Category: site|engadget

Auto Added by WPeMatico

The tech industry’s accessibility report card for 2021

In spite of all the advancements we’ve seen in tech, the industry as a whole has consistently neglected people with disabilities. There have been some improvements, including video call apps like FaceTime, Zoom, Microsoft Teams, Google Hangouts and more adding better support for sign language interpreters and closed captioning. And, this year, Instagram and TikTok finally added stickers that enable automated captioning for speech in videos, too. But major organizations continued to make decisions that exclude people with disabilities. The organizer of E3 2021, for example, failed its deaf and hard of hearing viewers during its live streamed show.

There are too many individual transgressions and improvements to exhaustively detail here. Due to their sheer size, though, tech’s largest companies wield the greatest influence over what the rest of the industry does. By holding them accountable, we have a better chance of seeing widespread change in the way tech thinks about inclusive design. Here’s how Apple, Google, Microsoft, Amazon, Meta (formerly Facebook) and more did to improve the accessibility of their products and services in 2021.

A blue iPhone 13 Pro Max and a starlight gold iPhone 13 Pro standing with their backs facing the camera.


Apple has led the way in inclusive design for years, and in 2021 the company continued to launch new features that made its products easier for those with disabilities to use. In addition to updating its screen reader, VoiceOver, to enable better descriptions of images for the visually impaired, Apple also launched several new products. In May, it rolled out a service called SignTime, which allowed customers to engage sign language interpreters on demand when communicating with customer service representatives (via their browsers at least). The feature is available in the US, UK and France and supports American, British or French sign languages in their respective countries.

Apple also introduced Assistive Touch for the Watch this year, allowing for touch-free interaction with its wearable. The idea is that users can clench their fists or pinch their fingers together to navigate the smartwatch. In practice, Assistive Touch took some learning, and it still may not be feasible for those who don’t have the dexterity or strength to clench their fists to trigger the “double clench” action. But it’s a start, and one that few other smartwatches offer.

For those with very limited range of motion, this year also saw the launch of the first medically certified eye-controlled iPad by Tobii Dynavox. Called the TD Pilot, it’s a case for iPads as large as the 12.9-inch Pro model and comes with Tobii’s powerful eye-tracking sensor, large speakers, additional batteries and a wheelchair mount. Together with iPadOS 15, this will allow those with cerebral palsy, for example, to interact not only with the tablet, but also communicate with others more easily. A window on the other side of the case can display words to show what the user is saying.

Apple also added improvements for hearing aid users with iPhones this year, allowing for bi-directional communication. This means that those who connect compatible hearing aids to their iPhones no longer have to use their handset’s mic to be heard by their callers — the hearing aid can pick up the speaker’s voice, too. So far, Starkey and ReSound have released compatible “made for iPhone” devices.

On macOS, Apple also made it possible to customize the outline and fill color of the cursor so those with visual impairments can more easily tell when the mouse moves or changes shape. The company also expanded its keyboard shortcuts to allow users to control everything on a Mac with a keyboard (no need for a mouse or trackpad).

Finally, Apple added tools for developers using SwiftUI to make their apps more accessible. With this simplified workflow, there are now fewer obstacles in the way when trying to make more inclusive products.

Unfortunately, when Apple released iOS 15, it removed some features from Siri that were “used by many individuals for accessibility purposes,” according to Clark Rachfal. He’s the director of advocacy and governmental affairs for the American Council of the Blind. Rachfal told Engadget that users “could no longer access their calling history, voicemails, emails and messages through Siri” when the OS was updated. The council and its members alerted Apple of these issues, he said, adding that the company said it’s working on “restoring this functionality to Siri.”

Google Pixel 6 and 6 Pro.


Google continued to add tools for people with disabilities across its broad portfolio of products and services in 2021. One highlight was the launch of Project Relate, an Android app that would generate custom voice recognition models for people with severe speech impairments. Then, the app can transcribe, display and read out what the user said. Project Relate is currently in beta, with Google inviting those with atypical speech to sign up as testers.

The company did plenty to improve its existing products, too. In February, it revamped its Talkback screen reader to offer new gestures and voice commands. In March, it announced that the Chrome browser could transcribe audio from the web for users who are deaf or hard of hearing. That transcription would be performed on-device, meaning you could get your captions without worrying about connecting to the cloud.

