> Paper ballot design can tip elections -- most famously in 2000, and elsewhere. Voting machines also suffer from myriad issues - e.g. unresponsive touchscreen

> Paper ballot design can tip elections -- most famously in 2000, and elsewhere. Voting machines also suffer from myriad issues - e.g. unresponsive touchscreens. And there are countless paper voting systems without the issues of the 2000 election - in fact they are the norm. Framing this as an inherent problem of paper that can only be solved by computers is dishonest. > Physical security is standard practice for computer systems. It's widely understood that somebody with physical access can make changes far beyond what anybody can do remotely. My point exactly. No black box can be trusted. > Is it worth the tradeoff? Let me remind you that the tradeoff is having elections silently and near-undetectably [1] stolen, and sow enough justifiable doubt to destabilize a democracy. [1] Exit polls have been as much as 30% off, especially in smaller elections through which nearly all politicians must pass, and that just passes the buck to securing polls. To dispel any naive ideas that polls would tell us about tampering. And that's in countries where people can answer those polls with relatively little worry. Imagine the disparity if they feared a militia would find they voted wrong and execute them. That's not a hypothetical. Voting systems must be resistant to such circumstances, because they may befall your country as well.