When choosing a library ton include in your project, which parameters do you take in account?
Do stars matter? Yes, of course.
How I proceed?
As for me, I pay much attentions to good design practices. Like moving from React to Vue.JS
Age, stability, and "community standing". The last one is github stars, whether the project is hosted by Apache, Google or Joe Blow, repo activity and all the rest.
Of course, what really matters is code quality, but who has the time to go through all the code.
And any library can be swapped out. It's frameworks that are really important, and those decisions are usually harder to make...
Stars, forks, most recent commit, number of open and closed issues, number of open and merged PRs, number of contributors... all forms a picture.
Also... quality (or in fact existence) of documentation; number of NPM downloads if it's a package; general level of community activity (hard to judge accurately - but you want to see some level of interest for basic google keywords).
If it's a big enough choice like a framework - look at developer surveys, popularity of the stack, availability of training resources... it's a bigger choice so do more background research.
But for a small library, a reasonable amount of NPM downloads and github activity is usually enough - I want to see some evidence that it's still actively maintained; and whether issues are getting resolved.
Whilst it can give you an overall impression, I find more often than not that just because something is popular, that doesn't necessarily mean it is any good. Just look at 99%+ of the talentless tone-deaf auto-tuned hacks that make up modern pop music -- or even the history of pop music as a whole. Some real CRAP has made it into the Billboard top 100 of the past 50 years!
Hence how mind-numbingly dumbass bloated broken BS like jQuery and bootcrap have mouth-breathers singing their praises out of ignorance and incompetence.
To that end I use factors like how often it is updated (though that's NOT entirely a deal breaker), how well it is written by actually looking at the code (that is an immediate deal breaker for me), as well as where/how it is hosted. If the project is one for web technologies, and it has its own website, and that website is poorly written bloated crap that screams "I need to learn HTML", I'm not going to trust that they have any business telling me how to do something online.
queue broken record --
As I'm always saying, see garbage like this:
<nav class="navbar navbar-toggleable-md navbar-inverse fixed-top bg-inverse">
<button class="navbar-toggler navbar-toggler-right" type="button" data-toggle="collapse" data-target="#navbarsExampleDefault" aria-controls="navbarsExampleDefault" aria-expanded="false" aria-label="Toggle navigation">
<span class="navbar-toggler-icon"></span>
</button>
<a class="navbar-brand" href="#">Navbar</a>
<div class="collapse navbar-collapse" id="navbarsExampleDefault">
<ul class="navbar-nav mr-auto">
<li class="nav-item active">
<a class="nav-link" href="#">Home <span class="sr-only">(current)</span></a>
Where if you don't know what's wrong with that, do the world a favor, back the **** away from the keyboard, and go take up something a bit less detail oriented like Macramé .
If they cannot handle something as simple as HTML, I'm not going to think that anything they wrote anywhere else -- like PHP, JavaScript, Python, Perl, Ruby, whatever -- is going to be worth a damn!
To that end, what other users think ends up pretty damned far down my list! Wizard's First Rule...
-- edit -- ok, why is hashnode screwing with code blocks stripping out linefeeds? Lemme guess, related to the stupid malfing font change?
When I need a library, I do some search, trying to collect 2-3 candidates. After that, here is my list.
Stars represent attention, but most people simply star projects they use; it doesn't necessarily mean that everyone who starred will contribute by any means.
No, stars don’t matter.
I’m mostly late to the party at these things like stars on GitHub but as far as I understand it, it’s nothing more than a like button. Please correct me if I’m wrong.
When I consider a library then I look on the activity in the repo. Are there issues? If, are there conversions going on and how old are the issues? Are there PRs? How old and how long did it take to merge if merged at all (seen half year old PRs that didn’t mergen despite no issues)
That gives you a feeling what’s going on there and what chances are things get fixed when something is coming up.
Stars matter, and due to that they have mass.
Leaving the pun to the side though, I do take into consideration how many people are using a particular library, as the common thought is that the more eyes that are on a library, the more likely someone is going to look at the code they're using and catch a security issue, or the more likely the library will be maintained well into the future.
Mark
formerly known as M
Mark
I'd say yes. When choosing a library, I'll look at:
Other things, like automated tests and having multiple frequent contributors, are a nice bonus. I don't really check if the code is well-written at first inspection, I probably should though.
In conclusion: if you use a library and like it, leave a star.