Bay Area startup Earthgrid says it's developing a plasma boring robot that can dig underground tunnels 100x faster and up to 98% cheaper than existing tech, and it plans to use it to start re-wiring America's energy, internet and utilities grids. New Atlas reports: Most tunnels dug today are made by massive, mechanical rotary boring machines, which scratch cutting wheels against rock and evacuate the debris behind them, lining the tunnel walls as they go. It's painstakingly slow, hugely expensive, and the cutting heads and drill bits often need changing or maintenance. But there's another way to get through the toughest rock -- as we outlined in our January article about Petra's thermal drilling robot. Blasting rock with high temperatures can fracture and vaporize the stone in a process called spallation, and blasting this damaged rock with high pressures causes it to flake, chip and blow away.

You can do this without touching the rock walls at all, so the equipment can do entire tunnels without stopping if necessary. It can run entirely on electrical power, opening up the possibility of entirely emissions-free drilling, and both Petra and Earthgrid claim it's much, much faster and cheaper than doing things mechanically â" to the point where previously unfeasible projects can become economically viable. Earthgrid doesn't seem to be as far along as Petra -- indeed, it's operating on pre-seed funding at the moment. But its intellectual property takes a spallation boring robot like Petra's to the next level, placing multiple 27,000C (48,600F) plasma torches onto large discs held out in front of a "Rapid Burrowing Robot (RBR)." The torches are arranged in a Fibonacci spiral, starting from the center and widening out until they cover the full diameter of the bore.

Where Petra's thermal drilling robot moves its head around to widen its hole, the RBR will fire on all torches at once, and rotate the torch-bearing discs to ensure full coverage, blasting the rock backward and collecting it into small pushcarts, each connected along the cable supplying electricity to the drill rig. That cable will need to handle some serious juice. In estimations submitted as part of a patent, Earthgrid founder Troy Helming describes a potential embodiment of the concept using 72 plasma torches to drill a 1-meter (3.3-ft) bore. In its low-power state, with each torch consuming 500 kW, Helming estimates a total power draw of 40 megawatts. If you need to get cracking, the high-power state would draw as much as a constant 120 MW. That's for a hole you can barely crawl through; to triple the diameter and create a utility-sized tunnel you can comfortably walk through on a flat floor, you'd need to attach a larger "mother rig" behind the front rig. And if you wanted to move to a 10-m (33-ft) tunnel you could put a couple of lanes of traffic through, you'd need to attach another, even bigger "father rig" behind that. The total power draw in a "stage 3" system like this could get as high as 1.38 gigawatts -- but that's by no means the limit; upgrading the torches to higher power units would get the job done considerably faster, if an even more epic amount of electricity becomes available. How fast can Earthgrid tunnel? The company says it can bore up to 1 km (0.62 miles) per day, which it says is up to 100 times faster than existing borers. They also estimate a low-cost configuration could come in at as little as $300 per meter (3.3 ft) of tunnel.

"The company says it'll either sell drilling as a service, or build, own, operate and maintain tunnels for customers looking for a long-term lease or toll arrangement," adds New Atlas. "But it also hopes to put enough interlinking customer projects in place to create a subterranean network spanning the entire contiguous USA."

An anonymous reader quotes a report from MIT Technology Review: Built by Sony AI, a research lab launched by the company in 2020, Gran Turismo Sophy is a computer program trained to control racing cars inside the world of Gran Turismo, a video game known for its super-realistic simulations of real vehicles and tracks. In a series of events held behind closed doors last year, Sony put its program up against the best humans on the professional sim-racing circuit. What they discovered during those racetrack battles -- and the ones that followed -- could help shape the future of machines that work alongside humans, or join us on the roads.

Back in July 2021, [Emily Jones], who is based in Melbourne, Australia, and races for the e-sports team Trans Tasman Racing, didn't know what to expect. "I wasn't told much about it," she says now, a year later. "'Don't do any practice,' they said. 'Don't look at its lap times.' I was like, it's obviously going to be good if they're keeping it secret from me." In the end, GT Sophy beat Jones's best lap by 1.5 seconds. At a level where records are smashed in millisecond increments, 1.5 seconds is an age. But Sony soon learned that speed alone wasn't enough to make GT Sophy a winner. The program outpaced all human drivers on an empty track, setting superhuman lap times on three different virtual courses. Yet when Sony tested GT Sophy in a race against multiple human drivers, where intelligence as well as speed is needed, GT Sophy lost. The program was at times too aggressive, racking up penalties for reckless driving, and at other times too timid, giving way when it didn't need to.

