Smart TVs have turned their makers from hardware specialists into advertising conduits. CC-licensed photo by Keith Williamson 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 10 links for you. Not an advert. I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. Observations and links welcome.
How Google Search ranking works • Search Engine Land
Mario Fischer:
»
The structure of organic search results is now so complex – not least due to the use of machine learning – that even the Google employees who work on the ranking algorithms say they can no longer explain why a hit is at one or two. We do not know the weighting of the many signals and the exact interplay.
Nevertheless, it is important to familiarize yourself with the structure of the search engine to understand why well-optimized pages do not rank or, conversely, why seemingly short and non-optimized results sometimes appear at the top of the rankings. The most important aspect is that you need to broaden your view of what is really important.
All the information available clearly shows that. Anyone who is even marginally involved with ranking should incorporate these findings into their own mindset. You will see your websites from a completely different point of view and incorporate additional metrics into your analyses, planning and decisions.
To be honest, it is extremely difficult to draw a truly valid picture of the systems’ structure. The information on the web is quite different in its interpretation and sometimes differs in terms, although the same thing is meant.
An example: The system responsible for building a SERP (search results page) that optimizes space use is called Tangram. In some Google documents, however, it is also referred to as Tetris, which is probably a reference to the well-known game.
Over weeks of detailed work, I have viewed, analyzed, structured, discarded and restructured almost 100 documents many times.
This article is not intended to be exhaustive or strictly accurate. It represents my best effort (i.e., “to the best of my knowledge and belief”) and a bit of Inspector Columbo’s investigative spirit. The result is what you see here.
«
This is not brief. But also, you’ll end up a bit stunned by how much goes into a search ranking.
unique link to this extract
Sonos considers relaunching its old app • The Verge
Chris Welch:
»
Sonos has explored the possibility of rereleasing its previous mobile app for Android and iOS — a clear sign of what an ordeal the company’s hurried redesign has become. The Verge can report that there have been discussions high up within Sonos about bringing back the prior version of the app, known as S2, as the company continues toiling away at improving the performance and addressing bugs with the overhauled design that rolled out in May to a flood of negative feedback. (The new Sonos app currently has a 1.3-star review average on Google Play.)
Letting customers fall back to the older software could ease their frustrations and reduce at least some of the pressure on Sonos to rectify every issue with the new app.
«
There is a lesson here. Sonos did a complete ground-up rewrite and changed the UI. So, interestingly, did Marco Arment of his very popular Overcast podcast app. “My business is on fire,” Arment declared a couple of weeks after the release: people hated the new UI and they hated the fact that elements they used all the time had been moved to somewhere illogical.
Arment wailed that moving those elements back was going to spoil his lovely UI redesign. But! He was prepared to do it, because the customer is right. Arment is a one-man band, and while he isn’t going back to the old app, it’s a lot easier for him to get approval to change things.
Sonos burnt its bridges and it’s hard to turn the supertanker around (so to speak). Meanwhile, it’s just let 100 people – about 6% of its workforce – go. Hardware is hard.
unique link to this extract
UK tech entrepreneur Mike Lynch among missing in Sicily yacht sinking • The Guardian
Jamie Grierson and Lorenzo Tondo:
»
The British tech entrepreneur Mike Lynch is missing after a superyacht sank off the coast of Sicily during a violent storm.
The British-flagged Bayesian, a 56-metre sailboat, was carrying 22 people and anchored just off shore near the port of Porticello when it was hit by a tornado in the early hours, the Italian coastguard said in a statement.
One man, understood to be the vessel’s chef, was confirmed dead and six others, including Lynch and his 18-year-old daughter, Hannah, remained unaccounted for on Monday evening. The coastguard said the missing had British, American and Canadian nationalities.
Fifteen people were rescued, including Lynch’s wife, Angela Bacares, who owned the boat, and a one-year-old girl who was saved by her mother.
