However It would certainly be nice to vote online yet anonymously, confident that I can see in a public secure ledger that my voted was counted as a intended and that no cheating or voting manipulation (on the ledger level) took place.
This is the same problem with Bitcoin privacy advocacy: it's a brittle design and the failure mode is complete, undeniable loss of privacy.
The coercion attack is an interesting problem but something that is already illegal but cannot be fully prevented. IMO it outweights the disadvantages of a centralized voting database that can be easily changed by a sys admin.
Your second paragraph ignores the fact that the current system of anonymous ballots prevents the problem quite well, as evidenced by fraud rates measured in single-digit-per-billion levels, and paper ballots allow recounts without any use of computers and have the nice effect of making tapering a physically-diverse hard problem requiring a conspiracy involving far more people.
Let's not make this an accidental strawman.. My second paragraph is referring to electronic and internet voting solutions vs decentralized/blockchain solutions and NOT traditional in person voting with independent observers, etc..
Can you specify precisely how those work to prevent this problem? Remember, this is harder than trying to prevent a third-party from identifying users because the system has to survive collusion by the user.
Again this is not about perfect pie in the sky solution to all problems, just better digital voting solutions. As I said in the original comment, I don't think it will solve problems that have always been with us, but "It would certainly be nice to vote online yet anonymously, confident that I can see in a public secure ledger that my voted was counted as a intended and that no cheating or voting manipulation (==> ON THE LEDGER LEVEL <==) took place."
This is one of the reasons why we don't have internet voting: it's inferior to the status quo in critical areas. “blockchain” is not magic pixie dust which fixes that problem and the people backing it don't seem to have studied the problem very deeply, as evidenced by your repeated assertions that “the voting would using a cryptographic system to ensure confidentiality” before finally admitting that nobody has proposed a system resistant to a key class of attacks.
> Even in normal voting, you can take a photo of the ballot to then prove who you voted for and get paid for it (and/or not get beaten).
The difference is that it's harder to do that – or even illegal, see http://bigstory.ap.org/article/709338e5557a49e7ad5c68109ffec... – and it's not foolproof since you could take a picture of a ballot but void it and cast a different one, both of which make widespread coercion or selling unreliable.
> As I said in the original comment, I don't think it will solve problems that have always been with us
Simply asserting that it's always been with us doesn't change the fact that this is a problem which the current status quo was actively designed to prevent and that the proposed system makes that problem far worse.
The day might come though that internet voting will be widespread. When that day comes I rather have a decentralized solution solution than a centralized one that can be easily manipulated.
How do you get valid voting credentials to nontechnical voters such that they can anonymously vote and verify?
> How do you get valid voting credentials to nontechnical voters such that they can anonymously vote and verify?
That has to do with identity systems (see uPort, onename) and is an interesting & open problem.
Imagine a log of transactions (a blockchain) confirming that Clinton got 1.5 million popular votes. That has ZERO effect on the Electoral College selecting Trump. So why even bring the election up? This guy might as well say that "a blockchain could stop the Patriot Act and the NSA wiretaps". No, it couldnt. Real political change happens through laws, constitutional ammendments, and court decisions. No amount of electronic blockchains will change that fact.