Kernel prepatch 6.16-rc2

Linus Torvalds has released 6.16-rc2,
which is “admittedly
even smaller than usual
“, though rc2 is not uncommonly one of the smaller
release candidates.

It may be that people are taking a breather after a fairly sizable
merge window, but it might also be seasonal, with Europe starting to
see summer vacations… We’ll see how this goes.

The diffstat looks somewhat unusual, with a lot of one-liners with
both ARC and pincontrol having (presumably independently) ended up
doing some unrelated trivial cleanups.

But even that is probably noticeable only because everything else is
pretty small. That “everything else” is mostly network drivers (and
bluetooth) and bcachefs, with some rust infrastructure and core
networking changes thrown in.

Mitsubishi Launches EV Battery Swap Network in Tokyo – for Both Cars and Trucks

In Tokyo Mitsubishi is deploying “an innovative new battery swap network for electric cars” in a multi-year test program reports the EV news site Electrek.

But it’s not just for electric cars. Along with the 14 modular battery swapping stations, Mitsubishi is also deploying “more than 150 battery-swappable commercial electric vehicles” from truck maker Fuso:

A truck like the Mitsubishi eCanter typically requires a full night of AC charging to top off its batteries, and at least an hour or two on DC charging in Japan, according to Fuso. This joint pilot by Mitsubishi, Mitsubishi Fuso Trucks, and [EV battery swap specialist] Ample aims to circumvent this issue of forced downtime with its swappable batteries, supporting vehicle uptime by delivering a full charge within minutes.

The move is meant to encourage the transport industry’s EV shift while creating a depository of stored energy that can be deployed to the grid in the event of a natural disaster — something Mitsubishi in Japan has been working on for years.
The article’s author also adds their own opinion about battery-swapping technology. “When you see how simple it is to add hundreds of miles of driving in just 100 seconds — quicker, in many cases, than pumping a tank of liquid fuel into an ICE-powered car — you might come around, yourself.”


Read more of this story at Slashdot.

Meta’s Llama 3.1 Can Recall 42% of the First Harry Potter Book

Timothy B. Lee has written for the Washington Post, Vox.com, and Ars Technica — and now writes a Substack blog called “Understanding AI.”
This week he visits recent research by computer scientists and legal scholars from Stanford, Cornell, and West Virginia University that found that Llama 3.1 70BÂ(released in July 2024) has memorized 42% of the first Harry Potter book well enough to reproduce 50-token excerpts at least half the time…
The paper was published last month by a team of computer scientists and legal scholars from Stanford, Cornell, and West Virginia University. They studied whether five popular open-weight models — three from Meta and one each from Microsoft and EleutherAI — were able to reproduce text from Books3, a collection of books that is widely used to train LLMs. Many of the books are still under copyright… Llama 3.1 70B — a mid-sized model Meta released in July 2024 — is far more likely to reproduce Harry Potter text than any of the other four models….
Interestingly, Llama 1 65B, a similar-sized model released in February 2023, had memorized only 4.4 percent of Harry Potter and the Sorcerer’s Stone. This suggests that despite the potential legal liability, Meta did not do much to prevent memorization as it trained Llama 3. At least for this book, the problem got much worse between Llama 1 and Llama 3. Harry Potter and the Sorcerer’s Stone was one of dozens of books tested by the researchers. They found that Llama 3.1 70B was far more likely to reproduce popular books — such as The Hobbit and George Orwell’s 1984 — than obscure ones. And for most books, Llama 3.1 70B memorized more than any of the other models…

For AI industry critics, the big takeaway is that — at least for some models and some books — memorization is not a fringe phenomenon. On the other hand, the study only found significant memorization of a few popular books. For example, the researchers found that Llama 3.1 70B only memorized 0.13 percent of Sandman Slim, a 2009 novel by author Richard Kadrey. That’s a tiny fraction of the 42 percent figure for Harry Potter… To certify a class of plaintiffs, a court must find that the plaintiffs are in largely similar legal and factual situations. Divergent results like these could cast doubt on whether it makes sense to lump J.K. Rowling, Richard Kadrey, and thousands of other authors together in a single mass lawsuit. And that could work in Meta’s favor, since most authors lack the resources to file individual lawsuits.

Why is it happening? “Maybe Meta had trouble finding 15 trillion distinct tokens, so it trained on the Books3 dataset multiple times. Or maybe Meta added third-party sources — such as online Harry Potter fan forums, consumer book reviews, or student book reports — that included quotes from Harry Potter and other popular books…”

“Or there could be another explanation entirely. Maybe Meta made subtle changes in its training recipe that accidentally worsened the memorization problem.”


