SoylentNews

janrinok writes:

New research indicates that complex life began forming almost a billion years earlier than previously thought:

New research has uncovered that complex life began forming much earlier, and across a longer timeframe, than scientists had previously assumed. The findings offer fresh insight into the environmental conditions that shaped early evolution and call into question several longstanding scientific ideas in this field.

Led by the University of Bristol and published today in Nature (December 3), the study reports that complex organisms arose well before oxygen became abundant in Earth's atmosphere. Oxygen had long been thought to be essential for the development of advanced life, but the results indicate that this requirement may not hold for the earliest stages of evolution.

"The Earth is approximately 4.5 billion years old, with the first microbial life forms appearing over 4 billion years ago. These organisms consisted of two groups – bacteria and the distinct but related archaea, collectively known as prokaryotes," said co-author Anja Spang, from the Department of Microbiology & Biogeochemistry at the Royal Netherlands Institute for Sea Research.

Prokaryotes dominated the planet for hundreds of millions of years before more complex eukaryotic cells emerged. This latter group includes algae, fungi, plants, and animals.

Davide Pisani, Professor of Phylogenomics in the School of Biological Sciences at the University of Bristol and co-author, explained: "Previous ideas on how and when early prokaryotes transformed into complex eukaryotes have largely been in the realm of speculation. Estimates have spanned a billion years, as no intermediate forms exist and definitive fossil evidence has been lacking."

To address these uncertainties, the international team expanded upon the existing 'molecular clocks' technique, which estimates when species last shared a common ancestor.

"The approach was two-fold: by collecting sequence data from hundreds of species and combining this with known fossil evidence, we were able to create a time-resolved tree of life. We could then apply this framework to better resolve the timing of historical events within individual gene families," added co-lead author Professor Tom Williams in the Department of Life Sciences at the University of Bath.

By comparing more than 100 gene families across multiple biological systems and focusing on traits that differentiate eukaryotes from prokaryotes, the researchers began reconstructing the sequence of events that shaped the rise of complex life.

janrinok writes:

https://appleinsider.com/articles/25/12/01/us-wants-laws-to-force-app-store-age-checks-despite-apples-existing-protections

The United States wants big tech companies like Apple to protect children online by adding age verification safeguards to the App Store. It's a political push that completely ignores what protections Apple already provides to parents and children.

Lawmakers have been particularly keen to protect children from online dangers, and have repeatedly demanded big tech companies like Apple and Google do more to help. In the latest attempt to make big tech bend to its demands, the U.S. government is going after the App Store.

The App Store Accountability Act (ASA) was introduced in May as a way for parents to get more tools to protect their children online. In late November, the ASA was brought up in Congress as part of a raft of measures to keep kids safe online, led by the Kids Online Safety Act (KOSA).

It's also due to be discussed as part of a House Energy and Commerce Committee hearing. A report on Monday by The Verge says that the discussion will look at a nine-bill package of measures.

Under the ASA, introduced by Sen. Mike Lee (R-UT) and Rep. John James (R-MI), app storefronts, like Apple's App Store and Google Play, will be required to verify the age of all users in a privacy-protecting way. The result can then be used to limit what apps would be accessible to the user, if they are deemed too young.

Accounts that are used by minors must be linked to a parental account, which would need to provide parental consent for downloading apps or making purchases. App Stores also have to meet standards like providing secure age verification and accurate app age ratings.

While there are some measures in place, such as California's age check law, the intention of the bills are to make things the same across the United States. Instead of dealing with various laws and measures in different states, there would be one set of overriding rules that offer close to the same protections.

Much like many attempts to legislate technology, there is a difference between intention and looking at reality. Had lawmakers looked closer, they would see that Apple already has something in place that does just what they asked.

An Anonymous Coward writes:

https://www.phoronix.com/news/CDE-2.5.3-Desktop

Two years and one week since the prior point release, Common Desktop Environment 2.5.3 is now available as the latest iteration of this Unix desktop environment built around the Motif toolkit. CDE has been open-source for more than a decade now but its development not exactly brisk. But for those resisting the likes of Wayland and other modern display tech -- especially with KDE announcing today Plasma 6.8 will be Wayland-exclusive -- CDE 2.5.3 is now available.

