Slashdot Mirror

MIT Technology Review reports: David Mitchell, a lanky, soft-spoken atmospheric physicist, believes frigid clouds in the upper troposphere may offer one of our best fallback plans for combating climate change... Fleets of large drones would crisscross the upper latitudes of the globe during winter months, sprinkling the skies with tons of extremely fine dust-like materials every year. If Mitchell is right, this would produce larger ice crystals than normal, creating thinner cirrus clouds that dissipate faster. "That would allow more radiation into space, cooling the earth," Mitchell says...

Increasingly grim climate projections have convinced a growing number of scientists it's time to start conducting experiments to find out what might work. In addition, an impressive list of institutions including Harvard University, the Carnegie Council, and the University of California, Los Angeles, have recently established research initiatives... By this time next year, Harvard professors David Keith and Frank Keutsch hope to launch a high-altitude balloon from a site in Tucson, Arizona. This will mark the beginning of a research project to explore the feasibility and risks of an approach known as solar radiation management. The basic idea is that spraying materials into the stratosphere could help reflect more heat back into space, mimicking a natural cooling phenomenon that occurs after volcanoes blast tens of millions of tons of sulfur dioxide into the sky.
"I don't really know what the answer is," says a former associate director at Lawrence Livermore National Laboratory. "But I do believe we need to keep saying what the truth is, and the truth is, we might need it."

MIT Technology Review reports: David Mitchell, a lanky, soft-spoken atmospheric physicist, believes frigid clouds in the upper troposphere may offer one of our best fallback plans for combating climate change... Fleets of large drones would crisscross the upper latitudes of the globe during winter months, sprinkling the skies with tons of extremely fine dust-like materials every year. If Mitchell is right, this would produce larger ice crystals than normal, creating thinner cirrus clouds that dissipate faster. "That would allow more radiation into space, cooling the earth," Mitchell says...

Increasingly grim climate projections have convinced a growing number of scientists it's time to start conducting experiments to find out what might work. In addition, an impressive list of institutions including Harvard University, the Carnegie Council, and the University of California, Los Angeles, have recently established research initiatives... By this time next year, Harvard professors David Keith and Frank Keutsch hope to launch a high-altitude balloon from a site in Tucson, Arizona. This will mark the beginning of a research project to explore the feasibility and risks of an approach known as solar radiation management. The basic idea is that spraying materials into the stratosphere could help reflect more heat back into space, mimicking a natural cooling phenomenon that occurs after volcanoes blast tens of millions of tons of sulfur dioxide into the sky.
"I don't really know what the answer is," says a former associate director at Lawrence Livermore National Laboratory. "But I do believe we need to keep saying what the truth is, and the truth is, we might need it."

Last year an experimental vehicle, developed by researchers at the chip maker Nvidia was unlike anything demonstrated by Google, Tesla, or General Motors. The car didn't follow a single instruction provided by an engineer or programmer. Instead, it relied entirely on an algorithm that had taught itself to drive by watching a human do it. Getting a car to drive this way was an impressive feat. But it's also a bit unsettling, since it isn't completely clear how the car makes its decisions, argues an article on MIT Technology Review. From the article: The mysterious mind of this vehicle points to a looming issue with artificial intelligence. The car's underlying AI technology, known as deep learning, has proved very powerful at solving problems in recent years, and it has been widely deployed for tasks like image captioning, voice recognition, and language translation. There is now hope that the same techniques will be able to diagnose deadly diseases, make million-dollar trading decisions, and do countless other things to transform whole industries. But this won't happen -- or shouldn't happen -- unless we find ways of making techniques like deep learning more understandable to their creators and accountable to their users. Otherwise it will be hard to predict when failures might occur -- and it's inevitable they will. That's one reason Nvidia's car is still experimental.

schwit1 shares an excerpt from a report via MIT Technology Review: No one really knows how the most advanced algorithms do what they do. That could be a problem. Last year, a strange self-driving car was released onto the quiet roads of Monmouth County, New Jersey. The experimental vehicle, developed by researchers at the chip maker Nvidia, didn't look different from other autonomous cars, but it was unlike anything demonstrated by Google, Tesla, or General Motors, and it showed the rising power of artificial intelligence. The car didn't follow a single instruction provided by an engineer or programmer. Instead, it relied entirely on an algorithm that had taught itself to drive by watching a human do it. Getting a car to drive this way was an impressive feat. But it's also a bit unsettling, since it isn't completely clear how the car makes its decisions. Information from the vehicle's sensors goes straight into a huge network of artificial neurons that process the data and then deliver the commands required to operate the steering wheel, the brakes, and other systems. The result seems to match the responses you'd expect from a human driver. But what if one day it did something unexpected -- crashed into a tree, or sat at a green light? As things stand now, it might be difficult to find out why. The system is so complicated that even the engineers who designed it may struggle to isolate the reason for any single action. And you can't ask it: there is no obvious way to design such a system so that it could always explain why it did what it did. The mysterious mind of this vehicle points to a looming issue with artificial intelligence. The car's underlying AI technology, known as deep learning, has proved very powerful at solving problems in recent years, and it has been widely deployed for tasks like image captioning, voice recognition, and language translation. There is now hope that the same techniques will be able to diagnose deadly diseases, make million-dollar trading decisions, and do countless other things to transform whole industries. But this won't happen -- or shouldn't happen -- unless we find ways of making techniques like deep learning more understandable to their creators and accountable to their users. Otherwise it will be hard to predict when failures might occur -- and it's inevitable they will. That's one reason Nvidia's car is still experimental.

schwit1 adds: "To be fair, we don't really understand how natural intelligence works, either."

"Nobody wants their data spread far and wide," write two associate editors at MIT Technology Review, "but the FCC's rules were an inconsistent solution to a much larger problem." An anonymous reader writes: They point out the rules passed in October "weren't even yet in effect," but more importantly -- they only would've applied to ISPs. "[T]he reality is that the U.S. doesn't have a baseline law that governs online privacy," and the truth is, it never did. "The FCC's new privacy rules would have been dramatic, to be sure -- but they would only have addressed one piece of the problem, leaving companies like Facebook and Google free to continue doing much the same thing.
While the repeal still needs approval in the U.S. House of Representatives and the president's signature, their article argues that what's really needed is "a more consistent approach to privacy."

