A “black box” is, in technological terms, a system that can only be viewed by its inputs and outputs. The inner-workings of a black box remain a mystery to any outside observer. In the world of voting, Direct-Recording Electronic (DRE) voting machines are often referred to as technological “black boxes.” The software and hardware of these machines are the property of election vendors and their functionality remains enigmatic to their users. This is true for the vast majority of people who use these machines. We show up to the polling place, make our selections on a screen in front of us, and assume that our selections will be counted correctly.
The mysterious nature of these machines leaves voters to trust that the voting system vendors responsible for their construction have built something secure and reliable. This dynamic of trust is consistently criticized by academics and security experts for being inherently undemocratic. These systems make hacking easy and auditing difficult.
We at Votem agree that the technology for all types of voting systems, including a mobile voting system, should be auditable. We do not agree that the notion of ‘security by obscurity” is enough to thwart malicious attacks from hackers and other nefarious actors. Where we believe the paper-based proponents argue incorrectly, however, is the statement that the current “black box” nature of most voting machines negates the possibility to use any form of digital voting. Halderman and many of his colleagues advocate for a system that is “software independent”, or a system in which “an undetected change or error in its software cannot cause an undetectable change or error in an election outcome.” As a mobile voting company, we naturally take issue with a strategy that is completely independent of software, but the underlying theory is sound.
As we have discussed previously in this series, paper-based voting systems pose a variety of risks. If we want to create a system that moves forward instead of backward, it is important to incorporate the tenets of the software independent theory while also creating a solution that is more accessible and verifiable than anything currently in mainstream use.
Two months ago, we published an overview of blockchain voting and its benefits. As explained, each transaction written on the blockchain, votes in this case, are permanently and immutably written onto the digital ledger if, and only if, the majority of the approved parties approve of the transaction. Each and every vote cast can be audited for accuracy, both by administrative parties and individual voters themselves. The blockchain solution is, of course, not completely software independent, but it incorporates some of the most important tenets of the philosophy. Below is a quote from the first paper, by Ron Rivest and John Wack, to define software independence:
“The answer is that a voting system is software-independent if, after consideration of its software and hardware, it enables use of any election procedures needed to determine whether the election outcome is accurate without having to trust that the voting system software is correct.”
While a blockchain system is still rooted in software, the immutable nature of the ledger it creates allows an election administrator or a voter to verifiably audit an election outcome. The system also opens the doors to mobile solutions that will lessen, and eventually eliminate, the accessibility and security problems inherent to paper-based systems.
Many experts in the cybersecurity space may detest the idea, a blockchain system is the most democratic way to keep a voting system accountable to the collective public it serves while offering an unmatched level of transparency. In a future where our lives will be continually impacted by software, blockchain represents a path forward for democratic systems all over the world.