Like clockwork. Literally every US voting season, ever since Diebold machines became fixtures. I tire of seeing these problems make headlines yet again when a technical solution exists.
I highly recommend HBO’s new documentary on election security. It’s called ‘Kill Chain: The Cyber War on America's Elections’ - really opened my eyes to the low level details of how bad our system is. Scary stuff
One party is fixated on voter id fraud and ignores the cyber security, the other party i fixated on cyber security and ignores voter id fraud.
Both are significant issues which we should be addressing. An accurate election results should be in everyone's interest.
Are they really both significant issues? A successful cyber attack could enable a single individual to decide multiple elections. A voter id attack could enable at most, what, 10 votes per conspirator? An undetected voter id fraud changing the outcome of an election seems absurd.
Lack of voter ID, combined with mail-in voting, can lead to fraudulent “ballot harvesting,” where boxes of votes are “found” at the last minute, for example. Also, many people aren’t aware that ballot harvesting is actually legal in some jurisdictions including California. [0]
"Went missing" makes this sound way more nefarious than it is. It just means that the ballot wasn't returned (e.g., the person didn't vote. I vote by mail and there are elections where my ballot would have gone missing because I didn't have the time to make myself well informed enough about the candidates to place an informed vote.) I don't exactly see what's so implausible about this.
Hard to do without getting caught, though. Still one ballot per person, so if you submit a ballot on behalf of someone who also tries to vote for themselves, the double vote will get flagged.
Even assuming 1 person could make it through 10-20 precincts per day, you're going to need thousands of attackers to change even a close national election.
Coordination on that scale is completely impossible to pull off without detection. This just isn't a feasible attack.
If there are enough double votes that throwing them all out changes the election. Then yes, do-over. The possibility of that happening is basically non-existent though because it would require a coordinated attack by thousands of people to change a national election (even assuming they strategically attack swing states).
There's a virtually zero chance that a group effort involving thousands of people isn't uncovered before the attack happens.
You invalidate duplicates and notify the real voter they need to re-cast their ballot. You then hand off the evidence to the police for prosecution.
But that is getting ahead of ourselves. The problem isn't how do you solve duplicate voting, it is how do you detect it. That is a solved problem, already, and the answer is that there are very few attempts at voter fraud in a vote-by-mail system.
Now, the electronic machines are a whole 'nother matter. I would fight hard against my state implementing such a disaster. Vote by mail is great.
Given that last election was actually decided in just a few states, by a low margin, voter ID attack sounds feasible, given well founded and determined adversary.
Of course cyber hacking is likely easier, but it’s not fair to fully dismiss even possibility of other attacks.
Election hacking is an actual, imminent national security threat. Intelligence agencies are scrambling to prevent or mitigate it right now.
In-person voting fraud on the other hand, as a meaningful threat, is entirely hypothetical.
Imagine if, at the height of the Cuban missile crisis, someone said “ok Soviet missiles within reach of the east coast is a threat; but what if vegan extremists blew up the white house tomorrow? Theoretically it could happen, so let’s not dismiss the threat!” That’s how different those two threat profiles actually are.
I do think majority of the effort should be spent on protection from hacking, but dismissing other very real attack vector as entirely hypothetical is just naive.
Last elections were decided by 80k votes, in a country with 250M eligible voters. With such a tight margins we have a lot of valid attack vectors, not only most obvious hacking ones.
Effective organizations are (and should be) able to deal with many risks at once. Do you think that during Cuban missiles crisis all other investigations stopped, and 100% of CIA, FBI, Police, Army etc focused on Cuba?
>With such a tight margins we have a lot of valid attack vectors
In person voter fraud--the kind stopped by requiring poll workers to check ID--isn't one of them. You would need thousands of coordinated attackers to change 80k votes. It's just not a feasible attack.
Pretty much every country in the world required voter-id. Even India with a population of over 900 million voters, securely counted over 600 million votes which was the largest turnout ever few months ago.
Get real. Any credible research on the topic of "voter fraud" suggests it's not a significant issue and efforts to combat "voter fraud" are really just disguised efforts to suppress voting by low income and minority voters.
There is no evidence that voting machines have been hacked, either.
Why is the standard of evidence for one different than the other?
