SoylentNews

fliptop writes:

Newspaper could join legacy media brands in embracing newsletter platform:

The Washington Post has held talks with Substack about hosting pieces by its writers, the site's co-founder has said, as a host of legacy media brands embrace the newsletter platform in the battle for readers.

In an interview with the Guardian, Substack's Hamish McKenzie said he had spoken to the Post about its plans to widen the types of opinion pieces on its website.

He said there had been a "change in mindset" from traditional media, which once viewed Substack with suspicion. He said many now saw the platform as an opportunity to adapt to what he described as "the most significant media disruption since the printing press".

[...] Substack has become increasingly influential since its launch in 2017. It allows anyone to publish and distribute digital content, primarily through newsletters, and charge a subscription. It has also been branching out into podcasts and video.

[...] McKenzie said Substack was trying to find new workable models for media amid the struggles of traditional outlets to hold on to rapidly fragmenting audiences. "It's not a problem with demand for quality journalism," he said. "It's a problem with the business model and so there has to be a reinvention. We're almost at the point where the fire has razed through the forest and there are a few trees still standing. It's time to replant the forest. We're living through the most significant media disruption since the printing press."

Originally spotted on The Honest Broker.

Original Submission

Arthur T Knackerbracket has processed the following story:

Is this admission a sign of what Beijing wants?

Huawei CEO Ren Zhengfei has admitted that its Ascend chip family is not as powerful as Washington thinks. As reported by its parent publication, Global Times, Zhengfei said this during an interview with the People’s Daily, the official newspaper of the Communist Party of China (CPC). The story itself was published on the front page of the newspaper on June 10, 2025, reflecting the importance of what the Huawei CEO said to the party and to state policy.

The second question the interviewer asked Zhengfei was about the U.S.’s crackdown on Huawei’s Ascend chips that allegedly violate export control and its impact on the company. “There are many companies in China making chips, and many are doing well; Huawei is just one of them. The U.S. has exaggerated Huawei’s achievements — the company isn’t that powerful yet,” said Ren. “We need to work hard to live up to their evaluations. Our single chips still lag behind the U.S. by a generation.”

This is a massive admission for a Chinese company that’s one of the key players in China’s ambition of global technological dominance; more so that it was done right on the front page of the CPC’s primary mouthpiece. Given that the People’s Daily is directly controlled by the party (not by the Chinese government), it’s often considered a tool that helps shape public opinion, share party policies, and even be used to gauge domestic feedback and the reaction of the wider international community before Beijing implements official changes.

China has been adamant for years that the U.S.’s bans on its technological sector won’t affect its progress, with China’s president telling the Dutch PM last year that it does not need ASML. However, no country is an island — and even if China does not or could not export its goods to the U.S., it still needs an international market to ship its goods and services to. Aside from that, this interview is a rare event where Beijing is strategically showing vulnerability, especially as the restriction set by the White House on the Huawei Ascend processors seems to have far-reaching consequences going beyond the borders of the United States and its allies.

[...] This isn’t to say that China doesn’t have an ace up its sleeve. The country is one of the biggest suppliers of rare earth metals needed for chip making, and it has restricted its exports as a response to Trump’s tariffs. Without them, the semiconductor supply chain would hit a snag, making it difficult, if not impossible, for fabs to continue making chips at current cost. Nevertheless, the country has recently started easing these export limits for domestic and European chip firms. It could be that China is willing to make a major concession, like giving U.S. companies access to its rare earths once again, in exchange for Washington not sanctioning Huawei’s customers.

Original Submission

upstart writes:

Preserved beneath the ice sheet's crushing weight, the land remained unchanged since long before glaciation began:

Beneath the thick ice of East Antarctica lies a hidden world—untouched for over 34 million years. This frozen expanse, more than 10 million square kilometers wide, has long concealed a forgotten landscape. Now, using cutting-edge satellite tools, researchers have pulled back the curtain on a time when Antarctica teemed with life.

A team led by Stewart Jamieson at Durham University made the discovery with help from RADARSAT, a Canadian satellite system. The technology allowed them to detect small changes in the ice surface, revealing the shape of the land buried below. What they found was extraordinary: an ancient river-carved terrain about the size of Wales, locked under nearly two kilometers of ice.

"It's like uncovering a time capsule," Jamieson said. The untouched condition of the landscape points to its extreme age. Preserved beneath the ice sheet's crushing weight, the land remained unchanged since long before glaciation began. This hidden world dates back to a period when Antarctica was not the icy desert we know today.