An anonymous reader quotes a report from MIT Technology Review: A streamlined set of goals for reducing carbon emissions could simplify the way nations approach the quest to reduce human impact on the planet. A group of European researchers have a refreshingly straightforward solution that they call a carbon law -- or, as the Guardian has coined it, a "Moore's law for carbon." The overarching goal is simple: globally, we must halve carbon dioxide emissions every decade. That's essentially it. The rule would ideally be applied "to all sectors and countries at all scales," and would encourage "bold action in the short term." Dramatic changes would naturally have to occur as a result -- from quick wins like carbon taxes and energy efficiency regulations, to longer-term policies like phasing out combustion-engine cars and carbon-neutral building regulations. If policy makers followed the carbon law, adoption of renewables would continue its current pace of doubling energy production every 5.5 years, and carbon dioxide sequestration technologies would need to ramp up in order for the the planet to reach net-zero emissions by the middle of the century, say the researchers. Along the way, coal use would end as soon as 2030 and oil use by 2040. There are, clearly, issues with the idea, not least being the prospect of convincing every nation to commit to such a vision. The very simplicity that makes the idea compelling can also be used as a point of criticism: Can such a basic rule ever hope to define practical ideas as to how to change the world's energy production and consumption? The study has been published in the journal Science.

"The first quantum computer to start paying its way with useful work in the real world looks likely to do so by helping chemists," writes MIT Technology Review, "trying to do things like improve batteries or electronics." An anonymous reader quotes their report: So far, simulating molecules and reactions is the use case for early, small quantum computers sketched out in most detail by researchers developing the new kind of algorithms needed for such machines... "From the point of view of what is theoretically proven, chemistry is ahead," says Scott Crowder, chief technology officer for the IBM division that today sells hardware including supercomputers and hopes to add cloud-hosted quantum computers to its product line-up in the next few years...

Researchers have long used simulations of molecules and chemical reactions to aid research into things like new materials, drugs, or industrial catalysts. The tactic can reduce time spent on physical experiments and scientific dead ends, and it accounts for a significant proportion of the workload of the world's supercomputers. Yet the payoffs are limited because even the most powerful supercomputers cannot perfectly re-create all the complex quantum behaviors of atoms and electrons in even relatively small molecules, says Alan Aspuru-Guzik, a chemistry professor at Harvard. He's looking forward to the day simulations on quantum computers can accelerate his research group's efforts to find new light-emitting molecules for displays, for example, and batteries suitable for grid-scale energy storage.
Microsoft is already focusing on chemistry and materials science in its quantum algorithm effort, saying a hybrid system combining conventional computers with a small quantum computer "has great promise for studying molecules." Meanwhile, the article argues that breaking encryption, "although a genuine threat, is one of the most distant applications of the technology, because the algorithms involved would require an extremely large quantum processor."

"The first quantum computer to start paying its way with useful work in the real world looks likely to do so by helping chemists," writes MIT Technology Review, "trying to do things like improve batteries or electronics." An anonymous reader quotes their report: So far, simulating molecules and reactions is the use case for early, small quantum computers sketched out in most detail by researchers developing the new kind of algorithms needed for such machines... "From the point of view of what is theoretically proven, chemistry is ahead," says Scott Crowder, chief technology officer for the IBM division that today sells hardware including supercomputers and hopes to add cloud-hosted quantum computers to its product line-up in the next few years...

Researchers have long used simulations of molecules and chemical reactions to aid research into things like new materials, drugs, or industrial catalysts. The tactic can reduce time spent on physical experiments and scientific dead ends, and it accounts for a significant proportion of the workload of the world's supercomputers. Yet the payoffs are limited because even the most powerful supercomputers cannot perfectly re-create all the complex quantum behaviors of atoms and electrons in even relatively small molecules, says Alan Aspuru-Guzik, a chemistry professor at Harvard. He's looking forward to the day simulations on quantum computers can accelerate his research group's efforts to find new light-emitting molecules for displays, for example, and batteries suitable for grid-scale energy storage.
Microsoft is already focusing on chemistry and materials science in its quantum algorithm effort, saying a hybrid system combining conventional computers with a small quantum computer "has great promise for studying molecules." Meanwhile, the article argues that breaking encryption, "although a genuine threat, is one of the most distant applications of the technology, because the algorithms involved would require an extremely large quantum processor."

"To many in Silicon Valley, the brain looks like an unconquered frontier whose importance dwarfs any achievement made in computing or the Web," including Bryan Johnson, the founder of Braintree online payments, and Elon Musk. An anonymous reader quotes MIT Technology Review: Johnson is effectively jumping on an opportunity created by the Brain Initiative, an Obama-era project which plowed money into new schemes for recording neurons. That influx of cash has spurred the formation of several other startups, including Paradromics and Cortera, also developing novel hardware for collecting brain signals. As part of the government brain project, the defense R&D agency DARPA says it is close to announcing $60 million in contracts under a program to create a "high-fidelity" brain interface able to simultaneously record from one million neurons (the current record is about 200) and stimulate 100,000 at a time...

According to neuroscientists, several figures from the tech sector are currently scouring labs across the U.S. for technology that might fuse human and artificial intelligence. In addition to Johnson, Elon Musk has been teasing a project called "neural lace," which he said at a 2016 conference will lead to "symbiosis with machines." And Mark Zuckerberg declared in a 2015 Q&A that people will one day be able to share "full sensory and emotional experiences," not just photos. Facebook has been hiring neuroscientists for an undisclosed project at Building 8, its secretive hardware division.
Elon Musk complains that the current speeds for transferring signals from brains are "ridiculously slow".

Curated advertising may be coming to a highway near you. "A startup called Synaps Labs has brought it to the physical world by combining high-speed cameras set up a distance ahead of the billboard (about 180 meters) to capture images of cars," reports MIT Technology Review. "Its machine-learning system can recognize in those images the make and model of the cars an advertiser wants to target. A bidding system then selects the appropriate advertising to put on the billboard as that car passes." From the report: There is a lot an advertiser can tell about you from the car you drive, says Synaps. Indeed, recent research from a group of university researchers and led by Stanford found that -- using machine vision and deep learning -- analyzing the make, model, and year of vehicles visible in Google Street View could accurately estimate income, race, and education level of a neighborhood's residents, and even whether a city is likely to vote Democrat or Republican. Synaps's business model is to sell its services to the owners of digital billboards. Digital billboard advertising rotates, and more targeted advertising can rotate more often, allowing operators to sell more ads. According to Synaps, a targeted ad shown 8,500 times in one month will reach the same number of targeted drivers (approximately 22,000) as a typical ad shown 55,000 times. In Russia, Synaps expects to be operating on 20 to 50 billboards this year. The company is also planning a test in the U.S. this summer, where there are roughly 7,000 digital billboards, a number growing at 15 percent a year, according to the company. (By contrast, there are 370,000 conventional billboards.) With a row of digital billboards along a road, they could roll the ads as the cars move along, making billboard advertising more like the storytelling style of television and the Internet, says Synaps's cofounder Alex Pustov.