And more to the point, since when does patching a vulnerability require evidence of exploitation. We wouldn't accept Microsoft saying "There's no evidence of this Outlook vulnerability being exploited in the wild, we're not going to fix it". We certainly shouldn't accept that reasoning for our elections.
>Why is the standard of evidence for one different than the other?
Because hacking one voting machine potentially allows one person or a small group of people to alter thousands of votes.
That person (or persons) has no way to alter anyway near that many votes through in person voter fraud. The amount of effort it would take to alter the results of an election through in person voter fraud makes it extremely difficult to pull off--particularly without detection.
Additionally there are downsides to voter id laws. The most troubling is demonstrably lower minority turn out. There are no similar downsides to methods like requiring paper receipts for voting machines.
Basically the cost benefit analysis is in favor of hardening electronic voting, but not in favor of voter id laws.
Detection is much more gnarly than you give it credit. A pretty likely scenario is that one side will claim that fraud was detected, and the other side will deny it. And both side will try to suppress and deplatform each other.
Youtube, twitter, facebook will pick a side. And whoever loses will fume with resentment. Maybe they pick up weapons. If they are right that the election was stolen and that fact is suppressed, they will certainly be justified. If they are right - which is a big hypothetical. Because the detection will be drowned in noise.
You're talking a coordinated conspiracy involving thousands of people. The FBI would have hard evidence during the recruiting phase.
You can't involve that many people in something and keep it secret.
Even in the best case were we don't find out until elections day, an enormous increase in double votes will be all over the news the same day, and arrests will start happening shortly after.
It's a felony. A felony that requires you to commit it in person at a precinct is a pretty high bar for an internet troll to get people to jump.
Assuming some people decide to take them up on it, they have to coordinate or they'll end up voting as people who have already voted. Double voting is very easy to spot and many of them will be arrested at the precinct.
Plus you'd need an accurate list of close to 100k dead voters who haven't already been purged (more in a less tight election), and thousands of people willing to drive to the right precincts to match the list.
And we didn't know which states would tip the election until election night in 2016, so you'd need to duplicate your efforts several fold.
It's difficult to pull off in a coordinated way. But coordination attacks are not the only thing that matters.
It shouldn't be possible for a liberal kid's conservative parents who know he doesn't vote to vote for him. But it's trivially possible, right now. There is literally nothing stopping anyone from doing that.
>There is literally nothing stopping anyone from doing that.
Other than the fact that it's a serious felony.
Additionally how are voter ID laws going to change this? It already requires some level of cooperation by the kid in that they don't show up and vote themselves.
If you're talking about absentee ballots, whatever you're using to identify the kid, the parents likely have access to--copy of id, birth certificate etc... So voter ID isn't going to stop this either.
>But coordination attacks are not the only thing that matters.
Coordination of some kind is required to change the results of an election. If even a few thousand random parents around the country decide to risk jail time to vote for their kid, it isn't going to impact the results of the election, so hardening against this kind of attack is a very low priority. Particularly when there's no cost effective means to stop it.
The standard has nothing to do with whether the exploit has been used but whether it is possible. There is no credibly attack an adversary could carry out against the election by exploiting the lack a of voter ID requirement. There are however many proof of concept attacks performed against voting machines every time they are introduced.
No, there is no credible way that a single actor could carry out a large scale attack that way. But there are credible ways that individuals can vote illegally extremely easily, and would never get detected.
Secondly, we have no reliable data on how often it occurs, precisely because it's more or less undetectable. There are documented cases of it, that have been caught by luck. And people have tried to do surveys that are probably worthless and unreliable.
It’s not undetectable at all. When you go to vote, you give your name and address, and get crossed off a list. If someone did that fraudulently, it would be noticed when the person who’s identity you stole comes to vote, it is obvious.
There is a good reason not to require it. Not everyone who is eligible to vote has an id, or a straightforward way to get one.
> It’s not undetectable at all. When you go to vote, you give your name and address, and get crossed off a list. If someone did that fraudulently, it would be noticed when the person who’s identity you stole comes to vote, it is obvious.
Sure. But in most cases where someone would do this, they'd do it knowing the person, and knowing they weren't going to vote, for whatever reason.