CDE 2.5.3 ships with various bug fixes, dtwm now supporting more mouse buttons, some compiler fixes and resolving some warnings, a systemd service file for dtlogin is added, and other mostly minor changes. Besides the dtlogin systemd service file, perhaps most notable otherwise are the fixes for satisfying the GCC 15 compiler.

CDE 2.5.3 downloads and more details on this new point release can be found via SourceForge. Yes, another sign of its times.

While on the topic of CDE, you may also recall the separate NxCDE as a modern CDE effort. NxCDE hasn't seen any release since June 2023 nor any commits to its GitHub repository now in two years.

Original Submission

JoeMerchant writes:

Physicists have developed a "physics shortcut" that allows ordinary laptops to solve complex quantum dynamics problems, a feat previously reserved for supercomputers and AI models (Live Science). The breakthrough, from the University at Buffalo, is an extension of a decades-old method called the truncated Wigner approximation (TWA).

TWA is a semiclassical approach that simplifies quantum math by retaining necessary quantum behavior while discarding less critical details. Historically, applying TWA required re-deriving complicated math for every new problem, making it inaccessible. The team transformed this into a user-friendly "conversion table" that translates a quantum problem into solvable equations, allowing physicists to get usable results on a consumer laptop within hours (University at Buffalo).

This new, practical approach significantly lowers the computational cost and makes exploring certain quantum phenomena much easier. It's hoped that this will save supercomputing resources for the truly intractable quantum systems, while allowing more common quantum dynamics to be studied efficiently on accessible consumer-grade computers (ScienceDaily).

Original Submission

An Anonymous Coward writes:

https://www.npr.org/2025/12/03/g-s1-100212/san-francisco-sues-manufacturers-ultraprocessed-foods

The city of San Francisco filed a lawsuit against some of the nation's top food manufacturers on Tuesday, arguing that ultraprocessed food from the likes of Coca-Cola and Nestle are responsible for a public health crisis.

City Attorney David Chiu named 10 companies in the lawsuit, including the makers of such popular foods as Oreo cookies, Sour Patch Kids, Kit Kat, Cheerios and Lunchables. The lawsuit argues that ultraprocessed foods are linked to diseases such as Type 2 diabetes, fatty liver disease and cancer.

"They took food and made it unrecognizable and harmful to the human body," Chiu said in a news release. "These companies engineered a public health crisis, they profited handsomely, and now they need to take responsibility for the harm they have caused."

Ultraprocessed foods include candy, chips, processed meats, sodas, energy drinks, breakfast cereals and other foods that are designed to "stimulate cravings and encourage overconsumption," Chiu's office said in the release. Such foods are "formulations of often chemically manipulated cheap ingredients with little if any whole food added," Chiu wrote in the lawsuit.

[...] U.S. Health Secretary Robert F. Kennedy Jr. has been vocal about the negative impact of ultraprocessed foods and their links to chronic disease and has targeted them in his Make America Healthy Again campaign. Kennedy has pushed to ban such foods from the Supplemental Nutrition Assistance Program for low-income families.

An August report by the U.S. Centers for Disease Control and Prevention found that most Americans get more than half their calories from ultraprocessed foods.

[...] "Mounting research now links these products to serious diseases—including Type 2 diabetes, fatty liver disease, heart disease, colorectal cancer, and even depression at younger ages," University of California, San Francisco, professor Kim Newell-Green said in the news release.

The lawsuit argues that by producing and promoting ultraprocessed foods, the companies violate California's Unfair Competition Law and public nuisance statute. It seeks a court order preventing the companies from "deceptive marketing" and requiring them to take actions such as consumer education on the health risks of ultraprocessed foods and limiting advertising and marketing of ultraprocessed foods to children.

It also asks for financial penalties to help local governments with health care costs caused by the consumption of ultraprocessed foods.

Original Submission

janrinok writes:

A boycott is unlikely to work:

The chaotic state of RAM prices continues to impact the industry. According to a new report, motherboard sales have fallen by as much as 50% as a result of the crisis. It has also led to gamers calling for a RAM boycott in hopes of easing the situation, but the reality is that such a move is unlikely to work.

