What Will JavaScript Frameworks Look Like in the Next 10 Years?

Just to start off with: JavaScript will still matter in 2030. I don’t know whether it will have the dominance it has today (the boom in serverless architecture in particular is a huge threat to Node, and Node is a big part of why JS enjoys such a prestige position), but it’s far too ingrained into absolutely everything for it to just disappear. There’s still work maintaining COBOL in 2020, and—even if some newcomer comes along and sucker punches the entire JavaScript ecosystem into oblivion—there will be plenty of JS work in 2030.

I don’t think something new that does come along is going to blow JS out of the water, either. It’s more likely that many smaller technologies will nibble away at the edges, and that will gradually wear away at JS’s dominance. It’s hard to say what 2030 will look like, but I think JS will still be around, near the top of the pile, though starting to fray around the edges a bit.

It’s worth noting that Angular and jQuery together account for just under half of all JavaScript jobs going, and both of them have been considered on their way out for years now. While they lose ground every year, there’s still a ton of value in understanding them, and I think that’s a microcosm of what we’re going to see with JS over the next ten years: newer and better things might come along, but JS has so much momentum that it’s going to very hard for anything to pull it off-course.

This is all very broad, so let’s dig into the specifics.

React was always popular, but I think the last few years are when React got good. Really good. It was always a decent JavaScript framework, but the last few years of updates have really taken things to the next level; 16.4.0 added pointer events, and 16.8 finally added hooks as part of a stable release. I’m really happy with the React team’s recent work and I know I’m not alone there.

Whether or not this stays the case across the decade is going to depend on a lot of things outside the world of tech: there are rumblings in Washington about invoking the Sherman Act on Facebook and it’s unclear whether they are going to go anywhere. Facebook getting broken up could definitely knock React out of its top spot, though even then I think it would take a couple of years to really start to slump.

If Facebook is allowed to continue on as-is and no major changes happen in their ability to produce new software, I can’t see React going anywhere any time soon. It’s too big, it’s too good, and it’s better at adding new features than other big ticket releases like Angular.

JavaScript has not always been great about adding new features, but I think Typescript is reaching a point of popularity where JS is going to start thinking about cribbing from it. Once JS adds native features like better static typing, you’re probably going to see Typescript fall off a lot. I don’t think that’s going to happen in 2020, but by 2030? Absolutely. This is a pattern we’ve seen in programming a lot: language x has an issue, so developers create a framework or tool to fill the hole, it becomes popular, then the original language releases a new version adopting its features and the framework dies off. Typescript fills an important gap as of 2020, but once JS plugs that gap, Typescript is going to have to ask itself some serious questions.

I’m going to go out on a limb here and say something I’m not particularly happy about saying: I don’t think Vue has much of a future. I love Vue, but if it was going to have its moment it would’ve had it by now. It’s not the plucky newcomer any more, and hungrier JavaScript frameworks like Svelte are nipping at its heels. There’s probably a reality where Vue is the dominant JS framework, but in our reality, React came along and steamrolled everybody. Unless something extreme happens that makes React less viable in the future, like Facebook shutting down or getting broken up, it’s not likely to ever overtake React in popularity.

It’s hard to say whether Svelte is going to become the next big thing. It’s certainly a wonderful little JavaScript framework, but Vue is wonderful too, but it struggles to compete with React’s FAANG money; it’s going to take a lot of talent for anybody to unseat Facebook, and Facebook can almost certainly pay that talent better. It’s a sad fact that smaller open source projects—no matter how valuable—often struggle to get the industry penetration of releases from major companies. I want to see Svelte succeed, but I’m not at a point where I’d put money on it.

I’m lumping these JavaScript frameworks together because they’re all pretty similar at the end of the day: they’re pretty unpopular these days, but they’ve achieved a level of penetration that makes them hard to avoid. jQuery and AngularJS were as ubiquitous in the past as React is now; our teams still do a lot of maintenance work on both, and I don’t see that changing any time soon. I’ve talked about momentum a lot today and that’s for good reason: technologies definitely get supplanted, but the big ones have a habit of sticking around for a very long time afterwards.

Here’s my wildcard, but TensorFlow now takes JavaScript, and I think it’s going to change JS in the same way Node did in 2009. AI is going to be the big gamechanger in the 2020s, and TF is the best tool we have right now for building neural networks and machine learning algorithms in JavaScript. It’s a good tool, too: I’ve only dabbled, but it’s robust and versatile and I can’t wait to get my hands on it a bit more.

I don’t think AI and ML are quite there yet, but they’re definitely not far away, and I think their accessibility and use are going to explode over the next ten years. TensorFlow is open-source, it has a lot of talent and funding behind it, it’s got C++ and Python folks working on their own projects that you can take inspiration and get advice from. 

Right now the main barrier is how complex neural networks are, but I can see that barrier coming down. We see that a lot in tech: something like containerization existed for a very long time but there was a high skill barrier, then Docker came along and made it accessible to more developers and suddenly containers are everywhere. What AI/ML needs is a Docker moment that puts it in the hands of everyday developers, and I could see TensorFlow.js being part of that.

Ten year predictions are hard, and ten year predictions in something as volatile as JavaScript are especially hard. Node barely existed ten years ago, and it took JS in directions none of us expected. There’s always going to be a lot of hedging when you’re trying to make this prediction, because it’s impossible to tell which new technology, currently on some undergrad’s laptop, is going to blow the whole thing sideways.

That said, I’d bet heavily in AI/ML, and I think TensorFlow.js is likely to be very important in that change. I also think—if you hadn’t already—that learning React is a highly useful job skill that you’re going to be able to leverage for a long time. The 2010s were a decade of squabbling, where various JavaScript frameworks fought each other for dominance, but React seems to have well and truly won that fight and I suspect the 2020s are going to be more of a monolithic decade, dominated by a smaller number of larger technologies.

Overall, I think it’s going to be an interesting decade. The JavaScript surge has begun to slow and mature, and that means a decline is somewhere in our future, but I think it’s a ways off yet, and investing in JavaScript is still as smart an idea as it was in 2014. I’m not sure I’ll be giving the same advice in 2030, but for now, my money’s still on JS.