Also people keep claiming that over iD is somehow suppressing voting for low income and minority voters are delusional. US is one of the only countries which doesn’t have voted id. India which has very strict voter iD laws and yet out of the 900 million eligible population, over 600 million votes were counted with voter iD few months ago which was also the largest turnout ever for any democratic election in the world. So no, voter iD and efforts to combat voter fraud are not suppressing votes. That’s just a talking point with zero backing.
That is an orthogonal issue. Turnout is unrelated to security. Identifying voters, and ensuring the integrity of the voting mechanism are both important security issues.
> That is an orthogonal issue. Turnout is unrelated to security
We are both talking about election security.
The reason we are talking past each other is that you’re describing a neat hypothetical model that fits your belief of what is worth discussing. Whereas I’m describing a nexus of actual threats, carried out by actual attackers, in the context of an overall strategy. You’re welcome to ignore the full context of those threats, but that will result in your model being worthless.
> Identifying voters, and ensuring the integrity of the voting mechanism are both important security issues.
A meaningless statement without clarifying the context.
Let’s continue your list of possible threats to the integrity and fairness of the US electoral system:
- Voter ID fraud
- Hacking of voting machines
- Hacking of the tally system
- Hacking of government contractors developing and operating the election infrastructure
- Illegal purging of voter rolls
- Corruption of election officials
- Voter intimidation
- gerrymandering
- Targeted shuttering of voting stations
- Targeted disinformation in crucial counties
- Exfiltration of polling data for precise targeting of attacks
- Hacking of political parties (headquarters and campaign staff)
All these threats are being carried out right now against the US election system, in a coordinated fashion, by the same nexus of attackers. They are interlocking attacks supporting the same goal: minority rule and a transition to an authoritarian regime. This threat is as serious as it gets.
Of course there is one exception in the list: voter ID fraud, in person or by mail, is not a threat. No serious election security expert worries about it. That would be like worrying about a Martian invasion in the middle of WW2: there are more pressing and realistic concerns to worry about... There is no serious coordinated effort to commit such fraud, and as far as I know, there never has been.
Hopefully this context will help you understand why you are mistaken in framing election security around the non-issue of voter ID fraud.
> The reason we are talking past each other is that you’re describing a neat hypothetical model that fits your belief of what is worth discussing. Whereas I’m describing a nexus of actual threats, carried out by actual attackers, in the context of an overall strategy. You’re welcome to ignore the full context of those threats, but that will result in your model being worthless.
We haven't been talking at all. I'm not the grandparent commenter. But more to the point, let's apply your security model to software. How would you feel about it if you reported a 0day to Microsoft, and they said "Listen, there's no evidence of this being exploited in the wild, we're not going to patch it".
> All these threats are being carried out right now against the US election system, in a coordinated fashion, by the same nexus of attackers. They are interlocking attacks supporting the same goal: minority rule and a transition to an authoritarian regime. This threat is as serious as it gets.
> Of course there is one exception in the list: voter ID fraud, in person or by mail, is not a threat. No serious election security expert worries about it. That would be like worrying about a Martian invasion in the middle of WW2: there are more pressing and realistic concerns to worry about... There is no serious coordinated effort to commit such fraud, and as far as I know, there never has been.
You are setting up a straw man argument, and knocking it down. Good for you. Integrity is about more than just protection from coordinated assault. It is also about ensuring the system functions as it is intended to.
No, you probably cannot, as an individual actor, systematically throw an election via voter impersonation. But that doesn't mean that the practice cannot alter outcomes. And it certainly doesn't mean that we should accept a low grade, persistent level of fraud that is trivially preventable.
One person cheating on the SAT isn't going to bring down our entire education system. But you have to show ID for that, because it matters. I'd say our elections matter quite a bit more than the SAT does.
> How would you feel about it if you reported a 0day to Microsoft, and they said "Listen, there's no evidence of this being exploited in the wild, we're not going to patch it"
But in-person voter ID fraud is not a 0day. The electoral system is not vulnerable to it. There is nothing to patch.
Meanwhile the made-up threat of ID fraud is used as a bad faith argument to justify actual attacks on the election system, taking place right now.
The “what about id fraud” narrative is part of the attack. By amplifying it you are indirectly providing cover to the attackers, while offering nothing of value to the conversation about election security.
