All the pieces of Snap’s grand plan are coming together, the latest of which appears to be putting its camera technology atop drones. 

At least that’s the obvious conclusion based on Snap’s acquisition of Ctrl Me Robotics for “under $1 million,” a deal that was first reported by Buzzfeed on Friday. Snap has not commented on the deal or the acquisition price.  

Although the report calls the Venice Beach-based company a drone manufacturer, a subsequent report from Variety indicates that Ctrl Me Robotics’ true expertise is customizing existing drones to be able to carry cameras rather than specializing in producing drones. Additionally, the Variety report claims that the deal actually occurred late last year. 

Headquartered in the same “Silicon Beach” neighborhood of Venice as Snap, the drone-focused startup may help Snap further its mission of moving from a mere app company to what it likes to call itself now: a camera company. 

In that respect, it’s beginning to look like Snap make be encroaching on GoPro’s territory, but with far greater reach into the mainstream consumer space as opposed to GoPro’s relatively “enthusiast” action camera user profile. 

Although it’s unclear what Snap’s plans are for the company’s expertise, a least one hint may rest in an Instagram post from Ctrl Me Robotics from almost two years ago. In it, the company shows off a drone fitted with a smartphone to “Snapchat from your drone.”

Monday, 29 May 2017 00:00

Facebook and the ethics of moderation

On The Listening Post this week: With two billion users and 1.3 million posts a minute, Facebook’s content moderation challenges are huge. Plus, we look at the people monitoring and moderating the social web.

Facebook’s status: Tech or media company?

This week, the British newspaper The Guardian revealed hundreds of internal Facebook documents that outline the site’s ground rules for content moderators. From violence to racism, revenge porn to child abuse – the picture that emerges is one of a company struggling with its responsibilities as a media platform and how to cope with more 1.3 million posts per minute, in multiple languages.


Alex Hazlett, deputy managing editor, Mashable
Olivia Solon, senior technology reporter, Guardian US
Jennifer Pybus, senior lecturer, London College of Communication
Richard Millington, founder, FeverBee
Natasha Tiku, Silicon Valley reporter, Wired

On our radar:

  • Turkish journalist Murat Celikkan has been sentenced to 18 months in prison for producing “propaganda for a terrorist organisation”. He was one of 56 people who took part in an “editor for a day” solidarity project in 2016 at a pro-Kurdish paper, Ozgur Gundem.
  • It’s been open season on journalists in Mexico and yet another has been abducted there. Salvador Adame Pardo, director of television station 6TV in the central city of Nueva Italia, was captured by armed men and forced into an SUV on May 18. He has not been heard from since.
  • In the Gulf state of Qatar, there’s been what the government says is a hack of the state news agency, QNA, this past week, after a fake statement was published on its website ascribing false comments to the country’s ruler. Qatar says it has opened an investigation into the hack and “will hold all those involved accountable”.

Scrubbing the net: The content moderators

Many social media users assume that content moderation is automated, that when an inappropriate image or video is uploaded to the net a computer removes it. In reality, there are thought to be more than 150,000 content moderators working around the world today. It can be unpleasant but necessary work, and many social media companies based in the West now outsource it to countries such as the Philippines or India.

But the question is: do they do that responsibly? Or do they just take advantage of the cheap labour with little consideration for the person doing the work?

Most malware requires some form of active user interaction in order to infect a device — a click on a link in a phishing email, or the installation of software from an unverified source. 

But a new type of attack, dubbed Cloak and Dagger, can basically take over your Android phone without your (conscious) help. Worse, no major version of Android is safe at this time. 

Described by a team of researchers from the University of California and the Georgia Institute of Technology, Cloak and Dagger relies on the way Android UI handles certain permissions. 

f an app is downloaded from Google’s Play Store, researchers claim, it is automatically granted the SYSTEM_ALERT_WINDOW permission, aka “draw on top.” You’ve likely seen this permission in action — it’s used by Facebook’s chat heads, which float over other content on your screen.

This can be used to hijack the user’s clicks and lure her into giving the app another permission, called BIND_ACCESSIBILITY_SERVICE or a11y, which can be used for stealing your passwords and pins, for example. 

