Infosec is a pox. Nowhere is this more clear than 'election security'. Whether election machines can be hacked is NOT the most important concern. The reason we went to electronic machines in the first place is the ambiguity of human marked paper ballots in the Bush/Gore election.https://twitter.com/ericgeller/status/1164239438114893824 …
-
Show this thread
-
With purely paper ballots, there are no unambiguous rules on how to count them. Instead, after the election, partisans fight for whichever rules favor them, such as in the ballot below:pic.twitter.com/eAxuFocTCP
7 replies 1 retweet 12 likesShow this thread -
Replying to @ErrataRob
So print the ballots from a machine. No big deal. Mandating paper does not mean mandating manual marking and voting.
1 reply 0 retweets 5 likes -
Replying to @colmmacc @ErrataRob
How do you know the machine printed what you wanted, or what it stored internally? Rob's off base here. Machine printed ballots create too high of a risk / cost not to use hand-marked ballots. Hand-marked paper ballots are still the way to go.
1 reply 0 retweets 0 likes -
Because it's the verified paper ballot that gets counted, not what's inside the machine.
1 reply 0 retweets 3 likes -
Replying to @ErrataRob @colmmacc
And what is the error checking rate of humans testing the output of a long ballot to their actual inputs? You're arguing in favor of humans manually double-checking their work as a solution for written ballot ambiguity? Which risk is more probable? I'll take ambiguity any day.
1 reply 0 retweets 1 like -
Replying to @meyerjef @ErrataRob
Machine input tends to be less error prone than writing, but obviously the UX is important for that. From the security perspective: it only takes a small number of people to notice that their intent has been altered to flag a problem.
1 reply 0 retweets 0 likes -
Replying to @colmmacc @ErrataRob
I get your point, but how about 1 or 2 errors per precinct per half day? People are more likely to fault themselves than the machines and the density of errors makes the likelihood of trend identification low, but impact overall could be high.
1 reply 0 retweets 0 likes -
To me, if we're trying to ensure accuracy, having a verifiable source is more important than processing efficiency.
1 reply 0 retweets 0 likes -
Replying to @meyerjef @ErrataRob
Absolutely! I'm just telling you that machines with good UX tend to do better than paper for input errors. People mark the wrong row all the time on paper. UX can guide people through races one by one, more space for bigger fonts, put some check points in, etc.
1 reply 0 retweets 2 likes
Additionally, the blind and those with motor control challenges are often forced to use an aide. But well designed voting machines can avoid that.
-
-
Replying to @colmmacc @ErrataRob
All good points & accessibility is likely an issue with any mode of input. What's your confidence in the UX design of the leading major voting machine vendors?
1 reply 0 retweets 0 likes -
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.