Art Bots and Talking Blenders: A Stroll Through Ars Electronica

For one week every September, things get a little weird in Linz, Austria, a picturesque, pastel city on the Danube. For instance, this year, robots roamed the main shopping avenue. Lasers “painted” the façade of the majestic Mariendom cathedral. And the city’s cobbled streets were filled with an eclectic mix of artists, technologists, and scientists, as they typically are at this time each year.

This annual disturbance is the Ars Electronica Festival, one of the world’s premiere digital and media art fairs. For 35 years, Ars Electronica has been a showcase for art at the intersection of science, technology, and society. The 2014 festival wrapped earlier this month, having featured 579 artists—from 59 countries—who staged 427 artworks around the city center. Here are just a few of the mind-bending installations visitors might have glimpsed on a stroll through Linz.

German artist Boris Petrovsky traces the inspiration for his interactive artwork Das Vergerät back to the whirring appliances in his mother’s kitchen. (“Das Vergerät” translates roughly to “This Appliance.”) The installation consists 104 household machines—everything from hair dryers and shavers to lawnmowers, epilators, and blenders. The goal is to make them speak our language.

Here's how it works: A visitor pushes a button and speaks into a mic. A few seconds later, while donning a pair of headphones, she hears the appliances “speak” her sentence back to her using the sounds of their own internal motors and mechanisms. 

Speech, Petrovsky explains, is really just an assemblage of phonemes: the sounds like “p,” “b,” “sh,” and “r” that help us to distinguish one word from another. To create Das Vergerät, he first had to determine what phonemes each of his appliances could produce, and at what pitch. (An adult’s voice, for example, will be pitched lower than a child’s. Petrovsky wanted to account for those speech differences too.) Then Petrovsky and his collaborator, programmer Georg Nagel, hacked the machine’s on/off buttons to run on software that enabled them to collectively “listen” to a human participant speaking through the mic, then parrot her phrase back to her. The “speech” a participant hears through her headphones is actually a time-cut collage of individual machines’ “phonemes.”

The appliances’ “speech” isn’t perfect. If a passerby doesn't hear what a participant initially speaks into the mic, she’s unlikely to understand what the machines are saying. But Petrovsky insists that ambiguity is also part of the artwork. “Is there really an understanding between man and machine, or is it a big misunderstanding?” he asks. Das Vergerät might be humorous and absurd by design, he says, but such questions will be increasingly relevant as the “Internet of things” develops.

Don’t rub your eyes. American artist Jacob Tonski’s Balance From Within, pictured above, is exactly what it looks like: a 170-year-old sofa balanced perfectly on one leg. It’s a metaphor, Tonski explains, for the balancing acts we all do to maintain our personal relationships—relationships that often play out in conversations and time spent on a sofa.

Tonski got the idea for Balance From Within while playing with a chair, trying to balance it on one leg. “When I would let go of it, there was this really magical moment—‘it looks like it’s going to stay there!’—before it falls,” he told SciFri. “There is so much potential energy psychologically in that moment…I thought, ‘Wow, what if you did that with something that’s as tall as a person?’”

To achieve this balancing act, Tonski embedded two strong motors inside the sofa, which each turn a wheel. As the wheels rotate in one direction, they exert a force on the sofa to turn in the opposite direction—only, that doesn’t happen. “That force trying to rotate the sofa in space pushes against the foot on the floor,” says Tonski. “It’s kind of glued by friction and gravity.” 

Tonski borrowed this elegant solution from satellite technology. The principle that a rotating internal wheel can cause the rotation of an external frame is called a “reaction wheel.” It’s what allows engineers to re-orient satellites in space. “The theory is satellite technology, [but] the motor inside there wasn’t from a satellite,” Tonski admits, gesturing towards the sofa. “It’s actually a motor from an electric bicycle.”

The Collider, by Danqing Shi, Ke Fang, Junjie Yu, Yunzi Qian, and Yin Li

Fans of the Harry Potter series will remember Platform 9 3/4, the illusive entrance to the Hogwarts Express. The “platform” looks like a solid wall, but when Harry and friends run toward it, they’re magically transported to a secret train station. The Collider—an installation by information, art, and design students from Beijing’s Tsinghua University—works on the same principle, but with the help of technology, not magic.

In the piece, participants run down a 15-meter track towards a pair of doors held together by electromagnets. Sensors along the track record their velocity. If, by the end of the 15 meters, the runner has maintained or increased her speed, the doors open. If she’s slowed down, however, she risks an abrupt impact.

Tsinghua University professor Danqing Shi explains that his students were inspired by new, autonomous technologies like the Google self-driving car that ask users to entrust their safety to machines. The Collider, he says, is designed as a test: Just how much do we really trust technology? “People say [the technology] is advanced,” he continues, “but is it advanced enough to 100-percent secure our lives?”

At least one Ars Electronica visitor might answer “no” after his encounter with The Collider. On the first day of the festival, Shi says, a man tried The Collider without supervision—and went away with a bloody face. (Shi explains that the man had walked down the track, then sprinted towards the finish line, depriving the machine of the necessary information to open the doors on time.)

Even after that failure, there was a long line to try out the installation the next day. “There are still brave people who want to try [it],” Shi told SciFri.

More than a decade into a painting career, London-based artist Patrick Tresset came to the heart-breaking realization that he no longer had the passion to paint. So he turned to an interest that was almost as old as his love of painting: computing. “I started to work on systems that could do what I couldn’t do by hand anymore,” he says.

The result is the installation “Five Robots Named Paul.” In a scene reminiscent of a portrait drawing class, five robots observe a sitter, then sketch her face using low-tech ballpoint pens. The finished drawings are spare, evocative, stylized, and almost always reasonably life-like. (Imagine what someone spotted through a thick fog looks like—that’s how the pictures appear.)

Each of the Paul robots consists of a wooden school desk, a robotic arm, and a digital camera or webcam “eye.” They run off of software Tresset wrote to mimic what happens when an artist draws from observation. At the beginning of a new session, each Paul takes a picture of the sitter and begins sketching the broad outlines of a face, “like a draftsman would do,” Tresset explains. In stage two, the machine compares its output with the original picture. “It looks at what it has done, and the level of grey of the tones, and then adds more things [to match],” Tresset says.

One of the big surprises of the installation is that the Pauls have recognizably different drawing styles. (Observers at Ars Electronica could be heard debating which of the Pauls was their favorite.) Tresset says those differences occur primarily because the Pauls’ camera “eyes” range in quality—the machines literally “see” things differently.

Could the Pauls be considered artists in their own right? “Not yet,” according to Tresset. But he says that he hopes to work with machine-learning researchers so that one day, his Pauls can “observe human culture just like an artist does,” and then “evolve meaningful styles” that might resonate with human audiences.