We've covered the memory-pricing crisis since it began, including this deep dive into the problem and how it's caused by demand from AI data centers that require massive amounts of DRAM.

A new report Japanese from outlet Gazlog[site in Chinese] states that out-of-control DDR5 prices are impacting motherboard sales, forcing manufacturers such as Asus, MSI, and Gigabyte to significantly lower their sales targets.

DDR5 prices are simply stupid right now. A 64GB kit is now more expensive than a PS5 console or an RTX 5070. We've even seen several stores remove fixed pricing signs from DDR5 displays, relying on market rates because costs are changing so rapidly each day.

The problem for motherboard makers is that people upgrading from DDR4 or older systems – along with first-time builders --need DDR5 to pair with their shiny new boards. But with prices so high, it's a bad time to buy.

The result is a 40-50% decrease in motherboard sales compared to the same period a year earlier, writes Gazlog, leading to a lowering of sales targets. It's expected that CPU sales will eventually experience a similar fall in sales due to the RAM situation.

In an attempt to fight back, there are now calls on Reddit for gamers to boycott RAM completely in the hope that prices will return to normal

Unfortunately, the rallying cry is likely to have very little, if any, effect. The biggest issue, as we know, is that DRAM supply and future manufacturing capacity have already been bought out by companies to support their aggressive data center-building plans, causing the shortage.

FunkyLich writes:

A Debian developer gave out the news. Julian Andres Klode wrote in the mailing lists that APT (the package manager tool of Debian Linux) begin requiring a Rust compiler. It goes like:

"I plan to introduce hard Rust dependencies and Rust code into APT, no earlier than May 2026. This extends at first to the Rust compiler and standard library, and the Sequoia ecosystem.
In particular, our code to parse .deb, .ar, .tar, and the HTTP signature verification code would strongly benefit from memory safe languages and a stronger approach to unit testing."

source: https://lists.debian.org/deity/2025/10/msg00071.html

I admit to not knowing enough of the "how"-s and "gotcha!"-s of Debian's development to be able to give an educated opinion about this news. But in the overall and what I have seen happening as of late, anything Rust has felt a little too eager to be wedged into any cog and bolt of Linuxlandia as can be possible.

My dormant prophetic eye almost sees a vague vision of something that resembles the Unix Wars of the 90s, but this time happening to Linux, as soon as Torvalds will eventually retire from his role as the benevolent emperor of Linux. The commercial vultures will come from all sides to "improve" it in any way that would suit them in their business, injecting in it endless bloat and useless features that noone asked about, and there will be noone there to stop it with a firm "no!" (and a middle finger up when deserved) like Torvalds does and has guarded Linux from such assaults so far.

Original Submission

An Anonymous Coward writes:

The Linux phone features 12GB RAM, up to 2TB storage, a 6.36-inch FullHD AMOLED display, and a user-replaceable 5,500mAh battery:

Jolla kicked off a campaign for a new Jolla Phone, which they call the independent European Do It Together (DIT) Linux phone, shaped by the people who use it.

The new Jolla Phone is powered by a high-performing Mediatek 5G SoC, and features 12GB RAM, 256GB storage that can be expanded to up to 2TB with a microSDXC card, a 6.36-inch FullHD AMOLED display with ~390ppi, 20:9 aspect ratio, and Gorilla Glass, and a user-replaceable 5,500mAh battery.

The Linux phone also features 4G/5G support with dual nano-SIM and a global roaming modem configuration, Wi-Fi 6 wireless, Bluetooth 5.4, NFC, 50MP Wide and 13MP Ultrawide main cameras, front front-facing wide-lens selfie camera, fingerprint reader on the power key, a user-changeable back cover, and an RGB indication LED.

On top of that, the new Jolla Phone promises a user-configurable physical Privacy Switch that lets you turn off the microphone, Bluetooth, Android apps, or whatever you wish.

The device will be available in three colors, including Snow White, Kaamos Black, and The Orange. All the specs of the new Jolla Phone were voted on by Sailfish OS community members over the past few months.

