In a two-story building in an industrial district of Cambridge, Massachusetts, Ara Knaian shows off prototypes of what could be the industry’s first completely modular smartphone.
On workbenches sit prototypes of memory modules, battery modules, and processor modules, all designed to slide easily in and out of an aluminum smartphone “endoskeleton.” A prototype infrared imaging lens module for night photography would protrude about a half-inch from the device. Another module would let you read your blood oxygen levels with a swipe of your finger.
Knaian runs a small electrical engineering company called NK Labs, one of the main contractors working on “Project Ara” (which is named after him). By now, though, more than 100 people at a dozen companies are involved in this modular smartphone venture from Motorola’s Advanced Technology and Projects (ATAP) group, a division Google retained when it agreed to sell the handset maker to Lenovo earlier this year (see “Why Google Kept Motorola’s Research Lab”).
Hardware modularity has been tried in the phone market before, but the awkward, bulky results fell well short of displacing sleek all-in-one devices that need frequent replacement. In 2007, Modu, an Israeli startup, developed smartphones that fit into electronic jackets to become cameras, fitness trackers, or music players. The idea failed in part because of the proprietary interface but also because of a clunky design and the limited number of available modules. The company folded soon after launching (though Google bought its intellectual property a few years later).
Google thinks modularity may succeed now thanks to the shrinking cost and size of the underlying electronics and because innovation in conventional mobile hardware is slowing down (see “The New Smartphone Incrementalism”). Also, by fostering open hardware innovation in smartphones and other mobile devices, Google believes it could gain footholds for its software and services in fresh markets and fresh industries.
“We believe that the smartphone hardware ecosystem should be, and can be, a lot more like the Android app ecosystem: with a low barrier to entry, lots and lots of developers, and faster, richer innovation,” says Paul Eremenko, a former office head at the Defense Advanced Research Projects Agency who leads the project (his boss, Regina Dugan, was DARPA’s director and now runs ATAP for Google).
The hardware ecosystem Eremenko envisions would be entirely open. Google would provide the endoskeleton, which has eight rear slots for modules, two front-facing slots for components such as a screen and a button panel, and onboard power and data transmission. Parts could be replaced or upgraded without discarding the rest of the phone, and the finished device could be adapted to serve any number of special functions—professional photography, environmental sensing, medical monitoring—depending on what hardware emerges. Though the project is still in the research and development stage, a working prototype is expected to be ready this month.
It’s an alluring idea, but Google will have to persuade both customers and hardware makers to think different.
At least now smartphone components are much smaller and cheaper than they once were. The electropermanent magnets that connect the modules without snaps or hinges and the simple wireless interfaces “help us make modules with as little added complexity, cost, and weight as possible,” Knaian says.
In fact, the size, power, and weight penalty that comes with making something modular is now under 25 percent, a level that is an acceptable tradeoff for the benefits that flexibility will bring, Eremenko argues. “Modular things tend to be brick-like,” he says. “We think we’re at an inflection point where the penalty is down to something that can comport with things that would be beautiful.”
At least one other smartphone maker seems to agree that modularity’s time has come. In January the Chinese smartphone giant ZTE proposed design concepts of quasi-modular phones, called Eco-Mobius. ZTE’s concept is more limited. It allows users to change only four types of components—screen, battery, camera, and a combination of processor and memory—but not to add new kinds of components.
Google may have an easier time convincing buyers to try a modular device if they aren’t yet accustomed to trading in their smartphone every few years. Customers in poorer parts of the world represent the next huge wave of smartphone adopters, and by next year Google hopes to conduct a pilot test of Ara devices with a Wi-Fi module, basic processor and memory, battery, and screen. These are projected to cost $50 apiece to make (the retail price has yet to be determined). Google expects to conduct the test in a South or Central American country where cellular minutes are expensive but Wi-Fi hotspots are common.
For customers to embrace modular hardware, Google will need to convince hardware companies to build a sufficient variety of Ara modules to make the idea of a hardware ecosystem credible and satisfying. Without a proven market, that may prove difficult, but there’s evidence the concept is gaining traction. Some 3,328 registrants—from companies that make medical diagnostics and imaging sensors to those making displays and batteries—have signed up for the first Ara Developers Conference, scheduled for April 15 and 16 in Mountain View, California, says Eremenko.
Peter Semmelhack, founder and CEO of Bug Labs, a New York City–based developer of software platforms for Internet-connected devices, says outside hardware makers will be important. “You have to drive enough sales to the third parties” who make the hardware, he says. “They aren’t going to make an investment without that. But Google, being Google, might be able to break through that because of their size.”
Even before Google got into the game, some people were agitating for longer-lasting mobile devices. David Hakkens, a 25-year-old industrial designer based in the Netherlands, leads a community of enthusiasts who want the smartphone industry to change its ways and come up with common designs and interchangeable parts (see “Where Cell Phones Go to Die”). Hakkens and his comrades are actively championing Project Ara. “My main goal is I just want to have a modular phone—and I don’t care who makes it,” he says.
As modular hardware becomes more sophisticated, it could perhaps include custom manufactured components. Google has partnered with the manufacturer 3D Systems in Andover, Massachusetts, to develop high-speed-3-D-printed plastic cases. This customization will allow cases to be manufactured in a wide range of colors and designs chosen by consumers. As the technology advances, the plastic casings could include some electronic components such as printed antennas or batteries.
Going bald? Lab-grown hair cells could be on the way
These biotech companies are reprogramming cells to treat baldness, but it’s still early days.
Meet Altos Labs, Silicon Valley’s latest wild bet on living forever
Funders of a deep-pocketed new "rejuvenation" startup are said to include Jeff Bezos and Yuri Milner.
Meta’s new learning algorithm can teach AI to multi-task
The single technique for teaching neural networks multiple skills is a step towards general-purpose AI.
A horrifying new AI app swaps women into porn videos with a click
Deepfake researchers have long feared the day this would arrive.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.