Later in the year, Google also added 10 languages to its auto-generated image descriptions tool, brought more natural-sounding voices to the “Select to speak” feature in Chromebooks and made it easier to interact with Android devices using facial expressions.

In addition to improving its existing products, Google also explored accessible experiences that could produce learnings for the industry at large. It teamed up with the Royal National Institute of Blind People (RNIB) and The Guardian on Auditorial, which it describes as an experiment in storytelling that “adapts to suit the reader.” It’s a fully customizable experience that Google said will “offer those with visual disabilities an experience that is as comfortable, rich and creative as any other reader.”

According to the company, Auditorial “is intended to pose a question about how much more accessible the world’s information could be, if you could simply tailor every website to suit your personal sensory needs and preferences.” The hope is that this triggers a discussion on how the web could become more inclusive instead of “a one-size-fits-all approach.” Google published its findings in an “Auditorial Accessibility Notebook,” in order to help other publishers learn tips on how to “open up online storytelling to millions of blind and low vision users.”

Google also launched a browser-based game this year called SignTown, which uses your camera to teach you sign language and assess your progress. The game is just “one component of a broader effort to push the boundaries of technology for sign language and Deaf culture.” The company said it’s also exploring building “a more comprehensive dictionary across more sign and written languages, as well as collaborating with the Google Search team on surfacing these results to improve search quality for sign languages.”

Microsoft Surface Adaptive Kit packaging and label.


In 2021, Microsoft surprised us by releasing Windows 11, which it called the “most inclusively designed version of Windows yet.” The new OS brings nicer-looking dark and high contrast themes, plus updated sounds that are more soothing and can be heard by more users. The company also renamed its “Ease of Access” section to “Accessibility” to make assistive tools easier to find. Windows Voice Typing also makes it easier to dictate your messages.

Prior to launching Windows 11, though, Microsoft announced a five-year commitment to “help bridge the ‘Disability Divide’.” It focuses on hiring and educating people with disabilities, as well as building more accessible products. That includes using AI in Word to detect and convert heading styles for blind and low-vision readers, a new navigation pane in Excel for screen readers and expanding Immersive Reader to better convey what’s on PowerPoint slides and notes. It added a new accessibility checker that works in the background and prompts users to fix issues across Microsoft Office apps and Outlook.

The company not only expanded live captioning and transcription capabilities in Teams, but also rolled out support for CART (Communication Access Realtime Translation) captions, as well as the ability to pin and spotlight multiple presenters. Auto-captioning is also available for LinkedIn Live broadcasts.

To make its hardware easier to use, the company launched a new Surface Adaptive Kit in September. The bundle includes tags, labels, keycaps and more to make PC parts and important buttons more identifiable. Like Apple, Microsoft also added sign language support (specifically ASL) to its Stores to assist deaf shoppers. Unlike Apple, however, Microsoft doesn’t appear to offer this for after-sales support yet.

Microsoft is one of few companies in tech that’s transparent about its efforts to improve training and hiring processes for people with disabilities. It made investments via its Urban Airband initiative “to provide affordable broadband, hardware and software to people with disabilities starting in Los Angeles and New York.” Following a pilot at the University of Illinois, Microsoft said it’s expanding to additional tertiary education institutes to “increase graduation rates of students with disabilities in STEM education.” It’s also working to “create best-in-class Universal Design Learning (UDL) environments in STEM education.”

To connect people with disabilities to employers, Microsoft announced that it’s adding new accessibility resources and features to LinkedIn, including a LinkedIn Learning course focusing on “accessibility in the modern workplace.” There were also LinkedIn Coaches events specifically aiming to help job seekers with disabilities find employment opportunities. The company also partnered with Be My Eyes, an app that connects blind and low vision users with sighted volunteers, to make LinkedIn staff available for visual assistance on video calls.

Microsoft also launched an AI for AccessibilityLow-Cost Assistive Technology Fund to make assistive technology reachable to those who can’t afford it. Considering how expensive assistive technology currently is, this is a promising step towards getting people the gear they need. Though it’s still limited in its reach, the Fund is at least an acknowledgement of the high price that people with disabilities continue to have to pay to be part of the world able-bodied people take for granted.

