If you need to start digging through a repository for more than just the README and issues, clone it locally. Even if the online documentation is thorough and well designed. If you'll be using it more than once, clone it.
You know all those random questions you'd never bother looking up if search engines didn't exist and you had to go to the library? Friction is real. You don't realize how much resistance you have to actually verifying how the code works until it's sitting on your computer. Nothing beats the speed of local search in a tool you know well, and if you know it's that fast you'll do it more instead of fiddling with code in hopes of being correct.
Online code search is awful, but at least within documentation links to related information usually work. Opening a chain of five class definitions and waiting that fraction of a second for each page to load before you can find the thing you need though, that grates. It gets better by the day, but there's still many repositories where I know finding the raw documentation is easier than the website. Even copy paste is easier when you aren't stuck in HTML code blocks.
You'll learn more this way too. Repositories clump everything together. Source code, examples, documentation. Seeing half a dozen snippets of how the implementation uses an function is often more helpful than reading the examples. You'll get background information you didn't know would be relevant.
Give it a shot, you can always go back to the websites if you prefer.
- Rew
Nightly Notes
I think this is the most programming specific post so far. I'm honestly surprised it took this long, it occupies so much of my life.
I considered writing the broader version of this entitled "Download Everything", but the gains for programming specifically when you clone repositories you need to reference is so much higher that I think the point would be lost.
- Rew