Articles d'hiver intéressants

[Hugh]

https://brunobertez.com/2025/01/22/editorial-leurope-doit-reaccumuler-de-lor-virer-lagarde-leuro-nest-pas-une-monnaie-souveraine-cest-une-monnaie-satellite-on-vous-trahit-la-resistance-a-trump-est-impossible-sans-souver/

 

Bruno Bertez 

 

"EDITORIAL. L’EUROPE DOIT RÉACCUMULER DE L’OR. VIRER LAGARDE. L’EURO N’EST PAS UNE MONNAIE SOUVERAINE, C’EST UNE MONNAIE SATELLITE. ON VOUS TRAHIT. LA RÉSISTANCE À TRUMP EST IMPOSSIBLE SANS SOUVERAINETÉ MONÉTAIRE."

 

C'est un article intéressant, je suis favorable à l'or, mais je désapprouve le reste de la conclusion. Je le lui (Bruno Bertez) ai dit et il m'a répondu:

 

Révélation

Evitez d'être péremptoire dans des domaines dont la logique vous échappe.

 

😐

[Arturus]

Le monde rêvé des libertariens est notre cauchemar

[Adrian]

How population stratification led to a decade of sensationally false genetic findings

On parlait du livre de Reich sur l'autre topic :

 

Citation

This height example might seem far-fetched, but pretty much exactly what I described actually happened, and it led to a decade-long mess where the field was convinced that Europeans had undergone rapid natural selection on height (and other phenotypes correlated with height like … head circumference) only to learn in 2019 that it was all or nearly all explained by stratification (see Berg et al. and Sohail et al. eLife; or press coverage that concludes “this is a major wake up call … a game changer”). But prior to learning this error, the possibility of selection on head circumference got people speculating what else about the head could be under rapid recent selection. That speculation included an famous opinion piece by esteemed population geneticist David Reich raising concern that genetic analyses may soon reveal substantial biological differences among human populations on traits like intelligence; differences that we as a society were unprepared to grapple with1. Naturally, in some circles, Reich’s cautious and circumscribed warnings that we may eventually find challenging genetic differences were read as a kind of Straussian message, a cryptic admission of precisely the “racist prejudices and agendas” Reich was attempting to head off (and, I should note, that he spent another two chapters in his book explicitly denouncing). Snippets from his editorial were further stripped of context, sometimes reworded entirely, and became meme fodder for open racists: Harvard’s superstar geneticist is secretly on our side, the truth about the inferior races will soon be revealed. And these memes continue to get passed around today, more than five years since the motivating height result was shown to be an artifact (in a paper on which Reich is a corresponding author no less). All of which is to say that poor control for population structure can have, well, some pretty big consequences.

 

 

Citation

For educational attainment, they find no significant genetic correlation with number of children and a point estimate at roughly zero (though with large uncertainty). But for Cognitive Performance / IQ scores, they do find a significantly positive genetic correlation with number of children. That’s right, positive. The same variants that appear to increase cognitive performance also appear to increase the number of offspring (implying, if all of the model assumptions hold up, that higher cognitive performance may actually be under some amount of positive selection). Not only is this the complete opposite of what has been observed in prior analyses from polygenic scores, it also runs counter to the environmental observation. Have we stumbled on a paradox? Not quite, as Beauchamp also noted in a commentary about his own findings:

First, there is nothing paradoxical about my findings. Phenotypes arise from the interplay of genetic and environmental factors, and environmental factors can induce phenotypic changes that run counter to those induced by natural selection. Although the slightly lower fertility of individuals carrying genetic variants associated with higher EA [educational attainment] implies that natural selection has been slowly favoring lower EA, countervailing cultural, economic, policy, and other environmental factors are almost certainly responsible for the vast increase in average EA observed in the past century.

It turns out that this view may have been conceptually correct but directionally wrong. When stratification is better controlled, there appears to be no direct genetic relationship between EA and fertility. And if natural selection is acting at all, it is slowly favoring higher Cognitive Performance. This is not yet a definitive answer — the Tan et al. / LDSC results still come with substantial statistical uncertainty and model assumptions — but it is clear that as we do a better job of addressing stratification, the results can change completely.