UNSPECIFIED - SEPTEMBER 28: In this screengrab, Amazon Astro is introduced during Amazon Devices and Services Announcement on September 28, 2021. (Photo by Jamie McCarthy/Getty Images)


Amazon’s accessibility efforts aren’t just aimed at people with disabilities. The company says it pays attention to aging individuals and helping them feel more confident living independently. This year, it introduced two programs as part of its Alexa Smart Properties service that enable administrators to offer voice-assisted experiences in places like senior living facilities and hospitals. The company also launched Alexa Together to keep caregivers and elderly individuals connected via an Alexa-enabled device. It would offer features like fall detection and remote assist to give loved ones peace of mind.

Amazon also updated the Alexa app to offer light and dark modes, as well as text scaling. It rolled out a new option to give people more time to finish speaking before Alexa gives a response, which it said is designed to help people with certain speech impediments. The company also included cards with braille text in the boxes for the Echo Frames 2nd gen, guiding users to a website with screen-reader-friendly setup instructions. On the Kindle app for iOS, Amazon launched a dictionary audio feature to read out individual selected words and help those with learning disabilities or foreign language speakers better understand pronunciations.

This year, the company introduced a new home robot called Astro that follows you around your home and provides easy access to helpful info via its display. It works with Alexa Together to help caregivers look out for loved ones remotely and keeps an eye on your home while you’re away. The robot features similar accessibility features to the Echo Show smart displays and has been trained “to work for customers who use wheelchairs, walkers or canes.” It will also play specialized driving sounds to stop it becoming a tripping hazard.

A Cognixion headset on a white mannequin head.

Amazon also invests in several accessibility-minded projects through its Alexa Fund, including Labrador Systems, which makes home robots to help people with limited mobility live more independently. The company has also worked with neural interface startup Cognixion to add Alexa support to its brain-computer interface headsets for easier smart home device control. Amazon and Voiceitt also released a new speech recognition app this year to help users with atypical speech converse with Alexa and other people.

Though its Alexa-focused products have received many updates to improve accessibility, Amazon’s Prime content appears to have been neglected. According to Rachfal, though Prime TV offers audio descriptions on a large amount of content, many titles use text to speech with synthetic voices. Rachfal added that these “lack the quality of human narrated audio description and the overall quality of the content suffers, making it less enjoyable for blind and low vision consumers.”

To Rachfal, this is an example of something that’s done for people with disabilities “without input, feedback and collaboration with the disability community.”

A sign of Meta, the new name for the company formerly known as Facebook, is seen at its headquarters in Menlo Park, California, U.S. October 28, 2021. REUTERS/Carlos Barria

Meta / Facebook

Amid all the drama surrounding Facebook, its whistleblower and its rebrand this year, it’s easy to overlook the company’s accessibility-related updates. At the start of 2021, the company updated its Automatic Alt Text (AAT) system to recognize over 1,200 objects and concepts in photos on Instagram and Facebook. According to Meta, this represented a 10x increase since AAT’s debut in 2016. It also rolled out additional features to Facebook on iOS that provided more detailed descriptions like positions of objects in a picture and their relative sizes.

Unfortunately, as it pushed out these updates, Facebook may have broken some accessibility features along the way. Rachfal said that when the company turned off its facial recognition system this year, it led to less-informative descriptions for users who are blind or have low-vision. Rachfal said this change “was done due to privacy concerns,” and he believes these decisions were made without considering accessibility and the disability community. “Nor were they given the same weight and consideration as privacy concerns,” Rachfal added.

Facebook published a post addressing this issue in November. In it, the company’s vice president of artificial intelligence Jerome Pesenti wrote, “We need to weigh the positive use cases for facial recognition against growing societal concerns, especially as regulators have yet to provide clear rules.”

A banner showing an example of Automatic Alternative Text on Instagram, with a green box around a picture and words saying

In the post, Pesenti acknowledges the critical role face recognition plays in AAT to help blind and low-vision users identify their friends in pictures. But while some facial recognition tools, like identity verification, will remain, for the most part features like alerting users to photos potentially including them or automatically labeling their friends are going away. That’s for both sighted and visually impaired users.

“We know the approach we’ve chosen involves some difficult tradeoffs,” Pesenti wrote, adding that “we will continue engaging in that conversation and working with the civil society groups and regulators who are leading this discussion.”

