Jerónimo López

Picador de código en @ClarityAIEng

[Luis Garicano 🇪🇺🇺🇦 @lugaricano]

¿La directora general de la Guardia Civil y el Jefe del Estado Mayor conspirando con las cloacas del partido socialista para parar los pies a la principal unidad anticorrupción de España? No existe ninguna explicación plausible de este comportamiento que no involucre al Presidente del Gobierno.
(fuente @elpais)

[José Luis Antúnez @jlantunez]

Doblegar y chantajear a las altas instituciones del Estado para que no te investiguen a ti como presidente y al gobierno es más grave que la financiación ilegal de tu partido.

No vale decir es que esto siempre ha pasado para quitarle gravedad.

[Jerónimo López @jerolba]

Para la charla de @commitconf me han salido 58 slides para 45 minutos. Tendré que ponerme en modo @ilopmar, a 2x para darla entera en tiempo 😂

[Google Gemma @googlegemma]

Meet Gemma 4 12B!

A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license.

Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

[Jerónimo López @jerolba]

El Gobierno debería caer sólo con que la mitad de la mitad de la mitad de lo que se cuenta aquí fuera cierto.
Pero todos sabemos que no pasará nada. Qué desgracia.

[El Equidistante @elequidistante]

Venga, vamos al lío.

Hace unos días os conté lo que decía el auto del juez Pedraz sobre la trama de Leire Díez y Santos Cerdán.

Aquello era el plan sobre el papel. Hoy os traigo lo que pasó cuando ese plan se puso en marcha.

He podido acceder a dos piezas nuevas:

Una es el atestado 93/2026 de la UCO: lo que la Guardia Civil hizo y encontró en los registros del 27 y 28 de mayo.

La otra son las actas de declaración de 16 testigos, de esos mismos días.

Y hay tela. Cosas muy graves. Y alguna que da hasta vergüenza ajena.

Lo de siempre, que no se me olvide: esto son indicios y declaraciones, no una sentencia. Lo valorará el juez.

Os aviso de que es largo, porque el caso lo es.

Voy a intentar contarlo lo mejor que pueda.

Empezamos. 🧵

[Jerónimo López @jerolba]

@javilop primo!, compartimos apellido, mis padres buscaron un nombre original....

[Jerónimo López @jerolba]

@diegomarino ¿Una de estas? 😂

[Growing Up Retro @GrowingUpRetro]

The A-Team saves the day.

[Boro Mas @boromas]

Estoy viviendo una situación kafkiana con la @ComunidadMadrid. La cuento por aquí a ver si a @IdiazAyuso la lee y se le cae la cara de vergüenza y alguien lo soluciona:

[Anish Moonka @anishmoonka]

Matthew McConaughey was paid less than $200,000 for the role that won him an Oscar. A year earlier, he'd turned down $14.5 million to keep playing the same charming guy in the same kind of movie. He walked away from that money on purpose.

For most of the 2000s, he was the actor you hired when you needed a handsome man to chase a woman around a beach for ninety minutes. The romantic comedies were the only films of his that reliably made money, so they were the only scripts coming in. Movies like The Wedding Planner and How to Lose a Guy in 10 Days, each one clearing $100 million. He was a safe bet, and he was bored out of his mind.

So in 2010 he told his agent he was finished. No more rom-coms. He went home to his ranch in Texas and waited.

Then the offers started climbing. An action comedy came in at $8 million. He passed. They raised it to $10 million, then $12 million, then $14.5 million for the same script with funnier jokes. He read that one twice and still said no. After that, the phone went quiet for almost two years. He stayed on the ranch and waited it out.

The part that finally pulled him back was a real man named Ron Woodroof, a Texas electrician who was diagnosed with HIV in 1985 and told he had a month to live. The film, Dallas Buyers Club, was dirt poor. Its entire makeup budget was $250. The makeup artist couldn't afford fake skin, so she drove to her mother's house, grabbed grits and cornmeal, and used them to paint rashes onto healthy actors to make them look like they were dying. She later won an Oscar for that work.

McConaughey lost 47 pounds for the role, dropping from 183 down to 136. He was paid under $200,000. He'd just said no to $14.5 million, and now he was taking home less than two cents for every one of those dollars. The whole movie cost $5 million to make. It went on to earn $55 million and win three Academy Awards, with Best Actor going to him in early 2014.

The math only works if you look at what came next. That small, brutal role changed what every director in town thought he could do. The same year he won the Oscar, he was the lead in Christopher Nolan's Interstellar, a space movie that earned $746 million. His five highest-grossing films have now pulled in around $2.3 billion between them, and almost none of them are the kind of movie that made him famous.

He gave up $14.5 million to earn $200,000. That trade bought him a career he could never have paid for.

[TRGCON @tarugoconf]

¡Hola Mundo!

