The growth of touchscreens may finally be going into reverse as their downsides become clear. CC-licensed photo by Graeme Maclean 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. Untouchable. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.

---

Have we reached peak touchscreen? Maybe yes • User Mag

Taylor Lorenz:

»

After years of futilely mashing our fingers onto touch screens, buttons on technology products are making a comeback. I’ve been fascinated by this re-buttonization of tech and once you see it, you’ll notice it everywhere.

There has been a proliferation of tools like the Clicks Keyboard Case, which appends a physical keyboard to your iPhone, BlackBerry style. E-readers like the Nook have started to put page-turning buttons back onto their devices.

Apple even added buttons back to the top of its MacBook Pro keyboards after backlash to the “Touch Bar” that it had it rolled out in 2016. In Apple’s announcement, the company noted, “Physical function keys… replace the Touch Bar, bringing back the familiar, tactile feel of mechanical keys that pro users love.”

Historically, producing custom buttons has been expensive, and as technology advanced in recent decades it became much cheaper for companies to produce touch screen interfaces. Touchscreens also have some benefits: they allow for a more flexible user interface design and make it easy to push updates to products remotely.

But now that touch screens have become ubiquitous, it’s becoming very clear that they suck. Touch screen interfaces can crash, rendering products unusable, they can wear out over time, and poorly designed touch screens often don’t sense a user’s swipes or taps. Not to mention, to navigate a touch screen interface you have to look at it, making things more dangerous when you integrate touchscreens into things like a car or heavy machinery.

«

I mean, buttons can come off in your hand, they can lose their functionality, but that’s nothing compared to the problems with touchscreens.
unique link to this extract

---

Chinese ship’s crew suspected of deliberately dragging anchor for 100 miles to cut Baltic cables • WSJ

Bojan Pancevski:

»

A Chinese commercial vessel that has been surrounded by European warships in international waters for a week is central to an investigation of suspected sabotage that threatens to test the limits of maritime law—and heighten tensions between Beijing and European capitals.

Investigators suspect that the crew of the Yi Peng 3 bulk carrier—225 meters long, 32 meters wide and loaded with Russian fertilizer—deliberately severed two critical data cables last week as its anchor was dragged along the Baltic seabed for over 100 miles (160km).

Their probe now centers on whether the captain of the Chinese-owned ship, which departed the Russian Baltic port of Ust-Luga on Nov. 15, was induced by Russian intelligence to carry out the sabotage. It would be the latest in a series of attacks on Europe’s critical infrastructure that law-enforcement and intelligence officials say have been orchestrated by Russia.

“It’s extremely unlikely that the captain would not have noticed that his ship dropped and dragged its anchor, losing speed for hours and cutting cables on the way,” said a senior European investigator involved in the case.

The ship’s Chinese owner, Ningbo Yipeng Shipping, is cooperating with the investigation and has allowed the vessel to be stopped in international waters, according to people familiar with the probe. The company declined to comment. 

The damage to undersea cables occurred in Swedish waters on Nov. 17-18, prompting that country’s authorities to open a sabotage investigation. Russia has denied wrongdoing. 

«

Ningbo Yipeng Shipping only owns one other vessel, and is based near the eastern Chinese port city of Ningbo. Shipping companies: terribly useful for all sorts of fronts.
unique link to this extract

---

Elon Musk’s Twitter backers gain windfall from xAI deal • Financial Times

Tabby Kinder and George Hammond:

»

Investors in Elon Musk’s takeover of Twitter are set to make a huge windfall from a surge in the valuation of his artificial intelligence company, reaping rewards from being loyal backers of the billionaire’s business empire.

Musk has given investors that backed his $44bn Twitter acquisition 25% of the shares in xAI, which he founded last year to take on rivals such as OpenAI and Anthropic.

xAI is set to close a new $5bn fundraising round as early as Wednesday, according to people with knowledge of the talks, doubling its valuation to $50bn in just six months.

That has meant some of Musk’s backers, who were sitting on billions of dollars of unrealised losses from the Twitter takeover, could be made “whole” through shares in xAI thanks to the startup’s massive rise in value.

Those set to benefit as investors in both Musk companies include Fidelity, Oracle co-founder Larry Ellison, Saudi Arabia’s Prince Alwaleed bin Talal, Twitter founder Jack Dorsey and Silicon Valley venture firms Sequoia Capital and Andreessen Horowitz.

The connections between the Musk businesses are the latest example of the overlapping incentives for those who support his ventures, which also include electric-car maker Tesla and rocket builder SpaceX.

Many of his financial backers have justified their support of the takeover of Twitter, since renamed X, as a bet on Musk and a means to stay within his orbit. That thinking has been considered especially prescient as Musk has become a close confidant of president-elect Donald Trump.

“There are few adages in tech that really hold up,” said one investor in Musk’s companies. “Never bet against Elon is one.”

