Electronics are superior to paper for many tasks but voting for your share of representation in government ain’t one of them. Without indentifying who voted for who (red flag) and verifying it then you can only hope the software was counting properly at run time and not subverted. You can watch people miscount paper in real-time and call them out.
I actually don’t like this XKCD, this is a bad answer to why electronic voting is bad.
Mostly because it is a largely obtuse process that most people can’t see or understand.
“Trust the magic rock is counting fairly and that bad people who would want to manipulate it in order to obtain power and conquer you haven’t had a chance to do so.” You cannot code your way out of that.
Everyone understands marks on paper. No one understands buffer overrun RCE, resistive touchscreen calibration, or database triggers and log sharding.
While I see what you’re getting at, I still like this XKCD. I work as a developer, and have also worked in more “handy” fields. The thing with planes, elevators, and basically all other physical things is that they’re limited by physics. A steel beam can’t suddenly decide to spontaneously fail or disappear.
With code, that can feel pretty different. With experience, I’ve basically learned to assume that there is always some edge-case I haven’t considered, that could trigger a bug. In a building, you can have redundant bolts, and over-dimensioned supports. A small mistake somewhere, a single missing bolt, won’t cause a catastrophic failure. With code, it’s different: A tiny, hard to notice mistake, can bring the whole think crashing down. Imagine if a plane could crash because the paint had a slightly non-uniform thickness…
And with voting machines there is no verifiable oversight.
You just kind of have to trust that the software that is running on the voting machine is actually correctly tallying your vote, and not doing shenanigans behind the scene. Even if the code is open source, and everyone knew how to read code, you cannot reasonably guarantee that that is the software that is running inside the black box that is a voting machine.
With paper voting you can observe the entire process from start to finish. There are no black boxes which just spit out an answer that you simply have to trust.
The machine can print a human and machine readable copy. Then feed that into another machine after verifying and you have two independent digital counts that can verify each other. You also have the paper that can be manually counted if you need to be extra sure.
This is what we had for a few years. Now the first step is replaced by manual bubbling. Still have the scan for the instant digital count though.
And now you have moved the problem.
Now you need to verify that the verification machine is also running the software you want and has also not been compromised. And you need to do this for every single voting machine in use, all of them.
With trillions of dollars riding on the result of an election, the motive for a group/nation to interfere is immense. Attacks against digital systems scale incredibly well, changing a handful of votes is barely different from changing all of them. This is not the case with physical votes, the more votes you want to change the more people you have to involve, and the likelihood you are caught goes up.
On your point about printing off a human readable copy that can be verified manually, you have now invented the worlds most expensive pencil. You’ll always want to verify the manual copy, so why bother with the computer one?
A fun fact about why pencils are used in voting in the UK is due to paranoia about pens being replaced with ones containing vanishing ink. There is no evidence this exists or has ever been done, but it demonstrates the levels some countries work at to ensure that all votes are accurately counted, and probably so.
On your point about printing off a human readable copy that can be verified manually, you have now invented the worlds most expensive pencil. You’ll always want to verify the manual copy, so why bother with the computer one?
This way the state can count the votes quickly and if there are any audits they have a physical analog to compare. Honestly if the state just randomly audits 1 county per election to check for issues they will catch anything weird going on and save a tremendous amount of time and money.
If the audit discovers something weird they can then count up all the paper ballots and fix the software.
Additionally in 2024 the UK had 28,809,340 votes cast. The US had 158,427,986 votes cast.
I don’t know if I wholly agree. I think there’s something interesting about the idea of using a (standardized, open-source) cryptographic system to prove that your vote was counted without revealing who you voted for.
But that would have to take place in a political system where people were acting in good faith, which is not what America has.
That can never work because the incentive to manipulate that system and then claim the results are fair because of it’s theoretical power of accountability, which it was sold on, cannot be broken and therefore we MUST accept the results is too great.
The moment someone cracks it, they become the new lords. Same problems as the current system.
Paper doesn’t have this problem, because you can see the paper. You can count paper with machines, but you can also verify with humans and - importantly - humans without PHDs.
Voting is the literal keys to the empire. I’m not letting some billionaire vibecode his way into the white house…again
I’m not letting some billionaire vibecode his way into the white house…again
That literally never happened. He simply told people what they wanted to hear and told them who to blame for all of their issues. Those being easily verifiable lies didn’t make a difference to enough people.
There wasn’t a high tech exploit, or even a low tech exploit. Just the human exploit. The same thing that results in 99% of technology “hacks”, social engineering. The human is always the weakest link.
I agree, I live in a country that still uses papel ballouts (Portugal if you’re curious) and on election day representatives from.diferent parties, volunteers, and government officials count the votes by hand, the whole process is bureaucratic, time consuming, and incredibly complex, but that’s exactly what makes it almost foll proof, there is no way you can alter votes in a meaningful way without a lot of people noticing, and even if by some miracle you did, the very next day all the votes get re counted in front of a judge and other set of volunteers chosen randomly.