Elsewhere in Meta’s family of products, the company added an Accessibility tab to the Oculus Settings menu to make assistive features easier to find. It also brought Color Correction and Raise View tools to offer more legible palettes and enable a standing perspective for seated users respectively. Meta said it’s still iterating on Raise View, working with the Oculus community to improve the feature and will permanently add it to the Accessibility menu when ready.

Meta also collaborated with ZP Better Together, a company that makes technology for deaf and hard-of-hearing users, to bring sign language interpreters into video calls on Portal devices. As of December, people who are deaf or hard-of-hearing can also apply on ZP’s website to get free Portals which will come with the ZP apps.

New models of Facebook's Portal video-calling devices are pictured in this undated handout image obtained by Reuters. Facebook/Handout via REUTERS   ATTENTION EDITORS - THIS IMAGE HAS BEEN SUPPLIED BY A THIRD PARTY. MANDATORY CREDIT
Handout . / reuters

Facebook launched Clubhouse-like audio rooms in the US this year and, notably, did so with live captioning included from the start. It also included a visual cue to indicate who’s speaking, and offers captions for other audio products like Soundbites and Podcasts on iOS and Android.

Let’s not forget the company’s renaming to Meta this year and its new focus on the metaverse. According to head of accessibility Mike Shebanek, “we're already working to bring the metaverse to life and are excited to explore the breakthrough possibilities it presents to make the digital world even more accessible and inclusive for people with disabilities.”

We’ll have to wait and see if and how that comes true, but in the meantime, Meta must continue to engage with the accessibility community to make sure that its expansion of the metaverse is inclusive from the start.


Twitter only set up its two accessibility teams last year, after an embarrassing launch of Voice Tweets that excluded its deaf and hard of hearing users due to a lack of captions. Since then, though, the company has shown noteworthy improvement. In 2021, Twitter introduced captions for voice tweets, added captions and accessibility labels in Spaces and brought automatic video captions. That last one is “available globally in most languages,” according to the company and supported on Android, iOS and the web.

Though this may seem like a small set of updates compared to the rest of the companies in this roundup, Twitter also has a smaller portfolio of products. Still, it has managed to make significant changes. Rachfal praised Twitter as being “the first social media platform to conspicuously prompt users to include alt text with images,” though he did note that filling out the field is still optional.

Other noteworthy developments in tech this year

Alt text and captioning continue to be tricky accessibility features for the industry. They’re labor-intensive processes that companies tend to delegate to AI, which can result in garbled, inaccurate results. This was especially evident at this year’s virtual E3 gaming convention, where illegible closed captions sometimes made the show incomprehensible for those who relied on subtitles to understand the announcements.

There are also large parts of the online world that are in dire need of accessibility-related upgrades. According to a February 2021 study by WebAIM (Web Accessibility In Mind), for example, a whopping 97.4 percent of websites had mistakes that failed the Web Content Accessibility Guidelines (WCAG 2). The most common errors included missing alt text, low contrast text, missing form input labels and more.

It’s not just websites that need work: Other media formats also need to be more inclusively designed. The National Association of the Deaf (NAD), for example, filed a lawsuit with Disability Rights Advocates (DRA) this year against three major podcast providers: SiriusXM, Stitcher and Pandora.

A banner showing three Spotify screenshots with an example of a transcript on the left.

According to the NAD, because the three defendants “do not make transcripts or captions available for any of the podcasts offered on their platforms, more than 48 million deaf and hard of hearing Americans are denied full and equal enjoyment of the content they offer their hearing users.” Meanwhile, Spotify announced this year that it will start offering automatically generated transcripts for podcasts, and Amazon Music launched synchronized transcripts in November.

Then there are entire industries that could use accessibility improvements. Rachfal notes that healthcare is a continually problematic area for people who are blind or have visual impairments. “This is still an entire sector that we hear about far too often from our members,” he said. Given that we are currently in the mires of the third wave of COVID-19, it’s inexcusable to continue excluding people with disabilities when it comes to things like scheduling vaccination or testing appointments.

In November this year, the Justice Department announced it had reached a settlement with Rite Aid to make COVID-19 testing and vaccination websites accessible. Rite Aid’s vaccine registration portal was not compatible with some screen readers and was not accessible to “those who have a hard time using a mouse.” The calendar on its website, for example, “did not show screen reader users any available appointment times,” while people relying on keyboard-based navigation instead of a mouse could not use the tab key to complete a consent form required to schedule an appointment.

