Uber ugliness unmasks Silicon Valley's bro culture

Uber ugliness unmasks Silicon Valley's bro culture

Recent revelations about sexual harassment and gender discrimination at Uber are the tip of the iceberg for a ‘bro culture’ run amok

The world was shocked—shocked, I say—to learn this week that female employees are routinely sexually harassed and discriminated against in Silicon Valley.

Susan J. Fowler’s blog recounting her “strange, fascinating, and slightly horrifying” year as an engineer at Uber unleashed a fury of condemnations, hastily chased by a maelstrom of mea culpas from CEO Travis Kalanick—who’s grown well-practiced in the art ever since last month’s #DeleteUber campaign began.

Terrible, horrible, no good, very bad Uber

Uber, with its history of flouting regulations and ignoring ethical standards, is the Valley’s made-to-order villain. “We definitely see Uber as a bottom-feeder of the tech industry when it comes to issues of fairness and equity,” Orson Aguilar, president of economic justice organization Greenlining Institute, told The Guardian.

Details of Fowler’s treatment at Uber are appalling. She documents an ultra-aggressive work culture where complaints about discrimination and sexual harassment were “addressed” by counseling the women to transfer to another team or expect poor performance reviews in future. “One HR rep even explicitly told me that it wouldn’t be retaliation if I received a negative review later, because I had been ‘given an option,’” Fowler wrote.

Male “high performers” (otherwise known as “brilliant jerks”) were above consequences. Even after her supervisor made highly illegal threats to fire Fowler for reporting a manager, “both HR and the CTO admitted that this was illegal but none of them did anything.” And in text-perfect victim-blaming mode, one meeting with an HR rep began with her “asking me if I had noticed that *I* was the common theme in all of the reports I had been making, and had I ever considered that I might be the problem.”

Everything old is new again

But is there anyone in Silicon Valley for whom any of this comes as a surprise? “If you’re not hearing about the mistreatment of women [in Silicon Valley], it’s because you’re not listening very hard,” Huffington Post wrote ... in 2012. That was the year Ellen Pao filed gender bias and sexual harassment charges against venture capital group Kleiner Perkins Caufield & Byers, then endured an avalanche of character assassination and mudslinging after speaking out.

“[Pao] has been tone policed. She has been slut-shamed. She has been labeled a gold digger. She has been accused of being untalented, amateurish, and unprofessional,” Quartz wrote at the time. “The message Kleiner’s lawyers are trying to communicate is clear: Ellen Pao is a lone voice trying to capitalize off an imagined gender problem in Silicon Valley.”

Only it’s not imagined.

Babes, boobs, butts, and beer

The gender problem in the tech industry goes far beyond Microsoft CEO Satya Nadella advising women to not ask for raises, “knowing and having faith that the system will give you the right raise.”

From “booth babes” at computer conferences to Kalanick’s reference to Uber as “Boob-er” because it helped him attract women to offensive tweets sent by tech companies in praise of their booth girls’ butts to a coding marathon that advertised “friendly (female) event staff” who serve you beer—Silicon Valley has a problem.

When Observer began reaching out to women in the tech world for their reactions to Fowler’s experience at Uber, it had no trouble quickly amassing 12 stories like Fowler’s, told by women who currently work in tech. “We knew this runs rampant in the industry, but we didn’t realize [stories of harassment would] be this easy to find it,” Observer wrote.

Why were they surprised? Newsweek did an in-depth report two years ago, “What Silicon Valley Thinks of Women,” and found the forward-looking technology community was stunningly backward when it came to gender relations. It described a misogynistic bro culture rampant with “blogs, screeds, letters, videos and tweets about threats of violence, sexist jokes and casual misogyny, plus reports of gender-based hiring and firing, major-league sexual harassment lawsuits and a financing system that rewards young men and shortchanges women.”

(Anyone still unconvinced that harassment exists should read tech journalist Sam Biddle’s blog, which provides “every f**ked-up text” from Tinder’s 2014 sexual harassment suit and starkly reveals “how ugly it can be to face a macho-dominated startup office.”)

Perhaps only in the “bro culture” could Gurbaksh Chahal, CEO and founder of adtech startup Gravity4, be captured on a home security video punching and kicking his girlfriend 117 times in a half-hour attack, then later assault another woman and threaten to report her to immigration authorities when she spoke of filing a restraining order against him—and be rewarded within the industry. In December Chahal was appointed as an adviser at a venture capital firm.   

“Paula Deen made racially insensitive comments and lost a show, lost very real money. Donald Sterling was forced to sell an NBA team,” Sarah Lacy told Recode. “And yet [in tech] we believe that frighteningly misogynist comments … are ‘boys being boys’ and that ‘they’re geniuses and this is what it takes to build a company.’ It’s really scary that there’s a company culture where objectification and violence against women is condoned,” she said.

(Lacy was the target of a smear campaign by Uber senior exec Emil Michael, who bragged about his idea of building a team of researchers to dig up personal dirt on reporters who were critical of the company. He remains a senior vice president at Uber.)

The elephant in the room

A New York Times article traced the fates of the Stanford class of 1994 and found many women with STEM degrees “simply bailed out.” Shanley Kane, a tech industry observer and founder of Model View Culture, noted: “[Women] are not getting hired, and we are not getting promoted, and we are being systematically driven out of the industry.”

Last year a study, “Elephant in the Valley,” asked what happens to the women who managed to stick it out. The survey focused on 200-plus women with at least 10 years of experience, largely in the Bay Area. The results were depressingly predictable:

  • 90 percent had witnessed sexist behavior at company off-sites and/or industry conferences
  • 88 percent had clients and colleagues address questions to male peers that should have been addressed to them
  • 87 percent had heard demeaning comments from their colleagues
  • 84 percent had been told that they were “too aggressive” (“At Company X we had a joke that there were only two reviews for women – you are either too reticent or you are too bossy – no middle ground,” said one respondent.)
  • 75 percent were asked about family life, marital status, and children in job interviews
  • 66 percent felt excluded from social and networking opportunities because of their gender
  • 60 percent reported unwanted sexual advances—two-thirds of which came from a superior. Among those women who were harassed, 60 percent reported the harassment and were dissatisfied with the results; 39 percent did nothing for fear of retribution
  • 47 percent had been asked to do lower-level tasks, like taking notes or ordering food, that were not expected of their male colleagues

The scathing “Open Letter to Tech”—written back in 2014—made clear how tired women in tech are of “pretending this stuff doesn’t happen.”

We regularly receive creepy, rapey emails where men describe what a perfect wife we would be and exactly how we should expect to be subjugated. Sometimes there are angry emails that threaten us to leave the industry, because “it doesn’t need anymore c**ts ruining it.”  

We have watched companies say that diversity is of highest importance and have invited us to advise them. After we donate much of our time they change nothing, do nothing, and now wear speaking to us as a badge of honor. Stating, “We tried!”

Note to the tech industry: Try harder.

Controversial LTE-U wireless tech OK'd by FCC

Controversial LTE-U wireless tech OK'd by FCC

Ericsson, Nokia get go-ahead for LTE-U base stations despite early fears they might interfere with Wi-Fi

The Federal Communications Commission today approved two cellular base stations—one each from Ericsson and Nokia—to use LTE-U, marking the first official government thumbs-up for the controversial technology.

FCC chairman Ajit Pai said in a statement that the unlicensed spectrum—historically, the territory of Wi-Fi—can now be used to help ease the load on carrier mobile networks. “It means they get to enjoy the best of both worlds: a more robust, seamless experience when their devices are using cellular networks and the continued enjoyment of Wi-Fi, one of the most creative uses of spectrum in history.”

T-Mobile has already announced that it will be deploying LTE-U technology, having participated in testing last year, and other networks are likely to follow suit.

Aside from equipment vendors like Ericsson and Nokia, the only companies likely to greet the news of LTE-U’s approval with enthusiasm are the carriers themselves and Qualcomm, which essentially invented the technology. Other major tech sector players, including Google, Comcast, and Microsoft, have expressed serious concerns that LTE-U doesn’t play as nicely with Wi-Fi as advertised, though collaborative testing has ratcheted tensions down of late.

Pai is seen to be close with some of the industries the FCC is supposed to regulate—more so than outgoing chairman Tom Wheeler, who had urged the various parties in the LTE-U debate to work together, rather than risk government regulatory intervention. That possibility appears to have vanished, given the FCC’s approval today, but proof of LTE-U’s compatibility will come—or not—with large-scale deployment.

This story, "Controversial LTE-U wireless tech OK'd by FCC" was originally published by Network World.

Netflix Stethoscope gives users a BYOD security checkup

Netflix Stethoscope gives users a BYOD security checkup

The IT-compiled web application provides users a personalized list of security recommendations for their own computers, smartphones, and tablets

Medical professionals use stethoscopes to help diagnose problems inside the body. With Netflix's newly open-sourced Stethoscope web application, users receive a security checkup for their mobile and computing devices without having to call IT.

Although device management platforms like MobileIron or VMware AirWatch for mobile devices, JAMF for Macs, and LANdesk for Windows let IT teams apply security controls to employee devices, there are few tools available for users interested in securing their personal devices, and most are focused on viruses or are platform-specific.

