Almost three years after announcing a plan to delivery Internet access via high-altitude balloons, Alphabet X’s Project Loon is set to begin testing service with carriers in Indonesia later this year.
Astro Teller, the head of Google’s top-secret innovation lab, who we profiled in 2014, announced the news during a talk at the annual TED Conference taking place in Vancouver, as well as a post on Medium.
During the talk he revealed that Project Loon experienced a lot of design and engineering challenges, and for a while there was doubt if the project would ever actually—excuse the pun—get off the ground.
"We busted a lot of balloons," Teller told the audience, and showed them multiple iterations of the designs of failed Loon balloons, including ones that looked like jelly fish as well as ones that looked like giant pillows. Teller said besides being able to keep the balloons in the sky for an extended period of time—months on end—the major challenge was finding a balloon that could be made cheaply and navigate precisely.
But the members of Project Loon were able to do just that, with a balloon they engineered last year that made a staggering 19 round-trips across the world over a period of 187 days. The team also made improvements in the speed of Internet connectivity the balloons deliver over their aerial wireless network. Teller said that originally the connection was very slow, but now it’s up to 15 megabits per second, which he noted would be enough speed to deliver live streaming video—like the one of his TED Talk.
But all isn’t smooth flying for the Project Loon balloons yet. The Indonesia launch is a test to see how the service works delivering actual Internet services to customers. New challenges could arise once the widespread testing begins. Re/code reports that testing of Project Loon will happen in other countries besides Indonesia. It also notes that Alphabet has reached a deal with Sri Lanka to get access to a needed radio frequency for the aerial wireless network in exchange for a stake in Project Loon.
Teller also revealed some Alphabet X projects, or "moonshots," that have been abandoned. One would have enabled landlocked countries to ship goods at greatly reduced costs using, as Jetsons as this sounds, "a lighter-than-air, variable buoyancy cargo ship"—essentially a rocket ship that could take off and land without needing a runway. That project was abandoned because of cost, says Teller, who revealed that building just the first rocket would have cost $200 million—too much of a risk for an unproven system, even for wealthy Google.
Teller also revealed a second project that was abandoned: vertical farming. The idea was to create a farming method that used one-one-hundredth of the land (by "stacking" piles of earth and crops) and one-tenth of the water traditional agriculture requires. Though Alphabet X was able to grow lettuce, it never found ways to grow critical crops like rice and grain.
Teller said that Alphabet X starts with testing out the hardest part of a moonshot project; if it can’t survive the most demanding tests, the project is killed. "If there is an Achilles' heel in one of our projects, we want to know it right now," Teller said.