Honouring the original Jolla Phone form factor and design, the new model ships with the Sailfish OS (with support for Android apps), a Linux-based European alternative to dominating mobile operating systems, and promises a minimum of 5 years of support, no tracking, no calling home, and no hidden analytics.

"Mainstream phones send vast amounts of background data. A common Android phone sends megabytes of data per day to Google even if the device is not used at all. Sailfish OS stays silent unless you explicitly allow connections," said Jolla.

The new Jolla Phone is now available for pre-order for 99 EUR and will only be produced if at least 2000 pre-orders are reached in one month from today [goal met on 7 December -- Ed.], until January 4th, 2026. The full price of the Linux phone will be 499 EUR (incl. local VAT), and the 99 EUR pre-order price will be fully refundable and deducted from the full price.

The device will be manufactured and sold in Europe, but Jolla says that it will design the cellular band configuration to enable global travelling as much as possible, including e.g. roaming in the U.S. carrier networks. The initial sales markets are the EU, the UK, Switzerland, and Norway.

TECH SPECS:

*SPOILER* (click to show) *SPOILER* (click to hide)

        SoC: High performant Mediatek 5G platform
        RAM: 12GB
        Storage: 256GB + expandable with microSDXC
        Cellular: 4G + 5G with dual nano-SIM and global roaming modem configuration
        Display: 6.36" ~390ppi FullHD AMOLED, aspect ratio 20:9, Gorilla Glass
        Cameras: 50MP Wide + 13MP Ultrawide main cameras, front facing wide-lens selfie camera
        Battery: approx. 5,500mAh, user replaceable
        Connectivity: WiFi 6, BT 5.4, NFC
        Dimensions: ~158 x 74 x 9mm
        Other: Power key fingerprint reader, user changeable backcover, RGB indication LED, Privacy Switch

Privacy by Design

        No tracking, no calling home, no hidden analytics
        User configurable physical Privacy Switch - turn off you microphone, bluetooth, Android apps, or whatever you wish

Scandinavian styling in its pure form

        Honouring the original Jolla Phone form factor and design
        Replaceable back cover
        Available in three distinct colours inspired by Nordic nature

Performance Meets Privacy

        5G with dual nano-SIM
        12GB RAM and 256GB storage expandable up to 2TB
        Sailfish OS 5
        Support for Android apps with Jolla AppSupport
        User replaceable back cover with colour options
        User replaceable battery
        Physical Privacy Switch

Original Submission

SomeGuy writes:

In 1983, before the release of Microsoft Windows, Digital Research GEM, or Apple Macintosh, the office software giant VisiCorp released a graphical multitasking operating system for the IBM PC called Visi On.

It was an "open system", so anyone could make programs for it. Well, if they owned an expensive VAX computer and were prepared to shell out $7,000 on the Software Development Kit.

42 years later, although the mainframe based development environment has been lost to time, enthusiast Nina Kalinina has pulled apart Visi Corp Visi On to reveal some of the strange and curious internals.

https://git.sr.ht/~nkali/vision-sdk/tree/main/note/index.md

In this article, they document some of the internals, clear up some marketing misconceptions, discover some interesting Visi On quirks, and even provide a new application for it.

Original Submission

janrinok writes:

https://boingboing.net/2025/12/03/waymo-drives-straight-into-active-police-scene-ignores-chaos.html

Waymo self-driving cars may know how to get you from point A to point B most of the time, but there are some things Waymos still don't understand. One of these things is that you shouldn't drive directly into an intense scene where multiple police cars are surrounding a vehicle. Unfortunately, that's precisely what Waymo did in this video [Instagram].[Not reviewed --JE]

As the Waymo cruises by, a gentleman gets out of his car and onto the ground at police command. The Waymo doesn't care for this chaos — it simply wants to keep truckin'. I really wish we could see what the passenger was doing during this whole ordeal.

Videos like these are why I refuse to ride in Waymo. These cars seem to lack critical thinking when it's most needed. Luckily, this kind of situation (when it ends safely) makes for a hilarious video.

