The main problem here is that an ordinary voter should be able to check the integrity of a voting system. With paper votes, (which we still have in the UK) an ordinary voter can do this - they can watch the ballot boxes being carried about, count people voting, and watch the votes being counted.
With electronic voting, an ordinary person has to take it on faith. The more clever technology you put in, the less it can solve this problem - the harder it is for an ordinary person to check that the vote is being counted correctly.
He didn't include any crypto, even though he's one of the world's most famous crypto experts, precisely because that would make it opaque to anyone who isn't a cryptographer.
This is a bitcoin-based system. The integrity of the keychain can be in danger only if a 51% attack takes place (creates a larger keychain). That's highly unlikely IMHO in a national election system, where any geek could setup a node not to mention political parties, organizations, etc.
There is a public ledger. The votes are easily verifiable by anyone, on the fly, all you have to do is a client with a nice GUI for the user. The user doesn't care about the crypto, all he wants to know is that his vote counted and that 40 years later, it's verifiable and private.
Bitcoin can do all those things. Actually it solved the most important problems related to e-voting IMHO.
ps. The only not-solved problem is selling votes. With this sort of scheme the buyer could request a proof of vote (e.g. a picture, or something more sophisticated like your user/pass). But this is already happening in Greece and I guess elsewhere too. I'm not sure though if an e-voting scheme would allow this to take a much larger scale. Technology can overcome the lack of social/political education, unfortunately.
I'm not sure if this is state of the art[0] since the talk is 7 years old. This scheme provides proof to the voter, but not enough so he can prove his vote to someone else, avoiding vote selling altogether.
The biggest challenge IMHO is trust in the system.
[0] I'd appreciate criticism or any other progress in the field.
I've seen the video you posted. It's a 2007 video, otherwise I'm sure the author would have mentioned the Bitcoin protocol time and again, because it solves all the problems mentioned + it's decentralized.
Decentralization is a key-aspect here. Votes count equally for all citizens. So every citizen has the opportunity to operate a node for the time (48hours) that the elections will take place, in order to inhibit possible corruption. Given the No of organizations, individuals (e.g. political activists, freedom believes, anarchists, you-name-it, conservatives, liberals, you-name-it) and geeks (who will do it for fun, get an RPi, setup a node for the elections, measure traffic, do a JS/CSS3 front-end to display traffic, real-time results etc.)
The government (formed by all parties) should be able to setup an X number of nodes that would allow the elections to take place anyway, without relying to third parties. But third parties will join to ensure reliability of the network.
A government could block the access of nodes from IP netmasks that are located outside the country's IP blocks, to avoid interference of foreign governments[1].
Now, I'm not 100% about the interface and the interaction which is equally important:
* Should the voters use PKI? Should an account be linked with their IDs?
* Should we use a dedicated machine with touch screen and accessibility options in a private chamber?
* Should we use electronic signatures generated by a special 'card' which every citizen should have one?
* etc
Estonia has an electronic card[2] for example that could turn handy into situations like these.
There many ways of deployment with pros/cons to be considered on a country by country scenario.
Bitcoin protocol solves the double voting, the possibility to re-count votes, the possibility (which is nowhere today) for citizens to review their votes even years later.
Also we should not underestimate the ability to vote from the comfort of your home, using any device: Imagine how greatly that would improve the % of people participating in the elections for example.
[1] I can't think of what the NSA could come with in this scenario. Would make an interesting discussion.
I don't see the point in using a Bitcoin-esque protocol for a voting system (ie a hashing based blockchain). In fact the existence of a double spent vote could simply invalidate both (or all) votes; the user attempted to cheat, no need to decide which is valid. Otherwise vote acceptance could be based on the votes of the ids themselves rather than by hashing power. You say decentralization is a key aspect, but the distribution of whatever underlying identity the users are voting with still has to be centralized. Bitcoin has opened people up to these sorts of ideas, but that doesn't mean it has to form the basis for everything in the future.
The Wikipedia entry is quite unclear. The paper itself states at this point it's a proposal of academic and not practical interest, since a vote buying attack has been discovered.
Yes, I should have made it clear that ThreeBallot is not itself a contender. The point is though, that lack of crypto is a desirable property of a voting system.
I was thinking about a Bitcoin-based e-voting system (Specifically made for political nation-scale elections) for some time. It was a project that I'd like to get involved after I get my degree (~ Jan 2015). I believe it is the future because it can incorporate everything that an e-voting system needs (privacy, votes are re-countable, data is publicly accessible, nodes can be run from individuals to organizations to parties, etc.)
This looks not bad but what I'm failing to understand is:
* Is this an organization or a company?
* Is code open source? Where is it?
* Are they interested in 'corporate voting' schemes or in public voting schemes?
I see nothing more than a landing page which says a several things but nothing specific and NO I don't want to register first in order to know exactly what this is.
