If you want to buy a Hermes Birkin bag, be prepared to be treated with suspicion by the company itself. And don’t wear the wrong watch. CC-licensed photo by Yvette Ilagan 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. Do parrots like them though? I’m @charlesarthur on Twitter. On Threads: charles_arthur. On Mastodon: https://newsie.social/@charlesarthur. On Bluesky: @charlesarthur.bsky.social. Observations and links welcome.
How WhatsApp took over the global conversation • The New Yorker
Sam Knight:
»
British politics and, arguably, the British state are coördinated by WhatsApp. Ninety-two% of U.K. internet users are on the platform. Police officers banter on it. The National Health Service relies on it. On the afternoon of March 13, 2020—ten days before the U.K. entered its first covid lockdown—Dominic Cummings, a senior adviser to Prime Minister Boris Johnson, formed a five-man WhatsApp group that came to more or less run the country.
That fall, a reporter from the Daily Mail asked a government spokesperson, via WhatsApp, whether it was true that national policies were being conceived this way. The spokesperson WhatsApped Simon Case, the country’s most senior civil servant, with a suggested response: “the PM does not make government decisions via WhatsApp.” Case replied on WhatsApp less than a minute later: “Erm—is that true? I am not sure it is. I think we will have to ignore.”
…WhatsApp is phatic before it is anything else. It is an architecture of presence. It winks with life, informing you who is online and when they were last seen. Tiny bundles of data—relayed on the app’s servers through sockets, or continuous connections—tell you that your best friend is typing. Koum introduced “read receipts,” to show that texts were being sent and seen. At first, he imagined miniature icons that would represent a message’s odyssey through the network—showing servers and hard drives—but Borzov suggested something simpler: one check mark to show that WhatsApp had received your message and two to show that it had been delivered. When the message was opened, the check marks turned blue.
Blue check marks have saved some lives (WhatsApp is often the platform of choice for disaster responders) and tested many relationships. Whether to respond to a message that someone knows you have read with a heart, a thumbs-up, or a crying-face emoji is a modern-day imponderable, although I’m pretty sure that Malinowski would have taken a hard line on the subject. (In phatic conversation, he notes, “taciturnity means not only unfriendliness but directly a bad character.”)
…Sociologists who study WhatsApp family groups sometimes call them W.F.G.s. In 2023, Galit Alkobi and Natalia Khvorostianov, of the Ben-Gurion University of the Negev, published a study of Israeli W.F.G.s and suggested that there were three archetypal roles in these groups: kin-keepers, who are committed to online family life; flickerers, who are seemingly indifferent; and silent warm experts, who are problem solvers. We all know who we are. Alkobi conducted forty-three interviews with family members about their W.F.G.s and found that groups encompassing three generations showed extremely similar traits: problematic-discourse avoidance, an exaggerated writing style (exuberant celebrations, morose commiserations), and routine ejections.
«
And there’s plenty more. For the American readers of the New Yorker, of course, all this talk of WhatsApp is puzzling: they tend to use iMessage because the iPhone predominates, or possibly Signal. But now, as Knight points out, it’s growing fast, and there are more than 100 million users there.
unique link to this extract
Grok was finally updated to stop undressing women and children, X Safety says • Ars Technica
Ashley Belanger:
»
Late Wednesday, X Safety confirmed that Grok was tweaked to stop undressing images of people without their consent.
“We have implemented technological measures to prevent the Grok account from allowing the editing of images of real people in revealing clothing such as bikinis,” X Safety said. “This restriction applies to all users, including paid subscribers.”
The update includes restricting “image creation and the ability to edit images via the Grok account on the X platform,” which “are now only available to paid subscribers. This adds an extra layer of protection by helping to ensure that individuals who attempt to abuse the Grok account to violate the law or our policies can be held accountable,” X Safety said. Additionally, X will “geoblock the ability of all users to generate images of real people in bikinis, underwear, and similar attire via the Grok account and in Grok in X in those jurisdictions where it’s illegal,” X Safety said.
X’s update comes after weeks of sexualized images of women and children being generated with Grok finally prompting California Attorney General Rob Bonta to investigate whether Grok’s outputs break any US laws.
In a press release Wednesday, Bonta said that “xAI appears to be facilitating the large-scale production of deepfake nonconsensual intimate images that are being used to harass women and girls across the Internet, including via the social media platform X.” Notably, Bonta appears to be as concerned about Grok’s standalone app and website being used to generate harmful images without consent as he is about the outputs on X.
