Up to date version:
That's why we have IA...
We automate stupidity at a level that as individual, mortal human beings, cannot do.
We can do gigatokens level of stupidity.
We can do agentic stupidity.
Humans can't screw this up so quickly, we need IA for that.
@KevlinHenney
"There's been some interesting studies about vibe coding (...), using a LLM-backed tool made them 20% slower, but (...) they all said they thought it had made them 20% faster."
- Sarah E. Chasins, Assistant Professor at UC Berkeley EECS
https://t.co/UItHTN4U8Z
El Vibe Coding está muerto, el futuro está en el "Bribe Coding", empresas que sobornen con dinero a personas (nada de IAs que fallan mucho y son muy caras) para que programen lo que quieran y así puedan deshacerse por fin de los programadores.
Lately I've seen this pattern quite often: information cherry-picking presented as universal facts. When people complains they say everyone is stupid for "missing the point", but the point is misleading and no matter at all for most cases. They're in for notoriety and engagement.
PHP just leveled up! 💪
Meet PIE (PHP Installer for Extensions) – the hot new tool making extensions EASY with Composer vibes! 🎉
Try it: https://t.co/GPK4fXKoRX
AI coding is used to generate a lot of bulk code that is often blindly accepted, but it seems there is at least as much opportunity for AI to help make codebases more beautiful.
Keeping a codebase in great shape or cleaning up a hairball takes a lot of effort beyond raw functionality, and a tireless AI assistant continuously pouring over everything looking for places to suggest changes should be valuable. AI as a diligent team member instead of your coding genie.
While there are factors peculiar to each’s understanding, I believe there are common coding behaviors that improve understanding for both humans and LLMs. It should be possible to run actual objective experiments with “style guides” for LLMs, then intersect them with the politics and fashion of human style guides.
Are there any tweaks to be made in LLM tokenization to more closely mirror programming language lexing?
Does saving context length settle tabs vs spaces?
I would like to see how a notoriously picky team like the @OpenBSD developers could onboard an AI team member.
@ARehmanMahi @cerbero90 It works no matter the web server, choose which one you prefer. PHP Fibers provides async concurrency on a single threaded process. Concurrency is not parallelism nor multithreading.
@devruso En mi experiencia los puntos de complejidad solo sirven para ver si una tarea puede ser desglosada en otras más pequeñas, pero nada más. Todo lo que tenga que ver con estimar tiempos de entrega huele a waterfall y microgestión.
@jenklore Cuando estaba en el instituto (hace casi de 30 años) usaba un programa de síntesis de voz para pasar los apuntes de historia cintas de casete y escucharlos en mi Walkman allá donde fuera. Aún hoy puedo recitar del tirón trozos enteros que recuerdo de memoria...
About the Clean Code debate, yes or no?
The right answer is: "Maybe, it depends on the context"
Never argue from a dogmatic approach, you've to know why and when to apply a practice, technique or methodology.
Different problems requires different solutions.