Paper Ballots vs. Voting Machines vs. both

[Bootstrap]

A useful answer depends on tests that compare the manual process to automated ones, and the results depend a lot on which two systems you are comparing. Which machines are you opposed to? All machines? Are OCR machines OK? How about punch cards? Which manual processes are you advocating? When you say "paper ballot", what exactly do you mean? Without that, it's hard to set up the comparison.

But if you take human error and auditing into account, I think you will find that:

1. Human beings can misread, miscount, or mishandle ballots, leading to inaccuracies in the election results. Most of the systems used in elections have proven that they can do better than human beings do. Those test results were an important reason they were adopted in the first place.
2. Paper ballots are physically vulnerable. They can be damaged, lost, or tampered with. And sometimes lost ballots have been a real issue.
3. Paper ballots take up space, and it's hard to pull them up quickly for reference or auditing. Especially if you want to archive them for decades to allow research or later audits.
4. Auditing paper ballots is very labor-intensive. Digital records allow a lot more audit trails and analytics.

Whether or not you have paper ballots, you need electronic systems too, and you need to be able to make sure that they correspond.

I think we are drawing a distinction between VOTING MACHINES and VOTE COUNTING MACHINES. They are not the same thing at all.

Electronic voting machines are some sort of electronic touch-screen machine in which votes are recorded electronically onto some sort of digital file that is electronically transmitted to a central election headquarters. Or sometimes stored onto a flash drive and hand-carried to a vote center. There are a variety of issues with voting this way not the least of which is digital security.

But if they are (1) not connected to the Internet, and (2) generate a paper ballot that the voter can check, then the result is the same as for a vote counting machine - at the end of the day, you have a paper trail and an electronic trail, and they agree. And for most purposes, the electronic record is going to be used in a lot more ways for auditing.

Of course, it's important to have user interfaces that people understand and can use accurately.

Vote counting machines are optical scanning devices used in elections headquarters to scan and count paper ballots. They are usually not connected to the internet and the technology is simple and often open-source and simply automate the tedious process of counting paper ballots.

That's what we have in North Carolina, at least where I vote.

To me, though, it seems to reach the same result. You need something that is not connected to the Internet so it can't be hacked. You need both an electronic record and a paper record that users can see and verify. You need ways to audit both. If you have that, I don't really understand why it matters whether you start with paper and scan it in or start with a screen and print it out.

[Bootstrap]

So what do states currently use? Looks like optical scanning is the big winner. Direct digital voting is a much smaller percent, and has been going down recently. But that doesn't mean no paper:

DREs also include a paper record called a Voter-Verified (or Verifiable) Paper Audit Trail (VVPAT) that can be used to audit or recount the election.

This source doesn't say what percent of DREs include VVPAT - does anyone know where to find that? To me, optical scanning and a DRE with a VVPAT both get to the same place, no?

https://electionlab.mit.edu/research/voting-technology

[ken_sylvania]

A useful answer depends on tests that compare the manual process to automated ones, and the results depend a lot on which two systems you are comparing. Which machines are you opposed to? All machines? Are OCR machines OK? How about punch cards? Which manual processes are you advocating? When you say "paper ballot", what exactly do you mean? Without that, it's hard to set up the comparison.

But if you take human error and auditing into account, I think you will find that:

1. Human beings can misread, miscount, or mishandle ballots, leading to inaccuracies in the election results. Most of the systems used in elections have proven that they can do better than human beings do. Those test results were an important reason they were adopted in the first place.
2. Paper ballots are physically vulnerable. They can be damaged, lost, or tampered with. And sometimes lost ballots have been a real issue.
3. Paper ballots take up space, and it's hard to pull them up quickly for reference or auditing. Especially if you want to archive them for decades to allow research or later audits.
4. Auditing paper ballots is very labor-intensive. Digital records allow a lot more audit trails and analytics.

Whether or not you have paper ballots, you need electronic systems too, and you need to be able to make sure that they correspond.

I think we are drawing a distinction between VOTING MACHINES and VOTE COUNTING MACHINES. They are not the same thing at all.

Electronic voting machines are some sort of electronic touch-screen machine in which votes are recorded electronically onto some sort of digital file that is electronically transmitted to a central election headquarters. Or sometimes stored onto a flash drive and hand-carried to a vote center. There are a variety of issues with voting this way not the least of which is digital security.

But if they are (1) not connected to the Internet, and (2) generate a paper ballot that the voter can check, then the result is the same as for a vote counting machine - at the end of the day, you have a paper trail and an electronic trail, and they agree. And for most purposes, the electronic record is going to be used in a lot more ways for auditing.

Of course, it's important to have user interfaces that people understand and can use accurately.