Different kind of complex 😅 to be honest is more complicated than complex. Believe it or not it is done the old fashion way, we use string and lacker to seal the envelops like in medieval times, and the protocols dictate that all present must sign and write their recount of events. It’s a lot of fun and paper
In the Netherlands we had switched to electronic voting in the past, but we switched back to paper after some very serious security flaws were pointed out. These days there is some discussion on whether electronic counting of paper ballots should be allowed, but at least there is still a paper trail in that case and you could hypothetically double check everything by hand.
Some things are meat to be time consuming and hard. You cannot hurry things like learning or maturity, our current world has convince us that faster is always better, but that is not always the case. I always volunteer for voting counting and despite the fact that it takes from 7am to 10 pm on a Sunday I enjoy doing it, you feel like you are part of the cogs of democracy, doing your part to ensure the democracy works as intended, like you take part In something greater than yourself, you set aside your political filiation and do your part
I think using electronic counting (of paper ballots) can be an acceptable way to speed up how soon we get to know what the result will likely be after the polls close. But it is important that there always happens a manual count, and that that manual count should ultimately be the answer we trust.
The idea of using zero-knowledge proofs to avoid identifying voters has been proposed. Don’t know if this is just theoretical at this point or if machines implement it, but that would solve the issue you mention.
The problem is not that its impossible to establish trust in an electronic voting system. A qualified individual with the necessary knowledge in formal verification, cryptography, and computer science (maybe I am missing a field or two here) might be able to audit a system and verify that it adheres to certain standards and criteria.
But I cannot do that, the average adult certainly cannot, and the bottom 5% percentile (of whatever criterion/metric might be applicable here) is so far removed from the problem that they are probably already having trouble operating such a machine.
We were able to organize our own elections in elementary school to elect class representatives, and every kid understood how they work, and was able to observe the election process themselves, establishing trust in the system. If I have any doubts if my vote is going to be counted correctly in an election, I can go to my polling station and monitor the election as an independent observer or join the election board and do the counting myself. Every citizen eligible to vote has all the necessary tools available, both in terms of access to the polling station and counting of the ballots (which is public), and in terms of mental capacity and required prior knowledge. (Well, the last two points at least apply to a large majority of voters). I don’t need to trust the local government or dubious “experts”. The public’s ability to establish trust in the election system is essential in a democracy, and establishing trust cannot be delegated.
Electronics are superior to paper for many tasks but voting for your share of representation in government ain’t one of them. Without indentifying who voted for who (red flag) and verifying it then you can only hope the software was counting properly at run time and not subverted. You can watch people miscount paper in real-time and call them out.
My state prints out a physical slip that shows the vote that’s deposited into a box. It’s tied to your electronic vote. Ideal system imo.
We get the speed and convenience of electronic voting and there’s a secure analog form to audit
as xkcd put it:
I actually don’t like this XKCD, this is a bad answer to why electronic voting is bad.
Mostly because it is a largely obtuse process that most people can’t see or understand.
“Trust the magic rock is counting fairly and that bad people who would want to manipulate it in order to obtain power and conquer you haven’t had a chance to do so.” You cannot code your way out of that.
Everyone understands marks on paper. No one understands buffer overrun RCE, resistive touchscreen calibration, or database triggers and log sharding.
While I see what you’re getting at, I still like this XKCD. I work as a developer, and have also worked in more “handy” fields. The thing with planes, elevators, and basically all other physical things is that they’re limited by physics. A steel beam can’t suddenly decide to spontaneously fail or disappear.
With code, that can feel pretty different. With experience, I’ve basically learned to assume that there is always some edge-case I haven’t considered, that could trigger a bug. In a building, you can have redundant bolts, and over-dimensioned supports. A small mistake somewhere, a single missing bolt, won’t cause a catastrophic failure. With code, it’s different: A tiny, hard to notice mistake, can bring the whole think crashing down. Imagine if a plane could crash because the paint had a slightly non-uniform thickness…
Every system can be subverted. Ballot boxes or stacks of votes stuff or lost… The integrity is in the checks and balances, verifiable oversight.
And with voting machines there is no verifiable oversight.
You just kind of have to trust that the software that is running on the voting machine is actually correctly tallying your vote, and not doing shenanigans behind the scene. Even if the code is open source, and everyone knew how to read code, you cannot reasonably guarantee that that is the software that is running inside the black box that is a voting machine.
With paper voting you can observe the entire process from start to finish. There are no black boxes which just spit out an answer that you simply have to trust.
The machine can print a human and machine readable copy. Then feed that into another machine after verifying and you have two independent digital counts that can verify each other. You also have the paper that can be manually counted if you need to be extra sure.
This is what we had for a few years. Now the first step is replaced by manual bubbling. Still have the scan for the instant digital count though.
And now you have moved the problem. Now you need to verify that the verification machine is also running the software you want and has also not been compromised. And you need to do this for every single voting machine in use, all of them.
With trillions of dollars riding on the result of an election, the motive for a group/nation to interfere is immense. Attacks against digital systems scale incredibly well, changing a handful of votes is barely different from changing all of them. This is not the case with physical votes, the more votes you want to change the more people you have to involve, and the likelihood you are caught goes up.
