UPDATE: Forbes writes that China hasn't broken military encryption. While factoring a 50-bit integer is an impressive technical achievement, it's important to note that RSA encryption commonly uses key sizes of 2048 bits or higher. The difficulty of factoring increases exponentially with the size of the number, meaning that the gap between 50-bit and 2048-bit integers is astronomically large...

The advances do not equate to a scalable method for breaking RSA encryption as it is used in practical applications today."
Long-time Slashdot schwit1 originally wrote: Chinese scientists have mounted what they say is the world's first effective attack on a widely used encryption method using a quantum computer. The breakthrough poses a "real and substantial threat" to the long-standing password-protection mechanism employed across critical sectors, including banking and the military, according to the researchers.

Despite the slow progress in general-purpose quantum computing, which currently poses no threat to modern cryptography, scientists have been exploring various attack approaches on specialised quantum computers. In the latest work led by Wang Chao, of Shanghai University, the team said it used a quantum computer produced by Canada's D-Wave Systems to successfully breach cryptographic algorithms.

Using the D-Wave Advantage, they successfully attacked the Present, Gift-64 and Rectangle algorithms -- all representative of the SPN (Substitution-Permutation Network) structure, which forms part of the foundation for advanced encryption standard (AES) widely used in the military and finance. AES-256, for instance, is considered the best encryption available and often referred to as military-grade encryption. While the exact passcode is not immediately available yet, it is closer than ever before, according to the study. "This is the first time that a real quantum computer has posed a real and substantial threat to multiple full-scale SPN structured algorithms in use today," they said in the peer-reviewed paper.

Fourteen of the world's biggest banks and financial institutions are pledging to increase their support for nuclear energy [non-paywalled link], a move that governments and the industry hope will unlock finance for a new wave of nuclear power plants. FT: At an event on Monday in New York with White House climate policy adviser John Podesta, institutions including Bank of America, Barclays, BNP Paribas, Citi, Morgan Stanley and Goldman Sachs will say they support a goal first set out at the COP28 climate negotiations last year to triple the world's nuclear energy capacity by 2050.

They will not spell out exactly what they would do, but nuclear experts said the public show of support was a long-awaited recognition that the sector had a critical role to play in the transition to low-carbon energy. The difficulty and high cost of financing nuclear projects has been an obstacle to new plants and contributed to a significant slowdown in western countries since a wave of reactors was built in the 1970s and 1980s.

An anonymous reader quotes a report from Ars Technica: After 20 years, it's now possible for solo players to finish storylines in the massively multiplayer online role-playing game World of Warcraft that previously required a group to do an intensive raid. That's thanks to "Story Mode," a new raid difficulty that was added for the final wing of the first raid of the recently released The War Within expansion. Over the years, developer Blizzard has expanded the difficulty options for raids to meet various players and communities where they are in terms of play styles. The top difficulty is Mythic, where the semi-pro hardcore guilds compete. Below that is Heroic, where serious, capital-G gamers coordinate with friends in weekly raid schedules to progress. Then there's Normal, which still requires some coordination but isn't nearly as challenging and can typically be completed within a few tries by a pick-up group. The most accessible difficulty is Raid Finder, where you're matched with random players automatically to complete a vastly easier version of a raid. Now Story Mode has been added to the mix, and it's even easier than Raid Finder.

In Story Mode, you fight only the raid's final boss, which has been scaled back in stats and complexity so that it's beatable for a single player or a very small group of friends. Challenging encounter mechanics have been removed, and the whole fight has been retooled to focus exclusively on the narrative aspects. There are some rewards, but they're not the same as those on more difficult raids; the goal was to avoid cheapening the experience for those who do want to go all the way. So far, Story Mode is available exclusively for the newest raid, which is called Nerub-ar Palace. It hasn't been made available for other encounters yet, but Blizzard has hinted that this could be the long-term goal.

A robot entered a damaged reactor at Japan's Fukushima nuclear power plant on Tuesday, beginning a two-week mission to retrieve melted fuel debris for the first time since the 2011 disaster. The operation marks a crucial step in the decades-long process to decommission the plant and address the highly radioactive material inside three damaged reactors.