 

[Adrian]

 

[Sloonz]

Caplan nous tease son prochain livre, et ça a l’air très prometteur :

 

https://www.betonit.ai/p/capitalism-socialism-and-social-desirability

[Adrian]

Citation

The brutally honest argument against government-funded health care, in contrast, is that the cost of saving a marginal life is too damn high. Spending $100,000 a month to keep an 85-year-old on life support is a terrible use of taxpayer money; a faithful steward of the public interest would pull the plug

 

Prometteur en effet /s

[Tramp]

Case in point. 

J’imagine que tu donnes tout ton argent chaque moi pour donner quelques semaines de plus à vivre à des gens de 90 ans. 

[Cthulhu]

Ça n'a pas l'air très bien argumenté sur ce court extrait (et ça ne m'étonnerait pas que ce soit le cas pour Caplan) mais on romantise beaucoup la vie de personnes âgées.

 

Alzheimer/Parkinson à cet âge, tu ne sais plus où tu es 90% du temps, tu n'est plus forcément en état d'aller aux toilettes de façon indépendante (je vous épargne les détails). Si tu as un cancer, c'est douleur quasi-permanente et la chimio qui te bouffe de l'intérieur. Potentiellement cablé en permanence à un appareil pour pouvoir respirer.

 

Et encore, je ne parle que des cas que je connais, j'imagine qu'il y a beaucoup de variantes toutes aussi horrifiques.

[GilliB]

Parmi l’ensemble des remboursements de l’assurance maladie en 2008, 10,5 % sont associés à la dernière année de vie.

https://www.sciencedirect.com/science/article/pii/S039876201200613X

 

 

Edit

Entre 15 et 75 ans, les remboursements moyens au cours de la dernière année de vie des personnes décédées en 2008 sont plus élevés pour les femmes, puisqu’ils s’élèvent à 30 731 € ([30 438 € ; 31 025 €]), contre 26 076 € ([25 878 € ; 26 273 €]) pour les hommes (p < 0,0001) (Fig. 2). Cette tendance s’inverse après 75 ans, avec des remboursements annuels moyens de 20 747 € ([20 613 € ; 20 880 €]) pour les hommes et 16 366 € ([16 270 € ; 16 462 €]) pour les femmes (p < 0,0001).

Modifié 30 avril 2025 par GilliB

[Lameador]

2 minutes ago, GilliB said:

Parmi l’ensemble des remboursements de l’assurance maladie en 2008, 10,5 % sont associés à la dernière année de vie.

https://www.sciencedirect.com/science/article/pii/S039876201200613X

 

 

Et 99% vont à des gens en mauvaise santé.

No shit sherlock.

[Hugh]

https://climatedrift.substack.com/p/why-solarpunk-is-already-happening

 

"Why Solarpunk is already happening in Africa"

 

Incomplet:

 

Citation

You know that feeling when you’re waiting for the cable guy, and they said ‘between 8am and 6pm, and you waste your entire day, and they never show up?

Now imagine that, except the cable guy is ‘electricity,’ the day is ‘50 years,’ and you’re one of 600 million people. At some point, you stop waiting and figure it out yourself.

 

What’s happening across Sub-Saharan Africa right now is the most ambitious infrastructure project in human history, except it’s not being built by governments or utilities or World Bank consortiums. It’s being built by startups selling solar panels to farmers on payment plans. And it’s working.

 

Over 30 million solar products sold in 2024. 400,000 new solar installations every month across Africa. 50% market share captured by companies that didn’t exist 15 years ago. Carbon credits subsidizing the cost. IoT chips in every device. 90%+ repayment rates on loans to people earning $2/day.

 

 

The traditional development playbook goes something like this: Chapter 1, build centralized power generation. Chapter 2, string transmission lines across hundreds of kilometers. Chapter 3, distribute to millions of homes. Chapter 4, collect payments. Chapter 5, maintain the whole thing forever.

 

This worked great if you were electrifying America in the 1930s, when labor was cheap, materials were subsidized, and the government could strong-arm right-of-way access. It works less great when you’re trying to reach a farmer four hours from the nearest paved road who earns $600 per year.

