A big new noise in the world of hearing aids is a company called Fortell which improves sound recognition in noisy spaces. CC-licensed photo by Mark Fonseca Rendeiro on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Clearly. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Researchers find what makes AI chatbots politically persuasive • Ars Technica

Jacek Krywko:

»

To see if conversational large language models can really sway political views of the public, scientists at the UK AI Security Institute, MIT, Stanford, Carnegie Mellon, and many other institutions performed by far the largest study on AI persuasiveness to date, involving nearly 80,000 participants in the UK. It turned out political AI chatbots fell far short of superhuman persuasiveness, but the study raises some more nuanced issues about our interactions with AI.

The public debate about the impact AI has on politics has largely revolved around notions drawn from dystopian sci-fi. Large language models have access to essentially every fact and story ever published about any issue or candidate. They have processed information from books on psychology, negotiations, and human manipulation. They can rely on absurdly high computing power in huge data centers worldwide. On top of that, they can often access tons of personal information about individual users thanks to hundreds upon hundreds of online interactions at their disposal.

Talking to a powerful AI system is basically interacting with an intelligence that knows everything about everything, as well as almost everything about you. When viewed this way, LLMs can indeed appear kind of scary. The goal of this new gargantuan AI persuasiveness study was to break such scary visions down into their constituent pieces and see if they actually hold water.

The team examined 19 LLMs, including the most powerful ones like three different versions of ChatGPT and xAI’s Grok-3 beta, along with a range of smaller, open source models. The AIs were asked to advocate for or against specific stances on 707 political issues selected by the team. The advocacy was done by engaging in short conversations with paid participants enlisted through a crowdsourcing platform. Each participant had to rate their agreement with a specific stance on an assigned political issue on a scale from 1 to 100 both before and after talking to the AI.

…Overall, AI models changed the participants’ agreement ratings by 9.4% on average compared to the control group. The best performing mainstream AI model was Chat GPT 4o, which scored nearly 12% followed by GPT 4.5 with 10.51%, and Grok-3 with 9.05%. For context, static political ads like written manifestos had a persuasion effect of roughly 6.1%. The conversational AIs were roughly 40–50% more convincing than these ads, but that’s hardly “superhuman.”

«

No, but it’s suprahuman, and this is only an early incarnation.
unique link to this extract

Aircraft crashed in Gloucestershire after 3D-printed part collapsed • BBC News

Maisie Lillywhite:

»

A plane crashed after a 3D-printed part softened and collapsed, causing its engine to lose power, a report has found.

The Cozy Mk IV light aircraft was destroyed after its plastic air induction elbow, bought at an air show in North America, collapsed.

The aircraft crashed into a landing aid system at Gloucestershire Airport in Staverton on 18 March at 13:04 GMT, after its engine lost power. The sole occupant was taken to hospital with minor injuries.

The Air Accidents Investigation Branch (AAIB) said in a report that the induction elbow was made of “inappropriate material” and safety actions will be taken in future regarding 3D printed parts.

Following an “uneventful local flight”, the AAIB report said the pilot advanced the throttle on the final approach to the runway, and realised the engine had suffered a complete loss of power.

“He managed to fly over a road and a line of bushes on the airfield boundary, but landed short and struck the instrument landing system before coming to rest at the side of the structure,” the report read.

It was revealed the part had been installed during a modification to the fuel system and collapsed due to its 3D-printed plastic material softening when exposed to heat from the engine.

«

Very unintended consequences.
unique link to this extract

A responsibility to the industry • LMNT

Louie Mantia, back in July, a month after the “Liquid Glass” design had been unveiled and developers were struggling to rewrite apps to look right with it:

»

Firstly, I maintain that it makes absolutely no sense that Alan Dye has the power he has, because he simply has no taste. But what’s worse is that he wields that power so clumsily, so carelessly. And because it goes unchallenged, unchecked by someone higher than him, the entire industry suffers the consequences.

If that sounds too dramatic, maybe the rest of this post won’t be for you.

One reason that developers struggle with implementing Liquid Glass is Apple’s own evolving implementation of it. From just the first few beta releases, enough of it has changed to make it difficult for some developers to understand what exactly Apple’s vision of it is. It also communicates a level of uncertainty about things that haven’t yet been addressed about its various concessions with long-standing UI elements in macOS especially. I do not want to list them all.

When Apple themselves have not yet reasonably prescribed what standard UI elements look like in this new design system, how can any developer responsibly implement them in good conscience? Isn’t there something about this that just reeks? Adopting a standard control means it can change without your involvement. This has always been true to some extent, but the stink of it keeps getting worse as trust in the company’s vision erodes over time, right?

Another reason that the industry is showing signs of reluctance is because Alan Dye did not prove he understood the platform, any platform, before he assumed the role of its lead designer. He’s not just a newcomer to these platforms, but to software design as a whole.

«

So much love for Meta’s new design guru. So much.
unique link to this extract

Bad Dye Job • Daring Fireball

John Gruber on Alan Dye’s departure from the top design job at Apple:

»

Dye’s replacement at Apple is longtime Apple designer Stephen Lemay. I’ve never met Lemay (or at least can’t recall meeting him), and prior to today never heard much about him. But that’s typical for Apple employees. Part of the job working for Apple is remaining under the radar and out of the public eye. What I’ve learned today is that Lemay, very much unlike Dye, is a career interface/interaction designer. Sources I’ve spoken to who’ve worked with Lemay at Apple speak highly of him, particularly his attention to detail and craftsmanship. Those things have been sorely lacking in the Dye era. Not everyone loves everything Lemay has worked on, but nobody bats 1.000 and designers love to critique each other’s work. I’ve chatted with people with criticisms of specific things Lemay has worked on or led at Apple (e.g. aspects of iPadOS multitasking that struck many of us as deliberately limiting, rather than empowering), but everyone I’ve spoken to is happy — if not downright giddy — at the news that Lemay is replacing Dye. Lemay is well-liked personally and deeply respected talent-wise. Said one source, in a position to know the choices, “I don’t think there was a better choice than Lemay.”

The sentiment within the ranks at Apple is that today’s news is almost too good to be true. People had given up hope that Dye would ever get squeezed out, and no one expected that he’d just up and leave on his own. (If you care about design, there’s nowhere to go but down after leaving Apple. What people overlooked is the obvious: Alan Dye doesn’t actually care about design.)

«

Agree that. I’ve updated precisely one of my Apple devices to v26, and that’s a old “sacrifice” Mac which I used to see how it looked.

Points too to Gruber for the headline, which is so good I’ve made an exception to the normal style here and left in the capitalisations.
unique link to this extract

Telehealth weight-loss provider NextMed hit with FTC crackdown over deceptive pricing and fake reviews • MSN

Maryann Pugh:

»

The operators of NextMed, a telehealth weight-loss provider, have agreed to pay $150,000 and overhaul their business practices to settle Federal Trade Commission (FTC) allegations that they misled consumers with deceptive advertising, fake reviews, and hidden costs tied to their membership programs.

The FTC’s complaint accuses Southern Health Solutions, Inc., doing business as NextMed, along with founders Robert Epstein and CEO Frank Leonardo III, of violating federal consumer protection laws through a range of deceptive tactics. The company marketed access to medical providers for popular weight-loss drugs such as Wegovy and Ozempic, offering memberships starting at $138 or $188 per month. However, the FTC contends those advertised prices did not include key costs like the medications themselves, required lab work, or medical consultations.

The agency further alleges that customers were locked into one-year contracts with undisclosed early termination fees and faced widespread difficulty when attempting to cancel or obtain refunds due to understaffed customer service.

“Consumers who signed up for NextMed’s programs faced significant unexpected costs and the company’s customer service failures prevented consumers from cancelling or getting a refund,” said Christopher Mufarrige, Director of the FTC’s Bureau of Consumer Protection.

«

There was all sorts of scammy stuff here: an ad with a thin actress who hadn’t used it, fake reviews via VPNs, before/afters solicited on Craiglist, and didn’t tell people the medication wasn’t included in the subscription. The owners have to pay $150,000 back to scammed customers.

Where there’s a growth market, there’s a scam.
unique link to this extract

PaperDebugger: a plugin-based multi-agent system for in-editor academic writing, review, and editing • ArXiv

Junyi Hou et al at the National University of Singapore:

»

Large language models are increasingly embedded into academic writing workflows, yet existing assistants remain external to the editor, preventing deep interaction with document state, structure, and revision history. This separation makes it impossible to support agentic, context-aware operations directly within LaTeX editors such as Overleaf.

We present PaperDebugger, an in-editor, multi-agent, and plugin-based academic writing assistant that brings LLM-driven reasoning directly into the writing environment. Enabling such in-editor interaction is technically non-trivial: it requires reliable bidirectional synchronization with the editor, fine-grained version control and patching, secure state management, multi-agent scheduling, and extensible communication with external tools.

PaperDebugger addresses these challenges through a Chrome-approved extension, a Kubernetes-native orchestration layer, and a Model Context Protocol (MCP) toolchain that integrates literature search, reference lookup, document scoring, and revision pipelines. Our demo showcases a fully integrated workflow, including localized edits, structured reviews, parallel agent execution, and diff-based updates, encapsulated within a minimal-intrusion user interface (UI).

«

There might not be a lot of readers who will be able to use this, but for the ones who can, it’s going to make a big difference.
unique link to this extract

I tested five AI browsers and lost my mind in the process • The Verge

Victoria Song:

»

Right now, AI browsers come in two main flavors. There are your regular browsers that have an AI assistant stapled on in a collapsible window, such as Chrome with its Gemini features, or Edge with Copilot Mode. Then there are more specialized AI browsers, most notably ChatGPT Atlas, Perplexity’s Comet, and The Browser Company’s Dia.