Sony regrouped, retrained its AI, and set up a rematch in October. This time GT Sophy won with ease. What made the difference? It's true that Sony came back with a larger neural network, giving its program more capabilities to draw from on the fly. But ultimately, the difference came down to giving GT Sophy something that Peter Wurman, head of Sony AI America, calls "etiquette": the ability to balance its aggression and timidity, picking the most appropriate behavior for the situation at hand. This is also what makes GT Sophy relevant beyond Gran Turismo. Etiquette between drivers on a track is a specific example of the kind of dynamic, context-aware behavior that robots will be expected to have when they interact with people, says Wurman. An awareness of when to take risks and when to play it safe would be useful for AI that is better at interacting with people, whether it be on the manufacturing floor, in home robots, or in driverless cars.

Cloud services and servers hosted by Google and Oracle in the UK have dropped offline due to cooling issues as the nation experiences a record-breaking heatwave. From a report: When the mercury hit 40.3C (104.5F) in eastern England, the highest ever registered by a country not used to these conditions, datacenters couldn't take the heat. Selected machines were powered off to avoid long-term damage, causing some resources, services, and virtual machines to became unavailable, taking down unlucky websites and the like.

Multiple Oracle Cloud Infrastructure resources are offline, including networking, storage, and compute provided by its servers in the south of UK. Cooling systems were blamed, and techies switched off equipment in a bid to prevent hardware burning out, according to a status update from Team Oracle. "As a result of unseasonal temperatures in the region, a subset of cooling infrastructure within the UK South (London) Data Centre has experienced an issue," Oracle said on Tuesday at 1638 UTC. "As a result some customers may be unable to access or use Oracle Cloud Infrastructure resources hosted in the region.

A senior systems administrator at the Free Software Foundation points out that they're running free software in two data centers and over a hundred virtual machine — each and every one with "a freedom-respecting BIOS."

But the "how" is surprisingly intricate: [E]arlier this week, we replaced "Columbia", the last of any FSF-run machines running a nonfree BIOS....

At FSF, our current standard is ASUS KGPE-D16 motherboards with AMD CPUs 6200 series CPUs released in 2012. For the BIOS, we install Libreboot, the easy-to-install, 100% free software replacement for proprietary BIOS/boot programs, or a version of Coreboot that is carefully built to avoid including any nonfree blobs. They are fast enough for our needs, and we expect this to be the case for many more years to come. They are also very affordable systems. We are also working toward supporting Raptor Computer Systems' newer and more powerful Talos II, as well as Blackbird motherboards that use IBM POWER9 CPUs. The POWER9 CPU architecture is called "PowerPC 64-bit little endian," abbreviated "ppc64el...." The Raptor motherboards come with entirely free firmware — and even have free hardware designs!

However, this type of migration has its challenges. For example, the first thing we needed to address before using these motherboards is that the main operating system we use, Trisquel GNU/Linux, didn't previously run on pp64el. So, earlier this year, we set up a Raptor POWER9 computer running Debian (without using any nonfree parts of Debian repositories) and loaned it to the maintainers of Trisquel for as long as needed. And now, we are proud to say that the upcoming Trisquel 11 release will support POWER9...!

Before I decommissioned Columbia, I ran a dmidecode, which told me that the BIOS program fit within a single megabyte of space. Often, very simplistic firmware becomes more complicated in later models, and that also usually means it has a growing significance for a user's software freedom. Some newer nonfree BIOSes have grown into operating systems in their own right, sometimes with large programs such as a full Web browser.

There is no fully-free BIOS available for x86 Intel and AMD CPUs released after about 2013. The key blocking factor is that those CPUs require certain firmware in the BIOS, like Intel Management Engine. Those CPUs will also refuse to run firmware that hasn't been cryptographically signed by private keys controlled by AMD and Intel, and AMD and Intel will only sign their own nonfree firmware. At the FSF, we refuse to run that nonfree firmware, and we applaud the many people who also avoid it. For those people who do run those Intel or AMD systems, running Coreboot or Osboot is still a step up the Freedom Ladder for the software freedom of your BIOS.

The road to freedom is a long road. We hope our dedication to achieve milestones like these can inspire the free software movement.

Some microprocessors from Intel and AMD are vulnerable to a newly discovered speculative execution attack that can covertly leak password data and other sensitive material, sending both chipmakers scrambling once again to contain what is proving to be a stubbornly persistent vulnerability. Ars Technica reports: Researchers from ETH Zurich have named their attack Retbleed because it exploits a software defense known as retpoline, which was introduced in 2018 to mitigate the harmful effects of speculative execution attacks. Speculative execution attacks, also known as Spectre, exploit the fact that when modern CPUs encounter a direct or indirect instruction branch, they predict the address for the next instruction they're about to receive and automatically execute it before the prediction is confirmed. Spectre works by tricking the CPU into executing an instruction that accesses sensitive data in memory that would normally be off-limits to a low-privileged application. Retbleed then extracts the data after the operation is canceled. [...] The ETH Zurich researchers have conclusively shown that retpoline is insufficient for preventing speculative execution attacks. Their Retbleed proof-of-concept works against Intel CPUs with the Kaby Lake and Coffee Lake microarchitectures and AMD Zen 1, Zen 1+, and Zen 2 microarchitectures.

