Echo JS 0.11.0

<~>
pluma 3400 days ago. link parent 1 point
I think the examples are not very useful. I don't think many people make high-level tool or framework choices based on arbitrary metrics -- APIs are far more important than GitHub stars.

But this might be useful when trying to find a library where the APIs are mostly irrelevant (e.g. because you just want to solve a specific problem), e.g. comparing lodash vs underscore vs ramda (vs ...). Or even comparing various different packages providing implementations of underscore's/jQuery's extend function when you don't want to add a major dependency like underscore or jQuery just for that function (my current preference is the `extend` package -- the obvious name certainly helps).

I've been in that situation several times -- trying to decide between multiple nearly identical packages and figuring out which one is the safest bet. However I think the most time consuming part is *finding* the different alternatives, not comparing their stats. A difference between 2M and 3M downloads barely matters while a difference between 10k and 10M certainly does.

Replies

dcohenb 3398 days ago. link 1 point
You Are defiantly right.

One of my main goals is collecting information about what packages are actually related to one another, that way you can say "I use underscore, have anything better for me?"

That kind of a discovery tool would be very useful if you ask me and it will be available as soon as i collect some more user searches.

Any thoughts?