DEV Community

loading...

Discussion on: The Shocking Immaturity of JavaScript

Collapse
zaydek profile image
Zaydek MG

I really appreciate you writing this article, and yes I can understand your hesitation to write it in the first place.

What I will say is that I’ve becoming increasingly enthusiastic about esbuild (esbuild.github.io) because it seems to be a real way out of the chaos while providing backwards compatibility. I was somewhat optimistic about Deno, but at the same time Deno is a breaking change oriented kind of technology, and I’m not 100% on board with that.

Anyway, yes. Frontend I think is plagued with too much hype, but I think that is also one of its best qualities. A lot of developers get their first footing into the world of programming because of how accessible JS is, but that also makes JavaScript, for better or for worse, more susceptible to lower-quality code / maintenance. Like you, I’m not pointing anyone or any project out, but I think it’s a symptom of the bigger picture that a lot of devs get into programming because of frontend. Me included.

I wish I had more to contribute other than simply writing a reply. But I am grateful you’re acknowledging one of the dirty truths about frontend-land.

Collapse
jaredcwhite profile image
Jared White Author

To be fair, I'm not letting JS backend off the hook either 😬 …but at least with backend tech there are numerous ecosystems to choose from. On the frontend, JS is basically it. Anyway, thanks for the reply, and I agree there are lots of interesting projects springing up utilizing esbuild that seem to promote speed and simplicity as core tenants. Encouraging stuff.

Collapse
zaydek profile image
Zaydek MG

I can’t handle JS backend. I mean, for small things, yeah -- why not. But for large things, why punish yourself? To each their own but JS backend is definitely not for me. I’m barely happy enough to stick with JS frontend, and that’s only thanks to the ecosystems that have been propped up to make it manageable.

I think frontend has suffered from a lack of technical competition. Framework competition isn’t meaningful in the same way language competition is. And unfortunately I don’t see WASM changing the game (because of DOM overhead). I would love to be wrong here, but I think frameworks will still be the way forward for a bit.

Dart / Flutter seemed to offer a way out -- by thinking about more reconcilers than just the DOM -- i.e. canvas, but for the time being appears to be underpowered and not ready for primetime.

Thread Thread
jaredcwhite profile image
Jared White Author • Edited

What makes me feel all warm and fuzzy is just how good browsers are getting at the core level, so our frontend code can become simpler and sit closer to the "metal" — esbuild being a showcase for that trend. After all the crazy framework wars of the 2010s, it feels like for many kinds of sites you can pull in a few simple libraries, write some web components, use a great backend with reactive HTML-over-the-wire techniques, and get a fantastic UX out of it all. That's my personal "escape hatch" anyway…I expect it'll take a while for the industry as a whole to embrace some of these trends.

Thread Thread
zaydek profile image
Zaydek MG

I’m somewhat open-minded to the whole server components approach but I also don’t like the idea that we suddenly need to be implementing and deploying servers rather than stateless assets to the web (assuming I understood this right). And I think web components had a chance to help resolve a lot of these woes but ultimately the SEO concerns I think outweighed whatever ergonomic benefits web components offer.

I really like how Svelte can be used as a compiler to generate web components and I think that was an incredibly smart idea for how to macro fix a lot of these systemic problems.

I really think that static site generators that are not JS-heavy is really the right approach for most sites. But once you want to develop and interactive frontend app, the water gets murky quickly because of just how much more tooling / overhead is needed. I will say that writing React in 2020+ is much more accessible than it felt to me before 16.8 (hooks) but I still think the experience overall could be improved.

Thread Thread
faraazahmad profile image
Syed Faraaz Ahmad • Edited

I really hope WebAssembly enables competition in the frontend for JS to get its stuff together, and ultimately gives everyone better tools across the board. I feel like something like Svelte and using your own preferred language in it that compiles down to webassembly could really improve the frontend DX

Collapse
hemiltherebel profile image
Hemil Ruparel

Exactly! Throw in a bunch of noobs and be ready to pay hefty prices to senior engineer/consultants to fix the mess they created (assuming you last that long of course :P)

Forem Open with the Forem app