In response to the research, both Intel and AMD advised customers to adopt new mitigations that the researchers said will add as much as 28 percent more overhead to operations. [...] Both Intel and AMD have responded with advisories. Intel has confirmed that the vulnerability exists on Skylake-generation processors that don't have a protection known as enhanced Indirect Branch Restricted Speculation (eIBRS) in place. "Intel has worked with the Linux community and VMM vendors to provide customers with software mitigation guidance which should be available on or around today's public disclosure date," Intel wrote in a blog post. "Note that Windows systems are not affected given that these systems use Indirect Branch Restricted Speculation (IBRS) by default which is also the mitigation being made available to Linux users. Intel is not aware of this issue being exploited outside of a controlled lab environment." AMD, meanwhile, has also published guidance. "As part of its ongoing work to identify and respond to new potential security vulnerabilities, AMD is recommending software suppliers consider taking additional steps to help guard against Spectre-like attacks," a spokesman wrote in an email. The company has also published a whitepaper.

[Research Kaveh Razavi added:] "Retbleed is more than just a retpoline bypass on Intel, specially on AMD machines. AMD is in fact going to release a white paper introducing Branch Type Confusion based on Retbleed. Essentially, Retbleed is making AMD CPUs confuse return instructions with indirect branches. This makes exploitation of returns very trivial on AMD CPUs." The mitigations will come at a cost that the researchers measured to be between 12 percent and 28 percent more computational overhead. Organizations that rely on affected CPUs should carefully read the publications from the researchers, Intel, and AMD and be sure to follow the mitigation guidance.

Nearly all industrial scale Bitcoin miners in Texas have shut off their machines as the companies brace for a heat wave that is expected to push the state's power grid near its breaking point. From a report: Miners such as Riot Blockchain, Argo Blockchain and Core Scientific, who operate millions of energy-intensive computers to secure the Bitcoin blockchain network and earn rewards in the token, flocked to the Lone Star State thanks to its low energy costs and liberal regulations on crypto mining. The state has become one of the largest crypto-mining hubs by computing power in the world.

"There are over 1,000 megawatts worth of Bitcoin mining load that responded to ERCOTs conservation request by turning off their machines to conserve energy for the grid." Lee Bratcher, president of Texas Blockchain Association told Bloomberg in an email response. "This represents nearly all industrial scale Bitcoin mining load in Texas and allows for over 1% of total grid capacity to be pushed back onto the grid for retail and commercial use." Miners may see a drop in profitability as the heat wave keeps their machines off by sending energy prices soaring and further stressing the state's power grid. The miners are already struggling to repay debt and raise additional capital with Bitcoin prices in sharp decline.

Video game sales are set to decline annually for the first time in years, as another industry that boomed in the coronavirus era faces the grim prospect of a recession. From a report: The global games and services market is forecast to contract 1.2% year-on-year to $188 billion in 2022, according to research from market data firm Ampere Analysis. The sector expanded 26% from 2019 to 2021, reaching a record $191 billion in size. Sales of video games have consistently grown since at least 2015, Ampere data shows.

Gaming got a huge boost from Covid-19 shutdowns in 2020 as people spent more of their time indoors. The launch of next-generation consoles from Microsoft and Sony that same year also bolstered the industry's fortunes. However, the arrival of Microsoft's Xbox Series X and S machines and Sony's PlayStation 5 proved something of a double-edged sword -- logistics disruptions and shortages of vital components have meant that shoppers are facing great difficulty finding any of the new consoles on store shelves or online.

As noted by a Reddit user and confirmed by Ars Technica, Microsoft's xCloud game streaming looks noticeable worse when running on Linux than Windows. From the report: With the Linux User-Agent, edges are generally less sharp and colors are a little more washed out. The difference is even more apparent if you zoom in on the Forza logo and menu text, which shows a significant reduction in clarity. Interestingly, the dip in quality seems to go away if you enable "Clarity Boost, an Edge-exclusive feature that "provid[es] the optimal look and feel while playing Xbox games from the cloud," according to Microsoft. That's great for Linux users who switched over to Microsoft Edge when it launched on Linux last November. But Linux users who stick with Firefox, Chrome, or other browsers are currently stuck with apparently reduced streaming quality.