Netflix is an extreme-BYOD company, with employees free to use almost anything. That's why it developed Stethoscope help employees self-diagnose their devices. Stethoscope doesn't apply any policies to user devices; it tells employees in fairly clear English what security deficiencies it found and how they might address them.

But Stethoscope is not a product you can use. Instead, it's a web-based tool that any organization can compile itself from the code that Netflix has provided on GitHub via an open source Apache 2.0 license. IT will need to compile the code and host the app for its users to access. That do-it-yourself approach does have the advantage of preventing user or company information being gathered by third-party scanners.

Once compiled and hosted, the Stethoscope application can collect information from desktop and laptop computers, as well as smartphones and tablets. At the moment, the application tracks and makes recommendations for disk encryption, firewall configuration, status of automatic updates, software and operating system updates, screen locks, and presence of a security software stack (such as Carbon Black). The application can also check whether a mobile device has been jailbroken or rooted. (This is similar to what mobile management tools can inspect.)

Each recommendation is assigned a rating to show how important it is, with more important actions sorted to the top. Critical recommendations are highlighted in red and collected in a banner to make it easier for users to see what they should be focusing on.

As a self-service application, Stethoscope also displays additional information about each security recommendation and a link to detailed instruction. This way, the user can understand the reasoning behind the recommendations being made. The app also can present notifications and let users respond, such as device access warnings alerting users to logins from unexpected IP addresses.

Stethoscope's back end is Python using Twisted + Klein, and the front end is React. Nginx serves up static files and proxies requests to the API server. Stethoscope doesn't have its own data store, but it merges user information obtained from external data sources, such as authentication providers and device management platforms. The plugin architecture lets IT add other services and data stores to integrate with Stethoscope, such as Elasticsearch imports and Google accounts. The Netflix team is currently working on adding OSquery support.

Verizon plans 5G trial service in 11 cities this year

Verizon plans 5G trial service in 11 cities this year

The pre-commercial service for selected customers will connect homes with high-speed wireless

Lab tests of pre-standard 5G wireless with multi-gigabit speeds are evolving into trial services that users can actually enjoy in the real world – though not necessarily while walking around with a smartphone.

Verizon said Wednesday it will launch pre-commercial 5G service in 11 markets around the U.S. by the middle of this year, joining rival AT&T in aggressively deploying the future technology.

At Mobile World Congress in Barcelona next week, carriers are expected to announce more upcoming 5G trials.

The Verizon trial, planned for select users in cities including Atlanta, Denver, Miami, and Seattle, will use fixed wireless equipment for home broadband instead of mobile networks for roving devices. But it’s an early use of millimeter-wave frequencies that carriers say they will need for future high-speed mobile broadband.

Verizon hasn’t said how fast the 5G trial networks will be, but a Verizon video about tests of the technology showed speed readouts over 3Gbps.

AT&T, which has already announced a trial of 5G service for business use at an Intel facility in Austin, Texas, also plans to launch fixed-wireless 5G for homes in Austin and Indianapolis later this year. In those trials, also for select customers, AT&T will stream DirecTV service to antennas on users’ homes.

The full specification for 5G is not expected to be finished until at least next year. But wireless equipment vendors and service providers have been running limited tests and trials using elements of the emerging technology for more than a year. Some have shown speeds in the tens of gigabits per second. In addition to high speed, other 5G goals include lower latency and longer battery life.

Verizon says the pre-commercial service will use spectrum around 28GHz, one of the millimeter-wave bands many companies are considering for 5G. These frequencies, much higher than the ones used for cellular service today, need a new generation of radios and antennas to overcome what would naturally be very short range.

Aiming those signals at a stationary antenna on a house is easier than tracking a mobile device with a beam, something vendors and carriers are also working on.

Verizon’s fixed wireless technology is part of a set of open specifications for 5G that the company developed along with partners in the Verizon 5G Technology Forum. That group includes Ericsson, Intel, Qualcomm, and Samsung.

Other cities involved in the Verizon trial will include Houston, Dallas, Sacramento, and Washington, D.C.

Windows 10's next big update, Redstone 3, hits the radar

Windows 10's next big update, Redstone 3, hits the radar

A new Windows 10 version, likely called 1710 or 1711, build 15141, has been posted on BuildFeed

Think of the BuildFeed.net site as an early warning system for new versions of Windows. Earlier this week, a brand-new version of Win10 got caught in the BuildFeed flypaper. Build 10.0.15141.1000 (rs_prerelease.170219-2340) appears to be the first in a new line of Win10 releases, code-named "Redstone 3."

When fully developed, Redstone 3 will supplant the Win10 version currently in beta, the Creators Update aka version 1703 ("Redstone 2"). Look for the coming out ceremony later this year.

If Microsoft continues its current pace of releasing new Win10 versions every eight months or so, Redstone 3 will likely hit in October or November. Given the pressure to get new releases out prior to the holiday buying season, I think it likely that Redstone 3 will end up as version 1710 (as in October 2017) or possibly 1709 or 1711.

We knew it was coming. Earlier this week, multitudes of Windows blogs bloviated over the discovery of a slide in a Ch9 video from Bill Karagounis' presentation at Microsoft's Ignite Australia conference. (Look at 22 to 24 minutes, if you're really interested.) The slide shows that Microsoft plans on shipping another version of Windows 10 in late 2017.

win10 release cadence Microsoft

The headlines about Microsoft "confirming" there will be a second version of Win10 later this year are silly. Of course Microsoft's working on a new version of Windows 10, and we've known for months about plans to slip another version out this year. Win10 has received version bumps every eight months or so since the first bump appeared four months after the original version of Win10. This continues the trend.

(A cynical soul might note that the "Pilot" bars in the slide correspond to the time after a version has been released, but before it's declared Current Branch for Business. That's the period after new versions go to the unwashed masses, but before they're anointed as being business-ready.)

We were originally told that Win10 version changes ("feature updates") would appear two to three times a year. Two weeks ago, Dani Halfin posted on TechNet in his Overview of Windows as a Service:

Windows as a service will deliver smaller feature updates two to three times per year

The rhythm now is definitely set at eight months, give or take a bit. That would equate to two versions of Win10 in 2017 and one in 2018.

Given the cadence shown in the slide, unless Microsoft changes its end-of-life calculation scheme -- it changed the rules last week -- you should expect that version 1511 (the November update) will hit end-of-life in Sept. 2017, version 1607 (the Anniversary Update) will last until May 2018 or so, and version 1703 (the Creators Update, due next month) will continue to receive security patches until early 2019.

In the interim, Windows 10 Creators Update, version 1703, has entered the final stretch of bug fixes before release.

What to expect in version 1710? It's still much too early to nail anything down, of course, but here's what I expect to see:

  • The My People feature, also called the "People Experience," was announced in October, but it didn't make the Creators Update. "With Windows My People, you can pin your favorite contacts to the Windows task bar and easily drag and drop any document, photo or video right on top of the contact for easy sharing. You can receive unique notifications, Shoulder Taps, and easily view and open emails, IMs, shared documents and more, all in one place."
  • Project NEON is a built-in ability to animate and blur backgrounds and sidebars. See Mehedi Hassan's description in MSPoweruser.
  • Composable Shell, or cshell, will unify the main Windows interface so that it can be run across PCs, Mobile, HoloLens, and Xbox. See Zac Bowden's description at Windows Central.

There's still some life left in the old-timer.

Discussion continues on the AskWoody Lounge.

FCC rolls back net neutrality ISP transparency rules

FCC rolls back net neutrality ISP transparency rules

Small ISPs will no longer be required to inform customers of their network management practices

The U.S. Federal Communications Commission has voted to roll back some net neutrality regulations that require broadband providers to inform customers about their network management practices.

The Republican-controlled FCC on Thursday suspended the net neutrality transparency requirements for broadband providers with fewer than 250,000 subscribers. Critics called the decision anticonsumer.

The transparency rule, waived for five years in a 2-1 party-line vote Thursday, requires broadband providers to explain to customers their pricing models and fees as well as their network management practices and the impact on broadband service.

The commission had previously exempted ISPs with fewer than 100,000 subscribers, but Thursday’s decision expands the number of ISPs not required to inform customers. Only about 20 U.S. ISPs have more than 250,000 subscribers.

The five-year waiver may be moot, however. FCC Chairman Ajit Pai and Republicans in Congress are considering ways to scrap a large chunk of the net neutrality regulations approved by the agency just two years ago.

Small ISPs have argued the transparency rules amount to burdensome and costly regulations, while consumer advocates have argued that broadband customers have the right to know when a provider is throttling traffic. The exemption will allow small ISPs to redirect the money they spend on compliance, Pai said.

“I firmly believe that these ISPs should spend their limited capital building out better broadband to rural America, not hiring lawyers and accountants to fill out unnecessary paperwork demanded by Washington, D.C.,” he said.

However, the White House Office of Management and Budget, under former President Barack Obama, found that compliance with the transparency rules takes each broadband provider less than seven hours a year, noted Commissioner Mignon Clyburn, a Democrat. The new waiver exempts some large companies with smaller broadband subsidiaries from the transparency rules, she said.