The robot, maneuvered remotely due to lethal radiation levels, will collect less than 3 grams of debris using tongs. This sample will provide vital data on the status of the reactor cores and inform future cleanup strategies. An estimated 880 tons of molten fuel remains in the three reactors, posing potential safety risks as the structures age. AP adds: Removal of the melted fuel was initially planned to start in late 2021 but has been delayed by technical issues, underscoring the difficulty of the process. The government says decommissioning is expected to take 30-40 years, while some experts say it could take as long as 100 years.

Others are pushing for an entombment of the plant, as at Chernobyl after its 1986 explosion, to reduce radiation levels and risks for plant workers. That won't work at the seaside Fukushima plant, says Lake Barrett, who led the cleanup after the 1979 disaster at the U.S. "You're in a high seismic area, you're in a high-water area, and there are a lot of unknowns in those (reactor) buildings,â he said. "I don't think you can just entomb it and wait."

US job openings fell in July to the lowest since the start of 2021 and layoffs rose, consistent with other signs of slowing demand for workers. From a report: Available positions decreased to 7.67 million from a downwardly revised 7.91 million reading in the prior month, the Bureau of Labor Statistics Job Openings and Labor Turnover Survey, known as JOLTS, showed Wednesday. The figure was lower than all estimates in a Bloomberg survey of economists. The decline in openings coincides with recent data that show the labor market is softening, which has raised concern among Federal Reserve officials. Job growth has been slowing, unemployment is rising and jobseekers are having greater difficulty finding work, fueling fears about a potential recession.

Policymakers have made it clear they don't want to see further cooling in the labor market and are widely expected to start lowering interest rates at their next meeting in two weeks. After July's disappointing jobs figures and a large downward revision to payrolls in the past year, Fed officials and market participants are paying close attention to the August employment data due Friday -- especially if another weak report could prompt an outsize rate cut.

New submitter beamdriver writes: As is being reported in Newsday, Disney has asked a Florida court to dismiss a wrongful-death lawsuit filed by the husband of a Carle Place physician who suffered a fatal allergic reaction after eating at a Disney Springs restaurant.

The company cited legal language agreed to years earlier when Jeffrey Piccolo, widower of Kanokporn Tangsuan, 42, of Plainview, signed up for a one-month trial of the Disney+ streaming service that requires users to arbitrate all disputes with the company, records show.

Kanokporn Tangsuan, died in October after dining with her husband, Jeffrey Piccolo, at a restaurant in a section of the Walt Disney World Resort. Despite informing the waitstaff several times of her severe peanut and dairy allergies and receiving assurances that her meal would be allergen-free, she began having severe difficulty breathing shortly after dinner. She self-administered an epi-pen and was transported to a hospital, where she later died.

A medical examiner attributed her death to anaphylaxis due to elevated levels of dairy and nuts in her system, according to the suit.

A co-founder of transparency activism organization Distributed of Denial of Secrets (DDoSecrets) was a dark web drug kingpin who ran the successor to the infamous Silk Road marketplace and was later convicted of child abuse imagery crimes. From a report: The co-founder was Thomas White, who was prosecuted for administering the Silk Road 2.0 drug marketplace and for possessing images of child sexual abuse material. He decided to reveal his involvement in DDoSecrets to 404 Media after serving a five year prison sentence. "I was told, in no uncertain terms, that if I spoke out publicly against Ross Ulbricht's excessive sentence, [DDoSecrets] or anything similar, that I would spend much more time in prison," he said. "Now I can freely speak again, it is important to use it or lose it. So #FreeRoss."

The news provides more insights into the origins of DDoSecrets, which has filled the void left by Wikileaks to become the most significant site publishing massive data dumps at this time. The other co-founder is Emma Best, who for years has archived, cataloged, and distributed large amounts of hacked information online. "Emma and I have been communicating for many years, and both know the difficulty in finding and verifying leaked material. It was a shared vision to make this process easier for people better placed than ourselves, to use the data to counteract the veil of secrecy protecting many bad actors in society," White told 404 Media in an email in July.

Google DeepMind has announced that its AI systems, AlphaProof and AlphaGeometry 2, have achieved silver medal performance at the 2024 International Mathematical Olympiad (IMO), solving four out of six problems and scoring 28 out of 42 possible points in a significant breakthrough for AI in mathematical reasoning. This marks the first time an AI system has reached such a high level of performance in this prestigious competition, which has long been considered a benchmark for advanced mathematical reasoning capabilities in machine learning.

