In this post I'll tell the other side of the story. The parts the React Documentary failed to mention.

The recent React documentary was not a good look, but you wouldn't know it looking at the comments on Youtube. Some may refrain from posting anything critical, fearing retribution by members of the community. It's a real thing. Haters gonna hate.

An ethical way to make a documentary is to be objective, but the narrator is clearly crafting a narrative that is biased toward React. Why do we even need a narrator? Are we caught in a cult of personality? The documentary is a warped take on reality. There is no counterpoint, a complete lack of diverse perspectives, so we don't get the complete story.

Who am I to tell you anything about React? I'm not a React maintainer. I'm a web developer that used React in an enterprise org, someone who taught functional programming with React to others, implemented a large React UI library, and coded a React app. I've been developing for the web for over 20 years. At least I'm not begging you to subscribe to my Youtube channel.

What's the problem with React?

Churn

My time with React involved engineers migrating class components to functional, enzyme to react testing library, fixing issues with build configurations, tracking a multitude of dependencies, patching security exploits. Time that could have been spent coding features and enhancements wasted. The job of an engineer at a corporation isn't to merely maintain the codebase, our job is to deliver value for the business, to delivering features and bug fixes for customers.

After many months of work, React finally upgraded their documentation. If you haven't already migrated your class components to functional, you soon will because in the React docs class components are now deprecated and will be removed in a later version. This is just one example of the churn.

Several developers first learned how to code React using the help of Create React App, but probably found they had to roll their own build with Rollup, Webpack or Vite while working at a corporation. There is no mention of how to bootstrap client-side React apps in the new React documentation. Create React App is gone from the docs, probably for good reason because the project was unmaintained. React doesn't even recommend client-side development in the new docs, favoring several server-side implementations including Next.js, Gatsby, Expo, and React Server Components.

If the recommendation is now all React development should be done with a meta framework, the problem with going all-in server-side is that some front end engineers at corporations have no control over the server. I was one of these engineers. No argument could pull Java away from the stack in favor of Node.js, an often more performant tool that could have even saved the corporation operational costs. React has effectively abandoned engineers like me. We have no guidance for bootstrapping a React app solely for the client.

This is what happened to me after trying to convince others at a large enterprise organization to move some logic to the server.

This problem of valuing DX over UX isn't unique to React, other libraries and frameworks suffer from the same thing. What's possibly unique to React is revealed by that famous quote from Mark Zuckerberg. "Move fast and break things." I'm sorry, but at this won't fly at other corporations. It's not like React even gets the DX right. DX is a hot mess, especially when multiple parts of the ecosystem are ever changing.

So What?

Just because React is popular now doesn't mean it always will be. I think we already hit peak React, but haven't seen any conclusive analytical evidence to validate that, just a curve trending downward the last couple years in the State of JS survey.

The job market is heavily React, but it won't always be this way. A decade ago, before React even gained popularity, KnockoutJS jobs were a rare opportunity in a sea of jQuery. It's happened before and it's happening again.

There is a great sea change happening in web development. Several libraries are springing up to support UI component development with Web Components. Frameworks like Qwik are changing the game by removing hydration and moving a majority of work to the server. Change is happening now. React engineers rode a wild wave that is about to come crashing down.

By the way, I realize jQuery is still around. I just dipped into a codebase that uses jQuery. React will most likely be around in a decade. I'm not saying React is going away entirely, I'm saying React in ten years will be regarded like jQuery is today. The two libraries have similar faults. jQuery’s faults were a lack of framework and failure to coordinate the open source community. React repeated similar mistakes and how everyone in the community has to cope with those mistakes. It’s time to start preparing because React is in decline. That’s not to say there won’t always be novel use cases for React: React Native or even using Virtual DOM to provide an XML like interface for complex libraries like three.js. React will still be around, but not nearly as popular as it is today.

How To Cope With Change

If you hope to make it for a long time in this profession, it's time to start learning concepts outside of the React ecosystem. Make sure you learn the fundamentals that apply to every library and framework. That means HTML, CSS, and JavaScript. Go beyond functional programming to understand prototypical inheritance and Object-Oriented programming, which also have applications in front-end development. Take some time to learn new standards landing in the browsers. Learn TypeScript, because it's here to stay, whether as an extension of the language or at some point, types baked into the specification. Learn how servers work, http, WebSocket, and so forth. These fundamentals help you fix issues, contribute to open source, spur innovation.

It's time to start thinking, "OK, how would I do that without React?"

P.S. If you have had a great experience with React so far, I’m happy for you. This post is more about how engineering decisions effect a business. This isn’t an attack on your choice of using React. On the contrary, this is a warning. The ecosystem is too brittle to survive, causes too much churn for businesses that have invested in it. React is the victim of it’s own eventual demise, much like jQuery. It’s coming, so be prepared. If I’m wrong in ten years, I’ll admit I’m wrong, but have a hard time believing history won’t repeat itself with so many signs it already is.