POSSESSION SOUND, Wash. — Steering a five-person submersible is like playing a video game, except for the fact that you’re piloting a nine-ton piece of hardware at watery depths that are inaccessible to . I got my chance to play this week during a survey dive in a pocket of Puget Sound known as Possession Sound, courtesy of , a manufacturer and operator of submersibles that’s headquartered in Everett, Wash. During our three-hour tour, GeekWire photographer Kevin Lisota and I were taken around the sound at depths ranging as low as 350 feet, in OceanGate’s Cyclops submersible. We even played a supporting role in finding a colony of anemones in an unexpected underwater setting. The trip was part of a summertime expedition to get a better sense of the ecosystem on the bottom of Puget Sound, in collaboration with researchers from the Washington State Department of Fish and Wildlife and the National Oceanic and Atmospheric Administration. If things had turned out differently, OceanGate would just now be wrapping up a series of submersible survey dives to the wreck of the Titanic in the North Atlantic. But , the company had to delay those trips until next year. That’s why OceanGate pivoted to the Puget Sound survey, and why Kevin and I found ourselves scrunched alongside marine biologist Tyler Coleman, pilot-in-training Mikayla Monroe and OceanGate CEO Stockton Rush on Tuesday. We began the morning with a safety briefing at the dock at Everett’s marina, led by Dan Scoville, OceanGate’s director of systems integration and marine operations. One of his bits of advice had to do with keeping calm if you hear thumps and bumps on Cyclops’ hull. “If you can hear it, you’re OK,” he said. If there’s a catastrophic collision and breach, you wouldn’t be around long enough to hear it. Then the Cyclops was towed out on its launch-and-recovery platform to Possession Sound by one of the boats in OceanGate’s fleet, the Kraken. A little more than an hour later, the Cyclops was in position, and we headed out to meet it on a faster boat, the Vito. Once we were dropped off on the floating platform, we handed up our backpacks, took our shoes off and climbed into Cyclops’ 5-foot-wide cabin. Mikayla sat on a mat toward the back, flanked by video screens that showed camera views and sonar readings. Stockton sat next to her, ready to give guidance. Tyler sat in the middle. Kevin and I had front-row seats, looking through Cyclops’ hemispherical acrylic viewing window. We let our stocking-clad feet rest on the window’s bottom, even though we were warned that we might feel the chill of the water on the other side. Once all the final checks were made, the crew members on the Vito, the Kraken, the platform and in the submersible took a five-minute timeout, known as a “stopski,” just to make completely sure all systems were go. (The idea — suggested by Scott Parazynski, a former NASA astronaut — was inspired by the built-in holds that are included in space launch countdowns.) Then it was time to dive. First the launch and recovery platform blew the compressed air out of its flotation tanks, in a process that had us dipping down backward into the water at a 20-degree angle. Green-tinted water sloshed wildly over our field of view. “Is this a freakout moment for some people?” I asked Stockton. “I haven’t run across that yet,” he replied. “You could be our first.” GeekWire’s Alan Boyle takes notes as he looks out the window of OceanGate’s Cyclops submersible. (GeekWire Photo / Kevin Lisota) Within minutes, we were sinking below the photic zone, where sunlight could penetrate to fuel the green plankton that tinted the waters. The view outside was total darkness, until Mikayla turned on the floodlights on each side of our window. Even then, plankton and the other particles floating in nutrient-rich Puget Sound cut the visibility to just a few feet around us. Mikayla relied on sonar readings to determine our depth, and on GPS readings to determine our heading. Our first destination was right beneath us: We headed for a wire cage containing a pile of salmon guts, which was dropped down on a line from a buoy to attract whatever creatures were foraging at the bottom. When we pulled up to the cage, we saw a smattering of rockfish (of the quillback and canary varieties), with 4-inch-long prawns and an occasional crab skittering through the scene, looking for a meal. The prime targets for OceanGate’s survey are shark species, and especially the rare, crowd-pleasing sixgill shark. We hoped to follow in the footsteps of Seattle rap musician Macklemore, who when he went looking for Puget Sound sixgills in a different Oceangate sub. We saw no sixgills, but we did catch sight of a slim, spiny dogflsh shark as it threaded its way around the bait box. “So we had our first official shark?” I asked Tyler. “Yup,” he said. Marine biologist Tyler Coleman identified this fish as a dogfish shark. (GeekWire Photo / Alan Boyle) Then we headed off to a wide stretch of muddy bottom, punctuated by holes that provided shelter for the prawns and other bottom-feeding critters. Stockton sidled over to me, holding the modified Sony PlayStation game controller that’s used to steer the sub. “Want to drive?” he said. It took me a while to get the hang of the controls: The front buttons serve as “dead man’s switches,” which have to be pressed in order to activate the controller’s dual joysticks. The left joystick controls the up and down thrusters, and the right joystick controls horizontal thrusters for forward and back, left and right. Simple, right? Nevertheless, I occasionally rose high enough to lose sight of the bottom, and sank low enough to plop the sub into the mud and send clouds of sediment rising up in front of our window. To get ourselves out of those obscuring clouds, I had to drive the sub out of the haze into clearer waters. At least there were no rocks to run into, which is why Stockton and Mikayla brought us to a field of mud before they handed me the controller. After a few minutes of meandering, Mikayla reported that there was something showing up on the sonar, about 15 meters dead ahead. Stockton took back the controller, and guided by Mikayla’s callouts, he brought us right up to what looked like a garden of cauliflowers, plunked in the middle of an underwater desert. It turned out that a tree stump had sunk 350 feet to the bottom, heaven knows how many years ago, and a colony of anemones had taken root there. Stockton was impressed, and he told Mikayla to take note of the coordinates. “The visibility is probably 10 feet today, but we can get 5 feet away, so that’s OK,” Stockton told me. “Imagine trying to find this if you were diving. … Nobody’s ever seen this log before, I’ll bet you even money.” Toward the end of the tour, we returned to the area where bait had been dropped to the bottom. Mikayla turned the lights off, waited for a school of rockfish to swim in front of our window, and then turned the lights back on so we could snap photos. When it was time to ascend, we rose through the dark murk and back into the sunlit green haze near the surface. Kevin and I were deputized to watch for the whitish outline of OceanGate’s launch-and-recovery platform, anchored a few meters below. It took a couple of tries to get properly “locked in” on the platform, due to a balky thruster. I was feeling grateful that Mikayla and Stockton were at the controls (and hoping I hadn’t damaged the thruster during my training session). At last we were locked in and lifted up. The sun seemed unusually bright as we climbed back up through the hatch and were motored back to shore. On the way back, Stockton talked about OceanGate’s plans to bring the submersible experience to a wider audience. “Diving’s no fun after you’ve been in a sub,” he said. Taking people down to the Titanic is still OceanGate’s prime objective: The submersible that’s designed for that role, which was initially called Cyclops II but is now known as Titan, proved it could safely get to Titanic-worthy depths this year . The postponement of the means there’s not a lot for Titan to do until next summer. It’s currently being prepped for an extra round of stress tests, plus equipment upgrades that should smooth the way for the 2020 season. OceanGate’s Titanic customers are paying to participate in the adventure as mission specialists, and most of them are keeping their reservations despite the delay. Stockton said that OceanGate’s subs — including Cyclops and Titan as well as the two-person Antipodes — are currently certified for research missions such as the Titanic expedition, but not for more casual tourist jaunts. Another perspective: Now OceanGate is seeking waivers from the Coast Guard that would allow the company to offer submersible tours for something like $1,000 or $2,000 per person. That’s more than operators in Hawaii charge for submarine tours, but those tours go only 100 feet beneath the surface and last only 45 minutes or so. OceanGate’s tourists would get an experience even more thrilling than ours — assuming that the regulatory go-ahead is given. “It’ll probably be six to 12 months before we get approval,” Stockton told me. Stockton and his team of 27 employees are also looking into whether their subs can be used for infrastructure inspection and environmental surveys. And they’re planning to build a bigger, better submersible called Cyclops III, which could handle depths of 6,000 meters (20,000 feet). To help fund those projects, OceanGate is in the midst of a that was reported to the Securities and Exchange Commission in April. So what’s tougher? Navigating the depths of Puget Sound, or negotiating the shoals of the startup world? , Stockton Rush is clearly adept at doing both. But personally, I’d rather be steering the sub. Correction for 4:36 p.m. PT Sept. 1: In a previous version of this report, we incorrectly identified a ratfish as a dogfish shark. We’ve amended the ID for the ratfish, and added a video screengrab of the dogfish. Woof! Also, we’ve corrected the anticipated cost of a submersible tour to be $1,000 to $2,000 per person, instead of per day.