Read more of this story at Slashdot.

Apple Migrates Its Password Monitoring Service to Swift from Java, Gains 40% Performance Uplift

Meta and AWS have used Rust, and Netflix uses Go,reports the programming news site InfoQ. But using another language, Apple recently “migrated its global Password Monitoring service from Java to Swift, achieving a 40% increase in throughput, and significantly reducing memory usage.”

This freed up nearly 50% of their previously allocated Kubernetes capacity, according to the article, and even “improved startup time, and simplified concurrency.”

In a recent post, Apple engineers detailed how the rewrite helped the service scale to billions of requests per day while improving responsiveness and maintainability… “Swift allowed us to write smaller, less verbose, and more expressive codebases (close to 85% reduction in lines of code) that are highly readable while prioritizing safety and efficiency.”
Apple’s Password Monitoring service, part of the broader Password app’s ecosystem, is responsible for securely checking whether a user’s saved credentials have appeared in known data breaches, without revealing any private information to Apple. It handles billions of requests daily, performing cryptographic comparisons using privacy-preserving protocols. This workload demands high computational throughput, tight latency bounds, and elastic scaling across regions… Apple’s previous Java implementation struggled to meet the service’s growing performance and scalability needs. Garbage collection caused unpredictable pause times under load, degrading latency consistency. Startup overhead — from JVM initialization, class loading, and just-in-time compilation, slowed the system’s ability to scale in real time. Additionally, the service’s memory footprint, often reaching tens of gigabytes per instance, reduced infrastructure efficiency and raised operational costs.
Originally developed as a client-side language for Apple platforms, Swift has since expanded into server-side use cases…. Swift’s deterministic memory management, based on reference counting rather than garbage collection (GC), eliminated latency spikes caused by GC pauses. This consistency proved critical for a low-latency system at scale. After tuning, Apple reported sub-millisecond 99.9th percentile latencies and a dramatic drop in memory usage: Swift instances consumed hundreds of megabytes, compared to tens of gigabytes with Java.

“While this isn’t a sign that Java and similar languages are in decline,” concludes InfoQ’s article, “there is growing evidence that at the uppermost end of performance requirements, some are finding that general-purpose runtimes no longer suffice.”


Read more of this story at Slashdot.

How to set up a WhatsApp account without Facebook or Instagram

There’s no shortage of reasons to stay off the Meta ecosystem, which includes Facebook and Instagram, but there are some places where WhatsApp remains the main form of text-based communication. The app is a great alternative to SMS, since it offers end-to-end encryption and was one of the go-to methods to send uncompressed photos and videos between iPhone and Android users before Apple adopted RCS. Even though Facebook, which later rebranded to Meta, acquired WhatsApp in 2014, it doesn’t require a Facebook or Instagram account to get on WhatsApp — just a working phone number.

How to create a WhatsApp account without Facebook or Instagram

To start, you need to download WhatsApp on your smartphone. Once you open the app, you can start the registration process by entering a working phone number. After entering your phone number, you’ll receive a unique six-digit code that will complete the registration process. From there, you can sort through your contacts on your attached smartphone to build out your WhatsApp network, but you won’t have to involve Facebook or Instagram at any point.

Alternatively, you can request a voice call to deliver the code instead. Either way, once you complete the registration process, you have a WhatsApp account that’s not tied to a Facebook or Instagram account.

How to link WhatsApp to other Meta accounts 

If you change your mind and want more crossover between your Meta apps, you can go into the app’s Settings panel to change that. In Settings, you can find the Accounts Center option with the Meta badge on it. Once you hit it, you’ll see options to “Add Facebook account” and “Add Instagram account.” Linking these accounts means Meta can offer more personalized experiences across the platforms because of the personal data that’s now interconnected.

You can always remove your WhatsApp account from Meta’s Account Center by going back into the same Settings panel. However, any previously combined info will stay combined, but Meta will stop combining any personal data after you remove the account.

This article originally appeared on Engadget at https://www.engadget.com/social-media/how-to-set-up-a-whatsapp-account-without-facebook-or-instagram-210024705.html?src=rss

OrangePi Equips Gateway Board with RISC-V Processor, Four RJ45 Ports, and OpenWRT Support

This week, OrangePi introduced the R2S, a compact board for embedded networking and gateway applications, running OpenWRT by default. Key features include dual 2.5G Ethernet ports, dual Gigabit Ethernet ports, and USB Type-C power input. The OrangePi R2S is powered by the Ky X1 8-core RISC-V AI processor, the same chip found in the OrangePi […]

Could This City Be the Model for How to Tackle the Both the Climate and Housing Crisis?