Back then, the continent was part of Gondwana—a supercontinent shared with Africa, South America, and Australia. Instead of ice, Antarctica featured flowing rivers, forests, and roaming dinosaurs. That changed about 20 million years ago when glaciers took hold, freezing the region's history beneath a growing sheet of ice.

The ancient landscape now uncovered is more than a prehistoric curiosity. It helps scientists understand how Antarctica has changed over millions of years. These findings could also shed light on how the ice sheet might respond to rising global temperatures in the future.

The research also opens a new window into how rivers once shaped the bedrock before the climate shifted. It suggests that massive ice coverage can preserve entire ecosystems in place, offering a rare glimpse into ancient environments that no longer exist. The survival of these features helps scientists map how Earth's surface responds to extreme changes in climate.

With every pass of the satellite, new details emerged. What started as faint surface cues turned into a clear picture of valleys, ridges, and channels below. As technology improves, more hidden corners of the Earth's past may be revealed. But for now, this glimpse beneath the Antarctic ice connects us to a greener, wilder world long gone—but not forgotten.

"We've had a longtime interest in, effectively, the shape of the land beneath the ice sheet," Jamieson said. "The implication is that this must be a very old landscape carved by rivers before the ice sheet itself grew."

slon writes:

from https://news.mit.edu/2025/photonic-processor-could-streamline-6g-wireless-signal-processing-0611

As more connected devices demand an increasing amount of bandwidth for tasks like teleworking and cloud computing, it will become extremely challenging to manage the finite amount of wireless spectrum available for all users to share.

Engineers are employing artificial intelligence to dynamically manage the available wireless spectrum, with an eye toward reducing latency and boosting performance. But most AI methods for classifying and processing wireless signals are power-hungry and can't operate in real-time.

Now, MIT researchers have developed a novel AI hardware accelerator that is specifically designed for wireless signal processing. Their optical processor performs machine-learning computations at the speed of light, classifying wireless signals in a matter of nanoseconds.

The photonic chip is about 100 times faster than the best digital alternative, while converging to about 95 percent accuracy in signal classification. The new hardware accelerator is also scalable and flexible, so it could be used for a variety of high-performance computing applications. At the same time, it is smaller, lighter, cheaper, and more energy-efficient than digital AI hardware accelerators.

The device could be especially useful in future 6G wireless applications, such as cognitive radios that optimize data rates by adapting wireless modulation formats to the changing wireless environment.

By enabling an edge device to perform deep-learning computations in real-time, this new hardware accelerator could provide dramatic speedups in many applications beyond signal processing. For instance, it could help autonomous vehicles make split-second reactions to environmental changes or enable smart pacemakers to continuously monitor the health of a patient's heart.

"There are many applications that would be enabled by edge devices that are capable of analyzing wireless signals. What we've presented in our paper could open up many possibilities for real-time and reliable AI inference. This work is the beginning of something that could be quite impactful," says Dirk Englund, a professor in the MIT Department of Electrical Engineering and Computer Science, principal investigator in the Quantum Photonics and Artificial Intelligence Group and the Research Laboratory of Electronics (RLE), and senior author of the paper.

He is joined on the paper by lead author Ronald Davis III PhD '24; Zaijun Chen, a former MIT postdoc who is now an assistant professor at the University of Southern California; and Ryan Hamerly, a visiting scientist at RLE and senior scientist at NTT Research. The research appears today in Science Advances.

Arthur T Knackerbracket has processed the following story:

14,000 vulnerable feeds found in the U.S.

A major privacy concern involving more than 40,000 security cameras worldwide has been revealed by Cybersecurity firm Bitsight. According to the company's TRACE research division, these cameras are live-streaming video feeds that are fully exposed to the internet — meaning that one can gain access without needing any sort of authentication, encryption, or even a basic password. In most cases, a person can access real-time footage from these exposed cameras simply by knowing their IP address.

Bitsight initially flagged the issue back in 2023, but recent research suggests that the situation “hasn’t gotten any better.” According to the latest research, these vulnerable cameras are not limited to one region or industry. The United States has close to 14,000 cameras that are potentially exposed, with states like California, Texas, Georgia, and New York having the highest numbers. Next on the list is Japan, with 7,000 exposed cameras, followed by Austria, Czechia, and South Korea, each of which have close to 2,000 vulnerable devices.