A Skyfront Perimeter drone takes off from the Alyeska trans-Alaska pipeline right of way near Fox for a milestone flight beyond the operator’s visual line of sight. The drone flew 3.87 miles along the pipeline corridor. (University of Alaska Photo / Sean Tevebaugh) A public-private consortium led by the University of Alaska has conducted the first-ever federally authorized test flight of a drone beyond the operator’s line of sight without on-the-ground observers keeping watch – with Echodyne, the radar venture that’s backed by Microsoft co-founder Bill Gates and headquartered in Kirkland, Wash., playing a supporting role. Autonomous flight beyond visual line of sight will be key to the kinds of drone delivery operations envisioned by Amazon, Walmart and other retailers. During Wednesday’s flight, a multirotor drone as part of the University of Alaska’s Unmanned Aircraft Systems Integration Pilot Program, from the Federal Aviation Administration last year. The big thing about this flight is that the drone made use of , paired up with Echodyne’s ground-based MESA airspace management radar system, without having a human on the route. Current FAA regulations limit drone flights to the operator’s visual line of sight. Pilot projects have been experimenting with technologies that can ensure safe operations beyond the visual line of sight, known as BVLOS. But until now, the FAA’s waivers still required a ground-based observer to look out for non-cooperative aircraft coming into the test area. This week’s flight of a drone totally on its own was authorized after it flew the same route with visual observers. “The test mission designed by the team at the University of Alaska at Fairbanks is an excellent demonstration of the potential for commercial UAS,” Eben Frankenberg, the founder and CEO of Echodyne, . “With Iris Automation and Echodyne sensor technologies, routine commercial missions like linear inspection and medical deliveries to remote communities are both practical and safe.” The radar system developed by Echodyne relies on metamaterials technology, which uses specially structured electronics to bend electromagnetic waves. Circuits based on metamaterials can allow for the construction of flat-panel radar devices that match the performance of larger, more expensive phased array antennas. Echodyne is one of several metamaterials-based startups that have been spun out from Bellevue, Wash.-based Intellectual Ventures with Gates’ financial backing. Its most recently reported funding round from Gates as well as Silicon Valley’s New Enterprise Associates, the Kresge Foundation, Lux Capital, Seattle’s Madrona Venture Group and Vulcan Capital. The company is playing a role in several tests of detect-and-avoid technologies for drones. In March, Echodyne announced that its as part of NASA’s UAS Traffic Management program, in Texas and Nevada. Meanwhile, the tests in Alaska will continue, focusing on pipeline inspection as well as other applications potentially including medical device delivery, search and rescue, road monitoring and surveys of fish and wildlife.
Voom, a subsidiary of Airbus, offers on-demand helicopter booking services in Brazil and Mexico. (Airbus Photo) Add the Airbus subsidiary to the — and to the list of pioneers in co-located and distributed workplaces. Both of those talking points are highlighted in a on working remotely, written last month by Robert Head, a senior software engineer at Voom. The posting was brought to light today by the . The California-based startup has been offering its app-based, on-demand helicopter taxi service in Mexico City and São Paulo, and last month it in league with . In his blog posting, Head, who works remotely from Ashland, Ore., talked about software development rather than flight plans. “When Voom decided to grow our own internal team of developers, we chose to locate the office not in San Francisco or Silicon Valley, but rather in Seattle, which has a similarly booming technology scene and an ecosystem of great talent,” he wrote. Today LinkedIn , and the company’s careers webpage has, including a spot for a vice president of engineering. But the point of Head’s posting wasn’t how Voom conducts its operations in Seattle. Instead, he focused on how the Seattle office serves as a springboard for a far more widely dispersed team. “For the first year, all new hires were local to the Seattle office and the rest of the team was ‘on the big screen,’ as we say,” Head wrote. “After a year, recruiting was getting tougher and we were at a crossroads. We knew we wanted access to a wider, more diverse range of experienced colleagues, but it’s tough to find that within one city.” Voom’s solution was to go to a blended pair programming model, facilitated by screen-sharing and videoconferencing. Co-workers can pop into a virtual shared space and pair up with colleagues. “A good pairing session gets into a rhythm, a give and take,” Head said. “In typical pairing terminology, one person is ‘driving’ (using the mouse and keyboard) and the other is ‘navigating’—holding mental context, noticing opportunities, making suggestions. It’s important to a healthy, egalitarian environment that these roles are switched frequently.” Thanks to pair programming, Voom no longer limits its job pool to local hires, Head said. Additional details about the workplace model are laid out in , and on . Does Voom’s Seattle presence suggest that its helicopter ride-hailing service will be swooping in anytime soon? We didn’t immediately get an answer to that question from Head or from Voom’s HQ in San Francisco when we contacted them, but we’ll update this item with anything substantive we hear back. In the meantime, it’s worth considering that a lot of companies tap the Seattle area’s software engineering talent even though they put their products and services through real-world tests elsewhere. , which is working on ride-hailing services that make use of self-driving cars, serves as a good example. Even though LinkedIn as working in the Seattle area, and has here, the company says it has no immediate plans to test-drive its cars in the Emerald City. For now, you’ll have to travel to San Francisco, Phoenix or Detroit to see Cruise’s cars in action. And unless we hear differently, you’ll have to go to Mexico City, São Paulo or maybe San Francisco to see the results of the paired programming work that Voom is doing in Seattle.
Abundant Robotics’ apple picking system was developed in collaboration with Washington state apple growers. (Abundant Robotics Photo) Next fall, as you browse the produce section at your local grocery store, pay close attention to the apples. You might be witnessing American history. For the first time, some of the apples sold in the U.S. will be picked by a robot rather than human hands. That’s thanks to agricultural automation startup , the maker of apple harvesting machines that will partake in Washington state’s next harvest. “This will be the first season that we’re actually ready to harvest commercially,” said Abundant CEO . “It’s incredibly exciting.” Abundant’s picker has more in common with a really smart Hoover vacuum than a human hand. The robot moves down rows of orchards and uses artificial intelligence with a dash of LIDAR to search for ripe apples. Once spotted, a robotic arm with a vacuum gently sucks the apples from the tree into a bin. The achievement is owed to advances not only in machine learning and robotics but also in agriculture. The architecture of apple trees has evolved over the decades, and it’s now common to grow them on trellises like you would tomatoes or cucumbers. Modern apple trees are also smaller, derived from dwarf varietals that yield more per acre and produce fruit more quickly after being planted. These horticultural leaps have allowed farmers to double their apple yields. They’ve also made the job of picking easier for humans and, now, for robots. Karen Lewis, a tree fruit specialist at Washington State University who has worked with Abundant and other robotics startups, said that apple trees have reached a “sweet spot” for robotic harvesting. Orchards are now sufficiently uniform and predictable for machines to reliably pick fruit, and canopies are narrow enough for sunlight, the human eye and vision systems to penetrate. Successful tech companies, she said, are the ones that listen to what farmers need. “We’re not going to let technology be the driver here. Horticulture needs to be the driver.” The U.S. debut comes following a rollout in New Zealand, where Abundant began a commercial harvest earlier this year. Steere said the decision to make the global debut in New Zealand rather than Washington was based purely on seasonal luck. He added that Abundant owed a lot to the Washington growers, who gave the startup crucial support and feedback in its early years. “The special thing about Washington is the scale, the sophistication and the openness to supporting innovation,” he said. Steere declined to say how many machines would be put to use this fall or which growers Abundant is working with. Menlo Park, Calif.-based Abundant has raised $12 million with backing from GV, formerly Google Ventures, among others. The startup formed out of the robotics division at SRI International, a research lab in California. Abundant’s main competition is Fresh Fruit Robotics, an Israeli startup that’s developing a picker that uses a claw-like appendage. Both companies have received funding from the Washington State Tree Fruit Association. Steere and his team have been developing the robots with help from growers in Washington state for the past six years. The process has involved both give and take: Abundant received feedback on how to improve the machines, and growers have adapted their practices to work with automation. “It’s not that people haven’t wanted to automate harvesting fruit, it’s that it’s never been possible,” Steere said. “Now we’re making it possible.” Modern apple orchards often feature smaller trees with narrow canopies. (WSU Photo) Wherever automation appears, so does the question of whose jobs might be displaced. But American farmers have for years an are increasingly dependent on foreign seasonal labor. Approvals for H-2A visas, which allow foreigners to do agriculture-related work in the U.S. temporarily, increased by a factor of five over the past 13 years, according to the USDA. The program now accounts for around 8 percent of the total agricultural workforce. In Washington state, those seasonal worker visas exploded from 3,014 to 24,862 in the past eight years. “We’re being squeezed,” said Lewis. “There’s substantial pressure from not being able to attract and retain a workforce.” That labor shortage has been accompanied by higher wages. In Washington state, the minimum wage is set to jump by $1.50 to $13.50 an hour next year, an increases that Lewis said could amount to a quarter of a million dollars for a grower that manages 250 acres. The typical American farm worker makes $11.84 per hour. (USDA data / GeekWire chart) Farmers have also been battered by trade wars, which have , and they face threats from . Robots could help carry some of the weight of Washington state’s enormous apple harvest, which supplies nearly in the U.S., according to the U.S. Apple Association. The U.S. ranks second globally in apple production, behind China. Agricultural robot shipments are from 60,000 units today to more than 727,000 in 2025, according to market research firm Tractica. Still, replacing people with new technology is “scary and expensive,” Lewis said. Lewis is hopeful that the robots are finally reaching a point where they can make growers more profitable. “We’ve had a bit of over promise, under deliver” from tech companies, she said. “That has led to fatigue among growers.” Abundant’s leadership boasts some agricultural street cred. Steere grew up in a family of farmers, spending time on his grandfather and uncle’s cotton and soybean farms. Abundant co-founder Michael Eriksen was raised on a dairy farm in Denmark. And Curt Salisbury, another co-founder, is a native of the Columbia Basin, one of Washington state’s main apple-growing regions. “When I was a kid, I was fascinated with combines, the big machines that would go through and harvest soybeans, or cotton pickers that would drive through and pick cotton,” Steere said. “We get to make these machines for an industry that’s never had that kind of automation before.”