The ACB also worked with CVS to offer accessible prescription information in all locations in the country. This includes a Spoken RX feature that would read out prescription labels via the CVS Pharmacy app.

Though there have been many transgressions in the past year, we also saw many promising developments in ensuring technology is inclusive. The FCC, for example, proposed rules in December to make emergency alerts more useful and informative for people who are deaf or hard of hearing.

A TV monitor showing the new Emergency Alert System(EAS) test is seen on November 9, 2011 in Washington, DC. The FCC and the Federal Emergency Management Agency, FEMA, conducted the first-ever nationwide test of the national Emergency Alert System. EAS participants broadcast alerts and warnings regarding severe weather alerts, child abductions and other types of emergencies. The EAS alerts are transmitted over radio and television broadcast stations, cable television and other media services.   AFP PHOTO / KAREN BLEIER (Photo credit should read KAREN BLEIER/AFP via Getty Images)
KAREN BLEIER via Getty Images

Meanwhile, HBO Max launched 1,500 hours of audio-described content starting in March 2021 and committed to including the descriptions to all newly produced original content as well as adding more to its back catalog. Also, in collaboration with the Coalition for Inclusive Fitness, Planet Fitness said it will buy and install accessible exercise equipment in its stores across the country.

I’ve only scratched the surface in this roundup of updates. What’s most encouraging, though, is the increasing willingness of companies to work with disability rights groups and advocates at the earliest stages of product design. Lizzie Sorkin, director of engagement for the NAD, said it’s “seeing more and more companies reach out to us in the beginning phases for input rather than late in the process.” Rachfal also noted a “growing commitment to accessible media and content” that’s “born out of the advocacy work of ACB and the Audio description Project through collaborative discussions with industry.”

Read more

Netflix will be required to stream 20 state TV channels in Russia

Starting in March, Netflix will have to stream 20 state television channels in Russia. Roskomnadzor, the country's media watchdog, registered the platform as an "audiovisual service" this week. Among the channels Netflix will have to carry are the flagship Channel One, entertainment network NTV and a Russian Orthodox Church channel called Spas (which means "Saved").

Streaming services with more than 100,000 daily users in Russia are included on the register, which was established late last year. Not only must registered platforms offer state TV channels, they need to set up a Russian company, according to The Moscow Times.

Companies on the register also have to abide by Russian laws. For one thing, Netflix will not be allowed to promote "extremism." Critics claim that provision has been wielded against those who support the Kremlin's opponents.

Other video services in the country reportedly argued that Netflix should be added to the register to level the playing field, since it meets the requirements. The Russian version of Netflix is operated by Entertainment Online Service, a subsidiary of National Media Group, which has a stake in Channel One.

Engadget has contacted Netflix for comment.

In November, it emerged Russia was investigating a complaint over LGBTQIA+ content on Netflix. The company told Engadget such content was rated appropriately. That same month, Russia ordered several tech giants (including Apple, Google, Meta, TikTok and Twitter) to set up offices within its borders by the end of this year.

Read more

Hyundai built an air purifier out of recycled Ioniq 5 EV parts

Every car is the result of a long development process in which automakers build many pre-production vehicles that never see the light of day. Rather than scrapping one of its Ioniq 5 test vehicles entirely, Hyundai repurposed the car's parts to make an air purifier.

According to a YouTube video description, the model "went through numerous tests to ensure our safety." The video notes that, over the course of a year, the vehicle was used to test the likes of the Acoustic Vehicle Alerting System, pass-by noise regulation and wind tunnel noise.

The video shows Hyundai engineers stripping the Ioniq 5 to its bones, then designing a completely different product using the components. Among other parts, they used the cooling fan, door panels, LED tail lamp, infotainment unit and, of course, the filter unit. The engineers put a 20-inch alloy wheel on the top of the case (so the purifier is probably pretty large), while the car's emblem adds some professional branding.

Although many car parts are already recyclable, including batteries, this is a neat experiment. It suggests there are other sustainable ways to repurpose a car that's otherwise outlived its usefulness. Meanwhile, Hyundai started deliveries of the Ioniq 5 in the US this month.