It is true that not every camera hooked up to the internet is a cause for concern, and some livestreams are set up intentionally to showcase scenes, like a beach or a birdhouse, for public viewing. However, some of these vulnerable cameras have been found in more private environments — including residential setups monitoring front doors, backyards, and even living rooms.

Cameras in office spaces, factories, as well as public transportation systems were also found. Bitsight researchers were able to observe sensitive spaces, monitor foot traffic, and, in some cases, even see details written on whiteboards — all in real time. The majority of the exposed devices are said to be using HTTP, while the rest stream through RTSP (Real-Time Streaming Protocol), which is a common protocol for controlling and managing streaming media over IP networks.

In addition to raising privacy and surveillance concerns, these exposed devices pose serious security risks. Information collected by Bitsight’s Cyber Threat Intelligence team suggests that users are openly discussing the feeds on dark web forums, where users are sharing tools and techniques to gain unauthorized access, and even selling access, to unprotected video streams.

Users and organizations are advised to double-check on how their cameras are configured: Disable remote access if not in use, update to the latest firmware, and make sure the device is protected behind a firewall or connected to a secure network. A simple way to check whether your camera is exposed or not is by accessing it from outside your home network. If you are able to view the camera feed without logging into a secure app or using a VPN (Virtual Private Network), it’s likely open to anyone on the internet. Additionally, one should replace any default usernames and passwords as many camera devices ship with a default set of credentials that are easy to crack.

Original Submission

upstart writes:

Nematode Tower power:

Living worm towers are recorded in the wild for the first time, a rare example of collective hitchhiking in nature

• First evidenceof "living towers" in nature: observed in rotting apples and pears from local orchards in Konstanz, Germany
• Tower function confirmed: towers can attach to passing insects and can bridge physical gaps to disperse
• A powerful model:C. elegans are a new a tool for studying the ecology and evolution of collective dispersal

Nematodes are the most abundant animal on earth, but when times get tough, these tiny worms have a hard time moving up and out. So, they play to the strength of their clade. If food runs out and competition turns fierce, they slither towards their numerous kin. They climb onto each other and over one another until their bodies forge a living tower that twists skyward where they might hitch a ride on a passing animal to greener and roomier pastures.

At least that's what scientists assumed. For decades, these worm structures were more mythical than material. Such aggregations, in which animals link bodies for group movement, are rare in nature. Only slime molds, fire ants, and spider mites are known to move in this way. For nematodes, nobody had even seen the aggregations—known as towers— forming anywhere but within the artificial confines of laboratories and growth chambers; and nobody really knew what they were for. Did towers even exist in the real world?

Now, researchers in Konstanz, Germany, have recorded video footage of worms towering in fallen apples and pears from local orchards. The team from the Max Planck Institute of Animal Behavior (MPI-AB) and the University of Konstanz combined fieldwork with laboratory experiments to provide the first direct evidence that towering behavior occurs naturally and functions as a means of collective transport.

"I was ecstatic when I saw these natural towers for the first time," says senior author Serena Ding, group leader at the MPI-AB, of the moment when co-author Ryan Greenway sent her a video recording from the field. "For so long natural worm towers existed only in our imaginations. But with the right equipment and lots of curiosity, we found them hiding in plain sight."

Greenway, a technical assistant at the MPI-AB, spent months with a digital microscope combing through decaying fruit in orchards near the university to record natural occurrences and behavior of worm towers. Some of these whole towers were brought into the lab. What was inside the towers surprised the team. Although the fruits were crawling with many species of nematodes, natural towers were made only of a single species, all at the tough larval stage known as a "dauer."

"A nematode tower is not just a pile of worms," says the first author Daniela Perez, a postdoctoral researcher at MPI-AB. "It's a coordinated structure, a superorganism in motion."

Arthur T Knackerbracket has processed the following story:

We may already have had our first-ever encounter with dark matter, according to researchers who say a mysteriously high-energy particle detected in 2023 is not a neutrino after all, but something far stranger

An extremely high-energy particle that was spotted tearing through Earth has left scientists flummoxed ever since it was discovered. While many researchers believe the particle was an unusual neutrino, some are now suggesting it may be something even wilder: a particle of dark matter travelling across the cosmos.

