This election year has me worried, Optimus.
DT is actively inciting his followers to violence if he loses, which is exactly the kind of mob mentality that makes me dislike political parties in the first place. I'm a proud unaffiliated voter, and always have been. During college my friends and I mocked the fraternity system by calling ourselves GDIs – God Damned Independents. We weren't quite organized enough for making T-shirts, which is one of the costs of being fractious like that.
In any case, my newest career as writer and entrepreneur allows me some flexibility in my schedule. Unlike certain other states, NC actually pays its poll workers ($11.60 an hour, base, which is not a lot, but better than the minimum wage of $7.25). This year's ugly rhetoric was enough to push me over the edge into applying a couple of months ago. The training was minimal (and paid), consisting of how to look up addresses at the check-in station and how to set up the voting machines for individual voters. The rest we would pick up on the job from the more experienced site supervisors (and that is happening; see below). Interestingly, the trainings were scheduled per-site, so that we could meet as many of our co-workers ahead of time as possible. This seems like a little thing, but it could turn out to be important for teamwork reasons.
The site was physically set up in the recreation center gym when we arrived, but we had to boot up all the laptops and printers at the check-in station, as well as the 20+ electronic voting machines. There were people waiting in line to vote when we opened at 8am. Check-in is quick for most people unless they're doing same-day registration. With all 5 check-in stations going, and a 5-page ballot, the bottleneck in time was the wait for an open voting machine. Only 4 or 5 of our voting machines booted up the first time, so we had a long line after check-in for about an hour and a half while the site supervisors (we had 2) got the others up and going. Most people were fairly patient about the wait. I can't be sure, but I imagine that being able to see the entire line and the entire bank of voting machines may have played into their recognition that we were working and quickly and efficiently as possible.
Where it got weird was walking voters over to the machines.
I started hearing all these rumors out of Texas about people seeing their votes changed by the machine (leading one county to switch to paper ballots). Now, I'm no longer a professional neuroscientist, doing experiments, but I'm well aware how flexible perception can be, and how emotion can play into that flexibility. So early on, I lapsed into ongoing lecture mode, talking people through all the stuff I was punching into the machine as I was doing it, careful to maintain a calm, confident, friendly voice:
“This is your precinct. That's just based on where you live. These two letters are the building we're in right now. This is the laptop over there that generated your check-in code. You're the 13th person through that line today. Have you used one of these machines before?”
Most people had, so I generally didn't deliver detailed instructions unless they asked. If an individual mentioned a concern about the validity of the record, I'd point out that the machine was storing the vote electronically, plus there was a paper backup that they could check after each selection to make sure it recorded, if they were worried about that.
Later, at the end of the day, I learned more about the process.
We actually maintain 3 electronic copies of each voter's ballot on-site.
- One is in the machine;
- at the end of the day, the supervisors create two electronic backups of each machine.
- Plus, there are the spooled paper copies of the ballots.
By my count, that's 4 copies of each voter's ballot.
Then there are individual sheets of paper that link the ballot (through a code) to the voter's identity, which both the voter and the check-in worker sign. These are the paper backups for downtown's networked database. That's a total of 6 records for each voter, in multiple formats. Plenty redundant, but not particularly convenient. I'm no security expert, and I'd like to eavesdrop on the Steemit community kibbutzing about how to improve on these systems. (Cory Doctorow made a big deal about transparency as a necessary component of continuous improvement in Walkaway.)
We served over a thousand voters the first day.
Responsibilities were fluid. When I saw a gap, I stepped into it and did what needed to be done, if I knew how. Everyone else seemed to be of the same mindset until around 4:30 or 5:00, when people started to get tired and hungry (more on that later). The two site supervisors did anything complicated, although when it slowed down towards the end of the day they started training whoever was free and willing on things like changing the paper rolls in the voting machines.
I am 46 years old. There were two people there who appeared to be younger than me. One was a kid still in braces who ran the Registrar's station, registering new voters and updating addresses in the database whenever necessary. He could also check in voters if he wasn't busy. The other was an IT guy in real life who spent the day on the check-in laptops until the last hour or so of the polls being open. I was on my feet the entire time until almost 6pm, running machines. As the youngest guy on the floor, I also ended up serving all the curbside voters.
Curbside is CRAZY!
