[Discussion] Are Bundlers Slowing Us Down More Than They’re Helping?
Posted by Stromedy1@reddit | programming | View on Reddit | 4 comments
I’ve been noticing a pattern across medium-to-large frontend teams:
- build times scaling out of control
- debugging becoming harder because of bundler magic
- deploy pipelines complicated for no real reason
- developers waiting instead of actually… developing
So I researched it deeply and wrote this article on why the bundler-first mindset might actually be killing productivity, and why many modern architectures are shifting toward runtime-driven frontends instead.
Topics I covered:
- Hidden cost of bundler-dependent architectures
- Why our build pipelines look like 200-step flowcharts
- What runtime routing / runtime module loading solves
- How big teams are avoiding redeploying entire apps for tiny changes
Curious to know what this sub thinks:
Are bundlers still worth the pain?
Or are we overdue for a runtime shift?
programming-ModTeam@reddit
Article is behind a paywall
walmartbonerpills@reddit
You shouldn't have to bundle to test locally.
R2_SWE2@reddit
That’s a tough spot to be in but it’s solvable now. Modern bundles do just-in-time compilation. I work on a sizable app and my vite HMR time is like 2 seconds (or less).
UnmaintainedDonkey@reddit
Paywalled.
Not sure what you are ralking about, as i could not read the article (please stop using/posting medium links). Bundlers for the web? Webpack? Vite?
They exist to solve a problem, the problem is basically to create a binary like youd get in Go. The solutions are bad, they are overly complex and do too much.
Then add transpiling. It get even more messy.
The solution would be for browsers to support more stuff out of the box, but there is no way this is happening any time soon.