Before today, X had not restricted the Grok app or website. X had only threatened to permanently suspend users who are editing images to undress women and children if the outputs are deemed “illegal content.” It also restricted the Grok chatbot on X from responding to prompts to undress images, but anyone with a Premium subscription could bypass that restriction, as could any free X user who clicked on the “edit” button on any image appearing on the social platform.
«
Marianna Spring, the BBC’s disinformation correspondent, remarked on a podcast that she could see the problem that was coming when Grok’s image-editing was made widely available, and that UK regulator Ofcom should have too – to say nothing of X and Elon Musk. But Musk is incapable of understanding the malice that exists out there, partly because he’s immersed in it; it’s like asking a fish to comment on the water.
unique link to this extract
The Atlantic, Penske, and Vox Media sue Google for adtech antitrust violations • The Verge
Lauren Feiner:
»
Lawsuits seeking damages from Google’s illegal ad tech monopoly are piling up following the Justice Department’s successful antitrust case. Vox Media, The Verge’s parent company, is the latest in a wave of media companies that have filed suit against Google, seeking to be reimbursed for the monopoly profits the tech company allegedly made at publishers’ expense.
“Absent Google’s conduct, Vox Media would be able to make available even more, higher quality impressions for purchase on Vox Media’s webpages and create more high-quality, premium journalism,” Vox Media alleges in its lawsuit, filed Wednesday in the Southern District of New York. The Atlantic, which is owned by Laurene Powell Jobs, filed a similar lawsuit in the same district this week, as did Penske Media, which is an investor in Vox Media and owns brands including Rolling Stone, Billboard, and The Hollywood Reporter. Later on Wednesday, two more publishers — McClatchy Media Company and Condé Nast owner Advance Publications — filed similar lawsuits. Google is also facing lawsuits from ad tech providers like PubMatic and OpenX, some of which testified in the trial about how Google’s dominance shut out competition.
…The publishers claim Google’s dominance lets it “depress prices for publisher inventory below competitive levels”
The lawsuits underscore the ways that the highly technical subject of the 2024 trial impacts an ecosystem of publishers and tech providers. For example, the media company complaints outline how Google’s illegal monopoly of the market for publisher ad servers, used to manage ad space for sale on publisher websites, has effectively stymied any viable competitors they might consider moving to.«
Microsoft is closing its employee library and cutting back on subscriptions • The Verge
Tom Warren:
»
Microsoft’s library of books is so heavy that it once caused a campus building to sink, according to an unproven legend among employees. Now those physical books, journals, and reports, and many of Microsoft’s digital subscriptions to leading US newspapers, are disappearing in a shift described inside Microsoft as an “AI-powered learning experience.”
Microsoft started cutting back on its employee subscriptions to news and reports services in November, with some publishers receiving an automated email cancellation of a contract. “This correspondence serves as official notification that Microsoft will not renew any existing contracts upon their respective expiration dates,” reads an email from Microsoft’s vendor management team. “We would like to take this opportunity to express our sincere appreciation for your partnership, collaboration, and continued support throughout our engagement.”
Strategic News Service (SNS), which has provided global reports to Microsoft’s roughly 220,000 employees and executives for more than 20 years, is no longer part of Microsoft’s subscription list. In an email to Microsoft employees that relied on SNS reports, the publisher notes that “Microsoft has just released an automated announcement that all library contracts, of which the SNS Global Report is perhaps the most strategic for your own use, are to be turned off.”
Microsoft employees I’ve spoken to recently have lost access to digital publications like The Information. They’re also unable to perform digital checkouts of business books from the Microsoft Library. While Microsoft often rotates the publishers it uses in its Library service, this time it’s part of a much broader change that seems like corporate cost cutting mixed with the continued push for AI.
«
Steven Sinofsky, who worked at Microsoft through its crucial growth years and led the Office team (among others), called it “a crown jewel of the early days”, which bought every PC book and two copies of any piece of software and would get any magazine or article you wanted, if they didn’t have it.
