Android Lollipop Now Running on 12.4 Percent of Active Devices

android_lollipop_interface_google.jpg
The market share of Google’s latest version of Android, Lollipop, has been slowly but steadily increasing since its launch last year. As per the latest numbers from Google, Android 5.0 Lollipop and Android 5.1 Lollipop now cumulatively account for 12.4 percent of active Android devices.Google publishes its Android distribution numbers every month, based on data gathered from the Google Play store app, which supports Android 2.2 and above. For its latest report, lists active devices that checked into Google Play during the seven-day period ending June 1.

As mentioned before, in this period, Google found 12.4 percent of active Android devices ran Android Lollipop, specifically, 11.6 percent ran Android 5.0, while 0.8 percent ran Android 5.1. This is up from 9.7 percent overall (an increase of 2.7 percent) in May, with 9 percent for Android 5.0, and 0.7 percent for Android 5.1.

android_distribution_numbers_june_2015_google.jpgDevices running Android 4.4.x KitKat registered 39.2 percent share of active devices, a fall of 0.6 percent from the 39.8 percent in May. Android Jelly Bean on the other hand (v4.1.x to v4.3.x) was spotted running on 37.4 percent of active devices, dropping 1.8 percent from 39.2 percent in May. Android 4.1.x is seen on 14.7 percent devices (down from 15.6 percent), Android 4.2.x on 17.5 percent (down from 18.1 percent), and Android 4.3 on 5.2 percent devices (down from 5.5 percent).

Android 4.0.x or Ice Cream Sandwich, in Google’s June latest numbers, registered a share of 5.1 percent, down 0.2 percent from May’s 5.3 percent. Android Gingerbread (v 2.3.3-2.3.7) and Android 2.2 Froyo’s device distribution shares are 5.6 percent (down 0.1 percent) and 0.1 percent (the same as May) respectively in the report.

The OpenGL version, on the other hand, saw OpenGL 2.0’s share drop from 64.1 percent in May to 63.0 percent in June, while OpenGL version 3.0 rose to 35 percent from to 34.5 percent in May. OpenGL version 3.1 registered a share of 2 percent.

Chinese Researchers Develop ATM With Facial Recognition Technology

atm_facial_recognition_china_nbc_24_youtube_screenshot.jpg
Chinese researchers have successfully developed the first automated teller machine (ATM) with facial recognition technology to reduce the risk of theft, media reports said.

The developers include Tsinghua University and Tzekwan Technology, a Hangzhou firm in eastern China’s Zhejiang province that provides security protection for financial transactions, South China Morning Post quoted Chinese official media as saying.

Tzekwan chairman Gu Zikun, an anti-counterfeit technology expert, believes the technology will curb ATM-related crimes.

China currently relies mostly on imported ATM technology, the report said, but the new machine – which combines high-speed banknote handling, improved counterfeit-bill recognition and facial recognition – was wholly Chinese.

Gu said the product had passed the authorities certification and would soon be available on the market. It is unclear who will manufacture the ATMs and how it will collect facial data.

The news come a week after the state launched its “Made in China” campaign, which aims to transition the mainland from a manufacturing hub for low-end goods to high-quality products within the next 10 years.

Cash machines using fingerprint authentication have sprung up in countries like Chile and Colombia. These biometrics ATMs are not being used by some countries, such as the United States, because of privacy concerns and its high cost.

The new ATMs are expected to connect with the country’s banks and public security networks, which guarantees that only cardholders withdraw money, even if someone else knows the password.

But opponents to the technology have taken their concerns about privacy and accuracy online. “What happens if someone had plastic surgery to look like someone else,” one user asked.

Google’s Project Vault Puts a Secure, Encrypted Computer Inside a microSD Card

google_atap_vault_io2015_screengrab_google.jpg
One of the most interesting announcements made at Google I/O was Project Vault, a security-focused computer concealed within an ordinary microSD card. When plugged into a compatible PC, smartphone or other device, it can enable completely secure end-to-end communications. Google’s Advanced Technology and Projects (ATAP) group designed Project Vault to protect a user’s most sensitive data, under the assumption that if not everything can be safe, at least some of it can.The fingernail-sized device runs on a custom ARM processor and has 4GB of secure storage space. It runs its own security-centric real-time OS with built in cryptographic infrastructure including a suite of encryption services and a hardware random-number generator. It has its own NFC hardware including an antenna, which can be used to authenticate users based on a physical token in their possession. The microSD interface suffices for all data input and output.

