Consider removing Internet Explorer support

[nickserv]

Its official support is ending, so we could probably save a bit on Babel resource usage. This would technically be a breaking change though, not sure how we'd want to deal with that.

[Methuselah96]

Would love to finally get a v5 out there with some cleaned up types. This could be a good ~excuse~ reason to do so.

[markerikson]

@nickmccurdy There's barely anything IE relevant in this codebase :) Closest I can see looking at a build artifact is:

  for (var _len = arguments.length, funcs = new Array(_len), _key = 0; _key < _len; _key++) {
    funcs[_key] = arguments[_key];
  }

I agree it'd be nice to put out v5 at some point. My biggest questions are still around ecosystem compat with the types.

[nickserv]

I'm mostly referring to how the Babel/browserslist config explicitly support IE in the build. Is that explicitly tested though?

[markerikson]

Yeah, I'm just saying that Redux's code is so minimal, and it's so FP-oriented, that there's almost nothing in the current code that would matter for IE beyond a couple bits of ES6 syntax. So, sure, we ought to remove that setting in the next major. There just won't be much of a size difference because it's a small lib and there's only a couple tiny bits of syntax that would be affected.

[nickserv]

That's fair. Would a draft PR to test changing the build, or would you rather wait until the types are more ready?

[timdorr]

We can just do it now and see if there's any notable impact. If it doesn't actually do anything to the bundle (and I suspect it may not), it probably doesn't matter enough to bother.

[markerikson]

Dead in the 5.0 alphas!