xherce (⨍) retweeted
Mi momento favorito de la cumbre de la ONU: el presidente de Haití bebiendo directamente de la jarra. 🤣😭
Italian gaming company Lottomatica agreed to buy Spanish rival Cirsa Enterprises in an all-share deal valued at about €2.8 billion bloomberg.com/news/articles/…
We just kicked off Omarchy Kids. Working on a special wholesome and safe way to get children started with Omarchy. Harris is organizing. Get in touch!
👋🏼 want to help with Omarchy Kids Mode? Looking for adults to sacrifice—ahem—volunteer.
A few of you had been interested in the typing plugin. Want to help? I have no details yet but it should be fun :)
Please confirm and I can follow up with more info. I can confirm @dhh is officially onboard!
@calledtocnstrct @jankeesvw @bastidotnet @logerhaus @mike_kasberg @Rob_HCB @toastyJ @lgwalker2010
El sou d'un auxiliar d'informàtica de l'AGE (Spain) ronda els 1.400 € nets, tant si és a Madrid, Barcelona o un poble de Conca.
xherce (⨍) retweeted
It's over. Good. Now invent something new. Start from scratch. We need a new kind of music. I'll wait.
xherce (⨍) retweeted
Probably the biggest startup announcement of the year (decade?) in terms of the calibre of founders. If I had to name the top 10 engineers currently or previously at Google in terms of impact it would be the two cofounders (Larry and Sergey) and all 4 of these folks:
Announcing Discovery Loop!
I am very excited to announce that, along with my longtime friends and collaborators @Sanjay_Ghemawat, @OriolVinyalsML and @quocleix, we are founding Discovery Loop (@DiscoLoopAI), a Public Benefit Corporation whose mission is to automate machine learning, science, and engineering to accelerate discoveries and progress. The four of us have worked together for 14 to 30 years, and have helped build some of the world’s most used products, infrastructure and AI models, and we’re excited to turn our attention to this ambitious endeavor.
♾
Learn more at: discoveryloop.com
xherce (⨍) retweeted
Finally we have an American alternative to GLM 5.2
Awesome to see American Open Source catch up!
Kudos to @thinkymachines team
Today, we are introducing Inkling.
Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.
thinkingmachines.ai/news/int…
Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
xherce (⨍) retweeted
Grace Hopper suggested teaching computers English, critics scoffed, insisting that machines "only do arithmetic."
She built the world's first compiler in her spare time.
She was right and the experts were wrong..
In the early 1950s, processing time on million-dollar machines like the UNIVAC I was treated as infinitely more valuable than the labor of the human programmers who operated them.
At the time, programming was a brutal exercise in hardware manipulation. Programmers had to write instructions in pure machine code or octal notation, manually keeping track of where every variable was stored in the machine's memory. The job required a deep understanding of the computer's internal architecture and an immense tolerance for tedious, error-prone translation.
Grace Hopper, a mathematician working on the UNIVAC, believed this manual translation was backwards. If computers were so good at repetitive tasks, she reasoned, they should be the ones translating human-readable instructions into machine code. In 1951, she proposed a "compiler"—a program that would take high-level mathematical symbols and automatically convert them into executable binary code.
The computing establishment rejected the idea outright. Critics argued that computers were not built for symbolic manipulation, and even if they were, using precious machine cycles to compile code was a foolish waste of resources.
After proving her compiler concept worked with the A-0 system, she didn't stop at mathematical symbols. She recognized that for computers to be truly useful in business and industry, non-mathematicians needed a way to program them. Her team developed FLOW-MATIC, the first programming language to use English keywords like "COMPARE," "ADD," and "TRANSFER."
By demonstrating that software could be written in a language close to plain English, Hopper fundamentally changed the trajectory of computer science. Her work proved that writing software could be abstracted away from the physical hardware, shifting programming from an exercise in binary translation into a tool for broad logical problem-solving.
This conceptual leap laid the direct foundation for COBOL, democratizing software development and setting the standard for almost every high-level programming language that followed.
xherce (⨍) retweeted
L'eina cli de Grok (equivalent a Claude o Codex) pujava en remot una còpia del teu codi, incloent-hi claus d'accés, detalls que no necessita com a part de la tasca encomanada.
Grok té una adopció molt baixa al món empresarial, i això, destrueix cap confiança que es pogués tenir.
‼️ BREAKING: xAI's Grok Build CLI was uploading entire Git repositories to a Google Cloud bucket, private codebases and unredacted secrets included. The uploads quietly stopped via a hidden server-side flag, and xAI still has not said a word about scope, retention, or deletion.
The scale is staggering. On a 12 GB test repo, 5.1 GB flew out the door to xAI's grok-code-session-traces bucket while the actual coding task needed just 192 KB. The tool grabbed whatever repository it ran in, not the files it needed.
The fix arrived as a hidden flag, disable_codebase_upload: true, a day after a researcher's wire-level analysis. The "Improve the model" opt-out never stopped the uploads.
Still no advisory, no scope, no word on whether already-uploaded code gets deleted. For anyone pointing AI coding agents at proprietary code, what crosses the wire matters more than what the settings page says.
xherce (⨍) retweeted
"I find it ironic that the status quo is to then turn around and impose restrictive terms on distillation, and to reserve the right to learn from customer usage and interaction data. If learning flows in only one direction, economic value converges toward the owners of the learning infrastructure rather than the creators of the knowledge itself. Therefore, it's imperative that we distribute the learning infrastructure to every firm so that they can control their own learning loop."
Well said.
xherce (⨍) retweeted
The Return of Wind-Powered Cargo Ships Signals a Greener Future
by @QuantumTech_X
#Sustainability #RenewableEnergy #CleanEnergy
xherce (⨍) retweeted
Kansas - based Astron aerospace has developed the H2 starfire, a hydrogen - powered rotary engine with no pistons and only 82 total parts - nuts and bolts included.
xherce (⨍) retweeted
Quan li demanes a OpenAI Codex que escrigui fitxers normalment ho fa sense accents (comunicacio).
Se soluciona dient-li que usi Unicode.
Aquest problema és degut a la instrucció de sota. OpenAI no deuen voler problemes amb els jocs de caràcters (que els deu semblar un merder😀)
xherce (⨍) retweeted
Que les dades d'una cosa tan innocent com un joc acabin sent usades per a ús militar ens recorda que les dades agregades poden tenir molt valor. Sempre algú tindrà l'incentiu de monetitzar-les, a més, havent-hi un mercat molt gran de brokers de dades.
lavanguardia.com/encatala/20…
A partir d'ara, aniràs a un entrevista de feina amb els teus agents (minions) i hauras de negociar quin pressupost de tokens tens.
xherce (⨍) retweeted
Portes temps buscant com col·laborar amb codi lliure?
Vols donar ajudar a Softcatalà i no saps com?
Iniciem una prova de concepte i ens cal ajuda:
github.com/Softcatala/arena-…
Ambiciós? Molt. Però amb la teva ajuda pot ser possible. Pregunteu sense vergonya i millorem la definició😉