Original Submission

upstart writes:

New research offers clues about why some prompt injection attacks may succeed:

Researchers from MIT, Northeastern University, and Meta recently released a paper suggesting that large language models (LLMs) similar to those that power ChatGPT may sometimes prioritize sentence structure over meaning when answering questions. The findings reveal a weakness in how these models process instructions that may shed light on why some prompt injection or jailbreaking approaches work, though the researchers caution their analysis of some production models remains speculative since training data details of prominent commercial AI models are not publicly available.

The team, led by Chantal Shaib and Vinith M. Suriyakumar, tested this by asking models questions with preserved grammatical patterns but nonsensical words. For example, when prompted with "Quickly sit Paris clouded?" (mimicking the structure of "Where is Paris located?"), models still answered "France."

This suggests models absorb both meaning and syntactic patterns, but can overrely on structural shortcuts when they strongly correlate with specific domains in training data, which sometimes allows patterns to override semantic understanding in edge cases. The team plans to present these findings at NeurIPS later this month.

As a refresher, syntax describes sentence structure—how words are arranged grammatically and what parts of speech they use. Semantics describes the actual meaning those words convey, which can vary even when the grammatical structure stays the same.

Semantics depends heavily on context, and navigating context is what makes LLMs work. The process of turning an input, your prompt, into an output, an LLM answer, involves a complex chain of pattern matching against encoded training data.

To investigate when and how this pattern-matching can go wrong, the researchers designed a controlled experiment. They created a synthetic dataset by designing prompts in which each subject area had a unique grammatical template based on part-of-speech patterns. For instance, geography questions followed one structural pattern while questions about creative works followed another. They then trained Allen AI's Olmo models on this data and tested whether the models could distinguish between syntax and semantics.

The analysis revealed a "spurious correlation" where models in these edge cases treated syntax as a proxy for the domain. When patterns and semantics conflict, the research suggests, the AI's memorization of specific grammatical "shapes" can override semantic parsing, leading to incorrect responses based on structural cues rather than actual meaning.

In layperson terms, the research shows that AI language models can become overly fixated on the style of a question rather than its actual meaning. Imagine if someone learned that questions starting with "Where is..." are always about geography, so when you ask "Where is the best pizza in Chicago?", they respond with "Illinois" instead of recommending restaurants based on some other criteria. They're responding to the grammatical pattern ("Where is...") rather than understanding you're asking about food.

An Anonymous Coward writes:

By my count, Linux has over 11% of the desktop market. Here's how I got that number - and why people are making the leap:

My colleague Jack Wallen and I have been telling you for a while now that you should switch from Windows to the Linux desktop. Sounds like some of you have been listening.

The proof of the pudding comes from various sources. First, with Windows 10 nearing the end of its supported life, we told you to consider switching from Windows to Linux Mint or another Windows-like Linux distribution. What do we find now?

Zorin OS, an excellent Linux desktop, reports that its latest release, "Zorin OS 18 has amassed 1 million downloads in just over a month since its release." What makes it especially interesting is that over "78% of these downloads came from Windows" users.

[...] Many have already been making the leap. By May 2025, StatCounter data showed the Linux desktop had grown from a minute 1.5% global desktop share in 2020 to above 4% in 2024, and was at a new American high of above 5% by 2025.

In StatCounter's latest US numbers, which cover through October, Linux shows up as only 3.49%. But if you look closer, "unknown" accounts for 4.21%. Allow me to make an educated guess here: I suspect those unknown desktops are actually running Linux. What else could it be? FreeBSD? Unix? OS/2? Unlikely.

In addition, ChromeOS comes in at 3.67%, which strikes me as much too low. Leaving that aside, ChromeOS is a Linux variant. It just uses the Chrome web browser for its interface rather than KDE Plasma, Cinnamon, or another Linux desktop environment. Put all these together, and you get a Linux desktop market share of 11.37%. Now we're talking.