“In an ongoing quest to dismantle basic consumer protections for broadband services, the majority has decided to exempt billion-dollar public companies from being transparent with consumers,” Clyburn added. “This represents yet another in a series of steps being taken to jettison pro-consumer initiatives.”

The commission’s two Republicans defended the change. In January, the House of Representatives passed a bill with a similar exemption in a voice vote with no dissenters, Commissioner Michael O’Rielly noted.

The vote represents a “sensible and soundly justifiable exemption for small internet service providers from unnecessary and expensive” regulations, O’Rielly said.

UN steps in to end marketing war over what 5G means

UN steps in to end marketing war over what 5G means

It means you won't get real 5G service until 2020, according to the ITU, the UN's telecom standards body

With mobile operators’ marketing departments already throwing around claims about their 5G services, the United Nations is weighing in with its definition of what qualifies a network as next-generation.

Verizon Wireless will begin delivering “5G” service to select users in 11 U.S. cities in mid-2017, even though some places don’t yet have access to 4G. And at the Mobile World Congress 2017 trade show in Barcelona, companies including Intel, Qualcomm and Ericsson will be promoting their moves towards 5G.

But what marks the difference between one generation of mobile technology and the next? 

There are 13 technical requirements for next-generation networks on the draft list published by the International Telecommunication Union, the UN agency that sets rules for radio spectrum usage and telecommunications interoperability.

Among the requirements are peak download speeds of up to 20 gigabits per second and peak upload speeds of up to 10Gbps.

We all know that those “up to” speeds are rarely seen outside the marketing department, so the ITU has helpfully set out what we should expect to see in the real world. Users should experience typical download speeds of 100 megabits per second, and typical upload speeds of 50Mbps, it said.

When things get really busy, it also wants networks to be able to deliver a minimum level of service even when there are 1 million connected devices per square kilometer (2.59 million per square mile), handy for texting from crowded presidential inaugurations or keeping tabs on the worst excesses of IoT clutter around your home.

It also calls for networks to offer minimum levels of service to people moving at different speeds, and sets out different service level requirements for users depending on whether they are stationary, walking, or in a vehicle. In dense urban environments it doesn’t expect those vehicles to exceed 30km/h, but in rural areas it sets two service levels for vehicles: those traveling between 10 and 120km/h, and those traveling up to 500km/h. If anyone ever builds a hyperloop (the average speed of which is expected to approach 1,000km/h), internet access will not be guaranteed aboard.

After that, things get a lot more technical, with minimum requirements for spectrum availability, traffic density and several measures of spectral efficiency.

Those requirements are all fine and dandy, but unfortunately for the puzzled consumer, the ITU’s draft report is titled “Minimum requirements related to technical performance for IMT-2020 radio interface(s)” and makes no reference to 5G, although that’s what it really means.

The ITU has a history of choosing such cryptic names: IMT-2020 is the follow-on to its “IMT-Advanced” specification, known to the rest of the world as 4G, which was itself preceded by IMT-2000, more widely known as 3G.

The 2020 in the title refers to the year the ITU expects the IMT-2020, or 5G, specification to be ready.

Standardization work isn’t just going on at the ITU: The Third Generation Partnership Project or 3GPP is working on a “5G” specification that it intends to submit to the ITU by October 2020, and earlier this month unveiled one of the first deliverables, a new logo.

3gpp 5g logo 3GPP

The Third Generation Partnership Project (3GPP)’s “5G” logo.

If you see it on a phone before the end of 2018 though, when 3GPP plans to publish the first release of its specification, then it’s probably a fake.

Container-friendly Alpine Linux may get Java port

Container-friendly Alpine Linux may get Java port

Project Portola would port the JDK to the security-focused, lightweight Linux distribution

Alpine Linux, a security-focused lightweight distribution of the platform, may get its own Java port. Alpine is popular with the Docker container developers, so a Java port could pave the way to making Java containers very small.

A proposal floated this week on an OpenJDK mailing list calls for porting the JDK (Java Development Kit), including the Java Runtime Environment, Java compiler and APIs, to both the distribution and the musl C standard library, which is supported by Alpine Linux. The key focus here is musl; Java has previously been ported to the standard glibc library, which you can install in Alpine, but the standard Alpine release switched two years ago to musl because it’s much faster and more compact

The Portola porting effort would be led by Mikael Vidstedt, a JDK 9 committer at Oracle. The proposal is still in its early stages of discussion, and the proposal has sparked questions. One member of the mailing list questioned the scope of the port, asking whether it would be handled as a new OS or a Linux version. He also questioned which architectures would be supported.

Vidstedt responded that project members would discuss how exactly to model the Alpine/musl port. But his preliminary investigation on Linux/64 did not uncover anything inherently specific to the CPU architecture, “so until proven otherwise I would assume that the musl-supported platforms is the limiting factor,” he said.

Although the next version of Java, due this summer, is JDK 9, the initial source of the Portola project would be based on a clone of a JDK 10 repository. As with the Project Lamdba and Project Valhalla Java efforts, proponents would follow a “commit first, review later” policy, with code not flowing directly from Portola repositories into JDK repositories. A curated merge would be done instead, with select changes extracted into new changesets for inclusion in JDK repositories when ready.

Disney Research demonstrates open-air wireless charging

Disney could someday deploy the far-field wireless charging tech in its amusement parks

Disney Research has invented a new method of wirelessly charging mobile devices that could someday allow amusement park patrons to walk about freely while also getting their mobile devices charged.

The technology, which Disney has dubbed "quasistatic cavity resonance" (QSCR), enables purpose-built structures such as cabinets, rooms and warehouses to generate quasistatic magnetic fields that "safely deliver kilowatts of power to mobile receivers contained nearly anywhere within."

Disney's wireless charging technology research was published in the peer-reviewed journal PLOS ONE.

Disney Research isn't the first company to create wireless charging at distance, or far-field wireless charging. Start-ups Ossia and Energous have demonstrated wireless charging beyond 15 feet.

Both Energous' WattUp and Ossia's Cota mobile device charging systems work much like a wireless router, sending radio frequency signals that can be received by enabled mobile devices, such as wearables and mobile phones. A small RF antenna in the form of a PCB board, an ASIC and software comprise the wireless power receivers.

Disney Research uses a magnetic resonator coil, not unlike those used by conventional resonant wireless charging or "near-field" technologies sold today, to project a field at distance through which devices can receive power. In the center of an experimental room stands a copper pipe with capacitors that transmit the electrical current.

An experimental demonstration showed that in a 16×16x7.5-ft room with a floor, ceiling and walls made of painted aluminum sheet metal, the QSCR could deliver power to small coil receivers in nearly any position with 40% to 95% efficiency. Up to 1900 watts of power can be transmitted safely between transmitters and a coil receiver in a mobile device, "enabling safe and ubiquitous wireless power," the company said in its research.

In the future, Disney said, the amount of aluminum used in a space to create the magnetic field can be reduced by optimizing the QSCR, "and retrofitting of existing structures will be possible via modular panels or conductive paint.

"Ultimately, QCSR based wireless power offers a viable method for eliminating the wires and batteries that have limited many innovative solutions in the industrial, medical, and consumer electronic spaces while providing an unprecedented amount of spatial charging freedom," the company said.

At the turn of the last century, Nikola Tesla routinely demonstrated far field wireless power transfer using what is typically called the "Tesla coil."

In recent years, regulatory agencies have adopted strict safety guidelines for the use of far-field magnetic resonance devices like the Tesla coil.

"This has resulted in a long-standing tradeoff between the range at which a device can be wirelessly powered and the maximum amount of power that can be safely delivered," Disney Research said. "These far-field wireless power techniques have not found widespread use, since they are limited to delivering only a few milliwatts of power due to health and safety concerns."

Near-field magnetic resonance is a highly localized -- traveling a few millimeters to a few feet before transfer efficiency drops off rapidly as the source and receiver are separated by more than a coil diameter. "Likewise, it is not possible to strongly couple coils of drastically different sizes," Disney said.

Disney Research used far-field electromagnetic waves to generate uniform field patterns in a metallic chamber, which can create near-field standing waves for strong coupling to small receivers.

"This is accomplished by stimulating the resonant electromagnetic mode of a specially designed, enclosed metallic structure such that induced currents flowing through the walls, ceiling and floor are channeled through discrete capacitors. These oscillating currents in turn generate magnetic fields that permeate the interior of the structure, thus enabling wireless power transfer to receivers contained within, while simultaneously isolating the potentially harmful electric fields in capacitors."

This story, "Disney Research demonstrates open-air wireless charging" was originally published by Computerworld.

Apple grabs icloud.net domain to solidify links to sync and storage service

Apple grabs icloud.net domain to solidify links to sync and storage service

Acquires icloud.net from Asian social network, which closes at month's end

Apple this week took administrative control of the icloud.net domain, the last notable web address it did not govern that users could have linked with its online sync and storage service.

According to WHOIS searches today, Apple acquired control of icloud.net on Tuesday.

Apple already ruled the primary top-level domains for iCloud, the cross-device, cross-OS service that stores files generated by iOS and macOS, and more importantly, synchronizes everything from Safari browser bookmarks to photographs between iPhones, iPads and Macs. Apple is on record as the owner of the domains icloud.com, icloud.org, icloud.us and icloud.eu, for example.

icloud.com was acquired by Apple in June 2011, just days before then-CEO Steve Jobs unveiled his firm's new cloud-based service. Apple purchased the domain for more than $5 million from a Swedish company, Xcerion, which had used the URL for an online file-storage service.

