Unabridged Conclusion to the State of JS

I was invited by Sacha Greif to write the conclusion for the State of JS 2021 survey, which was just released today. I felt it was a huge honor to be asked so I went ham on the “financial analyst” side of me to interpret the results for readers!

Ultimately though, I went way too long and a lot of it had to be cut, mostly because of the need to translate these words into 20 different languages.

Still… I already did the writeup, so here’s the full text!

JavaScript is in a tremendously better state today compared to the first survey in 2016. Back then, only 21% of you used TypeScript, compared to a nice 69% today. Where in 2016 we used to joke about a new frontend framework releasing every day, React has now held the top spot for 6 years running. The percentage of people reporting that “JS is moving too fast” is down from 59% then, to 38% today.

This is not to say that JS has stood still. Due kudos must be given to the champions of the Optional Chaining and Nullish Coalescing APIs, which have rocketed up to be by far the most used new JS APIs in our survey, despite only being 2 years old. The 6.2% and 13.8% of you who have never heard of them should probably give them a try! More good things are on the way from TC39: you reported that date management is a top missing feature and the Temporal API recently reached Stage 3.

Over in user-land, Solid made a solid debut tying with Svelte in first place for the coveted top spot in frontend satisfaction. By usage, Vue.js, now on version 3, is on track to overtake Angular as the second place framework (after React) sometime in 2022-23. Backend/Fullstack frameworks have seen much more innovation, with SvelteKit, Astro, and Remix all debuting with very high satisfaction. However, real usage changes much slower than satisfaction, with Express still the undisputed king and Next.js the distant but fastest growing second place. The difference in pace between front and back can be explained by developer happiness: 66% of you are happy with frontend frameworks, but less than 50% are happy with backend ones.

In terms of build tools, 2021 has been the year of Vite. Vite not only debuted at 98% satisfaction, the highest scoring library across all of JavaScript, but usage is already at 30% in its first year on our survey, eclipsing esbuild. Vite isn’t alone in trying to improve the state of JavaScript Build Tools - 8 of the 12 tools we track today are less than 2 years old! Also notable: for the first time, new entrants like Deno, Rome, swc, and Snowpack are now supported by venture-backed companies, perhaps charting a viable alternative to the fragile open source funding environment.

As the Third Age of JavaScript enters its third year, the opportunities for aspiring open sourcerors and entrepreneurs continue to unfold:

• In our first year of tracking monorepo tooling, pnpm, Turborepo, and Nx took the top spots, but only 25% report being happy with the state of monorepo tooling.
• Looking at runtimes, Deno is stable at 5.6% usage but faces fierce competition from the various Serverless and Workers platforms.
• Teaching online continues to be a source of incredible opportunity, as most of you report self directed learning, free courses, and paid screencasts and books to be your top choices in learning.

Lastly, it’s worth mentioning our top write-ins, as it can help correct unintentional biases. This year’s top write-ins are Elm, Stencil, and Platzi. If your favorite library or resource didn’t do well this year, keep spreading the word!

After all, last year’s 3rd place write in was… Vite.

I was also briefly on the State of JS results livestream, which you can see here