Washington Hyperloop team members show off their Husky spirit at an on-campus unveiling of this year’s pod racer. Veteran team member Mitchell Frimodt peeks out from within the pod’s carbon composite shell, while the guts of the racer are on display on a table at left. (Margo Cavis Photo) Could this year be the year for ? For the fourth time, the students on University of Washington’s pod-racing team are taking aim at the top prize in tech titan Elon Musk’s competition, and this time they’ve got their racer down to fighting weight. This year’s purple pod racer, which looks like a cross between a bobsled and a miniaturized bullet train, was unveiled Friday night at UW’s Husky Union Building. “Our pod this year is about 60 percent of the weight of last year’s pod, with the same propulsion specs,” engineering senior Mitchell Frimodt, one of the veterans on the Hyperloop team, told GeekWire. “That’s our performance boost.” Propulsive oomph per pound is a key factor in what’s become an annual tradition that plays out at SpaceX’s headquarters in Hawthorne, Calif. This year, Washington Hyperloop and a dozen other collegiate teams are . Competitors will show off the racers they’ve built, and the best of the pack will face off in time trials conducted in a mile-long tube that’s been built just across the street from SpaceX’s rocket factory. The fastest team wins. And in the previous three competitions, the fastest team has been WARR Hyperloop from the Technical University of Munich in Germany. This year, Munich’s student engineers are racing under a different team name — — but they’re expected to be every bit as formidable. “We’re looking to give them a run for the money,” Frimodt said. Frimodt is speaking figuratively: There’s no prize money as such, but the Hyperloop contest gives those who do well an enviable spotlight in the engineering world. Frimodt said some of Washington Hyperloop’s alumni are now working at SpaceX, Tesla, Apple and other cutting-edge ventures — including Musk’s very own tunneling venture, . “One of our old business guys is actually with the Boring Company currently, down in L.A.,” he said. “I haven’t heard exactly what he’s up to down there, but that’s pretty cool. We’ve had one alum go on to .” Musk as a way to short-circuit traffic snarls within and between urban areas. The original concept called for pods to travel through low-pressure tubes at near-supersonic speeds, cutting the travel time between the San Francisco Bay Area and Los Angeles to about a half-hour. To cite another example, a Pacific Northwest Hyperloop could . Since then, the intercity Hyperloop concept has given rise to commercial ventures that don’t involve Musk, including and . For now, Musk is concentrating on the Boring Company’s somewhat less speedy tunnel travel concept, which is known as the Loop. Loop projects are in various stages of development in , , and the . Frimodt said most of the 40 or so members of the Washington Hyperloop team see their involvement as a way to exercise their general engineering skills, rather than creating a commercially viable Hyperloop pod. “For us, it’s more on pushing the edge of technology. … Our focus is very much on the competition,” he said. Only about half of the team’s members are veterans of past Hyperloop contests. Thanks to the team’s previous successes, including an , Washington Hyperloop’s organizers received “a ton of applicants” seeking to join the team last fall, Frimodt said. So what’s the secret sauce for this year? It’s not necessarily the propulsion technology: The team is going with a cold-gas thruster system that’s similar to the one that won them fourth place in last year’s competition. Basically, pressurized nitrogen blasts through a rocket-like nozzle to shoot the pod down the Hyperloop track. “It’s a rocket without the combustion,” Frimodt explained. One big change from last year has to do with weight reduction. This year’s team members were rigorous about enforcing weight budgets for each of the pod’s components, and were lots more liberal about using lightweight carbon composites rather than metal parts. Another change has to do with advance preparation, particularly when it comes to the software side of the project. Last year, “people were working on our code on the day of the competition,” said Fedor Paretsky, a UW junior in applied physics who works on the control and power subsystem team. This year, the Washington Hyperloop team is ahead of the game, development-wise. And one of the reasons for that has to do with one of the team’s sponsors, a California-based tech company named , which is providing the team with financial support as well as help with an open-source programming language for cloud-based applications known as . “Ballerina acts as our data relay,” Paretsky said. “It relays data every 25 milliseconds. … If we decide that something’s going wrong, or we need to do an emergency stop, Ballerina ensures that the pod is communicated with, efficiently.” Ballerina also serves to coordinate software development. “Anybody can literally just pull the image [for the pod’s software] off the cloud,” he said. In addition to WSO2, Ballerina and various UW departments, the team has recruited ranging from Boeing to Pagliacci Pizza (and it doesn’t take an engineer to guess what Pagliacci is contributing). Their logos are all emblazoned on what Washington Hyperloop hopes will be this year’s top pod. It could be a nail-biter: Last year’s competition at SpaceX saw the WARR team set a new world record for Hyperloop travel, with a . Can Washington Hyperloop improve on that mark? “We’re not 100 percent sure yet, until we get a final pod weight based off everything we put on there, and do our static-fire testing to characterize our propulsion system,” Frimodt said. “But it’s probably looking like somewhere from 260 miles an hour to 300 miles an hour.”
Washington Hyperloop team members show off their Husky spirit at an on-campus unveiling of this year’s pod racer. Veteran team member Mitchell Frimodt peeks out from within the pod’s carbon composite shell, while the guts of the racer are on display on a table at left. (Margo Cavis Photo) Could this year be the year for ? For the fourth time, the students on University of Washington’s pod-racing team are taking aim at the top prize in tech titan Elon Musk’s competition, and this time they’ve got their racer down to fighting weight. This year’s purple pod racer, which looks like a cross between a bobsled and a miniaturized bullet train, was unveiled Friday night at UW’s Husky Union Building. “Our pod this year is about 60 percent of the weight of last year’s pod, with the same propulsion specs,” engineering senior Mitchell Frimodt, one of the veterans on the Hyperloop team, told GeekWire. “That’s our performance boost.” Propulsive oomph per pound is a key factor in what’s become an annual tradition that plays out at SpaceX’s headquarters in Hawthorne, Calif. This year, Washington Hyperloop and a dozen other collegiate teams are . Competitors will show off the racers they’ve built, and the best of the pack will face off in time trials conducted in a mile-long tube that’s been built just across the street from SpaceX’s rocket factory. The fastest team wins. And in the previous three competitions, the fastest team has been WARR Hyperloop from the Technical University of Munich in Germany. This year, Munich’s student engineers are racing under a different team name — — but they’re expected to be every bit as formidable. “We’re looking to give them a run for the money,” Frimodt said. Frimodt is speaking figuratively: There’s no prize money as such, but the Hyperloop contest gives those who do well an enviable spotlight in the engineering world. Frimodt said some of Washington Hyperloop’s alumni are now working at SpaceX, Tesla, Apple and other cutting-edge ventures — including Musk’s very own tunneling venture, . “One of our old business guys is actually with the Boring Company currently, down in L.A.,” he said. “I haven’t heard exactly what he’s up to down there, but that’s pretty cool. We’ve had one alum go on to .” Musk as a way to short-circuit traffic snarls within and between urban areas. The original concept called for pods to travel at near-supersonic speeds, cutting the travel time between the San Francisco Bay Area and Los Angeles to about a half-hour. To cite another example, a Pacific Northwest Hyperloop could . Since then, the intercity Hyperloop concept has given rise to commercial ventures that don’t involve Musk, including and . For now, Musk is concentrating on the Boring Company’s tunnel travel concept, which is known as the Loop. Loop projects are in various stages of development in , , and the . Frimodt said most of the 40 or so members of the Washington Hyperloop team see their involvement as a way to exercise their general engineering skills, rather than creating a commercially viable Hyperloop pod. “For us, it’s more on pushing the edge of technology. … Our focus is very much on the competition,” he said. Only about half of the team’s members are veterans of past competitions. Thanks to the team’s past successes, Washington Hyperloop’s organizers received “a ton of applicants” seeking to join the team last fall, Frimodt said. So what’s the secret sauce for this year? It’s not necessarily the propulsion technology: The team is going with a cold-gas thruster system that’s similar to the one that won them fourth place in last year’s competition. Basically, pressurized nitrogen blasts through a rocket-like nozzle to shoot the pod down the Hyperloop track. “It’s a rocket without the combustion,” Frimodt explained. One big change from last year has to do with weight reduction. This year’s team members were rigorous about enforcing weight budgets for each of the pod’s components, and were lots more liberal about using lightweight carbon composites rather than metal parts. Another change has to do with advance preparation, particularly when it comes to the software side of the project. Last year, “people were working on our code on the day of the competition,” said Fedor Paretsky, a UW junior in applied physics who works on the control and power subsystem team. This year, the Washington Hyperloop team is ahead of the game, development-wise. And one of the reasons for that has to do with one of the team’s sponsors, a California-based tech company named , which is providing the team with financial support as well as help with an open-source programming language for cloud-based applications known as . “Ballerina acts as our data relay,” Paretsky said. “It relays data every 25 milliseconds. … If we decide that something’s going wrong, or we need to do an emergency stop, Ballerina ensures that the pod is communicated with, efficiently.” Ballerina also serves to coordinate software development. “Anybody can literally just pull the image [for the pod’s software] off the cloud,” he said. In addition to WSO2, Ballerina and various UW departments, the team has recruited ranging from Boeing to Pagliacci Pizza (and it doesn’t take an engineer to guess what Pagliacci is contributing). Their logos are all emblazoned on what Washington Hyperloop hopes will be this year’s top pod. It could be a nail-biter: Last year’s competition at SpaceX saw the WARR team set a new world record for Hyperloop travel, with a . Can Washington Hyperloop improve on that mark? “We’re not 100 percent sure yet, until we get a final pod weight based off everything we put on there, and do our static-fire testing to characterize our propulsion system,” Frimodt said. “But it’s probably looking like somewhere from 260 miles an hour to 300 miles an hour.”
