It looked nothing like an iPhone, or anything Apple might dare to make.
On September 10, 2013, the day Apple unveiled the iPhone 5S, Dutch designer Dave Hakkens uploaded a video to YouTube to promote his college graduation project: Phonebloks, “a phone worth keeping.”
Hakkens imagined a smartphone made of interchangeable blocks, and each block — the screen, the battery, the processor, and so on — could be easily upgraded or repaired, so it wouldn’t end up in a landfill after two years.
“In the beginning, we were asking for like 500 supporters,” said Hakkens, but 48 hours later, the video went viral, racking up millions of views — and delivering a jolt to a small technology incubator called ATAP, Motorola’s Advanced Technology and Projects group.
“Phonebloks was announced, and the world went berserk. If Dave Hakkens hadn’t had this viral video, we wouldn’t have announced Ara probably for at least another year,” said former ATAP design lead Dan Makoski in a VentureBeat interview. “We were outed by Dave, inadvertently.”
The ATAP studio had worked in secret for more than a year on a similar device: a phone made of infinitely customizable parts they called modules. It wasn’t supposed to be a phone you’d queue up in line every year to buy. It was something radically different. Block by block, this was to be a phone you could create.
“Ara Knaian of NK Labs, who the project’s named after, he and his crew had figured out how to actually make it work,” recalls Makoski. The team had already finalized the initial industrial designs for it. “If we [waited] to tell the world that we’ve been working on this, everybody’s going to think that we copied Dave. So we said, ‘Shit, OK, what should we do?'”
“So,” Makoski said, “we just called Dave.”
48 days later, Motorola revealed it: Project Ara.
“Thinking back, I never called it a modular phone,” said Hakkens in an interview with VentureBeat.
“I had an old camera that I broke and I couldn’t really fix it. So I took it apart and I noticed all the components were still pretty good, except for one thing.”
“I thought: Isn’t that weird that we throw everything away just because one part is broken?” said Hakkens.
“At first, I wanted to make a phone that lasts 100 years. But then I realized, I kind of like technology — that it evolves, that it gets better. The only downside is that after it gets better, we throw everything away. I started looking into it, and it generates a lot of e-waste … I mean now we have some devices, but in the future it’s thermostats, fridges, microwaves — everything will be connected. So what if a chip breaks in your fridge? Do you just throw the entire thing away?”
The Phonebloks story spread like wildfire. Gadget blogs covered it en masse, hordes of supporters signed up to support, tweet, and share the idea with a viral marketing tool called Thunderclap, and developers fired back, saying it couldn’t be done — that it was impossible to build. Perhaps they had a point.
“He was just getting destroyed by engineers on Reddit saying ‘There’s no way this could work, right? This is just a pipe dream.’ The way he had designed it was totally unrealistic because he’s an industrial designer and doesn’t, you know, understand the complexity,” Makoski said.
“I didn’t really put a lot of effort into how the thing would look, because I didn’t really care about that. So I sort of stripped it down to the most basic version, which would be blocks,” said Hakkens. “I figured, maybe it’s a good thing to just put it out online to gather people that support it. Then, if you had a lot of support, companies would make it.”
The idea struck a nerve. Only Hakkens needed someone to build it.
Then ATAP called.
When Google’s $12.5 billion Motorola Mobility acquisition closed in May 2012, the maker movement looked like the dawn of a manufacturing revolution. Food trucks swarmed Motorola’s new Sunnyvale, California campus because the cafeteria was still under construction. Incoming CEO Dennis Woodside sought someone to run innovation at the new, Googly Motorola, and the job went to former DARPA director Regina Dugan. “Regina was the soul of ATAP,” recalls Makoski.
Makoski joined with a ticking clock: He had just two years at ATAP because of a rule inspired by Dugan’s time at DARPA, where newcomers were handed a name tag with an expiration date.
“When Regina hired me, I had no idea what I’d work on. You know, I had some thoughts — I had some interests — and I actually sketched out on my first week six different possible ways to go. One of those little sharpie marker sketches, it was kind of a question.”
“When I look at the highest form of admiration in the smartphone space it was for the iPhone. I love Apple’s design, but I had seen the maker movement, and Maker Faire, and Tech Shop, and these beginning glimmerings of a very different model than what Apple was celebrating. Apple was celebrating the craftsmanship of the perfect object. To me, it was like the ultimate expression of the Henry Ford vision of making one object a million times perfectly.”