1. Both parties are so similar that there’s essentially only one party.
2. You don’t know anything about US politics yet are compelled to share your opinion anyway.
3. You know exactly which party is which, but your political allegiances are in conflict with that reality. Instead of confronting the dissonance within you, you seek to hide it with a lie.
when I can apply for a mortgage, car loan, credit card, and all kinds of other financial legally binding contracts online in complete security?
I enter my SSN, my credit report is pulled.
Why can't I enter my SSN, it gets validated in a database on whether I'm alive/have already voted/am currently a legal citizen eligible to vote, and vote?
We have achieved problems way, way, way harder than "securely identify a person based on their SSN and allow them to vote A, B, or C on a subject"
> when I can apply for a mortgage, car loan, credit card, and all kinds of other financial legally binding contracts online in complete security? I enter my SSN, my credit report is pulled. Why can't I enter my SSN, it gets validated in a database on whether I'm alive/have already voted/am currently a legal citizen eligible to vote, and vote?
Come back once onlyyou can do all those things on your behalf, and not anyone else having your SSN or other PII.
But that’s a standard that even paper ballots don’t meet.
In my state you walk into the polling place, give them your name and address, sign your name, and that’s it. Your polling place is even listed publicly online for anyone who knows your name.
"It's 2020" is not an argument. Show us a better system that stays anonymous, I don't care which year is it. All I can find so far is examples of technologies going wrong for voting.
At some point the data hits a computer system, and that will be a vulnerability. You can always go back the the scans, but what if the scanners themselves are compromised? You could require a certain sample size to be manually counted and cross-referenced with the scans to be increase confidence, but I'm not sure there's a clear sweet spot, or maybe it's that I think your sweet spot needs some elaboration: Filled out on paper & scanned, sure, but then every digital link from scanning to tabulation and data storage should be designed to specs similar a similar level of security used in other very high security computing environments.
>You can always go back the the scans, but what if the scanners themselves are compromised? [...]
I think what the parent poster meant by "scan" was "tabulate". ie. you fill out a paper ballot, you put it into a scanner which scans it and records the result, and the ballot itself drops into a ballot box. It's still possible to hack the scanner to give a fake result, but it's easily auditable/detected since you still have the original ballots. You can even have the tabulators be entirely offline (since all they do is spit out a number), and their results can be easily cross-checked by random sampling.
Paper ballots have this wonderful property where basically any voter can reasonably understand what is going on and even be a precinct observer and be reasonably sure relying on first hand evidence that the election was fair.
No other system has this property. Even if you had a formally verified election system, it and the proofs of it could perhaps only be understood by less than 100 people in the world.
Yes, I think that's critical. Anyone can understand the security of paper ballots. And if there is fraud, people can understand that too. There have been incidents of ballot boxes hidden, and when it came to light the public could understand what happened and what a reasonable remedy would be. But if there's a hack of an e-voting system, how's a normal person supposed to understand it?
Here in Germany, you’re allowed to stick around and watch the counting up close. I’ve done so, and was able to count along the votes for the two major parties.
Six hours later, I checked the spreadsheet on the official website, scrolled to the row for my polling location, and verified these two numbers were identical to my count.
Sure, I can only check 1/3rd or so in one of 50,000 polling places. But get just 20 people you trust to do the same, and it becomes statistically unlikely that there is widespread fraud.
You can hang out in the polling location for the whole day if you want. If you get there before any voters, you can also check that the ballot box is empty. Each location serves less than 1,000 voters with three volunteer workers, so you’ll always be able to know everything that’s going on. Oh, and the longest wait I’ve ever had to endure myself was less than 5 minutes
The only possible problem (aside from Republicans) in the US would be the large number of individual races and ballot initiatives. It works well with five or so. But if you’re voting for ten deputy dog-catchers of the peace, it might get tricky.
Good that you did that! But for most locations that is not done. See the last NRW elections where it was later discovered that votes of the voldemord party have been zeroed in many locations. Or the endless stories of tampered mail votes.
I'm not against paper voting at all, just saying that they aren't automatic tamper proof.
That's why the public observability is so important. Hard to transmit the wrong result if multiple people representing different interest were there and saw the actual result.
