You are right to consider the possibility of compromised hardware. My information is that all commercially available chips are backdoored at the factory, via the security block.
However, I do not believe this compromise is competent to actually insert code. I believe this mechanism is suitable only for tracking and surveillance purposes.
In addition to the hardware backdoor thing, there's also this:
"Voters being able to verify that their vote was counted sounds nice, but I’m thinking that would ultimately be quite meaningless, since the establishment could just make up voters who vote in its favor. “But a state only has so many people living in it, so the establishment can’t just make up voters.” Well, maybe not after the fact…"
Steem shows the way. Distributed ledgers are secure, private, and auditable.
Each voter can verify their own vote, and further verification can enable each cast vote to be verified by the voter that cast it. The latter audit would be expensive, but definitive, as any votes cast in the names of the dead, illegals, or folks that just didn't vote (or voted otherwise) would be not verifiable.
Simply canvassing those that cast votes would verify that those votes were actually cast by those voters. Even more exhaustive audits could ensure that one voter didn't cast multiple votes, by using voice recognition during canvassing, and then personally interviewing voters suspected of casting multiple votes while verifying those particular votes.
It could be the most secure and auditable voting system every devised.
The problem is though: who would know which vote belongs to which voter? The government would, right? Well, we both know the government is beyond corrupt, meaning I doubt the audits would be accurate.
Distributed ledgers are publicly auditable. I see no reason that individuals couldn't audit the vote themselves, personally.
When I worked in market research, we did audits of our marketing efforts. We'd take 10% of the calls an individual made and verify them by calling them back, and this provided a statistically accurate representation of each marketer's work.
Individual people wouldn't have to audit the entire vote to have a very reliable grasp of the accuracy of the whole. Concerned groups would be more able to conduct audits.
I am not a fan of government agencies, such as FDA, EPA, and so on. I know they are but vectors for corruption, which allows the very problems they're created to avoid to flourish.
I think concerned people should be able to just hire some work done, and if the science shows problems, then tort actions should be brought against offending parties. If Bayer knew they'd be on the hook for the medical bills for the victims of glyphosate, they wouldn't have touched Monsanto with a pole a thousand feet long.
The less government the better, and the less corruption there will be. Keep government out of audits. Then they'll be honest.
Not unless that was necessary to cast the vote in the first place. For audit purposes, only such identifying information as is required on Steemit would be necessary.
It would be necessary to call the voter and verify that their vote was cast and recorded correctly. To prevent one person from casting multiple votes, it would be wise to use voice recognition software to monitor the calls, and where multiple votes were verified by what appeared to be the same person, then requesting the opportunity to meet with them in person to verify a vote would enable observing them while verifying the votes involved.
Unfortunately, should such interview be declined, affected votes would necessarily be unverifiable.
Also, sensitive identifying information needn't be plaintext, but could be encrypted. A means of audio contact would be all that was necessary in order to verify the authenticity of votes.
Had a conversation about that here: https://medium.com/@kevinsarpei/while-id-definitely-prefer-an-open-source-blockchain-based-voting-system-over-the-current-one-the-cf1720dc7539
You are right to consider the possibility of compromised hardware. My information is that all commercially available chips are backdoored at the factory, via the security block.
However, I do not believe this compromise is competent to actually insert code. I believe this mechanism is suitable only for tracking and surveillance purposes.
In addition to the hardware backdoor thing, there's also this:
"Voters being able to verify that their vote was counted sounds nice, but I’m thinking that would ultimately be quite meaningless, since the establishment could just make up voters who vote in its favor. “But a state only has so many people living in it, so the establishment can’t just make up voters.” Well, maybe not after the fact…"
And this (52:12-1:03:26):
Steem shows the way. Distributed ledgers are secure, private, and auditable.
Each voter can verify their own vote, and further verification can enable each cast vote to be verified by the voter that cast it. The latter audit would be expensive, but definitive, as any votes cast in the names of the dead, illegals, or folks that just didn't vote (or voted otherwise) would be not verifiable.
Simply canvassing those that cast votes would verify that those votes were actually cast by those voters. Even more exhaustive audits could ensure that one voter didn't cast multiple votes, by using voice recognition during canvassing, and then personally interviewing voters suspected of casting multiple votes while verifying those particular votes.
It could be the most secure and auditable voting system every devised.
The problem is though: who would know which vote belongs to which voter? The government would, right? Well, we both know the government is beyond corrupt, meaning I doubt the audits would be accurate.
Distributed ledgers are publicly auditable. I see no reason that individuals couldn't audit the vote themselves, personally.
When I worked in market research, we did audits of our marketing efforts. We'd take 10% of the calls an individual made and verify them by calling them back, and this provided a statistically accurate representation of each marketer's work.
Individual people wouldn't have to audit the entire vote to have a very reliable grasp of the accuracy of the whole. Concerned groups would be more able to conduct audits.
I am not a fan of government agencies, such as FDA, EPA, and so on. I know they are but vectors for corruption, which allows the very problems they're created to avoid to flourish.
I think concerned people should be able to just hire some work done, and if the science shows problems, then tort actions should be brought against offending parties. If Bayer knew they'd be on the hook for the medical bills for the victims of glyphosate, they wouldn't have touched Monsanto with a pole a thousand feet long.
The less government the better, and the less corruption there will be. Keep government out of audits. Then they'll be honest.
So every cast vote would have the voters name and address next to it?
Not unless that was necessary to cast the vote in the first place. For audit purposes, only such identifying information as is required on Steemit would be necessary.
It would be necessary to call the voter and verify that their vote was cast and recorded correctly. To prevent one person from casting multiple votes, it would be wise to use voice recognition software to monitor the calls, and where multiple votes were verified by what appeared to be the same person, then requesting the opportunity to meet with them in person to verify a vote would enable observing them while verifying the votes involved.
Unfortunately, should such interview be declined, affected votes would necessarily be unverifiable.
Also, sensitive identifying information needn't be plaintext, but could be encrypted. A means of audio contact would be all that was necessary in order to verify the authenticity of votes.