Let's say you physically can't get out of your car (or the car of the people who brought you) and walk into the building. On election day, at your own precinct, I could simply bring you a customized paper ballot, which would be pretty quick. During early voting, on the other hand, where we have people from all over the county, I have to run your address inside to check you in. Then I have to wheel a touch screen voting machine outside and unfold it like an origami crane, or a Transformer, and set it up on the sidewalk. The machines are not heavy, but they are delicate, and not waterproof (so rain, or sleet, would require me to send you home to try again another day; dark of night would actually be helpful, because the screens are not all that bright out in the sun).
Their rechargeable batteries have a limited life, and if you have not marked up a sample ballot, it can take a long time to work your way through five pages. Straight-party ticket voting was outlawed by the legislature about four years ago, so you do have to work your way through all five pages, even if you choose to skip individual races. That's assuming you can stand outside your car long enough to do it, rather than having your driver punch the buttons for you. I have to stay there there the whole time to guard the machine, although I stand back to give you some privacy.
Given the horrific inefficiency of this process, why the hell would anyone choose not to get an absentee ballot and vote from home? One reason I heard is that the voter feels the public vote is ritually important. Noone stated it in those words; I'm translating. The other is that old people often have terrible memories, and the window to request an absentee ballot closes 30 days before early voting starts.
I spent a total of probably two hours yesterday serving half a dozen individual curbside voters.
I mention that not as a complaint. I mention it to demonstrate that the poll workers that I was with yesterday are crazy dedicated. We are willing to go to ridiculous lengths to help you vote. If you want to blame someone for trying to suppress your vote, or otherwise rig the election, look to your legislature, not to some conspiracy among the people running the election.
After the polls closed at 6:30pm, we spent another hour and a half putting all the paper copies of the check-in records in numerical order and counting them to make sure they matched the number of ballots cast on the machines. That made for a twelve-hour shift, and by the end people were tired and hungry, which slowed the process considerably. I understand that a shift change, bringing in new people in the middle of the day, would have been a logistical nightmare (shift change is when most hospital mistakes happen). The problem was that most of us, including me, had only brought lunch, not supper. That was dumb. Fortunately, like Steemit's own fasting researcher @cristi, I personally am pretty used to skipping meals, so I put myself and my ketone-adapted brain in charge of un-shuffling some of the stacks of paper.
Lesson one: first put them in order and determine if any sequential numbers are missing, which could happen if one gets printed with the wrong address. Usually the operator will recognize that there's a discrepancy and resolve it before printing, but not every time. The check-in laptops apparently NEVER over-write a number on print-out; if it takes three prints to correctly match a voter's name and address, that will generate three sequential check-in codes, two of which are canceled manually without progressing to a ballot. Once in a while the voter will sign it before noticing a discrepancy, and then tear it up or throw it away, while the check-in person is busy printing the corrected one. Stuff like that happened about ten times out of over a thousand voters.
Ten needles in a haystack of a thousand straws makes you wish for a magnet. You can find them by hand, but it's a pain in the ass.
Lesson two: Look for missing codes by counting down, not up. This is basic cognitive neuroscience. Expectations can cause you to see what you expect to see, especially when you're tired, and most people are less used to counting down. Repeating a sort or a count with the same expectations leads to the same result, multiple times. Not helpful.
Lesson three: After the sort, if a count is necessary, count the papers from their blank backs, not from their printed fronts, where looking at the numbers can substitute the number you're looking at for the number you're trying to hold in your tired-ass short-term memory. Kind of like the classic Stroop effect.
None of these led to end-of-shift discrepancies.
We reconciled every check-in record with every ballot count, but it took longer than it should have because of fatigue and temporary mistakes. We had to sort and count some stacks multiple times, with multiple people checking one another's work to find those temporary mistakes. Again, we found and reconciled every single thing, but with better procedures (and snacks!) it could have gone a lot faster.
That was day one.
Today I'm resting up and writing up, before I go back tomorrow. Saturday may be even busier.
When he's not saving democracy from low blood sugar, your friendly neighborhood neuroscientist Randall Hayes blogs here on Steemit and writes the monthly column PlotBot for online SF magazine Orson Scott Card's Intergalactic Medicine Show.
for a counter-argument (which seems probable, and a balm to my ears)
but from the first article: It only takes one. Anyone remember Timothy McVeigh, who bombed the federal building in Oklahoma? In that case there were three co-conspirators convicted, but you get the idea.
"Republicans say the move has nothing to do with race. They argue the straight-ticket option prevents voters from thinking critically about individual candidates. They also say it encourages partisan polarization."
They're not mutually exclusive. You can do all three of those things with the same measure.