A spokesperson for Lynch declined to comment. A spokesperson for the UK Foreign Office said: “We are providing consular support to a number of British nationals and their families following an incident in Sicily, and are in contact with the local authorities.”
Lynch co-founded Autonomy, a software firm that became one of the shining lights of the UK tech scene, in the mid-90s. Once described as Britain’s Bill Gates, Lynch spent much of the last decade in court defending his name against allegations of fraud related to the sale of Autonomy to the US tech company Hewlett-Packard for $11bn.
The 59-year-old was acquitted by a jury in San Francisco in June, after he had spent more than a year living in effect under house arrest. Upon his acquittal, he told reporters: “I am looking forward to returning to the UK and getting back to what I love most: my family and innovating in my field.”
«
This has all the beats of a Greek tragedy. But as real life.
unique link to this extract
OpenAI’s new voice mode let me talk with my phone, not to it • TechCrunch
Maxwell Zeff:
»
I’ve been playing around with OpenAI’s Advanced Voice Mode for the last week, and it’s the most convincing taste I’ve had of an AI-powered future yet. This week, my phone laughed at jokes, made them back to me, asked me how my day was, and told me it’s having “a great time.” I was talking with my iPhone, not using it with my hands.
OpenAI’s newest feature, currently in a limited alpha test, doesn’t make ChatGPT any smarter than it was before. Instead, Advanced Voice Mode (AVM) makes it friendlier and more natural to talk with. It creates a new interface for using AI and your devices that feels fresh and exciting, and that’s exactly what scares me about it. The product was kinda glitchy, and the whole idea totally creeps me out, but I was surprised by how much I genuinely enjoyed using it.
Taking a step back, I think AVM fits into OpenAI CEO Sam Altman’s broader vision, alongside agents, of changing the way humans interact with computers, with AI models front and center.
…On Wednesday, I tested the most tremendous upside for this advanced technology I could think of: I asked ChatGPT to order Taco Bell the way Obama would.
“Uh, let me be clear — I’d like a Crunchwrap Supreme, maybe a few tacos for good measure,” said ChatGPT’s Advanced Voice Mode. “How do you think he’d handle the drive-thru?” said ChatGPT, then laughing at its own joke.
«
This is my unimpressed face. I know everyone’s saying this is the next step for smartphones, but seriously, I’m fine with a device that can interpret commands. Maybe it’ll be like TARS in Interstellar, where you can fine-tune the humour level.
unique link to this extract
Your TV set has become a digital billboard. And it’s only getting worse • Ars Technica
Scharon Harding:
»
Over the past few years, TV makers have seen rising financial success from TV operating systems that can show viewers ads and analyze their responses. Rather than selling as many TVs as possible, brands like LG, Samsung, Roku, and Vizio are increasingly, if not primarily, seeking recurring revenue from already-sold TVs via ad sales and tracking.
How did we get here? And what implications does an ad- and data-obsessed industry have for the future of TVs and the people watching them?
Success in the TV industry used to mean selling as many TV sets as possible. But with smart TVs becoming mainstream and hardware margins falling, OEMs have sought new ways to make money. TV OS providers can access a more frequent revenue source at higher margins, which has led to a viewing experience loaded with ads. They can be served from the moment you pick up your remote, which may feature streaming service ads in the form of physical buttons.
Some TV brands already prioritize data collection and the ability to sell ads, and most are trying to boost their appeal to advertisers. Smart TV OSes have become the cash cow of the TV business, with providers generating revenue by licensing the software and through revenue sharing of in-app purchases and subscriptions.
A huge part of TV OS revenue comes from selling ads, including on the OS’s home screen and screensaver and through free, ad-supported streaming television channels. GroupM, the world’s largest media investment company, reported that smart TV ad revenue grew 20% from 2023 to 2024 and will grow another 20% to reach $46bn next year. In September 2023, Patrick Horner, practice leader of consumer electronics at analyst Omdia, reported that “each new connected TV platform user generates around $5 per quarter in data and advertising revenue.”