75 comments
[ 2.3 ms ] story [ 143 ms ] thread[0] https://en.wikipedia.org/wiki/Ballot_harvesting
https://www.realclearpolitics.com/articles/2020/04/24/28_mil...
https://www.whitehouse.gov/sites/whitehouse.gov/files/docs/p...
Coordination on that scale is completely impossible to pull off without detection. This just isn't a feasible attack.
There's a virtually zero chance that a group effort involving thousands of people isn't uncovered before the attack happens.
But that is getting ahead of ourselves. The problem isn't how do you solve duplicate voting, it is how do you detect it. That is a solved problem, already, and the answer is that there are very few attempts at voter fraud in a vote-by-mail system.
Now, the electronic machines are a whole 'nother matter. I would fight hard against my state implementing such a disaster. Vote by mail is great.
Of course cyber hacking is likely easier, but it’s not fair to fully dismiss even possibility of other attacks.
Election hacking is an actual, imminent national security threat. Intelligence agencies are scrambling to prevent or mitigate it right now.
In-person voting fraud on the other hand, as a meaningful threat, is entirely hypothetical.
Imagine if, at the height of the Cuban missile crisis, someone said “ok Soviet missiles within reach of the east coast is a threat; but what if vegan extremists blew up the white house tomorrow? Theoretically it could happen, so let’s not dismiss the threat!” That’s how different those two threat profiles actually are.
Last elections were decided by 80k votes, in a country with 250M eligible voters. With such a tight margins we have a lot of valid attack vectors, not only most obvious hacking ones.
Effective organizations are (and should be) able to deal with many risks at once. Do you think that during Cuban missiles crisis all other investigations stopped, and 100% of CIA, FBI, Police, Army etc focused on Cuba?
In person voter fraud--the kind stopped by requiring poll workers to check ID--isn't one of them. You would need thousands of coordinated attackers to change 80k votes. It's just not a feasible attack.
Why is the standard of evidence for one different than the other?
And more to the point, since when does patching a vulnerability require evidence of exploitation. We wouldn't accept Microsoft saying "There's no evidence of this Outlook vulnerability being exploited in the wild, we're not going to fix it". We certainly shouldn't accept that reasoning for our elections.
Because hacking one voting machine potentially allows one person or a small group of people to alter thousands of votes.
That person (or persons) has no way to alter anyway near that many votes through in person voter fraud. The amount of effort it would take to alter the results of an election through in person voter fraud makes it extremely difficult to pull off--particularly without detection.
Additionally there are downsides to voter id laws. The most troubling is demonstrably lower minority turn out. There are no similar downsides to methods like requiring paper receipts for voting machines.
Basically the cost benefit analysis is in favor of hardening electronic voting, but not in favor of voter id laws.
Youtube, twitter, facebook will pick a side. And whoever loses will fume with resentment. Maybe they pick up weapons. If they are right that the election was stolen and that fact is suppressed, they will certainly be justified. If they are right - which is a big hypothetical. Because the detection will be drowned in noise.
You can't involve that many people in something and keep it secret.
Even in the best case were we don't find out until elections day, an enormous increase in double votes will be all over the news the same day, and arrests will start happening shortly after.
Assuming some people decide to take them up on it, they have to coordinate or they'll end up voting as people who have already voted. Double voting is very easy to spot and many of them will be arrested at the precinct.
Plus you'd need an accurate list of close to 100k dead voters who haven't already been purged (more in a less tight election), and thousands of people willing to drive to the right precincts to match the list.
And we didn't know which states would tip the election until election night in 2016, so you'd need to duplicate your efforts several fold.
This idea is pure fantasy.
It shouldn't be possible for a liberal kid's conservative parents who know he doesn't vote to vote for him. But it's trivially possible, right now. There is literally nothing stopping anyone from doing that.
Other than the fact that it's a serious felony.
Additionally how are voter ID laws going to change this? It already requires some level of cooperation by the kid in that they don't show up and vote themselves.
If you're talking about absentee ballots, whatever you're using to identify the kid, the parents likely have access to--copy of id, birth certificate etc... So voter ID isn't going to stop this either.
>But coordination attacks are not the only thing that matters.