Today, icloud.net continued to direct visitors to a small Asia-based social network. But that will not last long: The site's homepage displayed a message that said the service would be shuttered at the end February and that all user data would be "destroyed at [sic] March 1, 2017." An earlier message from the site's owner, Tong Lei, had designated Feb. 20 as the shut-down deadline.

"We decide to close iCloud.net and stop all its services," Tong wrote on Feb. 14, omitting any mention of a transfer to Apple.

Questions left in the comments section of Tong's post that asked how many users the social network had and how much Apple paid for the domain, went unanswered.

This story, "Apple grabs icloud.net domain to solidify links to sync and storage service" was originally published by Computerworld.

Windows 10's Mail app users start seeing the Focused Inbox organization feature

Some Windows 10 users are starting to see a Focused Inbox feature show up in the Mail app.

Sick of manually organizing your messages? Microsoft was on the case, apparently testing its Focused Inbox feature in Windows 10's Mail app. As of Feb. 22, though, it's live for both Office 365 subscribers and Outlook.com accounts.

A number of users recently told MSPowerUser they were seeing the feature on their PCs. The testing isn’t restricted to Windows Insider users either, as some are apparently reporting the new feature as showing up on the mainstream build of Windows 10.

Microsoft previously said that Outlook.com users included in the early rollout of Focused Inbox would start to see the new feature in Windows 10 Mail in February.

The feature first appeared on Outlook for Android and iOS. Focused Inbox is designed to cut out extraneous mail (such as newsletters and promotional blasts) to put more important messages front-and-center. It also learns from your mail habits as you move messages back and forth between the “focused” and “other” inboxes.

The story behind the story: Although Microsoft is rolling out this feature in the Mail app it’s not clear when all users will see it. It’s likely this feature will not roll out for some time unless Microsoft plans to include it as a built-in feature for the Windows 10 Creators Update this spring. In addition to the Windows 10 Mail app, Focused Inbox will eventually land on Windows 10 Mobile, Office 365 online, and Office 365 subscribers on Windows and Mac. 

More to come

For commercial Office 365 customers, Microsoft expects to finish the roll out to the Mac and the web version by May. Select commercial users on Windows 10 Mobile, meanwhile, will start receiving the Focused Inbox in April with rollout to all users to be completed at an unscheduled time after that. Outlook 2016 for Windows is even worse off than its mobile counterpart, with no set date for a Focused Inbox release.

In its Feb. 22 update, Microsoft also added additional features:

  • If you mention someone by name (via a "@" character) in the body of your email, that person will be "mentioned" and will be notified that there's something he or she needs to be aware of.
  • Calendar apps now can be color-coded to help you organize them better.
  • You can now subscribe to "interesting calendars," such as the schedule for your local sports teams. When you create a new calendar entry, you may also see little assistants pop up, such as an option to add Skype for online meeetings or location suggestions from Bing.
  • Mail now includes cards to help you track packages or check in for a flight.

This story was updated on Feb. 22 by Mark Hachman to note that Focused Inbox has gone live for Windows 10 Mail, and to note new features of the Mail and Calendar apps.

This story, "Windows 10's Mail app users start seeing the Focused Inbox organization feature" was originally published by PCWorld.

Amazon Kindle Paperwhite Is Just $79.99 For a Limited Time If You Buy Refurbished - Deal Alert

Amazon Kindle Paperwhite Is Just $79.99 For a Limited Time If You Buy Refurbished - Deal Alert

Amazon has discounted its refurbished Paperwhite models for a limited time, so with the current deal you can get one for just $79.99. That saves you $30 on the typical price of a refurbished model, and saves even more over buying it brand new ($119.99 new). A Refurbished Paperwhite E-reader is refurbished, tested, and certified by Amazon to look and work like new. The popular Kindle Paperwhite has a higher res 300ppi screen, WiFi, a built-in adjustable light, a long lasting battery, and best of all it's glare-free even in bright sunlight so you can use it literally anywhere, night or day. See the discounted refurbished Paperwhite model on Amazon to learn more and explore buying options.

This story, "Amazon Kindle Paperwhite Is Just $79.99 For a Limited Time If You Buy Refurbished - Deal Alert" was originally published by TechConnect.

Alleged Sony Xperia XA2 passes through the FCC

Alleged Sony Xperia XA2 passes through the FCC

Alleged Sony Xperia XA2 passes through the FCC

We have already seen spy shots of what is said to be the Xperia XZ successor, as well as the Xperia XA one. Now, the latter apparently made a stop over at the FCC.

Sadly, this yields little to no actual new information on the handset itself. It is somewhat interesting to note that Sony appears to be sticking with its usual placement for the FCC ID on a hidden label, next to the SIM slot. If, however, the document does indeed describe the Xperia XA2 or XA (2017), it makes an unveiling at MWC 2017 a lot more plausible.

Sony Xperia XA2 (allegedly): Square-shaped and Nougat-powered Sony Xperia XA2 (allegedly): Square-shaped and Nougat-powered Sony Xperia XA2 (allegedly): Square-shaped and Nougat-powered Sony Xperia XA2 (allegedly): Square-shaped and Nougat-powered
Sony Xperia XA2 (allegedly): Square-shaped and Nougat-powered

As for device specs, we already have leaked images, renders and videos to work with. Allegedly, the XA2 will be powered by the MediaTek Helio P20 chipset coupled to 4GB of RAM. On the back it will have a2 3MP f/2.0 camera, on the front a 16MP f/2.2 one. Other suggested specs include a USB Type-C port and a 3.5mm audio jack and Android 7.0 Nougat. Rumors put the price at about CNY 2,200 ($320/€300).

Samsung Galaxy S7 and S7 edge on AT&T start getting Nougat update

Samsung Galaxy S7 and S7 edge on AT&T start getting Nougat update

Samsung Galaxy S7 and S7 edge on AT&T start getting Nougat update

Shortly after T-Mobile started pushing it out, AT&T has also begun rolling the Samsung Galaxy S7/S7 edge Nougat update. Weighing in at around 1.6GB, the update arrives as version G930AUCU4BQA6 for the S7 and G935AUCU4BQA6 for the edge variant.

Aside from bumping the Android version to 7.0, the update also brings along some other changes, including network performance enhancements, Samsung Cloud and Samsung Pass, as well as Android security fixes for the month of February.

5.5” Xiaomi also spotted on TENAA, is this another Redmi Note device?

5.5” Xiaomi also spotted on TENAA, is this another Redmi Note device?

5.5” Xiaomi also spotted on TENAA, is this another Redmi Note device?

Earlier today we told you about a Xiaomi device spotted at TENAA named Xiaomi MAE136, but apparently the Chinese regulatory agency approved another, bigger phone. Its codename is Xiaomi MBE6A5 and the specs reveal 5.5” FullHD display and a deca-core Helio X20/Helio X25 CPU with Mali-T880MP4 GPU. The RAM is 3GB and the internal storage is 64 GB and the MBE6A5 also has a microSD slot.


The listing also reveals that the device will run MIUI 8.0 based on Android 6.0 Marshmallow and feature two SIM slots - microSIM and a hybrid nano/microSD. There is one camera on the back with 13 MP sensor and dual-tone LED flash and a 5MP front one.

The fingerprint sensor is below the camera but there are also sensor buttons on the front. The battery is 4000 mAh and the color options are Gold, Rose Gold, Matte Black and Dark Gray. It is not yet clear when the phone should be expected.

Apple to start manufacturing iPhone SE in India

Apple to start manufacturing iPhone SE in India

Apple to start manufacturing iPhone SE in India

Last month Apple reportedly selected Wistron as the primary manufacturer for all its iPhones in India. The Taiwanese partner of the Cupertino-based company set up a plant in Bengaluru in order to start the local production.

According to industry sources speaking to Reuters, iPhone SE assembly will begin in the coming months. Apple’s midranger is almost an year old but is still the cheapest devices the company offers in India.

Apple shipped 2.5 million iPhones to India last year which is the most the company has ever sold. It ranks at 10th at India’s smartphone market but last year lead the premium segment with a 62% market share.

BlackBerry reacts to the Gartner report that said it has 0.0% market share

BlackBerry reacts to the Gartner report that said it has 0.0% market share

BlackBerry reacts to the Gartner report that said it has 0.0% market share

Yesterday market research firm Gartner released its latest report on the state of the mobile world. One interesting tidbit in there was that for the fourth quarter of 2016, sales of devices running BlackBerry OS reached an unflattering 0.0%.

Today the Canadian company has sent us its official reaction to that report. Here it is:

As has been widely communicated, our current strategy focuses on providing state-of-the-art software and security for enterprise and devices. As part of this aggressive pivot to software, we have shifted away from hardware and toward a licensing model where partners develop hardware and distribute and market the BlackBerry brand. We are very excited about the new BlackBerry branded smartphones that our licensing partners - TCL, BB Merah Putih and Optiemus - will soon be introducing.

