The Chagos Islands have beautiful seas – but if they change hands, the .io domain could cease to exist. Then what? CC-licensed photo by Foreign, Commonwealth & Development Office 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 11 links for you. Subject to treaty. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.

---

India holds record for internet shutdowns with 771 since 2016 • Rest of World

Ananya Bhattacharya:

»

On September 10, when the northeast Indian state of Manipur announced a five-day internet shutdown in response to student-led protests, citizens were livid — but not surprised.

India has been a leader in internet shutdowns, by a huge margin, for nearly a decade, according to data shared by digital rights watchdog Access Now.

While some authoritarian regimes, such as those in China, North Korea, and Russia systematically censor, surveil, or limit the internet, India’s citizens have relatively free access. But India is unique among democratic countries for its frequent enforcement of blackouts. Between 2016 and 2023, India shut down the internet 771 times, Access Now’s data shows.

Myanmar and Ukraine ranked second on the list of internet shutdowns in 2023 and 2022, respectively. “These are regions obviously embroiled in conflict. And if you look at the margin between India and the second spot — massive,” Namrata Maheshwari, senior policy counsel at Access Now, told Rest of World. “There is no form of an internet shutdown that is proportionate or necessary.”

Reasons for the blackouts in India have included the government’s attempts to control agitation surrounding the Citizenship Amendment Act, suppress the farmers’ protests, and curb cheating during exams, according to Software Freedom Law Center, India’s tracker on internet shutdowns.

The majority of shutdowns in India have historically occurred in Jammu and Kashmir, a region at the center of a decades-long dispute between India, Pakistan, and China. In August 2019, it experienced  552 consecutive days of internet blackout, the world’s longest shutdown in history.

«

The irony – or perhaps explanation – is that India’s population absolutely loves the internet, and being connected. So being cut off this much is one of the most frustrating things they can experience.
unique link to this extract

---

The disappearance of an internet domain • Every

Gareth Edwards:

»

On October 3, the British government announced that it was giving up sovereignty over a small tropical atoll in the Indian Ocean known as the Chagos Islands. The islands would be handed over to the neighboring island country of Mauritius, about 1,100 miles off the southeastern coast of Africa. 

The story did not make the tech press, but perhaps it should have. The decision to transfer the islands to their new owner [by signing a treaty] will result in the loss of one of the tech and gaming industry’s preferred top-level domains: .io.

Whether it’s Github.io, gaming site itch.io, or even Google I/O (which arguably kicked off the trend in 2008), .io has been a constant presence in the tech lexicon. Its popularity is sometimes explained by how it represents the abbreviation for “input/output,” or the data received and processed by any system. What’s not often acknowledged is that it’s more than a quippy domain. It’s a country code top-level domain (ccTLD) related to a nation—meaning it involves politics far beyond the digital world.

…Once this treaty is signed, the British Indian Ocean Territory will cease to exist. Various international bodies will update their records. In particular, the International Standard for Organization (ISO) will remove country code “IO” from its specification. The Internet Assigned Numbers Authority (IANA), which creates and delegates top-level domains, uses this specification to determine which top-level country domains should exist. Once IO is removed, the IANA will refuse to allow any new registrations with a .io domain. It will also automatically begin the process of retiring existing ones. (There is no official count of the number of extant .io domains.)

Officially, .io—and countless websites—will disappear. At a time when domains can go for millions of dollars, it’s a shocking reminder that there are forces outside of the internet that still affect our digital lives.

«

IANA will probably – probably – do something to keep this going, since there’s a lot of money sloshing around in those domains.
unique link to this extract

---

US states sue TikTok, claiming its addictive features harm youth mental health • The Guardian

Johana Bhuiyan, Nick Robins-Early and agencies:

»

More than a dozen states and the District of Columbia filed lawsuits against TikTok on Tuesday, alleging the popular short-form video app is damaging children’s mental health with a product designed to be used compulsively and excessively.

The lawsuits stem from a national investigation into TikTok, which was launched in March 2022 by a bipartisan coalition of attorneys general from several states, including California, Kentucky and New Jersey. All of the complaints were filed in state courts and claim that TikTok’s algorithm is especially dangerous given the platform’s widespread use among young people and its ability to deliver quick hits of dopamine. Design choices such as infinite scrolling, push notifications and in-app purchases prey on youth and create addictive habits among users, prosecutors allege. There are over 170m monthly active TikTok users in the US, and over a billion worldwide.

At the heart of each lawsuit is the TikTok algorithm, which powers what users see on the platform by populating the app’s main “For You” feed with content tailored to people’s interests.

