alt Hacker NewsFully software, even a combo of software and hardware, cannot by itself make elections secure. Think of the Reflections On Trusting Trust problem, the need to verify firmware, etc. You can't ensure that machines _will_ work correctly, but you _can_ verify that they _did_. To do verify the machines' functioning after the fact you have to do it very soon after closing of the polls, and you have to devise a manual, statistical random sampling method where you manually check randomly sampled, marked paper ballots, against the digital cast vote record, then do a digital risk limiting audit to check that the CVR totals up to the machines' reported results. The sooner after closing of the polls you do this, the better the likelihood that any cheating will be caught. I.e., an immediate manual, partial audit of enough randomly-selected polling locations.
That still leaves tons of other technical problems to address, like security of mail-ballots / absentee ballots, the security of ballot boxes used for early voting (because you probably cannot run the above manual partial audit due to then having nightly results available to the public; or maybe you accept that), voter registration, etc.
Replies
But Brazilian gov says their electronic election system is unhackable... Yeah.
Only a selected few orgs and groups are allowed to audit and in a very controlled environment because they also rely on security through obscurity. The source code is not available to the citizens.
If by any chance I'm wrong and someone can point me a source code repo, please enlighten me.
https://international.tse.jus.br/en/electronic-ballot-box/pr...
While these all are problems, the REAL problem is that the goal of elections isn't to get scientifically or mathematically verifiably correct results. The goal is to produce agreeable consent.
That means it doesn't matter one bit how correct you are if the regular Joe does believe you tricked. Adding computers into the mix always makes this worse. And again, this is not something you can fix with clever solutions and cryptography (that admittedly fascinates me nonetheless), the problem is that people can't see electricity generally and I something at the scale of a computer, even if you could it would be utterly meaningless.
This is why the paper ballot system is so glorious in its stupid inefficiency, because it spreads the trust in a way any adult who has mastered object permanence can verify for themselves. Invite the conflicting parties to monitor and have them begrudgingly admit their defeat in the face of the simple, inefficient but agreeable process.
This is why I as a cryptography fan, programmer and hardware guy think electronic voting is not up to the task for high stakes elections.
> Think of the On Trusting Trust problem, the need to verify firmware, etc.
I mean that's too abstract and philosophical of a problem when talking about voting machines/software: there are much more prevalent and concrete risks including but not limited to APT-grade breach/espionage if your country is important enough, and the various other issues (some in your comment itself).
You solve all of these with cryptography.
Your ID card contains a private key, you use that key to sign a transaction. It doesn't matter that machine you are running on is compromised as these details can be verified at a later date.
Example A: You vote for candidate A, but software changes to candidate B and software displays that you voted for candidate A. At any later time you can check with an independent device and look at the transaction finding out that it was altered with and retract your vote / change it.
Example B: How do you know your vote was counted? Publish a hash and the signature of every vote, recompute the final tally and compare with all the votes. If your vote is not in the final tally the hash will be different and you can visibly see your vote in the dataset as every transaction is published. To preserve privacy each round has a number, private key + number -> new key so each fingerprint will be unique between votes.