4 ms·

> This issue would have been avoided by a electronic machine producing clean paper records verified by voters and are not ambiguous in a recount What happens i

> This issue would have been avoided by a electronic machine producing clean paper records verified by voters and are not ambiguous in a recount What happens if paper ballot is correct but machine then changes some small percentage of votes randomly when sending upstream? You say there will be a recount, how would that work?, what would trigger it?, and what could a malicious actor do to prevent it from working? Malicious actor that can have firmware modified (that seems like a easy target - compromise company making them), that can work up large crowds and have resources for lawsuits. What would the end result be? What would society look like after a few months of this?

nebula8804 2y ago
>What happens if paper ballot is correct but machine then changes some small percentage of votes randomly when sending upstream? If it does not change the outcome of the election then it does not have the intended effect. If the margin was so close as to change the election the laws on the books would force a complete manual count regardless. Nevertheless, you are proving my point about the risk limiting audits which would help to catch anomalies like this. >You say there will be a recount, how would that work?, what would trigger it?, and what could a malicious actor do to prevent it from working? If the margin is within a certain percentage (depends on the state) it forces a manual count of the ballots. I am advocating the risk limiting audits to help catch the machine acting malicious while printing the correct result on paper. >Malicious actor that can have firmware modified (that seems like a easy target - compromise company making them), that can work up large crowds and have resources for lawsuits. The lawsuits would still have to respect the law on the books. No machine is perfect but humans are not perfect either. In this thread I already discussed the incident in NY where they have paper ballots filled out by voters and then scanned. Humans messing up the ballots lead to the discarding of votes because they could not be validated. A combination of machine verifying human while human verifying machine is best.
- adriancr 2y ago
  You are missing my point. We do not leave in an ideal world. Saying something should happen does not translate to it being done. Example, Bush vs Gore was decided by the courts, not by recount. So if that election was stolen, we will never know. In that case, your whole argument falls apart, there should have been a recount but that never happened. In that case, if firmware altered enough votes it worked. " Justice Antonin Scalia, convinced that all the manual recounts being performed in Florida's counties were illegitimate, urged his colleagues to grant the stay immediately.[1] On December 9, the five conservative justices on the Court granted the stay, with Scalia citing "irreparable harm" that could befall Bush, as the recounts would cast "a needless and unjustified cloud" over Bush's legitimacy. " Now replace Bush with Trump, have a blatant steal, work a lawsuit up to supreme court, have supreme court decide the same way.
  - nebula8804 2y ago
    >Example, Bush vs Gore was decided by the courts, not by recount. That whole event was triggered by the "hanging chad" nonsense found in the poorly designed paper ballots that led to ambiguity. >Now replace Bush with Trump, have a blatant steal, work a lawsuit up to supreme court, have supreme court decide the same way. If you cannot trust the law then a speedy result works in your favor. If you have a quick electronic result plus have a risk limiting audit done the night of the election as standard practice it makes it harder to challenge the result. In a paper only ballot by definition will take longer and leaves the door open for every election to be challenged.
    - adriancr 2y ago
      > That whole event was triggered by the "hanging chad" nonsense found in the poorly designed paper ballots that led to ambiguity. You say nonsense but wikipedia says: "a statewide recount would have shown that Gore received the most votes, according to the Florida Ballot Project." So it seems that election result was changed successfully by machines. That would not have happened with regular vote counting. > In a paper only ballot by definition will take longer and leaves the door open for every election to be challenged. Paper and counting with observers is much harder to compromise and more efficient then some machines. (see EU parliament elections) Alas, you still argue machines are better when there's precedent of them changing an election result. I don't have anything more to add here, thanks for the discussion, was interesting digging through wikipedia on this.
      - nebula8804 2y ago
        >So it seems that election result was changed successfully by machines. Did you read the right wikipedia article? The ballots were paper ballots! The poor design caused people to vote for the incorrect candidate. What machine are you talking about? Certainly not DRE. There was no DRE involved in this state. >Alas, you still argue machines are better when there's precedent of them changing an election result. I am arguing DRE with risk limiting audits solves all issues. Fast turnaround with two way auditing. Prominent election security experts are advocating the same thing: https://www.youtube.com/watch?v=U-184ssFce4&t=1578s https://www.youtube.com/watch?v=U-184ssFce4&t=1578s >I don't have anything more to add here, thanks for the discussion, was interesting digging through wikipedia on this. Its so funny to see how HN people tend to be so ignorant of the real world because they think they know best. After all they are hackers right? Sooner or later they always get smacked with reality.