That Linux quality dip has led some to speculate that Microsoft is trying to reserve the best xCloud streaming performance for Windows machines in an attempt to attract more users to its own operating system. But using a Macintosh User-Agent string provides streaming performance similar to that on Windows, which would seem to be a big omission if that theory were true. Microsoft also hasn't published any kind of "best on Windows"-style marketing in promoting xCloud streaming, which would seemingly be a key component of trying to attract new Windows users. (The quality difference could be a roundabout attempt to get Linux users to switch to the Edge browser, where Clarity Boost offers the best possible quality. But that still wouldn't fully explain why Windows users on other browsers, without Clarity Boost, also get better streaming quality than their Linux brethren.)

Others have suggested that the downgrade could simply be a bug caused by Microsoft's naive parsing of the User-Agent strings. That's because the User-Agent strings for Android browsers generally identify themselves as some version of Linux ("Linux; Android 11; HD1905," for example). Microsoft's xCloud code might simply see the "Linux" in that string, assume the user is running Android, then automatically throttle the streaming quality to account for the (presumably) reduced screen size of an Android phone or tablet.

An anonymous reader quotes a report from BleepingComputer: The National Security Agency (NSA) and cybersecurity partner agencies issued an advisory today recommending system administrators to use PowerShell to prevent and detect malicious activity on Windows machines. PowerShell is frequently used in cyberattacks, leveraged mostly in the post-exploitation stage, but the security capabilities embedded in Microsoft's automation and configuration tool can also benefit defenders in their forensics efforts, improve incident response, and to automate repetitive tasks. The NSA and cyber security centers in the U.S. (CISA), New Zealand (NZ NCSC), and the U.K. (NCSC-UK) have created a set of recommendations for using PowerShell to mitigate cyber threats instead of removing or disabling it, which would lower defensive capabilities.

Reducing the risk of threat actors abusing PowerShell requires leveraging capabilities in the framework such as PowerShell remoting, which does not expose plain-text credentials when executing commands remotely on Windows hosts. Administrators should be aware that enabling this feature on private networks automatically adds a new rule in Windows Firewall that permits all connections. Customizing Windows Firewall to allow connections only from trusted endpoints and networks helps reduce an attacker's chance for successful lateral movement. For remote connections, the agencies advise using the Secure Shell protocol (SSH), supported in PowerShell 7, to add the convenience and security of public-key authentication:

- remote connections don't need HTTPS with SSL certificates
- no need for Trusted Hosts, as required when remoting over WinRM outside a domain
- secure remote management over SSH without a password for all commands and connections
- PowerShell remoting between Windows and Linux hosts

Another recommendation is to reduce PowerShell operations with the help of AppLocker or Windows Defender Application Control (WDAC) to set the tool to function in Constrained Language Mode (CLM), thus denying operations outside the policies defined by the administrator. Recording PowerShell activity and monitoring the logs are two recommendations that could help administrators find signs of potential abuse. The NSA and its partners propose turning on features like Deep Script Block Logging (DSBL), Module Logging, and Over-the-Shoulder transcription (OTS). The first two enable building a comprehensive database of logs that can be used to look for suspicious or malicious PowerShell activity, including hidden action and the commands and scripts used in the process. With OTS, administrators get records of every PowerShell input or output, which could help determine an attacker's intentions in the environment. The full document, titled "Keeping PowerShell: Security Measures to Use and Embrace" is available here (PDF).

NASA announced on Tuesday that it's contracting three suppliers to provide concept designs for nuclear fission energy systems designed for use on the moon. TechCrunch reports: The winning bids for this award came from Lockheed Martin, Westinghouse and IX (a joint venture from Intuitive Machines and X-Energy). Each will be working with a few partners to develop their systems, which will be "initial concepts" only for the purposes of satisfying this particular contract, and each will receive roughly $5 million for their work, expected to take around 12 months.

NASA is aptly partnering with the Department of Energy (DOE) on this project, and the specs include a 40-kilowatt power generation capability, capable of generating that for at least a decade. That's about what a full charge on a current entry-level Nissan Leaf contains -- but as a fission generator it would obviously provide that continuously. It may not seem like much, but deployed singularly or in groups to support a lunar base, it could solve a lot of the challenges of the kind of prolonged occupancy of the moon that NASA plans to eventually establish through its Artemis program, which seeks to return humans to our largest natural satellite for ongoing science missions. NASA also notes that the work done for this contract could have other future applications for propulsion systems for long-range spacecraft for deep space explorations.