In its filings, the District of Columbia called the algorithm “dopamine-inducing”, and said it was created to be intentionally addictive so the company could trap many young users into excessive use and keep them on its app for hours on end. TikTok does this despite knowing that these behaviors will lead to “profound psychological and physiological harms”, such as anxiety, depression, body dysmorphia and other long-lasting problems, the complaint said.

«

Popcorn time, though probably won’t reach a court until some time in 2025, and properly until 2026. There’s still a US ban due in January 2025 unless ByteDance sells the US division to a US owner.
unique link to this extract

---

Starlink was offered for free to those hit by Hurricane Helene. It is not actually free • The Register

Brandon Vigliarolo:

»

The free Starlink service Elon Musk and SpaceX so graciously promised for communities devastated by Hurricane Helene in the US is not actually entirely free, according to those living in the aftermath – and the satellite operator’s own signup page.

There is a significant caveat: you are still expected to foot the bill for the hardware.

Starlink’s Twitter account declared last week, in a post with tens of millions of views, that “Starlink is now free for 30 days.” The world’s richest man, with a net worth of approximately $260bn, followed up by saying, in quite the PR coup, that all Starlink terminals would now work automatically “without [the] need for payment in the areas affected by Hurricane Helene.” 

But try to sign up for the ostensibly “free” service in an area Starlink has designated as a Helene disaster zone, and surprise: You still have to pay for the terminal (normally $350, but reportedly discounted to $299 for disaster relief, though that’s not reflected in Starlink’s signup page), plus shipping and tax, bringing the grand total to just shy of $400.

You can see for yourself in the video: putting in the address of city hall in Boone, North Carolina, one of the areas wrecked by the lethal super storm, shows folks recovering from the disaster are still expected to pay hundreds for that that free, month-long Starlink service. Though better than nothing at all, it is not quite the humanitarian aid it was promoted and heralded as.

«

There’s also the question of being signed up to an auto-renewing contract – so one needs to cancel within 30 days.
unique link to this extract

---

How to delete Your 23andme data • Lifehacker

Beth Skwarecki:

»

So far there is no definitive word about the company being sold, with or without its data. However, it’s reasonable to expect that the company could be sold, and that owner could inherit the data. Something similar happened when MyHeritage bought Promethease, another DNA analysis company, in 2020.

Currently, your data may already be shared with other companies. If you signed up to participate in research studies through 23andme, “de-identified” data about you (including genetic data) has likely been given to research institutions and pharma companies. For example, 23andme has a data licensing agreement with GSK (formerly GlaxoSmithKline) to use the 23andme database to “conduct drug target discovery and other research.” 

This isn’t a possible future scenario, but rather the current operation of the business. Licensing agreements like these are a big part of how 23andme makes money. Or intends to make money. Or possibly once made money. They’re not doing so great at the money making thing these days. 

Deleting your 23andme data doesn’t necessarily withdraw it from studies, especially since the data was “de-identified,” that is, stripped from your name and personal information. It does mean that your data will not be used in future research projects.

«

OK, but how do you delete your data?

»

If you would like to keep any of your data, sign in to your account and visit your user settings page. (You can also opt out of research studies there.). Click View on the 23andme Data card.

You’ll be asked to enter your date of birth to confirm your identity. In theory this is where you can download your data, but I can’t test this—I have a 23andme account, but I must have given the company a fake date of birth all those years ago. The website just tells me to contact Customer Care.

«

Well, huh. Kind of sold us a pup there, Lifehacker.
unique link to this extract

---

Monday, October 7 2024 • Scripting News

Dave Winer has hit 30 years of blogging:

»

Blogging started out as a programming adventure and eventually became a form of literature. How about that. I’m up for doing more of that if you all are. But please expect to make contributions, don’t expect it all to come to you for free, because as we know nothing really is free. #

Today’s the big day. Thanks to John Naughton’s wonderful piece in the Guardian, I’m hearing from people all over the world about what blogging means to them. I appreciate all of the messages, but would appreciate them even more if they were on your blog. We need to keep using the tech. Blogging is kind of lost, and I would like to see that change. Every time you post something you’re proud of on a social media site, how about taking a moment and posting it to your blog too. And while there, if appropriate, link to something from some part of your post, even though the social media sites don’t support linking, the web is still there and it still does. #

Interestingly, the clock at the bottom of the nightly emails does not agree with the clock on the home page of Scripting News. It’s a hard thing to test in real life. And it’s completely fitting, given the motto of the blog is: it’s even worse than it appears, which could be the motto of all programmers everywhere, and probably bloggers too. We always focus on the bad news, of course — that’s human nature — but always remember, it could actually be worse. #

«

Is blogging going to make a comeback? There are many, many more ways to express oneself nowadays. Is Substack blogging? (I think so.) Is social media? (Arguably not, too difficult to roll back through time.)
unique link to this extract

---

What happens if a hurricane smashes Tampa, Florida? • The Big Newsletter

Matt Stoller:

»

Tampa hasn’t been hit directly by a big storm since 1921, and it has grown a lot since then.