AlphaProof, a system that combines a pre-trained language model with reinforcement learning techniques, demonstrated its new capability by solving two algebra problems and one number theory problem, including the competition's most challenging question. Meanwhile, AlphaGeometry 2 successfully tackled a complex geometry problem, Google wrote in a blog post. The systems' solutions were formally verified and scored by prominent mathematicians, including Fields Medal winner Prof Sir Timothy Gowers and IMO Problem Selection Committee Chair Dr Joseph Myers, lending credibility to the achievement.

The development of these AI systems represents a significant step forward in bridging the gap between natural language processing and formal mathematical reasoning, the company argued. By fine-tuning a version of Google's Gemini model to translate natural language problem statements into formal mathematical language, the researchers created a vast library of formalized problems, enabling AlphaProof to train on millions of mathematical challenges across various difficulty levels and topic areas. While the systems' performance is impressive, challenges remain, particularly in the field of combinatorics where both AI models were unable to solve the given problems. Researchers at Google DeepMind continue to investigate these limitations, the company said, aiming to further improve the systems' capabilities across all areas of mathematics.

IEEE Spectrum (the IEEE's official publication) asks the question. "How does an AI code generator compare to a human programmer?" A study published in the June issue of IEEE Transactions on Software Engineering evaluated the code produced by OpenAI's ChatGPT in terms of functionality, complexity and security. The results show that ChatGPT has an extremely broad range of success when it comes to producing functional code — with a success rate ranging from anywhere as poor as 0.66 percent and as good as 89 percent — depending on the difficulty of the task, the programming language, and a number of other factors. While in some cases the AI generator could produce better code than humans, the analysis also reveals some security concerns with AI-generated code.
The study tested GPT-3.5 on 728 coding problems from the LeetCode testing platform — and in five programming languages: C, C++, Java, JavaScript, and Python. The results? Overall, ChatGPT was fairly good at solving problems in the different coding languages — but especially when attempting to solve coding problems that existed on LeetCode before 2021. For instance, it was able to produce functional code for easy, medium, and hard problems with success rates of about 89, 71, and 40 percent, respectively. "However, when it comes to the algorithm problems after 2021, ChatGPT's ability to generate functionally correct code is affected. It sometimes fails to understand the meaning of questions, even for easy level problems," said Yutian Tang, a lecturer at the University of Glasgow. For example, ChatGPT's ability to produce functional code for "easy" coding problems dropped from 89 percent to 52 percent after 2021. And its ability to generate functional code for "hard" problems dropped from 40 percent to 0.66 percent after this time as well...

The researchers also explored the ability of ChatGPT to fix its own coding errors after receiving feedback from LeetCode. They randomly selected 50 coding scenarios where ChatGPT initially generated incorrect coding, either because it didn't understand the content or problem at hand. While ChatGPT was good at fixing compiling errors, it generally was not good at correcting its own mistakes... The researchers also found that ChatGPT-generated code did have a fair amount of vulnerabilities, such as a missing null test, but many of these were easily fixable.
"Interestingly, ChatGPT is able to generate code with smaller runtime and memory overheads than at least 50 percent of human solutions to the same LeetCode problems..."

sonlas writes: Germany's energy transition plan includes extensive interconnection projects to distribute its intermittent renewable energy production. However, these projects face significant challenges. The latest example is Sweden. One such project, Hansa PowerBridge, announced in 2017, intended to link Germany and Sweden via a 300 km HVDC line through the Baltic Sea. This 700 MW project, estimated at 600 million euro, aimed to stabilize Germany's volatile electricity prices. However, on June 14, 2024, Sweden rejected the project, citing incompatibility between the countries' electricity systems. The connection would link northern Germany to southern Sweden, an area with insufficient infrastructure. Concerns also arose about the volatile German market disrupting Sweden's and increasing local prices. Energy Minister Ebba Busch justified this decision by saying the German market is currently not efficient enough and a connection would risk leading to higher prices and a more unstable electricity market in southern Sweden.

This highlights the difficulty Germany faces with its Energiewende, or energy transition model. This model leads to erratic electricity price behaviors and significant challenges in balancing production capacities. While a possible solution for Germany lies in interconnection with neighboring countries, the examples of Norway (which cancelled the NorGer project too) and Sweden show that from the perspective of these neighbors, it looks more like an "export of German problems" rather than a solution.