«
Hardware is hard; advertising seems to be eating the world, or at least all its available attention. This isn’t really what we wanted when we got smart TVs, but the connectedness means that the makers refuse to let go of their device. So we see smart TVs showing advertising and car manufacturers holding back capabilities such as heated seats unless you pay them a subscription.
unique link to this extract
Humane’s daily returns are outpacing sales • The Verge
Kylie Robison:
»
Shortly after Humane released its $699 AI Pin in April, the returns started flowing in.
Between May and August, more AI Pins were returned than purchased, according to internal sales data obtained by The Verge. By June, only around 8,000 units hadn’t been returned, a source with direct knowledge of sales and return data told me. As of today, the number of units still in customer hands had fallen closer to 7,000, a source with direct knowledge said.
At launch, the AI Pin was met with overwhelmingly negative reviews. Our own David Pierce said it “just doesn’t work,” and Marques Brownlee called it “the worst product” he’s ever reviewed. Now, Humane is attempting to stabilize its operations and maintain confidence among staff and potential acquirers. The New York Times reported in June that HP is considering purchasing the company, and The Information reported last week that Humane is negotiating with its current investors to raise debt, which could later be converted into equity.
Humane’s AI Pin and accessories have brought in just over $9m in lifetime sales, according to the internal data seen by The Verge. But around 1,000 purchases were cancelled before shipping, and more than $1m worth of product has been returned.
«
Remarkable that there are that many people still finding them worthwhile. Or maybe they dropped them down the back of the sofa and forgot about them?
Anyway: it’s dead, Jim. Hardware is so, so difficult.
unique link to this extract
Bots on Twitter/X run wild amid global elections • Rest of World
Russell Brandom:
»
As Rwanda prepared for its national election on July 15, something strange was happening on X. Hundreds of accounts appeared to be operating in unison, posting identical or oddly similar messages in support of incumbent president Paul Kagame. A team of researchers at Clemson University started tracking the seemingly automated network and discovered more than 460 accounts involved, sharing what appeared to be AI-generated messages.
“The campaign, which exhibits several markers of coordinated inauthentic behavior, seems to be trying to affect discourse about the performance of the Kagame regime,” the researchers wrote in a paper tracing the network.
It was the kind of revelation that would normally send moderators scrambling, particularly in the weeks before a national election. But when the group reported their findings to X, nothing happened. Flagged accounts remained up, and the network continued to post.
It was a stunning result, given the sensitivity of the country’s election and how easy it would have been to stop the network in its tracks. “It’s obvious,” researcher Morgan Wack, who led the Clemson project, told Rest of World. “If you’re paying attention at all, it’s very clear you could take down several of these accounts. There’s just no effort to take down any of this.”
Wack’s experience is part of a larger shift in moderation on X, which has opened the door to influence operations around the world. In the two years since Elon Musk took ownership of the company, its trust and safety team has been decimated — and the result has been a steady drip of networks like the one uncovered in Rwanda. Simply finding the networks is harder than ever, as researchers must do without API access and face growing legal risks after their work is published.
«
Musk utterly doesn’t care. He’s got no interest in anything outside the US, and even there he only cares about his stupid little obsessions.
unique link to this extract
Sam Altman’s Worldcoin is battling with governments over your eyes • WSJ
Angus Berwick:
»
Sam Altman wants to save us from the AI-dominated world he is building. The trouble is, governments aren’t buying his plan, which involves an attempt to scan the eyeballs of every person on Earth and pay them with his own cryptocurrency.
Altman’s OpenAI is creating models that may end up outsmarting humans. His Worldcoin initiative says it is addressing a key risk that could follow: we won’t be able to tell people and robots apart.
But Worldcoin has come under assault by authorities over its mission. It has been raided in Hong Kong, blocked in Spain, fined in Argentina and criminally investigated in Kenya. A ruling looms on whether it can keep operating in the European Union.