This is all previously known information of course, but we assume BlackBerry thought it was useful to reiterate following the barrage of media articles saying that it's reached 0.0% market share. Things aren't as black and white as that number seems to imply, however. First off, we're talking about devices running BlackBerry OS here - not all BlackBerry-branded smartphones. The newest of those all run Android as you may know.

Then it's good to note that while Gartner did say the market share for the OS was 0.0% in Q4 2016, that doesn't mean there were 0 sales of handsets running it in that period. The 0.0% number is a rounding, there were in fact 207,900 devices with BlackBerry OS sold in the last three months of last year. But of course, BlackBerry itself is now focused on software for Android phones, and licensing that and its brand name to partners such as TCL, which is going to fully reveal the next BlackBerry smartphone at Mobile World Congress later this month.

Hydrogen OS 3.0 for OnePlus 3 and 3T is now live

Hydrogen OS 3.0 for OnePlus 3 and 3T is now live

Hydrogen OS 3.0 for OnePlus 3 and 3T is now live

Chinese manufacturer OnePlus just released version 3.0 of its Hydrogen Android OS. This new version is based on Android 7.0 Nougat and some users have been reporting slightly better battery life, compared to the latest Nougat OxygenOS update.

H2OS 3.0 H2OS 3.0 H2OS 3.0 H2OS 3.0 H2OS 3.0
H2OS 3.0

If this is the first time you are hearing of Hydrogen OS or H2OS, it is more than understandable. It is the Android ROM OnePlus develops specifically for its local market. It features some UI differences to better suite local tastes and also lacks Google Play Services and the rest of the Google Apps package, since those don't really work in China. The latter might very well be the reason why certain users have been reporting better battery endurance after flashing the eastern ROM.

Google Services running on H2OS 3.0 Google Services running on H2OS 3.0
Google Services running on H2OS 3.0

If you are already running H2OS on your OnePlus 3 or 3T, then the new version should be available as an OTA. OnePlus has been promising to merge the two branches for some time now, but no progress has been reported on the matter, yet. However, you can flash it on international units as well. The official OnePlus forum has the download links and there is an English language option for the UI. If you decide to mix things up and go for it, we suggest checking out this XDA thread as well. Also, if you do use Google services on your device, be sure to pick up a Gapps package to flash alongside the 1.2GB ROM. OnePlus has a version of the Google Play Store in the Hydrogen OS app store for an easy install as well.

LG G Pad III 8.0 starts getting Nougat update

LG G Pad III 8.0 starts getting Nougat update

LG G Pad III 8.0 starts getting Nougat update

The Nougat update for the LG G Pad III 8.0 has been released. Specifically, Canadian carrier Fido has started rolling out the update to units on its network. The update bumps the Android version to 7.0.

Given the fact that it's an OTA roll out, it might take some time for the update to hit your device. Meanwhile, if you feel impatient, you can manually check for the update by heading to your handset's Settings menu.

Sunday Q&A

Sunday Q&A

Sunday Q&A

Welcome to the week 7 edition of our Sunday Q&A. This week we talk the Samsung Galaxy S7 edge Nougat update, the resolution vs pixel size debate and benchmarks.

Ravi: With the release of the Huawei Mate 9 and the Cortex A73 cores inside it, the mighty Kryo cores inside Snapdragon 82x chips finally had a competitor that fared better in benchmarks. However, even the latest and greatest offering from ARM didn't even come close to the raw single-core computing power offered by A10 fusion cores inside iPhone 7/7+.

My question is how did Apple manage to create a chipset that is way ahead of its competitors when at the end of the day, even they rely on ARM licensed procedures and techniques to make their chipsets? Is it perhaps because of the optimised software, or is it because of plain old cheating like in the recently revealed cases of OnePlus and Meizu? Any insight will be greatly appreciated.

The most important thing to note here is that benchmark results aren’t directly comparable between platforms. Even those cross-platform benchmark tests warn that scores obtained under Android and iOS aren’t meant to be directly compared.

That said, there’s no doubt that the A10 is a beast of a chipset with beastly single-core performance. Part of the reason is that it only has two high-power cores with each of those much larger footprint than the Snapdragon and Kirin CPU cores. There’s also more to performance than just the architecture – stuff like L2 and L3 cache size and memory speed, which isn’t normally officially announced by manufacturers so we don’t have the full picture.

Finally in day to day operations software optimizations are just as important as sheer processing power and Apple having full control over the hardware and software is having quite the advantage there. When you know the specific hardware and software that you will run it’s much easier to extract the best performance. Compare that to ARM, who designs cores that should power an enormous number of devices – from high-end smartphones through entry-level ones to stuff like smart refrigerators and Wi-Fi routers.

Nikolas: 1. In iPhone 6s camera spec, you write 1.22 µm pixel size on iPhone 6s. How about iPhone 7? Do they have the same pixel size?

2. iPhone 5s has 8 MP camera, and 1.5 µm pixel size, while iPhone SE has 12 MP camera, but smaller pixel size (1.22 µm) Which one is better? And which one is more important for picture quality in general - resolution or pixel size?

1. The iPhone 7 has the exact same sensor size and resolution as the iPhone 6s, which means pixel size is also identical between the two.

2. That a question as old as digital photography itself. The short answer is higher resolution leads to more fine detail in the shots, while smaller but larger pixels bring wider dynamic range and less noise. So we can’t simply say that one is better than the other – it all depends on the specific use case. If you are doing most of your shooting in good lighting then a higher-resolution sensor might yield better results. If you shoot mostly in low-light environments you will lose most of the resolution advantage anyway so the larger, less noisy pixels of the lower-res sensor will give you better shots.

There’re also other things to consider – putting a very high-res sensor on a smartphone means you also need the optics that are able to resolve as much detail to take full advantage of it. And that’s really asking a lot from a tiny lens that can fit inside a modern ultra-slim smartphone.

Finally, there’s the thing that the images a smartphone produces have just as much to do with the decisions manufacturers make regarding their processing as with their sensors. To give you one example – making images more contrasty and with more saturated colors leads to an increase in the noise levels as well. It’s all about finding the right balance and it really isn’t a one-size-fits-all thing so some will inevitably prefer one cameraphone, while others will go for another.

SaRPeR: Can you test the Sony Xperia XZ's battery life again after the Nougat update? Lots of people are saying Xperia XZ's battery life improved drastically after updating to Nougat?

Unfortunately we don’t have the Sony Xperia XZ unit with us anymore as we had to return it following our review. That said, we are looking to obtain another unit so we can retest its battery life after the update and if we succeed we’ll bring you the updated scores as soon as possible.

Alien: A lot of trolls are saying that Sony's mobile division is dying. Do you agree with them or is it just jealousy? Would you use the Xperia XZ as a daily driver?

Judging by its financial results lately the Sony Mobile division is certainly going through a rough patch. I wouldn’t go as far as to say that it’s dying, but if it keeps losing money Sony might consider selling it to reduce its losses. Hopefully that won’t happen though as the company has brought plenty of innovation through the years and in recent times, so I hope it finds a way to improve its results.

As for the Xperia XZ – I’d gladly use it as a daily driver as it’s really an awesome. Its exorbitant initial pricing meant it was out of reach for many people, but not that it’s settled down a bit, it’s actually a pretty good value flagship.

Vineet: Is it really wise to update the Samsung Galaxy S7 Edge to Nougat, while it looses battery life with the latest OS. Besides looks of UI nothing has improved so much.

We have a whole article detailing the changes brought by the Nougat update on the Galaxy S7 edge. The short answer is yes – the update is worth it, but you should probably check the complete review to know about all of the changes and then decide for yourself if it’s worth it.

Anonymous: Hello guys. I have a really burning question that I need to ask and hope you can answer it, so I can compare phones specification. My question is which Soc is better for gaming? Snapdragon 430? Snapdragon 625? Snapdragon 650? HiSilicon Kirin 655? Mediatek Hellio X10? MTK Hellio X20? And which of this SoCs is power efficient when playing intense games?

It’s a rather tough question to answer in a straightforward way. Power-efficiency seems to be the easier bit – the Snapdragon 625 has an advantage here with its 14nm manufacturing process, with the Kirin 655 coming a close second with its 16nm build. At the other end of the spectrum the MediaTek Helio X10 and Snapdragon 430 and 650 use the archaic 28nm process so they use the most power for the same amount of work. The Helio X20 is in between, on 20nm, and since it has a tri-cluster structure it actually does decently.

In terms of sheer performance We’d suggest you check out our mobile chipset guide to get a better idea on how they compare. Depending on the use case, the Snapdragon 625, 650, the Helio X20 and the Kirin 655 could all be the best.

If you have any questions for us - you can send them by using the form over here.

Nougat update starts hitting Huawei Honor 8 units in India

Nougat update starts hitting Huawei Honor 8 units in India

Nougat update starts hitting Huawei Honor 8 units in India

Huawei has started rolling out a new update to Honor 8 units in India. Arriving as version FRD-L02C675B310, the update brings along Android Nougat (version 7.0) and EMUI version 5.0. The update was being beta tested until now, but now the beta phase has come to an end.

"Honor 8 with Nougat and Emui 5.0 is open to users in India now," the Chinese company said in a blog post. "FRD-L02C675B310 became open to public a week ago. The same version will be available to all users in batches by end of February."

