Feed aggregator

9to5Linux Weekly Roundup: August 24th, 2025  9to5Linux

9to5Linux Weekly Roundup: August 24th, 2025  9to5Linux

"The truly universal properties of languages are not independent of our physiology and cognition," argues the co-author of a new study. Instead he says their research "strengthens the idea that intonation units are a universal feature of language." Phys.org explains: Have you ever noticed that a natural conversation flows like a dance — pauses, emphases, and turns arriving just in time? A new study has discovered that this isn't just intuition; there is a biological rhythm embedded in our speech... According to the study, led by Dr. Maya Inbar, alongside Professors Eitan Grossman and Ayelet N. Landau, human speech across the world pulses to the beat of what are called intonation units, short prosodic phrases that occur at a consistent rate of one every 1.6 seconds. The research analyzed over 650 recordings in 48 languages spanning every continent and 27 language families. Using a novel algorithm, the team was able to automatically identify intonation units in spontaneous speech, revealing that regardless of the language spoken, from English and Russian to endangered languages in remote regions, people naturally break their speech into these rhythmic chunks. "These findings suggest that the way we pace our speech isn't just a cultural artifact, it's deeply rooted in human cognition and biology," says Dr. Inbar. "We also show that the rhythm of intonation units is unrelated to faster rhythms in speech, such as the rhythm of syllables, and thus likely serves a different cognitive role...." Most intriguingly, the low-frequency rhythm they follow mirrors patterns in brain activity linked to memory, attention, and volitional action, illuminating the profound connection between how we speak and how we think. The work is published in the journal Proceedings of the National Academy of Sciences.

Read more of this story at Slashdot.

"The ultimate test of climate projections is to compare them with what has played out..." says earth sciences professor Torbjörn Törnqvist, lead author on a new study published in the open-access journal Earth's Future (published by the American Geophysical Union). But after "decades of observations," he says his researchers "were quite amazed how good those early projections were, especially when you think about how crude the models were back then, compared to what is available now." "For anyone who questions the role of humans in changing our climate, here is some of the best proof that we have understood for decades what is really happening, and that we can make credible projections...." A new era of monitoring global sea-level change took off when satellites were launched in the early 1990s to measure the height of the ocean surface. This showed that the rate of global sea-level rise since that time has averaged about one eighth of an inch per year. Only more recently, it became possible to detect that the rate of global sea-level rise is accelerating. When NASA researchers demonstrated in October 2024 that the rate has doubled during this 30-year period, the time was right to compare this finding with projections that were made during the mid-1990s, independent of the satellite measurements. In 1996, the Intergovernmental Panel on Climate Change published an assessment report soon after the satellite-based sea-level measurements had started. It projected that the most likely amount of global sea-level rise over the next 30 years would be almost 8 centimeters (3 inches), remarkably close to the 9 centimeters that has occurred. But it also underestimated the role of melting ice sheets by more than 2 centimeters (about 1 inch). At the time, little was known about the role of warming ocean waters and how that could destabilize marine sectors of the Antarctic Ice Sheet from below. Ice flow from the Greenland Ice Sheet into the ocean has also been faster than foreseen. "The findings provide confidence in model-based climate projections," according to the paper. Again, its two key points: The largest disparities between projections and observations were due to underestimated dynamic mass loss of ice sheets Comparison of past projections with subsequent observations gives confidence in future climate projections Thanks to Slashdot reader Mr. Dollar Ton for sharing the news.

Read more of this story at Slashdot.