Curated advertising may be coming to a highway near you. "A startup called Synaps Labs has brought it to the physical world by combining high-speed cameras set up a distance ahead of the billboard (about 180 meters) to capture images of cars," reports MIT Technology Review. "Its machine-learning system can recognize in those images the make and model of the cars an advertiser wants to target. A bidding system then selects the appropriate advertising to put on the billboard as that car passes." From the report: There is a lot an advertiser can tell about you from the car you drive, says Synaps. Indeed, recent research from a group of university researchers and led by Stanford found that -- using machine vision and deep learning -- analyzing the make, model, and year of vehicles visible in Google Street View could accurately estimate income, race, and education level of a neighborhood's residents, and even whether a city is likely to vote Democrat or Republican. Synaps's business model is to sell its services to the owners of digital billboards. Digital billboard advertising rotates, and more targeted advertising can rotate more often, allowing operators to sell more ads. According to Synaps, a targeted ad shown 8,500 times in one month will reach the same number of targeted drivers (approximately 22,000) as a typical ad shown 55,000 times. In Russia, Synaps expects to be operating on 20 to 50 billboards this year. The company is also planning a test in the U.S. this summer, where there are roughly 7,000 digital billboards, a number growing at 15 percent a year, according to the company. (By contrast, there are 370,000 conventional billboards.) With a row of digital billboards along a road, they could roll the ads as the cars move along, making billboard advertising more like the storytelling style of television and the Internet, says Synaps's cofounder Alex Pustov.

holy_calamity writes: University of Nebraska student Brevan Jorgenson swapped the rear-view mirror in his 2016 Honda Civic for a home-built device called a Neo, which can steer the vehicle and follow traffic on the highway. Jorgenson used hardware designs and open-source software released by Comma, a self-driving car startup that decided to give away its technology for free last year after receiving a letter asking questions about its functionality from the National Highway Traffic Safety Administration (NHTSA). Jorgenson is just one person in a new hacker community trying to upgrade their cars using Comma's technology. "A Neo is built from a OnePlus 3 smartphone equipped with Comma's now-free Openpilot software, a circuit board that connects the device to the car's electronics, and a 3-D-printed case," reports MIT Technology Review. The report notes that Neodriven, a startup based in Los Angeles, has recently started selling a pre-built Neo device that works with Comma's Openpilot software, but it costs $1,495.

moon_unit2 quotes a report from MIT Technology Review: You can, for instance, train a deep-learning algorithm to recognize a cat with a cat-fancier's level of expertise, but you'll need to feed it tens or even hundreds of thousands of images of felines, capturing a huge amount of variation in size, shape, texture, lighting, and orientation. It would be lot more efficient if, a bit like a person, an algorithm could develop an idea about what makes a cat a cat from fewer examples. A Boston-based startup called Gamalon has developed technology that lets computers do this in some situations, and it is releasing two products Tuesday based on the approach. Gamalon uses a technique that it calls Bayesian program synthesis to build algorithms capable of learning from fewer examples. Bayesian probability, named after the 18th century mathematician Thomas Bayes, provides a mathematical framework for refining predictions about the world based on experience. Gamalon's system uses probabilistic programming -- or code that deals in probabilities rather than specific variables -- to build a predictive model that explains a particular data set. From just a few examples, a probabilistic program can determine, for instance, that it's highly probable that cats have ears, whiskers, and tails. As further examples are provided, the code behind the model is rewritten, and the probabilities tweaked. This provides an efficient way to learn the salient knowledge from the data.

wisebabo writes: Say goodbye to our little whiskered friends! There is an effort to wipe out not just any species, (there's been discussions to wipe out the mosquitos that carry Malaria), but a mammal. Specifically the house mouse which, along with other invasive species introduced by Westerners, have ravaged New Zealand's ecosystem. (Amongst other things they've rendered extinct many of the flightless birds there). They'll try using the "gene drive" in mammals, which is a new genetic weapon made possible by the editing system CRISPR-Cas9. Basically, it'll make all of the children of the genetically engineered mice male and then all of their children male and so on. This'll continue until there are no females left and the population will crash. If this is successful, they want to use this technique on other species until all of the predators on New Zealand are wiped out.

Goldman Sach's New York headquarters has replaced 600 of its traders with 200 computer engineers over the last two decades or so, thanks to automated trading programs. (Though, the effort to do so has accelerated over the past five years.) "Marty Chavez, the company's deputy chief financial officer and former chief information officer, explained all this to attendees at a symposium on computer's impact on economic activity held by Harvard's Institute for Applied Computational Science last month," reports MIT Technology Review. From their report: The experience of its New York traders is just one early example of a transformation of Goldman Sachs, and increasingly other Wall Street firms, that began with the rise in computerized trading, but has accelerated over the past five years, moving into more fields of finance that humans once dominated. Chavez, who will become chief financial officer in April, says areas of trading like currencies and even parts of business lines like investment banking are moving in the same automated direction that equities have already traveled. Today, nearly 45 percent of trading is done electronically, according to Coalition, a U.K. firm that tracks the industry. In addition to back-office clerical workers, on Wall Street machines are replacing a lot of highly paid people, too. Complex trading algorithms, some with machine-learning capabilities, first replaced trades where the price of what's being sold was easy to determine on the market, including the stocks traded by Goldman's old 600. Now areas of trading like currencies and futures, which are not traded on a stock exchange like the New York Stock Exchange but rather have prices that fluctuate, are coming in for more automation as well. To execute these trades, algorithms are being designed to emulate as closely as possible what a human trader would do, explains Coalition's Shahani. Goldman Sachs has already begun to automate currency trading, and has found consistently that four traders can be replaced by one computer engineer, Chavez said at the Harvard conference. Some 9,000 people, about one-third of Goldman's staff, are computer engineers.

"Now when a machine is compromised it takes days or weeks for someone to notice and then days or weeks -- or never -- until a patch is put out," says Carnegie Mellon professor David Brumley. "Imagine a world where the first time a hacker exploits a vulnerability he can only exploit one machine and then it's patched." An anonymous reader quotes MIT Technology Review: Last summer the Pentagon staged a contest in Las Vegas in which high-powered computers spent 12 hours trying to hack one another in pursuit of a $2 million purse. Now Mayhem, the software that won, is beginning to put its hacking skills to work in the real world... Teams entered software that had to patch and protect a collection of server software, while also identifying and exploiting vulnerabilities in the programs under the stewardship of its competitors... ForAllSecure, cofounded by Carnegie Mellon professor David Brumley and two of his PhD students, has started adapting Mayhem to be able to automatically find and patch flaws in certain kinds of commercial software, including that of Internet devices such as routers.