Coordination of some kind is required to change the results of an election. If even a few thousand random parents around the country decide to risk jail time to vote for their kid, it isn't going to impact the results of the election, so hardening against this kind of attack is a very low priority. Particularly when there's no cost effective means to stop it.
Secondly, we have no reliable data on how often it occurs, precisely because it's more or less undetectable. There are documented cases of it, that have been caught by luck. And people have tried to do surveys that are probably worthless and unreliable.
https://www.nytimes.com/1998/10/29/us/18-are-arrested-in-199...
https://abcnews.go.com/Politics/OTUS/voter-fraud-real-rare/s...
Indeed, the incidence is low. But it is importantly not zero, and the numbers we know are only the ones who got caught.
Most democracies require ID to vote. It's an anomaly that we do not. There is no good reason not to require it.
https://news.ycombinator.com/item?id=22992159
There is a good reason not to require it. Not everyone who is eligible to vote has an id, or a straightforward way to get one.
Sure. But in most cases where someone would do this, they'd do it knowing the person, and knowing they weren't going to vote, for whatever reason.
Are you implying that having an ID, such as a drivers license, is out of reach of either low income or minority citizens?
That seems classist and/or racist.
Are there any studies supporting the idea that either of those two groups of citizens don't have ID?
https://www.realclearpolitics.com/articles/2020/04/24/28_mil...
https://www.whitehouse.gov/sites/whitehouse.gov/files/docs/p...
Also people keep claiming that over iD is somehow suppressing voting for low income and minority voters are delusional. US is one of the only countries which doesn’t have voted id. India which has very strict voter iD laws and yet out of the 900 million eligible population, over 600 million votes were counted with voter iD few months ago which was also the largest turnout ever for any democratic election in the world. So no, voter iD and efforts to combat voter fraud are not suppressing votes. That’s just a talking point with zero backing.
The other party is fixated on getting more people to vote.
Any other framing of this issue is a distraction.
We are both talking about election security.
The reason we are talking past each other is that you’re describing a neat hypothetical model that fits your belief of what is worth discussing. Whereas I’m describing a nexus of actual threats, carried out by actual attackers, in the context of an overall strategy. You’re welcome to ignore the full context of those threats, but that will result in your model being worthless.
> Identifying voters, and ensuring the integrity of the voting mechanism are both important security issues.
A meaningless statement without clarifying the context.
Let’s continue your list of possible threats to the integrity and fairness of the US electoral system:
- Voter ID fraud
- Hacking of voting machines
- Hacking of the tally system
- Hacking of government contractors developing and operating the election infrastructure
- Illegal purging of voter rolls
- Corruption of election officials
- Voter intimidation
- gerrymandering
- Targeted shuttering of voting stations
- Targeted disinformation in crucial counties
- Exfiltration of polling data for precise targeting of attacks
- Hacking of political parties (headquarters and campaign staff)
All these threats are being carried out right now against the US election system, in a coordinated fashion, by the same nexus of attackers. They are interlocking attacks supporting the same goal: minority rule and a transition to an authoritarian regime. This threat is as serious as it gets.
Of course there is one exception in the list: voter ID fraud, in person or by mail, is not a threat. No serious election security expert worries about it. That would be like worrying about a Martian invasion in the middle of WW2: there are more pressing and realistic concerns to worry about... There is no serious coordinated effort to commit such fraud, and as far as I know, there never has been.
Hopefully this context will help you understand why you are mistaken in framing election security around the non-issue of voter ID fraud.
We haven't been talking at all. I'm not the grandparent commenter. But more to the point, let's apply your security model to software. How would you feel about it if you reported a 0day to Microsoft, and they said "Listen, there's no evidence of this being exploited in the wild, we're not going to patch it".
> All these threats are being carried out right now against the US election system, in a coordinated fashion, by the same nexus of attackers. They are interlocking attacks supporting the same goal: minority rule and a transition to an authoritarian regime. This threat is as serious as it gets.
> Of course there is one exception in the list: voter ID fraud, in person or by mail, is not a threat. No serious election security expert worries about it. That would be like worrying about a Martian invasion in the middle of WW2: there are more pressing and realistic concerns to worry about... There is no serious coordinated effort to commit such fraud, and as far as I know, there never has been.