Ya tenéis disponible la web de la #TRG26 😱 https://t.co/5O18SkC8xZ

[Jaime Novoa @jaimenovoa]

🔥🔥 Factorial

$150m in Series D funding at a $2.5b valuation, led by General Catalyst

Plus GC has committed up to an additional $540m through its Customer Value Fund

[Luis Garicano 🇪🇺🇺🇦 @lugaricano]

The most important chart in Spain's history: the top scenario (sudden, sharp fertility rebound) produced by ex-fiscal responsibility agency chair, future minister of pensions and social security and current CB governor Escrivá was the basis of the pension reform that abandoned dynamic adjustment and also of Mr Escrivá's political career.

[Millie Marconi @MillieMarconnni]

A Dutch computer scientist gave one lecture in 1988 arguing that programming is unlike anything humans have ever tried to do before, and the reason most software on earth is broken is that we are still teaching it as if it were a hobby.

His name was Edsger Dijkstra. He won the Turing Award in 1972. He invented the shortest path algorithm that every GPS on earth still runs on.

He wrote the paper that killed the goto statement in modern programming languages.

He spent 50 years quietly being one of the most consequential thinkers in the entire history of computer science, and he was in a very bad mood by the time he stood up at the ACM Computer Science Conference in 1988 to deliver the lecture that almost nobody at the conference wanted to hear.

The lecture was called On the Cruelty of Really Teaching Computer Science.

It is now one of the most cited papers in the entire history of computing education. It was filed in his archive as EWD1036, handwritten in his careful fountain-pen calligraphy because he refused to use a typewriter and famously refused to use email for the rest of his life.

The argument was simple and uncomfortable.

Programming, Dijkstra said, is a radical novelty. Not a new tool. Not a new skill. Not a faster version of something humans already knew how to do. A genuinely new category of intellectual activity that has no real precedent in the entire history of the human species, and our brains have not been built to handle it.

Here is what he meant by that.

When a programmer writes a line of high-level code and presses run, that single line might trigger a billion operations at the level of the silicon.

The ratio between the abstraction you are working in and the physical events you are actually causing is roughly one billion to one. No engineer in history before computing ever had to reason about a system spanning that kind of ratio inside their own head.

A bridge builder reasons about steel beams and the physics of weight. A surgeon reasons about organs and the physics of tissue. A chemist reasons about molecules and the physics of bonds.

All of them are working inside ratios of physical scale where the largest and smallest things they need to think about are within a few orders of magnitude of each other.

A programmer routinely writes one line that orchestrates a billion physical events on a chip, and is expected to predict the behavior of all of them in advance.

Dijkstra argued that the human brain was simply not built for this. Every intuition we have evolved over hundreds of thousands of years comes from a world of medium-sized objects behaving in continuous ways. Computing is the opposite. It is discrete, not continuous.

A program that runs perfectly a billion times can crash on the billion-and-first iteration because of a single bit. A single character missing from a line of code can take down a power grid. There is no margin. There is no graceful degradation. The system either works or does not, and the only way to know is to actually run it.

This was the part of the lecture where Dijkstra made everyone in the room uncomfortable.

He said the way computer science was being taught in universities was a quiet disaster. Professors were teaching programming the way carpenters teach woodworking. With examples. With metaphors. With analogies to things students already understood. Files are like folders. Memory is like a desk. A function is like a recipe.

Dijkstra said this was actively making it harder for students to think clearly. The whole point of a radical novelty is that there is nothing in your past experience to compare it to.

The moment you start reaching for metaphors, you are smuggling in old intuitions that do not apply, and those intuitions will betray you the first time you try to reason about a system the metaphor was not built to describe.

His exact line was this: the usual way in which we plan today for tomorrow is in yesterday's vocabulary. And yesterday's vocabulary, he argued, was killing the field.

The reason most software is broken is downstream of this single misunderstanding. Programmers are taught to think of code as a craft. Something you get a feel for.

Something you pick up through practice. Something where intuition gets sharper with experience.

Dijkstra said this is exactly backwards. Programming is not a craft. It is closer to mathematics than to carpentry, and the moment you treat it as a craft, you guarantee that the software you produce will be full of the kind of bugs that craftsmanship cannot catch.

The fix, in his view, was to teach programming the way mathematics is taught. You should be able to prove your program correct before you run it.

You should reason about your code formally, the way a mathematician reasons about a theorem, not the way a carpenter feels their way through a joint. The students who learned this way, he said, would walk out of their classes with a kind of confidence that no amount of typing practice could produce.

The lecture was published in Communications of the ACM in 1989. The field did not listen. Universities kept teaching programming the same way.

Software kept getting bigger. Bugs kept compounding. By 2026, almost every piece of software on earth has known security vulnerabilities, undefined behaviors, and edge cases that nobody has ever proven safe. The doom that Dijkstra warned about in 1988 is now the default condition of the digital world we have built.

The deeper lesson is the one most readers miss the first time through.

Dijkstra was not just talking about software. He was making a much bigger point about how humans learn anything that is genuinely new. The instinct to translate the unfamiliar into the familiar is the most natural thing in the world.

It is also the single biggest obstacle to actually understanding something that has no precedent. If you keep reaching for analogies, you will never see the new thing clearly. You will only see your old framework projected onto it.

This is happening right now with AI. The same instinct that made people learn programming through metaphors of files and folders is making people understand large language models through metaphors of brains and people.

