Month: January 2019

A quantum computer harnesses some of the almost-mystical phenomena of quantum mechanics to deliver huge leaps forward in processing power. Quantum machines promise to outstrip even the most capable of today’s—and tomorrow’s—supercomputers.

They won’t wipe out conventional computers, though. Using a classical machine will still be the easiest and most economical solution for tackling most problems. But quantum computers promise to power exciting advances in various fields, from materials science to pharmaceuticals research. Companies are already experimenting with them to develop things like lighter and more powerful batteries for electric cars, and to help create novel drugs.

The secret to a quantum computer’s power lies in its ability to generate and manipulate quantum bits, or qubits.

What is a qubit?

Today’s computers use bits—a stream of electrical or optical pulses representing 1s or 0s. Everything from your tweets and e-mails to your iTunes songs and YouTube videos are essentially long strings of these binary digits.

Quantum computers, on the other hand, use qubits, which are typically subatomic particles such as electrons or photons. Generating and managing qubits is a scientific and engineering challenge. Some companies, such as IBM, Google, and Rigetti Computing, use superconducting circuits cooled to temperatures colder than deep space. Others, like IonQ, trap individual atoms in electromagnetic fields on a silicon chip in ultra-high-vacuum chambers. In both cases, the goal is to isolate the qubits in a controlled quantum state.

Qubits have some quirky quantum properties that mean a connected group of them can provide way more processing power than the same number of binary bits. One of those properties is known as superposition and another is called entanglement.

What is superposition?

Qubits can represent numerous possible combinations of 1and 0 at the same time. This ability to simultaneously be in multiple states is called superposition. To put qubits into superposition, researchers manipulate them using precision lasers or microwave beams.

Thanks to this counterintuitive phenomenon, a quantum computer with several qubits in superposition can crunch through a vast number of potential outcomes simultaneously. The final result of a calculation emerges only once the qubits are measured, which immediately causes their quantum state to “collapse” to either 1or 0.

What is entanglement?

Researchers can generate pairs of qubits that are “entangled,” which means the two members of a pair exist in a single quantum state. Changing the state of one of the qubits will instantaneously change the state of the other one in a predictable way. This happens even if they are separated by very long distances.

Nobody really knows quite how or why entanglement works. It even baffled Einstein, who famously described it as “spooky action at a distance.” But it’s key to the power of quantum computers. In a conventional computer, doubling the number of bits doubles its processing power. But thanks to entanglement, adding extra qubits to a quantum machine produces an exponential increase in its number-crunching ability.

Quantum computers harness entangled qubits in a kind of quantum daisy chain to work their magic. The machines’ ability to speed up calculations using specially designed quantum algorithms is why there’s so much buzz about their potential.

That’s the good news. The bad news is that quantum machines are way more error-prone than classical computers because of decoherence.

What is decoherence?

The interaction of qubits with their environment in ways that cause their quantum behavior to decay and ultimately disappear is called decoherence. Their quantum state is extremely fragile. The slightest vibration or change in temperature—disturbances known as “noise” in quantum-speak—can cause them to tumble out of superposition before their job has been properly done. That’s why researchers do their best to protect qubits from the outside world in those supercooled fridges and vacuum chambers.

But despite their efforts, noise still causes lots of errors to creep into calculations. Smart quantum algorithms can compensate for some of these, and adding more qubits also helps. However, it will likely take thousands of standard qubits to create a single, highly reliable one, known as a “logical” qubit. This will sap a lot of a quantum computer’s computational capacity.

And there’s the rub: so far, researchers haven’t been able to generate more than 128 standard qubits (see our qubit counter here). So we’re still many years away from getting quantum computers that will be broadly useful.

That hasn’t dented pioneers’ hopes of being the first to demonstrate “quantum supremacy.”

What is quantum supremacy?

It’s the point at which a quantum computer can complete a mathematical calculation that is demonstrably beyond the reach of even the most powerful supercomputer.

It’s still unclear exactly how many qubits will be needed to achieve this because researchers keep finding new algorithms to boost the performance of classical machines, and supercomputing hardware keeps getting better. But researchers and companies are working hard to claim the title, running tests against some of the world’s most powerful supercomputers.

There’s plenty of debate in the research world about just how significant achieving this milestone will be. Rather than wait for supremacy to be declared, companies are already starting to experiment with quantum computers made by companies like IBM, Rigetti, and D-Wave, a Canadian firm. Chinese firms like Alibaba are also offering access to quantum machines. Some businesses are buying quantum computers, while others are using ones made available through cloud computing services.

