Future Imperfect #34: Algorithm as art
Welcome to Future Imperfect! This week I’ve been reading about algorithmic artists, the future of gravel on America’s highways, the impact of increasingly perfect athletes, glasses-free 3D movies, and a cyberpunk short story on “recoding” humans.
If this is your first time here, be sure to follow the FI publication to get this newsletter in your inbox each week. If there’s someone you know who is into future trends, forward this along! As a reward, I’ll send you ten Schrute bucks.
Algorithm as art
Time to get ready for a new definition of artist. Last week, Oliver Roeder wrote about the future of computer-generated art, or should I just say, art.
In December 1964, over a single evening session in Englewood Cliffs, New Jersey, John Coltrane and his quartet recorded the entirety of A Love Supreme. This jazz album is considered Coltrane’s masterpiece — the culmination of his spiritual awakening — and sold a million copies. What it represents is all too human: a climb out of addiction, a devotional quest, a paean to God.
Five decades later and 50 miles downstate, over 12 hours this April and fueled by Monster energy drinks in a spare bedroom in Princeton, New Jersey, Ji-Sung Kim wrote an algorithm to teach a computer to teach itself to play jazz. Kim, a 20-year-old Princeton sophomore, was in a rush — he had a quiz the next morning. The resulting neural network project, called deepjazz, trended on GitHub, generated a buzz of excitement and skepticism from the Hacker News commentariat, got 100,000 listens on SoundCloud, and was big in Japan.
This half-century gulf, bracketed by saxophone brass and Python code, has seen a rise in computer-generated music and visual art of all methods and genres. Computer art in the era of big data and deep learning, though, is a reckoning for algorithms, capital-A. We must now embrace — either to wrestle or to caress — computer art.
Why does this matter? If an algorithm writes a song, or paints a picture, is it still considered art? Is it art created by the human that writes the code, or is the art attributable to the algorithm itself?
It’s a damned good question. Ultimately I suppose it has to lie with the human until we figure out how to determine algorithmic sentience. Until then, Benjamin probably has to wait to create his IMDb page.
Back to basics (on the highway)
Sometimes the most advanced technological solution is not the most efficient or effective. Smart roads have their uses, but for many jurisdictions around the world paying billions of dollars on road paving and maintenance is just not realistic. Instead, let’s consider the future of gravel.
City Hall received a hollering from a couple living on Bliss Road in the Vermont capital who wanted to sell their home, but feared the horrifying pavement in front of the house would scare away buyers. They had reason to be pissed off: The city of 8,000 people ranks pavement on an index of one to 100. Bliss Road scored a one.
Repaving roads is expensive, so Montpelier instead used its diminishing public works budget to take a step back in time and un-pave the road. Workers hauled out a machine called a “reclaimer” and pulverized the damaged asphalt and smoothed out the road’s exterior. They filled the space between Vermont’s cruddy soil and hardier dirt and gravel up top with a “geotextile”, a hardy fabric that helps with erosion, stability and drainage.
In an era of dismal infrastructure spending, where the American Society of Civil Engineers gives the country’s roads a D grade, rural areas all over the country are embracing this kind of strategic retreat. Transportation agencies in at least 27 states have unpaved roads, according to a new report from the National Highway Cooperative Highway Research program. They’ve done the bulk of that work in the past five years.
Why does this matter? According to the American Road and Transportation Builders Association, it costs $1.25 million per mile to resurface a 4-lane road. That’s funding that many jurisdictions just don’t have.
Gravel isn’t a complete panacea. The article rightly points out the indirect cost of vehicle maintenance as a result of rougher roads. But there’s a strong argument to be made that pothole-ridden roads aren’t much better. If state and local governments are looking for cost savings—this is one way to do it.
Result: Draw
In How We Get To Next, a Medium publication about “inspiring stories about the people and places building our future,” James Bridle shares an account of athletes hurtling toward algorithmic perfection—or rather, a dull tedious heat death of sport.
The concept of the “ghost car” is familiar from computer games: an ethereal representation of one’s best performance, almost always infuriatingly, unattainably ahead, except in the rare moment you break through, scoot past, and set a new high score, a new yardstick of attainment. It’s a race against oneself, an attempt to better oneself, to improve, to self-perfect. But the Raptors’ ghost players are something very different. They are players who have never existed, might never exist, but whose performance can be benchmarked and emulated. Instead of bettering themselves, the players can be trained to better resemble the meticulous creation of the algorithm: zombies playing out an imperfect recreation of an already-determined scenario.
The gap between what can be predicted and what can be acted on is closing fast. Algorithmic commands don’t need to be pre-programmed but can be determined once the match is already in progress. Sportstech LLC, a company founded by an astrophysics professor and his programmer son, has filed patents for in-game predictions: a motion-tracking system capable of alerting players to a shot’s accuracy at the very moment it is fired. Potential applications include a flashing light to tell basketball players a shot is on target before it lands, so that the defenders could immediately go offensive without bothering to actually play any more defense, and a vibrating wristband for goalkeepers which would prevent them from conceding corners by deflecting balls which weren’t going in anyway. Who needs team practice when you have machine-augmented precognition beamed right into your sensorium?
