Hacker News Reader

Comment by Yetanfou | Hacker News Reader

Did they also invent a system which protects against voter coercion? If they did not, they might as well stop now since no matter how secure the voting system is its relevance falls down to zero if it is possible for the local gang leader to coerce voters to vote for 'his' candidate - "my man stands behind you to watch you vote for candidate X, if you so much as twitch your little girl gets whipped" (which works but doesn't scale that well) or "you show me proof of you voting for candidate X or we'll burn down your house" (which scales but depends on vote verifiability and as such is rather easily foiled).

View on HN

This is a ridiculous argument, because it could just as well apply to mail-in voting as blockchain-based voting. I also think this fear is far overstated (do we have any stories of this happening in the US?) and the more pernicious forms of voter suppression and misinformation are downplayed.

If blockchain systems allow 90% of the populace to vote unencumbered, with a 5% inaccuracy rate (which is far higher than to be expected), that's still more democratic than only 70% voting.

EDIT: There are also ways to guard against voter coercion, for example by allowing "fake votes" to be cast, which if not cast with a certain memo in the transaction, will not be counted. This memo can be a PIN that the user is given as their "legitimate vote PIN", without which the vote would be invalid. When the user wants to appease their "mob boss", they would simply vote with another pin, and the vote would show as having gone through.

This is a surface-level solution, but the technical architecture that can be built to avoid voter coercion in a digital system is far greater than that with mail-in voting. With mail-in voting, your "mob boss" just forces you to tick a box, and put the sealed ballot in the nearest USPS drop box.

It's certainly not a ridiculous argument. Anonymity is a cornerstone of fair voting systems, and any system that allows votes to be deanonymised increases the risk of coercion.

The same type of risk applies to postal voting, although with less severity, as with postal voting there is only one opportunity to check the coerced vote. Blockchain based votes can be checked after the fact.

Still, for this reason, postal voting is rare, and most countries that allow it do so only for citizens living abroad or who cannot travel to a polling station die to injury or illness.

Countries like the UK and Australia that allow any eligible voter to do so by mail are rare.

The results of blockchain-based elections can be checked after the fact, that's true. But if the votes of the individuals in the electorate can be checked after the fact, the system was badly architected. The USPS solution linked by OP does not link the identity of a voter to their blockchain identity, so coercion remains as much of a threat as in postal voting.

Blockchain-based voting systems can be either the least or the most anonymous voting systems. Electronic voting allows the abstraction of many voter-suppression tactics which are still in play in the US.

I may have misjudged the audience, because the postal voting argument is very US centric, at a time when the prevailing media narrative is that postal voting is an essentially infallible system which should not be questioned. I would find it hypocritical if people strongly supported mail-in voting while not considering that blockchain-based voting carries similar advantages and risks (which is why the USPS proposed this, I'm sure).

If you read my other comments in this thread, you'll see I'm not in favor of implementing a blockchain-based voting system yet. I just think the above argument was made from a fundamental misunderstanding of blockchain technology.

> I just think the above argument was made from a fundamental misunderstanding of blockchain technology.

It isn't blockchain which is the problem but wide-spread absentee voting. It doesn't matter whether this is done through mail-in voting, through some blockchain-based app or site, phone-in voting or anything else. The problem is that the person casting the vote can not be assured to do so with privacy and without coercion.

> I just think the above argument was made from a fundamental misunderstanding of blockchain technology.

Which misunderstanding of blockchains/merkle trees do you believe has caused the argument above to be incorrect?

The system as described in the patent provides no protection that I can see against voter coercion, and in some embodiments, allows a voter to verify that their vote has been counted as cast, which is significantly more ripe for abuse than in-person or even postal voting.

I did read your other comments in this thread, but it seems to me that you have a fundamental misunderstanding of blockchain technology and the problems that it can solve. Your proposed right-to-vote token solution is worse in every way than paper ballots cast in-person at polling stations with private areas, counted by hand in publicly observable count centres.

To be clear, most of my other comments in this thread are solutions to problems I took from the top of my head. I'm not suggesting we implement a voting system off of my comments.