“So my question was, What if we did the exact opposite?” said Makoski.
“We started thinking about this idea of extreme personalization. Is there evidence that this is working? We looked at everything from Etsy — we actually did some interesting research on the tattoo space, because that’s evidence of extreme personalization.”
That’s how Project Ara got its first name; before it was Ara, a name inspired by the project’s eventual electrical, mechanical, and software engineering lead, Ara Knaian, Makoski had called it Esprimo — Esperanto for “expression.”
In ATAP’s earliest months, Dugan and Makoski often walked to lunch together, traversing the parking lot to reach the best food truck: Sam’s Chowder Mobile. And during one of these lunch breaks, in line at Sam’s, Dugan had an epiphany.
“‘A food truck,'” Makoski recalls Dugan exclaiming. “‘We should make a factory food truck. Like you customize your meal right here, we should put 3D printers and hackable electronics into a truck, and that’s how we’re going to make this idea.'”
“And I thought she was insane. But you know, I thought about it more, and that was the beginning. It was all about how to create electronics that are malleable and adaptive and personal, and where every one is different,” said Makoski.
The slog began.
“We were working on both aesthetic customization and functional customization. I was leading a lot of those design and psychology experiments around, ‘how do we overcome the paradox of choice?'” recalls Makoski.
Ali Javidan, a Tesla engineer Dugan poached to run ATAP’s prototyping shop, was charged with developing it. “His part was much harder than my part, quite honestly. But he got to the point where we took these Razr phones and we took the main application processor and took the leafs off of that, connected it to an Arduino board.” They nicknamed this early prototype Moduino.
“We, from that August for the next six months, really struggled, quite honestly, because it’s just so freakin’ hard.”
Javidan — charged with developing dozens of projects at ATAP, from touch-sensitive fabric to mobile 3D mapping — was stretched incredibly thin. And Makoski, a designer at heart, did not consider himself technically capable of getting Ara off the ground. Something had to give.
“At that point, Ali and I, quite honestly, we didn’t feel like we were equipped to shepherd the project.”
“Regina liked this idea so much. And she just believed in people so much, she said, ‘Dan, hire the people, you can be the technical project lead.’ And I honestly — I sometimes regret making this decision — but I just said, ‘Regina, I’m not capable of shepherding this project to life.'”
So Dugan found someone else: Paul Eremenko, who worked on rapid manufacturing at DARPA, joined ATAP in April, 10 months into Makoski’s time at the studio. The project, not even a year old, was under new leadership.
And it was an omen of things to come. The sheer audacity that brought Ara to life — a phone you could create, block by block — made it impossible for Makoski, or anyone else, to succeed. But Eremenko would get close.
Javidan assumed responsibility for managing the shop. Makoski, meanwhile, partnered with Eremenko on the design, cultural, community, and psychological facets of Ara.
“I was in charge of the soul. Paul was in charge of body. That’s how we divvied our time.”
“Paul was always the one with red eyes, hadn’t gotten any sleep, and was always super focused with what he did,” recalls Hakkens. “It was Paul, for me, who was really pushing the project forward.”
“Paul was really ambitious. And he was actually hoping that we’d have something we could get out in people’s hands in early 2014, I guess, initially. That was back in the early days,” said Makoski.
The team hired another key contributor: Jason Chua. He was a Stanford graduate who had cofounded SparkTruck — a project to connect students with maker-movement supplies, like 3D printers and electronics. Chua and Makoski set out to build Dugan’s factory on wheels to test the team’s early experiments at scale.
That summer, as Makoski and Chua drove 12,683 miles across the U.S. in a Velcro-clad van testing Dugan’s food truck epiphany with real people, Eremenko stayed home, creating the technical infrastructure for a malleable smartphone.
Even as Makoski and Chua drew crowds testing the thesis behind Ara, back home in Sunnyvale, the audience was far from enthusiastic. According to Makoski, Motorola’s then-CEO Dennis Woodside was rather unimpressed by Ara’s initial prototypes.
Before Phonebloks went viral, “He was like, ‘Yeah, cool guys, that’s interesting, but you know this maker thing is kind of niche. No one’s really going to care,'” said Makoski.
Motorola’s engineering VP, Iqbal Arshad, had advised on the project early on, and his team helped Javidan and Makoski put together their first prototypes. But Ara would be shrugged off as a side project by Motorola.