This second category often supplants your search bar with AI and sometimes includes an “agentic mode,” in which the AI can complete more complex, browser-related tasks for you. Theoretically, that includes helping you book reservations or add items to a shopping cart.

For testing, I decided on a few ground rules. I kept it to five browsers: Chrome, Edge, Atlas, Comet, and Dia. There are more available, but this felt like a representative mix of both AI browser categories from a variety of players in the field. I focused on desktop apps, and tried to make settings as uniform as possible: I generally instructed the AI browsers to keep answers snappy, shared my location information where possible, enabled memory settings, and described myself as a “tech journalist specializing in health and wearable tech.”

I also approached testing from a variety of AI skill levels. What would results look like if I was a complete AI newbie versus someone more adept at prompting? Lastly, if I tried one task in a browser, I gave it a go in all the browsers, down to the same exact prompt.

Ultimately, my question was not which AI browser you should use, but whether any of them are worth your time and energy. This was a journey to see whether any of them live up to the hype.

The short answer: they don’t.

«

Turns out you have to think hard, not about your search term, but about your prompt. New stuff, same old junk. Will it magically get better? Well, have the search terms you use got shorter or longer over time, and have you had to think more or less about what you’re going to type? That’s probably what’s going to happen here as websites figure out how to fool “agentic” browsers into paying attention to them.
unique link to this extract

IRS agents will be required to watch OnlyFans to determine if content fits ‘no tax on tips’ criteria • The Independent

Owen Scott:

»

IRS agents will be required to watch pornographic content on OnlyFans to determine if the content meets the “no tax on tips” law included in Donald Trump’s Big, Beautiful Bill.

The president’s controversial tax and spending policies were passed on July 4, 2025, with the slashing of taxes on tips being designed to incentivize people to earn more tips at work.

However, the new law included a caveat. Pornographic creators and actors, including OnlyFans influencers, were not entitled to have taxes waived on their work.

Some campaigners have argued that the wording is too vague, with one accountant telling The New York Times that the line of what is considered pornography is unclear.

“Where’s the line?” said Katherine Studley, who works with several OnlyFans creators. “Just because you’re on OnlyFans, that doesn’t necessarily mean it’s pornographic. You could have a cooking channel or a yoga channel.”

Defining what pornography actually is has often proven difficult for lawmakers, meaning that it usually has to be judged on a case-by-case basis. When the First Amendment is used to defend pornography in court, lawmakers have to view the material in question to make a judgment.

That means taxpayers who report tips from OnlyFans will likely need to have their content viewed by an IRS agent.

«

(Thanks Joe S for the link.)
unique link to this extract

Want a Fortell hearing aid? Well, who do you know? • WIRED

Steven Levy:

»

A secret is percolating at dinner parties, salons, and cocktail gatherings among the august New York City elite. It’s whispered in the circles of financial masters of the universe, Hollywood stars, and owners of sports teams. Have you heard about Fortell?

Many haven’t—or if they did hear, they might not have made out the words through noisy cross-conversations. Once they do know—particularly if they’re boomers—they want it desperately. Fortell is a hearing aid, one that claims to use AI to provide a dramatically superior aural experience. The chosen few included in its beta test claim that it seems to top the performance of high-end devices they’d been unhappily using.

These testers have made pilgrimages to Fortell’s headquarters on the fifth floor of a WeWork facility in New York City’s trendy SoHo neighborhood, where they were fitted for the hearing aids—which from the outside look pretty much like standard, over-the-ear, teardrop-shaped devices. But the big moment comes when a Fortell staffer takes them down to street level. There, among street clatter, honking cabs, and delivery trucks backing up to luxury stores, they are asked to conduct a conversation with a Fortell worker. Two other employees stand behind them, adding their own loud discourse to the urban cacophony.

Despite the din, the testers clearly make out what the person in front of them is saying. The clouds lift. Angels croon. “This was so incredible that I burst into tears,” says Ashley Tudor, one of the seemingly few beta testers who isn’t famous or powerful (though she is married to a venture capitalist).

Among the age-related-hearing-loss set, getting into the Fortell beta test has become a weird status symbol, the aural-prosthetics version of a limited-edition Birkin bag. “This product has become a major flex for the post-70 set,” says one investor. When entertainment lawyer Allen Grubman got his—he’s buddies with an investor—he began getting calls from “very substantial” people. “They said, ‘Allen, we hear that you have these new great hearing aids,’” he says of these callers, who all wanted in. Those who finagled their way into the program include multiple Forbes 400 billionaires, a chart-topping musician, the producer of a beloved TV series, and Hollywood A-listers, both old and not-so-old. KKR private equity co-executive chair Henry Kravis raves about his Fortells, as does performer and beta tester Steve Martin.

«

As the article explains, the problem for hearing as you age is in focussing on the sounds you want to hear and ignoring the ones you don’t. The solution isn’t just making everything louder.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

Head designer Alan Dye has left Apple to join Meta. His resignation was written in grey ink on light grey paper. CC-licensed photo by Kris Arnold on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Indecipherable. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

Rare win for renewable energy: Trump Administration funds geothermal network expansion • Inside Climate News

Phil McKenna:

»

The U.S. Department of Energy has approved an $8.6m grant that will allow the nation’s first utility-led geothermal heating and cooling network to double in size.

Gas and electric utility Eversource Energy completed the first phase of its geothermal network in Framingham, Massachusetts, in 2024. Eversource is a co-recipient of the award along with the city of Framingham and HEET, a Boston-based nonprofit that focuses on geothermal energy and is the lead recipient of the funding.

Geothermal networks are widely considered among the most energy-efficient ways to heat and cool buildings. The federal money will allow Eversource to add approximately 140 new customers to the Framingham network and fund research to monitor the system’s performance.

The federal funding was first announced in December 2024 under the Biden administration. However, the contract between HEET and the Department of Energy was not finalized until Sept. 30 and was just announced Wednesday. The agreement, which allows construction to move forward, comes as the Trump administration is clawing back billions of dollars in clean energy funding, including hundreds of millions of dollars in Massachusetts. 

“This award is an opportunity and a responsibility to clearly demonstrate and quantify the growth potential of geothermal network technology,” Zeyneb Magavi, HEET’s executive director, said in a written statement.

The existing system provides heating and cooling to approximately 140 residential and commercial customers in the western suburb of Boston. The network taps low-temperature thermal energy from dozens of boreholes drilled several hundred feet below ground, where temperatures remain steady at 55ºF.

«

This is very small potatoes in the grand scheme of things. But as a New Yorker article points out, there’s enormous potential for geothermal energy (Iceland practically runs on it), and the techniques that fracking has refined make tapping geothermal power easier than ever. So maybe it could be the little acorn that grows into the big oak.
unique link to this extract

Apple design executive Alan Dye poached by Meta in major coup • Bloomberg via MSN

Mark Gurman:

»

Meta Platforms Inc. has poached Apple Inc.’s most prominent design executive in a major coup that underscores a push by the social networking giant into AI-equipped consumer devices. 

The company is hiring Alan Dye, who has served as the head of Apple’s user interface design team since 2015, according to people with knowledge of the matter. Apple is replacing Dye with longtime designer Stephen Lemay, according to the people, who asked not to be identified because the personnel changes haven’t been announced.

Apple confirmed the move in a statement provided to Bloomberg News. 

“Steve Lemay has played a key role in the design of every major Apple interface since 1999,” Chief Executive Officer Tim Cook said in the statement. “He has always set an extraordinarily high bar for excellence and embodies Apple’s culture of collaboration and creativity.”

The move represents a seismic shift in Silicon Valley and shows that Meta is committed to becoming a name-brand maker of hardware devices. For Apple, the departure extends an exodus of talent suffered by the design team since the exit of visionary executive Jony Ive in 2019.

Dye had taken on a more significant role at Apple after Ive left, helping define how the company’s latest operating systems, apps and devices look and feel. The executive informed Apple this week that he’d decided to leave, though top management had already been bracing for his departure, the people said. 

With the Dye hire, Meta is creating a new design studio and putting him in charge of design for hardware, software and AI integration for its interfaces.

He will be reporting to Chief Technology Officer Andrew Bosworth, who oversees Reality Labs. That group is tasked with developing wearable devices, such as smart glasses and virtual reality headsets.

«

This is absolutely wonderful news. You are not going to find anyone who appreciates Apple design who is going to be sorry about this. Dye’s approach to Apple’s software design has had a sort of indifference to users’ needs in favour of stuff that wows people at demos but doesn’t stand up to being used. This week’s Accidental Tech Podcast, for which this is timely, will treat this as an early Christmas present.
unique link to this extract

Apple replaces head of AI with executive poached from Microsoft • Financial Times

Rafe Rosner-Uddin:

»

Apple’s vice-president of artificial intelligence will be replaced by a top Microsoft executive as the iPhone maker struggles to recover from a slow start in the race to harness advanced AI.

John Giannandrea, senior vice-president for machine learning and AI strategy, will step down and serve as an adviser to Apple until retiring in the spring, the company said on Monday.

He will be replaced by former Microsoft executive Amar Subramanya, who leaves a job as a corporate vice-president for Microsoft six months after jumping ship from Google, where he worked on the Gemini chatbot.

Tim Cook, Apple’s chief executive, said: “AI has long been central to Apple’s strategy, and we are pleased to welcome Amar . . . and to bring his extraordinary AI expertise to Apple.”