So utilities do what any rational actor would do: they stop building where the math stops working. Which is exactly where the people are.

 

This has been the development sector’s dirty little secret for 50 years. “We’re working on grid extension!” Translation: we’re not working on grid extension because the economics are impossible, but we need to say we’re working on it so we keep getting donor money.

Meanwhile, 1.5 billion people spend up to 10% of their income on kerosene, diesel, and other dirty fuels. They walk hours to charge their phones. They can’t refrigerate medicine or food. Their kids can’t study after dark. Women inhale cooking smoke equivalent to two packs of cigarettes daily.

 

Battery costs also collapsed 90%. Inverters got cheap. LED bulbs got efficient. Manufacturing in China got insanely good. Logistics in Africa got insanely better.

All of these trends converged around 2018-2020, and suddenly the economics of off-grid solar just... flipped. The hardware became a solved problem.

 

But there was still a massive, seemingly insurmountable barrier: $120 upfront might as well be $1 million when you earn $2/day.

This is where the story gets interesting.

 

Quick history lesson: In 2007, Safaricom (Kenya’s telco) launched M-PESA, a mobile money platform that let people transfer cash via SMS.

 

By 2025: 70% of Kenyans use mobile money. Not in addition to banks. Instead of banks. Kenya processes more mobile money transactions per capita than any country on Earth.

 

It worked because it solved a real problem: Kenyans were already sending money through informal networks. M-PESA just made it cheaper and safer.

This is the unlock. This is the thing that makes everything else possible.

Here’s the model:

1. A company (Sun King, SunCulture) installs a solar system in your home

2. You pay ~$100 down

3. Then $40-65/month over 24-30 months

4. The system has a GSM chip that calls home

5. No payment = remotely shut off

6. Keep paying = keep power

7. After 30 months = you own it, free power forever

The magic is this: You’re not buying a $1,200 solar system. You’re replacing $3-5/week kerosene spending with a $0.21/day solar subscription (so with $1.5 per week half the price of kerosene) that’s cheaper AND gives you better light, phone charging, radio, and no respiratory disease.

 

The default rate? 90%+ of customers repay on time.

Why? Because the asset actually works. It delivers value every single day. The alternative is going back to kerosene lamps in the dark. Nobody wants that.

 

This is the “innovation” that everyone missed. The hardware got cheap, but PAYG made it accessible. And mobile money made PAYG economically viable.

 

And here’s what nobody outside Africa understands: Sun King has 50%+ market share in their category. They’re not scrappy startup. They’re a dominant infrastructure provider.

 

Why is the market concentrated? Because the full-stack is really fucking hard.

You need:

1. Hardware manufacturing expertise

2. Supply chain across fragmented markets

3. Last-mile distribution (29,500 agents for Sun King)

4. Mobile money integration

5. Credit scoring models for the unbanked

6. IoT/telemetry systems

7. Customer service in 10+ languages

8. Financing (equity, debt, securitization)

9. Carbon market relationships

10. Regulatory navigation across 40+ countries

Most companies can do 2-3 of these. The winners do all 10.

This creates massive barriers to entry and long-term moats. New entrants can’t just show up with cheaper panels. The moat is the full-stack execution.

 

 

But here’s the meta-point: This is the template for building infrastructure in the 21st century.

Not government-led. Not centralized. Not requiring 30-year megaprojects.

 

Instead: modular, distributed, digitally-metered, remotely-monitored, PAYG-financed, carbon-subsidized infrastructure deployed by private companies in competitive markets.

 

The grid that never came turned out to be a blessing. While development experts spent 50 years debating how to extend 20th-century infrastructure to rural Africa, something more interesting happened: Africa built the 21st-century version instead.

Modular. Distributed. Digital. Financed by the people using it, subsidized by the carbon it avoids.

The solarpunk future isn’t speculative fiction. It’s 23 million solar systems, 40 million people, and a template for how infrastructure gets built when you’re not stuck defending the past.

