As mentioned there, I have reason to distrust electronic voting machines, which stir the unreliability of PC-based computing into the boiling pot of election politics.
Attempting to open the polls with the Administrative Menu on the LCD produced this incorrect response. Fortunately, the next screen in the Reports section had a Cancel option, so I could back out and try again by tapping the screen well above the Open Poll button. That worked.
Later on in the day, for the first time in my experience as a Ballot Marking Device Election Inspector, a voter requested to use the BMD machinery to cast her ballot.
Here’s what happened when I tried to start the somewhat misleadingly named Audio Session that invokes the BMD: Ballot Review turns on a mode that presents the scanned values from the next ballot on that tiny little LCD, one contest at a time.
When I called the Board of Elections to get help, the tech said “Hmmm. That shouldn’t happen.” We did get the Audio Session started and the voter commenced entering her choices, eventually succeeding in producing a printed ballot that she found satisfactory.
The tech sent to fix the situation (we Election Inspectors are not encouraged to fiddle around with the machinery, for well and good reason) was stumped. Eventually we scanned a ballot, using a live vote as a debugging aid, and managed to get the option turned off again. Obviously we hit a corner case, but that’s not what you want in an election with voters lining up behind a dysfunctional scanner.
It was, of course, the one-and-only scanner in the polling place.
While this does not directly affect the election results, it certainly does not inspire confidence in the architecture, the programming, or the operator training of the election system.
Not a pleasant experience…
The Smell of Molten Projects in the Morning 
Comments
2 responses to “Voting Machines: More Distrust Thereof”
On the plus side, there were no lines at my polling place, which was usually congested in the past when we had just one lever machine per precinct.
The only problem I had was that the poll workers didn’t tell me about or give me a ballot privacy sleeve, so I had to carry my ballot folded over to the scanner.
a ballot privacy sleeve
After 3/4 of the first two dozen voters had their ballots rejected multiple times for either overvotes or ambiguous marks, due entirely to problems with the sleeves, we started instructing each person how to get it done. That reduced the problems to about 10%, essentially all of which were false rejects.
Scanning without the sleeves worked perfectly nearly every time.
The sleeves are a disaster, particularly when handled by short women with no experience using the machine who were wearing heavy coats and carrying pocketbooks. Evidently, all the testers (for whom the machines work perfectly) were males in a shirt-sleeve environment like, oh, the voting machine warehouse.
We’re trying to get the votes cast on that machine hand-verified, but that looks like it’s going to take some doing…