[falafelite]

A lot of skepticism and negativity here, with a focus on "open source means something to me, not the layman" as well as mentions of is that really helpful, is it auditable, etc. Valid concerns, but I guess my reaction is hopeful because hey there's at least one group of people who are trying to build something that is more transparent.

Maybe it's not perfect, but (at least to me) seems like a step in a good direction. Also for what it's worth the company making the tech has a product for auditing as well. Idk guess I'm feeling more optimistic this morning.

[DoingIsLearning]

> Maybe it's not perfect, but (at least to me) seems like a step in a good direction.

Is it? Provided you have a working democratic state with checks and auditing, what is the problem that voting machines solve?

The level of guarantees that you have with a paper vote is hard to surpass and the inconvenience of paper vote is not that big considering how often voting occurs (even in the extreme case of Swiss style referendums).

[davoneus]

>Provided you have a working democratic state with checks and auditing, what is the problem that voting machines solve?

It solves the problem of waiting days or weeks for a result. It solves the problem of multiple languages on a ballot. It solves the problem of visually-impaired voters accessing the voting machine, it solves the problem of incorrectly filled out voting sheets (hanging-chat).

Hell, that's just off the top of my head, and I'm sure others can contribute more.

[falafelite]

Is a paper-vote a reference to literally marking a piece of paper, or a human counting said piece of paper? Sorry if that's a dumb question, just want to make sure I follow. It seems like this tech works with marking a piece of paper that is then read by a machine. I suppose if the paper is kept, perhaps it's nice from the point of view of the election-handling-folks to automate the counting process, but be able to audit the paper ballots?

[tokai]

Voting machines are always a bad idea. There are issues of trust and security that can never be solved. So trying to open-wash this dangerous way of voting is at best just as bad as closed machines and at worst an angle to make people think its a safe and good idea.

[kube-system]

But it is a solved problem. You can use machines to mark paper ballots, and to quickly tabulate them. This is the best of all worlds: the results are auditable, the user can confirm their ballot was recorded properly, and the election officials can quickly count the ballots.

[falafelite]

Why are they always a bad idea in your view? I'd imagine that they're quite nice from the point of view of those who run elections and actually have to count them. I'd love to hear your thoughts!

[yifanl]

Consider: the number of available volunteers to count votes scales directly with the number of voters in a given electorate.

[Loughla]

Yes and no. There is never a situation where electronic voting with no paper back up that must be stored and catalogued for a prescribed time via law is a good idea.

But a system that uses electronics to tabulate votes that can be verified via paper ballots that are stored long-term, securely? Why not?

Edit: Maybe I didn't describe this well. The person makes the vote on paper. The paper is counted by a machine (like the article is saying). The paper is stored securely and catalogued for later reference and audit. What is the problem there?

[nyokodo]

> But a system that uses electronics to tabulate votes that can be verified via paper ballots that are stored long-term, securely? Why not?

If the paper ballots as backup are manually created by the voter then while it’s still possible it’s much harder to fake the votes en masse but there’s more scope for human error (what if they vote in two different ways). If the paper backup is automatically generated then can they check it? If not then there’s little improvement over purely digital voting. If yes, do they actually check it? Many won’t bother and maybe there’s an exploit there. Then there’s the fact that this system would require regular auditing, and lawsuits and close contests will force a certain number of audits every election. No one can reliably predict which districts will require audits so presumably they’ll need to hire sufficient people to do a manual recount anyway which eliminates the labor cost advantage.

If by using electronic voting we are opening up new potential exploits, even with paper backups, and not really gaining much of an advantage why would we go to the expense and bother of implementing electronic voting?

[Loughla]

Sorry, when I say backups, I mean paper voted created by the voter's own hand as the original vote, counted by machine, and held securely as a way to verify the vote if any of the candidates call the election into question.

There is no system in which electronically cast and electronically created backups are a good idea.

[nyokodo]

> when I say backups, I mean paper voted created by the voter's own hand as the original vote

Fair enough, however the issue of the thread is electronic voting machines not electronic paper vote tabulation.

[Loughla]

The story is about electronic paper vote tabulation.

[lm28469]

> Maybe it's not perfect, but (at least to me) seems like a step in a good direction.

It just add bigger SPOF. Compromise a single voting machine and you control hundred thousands of votes. Compromise one vote counter and you control thousands at most

There are literally no problems to solve in modern functioning democracies when it comes to votes, it's just some technocrat mentality that requires everything to be automated so it's faster/more efficient, etc.

So yeah you can make these things faster at the expense of basically everything else, including trust

[falafelite]

Interesting okay. Is a combo of paper and automated counting not ideal then? To my mind automatic counting is probably great for the speed you mentioned, and I'm sure those who have to perform the counting/run the election like it, but then keeping paper to hand-count as a backup or to verify seems sensible too. Or is the very possibility of a compromise too bad to entertain?

[simiones]

Speed of counting is not a problem. Most countries in the world finish counting their election votes by the end of the second day after the election, with pure paper voting. The great thing about elections is that election officials can very easily scale with the number of votes, so there is no reason why you'd need automation here.

[carapace]

If there is a CPU between you and your recorded vote you live in a "hackerocracy": you are ruled by the best hackers.

The only reason to have electronic voting machines is to subvert the voting process.

[simiones]

"Open source" voting machines is essentially like baby-proofing a machine-gun. All of the problems that electronic voting poses to true democracy are 100% still there, but it looks safer - which may, unfortunately, make it more attractive.

I'm putting open source in quotes, since only software can be open source, a voting machine is a piece of hardware that you do not own, so you have no idea what it software it is running.

[eru]

I wrote some skeptical remarks.

But yes: this is definitely progress compared to closed source machines.

[ianai]

Yes the negativity is stark and dangerous given the other conditions this year.