As image recognition, motion capture, and raw computational capacity improves, it’s obvious that the trajectory is toward not better endurance or ball skill, but better prediction and emulation. Players become actors — or reenactors — and the sports stars of the future will be those who can best and most quickly adapt themselves to the dictates of algorithmic play. Likewise, the successful teams — if they aren’t already, and some of them surely are — will be those with the most programmers on side, the most AI techs, the best simulation teams, and the best visualizers. Ultimately, perhaps, we will reach the perfect state described in our opening scenario: the ideal game which tends, inevitably and mathematically — and unfortunately for the spectators — toward the draw.
Perfect pass, perfect tackle, locked in a computationally impossible cycle. The only winning move is not to play.
Why does this matter? I’ve personally shifted my opinion on this over the years. When I was younger, I was very committed to the idea of added technology into sports. Even when human error worked to my team’s favor, it felt to me that it was right—it was fairer—to get the perfect call every time. And to a certain extent I still believe this. But I’m growing to question the impact of technology on sports. Not because I’m anti-technology or anti-analytics (I have an analytics background at Atlantic Media), but because I think it would be a net loss for the enjoyment of sports.
What is ultimately interesting about sports, anyway? It’s not just about your team winning (even if that’s a big part of the appeal). It’s about another team failing. It’s about seeing the unexpected. It’s about observing the limits of human capacity (both mental and physical). Instead, the “ideal game” is repeated victories by the favorites time after time. Goodbye Cinderella.
I think I’m willing to keep a bit of human error to make sure we keep that essence that makes sports fun to play (and watch).
Removing the physical interface
If you’re like 99% of movie fans out there, you hate 3D glasses. They’re clunky and limit your view of the screen. Thankfully, from MIT, an early look at what glasses-free 3D might look like at a theater near you.
The key insight with Cinema 3D is that people in movie theaters move their heads only over a very small range of angles, limited by the width of their seat. Thus, it is enough to display images to a narrow range of angles and replicate that to all seats in the theater.
What Cinema 3D does, then, is encode multiple parallax barriers in one display, such that each viewer sees a parallax barrier tailored to their position. That range of views is then replicated across the theater by a series of mirrors and lenses within Cinema 3D’s special optics system.
“With a 3-D TV, you have to account for people moving around to watch from different angles, which means that you have to divide up a limited number of pixels to be projected so that the viewer sees the image from wherever they are,” says Gordon Wetzstein, an assistant professor of electrical engineering at Stanford University, who was not involved in the research. “The authors [of Cinema 3D] cleverly exploited the fact that theaters have a unique set-up in which every person sits in a more or less fixed position the whole time.”
Why does this matter? What is the perfect interface? An interface without physical limitations. In the long term, screens have to disappear. This is a limited step in the right direction, but the cultural implications are bigger.
As people get to try this technology, they will expect to see it elsewhere. Why force them to experience AR through their phone when a headset (better than Google Glass) allows for a hands-free experience? Then, why force them to use a headset when augmented contact lenses could allow for an effectively invisible hardware interface? Lastly, why force them to use any external hardware interface when the interface could be an internal implant?
To recode, or not to recode
Check out Conversion Ratio, a cyberpunk short story from ReTech, published in the Exolymph newsletter. ReTech is worth a look too, framed as “the vision of one man lost in the noise of the post industrial collapse.” This story, however, follows a person looking for a bit of augmentation to get out of his present circumstances.
Thirteen days ago Swen held the hand of a 147-year-old woman who did not receive one call, one text, a single feed mention, nor have anyone claim her things after she died. This was not the sad part to Swen. Millions died like that every year. What made him maudlin was that he’d end up in a bed the same way, in a hundred or so years. The thought of some young forty-year-old sitting with him as he died, just because the kid had to, was repulsive enough.
But the thought of an adventureless life nauseated Swen.
Twelve days ago, he asked Sully if he still had friends that recoded. Swen didn’t try to get Sully drunk first. He didn’t do it over dinner or in some coy fashion, just-so-happening to mention the topic in conversation. Instead Swen walked into Sully’s apartment, smiled, said hello, kissed him lightly, and asked matter-of-factly: “Can you get me in touch with a recoder? I’m tired of being on basic and I want to make enough money so I’m not stuck anymore.”
Sully paused mid-breath for a moment. A slice of black hair slid down over his left eye. He didn’t bother to push it back. He didn’t even bother to breath until his brain reminded him to. Then, slowly, he sputtered: “Is this legal money or illegal?”
Swen’s smile broadened. “It’s legal if you win it.”
GIF of the Week: The Bernoulli Effect
Like Future Imperfect? Click that recommend button below so others can find it too! And send me feedback at lasky.joshua@gmail.com.