«

Rich people not becoming poor but instead staying or becoming even more rich! It’s a tale told again and again. Weird how it happens, eh.
unique link to this extract

---

Google’s plan to keep AI out of search trial remedies isn’t going very well • Ars Technica

Ashley Belanger:

»

Google got some disappointing news at a status conference Tuesday, where US District Judge Amit Mehta suggested that Google’s AI products may be restricted as an appropriate remedy following the government’s win in the search monopoly trial.

According to Law360, Mehta said that “the recent emergence of AI products that are intended to mimic the functionality of search engines” is rapidly shifting the search market. Because the judge is now weighing preventive measures to combat Google’s anticompetitive behavior, the judge wants to hear much more about how each side views AI’s role in Google’s search empire during the remedies stage of litigation than he did during the search trial.

“AI and the integration of AI is only going to play a much larger role, it seems to me, in the remedy phase than it did in the liability phase,” Mehta said. “Is that because of the remedies being requested? Perhaps. But is it also potentially because the market that we have all been discussing has shifted?”

To fight the DOJ’s proposed remedies, Google is seemingly dragging its major AI rivals into the trial. Trying to prove that remedies would harm Google’s ability to compete, the tech company is currently trying to pry into Microsoft’s AI deals, including its $13bn investment in OpenAI, Law360 reported. At least preliminarily, Mehta has agreed that information Google is seeking from rivals has “core relevance” to the remedies litigation, Law360 reported.

The DOJ has asked for a wide range of remedies to stop Google from potentially using AI to entrench its market dominance in search and search text advertising. They include a ban on exclusive agreements with publishers to train on content, which the DOJ fears might allow Google to block AI rivals from licensing data, potentially posing a barrier to entry in both markets. Under the proposed remedies, Google would also face restrictions on investments in or acquisitions of AI products, as well as mergers with AI companies.

«

unique link to this extract

---

OpenAI’s Sora tool leaked by group of aggrieved early testers • Forbes

Moin Roberts-Islam:

»

A storm has been brewing in the AI landscape following the unauthorized leak of OpenAI’s groundbreaking Sora model, a text-to-video generator that has been making waves for its ability to create short, high-fidelity videos with remarkable temporal stability. At the heart of the controversy is a multifaceted conflict involving technological advancement, ethical concerns and artistic advocacy.

The leak was posted on Hugging Face and was allegedly carried out by individuals involved in the testing phase — using the username “PR-Puppets” — and raises pressing questions about the relationship between innovation, labor and corporate accountability. The leaked model, released alongside an open letter addressed to the “Corporate AI Overlords,” can purportedly produce 10-second video clips at up to 1080p resolution.

…The leak of Sora’s model appears to stem from dissatisfaction among testers and contributors, particularly those in creative industries. Critics allege that OpenAI (currently valued at over $150 billion) exploited their labor by relying on unpaid or undercompensated contributions to refine the model. These testers, including visual artists and filmmakers, provided valuable feedback and creative input, only to allegedly find themselves excluded from equitable recognition or compensation.

“This wasn’t just about unpaid work—it was about respect,” noted one anonymous contributor quoted in the Hugging Face commentary. “OpenAI treated our input like raw material, not creative expertise. It’s not collaboration; it’s extraction.”

This act of rebellion serves as a protest against the broader commodification of creative expertise in AI development.

«

If you leak a product like this, does it really put power back into the hands of the artists, or does it just reduce the chance to monetise for OpenAI et al?
unique link to this extract

---

Nuclear electricity generation has hidden problems • Our Finite World

Gail Tverberg:

»

It is easy to get the impression that proposed new modular nuclear generating units will solve the problems of nuclear generation. Perhaps they will allow more nuclear electricity to be generated at a low cost and with much less of a problem with spent fuel.

As I analyze the situation, however, the problems associated with nuclear electricity generation are more complex and immediate than most people perceive. My analysis shows that the world is already dealing with “not enough uranium from mines to go around.” In particular, US production of uranium “peaked”about 1980.

For many years, the US was able to down-blend nuclear warheads (both purchased from Russia and from its own supply) to get around its uranium supply deficit.

Today, the inventory of nuclear warheads has dropped quite low. There are few warheads available for down-blending. This is creating a limit on uranium supply that is only now starting to hit.

Nuclear warheads, besides providing uranium in general, are important for the fact that they provide a concentrated source of uranium-235, which is the isotope of uranium that can sustain a nuclear reaction. With the warhead supply depleting, the US has a second huge problem: developing a way to produce nuclear fuel, probably mostly from spent fuel, with the desired high concentration of uranium-235. Today, Russia is the primary supplier of enriched uranium.

The plan of the US is to use government research grants to kickstart work on new small modular nuclear reactors that will be more efficient than current nuclear plants. These reactors will use a new fuel with a higher concentration of uranium-235 than is available today, except through purchase from Russia.

«

There’s a long laundry list of problems. None seems simple to solve.
unique link to this extract

---

Cold case solved: DNA evidence confirms the identity of a rapist and killer in a case dating back to 1979 • Riverside County District Attorney

»

In 1979, the body of a 17-year-old girl was found dumped in a snowpack off Highway 243 near Banning. Authorities determined she had been raped and bludgeoned to death.

