Comment by cryptonector
4 hours ago
> You solve all of these with cryptography.
I've spent a lot of time thinking about this problem, and my conclusion is that cryptography alone is not the solution, for reasons like:
1. The public needs to believe it, and hypertechnical solutions are a hard sell.
2. Any blockchain approach needs to preserve the secrecy of the ballot, otherwise you get all the problems you can expect of non-secret votes (coercion, vote buying, shaming network effects).
3. You still need to have the cryptographic rubber meet the physical road _somewehere_ -- it has to, otherwise the need for vote secrecy means that the results can be entirely fictitious and pre-programmed and there is no way to check!
I.e., you can have a paperless blockchain solution IFF you abandon vote secrecy. That is a very tough to swallow pill.
On the other hand, if you accept paper ballots and manual partial audits as part of the solution, then blockchain can have a role to play. For example, the machines could display a blockchain head every 50 ballots cast, and voters take a copy of the blockchain head visible to them, then after the election the public can verify the integrity of the blockchain, and the fact that you only get running tallies every 50 ballots means you get vote secrecy.
A system that uses paper ballots and partial manual audits can be explained to the public, and the public can develop confidence in it. Blockchains can then add a good deal of confidence: the blockchain tells you that the tallies were not modified after the fact, and the partial manual audit tells you to an acceptable level of statistical confidence that the digital cast vote record contains the actual paper ballots _as marked and cast_ by voters.
>I.e., you can have a paperless blockchain solution IFF you abandon vote secrecy. That is a very tough to swallow pill.
This is false. I know this is a common thought on HN because some formerly popular youtuber did a video about this topic that is still regularly reposted here, but he was already wrong in theory back then and it is certainly wrong in practice today. Secrecy/anonymity and verifiability are no longer mutually exclusive as anyone who is well versed in modern cryptography will be able to tell you.
I'd appreciate some references. That's not because I'm trying to put you down with a request for references, but clearly you have seen them, and I'd like to see them.
Also, you still have to convince the public.
I agree that convincing the public is a separate issue. But on a technical level, these problems are solved. See e.g. this modern implementation: https://github.com/vocdoni/davinci-paper This is a pretty recent one, but the mathematical foundation dates back to the late 80s.
There have also been earlier implementations like this one: https://github.com/benadida/helios (corresponding paper https://www.usenix.org/legacy/event/sec08/tech/full_papers/a...), although it still lacked coercion resistance. But it already uses zero knowledge proofs, which is at the core of anonymous yet verifiable systems.
1 reply →
the issue isn't that people won't believe it or that blockchain can't preserve secrecy of ballots (it can). it is, in part, that crypto eventually has to meet the real world, but mainly that crypto is hard and governments might not be up to the task. for example, last year the IACR (which had been doing online voting via crypto for more than a decade, and still is) had to void their elections because one of the trustees lost their private key and the results were impossible to decrypt. while it was ultimately fine because the IACR isn't really a high-stakes operation, how would you feel if your government just said "sorry folks, there's been some technical issues and elections will have to be redone"? would you believe them or suspect they're up to no good? what if this happens in a very contended and heated election?
> The public needs to believe it, and hypertechnical solutions are a hard sell.
Public doesn't need to know the details. Production of paper, printers is quite complex and so production of the pen. Yet people are not concerned about that when voting and we don't publish schematics of a printer that printed the ballots.
> Public doesn't need to know the details. Production of paper, printers is quite complex and so production of the pen. Yet people are not concerned about that when voting and we don't publish schematics of a printer that printed the ballots.
Except that people know what paper looks and feels like, they know how it absorbs ink, they know its various properties, they can _read_ whatever is printed on them, and they can see that the marks they make (or a ballot marking device does for them) are faithfully visible on the paper. With paperless machines there is nothing comparable.