|
Auditing the count is an entirely separate issue.
I described the problem of implementing a single voting machine user interface that would accommodate all voters and make it possible for them to cast their vote independently, regardless of their disability. I contend that it's not actually feasible: the best you can realistically hope for is to accommodate as wide a group as possible, and we should expect vendors to do exactly that. Thus I think it's reasonable to ask vendors to supply an audio interface for blind voters, and a sip-and-puff interface for voters who lack the necessary motor control for a mouth-held stylus. On the other hand, I don't think it's reasonable to expect the same voting machine to incorporate an ocular scanner or a brain-activity monitor to accommodate voters who can't use the sip-and-puff. Unfortunately that means there's probably someone who can't use the system unassisted. But I don't know what to do about that... there's no alternative system available that would let those voters vote independently.
As to your example above, I don't know of any system that actually creates an actual audio record of the voter's choices. All the systems I'm familiar with record the voter's choices as an electronic record. Diebold and Sequoia also offer units that print a paper record in addition to the electronic record (the printer units are optional in both cases). Neither company offers an option that lets a blind voter independently verify the paper record -- that's not possible since blind voters can't see what's on the paper. There were originally requirements proposed by California that required independent playback of whatever text was sent to the printer, but those requirements were dropped. Mostly because it's not actually a great solution: you'd end up with some kind of robotic speak-and-spell device. Quality would be extremely low (bear in mind that playback would be handled by an independent piece of hardware interposed between the voting machine and the printer, and not by a sophisticated computer), pronunciation of names would be frequently garbled and handling of non-English languages would be impossible. You also have the problem of who does the playback software -- the main point of the voter-verifiable printout is to eliminate the issue of trust (or mistrust) of the equipment manufacturer, so it makes no real sense for the manufacturer to write the code for the speak-and-spell unit. But if the manufacturer doesn't write it, then who will? And who will get it certified and tested? That sort of thing. So anyway, California removed the speak-and-spell requirement and blind voters are currently required to trust that the audio summary of their choices matches what's printed on the paper... at least until new updated standards are put in place to address the issue. I'm not sure what the right solution is to accommodate independent verifiability for everyone, but if I ever think of it I bet I could make a lot of money selling it, lol. :)
Hand-countable paper ballots are independently auditable, as you point out. You just need to pay for that independent auditability in terms of increased labor costs and lower accuracy. Most election officials would probably accept the latter, but the former's a problem given that their operating budgets are small and they can't get enough volunteers as it is. If you can convince some of them otherwise, then great, but it seems like a losing battle.
As to your last point, my impression is that the majority of disabled people want the ability to vote independently, without assistance, just like you or me. They liked the DREs when they first came out because those machines put everyone -- abled and disabled -- on the exact same footing for the first time. Advent of the voter-verified paper trail disrupts that: because it's not a "voter-verifiable" anything for people who can't see, and consequently the equality they enjoyed without the paper trail is gone. Ditto on the automark-style devices as well, for what it's worth: the device prints a nice paper ballot for them, and it's scanned by the same reader, but the blind voter has to take it 100% on faith that what's printed on the paper reflects what they really selected. So there's no equality, and that's a core problem for disability rights advocates. I'm not disabled myself, so this is all mostly just my own understanding of their viewpoint as I've heard it expressed. You'd be better off talking to people in the disabled community to get a better opinion than mine!
Neil
|