"Delta Air Lines Flight 1334 was flying from Atlanta to Fort Lauderdale last month when smoke and flames started pouring out of a backpack," reports CNN. "The pilots declared an emergency and diverted to Fort Meyers where the 191 people onboard safely evacuated." The culprit was a passenger's personal lithium-ion battery pack, which had been tucked away in the carry-on bag. At the FAA's William J. Hughes Technical Center for Advanced Aerospace in Atlantic City, New Jersey, fire safety engineers research and demonstrate just how bad it can be. "Lithium batteries can go into what's called thermal runaway," Fire Safety Branch Manager Robert Ochs, explained. "All of a sudden, it'll start to short circuit ... It will get warmer and warmer and warmer until the structure of the battery itself fails. At that point, it can eject molten electrolyte and flames and smoke and toxic gas...." These thermal runaways are difficult to fight. The FAA recommends flight attendants first use a halon fire extinguisher, which is standard equipment on planes, but that alone may not be enough. In the test performed for CNN, the flames sprung back up in just moments... "Adding the water, as much water from the galley cart, non-alcoholic liquids, everything that they can get to just start pouring on that device." The problems are not new, but more batteries are being carried onto planes than ever before. Safety organization UL Standards and Engagement says today an average passenger flies with four devices powered by lithium-ion batteries. "The incidents of fire are rare, but they are increasing. We're seeing as many as two per week, either on planes or within airports," Jeff Marootian, the president and CEO of the organization, told CNN... [T]he latest federal data shows external battery packs are the top cause of incidents, and as a result the FAA has banned them from checked baggage where they are harder to extinguish. But despite all of the warnings, UL Standards and Engagement says two in five passengers still say they check them.

Read more of this story at Slashdot.

The mysterious X-37B space-plane — the U.S. military's orbital test vehicle — "serves partly as a platform for cutting-edge experiments," writes Space.com And "one of these experiments is a potential alternative to GPS that makes use of quantum science as a tool for navigation: a quantum inertial sensor." This technology could revolutionize how spacecraft, airplanes, ships and submarines navigate in environments where GPS is unavailable or compromised. In space, especially beyond Earth's orbit, GPS signals become unreliable or simply vanish. The same applies underwater, where submarines cannot access GPS at all. And even on Earth, GPS signals can be jammed (blocked), spoofed (making a GPS receiver think it is in a different location) or disabled — for instance, during a conflict... Traditional inertial navigation systems, which use accelerometers and gyroscopes to measure a vehicle's acceleration and rotation, do provide independent navigation, as they can estimate position by tracking how the vehicle moves over time... Eventually though, without visual cues, small errors will accumulate and you will entirely lose your positioning... At very low temperatures, atoms obey the rules of quantum mechanics: they behave like waves and can exist in multiple states simultaneously — two properties that lie at the heart of quantum inertial sensors. The quantum inertial sensor aboard the X-37B uses a technique called atom interferometry, where atoms are cooled to the temperature of near absolute zero, so they behave like waves. Using fine-tuned lasers, each atom is split into what's called a superposition state, similar to Schrödinger's cat, so that it simultaneously travels along two paths, which are then recombined. Since the atom behaves like a wave in quantum mechanics, these two paths interfere with each other, creating a pattern similar to overlapping ripples on water. Encoded in this pattern is detailed information about how the atom's environment has affected its journey. In particular, the tiniest shifts in motion, like sensor rotations or accelerations, leave detectable marks on these atomic "waves". Compared to classical inertial navigation systems, quantum sensors offer orders of magnitude greater sensitivity. Because atoms are identical and do not change, unlike mechanical components or electronics, they are far less prone to drift or bias. The result is long duration and high accuracy navigation without the need for external references. The upcoming X-37B mission will be the first time this level of quantum inertial navigation is tested in space. The article points out that a quantum navigation system could be crucial "for future space exploration, such as to the Moon, Mars or even deep space," where autonomy is key and when signals from Earth are unavailable. "While quantum computing and quantum communication often steal headlines, systems like quantum clocks and quantum sensors are likely to be the first to see widespread use."

Read more of this story at Slashdot.