Now, more than 45 years later, using forensic genealogy, the Riverside County Regional Cold Case Homicide Team announced on Nov. 20, 2024, that they have confirmed the identity of the rapist and killer.

On Feb. 9, 1979, Esther Gonzalez was attacked and murdered while walking from her parents’ house in Beaumont to her sister’s house in Banning. Her body was found the next day off Highway 243, south of Poppet Flats Road.

Esther’s body was found after an unidentified man, described by deputies as argumentative, called the Riverside County Sheriff’s Station in Banning to report finding a body, saying he didn’t know if it was a male or female. Five days later, sheriff’s investigators were able to identify the caller as Lewis Randolph “Randy” Williamson and asked him to take a polygraph. He agreed and passed which, at the time, cleared him of any wrongdoing.

Investigators continued to work on this case for years and eventually uploaded a semen sample from the crime scene into the Combined DNA Index System (CODIS). In 2023, members of the cold case homicide team sent various items of evidence to Othram, Inc. in Texas, initiating a Forensic Investigative Genetic Genealogy investigation, in hopes of developing additional leads. Earlier this year, a crime analyst assigned to the cold case team determined that, although Williamson was seemingly cleared by the polygraph in 1979, he was never cleared through DNA because the technology had not yet been developed.

Williamson died in Florida in 2014.

«

If you needed any more evidence that polygraphs (aka lie detectors) aren’t reliable, there you go.
unique link to this extract

---

2022: Police used a baby’s DNA to investigate its father for a crime • WIRED

Emily Mullin, in 2022:

»

If you were born in the United States within the last 50 or so years, chances are good that one of the first things you did as a baby was give a DNA sample to the government. By the 1970s, states had established newborn screening programs, in which a nurse takes a few drops of blood from a pinprick on a baby’s heel, then sends the sample to a lab to test for certain diseases. Over the years, the list has grown from just a few conditions to dozens.

The blood is supposed to be used for medical purposes—these screenings identify babies with serious health issues, and they have been highly successful at reducing death and disability among children. But a public records lawsuit filed last month in New Jersey suggests these samples are also being used by police in criminal investigations. The lawsuit, filed by the state’s Office of the Public Defender and the New Jersey Monitor, a nonprofit news outlet, alleges that state police sought a newborn’s blood sample from the New Jersey Department of Health to investigate the child’s father in connection with a sexual assault from the 1990s.

Crystal Grant, a technology fellow at the American Civil Liberties Union, says the case represents a “whole new leap forward” in the misuse of DNA by law enforcement. “It means that essentially every baby born in the US could be included in police surveillance,” she says.

It’s not known how many agencies around the country have sought to use newborn screening samples to investigate crimes, or how often those attempts were successful. But there is at least one other instance of it happening. In December 2020, a local TV station reported that police in California had issued five search warrants to access such samples, and that at least one cold case there was solved with the help of newborn blood. “This increasing overreach into the health system by police to get genetic information is really concerning,” Grant says.

«

I came across this by chance, but it’s a remarkable use of DNA, which now feels like the police’s go-to for solving crimes of all sorts.
unique link to this extract

---

Taiwan gets slammed with 15,000 cyber attacks per second, says digital minister • Tom’s Hardware

Mark Tyson:

»

A government minister asserted that Taiwan experiences four times more cyber attacks than the average country. Earlier this week, Taiwanese Digital Minister Huang Yen-nun (黃彥男) told attendees at the CYBERDAY 2024 Information Security Industry Day in Tainan that hackers attempt to breach Taiwan’s digital defenses an astonishing 15,000 times per second. As well as being 4X the average figure, this digital onslaught is touted as the most intense worldwide.

As far as geopolitics goes, Taiwan is well known to be a political hot potato and a potential flashpoint for a major military conflict in East Asia. Military hardware like planes and ships dance carefully around each other all around the sweet potato-shaped island, but so far (touch wood) have never sparked a serious incident. In contrast, Taiwan is now seen as a “first-level war zone,” in the cyber world.

Huang Yen-nun heads up the Ministry of Digital Affairs, which was only set up in 2022. Revealing these figures to the public might be scary in some ways, but knowing politicians are well aware of the problems and are actively bolstering cybersecurity for government and business should encourage stakeholders.

The Taiwanese government has also reportedly tasked the National Security Bureau with setting up a national cybersecurity response center. Leveraging the country’s top intelligence agency this way, with a structure for information sharing, should also protect both private and government concerns on the island.

In addition to securing data in Taiwan, the above initiative demonstrates that the country is serious about building trust with anyone who works with or trades with the island.

«

Surprising if China isn’t behind it.
unique link to this extract

---

• Why do social networks drive us a little mad? • Why does angry content seem to dominate what we see? • How much of a role do algorithms play in affecting what we see and do online? • What can we do about it? • Did Facebook have any inkling of what was coming in Myanmar in 2016? Read Social Warming, my latest book, and find answers – and more.

---

Errata, corrigenda and ai no corrida: none notified