“[Arshad’s] a really good guy, but when Google had taken over Motorola, he was trying to focus his engineering resources on just a couple of things that were really going to move the needle. I mean, his whole focus was on the Moto X at that point,” recalls Makoski.
“He was like, ‘I don’t have time for your crazy modular thing. That’s totally insane, architecturally.'”
After a summer of working cheek by jowl, Makoski and Chua’s cross-country tour came to an end and Eremenko finished laying out the core technical details of Ara — a “packet-switched network on a device,” printed on 3D printers and held together by experimental EPM magnets under development by Ara Knaian.
Then Hakkens went public with Phonebloks. His bold idea took off like a rocket.
Flooded with interest, Hakkens spoke with several companies, but Google, in particular, got his attention. “Google was really persistent,” he said, “and they mentioned that ‘we’ve been working on a project like this — would you like to come and see and maybe we can do something together?'” ATAP flew Hakkens from Rotterdam to Motorola’s Sunnyvale campus.
“We just walked him through what we were working on, and he was just like, ‘Holy shit, you guys have actually been working on it. Like, this could be real,'” said Makoski. It was September 2013, and Makoski had less than ten months remaining in his two-year tour at ATAP.
“I know Google doesn’t necessarily care about the environment, but still I liked what they were doing. So instead of working for Google, it seemed to make sense to stay independent from them,” said Hakkens. “In the end, [I] wanted to support other modular phones. Not just Google.”
“I’ll just say we left those two days of meetings with Dave suggesting a [consulting] price that was just ridiculous to us, and us offering Dave to join Motorola,” said Makoski.
Hakkens turned them down.
That seemed like it would be the last ATAP would hear from him.
“He just wanted to stay independent. And we said ‘good luck.’ It was friendly, but not finding a way to collaborate,” said Makoski.
Then, “literally 48 hours before the Thunderclap [marketing push] was to go out, Dave called us and said, ‘Hey guys, I have a proposal: How about you guys just pay me for my time to tell your story, and I use the Phonebloks community as a place where you guys can get feedback on what you’re trying to do’,” said Makoski. “And that was acceptable to us. So we said great.”
On October 28, 2013, Motorola unveiled Ara.
In Motorola’s announcement, Eremenko posed a gutsy question: “How do we bring the benefits of an open hardware ecosystem to 6 billion people?” Six billion — a very, very large number.
He painted the project in brilliant, broad strokes on the company’s blog.
“Our goal is to drive a more thoughtful, expressive, and open relationship between users, developers, and their phones. To give you the power to decide what your phone does, how it looks, where and what it’s made of, how much it costs, and how long you’ll keep it.”
“We all went home after we announced,” recalls Makoski, “and then we all came back two hours later and the world just went psycho. I don’t think we expected the amount of interest. Even Woodside — who was lukewarm … when we had showed him stuff before we announced publicly — the week after we announced, Dennis was like: ‘We’ve been proudly working on this…'”
Ara had appeared out of nowhere — just 48 days after Phonebloks’ viral hit — and the optics couldn’t be stranger. From outside ATAP, the arrangement was incredibly opaque. Motorola appeared to have copied Hakkens, even as Eremenko said otherwise. But Hakkens had never contributed to Ara, the phone — neither the hardware nor the software.
“He basically became a journalist,” said Makoski. “He would come in every couple of months and update his readership and the broader world about what we were doing. He’s a storyteller.” Hakkens hoped other companies would partner with Phonebloks, but the surrounding confusion muddied his vision.
“One of the things we did wrong, or could have done better, was that everybody thought Ara was Phonebloks,” said Hakkens. “I think everybody thought we were bought out by Google … so we kind of lost our independent image. And companies wouldn’t get in touch that much anymore.”
But Hakkens did something crucial for Ara; he dragged it out of the shadows and forced ATAP to develop the project in the open. Within the tech industry, all eyes fell on Ara and ATAP.
When Google sold Motorola to Lenovo that January, it held onto ATAP and Ara — a shining light of innovation at an advertising behemoth.
As one might expect, Google’s absorption of ATAP had side effects.
“In some ways, it let us preserve our charter,” said Makoski. “And actually, there were some internally that said it was this project, Project Ara, that was the one that ended up keeping us [at Google], because it generated so much enthusiasm in the technical world.”