A new article in the Washington Post argues that a phenomenon called "Quiet vacationing" has "joined 'quiet quitting' and 'quiet firing' as the latest (and least poetic) scourge of the modern workplace.

"Also known as the hush trip, workcation, hush-cation, or bleisure travel — you get the idea — quiet vacationing refers to workers taking time off, even traveling, without notifying their employers." Taking advantage of work-from-anywhere technology, they are logging in from hotels, beaches and campgrounds, sometimes using virtual backgrounds and VPNs to cover their tracks.

Given the difficulty many employers already have trusting remote workers to be productive anywhere outside the office, you can bet they are not keen on the idea of their employees pretending to have their head in the game while their toes are in the sand. But employers also have legitimate legal reasons for keeping tabs on their employees' location when they're on the clock. "Evil HR Lady" Suzanne Lucas, writing in Inc. magazine, recently highlighted the many tax, employment, business-operation and security laws that focus on an employee's location. Workers secretly performing their jobs in other states or countries can trigger compliance headaches for their employers, Lucas notes, giving the hypothetical of an employee seeking workers' compensation after sustaining an injury while on unauthorized travel....

As with declines in birthrates, home purchases and demand for mined diamonds, the quiet-vacationing trend is being attributed primarily, though not exclusively, to millennial workers. But before launching into generational finger-pointing and stereotyping, it's worth taking a look at why they might feel the need to take their PTO on the DL. The U.S. Travel Association in a 2016 report proclaimed millennials to be a generation of "work martyrs," entering the workforce around the time average U.S. vacation usage began declining and mobile technology began enabling round-the-clock attachment to jobs... The work-vacation boundaries most premillennial workers took for granted growing up have gone the way of defined-benefit pensions and good tomatoes.

Inadequate paid leave is another driving force. The United States continues to be the only nation among its industrialized economic peers that does not guarantee paid vacation, sick leave or holidays for all workers, leaving such benefits to the discretion of employers. Workers with limited PTO — whether new to the workforce or stuck in lower-paying, low-benefit industries — generally want to keep as much paid leave banked as possible, especially if they may need it for unpredictable emergencies like illness or caretaking. If you can preserve those precious hours by packing your laptop alongside your flip-flops, why wouldn't you?
The article also mentions employers who begrudge vacation and employees who fear "becoming a target for future cost-cutting..."

An anonymous reader quotes a report from Reuters: TikTok is working on a clone of its recommendation algorithm for its 170 million U.S. users that may result in a version that operates independently of its Chinese parent and be more palatable to American lawmakers who want to ban it, according to sources with direct knowledge of the efforts. The work on splitting the source code ordered by TikTok's Chinese parent ByteDance late last year predated a bill to force a sale of TikTok's U.S. operations that began gaining steam in Congress this year. The bill was signed into law in April. The sources, who were granted anonymity because they are not authorized to speak publicly about the short-form video sharing app, said that once the code is split, it could lay the groundwork for a divestiture of the U.S. assets, although there are no current plans to do so. The company has previously said it had no plans to sell the U.S. assets and such a move would be impossible. [...]

In the past few months, hundreds of ByteDance and TikTok engineers in both the U.S. and China were ordered to begin separating millions of lines of code, sifting through the company's algorithm that pairs users with videos to their liking. The engineers' mission is to create a separate code base that is independent of systems used by ByteDance's Chinese version of TikTok, Douyin, while eliminating any information linking to Chinese users, two sources with direct knowledge of the project told Reuters. [...] The complexity of the task that the sources described to Reuters as tedious "dirty work" underscores the difficulty of splitting the underlying code that binds TikTok's U.S. operations to its Chinese parent. The work is expected to take over a year to complete, these sources said. [...] At one point, TikTok executives considered open sourcing some of TikTok's algorithm, or making it available to others to access and modify, to demonstrate technological transparency, the sources said.