The KM3NeT detector, off the coast of Italy, spotted this “impossible” neutrino in 2023 while it was still under construction. The particle in question was of immense proportions, 35 times more energetic than any seen before. Where it came from remains a mystery, with possible sources including a galaxy with a very active central black hole known as a blazar, or a background source of high-energy neutrinos pervading the universe.

[...] IceCube has seen evidence for hundreds of cosmic neutrinos since 2011, but never something as energetic as KM3NeT’s discovery. That was confusing, because whatever source KM3NeT was seeing, IceCube should have seen it too.

Dev says that if the incoming particle was dark matter and not a neutrino, it could explain this mystery. The shallow predicted path of the incoming particle meant that it had to travel through more of Earth to reach KM3NeT than IceCube, increasing the chance of it being scattered into a muon. “The dark matter goes through lots of Earth’s matter,” says Dev, “and we can explain why IceCube didn’t see it.”

The particle would have been produced in a blazar and then fired towards Earth in a beam. Dev favours this idea because high-energy protons in a blazar more efficiently transfer their energy into dark matter than neutrinos, he says. The vast majority of the other events detected by KM3NeT and IceCube would probably still have been neutrinos.

Not everyone is convinced just yet. “From an Occam’s razor perspective, this is probably just an ordinary neutrino that’s exceptional in energy,” says Dan Hooper at the University of Wisconsin–Madison. However, if correct, it would give us a method to find and study dark matter particles, which have never previously been detected. “Everybody would be pretty thrilled if these machines can study not only neutrinos but also dark matter,” says Hooper.

Journal Reference: arXiv DOI: 10.48550/arXiv.2505.22754

Original Submission

Arthur T Knackerbracket has processed the following story:

William Dana Atkinson was one of the core people on the teams that created the Lisa and then Macintosh computers at Apple in the late 1970s and early 1980s. He was one of the most important and influential computer programmers who has ever lived. It is no exaggeration to say that all computer UIs designed in the last 40 years were shaped and influenced by Atkinson's brilliance and originality.

He dropped out of a doctorate in neuroscience to join Apple in 1978. He became employee # 51. In 2018, he said:

Some say Steve used me, but I say he harnessed and motivated me, and drew out my best creative energy. It was exciting working at Apple, knowing that whatever we invented would be used by millions of people.

Since the weekend when the news first broke, a remarkable range of tributes to the man and his work have appeared. For the Reg FOSS desk, one of the things that has particularly struck us is the range of different aspects of Atkinson's creativity that resonated with different people.

For instance, those working mainly online, such as TechCrunch's Antony Ha, call out HyperCard. Atkinson designed and wrote HyperCard, which introduced the wider world to the concept of hyperlinks as invented by Ted Nelson for his Xanadu hypertext system. Hypercard made it easy to create "stacks" of documents and navigate through them by clicking on links. As The Register has described more than once, it's inspired JavaScript recreations and parts of the Windows 10 UI, but most significantly of all it inspired the creators of the World Wide Web. It is specifically mentioned in the original proposal. Atkinson himself described the inspiration for Hypercard as a 1985 LSD trip.

Anyone who creates or edits pictures on computers uses tools that follow in the footsteps of an earlier Bill Atkinson app: he wrote MacPaint. Between it and his earlier version, LisaSketch, this introduced ideas like tool palettes, which became toolbars; the lasso tool to select objects; zoomed-in pixel editing, which Atkinson called FatBits; and the paint-bucket fill tool, among many other things.

Programmers who know his work nod with respect to lower-level stuff. The story of Apple's visit to Xerox, where the Xerox Alto graphical workstation inspired the Apple Lisa's overlapping windows, is well known. What's less well understood is that Smalltalk could only write or draw into one window, the topmost. It was impossible to update the contents of background windows. Atkinson was on that visit, but he didn't know that. He just assumed the Smalltalk team must have solved that, and all he had to do was work out how. He called his resulting algorithm regions and after a serious 1982 car accident and head injuries, Atkinson's first words to his visiting boss were "Don't worry, Steve, I still remember how to do regions."

Lower-level still, he is remembered for his remarkably efficient dithering algorithm, which you can try here.

As "Apple acolyte" Steven Levy wrote in Rolling Stone, Atkinson also brought Burrell Smith, designer of the original Macintosh hardware, to the project. In his book Insanely Great, Levy described the coming up with regions:

Atkinson worked at the problem for months - not only in long hours at a desk, but literally in his dreams. Upon arising he would record his somnambulant labors in a notebook. Eventually wave after wave of Atkinson's brainpower eroded the problem. He had set our to reinvent the wheel; actually he wound up inventing it.