Tests are underway with undisclosed partners, including an Internet device manufacturer, to see if Mayhem can help companies identify and fix vulnerabilities in their products more quickly and comprehensively. The focus is on addressing the challenge of companies needing to devote considerable resources to supporting years of past products with security updates... Last year, Brumley published results from feeding almost 2,000 router firmware images through some of the techniques that powered Mayhem. Over 40%, representing 89 different products, had at least one vulnerability. The software found 14 previously undiscovered vulnerabilities affecting 69 different software builds. ForAllSecure is also working with the Department of Defense on ideas for how to put Mayhem to real world use finding and fixing vulnerabilities.

Pong's creator is now "a grizzled guy in his mid-70s" who believes there's a market for people who'd prefer to try out virtual reality headsets at videogame arcades. An anonymous reader quotes MIT Technology Review: In 1972, Atari founder Nolan Bushnell invented Pong, a version of table tennis that, in many ways, launched the video-game industry. Forty-five years later, Bushnell is using that same simple game to test the waters for virtual-reality arcade gaming. Bushnell's latest venture is a company called Modal VR, which is building its own wireless virtual-reality headsets and games that it plans to roll out in places like arcades, malls, and movie theaters in the coming months.
Bushnell's company has built three games -- a fighting game called Mythic Combat and Project Zenith a first-person shooter set in outer space. (More than 16 players can gather in the same virtual space.) Their third game, a VR adaptation of Pong "was originally put together as a joke, in homage to Bushnell's past -- but the company decided to use the simple two-player game anyway to demonstrate what it's working on at the World's Fair Nano technology fair in San Francisco in late January."

The article describes players who "donned a prototype bulky black headset and played Pong in virtual reality, running from side to side to control the game's simple white paddles -- which a smiling Bushnell said was fitting because "we're at the Pong stage of VR."

Neuroscientists have designed a brain-reading device to hold simple conversations with "locked-in" patients that promises to transform the lives of people who are too disabled to communicate. Details of four patients who were able to communicate using what is being touted as a groundbreaking system were made public this week. From a report on MIT Technology Review: Now researchers in Europe say they've found out the answer after using a brain-computer interface to communicate with four people completely locked in after losing all voluntary movement due to Lou Gehrig's disease, or amyotrophic lateral sclerosis. In response to the statement "I love to live" three of the four replied yes. They also said yes when asked "Are you happy?" Designed by neuroscientist Niels Birbaumer, now at the Wyss Center for Bio and Neuroengineering in Geneva, the brain-computer interface fits on a person's head like a swimming cap and measures changes in electrical waves emanating from the brain and also blood flow using a technique known as near-infrared spectroscopy. To verify the four could communicate, Birbaumer's team asked patients, over the course of about 10 days of testing, to respond yes or no to statements such as "You were born in Berlin" or "Paris is the capital of Germany" by modulating their thoughts and altering the blood-flow pattern. The answers relayed through the system were consistent about 70 percent of the time, substantially better than chance.

An anonymous reader writes: After four years of development, Sansar, the new virtual reality world from Second Life's creators will arrive later this year on Oculus Rift and HTC Vive headsets. "It is trying to solve some of the big problems that plagued Second Life for years," reports MIT Technology Review, "such as that most users come in through what is essentially a front door and have a hard time finding things to do once they get in... In the demos I tried, I navigated via an atlas that shows a simple clickable thumbnail image of each destination along with its name."

But it still has to prove itself to users like John Artz, an associate professor at George Washington University who once taught a class about using Second Life for business applications. Artz "thinks Sansar will still suffer from the same fundamental issue that dogs Second Life: while the technology behind it is good, he says, it just got boring after a while."
Second Life still has 800,000 monthly users -- and in Sansar, virtual land will be cheaper, with Linden Lab concentrating "more on making money from selling virtual objects like clothing for avatars and furniture."

A tiny new Robotic Retinal Dissection Device -- nicknamed "R2D2" -- can crawl into an incision in the eye and lift a membrane no more than a hundredth of a millimeter. "The cables that enable the robot to navigate are each 110 microns across, a little over the diameter of a human hair," reports the MIT Technology Review. The robot is controlled by a joystick (while providing a live camera feed to the ophthalmologist). In September an Oxford professor used it to perform the first operation inside the human eye, and since then five more patients have undergone robot-assisted operations at an Oxford hospital. In one procedure, a gene-therapy virus that stops retinal degeneration "was planted on the retina itself, a procedure only made possible by R2D2's unprecedented precision."
Robotic surgery is already happening. The article points out that Da Vinci, an elephant-sized surgical robot that repairs heart valves, "has operated on more than three million patients around the world." But the Oxford professor believes these tiny eye robots "will open the door to new operations for which the human hand does not have the necessary control and precision."

The embattled blood-testing company Theranos is laying off 155 people, about 41 percent of its workforce, as it struggles to recover from the backlash generated when the company failed to provide accurate results to patients using its proprietary blood test technology. The job cuts announced today are similar to the cuts announced last year in October, when the company said it would shut down its blood-testing facilities and shrink its workforce by more than 40%. CNBC reports: The start-up will let go of its workers after months of regulatory setbacks as well as lawsuits and scrutiny. That would leave 220 workers to focus on its business plans, primarily its blood testing product called the miniLab. "These are always the most difficult decisions; however, this move allows Theranos to marshal its resources most efficiently and effectively," Theranos said in a statement.

schwit1 quotes a report from MIT Technology Review: In the southern Indian city of Tuticorin, locals are unlikely to suffer from a poorly risen cake. That's because a coal-fired thermal power station in the area captures carbon dioxide and turns it into baking soda. Carbon capture schemes are nothing new. Typically, they use a solvent, such as amine, to catch carbon dioxide and prevent it from escaping into the atmosphere. From there, the CO2 can either be stored away or used. But the Guardian reports that a system installed in the Tuticorin plant uses a new proprietary solvent developed by the company Carbon Clean Solutions. The solvent is reportedly just slightly more efficient than those used conventionally, requiring a little less energy and smaller apparatus to run. The collected CO2 is used to create baking soda, and it claims that as much as 66,000 tons of the gas could be captured at the plant each year. Its operators say that the marginal gain in efficiency is just enough to make it feasible to run the plant without a subsidy. In fact, it's claimed to be the first example of an unsubsidized industrial plant capturing CO2 for use. schwit1 notes: "A 'climate change' project that doesn't involve taxpayer dollars? Is that even allowed?"

schwit1 quotes a report from MIT Technology Review: Mining companies are rolling out autonomous trucks, drills, and trains, which will boost efficiency but also reduce the need for human employees. Rio Tinto uses driverless trucks provided by Japan's Komatsu. They find their way around using precision GPS and look out for obstacles using radar and laser sensors. The company's driverless trucks have proven to be roughly 15 percent cheaper to run than vehicles with humans behind the wheel -- a significant saving since haulage is by far a mine's largest operational cost. Trucks that drive themselves can spend more time working because software doesn't need to stop for shift changes or bathroom breaks. They are also more predictable in how they do things like pull up for loading. "All those places where you could lose a few seconds or minutes by not being consistent add up," says Rob Atkinson, who leads productivity efforts at Rio Tinto. They also improve safety. The driverless locomotives, due to be tested extensively next year and fully deployed by 2018, are expected to bring similar benefits. They also anticipate savings on train maintenance, because software can be more predictable and gentle than any human in how it uses brakes and other controls. Diggers and bulldozers could be next to be automated.