I also am not in favor of blockchain-based voting. Of course a "right-to-vote" token is, at the moment, a far worse solution than paper ballots at polling stations. Mostly, this limitation is pragmatic — we don't have good ways to store private keys, low-level hardware and software is not easily auditable, UX/UI issues, etc.

In the US, we are considering having the largest vote-by-mail election ever. This is a politically charged issue, where Trump claims mail-in-voting will lead to massive voter fraud, and the Democratic party claims otherwise. If you suggest that mail-in-voting may be insecure, you're labelled as a Trump supporter.

With this political landscape in mind, electronic voting vs. postal voting is certainly a pertinent discussion. This comes especially as Trump has made efforts to "DDoS" the USPS by kneecapping its throughput. This would, depending on the system's design, be harder to do with electronic voting. Estonia is a good example of how digital voting can be implemented securely, with the approval and understanding of the populous.

But my main contention is that there are ways that blockchain voting can be fully anonymous, even if we don't have the implementation capability right now. Zero-Knowledge proofs, combined with a system where the voter's real identity is never mapped to their on-chain public key, allow for this. The individual voter would receive a private key in the mail, or using a Monero-like blockchain they would generate their private key and redeem one "Right to Vote" token, which would be sent to their account. They would then vote with this token, and using a "view key" would be able to audit that their vote was cast for X party.

Nowhere in this system would real identity be mapped to blockchain identity. The risk of "deanonymization" is therefore about the same as with mail-in voting. A coercer could force you to show them the results of your vote afterwards, true, but there are ways around this as well. Most simply, all view keys could be revoked after casting a vote. The public "token balances" of each candidate would be viewable, but the results of an individual vote not.

For example the system used for postal voting or more exactly early voting in Finland is one where there is time period(around week) when voters can vote in any early voting location around country, or in such places like embassies around the world.

The process is same as regular voting. Only difference is that vote is enclosed in envelope which is enclosed in second envelope and shipped to voting precinct to be opened later and counted with rest of the votes. During the voting process identity of voter is verified.

There is some risks here, but there really isn't much to fix that. In the end if you trust enough the step between voting and shipping votes it's very decent. Anonymity is there and also voter is verified.

> This is a ridiculous argument,

It is most certainly not. It is the reason voter secrecy in considered a cornerstone of a legitimate public vote.

There are whole organizations, like the OSCE Election Observation, who audit these sort of things internationally.

> it could just as well apply to mail-in voting

It does, very much so. Every mail-in voting has to take this into account. Any introductory political science course will touch on these things. Commonly this is done by way of a method to override or revoke votes. A mail-in voting system would be problematic if it was the only way to vote. This is the most common objection to electronic voting systems, that it's hard for them to stand on their own.

Why encourage different amounts of electronic voting than mail in paper voting ?

"blockchain systems" - which I notice you're invested in by virtue of leading a startup which in some way utilises blockchain technology - may be able to help in reducing the opportunities to cheat on the technology side. This is not the main problem when combating voter coercion, a problem which you noted also exists with mail-in voting. This is true and it is one of the reasons why mail-in voting (as compared to absentee ballots which have to be personally requested by the voter) is rife with problems. Technical solutions can only go so far, even that pin-code system with "fake votes" can be thwarted by using the same threat of force to obtain the "legitimate vote PIN". As far as I can see there are currently no reliable technical solutions to this problem and with that the personal vote at the voting station, using a ballot form and a pencil, is still the best solution.

Sometimes a low-tech solution is the best because it provides a low attack surface and is not easily abused at a massive scale. While there are still some opportunities to mess up things - lost ballot boxes, missing ballot forms, etc - these are easily detected and can be rectified.

For now, I vote for simplicity when it comes to casting my vote.

I vote for simplicity too. As I've tried to be clear in mentioning, I am not in favor of a voting system like this.

For the foreseeable future, paper ballots cast at a polling station are all that make sense.

The discussion is in part relevant because the US is on the verge of having the largest mail-in election ever. Trump claims this is insecure, the Democratic party claims otherwise. It's a political hot-button issue, and seems likely to happen. For this reason, I think exploring alternate systems is worthwhile.

I hope that makes sense.

Reply on news.ycombinator.com