This is it. This is what writing javascript is like in 2020. It's not framework fatigue. It's none of the parts fitting together anymore, and all the defaults being ones to support legacy systems.

http://lea.verou.me/2020/05/todays-javascript-from-an-outsiders-perspective/

And all the blog posts you can find are from developers building what will eventually, hopefully, become the status quo but right now is terribly broken or wide-eyed code-school grads trying desperately to write relevant content to seem competent enough to get a job.

In either case none of it reflects the industrial reality that everything is terribly, terribly broken.

I'll admit I'm part of this problem. I'd rather use the new stuff, the status quo I'm hoping is coming, than deal with the legacy. At points I feel justified because the legacy is broken too, in its own ways.

I don't know how to fix it.

Honestly, I think part of the appeal of deno is that it's not compatible with node. It's _hard_ to port. Lots of tools are missing. It'll be a greenfield hell for a while.

Maybe that's what we need: make backward compatibility a shim, not default.

Maybe this will make it clear when something is not yet solved. The answer to "can I use X yet?" is actually no. It'd be nice if it _looked_ like no when you investigated.