 

 

[Adrian]

No meta-analytical effect of economic inequality on well-being or mental health
 

Citation

Exposure to economic inequality is widely thought to erode subjective well-being and mental health, which carries important societal implications. However, existing studies face reproducibility issues, and theory suggests that inequality only affects individuals in disadvantaged contexts. Here we present a meta-analysis of 168 studies using multilevel data (11,389,871 participants from 38,335 geographical units) identified across 10 bibliographical databases (2000–2022). Contrary to popular narratives, random-effects models showed that individuals in more unequal areas do not report lower subjective well-being (standardized odds ratio (OR_+0.05) = 0.979, 95% confidence interval = 0.951–1.008). Moreover, although inequality initially seemed to undermine mental health, the publication-bias-corrected association was null (OR_+0.05 = 1.019; 0.990–1.049)¹⁷. Meta-analytical effects were smaller than the smallest effect of interest, and specification curve analyses confirmed these results across ≈95% of 768 alternative models. When assessing study quality and certainty of evidence using ROBINS-E and GRADE criteria, ROBINS-E rated 80% of studies at high risk of bias, and GRADE assigned greater certainty to the null effects than to the negative effects. Meta-regressions revealed that the adverse association between inequality and mental health was confined to low-income samples. Moreover, machine-learning analyses¹⁹ indicated that the association with well-being was negative in high-inflation contexts but positive in low-inflation contexts. These moderation effects were replicated using Gallup World Poll data (up to 2 million participants). These findings challenge the view that economic inequality universally harms psychological health and can inform public health policy.

J'ai pas creusé l'étude mais j'ai tout de suite pensé aux sociologues français qui ne seraient sûrement pas d'accord  . Les résultats en France seraient peut-être différents vu la sensibilité très forte aux inégalités chez les gens de gauche en France.

[Lancelot]

L'inégalité économique c'est mal parce que quand on persuade les gens que l'inégalité économique c'est mal, ils sont persuadés que l'inégalité économique c'est mal. Logique de tarés.

[Rincevent]

Il y a 9 heures, Vilfredo a dit :

Sur Oliver Sacks. Bouleversant

https://archive.ph/20251208231810/https://www.newyorker.com/magazine/2025/12/15/oliver-sacks-put-himself-into-his-case-studies-what-was-the-cost

La phrase la plus importante je crois : 

 

Citation

It speaks to the power of the fantasy of the magical healer that readers and publishers accepted Sacks’s stories as literal truth. 

 

[Lancelot]

J'ai tenu une quinzaine de paragraphes qui peuvent être résumés par :

 

 

Après ça, ne voyant toujours pas où l'article voulait en venir et constatant que je n'en étais même pas au quart, j'ai ragequit.

[Adrian]

Grok
 

Révélation

Lazaridis dit deux choses simples mais puissantes :

• Les humains ont eu des cerveaux de taille moderne (dans la fourchette actuelle) depuis au moins 500 000 ans (Homo heidelbergensis et descendants).
• Rien n’indique qu’on ait développé un « nouveau module génétique » (style mutation unique genre FOXP2 « langue » ou « Merge » de Chomsky/Berwick) qui aurait soudainement rendu nos ancêtres « modernes » vers 50 000 ans (révolution du Paléolithique supérieur) ou même 2 000 ans.

C’est cohérent avec ce qu’on sait aujourd’hui :

• L’anatomie crânienne montre que la capacité encéphalique est stable depuis longtemps.
• L’ADN ancien (y compris chez les Néandertaliens, qui partageaient déjà beaucoup de nos variants neuronaux) ne révèle pas de grand balayage sélectif unique sur des gènes de « cognition supérieure » après ~300 000 ans.
• Les traces de comportement symbolique (ocre, coquillages gravés, art) existent sporadiquement en Afrique bien avant 50 000 ans (Blombos Cave ~100-75 000 ans, etc.). L’« explosion » européenne du Paléolithique supérieur ressemble plus à une accélération culturelle qu’à une mutation biologique.

C’est exactement comme il le dit : on aurait pu aller sur la Lune il y a 50 000 ans (cognitivement parlant). On ne l’a pas fait parce que la culture cumulative n’était pas encore là.