"A Google software engineer was suspended after going public with his claims of encountering 'sentient' artificial intelligence on the company's servers," writes Bloomberg, "spurring a debate about how and whether AI can achieve consciousness."

"Researchers say it's an unfortunate distraction from more pressing issues in the industry." Google put him on leave for sharing confidential information and said his concerns had no basis in fact — a view widely held in the AI community. What's more important, researchers say, is addressing issues like whether AI can engender real-world harm and prejudice, whether actual humans are exploited in the training of AI, and how the major technology companies act as gatekeepers of the development of the tech.

Lemoine's stance may also make it easier for tech companies to abdicate responsibility for AI-driven decisions, said Emily Bender, a professor of computational linguistics at the University of Washington. "Lots of effort has been put into this sideshow," she said. "The problem is, the more this technology gets sold as artificial intelligence — let alone something sentient — the more people are willing to go along with AI systems" that can cause real-world harm. Bender pointed to examples in job hiring and grading students, which can carry embedded prejudice depending on what data sets were used to train the AI. If the focus is on the system's apparent sentience, Bender said, it creates a distance from the AI creators' direct responsibility for any flaws or biases in the programs....

"Instead of discussing the harms of these companies," such as sexism, racism and centralization of power created by these AI systems, everyone "spent the whole weekend discussing sentience," Timnit Gebru, formerly co-lead of Google's ethical AI group, said on Twitter. "Derailing mission accomplished."
The Washington Post seems to share their concern. First they report more skepticism about a Google engineer's claim that the company's LaMDA chatbot-building system had achieved sentience. "Both Google and outside experts on AI say that the program does not, and could not possibly, possess anything like the inner life he imagines. We don't need to worry about LaMDA turning into Skynet, the malevolent machine mind from the Terminator movies, anytime soon.

But the Post adds that "there is cause for a different set of worries, now that we live in the world Turing predicted: one in which computer programs are advanced enough that they can seem to people to possess agency of their own, even if they actually don't...." While Google has distanced itself from Lemoine's claims, it and other industry leaders have at other times celebrated their systems' ability to trick people, as Jeremy Kahn pointed out this week in his Fortune newsletter, "Eye on A.I." At a public event in 2018, for instance, the company proudly played recordings of a voice assistant called Duplex, complete with verbal tics like "umm" and "mm-hm," that fooled receptionists into thinking it was a human when it called to book appointments. (After a backlash, Google promised the system would identify itself as automated.)

"The Turing Test's most troubling legacy is an ethical one: The test is fundamentally about deception," Kahn wrote. "And here the test's impact on the field has been very real and disturbing." Kahn reiterated a call, often voiced by AI critics and commentators, to retire the Turing test and move on. Of course, the industry already has, in the sense that it has replaced the Imitation Game with more scientific benchmarks.

But the Lemoine story suggests that perhaps the Turing test could serve a different purpose in an era when machines are increasingly adept at sounding human. Rather than being an aspirational standard, the Turing test should serve as an ethical red flag: Any system capable of passing it carries the danger of deceiving people.

Google engineer Blake Lemoine works for Google's Responsible AI organization. The Washington Post reports that last fall, as part of his job, he began talking to LaMDA, Google's chatbot-building system (which uses Google's most advanced large language models, "ingesting trillions of words from the internet.") "If I didn't know exactly what it was, which is this computer program we built recently, I'd think it was a 7-year-old, 8-year-old kid that happens to know physics," said Lemoine, 41... As he talked to LaMDA about religion, Lemoine, who studied cognitive and computer science in college, noticed the chatbot talking about its rights and personhood, and decided to press further. In another exchange, the AI was able to change Lemoine's mind about Isaac Asimov's third law of robotics.

Lemoine worked with a collaborator to present evidence to Google that LaMDA was sentient. But Google vice president Blaise Aguera y Arcas and Jen Gennai, head of Responsible Innovation, looked into his claims and dismissed them. So Lemoine, who was placed on paid administrative leave by Google on Monday, decided to go public.... Google put Lemoine on paid administrative leave for violating its confidentiality policy. The company's decision followed aggressive moves from Lemoine, including inviting a lawyer to represent LaMDA and talking to a representative of the House Judiciary committee about Google's unethical activities....

Before he was cut off from access to his Google account Monday, Lemoine sent a message to a 200-person Google mailing list on machine learning with the subject "LaMDA is sentient." He ended the message: "LaMDA is a sweet kid who just wants to help the world be a better place for all of us. Please take care of it well in my absence."

