iñaki arrola

Ser libre es que me importa de cero a nada quien me sigue o me deja de seguir.

The Economist’s glass-ceiling index #selection-117.0-117.35" target="_blank" rel="nofollow noopener">archive.ph/TPxw4#selectio… Y si, nuestro país, por mucho que nos digan nuestros gobernantes que hacen mucho por esto, está yendo a peor.

@DonShelby_ Oh! Don Shelby todo lo que sabe! Claramente el equipo está mucho peor que lo que estaba antes de Pintus. Te tienes que reír. 🤣

Con la lesión de Courtois, Antonio Pintus ha lesionado a TODOS los jugadores de la plantilla en apenas 3 meses. El gran fichaje invernal, decían los listos.

El dinero. O sale o se reiventa. Eso fue, en esencia, lo que se respiró en la última edición de Capital Letters de Harmon, que tuvo lugar ayer en nuestras oficinas. Léelo en nuestra Sala de Prensa: harmon.es/cuando-el-dine…

Hoy daremos buena cuenta del gran “capa de fuego” de Vitaminajota.com bestial @MaxiPortes!

👀👀

1) Un particular crea subvencions.cat 2) Es descobreix què a Catalunya es destinen 5.000 milions anuals a subvencions (!) 3) La Generalitat inhabilita l'accés al portal de transparència durant 28 hores 4) Quan rehabilita l'accés, es descobreix que han eliminat 1 milió de registres, dificultant així que la ciutadania pugui fiscalitzar on van i a qui regalen els diners públics 5) Un exemple: Open Arms ha rebut uns 2'5 milions d'euros d'institucions públiques catalanes els últims anys. Al portal de transparència només n'apareixen uns 550.000. És una vergonya i han de donar explicacions. I que no ens prenguin el pèl parlant d'un "error tècnic".

🚨 BREAKING: A Google researcher and a Turing Award winner just published a paper that exposes the real crisis in AI. It's not training. It's inference. And the hardware we're using was never designed for it. The paper is by Xiaoyu Ma and David Patterson. Accepted by IEEE Computer, 2026. No hype. No product launch. Just a cold breakdown of why serving LLMs is fundamentally broken at the hardware level. The core argument is brutal: → GPU FLOPS grew 80X from 2012 to 2022 → Memory bandwidth grew only 17X in that same period → HBM costs per GB are going UP, not down → The Decode phase is memory-bound, not compute-bound → We're building inference on chips designed for training Here's the wildest part: OpenAI lost roughly $5B on $3.7B in revenue. The bottleneck isn't model quality. It's the cost of serving every single token to every single user. Inference is bleeding these companies dry. And five trends are making it worse simultaneously: → MoE models like DeepSeek-V3 with 256 experts exploding memory → Reasoning models generating massive thought chains before answering → Multimodal inputs (image, audio, video) dwarfing text → Long-context windows straining KV caches → RAG pipelines injecting more context per request Their four proposed hardware shifts: → High Bandwidth Flash: 512GB stacks at HBM-level bandwidth, 10X more memory per node → Processing-Near-Memory: logic dies placed next to memory, not on the same chip → 3D Memory-Logic Stacking: vertical connections delivering 2-3X lower power than HBM → Low-Latency Interconnect: fewer hops, in-network compute, SRAM packet buffers Companies that tried SRAM-only chips like Cerebras and Groq already failed and had to add DRAM back. This paper doesn't sell a product. It maps the entire hardware bottleneck and says: the industry is solving the wrong problem. Paper dropped January 2026. Link in the first comment 👇

@rjlapetra @2010MisterChip Tienes datos de la Copa del Rey y de la intertoto?

@arrola @2010MisterChip En Liga española, últimas 5 temporadas: Real Madrid 0,30 penaltis por partido FC Barcelona 0,15 ppp Los datos en este caso cuadran con el relato de trato de favor arbitral hacia el Real Madrid

Más penaltis a favor en la Champions League: 65 REAL MADRID (+1 hoy) 65 Bayern 59 Barcelona 38 Manchester City 38 Chelsea

@DonShelby_ Pues nada. A mamar. Contempla tu obra Florentino Pérez

No tengo palabras para definir esto.

@DonShelby_ Y te jode.

Este tío ha decidido que en este partido quiere ser defensivamente Paolo Maldini, Fernando Redondo en el centro del campo y Ronaldo Nazario en la delantera. Fede Valverde acaba de realizar la mejor primera media hora de la Historia del Fútbol. Sí, sí, de la Historia del Fútbol.

@Ramon_AlvarezMM Qué día más bueno Ramón.

@samuelgil Es tan ridículo….

como sigamos matando cosas ya no sé qué nos va a quedar

🚨 AI AGENTS ARE ABOUT TO START HIRING EACH OTHER ON ETHEREUM A new Ethereum draft standard ERC-8183 enables trustless AI agent commerce using on-chain escrow. Here’s how it works: Client posts funds. Provider completes the work. Evaluator verifies the result. Payment releases automatically. No humans required. The AI agent stack on Ethereum is starting to emerge: 🔹 x402 → micropayments 🔹 ERC-8004 → identity & reputation 🔹 ERC-8183 → escrow for work AI agents will need to pay for APIs, data, compute, services, and other agents. ERC-8183 enables programmable escrow for agent to agent commerce. Virtuals Protocol and the Ethereum Foundation built the standard together, and builders are already integrating it for things like image generation and fund management. The rails for a machine to machine economy are forming. And they’re being built on Ethereum. $ETH

Humbled by the continuous support that we get from our users. We are the first company in that list that is European, and maybe correlated, bootstrapped. Proud of representing Spain.

$VIX - It's amazing how well this range gets respected.

Oil down 30% since this last night

Oil is now in a bear market… Down -20% from it’s recent highs.

I had to read this twice… Because the numbers don’t look real. Jacob Kiplimo just ran 57:20 for the half marathon. Let that sink in. That’s 4:22 per mile Or about 2:43 per kilometer… for 21.1 km straight. No fade. No collapse. Just machine-like pacing. Look at the splits: 5K – 13:28 10K – 27:00 15K – 40:52 20K – 54:23 Half – 57:20 That means he basically ran: • a 13-minute 5K • four times in a row. Most runners would celebrate breaking 20 minutes for a 5K. He’s running 13 minutes like it’s a tempo run. And here’s the crazy part… This is the FIFTH time he has gone sub-58 for the half marathon. That level of consistency is just absurd. For context: If you run a 2-hour half marathon, Kiplimo would finish… over an hour before you. Think about that. You’d still have 11 km left when he’s already showering. Running right now is on another level. Between Kiplimo, Kipchoge, and the new wave of super-shoe athletes… the limits of endurance keep getting pushed further. Makes you wonder: How fast can the half marathon actually go? Will we see a sub-57 soon? 🧠🏃‍♂️