Overwatch Imaging’s real-time fire perimeter mapping increases safety. (Overwatch Image) , an Oregon venture that specializes in airborne imaging systems, says it has won a multimillion-dollar investment from , which focuses on aviation solutions that are specialized to suit the needs of its clients in government and the commercial sector. The Series A funding deal, announced today, builds on an existing partnership between Overwatch and Tenax, a privately held company that’s based in Mississippi. It marks the first outside investment taken in by Overwatch, which was founded in 2016. Overwatch CEO and co-founder Greg Davis said the size of the investment amounts to millions of dollars, but he declined to be more precise. The money will go toward expanding Overwatch’s production operations into a larger facility in Hood River, Ore., and accelerating development of the company’s AI software for autonomous imagery collection and analysis. In a news release, Tenax Aerospace’s president, Taran Bakker, called Overwatch “an emerging leader in artificial intelligence and autonomy in airborne imaging.” “Overwatch Imaging has developed an exciting new technology that will be very valuable to customers with special missions involving surveillance, mapping or threat detection,” Bakker said. Tenax Aerospace provides special mission aircraft and related services to customers including the Federal Aviation Administration and the departments of Defense, Justice; Agriculture and Homeland Security. The company focuses on applications that are critical to national security and the public interest, including aerial fire suppression, aerial intelligence gathering and airborne data acquisition. Tenax and Overwatch are already working together on a U.S. Forest Service project related to monitoring and fighting forest fires. That project involves the use of Overwatch’s imaging system on Tenax’s aircraft. Future projects could focus on applications such as border surveillance and maritime traffic monitoring. Davis said Tenax Aerospace emerged as the ideal partner for Overwatch Imaging’s expansion campaign during a six-month process to assess potential investors. As a result of that process, Tenax will be contributing more than money: Bakker will be joining Davis and co-founder Nick Anderson on Overwatch’s board. “We immediately shared a common vision for the future,” said Davis, who’s a veteran of . “I am excited to have Taran’s expertise and enthusiasm on our board as we grow.”
Boeing’s first 737 MAX 9 jet makes its appearance at the company’s Renton plant in 2017. (Boeing Photo) Boeing says a warning alert system that figures in the investigation of two catastrophic 737 MAX crashes didn’t work the way it was supposed to because of a software flaw that engineers identified a year before the accidents. The revelation adds a new twist to the debate over the company’s safety practices. In this case, the debate focuses on a feature known as the “AOA Disagree” alert, which is supposed to light up in the cockpit if there’s a mismatch in data coming from two angle-of-attack sensors on the plane. Investigators suggest that bad sensor data played a key role in October’s Lion Air crash in Indonesia, which killed all 189 people aboard the plane; and March’s Ethiopian Airlines crash, which killed 157. Within days of the Ethiopian crash, all 737 MAX airplanes were grounded worldwide. Boeing engineers knew about a problem with the “AOA Disagree” alert well before that. The alert was originally intended to be a standard feature on the 737 MAX and the previous generation of 737 planes, known as the 737 NG (for “Next Generation”). But in a , Boeing said that in 2017, several months after deliveries began, engineers became aware that the 737 MAX display system software didn’t meet the original requirements. “The software delivered to Boeing linked the AOA Disagree alert to the AOA indicator, which is an optional feature on the MAX and the NG. Accordingly, the software activated the AOA Disagree alert only if an airline opted for the AOA indicator,” Boeing said. The company said it followed standard procedure for reviewing the issue, and determined that neither the software-based angle-of-attack indicator nor the alert was necessary for safe operation of the airplane. “Accordingly, the review concluded, the existing functionality was acceptable until the alert and the indicator could be delinked in the next planned display system software update,” Boeing said. “Senior company leadership was not involved in the review and first became aware of this issue in the aftermath of the Lion Air accident.” That’s when Boeing discussed the issue with the Federal Aviation Administration. About a week after the Indonesia crash, Boeing and the FAA issued bulletins noting that both the angle-of-attack indicator and the alert system were optional. Last December, Boeing conducted a follow-up internal safety review, which confirmed the view that the absence of a warning alert did not present a safety issue. The findings of that review were shared with the FAA, Boeing said. Three months after the review, the Ethiopian crash occurred. In the wake of that crash, Boeing has said the “AOA Disagree” issue will be fixed before the 737 MAX returns to flight. “Boeing is issuing a display system software update, to implement the AOA Disagree alert as a standard, standalone feature before the MAX returns to service,” the company said in its statement. “When the MAX returns to service, all MAX production aircraft will have an activated and operable AOA Disagree alert and an optional angle of attack indicator. All customers with previously delivered MAX airplanes will have the ability to activate the AOA Disagree alert.” The statement adds context to earlier reports that the angle-of-attack alert system was available only as part of an optional software package. It’s debatable whether either of the crashes could have been avoided if the alert was available as a standard feature on the planes involved. In the Ethiopian case, the investigation suggests that the pilots knew about the issue surrounding the angle-of-attack sensors and their effect on the 737 MAX’s automatic flight control system — but were nevertheless unable to pull the plane out of its final dive. In any case, the questions surrounding what Boeing engineers knew and when they knew it seem sure to figure in the multiple investigations sparked by the crashes. U.S. Rep. Rick Larsen, D-Wash., told KOMO News that the FAA’s actions before and after the crashes will also be examined during a May 15 hearing before the House Subcommittee on Aviation, which he chairs. ”We need to first find answers to ensure the safety of the airplane, to ensure the FAA is doing the right thing, that Boeing is doing the right thing, and that’ll be our focus,” Larsen told KOMO on Saturday, a day before Boeing issued its statement.
Islamic State forces blew up the Al-Nuri Mosque in Mosul, Iraq, as they withdrew from the city in 2017. (Photo Courtesy of History Blocks) Can a video game reclaim centuries’ worth of lost cultural heritage in the Middle East? Microsoft’s Minecraft Education Edition is being used to do just that, in league with UNESCO and schools around the world. History Blocks takes advantage of the educationally oriented Minecraft platform to build virtual versions of ancient monuments — starting with sites that were destroyed by the Islamic State group in Syria and Iraq, and by the Taliban in Afghanistan. The project was conceived and developed by Agencia Africa in Brazil, and put to its first test this February at Escola Bosque, a private school in São Paulo. “It is surprising to see the level of the students’ engagement in the History Blocks project,” Escola Bosque’s pedagogical director, Silvia Scuracchio, said today in a news release. “At the same time that they solve complex geometry, logic and abstract challenges, it’s possible to see how they get involved with the culture and history behind the monuments and their destruction. For many of them, it was their first contact with concepts such as cultural destruction and ideology oppression.” Students aged from 9 to 13 built up their models from historical images of the , the and the entrance to the in Syria, as well as the and the in Iraq, and Afghanistan’s . Since February, the History Blocks project has been picked up by schools in more than 30 countries using the Minecraft Education Edition. “Technology is a tool to transform education and bring to life methods that used to be unthinkable when it comes to teaching,” said Daniel Maia, manager for academic projects at Microsoft Brazil. “The project on UNESCO’s world heritage sites opens the door for students all over the world to study important monuments of our history.” Minecraft and History Blocks are great teaching tools, but if you’re looking for high-fidelity models of heritage hotspots ranging from to Paris’ Notre Dame Cathedral, they’re covered by other software and survey programs. The International Council on Monuments and Sites, a U.N. advisory panel also known as ICOMOS, is one of the leaders in the effort to document cultural sites. Over the past few years, ICOMOS’ (from the Arabic word for “phoenix”) has been conducting surveys of sites in Syria, starting with six representative buildings in Damascus. You can . A historical conservation initiative called is playing a key role in 3-D documentation, for Project Anqa as well as s around the world. CyArk’s detailed digital scans feed into Google Arts and Sciences’ . For a powerful demonstration of the technology, check ou in South Dakota. (But make sure your computer is powerful enough for the task.) Could virtual models provide enough information to rebuild lost monuments? Historians and architects certainly hope so: They’re banking on surveys of Notre Dame, including conducted several years ago under the leadership of the late art historian , to serve as a guide for the reconstruction ahead.