Read more

Premier League games will stay on Peacock as NBC Sports Network shuts down

When the clock strikes midnight to ring in 2022 on the east coast, NBC Sports Network will shut down. NBCUniversal will shift much of its sports coverage, including Premier League matches, over to the USA Network.

Assuming more games aren't suspended due to teams having too many COVID-19 cases, Premier League coverage will get off to a flying start on USA Network with three matches on New Year's Day. Arsenal will square off against league leaders Manchester City at 7:30AM ET before Watford host Tottenham and Crystal Palace take on West Ham. Sunday brings a match between Everton and Brighton, before Chelsea host Liverpool in a clash between title contenders.

The shift to USA Network shouldn't change much for Peacock users, though. Overflow games and streaming-only matches will remain on that platform, including three matches that are scheduled to stream on Peacock Premium on Sunday. Some of the bigger games will still air on the main NBC network, and NBCU will continue to broadcast every Premier League match across its multitude of platforms.

It's not clear as yet whether there are plans to broadcast overflow games on other NBCU networks. On the last day of the 2020-21 season, when all 10 matches kicked off at the same time, USA Network, CNBC and the Golf Channel each aired one game.

Read more

CES 2022 will end one day early amid COVID-19 surge

After several companies cancelled on-site plans entirely for CES next week, the organization that runs the show announced today that the event will close one day early. The Consumer Technology Association (CTA) said CES 2022 will now run January 5th-7th "as an additional safety measure to the current health protocols that have been put in place." January 3rd and 4th are media-only days when press conferences and keynotes take place. 

Protocols include masks and proof of vaccination to attend in-person events. The CTA also recently announced that it would provide Abbott BinaxNOW COVID-19 tests at badge pickup. The association has asked attendees to complete a rapid test 24 hours before entering a CES venue. If someone experiences symptoms while in a CES venue, the CTA has advised them to report to a first aid station for testing. The organization says it will also provide PCR tests for any attendees who are required to have one by their destination prior to departure from Las Vegas. 

In the last week, high-profile exhibitors like Amazon, Google, Intel, Lenovo, Meta and more have cancelled in-person plans for CES 2022. As part of today's announcement, the CTA explained that over 2,200 companies will still show off their goods on the ground in Vegas, noting that it has added 143 more in the last two weeks. 

CTA CEO Gary Shapiro has defended the decision to keep the show an in-person event as the omicron variant causes COVID-19 case numbers to surge in several parts of the world. “As the world’s most influential technology event, CES is steadfast in its pledge to be the gathering place to showcase products and discuss ideas that will ultimately make our lives better,” said Shapiro. “We are shortening the show to three days and have put in place comprehensive health measures for the safety of all attendees and participants.” Shapiro has pointed to the smaller companies that rely on the event as a potential springboard for products as a key reason for not cancelling the event. CES 2021 was entirely virtual.

"We will all be taking risks," he said in a recent op-ed for the Las Vegas Review-Journal. "But without risk there is no innovation."

Read more

Quest headset owners can capture VR gameplay using their phones

Meta’s Reality Labs unit is rolling out one last major software update for the Quest and Quest 2 before the end of the year. And it’s one you’ll want to download as soon as you can because it adds some handy features.

One of them allows you to record yourself while inside a game or app. If you own a VR headset, you’ve probably seen videos like the one above where you can see how a game physically plays. Before today, you needed special equipment to capture footage from that mixed reality perspective. With the new update, you can use your phone instead.

Mobile mixed reality camera
Reality Labs

Naturally, the final result isn’t as polished as the above video, but you can still get an idea of how games like Beat Saber play out in the real world. You’ll need an iPhone XS or above with iOS 11 or higher to use the new mobile mixed reality camera. With today’s release, about a dozen games support the feature, including Superhot VR, Pistol Whip and Synth Riders.

The update also includes a number of features Meta said were coming “soon” at its Connect conference in late October. To start, you can now make voice calls through the Messenger app on Quest and Quest 2 headsets. The feature allows you to not only call other Quest users, but you can also dial up your Facebook friends.

Reality Labs app sharing
Reality Labs

Starting today, some games will also allow you to back up your save data to the cloud. The feature may not be immediately available on your headset after downloading the update. That’s because Reality Labs says it’s rolling it out at a slower pace to make sure it works correctly. Additionally, it’s an opt-in feature for developers, so not every game may support it even after it’s broadly available. While not new to the Quest platform, today’s update also removes the experimental tag that had been applied to the multi-user and app sharing features that were introduced at the start of the year.