But don’t worry, CoPilot is here to make them up!
unique link to this extract
My work went from air-conditioned offices to delivering food on a bike. The culture shock is significant • The Guardian
David Rayfield started working as a bike messenger in Melbourne:
»
When you deliver for companies like Uber Eats, no two shifts are the same. I’d say 80% of my deliveries are fast food – a combination of McDonald’s, KFC, Hungry Jacks and coffee. Lots of coffee. My new workplace couldn’t be more different than the grey blur of an office, surrounded by the same walls and same people week after week. The culture shock of going from comfy chairs and morning teas to trying not to get killed in traffic was significant.
After being made redundant four times in six years, months passed with hundreds of unsuccessful job applications. Bills needed to be paid. The big ones were still looming, but the small costs could be covered by delivering for Uber Eats. My Xbox was gathering dust so I sold it at Cash Converters and put the money towards a half-decent mountain bike.
The first thing that hits you riding is the feeling of independence. In between delivering Grey Goose vodka and KFC Zinger burgers, there’s the realisation that it’s just you out there. After four months, I haven’t talked to anyone at Uber Eats. I signed up on my phone, they sent me a fluoro thermal bag, and off I went. I don’t have to worry about anybody making me redundant because there is no anybody – there’s just me. If I want to spend hours in the park eating lunch, I will. Need more sleep? That’s between me and my doona. Granted, my wages are much lower, but what value can be placed on listening to Wu-Tang Clan with wind in my hair versus dreading another company restructure?
Just to be clear, my hair is secure underneath a helmet. Which was good for when I collided with a car and ended up with a face full of road. There are lots of bike paths in Melbourne, but far too many of them are squeezed in between moving traffic and parked traffic. They’re less than a metre wide and often occupied by wavering cars, massive trucks or in this case, part of a parked Ford Fiesta. Before I could react, a driver had opened their door into the bike lane without looking. I couldn’t do anything except stop my bike dead while my body crashed into the bitumen.
Things like this happen too often.
…I always knew urban planning was designed around cars but now I feel it in my bones. Not only is the amount of space dedicated to cars immense, society focuses on cars first. Shopfronts and advertisements face the road, hundreds of car parks take up thousands of metres and I avoid countless potholes in bike lanes while the cars right next to me drive on pristine asphalt.
«
The third audience is generative AI • Dries Buytaert
Dries Buytaert is the founder of Drupal, the website and blogging platform:
»
I used Claude Code to build a new feature for my site this morning. Any URL on my blog can now return Markdown instead of HTML.
I added a small hint in the HTML to signal that the Markdown version exists, mostly to see what would happen. My plan was to leave it running for a few weeks and write about it later if anything interesting turned up.
Within an hour, I had hundreds of requests from AI crawlers, including ClaudeBot, GPTBot, OpenAI’s SearchBot, and more. So much for waiting a few weeks.
For two decades, we built sites for two audiences: humans and search engines. AI agents are now the third audience, and most websites aren’t optimized for them yet.
We learned how to play the SEO game so our sites would rank in Google. Now people are starting to invest in things like Generative Engine Optimization (GEO) and Answer Engine Optimization (AEO).
…The speed of adoption tells me AI agents are hungry for cleaner content formats and will use them the moment they find them. What I don’t know yet is whether this actually benefits me. It might lead to more visibility in AI answers, or it might just make it easier for AI companies to use my content without sending traffic back.
«
That’s the worry, isn’t it? The first audience is real humans. The second is search engines. And now the third one is generative AI, eagerly slurping up anything new or changed. But as Buytaert says, will that bring traffic?
unique link to this extract
Home address, social media checks: how Hermes stalks would-be buyers before (and after) selling a Birkin • NDTV
Dristi Sharma:
»
At first glance, buying a luxury handbag should be simple enough if you have the money. Walk into a boutique, choose what you like, pay, leave. But at Hermes, one of the world’s most powerful luxury houses, money alone rarely guarantees access. Instead, shoppers are quietly inducted into what even loyal clients jokingly call the “Hermes game” a complex, opaque system where patience, perception and behaviour matter as much as spending power.
The conversation around this unwritten game sharpened last week when Hermes hosted an ultra-exclusive private sale in Paris, open only to a shrinking and tightly vetted circle of clients. At the same time, prices for leather goods and ready-to-wear quietly rose, according to Glitz. The timing was telling. This was not a routine price hike, but another signal of how firmly Hermes is tightening control over access, scarcity and status.
…Ironically, while Hermes emphasises “relationships”, clients increasingly find it difficult to build genuine rapport with sales associates. Staff are encouraged to avoid overt familiarity, as relationships deemed too close can trigger managerial suspicion around favouritism or resale risks.