Tesla CEO Elon Musk checks out the Model Y during its unveiling in March. (Tesla via YouTube) Tesla is aiming to raise up to $2.3 billion in newly announced offerings of stock and convertible notes, just a week after CEO Elon Musk told analysts that the . Musk himself will purchase an additional $10 million of common stock, Tesla said today in a . That would add to his status as the electric-car company’s largest shareholder, with roughly 20 percent of Tesla’s shares. The share price was more than 3.5 percent above the previous day’s close during midday trading today. Wedbush Securities analyst Dan Ives said in a note to investors that the offerings were a “clear net positive for Tesla” because they cleared up long-lingering uncertainty over whether Tesla would have enough cash on hand to meet upcoming debt payments. One of the offerings announced today will make $650 million in common stock available, while the second offering calls for the issuance of up to $1.35 billion in convertible senior notes due in 2024. There’s also a 30-day option for underwriters to purchase up to an additional 15% of each offering. If all the options are exercised, the gross proceeds would come to about $2.3 billion before discounts and expenses, Tesla said. Goldman Sachs and Citigroup are acting as joint lead managers for the offering, with involvement as well from BofA Merrill Lynch, Deutsche Bank Securities, Morgan Stanley, Credit Suisse, Societe Generale and Wells Fargo Securities. The company said it would “use the net proceeds to further strengthen its balance sheet, as well as for general corporate purposes.” In last week’s financial report, Tesla for the first quarter of the year, after posting profits for the previous two quarters. Looking ahead, the company has ambitious plans to ramp up production of its Model 3 electric car and move ahead with projects ranging from its Semi truck, Model Y crossover SUV and all-electric pickup truck to electricity-generating solar roofs, a and car insurance. Some analysts worry about the effect of Tesla’s financial losses, the gradual fade-out of federal tax credits and rising competition in the electric-vehicle market. Such uncertainties, coupled with , have led to dramatic ups and downs in the share price over the past year. A year ago, for Tesla’s investors: “Do not buy if volatility is scary,” he said.
Boeing CEO Dennis Muilenburg takes questions at a news conference in Chicago. (AP via YouTube) Boeing CEO Dennis Muilenburg stuck to his positions on the safety of the 737 MAX airplane today during a contentious annual shareholders’ meeting and news conference in Chicago. Muilenburg took questions in a face-to-face public forum for the first time since last month’s due to concerns raised by two catastrophically fatal crashes last October and this March. At one point, a reporter asked Muilenburg whether he’d resign. “My clear intent is to continue to lead on the front of safety and quality and integrity,” he replied. “That’s who we are as a company.” Muilenburg said that he’s been talking with factory workers in Renton, Wash., and with Boeing test pilots over the past few weeks. “To the core of our people, they care about this business and the safety of our airplanes,” he said. “That’s what I’m focused on.” Investigations into October’s Lion Air crash in Indonesia, and March’s Ethiopian Airlines crash in Ethiopia, have focused on an automatic flight control system known as the Maneuvering Characteristics Augmentation System, or MCAS. The system was designed to replicate the operating conditions of previous-generation 737 planes on the 737 MAX, which is equipped with bigger jet engine. But preliminary findings suggested that during each fatal flight, spurious data from an angle-of-attack sensor repeatedly forced the plane into a steep dive. Boeing had laid out procedures to regain control, but in the two tragedies, the procedures either weren’t followed to the letter or didn’t work. Muilenburg resisted characterizing the MCAS issue as a design flaw or a mistake. Instead, he said the issue was a “link in the chain” that will be broken thanks to a software update that’s being tested for certification by the Federal Aviation Administration and other regulatory agencies around the world. The process by which the 737 MAX was originally certified for flight in 2017 is currently under investigation by an internal Boeing team and the FAA, as well as by the Justice Department and the FBI. Muilenburg insisted that MCAS system “was designed per our standards” and followed proper certification procedures. “We haven’t seen a technical slip or gap in terms of the fundamental design and certification of the approach,” Muilenburg told reporters. “That said, we know this is a link in both accidents that we can break. That’s a software update that we know how to do. We own it, and we will make that update, and this will make the airplane even safer going forward.” During today’s meeting, Boeing shareholders voted down a proposal to remake the company’s chairmanship as an independent position that would rule out Muilenburg’s dual role as CEO and chairman. Relatives of some of the victims of the 737 MAX crashes traveled to Chicago to take part in a news conference aimed at drawing attention to lawsuits being filed against Boeing. One of the speakers was Manant Vaidya, a Canadian of Indian descent who lost six family members in the Ethiopia crash. Vaidya was sharply critical of Muilenburg’s comments. “He said that all design and certifications were followed. At the end of day, if all certifications were done, how could the crash still have occurred?” he said. “I am completely lost right now. I want to make sure this doesn’t happen to anyone else in the world.” Over the weekend, a number of media outlets reported that some 737 MAX planes didn’t have an indicator known as a “disagree alert,” which might have given pilots an early indication that an angle-of-attack sensor was feeding bad data to the MCAS system. Today, Boeing : “Boeing included the disagree alert as a standard feature on the MAX, although this alert has not been considered a safety feature on airplanes and is not necessary for the safe operation of the airplane. Boeing did not intentionally or otherwise deactivate the disagree alert on its MAX airplanes. “The disagree alert was intended to be a standard, stand-alone feature on MAX airplanes. However, the disagree alert was not operable on all airplanes because the feature was not activated as intended. “The disagree alert was tied or linked into the angle of attack indicator, which is an optional feature on the MAX. Unless an airline opted for the angle of attack indicator, the disagree alert was not operable. [The angle-of-attack indicator is a software-based information feature that’s distinct from the angle-of-attack sensor hardware. For more on the distinction, .] “On every airplane delivered to our customers, including the MAX, all flight data and information needed to safely operate the aircraft is provided in the flight deck and on the flight deck display. This information is readily accessible to pilots, and it always has been. “The air speed, attitude, and altitude displays, together with the stick shaker, are the primary flight information indicators in the flight deck. All recommended pilot actions, checklists, and training are based upon these primary indicators, not on the AOA disagree alert or the angle of attack indicator. “As the MAX safely returns to the air after the software modifications are approved and certified, all MAX production aircraft will have an activated and operable disagree alert and an optional angle of attack indicator. All customers with previously delivered MAX airplanes will have the ability to activate the disagree alert per a service bulletin to airlines. “We are confident that when the MAX returns to the skies, it will be one of the safest airplanes ever to fly.” For what it’s worth, Boeing’s shares finished the trading day down 0.46%, at $379.05.