A hacker that combines both these vulnerabilities could silently install a “God-mode” app with all permissions enabled, including access to your messages and calls.  

Even though a lot of this is intended behavior and not an actual exploit, it can definitely be used to take over someone’s device. The researchers claim they tested it on 20 human subjects, none of which had realized what was going on. 

The one thing that protects users right now is the fact that to do all this, the malicious app must be downloaded from Google’s official Play Store, meaning that it has to pass Google’s security checks. But from past examples we know it’s definitely possible for malicious hackers to slip in a malware-infested app into Play Store. 

“It is trivial to get such an app accepted on the Google Play Store.”

“A quick experiment shows that it is trivial to get such an app accepted on the Google Play Store,” the researchers claim. “We submitted an app requiring these two permissions and containing a non-obfuscated functionality to download and execute arbitrary code (attempting to simulate a clearly-malicious behavior): this app got approved after just a few hours (and it is still available on the Google Play Store),” they wrote. 

While Google has partially fixed the issue in the latest version of Android (7.1.2), the researchers claim it’s still fully possible to take advantage of the vulnerabilities described above. According to the researchers, these aren’t “simple bugs” but “design-related issues,” meaning it will take more time to fix them; moreover, Google considers some of these issues as features, and does not currently plan to fix them. 

To protect their devices, the only thing users can do right now is check which apps have access to the “draw on top” and a11y permissions. The steps to do this vary in different versions of Android; they are listed here

You can breathe easier now, iPhone fans: Apple might have solved the biggest challenge facing your favorite upcoming smartphone.

Apple has reportedly figured out how to integrate an optical Touch ID sensor directly into the iPhone 8‘s display, clearing the one hurdle that tripped up Samsung’s Galaxy S8

The report comes from Chinese language Economic Daily News (EDN) via Mac Rumors, citing sources from within the Taiwan Semiconductor Manufacturing Company (TSMC), an Apple supplier.

Previous reports suggested the iPhone 8 would be delayed beyond its expected September unveiling and release because of difficulty with an embedded fingerprint sensor. But EDN‘s sources claim that Apple will be ready for rollout in the second half of 2017, right on schedule.  

The report also claims the new iPhone will also feature “invisible infrared image sensors” to boost the performance of its high-pixel camera and enable new AR applications. We’ve heard about this before, too — the IR sensors might be used for a new way to unlock the phone via facial recognition, along with more fun applications for games and tricked-out AR selfies. 

The iPhone 8’s screen size could get a boost from more than just the removal of the physical home button and the edge-to-edge design. EDN‘s sources echoed another recent leak claiming Apple will make the move to a larger aspect ratio from its current 16:9 setup, shifting to an 18:5:9 aspect ratio, just like the Galaxy S8.  

An embedded Touch ID will avoid marring the iPhone’s otherwise sleek, largely button-free design with an awkward rear-positioned sensor. Recently leaked images out of China claimed to show off a finalized version of the iPhone 8 sporting a Touch ID button on its back — but if this new report is true, those images can be dismissed.

Instead, the design we’re more likely to drool over at the iPhone 8’s unveiling will align with the other dummy version floating around the rumor mill, complete with an edge-to-edge display, no visible bezels or buttons, and a powerful, vertically-oriented dual lens camera on the back. Wireless charging, a bigger battery, and a super premium price are also expected for the deluxe anniversary version of the phone, too.   

If Apple really did crack the embedded Touch ID code, the iPhone 8 will look even more attractive when compared to the Galaxy S8, which is a damn good phone even with the awkward sensor. The placement did little to hurt S8’s sales — but every advantage Apple can claim over its competitors could make the expected super cycle of iPhone sales even stronger. 

Well that’s no good.

It turns out vaping may be bad for more than just your look. 

With a few tweaks of the pen, a security researcher has demonstrated that vaporizers can be modified in such a way as to pass code to your computer. 

The problem, as with many things security related, comes down to the USB port. Used for both charging and data transfer, the port is a convenient place to plug in phones or other devices that need a battery boost—devices like vape pens. 