Lastly, while not directly related to today’s update, in the “coming weeks” Reality Labs plans to introduce new customization options to Horizon Workrooms. To start, you’ll have the option to choose from multiple virtual office environments and the ability to decorate the space with custom posters and your company's logo.

Read more

Qualcomm’s Snapdragon 8 Gen 1 will power the next generation of Android flagships

Every December for the last few years, Qualcomm has held an annual event in Hawaii to announce its latest flagship mobile chipset. This year was no different with the company taking the opportunity to unveil the Snapdragon 8 Gen 1. That’s right, for the second year in a row, Qualcomm is moving away from the sequential numbering scheme that has defined its processors for years. Just as the Snapdragon 865 gave way to the 888, the company will now replace the 888 with the Gen 1. 

The system-on-a-chip includes Qualcomm’s own X65 5G modem. The company says it’s capable of theoretical download speeds of 10Gbps. That’s one of those specs that’s impressive on paper, but won’t mean much out in the real world since some of the fastest 5G networks can’t deliver speeds greater than 4Gbps in ideal conditions. If you have access to a WiFi 6 or 6E router, the Gen 1 can sustain download speeds of 3.6 Gbps over WiFi.

As with its past flagship chipsets, Qualcomm has put significant effort into improving the camera experience. The Gen 1 features an 18-bit image signal processor. That’s a first for the company, and something it says allows the component to process 4,000 times more data than the 14-bit Spectra ISP found on the Snapdragon 888. Additionally, phones with the Gen 1 will have the ability to capture photos at 3.2 gigapixels per second. In practice, that means the Gen 1 can process data from three 36-megapixel cameras simultaneously without any shutter lag, according to Qualcomm.

In another first for a mobile device, the company says the chipset can record 8K HDR footage at 30 frames per second. Again, that’s not the most practical feature for a phone in 2021 since 4K is the top end for most content. On that note, the Gen 1 supports UHD capture at 120 frames per second and can record slow motion footage at 960 frames per second at 720p. Separate from its Spectra ISP, the Gen 1 includes a always-on image signal processor that can power a camera while consuming very little battery power. It’s a feature that will allow Gen 1-equipped devices to offer always-on face detection for biometric authentication.

The Gen 1 won’t offer greatly improved CPU performance over what was already possible with the Snapdragon 888 Plus. What it does promise is faster performance when it comes to AI-related tasks. That’s thanks to Qualcomm’s new seventh-generation AI engine, which the company says is up to four times faster than its predecessor thanks to more shared memory and a faster tensor accelerator. Gaming performance is another highlight of the Gen 1. According to Qualcomm, its latest Adreno GPU offers 30 percent faster rendering performance while consuming 25 percent less power. Over on the audio front, the Gen 1 includes support for Qualcomm’s recently announced aptX Lossless Bluetooth codec. It can deliver up to CD-quality 16-bit 44.1kHz audio streaming over a wireless connection.

Rounding out the Gen 1’s feature list is a dedicated Trust Management Engine. The Gen 1 is the first mobile chipset to support Google’s Android Ready SE standard out of the box, which means it has the capability to store things like digital car keys and IDs.

With its mix of performance improvements and new features, the Snapdragon 8 Gen 1 offers an intriguing look at the next generation of Android flagships. Now we have to wait to see what capabilities manufacturers decide to enable in their latest devices. The first Gen 1-equipped phones will arrive later this year, with more expected to come in the first half of 2022.

Separately, Qualcomm announced it’s partnering with Google to bring the company’s Neural Architecture Search platform to its product portfolio. The technology, which will be available first on the Snapdragon 8 Gen 1, allows companies to create and optimize AI models automatically instead of manually. According to Google, NAS optimize AI models in weeks rather than months.

Read more

Twitch will use machine learning to catch ban-dodging trolls

Twitch is introducing a new machine learning feature to help streamers protect their channels from people attempting to avoid bans. Dubbed "Suspicious User Detection," the tool will automatically flag individuals it suspects may be "likely" or "possible" ban dodgers.

In cases involving the former, Twitch will prevent any messages they send from showing up in chat. It will also identify those individuals for streamers and any mods helping them with their channel. At that point, they can decide if they want to ban that person. By default, possible repeat trolls can send messages in chat, but they too will be flagged by the system. Additionally, Twitch says creators have the option to prevent them from sending any messages in the first place.