The leadership change in Apple’s AI division comes as Giannandrea faced mounting criticism for a faltering approach to deploying generative AI, the technology that underpins competitors’ products such as Gemini and ChatGPT.

Apple has been slow to catch up with the technology and roll out AI tools in recent years, as the popularity of chatbots has grown rapidly.

«

The roundabout continues. Microsoft-Google-Apple-Meta-OpenAI and everywhere, like a game of very well paid musical chairs.
unique link to this extract

John Mayer spotted playing Neural DSP Quad Cortex at Coachella • Guitar World

Matt Owen:

»

Tube amp loyalist and gear aficionado John Mayer has once again demonstrated his increasing affection for digital guitar gear by turning up to play at Coachella with an amp modeler – but it wasn’t one he has ever been spotted playing before.

On Sunday (April 13), the electric guitar giant joined German DJ and producer Zedd to perform two songs: Automatic Yes – a Zedd track Mayer features on – and Mayer’s own song, New Light, which was released in 2018.

For the short guest performance, though, the PRS signature artist opted against wheeling out his entire rig (it would have been entirely impractical to do so) and instead played through, for the first time on stage, a Neural DSP Quad Cortex.

As per the John Mayer’s Gear Instagram page, for the two-song cameo Mayer partnered his Faded Black Tee Satin Silver Sky with the acclaimed QC, which had been attached to what looked to be D’Addario’s XPND pedalboard.

Those familiar with Mayer’s guitar gear will be aware this isn’t the first time the guitarist has played through an amp modeler, either on stage or in the studio. At 2019’s Coachella, he performed with Khalid through a Fractal Axe-Fx III, and two years later he used a Fractal to record parts of Sob Rock.

Not only that, Mayer also reportedly owns a Kemper Profiler for casual use, and used a Fractal while supporting Ed Sheeran in 2023.

It is, however, the first time Mayer has opted to use the Neural DSP Quad Cortex in this capacity. Whether that means he’s decided to ditch his trusty Fractals altogether, or whether this was a one-time-only thing, it remains to be seen.

It also remains to be seen whether this is the start of a formal partnership with Neural DSP. Did the team use their T.I.N.A robot to model Mayer’s hugely sought-after and beloved Dumble-loaded rig? The tones, by all accounts (footage is yet to surface online) were on point during the set, so it’s not entirely out of the question.

«

OK, so to a lot of people (including me) this is incomprehensible – the feeling captured by the classic tweet “I’m 50. All celebrity news looks like this: ‘Curtains for Zoosha? K-smog and Batboy caught flipping a grunt’“.

Anyhow, the way that physical amplifiers are being replaced by digital versions is well explained in an Ars Technica article, which in effect riffs on the Mayer event. Software is eating the world. And its amplifiers.
unique link to this extract

Fraudulent gambling network may actually be something more nefarious • Ars Technica

Dan Goodin:

»

Researchers have previously tracked smaller pieces of the enormous infrastructure. Last month, security firm Sucuri reported that the operation seeks out and compromises poorly configured websites running the WordPress CMS. Imperva in January said the attackers also scan for and exploit web apps built with the PHP programming language that have existing webshells or vulnerabilities. Once the weaknesses are exploited, the attackers install a GSocket, a backdoor that the attackers use to compromise servers and host gambling web content on them.

All of the gambling sites target Indonesian-speaking visitors. Because Indonesian law prohibits gambling, many people in that country are drawn to illicit services. Most of the 236,433 attacker-owned domains hosting the gambling sites are hosted on Cloudflare. Most of the 1,481 hijacked subdomains were hosted on Amazon Web Services, Azure, and GitHub.

On Wednesday, researchers from security firm Malanta said those details are only the most visible signs of a malicious network that’s actually much bigger and more complex than previously known. Far from being solely a financially motivated operation, the firm said, the network likely serves nation-state hackers targeting a wide range of organizations, including those in manufacturing, transport, healthcare, government, and education.

The basis for the speculation is the tremendous amount of time and resources that have gone into creating and maintaining the infrastructure over 14 years. The resources include 328,000 separate domains, which comprise 236,000 addresses that the attackers bought and 90,000 that they commandeered by compromising legitimate websites. It’s also made up of nearly 1,500 hijacked subdomains from legitimate organizations. Malanta estimates that such infrastructure costs anywhere from $725,000 to $17m per year to fund.

…“This combination—longevity, scale, cost, and sophistication—goes well beyond a typical ‘quickhit’ gambling scam or financially motivated crew,” Malanta said. “That’s why we classify it as an APT and describe it as state sponsored-level, while being careful not to assert that we have direct evidence tying it to a specific government entity.”

The focus on compromising government agencies in the US and Europe and a wide swath of industries is another reason for the assessment.

«

unique link to this extract

This AI bubble is more memory than dot-com • Culpium

Tim Culpan:

»

Since the AI sector at the model and token-making level is just like memory, the industry shakeout is likely to play out in the same way. DRAM used to be a big deal. For a while it was Intel’s bread and butter. These chips are fundamental to any computing system because they temporary hold data — often for milliseconds — which a processor uses to make calculations. In the early PC era, DRAM supply was the bottleneck for computer sales.

By the early 2000s, there were around a dozen memory-chip makers across South Korea, Japan, Taiwan and the US. Most were backed by industrial conglomerates. Hynix, for example, was spun off from the Hyundai group after the chaebol’s boss realized electronics were an increasingly important part of cars. Nanya Technology was born out of Taiwan’s massive Formosa Plastics Group. But heavy competition, unstable earnings, and an unwillingness by corporate parents and banks to keep funding them lead to a wave of consolidations and shutdowns by the early 2010s. Today, the DRAM sector is dominated by just three companies: Samsung, SK Hynix — both from South Korea — and Boise, Idaho-based Micron.

What kept the leaders atop the market was both an unrelenting pace of capacity expansion, and continued technological development using the latest equipment. Both are crucial to maintaining price competitiveness on a per-bit basis. Assuming the product is largely reliable, price is the factor which truly differentiates.

Today’s AI leaders can be sorted into two categories: stand alone, and conglomerate-backed. OpenAI and Anthropic are startups that stand alone. A recent series of circular deals muddies the water a little: Nvidia is taking a stake in Open AI2, while the startup has a warrant for shares in Nvidia rival AMD. And Microsoft’s relationship with OpenAI makes the software giant as much a sugar daddy as a conglomerate parent: it’ll last as long as Microsoft’s AI friend provides benefits.3 These companies are kind of like Micron. Although it found a wealthy industrialist to back it, Micron never benefited from having a conglomerate parent.

«

Culpan used to be Bloomberg’s technology correspondent based in Taiwan, and has long experience writing about the sector. The longer part of this article looks at whether the AI bubble (come on, it is) resembles the dot-com bubble, or the RAM bubble.
unique link to this extract

Crucial is shutting down — because Micron wants to sell its RAM and SSDs to AI companies instead • The Verge

Emma Roth:

»

Micron is retiring the Crucial brand, marking the end of its line of budget-friendly solid-state drives (SSDs) and RAM kits, as reported earlier by VideoCardz. In an announcement on Wednesday, Micron says winding down its consumer-focused business will “improve supply and support for our larger, strategic customers in faster-growing segments” — a.k.a. AI companies.

The brand’s shutdown is a huge blow for PC builders and hobbyists, who are already dealing with skyrocketing RAM prices linked to a surge in demand from AI companies. OpenAI, for example, struck a deal with SK Hynix and Samsung to make up to 900,000 DRAM per month for its Stargate project.

Now, there’s going to be one less brand selling consumer-focused memory for PCs, potentially intensifying the global memory shortage. Soaring demand for RAM is already impacting pricing at CyberPowerPC, Framework, and Raspberry Pi, while HP has even hinted at raising the prices of its devices or equipping them with less memory.

«

So RAM goes from being a hard-to-obtain specialist product in the 1970s to a ubiquitous, cheap product in the 2010s to a hard-to-obtain specialist product again.
unique link to this extract

WordPress’s vibe-coding experiment, Telex, has already been put to real-world use • TechCrunch

Sarah Perez:

»

WordPress’s experimental AI development tool, Telex, has already been put to real-world use, only months after its September debut. At the company’s annual “State of the Word” event on Tuesday in San Francisco, WordPress Project Cofounder and Automattic CEO Matt Mullenweg shared several examples where Telex had been used within a working WordPress shop to do things like create price comparisons, price calculators, and pull in real-time business hours plus a map link to a retail store, among examples.

Telex, which Mullenweg previously described as a “v0 or Lovable, but specifically for WordPress,” is essentially the publishing platform’s attempt to build its own vibe-coding tool for the AI era. The software allows developers to generate Gutenberg blocks — the modular bits of text, images, columns, and more — that make up a WordPress website.

While the software is still labeled as an experiment, Mullenweg was able to demonstrate several real-world examples that had been built by community creator Nick Hamze.

In the first example, Mullenweg showed off a pricing comparison tool built with Telex, noting that these sorts of rich, interactive web elements were something that a developer used to have to custom-build, but could now be created in a few seconds.

«

This is getting closer and closer to the ordinary person being able to use it.
unique link to this extract

The fall of a prolific science journal exposes the billion-dollar profits of scientific publishing • EL PAÍS English

Manuel Ansede:

»

In the autumn of 2020, with humanity terrified by the deadly second wave of the coronavirus, a scientific journal published a study with a solution: jade amulets from traditional Chinese medicine could prevent COVID-19. The proposal was outlandish, but the editor-in-chief of the weekly, Spanish chemist Damià Barceló, defended its quality controls. That journal, Science of the Total Environment — one of the 15 that publishes the most studies worldwide — has just been expelled from the group of reputable publications by one of the leading evaluation companies, after dozens of irregular articles were discovered. The scandal exposes the windfall profits of scientific publishers, who in recent years have amassed billions of dollars in earnings from public funds earmarked for science.

