9 ms·
In regards to the GitHub stars, I don't think that a high amount of stars actually correlates to the amount of people using the repository. Cachet (https://git
by jbrooksuk 11y ago
In regards to the GitHub stars, I don't think that a high amount of stars actually correlates to the amount of people using the repository.
Cachet (https://github.com/cachethq/cachet https://github.com/cachethq/cachet) has 2.6k stars but I know that the amount of installs is actually far higher.
- jnbiche 11y agoFair enough, and to be clear, I'm nowhere near 2.6k stars, although the project has grown quickly with no promotion. The only reason I pulled that particular indicator out of the hat was to provide some reasonable benchmark for comparison. But perhaps it was not a reasonable benchmark, after all.
- e12e 11y agoI would think merged pull-request (aka community patches that doesn't suck) would be a better indicator on gh - at least for open source project. Fixed issues a close second.
- jnbiche 11y agoLinux and LLVM have no merged pull requests on Github. And a lot of these "Awesome x" (Node.js, Rust, Go, React, etc) will merge just about anything. Also, I know of several project that fix tons of issues, but are relatively unpopular (perhaps because they have so many issues?). I think the bottom line is there are no perfect indicators of open source project popularity.
- e12e 11y agoLinux has lots of merged patches though. That they don't have merged prs isn't surprising as they handle patches via email.
- SwellJoe 11y agoOur project has 295 stars and 1+ million installs. Stars is definitely not a reliable indicator.
- to3m 11y agoDon't know how common it is but I for one use the starring function to tag projects that I'm not using! This is so I've got a note of it to look at later. For projects I'm actively using, I'll either have a fork, or just a clone of the repo locally.