Twitch Suspicious User Detection

"The tool is powered by a machine learning model that takes a number of signals into account — including, but not limited to, the user's behavior and account characteristics — and compares that data against accounts previously banned from a Creator's channel to assess the likelihood the account is evading a previous channel-level ban," a Twitch spokesperson told Engadget when we asked about the signals the system uses to detect potential offenders.

While Twitch plans to turn on Suspicious User Detection for everyone, the tool won't automatically ban users for streamers. That's by design because it's impossible to create a machine learning tool that is 100 percent accurate in every context. "You're the expert when it comes to your community, and you should make the final call on who can participate," the company said in a blog post. "The tool will learn from the actions you take and the accuracy of its predictions should improve over time as a result."

The introduction of the tool follows a summer in which Twitch struggled to contain a phenomenon called "hate raids." The attacks saw malicious individuals use thousands of bots to spam channels with hateful language. In many cases, they targeted creators from marginalized communities. Hate raids became such a frequent feature of the platform that some creators walked away from Twitch for a day in protest of the company's lack of action.

Read more

Three Google workers sue over alleged violations of ‘don’t be evil’ motto

Google's classic "don't be evil" mantra may have been more of a philosophical statement than a practical guideline, but former staff members now want to hold the company accountable for it. NPR and The Verge say ex-engineers Paul Duke, Rebecca Rivers and Sophie Waldman have sued Google for allegedly violating the "don't be evil" segment of the company's code of conduct. They claim Google fired them for organizing worker opposition to controversial projects, like working with the Trump-era Customs and Border Protection. They were supposedly punished for pointing out evil like Google as instructed, in other words.

The one-time employees claimed Google rejected the famous phrase as it was both expensive and leading workers to organize. The internet firm supposedly decided it was better to fire people than admit its approach had changed and give up the "accompanying benefits" that came with its well-known motto.

There are concerns the lawsuit is too vague. What defines evil, exactly? However, plaintiff lawyer Laurie Burgess argued "don't be evil" was specific enough that it could be enforceable. The saying "must have meaning" if it was in the company code and thus binding, Burgess said.

We've asked Google for comment. It has previously accused all of the workers (plus Laurence Berland) of repeatedly violating data security policies by obtaining or sharing confidential data, but the workers and other critics have said this was just a cover for retaliatory action.

The lawsuit won't necessarily lead to stiff penalties. Google settled with Berland over his departure, for instance. There's a lot of pressure on Google to avoid a drawn-out legal battle when the National Labor Relations Board is still investigating the other firings. Still, this could be an important case — even if there is a settlement, it might open the door to other complaints about the company's ethical standards.

Read more

Twitch now works with SharePlay on the iPhone and iPad

Twitch has rolled out another feature designed to bring viewers closer together. The livestreaming service now supports SharePlay on iPhone and iPad, so up to 32 people can watch the same stream while they're on a FaceTime call.

Everyone on the call will need to log in to the Twitch app — the service confirmed to Engadget that each person will count as an individual viewer. The first time you open a stream while you're on FaceTime, Twitch will ask whether you want to play it for yourself or everyone on the call, and it will remember your choice. If you choose to share it with everyone, SharePlay will sync the stream on everyone's devices, so they're all watching the same moment simultaneously. Play and pause controls will sync across devices too.

Anyone on the call can move everyone over to another Twitch channel. Everyone will be able to interact with the streamer's chat, follow or subscribe to them and send Bits from their own account. You can watch the stream in either portrait or landscape orientation but, at least for now, you can't continue a SharePlay session on Twitch's Apple TV app.

A SharePlay session ends when the stream is closed, you leave the FaceTime call or end SharePlay. If you close the stream, you'll be asked if you want to end it for yourself or everyone. Choosing the latter won't actually close the stream on everyone else's devices, but playback won't be synced.

Twitch is one of the biggest streaming platforms around. It's a welcome addition to the growing lineup of services that support SharePlay, which Apple rolled out last month in iOS 15.1. Corralling a bunch of friends on a FaceTime call to watch some killer speedruns at Awesome Games Done Quick sounds like a fun way to spend time together, even when you're in your own homes.

Read more