Remix.run Logo
weitendorf 21 hours ago

The way I process this kind of content is by skimming or ctl-Fing for the material I'm interested in reading and usually just reading the example code or the specific explanation for the content I'm after

For example this site ranks on the first page for "go 1.27 generics" and "go 1.27 uuid"[0] and if I were looking for uuid content I'd probably click on the toc for uuid and go to here [1] and look at the examples and v4 vs v7 semantics and then bounce. For this particular article the thing I'd be most interested in applying is probably REPACK [2]

  REPACK (CONCURRENTLY) events;
And all of the content around their code snippet showing that is pretty prescient/semantically dense and useful. What I wouldn't do is read the whole thing front to back, or the prose at the top/bottom with the LLMisms: it's way too long and dense for that.

For comparison here are the official new postgres docs about REPACK [3]. Is it human-written and more informative? Maybe for some people, or for me if I needed to reimplement a postgres-compliant spec or something, but I'd prefer the LLM-assisted (and I say assisted because IME it's actually a decent amount of work to get LLMs to write content like this) article most of the time.

[0] https://www.google.com/search?q=go+1.27+generics

[1] https://victoriametrics.com/blog/go-1-27/#the-uuid-package

[2] https://victoriametrics.com/blog/postgres-19/index.html#repa...

[3] https://www.postgresql.org/docs/19/sql-repack.html