Global roll out for the Honor 8 Nougat update began last month, and units in the US started getting it last week.

ZTE Hawkeye Kickstarter campaign has been cancelled

ZTE Hawkeye Kickstarter campaign has been cancelled

ZTE Hawkeye Kickstarter campaign has been cancelled

ZTE has announced that they have decided to phase out the Kickstarter campaign for the Hawkeye smartphone that was revealed by the company last month. The announcement came in the form of an update to the phone's Kickstarter page.

The company, however, said that the project isn't over. "We are reevaluating the device for the winning Project CSX idea – an eye-tracking feature with self adhesive backing – and it will be implemented based on your feedback."

This comes less than a month after the Chinese firm acknowledged that their decision to implement eye-tracking feature with self adhesive backing on a mid-range phone didn't meet expectations of those who backed the project.

The handset carried a price tag of $199, and less than 200 people had pledged support, which translated into a total of $36,245, compared to the project goal of $500,000. Those who pledged support will get refunds.

As for what's next, ZTE is likely to implement the hands-free features on a higher spec’d device, but there's no word on when such a product would arrive.

"Based on your feedback we have determined it is best to push back the date of the release in order to deliver a higher spec’d device. While we are still finalizing the new date, we will continue providing updates and collaborate with the community throughout the entire process."

Silicon Valley execs grow a backbone

Silicon Valley execs grow a backbone

After a period of uneasy silence, tech leaders are moved to rally in opposition to the Muslim ban

Since November, leaders in Silicon Valley have been trapped in an uneasy détente, pussyfooting around a president-elect who opposes the tech industry on myriad key issues. While their employees organized and protested, most tech execs remained silent or issued platitudes about cooperation.

When Silicon Valley leaders were summoned to the White House for a tech summit in December, Kara Swisher berated them for climbing the "Trumplethinskin Tower" in silence. "Now [they] have to become reality show star[s] in a new episode of 'The Apprentice: Nerd Edition,' bowing and scraping to that luddite Trump, who will probably simultaneously berate [them] in person and bully [them] on Twitter later with a lot of poop emoticons," she wrote.

After denouncing the "symphony of compromise," Swisher excoriated the execs to "realize again that you have the smarts and invention and the innovative spirit to do whatever you like. Realize you have untold money and power and influence and massive platforms to do what you think is right. Realize that you are inventing the frigging future."

Trump disagrees with the tech industry on net neutrality, encryption, privacy, trade, and many social issues. Venture capitalist Sam Altman cautioned: "Think now about what action President Trump might take that you would consider crossing a line -- and write it down."

That line in the sand proved to be immigration.

When push comes to shove

More than 125 companies signed an amicus brief filed in support of a Washington state lawsuit against a Trump executive order that bans travelers from seven Muslim-majority countries from entering the country. Tech execs fell over each other in their rush to issue statements opposing the order.  

Trump's ban strikes uncomfortably close to home for tech companies, many of which have CEOs who were immigrants. Immigrants or their children founded more than 200 of the companies on the Fortune 500 list, companies that collectively generate annual revenue of $4.2 trillion and employ millions of Americans.  

The brief spells out the threat to American business: "The Order makes it more difficult and expensive for U.S. companies to recruit, hire, and retain some of the world's best employees. It disrupts ongoing business operations. And it threatens companies' ability to attract talent, business, and investment to the United States."  

The Computing Research Association also issued a statement, warning of the potential brain drain.  "[The ban] may discourage foreign-born researchers from bringing their talents to the U.S. in the future, which would have significantly detrimental impacts on our national competitiveness."

Leading the leaders

But well before tech company leaders were moved to take a stand, frustration among their employees was fueling protests and walkouts at work campuses ranging from Google to Palantir.  The election "awakened the nerdy set" who fear tech companies might end up serving Trump's interests, Recode wrote. In December, more than 2,840 tech workers signed the Never Again pledge, which stated their refusal to cooperate in creating a database of Muslims -- or people of any religious belief -- or aiding in mass deportations.

IBM employees launched a petition to put pressure on their company (which did not sign the amicus brief). Many are skeptical they'll see action from CEO Ginni Rometty, who sits on Trump's business advisory council. "There are employee petitions that have gone ignored. People have openly quit because of [Rometty's] stance on Trump. People have quit without jobs," a source with knowledge of the inner workings of IBM told Gizmodo. 

Employee protest can prove powerful, however. "Tech companies go to extraordinary lengths to recruit and retain employees; those employees have a lot of leverage," Y Combinator's Altman wrote. "If employees push companies to do something, I believe they'll have to."  

Activist group Tech Against Trump is currently planning a tech-employee walkout on March 14, and grassroots groups like Tech Solidarity are also organizing to fight Trump. The building momentum will be "impossible for tech companies' management to walk back," Valerie Aurora, a former Intel engineer and an activist in Silicon Valley, told Quartz. "It feels like the tipping point was last week. I'm sad that it took this direct threat to the business models for direct management to act, but the workers were already there," she said.

Principle leavened with pragmatism

When principle aligns with self-interest, Silicon Valley companies have stood together. Tech companies united behind Apple last year in its fight with the FBI over encryption. The unifying factor then was a "shared resistance to government-imposed orders on software security built into devices," writes Forbes. "Now, it's the widely held belief in Silicon Valley that immigrants have and continue to be a driving force in the American economy -- or at least, for the Valley's own bottom line."

It's also possible the industry believes it can reap political capital from opposition to the travel order.  Computerworld's Patrick Thibodeau calls the ban "a PR gift" for the industry, which will use it as "a cudgel against H-1B reform."

The industry doesn't like Trump's ideas for visa reform, he continued, because they would raise wages and "make H-1B visa workers less indentured and give them more flexibility to quit one firm and move to another without risk of deportation." The industry wants H-1B caps to increase, with no change in the rules.

"If [Silicon Valley] really wants immigration, then [it should] support green cards, not temporary work visas," Thibodeau said. "The industry says it supports both. It's lobbying, however, is direct[ed] heavily on H-1Bs."

Anaconda Fusion melds Excel with Python data science

Anaconda Fusion melds Excel with Python data science

A new offering from the creator of a Python distribution for data science uses Microsoft Excel as a front end for Jupyter notebooks and other data-centric apps

Some of the most creative uses for applications are never intended by their makers. Microsoft Excel may not have been devised as a scratchpad for data scientists, but it's been pressed into the role with such zeal that Microsoft has gone with the flow.

Continuum Analytics, makers of a Python distribution called Anaconda preferred by data science users, has released Anaconda Fusion, a system for connecting the enterprise-grade version of Anaconda with Microsoft Excel 2016 and higher. This bridges Excel data accessible to Python and makes Anaconda's visualizations and other functions available inside Excel.

Spreadsheets are familiar territory for those in business analytics -- environments like Python and R, less so. If data scientists have work they want to share with Excel jockeys, it'll be easier to do so through Excel than to have them come to the Python side.

Data scientists can expose their work to Excel users with Python code and data available in Jupyter notebooks. Functions in a notebook can be decorated by a standard Python decorator syntax to indicate they're available to Excel users, and data supplied to those functions can be open-ended (that is, any data range that can be listed in Excel) or closed (a choice from a list).

When an Excel user interacts with those functions, the results -- plain-text data, or graphical visualizations such as those created by packages like Bokeh -- are shown in a tabbed side panel in Excel. With the side panel, it's also easier for Excel users to explore notebooks already available or to upload existing notebooks into Fusion. This is not simply a convenience; it acclimates Excel users little by little to Fusion and to Anaconda and Python is general.

Most of the previous integrations between Excel and Python have allowed Excel to interoperate with Python as a data source, such as the Xlwings project. Fusion might also be thought of as the obverse: It lets Python use Excel as a front end for data-driven applications and functions built with Python.

AMD: Sorry, there will be no official Ryzen drivers for Windows 7

AMD: Sorry, there will be no official Ryzen drivers for Windows 7

Some PC users will again be forced to choose: the latest hardware or Windows 7?

AMD won’t be providing Windows 7 drivers for its upcoming Ryzen processors, the company said, contradicting recent reports that indicated AMD would support Microsoft’s older operating system.

AMD confirmed that it has tested and validated Ryzen on Windows 7, but that it won’t officially support the OS. It puts to rest the suggestion of a recent translated report from Computerbase that AMD would reverse its stance and ship Windows 7 drivers for Ryzen.

“To achieve the highest confidence in the performance of our AMD Ryzen desktop processors (formerly code-named ‘Summit Ridge’), AMD validated them across two different OS generations, Windows 7 and 10,” AMD said in a statement in response to a question from PCWorld. “However, only support and drivers for Windows 10 will be provided in AMD Ryzen desktop processor production parts.”

AMD is maintaining a position that it, along with Intel and Microsoft, has held for the last year. In January 2016, Microsoft said that Intel’s Kaby Lake and AMD’s Ryzen would only be supported under Windows 10, and reiterated that position last August. 