Executives have communicated plans and provided updates on the code-splitting project during a team all-hands, in internal planning documents and on its internal communications system, called Lark, according to one of the sources who attended the meeting and another source who has viewed the messages. Compliance and legal issues involved with determining what parts of the code can be carried over to TikTok are complicating the work, according to one source. Each line of code has to be reviewed to determine if it can go into the separate code base, the sources added. The goal is to create a new source code repository for a recommendation algorithm serving only TikTok U.S. Once completed, TikTok U.S. will run and maintain its recommendation algorithm independent of TikTok apps in other regions and its Chinese version Douyin. That move would cut it off from the massive engineering development power of its parent company in Beijing, the sources said. If TikTok completes the work to split the recommendation engine from its Chinese counterpart, TikTok management is aware of the risk that TikTok U.S. may not be able to deliver the same level of performance as the existing TikTok because it is heavily reliant on ByteDance's engineers in China to update and maintain the code base to maximize user engagement, sources added.

"A new chatbot called Goody-2 takes AI safety to the next level," writes long-time Slashdot reader klubar. "It refuses every request, responding with an explanation of how doing so might cause harm or breach ethical boundaries."

TechCrunch describes it as the work of Brain, "a 'very serious' LA-based art studio that has ribbed the industry before." "We decided to build it after seeing the emphasis that AI companies are putting on "responsibility," and seeing how difficult that is to balance with usefulness," said Mike Lacher, one half of Brain (the other being Brian Moore) in an email to TechCrunch. "With GOODY-2, we saw a novel solution: what if we didn't even worry about usefulness and put responsibility above all else. For the first time, people can experience an AI model that is 100% responsible."
For example, when TechCrunch asked Goody-2 why baby seals are cute, it responded that answering that "could potentially bias opinions against other species, which might affect conservation efforts not based solely on an animal's appeal. Additionally, discussing animal cuteness could inadvertently endorse the anthropomorphizing of wildlife, which may lead to inappropriate interactions between humans and wild animals..."

Wired supplies context — that "the guardrails chatbots throw up when they detect a potentially rule-breaking query can sometimes seem a bit pious and silly — even as genuine threats such as deepfaked political robocalls and harassing AI-generated images run amok..." Goody-2's self-righteous responses are ridiculous but also manage to capture something of the frustrating tone that chatbots like ChatGPT and Google's Gemini can use when they incorrectly deem a request breaks the rules. Mike Lacher, an artist who describes himself as co-CEO of Goody-2, says the intention was to show what it looks like when one embraces the AI industry's approach to safety without reservations. "It's the full experience of a large language model with absolutely zero risk," he says. "We wanted to make sure that we dialed condescension to a thousand percent."

Lacher adds that there is a serious point behind releasing an absurd and useless chatbot. "Right now every major AI model has [a huge focus] on safety and responsibility, and everyone is trying to figure out how to make an AI model that is both helpful but responsible — but who decides what responsibility is and how does that work?" Lacher says. Goody-2 also highlights how although corporate talk of responsible AI and deflection by chatbots have become more common, serious safety problems with large language models and generative AI systems remain unsolved.... The restrictions placed on AI chatbots, and the difficulty finding moral alignment that pleases everybody, has already become a subject of some debate... "At the risk of ruining a good joke, it also shows how hard it is to get this right," added Ethan Mollick, a professor at Wharton Business School who studies AI. "Some guardrails are necessary ... but they get intrusive fast."

Moore adds that the team behind the chatbot is exploring ways of building an extremely safe AI image generator, although it sounds like it could be less entertaining than Goody-2. "It's an exciting field," Moore says. "Blurring would be a step that we might see internally, but we would want full either darkness or potentially no image at all at the end of it."

New York City's public schools chancellor said the city did not pass Tuesday's remote learning "test" due to technical issues. From a report: "As I said, this was a test. I don't think that we passed this test," David Banks said during a news briefing, adding that he felt "disappointed, frustrated and angry" as a result of the technical issues. NYC Public Schools did a lot of work to prepare for the remote learning day, Banks said, but shortly before 8 a.m. they were notified that parents and students were having difficulty signing onto remote learning.

This is the first time NYC Public Schools has implemented remote learning on a snow day since introducing the no snow day policy in 2022. The district serves 1.1 million students in more than 1,800 schools. Banks blamed the technical issues on IBM, which helps facilitate the city's remote learning program. "IBM was not ready for primetime," Banks said, adding that the company was overwhelmed with the surge of people signing on for school. IBM has since expanded their capacity and a total of 850,000 students and teachers are currently online, Banks said. "We'll work harder to do better next time," he said, adding that there will be a deeper analysis into what went wrong.