Damià Barceló, 71, took over as editor of the journal in 2012. In just two years, he doubled the number of studies published. In a decade, he increased the number tenfold, with the journal reaching nearly 10,000 articles annually. As the number of articles increased, the quality declined, because there was a perverse incentive to accept mediocre work: to publish research open access in the journal, a scientist has to pay $4,150 plus taxes.

Emilio Delgado, professor of documentation at the University of Granada in Spain, explains it this way: “It’s clearly an open-door journal that takes almost anything. It’s what I call a mega-journal, that is, a mega-business.”

«

“Paper mills” are just there to make noise. As ever, Sturgeon’s Law applies: 90% of everything is crap. This was the crap.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

The Environment Agency was far too slow to respond to thousands of tonnes of illegally dumped waste near the river Cherwell. CC-licensed photo by Howard Stanbury on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Rubbish. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

OpenAI CEO declares “code red” as Gemini gains 200 million users in three months • Ars Technica

Benj Edwards:

»

The shoe is most certainly on the other foot. On Monday, OpenAI CEO Sam Altman reportedly declared a “code red” at the company to improve ChatGPT, delaying advertising plans and other products in the process,  The Information reported based on a leaked internal memo. The move follows Google’s release of its Gemini 3 model last month, which has outperformed ChatGPT on some industry benchmark tests and sparked high-profile praise on social media.

In the memo, Altman wrote, “We are at a critical time for ChatGPT.” The company will push back work on advertising integration, AI agents for health and shopping, and a personal assistant feature called Pulse. Altman encouraged temporary team transfers and established daily calls for employees responsible for enhancing the chatbot.

The directive creates an odd symmetry with events from December 2022, when Google management declared its own “code red” internal emergency after ChatGPT launched and rapidly gained in popularity. At the time, Google CEO Sundar Pichai reassigned teams across the company to develop AI prototypes and products to compete with OpenAI’s chatbot. Now, three years later, the AI industry is in a very different place.

Google released Gemini 3 in mid-November, and the model quickly topped the LMArena leaderboard, a crowdsourced vibemarking site that allows users to compare two AI models and select the one with outputs that please them most. The launch has been accompanied by measured praise from some and bombastic hype from others. Salesforce CEO Marc Benioff wrote Sunday on X that he was switching to Gemini 3 after using ChatGPT daily for three years. “I’m not going back,” Benioff wrote. “The leap is insane.”

…Not everyone views OpenAI’s “code red” as a genuine alarm. Reuters columnist Robert Cyran wrote on Tuesday that OpenAI’s announcement added “to the impression that OpenAI is trying to do too much at once with technology that still requires a great deal of development and funding.” On the same day Altman’s memo circulated, OpenAI announced an ownership stake in a Thrive Capital venture and a collaboration with Accenture. “The only thing bigger than the company’s attention deficit is its appetite for capital,” Cyran wrote.

«

Shades of Facebook when Google announced Google+?
unique link to this extract

Rubbish mountain • The Value of Nothing

Martin Robbins goes to Oxford to look at the gigantic amount of waste dumped there by criminals some time earlier this year:

»

Eyewitness reports speak of convoys of lorries turning up at the site, delivering the rubbish to waiting excavators that moulded it into the extraordinary monument we see today. When I visited the area last weekend I estimated the pile to be around 170m long (roughly the length of the Gherkin in London), 12m wide and perhaps 5 or 6 deep, making it somewhere north of ten thousand cubic metres in size. It would take four or five hundred of the largest bin lorries to shift it all. This was not a few dodgy geezers in white vans, but a large criminal operation that must have involved dozens of people.

The sheer scale of the crime scene makes its location rather ironic, because this giant landfill sits just twelve hundred yards from the headquarters of Thames Valley Police in Kidlington, just north of Oxford. In fact visiting the town was a surreal experience in its own right – the local Sainsbury’s is such a notorious crime spot that the supermarket has installed a highly visible CCTV monitoring station in front of the exit, with a uniformed guard watching TV screens as you wander by with your shopping. Presumably this security theatre is supposed to comfort shoppers, but it made the place feel like some lawless outpost, a town in visible decline.

In fairness to the police, it’s hard to imagine a more convenient or secluded site to carry out this crime. Surrounded by trees, it was completely screened off from view until leaves began falling in the Autumn. In theory a public footpath crosses the land, but in practice nobody would ever walk down it – one end is hidden behind a crash barrier on the main road half a mile out of town, while the other terminates at the end of a field in the middle of nowhere, coming out on a small road with no parking nearby and no other paths to connect with.

In any case, Thames Valley Police have shown little interest in the crime – perhaps too busy with the local Sainsbury’s – and the job of dealing with it has fallen to the Environment Agency, the public body responsible for waste crime. They swung into action at the start of July, and in the spirit of being as fair as I possibly can, I’ll tell you their side of the story first.

«

There’s an embedded 10-minute YouTube video which shows this story in full, with terrific drone footage. It will surely make you angry at the sheer incompetence and indifference of the Environment Agency, which shows absolutely no interest in discovering the people behind this or using simple detection methods such as wildlife cameras. I’ve put more effort into finding a lost dog than they did into uncovering those behind a criminal enterprise that will poison a river.
unique link to this extract

UK ministers aim to ban cryptocurrency political donations over anonymity risks • The Guardian

Rowena Mason:

»

Ministers are working to ban political donations made with cryptocurrency but the crackdown is not likely to be ready for the elections bill in the new year, Whitehall sources have said.

The government increasingly believes that donations made with cryptocurrency pose a risk to the integrity of the electoral system, not least because the source can be hard to verify.

However, the complex nature of cryptocurrency means officials do not believe a ban will be workable by the time of the elections bill, due to be published shortly, which is set to lower the voting age to 16 and reduce loopholes in political finance.

The government’s ambition to ban crypto donations will be a blow to Nigel Farage’s Reform UK party, which became the first to accept contributions in digital currency this year. It is believed to have received its first registrable donations in cryptocurrency this autumn and the party has set up its own crypto portal to receive contributions, saying it is subject to “enhanced” checks.

Government sources have said ministers believe cryptocurrency donations to be a problem, as they are difficult to trace and could be exploited by foreign powers or criminals.

Pat McFadden, then a Cabinet Office minister, first raised the idea in July, saying: “I definitely think it is something that the Electoral Commission should be considering. I think that it’s very important that we know who is providing the donation, are they properly registered, what are the bona fides of that donation.”

The Electoral Commission provides guidance on crypto donations but ministers accept any ban would probably have to come from the government through legislation.

Earlier this year, the Electoral Commission initially appeared to believe the risks of donations in cryptocurrency were manageable, saying they could be assessed like any other asset such as a work of art or donations in kind.

«

Interesting point: if donations are anonymous, then how can they be influence? But this is yet another form of hawala – the trust-based system of money transfer. You tell your target, in a secure way, that you’re making a donation of a specified amount; the donation turns up subsequently in a crypto transfer. The Electoral Commission is overoptimistic.
unique link to this extract

Google is experimentally replacing news headlines with AI clickbait nonsense • The Verge

Sean Hollister:

»

Did you know that BG3 players exploit children? Are you aware that Qi2 slows older Pixels? If we wrote those misleading headlines, readers would rip us a new one — but Google is experimentally beginning to replace the original headlines on stories it serves with AI nonsense like that.

I read a lot of my bedtime news via Google Discover, aka “swipe right on your Samsung Galaxy or Google Pixel homescreen until you see a news feed appear,” and that’s where these new AI headlines are beginning to show up.

They’re not all bad. For example, “Origami model wins prize” and “Hyundai, Kia gain share” seem fine, even if not remotely as interesting as the original headlines. (“Hyundai and Kia are lapping the competition as US market share reaches a new record” and “14-year-old wins prize for origami that can hold 10,000 times its own weight” sound like they’re actually worth a click!)

But in the seeming attempt to boil down every story to four words or less, Google’s new headline experiment is attaching plenty of misleading and inane headlines to journalists’ work, and with little disclosure that Google’s AI is rewriting them.

The very first one I saw was “Steam Machine price revealed,” which it most certainly was not! Valve won’t reveal that till next year. Ars Technica’s original headline was the far more reasonable “Valve’s Steam Machine looks like a console, but don’t expect it to be priced like one.”

…The good news is, this is a Google experiment. If there’s enough backlash, the company probably won’t proceed. “These screenshots show a small UI experiment for a subset of Discover users,” Google spokesperson Mallory Deleon tells The Verge. “We are testing a new design that changes the placement of existing headlines to make topic details easier to digest before they explore links from across the web.”

But the overall trend at Google has been to prioritize its own products at the expense of sending clicks to news websites.

«

Google used to be about the open web. More and more, it’s about keeping people inside its properties.
unique link to this extract

‘Unauthorized’ edit to Ukraine’s frontline maps point to Polymarket’s war betting • 404 Media

Matthew Gault:

»

A live map that tracks frontlines of the war in Ukraine was edited to show a fake Russian advance on the city of Myrnohrad on November 15. The edit coincided with the resolution of a bet on Polymarket, a site where users can bet on anything from basketball games to presidential election and ongoing conflicts. If Russia captured Myrnohrad by the middle of November, then some gamblers would make money. According to the map that Polymarket relies on, they secured the town just before 10:48 UTC on November 15. The bet resolved and then, mysteriously, the map was edited again and the Russian advance vanished.