This creates a structural mistrust that stiffens interactions and undermines the warm, personalised experience luxury retail typically promises. From the brand’s perspective, this rigidity serves a purpose. Control is central to the Hermes model. The goal is not merely to sell bags, but to protect the symbolic power of scarcity that surrounds them.
The booming second-hand luxury market and stricter anti-money-laundering regulations have intensified this scrutiny. As one sales associate at a major Paris boutique told Glitz, “Every new client is automatically a suspect.” Staff now collect and assess far more data than before, from home addresses and their perceived prestige to social media activity and online presence. Sales associates are trained to evaluate whether a client’s buying journey appears coherent.
Rapid accumulation of non-quota bags to hit a spending threshold raises red flags, as does shopping across multiple boutiques or countries. Furniture purchases, interestingly, score highly, signalling long-term commitment rather than quick flips. Loyalty to one store, cross-category shopping and a clear alignment with the Hermes universe all work in a client’s favour.
Even subtle signals matter. Wearing an Audemars Piguet or Richard Mille watch is read positively, while a flashy Rolex may be judged ostentatious.
«
Please enjoy this glimpse into another, weirdly suspicious world.
unique link to this extract
How to manipulate prediction markets for your greater good • Polemic Paine
Polemic Paine:
»
It’s 8PM Eastern Time on January 14th, 2026. In four hours, a Polymarket bet will resolve.
Earlier today, an account called “mutualdelta”, so fresh it still had that new-car smell, dropped over $160,000 on “Yes” for a US military strike on Iran by end of day. They scaled it to over $240,000 as the day progressed. The odds, which had been languishing around 14%, shot up to 25%. Twitter lost its mind.
Real-time tracking threads. Speculation about insider knowledge. The whole circus.
The market closes at 11:59 PM ET tonight. As I write this, no strike has occurred. The bet is most likely going to zero.
And somewhere, I’m fairly certain, someone is counting their money.
Just not from Polymarket.
«
As you’ll now know, the attack didn’t happen, and so the money was lost. So why make the dramatic money-losing bet? In order to make money somewhere else, as the writer explains. It’s complicated, but makes total sense, and shows why you really shouldn’t pay too much attention to Polymarket (a predictions betting market) – or at least not take it entirely seriously.
unique link to this extract
New UK offshore wind farms could significantly cut power prices • Energy & Climate Intelligence Unit
»
The 8.4GW of offshore wind power announced this week in the Contracts for Difference (CfDs) auction Allocation Round 7 (AR7) is set to boost clean power output, such that – had these wind farms been operating over the last year – gas power generation could have been a third (35%) lower, cutting day-ahead wholesale electricity prices by up to £11 per MWh (13%), down to £72/MWh, according to analysis by the Energy and Climate Intelligence Unit (ECIU).
The price reductions caused by these new renewables would have been on top of the savings delivered by operational wind farms pushing gas off the system, which cut the average price by around £38/MWh in 2025, down from £121/MWh to £83/MWh. [3]
Taken together, these results suggest that, had Britain deployed no wind power over recent decades and had instead relied more on gas, power prices could have been up to £49/MWh (67% i.e. two-thirds) higher in 2025 compared to if renewables had been rolled out faster. [4]
Jess Ralston, Energy Analyst at the Energy and Climate Intelligence Unit, said: “It might not be itemised on bills, but more British renewables squeezing gas off the system has the effect of reducing wholesale power prices, lowering those costs for both industry and households.
“There’s lots of large numbers being bandied around, but the reality is that the offshore wind projects secured today are likely to see levies on bills break-even. And in the event of another gas price spike, which given uncertainty in petro-states worldwide is possible, could see billpayers paid back.”
«
Various 19th century loons have come out of the woodwork at the announcement of this nuclear power station’s worth of wind, which they’re sure could be more cheaply done with gas, ignoring the volatility of gas prices and its greenhouse effects.
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 
Every single classic scam in finance shows up whenever there’s money to be made. I’m sure someone has written on this, but I’d be hard-pressed to find it among all the noise. There’s got to be a good book on it. It’s like popular cryptography where we start with the Caesar Cipher and end up with stuff like RSA. Except here, start with the old process of clipping a little bit of precious metal off coins and end up with manipulating low-quality betting markets.