Linux 6.17-rc3 Released: "A Bit Larger Than Usual"  Phoronix

The BBC reports a growing trend in music: "for established (but not superstar) artists to be targeted by fake albums or songs that suddenly appear on their pages on Spotify and other streaming services." Even dead musicians have had AI-generated "new" material added to their catalogues... According to music industry analysts Luminate, about 99,000 songs are uploaded to streaming services every day, usually via dozens of distribution services, which ask the uploader to submit the artist's details. If that information is incorrect, and a song wrongly gets listed under an existing artist's name, it's down to them or their label to complain and get it removed. Spotify took three weeks to remove fakes of folk singer/songwriter Emily Portman, according to the article, "and she still hasn't regained control of her Spotify artist profile... Considering how the streaming era has already made a big dent in many artists' incomes, Emily Portman says this affair has felt like a "very low blow"... She suspects independent artists are being targeted because star names have more protection and more power to get fraudulent releases removed swiftly." But it's also happened to "a number of Americana and folk-rock artists who have had fake tracks posted using their names in recent weeks — apparently all from the same source," including Wilco frontman Jeff Tweedy, J Tillman (now known as Father John Misty), Sam Beam (aka Iron & Wine), Teddy Thompson and Jakob Dylan: All the releases used the same style of AI artwork and were credited to three record labels, two with apparently Indonesian names. Many listed the same name as a songwriter — Zyan Maliq Mahardika. That name has also been credited on other songs mimicking real US Christian musicians and metalcore bands. Spotify said it had flagged the issue with the distributor and removed these tracks as they "violated our policy against impersonating another person or brand." It added it would "remove any distributor who repeatedly allows this type of content on our platform".... Tatiana Cirisano from media and technology analysis company Midia Research says AI is "making it easier for fraudsters" to fool listeners, who are also more "passive" in the algorithmic age. She thinks bad actors posing as real-life artists are hoping their fraudulent tracks will "rack up enough streams" — hundreds of thousands — to earn them a nice payday. "I would think that the AI fakes are targeting lesser-known artists in the hopes that their schemes fly under the radar, compared to if they were to target a superstar who could immediately get Spotify on the line," she notes. But streaming services and distributors are "working hard" and getting better at spotting it, she stresses, "ironically, also by using AI and machine learning!

Read more of this story at Slashdot.

Sysdig on Linux: diagnose problems in minutes (2025) | by Aleksei Aleinikov | Aug, 2025  DataDrivenInvestor

Archer Aviation is "the official air taxi partner" of the 2028 Olympic Games in Los Angeles, Electrek reported in May. In June it entered "a key development phase ahead of full-fledged flight certification and commercial operations" by completing a piloted flight in its flagship Midnight aircraft, "demonstrating a conventional takeoff and landing instead of vertical (it can do both)." During that flight, which took place in the skies above Salinas, California, the eVTOL achieved a top speed of 125 mph and a maximum altitude of 1,500 feet above ground level. Most recently, Archer has taken its Midnight eVTOL above Salinas again, achieving its longest flight to date. Per Archer, the recent successful flight in California lasted 31 minutes, and the piloted Midnight eVTOL traveled 55 miles — the company's longest recorded flight yet with a pilot onboard... [Again with speeds exceeding 125 mph] United Airlines CFO Mike Leskinen, who led the airline's early investment in Archer Aviation, was present at the test facility to witness the milestone flight. Leskinen congratulated the Archer team on its longest eVTOL flight and expressed his satisfaction with the Midnight aircraft's quiet operation. Their aircraft even "reached speeds of nearly 150 miles per hour" the week before, according to Archer's announcement. They're calling it another milestone "as the company advances toward FAA certification in the U.S. and near-term commercialization in the United Arab Emirates." And Archer's Founder/CEO said crossing the 50-mile mark at speed "is another clear step toward commercialization that shows the maturity of our program."

Read more of this story at Slashdot.