On your point about printing off a human readable copy that can be verified manually, you have now invented the worlds most expensive pencil. You’ll always want to verify the manual copy, so why bother with the computer one?
A fun fact about why pencils are used in voting in the UK is due to paranoia about pens being replaced with ones containing vanishing ink. There is no evidence this exists or has ever been done, but it demonstrates the levels some countries work at to ensure that all votes are accurately counted, and probably so.
This way the state can count the votes quickly and if there are any audits they have a physical analog to compare. Honestly if the state just randomly audits 1 county per election to check for issues they will catch anything weird going on and save a tremendous amount of time and money.
If the audit discovers something weird they can then count up all the paper ballots and fix the software.
Additionally in 2024 the UK had 28,809,340 votes cast. The US had 158,427,986 votes cast.
I don’t know if I wholly agree. I think there’s something interesting about the idea of using a (standardized, open-source) cryptographic system to prove that your vote was counted without revealing who you voted for.
But that would have to take place in a political system where people were acting in good faith, which is not what America has.
That can never work because the incentive to manipulate that system and then claim the results are fair because of it’s theoretical power of accountability, which it was sold on, cannot be broken and therefore we MUST accept the results is too great.
The moment someone cracks it, they become the new lords. Same problems as the current system.
Paper doesn’t have this problem, because you can see the paper. You can count paper with machines, but you can also verify with humans and - importantly - humans without PHDs.
Voting is the literal keys to the empire. I’m not letting some billionaire vibecode his way into the white house…again
That literally never happened. He simply told people what they wanted to hear and told them who to blame for all of their issues. Those being easily verifiable lies didn’t make a difference to enough people.
There wasn’t a high tech exploit, or even a low tech exploit. Just the human exploit. The same thing that results in 99% of technology “hacks”, social engineering. The human is always the weakest link.
I agree, I live in a country that still uses papel ballouts (Portugal if you’re curious) and on election day representatives from.diferent parties, volunteers, and government officials count the votes by hand, the whole process is bureaucratic, time consuming, and incredibly complex, but that’s exactly what makes it almost foll proof, there is no way you can alter votes in a meaningful way without a lot of people noticing, and even if by some miracle you did, the very next day all the votes get re counted in front of a judge and other set of volunteers chosen randomly.
If you think vote counting is conplex, just wait until you hear about embedded microcontrollers
Different kind of complex 😅 to be honest is more complicated than complex. Believe it or not it is done the old fashion way, we use string and lacker to seal the envelops like in medieval times, and the protocols dictate that all present must sign and write their recount of events. It’s a lot of fun and paper
In the Netherlands we had switched to electronic voting in the past, but we switched back to paper after some very serious security flaws were pointed out. These days there is some discussion on whether electronic counting of paper ballots should be allowed, but at least there is still a paper trail in that case and you could hypothetically double check everything by hand.
Some things are meat to be time consuming and hard. You cannot hurry things like learning or maturity, our current world has convince us that faster is always better, but that is not always the case. I always volunteer for voting counting and despite the fact that it takes from 7am to 10 pm on a Sunday I enjoy doing it, you feel like you are part of the cogs of democracy, doing your part to ensure the democracy works as intended, like you take part In something greater than yourself, you set aside your political filiation and do your part
I think using electronic counting (of paper ballots) can be an acceptable way to speed up how soon we get to know what the result will likely be after the polls close. But it is important that there always happens a manual count, and that that manual count should ultimately be the answer we trust.
The idea of using zero-knowledge proofs to avoid identifying voters has been proposed. Don’t know if this is just theoretical at this point or if machines implement it, but that would solve the issue you mention.
It’s been proposed for age verification too, but shockingly it doesn’t seem to be a priority.
ZKP solves the math problem.
It does not solve the trust problem.
Well the implementation would still need to be audited, but you could argue this is more objective and practical than overseeing manual counting.
The problem is not that its impossible to establish trust in an electronic voting system. A qualified individual with the necessary knowledge in formal verification, cryptography, and computer science (maybe I am missing a field or two here) might be able to audit a system and verify that it adheres to certain standards and criteria.
But I cannot do that, the average adult certainly cannot, and the bottom 5% percentile (of whatever criterion/metric might be applicable here) is so far removed from the problem that they are probably already having trouble operating such a machine.
We were able to organize our own elections in elementary school to elect class representatives, and every kid understood how they work, and was able to observe the election process themselves, establishing trust in the system. If I have any doubts if my vote is going to be counted correctly in an election, I can go to my polling station and monitor the election as an independent observer or join the election board and do the counting myself. Every citizen eligible to vote has all the necessary tools available, both in terms of access to the polling station and counting of the ballots (which is public), and in terms of mental capacity and required prior knowledge. (Well, the last two points at least apply to a large majority of voters). I don’t need to trust the local government or dubious “experts”. The public’s ability to establish trust in the election system is essential in a democracy, and establishing trust cannot be delegated.
No, you could not. In any way.