elections we see this with people who don't like any choice, but don't want to spoil their ballot. So they pick their next best favorite. This means that if A is removed, people might pick B or C instead just because A isn't there. People are tricky. Hard to pin down what's exactly going to happen every time. There are distinguishable patterns and such; but ultimately you can't rely on them as if they are infallible.
  - contravariant 5y ago
    As long as you assume people aren't suddenly going to prefer A to B when C is there and prefer B to A when C isn't then you can safely recalculate the result. Of course people might not necessarily act rationally but you've got to draw a line somewhere, otherwise elections are simply undefined behaviour and talking about properties of elections becomes meaningless.
    - couchand 5y ago
      The behavior you describe isn't at all irrational: in a nutshell that's the effect that propelled Biden to the White House.
      - winkeltripel 5y ago
        I don't understand your point. In ranking presidential candidates, most left-or-center-leaning voters would rank any and every Democrat above any and every Republican, with Trump as their very last choice. So the presence or not of any specific candidate doesn't change the outcome. They would have voted in whomever the Democrat Nominee was.
        couchand 5y ago
        There was lots of reporting in 2020 about Democratic voters who preferred another candidate throwing their support behind Biden based on the idea that he was best positioned to win in the general election.
herodoturtle 5y ago
> Debian uses the Schulze method For those that are wondering (like I was); Google & Wikipedia: "The Schulze method selects a single winner using votes that express preferences. The method can also be used to create a sorted list of winners. Therefore, if several positions are available, the method can be used for this purpose without modification, by letting the k top-ranked candidates win the k available seats."
avianlyric 5y ago
Because this is a constitution change, which requires a 3:1 super majority to change, a concept the Schulze method doesn’t really provide for. The issues arises from how the 3:1 ratio is calculated. Which is by comparing the number of “change it” votes to the number of the “None of the above (NOTA)” votes (this is my understanding of from the email thread). As a result the number of NOTA directly impacts the number to approval votes needed to win, as a simple majority isn’t enough. Thus having a NOTA option, and an option that looks like a NOTA can create confusion which splits the NOTA vote, because voting “keep public vote” above NOTA, which reduces the number of counted NOTA votes, and increases the number of “change it” votes (as all non-NOTA votes are considered “change it” votes, regardless of practical impact, at least that my understanding from the thread). Thus reduces number of approval votes need to achieve the super majority. As this vote was won by a single vote, voter confusion on how choice ordering regarding the NOTA option could have swung vote against the intentions of the voting members.
- malthuswaswrong 5y ago
  But NOTA and Affirm the Public Vote are not the same. Choosing ATPV would have changed the constitution. NOTA would not have changed the constitution. If this were an American election full of idiots and people who have no idea what they are voting for, then the argument would have weight. These are people who follow the Debian source code debate close enough to give a shit and vote. They knew what they were doing.
  - mjw1007 5y ago
    ATPV would not have changed the constitution (and did not require a supermajority).

contravariant 5y ago

As far as I can tell the Schulze method does not satisfy the independence of irrelevant alternatives condition, and so adding an extra option can have an effect on the end result. However the upside is that you can actually recalculate what the result would have been if you remove one of the options, since the voting results are summarized as the number of people who prefer A to B, which shouldn't change if you remove one of the options.

nulbyte 5y ago
If you remove option A, it would necessarily be a different outcome for someone who would have voted for A over B.
- YetAnotherNick 5y ago
  I think what GP meant that if you have two identical option A and B, we could remove A after voting and it would have the same result as not having option in the first place.
  - sjmm1989 5y ago
    I think you got their intent in sharing their belief correct; but their belief is wrong. If you remove A as a choice, or B, people might go with their next preferred option instead of just not voting at all, etc. This means you can't know for sure that results would equal the same with or without that option, because you cannot actually assume all actors will vote the same every time. To even begin to think such a thing as true is just asinine when the world proves to us on a daily basis that people are willing to compromise on their beliefs all the time. (No offense.)
    - winkeltripel 5y ago
      But its a ranked ballot. The sort should be stable.
sjmm1989 5y ago
The problem with this is that it is thinking in absolutes during a situation where people are often very willing to compromise their own beliefs just to be able to provide an answer. In normal elections we see this with people who don't like any choice, but don't want to spoil their ballot. So they pick their next best favorite. This means that if A is removed, people might pick B or C instead just because A isn't there. People are tricky. Hard to pin down what's exactly going to happen every time. There are distinguishable patterns and such; but ultimately you can't rely on them as if they are infallible.
- contravariant 5y ago
  As long as you assume people aren't suddenly going to prefer A to B when C is there and prefer B to A when C isn't then you can safely recalculate the result. Of course people might not necessarily act rationally but you've got to draw a line somewhere, otherwise elections are simply undefined behaviour and talking about properties of elections becomes meaningless.
  - couchand 5y ago
    The behavior you describe isn't at all irrational: in a nutshell that's the effect that propelled Biden to the White House.
    - winkeltripel 5y ago
      I don't understand your point. In ranking presidential candidates, most left-or-center-leaning voters would rank any and every Democrat above any and every Republican, with Trump as their very last choice. So the presence or not of any specific candidate doesn't change the outcome. They would have voted in whomever the Democrat Nominee was.
      - couchand 5y ago
        There was lots of reporting in 2020 about Democratic voters who preferred another candidate throwing their support behind Biden based on the idea that he was best positioned to win in the general election.

herodoturtle 5y ago

> Debian uses the Schulze method For those that are wondering (like I was); Google & Wikipedia: "The Schulze method selects a single winner using votes that express preferences. The method can also be used to create a sorted list of winners. Therefore, if several positions are available, the method can be used for this purpose without modification, by letting the k top-ranked candidates win the k available seats."

avianlyric 5y ago

Because this is a constitution change, which requires a 3:1 super majority to change, a concept the Schulze method doesn’t really provide for. The issues arises from how the 3:1 ratio is calculated. Which is by comparing the number of “change it” votes to the number of the “None of the above (NOTA)” votes (this is my understanding of from the email thread). As a result the number of NOTA directly impacts the number to approval votes needed to win, as a simple majority isn’t enough. Thus having a NOTA option, and an option that looks like a NOTA can create confusion which splits the NOTA vote, because voting “keep public vote” above NOTA, which reduces the number of counted NOTA votes, and increases the number of “change it” votes (as all non-NOTA votes are considered “change it” votes, regardless of practical impact, at least that my understanding from the thread). Thus reduces number of approval votes need to achieve the super majority. As this vote was won by a single vote, voter confusion on how choice ordering regarding the NOTA option could have swung vote against the intentions of the voting members.

malthuswaswrong 5y ago
But NOTA and Affirm the Public Vote are not the same. Choosing ATPV would have changed the constitution. NOTA would not have changed the constitution. If this were an American election full of idiots and people who have no idea what they are voting for, then the argument would have weight. These are people who follow the Debian source code debate close enough to give a shit and vote. They knew what they were doing.
- mjw1007 5y ago
  ATPV would not have changed the constitution (and did not require a supermajority).