Where is a quantum computer likely to be most useful first?

One of the most promising applications of quantum computers is for simulating the behavior of matter down to the molecular level. Auto manufacturers like Volkswagen and Daimler are using quantum computers to simulate the chemical composition of electrical-vehicle batteries to help find new ways to improve their performance. And pharmaceutical companies are leveraging them to analyze and compare compounds that could lead to the creation of new drugs.

The machines are also great for optimization problems because they can crunch through vast numbers of potential solutions extremely fast. Airbus, for instance, is using them to help calculate the most fuel-efficient ascent and descent paths for aircraft. And Volkswagen has unveiled a service that calculates the optimal routes for buses and taxis in cities in order to minimize congestion. Some researchers also think the machines could be used to accelerate artificial intelligence.

It could take quite a few years for quantum computers to achieve their full potential. Universities and businesses working on them are facing a shortage of skilled researchers in the field—and a lack of suppliers of some key components. But if these exotic new computing machines live up to their promise, they could transform entire industries and turbocharge global innovation.

The latest beleaguered Google product to get a death date is Google+. Google’s controversial Facebook clone is shutting down on April 2. Google has been backing away from the service for years, but it gave the site a death sentence in October, after revelations of a data leak were made public. Now we have a concrete shutdown date for the service.

Google’s support page details exactly how the G+ shutdown will go down, and it’s not just freezing posts on the site. The whole site will be taken down, and everything will be deleted. “On April 2nd, your Google+ account and any Google+ pages you created will be shut down and we will begin deleting content from consumer Google+ accounts,” the page reads.

The whole deletion process sounds brutal. It won’t just be the entire Google+ site that will be scrubbed from the Internet—Google+-powered comments on Blogger and other third-party sites will all be deleted, too. Users of Google+ have until April to download and save everything themselves, which they can do via this page.

Google+ was Google’s panicked reaction to the rise of Facebook and other social networks. Back in 2011 Google saw Facebook as an existential threat. What if people stopped searching and just asked their friends for good websites and product recommendations? It sounds ridiculous today, but back then Google thought social was the future, and Google’s then-CEO Larry Page tied all employee bonuses to the success of Google+. As a result, Googlers forced Google+ into every other Google product, whether it made sense or not. At one point the service was described as “kind of like the next version of Google” by a Google executive. Now Google has to remove Google+’s tentacles from all of its products.

With all the integrations into other products, Google warns that shutting down Google+ won’t be a quick process. “The process of deleting content from consumer Google+ accounts, Google+ Pages, and Album Archive will take a few months, and content may remain through this time,” the Google support page reads. “For example, users may still see parts of their Google+ account via activity log and some consumer Google+ content may remain visible to G Suite users until consumer Google+ is deleted.”

While April 2 is the drop-dead date for profiles, the wind-down of Google+ will start pretty soon. As of February 4 you won’t be able to make new Google+ profiles or communities. Google says that third-party sites and apps that use the Google+ login button will stop working “in the coming weeks” and that third parties should switch to the non-Google+ version of Google’s sign-in button. Google+-powered comments will be stripped from Blogger on February 4 and from third-party sites on March 7.

Google has been backing away from Google+ for some time, and today the service is not as fully Google-integrated as it has been in the past. Google Photos and Google Hangouts were once parts of Google+, and now they are standalone services. At one point Google+ was mandatory for YouTube comments, and now it isn’t. The service was similarly de-integrated from the Google Play Store, Gmail, and Google Search.

Google+ is yet another product in a list of services Google is planning to shut down soon. Before Google+ is shuttered, Google Inbox will die in March. Google Hangouts will start to wind down in October for G Suite users, with a consumer shutdown expected sometime in 2020. In the future I suspect we’ll be talking about the death of Google Play Music, which is expected to merge with YouTube Music at some point. It’s going to be a tough year to be a Google user.

For the past three years, Facebook has paid consumers as young as 13 to download a “Facebook Research” application that gives the company wide-ranging access to their mobile devices, according to a TechCrunch investigation published Tuesday. In order to allow people with iPhones to participate, Facebook sidestepped the strict privacy rules imposed by Apple in its App Store by taking advantage of a business applications program designed for internal company use. Apple soon announced it was revoking Facebook’s access to its Developer Enterprise Program, which also allowed the company to share custom iOS apps with its own employees. Apple’s decision is reportedly wreaking havoc at the social network, rendering workers unable to access the apps they use for their jobs.

As Facebook deals with the fallout from yet another privacy scandal, it’s worth unpacking how its Research app worked—especially because it serves as a good reminder for other apps you might already be using, particularly virtual private networks. It wasn’t just Facebook: Google also disabled a similar app on iOS devices Wednesday as well. Both apps are still available on Android.

