I’m very sympathetic to desire to reduce cost of voting — especially given the very low level of funding for US elections compared to other countries — but at the moment we are working on the low hanging fruit, both of cost and ease of use. The current market (or really non-market) in elections has some compressible costs, for example, ballot printing costs in excess of a dollar per voted paper ballot, or for electronically cast ballots several dollars (or several dozen dollars depending on the real lifetime of voting computers) per ballot.

Ease of use also has some easy hits, for example, in ballots where the instructions are so obtuse that voters don’t know how to avoid canceling their own votes (e.g. straight party voting and emphasis votes). So there is plenty we can do to help with cost and ease of use in the election methods in use today.

And it is important to do so because eventually computing and networking is going to get thrown at remote voting. As you have no doubt experienced, throwing computing tech at a low-tech situation can make it a lot better or nightmarishly worse. If we can squeeze out some of the cost and difficulty first, we raise the chances that later on, more technology can actually make it faster, better, and cheaper – without compromise on trust, integrity, security, and transparency.

– EJS