Uber has launched a website for a service called Uber Freight. While there are little details about the company's expansion from ride-hailing, Uber Freight is meant to prepare the world for autonomous delivery trucks, according to Inverse. From the report: Uber acquired a startup called Otto, which planned to bring the first self-driving trucks to market, in August. Since then the company has used its trucks to deliver 50,000 cans of beer and hundreds of Christmas trees in San Francisco. This new service won't use those trucks, at least not at the beginning. Instead it will function much like Uber's existing platform: Some people will sign up to drive items across the country, and others will join so they can send packages without having to sign a contract with established shipping companies. The service will likely bring "surge pricing" to trucking, too. Uber Freight could also help Otto's trucks by using data gathered from drivers on the platform. This would allow the self-driving vehicles to learn from experienced people while regulators figure out how to govern autonomous trucks and the technology catches up to all of the promises made by its creators. Uber Freight's launch coincides with growing interest in trucking from many tech companies. Nikola Motor Company wants to use tech to make trucking more environmentally friendly and appealing to millennials; Tesla's working on self-driving trucks; the list could go on. Uber told Inverse it's going to wait until the new year to elaborate on how the system works. "We don't have any new information to share at the moment," a spokesperson said, "but hope to in the new year so please do stay in touch." It looks like the future of trucking -- or at least one potential future -- is going to take a little while longer to make its debut.

Reader Joe_NoOne writes: Like TV, social media now increasingly entertains us, and even more so than television it amplifies our existing beliefs and habits. It makes us feel more than think, and it comforts more than challenges. The result is a deeply fragmented society, driven by emotions, and radicalized by lack of contact and challenge from outside. This is why Oxford Dictionaries designated "post-truth" as the word of 2016: an adjective "relating to circumstances in which objective facts are less influential in shaping public opinion than emotional appeals." Traditional television still entails some degree of surprise. What you see on television news is still picked by human curators, and even though it must be entertaining to qualify as worthy of expensive production, it is still likely to challenge some of our opinions (emotions, that is). Social media, in contrast, uses algorithms to encourage comfort and complaisance, since its entire business model is built upon maximizing the time users spend inside of it. Who would like to hang around in a place where everyone seems to be negative, mean, and disapproving? The outcome is a proliferation of emotions, a radicalization of those emotions, and a fragmented society. This is way more dangerous for the idea of democracy founded on the notion of informed participation. Now what can be done? Certainly the explanation for Trump's rise cannot be reduced to a technology- or media-centered argument. The phenomenon is rooted in more than that; media or technology cannot create; they can merely twist, divert, or disrupt. Without the growing inequality, shrinking middle class, jobs threatened by globalization, etc. there would be no Trump or Berlusconi or Brexit. But we need to stop thinking that any evolution of technology is natural and inevitable and therefore good. For one thing, we need more text than videos in order to remain rational animals. Typography, as Postman describes, is in essence much more capable of communicating complex messages that provoke thinking. This means we should write and read more, link more often, and watch less television and fewer videos -- and spend less time on Facebook, Instagram, and YouTube.

Speculations have turned out be true. The Chinese government is now testing systems that will be used to create digital records of citizens' social and financial behavior. In turn, these will be used to create a so-called social credit score, which will determine whether individuals have access to services, from travel and education to loans and insurance cover. Some citizens -- such as lawyers and journalists -- will be more closely monitored. From a report on MIT Technology Review: Planning documents apparently describe the system as being created to "allow the trustworthy to roam everywhere under heaven while making it hard for the discredited to take a single step." The Journal claims that the system will at first log "infractions such as fare cheating, jaywalking and violating family-planning rules" but will be expanded in the future -- potentially even to Internet activity. Some aspects of the system are already in testing, but there are some challenges to implementing such a far-reaching apparatus. It's difficult to centralize all that data, check it for accuracy, and process it, for example -- let alone feed it back into the system to control everyday life. And China has data from 1.4 billion people to handle.

Drivers stuck in traffic in Mexico city are lately seeing a fleet of sign-toting drones buzzing at them, saying (in Spanish) "Driving by yourself? This is why you can never see the volcanoes." (It's a reference to the smog that often hovers over the city and obscures two nearby peaks.) Turns out, it's an ad for UberPOOL, part of Uber's big push into markets across Latin America. From an MIT Technology Review article: Uber already does more business in Mexico City than any other city it operates in, and Brazil is its third-largest market after the U.S. and India. Uber sees Latin American countries as generally easier targets for expansion than either of its top two markets.Umm, I get that Uber has self-driving cars now in Pittsburgh, but they don't fly (at least as of now). So wouldn't they be stuck in the traffic as well?

An anonymous Slashdot reader writes: The recently-ratified Paris Climate Accord calls on countries to keep the rise in average global temperatures under 2 degrees Celsius (a threshold which would bring extreme weather, water shortages and reduced agricultural production). But a recent article on Vox warns that "the world has to zero out net carbon emissions...for a good chance of avoiding 2 degrees, by around 2065. After that, emissions have to go negative... We are betting our species' future on our ability to bury carbon."

That's why everyone's watching the W.A. Parish Generating Station in Texas, which came online this week -- on schedule, and under budget. "The plant will use a newly installed system to capture 90 percent of the carbon dioxide created during combustion."
Alas, Slashdot reader Dan Drollette brings bad news from the Bulletin of the Atomic Scientists: To fight climate change with carbon capture and storage technology, we'd have to complete one new carbon capture facility every working day for the next 70 years. It's better to switch to a diet of energy conservation, efficiency, and renewables, rather than rely on this technology as a kind of emergency planetary liposuction.