Facebook reportedly paid users between the ages of 13 and 35 $20 a month to download the app through beta testing companies like Applause, BetaBound, and uTest. Participants found out about the opportunity via Snapchat and Instagram advertisements, according to TechCrunch. Minors were required to get consent from their parents. Once approved, participants downloaded the app via their browser—not through the Google Play Store or Apple App Store.

Apple typically doesn’t allow app developers to go around the App Store, but its enterprise program is one exception. It’s what allows companies to create custom apps not meant to be downloaded publicly, like an iPad app for signing guests into a corporate office. But Facebook used this program for a consumer research app, which Apple says violates its rules. “Facebook has been using their membership to distribute a data-collecting app to consumers, which is a clear breach of their agreement with Apple,” a spokesperson said in a statement. “Any developer using their enterprise certificates to distribute apps to consumers will have their certificates revoked, which is what we did in this case to protect our users and their data.” Facebook didn’t respond to a request for comment.

Facebook needed to bypass Apple’s usual policies because its Research app is particularly invasive. First, it requires users to install what is known as a “root certificate.” This lets Facebook look at much of your browsing history and other network data, even if it’s encrypted. The certificate is like a shape-shifting passport—with it, Facebook can pretend to be almost anyone it wants. If you visit the website for a clothing retailer, for instance, Facebook can use the root certificate to pretend to be the store and see the pants you were looking to buy. “You allow Facebook to pretend to be anyone they want to be on the internet—your device will trust the certificates they generate,” says David Choffnes, a professor and mobile networking researcher at Northeastern University.

Facebook couldn’t use its root certificate for every website or application, since some companies, like banks, protect hackers from using them for man-in-the-middle attacks using a technique called “certificate pinning.” The bank or other company essentially decides that it won’t accept any certificate but its own—it knows not to take phonies like Facebook’s. “This attack doesn’t work on everything, but there’s still a large fraction of apps that are vulnerable because it’s not a standard threat model,” says Choffnes.

Facebook’s app also established an on-demand private network connection, meaning it routed all of the participants’ traffic through its own servers before passing it along to its final destination. This is essentially what all VPNs do—they disguise traffic by rerouting it, allowing you to hide things like your location, perhaps to use Gmail in China or access streaming shows not available where you live. But VPNs typically can’t see your encrypted traffic, since they don’t have the right certificate. They can still look at your unencrypted traffic, which can be an issue, but the vast majority of internet traffic today happens over encrypted HTTPS connections. But with its root certificate installed, Facebook could decrypt the browsing history or other network traffic of the people who downloaded Research, possibly even their encrypted messages.

To use a non-digital analogy, Facebook not only intercepted every letter participants sent and received, but also had the ability to open and read them. All for $20 a month!

Using its VPN connection and root certificate, Facebook had the ability to gather extensive data from participants, including their browsing history, what apps they used and for how long, as well as the messages they sent. Facebook also requested some people screenshot their Amazon orders page, according to TechCrunch, suggesting the social network may have had an interest in consumer purchasing habits. But unless Facebook discloses what it sought to learn from Research, there’s no way to know exactly what the app might have been collecting.

“Capability versus actual things they did is a much bigger question,” says Mike Murray, the chief security officer of the mobile security firm Lookout. “Because that all happens on the backend, you can’t really tell what they did.”

In the past, Facebook has used a similar app to learn more about its rivals. In 2013, the social network acquired Onavo, an Israeli VPN maker, which it reportedly used to research popular emerging apps in order to either copy or buy them. It used Onavo to look into WhatsApp, for instance, which Facebook later acquired in 2014. Last year, Facebook began promoting Onavo in its iOS app under the banner “Protect,” but it later pulled the app from the App Store after Apple said it violated its new data-sharing policies, according to The Wall Street Journal.

Facebook isn’t the only company hungry for data on what consumers are doing on their phones. Google used Apple’s enterprise program to distribute an app called Screenwise Meter, which also acts like a VPN. In exchange for letting the tech giant collect and analyze their network traffic, Google provides participants with gift cards to various retailers. It’s part of a wider Google consumer behavior program where participants can install tracking software on their router, laptop browser, and television. The difference is that the Google app doesn’t require users to install a root certificate—meaning they can’t look at encrypted traffic. Still, Google wasn’t complying with Apple’s rules either, and it has now disabled the iOS version of Screenwise.