Google claims the Vault is completely transparent to its host devices and shows up as any ordinary storage destination. It is completely device- and OS-agnostic and only needs to be plugged in to work. Compatible apps can interact with two dedicated files: one that must be written to and the other that can only be read from. The rest of its file system is fake, and any interaction with it will result in a standard “bad sector” error, preventing even the host device from seeing what goes in and out of the Vault. No special drivers or user intervention are required. This means that a user’s security is maintained even if he or she moves the Vault between multiple host devices, no matter how insecure they might be.

The device and its software support text messaging, voice, and video streaming, as long as parties on both ends have their own Vault cards. Immutable hardware logging features would help owners determine if anyone has tried tampering with their Vault. Google is using 500 prototype cards internally and hopes to have commercial products out soon, with enterprise customers targeted first and consumer applications rolling out later. An open-source kit including development hardware and source code is already available.

Nazara Technologies Hires Former Reliance Games CEO Manish Agarwal

Manish_Agarwal_CEO_nazara.JPG
Nazara Technologies, one of India’s oldest mobile game companies has announced that erstwhile Reliance Games CEO Manish Agarwal will be joining the company as CEO. Nazara’s current CEO and founder, Nitish Mittersain will continue on as MD.

Last week we reported that Agarwal was parting ways with Reliance with Amit Khanduja taking his place. At the time industry sources indicated he would join Nazara Technologies, which is looking to raise between $50 to $100 million in funding, as confirmed by sources in Nazara, as well as other sources in the Indian gaming industry. Additionally, our sources have told us that Agarwal will receive equity in Nazara as a part of the deal.

“As an industry as well as a company, we are at an inflection point today. While on the demand-side, the adoption of smartphones and enhanced connectivity is helping the segment grow at an unprecedented pace, on the supply-side there is tremendous interest in the Indian market from global players, new entrants, content developers and investors. To remain competitive in this market, we need to scale up our operations and increase the management bandwidth, and this is where Manish will play a role. We are certain that his experience of running a large mobile gaming business, expertise of developing world-class games and passion for the mobile game developer ecosystem will help in taking Nazara to new heights,” said Mittersain in a prepared statement.

Speaking about his appointment, Agarwal, said in the same prepared statement, “Nazara has been a pioneer in the Indian mobile gaming space. With an in-depth understanding of the gaming ecosystem and a highly talented team, the company is well poised to shift gears and enter a phase of faster growth. I am glad to be a part of Nazara, and together with Nitish will help steer the company and set it on course to becoming a world class mobile gaming company.”

Uber, Ola Cabs Directed to Submit Details of Drivers, Vehicles to Ply in Delhi

indian_taxi_driver_with_uber_app_reuters.jpg
Giving a last chance to app-based taxi operators Uber and Ola Cabs, Delhi Government on Thursday directed them to submit full details of their drivers and vehicles within seven days if they want to “regularise” their services in the national capital.According to a senior government official, applications of Uber and Ola Cabs to get licences are currently pending before the transport department.

Transport Minister Gopal Rai on Thursday held a meeting with representatives of app-based taxi operators and other taxi associations at the Delhi Secretariat.

“Government has asked Uber and Ola Cabs to give it all details pertaining to their vehicles and drivers within seven days if they want their applications for licences to operate their services to be considered by the concerned authority,” said a government spokesperson.

All app-based cab service were banned in the national capital after a driver of Uber, the US-based online taxi-hailing company, was accused of raping a female passenger in December last year.

In January this year, the company had applied for a radio taxi licence to operate through its subsidiary Resource Expert India Pvt Ltd.

The city government has also directed these app-based taxi operators to file an undertaking stating that they would follow all laid down rules in future.