Almost every framework being used to describe AI in 2026 is borrowed from a previous domain. None of them quite fit. The few people who are actually building useful intuitions about how these systems work are the ones who have done what Dijkstra recommended forty years ago.

They have set down the old vocabulary. They have looked at the new thing on its own terms. They have accepted that the radical novelty is radical for a reason.

You are not slow. You were taught a discipline as if it were a hobby. The cruelty is real.

The fix is still available.

[DATHOS Science @DathosBD]

Tras mi operación me propuse cambiar algunas ideas y enfoques en mis tuits de X, pero lo ponen imposible.

"La población migrante utiliza menos la sanidad que la población nacida en España"; Dª Mónica García.

El informe 'Estado de salud y uso del sistema sanitario por la población migrante' en materia de data es muy malo entre otros en forma, contenido y análisis.

Pero se les da bien crear teorías y relatos que tapen escándalos, corrupción y vergüenzas políticas.

En la forma.
Un barómetro con 7.620 encuestas, fuente M. Sanidad y CIS, sobre población de 48,6 millones de personas (año 2024) en una época con millones de datos e IA es un despropósito que se resume en dos datos:

27 encuestas a personas nacidas en África sobre una población de 1,5 millones de personas.

674 encuestas a personas nacidas en Latinoamérica sobre una población de 4,1 millones de personas...

[Commit Conf @commitconf]

¡Pistolazo de salida en #CommitConf con @eferro que será el encargado de la keynote de apertura para hablarnos sin filtros sobre la #IA y el futuro real de los y las profesionales de IT. 🧠

📅 Viernes 5 de junio
⏰ 9:00 h
📍 Track 1 - Next Digital

¡No te lo pierdas!

[Emilio Montilla @EmilioMontilla_]

Este es José María Sánchez Sánchez.

El tribunal del caso de David Sánchez sopesa imputarlo por presunto falso testimonio y falsedad documental durante el transcurso del juicio.

¿El motivo?

Su papel fue fundamental para enchufar a Luis Carrero, amigo íntimo del hermanísimo y exasesor en Moncloa, en la jefatura de actividades transfronterizas de la Diputación de Badajoz.

Participó en la comisión que valoró su plaza pese a que legalmente no podía hacerlo, pues aún no estaba vigente su contrato con la Diputación.

Sin embargo, durante el juicio declaró que su presencia allí estaba totalmente justificada, aunque esa comisión se celebró antes de que él entrara oficialmente en su puesto en noviembre de 2023.

Y hay más.

Puede acabar imputado por posible falsedad documental por un escrito con membrete y firma digital de la Diputación que habría intentado justificar su presencia en dicha comisión, pese a que la propia institución habría reconocido que no fue tramitado reglamentariamente.

Lo más curioso es que, poco antes del juicio, ha sido beneficiado por la diputación socialista de Badajoz al hacerlo fijo como jefe del Servicio Provincial de Bibliotecas.

Ya es casualidad –que no causalidad– que este señor declare lo que al PSOE le conviene poco después de haber sido ascendido por el PSOE.

Y ahora puede acabar en la cárcel por mentirle al tribunal.

Espero que su caso sirva como ejemplo para todos aquellos esbirros que La Rosa Nostra ha comprado.

[Santiago Calvo @SantiCalvo_Eco]

El 22% de las empresas que pidieron ayuda al ICO por la pandemia han desaparecido y dejan un agujero de 6.800 millones.

De las 388.424 empresas que recibieron avales del ICO, una de cada cinco no devolverá el dinero. El propio Tribunal de Cuentas reconoce que sirvieron para sobrevivir, pero a costa de más endeudamiento y con efectos limitados y puntuales sobre el empleo.

Avalar en masa con el riesgo socializado mantiene con vida a empresas inviables y traslada la factura al contribuyente.

Por @carlossegovia_.

https://t.co/EA0yZtSBex

[Daniel Lemire @lemire]

I do a lot of optimization work.
What has been tempting to do is to ask the AI to run the benchmarks with and without a change, and to let me know which one is faster.
Seems simple enough.
Twice now, it has misled me. It reported its own synthesis of the results… but then when I asked for straight outputs from the code, the results did not check out. And the AI started looking for excuses for its misreporting.
Also, it tends to lie outright when it comes to deeper analysis. For example, in one case, I asked it to look at the assembly code of a function and check that two memory regions were loaded just once in the whole function. It insisted that everything was fine.
Then I asked to see the assembly for myself. It presented a branchy routine where, as I had suspected, one of the two memory regions was often reloaded twice, introducing additional latency.
When pushed, it told me, “Yeah, but it is faster this way,” blah blah blah. Of course, that wasn’t my original question, and the correct answer was, “Yeah, it will get reloaded twice in some cases.”
What this stresses is that you need to focus on “ground truth.” Use tools that are bound to give you the hard truth, and reject “AI interpretations.”

[Daniel Marín @Eurekablog]

Me siento totalmente identificado (también me pasa con el "pseudo-cirílico" de las películas).