You are setting up a straw man argument, and knocking it down. Good for you. Integrity is about more than just protection from coordinated assault. It is also about ensuring the system functions as it is intended to.
No, you probably cannot, as an individual actor, systematically throw an election via voter impersonation. But that doesn't mean that the practice cannot alter outcomes. And it certainly doesn't mean that we should accept a low grade, persistent level of fraud that is trivially preventable.
One person cheating on the SAT isn't going to bring down our entire education system. But you have to show ID for that, because it matters. I'd say our elections matter quite a bit more than the SAT does.
But in-person voter ID fraud is not a 0day. The electoral system is not vulnerable to it. There is nothing to patch.
Meanwhile the made-up threat of ID fraud is used as a bad faith argument to justify actual attacks on the election system, taking place right now.
The “what about id fraud” narrative is part of the attack. By amplifying it you are indirectly providing cover to the attackers, while offering nothing of value to the conversation about election security.
1. Both parties are so similar that there’s essentially only one party.
2. You don’t know anything about US politics yet are compelled to share your opinion anyway.
3. You know exactly which party is which, but your political allegiances are in conflict with that reality. Instead of confronting the dissonance within you, you seek to hide it with a lie.
I serve as an election officer in my county and have yet to hear of a compelling alternative.
paper ballots, really?
when I can apply for a mortgage, car loan, credit card, and all kinds of other financial legally binding contracts online in complete security?
I enter my SSN, my credit report is pulled.
Why can't I enter my SSN, it gets validated in a database on whether I'm alive/have already voted/am currently a legal citizen eligible to vote, and vote?
We have achieved problems way, way, way harder than "securely identify a person based on their SSN and allow them to vote A, B, or C on a subject"
Notably, none of these things require the actual data to remain anonymous like voting does. That's where all complexity comes from.
Assuming I’m not alone why not give people the choice and increase voter turnout?
Your idea is a shortcut to tyranny, and must be peacefully opposed by all holding liberty dear.
Come back once only you can do all those things on your behalf, and not anyone else having your SSN or other PII.
In my state you walk into the polling place, give them your name and address, sign your name, and that’s it. Your polling place is even listed publicly online for anyone who knows your name.
I think what the parent poster meant by "scan" was "tabulate". ie. you fill out a paper ballot, you put it into a scanner which scans it and records the result, and the ballot itself drops into a ballot box. It's still possible to hack the scanner to give a fake result, but it's easily auditable/detected since you still have the original ballots. You can even have the tabulators be entirely offline (since all they do is spit out a number), and their results can be easily cross-checked by random sampling.
There can always be tiny nits in the numbers, but the delta between the pollbook, the unused ballots, and the counts will be vanishingly small.
With enough people involved, and small enough pollbooks (local county maxes out ~4k per polling place) there is a good tradeoff between:
- maximizing participation
- keeping the integrity of the process
- minimizing fraud
As with encrypting data, there really isn't any way to make the process completely bullet proof; the task is to make real fraud infeasible.
> All of the states replaced these with secret ballots around 1890, popularly called "Australian ballots."
https://en.wikipedia.org/wiki/Secret_ballot
No other system has this property. Even if you had a formally verified election system, it and the proofs of it could perhaps only be understood by less than 100 people in the world.
The system has to be comprehensible to voters.
Six hours later, I checked the spreadsheet on the official website, scrolled to the row for my polling location, and verified these two numbers were identical to my count.
Sure, I can only check 1/3rd or so in one of 50,000 polling places. But get just 20 people you trust to do the same, and it becomes statistically unlikely that there is widespread fraud.
You can hang out in the polling location for the whole day if you want. If you get there before any voters, you can also check that the ballot box is empty. Each location serves less than 1,000 voters with three volunteer workers, so you’ll always be able to know everything that’s going on. Oh, and the longest wait I’ve ever had to endure myself was less than 5 minutes
The only possible problem (aside from Republicans) in the US would be the large number of individual races and ballot initiatives. It works well with five or so. But if you’re voting for ten deputy dog-catchers of the peace, it might get tricky.
I'm not against paper voting at all, just saying that they aren't automatic tamper proof.
I'd be a lot more concerned if large numbers of ballots didn't "go missing".