Hopefully, [Hurricane] Milton doesn’t hit the city directly, and it may not. These big storms almost always tend to avoid the most catastrophic hits. But let’s go over some of the consequences if it does. First, Tampa has an important port, managing 33 million tons of cargo a year. It’s the biggest exporter of fertilizer in America, and is the biggest importer of gasoline and jet fuel used in Florida. So that means we can expect significant supply shocks, and probably environmental damage. Utilities are already stretched because of Helene, so replacing electrical equipment is going to be difficult. All the major supplies for recovery, everything from lumber to ice to drinking water to skilled labor, are already being sucked into North Carolina to deal with the after-effects of Helene.

In addition, MacDill Air Force Base is in Tampa, which is where CENTCOM, the command center for U.S. forces in the Middle East, is located. So we could see modest disruptions to U.S. military operations. Tampa is near important NASA assets like Cape Canaveral, so there are space and defense contractors in the city.

There’s a lot more than that, of course, since Tampa is a major metropolitan area, an important hub for fishing, tourism, medicine, manufacturing and finance. It has convenient rail lines and highways to pair with its deepwater port, serving as a trans-shipment point for moving goods throughout Florida. Beyond Tampa, a good chunk of Florida is in the path of this storm, with unpredictable consequences. For instance, I wrote earlier about the shortage of IV fluids due to problems in North Carolina. It turns out another large IV solutions manufacturing plant in Daytona Beach is in the path of Milton. Yikes.

The insured losses could be massive, and we could face shortages of all sorts of random and important stuff. But more than these elements, we might lose an entire city, an apocalyptic level of destruction. And increasingly in Florida, there is no way to insure anything. In fact, the state itself, through its Citizens Property Insurance Corp., self-insures against natural disasters, because private insurers just won’t do business in Florida anymore. That means Florida property owners – who in aggregate own about $4 trillion – could enter a death spiral where they can’t get insurance, and so can’t get financing.

«

Milton is expected to make landfall on Thursday morning, UK time. Fingers crossed.
unique link to this extract

---

The connected TV industry’s unprecedented “surveillance” • Ars Technica

Scharon Harding:

»

The companies behind the streaming industry, including smart TV and streaming stick manufacturers and streaming service providers, have developed a “surveillance system” that has “long undermined privacy and consumer protection,” according to a report from the Center for Digital Democracy (CDD) published today and sent to the Federal Trade Commission (FTC). Unprecedented tracking techniques aimed at pleasing advertisers have resulted in connected TVs (CTVs) being a “privacy nightmare,” according to Jeffrey Chester, report co-author and CDD executive director, resulting in calls for stronger regulation.

The 48-page report, How TV Watches Us: Commercial Surveillance in the Streaming Era [PDF], cites Ars Technica, other news publications, trade publications, blog posts, and statements from big players in streaming—from Amazon to NBCUniversal and Tubi, to LG, Samsung, and Vizio. It provides a detailed overview of the various ways that streaming services and streaming hardware target viewers in newfound ways that the CDD argues pose severe privacy risks. The nonprofit composed the report as part of efforts to encourage regulation. Today, the CDD sent letters to the FTC [PDF], Federal Communications Commission (FCC), California attorney general [PDF], and California Privacy Protection Agency (CPPA) [PDF], regarding its concerns.

“Not only does CTV operate in ways that are unfair to consumers, it is also putting them and their families at risk as it gathers and uses sensitive data about health, children, race, and political interests,” Chester said in a statement.

…The report notes “misleading” privacy policies that have minimal information on data collection and tracking methods and the use of marketing tactics like cookie-less IDs and identity graphs that make promises of not collecting or sharing personal information “meaningless.”

“As a consequence, buying a smart TV set in today’s connected television marketplace is akin to bringing a digital Trojan Horse into one’s home,” it says.

«

unique link to this extract

---

Machine learning pioneers win Nobel prize in physics • The Guardian

Ian Sample:

»

Two researchers who helped lay the foundations for modern artificial intelligence – although one later warned of its potential harms – have been awarded the 2024 Nobel prize in physics.

Inspired by the workings of the brain, John Hopfield, a US professor emeritus at Princeton University, and Geoffrey Hinton, a British-Canadian professor emeritus at the University of Toronto, built artificial neural networks that store and retrieve memories like the human brain, and learn from information fed into them.

