,
13 tweets,
2 min read
Read on Twitter
Whatever happened here, this case illustrates why vote-by-mail is no panacea. And Internet-based vote-at-home schemes are even more vulnerable to this kind of mischief at even larger scale.
A lot of difficult tradeoffs here. Vote by mail can increase participation, but because voting is unsupervised, it can make certain kinds of fraud and tampering- as may have occurred in NC - easier.
I’ll just repeat again the widely recommended current best practice for US elections: precinct counted optical scan paper ballots coupled with statically rigorous risk-limiting audits after each election.
Widely expanded early voting options can help increase participation without introducing some of the risk of large scale unsupervised voting that you get with vote-by-mail.
*statistically
Anyway, let me take this as yet another opportunity to suggest the recent National Academies report on voting, which is an excellent overview of many if the issues here. nap.edu/catalog/25120/…
(You can download the pdf for free; you don’t need to buy the ebook)
I’ve said it before, I’ll say it again. In the (too many) decades I’ve been working in security, I’ve yet to encounter any practical problem as fundamentally hard as election security. Nothing even comes close.
Almost all the “simple” solutions ignore or compromise at least one basic requirement. And there’s far more to elections than technology for casting and counting votes.
And in spite of all that, we still have elections and accept their outcomes. There’s an enormous amount of hard work behind that minor miracle.
It’s an easy temptation to dismiss secret ballots as a frivolous nicety, especially if your preferred election tech weakens it. But remember that there are people for whom the ability to vote their conscience without fear of intimidation is what makes their franchise meaningful.
This is a central reason why any universal “vote at home” system - whether computers are involved or not - needs to be approached with extreme caution.
Important also to remember that the test of an election system is not whether it works well when run entirely by honest, highly competent people (as is probability the case right now in most of the US). It’s whether we trust the outcome even when it might not be.