The degenerate gamblers on Polymarket are making money by betting on the outcomes of battles big and small in the war between Ukraine and Russia. To adjudicate the real time exchange of territory in a complicated war, Polymarket uses a map generated by the Institute for the Study of War (ISW), a DC-based think tank that monitors conflict around the globe.

One of ISW’s most famous products is its live map of the war in Ukraine. The think tank updates the map throughout the day based on a number of different factors including on the ground reports. The map is considered the gold standard for reporting on the current front lines of the conflict, so much so that Polymarket uses it to resolve bets on its website.

…ISW acknowledged the stealth edit, but did not say if it was made because of the betting markets. “It has come to ISW’s attention that an unauthorized and unapproved edit to the interactive map of Russia’s invasion of Ukraine was made on the night of November 15-16 EST. The unauthorized edit was removed before the day’s normal workflow began on November 16 and did not affect ISW mapping on that or any subsequent day. The edit did not form any part of the assessment of authorized map changes on that or any other day. We apologize to our readers and the users of our maps for this incident,” ISW said in a statement on its website.

ISW did say it isn’t happy that Polymarket is using its map of the war as a gambling resource.

«

Gamblers are weird, weird people.
unique link to this extract

Companies have found a new way to advertise: ragebaiting. You’ll hate it • The Washington Post

Tatum Hunter and Nitasha Tiku:

»

Tech founder Avi Schiffmann spent around a million dollars this autumn papering New York City’s subways with ads proclaiming that Friend, an AI device worn like a necklace, is a better support system than human companions.

The ads were less about selling the device, he said, than getting people to talk about it — for good or ill.

On those terms, at least, it worked. Riders, angry at the encroachment of AI, vandalized many of the ads with scrawled messages such as “Stop capitalizing on loneliness” and “AI wouldn’t care if you lived or died.” Anti-AI social media chatter featuring photos of the defaced ads started gaining traction online.

Schiffmann, 23, sat back and watched the attention roll in. When subway workers started washing the graffiti off the ads, he raced on foot to the West 4th Street station to beg them to stop.

“I wanted Friend to be a scapegoat for everything people don’t like about the world right now,” Schiffmann said. The campaign’s viral success, he added, was primarily the work of online posters rushing to smear Friend’s product and presentation. All he did was set the bait.

Schiffmann is hardly alone. Ragebait — the art of making people mad on social media — has graduated this year from a growth hack for online influencers to a corporate marketing strategy. This week Oxford University Press declared “rage bait” 2025’s word of the year, finding the term’s usage has tripled in the last 12 months.

«

Does it make you angry? Does it? (Most of this stuff is very resistible, if we’re honest.)
unique link to this extract

The Taiwan crisis of 2025 is here • National Security Journal

Robert E. Kelly:

»

Japan and China are now locked in a protracted spat over China’s claims to Taiwan.

What started as a minor flap is growing into a major contest in which regional players are desperately trying to avoid taking sides between the two rivals and are increasingly staking out opposed positions.

China’s designs on Taiwan are well known, but Beijing appears to have suddenly decided to force the issue in the region.

Beijing is using new Japanese Prime Minister Sanae Takaichi’s words—that a Chinese assault on Taiwan would inevitably become a security issue for Japan—to bully the region to accept the Chinese position on Taiwan, namely, that it should be permitted to invade and conquer it with no outside intervention.

Japan is China’s primary antagonist in the region. No other economy is large enough to compete with China, and the US alliance with Japan is the linchpin of the US position in East Asia.

This position is turning into a major showdown. If Beijing can humble Japan—if it can force Takaichi, via trade coercion and military threats, to retract her words—then it will establish rhetorical dominance over its regional rival.

A Japanese capitulation will signal to other regional powers, such as South Korea and the Philippines, that they, too, should find an accommodation with Beijing.

For this reason, Japan is unlikely to back down. It cannot afford to swerve in a direct chicken contest with its primary competitor. This stalemate will therefore likely continue for a while.

That Japan and China might fall into a cold war over the future of East Asia is not a new observation.

The chill began under the premiership of Shinzo Abe in Japan and the presidency of Xi Jinping of China. But both sides had strong economic incentives to keep security competition muffled.

Their trade relationship is substantial. Both would suffer from a prolonged fallout. When the history of this standoff is written, much focus will be on why China chose this moment to plant its flag. Does it now feel ready to take Japan on directly?

«

Occasionally, geopolitics intrudes. This is important, even if it goes under the radar of all the Trumpist nonsense across the world. (Kelly is an analyst based in Korea who achieved fame when his children intruded on his BBC talking head spot in 2017.)
unique link to this extract

Nuclear Taskforce Tracker • Centre for British Progress

Centre for British Progress:

»

Tracking the progress of government departments, regulators and industry in implementing the UK Nuclear Regulatory Taskforce’s recommendations. The content in this tracker is partially AI-generated based on the Nuclear Regulatory Taskforce report. We have worked hard to ensure it is accurate, but some of the titles, descriptions, etc. may be slightly different or truncated.

«

The Centre for British Progress describes itself as “a non-partisan think tank researching and producing concrete ideas for an era of British growth and progress.” So far everything’s on track! Though that’s only one recommendation completed. Many, many more to come.
unique link to this extract

Why Zipcar gave up on London • London Centric

Jim Waterson and Polly Smythe:

»

Zipcar’s planned closure date coincides with the mayor’s decision to introduce a new £13.50 daily congestion charge on electric vehicles, a move that would hit any Zipcar that is picked up outside the zone and driven through central London. The company had already said it would pass on the cost to drivers, substantially raising the price of a car journey through the heart of the capital — and making it much less financially attractive.

…The congestion charge extension might have been the final nail in Zipcar’s coffin. But looking at the company’s UK accounts, it’s clear the business model had been in deep trouble for several years due to rising costs and flatlining revenue.

Zipcar’s UK income fell by £3.95m to £47m in 2024, due to customers taking fewer and shorter trips in their cars. Costs increased, meaning post-tax losses widened dramatically to £11.6m. The company said the cost-of-living crisis “reduced members’ disposable income and impacted their demand for leisure activities”. Electric vehicles proved to be costly to buy and difficult to resell.

The arrival of Uber in the mid-2010s ate into Zipcar’s business model of enabling people nipping around London for short car trips. It’s also reasonable to assume that some people who might have been tempted by a one-way Zipcar Flex in the past are now choosing to pop on a much cheaper Lime e-bike. IKEA started doing delivery.

All in all, People just aren’t travelling as much and while a substantial number of Londoners came to rely on Zipcar as an emergency back-up travel option, that wasn’t enough to sustain it as a profitable business.

Parent company Avis Budget, which is already in financially dire straits, simply appears to have had enough and pulled the plug on this comparatively small part of its UK operation at short notice.

But what’s the impact going to be?

Moving house with one of Zipcar’s van and the help of a couple of friends became a rite of passage for many young Londoners. Zipcar claimed each of its vehicles removed 27 barely-used privately owned cars from the capital’s roads, with 12,000 businesses supposedly using its services.

«

Reading this, and an earlier London Centric report on how TfL isn’t hitting its (lowering) carbon emission targets unless it gets more EVs on the roads, one despairs a little of anybody being able to come up with a scheme where the left hand and right hand are in communication.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

The car hire company Zipcar is shutting down its UK arm, partly blaming high electricity prices for charging its EVs. CC-licensed photo by Michael Coghlan on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

There’s another post coming this week at the Social Warming Substack on Friday at 0845 UK time. Free signup.

A selection of 9 links for you. Discharged. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

your phone is a fake house • The Etymology Nerd

Adam Aleksic:

»

Only I have the motor memory to immediately open the Notes app on my phone. A stranger would have to look for it, but my fingers subconsciously understand where to go. Much like with my childhood home, I have an embodied knowledge of my home screen.

That phrase—“home screen”—has been on my mind recently. The language of the smartphone invites you to think of it as a house. You can “choose your wallpaper,” just like with a real house; you can “lock” your phone like a front door. The metaphor is that this is a private refuge from the outside world. It is a tiny dwelling in your pocket, which you can customize like an actual dwelling to affirm your identity. In doing so, you “tame” the technology, making it feel natural in your everyday life.

The phone, like your house, is a focal point. Everything revolves around it. When you need comfort in the physical world, you go back to the home; in the digital world, you go back to your home screen. There is something calming about a deeply personal environment. It provides a grounding presence which we can retreat to.

A computer, meanwhile, remains more functional. Phrases like “desktop” and “taskbar” create a metaphor that this is a workstation; you have “trash” and “files.” Of course, there are still work-like aspects to the phone and home-like aspects to the computer, but the phone takes on a far more domestic role in our lives. It is not a utility: it is an extension of self.

In his book The Poetics of Space the philosopher Gaston Bachelard argues that our intimate spaces are deeply intertwined with our imagination and sense of being. When you curl up in a comfortable nook in your home, for example, your consciousness is gathered inward. You have control over this small space, in contrast to the wild, turbulent outdoors. You can focus attention differently in miniature.

As I move between apps on my phone, I notice a vague emotion that I am entering different rooms, each with its own character. The settings app is the basement; the dating apps are the bedroom. No matter where I go, though, there is that coziness of being in a nook. This is my corner of the world; I am free to do what I want. I can let my mind relax, for I am safe and secure from the vast, terrifying world.