Arthur T Knackerbracket has processed the following story:

US Navy Secretary John Phelan has told the Senate the service needs the right to repair its own gear, and will rethink how it writes contracts to keep control of intellectual property and ensure sailors can fix hardware, especially in a fight.

Speaking to the Senate Armed Services Committee on Tuesday, Phelan cited the case of the USS Gerald R. Ford, America's largest and most expensive nuclear-powered aircraft carrier, which carried a price tag of $13 billion. The ship was struggling to feed its crew of over 4,500 because six of its eight ovens were out of action, and sailors were barred by contract from fixing them themselves.

"I am a huge supporter of right to repair," Phelan told the politicians. "I went on the carrier; they had eight ovens — this is a ship that serves 15,300 meals a day. Only two were working. Six were out."

He pointed out the Navy personnel are capable of fixing their own gear but are blocked by contracts that reserve repairs for vendors, often due to IP restrictions. That drives up costs and slows down basic fixes. According to the Government Accountability Office, about 70 percent [PDF] of a weapon system's life-cycle cost goes to operations and support.

A similar issue plagued the USS Gerald Ford's weapons elevators, which move bombs from deep storage to the flight deck. They reportedly took more than four years after delivery to become fully operational, delaying the carrier's first proper deployment.

"They have to come out and diagnose the problem, and then they'll fix it," Phelan said. "It is crazy. We should be able to fix this."

The Navy is not alone in its concerns, as the US Army is peeved about the right to repair equipment it paid for too. In a rare display of bipartisanship, both Democrats and Republicans agreed that the Army shouldn't be waiting on contractors to fix its kit and Defense Secretary Pete Hegseth issued a memo directing the service to add right-to-repair provisions to its contracts.

"On a go-forward basis, we have been directed to not sign any contracts that don't give us a right to repair," Army Secretary Daniel Driscoll told the House Armed Services Committee on June 4. "On a go-back basis, we have been directed to go and do what we can to go get that right to repair."

canopic jug writes:

The Steve Jobs Archive, Wired, and MacRumors are covering the 20^th anniversary of an iconic commencement address to Standford University which Steve Jobs delivered in June, 2005. In it he quoted the Whole Earth Catalog by encouraging the graduates to "stay hungry, stay foolish".

At that point in time, YouTube was only months old, Twitter didn't exist, and Facebook didn't even have its news feed. The national media hadn't covered the speech. Apple sent out no press releases. But Stanford published the transcript on its primitive website, and people began discovering it. I recently checked my inbox for June 2005 and found multiple copies sent to me from different mailing lists. As the weeks and months went by, more and more people found the speech. Berlin describes it as going "slow-motion viral."

— Wired

It's not an obvious candidate for a classic. A commencement address by a college dropout. A talk aimed at 22-year-olds that warns "You will gradually become the old and be cleared away." A text as shadowed by reality as soaring with inspiration: "Your time is limited, so don't waste it living someone else's life."

It is a speech by a tech founder who scarcely mentions technology. A few years earlier, Steve told an interviewer, "People sometimes forget that they are very unique and that they have very unique feelings and perspectives. The whole computer industry wants to forget about the humanist side." Steve had not forgotten. At Stanford, under the guise of great simplicity—"Today I want to tell you three stories from my life"—Steve touches on fundamental truths that make us human: love, death, fear, authenticity, hope.

— The Steve Jobs Archive

The video, as published at The Steve Jobs Archive, has been recently enhanced from SD to HD.

For those that have attended commencement events, either their own or other people's, are there any memorable speakers who stand out?

Original Submission

Arthur T Knackerbracket has processed the following story:

Germany's long-awaited Jupiter supercomputer launched into the number four spot on the Top500 list of publicly ranked systems, dethroning Italy's HPC6 as Europe's biggest, baddest iron.

Developed as part of the EuroHPC Joint Undertaking and built-in collaboration with Nvidia and French supercomputer builder Eviden (formerly Atos), the system is designed to support research across a number of fields, including biophysics, cellular neuroscience, nuclear and elementary particle physics, astrophysics, and climate science.

Boffins [scientists] are also exploring the potential application of machine learning algorithms like those used in diffusion models to generate images to advance areas like medical imaging and autonomous vehicles.

