50 mentions · 22 chunks · 17 episodes
...tion behind a command-line option. Or requiring the user to write a string like React's famous "__SECRET_INTERNALS_DO_NOT_USE_OR_YOU_WILL_BE_FIRED". The idea was users couldn't stumble into doing something dangerous if they had to craw...
...lieve is what matters. It is their beliefs that set their world model that they react to. Norms arise out of interdependent beliefs and expectations about what others believe. Rules are schelling points, they help set the default of ho...
...uthing with real world results. That can be done automatically with things like React components. But anything that is the least bit complex has to be put out into the world and see how the world reacts. Complexity can't just be calcul...
...e now that we have this self-replicating technology? We're never getting rid of React! We're stuck for it for all time now. Apparently alphabets stopped morphing when dictionaries were published.
LLMs have a codex of lots of examples of React apps embedded within them. They can replicate them on demand, with tweaks for any given context. LLMs have absorbed this knowledge by being bombarded...
...ive a dependably good result. But RLAIF works well for writing code (especially React components) since it's easy to construct an auto-ground-truthing pipeline. Write the code, try to run it (iterating until no errors) then using Playw...
...Ms where they are for code will get the most out of them. LLMs can write simple React components with high quality They need a lot more hand-holding and direction to write more complex code. A system that makes it possible for LLMs to ...
Today there's two camps: pro tech and anti tech. Both react to tech as it exists today. What about tech as it should exist? Acknowledging where tech has gotten lost or stuck or diverged from its ideals, while ...
Claude 3.7 has clearly been specially trained to be good at generating React code and SVGs. They are almost certainly using RLAIF to post-train the model to be significantly better in those domains. It probably looks something...
It's possible for a model to over-fit to a specific framework, like React. If I'm right that Anthropic has specially focused on React, I'd imagine the model got at least incrementally worse on non-React code. The model like...
...use cases, or a carefully calibrated UX flow to help squeeze out higher quality React apps out of models. But then a new model comes out that was directly trained on doing that use case well and it leapfrogs all of your linear improvem...
If you use a linear process to react to a compounding curve, you will never catch up. Compounding curves are things that grow on their own, like a tree. But it takes time for the tree to...
Claude's love language is React components. Any programming-adjacent question you ask it, it hops immediately to writing a crappy little React component[adw] without fully understan...
...ften above the bar, just not always hitting it out of the park like they do for React style code. The meta insight is that LLMs' ability to write code on demand is not some smooth distribution over types of code, but spiky based on whi...
...re architectures: Serious and Playful. The Serious archetype is typified by the React architecture of convergent modern front-end development. The goal is to precisely implement the design with minimal variance. The least-bad architect...
Riffing on a metaphor for declarative vs imperative from React's documentation; Imagine you're asking someone to drive you somewhere. Imperative is like giving a friend instructions on how to get there, describin...
...e code in your particular custom framework. Instead, just have it write code in React, which it's most familiar with. Later, you can kludge in a fake React-shaped shim to interpret it to your underlying semantics. Meet LLMs where they ...
...t incremental, continuous, and interdependent ones. A small move, then wait and react, then repeat. This allows coordination, call and responses, and mutual reactions. If everyone made their plans individually and then executed them wi...
...y to do, to stand out from the noise, to thrive. Adaptability is the ability to react, to change, to adapt, to survive. You want both coherence and adaptability. Coherence is easy to see even quickly at a distance. Adaptability is subt...
In design, a probe is a thing to catalyze feedback, a concrete thing to react to. A probe is not designed to be correct. It's designed to attract feedback. Pick a maximally interesting "what if" question whose answer will be ma...