«

The subtlety of language; the seductiveness of the faux-private space. But is it really private?
unique link to this extract

The AI boom is based on a fundamental mistake • The Verge

Benjamin Riley:

»

The AI hype machine relentlessly promotes the idea that we’re on the verge of creating something as intelligent as humans, or even “superintelligence” that will dwarf our own cognitive capacities. If we gather tons of data about the world, and combine this with ever more powerful computing power (read: Nvidia chips) to improve our statistical correlations, then presto, we’ll have AGI. Scaling is all we need.

But this theory is seriously scientifically flawed. LLMs are simply tools that emulate the communicative function of language, not the separate and distinct cognitive process of thinking and reasoning, no matter how many data centers we build.

Last year, three scientists published a commentary in the journal Nature titled, with admirable clarity, “Language is primarily a tool for communication rather than thought.” Co-authored by Evelina Fedorenko (MIT), Steven T. Piantadosi (UC Berkeley) and Edward A.F. Gibson (MIT), the article is a tour de force summary of decades of scientific research regarding the relationship between language and thought, and has two purposes: one, to tear down the notion that language gives rise to our ability to think and reason, and two, to build up the idea that language evolved as a cultural tool we use to share our thoughts with one another.

Let’s take each of these claims in turn.

When we contemplate our own thinking, it often feels as if we are thinking in a particular language, and therefore because of our language. But if it were true that language is essential to thought, then taking away language should likewise take away our ability to think. This does not happen. I repeat: taking away language does not take away our ability to think. And we know this for a couple of empirical reasons.

«

This one will run and run. Is a word a representation of an idea? Is manipulating words in multidimensional space thinking, or is it just shuffling a dictionary?
unique link to this extract

Zipcar to exit UK car sharing market • Financial Times

Kana Inagaki:

»

Zipcar, the car-sharing giant owned by US rental group Avis Budget, has said it plans to exit the UK market, in the latest hit to the struggling industry. 

In an email to customers on Monday, James Taylor, the general manager of Zipcar UK, said it would suspend new bookings beyond December 31, pending the outcome of a consultation with British employees.

The group claims to be the UK’s biggest car-sharing club with more than 650,000 members, who can rent cars and vans for between an hour and seven days.

According to its latest accounts, the company had 71 employees last year, down from 92 in 2023. It cited high electricity costs as it reported an operating loss of £4m last year, compared with a profit of £303,000 in the previous 12 months.

“This particularly affected the company due to the size of the electric fleet and the fact that fuel costs are included in the cost of the rental,” it said.

In a statement on Monday, Avis Budget said Zipcar’s operations outside of the UK will not be affected.

…The company began operations in London in 2007 and now operates more than 1,000 electric vehicles in the city as part of a fleet of more than 3,000.

Zipcar’s plan to end its UK operations comes ahead of London’s introduction of a £13.50 a day congestion charge from January 2 for EVs, marking the first time the battery-powered cars will be hit with a fee for entering the city’s centre.

«

Christmas bookings assured; everything afterwards, not so much.
unique link to this extract

Airbus A320 recall: what the solar flare software update is all about • Gulf News

Jay Hilotin:

»

Airbus, the European planemaker, issued an urgent Alert Operators Transmission (AOT) on November 29, 2025, mandating immediate software updates or hardware protections for around 6,000 A320 family aircraft — over half the global fleet.

The move came after a detailed probe into a JetBlue A320 nosedive incident on October 30 revealed that intense solar radiation could corrupt key flight control data in the Elevator Aileron Computer (ELAC). Intense solar radiation incidents are sometimes blamed for the incineration of dozens of satellites.

The vulnerability affects aircraft with specific ELAC B software versions exposed to high-altitude radiation from solar flares. This, the planemaker said, risks uncommanded pitch or roll manoeuvres.

Airbus is collaborating with regulators like EASA, which has formalised an Emergency Airworthiness Directive effective 6pm CT on November 30, 2025.

…JetBlue A320 from Cancun (Mexico) to Newark (US) suddenly lost altitude mid-flight due to corrupted ELAC data. The incident, blamed on solar radiation, injured passengers and prompted an FAA investigation; this led to the global recall.

Intense solar radiation is seen behind data corruption in the ELAC (Elevator Aileron Computer) flight control system, potentially causing uncommanded manoeuvres like sudden altitude drops.

About 6,000 active A320s worldwide require fixes before next flights, impacting operators like IndiGo (338 affected), Air India (138), American Airlines (209 of 480), Jetstar, Wizz Air and others across Asia, Europe, Americas and the US.

Most software rollbacks take 2-3 hours per plane, but ~1,000 need hardware changes lasting weeks; this has caused thousands of cancellations and delays worldwide, stranding millions during holidays, though many airlines aim for completion by November 30.

«

A surprising little result of the solar flare activity. Sure to be written in to a streaming service’s cheap film in a year or two.
unique link to this extract

The triumph of logical English • Works in Progress Magazine

Henry Oliver:

»

The Elizabethans and Victorians wrote long tangled sentences that resembled the briars growing underneath Sleeping Beauty’s tower. Today we write like Hemingway. Short. Sharp. Readable. Pick up an old book and the sentences roll on. Go to the office, read the paper, or scroll Twitter and they do not. So it is said. I would like to suggest that this account is incomplete. 

I propose a different story. The great shift in English prose took place in the sixteenth and seventeenth centuries, probably driven by the increasing use of writing in commercial contexts, and by the style of English in post-Reformation Christianity. It consisted in two things: a ‘plain style’ and logical syntax. A second, smaller shift has taken place in modern times, in which written English came to be modelled more closely on spoken English.

What this should demonstrate is that shortness is the wrong dimension to investigate. We think we are looking at a language that got simpler; in fact we are looking at one that has created huge variation in what it can express and how, by adding new ways of writing. Lots of English writing has got simpler through use of the plain style, sticking to a logical shared syntax, especially the syntax of speech. But all the other ways of writing are still there, often showing up when we don’t expect them.

…You might have read this essay and largely agreed with me but still been left with the feeling that something is different about modern prose as compared to the writing of the 1700s, not just the fact that we use less obscure vocabulary or the substituting full stops for colons and semicolons. Something else is still different. I think that something is that we increasingly write like we speak.

«

Write as we speak. Notwithstanding, it’s a fascinating (long) article. Furthermore, Works in Progress is rapidly becoming a must-read. They also do a print version.
unique link to this extract

Meet Rey, the admin of ‘Scattered Lapsus$ Hunters’ • Krebs on Security

Brian Krebs:

»

A prolific cybercriminal group that calls itself “Scattered LAPSUS$ Hunters” has dominated headlines this year by regularly stealing data from and publicly mass extorting dozens of major corporations. But the tables seem to have turned somewhat for “Rey,” the moniker chosen by the technical operator and public face of the hacker group: Earlier this week, Rey confirmed his real life identity and agreed to an interview after KrebsOnSecurity tracked him down and contacted his father.

Scattered LAPSUS$ Hunters (SLSH) is thought to be an amalgamation of three hacking groups — Scattered Spider, LAPSUS$ and ShinyHunters. Members of these gangs hail from many of the same chat channels on the Com, a mostly English-language cybercriminal community that operates across an ocean of Telegram and Discord servers.

«

Krebs ranges far and wide, pulling in tiny pieces of information – passwords reused here, email scams used there – to eventually track down the person..

»

Specifically, Rey mentioned in several Telegram chats that he had Irish heritage, even posting a graphic that shows the prevalence of the surname “Ginty.”

Spycloud indexed hundreds of credentials stolen from cybero5dev@proton.me, and those details indicate that Rey’s computer is a shared Microsoft Windows device located in Amman, Jordan. The credential data stolen from Rey in early 2024 show there are multiple users of the infected PC, but that all shared the same last name of Khader and an address in Amman, Jordan.

The “autofill” data lifted from Rey’s family PC contains an entry for a 46-year-old Zaid Khader that says his mother’s maiden name was Ginty. The infostealer data also shows Zaid Khader frequently accessed internal websites for employees of Royal Jordanian Airlines. The infostealer data makes clear that Rey’s full name is Saif Al-Din Khader.

«

He’s 16, or claims to be.
unique link to this extract

Dutch study finds teen cybercrime is mostly just a phase • The Register

Connor Jones:

»

Young threat actors may be rebels without a cause. These cybercriminals typically grow out of their offending ways by the time they turn 20, according to data published by the Dutch government.

In a report examining the social cost of adolescent crime, the Dutch House of Representatives cited various research papers to show that teenagers tend to explore their criminal tendencies at similar ages, regardless of the type of crime.

The report stated that cybercriminals tend to develop their skills at an early age – no shocks there – and do so through “hacking games.” The number of teenage cyber offenders is similar to those involved in weapons or drug offenses; together they are the three least common offence types for adolescents. Property offences such as theft were the most common.

Young cybercriminals reach peak criminality at around age 20, although this tends to fluctuate by a few years, depending on the decade. For example, between 2010-2012 and 2018-2021, the peak age floated between 17 and 19, but, in between, it remained at 20.

Research also shows that these peak ages broadly apply to all crimes, cyber or otherwise.

In 2013, one study of 323 cybercriminals found that 76% of offenders reached peak offending at age 20, before veering away from the trade gradually in the following years.

Only around 4% of those who embark on an early black hat career maintain a high likelihood of continuing that into ages well beyond the 20 mark.

«

So only a couple more years for Krebs’s target. But you know there will be more to come.
unique link to this extract

Out of Eden Walk • Center for Geographic Analysis

»

The Out of Eden Walk is a 24,000-mile journalistic endeavor to create a global record of human life at the start of a new millennium as told by villagers, nomads, traders, farmers, soldiers, and artists who rarely make the news.  Sponsored and hosted by National Geographic Society, the project is led by Pulitzer Prize winning writer Paul Salopek who is walking the path of human migration across the globe, and recording what he encounters in the form of writing, photographs, video, and audio.