In a video demonstrating his work, the researcher, who goes by FourOctets, plugs an e-cigarette into a computer’s USB and the device immediately lights up as if to charge. A few seconds go by and the computer starts to react. 

“DO U EVEN VAPE BRO!!!!!,” reads a message that pops up on the screen. 

Essentially, the vaporizer issued a custom command to the computer, and the computer was all too happy to oblige. 

Take this as the weirdest example yet that you should never plug random devices into your USB ports.

While FourOctets has no ill-intent, it is easy to imagine someone less scrupulous loading a computer with something not quite as funny. Like, say, a keylogger. Or ransomware

So how did he make this happen? Thankfully for people worried about their e-cigs catching a virus, it required some hands-on work. 

“It started as more of a joke than anything,” FourOctets elaborated over Twitter direct message (he declined to give his real name). “This is done with extra hardware and a little bit of code.”

As to the point of the demonstration, other than the fact that it is legitimately hilarious? 

“Another goal usually when doing dumb stuff like this is that stuff is not always what it seems and that random stuff that can plug into a computer can be dangerous,” he explained. “A lot of folks aren’t aware that something like this is even possible whether it be with firmware or added hardware and a tiny bit of code found online.” 

So should you be worried that your vape pen is delivering malicious code to your laptop? 

“It’s probably pretty unlikely to ever get something like this from the factory that would do this,” FourOctets noted, “but the possibility is there and people need to be mindful of that.”

So, you know, something to maybe consider the next time you’re ripping that sweet cotton. 

Facebook can’t catch a break these days.

The Guardian published a comprehensive tell-all, dubbed the “Facebook Files,” revealing the social network’s guidelines for monitoring violence, hate speech, and revenge porn. Let’s just say it’s raised a lot of questions.

This week’s MashTalk is hosted by Tech Editor Pete Pachal, with commentary from Chief Correspondent Lance Ulanoff, Senior Tech Correspondent Raymond Wong, and Tech Correspondent Jack Morse.

So, the Facebook Files. Yeah, we know, Facebook’s guidelines provided to its moderators for determining what is and isn’t considered acceptable content is worrisome. 

Many people are demanding the company find better solutions to curb hate speech and situations  like people livestreaming themselves committing suicide.

What measures can Facebook take to better control its content? Hire more moderators? Hire local moderators? Can machine learning and AI help in any way? Jack breaks it down.

Features Editor Rebecca Ruiz also phones in to give her insight on whether there are cases where objectionable content such as livestreamed self-harm or suicide can actually be educational or helpful.

What measures can Facebook take to better control its content?

Next, Lance gives us all a preview of DJI’s smallest drone yet, the DJI spark. Priced at $500, the tiny drone — the body’s about the size of a soda can — is designed to appeal to beginners. 

Though it doesn’t have quite the same amount of features and performance as DJI’s foldable Mavic Pro or Phantom drones, the Spark is plenty capable with 16 minutes of flight time, a maximum speed of 31 mph, and a range of 1.24 miles.

And speaking of DJI, the company is now forcing all of its drone users (old and new) to activate their DJI accounts and update their drones to the latest firmware, or risk having its capabilities handicapped (i.e. limited flight up to 100 feet high, disabled livestreaming, etc.). Is DJI being fair or overstepping its reach?

Lastly, we discuss Microsoft’s new Surface Pro (no, not the Pro 5 because that doesn’t exist). Microsoft refreshed its flagship 2-in-1 tablet-PC, but interestingly enough it’s now advertising it as a laptop — which is odd because the company is shipping a proper clamshell laptop, the Surface Laptop, next month.

The guts are all new, the hinge improved, the Surface Pen stylus more accurate, and some of the corners rounded off, but where the hell is the USB-C port? Microsoft’s Panos Panay says the market’s not ready for it. But if a tech giant like Microsoft doesn’t help accelerate USB-C, who will? We hate dongles as much as anyone else, but come on, it’s been two years since Apple released the MacBook with its single USB-C port. If USB-C’s moment isn’t now, then when will it be?

Around 2,000 beacons have been installed across Gatwick Airport’s two terminals providing an indoor navigation system that is much more reliable than GPS and that enables augmented reality way finding for passengers – a world first for an airport.

