…
The push to hand-count ballots is ramping up, albeit with spotty success, as the 2024 election nears, according to a review by the Guardian and Votebeat. If more localities decide to try hand-counting in the November election, results could be inaccurate, untrustworthy or delayed, fostering more distrust in elections. In places that opt not to hand-count, supporters of the practice could use this choice as a reason to question or refuse to sign off on certification.
Either way, it raises the risk of throwing the 2024 election into chaos.
“It just gives additional grounds for calling into question the results of elections when there are no valid grounds,” said Heather Sawyer, executive director at American Oversight. “There’s no good reason to do it. And there’s lots of room for mischief and problems.”
The push hasn’t gained much ground in the large swing counties where Trump claimed votes were stolen from him. It’s been more effective in small or rural counties that voted heavily for Trump, where conservative activists have lined up at public meetings to repeat the conspiracies of Cook, Lindell and others. There – in Missouri, Nevada, Pennsylvania, Texas and Wisconsin – local officials voted to give hand-counting ballots a try in either their midterm or presidential primary elections.
I’ve volunteered as a candidate representative during Canadian federal elections. We hand count every ballot. We have representatives from each candidate present, so we can ensure we’re all in agreement about each ballot. If there are general concerns about the polling station or riding, it’s very easy to conduct a recount. It works extremely well.
It’s unfortunate that electronic voting is creeping into our municipal elections. The only advantage that I can see is that the polling station produces results an hour or two faster.
You can easily do both. Get a scanner to tally the results then pick like 1000 ballots (or some statistically significant number) to hand count. If the hand counted results are the same then you can be confident in the automatic count.
What’s the value in the automatic count?
I’ve only volunteered at urban polling stations. Each polling station has something like five or six polls. Each poll has something like two hundred possible voters (with half of them bothering to vote). We got them counted in an hour or two.
I don’t understand the push for electronic voting. Paper ballots are great in so many ways. The only drawbacks I’m aware of are the time to count (which is negligible) and the risk that voters will mark ballots so they can get paid (mitigated by discarding ballots that are marked with anything other than an x).
That isn’t urban then. No mention of mail in either?
Theoretically, electronic polling should generate a near instand result with no counting errors. How many legal documents can you file online nowadays? There never seems to be an issue with those “getting lost” or “duplicated.” I think back to a USA election a few years ago where boxes of ballots were recounted several times until the supreme Court declared a winner. Votes stored in a secure database wouldn’t just “get lost” or “get miscounted.”
Few legal documents have the anonymity requirements of ballots.
The problem is that voters need to trust all of the devices between the voter, the database, and the person who declares the winner. With software, that’s hard. How do you convince Jane Public that the iPad is reporting accurately across some Verizon network to the IBM Db2 instance which is telling the truth to the state election official? Even when it’s working perfectly, it’s opaque to most people.
The great thing about paper ballots is that representatives of all interested parties can watch the whole thing and audit the results. Representatives of each candidate are present at the ballot counting, and they forward the counts to the campaign HQ. It’s easy to watch and explain.
I’m not sure how an electronic system can provide that level of openness.
It shouldn’t be too hard to have two separate databases. One with personal identifying info attached and one without. It could even allow voters to look up their vote later if they wanted to confirm how it was cast. The database without identifying info could be made public so people could compare results on their own vote counting software.
There’s always a relevant xkcd:
How many people who are pissed with the results might claim an alternate cast vote later on just to bring the machines into question? It wouldn’t take many to do this to create a fair amount of distrust.
Opposed too now? People already claim the votes are counted wrong.
Perception. People trust a vote that is more transparent and completed by two independent people.
Personally I trust the electronic one entirely but I could see it a bug occurring that puts some future election results in question. Would you trust it if an AI wrote the program?
Theoretically, but in actuality the software is written by incompetent companies and there’s a much higher potential to interfere maliciously.
Hand counting our elections is worth the effort… even if this particular push is a disingenuous effort to muddy the election results. This is a “broken clock is right twice a day” situation.
Not necessarily true. Sure code is easy to mess up, source I work as a software engineer, but it’s also easy to proofread. An open source government sponsored vote counting software could easily be implemented. Heck the data base without personal identifying information could be made public for people to compare results on local builds of the software.
As a fellow software engineer, please realize that open SSH is used by pretty much everyone and had several severe security issues for decades. Open source software is much more secure than closed source but if the pool of people reviewing the software in detail is small and the stakes are high then it’ll be much cheaper for a foreign actor to expend 10 million per person for bribery or blackmail for a few dozen people then trying to infiltrate hundreds of municipalities.
I make software, I’ve been in this industry for decades and it’s awesome - this isn’t a problem space where it’s a good solution.
Voting machines seem like a solution in search of a problem. Yes. We can do it, maybe we can even do it well. But that doesn’t make it better than paper ballots.