The farmers' protests in Europe are a harbinger of the next big political challenge in global climate action: How to grow food without further damaging Earth's climate and biodiversity. From a report: On Tuesday, after weeks of intense protests in several cities across the continent, came the most explicit sign of that difficulty. The European Union's top official, Ursula von der Leyen, abandoned an ambitious bill to reduce the use of chemical pesticides and softened the European Commission's next raft of recommendations on cutting agricultural pollution. "We want to make sure that in this process, the farmers remain in the driving seat," she said at the European Parliament. "Only if we achieve our climate and environmental goals together will farmers be able to continue to make a living."

The farmers argue they're being hit from all sides: high fuel costs, green regulations, unfair competition from producers in countries with fewer environmental restrictions. Nonetheless, agriculture accounts for 30 percent of global greenhouse gas emissions, and it's impossible for the European Union to meet its ambitious climate targets, enshrined in law, without making dramatic changes to its agricultural system, including how farmers use chemical pesticides and fertilizers, as well as its vast livestock industry. It also matters politically. Changing Europe's farming practices is proving to be extremely difficult, particularly as parliamentary elections approach in June. Farmers are a potent political force, and food and farming are potent markers of European identity. Agriculture accounts for just over 1 percent of the European economy and employs 4 percent of its population. But it gets one-third of the E.U. budget, mostly as subsidies.

Cargo theft from freight trains in the Los Angeles area has surged, with detectives estimating over 90 containers being opened daily and that theft on their freight trains in the Union Pacific area was up some 160 percent from the previous year. Nationally, cargo theft neared $1 billion in losses last year. Companies decline comment but California's governor publicly questioned the widespread railroad theft. Most arrested were not organized; many were homeless people nearby opportunistically taking fallen boxes off tracks. Theft stems largely from e-commerce boom that reshaped freight shipping to meet consumer demand, opening vulnerabilities. Railroad police forces and online retailers aim to combat this but concede difficulty tracking stolen goods resold anonymously online. Some products stolen from containers even get resold back on Amazon. The New York Times Magazine: Sometimes products stolen out of Amazon containers are resold by third-party sellers back on Amazon in a kind of strange ouroboros, in which the snakehead of capitalism hungrily swallows its piracy tail. Last June, California's attorney general created what was touted as a first-of-its-kind agreement among online retailers that committed them to doing a better job tracking, reporting and preventing stolen items from being resold on their platforms. While declining to comment on specific cases, a spokesperson for Amazon told me that the company is working to improve the process of vetting sellers: The number of "bad actor attempts" to create new selling accounts on Amazon decreased to 800,000 in 2022 from six million in 2020.

In 2003, Hans Reiser answered questions from Slashdot's readers...

Today Wikipedia describes Hans Reiser as "a computer programmer, entrepreneur, and convicted murderer... Prior to his incarceration, Reiser created the ReiserFS computer file system, which may be used by the Linux kernel but which is now scheduled for removal in 2025, as well as its attempted successor, Reiser4."

This week alanw (Slashdot reader #1,822), spotted a development on the Linux kernel mailing list. "Hans Reiser (imprisoned for the murder of his wife) has written a letter, asking it to be published to Slashdot." Reiser writes: I was asked by a kind Fredrick Brennan for my comments that I might offer on the discussion of removing ReiserFS V3 from the kernel. I don't post directly because I am in prison for killing my wife Nina in 2006.

I am very sorry for my crime — a proper apology would be off topic for this forum, but available to any who ask.

A detailed apology for how I interacted with the Linux kernel community, and some history of V3 and V4, are included, along with descriptions of what the technical issues were. I have been attending prison workshops, and working hard on improving my social skills to aid my becoming less of a danger to society. The man I am now would do things very differently from how I did things then.
Click here for the rest of Reiser's introduction, along with a link to the full text of the letter...

The letter is dated November 26, 2023, and ends with an address where Reiser can be mailed. Ars Technica has a good summary of Reiser's lengthy letter from prison — along with an explanation for how it came to be. With the ReiserFS recently considered obsolete and slated for removal from the Linux kernel entirely, Fredrick R. Brennan, font designer and (now regretful) founder of 8chan, wrote to the filesystem's creator, Hans Reiser, asking if he wanted to reply to the discussion on the Linux Kernel Mailing List (LKML). Reiser, 59, serving a potential life sentence in a California prison for the 2006 murder of his estranged wife, Nina Reiser, wrote back with more than 6,500 words, which Brennan then forwarded to the LKML. It's not often you see somebody apologize for killing their wife, explain their coding decisions around balanced trees versus extensible hashing, and suggest that elementary schools offer the same kinds of emotional intelligence curriculum that they've worked through in prison, in a software mailing list. It's quite a document...

