What I don't get is the interest in complex tech products for complex tech products' sake. Any sort of "modern" voting device is going to have a higher barrier

What I don't get is the interest in complex tech products for complex tech products' sake. Any sort of "modern" voting device is going to have a higher barrier of entry. Even if you assume a completely non-adversarial environment, it's simply harder. How does your touchscreen machine work in a power outage? Will the minimally trained election staff bumble setup, causing long queues for the working machines? How much is the per-election cost of ownership, before it has to be replaced with a new model that doesn't run on WinXP? In my county, we use paper ballots with a "join the lines to select a candidate" model. The ballots can be scanned quickly to tally them-- the technology is basically a large-format version of the Scantron slips they've been using in schools for 40+ years. We had a comical "audit" which ended up blowing up in the instigators' face when even their chosen super-sketchy auditors couldn't find meaningful fraud. When this is the bar, what is the exact advantage being presented by these big ugly black-box voting machines? I never understood their appeal, especially to government bodies that are normally fairly risk-averse and perpetually underfunded. The other side of things is perceptual: the simpler and more tangible the system is, the easier it is for people to feel like they understand and trust it. This means any system-- whether it's the old lever machines or a touchscreen is always going to start at a disadvantage versus a manually-inscribed ballot.