Month: January 2020

Although the passage of time serves to make the past seem sweeter in recollection than it might have been in the moment, it's impossible to deny that there was something special about the gaming landscape of the 1990s. Every year in that decade brought a torrent of titles that were destined to become classics—including the often-imitated-but-ultimately-inimitible Myst.

Myst came to market in 1993, which was a banner year in PC gaming—1993 also brought us X-Wing, Doom, Syndicate, and Day of the Tentacle, among others. It's fascinating that Myst happened the same year that Doom launched, too—both games attempted to simulate reality, but with vastly different approaches. Doom was a hard and fast shotgun blast to the face, visceral and intense, aiming to capture the feeling of hunting (and being hunted by) demons in close sci-fi corridors; Myst was a love letter to mystery and exploration at its purest.

A few months back, Ars caught up with Myst developer Rand Miller (who co-created the game with his brother Robyn Miller) at the Cyan offices in Washington state to ask about the process of bringing the haunting island world to life. Myst's visuals lived at the cutting edge of what interactive CD-ROM technology could deliver at the beginning of the multimedia age, and, perhaps unsurprisingly, fitting the breadth of the Millers' vision onto CD-ROM didn't happen without some challenges.

Enlarge / Stepping outside the Ars Orbiting HQ for a brief moment to take a space selfie. (credit: Aurich Lawson / Getty)

We're running a new series on Ars over the next few weeks about “the future of work,” which will involve (among other things) some predictions about how folks in and out of offices will do their future officing. To start, let's take a tour of the fabled Ars Orbiting HQ—because we've learned a lot about how work works in the future, and we'd love to share some details about how we do what we do.

Ars bucks the trend of most digital newsrooms in that we truly are an all-digital newsroom. While we have mail stops at the Condé Nast mothership in New York, there is no physical Ars Technica editorial office. Instead, Ars Technica's 30-ish editorial staff work from their homes in locations scattered across the country. We’ve got folks in all US time zones and even a few contributors in far-flung locations across the Atlantic.

Marshaling this many remote staffers into a news-and-feature-writing machine can have its challenges, but Ars has operated this way for more than twenty years. We’ve gotten pretty good at it, all things considered. The main way to make it work is to hire self-sufficient, knowledge-hungry people, but another major part of our remote work philosophy is flexibility. Not everyone works the same way, and remote work should never be treated like a one-size-fits-all, time-clocked job. Also, tools matter—you can’t expect people to do collaborative jobs like writing and editing without giving them the right hardware and software.

Enlarge / The updated Super Cruise has a new UI. (credit: Cadillac)

General Motors' Super Cruise is widely recognized to be the best of the so-called "Level 2+" driver assists. It combines adaptive cruise control and lane-keeping with a geofence—so it only operates on divided-lane highways—plus an infrared driver monitoring system that only allows for hands-free operation when it knows the person behind the wheel has their eyes on the road ahead. On Tuesday morning, Cadillac announced that it's rolling out an enhanced version this year that includes the ability to change lanes on demand.

"This is our most extensive update we’ve made to Super Cruise since its debut," said Mario Maiorana, Super Cruise chief engineer. "We have made a number of improvements to make Super Cruise more intuitive, better performing and more accessible for our customers. In addition to the automated lane change functionality, we’ve made improvements to the user interface and hands-free driving dynamics."

Currently, if you're Super Cruising along one of the 200,000 miles (321,868km) of lidar-mapped highways in a Cadillac CT6 and you want to change lanes, it's all down to you to execute the maneuver. You check there's a gap, indicate (please remember to use your turn signal), and as you begin applying torque to the wheel the system temporarily disengages, giving you full control. You know you're in charge because the strip of LEDs in the steering wheel go from green to blue. Once you're traveling straight and true again, the system can re-engage, the LEDs turn green, and you can go back to vogueing, doing 'big fish little fish cardboard box," or even jazz hands, while all around you everyone else has to keep their mitts on the rim.

Enlarge / This image, taken from the interior of a Tesla Model X, shows a projected image of a car in front of the Model X. The inset in the bottom right, created by Nassi from the Model X's logs, shows the Model X detecting the projection as a real car. (credit: Ben Nassi)

Six months ago, Ben Nassi, a PhD student at Ben-Gurion University advised by Professor Yuval Elovici, carried off a set of successful spoofing attacks against a Mobileye 630 Pro Driver Assist System using inexpensive drones and battery-powered projectors. Since then, he has expanded the technique to experiment—also successfully—with confusing a Tesla Model X and will be presenting his findings at the Cybertech Israel conference in Tel Aviv.

The spoofing attacks largely rely on the difference between human and AI image recognition. For the most part, the images Nassi and his team projected to troll the Tesla would not fool a typical human driver—in fact, some of the spoofing attacks were nearly steganographic, relying on the differences in perception not only to make spoofing attempts successful but also to hide them from human observers.

• 

  This is a frame from an ad you might see on a digital billboard, with a fake speed-limit sign inserted. It's only present for an eighth of a second, and most drivers would miss it—but AI image recognition recognizes it. [credit: Ben Nassi ]

Nassi created a video outlining what he sees as the danger of these spoofing attacks, which he called "Phantom of the ADAS," and a small website offering the video, an abstract outlining his work, and the full reference paper itself. We don't necessarily agree with the spin Nassi puts on his work—for the most part, it looks to us like the Tesla responds pretty reasonably and well to these deliberate attempts to confuse its sensors. We do think this kind of work is important, however, as it demonstrates the need for defensive design of semi-autonomous driving systems.