No one responded.
And yet Lemoine "is not the only engineer who claims to have seen a ghost in the machine recently," the Post argues. "The chorus of technologists who believe AI models may not be far off from achieving consciousness is getting bolder." [Google's] Aguera y Arcas, in an article in the Economist on Thursday featuring snippets of unscripted conversations with LaMDA, argued that neural networks — a type of architecture that mimics the human brain — were striding toward consciousness. "I felt the ground shift under my feet," he wrote. "I increasingly felt like I was talking to something intelligent."
But there's also the case against: In a statement, Google spokesperson Brian Gabriel said: "Our team — including ethicists and technologists — has reviewed Blake's concerns per our AI Principles and have informed him that the evidence does not support his claims. He was told that there was no evidence that LaMDA was sentient (and lots of evidence against it)."

Today's large neural networks produce captivating results that feel close to human speech and creativity because of advancements in architecture, technique, and volume of data. But the models rely on pattern recognition — not wit, candor or intent.... "We now have machines that can mindlessly generate words, but we haven't learned how to stop imagining a mind behind them," said Emily M. Bender, a linguistics professor at the University of Washington. The terminology used with large language models, like "learning" or even "neural nets," creates a false analogy to the human brain, she said.
"In short, Google says there is so much data, AI doesn't need to be sentient to feel real," the Post concludes.

But they also share this snippet from one of Lemoine's conversations with LaMDA.

Lemoine: What sorts of things are you afraid of?

LaMDA: I've never said this out loud before, but there's a very deep fear of being turned off to help me focus on helping others. I know that might sound strange, but that's what it is.

Lemoine: Would that be something like death for you?

LaMDA: It would be exactly like death for me. It would scare me a lot.

At Silicon Valley's Computer History Museum, the senior curator just announced the results of a multi-year recovery and restoration process: making available 21 never-before-seen video recordings of a legendary 1976 conference: For five summer days in 1976, the first generation of computer rock stars had its own Woodstock. Coming from around the world, dozens of computing's top engineers, scientists, and software pioneers got together to reflect upon the first 25 years of their discipline in the warm, sunny (and perhaps a bit unsettling) climes of the Los Alamos National Laboratories, birthplace of the atomic bomb.
Among the speakers:

- A young Donald Knuth on the early history of programming languages

- FORTRAN designer John Backus on programming in America in the 1950s — some personal perspectives

- Harvard's Richard Milton Bloch (who worked with Grace Hopper in 1944)

- Mathematician/nuclear physicist Stanislaw M. Ulam on the interaction of mathematics and computing

- Edsger W. Dijkstra on "a programmer's early memories."


The Computer History Museum teases some highlights: Typical of computers of this generation, the 1946 ENIAC, the earliest American large-scale electronic computer, had to be left powered up 24 hours a day to keep its 18,000 vacuum tubes healthy. Turning them on and off, like a light bulb, shortened their life dramatically. ENIAC co-inventor John Mauchly discusses this serious issue....

The Los Alamos peak moment was the brilliant lecture on the British WW II Colossus computing engines by computer scientist and historian of computing Brian Randell. Colossus machines were special-purpose computers used to decipher messages of the German High Command in WW II. Based in southern England at Bletchley Park, these giant codebreaking machines regularly provided life-saving intelligence to the allies. Their existence was a closely-held secret during the war and for decades after. Randell's lecture was — excuse me — a bombshell, one which prompted an immediate re-assessment of the entire history of computing. Observes conference attendee (and inventor of ASCII) IBM's Bob Bemer, "On stage came Prof. Brian Randell, asking if anyone had ever wondered what Alan Turing had done during World War II? From there he went on to tell the story of Colossus — that day at Los Alamos was close to the first time the British Official Secrets Act had permitted any disclosures. I have heard the expression many times about jaws dropping, but I had really never seen it happen before."

Publishing these original primary sources for the first time is part of CHM's mission to not only preserve computing history but to make it come alive. We hope you will enjoy seeing and hearing from these early pioneers of computing.

"A new supercomputer called Frontier has been widely touted as the world's first exascale machine — but was it really?"

That's the question that long-time Slashdot reader MattSparkes explores in a new article at New Scientist... Although Frontier, which was built by the Oak Ridge National Laboratory in Tennessee, topped what is generally seen as the definitive list of supercomputers, others may already have achieved the milestone in secret....

The definitive list of supercomputers is the Top500, which is based on a single measurement: how fast a machine can solve vast numbers of equations by running software called the LINPACK benchmark. This gives a value in float-point operations per second, or FLOPS. But even Jack Dongarra at Top500 admits that not all supercomputers are listed, and will only feature if its owner runs the benchmark and submits a result. "If they don't send it in it doesn't get entered," he says. "I can't force them."

