Using Phoenix issues for a quick estimate on the amount of npm/brunch/webpack support required over the years:
Issues — 2,034
NPM — 591
Webpack — 79
Brunch — 171
Around 30% dedicated to JS packaging. This is a big win.
I say this in the fairest way possible after years of support & churn – I now consider placing the burden of node/npm/webpack on new users as actively harmful. Also, long-term reproducible builds are essential for maintainable software & node has not stood this test of time /1