We're all waiting for the thing that will replace Webpack so we don't have to keep using it. Maybe we should just go and make it.
And most importantly as for me, developers who works full-time on such projects can sustain the higher quality and improve it over time and if you chose it for now it may been relevant year later or 2-3... I hope at least.
It's easy to complain about something without offering an alternative. You just shot down every tool, so let's see this impressive tool that you surely must have in your head.
As a tool becomes sufficiently generalized, it always picks up what looks like cruft in the simple cases. But the simple cases cannot tell you how good or versatile a tool is because you can solve a simple problem with virtually anything. That's why people who need to concat 2 files think you can replace all of Webpack with `cat`.
The cycle continues when someone creates a new tool with a simple API because it doesn't generalize, but then they realize they need to extend it such that it generalizes over the real-world problems that people have when bundling, like hot reloading (different than live reloading) and code splitting and arbitrary transformations.