This is a highly problematic claim, I am not aware of any serious security specialist who guarantees any system is 100% fraud proof.
http://v-app.io/landing seems to be the real application behind this, which seems to be a much better and wider version of Doodle, the voting application. If anyone from v-app is listening I don't think you should mix v-initiative.org with v-app.io/organising things in a democratic way.
It's not just a highly problematic claim, it's 100% false.
Any democratic digital vote is in itself 100% fraud, because it lacks two fundamental elements:
1) Completely transparent and verifiable for all voters.
2) The guarantee that anyone can cast their vote free of immediate external pressure.
That's why we have paper ballots and voting booths. Anyone who claims otherwise, anyone who claims this system is merely "outdated" isn't talking about free and democratic elections.
Uncertain why you got downvoted - perhaps "fraud" is a bit too strong but you are mostly correct.
Even if you had a device and protocol that was perfectly secure, if you vote from home, someone else can demand to see you cast it. Or, if it's an app on your phone or a website, there will be an enormous incentive to either monitor your votes or hack them from the rest of the OS.
I've looked into this question a lot and as far as I can tell, there's no way around it. At best, you might be able to distribute relatively secure machines to many places. Maybe every place that has minimal security, like pharmacies or post offices. Or possibly you have some scheme that allows you to repudiate coerced votes later - but that also offers more opportunities for coercion!
That still leaves the problem of verifiability. We might imagine a relaxed standard of verifiability where almost any geeky person can verify the vote is good. I don't have any real ability to verify paper ballots for even a local election, as they are held in a warehouse somewhere. Maybe ubiquity of verification trumps ease of understanding.
I think the main issues to overcome will be making sure that only registered users vote, and only once per voting event.
This could be achieved by distributing an undivisible votecoin to each registered voter upon presenting their proof of registration or eligibility.
But then there's a weak point of someone giving/receiving more than one votecoin (bribe, threat, blackmail, etc)
Even if there are checks and balances with number of votes vs number of registered voters, there's again the weak point of human element of data entry.
Bitcoin isn't the correct starting point for this as the whole point of verifying blocks via miners is that you can't trust entities in the system to be unique for each user (i.e. Sybil attack), whereas with a democratic voting system trusting that each user has been given a unique starting key is a core requirement.
I agree that you need to know that voters aren't voting twice (assuming one voter one vote), but that doesn't mean that a blockchain isn't the way forward. If keys are issued / signed centrally, then there's no way to vote more than once.
That's just a record though. A bitcoin style blockchain would provide a hashing-power based method of deciding which votes count in the case of double voting, which isn't required or even wanted in this case.
It's not just a record. It is public and untamperable, and could form the core of vote verification.
I didn't imagine voting to be like spending a bitcoin. Although that could work too. Issue every voter a single indivisible votecoin. To vote for a candidate send the votecoin to their address.
But, perhaps the V initiative have something else in mind for the blockchain. I'm sure there are others.
I've outlined only two possibilities for using a blockchain here, but I'm sure there will be others.
You haven't given a reason to use a blockchain, you're just pointing out ways in which you could use a block chain.
As I said, why would you rely on a democracy of hashing power for anything when you have unique underlying identities you're trying to let vote? Why use a system designed to allow the decentralized use of a currency without double spending when double spending isn't an issue?
The two biggest issues solved by Bitcoin are double spending and distribution of a decentralized currency. Our would-be voting system doesn't need either of these things.
If it's a digital democracy it's neither secure nor a democracy.
The system can be rigged with the oldest means possible: a knife to the voters' throat.
I really wish fellow techies would cut this kind of dangerously ignorant and arrogant crap.
We've invented polling stations and voting booths for reasons other than being fucking Luddites. Only in the voting booth behind drawn curtains, an anonymous ballot paper and zero potential electronic surveillance equipment are you free to cast your vote without external pressure and manipulation.
And the result, the counting process, should be observably by any voter, including the ones that don't know shit about software. I.e., the manual counting of ballots.
That's a fucking democracy. Totally secure and yet totally transparent and verifiable.
The rest, like this project, is a self-important sideshow that only serves to undermine true democracy, and half the time that is the actual agenda.
I was thinking about digital democracy for some time but the open source model loses credibility when you use a centralized approach. I think making it decentralized with a cryptographically sound method is the key. Bitcoin is everywhere. I hope this project gains traction.
Had an idea for something similar several months ago but was struggling with the conceptual implementation [1]. Very glad to see someone is doing something here and working towards a solution.
I'm going to install this on my phone and play around with it, and see if I can get some folks in my community involved as well - and see where it goes. Looks very very early but I'd be happy to give it a shot.
A better route is the development of voter-verification, such as Rivest's ThreeBallot: https://en.wikipedia.org/wiki/ThreeBallot
He didn't include any crypto, even though he's one of the world's most famous crypto experts, precisely because that would make it opaque to anyone who isn't a cryptographer.