USPS Files Patent for a Blockchain-Based Voting System • Hacker News Board

Re: USPS Files Patent for a Blockchain-Based Voting System

#11

Post by sebmellen » Sun, Aug 16, 2020, 8:52 AM UTC

peignoir wrote: post #8

sebmellen wrote: post #5

My whole startup is built on blockchain, and I'm an Estonian e-Resident (Estonia allows their citizens to vote digitally), so I find blockchain voting fascinating. That said, there are some problems with it. This Tom Scott video explains why: https://www.youtube.com/watch?v=LkH2r-sNjQs . But this is a very cool idea: combining the USPS vote-by-mail infrastructure with a blockchain layer that sits on top, used mainly…

Oh cool, not related to the article directly but I m off to Estonia and working on a blockchain (tezos) voting system and community if interested : www.electis.io

Interesting. For lower stakes voting, this could be a great implementation. I have good friends building a Tezos startup right now (Tezsure) - certainly an interesting chain!

We're on Stellar mainly because of block times. You can see a beta of our timestamping service (mainly built for scientists... launch coming soon) here: https://assembl.app/chronos

Re: USPS Files Patent for a Blockchain-Based Voting System

#12

Post by nathias » Sun, Aug 16, 2020, 8:54 AM UTC

We should be experimenting with electronic voting for 30 years and innovate solutions to its problems. Nobody thinks it doesn't have problems, but to just stop at pen and paper voting just means to affirm the problems of pen and paper which are IMO much worse. (scarce voting, batched issues)

Re: USPS Files Patent for a Blockchain-Based Voting System

#13

Post by foepys » Sun, Aug 16, 2020, 8:55 AM UTC

colordrops wrote: post #10

foepys wrote: post #2

All blockchain (and electronic for that matter) voting systems are inherently bad because they are not easy to verify. Nothing beats pen and paper in verifiability and ease of use. If done right with many eyes on the ballots, manipulation requires thousands of co-conspirators. The only downside is the slow speed.

You give one citizen one vote on the chain. They can personally verify that their vote was counted because the chain is visible. Counts can't be rigged because the entire chain is public. Voters can only vote by going to a physical location where they use a multi-key signature, one by the voter, and one by the local polling station, avoiding people selling off their keys. This is anonymous, traceable, and the vote ca…

How does the average citizen verify what is actually happening and that their vote is actually being counted? That when they vote for X that the counter for X is increasing by 1? By verify I mean full understanding of the process without the need to believe somebody who calls themselves an "expert".

Re: USPS Files Patent for a Blockchain-Based Voting System

#14

Post by ianleeclark » Sun, Aug 16, 2020, 8:56 AM UTC

kfrzcode wrote: post #6

Electronic voting is a terrible idea. Attacks on paper & pen ballot systems are much, much harder to scale. Here's Tom Scott with a great explanation of the basics. https://www.youtube.com/watch?v=LkH2r-sNjQs

> Attacks on paper & pen ballot systems are much, much harder to scale.

As it turns out, you can just impose austerity.

Re: USPS Files Patent for a Blockchain-Based Voting System

#15

Post by Galanwe » Sun, Aug 16, 2020, 8:56 AM UTC

What troubles me with this is that at some point in the system there will need to be a mapping "real identity" -> "ID of voter in the blockchain".

What happens the day this mapping leaks?

Re: USPS Files Patent for a Blockchain-Based Voting System

#16

Post by foepys » Sun, Aug 16, 2020, 8:58 AM UTC

Galanwe wrote: post #15

What troubles me with this is that at some point in the system there will need to be a mapping "real identity" -> "ID of voter in the blockchain". What happens the day this mapping leaks?

Even if that's solved: what happens when voters lose they key? Or worse: get it stolen from them.

Re: USPS Files Patent for a Blockchain-Based Voting System

#17

Post by galkk » Sun, Aug 16, 2020, 8:59 AM UTC

colordrops wrote: post #10

foepys wrote: post #2

All blockchain (and electronic for that matter) voting systems are inherently bad because they are not easy to verify. Nothing beats pen and paper in verifiability and ease of use. If done right with many eyes on the ballots, manipulation requires thousands of co-conspirators. The only downside is the slow speed.

You give one citizen one vote on the chain. They can personally verify that their vote was counted because the chain is visible. Counts can't be rigged because the entire chain is public. Voters can only vote by going to a physical location where they use a multi-key signature, one by the voter, and one by the local polling station, avoiding people selling off their keys. This is anonymous, traceable, and the vote ca…

As I see, the problem is not that the vote can be altered, but that the vote can be bought and the fact that person voted in the right fashion could be verified by 3-rd party much more easily (screenshots/whatever).

The privacy of voting booth could give at least some chance to change mind/lie to the third party.

Re: USPS Files Patent for a Blockchain-Based Voting System

#18

Post by soulofmischief » Sun, Aug 16, 2020, 9:00 AM UTC

Galanwe wrote: post #15

What troubles me with this is that at some point in the system there will need to be a mapping "real identity" -> "ID of voter in the blockchain". What happens the day this mapping leaks?

Ideally you should be able to produce a unique hash each time you vote, derived from your master key. The server doesn't need to know your private key. This is how crypto works now, generating unique and theoretically unlinkable wallet addresses from a secret key.

Re: USPS Files Patent for a Blockchain-Based Voting System

#19

Post by sebmellen » Sun, Aug 16, 2020, 9:00 AM UTC

Galanwe wrote: post #15

What troubles me with this is that at some point in the system there will need to be a mapping "real identity" -> "ID of voter in the blockchain". What happens the day this mapping leaks?

In theory, this on-chain voter ID would never have to be "mapped" directly to a voter. Rather, IDs are simply private-public keypair accounts, and all that's provisioned by the holders of the election is the right to vote with one account. Then, the vote is cast by redeeming this right to vote token, without the voting account ever being linked to a "real identity". Of course, there are problems here too. Mainly with UX, and then at a low level the hardware used (the general public is very easy to hack if you have nation-state power).

Re: USPS Files Patent for a Blockchain-Based Voting System

#20

Post by barbegal » Sun, Aug 16, 2020, 9:02 AM UTC

You could easily replace the word Blockchain in this patent application with the word database and it would all still make sense.

Blockchains (I prefer to call them Merkle trees) were invented 40 years ago. The interesting part about cryptocurrency based Blockchains that makes them unforgeable is that they contain proof of work. The work to create an entire Blockchain is equal to the sum of the work contained in each node. Without proof of work Blockchains are easily forgeable.