The lack of satellite signals makes road-based navigation systems – such as Google or Apple maps – unreliable indoors, so Gatwick has deployed a beacon based positioning system to enable reliable ‘blue dot’ on indoor maps, which in time can be used within a range of mobile airport, airline or third party apps. 

The beacon system also enables an augmented reality way finding tool - so passengers can be shown directions in the camera view of their mobile device - making it easier for passengers to locate check in areas, departure gates, baggage belts etc.

The new navigation technology is currently being integrated into some of the Gatwick apps and the airport is also in discussion with airlines to enable the indoor positioning and way finding tools to also feature on their app services.

No personal data will be collected by Gatwick although generic information on ‘people densities’ in different beacon zones may help to improve airport operations including queue measurement, streamlining passenger flows and reducing congestion.



Airlines could go further - and with the consent of their passengers - may send reminders on their airline app to late running passengers, for example, or find out where they are and make an informed decision on whether to wait or offload their luggage so the aircraft can take off on time.

Retailers and other third parties may also use the beacon system to detect proximity and send relevant offers or promotional messages, if the passenger has chosen to receive them.


Battery powered beacons kept logistical complexity and costs low, with deployment taking just three weeks, followed by two months of testing and calibration.

The new technology is part of Gatwick’s £2.5 billion investment programme to transform the airport.

Abhi Chacko, head of IT commercial and innovation, Gatwick Airport, said: “By providing the infrastructure we’re opening the door for a wide range of tech savvy airport providers, including our airlines and retailers, to launch new real-time services that can help passengers find their way around the airport, avoid missing flights or receive timely offers that might save them money.

“We are proud to be the first airport to deploy augmented reality technology and we hope that our adoption of this facility influences other airports and transport providers so that it eventually becomes the norm.”

NASA’s Kepler telescope has confirmed crucial details about the orbital pattern of the seven newly discovered exoplanets in the TRAPPIST-1 system, as well as providing new insight into the system’s least understood planet, TRAPPIST-1h.

Astronomers from the University of Washington were able to confirm a regular orbital pattern using data collected by the telescope. The team found that, at six million miles from its dwarf star, TRAPPIST-1h – the outermost of the seven exoplanets – does not lie within the habitable zone and is likely far too cold for mere humans. The team also confirmed ‘planet h’ orbits its TRAPPIST-1 star every 19 days.


It’s incredibly exciting that we’re learning more about this planetary system elsewhere, especially about planet h, which we barely had information on until now,” said Thomas Zurbuchen, associate administrator of NASA’s Science Mission Directorate.

The international team analyzed raw Kepler data immediately after it was released by NASA on March 8 and published their findings in the Nature Astronomy journal on Monday.

This finding is a great example of how the scientific community is unleashing the power of complementary data from our different missions to make such fascinating discoveries,” Zurbuchen added.

Using data collected previously by the Spitzer Space Telescope, the team first identified a “complex but predictable” pattern in which the six innermost planets orbit the star.


Studying the orbital frequency, the scientists calculated six possible routes planet h could be taking – five of which were eventually ruled out by additional data. Meaning the scientists could predict the planet’s exact orbital period even before Kepler’s observations were released.

NASA first announced the discovery of seven Earth-sized planets, found some 39 light-years away, at a press conference in February. Since then, the race has been on to unlock the system’s secrets.

The space agency plans to study the seven new worlds with the James Webb Space Telescope, scheduled to launch in October 2018. It is hoped the super-powerful instrument will help scientists discover if the planets could possibly support life.

We've officially gone beyond the iPhone 8 specs and supply chain rumor phase — now we're in full-on production photo leak territory. 

New leaked images purported to be the molds used to make the new iPhone models have hit the web, surfacing first on Slashleaks. The pics give us a sneak peek at what could be the final design of the upcoming iPhone 8 — the molds fall in line with what we're expecting from Apple in its next line of phones, and appear to be identical to other recently leaked images of a dummy version of the phone.

