A lot of people are laughing at this account of Mitt Romney’s ORCA–and automated GOTV tracking system. Rather than the efficient new system that would leapfrog Obama’s turnout machine, the system crashed even before the evening rush started.
The entire purpose of this project was to digitize the decades-old practice of strike lists. The old way was to sit with your paper and mark off people that have voted and every hour or so, someone from the campaign would come get your list and take it back to local headquarters. Then, they’d begin contacting people that hadn’t voted yet and encourage them to head to the polls. It’s worked for years.
From the very start there were warning signs. After signing up, you were invited to take part in nightly conference calls. The calls were more of the slick marketing speech type than helpful training sessions. There was a lot of “rah-rahs” and lofty talk about how this would change the ballgame.
Working primarily as a web developer, I had some serious questions. Things like “Has this been stress tested?”, “Is there redundancy in place?” and “What steps have been taken to combat a coordinated DDOS attack or the like?”, among others. These types of questions were brushed aside (truth be told, they never took one of my questions). They assured us that the system had been relentlessly tested and would be a tremendous success.
Now a note about the technology itself. For starters, this was billed as an “app” when it was actually a mobile-optimized website (or “web app”). For days I saw people on Twitter saying they couldn’t find the app on the Android Market or iTunes and couldn’t download it. Well, that’s because it didn’t exist. It was a website. This created a ton of confusion. Not to mention that they didn’t even “turn it on” until 6AM in the morning, so people couldn’t properly familiarize themselves with how it worked on their personal phone beforehand.
From what I understand, the entire system crashed at around 4PM.
FWIW, Obama’s campaign had two innovations from 2008 this year. For vote trackers–the same purpose as this website was supposed to serve–they had bar code labels for each voter that the tracker would collect on a sheet to be picked up; I assume–but did not see–someone came and picked up those labels and used them later in the day.
For voting problems, they had a great website that showed the campaign where problems were across the country. That’s the website I used. The website worked great. We got advance access to it to practice. And the customer service was amazing: I had a login problem; I submitted a request to fix it, and it got it fixed 6 minutes after I made the request–I’m hoping OFA buys out Comcast.
There was, for me, one significant problem though: you could only enter problems via the polling place name, not the precinct or the address. I didn’t get my assignment until after I went to bed (very early) the night before the election, so I just got up, checked my phone, and drove there; never really processed what the name of my polling location was. Even if I had, it wouldn’t have been easy to work with: I was in Reform Christian Church Number Yadda Yadda. Given how common Reform Christian Churches in this part of MI and how generic their names are, its name was the functional equivalent of “McDonalds number 2,364.” So for user interface reasons, it didn’t work as well for me as planned. (By comparison, when I called in with problems in 2008 and said I was at “the firehouse,” the local person on the other end of the line knew precisely where I was.)
All that said, the actual website was very nice, and worked well.
So there’s a direct comparison to be made.
Even more, though, this account made me think of one thing: how Mitt Romney advisor Michael Hayden paid SAIC $1 billion to do what NSA could have done, far better, for $3 million. As with that program, Mitt apparently paid a lot of money to get a program that didn’t perform the function it was supposed to.
That–like Mitt’s habit of contracting things out at expensive rates, like his award of big bonuses to the top aides who were deluding him but not the actual workers who would do things like make the voter tracker system work–seems so typical of the GOP way of doing things. Expensive, ineffective overkill.
But hey. The contractors get paid even if the candidate doesn’t win!
Update: Politico has more.
It’s been reported the system crashed at 4 p.m., but multiple sources familiar with the war room operation said it had actually been crashing throughout the day. Officials mostly got information about votes either from public news sources tracking data, like CNN.com, or by calling the counties for information, the source said. Officials insisted the day after the election that they had still believed they were close, and that they had hit their numbers where they needed to, even as Fox News and other outlets called the race.
The numbers in the interface never moved, leaving officials in Boston and out in the states “flying blind” — a phrase used by several people. The workers on the ground didn’t know what doors to knock on or what efforts to make with which voter targets who had not yet turned out – some efforts were made but they were slow and more cumbersome. And the campaign officials also generally didn’t know which precincts to send auto-calls into to try to boost turnout — especially in precincts in Ohio, where there is no party affiliation in the general election. Instead of targeted information, all they really had to work with was the generic raw vote tallies in various counties.
“The whole point of this system was we were supposed to be able to identify who in these precincts had not turned out, who were our supporters,” said one source of the system, which was built at a “substantial” cost. The idea behind it was to use pre-canned, targeted messages to push the voters who hadn’t yet cast a ballot, one of the most basic aspects of Election Day GOTV, which is knowing which supporters have already voted and who still needs to be part of a pull operation.
FWIW, there were several Republican challengers in my poll over the course of the time I was there. They were in close–and effective–phone conversation with the campaign, presumably at the state level. Of course, they weren’t tracking voters at this poll–there were only 40 votes cast for Mitt over the entire day. Mostly, they were looking glumly at the long line of African American voters waiting to legally cast their vote–there was nothing they could do. Still–in a county with a very well run Republican Party, they were well organized, albeit entirely by phone.