Tesla CEO Elon Musk unveils the Model 3 electric car in 2016. (Tesla via YouTube) Tesla CEO Elon Musk has to keep his Twitter habit in check — and feels comfortable enough with the arrangement to refer to it in a teasing tweet. Friday’s settlement was a serious matter: Musk could have faced sanctions for contempt of court if he failed to patch up the rift with the SEC over whether he was following the terms of an . To refresh your memory, the SEC last September for incorrectly claiming on Twitter that he had secured funding to buy up publicly traded shares of the Tesla electric-car company and take the company private. Under the terms of the settlement of that fraud case, Musk and Tesla each paid a $20 million fine, Musk agreed to surrender his title as chairman, and he agreed to have tweets about Tesla reviewed in advance by a company overseer. (Some folks have referred to that person as Musk’s “Twitter sitter.”) During the months that followed, Musk made statements suggesting that he had a loose interpretation of what fell under the Twitter sitter’s purview, and in February he sent out a controversial Tesla-centric tweet that the company’s lawyers acknowledged wasn’t cleared in advance. Tesla made 0 cars in 2011, but will make around 500k in 2019 — Elon Musk (@elonmusk) That’s what set off the . U.S. District Judge Alison Nathan told Musk and the SEC to resolve their differences, and after a couple of extensions, the two parties finally filed the for Nathan’s review and likely approval. The consent motion gets specific about the topics that Musk will have to have the company overseer review before issuing written statements in venues ranging from Twitter to press releases and blog posts. (We’ve included the list at the end of this report.) Musk didn’t make any direct comments about the agreement, or how he’ll change his tweeting ways, but he did touch on the tiff in a Twitter conversation about his Twitter habit. (So meta!) Here’s how the exchange played out: sometimes i think Neuralink is already implanted in you… how can you keep up with all of your twitter mentions & pick which one is important or interesting to respond to!? — Evelyn Janeidy Arevalo (@JaneidyEve) Only see about 5% of mentions. No deep logic to those I answer. Aspirationally useful, but often whimsical. — Elon Musk (@elonmusk) It’s just me writing, so the foolish things I say are entirely my fault
Alaska Airlines CEO Brad Tilden and Nordstrom’s chief digital officer, Ken Worzel, share a laugh during the 2019 State of Technology Luncheon, presented by the Technology Alliance. (GeekWire Photo / Alan Boyle) There was a heavy aerospace spin to this year’s presented today by the Technology Alliance at the Seattle Sheraton. Alaska Airlines CEO was the keynote speaker for what’s been billed as “the premier annual event of Washington’s innovation community.” Three other aerospace executives had their time in the spotlight as well, and hundreds of representatives from the tech industry, academia and government were in attendance. Here are a few highlights from the event: We’re all tech companies now: During an onstage fireside chat, Nordstrom chief digital officer Ken Worzel asked Tilden whether he classified Alaska Airlines as a technology company. “Absolutely,” Tilden replied. “I bet every single person in this room thinks of where they work as a technology company. It’s hugely, hugely important to us.” What about my Wi-Fi? One of the biggest applause lines came when Tilden responded to a question about the slowness of in-flight Wi-Fi. he said. “We have 225 airplanes with Wi-Fi, 25 of them have satellite [connectivity]. Satellite is 20 times faster than ground-based Wi-Fi. All the airplanes will be done 12 months from now. … It’s not going to be as fast as your 1-gigabit home computer, but it will be as fast as your mobile phone, and you’ll be able to download and stream.” A vote of support for Boeing: Tilden said Alaska Airlines is remaining “hugely loyal to Boeing” as the airplane manufacturer works through issues that have arisen in the wake of two catastrophic crashes. Software updates are soon expected to address problems with an automated flight control system on the 737 MAX. “Our team has looked at them,” Tilden said of the updates, “and we’re satisfied that they’re the right changes.” Debut for Digital Winglets: CEO Tom Gibbons showed off his company’s app, which monitors and analyzes an airplane’s vital statistics and crowdsourced situational data in real time to optimize the plane’s fuel usage and route during a flight. “It’s Waze for your airplane,” Gibbons said, referring to the . Today the company that it was partnering with Alaska Airlines to develop NASA’s technology and deploy it across Alaska’s entire fleet. also makes sure airplanes are serviced in a timely manner during on-the-ground turnarounds, and keeps track of an airplane’s operational health. Electric aviation on the horizon: CEO Roei Ganzarski recapped his electric propulsion company’s recent deals to with electric motors, and . Ganzarski said that Harbour Air’s first converted all-electric plane would make its first test flight in November, and that Eviation’s Alice airplane would have its first flight by the end of the year. “The one thing I would love to see this state do is take the lead in the United States,” he said. Ganzarski suggested that Washington state could encourage short-haul airlines to go all-electric by, say, 2050. The space gold rush: CEO Curt Blake talked up his company’s logistical role in , this year’s launch that , and other satellite missions. “I was downtown in Pioneer Square not long ago, and I went by that ,” Blake said. “And I thought, that’s not really unlike what we’re doing. The people in Seattle were selling tickets to get up to Alaska, they were selling picks and shovels. … We’re basically doing the same thing: We’re providing the means for people to get to outer space and really fulfill their dreams, commercializing what’s up there — hopefully to make what’s down here a little better.”
An artist’s conception shows HAPSMobile’s Hawk30 aircraft in flight. (HAPSMobile / SoftBank Illustration) , a joint venture created by Japan’s and California-based , is jumping into the race to provide global broadband access from above, alongside SpaceX, Amazon, OneWeb and Telesat. Unlike those four companies, HAPSMobile plans to use high-flying, solar-powered planes rather than satellites to transmit signals wirelessly over a wide swath of the planet’s surface. In that respect, the concept has more in common with the aerial broadband concept that Facebook was pursuing . Like Facebook’s plan, HAPSMobile’s concept relies on high-altitude platform stations, or HAPS. The venture’s ultralight Hawk30 airplanes would fly above the clouds at an altitude of 13 miles (20 kilometers) for months at a time. Each aircraft would have a wingspan of roughly 260 feet (78 meters) and sport 10 electric-powered propellers. Average flight speed would be about 70 mph (110 kilometers per hour), . The plan calls for signals to be sent in the same frequency ranges used for terrestrial mobile telecom, making for a “seamless handover” between ground-based and aerial networks. HAPSMobile says it’s entered into a technology licensing agreement with Facebook for their advanced communication system. AeroVironment, which specializes in the development of unmanned aircraft systems for military and commercial applications, for testing. The ceiling value for HAPSMobile’s design development agreement with AeroVironment was recently raised by $39 million to reach a total of $126 million. HAPSMobile is targeting 2023 for mass production of the Hawk30 planes and the launch of commercial service. In the meantime, the venture will be working with Alphabet’s Loon, which started out as a Google project and is deploying balloon-based platforms to provide wireless telecommunications services. In an unusual arrangement, HAPSMobile says it has decided to invest $125 million in Loon, with the intention of using Loon’s balloon platforms and technology to build out a global broadband network. Loon has the right to invest the same amount in HAPSMobile sometime in the future, and will be able to use Hawk30 aircraft once they go into production. Loon and HAPSMobile will work on a common gateway or ground station network that could be deployed globally and used by both companies to provide connectivity over their respective platforms. The two companies could also share network connectivity in the air. Representatives of the two ventures said the linkup would further their shared vision of providing high-speed connectivity to billions of people around the world who are currently underserved. “Through HAPS, we aim to eliminate the digital divide and provide people around the world with the innovative network services that they need,” Junichi Miyakawa, who is the president and CEO of HAPSMobile as well as representative director and chief technology officer at SoftBank Corp., . Loon CEO Alastair Westgarth hailed “the beginning of a long-term relationship based on a shared vision for expanding connectivity to those who need it.” “We see joining forces as an opportunity to develop an entire industry, one which holds the promise to bring connectivity to parts of the world no one thought possible,” he said. HAPSMobile and Loon aren’t the only ventures that have a vision of providing broadband access from above. Here are some of the other contestants in the race: aims to put up hundreds of satellites to create a broadband constellation in low Earth orbit. The . OneWeb expects to demonstrate its service in 2020 and offer global, 24-hour coverage to customers starting in 2021. SoftBank Group and other investors just last month, and HAPSMobile says it could take advantage of OneWeb’s backhaul communication capability. eventually plans to have about 12,000 satellites in its . Two prototype satellites have been in orbit since last year, and the first operational satellites are due for launch as early as next month. Service could begin in the 2020-2021 time frame. SpaceX’s facility in Redmond, Wash., has taken a lead role in the satellite development effort. , Canada’s biggest satellite operator, had the first prototype satellite for its broadband constellation launched into low Earth orbit last year. The company expects to offer first-generation data services in the early 2020s. to use a data delivery platform that’s based on the system that Loon uses for its fleet of high-altitude balloons. has its own plan to deploy a 3,236-satellite broadband constellation, code-named Project Kuiper, . Many of the details surrounding the venture, including how and when the satellites would be deployed, have not yet been revealed. But Amazon already has . Virtually all of those jobs are based in Bellevue, Wash. The list doesn’t stop there: , and Luxembourg-based have laid out plans for space-based internet access as well. Even Lockheed Martin has mentioned the idea of having a And Aurora Flight Sciences, a Boeing subsidiary, is working on a that looks a lot like the Hawk30. How many satellites, balloons and signal-beaming drone aircraft can the world handle? How many of these ventures will come to fruition? In the next five or 10 years, we may well find out.
SalesPal CEO Ashvin Naik, Google Cloud’s Chanchal Chatterjee, Audioburst’s Rachel Batish and T-Mobile’s Chip Reno discuss the future of artificial intelligence at the Global AI Conference in Seattle. (GeekWire Photo / Alan Boyle) Artificial intelligence can rev up recommendation engines and make self-driving cars safer. It can even . But what else will it be able to do? At today’s session of the , a panel of techies took a look at the state of AI applications — and glimpsed into their crystal balls to speculate about the future of artificial intelligence. The panelists included Chanchal Chatterjee, AI leader at ; Ashvin Naik, CEO of , which markets AI-enabled sales analysis tools; Rachel Batish, vice president of product for , an audio indexing service; and Chip Reno, senior advanced analytics manager at . The moderator was Shailesh Manjrekar, head of product and solutions marketing for , a multi-cloud data storage and management company. Here are five AI frontiers that came up in today’s conversations, plus a couple of caveats to keep in mind: Smarter grocery stores: AI-enabled grocery shopping was pioneered right here in Seattle at , but the trend is catching on. Today called the Intelligent Retail Lab in Levittown, N.Y. Britain’s takes a different tack: Users fill up a virtual shopping cart, then schedule a one-hour delivery slot. Google Cloud helped Ocado develop the , including a recommendation engine that figures out customers’ shifting preferences, an algorithm that handles and prioritizes customer service emails, and a as Ocado’s previous system. Energy-saving server farms: Chatterjee pointed to how Google used its DeepMind machine learning platform to . Before AI was put on the case, 10 years’ worth of efficiency measures could reduce energy usage by merely 12 percent, he said. Within six months, AI brought about a 40 percent reduction. “That was a huge difference that AI made in a very short amount of time that we could not do with 10 years of research,” Chatterjee said. Financial market prediction: Hedge fund managers and bankers are already , detect market manipulation and assess credit risks. But Chatterjee said the models are getting increasingly sophisticated. AI is being used to predict how margin trades could play out, or whether undervalued financial assets are ripe for the picking. AI models could even anticipate . “When the lock-in period expires … that’s a great time to short,” Chatterjee said. Deeper, wider AI conversations: Chatterjee predicted that our conversations with voice assistants are likely to get wider, deeper and more personal as AI assistants become smarter. Audioburst’s Batish said conversational AI could provide a wider opening for smaller-scale startups and for women in tech. “Women are very much prominent in conversational applications and businesses,” she said. Salespal’s Naik agreed with that view — but he worried about the dearth of compelling applications, based on his own company’s experience with voice-enabled devices like Amazon Echo and Google Home. “They’re gathering dust. … We use them just to listen to music or set up alarms. That’s it,” he said. AI for good, or evil? Chatterjee said AI could be a powerful tool to root out fraud and corruption. AI applications could be built “to see what influence relationships have on outcomes — that tells you if there are any side deals being made,” he said. But Batish worried about the rise of , virtual and . “I’m actually afraid of what that could bring into our world,” she said. “It would be interesting to see how companies are trying to be able to monitor or identify fake situations that are being built out of very complicated AI.” Watch out for job disruption: Many studies have pointed out that automation is likely to disrupt employment sectors, especially in the service, manufacturing and transportation sectors. “Anything that is repetitive, that can be extracted from multiple sources, that doesn’t have a lot of creativity amd innovation, is at risk due to AI,” Chatterjee said. “That means that more people will have to move into other sectors.” Watch out for the hype: “I’d like to see people get away from the hype a little bit,” T-Mobile’s Reno said. “I’m on the client side, so I see all the pitches involving AI and ML or deep learning. … A lot of times, AI is not applicable to certain use cases where we’re applying it. Just good old-fashioned statistics or business intelligence is fine. So I think that the future of AI relies on getting past the hype and getting more into aligning these awesome tools and algorithms to specific business cases.”