C’est le point le plus sous-estimé dans ces débats. La cognition est une capacité, pas une garantie d’innovation immédiate. Exemples concrets :

• Les Romains (ou les Grecs classiques) avaient exactement la même cognition que nous. Ils avaient les maths, l’ingénierie, la philosophie. Ils auraient pu lancer la révolution industrielle… mais ils n’avaient ni le charbon facile à extraire en masse, ni la science expérimentale institutionnalisée, ni une population assez dense et connectée pour accumuler les idées à ce rythme.
• Même chose pour les chasseurs-cueilleurs du Paléolithique : petites bandes, isolement géographique, transmission orale fragile → une bonne idée meurt avec l’inventeur. Dès que la démographie explose (post-50 000 ans, puis surtout Néolithique), la culture devient cumulative et explosive.

C’est la grande leçon de l’anthropologie culturelle : la culture évolue beaucoup plus vite que les gènes, et elle peut rester bloquée pendant des dizaines de millénaires même avec un hardware parfait.

 

[Adrian]

Dignity neuroscience: universal rights are rooted in human brain science

[Adrian]

Je vous écris de Mésopotamie, il y a 4000 ans

 

Citation

La plus ancienne correspondance privée de l’humanité raconte les mariages et les divorces, le commerce et la fraude fiscale, le métier d’éleveuse ou de tisseuse. Des voix de femmes imprimées dans l’argile depuis 4000 ans, qui livrent des récits troublants d’actualité.

 

[Alchimi]

J'avais adoré lire sur une tablette mésopotamienne au British museum "je t'ai écrit le mois dernier, pourquoi n'as tu pas répondu?"

 

3000+ ans d'écart, même combat. La condition humaine dans toute son universalité.

[Lancelot]

https://en.wikipedia.org/wiki/Complaint_tablet_to_Ea-nāṣir

 

Quote

The complaint tablet to Ea-nāṣir (UET V 81)^[1] is a clay tablet that was sent to the ancient city-state Ur, written c. 1750 BCE. The tablet, which measures 11.6 centimetres (4+9⁄16 in) high and 5.0 centimetres (1+15⁄16 in) wide, documents a transaction in which Ea-nāṣir, a trader, allegedly sold sub-standard copper to a customer named Nanni. Nanni, dissatisfied with the quality, wrote a cuneiform complaint addressing the poor service and mistreatment of his servant.

Discovered by Sir Leonard Woolley in Ur, it is currently kept in the British Museum. Written in Akkadian cuneiform, this tablet is recognized as the "Oldest Customer Complaint" by Guinness World Records. The tablet's content and Ea-nāṣir in particular gained popularity as an internet meme, due to its relatable subject matter in expressing dissatisfaction with goods.^[2][3][4]

 

[Hugh]

https://www.houseandgarden.co.uk/article/why-the-newbuild-period-house-is-in-such-demand

 

"Why the newbuild ‘period’ house is in such demand"

 

Citation

the rise of the newbuild ‘period’ house is unmistakable. Built with modern methods but drawing on classical architectural language, these homes are more popular than ever in Britain and in other countries like the US. But in a landscape already full of Georgian terraces and Victorian villas, why would you choose to commission a new house in an old style, rather than simply buy the original?

For anyone who has ever fallen for a beautiful old house, the frustrations are familiar. Original period homes can be wonderfully atmospheric, but they often come with a long list of compromises: small or awkwardly placed rooms, winding corridors, a lack of storage, or draughts and cold spots that no number of blankets can fix. Retrofitting these buildings for modern life – whether that means installing underfloor heating, creating an open-plan kitchen or adding insulation – can be disruptive and costly. In many cases, listed status makes even simple changes a bureaucratic challenge.

This freedom is what makes newbuilds so appealing to today’s homeowners. With a period house, you inherit someone else’s decisions; with a newbuild, you start with a blank canvas.

A further, and hardly minor, advantage is that, unlike historic properties, a newbuild ‘period’ house is not constrained by protected status. There is no need to seek permission for every alteration, no council forms to fill out before adding a bathroom, and no painstaking negotiations with conservation officers. 

In the end, commissioning a newbuild ‘period’ house is not about nostalgia, but about stewardship and practicality, creating a home that honours the past yet is truly made for the future