Sperm bank? There's an app for that. The largest sperm bank in the United Kingdom -- the London Sperm Bank -- has released an official app that aims to "modernize the process of hooking prospective parents up with the biological material they need to make it happen," according to MIT Technology Review: The app is essentially just a mobile version of the filtered search function the London Sperm Bank offers on its website. But in doing something as simple as bringing its desktop services to mobile devices, the bank is making a play to further normalize reproductive technologies. The London Sperm Bank boasts that users will receive push notifications as soon as new donors are available, which could help speed things up for hopeful parents looking for a match. The road to conception can take years for people using reproductive technologies, so expediting any part of the process would be a welcome time-saver. But the bank has over 10,000 vials of sperm, so searching, even using filters, could still be a lengthy process. To combat this, the app also offers a wish list function that lets more focused users predetermine what they're looking for in a donor, and receive a notification when their criteria are met. The way the service works on mobile has been compared to Tinder, but there's actually no swiping involved. Its wish list function means it's more akin to apps like Anthology, which job seekers use to find their next career move. The report notes that, while there are other mobile sperm bank apps out there, the London Sperm Bank is the only one with several medical associations and the U.K. government's Human Fertilization and Embryology Authority on board. Also, the app is free to download, but the cost of ordering sperm is about $1,200 per order, which is the same as if you order through the London Sperm Bank catalogue.

Transportation is likely to surpass the electricity sector in 2016 as the largest source of carbon dioxide emissions in the United States, according to a new analysis of government data, MIT Technology reports. From the article: In 2008, the global financial crisis caused widespread declines in energy use. In the U.S., that coincided with the early stages of a large-scale shift away from coal toward cleaner-burning natural gas as a way to generate electricity. As a result, carbon dioxide emissions from the electricity sector have continued to decline from their 2007 peak, even as the economy has resumed growing. The trend line for the transportation sector is less encouraging. Transportation emissions have begun rising as the economy rebounds. John DeCicco at the University of Michigan Energy Institute, who wrote the study, attributes the rebound we've seen during the past four years to straightforward causes: economic recovery and more affordable fuel prices. Vehicle sales numbers have been rising for several years, in particular for trucks and SUVs, and people are traveling more miles.

An anonymous reader quotes a report from ScienceAlert: Quantum teleportation just moved out of the lab and into the real world, with two independent teams of scientists successfully sending quantum information across several kilometers of optical fiber networks in Calgary, Canada, and Hefei, China. Quantum teleportation relies on a strange phenomenon called quantum entanglement. Basically, quantum entanglement means that two particles are inextricably linked, so that measuring the state of one immediately affects the state of the other, no matter how far apart the two are -- which led Einstein to call entanglement "spooky action at a distance." In the latest experiments, both published in Nature Photonics (here and here), the teams had slightly different set-ups and results. But what they both had in common is the fact that they teleported their information across existing optical fiber networks -- which is important if we ever want to build useable quantum communication systems. To understand the experiments, Anil Ananthaswamy over at New Scientist nicely breaks it down like this: picture three people involved -- Alice, Bob, and Charlie. Alice and Bob want to share cryptographic keys, and to do that, they need Charlie's help. Alice sends a particle to Charlie, while Bob entangles two particles and sends just one of them to Charlie. Charlie then measures the two particles he's received from each of them, so that they can no longer be differentiated -- and that results in the quantum state of Alice's particle being transferred to Bob's entangled particle. So basically, the quantum state of Alice's particle eventually ends up in Bob's particle, via a way station in the form of Charlie. The Canadian experiment followed this same process, and was able to send quantum information over 6.2 km of Calgary's fiber optic network that's not regularly in use.

"Keeping your enemies offline can cripple their chances of overthrowing you," reports the MIT Technology Review. Slashdot reader schwit1 quotes their article: Whether or not your ethnic group has political power is a crucial factor determining your access to the Internet, according to a new analysis. The effect varies from country to country, and is much less pronounced in democratic nations. But the study, published today in Science, suggests that besides censorship, another way national governments prevent opposing groups from organizing online is by denying them Internet access in the first place, says Nils Weidmann, a professor of political science at the University of Konstanz in Germany.
Researchers used a geolocation database to create a map showing subnetwork activity for a large volume of internet traffic, then compared it with geographic data for the world's ethnic groups. "They concluded that excluded groups had significantly lower access compared to the groups in power, and that this can't be explained by other economic or geographic factors (like living in rural vs. urban areas)... 'You don't have to censor if the opposition doesn't get access at all.' "

If scientists could send Zika-carrying mosquitoes into extinction, should they do it? Several science and business journals are now exploring the question, and Slashdot reader retroworks asks if scientists will ultimately target "not just the most deadly species of the animal, but all 12 species of human-biting mosquitoes in the world, responsible for 500,000 deaths per year." The headline on today's [paywalled] Wall Street Journal article begs the question, "Why Not Kill Them All...?" [M]ore business journals are exploring private sector investments to eradicate the species of mosquito entirely, [and] most articles seem to find extinction of the indoors-attacking, dengue fever- and malaria-spreading Aedes aegypti a tantalizing prospect...

The BBC weighed the approach more carefully, noting that mosquitoes make rain forests uninhabitable (and consequences of human populations in rain forests are usually disastrous)... Will capitalism make the itch of mosquito bites be forgotten... Forever?

If scientists could send Zika-carrying mosquitoes into extinction, should they do it? Several science and business journals are now exploring the question, and Slashdot reader retroworks asks if scientists will ultimately target "not just the most deadly species of the animal, but all 12 species of human-biting mosquitoes in the world, responsible for 500,000 deaths per year." The headline on today's [paywalled] Wall Street Journal article begs the question, "Why Not Kill Them All...?" [M]ore business journals are exploring private sector investments to eradicate the species of mosquito entirely, [and] most articles seem to find extinction of the indoors-attacking, dengue fever- and malaria-spreading Aedes aegypti a tantalizing prospect...

The BBC weighed the approach more carefully, noting that mosquitoes make rain forests uninhabitable (and consequences of human populations in rain forests are usually disastrous)... Will capitalism make the itch of mosquito bites be forgotten... Forever?