Others say Google held Ara back. “The pace and scale of contracting that we did all of the sudden really slowed down [after the acquisition],” said one source, who requested anonymity. “The model of ATAP is based on really rapid contracts, like DARPA. There were 150 people working on Ara before we transitioned into Google. And only 3 of them, or 4 of them, were Google employees. All the rest were external contractors,” said the source.
“To us, it just felt like we had just kinda hit molasses. And I think, to Google, it felt like they were being uncomfortably sped up,” the source said.
When Ara was announced in October 2013, Eremenko told the world that ATAP would release Project Ara’s Module Developer’s Kit (MDK) early the next year. And in February, in a lengthy Time profile, he promised a basic commercial release for the first quarter of 2015.
Eremenko imagined doing “for hardware what the Android platform has done for software,” but more remarkable was Project Ara’s starting price: $50.
Eremenko sought to build a phone that was customizable beyond any electronic device ever created. You could pick the features, decide when and what to upgrade, and (within reason) pay whatever you wanted.
For $50, the starter phone would ship with only the most basic features — not even a 3G modem. But that same phone could be enhanced and evolve to incorporate technological improvements over time. Premium modules, like a bigger battery or a high-quality camera, could transform that smartphone from an obsolescence timebomb into a worthwhile investment.
“We had a philosophy that this phone was not for the iPhone-carrying, latest Samsung Galaxy-carrying smartphone owner in the U.S.,” said Makoski. “We wanted to bring access to the internet, to the smartphone space, to those who previously didn’t have it. And part of the ways to do that is to create a platform where an India telecom could put customized radios into Ara for a $50 price point or a $100 price point, or it could scale all the way up to something for Latin America or the U.S.”
ATAP released Project Ara’s developer kit one month behind schedule, in April 2014 — the same month Google teased its legion of fans with a prototype of Ara’s main Endo module.
Two months later, ATAP started accepting applications for Ara developer boards. The team was in full swing, determined to make Ara real. And Ara’s biggest supporters were certain it soon would be.
And then Ara lost one of its founders: Makoski’s two years were up. Meanwhile, Eremenko had 13 months to go.
On October 29, 2014, Google released a video of its first working Ara prototype. Soon after, in January, the company shared a second clip, this one showing Ara with a 3G modem and a swath of modules across every category, from miniature piano keyboards to heart rate monitors, all clad in vivid colors and personalized with photos.
Another goodbye brought Ara’s myriad designs within reach. Eremenko cut ties with one of Ara’s earliest supporters, 3DSystems, scrapping the project’s dependence on rapid 3D printing for a dye sublimation process. 3DSystems’ printers were too slow, and the new system could adorn modules with selfies and pets.
Eremenko fought to release Ara into the world before his final days at ATAP, but he lost faith as the project’s timeline shifted stubbornly into the future.
“I remember from the beginning that Phonebloks was like a 10-year vision or something like that. And then Google said, ‘Alright, we are going to do it in two years,’ which seemed super ambitious,” said Hakkens.
“Once we had figured out that we could pull it off technically, we just needed to line up all of the silicon providers, etc., to get it all working. And then there were all different types of roadblocks,” said Makoski. “Everything from the EPM magnets not going to the strength we wanted to some of our partners requiring all kinds of business relationships that maybe we weren’t ready to give. I can’t really comment on that stuff, partly because I don’t know it that well.”
ATAP planned to launch a market pilot outside the U.S. first — in a South American country that shared a timezone with Google’s headquarters, like Ecuador or Argentina. The pilot, as promised by Eremenko in the pages of Time magazine, was slated for release for sometime in 2015 — ATAP never said when exactly, but the studio eventually settled on a location, Puerto Rico.
“We just said, “Let’s use Puerto Rico’ because it has a similar regulatory regime as the U.S., and if we can do something in Puerto Rico than we kind of check the box that, if we wanted to, we can scale in the U.S. market (and also in Latin America),” explained Makoski.
But the phone wasn’t stable or sturdy enough; the project needed yet more nurturing, and Ara once again missed a shipping window.
In June 2015, Eremenko left ATAP. His two years were up, and Ara faced more uncertainty than ever.
The next month, Google canceled the Puerto Rico pilot and promised an Ara release in 2016. “It’s difficult for a project like this to have a good estimate of where you’re going and when it’s ready. I remember telling them a couple of times that they really shot themselves in the foot by mentioning something like that,” said Hakkens.