“We have also asked Uber and Ola Cabs to file an undertaking stating that they will follow all rules laid down rules by the transport department,” the official said.

The Delhi government in March had approached Department of Electronic & Information Technology (Deity) to block web-based apps of taxi operators Uber, Ola Cabs and TaxiForSure alleging that these companies are not complying with a ban order imposed by it.

Thereafter, the Department of Telecom had earlier this month issued a letter to Internet Service Providers to block websites of taxi operators Uber, TaxiForSure and Ola Cabs, following a request from the Delhi government.

Google Announces A/B Testing for Play Store Listings; Improved Developer Pages

google_play_store_ab_testing_androiddevblog.jpg
Amongst the many announcements made at its 2015 I/O Conference, Google has taken the wraps off a handful of improvements designed to help developers attract more traffic to their app listings and convert more visits into downloads. The two biggest features are A/B testing for listings and an improved developer page experience.Google will now allow ‘Experiments’ from the developer console, which let developers choose up to three alternative sets of graphics and text descriptions for their app listings. These alternative listings will be shown to some visitors, and statistics about engagement and installs will be shown to developers so that they can determine which version of the listing was most appealing and effective, thereby optimising potential. According to Google, some developers achieved double-digit install rate improvements in a pilot program conducted prior to the announcement.

Developer pages are also to be overhauled with a large banner graphic, company logo, descriptive text and improved catalogue display. Companies will also be able to feature one app or update prominently on their page to drive traffic. Those with lots of apps in the Play Store will benefit the most, and will be able to use their developer page as a homepage to promote all offerings in one place.

Google stated that it intends to make developer homepages more prominent and easier to find. One way to do this will be the creation of sections based on broad interests such as shopping, with subsections such as coupon apps and fashion apps. These will appear in search results to increase the chances of users discovering new and unfamiliar apps that suit their requirements.

There will also be a new easy way for people to discover family-friendly apps and content. A new star icon will lead users to a section highlighting apps approved for the Designed for Families program, sorted by suitable age ranges.

To further boost user acquisition, developers will be able to create and roll out ad campaigns right from the developer console. Only a budget needs to be set, after which Google will scale the campaign out across all its properties including Google Search, AdMob, YouTube and new Google Play search ads.

Developers will also be able to submit apps for testing across a wide variety of actual Android devices that they otherwise would not have access to. Given the fragmentation in terms of screen size, resolution, CPU power and OS version, developers would be able to identify and fix real-world issues that would otherwise not be identified till someone encountered them and complained.

New ‘Deep Learning’ Robot Learns Skills Through Trial and Error

pr2_brett_uc_berkeley_robot_learning_lab.jpg
Researchers at the University of California-Berkeley have developed algorithms that enable robots to learn motor tasks through trial and error – much like humans learn things.They demonstrated their technique, a type of reinforcement learning, by having arobot complete various tasks – putting a clothes hanger on a rack, assembling a toy plane, screwing a cap on a water bottle, and more – without pre-programmed details about its surroundings.

“What we’re reporting on here is a new approach to empowering a robot to learn,” said Pieter Abbeel.

“The challenge of putting robots into real-life settings, like homes or offices, is that those environments are constantly changing. The robot must be able to perceive and adapt to its surroundings,” added co-researcher Trevor Darrell.

The researchers turned to a new branch of artificial intelligence known as deep learning.

This helps the robot recognise patterns and categories among the data it is receiving.

In the experiment, the researchers worked with a Willow Garage Personal Robot 2 (PR2) which they nicknamed BRETT.

They presented BRETT with a series of motor tasks, such as placing blocks into matching openings or stacking Lego blocks.

When given the relevant coordinates for the beginning and end of the task, the PR2 could master a typical assignment in about 10 minutes.

When the robot is not given the location for the objects in the scene and needs to learn vision and control together, the learning process takes about three hours.

In the world of artificial intelligence, deep learning programs create “neural nets” in which layers of artificial neurons process overlapping raw sensory data, whether it be sound waves or image pixels.