The crew of OceanGate’s Titan submersible gets set for a dive of Titanic proportions. From left are Karl Stanley, Petros Mathioudakis, pilot Stockton Rush and Joel Perry. (OceanGate Photo) set a deep-diving record last week when a crew of four rode inside the Everett, Wash.-based company’s Titan submersible to the Titanic-level depth of 3,760 meters (12,336 feet) in the Bahamas. , which served as a test run for this summer’s trips to the wreck of the Titanic, marked the first time a non-military submersible carried more than three people to that depth, OceanGate said. “This dive was another important step toward deep-sea exploration to more people and places,” OceanGate CEO Stockton Rush, who served as Titan’s chief pilot for the trip, . “We are developing technologies and designing submersibles and infrastructure that is making underwater exploration more accessible than ever before.” OceanGate said it took two hours for Titan to descend to the ocean floor, outside Little Harbor near Great Abaco Island in the Bahamas. Accompanying Rush on the ride were Joel Perry, president of OceanGate Expeditions; Karl Stanley, owner of Roatan Institute of Deepsea Exploration; and Petros Mathioudakis, field technician for 2G Robotics. Titan’s crew members spent an hour capturing 4K video imagery and testing 2G Robotics’ underwater laser scanner as they surveyed the ocean floor. The imagery and data will serve to calibrate the submersible’s instruments for the , about 2,000 miles north of the Bahamas in the Atlantic Ocean. The test missions in the Bahamas serve to demonstrate the safety of the composite-hulled submersible, which is equipped with sensors and a to ensure hull integrity during every dive. A representative from Lloyd’s Register validated last week’s dive. Titan crew members are silhouetted inside the submersible during their deep-sea dive in the Bahamas. Click on the picture to watch the video. (OceanGate via YouTube) OceanGate Expeditions has scheduled a series of dives to the 107-year-old Titanic shipwreck from June to August. Titan can carry up to five people at a time down to the site. The teams will including researchers and diving experts as well as mission specialists who are paying more than $100,000 each to take part in the adventure. Mathioudakis said the dive was a “life-changing experience.” “It was incredibly exciting to be part of this record-breaking crew,” he said. “The dive provided hands-on experience ahead of the Titanic expedition, and the information we gathered throughout the dive will help streamline how we capture data, perform scans and catalog data while at sea this summer.” Other crewed submersibles , but none of those deep-sea craft could carry more than three people at a time. The Deepsea Challenger, which took filmmaker James Cameron to the , could accommodate only one person. OceanGate plans to check in on the Titanic wreck repeatedly to document the condition of the debris field with a series of high-resolution, 3-D virtual renderings. Only a few spots remain for mission specialists on this year’s series of dives, and applications are already being taken for the 2020 season via the website.
Acting Defense Secretary Patrick Shanahan speaks at an event presented by the Center for Strategic and International Studies. (Department of Defense Photo) The Department of Defense’s Office of Inspector General says acting Defense Secretary Patrick Shanahan is in full compliance with his ethics agreements and official obligations — turning back allegations that he took actions to promote Boeing, his former employer, and disparage its competitors. The findings of a weeks-long investigation, issued today, seem likely to clear what might have been an obstacle standing in the way of Shanahan taking on the Pentagon’s post on a permanent basis. Shanahan, who spent much of his 31 years at Boeing managing commercial airplane programs, won Senate confirmation to become assistant defense secretary in 2017 and after James Mattis’ departure at the end of last year. When Shanahan came to the Pentagon, he pledged to recuse himself from any matters involving Boeing. But in January, as saying that he repeatedly praised Boeing and trashed Lockheed Martin during high-level internal meetings. One former official quoted him as describing Lockheed Martin’s F-35 fighter program as “f—ed up” and complaining that the company “doesn’t know how to run a program.” Such reports led the inspector general’s office to . The office said investigators interviewed Shanahan and 33 witnesses, and reviewed more than 5,600 pages of unclassified documents as well as about 1,700 pages of classified documents related to the allegations and the relevant weapons systems. In the resulting , investigators said Shanahan’s criticism of the F-35 program didn’t appear to violate his ethics agreements: “Specifically, with regard to the alleged comments about the aircraft made by Boeing and its competitors, we concluded that Mr. Shanahan did not “repeatedly dump” on the F-35 aircraft in meetings. Rather, we determined that Mr. Shanahan’s comments related to the F-35 program and its performance, and were consistent with other comments about problems in the F-35 program made by other senior DoD officials.” The report said that some witnesses recalled hearing Shanahan refer to his experience with commercial aircraft programs during discussions about driving costs down and improving performance, but that the witnesses didn’t take such comments as promoting Boeing. “In my experience, he’s been very careful not to say something like, ‘At Boeing we would do it this way,’ ” Air Force Gen. Paul Selva, vice chairman of the Joint Chiefs of Staff, was quoted as saying in his testimony. “What he would do is say, ‘In commercial industrial engineering processes or system design processes, here are the things I have learned.’ ” The report also delved into the mechanics of meetings that Air Force Secretary Heather Wilson and Shanahan conducted with SpaceX CEO Elon Musk in December 2018, after SpaceX’s proposal for a launch services contract was turned down. Wilson thought Shanahan should have avoided meeting with Musk, because (by virtue of Boeing’s partnership with Lockheed Martin in United Launch Alliance). Other Pentagon officials, however, told Shanahan that he could meet with Musk as long as the contract wasn’t discussed. Here’s what the report had to say about Shanahan’s hourlong meeting with Musk on Dec. 6, 2018: “A member of Mr. Shanahan’s staff attended and summarized the meeting in a memorandum for the record (MFR). According to the MFR, Mr. Musk discussed increased competition from China, his plans to self-fund and launch communication satellites, and his production experience at Tesla. Mr. Musk also noted that SpaceX was not successful in the recent Air Force competition for a launch service contract and that SpaceX had written a poor proposal that ‘missed the mark.’ According to the MFR, Mr. Shanahan did not comment on the bid competition. “Mr. Shanahan told us that he met with Mr. Musk because he ‘thought it would be interesting to talk to him about his views of the future.’ Mr. Shanahan told us that Mr. Musk discussed his view on ‘electrification and autonomy,’ and how he saw ‘that space evolving here in the next 5 years because we have some really critical decisions to make in terms of our logistics system in the Army and the combat vehicles.’ ” The inspector general’s office said the meeting didn’t violate Shanahan’s ethical obligations. Shanahan’s spokesman, Lt. Col. Joe Buccino, said that the acting secretary refers any matters relating to Boeing to other Pentagon officials “and ensures no potential for a conflict of interest with Boeing on any matter.” Shanahan himself didn’t issue a statement relating to the report. Instead, he issued a family-friendly tweet relating to “Take Your Kids to Work Day.” Great to have our families together here at the Pentagon for ! Our families are the core of the Department and the heart of defense for our freedoms and American way of life. Military service is a family affair! — Acting SecDef Pat Shanahan (@ActingSecDef) Shanahan has been in charge of the Pentagon in an acting capacity, but he would have to be nominated by President Donald Trump and confirmed by the Senate to take on the defense secretary’s position on a formal basis.
