The results of the much-anticipated Democratic Iowa caucus have been delayed due to problems with the voting system. The Iowa Democratic Party has blamed it on a “coding issue” within a mobile phone app used to help count the votes.
Ken Birman, an expert in cloud computing in Cornell University’s Department of Computer Science, researches trustworthy, secure and scalable mobile applications. He says the delayed Iowa caucus results represent a warning for future elections and highlight the need for oversight of technology used in voting.
“The Iowa situation was to be expected based on the unwarranted trust that the public feels towards computing systems and ‘apps.’ Last night’s problems occurred because of human errors. But next November, we could easily see outright attacks, aimed both at changing outcomes and at destabilizing our democracy.
“The bottom line is that today’s mobile apps are seductively easy to build. Many are released without proper testing, and there is often minimal support if problems occur when installing them or setting up accounts. Voting is far too important to trust to such a haphazard process.
“The United States needs tough new legal protections and oversight spanning all forms of technology used in voting. We know how to protect air traffic control systems, electronic medical health records, and other kinds of critical data. The same level of oversight is needed for voting systems.
“We all know the expression ‘trust but verify.’ This is the second key lesson of Iowa: Today’s confusion will quickly be resolved, because there is a complete paper record that can be used to reconstruct the exact caucus outcomes. Electronic voting systems of every kind must maintain paper trails. Even if the system seems to be working flawlessly, random audits should always be used to check for any possible discrepancies.
“Technology mishaps are inevitable. Audited paper trails offer the best and most cost-effective remedy.”