Stan Stanart talks election security
Source: Charles Kuffner
I have a few thoughts as well.
vote-button
Over the course of the presidential race, concern has grown about digitally safeguarding election results.
New cyber security threats seem to emerge monthly. Republican Donald Trump has repeatedly contended the presidential election will be “rigged.” And suspected Russian hackers have broken into computer systems of the Democratic Party.
“With so much news out there, people are concerned,” acknowledged Harris County’s top election official, Stan Stanart, at a news conference Thursday.
However, Stanart sought to reassure the public that all necessary defenses are up and there is no way Harris County’s election will be hacked or rigged, because it is not connected to the internet.
[…]
“Our elections are too important to leave them open to attack,” said Dan Wallach, a Rice University computer science professor who testified in September to Congress on election cyber security. “We need to do better.”
The most attractive part of an election system for a malicious attack, he said,is the voter database – in Harris County, it’s a list of nearly 2.2 million registered voters. If hackers successfully deleted it, chaos would ensue.
But the county database is kept offline, invulnerable from the outside. Even so, Stanart, the county clerk, said his office, the county tax assessor’s office, and the Texas secretary of state save a backup copy every day.
“There are many eyes and there are many triggers in the whole system that would notify us, and we would observe if there were any issues with any registrations being changed,” Stanart said. “I assure you there’s no problem there.”
Wallach agreed that the daily database backups provided excellent protection.
The fact that the voting machines are not connected to the Internet is a good thing. Dan Wallach (who is a friend of mine) has some criticism of the “secure network” setup for transferring the voting data from the individual memory cards to the central network, but I agree with him that this is an unlikely target for attack. The main vulnerability here is what it has always been, with the cards themselves and their handling. If a card becomes corrupted or lost before its contents can be uploaded, there’s no backup. This is why people like Wallach have been calling for paper receipts to be included. That problem, and the accompanying risk, cannot be solved with the current voting machines. I don’t know how big that risk is – in over a decade of using the eSlate machines, we have not had this problem, but the downside if it happens even once is enormous, and these machines are at the end of their lifecycle with no obvious path forward. But hey, maybe we’ll make it through another election.
As for the voter registration data, it’s really a question of the county’s network security overall. There are a lot of pieces to this, so I’ll just focus on the question of monitoring. As long as they monitor all changes to the voter registration file – what, when, by whom – and they have someone keeping an eye on that, then they’re probably OK.
So I tend to agree that at the very least there’s nothing new or unusual to worry about this year, and I appreciate Stanart making the effort to address that. We should always be vigilant, but let’s not lose perspective, and let’s not worry about things that aren’t worth worrying about. If only Stanart took that same approach to the far smaller risk of in person vote fraud.
| }
|