Why PC enthusiasts should care: Though Windows 10 is Microsoft’s most modern OS, with support for the latest APIs and drivers, a dedicated base of users has stuck with Windows 7. If AMD supplied Ryzen drivers for Windows 7, all would be well: Windows 7 users could continue to use their older, stabler OS. AMD’s messaging, however, is a return to the status quo: If you want to run AMD’s latest chip, you should have Windows 10, too.

windows 10 pc gaming game mode Microsoft

Microsoft’s busy building features like Game Mode into Windows 10 to lure PC gamers onto its most modern OS. 

The latest and greatest

The reason Microsoft gave for tying Windows 10 to Kaby Lake and Ryzen was simple: Silicon and software needs to be designed and shipped in close conjunction, so that the software can support the features of the processor and vice versa. Windows 7, for example, simply couldn’t anticipate features that chip vendors would include later. Microsoft stopped mainstream support for Windows 7 in 2015, and extended support will end in 2020. (As it is, Microsoft only grudgingly added support for Intel’s prior Skylake chip on Windows 10 after customer complaints.)

“As new silicon generations are introduced, they will require the latest Windows platform at that time for support,” a Microsoft spokeswoman said last August. “This enables us to focus on deep integration between Windows and the silicon, while maintaining maximum reliability and compatibility with previous generations of platform and silicon.” 

At this point, AMD has “verified”, or tested, that a Windows 7 PC powered by Ryzen will boot. But will it take advantage of all of Ryzen’s capabilities? That seems less likely. Consider: Gamers are always seeking the most up-to-date GPU drivers to eke out the last little bit of performance from their graphics card. A Ryzen desktop won’t have to worry about properly supporting the sleep states that a notebook might, but there’s still a decent chance of something breaking.

As my colleague Brad Chacos notes, systems powered by Kaby Lake have booted an unsupported OS, namely Windows 8. Ryzen may as well. But there are two issues: compatibility, and performance. Even if Windows 7 boots Ryzen, a lack of support means that any bugs will probably not be patched. And it’s totally unknown how a Ryzen system will perform compared to a Windows 10 Ryzen PC, and if all of Ryzen’s features will be available on Windows 7.

Unfortunately, now that AMD has clarified its position, users remain at a crossroads: Upgrading a Windows 7 PC with a Ryzen CPU—but without proper drivers—seems like cutting off one’s nose to spite Microsoft’s face.

Updated at 9:23 PM with additional details.

This story, "AMD: Sorry, there will be no official Ryzen drivers for Windows 7" was originally published by PCWorld.

Intel's new Xeon chip has 24 cores and sells for $8,898

Intel's new Xeon chip has 24 cores and sells for $8,898

The Xeon E7-8894 v4 is a monster server chip with 60MB of cache and a maximum clock frequency of 3.4GHz

No Intel chip has ever been as expensive as the new Xeon E7-8894 v4 server processor.

The $8,898 Xeon chip has massive horsepower with 24 cores, 60MB of cache and a maximum clock frequency of 3.4GHz. Intel said this is the company's fastest server chip, breaking enterprise application speed records.

The company's next expensive chip after the E7-8894 v4 is its other 24-core processor, the Xeon E7-8890 v4, which is priced at $7,174. The chips have similar features except for the base clock speed. The new chip starts at 2.4GHz compared to 2.2GHz for the less expensive chip.

The $8,898 chip even outprices Intel's fastest supercomputing chip, the Xeon Phi 7290F, which is priced at $6,401. It is also over four times more expensive than the costliest PC chip, the $1,723 Core i7-6950X for gaming desktops.

Some price competition could come from AMD, which is reentering the server market with Zen-based chips in the second quarter. Its initial server chips code-named Naples have up to 32 cores. AMD has not revealed the target market for Naples, but it could be cloud providers.

During an earnings call, AMD indicated Naples chips would be competitively priced, and downplayed its margins expectations. AMD is projecting Zen chips to be high performance, but the company is looking for volume shipments and may not participate in the low-volume market of servers with more than eight sockets.

The Xeon E7-8894 v4, which is based on the Broadwell architecture, is priced high for many reasons. It is targeted at fault tolerant servers used by financial or retail companies which need highly reliable systems for transactions and fraud detection. These companies could lose hundreds of millions of dollars if a server crashed.

The new chip also has features not found in regular PC or server chips, like error correction and RAS (reliability, availability and serviceability), which can diagnose and troubleshoot issues without crashing servers. The chips also have high levels of I/O and networking integration.

Right now server buyers don't have bargaining power with Intel, which has more than a 90 percent market share in server chips. Intel charges a hefty premium for the performance and adjacent technologies it bundles, including server and networking gear.

The E7 v4 chips aren't high in volume shipments, and are made in smaller batches. But these are highly profitable chips and fuel revenue for Intel, which is now relying more on data center equipment than PCs for growth in the future.

Server chips already have a high markup, but average prices have been going up in recent years and will continue to rise, said Diane Bryant, executive vice president and general manager of the Data Center Group for Intel, during a speech at a investor day meeting in Santa Clara, California, on Thursday.

Prices have been going up due to new applications like machine learning and analytics and the growth of business cloud services, Bryant said.

"Because they see value in our high-end products, they are buying up the stack," Bryant said.

But server purchases have stalled after years of growth, and Bryant projected a 5 percent decline in server CPU shipments going into 2021. The company isn't expecting to grow from server CPU shipments, but from adjacent product shipments to telecom and cloud providers.

Competition from AMD has been worked into the server chip price and shipment projections, Bryant said.

Intel plans more potentially powerful and expensive chips this year. The company will ship a deep learning chip called Knights Mill and a separate chip called Lake Crest that will integrate Xeon with a deep-learning chip based on technology acquired from Nervana Systems.

Microsoft lawsuit against indefinite gag orders can proceed

Microsoft lawsuit against indefinite gag orders can proceed

A federal judge has ruled that Microsoft can sue the US to tell customers about searches

A Microsoft lawsuit against the U.S. Department of Justice over indefinite gag orders attached to search warrants can proceed, following a federal judge’s ruling on Thursday.

The tech titan sued last year to end the government’s practice of indefinitely blocking it from informing customers of search warrants for their information. Microsoft alleged that such orders violate its First Amendment frees speech rights and the Fourth Amendment privacy rights of its users.

The Justice Department argued that Microsoft couldn’t bring either of the claims in a motion argued in front of the judge two weeks ago.

Judge James Robart from the U.S. District Court for the Western District of Washington said Microsoft’s First Amendment claims could proceed, but its Fourth Amendment claims could not.

Microsoft said that it was pleased with the ruling. The DOJ did not immediately respond to a request for comment.

“We’re pleased this ruling enables our case to move forward toward a reasonable solution that works for law enforcement and ensures secrecy is used only when necessary,” Microsoft President and Chief Legal Officer Brad Smith said in an emailed statement.

At issue in the case are gag orders allowed by a section of the Electronic Communications Privacy Act, which allows the government to indefinitely prohibit a cloud vendor like Microsoft from telling one of its users about a search warrant for their digital information.

Robart said Microsoft had standing to continue with its argument that the gag orders are an unconstitutional prohibition on its speech. However, because of legal precedent, Microsoft could not assert a Fourth Amendment claim on behalf of its users, the judge said. That decision presents a conundrum, as Robart noted in his opinion.

“As Microsoft alleges, the indefinite nondisclosure orders allowed under [the law] mean that some customers may never know that the government has obtained information in which those customers have a reasonable expectation of privacy,” he wrote.

It’s been a busy month for Robart. He issued a restraining order blocking parts of President Donald Trump’s executive order banning immigration from seven countries with large Muslim populations.

Apple proposes GPU API for 3D graphics on the web

Apple proposes GPU API for 3D graphics on the web

WebGPU proposal would create a standard API for exposing GPU features to web browsers

Apple is promoting the development of a standard API that would allow web browsers to access GPU features for 3D graphics. 

The team behind Apple's WebKit browser engine proposed a community group at the W3C (World Wide Web Consortium) called the W3C Community Group for GPU on the Web that would be charged with providing an interface between the web browser and modern 3D graphics and computation capabilities in native system platforms.

"The goal is to design a new web API that exposes these modern technologies in a performant, powerful, and safe manner," the community group proposal states. "It should work with existing platform APIs, such as Direct3D 12 from Microsoft, Metal from Apple, and Vulkan from the Khronos Group." Beyond generic computational capabilities being exposed in existing GPUs via the API, Apple also wants to investigate shader languages to produce a cross-platform solution.

Browser engine developers, GPU hardware vendors, 3D software engineers, and anybody else interested in 3D graphics are invited to participate. As part of the launch of this effort, Apple is sharing its WebGPU API proposal. It will also be releasing a prototype of the API to the WebKit open source project.

The WebKit team anticipated a next-generation graphics API several years ago and began prototyping in WebKit to show it could expose a low-level GPU API to the Web and still improve performance. Apple's effort so far is not expected to become the actual API in the standard.

W3C released a statement in support of the effort. Google and Microsoft did not comment Wednesday on whether they might participate in Apple's effort. Mozilla said Apple's proposal would be useful in the design process, but it was far too early to say more.

The current standard for rendering graphics on the web is WebGL. However, since WebGL was created, both GPU technology and software APIs for tapping the power of these processors have improved. These newer APIs -- such as Direct3D, Metal, and Vulkan -- generally offer better performance than WebGL, but they are not available across all platforms. Apple's proposal attempts to remedy this by creating a standard API that can be implemented on top of many systems. 