In its first appearance on the biannual ranking, the Jülich Supercomputing Center's (JSC) flagship system managed 793 petaFLOPS of double precision (FP64) grunt in the time-honored High-Performance Linpack (HPL) benchmark.

However, the run is just a teaser for what's to come. Much like the US Department of Energy's Aurora supercomputer in its first appearance on the Top500 in late 2023, Jupiter's is also a partial run. The system is widely expected to be the European continent's first true exascale supercomputer.

Built by Eviden for the EuroHPC Joint Undertaking and deployed at the Jülich Supercomputing Center in Germany, Jupiter is expected to be Europe's first exascale-capable supercomputer.

While Jupiter ranks as Europe's new supercomputing beast, it still falls far short of its American counterparts. The US DoE's El Capitan, Frontier, and Aurora systems remain uncontested as the only three exascale-capable supercomputers on the Top500 at 1.74, 1.35, and 1.01 exaFLOPS, respectively.

That's likely to change in the not-so-distant future once Jülich can bring the full force of the Jupiter system to bear on the semiannual benchmark. The machine will only need to pack on another 207 petaFLOPS to its next HPL run to herald Europe's entrance into the exascale era, and just 220 more petaFLOPS to fly above Aurora. That's assuming, of course, the boffins at Argonne don't find a way to eke out a bit more compute from the Intel-based monster.

fliptop writes:

The company's chairman insists that going all-in on electric cars is wrong:

Akio Toyoda is a man who speaks his mind. He's been saying for years that forcing everyone to buy EVs isn't the way forward. Toyota's chairman is adamant that the transition can't be rushed and that going all-in on electric vehicles would have massive repercussions across the automotive industry. He believes millions of jobs throughout the supply chain could be at risk if the combustion engine is phased out too quickly. On the environmental front, Toyoda maintains that EVs are still much dirtier than hybrids.

The grandson of Toyota founder Kiichiro Toyoda claims the company has sold around 27 million hybrids since launching the first-generation Prius in 1997. According to him, those hybrids have had the same carbon footprint as nine million fully electric vehicles when adding battery and vehicle production into the equation.

Toyoda argues that a single EV is as dirty as three hybrids. However, while it's true that producing EVs and their batteries creates more carbon emissions than building gas cars, over their life cycles, EVs are responsible for far fewer overall emissions.

From InsideEVs:

The biggest anti-EV argument stems from the emissions generated during the mining, refining and processing of the raw materials used in high-voltage batteries. EV batteries use materials such as lithium, cobalt and nickel that require hazardous, water-intensive mining processes.

So when an EV rolls off a production line, it's already born "dirtier" than the average gas or hybrid vehicle, for now. It comes with a bigger "carbon debt," a term that researchers use to calculate the emissions vehicles gather before even hitting the road.

A research paper published in the scientific journal IOP Science says that gas and hybrid vehicles create six to nine metric tons of carbon dioxide emissions in their manufacturing, depending on the vehicle segment. EVs, on the other hand, generate 11 to 14 metric tons of CO2 emissions before going into the hands of customers.

But that's only part of the story. Once EVs hit the road, they begin paying off that carbon debt and their overall "emissions" start decreasing. Hybrids and gas vehicles, on the other hand, head in the opposite direction, growing their carbon emissions over time. After a certain number of miles, an EV can potentially clear that debt entirely.

How long that takes, exactly, can depend on who you ask. A 2023 Argonne National Laboratory study found that it can take an electric car 19,500 miles to mitigate the emissions made during manufacturing. That's less than two years of typical American driving, according to FactCheck.org. Another study in the journal Nature put that number higher, with carbon reductions beginning around 28,000 miles. Either way, considering how long Americans keep their cars, EVs become the far cleaner option over time.

Original Submission

Arthur T Knackerbracket has processed the following story:

By affecting cows’ diets, climate change can affect cheese’s nutritional value and sensory traits such as taste, color and texture. This is true at least for Cantal — a firm, unpasteurized cheese from the Auvergne region in central France, researchers report February 20 in the Journal of Dairy Science.

Cows in this region typically graze on local grass. But as climate change causes more severe droughts, some dairy producers are shifting to other feedstocks for their cows, such as corn, to adapt. “Farmers are looking for feed with better yields than grass or that are more resilient to droughts,” but they also want to know how dietary changes affect their products, says animal scientist Matthieu Bouchon.

