4 ms·

Absolutely not. Until cryptographically-sound voting systems are pervasive and well-vetted, there is absolutely no hope of making online balloting both reliable

Absolutely not. Until cryptographically-sound voting systems are pervasive and well-vetted, there is absolutely no hope of making online balloting both reliable and secret.

schainks 13y ago
have you seen this Ted Talk: http://www.youtube.com/watch?v=izddjAp_N4I http://www.youtube.com/watch?v=izddjAp_N4I It's not totally "user friendly", but I'd say it's a huge step towards "secure and reliable" voting.
- nknighthb 13y ago
  I know of various proposals that offer a theoretical solution. The problem is getting it implemented, fully vetting the implementation, and getting everybody to use it. There were lots of ways we could have made touchscreen voting machines highly secure. Instead we gave Diebold millions of dollars. I have no reason to believe that sound security principles will be followed by any large organization, including governments, in my lifetime.
dalek_cannes 13y ago
Will something like this work? 1. Voter (physically) visits a station, presents his ID and picks an auth token out of a hat (or some other physical, non-traceable method) -- say a paper scratch card with only a randomly generated number. 2. Voter goes home and uses the token to anonymously authenticate himself as a voting citizen and participates in one or more rounds of voting CONS: 1. Voter misplaces his card and has to come back to the station to identify himself and revoke & re-issue. 2. In some neighborhoods, voter has to beat back thugs trying to steal his card on his way out of the issuing station. Or... 1. Voters have regular online accounts with their credentials attached. 2. Source code dealing with vote counting and anonymizing is made open source and publicly readable at the hosted directory itself (if such thing is even possible) CONS: 1. Hackers study the source code and exploits vulnerabilities. 2. Server serves the good copy when source code is requested, but executes a different copy that collects voter identities. Note: not an expert, just asking.
- nknighthb 13y ago
  schainks linked to a video you should watch. Finding theoretically secure systems is not the problem. Properly implementing them, vetting the implementation, and gathering the political will to mandate the properly vetted implementation is the problem. As always, the weakest link in security is the human factor.

Absolutely not. Until cryptographically-sound voting systems are pervasive and well-vetted, there is absolutely no hope of making online balloting both reliable and secret.

schainks 13y ago

have you seen this Ted Talk: http://www.youtube.com/watch?v=izddjAp_N4I http://www.youtube.com/watch?v=izddjAp_N4I It's not totally "user friendly", but I'd say it's a huge step towards "secure and reliable" voting.

nknighthb 13y ago
I know of various proposals that offer a theoretical solution. The problem is getting it implemented, fully vetting the implementation, and getting everybody to use it. There were lots of ways we could have made touchscreen voting machines highly secure. Instead we gave Diebold millions of dollars. I have no reason to believe that sound security principles will be followed by any large organization, including governments, in my lifetime.

dalek_cannes 13y ago

Will something like this work? 1. Voter (physically) visits a station, presents his ID and picks an auth token out of a hat (or some other physical, non-traceable method) -- say a paper scratch card with only a randomly generated number. 2. Voter goes home and uses the token to anonymously authenticate himself as a voting citizen and participates in one or more rounds of voting CONS: 1. Voter misplaces his card and has to come back to the station to identify himself and revoke & re-issue. 2. In some neighborhoods, voter has to beat back thugs trying to steal his card on his way out of the issuing station. Or... 1. Voters have regular online accounts with their credentials attached. 2. Source code dealing with vote counting and anonymizing is made open source and publicly readable at the hosted directory itself (if such thing is even possible) CONS: 1. Hackers study the source code and exploits vulnerabilities. 2. Server serves the good copy when source code is requested, but executes a different copy that collects voter identities. Note: not an expert, just asking.

nknighthb 13y ago
schainks linked to a video you should watch. Finding theoretically secure systems is not the problem. Properly implementing them, vetting the implementation, and gathering the political will to mandate the properly vetted implementation is the problem. As always, the weakest link in security is the human factor.