Join 📚 Josh Beckman's Highlights
A batch of the best highlights from what Josh's read, .
My gut tells me this pattern, which doesn’t really have a name, is going to be *the* pattern for working with these technologies to do complex things. The best name I can find for it is “specification repair”—see, for example, [this paper](https://arxiv.org/html/2309.16120v3). You ask a question and the LLM transforms it into a more detailed specification of a problem, then writes code to test that specification, then creates more specification, then tests that.
Is “Specification Repair” the AI Endgame?
Paul Ford
Tree-sitter’s parsers already let us efficiently parse the code that our users upload. For instance, the Tree-sitter parser for Python produces a concrete syntax tree (CST)
Introducing Stack Graphs | the GitHub Blog
github.blog
In fast evolving fields it’s always all about sociology, not canon or pedagogy. Meaning in new fields is created in community (constructionism).
You need to plug into the community and overhear what people are talking about (HN is such a community). You’ll also get a sense of the linguistic subculture (acronyms, lingo etc) much like you learn to talk hip hop if you’re into the hip hop subculture. Much of it will be noise but overall you’ll get a sense of what the community cares about, which helps you narrow what you need to focus on.
In Fast Evolving Fields It’s Always All About Sociology, Not Canon or Pedagogy. Meaning in New Fields Is Created in Community (Constructionism).
ycombinator.com
...catch up on these, and many more highlights
Readwise