If you want to look at the broader world of end-user operating systems, including phones and tablets, Linux comes out even better. In the US, where we love our Apple iPhones, Android -- yes, another Linux distro -- boasts 41.71% of the market share, according to StatCounter's latest numbers. Globally, however, Android rules with 72.55% of the market.

janrinok writes:

https://www.geekwire.com/2025/uw-nobel-winners-lab-releases-most-powerful-protein-design-tool-yet/

David Baker's lab at the University of Washington is announcing two major leaps in the field of AI-powered protein design. The first is a souped-up version of its existing RFdiffusion2 tool that can now design enzymes with performance nearly on par with those found in nature. The second is the release of a new, general-purpose version of its model, named RFdiffusion3, which the researchers are calling their most powerful and versatile protein engineering technology to date.

Last year, Baker received the Nobel Prize in Chemistry for his pioneering work in protein science, which includes a deep-learning model called RFdiffusion. The tool allows scientists to design novel proteins that have never existed. These machine-made proteins hold immense promise, from developing medicines for previously untreatable diseases to solving knotty environmental challenges.

Baker leads the UW's Institute for Protein Design, which released the first version of the core technology in 2023, followed by RFdiffusion2 earlier this year. The second model was fine-tuned for creating enzymes — proteins that orchestrate the transformation of molecules and dramatically speed up chemical reactions.

The latest accomplishments are being shared today in publications in the leading scientific journals Nature and Nature Methods, as well as a preprint last month on bioRxiv.

In the improved version of RFdiffusion2, the researchers took a more hands-off approach to guiding the technology, giving it a specific enzymatic task to perform but not specifying other features. Or as the team described it in a press release, the tool produces "blueprints for physical nanomachines that must obey the laws of chemistry and physics to function."

"You basically let the model have all this space to explore and ... you really allow it to search a really wide space and come up with great, great solutions," said Seth Woodbury, a graduate student in Baker's lab and author on both papers publishing today.

In addition to UW scientists, researchers from MIT and Switzerland's ETH Zurich contributed to the work.

The new approach is remarkable for quickly generating higher-performing enzymes. In a test of the tool, it was able to solve 41 out of 41 difficult enzyme design challenges, compared to only 16 for the previous version.

"When we designed enzymes, they're always an order of magnitude worse than native enzymes that evolution has taken billions of years to find," said Rohith Krishna, a postdoctoral fellow and lead developer of RFdiffusion2. "This is one of the first times that we're not one of the best enzymes ever, but we're in the ballpark of native enzymes."

The researchers successfully used the model to create proteins calls metallohydrolases, which accelerate difficult reactions using a precisely positioned metal ion and an activated water molecule. The engineered enzymes could have important applications, including the destruction of pollutants.

The promise of rapidly designed catalytic enzymes could unleash wide-ranging applications, Baker said.

"The first problem we really tackled with AI, it was largely therapeutics, making binders to drug targets," he said. "But now with catalysis, it really opens up sustainability."

The researchers are also working with the Gates Foundation to figure out lower-cost ways to build what are known as small molecule drugs, which interact with proteins and enzymes inside cells, often by blocking or enhancing their function to effect biological processes.

The Nature paper, titled "Computational design of metallohydrolases," was authored by Donghyo Kim, Seth Woodbury, Woody Ahern, Doug Tischer, Alex Kang, Emily Joyce, Asim Bera, Nikita Hanikel, Saman Salike, Rohith Krishna, Jason Yim, Samuel Pellock, Anna Lauko, Indrek Kalvet, Donald Hilvert and David Baker.

The Nature Methods paper, titled "Atom-level enzyme active site scaffolding using RFdiffusion2," was authored by Woody Ahern, Jason Yim, Doug Tischer, Saman Salike, Seth Woodbury, Donghyo Kim, Indrek Kalvet, Yakov Kipnis, Brian Coventry, Han Raut Altae-Tran, Magnus Bauer, Regina Barzilay, Tommi Jaakkola, Rohith Krishna and David Baker.

Original Submission

An Anonymous Coward writes:

https://www.osnews.com/story/143942/freebsd-15-0-released-with-pkgbase/

The FreeBSD team has released FreeBSD 15.0, and with it come several major changes, one of which you will surely want to know more about if you're a FreeBSD user. Since this change will eventually drastically change the way you use FreeBSD, we should get right into it.