NPR looks at the “high-quality, climate-friendly apartments” in Vienna, asking if it’s a model for addressing both climate change and the housing crisis.
About half the city’s 2 million people live in the widespread (and government-supported) apartments, with solar panels on top and very thick, insulated walls that reduce the need for heating and cooling. (One resident tells NPR they don’t even need an air conditioner because “It’s not cold in winter times. It’s not hot in summer times.”)

Vienna council member Nina Abrahamczik, who heads the climate and environment committee, says as the city transitions all of its buildings off planet-heating fossil fuels, they’re starting with the roughly 420,000 housing units they already own or subsidize…. As Vienna makes an aggressive push to completely move away from climate-polluting natural gas by 2040, it’s starting with much of this social housing, says Jürgen Czernohorszky, executive city councilor responsible for climate and environment. City-owned buildings are now switching from gas to massive electric heat pumps, and to geothermal, which involves probing into the ground to heat homes. Another massive geothermal project that drills even deeper into the earth to heat homes is also underway.
The city is also powering housing with solar energy. As of a year and a half ago, Vienna mandates all new buildings and building extensions to have rooftop solar. And Vienna’s older apartment buildings are getting climate retrofits, says Veronika Iwanowski, spokesperson for Vienna’s municipal housing company, Wiener Wohnen. That includes new insulation, doors and windows to prevent the city’s wind from getting in the cracks. The increase in energy efficiency and switching from gas to renewables doesn’t just have climate benefits from cutting fossil fuel use. It also means housing residents are paying less on electric bills…
With city-subsidized housing, housing developers can compete to get land and low-interest loans from the city. Officials say those competitions are a critical lever for climate action. “As we can control the contents of the competitions, we try to make them fit to the main goals of the city,” says Kurt Hofstetter, city planner for Vienna, “which is of course also ecological….” Now the housing judges give out points for things like increased energy efficiency, green roofs and sustainable building materials… Now the climate innovations in subsidized housing are inspiring the private market as well, Hofstetter says…
The article notes that most of the city’s funding is provided in the form of low-interest loans, according to a researcher at the Austrian Federation of Limited-Profit Housing Associations. (And the average social housing rents are about $700 for a large one-bedroom apartment, says Gerald Kössl, researcher at the Austrian Federation of Limited-Profit Housing Associations.)


Read more of this story at Slashdot.

The Kubuntu Focus Team Announces The M2 GEN 6

The Kubuntu Focus team has some exciting news to share. We have launched the sixth generation of the flagship M2 laptop, featuring the Intel Core Ultra 9 275HX CPU with 24 threads, Nvidia® GeForce® RTX 50-series GPUs, and many more upgrades.
See more about this laptop in the press release, and see the press photos here!

The post The Kubuntu Focus Team Announces The M2 GEN 6 appeared first on Linux Today.

Delightfully irreverent Underdogs isn’t your parents’ nature docuseries

Narrator Ryan Reynolds celebrates nature’s outcasts in the new NatGeo docuseries Underdogs.

Most of us have seen a nature documentary or two (or three) at some point in our lives, so it’s a familiar format: sweeping majestic footage of impressively regal animals accompanied by reverently high-toned narration (preferably with a tony British accent). Underdogs, a new docuseries from National Geographic, takes a decidedly different and unconventional approach. Narrated by with hilarious irreverence by Ryan Reynolds, the five-part series highlights nature’s less cool and majestic creatures: the outcasts and benchwarmers, more noteworthy for their “unconventional hygiene choices” and “unsavory courtship rituals.” It’s like The Suicide Squad or Thunderbolts*, except these creatures actually exist.

Per the official premise, “Underdogs features a range of never-before-filmed scenes, including the first time a film crew has ever entered a special cave in New Zealand—a huge cavern that glows brighter than a bachelor pad under a black light thanks to the glowing butts of millions of mucus-coated grubs. All over the world, overlooked superstars like this are out there 24/7, giving it maximum effort and keeping the natural world in working order for all those showboating polar bears, sharks and gorillas.” It’s rated PG-13 thanks to the odd bit of scatalogical humor and shots of Nature Sexy Time

Each of the five episodes is built around a specific genre. “Superheroes” highlights the surprising superpowers of the honey badger, pistol shrimp, and the invisible glass frog, among others, augmented with comic book graphics; “Sexy Beasts” focuses on bizarre mating habits and follows the format of a romantic advice column; “Terrible Parents” highlights nature’s worst practices, following the outline of a parenting guide; “Total Grossout” is exactly what it sounds like; and “The Unusual Suspects” is a heist tale, documenting the supposed efforts of a macaque to put together the ultimate team of masters of deception and disguise (an inside man, a decoy, a fall guy, etc.).  Green Day even wrote and recorded a special theme song for the opening credits.