Tesla CEO Elon Musk introduces the Tesla Semi truck and an updated version of the Tesla Roadster at a 2017 unveiling. (Tesla via YouTube) Tesla today reported wider-than-expected financial losses in the first quarter — due to what the company said were delivery challenges, a seasonal dip in demand and the unanticipated effects of pricing decisions. Despite the downturn from what had been a profitable couple of quarters, Tesla CEO Elon Musk was bullish on several fronts, including rollouts for the company’s and , plus the opening of Tesla’s Gigafactory in Shanghai, China. Musk is even planning to offer car insurance policies starting next month, with pricing determined by the data that’s received from the company’s cars. “We have direct knowledge of the risk profile of customers and the car,” he explained during today’s teleconference with financial analysts. “If they want to buy Tesla insurance, they have to agree to not drive the car in a crazy way. Or they can, but then the insurance rate is higher.” If it’s done right, in-house insurance could add another revenue stream to Tesla’s bottom line. That could help ease the pain for Tesla’s accountants as well as for investors, who have seen share prices slump due to concerns about long-term profitability. (The price slipped nearly 2 percent during today’s trading, to $258.66 at the close.) Net losses amounted to $702 million, and adjusted net losses per share were $2.90. That’s 13 percent worse than the year-ago figure and . Revenue was $4.5 billion, which was better than the year-ago figure but not as high as analysts thought it would be. In its , Tesla said there was a production jam-up that forced a large number of deliveries to be deferred into the second quarter. “This is the most difficult logistics problem I’ve ever seen, and I’ve seen some tough ones,” Musk said. In all, 63,000 electric cars were delivered during the first quarter, which fell far short of expectations. In addition to the logistical challenges, Tesla said pricing changes for its Model S and Model X cars caused a higher-than-anticipated return rate. One disincentive to sales was the gradual phase-out of federal tax credits for electric vehicles. Previously: The good news is that powertrain improvements have boosted the performance and range of those two models: The maximum range was extended to 370 miles on a full charge for the Model S, and 325 miles for the Model X SUV. For the past two years, Tesla has been focused on ramping up production of the Model 3, which finally . Tesla reported producing 63,000 Model 3 cars during the quarter and is aiming to raise that figure higher. “If our Gigafactory Shanghai is able to reach volume production early in Q4 this year, we may be able to produce as many as 500,000 vehicles globally in 2019,” the company said in its shareholders’ letter. “This is an aggressive schedule, but it is what we are targeting.” Musk said that the Shanghai construction project was “going incredibly well,” and that he was receiving “midnight Gigafactory email” on an almost nightly basis. On other financial fronts: Tesla’s cash on hand fell by $1.5 billion over the course of the quarter, to $2.2 billion. A $920 million convertible bond repayment accounted for most of that reduction, and the delivery snag was an additional factor. Another, linked to Tesla’s SolarCity subsidiary, is said to be due this month. One analyst asked Musk whether he wished that he had persevered with efforts to take Tesla private last year — efforts that ended up getting him in hot water with the Securities and Exchange Commission. “I would prefer that we were private,” Musk replied, “but unfortunately that ship has sailed.” Musk told analysts that “at this point I do think there is some merit to raising capital,” but he didn’t provide further details.
Boeing’s first 737 MAX makes its way through the assembly plant in Renton, Wash., in 2015. (Boeing Photo) Boeing’s first-quarter financial stats took a serious hit in the wake of two fatal crashes involving its bestselling plane, the 737 MAX. The company estimated the additional costs associated with grounding the 737 MAX fleet at $1 billion, but more uncertainty lies ahead. Boeing executives held off on providing updated financial guidance until the impact of the 737 MAX issue becomes clearer. Revenue: Boeing reported first-quarter revenue of $22.9 billion, which is 2 percent below the year-ago figure but in line with analysts’ expectations. The shortfall is due primarily to fewer deliveries of 737 jets. Earnings: Net earnings for the quarter amounted to $2.15 billion, which translates to adjusted earnings per share of $3.16. That’s 13 percent below what they were a year ago but from Zacks Investment Research. Word from the top: “Across the company, we are focused on safety, returning the 737 MAX to service, and earning and re-earning the trust and confidence of customers, regulators and the flying public,” Boeing CEO Dennis Muilenburg said in the company’s . “As we work through this challenging time for our customers, stakeholders and the company, our attention remains on driving excellence in quality and performance and running a healthy sustained growth business built on strong, long-term fundamentals.” Word on the Street: Boeing’s share price rose 0.36 percent to close at $375.46, apparently reflecting investor relief that the quarterly results weren’t worse. The stock’s value has fallen about 8.5 percent since the beginning of March, just before the fatal crash of an Ethiopian Airlines 737 MAX. Quarter highlights Boeing said the $1 billion in additional costs comes primarily from reducing the 737 production rate from 52 to 42 airplanes a month, a move that spreads fixed costs over a smaller number of planes. That toll could rise if the uncertainty surrounding the 737 MAX’s fate continues into the long term. Muilenburg said Boeing is making “steady process on the path to final certification for a software update for the 737 MAX.” That update is meant to address problems with an automatic flight control system that has been linked to the Ethiopia crash as well as last October’s fatal crash of a Lion Air jet in Indonesia. Muilenburg said the fix has been tested on more than 135 flights but didn’t give a timetable for winning certification from the Federal Aviation Administration. This week as saying certification could come as early as late May, potentially leading to a return to flight in mid-June. The process that led to the 737 MAX’s certification in 2017 is the subject of multiple investigations, including inquiries by the FAA, the Justice Department and the FBI, and a panel of experts convened by Boeing. But Muilenburg said “there was no surprise, or gap, or unknown here, or something that somehow slipped through a certification process.” Because of the focus on the 737 MAX situation, Muilenburg said Boeing hasn’t yet decided whether to go ahead with plans for a next-generation midsize airplane known as the New Mid-Market Airplane, the NMA or 797. But 777X production is moving ahead as planned, with the model’s first flight expected within months and the first 777X delivery due next year. Boeing Global Services, which was broken out as a separate business unit in 2017, brought in glowing financial results. First-quarter revenue increased 17 percent over the year-ago figure, to $4.6 billion. Boeing said the increase was driven primarily by higher volume across the portfolio, including , an aerospace parts distributor. Boeing Defense, Space & Security also served as a bright spot in the quarterly report: Revenues for that business unit amounted to $6.6 billion, which represents a 2 percent increase over the previous year’s first-quarter figures. The first seven KC-46 tankers were during the quarter — a milestone dimmed only slightly by . Boeing said its CST-100 Starliner space taxi has gone through “successful environmental testing,” calls for an uncrewed Starliner to make its first demonstration flight to the International Space Station for NASA no earlier than August, with the first crewed flight to follow by as early as late 2019.
Erez Barak, senior director of product for Microsoft’s AI Division, speaks at the Global Artificial Intelligence Conference in Seattle. (GeekWire Photo / Alan Boyle) Artificial intelligence can work wonders, but often it works in mysterious ways. Machine learning is based on the principle that a software program can analyze a huge set of data and fine-tune its algorithms to detect patterns and come up with solutions that humans may miss. That’s how Google DeepMind’s Alpha Go AI agent (and other games) well enough to beat expert players. But if programmers and users can’t figure out how AI algorithms came up with their results, that black-box behavior can be a cause for concern. It may become impossible to judge whether AI agents have picked up . That’s why terms such as transparency, explainability and interpretability are playing an increasing role in the AI ethics debate. The European Commission includes transparency and traceability among its , in line with the laid out in data-protection laws. The French government that powers the algorithms it uses. In the United States, the Federal Trade Commission’s has been charged with providing guidance on algorithmic transparency. Transparency figures in Microsoft CEO Satya Nadella’s as well — and , senior director of product for Microsoft’s AI Division, addressed the issue head-on today at the Global Artificial Intelligence Conference in Seattle. “We believe that transparency is a key,” he said. “How many features did we consider? Did we consider just these five? Or did we consider 5,000 and choose these five?” Barak noted that a is built right into Microsoft’s Azure Machine Learning service. “What it does is that it takes the model as an input and starts breaking it down,” he said. The model explanation can show which factors went into the computer model, and how they were weighted by the AI system’s algorithms. As a result, customers can better understand why, for instance, they were turned down for a mortgage, passed over for a job opening, or denied parole. AI developers can also use the model explanations to make their algorithms more “human.” For instance, it may be preferable to go with an algorithm that doesn’t fit a training set of data quite as well, but is more likely to promote fairness and avoid gender or racial bias. As AI applications become more pervasive, calls for transparency — perhaps enforced through government regulation — could well become stronger. And that runs the risk of exposing trade secrets hidden within a company’s intricately formulated algorithms, said , a partner at Seattle’s Perkins Coie law firm who specializes in trade regulations. “Algorithms tend to be things that are closely guarded. … That’s not something that you necessarily want to be transparent with the public or with your competitors about, so there is that fundamental tension,” Castillo said. “That’s more at issue in Europe than in the U.S., which has much, much, much stronger and aggressive enforcement.” Microsoft has already taken a strong stance on responsible AI — to the point that the company . After his talk, Barak told GeekWire that Azure Machine Learning’s explainability feature could be used as an open-source tool to look inside the black box and verify that an AI algorithm doesn’t perpetuate all-too-human injustices. Over time, will the software industry or other stakeholders develop a set of standards or a “seal of approval” for AI algorithms? “We’ve seen that in things like security. Those are the kinds of thresholds that have been set. I’m pretty sure we’re heading in that direction as well,” Barak said. “The idea is to give everyone the visibility and capability to do that, and those standards will develop, absolutely.”