Up until now, a full, system-wide update for FreeBSD – as in, updating both the base operating system as well as any packages you have installed on top of it – would use two separate tools: freebsd-update and the pkg package manager. You used the former to update the base operating system, which was installed as file sets, and the latter to update everything you had installed on top of it in the form of packages.

With FreeBSD 15.0, this is starting to change. Instead of using two separate tools, in 15.0 you can opt to deprecate freebsd-update and file sets, and rely entirely on pkg for updating both the base operating system as well as any packages you have installed, because with this new method, the base system moves from file sets to packages. When installing FreeBSD 15.0, the installer will ask you to choose between the old method, or the new pkg-only method.

Packages (pkgbase / New Method): The base system is installed as a set of packages from the "FreeBSD-base" repository. Systems installed this way are managed entirely using the pkg(8) tool. This method is used by default for all VM images and images published in public clouds. In FreeBSD 15.0, pkgbase is offered as a technology preview, but it is expected to become the standard method for managing base system installations and upgrades in future releases.

↫ FreeBSD 15.0 release announcement

As the release announcement notes, the net method is optional in FreeBSD 15 and will remain optional during the entire 15.x release cycle, but the plan is to deprecate freebsd-update and file sets entirely in FreeBSD 16.0. If you have an existing installation you wish to convert to using pkgbase, there's a tool called pkgbasify to do just that. It's sponsored by the FreeBSD Foundation, so it's not some random script.

Of course, there's way more in this release than just pkgbase. Of note is that the 32bit platforms i386, armv6, and 32-bit powerpc have been retired, but of course, 32bit code will continue to run on their 64bit counterparts. FreeBSD 15.0 also brings a native inotify implementation, a ton of improvements to the audio components, improved Intel Wi-Fi drivers, and so, so much more.

Original Submission

janrinok writes:

https://www.tomshardware.com/tech-industry/ibm-ceo-warns-trillion-dollar-ai-boom-unsustainable-at-current-infrastructure-costs

...says there is 'no way' that infrastructure costs can turn a profit

IBM CEO Arvind Krishna used an appearance on The Verge's Decoder podcast to question whether the capital spending now underway in pursuit of AGI can ever pay for itself. Krishna said today's figures for constructing and populating large AI data centers place the industry on a trajectory where roughly $8 trillion of cumulative commitments would require around $800 billion of annual profit simply to service the cost of capital.

The claim was tied directly to assumptions about current hardware, its depreciation, and energy, rather than any solid long-term forecasts, but it comes at a time when we've seen several companies one-upping one another with unprecedented, multi-year infrastructure projects.

Krishna estimated that filling a one-gigawatt AI facility with compute hardware requires around $80 billion. The issue is that deployments of this scale are moving from the drawing board and into practical planning stages, with leading AI companies proposing deployments with tens of gigawatts — and in some cases, beyond 100 gigawatts — each. Krishna said that, taken together, public and private announcements point to roughly one hundred gigawatts of currently planned capacity dedicated to AGI-class workloads.

At $80 billion per gigawatt, the total reaches $8 trillion. He tied those figures to the five-year refresh cycles common across accelerator fleets, arguing that the need to replace most of the hardware inside those data centers within that window creates a compounding effect on long-term capex requirements. He also placed the likelihood that current LLM-centric architectures reach AGI at between zero and 1% without new forms of knowledge integration.

Krishna pointed to depreciation as the part of the calculation most underappreciated by investors. AI accelerators are typically written down over five years, and he argued that the pace of architectural change means fleets must be replaced rather than extended. "You've got to use it all in five years because at that point, you've got to throw it away and refill it," he said.

Recent financial-market criticism has centred on similar concerns. Investor Michael Burry, for example, has raised questions about whether hyperscalers can continue stretching useful-life assumptions if performance gains and model sizes force accelerated retirement of older GPUs.

The IBM chief said that ultimately, he expects generative-AI tools in their current form to drive substantial enterprise productivity, but that his concern is the relationship between the physical scale of next-gen AI infrastructure and the economics required to support it. Companies committing to these huge, multi-gigawatt campuses and compressed refresh schedules must therefore demonstrate returns that match the unprecedented capital expenditure that Krishna outlined.

Original Submission