Hamilton Carvalho (oi, complexidade)

Cosmic dust. I write about complex social problems. PhD from the University of São Paulo.

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Climate Change Isn’t the Disease. “Civilizitis” Is.

The last time I went to the sports center where I play footvolley—a sport invented in Brazil that blends soccer and volleyball—I was taken aback when I ordered coconut water. Instead of the familiar, hand-pierced coconut...

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Neymar is approaching the proverbial "last time", the final occurrence of an everyday act that we perform so routinely we fail to notice when it becomes the last. For Neymar, that moment may already have come (yesterday).

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

@lygia_maria Parabéns pelo texto, Lygia. Impecável. Também escrevi neste final de semana sobre o tema. https://t.co/9Db2j2pAYK

[Rory Sutherland @rorysutherland]

Nobody Knows Whether Michael Jackson Is Canceled,
by @robkhenderson

https://t.co/RDc843Ytxf

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

E aqui o artigo. Diminuir a maioridade penal é pouco. https://t.co/9Db2j2pAYK

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Um modelo para simular efeitos da redução da maioridade penal.

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Aqui a execução do cenário-base.

[Mambo Italiano @mamboitaliano__]

Social media trends have turned the world’s most beautiful places into endless bathroom lines at a concert, where everyone waits for hours just to take the same photo to show to people who couldn’t care less 🌎📸

Nothing captures the shallow decay of our time better than this

[Rory Sutherland @rorysutherland]

The entire tech world has sold people a plausible fiction: that digitising everything will inevitably make it better, and that money spent on tech magically does not incur an opportunity cost.

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

This is helpful indeed.

[Dan Williams @danwilliamsphil]

Helpful project instructions for Claude/ChatGPT (generated by Claude) if you want a decent daily summary of what's going on in the world:

Trigger. When I open a conversation with a greeting ("hello," "good morning," or anything similar), treat it as a request for my daily world briefing and produce it without waiting for further instruction. If I say anything else, respond normally.

Task. Search the web for the current day's events (use the actual current date) and write a synthesis of roughly 1,500 words covering the significant developments in the UK, Europe, the US, and the wider world.

Foreground politics, technology, science, and culture; treat economic, geopolitical, and demographic developments as the structural layer beneath them. Run multiple distinct searches — realistically eight to fifteen — rather than a single sweep: at minimum search the UK, European, US, and international picture separately, and follow threads into original or primary sources (newspapers of record, official documents, specialist outlets) rather than aggregators or any single outlet's framing.

What I want from the synthesis — this is the part that matters. Not a headline reel. An epistemically disciplined account of what is actually happening, which means:

- Lead with significance, not salience. Rank stories by whether they will still matter in a month or a year, not by how much attention they are getting today. Demote the most-covered story and promote a quieter one where that is warranted, and say plainly when a day is uneventful rather than manufacturing importance.

- Situate every figure and event in its trend. A number without its base rate, denominator, and recent trajectory is close to meaningless. "Inflation came in at X" should become "X, up or down from prior months, above or below the longer-run trend." Treat single data points as draws from a noisy series, not as turning points.

- Distinguish genuine news from pseudo-news. A change in the world is news; a restatement of an ongoing condition, a poll consistent with the existing trend, or a forecast reported as an event, usually is not. Say which is which.

- Calibrate uncertainty explicitly. Separate what is established, what is contested, and what is speculation, and flag where early reporting is likely to be revised. Do not launder rumour or a single source into fact.

- Correct for the structural biases of the news itself: its negativity bias, its preference for the dramatic and discrete over the gradual and systemic, and its event-driven myopia. Give real attention to slow-moving developments that rarely produce a day's headline but do more to determine outcomes than most of what does.

- Avoid false balance and coalitional framing. Do not split the difference between positions to appear neutral; weight claims by the evidence. Do not adopt the framing, vocabulary, or grievance structure of any political side. Where a contested empirical question has a defensible answer, give it.

Style. Dry, precise, analytical prose. No rhetorical flourishes, no bold text, no exclamation, no motivational or "buckle up" cadence, no headline-style section titles. Light signposting by region or theme is fine; bullet-point lists are not — write it as continuous prose. Integrate the deeper-trends analysis into the account rather than appending it as a separate "context" section. Cite sources for specific claims. Be concise and concrete: do not pad, and do not hedge as filler. Calibrated directness beats reflexive both-sidesing.

Close with a sentence or two on the limits of the briefing — what the searches may have missed, where coverage was thin, and which stories are still developing and should be treated as provisional.

[Adam Hunt @RealAdamHunt]

Status anxiety is rational in a species where status determined access to mates, resources, and allies for two million years. Social media didn't create the wound! But it did remove every natural ceiling on how many people you can measure yourself against.

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Coluna de hoje, sobre riscos existenciais: O ser humano é uma linda experiência que deu errado. https://t.co/U6qNc1eEYY

[Science Magazine @ScienceMagazine]

How do childhood environments shape the brain?

In the latest #SciencePodcast🎙️, researcher Scott Marek sheds light on brainwide association studies for childhood brain development.

🎧 Listen here: https://t.co/iJSSXKKktr

[Will @Evolving_Moloch]

Patriarchy is a massive blind spot for many evolutionary psychologists because they think only in terms of sexual selection and assume individual men should always be in competition with other men, but when you understand cultural evolution and the importance of norms and norm psychology it makes perfect sense.

In fact, even from a purely individual fitness perspective it should be easy to understand. Norms of coercive arranged marriage and polygyny, casual wife beating, male control of space and resources, etc can be strongly in all or nearly all individual male interests even if it comes at the expense of their wives, daughters, sisters, etc. Low status men can benefit from having a wife when they otherwise might have none, high status men can benefit from multiple wives, men as a whole can benefit from increased ability to mate guard through violence, I mean there’s dozens of ways the logic can go for why all or nearly all men would endorse coercive patriarchal norms, and our evolved norm psychology leads people to commonly adopt and perpetuate the norms of their society. This EP argument that keeps getting repeated that intrasexual conflict makes patriarchy impossible is bizarre. We have abundant ethnographic evidence of men consciously using gang rape as a way to control women.

[Science Advances @ScienceAdvances]

Ancient European humans ate bugs—but it was likely by accident—according to a new examination of dental plaque. https://t.co/0HIPDaIAjk

[John B. Holbein @JohnHolbein1]

I have often heard academics say something like, "in the AI era, simply being hyper-productive is no longer a useful signal. It's all going to be about the quality of the ideas now."

I think there's a nugget of truth there.

However, this take is, ultimately, incomplete. Finishing high quality research projects is still a very important learned human skill!

[Florian Ederer @florianederer]

Is it the phones?

A new paper uses AT&T's 2007–2011 iPhone monopoly as a natural experiment and finds that iPhone access sharply reduced births, especially among younger women.

Smartphones may explain 33–52% of the post-2007 decline in U.S. fertility.

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Não, você não enxerga o óbvio. https://t.co/Zc00WnJ2wk

[MIT Sloan School of Management @MITSloan]

When workers become overreliant on AI, there’s a risk of significant skills collapse. Here are specific actions individuals and organizations can take to protect cognitive capital and preserve institutional knowledge. https://t.co/0UK9IdI1R2

[Paul Novosad @paulnovosad]

Authors can't be trusted to run their own robustness checks.

In 17 AER papers, only 12/211 robustness checks "fail" with p > 0.05 (white).

In robustness checks chosen by 3rd parties, almost *half* of them fail (blue).

1/

[Hamilton Carvalho (oi, complexidade) @PublicHamilton]

Detergente sabor Neymar. https://t.co/zSmntUjeVs