CachyOS ISO Snapshot for August 2025 Introduces a New Package Dashboard  9to5Linux

"Good news, everyone! The new version of Mozilla's browser now makes even more extensive use of AI," writes the Register, "providing summaries of linked content and offering developers the ability to add LLM support to extensions." Firefox 142 brings some visible shininess, but due to the combination of regional restrictions and Mozilla's progressive rollout system, not everybody can see all the features just yet... Not geofenced but subject to phased rollout are link previews, for various native-English-speaking regions. Hover over, long-press, or right-click a link and pick Preview Link, and a summary should appear. Mozilla's summary says: "Previews can optionally include AI-generated key points, which are processed on your device to protect your privacy." "Link Previews is gradually rolling out to ensure performance and quality," Firefox says in their release notes, "and is now available in en-US, en-CA, en-GB, en-AU for users with more than 3 GB of available RAM." (The notes also add a welcome for "the developers who contributed their first code change to Firefox in this release, 20 of whom were brand new volunteers!") The Register notes that Firefox 142 also gives developers the ability to add LLM support to extensions using wllama, a Wasm binding interfacing with llama.cpp, which lets you run Meta's Llama LLM and other models, locally or in the cloud.

Read more of this story at Slashdot.

Best Non-Violent Games Available on Steam for Linux Users  It's FOSS

CachyOS Introduces Packages Dashboard, GRUB+Btrfs Bootable Snapshots  Phoronix

The Framework Desktop and Linux have shown me the path to PC gaming in the living room  The Verge

The Hill reports: Russian state-sponsored hackers have targeted thousands of networking devices associated with U.S. critical infrastructure sectors over the past year, the FBI warned Wednesday. The cyber actors are associated with the Russian Federal Security Service's (FSB) Center 16 and have taken aim at a vulnerability in certain Cisco devices, according to an agency public service announcement. In some cases, hackers have been able to modify configuration files to enable unauthorized access, which they have used to conduct reconnaissance on networks. This has "revealed their interest in protocols and applications commonly associated with industrial control systems," the FBI said. Cisco's threat intelligence research arm, Talos, explained in a separate advisory that a subcluster of this group, which it has named "Static Tundra," is targeting a seven-year-old vulnerability in the company's Smart Install feature. The firm has offered a patch for the vulnerability, but it remains a problem in unpatched and end-of-life network devices, it warned. "Once they establish initial access to a network device, Static Tundra will pivot further into the target environment, compromising additional network devices and establishing channels for long-term persistence and information gathering," warns the Talos blog. "This is demonstrated by the group's ability to maintain access in target environments for multiple years without being detected." In a statement emailed to The Register, a Cisco spokesperson "said the company is aware of ongoing exploitation targeting this flaw." "We strongly urge customers to immediately upgrade to fixed software versions as outlined in the security advisory and follow our published security best practices," the spokesperson said, directing customers to the FBI's announcement and Cisco Talos blog for additional details. The ongoing campaign targets telecommunications, higher education, and manufacturing organizations across North America, Asia, Africa, and Europe, "with victims selected based on their strategic interest to the Russian government," according to Talos researchers Sara McBroom and Brandon White. "We assess that the purpose of this campaign is to compromise and extract device configuration information en masse, which can later be leveraged as needed based on then-current strategic goals and interests of the Russian government," McBroom and White wrote. And while both security alerts focus on the FSB's latest round of network intrusions, "many other state-sponsored actors also covet the access these devices afford," the Talos team warned. "Organizations should be aware that other advanced persistent threats (APTs) are likely prioritizing carrying out similar operations as well." Some context from Hot Hardware: Cisco indicated in its advisory that "Only Smart Install client switches are affected by the vulnerability". The list of affected devices is in Table A-1 here. For a successful attack, hackers exploit a vulnerability tracked as CVE-2018-0171. This was a vulnerability that was patched way back in 2018.

Read more of this story at Slashdot.

Leaving Windows Behind: 5 Signs You’ve Shifted to a Linux Mindset  How-To Geek

Years Later, EDAC Linux Driver Coming For The ARM Cortex-A72  Phoronix

Qualcomm Adreno X1-45 GPU Support Appears Ready For The Linux Kernel  Phoronix

Valve Proton marks 7th anniversary  VideoCardz.com