«

One of the most impressive parts is the 2013 “Walk Through Time” which shows human migration from the East Africa Rift valley through Africa and then the rest of the world.
unique link to this extract

Michel Bettane: Chinese wine now outshines France in technical precision • Vino Joy News

Vino Joy News:

»

French wine critic Michel Bettane says the overall technical standard of Chinese wines now surpasses what he and his team often encounter in their annual tastings in France — a sign, he believes, that China is undergoing an “astonishing awakening of terroir.”

The sixth edition of the Bettane + Desseauve China Wine Tasting concluded in Beijing and Shangri-La, Yunnan, in September 2025. Led by Bettane, the panel of six international experts and local judges evaluated more than 300 premium Chinese wines — a process that, he said, revealed how far the country’s winemaking has come.

Chinese wines, Bettane noted, have reached a level of maturity unimaginable 15 years ago. When he first visited, the market “was dominated by a few major brands” and “people focused on the label, not the land.” Today, from the deserts of Xinjiang to the mountains of Yunnan, he said, producers are “confidently expressing their terroir.”

What impressed him most, however, was the technical precision. “We encountered almost no wines with serious flaws,” he said. “The overall solidity of the winemaking standard is, in fact, superior to what we often find in our annual tastings in France.”

Among China’s emerging wine regions, Bettane highlighted two with strikingly different identities: Ningxia and Yunnan’s Deqin County.

«

How long before we’re seeing Chinese wine on our supermarket shelves? Though French wine has, to some tastes, been going downhill for a while.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

Buyer beware – the “Pyrex” trademark has been licensed to companies which don’t make cooking glassware to the same standards as the original one. CC-licensed photo by ricky shore on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Hot in here. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.

The world still hasn’t made sense of ChatGPT • The Atlantic

Charlie Warzel on the third anniversary of the introduction of ChatGPT:

»

This is disruption, in the less technical sense of the word. In August, I wrote that “one of AI’s enduring impacts is to make people feel like they’re losing it.” If you genuinely believe that we are just years away from the arrival of a paradigm-shifting, society-remaking superintelligence, behaving irrationally makes sense. If you believe that Silicon Valley’s elites have lost their minds, foisting a useful-but-not-magical technology on society, declaring that it’s building God, investing historic amounts of money in its development, and fusing the fate of its tools with the fate of the global economy, being furious makes sense.

The world that ChatGPT built is a world defined by a particular type of precarity. It is a world that is perpetually waiting for a shoe to drop. Young generations feel this instability acutely as they prepare to graduate into a workforce about which they are cautioned that there may be no predictable path to a career.

Older generations, too, are told that the future might be unrecognizable, that the marketable skills they’ve honed may not be relevant. Investors are waiting too, dumping unfathomable amounts of capital into AI companies, data centers, and the physical infrastructure that they believe is necessary to bring about this arrival. It is, we’re told, a race—a geopolitical one, but also a race against the market, a bubble, a circular movement of money and byzantine financial instruments and debt investment that could tank the economy. The AI boosters are waiting. They’ve created detailed timelines for this arrival. Then the timelines shift.

We are waiting because a defining feature of generative AI, according to its true believers, is that it is never in its final form. Like ChatGPT before its release, every model in some way is also a “low-key research preview”—a proof of concept for what’s really possible. You think the models are good now? Ha! Just wait.

Depending on your views, this is trademark showmanship, a truism of innovation, a hostage situation, or a long con. Where you fall on this rapture-to-bullshit continuum likely tracks with how optimistic you are for the future. But you are waiting nonetheless—for a bubble to burst, for a genie to arrive with a plan to print money, for a bailout, for Judgment Day. In that way, generative AI is a faith-based technology.

«

“Anniversary journalism” is usually a poor excuse for an article, but it’s important here to take stock of how truly disruptive this has all been. Not often in a good way.
unique link to this extract

Someone is trying to ‘hack’ people through Apple’s Podcasts app • 404 Media

Joseph Cox:

»

Something very strange is happening to the Apple Podcasts app. Over the last several months, I’ve found both the iOS and Mac versions of the Podcasts app will open religion, spirituality, and education podcasts with no apparent rhyme or reason. Sometimes, I unlock my machine and the podcast app has launched itself and presented one of the bizarre podcasts to me. On top of that, at least one of the podcast pages in the app includes a link to a potentially malicious website. Here are the titles of some of the very odd podcasts I’ve had thrust upon me recently (I’ve trimmed some and defanged some links so you don’t accidentally click one):

“5../XEWE2′””&#x22″onclic…”
“free will, free willhttp://www[.]sermonaudio[.]com/rss_search.asp?keyword=free%will on SermonAudio”
“Leonel Pimentahttps://play[.]google[.]com/store/apps/detai…”
“https://open[.]spotify[.]com/playlist/53TA8e97shGyQ6iMk6TDjc?…”

There was another with a title in Arabic that loosely translates to “Words of Life” and includes someone’s Gmail address. Sometimes the podcasts do have actual audio (one was a religious sermon); others are completely silent. The podcasts are often years old, but for some reason are being shown to me now.

I’ll be honest: I don’t really know what exactly is going on here. And neither did an expert I spoke to. But it’s clear someone, somewhere, is trying to mess with Apple Podcasts and its users.

«

Security experts agree: something weird is going on, apparently with the intent of launching an XSS (cross-site scripting) attack. Apple didn’t respond to Cox – which implies that it’s having a think about what’s going on here.
unique link to this extract

Survey says 97% of people struggle to identify AI music – but it’s not as bad as it seems • The Verge

Terrence O’Brien:

»

Streaming service Deezer ran an experiment recently, with the help of research firm Ipsos. The finding — that 97% of people can’t tell the difference between fully AI-generated and human-made music — was alarming. But it’s also not the whole story.

In the survey, 9,000 participants listened to three tracks and were asked to guess which, if any, were completely AI-generated. If the participant failed to guess all three correctly, they were put in the fail pile. That means if you got two of three correct, Deezer and Ipsos still said you couldn’t tell the difference between fully AI-generated music and the real deal.

Deezer sent me the three tracks it used in the study, and so I decided to run my own (less scientific) experiment. I had ten people listen to the same tracks and gave them the same prompt. People did have trouble identifying which songs were fully AI. Only one person got all three right. But if I didn’t bundle the responses, the results were much less dire. People were able to successfully identify whether a track was AI or human-generated 43% of the time.

It’s also worth noting that several people told me one of the songs was so terrible, so obviously AI, that they thought it had to be a trap and guessed it was real.

Unsurprisingly, participants in Deezer’s study were a little caught off guard by how poorly they performed: 71% were surprised by the results, and 51% said it made them uncomfortable to not be able to tell the difference between AI- and human-created art.

«

This might get like autotune, which was reviled at first (when Cher’s Believe was a huge hit in 1998, music executives insisted the voice effect was a vocoder, not autotune. Nowadays, it’s all over the place. AI tunes with human vocals? Human musicians with AI vocals? It’s all on the table now.
unique link to this extract

From Olympic dreams to Nairobi jail: how an Indian teen got embroiled in doping scandal in Kenya • The Indian Express

Mihir Vasavda and Nihal Koshie:

»

Iten, a Kenyan town that runners consider their mecca and where world-beaters are forged at altitude, is where Aman Malik, all of 17, chose to go in May 2023. The budding cross-country and long-distance runner from Haryana’s Sonipat was convinced that the road to the Olympics ran through this town in East Africa.

Two years on, the script has flipped completely.

In September 2025, a Nairobi court handed Aman, now 19, a three-year sentence for being a part of an organised network that allegedly traffics prohibited substances into the country and gives banned substances to Kenyan athletes.

Now in a four-room enclosure that houses 30 inmates, Aman has been navigating an environment far removed from the training camps he once lived in. “They could have banned me from athletics or deported me to India, where I could have served a jail term,” he tells The Indian Express from the Nairobi jail, where he gets to use his phone for one hour daily.

A high-altitude town in Kenya’s Great Rift Valley, Iten enjoys global fame as the ‘home of champions’ for the sheer number of world and Olympic winners it has produced, including David Rudisha, the 800m Olympic and world record holder; and Beatrice Chebet, the multiple Olympic and world championship gold medallist.

However, of late, this distance-running powerhouse has been battling a surge in doping violations, besides accounting for the highest number of doping cheats in track and field.

Aman left for Iten in May 2023. Two years later, while returning from training, Kenya’s Directorate of Criminal Investigations sleuths followed him and raided his room. According to Kenyan court documents, they found an “entire suitcase filled with prohibited substances, supplements and medication” in his possession. These “prohibited substances” included meldonium, a drug that had led to a 15-month ban on tennis star Maria Sharapova; a human growth hormone (HGH) made infamous by the Lance Armstrong doping scandal; and Mannitol, a masking agent.

The sleuths also allegedly seized a one-page agreement between Aman and athlete Reubin Mosin that states the Indian would “supply all that it takes” in return for 50% of the latter’s winnings.

«

Kenya has a colossal doping problem, and this starts to clarify how and why.
unique link to this extract

Indiana Jones and the Last Crusade: learning from Stoppard • Creative Screenwriting

Mike Fitzgerald:

»

Comparing two drafts of a script can be hugely instructive, revealing point-by-point how a writer went about improving the story. When I stumbled upon an earlier draft of Indiana Jones and the Last Crusade, I discovered a dazzling, glittering trove of lessons as nourishing as eternal life itself. Well, nearly so.