“The Screenwise Meter iOS app should not have operated under Apple’s developer enterprise program—this was a mistake, and we apologize,” a Google spokesperson said in a statement. “We have disabled this app on iOS devices. This app is completely voluntary and always has been. We’ve been upfront with users about the way we use their data in this app, we have no access to encrypted data in apps and on devices, and users can opt out of the program at any time.”

While Facebook’s app is particularly invasive, a number of other companies also pay or reward users in exchange for information about what they do online, like the data giant Nielsen. In every case, people voluntarily download these apps and programs, though they may not always understand the full extent of the access they are granting—especially if they’re not even 18.

Even if you don’t plan to make money by selling your data, Facebook’s latest privacy scandal is a good reminder to be wary of mobile apps that aren’t available for download in official app stores. It’s easy to overlook how much of your information might be collected, or to accidentally install a malicious version of Fortnite, for instance. VPNs can be great privacy tools, but many free ones sell their users’ data in order to make money. Before downloading anything, especially an app that promises to earn you some extra cash, it’s always worth taking another look at the risks involved.

---

More Great WIRED Stories

This is not a story about how the polar vortex is bad—bad for the human body, bad for public transportation, bad for virtually everything in its path. This is a story about how one being among us is actually taking advantage of the historic cold snap: Cassie the bipedal robot. While humans suffer through the chill, this trunkless pair of ostrich-like legs is braving the frozen grounds of the University of Michigan, for the good of science.

“When we saw the announcement for the polar vortex, we started making plans to see how long we could operate in that kind of weather,” says roboticist Jessy Grizzle. “We were going to tie a scarf on her just so it looked cute, but we decided people would think that was keeping her warm and affecting the experiment, so we didn’t.”

Scarves aside, this is vital research for a future in which robots tackle not just polar vortices but any number of other brutal environments. Because when you or I go out in -50 degree weather, the concern is that frostbite will set in pretty much immediately. Robots, of course, don’t have to worry about their tissue freezing, but they do have to worry that their components might not be engineered to withstand the cold.

The most obvious concern is the toll on their batteries: The colder it gets, the slower the chemical reaction that ultimately powers machines, meaning the batteries drain more quickly. You see that in gadgets like phones, as well as in electric cars. But Cassie? Weirdly, no, not even in subzero temperatures.

“We thought the walking would be limited by the duration of the battery outdoors,” Grizzle says. “It turns out it doesn’t seem that that was the case.” Cassie managed a total of an hour and a half of operation out in the cold, with battery power to spare. Why it fared so well, Grizzle isn’t sure.

But the remote-controlled Cassie had other problems, which is perhaps not surprising given that it’s a research platform meant for the comfort of labs, not the ravages of polar vortices. Twenty minutes into the robot’s meanderings, something went haywire with the electronics.

“We couldn’t quite tell because the power shut down and the robot just fell to the ground,” says Grizzle. “So we took Cassie back into the lab and took some caps apart and reconnected the wires.” No fall for a robot worth hundreds of thousands of dollars is a good fall, but the weather only made matters worse here. “She cracked the protective cover on the battery because the plastic became brittle at this temperature, so it shattered almost like glass. So we kind of taped it back together.” Good as almost new, Cassie went back out and walked for another hour straight.

She was sounding a bit funny, though: Her actuators were squeaking. These electric motors are complicated, after all—precisely honed metal whirring at high speeds to propel the robot. But other than the noise, nothing with the locomotion was out of the ordinary. “Otherwise, she seemed fine,” Grizzle says.

Cassie’s cold-weather jaunt was a rare opportunity to push those legs to one kind of extreme. As robots like Cassie edge closer to walking among us, they have to learn to adapt to the real world—not just by withstanding plunging temperatures but by adapting themselves to walk on snow or ice.

“Walking in the snow, you know you have to take higher steps and more of a vertical up and down action,” Grizzle says. “Otherwise you end up dragging your foot through the snow, so you have to change the style of gait.”

Roboticists usually make these adjustments manually, coding a robot to walk a certain way in particular conditions. But the goal is to get Cassie and other bipeds to automatically detect when, say, they’ve transitioned from the wood slats of a boardwalk onto sand. And there’s no better way to get the machines ready than to subject them to punishing conditions. Last year, for example, Grizzle’s lab had Cassie traipse through flames. (Very Game of Thrones, by the way, all the fire and now ice.)

“We all want to use robots in conditions where people could be at risk,” Grizzle says. “If we can have Cassie walking in inclement weather, she could then, with her mapping system, help identify anyone caught out in the cold.”

Better Cassie than you or me, after all.

---

More Great WIRED Stories