An anonymous reader quotes a report from MIT Technology Review: Satellite operator DigitalGlobe is teaming up with Amazon, the venture arm of the CIA, and NVIDIA to make computers watch the Earth from above and automatically map our roads, buildings, and piles of trash. MIT Technology Review reports: "In a joint project, DigitalGlobe today released satellite imagery depicting the whole of Rio de Janeiro to a resolution of 50 centimeters. The outlines of 200,000 buildings inside the city's roughly 1,900 square kilometers have been manually marked on the photos. The SpaceNet data set, as it is called, is intended to spark efforts to train machine-learning algorithms to interpret high-resolution satellite photos by themselves. DigitalGlobe says the SpaceNet data set should eventually include high-resolution images of half a million square kilometers of Earth, and that it will add annotations beyond just buildings. DigitalGlobe's data is much more detailed than publicly available satellite data such as NASA's, which typically has a resolution of tens of meters. Amazon will make the SpaceNet data available via its cloud computing service. Nvidia will provide tools to help machine-learning researchers train and test algorithms on the data, and CosmiQ Works, a division of the CIA's venture arm In-Q-Tel focused on space, is also supporting the project." "We need to develop new algorithms for this data," says senior vice president at DigitalGlobe, Tony Frazier. He goes on to say that health and aid programs are to benefit from software that is able to map roads, bridges and various other infrastructure. The CEO of Descartes Labs, Mark Johnson, a "startup that predicts crop yields from public satellite images," says the data that is collected "should be welcome to startups and researchers," according to MIT Technology Review. "Potential applications could include estimated economic output from activity in urban areas, or guiding city governments on how to improve services such as trash collections, he says."

MIT recently identified the states "at the greatest risk of having their voting process hacked". but added this week that "Maintaining the secrecy of ballots returned via the Internet is 'technologically impossible'..." Long-time Slashdot reader Presto Vivace quotes their article: That's according to a new report from Verified Voting, a group that advocates for transparency and accuracy in elections. A cornerstone of democracy, the secret ballot guards against voter coercion. But "because of current technical challenges and the unique challenge of running public elections, it is impossible to maintain the separation of voters' identities from their votes when Internet voting is used," concludes the report, which was written in collaboration with the Electronic Privacy Information Center and the anticorruption advocacy group Common Cause.
32 states are already offering some form of online voting, apparently prompting the creation of Verified Voting's new site, SecretBallotAtRisk.org.

MIT recently identified the states "at the greatest risk of having their voting process hacked". but added this week that "Maintaining the secrecy of ballots returned via the Internet is 'technologically impossible'..." Long-time Slashdot reader Presto Vivace quotes their article: That's according to a new report from Verified Voting, a group that advocates for transparency and accuracy in elections. A cornerstone of democracy, the secret ballot guards against voter coercion. But "because of current technical challenges and the unique challenge of running public elections, it is impossible to maintain the separation of voters' identities from their votes when Internet voting is used," concludes the report, which was written in collaboration with the Electronic Privacy Information Center and the anticorruption advocacy group Common Cause.
32 states are already offering some form of online voting, apparently prompting the creation of Verified Voting's new site, SecretBallotAtRisk.org.

An anonymous reader writes from a report via MIT Technology Review: Spencer Wheatley and Didier Sornette at ETH Zurich in Switzerland and Benjamin Sovacool at Aarhus University in Denmark have compiled the most comprehensive list of nuclear accidents ever created and used it to calculate the chances of future accidents. They say there is a 50:50 chance that a major nuclear disaster will occur somewhere in the world before 2050. "There is a 50 percent chance that a Chernobyl event (or larger) occurs in the next 27 years," they conclude. Since the International Atomic Energy Agency doesn't publish a historical database of the nuclear accidents it rates using the International Nuclear Event Scale, others, like Wheatley and co, have to compile their own list of accidents. They define an accident as "an unintentional incident or event at a nuclear energy facility that led to either one death (or more) or at least $50,000 in property damage." Each accident must have occurred during the generation, transmission, or distribution of nuclear energy, which includes accidents at mines, during transportation, or at enrichment facility, and so on. Fukushima was by far the most expensive accident in history at a cost of $166 billion, which is 60 percent of the total cost of all other nuclear accidents added together. Wheatley and co say their data suggests that the nuclear industry remains vulnerable to dragon king events, which are large unexpected events that are difficult to analyze because they follow a different statistical distribution, have unforeseen causes, and are few in number. "There is a 50% chance that a Fukushima event (or larger) occurs in the next 50 years," they say.

The New York Times reports: More than 100 Nobel laureates have a message for Greenpeace: Quit the G.M.O.-bashing. Genetically modified organisms and foods are a safe way to meet the demands of a ballooning global population, the 109 laureates wrote in a letter posted online and officially unveiled at a news conference on Thursday in Washington, D.C...

"Scientific and regulatory agencies around the world have repeatedly and consistently found crops and foods improved through biotechnology to be as safe as, if not safer than those derived from any other method of production," the group of laureates wrote. "There has never been a single confirmed case of a negative health outcome for humans or animals from their consumption. Their environmental impacts have been shown repeatedly to be less damaging to the environment, and a boon to global biodiversity."
Slashdot reader ArmoredDragon writes: As an echo to that comment, one of the key benefits of GMO is increased crop yield, which means a reduced need for deforestation to make way for farmland. GMO food such as Golden Rice, which improves the micronutrient content of rice, and Low Acrylamide Spuds, which are potatoes engineered to have reduced carcinogen content compared to their natural counterparts, can possibly solve many health problems that are inherent with consuming non-GMO produce. And for those concerned about patent-related issues, many of these patents have recently expired, which means anybody can freely grow them and sell the seeds without the need to pay any royalties.

schwit1 quotes the MIT Technology Review: The FBI has access to nearly 412 million photos in its facial recognition system—perhaps including the one on your driver's license. But according to a new government watchdog report, the bureau doesn't know how error-prone the system is, or whether it enhances or hinders investigations.

Since 2011, the bureau has quietly been using this system to compare new images, such as those taken from surveillance cameras, against a large set of photos to look for a match. That set of existing images is not limited to the FBI's own database, which includes some 30 million photos. The bureau also has access to face recognition systems used by law enforcement agencies in 16 different states, and it can tap into databases from the Department of State and the Department of Defense. And it is in negotiations with 18 other states to be able to search their databases, too...

Adding to the privacy concerns is another finding in the GAO report: that the FBI has not properly determined how often its system makes errors and has not "taken steps to determine whether face recognition systems used by external partners, such as states and federal agencies, are sufficiently accurate" to support investigations.

Security experts are claiming that iOS 10 preview, which Apple made available to enthusiasts last week, is not secure. iOS 10 is the latest version of Apple's mobile operating system. It will be available to standard customers later this year (likely around September). According to security experts, iOS 10's kernel is not encrypted. MIT News reports: Why Apple has suddenly opened up its code is unclear. One hypothesis in the security community is that, as author Jonathan Levin puts it, someone inside the company "screwed up royally." But he and security researcher Mathew Solnik both say there are reasons to think it may have been intentional. Encouraging more people to pore over the code could result in more bugs being disclosed to Apple so that it can fix them.