Got secrets? Docker Datacenter can help your applications keep them

Got secrets? Docker Datacenter can help your applications keep them

Docker Datacenter now supports secrets in containerized apps -- API keys, passwords, and encryption keys -- and ensures they are securely stored and transmitted

The latest version of Docker Datacenter, Docker’s end-to-end container service for enterprises, now solves a common problem with containers: How to store secrets—API keys, passwords, encryption keys—in containerized apps without betraying those secrets.

Docker's solution, Docker Secrets Management, provides a standardized interface for storing and retrieving secrets in containerized apps. It's compatible with both container-native apps and legacy apps that have been containerized.

A nice, safe place for everything

By providing a safe place to store secrets, Docker enhances enhances not only security but convenience. Adding secrets directly to application source code is tough to maintain and highly insecure, whether your apps are inside containers or not. It’s also convenient to have a single, consistent manner to store and retrieve secrets throughout an organization. Docker believed these items were the responsibility of the platform provider and set out to address them.

Docker Secrets runs wherever Docker Datacenter runs—in a public cloud, on-premises cloud or bare metal, hybrid cloud, and so on. Secrets are always encrypted when at rest, are only delivered by way of a TLS connection, and are only provided to applications that are authorized to receive them.

When secrets are delivered to an app, they’re exposed by what amounts to a virtual file system, so legacy apps that store secrets in a configuration file can make use of the system. That said, secrets are never written to disk unencrypted. In fact, when used by an application, they’re not written to disk at all; the virtual file system is functionally similar to a RAM disk.

Who else has a secret to share?

If you’re running Docker apart from Docker Datacenter, it’s still possible to add secrets using the docker secret command. Docker Secrets doesn’t require you to integrate the secret into a container, which is handy if you're dealing with secrets that change often, such as API keys. Secrets is also intended to be end-to-end secure, so secrets are never exposed in plaintext at any point.

Third-party products like HashiCorp Vault or Kubernetes’ secrets system can also provide secrets to apps in Docker containers. However the latter, in particular, has shortcomings that Docker Secrets seems aimed at addressing, namely that secret data is stored as plaintext; peer-to-peer communication of secrets isn’t by default secured with TLS (it’s possible to do so, but it’s not the default); and secrets can be read too easily by applications they’re not intended for.

Docker Secrets is currently designed only to use its own internal store, presumably for the sake of keeping tight control over the delivery chain for secrets. However, in a Hacker News thread devoted to the product, Docker security team member Diogo Mónica noted that support for external stores is in the works, “First implementation will probably be w/Vault, but we would love for this to come from the community,” Mónica said.

Ford to invest $1B in A.I. startup toward self-driving cars

Ford to invest $1B in A.I. startup toward self-driving cars

The automaker is far from alone in spending on Silicon Valley software makers

Ford announced today it's investing $1 billion over the next five years in an artificial intelligence (A.I.) startup founded by former Google and Uber employees to further the development of autonomous vehicle technology.

The massive investment will make Ford the majority stakeholder in Argo AI, but the automaker said the software company has been structured to operate with "substantial independence."

Ford said its relationship with Argo AI, which was founded last year in Pittsburgh, will combine its existing autonomous vehicle development program with Argo AI's robotics and "startup speed" on artificial intelligence software.

mark fields autonomous driving Ford Ford

Ford CEO Mark Fields poses with the autonomous Fusion at the company's research lab in Palo Alto, Calif.

Argo AI founders CEO Bryan Salesky, and COO Peter Rander are alumni of Carnegie Mellon National Robotics Engineering Center and former leaders on the self-driving car teams of Google and Uber, respectively.

Argo AI's team will include roboticists and engineers from inside and outside of Ford working to develop a new software platform for Ford's fully autonomous vehicle, expected in 2021. Ford said it could also license the software to other carmakers.

Ford’s model is similar to what was announced a few months earlier by Volvo and Autoliv (a huge auto supplier) to develop autonomous vehicle software that would be used by Volvo but also sold to other companies, according to Michael Ramsey, a research director at Gartner.

"In the end, the car companies don’t want to license anything from another company unless they have to. In a few years, there’s going to be some big consolidation because there are more makers than buyers," Ramsey said.

Ford's acquisition of Argo AI should be viewed more of as an organizational structure move and not an investment in this company.

"The company and its leaders have been acquired to run Ford’s driver-control module for autonomous vehicles. The investment is plugged in over five years, and designed to give an equity reward to employees," Ramsey stated in an email to Computerworld. "Ford is trying to create a system where they can effectively hire top talent and reward it, outside of the bureaucracy that exists inside the company. It shouldn’t be seen as a $1 billion gift to two guys."

The current team developing Ford’s virtual driver system – the machine-learning software that acts as the brain of autonomous vehicles – will be combined with the robotics talent and expertise of Argo AI. This partnership, Ford said, will work to deliver the virtual driver system for Ford’s SAE level 4 self-driving vehicles.

sae automation levels detailed graphic SAE

The Society of Automotive Engineers (SAE) International, a U.S.-based industry standards organization, has established six autonomous driving categories where Level 0 represents no automation and Level 5 is a fully autonomous vehicle that controls all aspects of driving previously performed by humans.

While Ford already has a substantial in-house autonomous vehicle program, Ramsey said the carmaker needs leadership and the ability to attract and retain more talent.

"Argo will act somewhat independently and hopefully be able to ramp up talent," Ramsey said.

Ford will become a majority stakeholder in Argo, and John Casesa, Ford's group vice president of global strategy, and Raj Nair, the automaker's product chief and chief technical officer, will sit on Argo's five-person board.

By the end of this year, Argo will have offices in southeastern Michigan and California, according to Ford, along with its Pittsburgh headquarters. It will employ more than 200 workers at those three sites combined.

chariot ford autonomous Ford

Last year, Ford purchased Chariot, a San Francisco-based on-demand shuttle service, which also operates in Austin, Texas and is planning to expand to eight other cities.  

Another subsidiary created last March, Ford Smart Mobility LLC, will take the lead on the commercialization of Ford's self-driving vehicles, which includes options for using autonomous vehicles to move goods and people, such as ride sharing, ride hailing or package delivery fleets.

Ford Smart Mobility also purchased Chariot, a San Francisco-based on-demand shuttle service, which also operates in Austin, Texas and is planning to expand to eight other cities.

"The next decade will be defined by the automation of the automobile, and autonomous vehicles will have as significant an impact on society as Ford's moving assembly line did 100 years ago," Ford CEO Mark Fields said in a statement. "As Ford expands to be an auto and a mobility company, we believe that investing in Argo AI will create significant value for our shareholders by strengthening Ford's leadership in bringing self-driving vehicles to market in the near term."

autonomous fusion mcity 39a9958 hr 100677146 orig Ford

A Ford self-driving prototype being tested at the University of Michigan's Mcity prvoing grounds.

Software is the key differentiator in the functionality of semi-autonomous and fully autonomous vehicles. The average vehicle today has about 50 computer processors that control everything from engine control units (ECUs) to Advanced Driver Assistance Systems (ADAS) such as parking assist and adaptive cruise control.

ADAS and fully autonomous vehicle technology are the sweet spots for Silicon Valley software developers, whose code can tie together the myriad cameras and sensors needed for a car to drive itself.

Ford has already made significant investments in autonomous driving technology, including opening an R&D center in the heart of Silicon Valley.

Ford is only one in a long list of carmakers that have invested in the area. Since 2011, BMW, Honda, Hyundai, Mercedes-Benz, Nissan-Renault and Toyota have all opened R&D centers in Silicon Valley.

silicon valley auto autonomous vehicles IDG, Wall Street Journal, Carmaker data

Automakers who've set up shop or invested in Silicon Valley software companies to develop self-driving car technology.

General Motors opened its Advanced Technology Silicon Valley Office in Palo Alto to develop an HTML browser for its Cadillac CUE in-vehicle infotainment (IVI) system.

Nissan is focused on developing autonomous vehicles at its Silicon Valley facility, while Honda's operations there are working on human-machine interface technology, big data, connected vehicles and cybersecurity.

Not only are the companies opening up R&D facilities, they're also recruiting rock star security and system engineers from the mecca of computer development, Jon Allen, a principal at management and tech consulting service Booz Allen Hamilton, said in an earlier interview with Computerworld.

Meanwhile, a former mid-level Apple engineer heads Ford's new Palo Alto R&D center, just down the street from electric-car maker Tesla Motors.

GM silicon valley General Motors

GM opened its Advanced Technology Silicon Valley Office in Palo Alto to develop an HTML browser for its Cadillac CUE in-vehicle infotainment (IVI) system.

"This isn't your grandfather's automotive company any more. The [car makers] are moving away from simply being hardware manufacturers to becoming software developers," Allen said. "Fundamentally, the auto industry cannot be seen as just automakers any more. They're mobile developers."

"And, look at who Apple is hiring. They're hiring auto executives," Allen added.

This story, "Ford to invest $1B in A.I. startup toward self-driving cars" was originally published by Computerworld.

Kategori

Kategori