Virtual Reality, Hot Medium In 1964, Marshall McLuhan published “Understanding Media.” In it, introduced the concept of hot media and cool media. A cool media, like television provides high amounts of stimulus to the watcher, but requires very little involvement. A hot media may provide high amounts of stimulus, but it also requires a lot of the consumer. McLuhan suggested radio was a “hot medium” because the listener was required to imagine the visual element. The absence of pictures in radio paradoxically makes the listener’s mental images that much more important. I believe that virtual reality is a hot medium, but it is unfortunately being developed as a cool medium. VR, like AR will require a lot of the user. Unlike in other visual mediums, they must provide the impetus for action, the visual framing and the narrative consistency. VR developers attempt to provide a totalizing visual experience. Naturally, this feels like it should be the whole point. But experience tells us that our experience and engagement with reality is rarely purely visual. Sound matters. Why does this matter: I was reminded of McLuhan and the power of sound reading a recent Ars Technica article by Sam Machkovech about the new Dolby Atmos system. Machkovech describes how this new “spatial sound” shocks the listener with something much closer to three dimensions of sound. In his article, he recounts listening to a Dolby Atmos remix of REM’s Automatic for the People. This isn’t as simple (or tiresome) as just having sound whizz across the theater in Dolby surround sound. The sounds have a subtlety and distinctiveness. Each instrument in a song exists as a separate element in dimensional sonic space. Microsoft, as usual, has a sneaky head start on integrating spatial sound since their Xbox One X supports Dolby Atmos. I wonder which VR company will be the first to relearn McLuhan’s lessons about the role sound can play in a hot medium. Wouldn’t it be ironic if the secret to truly brilliant VR wasn’t visuals at all. In a nutshell: Hearing is arguably more important than sight in giving us an awareness of reality. Read More Hack your face Wired reports that Vietnamese computer security firm Bkav has successfully hacked the facial recognition system on the new iPhone. They broke through the system using a mask constructed out of a 3D scan of a person’s face with some two dimensional pictures of their eyes nose and mouth taped onto it. This is a shockingly simplistic approach that seems to imply that Apple’s facial recognition technology features far fewer protocols than their patent activity seems to imply. For example, Apple patented a technology that detected eye movement in facial recognition. This would be important to keep someone from hacking a phone when someone was asleep, unconscious, or dead. But the fact that the system could be hacked with photographs of someone’s eyes seems to imply that they aren’t making use of that patent. Why does this matter? Security is an illusion. Given sufficient determination and resources any system can be hacked, whether it is an iPhone or the NSA. It is not even clear if total security for a consumer application is desirable. We tend to think of security features as representing a means to distinguish between hackers and legitimate users. But certain features create such an inconvenience or loss of utility to the user that they cannot be implemented regardless of the potential exposure. A full face scan would be time-consuming and require large amounts of processing power. My guess is that the researchers at Bkav calculated the cost of actual facial recognition and then made intelligent guesses about where Apple must have taken shortcuts. Eyes, nose, mouth and head shape were intelligent guesses since eyebrows can vary in shape, as can hairline and people rarely bother to frame their ears and chin into the facial scan. Given that any consumer level facial recognition technology will involve shortcuts, Apple might have been better advised to randomize them – scan for eyes one time, nose and mouth the next, etc. In a nutshell: Security should be evaluated on whether it is “good enough” or “not good enough.” Read More Beyond Quantum Simulation In order to simulate a quantum computer, a standard binary computer needs to pretend to provide for multiple quantum states. This is incredibly computationally intensive. Part of the reason we know quantum computers will be so much more powerful than standard computers is that standard computers, even supercomputers, have a very hard time pretending to be quantum. For a long time, computer scientists believed that standard computers can’t simulate beyond 45 qubits (like computer bits, but each one can maintain quantum states.) In the last few months, scientists have lifted this ceiling to 49 qubits. At that point, they suggest, the cost and requirements of a supercomputer would be economically not feasible. Now IBM has introduced a 50 qubit quantum computer capable of maintaining quantum states for up to 90 seconds. That means that effectively the era of quantum simulation is over. The companies and organizations capable of affording a standard computer that could perform high level simulations are necessarily also capable of affording a superior (and real) quantum computer. Why does this matter? I am typing this on an obsolete computer. Your computer is obsolete as well. So is your phone. So is every computer and device owned by everyone you know. Not yet, of course. It will take years for quantum computing to reach consumer devices. But it cannot be denied that everything we currently do with computers works off a computing paradigm that is no longer the state of the art. With apologies to William Gibson, the future is already here, it’s just mostly distributed at Google and IBM and Rigetti. In a nutshell: The most powerful computer on earth is now a quantum computer. Read More