An anonymous reader writes: For the first time data may show that a human being has been successfully rejuvenated by gene therapy, claims Bioviva USA. "In September 2015, then 44 year-old CEO of Bioviva USA Inc. Elizabeth Parrish received two of her own company's experimental gene therapies: one to protect against loss of muscle mass with age, another to battle stem cell depletion responsible for diverse age-related diseases and infirmities." Bypassing America's FDA, the controversial therapies were described by the MIT Technology Review as "do-it-yourself medicine," saying it "raises ethical questions about how quickly such treatments should be tested in people and whether they ought to be developed outside the scrutiny of regulators." "The treatment was originally intended to demonstrate the safety of the latest generation of the therapies," reports Bioviva's web site. "But if early data is accurate, it is already the world's first successful example of telomere lengthening via gene therapy in a human individual."

MarkWhittington writes: The EmDrive, the so-called "impossible" space drive that uses no propellant, has roiled the aerospace world for the past several years ever since it was proposed by British aerospace engineer Robert Shawyer. In essence, the claim advanced by Shawyer and others is that if you bounced microwaves in a truncated cone, thrust would be produced out the open end. Most scientists have snorted at the idea, noting correctly that such a thing would violate physical laws. However, organizations as prestigious as NASA have replicated the same results, that prototypes of the EmDrive produces thrust. How does one reconcile the experimental results with the apparent scientific impossibility? MIT Technology Review suggested a reason why.

Reader holy_calamity writes: Pentagon research agency DARPA has funded the creation of a chip incapable of correct arithmetic, in the hope of making computers better at understanding the real world. A chip that can't guarantee that every calculation is perfect can still get good results on many problems but needs fewer circuits and burns less energy, says Joseph Bates, cofounder and CEO of Singular Computing. The S1 chip can process noisy data like video very efficiently because it doesn't need the extra circuits or operations needed to ensure every mathematical operation is performed perfectly. This summer DARPA will put five prototype computers, each equipped with 16 of the inexact S1 chips, online for researchers to experiment with.

An anonymous reader links to an article on MIT Technology Review: It's worth taking a minute to appreciate the sheer scale of what China is doing in solar right now. In 2015, the country added more than 15 gigawatts of new solar capacity, surpassing Germany as the world's largest solar power market. China now has 43.2 gigawatts of solar capacity, compared with38.4 gigawatts in Germany and 27.8 in the United States. According to new projections, it seems that trend is going to continue. Under its 13th Five Year Plan, China will nearly triple solar capacity by 2020, adding 15 to 20 gigawatts of solar capacity each year for the next five years, according to Nur Bekri, director of the National Energy Administration. That will bring the country's installed solar power to more than 140 gigawatts. To put that in context, world solar capacity topped 200 gigawatts last year and is expected to reach 321 gigawatts by the end of 2016.

moon_unit2 writes: Apparently, playing computer games might provide a shortcut to greater intelligence. MIT Technology Review has a story about researchers using virtual game environments to train deep neural networks to recognize real-world objects. It's an important idea because deep learning usually requires huge quantities of annotated data, which isn't always available. So researchers from Xerox Europe, led by Adrien Gaidon, showed that training a deep learning system on a photo-realistic street scene could enable it to identify cars on real roads. "The nice thing about virtual worlds is you can create any kind of scenario," Gaidon says. Perhaps video games could play a bigger role in the future of AI than anyone realized.

HughPickens.com writes: Ben Popper writes at The Verge that bitcoin's nightmare scenario has come to pass as the bitcoin network reached its capacity, causing transactions around the world to be massively delayed, and in some cases to fail completely. The average time to confirm a transaction has ballooned from 10 minutes to 43 minutes. Users are left confused and shops that once accepted Bitcoin are dropping out. For those who want the Bitcoin system to continue to grow and thrive, this is troubling. Merchants can't rely on digital transactions that can take minutes or hours to validate. A number of prominent voices in the Bitcoin community have been warning over the past year that the system needed to make fundamental changes to its core software code to avoid being overwhelmed by the continued growth of Bitcoin transactions. A schism has developed between the team in charge of the original codebase for Bitcoin, known as Core, and a rival faction pushing its own version of that open source code with a block size increase added in, known as Classic. "Many in the US Bitcoin community had hoped that hitting this crisis point — a network maxed out, transactions faltering — would result in closure, with miners quickly moving to adopt whichever chain proved more valuable to their economic interests," says Popper. "But so far the debate is dragging on without one side claiming a clear victory, leaving tens of thousands of consumer transactions stranded in limbo."

moon_unit2 writes: DJI's new Phantom 4 drone may be able to stop if there's an obstacles directly in front of it, but MIT Technology Review has a story about a much more sophisticated self-flying drone, from a startup called Skydio (basically using high-speed visual SLAM, which is no mean feat in such a tiny package). The company's prototype uses several video cameras to navigate around obstacles at high speeds through busy airspace. The technology could make consumer drones much harder to crash, and it could let drones do more complex surveillance tasks. Skydio, founded last year, has so far raised $25 million in funding in a round led by Andreessen Horowitz and Accel Partners.

schwit1 writes: Here's a tricky task. Pick a photograph from the web at random. Now try to work out where it was taken using only the image itself. If the image shows a famous building or landmark, such as the Eiffel Tower or Niagara Falls, the task is straightforward. But the job becomes significantly harder when the image lacks specific location cues or is taken indoors or shows a pet or food or some other detail. Nevertheless, humans are surprisingly good at this task. To help, they bring to bear all kinds of knowledge about the world such as the type and language of signs on display, the types of vegetation, architectural styles, the direction of traffic, and so on. Humans spend a lifetime picking up these kinds of geolocation cues. So it's easy to think that machines would struggle with this task. And indeed, they have. Today, that changes thanks to the work of Tobias Weyand, a computer vision specialist at Google, and a couple of pals. These guys have trained a deep-learning machine to work out the location of almost any photo using only the pixels it contains.

At MIT Technology Review, Rachel Metz lists four factors she believes will mean great advancements for virtual reality in the next year. More and better games; wider adoption of specialized cameras for capturing the deep imagery that immersive worlds require; specialized presentation techniques that supplement VR with physical cues like temperature or direction; and availability of better viewing hardware. That better hardware seems poised to take off. According to the article, Facebook-owned Oculus’s first consumer headset, Rift, is slated for release in the first quarter of the year, while the HTC Vive—a headset created by smartphone maker HTC and video-game company Valve—is set to be available to consumers in April. Sony, meanwhile, is building its own headset, called PlayStation VR, which the company says will be released in the first half of the year.