More than a dozen jurisdictions have either suspended Worldcoin’s operations or looked into its data processing. Among their concerns: How does the Cayman Islands-registered Worldcoin Foundation handle user data, train its algorithms and avoid scanning children?
…Worldcoin verifies “humanness” by scanning irises using a basketball-sized chrome device called the Orb. Worldcoin says irises, which are complex and relatively unchanging in adults, can better distinguish humans than fingerprints or faces.
Users receive immutable codes held in an online “World ID” passport, to use on other platforms to prove they are human, plus payouts in Worldcoin’s WLD cryptocurrency.
Worldcoin launched last year and says it has verified more than six million people across almost 40 countries. Based on recent trading prices, the total pool of WLD is theoretically worth some $15bn.
The project says its technology is completely private: Orbs delete all images after verification, and iris codes contain no personal information—unless users permit Worldcoin to train its algorithms with their scans. Encrypted servers hold the anonymised codes and images.
«
It’s not scanned a lot of people, to be honest. But also: if the worry is that superhuman AI will pose as humans, why wouldn’t the superhuman AI just hack into the servers? It makes no sense. (Thanks Karsten for the link.)
unique link to this extract
US wind and solar on track to overtake coal this year • Microgrid Media
Jonas Muthoni:
»
For the first time in history, wind and solar energy have generated more electricity than coal in the United States through July. This shift marks a significant milestone as these renewable resources continue their upward trajectory in the power generation sector.
According to federal data, the combined output of wind and solar was 393 terawatt-hours (TWh), slightly ahead of coal’s 388 TWh. This development is particularly notable as it does not include other renewable sources like hydropower, which have also seen substantial usage in past comparisons.
Factors influencing the shift:
• Increase in solar output: solar facilities saw a 36% increase in production over the previous year, reaching 118 TWh.
• Steady growth in wind energy: wind energy production also increased by 8%, totaling 275 TWh.
• Decline of coal: The decline in coal is attributed to the closure of numerous coal plants and an overall decrease in its economic viability.«
And yet I do come across people who insist that wind and solar are no use and that we should be piling in to the North Sea for its oil and gas reserves rather than onshore wind or solar. Hard to argue with “logic” like that.
unique link to this extract
When will climate change turn life in the US upside down? • Yale Climate Connections
Jeff Masters is a former meteorologist:
»
In the US, the most likely major economic disruption from climate change over the next few years might well be a collapse of the housing market in flood-prone and wildfire-prone states. Billion-dollar weather disasters — which cause about 76% of all weather-related damages — have steadily increased in number and expense in recent years and would be even worse were it not for improved weather forecasts and better building codes. The recent increase in weather-disaster losses has brought on an insurance crisis — especially in Florida, Louisiana, California, and Texas — which threatens one of the bedrocks of the U.S. economy, the housing and real estate market.
In California, the insurer of last resort, the FAIR plan, had only about $250m in cash on hand as of March 2024.
“One major fire near Lake Arrowhead, where the Plan holds $8 billion in policies, would plunge the whole scheme into insolvency,” observed Harvard’s Susan Crawford, author of “Charleston: Race, Water, and the Coming Storm.”
It is widely acknowledged that higher weather disaster losses result primarily from an increase in exposure: more people with more stuff moving into vulnerable places, including those at risk of floods. Martin Bertogg, Swiss Re’s head of catastrophic peril, said in a 2022 AP interview that two-thirds, perhaps more, of the recent rise in weather-related disaster losses is the result of more people and things in harm’s way.
But this balance will likely shift in the coming decades. Increased exposure will continue to drive increased weather disaster losses, but the fractional contribution of climate change to disaster losses — at least for wildfire, hurricane, and flood disasters — is likely to increase rapidly, making the insurance crisis accelerate.
«
Long essay, but interesting. He thinks there’s another 15 years or so before things go haywire.
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
The Overspill: when there's more that I want to say 