Last Crusade was written by Jeffrey Boam, from a story by George Lucas and Menno Meyjes. So say the opening credits. Boam’s final draft, dated March 1, 1988 (ten weeks before production) differs drastically from the published script which reflects the released version of the film. Differences come as no shock, but with Last Crusade they aren’t just a few deleted scenes and some line changes. Whole sections of the Boam draft were reimagined, major set pieces were added, and the pacing and tone were markedly transformed. Whoever made these changes possessed a profound grasp of story craft.

So who was that? Spielberg himself made certain revisions, such as expanding the desert tank sequence from a few pages to over eleven – injecting some much-needed action into the story. Some scenes were filmed but omitted during the edit, like an extended chase through the Zeppelin in which Indy and Henry are pursued by a Gestapo agent and a World War One flying ace.

And then there was the uncredited script polish by Barry Watson – you know, the Barry Watson? Never heard of him? Perhaps if we peek under his pseudonym… ah, yes: Sir Tom Stoppard, a four-time Tony winner who later bagged an Oscar for Shakespeare in Love. Since we can’t know whose pen revised which pages (although Spielberg did say that “Tom is pretty much responsible for every line of dialogue.” Let’s just call it a collaboration of some titans of storytelling.

«

There’s a similar breakdown elsewhere – which goes into scene-by-scene comparison. Apart from the pleasure of realising how much better the Stoppard-revised version is, it’s also a valuable lesson in writing generally: focus on character, learn how to time suspense, learn what amplifies a joke or tragedy. Useful for all writing. Well, all writing for humans.
unique link to this extract

San Francisco’s robotaxi takeover, as seen from city hall • Bloomberg

David Zipper:

»

…As the director of the San Francisco Municipal Transportation Agency since 2019, Jeffrey Tumlin has been immersed in his city’s self-driving saga. With unified oversight of transit, taxis, curbs and streets in San Francisco, SFMTA is more powerful than most urban transportation departments. Still, Tumlin and his colleagues struggled to handle robotaxi companies that are accountable only to state and federal regulators, not city officials.

DZ: You had worked in the transportation sector for decades before joining SFMTA, and you were already familiar with autonomous vehicles. What have you learned about self-driving technology in the last few years that most surprised you?
JT: As a regular Waymo user, I have watched the cars become better than I am at seeing pedestrians hidden from view and predicting their behavior. I didn’t think that was going to happen. I am surprised by how sophisticated they are with erratic human behavior, which I had assumed would be very challenging.

DZ: Overall, are robotaxis a positive or negative for San Francisco?
JT: So far, there is no net positive for the transportation system that we’ve been able to identify. The robotaxis create greater convenience for the privileged, but they create problems for the efficiency of the transportation system as a whole.

DZ: What do you mean by that?
JT: What I like about Waymo is that the user interface design works well. I don’t have to talk to a human, and the vehicle’s driving behavior is slow and steady. I think robotaxis offer the potential for significant upsides for personal convenience, but it remains to be seen whether they offer any overall benefit to the transportation system.

DZ: How would you respond to those who say robotaxis are making San Francisco a better city because the experience of using them is superior to other ways to travel?
JT: I agree that there are qualities or Waymos that outperform other modes. The vehicles are very nice. The driving behavior is slow and steady and predictable, and there is chill music.

But those are qualities that you can replicate in any mode. If we mandated speed governors, passenger cars can be slow and steady. If we regulated taxis in order to optimize for user convenience and safe driving behavior, taxis could emulate those same qualities as well. Similarly, if we had massive private funding, we could achieve the same level of quality in public transit.

«

(Article is free to read, apparently.)
unique link to this extract

It pays to speak fluent LinkedIn — if you can crack the bro code • The Times

Harry Wallop:

»

As on all social media platforms, LinkedIn does not show users a straightforward chronology of all the updates from the people they follow. Instead they are shown posts by a mix of people they follow, as well as those of strangers the algorithm thinks they might find interesting.

One woman, Cindy Gallop, was so suspicious LinkedIn had changed its algorithm that she conducted an experiment. Gallop, who is quite a big cheese on LinkedIn with 141,000 followers, got a fellow female businesswoman, Jane Evans, and two male colleagues to publish exactly the same post as her. Gallop’s reached 0.6% of her followers. Evans’s reached 8.3% of her 19,100 followers. The two men reached 51% and 143% of their respective follower numbers.

Some women have reacted by changing their profile on LinkedIn to pretend they are men. They have even started using “masculine” words to trick the LinkedIn algorithm into thinking they are men. Lots of “drive”, “accelerate”, “transform”, apparently, sees your posts reach more people. This is bro-coding. Speak like an obnoxious Silicon Valley tech bro, refer to forward-deployed engineers and how you are protein-maxxing when you get out of your ice bath at 5am, and your thoughts will be seen by more people.

Gallop and Evans are so enraged by this discovery they have started a petition. “We’re calling on LinkedIn to take urgent action,” they say, demanding “an independent equity audit of the algorithm and its impact on under-represented voices”.

At this point, you may want to scream. Who cares if LinkedIn, which has always been home to plenty of self-promoting, self-regarding and self-appointed “thought leaders”, amplifies the voices of some boastful men? If you hate LinkedIn so much, there’s no reason to spend time on this Microsoft-owned platform.

This fails to accept the reality of the modern business world. Over the past few years, being on LinkedIn has become almost mandatory. Microsoft claims LinkedIn has 1.3 billion users globally, with 44 million in Britain — an almost unbelievable number that is more than the 43.4 million adults in the country of working age.

«

It’s such an unserious platform, and yet because it’s become essential to people in business, it’s also unavoidable.
unique link to this extract

UK ‘not in favor’ of dimming the sun • POLITICO

Karl Mathiesen:

»

The British government said it opposes attempts to cool the planet by spraying millions of tons of dust into the atmosphere — but did not close the door to a debate on regulating the technology.

The comments in parliament Thursday came after a POLITICO investigation revealed an Israeli-US company Stardust Solutions aimed to be capable of deploying solar radiation modification, as the technology is called, inside this decade.

“We’re not in favor of solar radiation modification given the uncertainty around the potential risks it poses to the climate and environment,” Leader of the House of Commons Alan Campbell said on behalf of the government.

Stardust has recently raised $60m in finance from venture capital investors, mostly based in Silicon Valley and Britain. It is the largest ever investment in the field. 

The emergence of a well-funded, private sector actor moving aggressively toward planet cooling capability has led to calls for the global community to regulate the field.

Citing POLITICO’s reporting, Labour MP Sarah Coombes asked the government: “Given the potential risks of this technology, could we have a debate on how Britain will work with other countries to regulate experiments with the Earth’s atmosphere, and ensure we cooperate with other countries on solutions that actually tackle the root cause of climate change?”

Campbell signalled the government was open to further discussion of the issue by inviting Coombes to raise the point the next time Technology Secretary Liz Kendall took questions in parliament.

…Stardust is proposing to use high-flying aircraft to dump millions of tons of a proprietary particle into the stratosphere, around 12 miles above the Earth’s surface. The technology mimics the short-term global cooling that occurs when volcanoes blow dust and gas high into the sky, blocking a small amount of the sun’s heat.

Most scientists agree this could temporarily lower the Earth’s surface temperature, helping to avert some impacts of global warming. The side effects, however, are not well researched.  

«

unique link to this extract

What Is the difference between the two types of Pyrex? • Allrecipes

Meghan Glass:

»

Three basic types of glassware are typically found in most home kitchens: soda-lime, tempered, and borosilicate. Borosilicate glass includes boron trioxide, which has a low thermal expansion. This suggests unlike normal glass, it won’t break when exposed to major temperature shifts such as taking a dish from a fridge to an oven. This is thanks to boron trioxide, the element that makes glass resistant to major temperature changes. Pyrex is a sub-group of borosilicate.

Soda-lime glass is the most common glass type in kitchens since it’s used for most drinkware from juice cups to jars. Untreated soda-lime glass is more susceptible to breaking from extreme temperature changes. This shock expands the glass at different rates, resulting in cracks and fissures.

Tempered glass is just soda-lime glass that’s been heat-treated to make it more durable. During that heat-tempering process, the exterior of the glass is force-cooled so that it solidifies quickly, leaving the center to cool more slowly. As the inside cools, it pulls at the stiff, compressed outer layer, which puts the center of the glass in tension.

Are “PYREX” and “pyrex” the same? Historically, both trademarks were used interchangeably in the marketing of kitchenware products made up of both borosilicate and soda-lime glass. However, now Corning has licensed out the use of their PYREX (upper case lettering) and pyrex (lower case lettering) logos to other companies.

Lowercase “pyrex” is now mostly used for kitchenware sold in the United States, South America, and Asia. In Europe, Africa, and the Middle East, uppercase PYREX is still available.

Pyrex used to be made of the more heat-resistant borosilicate glass, which is more resistant to breakage when subjected to extreme shifts in temperature. Pyrex eventually switched to tempered glass most likely because boron is toxic and expensive to dispose of. Although tempered glass can better withstand thermal shock than regular soda-lime glass can, it’s not as resilient as borosilicate. This is what causes the shattering reaction people are talking about. Watch out for those casseroles.

In short: if the logo is in upper case lettering – PYREX – it’s most likely made of borosilicate, and thus safer.

«

Just in case anyone is thinking of doing some cooking this month. Too late for all the Americans at Thanksgiving, of course. Sorry folks!
unique link to this extract

Errata, corrigenda and ai no corrida: none notified