Vote counting machines are optical scanning devices used in elections headquarters to scan and count paper ballots. They are usually not connected to the internet and the technology is simple and often open-source and simply automate the tedious process of counting paper ballots.

That's what we have in North Carolina, at least where I vote.

To me, though, it seems to reach the same result. You need something that is not connected to the Internet so it can't be hacked. You need both an electronic record and a paper record that users can see and verify. You need ways to audit both. If you have that, I don't really understand why it matters whether you start with paper and scan it in or start with a screen and print it out.

In a county not far from where I live, the electronic voting machines had issues four years ago where the touch-screens were too sensitive and caused all kinds of problems. This year, the paper printouts that the machines printed didn't match the votes that the machines actually registered. So some polling places had to be closed while they tried to figure out the problem, and of course the big question is "Why should the voters trust the manufacturer's claim that the votes were actually recorded correctly?".

[ken_sylvania]

So what do states currently use? Looks like optical scanning is the big winner. Direct digital voting is a much smaller percent, and has been going down recently. But that doesn't mean no paper:

DREs also include a paper record called a Voter-Verified (or Verifiable) Paper Audit Trail (VVPAT) that can be used to audit or recount the election.

This source doesn't say what percent of DREs include VVPAT - does anyone know where to find that? To me, optical scanning and a DRE with a VVPAT both get to the same place, no?

https://electionlab.mit.edu/research/voting-technology

Screenshot 2023-11-27 at 17.52.11.png

No, not if the DRE prints out a VVPAT that doesn't match what the machine recorded, especially if the printout doesn't match what the voter selected. Chances are high that a votor might not notice if the VVPAT doesn't match what they voted for.

[Ken]

So what do states currently use? Looks like optical scanning is the big winner. Direct digital voting is a much smaller percent, and has been going down recently. But that doesn't mean no paper:

DREs also include a paper record called a Voter-Verified (or Verifiable) Paper Audit Trail (VVPAT) that can be used to audit or recount the election.

This source doesn't say what percent of DREs include VVPAT - does anyone know where to find that? To me, optical scanning and a DRE with a VVPAT both get to the same place, no?

https://electionlab.mit.edu/research/voting-technology

Screenshot 2023-11-27 at 17.52.11.png

No, not if the DRE prints out a VVPAT that doesn't match what the machine recorded, especially if the printout doesn't match what the voter selected. Chances are high that a votor might not notice if the VVPAT doesn't match what they voted for.

Yes.

Voting machines add all sorts of unnecessary complexity to what should be a very simple process. They are also not scalable which can be an enormous problem. If you have big crowds at the polls you can't roll out more machines than you already have. But if you are using paper ballots you can pass them out as fast as you can record people in the voting register.

If you want to make the process more electronic, then put a vote scanning machine in every precinct such that when people turn in their completed ballots the election workers put them in the scanning ballot box where they are scanned and then scored securely for later canvassing and recounts. that way you don't have enormous piles of ballots to scan at some central location.

The only good reason for voting machines is for disabled voters who can't use a paper ballot (blind, paralyzed, etc.) In such cases a voting machine is appropriate but it only really needs to be a disabled accessible computer (like a braile computer) that prints out a regular ballot for submission.

[Bootstrap]

To me, though, it seems to reach the same result. You need something that is not connected to the Internet so it can't be hacked. You need both an electronic record and a paper record that users can see and verify. You need ways to audit both. If you have that, I don't really understand why it matters whether you start with paper and scan it in or start with a screen and print it out.

In a county not far from where I live, the electronic voting machines had issues four years ago where the touch-screens were too sensitive and caused all kinds of problems. This year, the paper printouts that the machines printed didn't match the votes that the machines actually registered. So some polling places had to be closed while they tried to figure out the problem, and of course the big question is "Why should the voters trust the manufacturer's claim that the votes were actually recorded correctly?".

Voting machines add all sorts of unnecessary complexity to what should be a very simple process. They are also not scalable which can be an enormous problem. If you have big crowds at the polls you can't roll out more machines than you already have. But if you are using paper ballots you can pass them out as fast as you can record people in the voting register.

If you want to make the process more electronic, then put a vote scanning machine in every precinct such that when people turn in their completed ballots the election workers put them in the scanning ballot box where they are scanned and then scored securely for later canvassing and recounts. that way you don't have enormous piles of ballots to scan at some central location.

The only good reason for voting machines is for disabled voters who can't use a paper ballot (blind, paralyzed, etc.) In such cases a voting machine is appropriate but it only really needs to be a disabled accessible computer (like a braile computer) that prints out a regular ballot for submission.

These are helpful answers. Thanks.