nebula8804 2y ago

>What happens if paper ballot is correct but machine then changes some small percentage of votes randomly when sending upstream? If it does not change the outcome of the election then it does not have the intended effect. If the margin was so close as to change the election the laws on the books would force a complete manual count regardless. Nevertheless, you are proving my point about the risk limiting audits which would help to catch anomalies like this. >You say there will be a recount, how would that work?, what would trigger it?, and what could a malicious actor do to prevent it from working? If the margin is within a certain percentage (depends on the state) it forces a manual count of the ballots. I am advocating the risk limiting audits to help catch the machine acting malicious while printing the correct result on paper. >Malicious actor that can have firmware modified (that seems like a easy target - compromise company making them), that can work up large crowds and have resources for lawsuits. The lawsuits would still have to respect the law on the books. No machine is perfect but humans are not perfect either. In this thread I already discussed the incident in NY where they have paper ballots filled out by voters and then scanned. Humans messing up the ballots lead to the discarding of votes because they could not be validated. A combination of machine verifying human while human verifying machine is best.

adriancr 2y ago
You are missing my point. We do not leave in an ideal world. Saying something should happen does not translate to it being done. Example, Bush vs Gore was decided by the courts, not by recount. So if that election was stolen, we will never know. In that case, your whole argument falls apart, there should have been a recount but that never happened. In that case, if firmware altered enough votes it worked. " Justice Antonin Scalia, convinced that all the manual recounts being performed in Florida's counties were illegitimate, urged his colleagues to grant the stay immediately.[1] On December 9, the five conservative justices on the Court granted the stay, with Scalia citing "irreparable harm" that could befall Bush, as the recounts would cast "a needless and unjustified cloud" over Bush's legitimacy. " Now replace Bush with Trump, have a blatant steal, work a lawsuit up to supreme court, have supreme court decide the same way.
- nebula8804 2y ago
  >Example, Bush vs Gore was decided by the courts, not by recount. That whole event was triggered by the "hanging chad" nonsense found in the poorly designed paper ballots that led to ambiguity. >Now replace Bush with Trump, have a blatant steal, work a lawsuit up to supreme court, have supreme court decide the same way. If you cannot trust the law then a speedy result works in your favor. If you have a quick electronic result plus have a risk limiting audit done the night of the election as standard practice it makes it harder to challenge the result. In a paper only ballot by definition will take longer and leaves the door open for every election to be challenged.
  - adriancr 2y ago
    > That whole event was triggered by the "hanging chad" nonsense found in the poorly designed paper ballots that led to ambiguity. You say nonsense but wikipedia says: "a statewide recount would have shown that Gore received the most votes, according to the Florida Ballot Project." So it seems that election result was changed successfully by machines. That would not have happened with regular vote counting. > In a paper only ballot by definition will take longer and leaves the door open for every election to be challenged. Paper and counting with observers is much harder to compromise and more efficient then some machines. (see EU parliament elections) Alas, you still argue machines are better when there's precedent of them changing an election result. I don't have anything more to add here, thanks for the discussion, was interesting digging through wikipedia on this.
    - nebula8804 2y ago
      >So it seems that election result was changed successfully by machines. Did you read the right wikipedia article? The ballots were paper ballots! The poor design caused people to vote for the incorrect candidate. What machine are you talking about? Certainly not DRE. There was no DRE involved in this state. >Alas, you still argue machines are better when there's precedent of them changing an election result. I am arguing DRE with risk limiting audits solves all issues. Fast turnaround with two way auditing. Prominent election security experts are advocating the same thing: https://www.youtube.com/watch?v=U-184ssFce4&t=1578s https://www.youtube.com/watch?v=U-184ssFce4&t=1578s >I don't have anything more to add here, thanks for the discussion, was interesting digging through wikipedia on this. Its so funny to see how HN people tend to be so ignorant of the real world because they think they know best. After all they are hackers right? Sooner or later they always get smacked with reality.