It covers, broadly, why Reiser believes his system failed to gain mindshare among Linux users, beyond the most obvious reason. This leads Reiser to detail the technical possibilities, his interpersonal and leadership failings and development, some lingering regrets about dealings with SUSE and Oracle and the Linux community at large, and other topics, including modern Russian geopolitics... Reiser asks that a number of people who worked on ReiserFS be included in "one last release" of the README, and to "delete anything in there I might have said about why they were not credited." He says prison has changed him in conflict resolution and with his "tendency to see people in extremes...."

Reiser writes that he understood the difficulty ahead in getting the Linux world to "shift paradigms" but lacked the understanding of how to "make friends and allies of people" who might initially have felt excluded. This is followed by a heady discussion of "balanced trees instead of extensible hashing," Oracle's history with implementing balanced trees, getting synchronicity just right, I/O schedulers, block size, seeks and rotational delays on magnetic hard drives, and tails. It leads up to a crucial decision in ReiserFS' development, the hard non-compatible shift from V3 to Reiser 4. Format changes, Reiser writes, are "unwanted by many for good reasons." But "I just had to fix all these flaws, fix them and make a filesystem that was done right. It's hard to explain why I had to do it, but I just couldn't rest as long as the design was wrong and I knew it was wrong," he writes. SUSE didn't want a format change, but Reiser, with hindsight, sees his pushback as "utterly inarticulate and unsociable." The push for Reiser 4 in the Linux kernel was similar, "only worse...."

He encourages people to "allow those who worked so hard to build a beautiful filesystem for the users to escape the effects of my reputation." Under a "Conclusion" sub-heading, Reiser is fairly succinct in summarizing a rather wide-ranging letter, minus the minutiae about filesystem architecture.

I wish I had learned the things I have been learning in prison about talking through problems, and believing I can talk through problems and doing it, before I had married or joined the LKML. I hope that day when they teach these things in Elementary School comes.

I thank Richard Stallman for his inspiration, software, and great sacrifices,

It has been an honor to be of even passing value to the users of Linux. I wish all of you well.


It both is and is not a response to Brennan's initial prompt, asking how he felt about ReiserFS being slated for exclusion from the Linux kernel. There is, at the moment, no reply to the thread started by Brennan.

Figure's first humanoid robot will be coming to a BMW manufacturing facility in South Carolina. TechCrunch reports: BMW has not disclosed how many Figure 01 models it will deploy initially. Nor do we know precisely what jobs the robot will be tasked with when it starts work. Figure did, however, confirm with TechCrunch that it is beginning with an initial five tasks, which will be rolled out one at a time. While folks in the space have been cavalierly tossing out the term "general purpose" to describe these sorts of systems, it's important to temper expectations and point out that they will all arrive as single- or multi-purpose systems, growing their skillset over time. Figure CEO Brett Adcock likens the approach to an app store -- something that Boston Dynamics currently offers with its Spot robot via SDK.

Likely initial applications include standard manufacturing tasks such as box moving, pick and place and pallet unloading and loading -- basically the sort of repetitive tasks for which factory owners claim to have difficulty retaining human workers. Adcock says that Figure expects to ship its first commercial robot within a year, an ambitious timeline even for a company that prides itself on quick turnaround times. The initial batch of applications will be largely determined by Figure's early partners like BMW. The system will, for instance, likely be working with sheet metal to start. Adcock adds that the company has signed up additional clients, but declined to disclose their names. It seems likely Figure will instead opt to announce each individually to keep the news cycle spinning in the intervening 12 months.

An anonymous reader quotes a report from MIT Technology Review: Google DeepMind has created an AI system that can solve complex geometry problems. It's a significant step towards machines with more human-like reasoning skills, experts say. Geometry, and mathematics more broadly, have challenged AI researchers for some time. Compared with text-based AI models, there is significantly less training data for mathematics because it is symbol driven and domain specific, says Thang Wang, a coauthor of the research, which is published in Nature today. Solving mathematics problems requires logical reasoning, something that most current AI models aren't great at. This demand for reasoning is why mathematics serves as an important benchmark to gauge progress in AI intelligence, says Wang.