For almost five months in 2021, Bouchon and colleagues at France’s National Research Institute for Agriculture, Food and Environment tested 40 dairy cows from two different breeds — simulating a drought and supplementing grass with other fodder, largely corn, in varying amounts.

[...] They found that a corn-based diet did not affect milk yield and even led to an estimated reduction in the greenhouse gas methane coming from cows’ belching. But grass-fed cows’ cheese was richer and more savory than that from cows mostly or exclusively fed corn. Grass-based diets also yielded cheese with more heart-healthy omega-3 fatty acids and higher counts of probiotic lactic acid bacteria. The authors suggest that to maintain cheese quality, producers should include fresh vegetation in cows’ fodder when it is based on corn.

Experts not involved with the study point out that warming climates impact cattle physiology as well as feed quality. “Cows produce heat to digest food — so if they are already feeling hot, they’ll eat less to lower their temperature,” says Marina Danes, a dairy scientist at the Federal University of Lavras in Brazil.

upstart writes:

Mistral releases a vibe coding client, Mistral Code:

French AI startup Mistral is releasing its own "vibe coding" client, Mistral Code, to compete with incumbents like Windsurf, Anysphere's Cursor, and GitHub Copilot.

Mistral Code, a fork of the open source project Continue, is an AI-powered coding assistant that bundles Mistral's models, an "in-IDE" assistant, local deployment options, and enterprise tools into a single package. A private beta is available as of Wednesday for JetBrains development platforms and Microsoft's VS Code.

"Our goal with Mistral Code is simple: deliver best-in-class coding models to enterprise developers, enabling everything from instant completions to multi-step refactoring through an integrated platform deployable in the cloud, on reserved capacity, or air-gapped, on-prem GPUs," Mistral wrote in a blog post provided to TechCrunch.

AI programming assistants are growing increasingly popular. While they still struggle to codequalitysoftware, their promise to boost coding productivity is pushing companies and developers to adopt them rapidly. One recent poll found that 76% of developers have used or were planning to use AI tools in their development processes last year.

Mistral Code is said to be powered by a combination of in-house models including Codestral (for code autocomplete), Codestral Embed (for code search and retrieval), Devstral (for "agentic" coding tasks), and Mistral Medium (for chat assistance). The client supports more than 80 programming languages and a number of third-party plug-ins, and can reason over things like files, terminal outputs, and issues, the company said.

[...] Mistral said it plans to continue making improvements to Mistral Code and contribute at least a portion of those upgrades to the Continue open source project.

Another day, another whatever these things are.

Original Submission

Arthur T Knackerbracket has processed the following story:

But it could just be a coincidence...

Just days after a new Washington crackdown on semiconductor design software exports to China, which banned companies like Synopsys from offering their services to clients in the country, access to some vital services appears to have been quietly restored. Notably, the turnabout comes within days of a high-level phone call between President Trump and Xi Jinping, according to Digitimes.

Digitimes reports that following the call, which took place on June 5, there has been a shift in the semiconductor market pertaining to the software used by companies in semiconductor design. Notably, several local Chinese IC design engineers and companies have reported that access to Synopsys' SolvNetPlus platform and Cadence's Support Portal has now been restored.

The report notes that it's unclear at this stage whether the change marks an isolated dispensation for certain clients or a broader relaxing of tension and restrictions between China and the US.

At the end of May, Synopsys paused its sales and services offerings in China and suspended its financial guidance, after receiving a BIS letter from the Bureau of Industry and Security of the U.S. Department of Commerce. The letter reportedly disclosed "new export restrictions related to China," and further reports claimed Synopsys had told staff to halt services and sales in the country and stop taking new orders to comply.

The EDA ban was expected to hit Chinese companies, notably Xiaomi and Lenovo, hard. Specifically, Chinese companies rely on American software for the manufacture and production of more advanced semiconductors like those used in AI processing. Reports at the time indicated that while China did have some homegrown EDA capacity, it was only "usable" on 7nm nodes and older, a weighty concern for future production.

These concerns appear to have been short-lived, however. Digitimes reports that following the phone call on June 5 (it is unclear if EDA access was specifically discussed), multiple Chinese IC firms reported successfully logging into SolvNetPlus with no issues.

Digitimes cites industry analysts who postulate whether the call might have garnered a software approach from the U.S. in regard to technology export restrictions and a gradual restoration of some services.

Original Submission