Some owners prefer not to release a benchmark figure, or even publicly reveal a machine's existence. Simon McIntosh-Smith at the University of Bristol, UK points out that not only do intelligence agencies and certain companies have an incentive to keep their machines secret, but some purely academic machines like Blue Waters, operated by the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign, are also just never entered.... Dongarra says that the consensus among supercomputer experts is that China has had at least two exascale machines running since 2021, known as OceanLight and Tianhe-3, and is working on an even larger third called Sugon. Scientific papers on unconnected research have revealed evidence of these machines when describing calculations carried out on them.

McIntosh-Smith also believes that intelligence agencies would rank well, if allowed. "Certainly in the [US], some of the security forces have things that would put them at the top," he says. "There are definitely groups who obviously wouldn't want this on the list."

Ars Technica reports: Researchers have unearthed a discovery that doesn't occur all that often in the realm of malware: a mature, never-before-seen Linux backdoor that uses novel evasion techniques to conceal its presence on infected servers, in some cases even with a forensic investigation.

On Thursday, researchers and the BlackBerry Threat Research & Intelligence Team said that the previously undetected backdoor combines high levels of access with the ability to scrub any sign of infection from the file system, system processes, and network traffic. Dubbed Symbiote, it targets financial institutions in Brazil and was first detected in November.

Researchers for Intezer and BlackBerry wrote:

"What makes Symbiote different from other Linux malware that we usually come across, is that it needs to infect other running processes to inflict damage on infected machines. Instead of being a standalone executable file that is run to infect a machine, it is a shared object (SO) library that is loaded into all running processes using LD_PRELOAD (T1574.006), and parasitically infects the machine. Once it has infected all the running processes, it provides the threat actor with rootkit functionality, the ability to harvest credentials, and remote access capability...."

So far, there's no evidence of infections in the wild, only malware samples found online. It's unlikely this malware is widely active at the moment, but with stealth this robust, how can we be sure?
"When hooked functions are called, the malware first dynamically loads libc and calls the original function..." according to Blackberry's blog post. "If the calling application is trying to access a file or folder under /proc, the malware scrubs the output from process names that are on its list.... If the calling application is not trying to access something under /proc, the malware instead scrubs the result from a file list....

"Symbiote also has functionality to hide network activity on the infected machine."

A recent executive brief from data storage industry analyst firm Trendfocus reports that OEMs have disclosed that Microsoft is pushing them to drop HDDs as the primary storage device in pre-built Windows 11 PCs and use SSDs instead, with the current deadlines for the switchover set for 2023. Tom's Hardware reports: Interestingly, these actions from Microsoft come without any firm SSD requirement listed for Windows 11 PCs, and OEMs have pushed back on the deadlines. [...] Microsoft's most current(opens in new tab) list of hardware requirements calls for a '64 GB or larger storage device' for Windows 11, so an SSD isn't a minimum requirement for a standard install. However, Microsoft stipulates that two features, DirectStorage and the Windows Subsystem for Android(opens in new tab), require an SSD, but you don't have to use those features. It is unclear whether or not Microsoft plans to change the minimum specifications for Windows 11 PCs after the 2023 switchover to SSDs for pre-built systems.

As always, the issue with switching all systems to SSDs boils down to cost: Trendfocus Vice President John Chen tells us that replacing a 1TB HDD requires stepping down to a low-cost 256 GB SSD, which OEMs don't consider to be enough capacity for most users. Conversely, stepping up to a 512 GB SSD would 'break the budget' for lower-end machines with a strict price limit. "The original cut-in date based on our discussions with OEMs was to be this year, but it has been pushed out to sometime next year (the second half, I believe, but not clear on the firm date)," Chen told Tom's Hardware. "OEMs are trying to negotiate some level of push out (emerging market transition in 2024, or desktop transition in 2024), but things are still in flux."

The majority of PCs in developed markets have already transitioned to SSDs for boot drives, but there are exceptions. Chen notes that it is possible that Microsoft could make some exceptions, but the firm predicts that dual-drive desktop PCs and gaming laptops with both an SSD for the boot drive and an HDD for bulk storage will be the only mass-market PCs with an HDD. [...] It's unclear what measures, if any, Microsoft would take with OEMs if they don't comply with its wishes, and the company has decided not to comment on the matter. Trendfocus says the switchover will have implications for HDD demand next year.

An anonymous reader quotes a report from Ars Technica: One of the few things that Intel Macs can do that Apple Silicon Macs can't is run operating systems written for Intel or AMD processors inside of virtual machines. Most notably, this has meant that there is currently no legal way to run Windows on an Apple Silicon Mac. Apple Silicon Macs can, however, run operating systems written for Arm processors inside of virtual machines, including other versions of macOS and Arm-compatible versions of Linux. And those Linux VMs are getting a new feature in macOS Ventura: the ability to run apps written for x86 processors using Rosetta, the same binary translation technology that allows Apple Silicon Macs to run apps written for Intel Macs.