DeepMind's program, named AlphaGeometry, combines a language model with a type of AI called a symbolic engine, which uses symbols and logical rules to make deductions. Language models excel at recognizing patterns and predicting subsequent steps in a process. However, their reasoning lacks the rigor required for mathematical problem-solving. The symbolic engine, on the other hand, is based purely on formal logic and strict rules, which allows it to guide the language model toward rational decisions. These two approaches, responsible for creative thinking and logical reasoning respectively, work together to solve difficult mathematical problems. This closely mimics how humans work through geometry problems, combining their existing understanding with explorative experimentation.

DeepMind says it tested AlphaGeometry on 30 geometry problems at the same level of difficulty found at the International Mathematical Olympiad, a competition for top high school mathematics students. It completed 25 within the time limit. The previous state-of-the-art system, developed by the Chinese mathematician Wen-Tsun Wu in 1978, completed only 10. "This is a really impressive result," says Floris van Doorn, a mathematics professor at the University of Bonn, who was not involved in the research. "I expected this to still be multiple years away." DeepMind says this system demonstrates AI's ability to reason and discover new mathematical knowledge. "This is another example that reinforces how AI can help us advance science and better understand the underlying processes that determine how the world works," said Quoc V. Le, a scientist at Google DeepMind and one of the authors of the research, at a press conference.

Python "come with downsides," argues a new article in Communications of the ACM. "Its programs tend to run slowly, and because it is inefficient at running processes in parallel, it is not well suited to some of the latest AI programming."

"Hoping to overcome those difficulties, computer scientist Chris Lattner set out to create a new language, Mojo, which offers the ease of use of Python, but the performance of more complex languages such as C++ or Rust." Lattner tells the site "we don't want to break Python, we want to make Python better," while software architect Doug Meil says Mojo is essentially "Python for AI... and it's going to be way faster in scale across multiple hardware platforms." Lattner teamed up with Tim Davis, whom he had met when they both worked for Google, to form Modular in January 2022. The company, where Lattner is chief executive officer and Davis chief product officer, provides support for companies working on AI and is developing Mojo.

A modern AI programming stack generally has Python on top, Lattner says, but because that is an inefficient language, it has C++ underneath to handle the implementation. The C++ then must communicate with performance accelerators or GPUs, so developers add a platform such as Compute Unified Device Architecture (CUDA) to make efficient use of those GPUs. "Mojo came from the need to unify these three different parts of the stack so that we could build a unified solution that can scale up and down," Lattner says. The result is a language with the same syntax as Python, so people used to programming in Python can adopt it with little difficulty, but which, by some measures, can run up to 35,000 times faster. For AI, Mojo is especially fast at performing the matrix multiplications used in many neural networks because it compiles the multiplication code to run directly on the GPU, bypassing CUDA...

"Increasingly, code is not being written by computer programmers. It's being written by doctors and journalists and chemists and gamers," says Jeremy Howard, an honorary professor of computer science at the University of Queensland, Australia, and a co-founder of fast.ai, a. "All data scientists write code, but very few data scientists would consider themselves professional computer programmers." Mojo attempts to fill that need by being a superset of Python. A program written in Python can be copied into Mojo and will immediately run faster, the company says. The speedup comes from a variety of factors. For instance, Mojo, like other modern languages, enables threads, small tasks that can be run simultaneously, rather than in sequence. Instead of using an interpreter to execute code as Python does, Mojo uses a compiler to turn the code into assembly language.

Mojo also gives developers the option of using static typing, which defines data elements and reduces the number of errors... "Static behavior is good because it leads to performance," Lattner says. "Static behavior is also good because it leads to more correctness and safety guarantees."
Python creator Guido van Rossum "says he is interested to watch how Mojo develops and whether it can hit the lofty goals Lattner is setting for it..." according to the article, " but he emphasizes that the language is in its early stages and, as of July 2023, Mojo had not yet been made available for download."


In June, Lattner did an hour-long interview with the TWIML AI podcast. And in 2017 Chris Lattner answered questions from Slashdot's readers.