Dan Cohen - Digital Humanities Blog

Syndicate content
Updated: 15 hours 6 min ago

Age of Asymmetries

Wed, 16/08/2017 - 18:56

Cory Doctorow’s 2008 novel Little Brother traces the fight between hacker teens and an overactive surveillance state emboldened by a terrorist attack in San Francisco. The novel details in great depth the digital tools of the hackers, especially the asymmetry of contemporary cryptography. Simply put, today’s encryption is based on mathematical functions that are really easy in one direction—multiplying two prime numbers to get a large number—and really hard in the opposite direction—figuring out the two prime numbers that were multiplied together to get that large number.

Doctorow’s speculative future also contains asymmetries that are more familiar to us. Terrorist attacks are, alas, all too easy to perpetrate and hard to prevent. On the internet, it is easy to be loud and to troll and to disseminate hate, and hard to counteract those forces and to more quietly forge bonds.

The mathematics of cryptography are immutable. There will always be an asymmetry between that which is easy and that which is hard. It is how we address the addressable asymmetries of our age, how we rebalance the unbalanced, that will determine what our future actually looks like.


Irrationality and Human-Computer Interaction

Thu, 22/06/2017 - 21:07

When the New York Times let it be known that their election-night meter—that dial displaying the real-time odds of a Democratic or Republican win—would return for Georgia’s 6th congressional district runoff after its notorious November 2016 debut, you could almost hear a million stiff drinks being poured. Enabled by the live streaming of precinct-by-precinct election data, the dial twitches left and right, pauses, and then spasms into another movement. It’s a jittery addition to our news landscape and the source of countless nightmares, at least for Democrats.

We want to look away, and yet we stare at the meter for hours, hoping, praying. So much so that, perhaps late at night, we might even believe that our intensity and our increasingly firm grip on our iPhones might affect the outcome, ever so slightly.

Which is silly, right?

*          *          *

Thirty years ago I opened a bluish-gray metal door and entered a strange laboratory that no longer exists. Inside was a tattered fabric couch, which faced what can only be described as the biggest pachinko machine you’ve ever seen, as large as a giant flat-screen TV. Behind a transparent Plexiglas front was an array of wooden pegs. At the top were hundreds of black rubber balls, held back by a central gate. At the bottom were vertical slots.

A young guy—like me, a college student—sat on the couch in a sweatshirt and jeans. He was staring intently at the machine. So intently that I just froze, not wanting to get in the way of his staring contest with the giant pinball machine.

He leaned in. Then the balls started releasing from the top at a measured pace and they chaotically bounced around and down the wall, hitting peg after peg until they dropped into one of the columns at the bottom. A few minutes later, those hundreds of rubber balls had formed a perfectly symmetrical bell curve in the columns.

The guy punched the couch and looked dispirited.

I unfroze and asked him the only phrase I could summon: “Uh, what’s going on?”

“I was trying to get the balls to shift to the left.”

“With what?”

With my mind.”

*          *          *

This was my first encounter with the Princeton Engineering Anomalies Research program, or PEAR. PEAR’s stated mission was to pursue an “experimental agenda of studying the interaction of human consciousness with sensitive physical devices, systems, and processes,” but that prosaic academic verbiage cloaked a far cooler description: PEAR was on the hunt for the Force.

This was clearly bananas, and also totally enthralling for a nerdy kid who grew up on Star Wars. I needed to know more. Fortunately that opportunity presented itself through a new course at the university: “Human-Computer Interaction.” I’m not sure I fully understood what it was about before I signed up for it.

The course was team-taught by prominent faculty in computer science, psychology, and engineering. One of the professors was George Miller, a founder of cognitive psychology, who was the first to note that the human mind was only capable of storing seven-digit numbers (plus or minus two digits). And it included engineering professor Robert Jahn, who had founded PEAR and had rather different notions of our mental capacity.

*          *          *

One of the perks of being a student in Human-Computer Interaction was that you were not only welcome to stop by the PEAR lab, but you could also engage in the experiments yourself. You would just sign up for a slot and head to the basement of the engineering quad, where you would eventually find the bluish-gray door.

By the late 1980s, PEAR had naturally started to focus on whether our minds could alter the behavior of a specific, increasingly ubiquitous machine in our lives: the computer. Jahn and PEAR’s co-founder, Brenda Dunne, set up several rooms with computers and shoebox-sized machines with computer chips in them that generated random numbers on old-school red LED screens. Out of the box snaked a cord with a button at the end.

You would book your room, take a seat, turn on the random-number generator, and flip on the PC sitting next to it. Once the PC booted up, you would type in a code—as part of the study, no proper names were used—to log each experiment. Then the shoebox would start showing numbers ranging from 0.00 to 2.00 so quickly that the red LED became a blur. You would click on the button to stop the digits, and then that number was recorded by the computer.

The goal was to try to stop the rapidly rotating numbers on a number over 1.00, to push the average up as far as possible. Over dozens of turns the computer’s monitor showed how far that average diverged from 1.00.

That’s a clinical description of the experiment. In practice, it was a half-hour of tense facial expressions and sweating, a strange feeling of brow-beating a shoebox with an LED, and some cursing when you got several sub-1.00 numbers in a row. It was human-computer interaction at its most emotional.

Jahn and Dunne kept the master log of the codes and the graphs. There were rumors that some of the codes—some of the people those codes represented—had discernable, repeatable effects on the random numbers. Over many experiments, they were able to make the average rise, ever so slightly but enough to be statistically significant.

In other words, there were Jedis in our midst.

Unfortunately, over several experiments—and a sore thumb from clicking on the button with increasing pressure and frustration—I had no luck affecting the random numbers. I stared at the graph without blinking, hoping to shift the trend line upwards with each additional stop. But I ended up right in the middle, as if I had flipped a coin a thousand times and gotten 500 heads and 500 tails. Average.

*          *          *

Jahn and Dunne unsurprisingly faced sustained criticism and even some heckling, on campus and beyond. When PEAR closed in 2007, all the post-mortems dutifully mentioned the editor of a journal who said he could accept a paper from the lab “if you can telepathically communicate it to me.” It’s a good line, and it’s tempting to make even more fun of PEAR these many years later.

The same year that PEAR closed its doors, the iPhone was released, and with it a new way of holding and touching and communicating with a computer. We now stare intently at these devices for hours a day, and much of that interaction is—let’s admit it—not entirely rational.

We see those three gray dots in a speech bubble and deeply yearn for a good response. We open the stocks app and, in the few seconds it takes to update, pray for green rather than red numbers. We go to the New York Times on election eve and see that meter showing live results, and more than anything we want to shift it to the left with our minds.

When asked by what mechanism the mind might be able to affect a computer, Jahn and Dunne hypothesized that perhaps there was something like an invisible Venn diagram, whereby the ghost in the machine and the ghost in ourselves overlapped ever so slightly. A permeability between silicon and carbon. An occult interface through which we could ever so slightly change the processes of the machine itself and what it displays to us seconds later.

A silly hypothesis, perhaps. But we often act like it is all too real.