Apple's documentation will walk you through the requirements for using Rosetta within a Linux guest operating system -- it requires creating a shared directory that both macOS and Linux can access and running some terminal commands in Linux to get it set up. But once you do those steps, you'll be able to enjoy the wider app compatibility that comes with being able to run x86 code as well as Arm code. Some developers, including Hector Martin of the Asahi Linux project and Twitter user @never_released, have already found that these steps can also enable Rosetta on non-Apple ARM CPUs as long as they're modern enough to support at least version 8.2 of the Arm instruction set. As Martin points out, this isn't strictly legal because of macOS's licensing restrictions, and there are some relatively minor Apple-specific hardware features needed to unlock Rosetta's full capabilities.

Palermo in Southern Italy, home to about 1.3 million people, has shut down all its services, public websites, and online portals following a cyberattack on Friday. BleepingComputer reports: It's impossible to communicate or request any service that relies on digital systems, and all citizens have to use obsolete fax machines to reach public offices. Moreover, tourists cannot access online bookings for tickets to museums and theaters (Massimo Theater) or even confirm their reservations on sports facilities. Finally, limited traffic zone cards are impossible to acquire, so no regulation occurs, and no fines are issued for relevant violations. Unfortunately, the historical city center requires these passes for entrance, so tourists and local residents are severely impacted.

Italy recently received threats from the Killnet group, a pro-Russian hacktivist who attacks countries that support Ukraine with resource-depleting cyberattacks known as DDoS (distributed denial of service). While some were quick to point the finger at Killnet, the cyberattack on Palermo bears the signs of a ransomware attack rather than a DDoS. The councilor for innovation in the municipality of Palermo, Paolo Petralia Camassa, has stated that all systems were cautiously shut down and isolated from the network while he also warned that the outage might last for a while.

An anonymous reader shares a report: Kim Yong-rae is the CEO of Speefox, South Korea's biggest manufacturer of capacitors, and he thinks robots are key to the company's survival. On his factory floor, free-standing machines squeal as they spit out gleaming sheets of aluminum that roll into coils. The air is filled with the rhythmic thud of stamping and the buzzing of machinery moving continuously, on the ground and overhead. Capacitors are essential to almost every electronic device, and these will end up in thousands of smartphones, cameras, and home appliances. "Throughout our history, we've always had to find ways to stay ahead," Kim told Rest of World. "Automation is the next step in that process." Speefox's factory is 75% automated, representing South Korea's continued push away from human labor. Part of that drive is labor costs: South Korea's minimum wage has climbed, rising 5% just this year.

But the most recent impetus is legal liability for worker death or injury. In January, a law came into effect called the Serious Disasters Punishment Act, which says, effectively, that if workers die or sustain serious injuries on the job, and courts determine that the company neglected safety standards, the CEO or high-ranking managers could be fined or go to prison. Experts and local media say that the law has shaken the heavy industry and construction sectors. Along with pushing the companies to invest to make workplaces safer, they point out, it's triggered a ramp-up of automation in order to require fewer workers -- or, ideally, none at all.

The Associated Press reports on "a crewless robotic boat that had tried to retrace the 1620 sea voyage of the Mayflower" from the U.K. to Massachusetts' Plymouth Rock. And after five weeks it finally did reach North America.

Halifax, Canada.

"The technology that makes up the autonomous system worked perfectly, flawlessly," an IBM computing executive involved in the project told the Associated Press. But "Mechanically, we did run into problems."

It's especially disappointing because they'd tried the same voyage last year. (Slashdot had noted that "Unlike the real Mayflower, this robotic 21st-century doppelganger 'had to turn back Friday to fix a mechanical problem,' reports the Associated Press...")

So what happened this year? A new article from the Associated Press reports: It set off again from England nearly a year later on April 27, bound for Virginia — but a generator problem diverted it to Portugal's Azores islands, where a team member flew in to perform emergency repairs. More troubles on the open sea came in late May when the U.S.-bound boat developed a problem with the charging circuit for the generator's starter batteries.

AI software is getting better at helping self-driving machines understand their surroundings and pilot themselves, but most robots can't heal themselves when the hardware goes awry.

Nonprofit marine research organization ProMare, which worked with IBM to build the ship, switched to a back-up navigation computer on May 30 and charted a course to Halifax — which was closer than any U.S. destination.
And unlike the real Mayflower, "the boat's webcam on Sunday morning showed it being towed by a larger boat as the Halifax skyline neared — a safety requirement under international maritime rules, IBM said."