And Project Ara, without a leader, changed hands again. This time, Dugan took charge, with Motorola veteran Rafa Camargo, and the team went quiet to rework Ara, leaving many developers, and Hakkens, in the dark.
At the start of 2015, Eremenko’s dream of a $50 phone had evaporated. “The evolution of what Ara was supposed to be had changed so much because of the big question mark around what consumers actually wanted,” a source who worked on Ara told VentureBeat. “And a $50 [smartphone] is just not technically possible. That’s the truth. Anybody who makes smartphones can tell you that.”
So Ara pivoted.
Under Dugan and Camargo, Ara morphed into a high-end phone with all the essentials built in. Ara lost its ability to serve different price points, but the project gained focus. It was designed to push the limits of what a smartphone could be with modules that add entirely new features, like super-powered Lego blocks.
Imagine the modules developers might dream up. There were the obvious ideas, like specialized cameras and high-end speakers. But modules could get stranger, wilder, too. One module idea, in particular, frequently derailed meetings inside ATAP’s walls, as studio leaders strained to picture a module gold rush akin to Apple’s App Store.
“One of the modules that we were working on was basically like a tiny aquarium for your phone,” said the source. “It was a little tiny biome that would go inside of a module and it would have a microscope on the bottom part, and it would have live tardigrades and algae — some people call them water bears. They are the tiniest living organism. We had this idea to build a tardigrade module and we’d build a microscope with it. So you’d have this app on your phone and you could essentially look at the tardigrades up close and watch them floating around.” Brooklyn-based art, design, and technology agency Midnight Commercial conceived the idea, and was commissioned by Google to build it, demonstrating the depth of what developers could create.
Enthusiasm inside ATAP climbed as 2015 advanced. “It was very much balls to the wall, let’s go. Let’s make this happen. It was very fast. It was super exciting. Really, really exciting,” said the source. “The way [Dugan] presented it and the way she talked about it, I felt like it was the biggest opportunity that I had ever been presented.”
But while Ara’s team grew, Larry Page and Sergey Brin sought to slim Google down. The pair created a parent company named Alphabet, split most of Google’s wild bets, like internet balloons and smart thermostats, into separate companies, and put Sundar Pichai in charge of Google.
ATAP remained at Google, closely guarded by Dugan, and Pichai hired Motorola president Rick Osterloh to make sense of Google’s fragmented hardware projects. Increasingly known for duds like Glass and the bowling-ballish Nexus Q, Google had to get serious about hardware.
Then in April 2016, one month before Ara’s biggest-ever reveal, planned for Google I/O, Dugan suddenly left.
She was hired to lead Building 8, an undefined company created by Facebook with “hundreds of people and hundreds of millions of dollars” at her disposal, Mark Zuckerberg wrote on his personal Facebook page announcing the appointment.
“When she left, there was essentially a leadership vacuum,” a source said. “Because there wasn’t a champion for Ara, someone who could really go to bat for it, I think it just fell down the priority list. And if you look at the LG [G5] phone, it just didn’t do very well. If you look at the other attempts at modularity and try to get a feel for the market, there’s just not a lot of good signals out there.”
In August, the Ara team was tracking to ship Ara to developers in late 2016 and to the public in 2017. Then Osterloh’s axe fell. Ara was “suspended.”
News of the cull rang out like a sudden alarm in the ears of Ara’s staff, who had continued to toil in Dugan’s absence. And Ara’s remaining loyal fans, worn out by years of delays, mourned the news on Twitter. Perhaps more than anyone else, the news pierced the heart of Dan Makoski, who walked the streets of Palo Alto that night in disbelief that Ara was really, truly gone.
Ara has laid dormant for five months — the prototypes locked away from developers as a fraction of Ara’s remaining team stayed to wind down the project. Some later rejoined Dugan at Building 8. Others went on to other projects at Google.
Google may someday license or sell Ara. The company has hinted it might, yet throughout Ara’s evolution there was always a missing piece. “Consumers don’t care about modularity,” said the source who worked on Ara. “And even today, I’m still not sure that it’s something that consumers want.”
Makoski thinks otherwise.
“I feel like Project Ara, if it does get shelved, it’s a shame that they didn’t have the courage to go all the way and fail a couple of times at a commercial release to get it over the hump. But I think the industry will. And hopefully I’ll be alive to see that. I’m only 42 … I still believe.”