Hinton, 76, who is often called “the godfather of AI”, made headlines last year when he quit Google and warned about the dangers of machines outsmarting humans.

The scientists’ pioneering work began in the 1980s and demonstrated how computer programs that draw on neural networks and statistics could form the basis for an entire field, which paved the way for swift and accurate language translation, facial recognition systems, and the generative AI that underpins chatbots such as ChatGPT, Gemini and Claude.

Hopfield, 91, was honoured for building “an associative memory that can store and reconstruct images and other types of patterns in data”, while Hinton invented a method that can “independently discover properties in data”, an important feature of the large artificial neural networks in use today.

In 1982, Hopfield built a neural network that stored images and other information as patterns, mimicking the way memories are stored in the brain. The network was able to recall images when prompted with similar patterns, akin to identifying a song heard only briefly in a noisy bar.

Hinton built on Hopfield’s research by incorporating probabilities into a multilayered version of the neural network, leading to a program that could recognise, classify and even generate images after being fed a training set of pictures.

«

Doesn’t feel like physics, though, does it. Are we in The Three-Body Problem? Has physics stopped?
unique link to this extract

---

How the US lost the solar power race to China • Bloomberg

David Fickling:

»

Washington blames China’s dominance of the solar industry on what are routinely dubbed “unfair trade practices.” But that’s just a comforting myth. China’s edge doesn’t come from a conspiratorial plot hatched by an authoritarian government. It hasn’t been driven by state-owned manufacturers, subsidized loans to factories, tariffs on imported modules or theft of foreign technological expertise. Instead, it’s come from private businesses convinced of a bright future, investing aggressively and luring global talent to a booming industry — exactly the entrepreneurial mix that made the US an industrial powerhouse.

The fall of America as a solar superpower is a tragedy of errors where myopic corporate leadership, timid financing, oligopolistic complacency and policy chaos allowed the US and Europe to neglect their own clean-tech industries. That left a yawning gap that was filled by Chinese start-ups, sprouting like saplings in a forest clearing. If rich democracies are playing to win the clean technology revolution, they need to learn the lessons of what went wrong, rather than just comfort themselves with fairy tales.

To understand what happened, I visited two places: Hemlock, Michigan, a tiny community of 1,408 people that used to produce about one-quarter of the world’s PV-grade polysilicon, and Leshan, China, which is now home to some of the world’s biggest polysilicon factories. The similarities and differences between the towns tell the story of how the US won the 20th century’s technological battle — and how it risks losing its way in the decades ahead.

… the core questions are often almost impossible to answer. Is [China’s] Tongwei’s cheap electricity from a state-owned utility a form of government subsidy? What about [US company] Hemlock’s tax credits protecting it from high power prices? Chinese businesses can often get cheap land in industrial parks, something that’s often considered a subsidy. But does zoning US land for industrial usage count as a subsidy too? Most countries have tax credits for research and development and compete to lower their corporate tax rates to encourage investment. The factor that determines whether such initiatives are considered statist industrial policy (bad), or building a business-friendly environment (good), is usually whether they’re being done by a foreign government, or our own.

«

Long piece, but worthwhile.
unique link to this extract

---

Threads knows it has an engagement bait problem • The Verge

Jess Weatherbed:

»

If you’ve noticed more engagement bait appearing across your Threads feed, you’re not alone. Meta is aware of the issue and looking at how to address it, according to Instagram boss Adam Mosseri. “We’ve seen an increase in engagement-bait on Threads and we’re working to get it under control,” Mosseri said on Threads in response to comments flagging the issue.

Engagement bait on Threads typically covers posts with banal questions or invites for open-ended discussions to encourage other users to interact. Because Threads, like Instagram, pushes users to see an algorithmic feed of posts by default, getting more interactions can snowball a simple post into virality.

When asked about comparisons between Threads and X in a recent Decoder interview, Meta CEO Mark Zuckerberg alluded to the fact that comment engagement was a fundamental aspect of a “very good discussion-oriented platform.”

“Not all comments or replies are good,” said Mosseri. “Mark’s comment is more about the Twitter pioneering a format where the reply can be elevated, which is a good thing, but that doesn’t mean that every reply should be.” The thread that Mosseri replied to cites one example of a bait post, featuring a seemingly AI-generated image paired with an incendiary take on politeness that had pulled in more than 17,000 responses.

«

You’d think – you’d think! – that by now Meta would know what does and doesn’t work for engagement, and what reaction clickbait would get. It’s hard to shake a tiny suspicion that Meta doesn’t mind the wild engagement for now because it keeps people on the site.
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