The noise around new tools

There is a recurring lesson in web development that many of us relearn every year: the tools dominating community discussion are not the tools most people actually use. Feeds and blogs create the impression that everyone has already moved on to the newest framework or state library. State management alone demonstrates the churn: Redux, then MobX, then Unstated, then Recoil, then Jotai, then Valtio. Following this can feel exhausting, as though you are permanently behind your peers.

The reality is different. The vast majority of web developers work with what could be called boring or old technology. The intuition behind this is simple: most of what exists on the web today was built before today, and maintaining it means working in whatever stack it was built with.

“The best bet for 2030” by CommitStrip

Why the old stack keeps its share

The key observation is that we all spend more time maintaining old things than building new ones. Someone experimenting with auto-reloading serverless static deploys is outnumbered by people still updating a Grunt configuration. A developer who is all-in on Tailwind 2.0 may still be maintaining a Bootstrap 2.3 site, and a JavaScript expert who switches state libraries weekly may still be maintaining a large application built on Flow.

Raw adoption numbers, such as the share of sites running WordPress versus React, do not settle the argument. The mismatch is about perception inside the community: one technology feels old-hat while the other does not.

Choosing dependability over novelty

Old technology has real virtues. It can be solid, dependable and predictable. There is a time and place for trying new things and telling others about them, but there is also a time to reach for the technology you know so you can get things done.

The automatic assumption that "nobody really uses $foo anymore" is persistent and will likely survive well past 2021. It remains wrong for whatever value of $foo you plug in: plenty of people are still using it, and it is still valuable.