Read full article

Comments

Tesla blows past stopped school bus and hits kid-sized dummies in Full Self-Driving tests

A revealing demonstration with Tesla’s Full Self-Driving mode is raising concerns about whether fully autonomous cars are ready to hit the streets. Tesla has reportedly pushed back the rollout of its upcoming all-electric, fully autonomous car called the Cybercab, while a recent demonstration in Austin, Texas showed a Tesla Model Y running through a school bus’ flashing lights and stop signs, and hitting child-size mannequins. The tests were conducted by The Dawn Project, along with Tesla Takedown and ResistAustin, and showed Tesla’s Full Self-Driving software repeating the same mistake eight times.

It’s worth noting that Tesla’s autonomous driving feature is formally known as Full Self-Driving (Supervised) and “requires a fully attentive driver and will display a series of escalating warnings requiring driver response.” Tesla even has a warning that says, “failure to follow these instructions could cause damage, serious injury or death.” However, it’s not the first time that Tesla’s FSD software has found itself in hot water. The Dawn Project, whose founder Dan O’Dowd is the CEO of a company that offers competing automated driving system software, previously took out ads warning about the dangers of Tesla’s Full Self-Driving and how it would fail to yield around school buses. In April 2024, a Model S using Full Self-Driving was involved in a crash in Washington, where a motorcyclist died.

With anticipation building up for an eventual Cybercab rollout on June 22, the company’s CEO posted some additional details on X. According to Elon Musk, Tesla is “being super paranoid about safety, so the date could shift.” Beyond that, Musk also posted that the “first Tesla that drives itself from factory end of line all the way to a customer house is June 28.”

This article originally appeared on Engadget at https://www.engadget.com/transportation/tesla-blows-past-stopped-school-bus-and-hits-kid-sized-dummies-in-full-self-driving-tests-183756251.html?src=rss

Facial Recognition Error Sees Woman Wrongly Accused of Theft

A chain of stores called Home Bargains installed facial recognition software to spot returning shoplifters. Unfortunately, “Facewatch” made a mistake.
“We acknowledge and understand how distressing this experience must have been,” an anonymous Facewatch spokesperson tells the BBC, adding that the store using their technology “has since undertaken additional staff training.”

A woman was accused by a store manager of stealing about £10 (about $13) worth of items (“Everyone was looking at me”). And then it happened again at another store when she was shopping with her 81-year-old mother on June 4th:

“As soon as I stepped my foot over the threshold of the door, they were radioing each other and they all surrounded me and were like ‘you need to leave the store’,” she said. “My heart sunk and I was anxious and bothered for my mum as well because she was stressed….”
It was only after repeated emails to both Facewatch and Home Bargains that she eventually found there had been an allegation of theft of about £10 worth of toilet rolls on 8 May. Her picture had somehow been circulated to local stores alerting them that they should not allow her entry. Ms. Horan said she checked her bank account to confirm she had indeed paid for the items before Facewatch eventually responded to say a review of the incident showed she had not stolen anything. “Because I was persistent I finally got somewhere but it wasn’t easy, it was really stressful,” she said. “My anxiety was really bad — it really played with my mind, questioning what I’ve done for days. I felt anxious and sick. My stomach was turning for a week.”
In one email from Facewatch seen by the BBC, the firm told Ms Horan it “relies on information submitted by stores” and the Home Bargains branches involved had since been “suspended from using the Facewatch system”. Madeleine Stone, senior advocacy officer at the civil liberties campaign group Big Brother Watch, said they had been contacted by more than 35 people who have complained of being wrongly placed on facial recognition watchlists.

“They’re being wrongly flagged as criminals,” Ms Stone said.
“They’ve given no due process, kicked out of stores,” adds the senior advocacy officer. “This is having a really serious impact.” The group is now calling for the technology to be banned. “Historically in Britain, we have a history that you are innocent until proven guilty but when an algorithm, a camera and a facial recognition system gets involved, you are guilty.

The Department for Science, Innovation and Technology said: “While commercial facial recognition technology is legal in the UK, its use must comply with strict data protection laws. Organisations must process biometric data fairly, lawfully and transparently, ensuring usage is necessary and proportionate.

“No one should find themselves in this situation.”

Thanks to alanw (Slashdot reader #1,822) for sharing the article.


Read more of this story at Slashdot.