Instagram Starts Sending ‘Highlights’ Emails to Users

instagram_new_logo_reuters.jpg
In a recent development, Instagram started sending an email digest called ‘Highlights’ featuring a few of the posts from people you follow.It could bring back users who have strayed from the photo network.

This is the first time Instagram has sent any type of promotional or re-engagement email, TechCrunch reported.

To accomplish its mission of sharing moments and fuel its ad-based business model, Instagram needs users to voraciously browse its feed.

Instagram’s strategy for that has been to avoid letting photos slip outside its walls. It decided not to let Twitter render its photos in-line, so people would have to use Instagram.

But now it is reaching out, hoping to meet users half way. ‘Highlights’ could remind people what they are missing when they do not open Instagram.

‘Highlights’ could solve an issue common amongst maturing social networks that show a live, reverse chronological feed of posts. It impacts networks like Instagram and Twitter, but not relevancy-sorted streams like Facebook’s.

The unfiltered feed problem happens when people follow too many oversharers that drown out their real friends who do not post as often.

But those effects can be offset, to a degree, with ways to surface the most popular content in your network.

Instagram earlier this month released its Layout app for Android, an app launched for iPhone back in March.

In April, Instragram launched the @Music Channel to highlight artists. The account @Music will each week highlight an up-and-coming artist or offer a behind-the-scenes look at a more established musician.

Facebook’s Oculus VR Founder Palmer Luckey Hit With Lawsuit

palmer_luckey_facebook.jpg
The founder of virtual reality glasses maker Oculus VR Inc, acquired by Facebook Inc for $2 billion, has been accused of taking confidential information he learned while working with another company and passing it off as his own, according to a lawsuit filed this week.

The plaintiff, Hawaii-based company Total Recall Technologies, said it hired Oculus founder Palmer Luckey in 2011 to build a prototype head mounted display. Luckey signed a confidentiality agreement, according to the lawsuit filed on Wednesday.

Throughout the latter half of 2011 and into 2012, Luckey received feedback and information to improve the design of the display. However, Luckey used information he learned from his partnership when he launched a Kickstarter campaign for his own head mounted display called the Oculus Rift, according to the lawsuit.

The lawsuit seeks compensatory and punitive damages but does not specify and amount.

Facebook’s $2 billion acquisition of Oculus last year was its first-ever hardware deal, as the company sought a way into the fast-growing wearable devices arena.

Oculus is listed as a defendant along with Luckey, and a Facebook representative on Friday declined to comment. The lawsuit was reported earlier by The Recorder, a California legal newspaper.

For its lawsuit Total Recall Technologies enlisted global law firm Quinn Emanuel Urquhart & Sullivan, which also counts Google and Samsung Electronics Co Ltd as major clients. Luckey is accused of breach of contract and fraud, among other claims.

CCI Orders Probe Against Sweden’s Ericsson

ericsson_new_officie_reuters.jpg
Fair trade regulator Competition Commission of India on Friday ordered a detailed probe against Swedish telecom giant Ericsson, after it ‘prima facie’ found the company in violation of the competition norms.

The order follows complaints that the practices adopted by Ericsson with regard to royalty rates were discriminatory as well as contrary to FRAND (fair, reasonable, and non-discriminatory) terms, CCI said.

“The royalty rate being charged by Ericsson has no linkage to the functionality of the patented product rather it has linkage to the final price of the manufactured product in which the patent is being used.

“Ericsson seems to be acting contrary to the FRAND terms by imposing royalties linked with the cost of manufacturing product. Charging of two different licence fees per phone for use of the same technology, prima facie, appears to be discriminatory,” it said.

CCI said that forcing a party to execute non disclosure agreement and imposing excessive and unfair royalty rates, prima facie, amount to abuse of dominance in violation of the relevant section of the Competition Act.

“… the Commission finds that a prima facie case of contravention of the provisions of section 4 of the Act is made out against the Opposite Parties and it is a fit case to be investigated by the Director General.

“Accordingly, the Commission directs the DG to cause an investigation into the matter and to complete the investigation within a period of 60 days from receipt of this order,” CCI said in its order.