The images on Slashleaks aren't the only photos of the iPhone 8 molds. Graphic designer and leaker Benjamin Geskin also posted photos of the molds, which look just like the dummy model he first shared back in April. He tweeted that the upcoming iPhone designs have been confirmed by these images — but that ultimately can't be determined until we hear from Apple itself. 

The molds show the three models expected from Apple in its next iPhone cycle: a 7S and 7S Plus  — with the same dimensions as the current 7 and 7 Plus handsets at 4.7 and 5.5 inches, respectively —and the deluxe iPhone 8 handset measuring in the middle at around 5 inches. The 8 is expected to take advantage of an edge-to-edge design for more screen space, with a projected 5.8-inch display, but there's no insight on the screen size from the molds, which only show the shells of the devices.    

These production molds don't show us anything we haven't already seen from other iPhone leaks. The 7S and 7S Plus rigs have the same camera orientations as the current models, while the 8 has the expected vertically-oriented design that was teased in the photos of the dummy units. 

The leaked images also suggest the next iPhone's fingerprint sensor won't be on the back of the device — Geskin is extremely confident Apple won't repeat Samsung's mistake — but, like the size of the screen, there's no way to know from just a production mold. Geskin speculates the sensor will actually make it under the display, which has been a long held rumor, or it could be integrated into the power button on the side of the device.  

These photos give the rumors of the "finalized" version of the iPhone 8 that've been floating around some more legitimacy, but remember, nothing about the new iPhones is set in stone — not the supposedly exorbitant price or even its name — until Apple officially unveils the upcoming devices. That probably won't be until September, barring a shocker in June at the World Wide Developers Conference (WWDC), so a real, 100 percent confirmation is still months away.

How Facebook moderates violent and racist posts is in the spotlight following the allegedly racially motivated killing of a young black man and document leaks showing what the social network allows to remain on its platform.

The Guardian revealed details of Facebook’s guidelines to moderators on Sunday, providing insight into how the company moderates graphic content including violence, hate speech, terrorism, pornography and racism.

The news organization reportedly saw more than 100 internal documents including Facebook training manuals, spreadsheets and flowcharts.

The guidelines leak comes as news emerged that a young black man was allegedly killed by a man who is a member of a white supremacist Facebook group called “Alt-Reich Nation".

Richard Collins III was killed on the University of Maryland campus on Friday night in what police described as an “unprovoked” attack. The 23 year old was due to graduate from Bowie State University this week.

Sean Christopher Urbanski, 22, has been charged with first-degree murder for his killing. The University of Maryland’s Police Chief David Mitchell revealed that Urbanski was a member of the racist group on Sunday.

“When I look at the information that's contained on that website, suffice it to say that it's despicable, it shows extreme bias against women, Latinos, persons of Jewish faith and especially African-Americans," Mitchell said, as cited by The Baltimore Sun.

The Alt-Reich Nation page appears to have been removed from Facebook following the high profile incident. However, numerous other white supremacist groups can easily be found on the social network, apparently operating within the company’s guidelines.

The leaked files provide detail on types of posts and groups that will be removed from the site as Facebook struggles to strike a balance between removing graphic content and allowing free speech.

Examples of groups allowed to remain on the site because they are not considered “credible threats” of violence are outlined in the files. “I hate it when I wake up and Sarah Palin is still alive” and “lets nuke the middle east” fell into this category.

Among other things the files show that the social network allows users to livestream acts of self-harm because it “doesn’t want to censor or punish people in distress". It also allows the sharing of non-sexual child abuse, and some depictions of animal cruelty. has reached out to Facebook for comment regarding the leaks, but received no response at the time of writing.

Facebook moderators interviewed by The Guardian said the moderation policies are "inconsistent" and "peculiar." They also said they were "overwhelmed" by the work and only had seconds to make decisions on each post.

Earlier this month, Facebook CEO Mark Zuckerberg announced the company would hire 3,000 additional people to help monitor live videos and remove inappropriate content. Zuckerberg said the new staff will help Facebook “get better” at removing content like “hate speech and child exploitation.”

Page 3 of 7
We use cookies to improve our website. By continuing to use this website, you are giving consent to cookies being used. More details…