New York State Begins Asking Employers to Offically Identify Layoffs Caused by AI

The state of New York is “asking companies to disclose whether AI is the reason for their layoffs,” reports Entrepreneur:
The move applies to New York State’s existing Worker Adjustment and Retraining Notification (WARN) system and took effect in March, Bloomberg reported. New York is the first state in the U.S. to add the disclosure, which could help regulators understand AI’s effects on the labor market.

The change takes the form of a checkbox added to a form employers fill out at least 90 days before a mass layoff or plant closure through the WARN system. Companies have to select whether “technological innovation or automation” is a reason for job cuts. If they choose that option, they are directed to a second menu where they are asked to name the specific technology responsible for layoffs, like AI or robots.


Read more of this story at Slashdot.

Sony unlocks regional restrictions for some PC games in more than 100 countries

It turns out that region-locking your games makes it harder to sell more copies. First discovered by Wario64, Sony has lifted regional restrictions for several of its titles that are available to play on PC through Steam. Sony hasn’t officially announced the removal of these region locks, nor the reason why, but God of War Ragnarok, The Last of Us Part II Remastered, Spider-Man 2, and the hit co-op shooter Helldivers 2 can be purchased once again in 177 countries that were previously prohibited, as seen in database changelogs.

To understand why Sony decided to limit sales of its games in certain countries, we have to revisit when the video game publisher announced it would require Helldivers 2 PC players to link their Steam accounts to the PlayStation Network. This controversial decision prohibited players in countries that didn’t have access to the PSN, so Sony also removed Helldivers 2 from being on sale in those locations. The publisher quickly backtracked on that move, even later removing the PSN requirements for other titles like Spider-Man 2.

More recently, the PC release of Stellar Blade on June 11 resurfaced Sony’s PSN login requirement debacle. The game’s developer, Shift Up, was able to work with Sony to loosen restrictions and make a PSN login optional, allowing for an explosive launch where the game saw a peak of nearly 200,000 players in its opening week. Now, PC gamers across the world can play Stellar Blade and Helldivers 2 without needing a PSN account. However, there are still some Sony titles on Steam that are heavily restricted, like Ghost of Tsushima and Until Dawn.

This article originally appeared on Engadget at https://www.engadget.com/gaming/playstation/sony-unlocks-regional-restrictions-for-some-pc-games-in-more-than-100-countries-162228948.html?src=rss

Gemini 435Le Features Active Stereo, Dual-Laser Modules, and 6-Axis IMU for 3D Vision

The Gemini 435Le is Orbbec’s newest 3D camera, designed to deliver robust, high-precision depth sensing for demanding industrial and outdoor robotics environments. Engineered with industrial-grade construction and IP67 protection, it supports logistics automation, robotic arms, and autonomous mobile robots operating in variable and dynamic conditions. The system employs stereo vision technology with a 95 mm […]

17-Year-Old Student Builds 3D-printed Drone In Garage, Interests DoD and MIT

“Cooper Taylor is only 17 years old, but he’s already trying to revolutionize the drone industry,” writes Business Insider:

His design makes the drone more efficient, customizable, and less expensive to construct, he says. He’s built six prototypes, 3D printing every piece of hardware, programming the software, and even soldering the control circuit board. He says building his drone cost one-fifth of the price of buying a comparable machine, which sells for several thousand dollars. Taylor told Business Insider he hopes that “if you’re a first responder or a researcher or an everyday problem solver, you can have access to this type of drone.”

His innovation won him an $8,000 scholarship in April at the Junior Science and Humanities Symposium, funded by the Defense Department. Then, on May 16, he received an even bigger scholarship of $15,000 from the US Navy, which he won after presenting his research at the Regeneron International Science and Engineering Fair…

It all started when Taylor’s little sister got a drone, and he was disappointed to see that it could fly for only about 30 minutes before running out of power. He did some research and found that a vertical take-off and landing, or VTOL, drone would last longer. This type of drone combines the multi-rotor helicopter style with the fixed wings of an airplane, making it extremely versatile. It lifts off as a helicopter, then transitions into plane mode. That way, it can fly farther than rotors alone could take it, which was the drawback to Taylor’s sister’s drone. Unlike a plane-style drone, though, it doesn’t need a runway, and it can hover with its helicopter rotors.
Taylor designed a motor “that could start out helicopter-style for liftoff, then tilt back to become an airplane-style motor,” according to the article.
And now this summer he’ll be “working on a different drone project through a program with the Reliable Autonomous Systems Lab at the Massachusetts Institute of Technology.”

Thanks to Slashdot reader Agnapot for sharing the news.


Read more of this story at Slashdot.