About

This blog always starts with the same two questions: is this true, and what is the Brazilian figure?

The second one is the one that usually goes missing. Much of what gets read in Brazil about technology, education and cognition arrives translated from an argument that happened somewhere else, carrying that country’s data. Sometimes the Brazilian figure exists, has been published, and tells a different story than the imported version. Sometimes it doesn’t exist at all — and saying so out loud beats borrowing a foreign number and hoping nobody checks.

The standard

Primary source or nothing. The study, the report, the agency’s own table. Not the article about the study.

Every number holding up an argument goes through a source dossier first, where each claim gets a status: confirmed in the original document, confirmed only from a search snippet, not found, or contradicted by the source. The distinction isn’t bureaucracy. “Confirmed from a snippet” means the number matches what the institution publishes, but that the methodology note around it — the sample, the base year, the “rural areas excluded” — went unread. No whole argument gets to hang on that.

Anything marked not found doesn’t become prose. It either leaves the text or appears with explicit attribution to whoever claimed it.

Large figures come with a sense of scale, because a number nobody can picture informs nobody. Foreign-currency amounts come with the exchange rate for the date used.

The dossiers are archived alongside the piece. The reasoning is simple: six months from now, if someone disputes a number, the answer has to exist somewhere other than my memory.

What you won’t find here

Motivational posts, coaching threads, lists of five takeaways, corporate press releases dressed up as analysis.

You also won’t find the opposing argument in its weakest form. Where there’s real disagreement, the other side goes in as strongly as I can put it, and only then comes the response. Irony aimed at ideas, never at people.

And you won’t find quiet corrections. When I get something wrong, the fix shows up in the text, dated.

The subjects

Technology and society, education, cognition, economics, inequality, labour, climate and the environment, culture and philosophy. With a preference for the Brazilian angle — which tends to be precisely the one nobody looked up.

Everything is published in Portuguese and in English. The language switcher in the menu takes you to the same page in the other language, not back to the home page.

Getting in touch

Instagram: @fernandodeoliveira.md

Found a number that’s wrong? Write in. A mistake pointed out with a source is the best kind of message this blog can get.