Based on a recent talk I gave, in this blog post I'll discuss my thoughts on the current state of JS and what we might see over the next 12 months.
I made 8 predictions on what I think we’ll see in the next 12 months, and most of these are influenced by the three core goals of ECMAScript 2015 (formerly ES6), which hopes to provide a better language for:
These are by no means the most bold of predictions, more so thoughts on what I think will happen in the next year or so.
Babel is slightly different in that all the new functionality it provides is part of ECMAScript 2015 or beyond, so everything it implements in theory will eventually be in the browser. Going forward, Babel, TypeScript and ClojureScript will probably be the three I’d back to become even more popular.
Alongside the larger, fully featured frameworks of Angular, Ember and others, there’s a myriad of smaller libraries that focus on doing one thing, and doing it really well. You could even argue that ReactJS is a good example of this; as a library it provides just the view layer for an application, and nothing more. Given that npm provides a (relatively) easy way to install and manage all these libraries, I think it will become more common for developers to construct their own stacks of small libraries that can be swapped in and out, over using a large framework where you’re stuck with what it provides.
Related very much to the previous point, I think that there will be an even bigger focus on the development and release of libraries that exist to solve one problem, and do it very well. This feels like a natural process as an ecosystem matures, and we figure out the best solutions to new problems (such as client side “MVC” approaches). Why write an entire framework when you could write a small library to plug the one problem you need to fix, and then couple it with some other libraries that provide the rest of the functionality you need.
The previous two thoughts might make you think that I’m predicting the demise of Angular, Ember and so on. This is definitely not the case. There will always be (and quite rightly so) a use case and need for these larger frameworks.
More and more developers will use compilers like Babel, but they won’t do it by directly installing and running Babel. Most will use it through some other system like jspm or webpack, generic tools that abstract away the compiling step and provide all the functionality you could ever need.
There are a lot of benefits to running the same application on your client and server. At GoCardless, we just launched the new gocardless.com, a ReactJS application that runs on client and server (we blogged about how we did it) and it’s gone really well. I expect that tools will grow out to serve this demand and that the approach will be refined over time.
Tools like Traceur and Babel (initially called 6to5) initially existed to let us write ES2015 ahead of it being fully supported across browsers. However they have since grown to support upcoming features of ECMAScript7 and beyond. I can’t see a time where we won’t run our code through something like Babel, because by the time ES2015 is fully implemented, the next version of the language will be well under way. In fact, this is a good thing, because it should let new proposed features be tested by developers before they are implemented. The feedback loop should be quicker as a result of people writing ES7 way before release and that can only be a benefit to everyone involved.
Framework booms are to be expected when a new approach to web development comes along. The swap to client side applications really began with BackboneJS, before many others came along. Every week it felt like a new framework hit the internet but recently to me it feels like that’s slowed down a little. Angular and Ember have emerged as the two most popular options, but we’ve not seen as many new options really catch on. I think as we’ve figured out the best approaches for building and architecting client side applications, we’ve picked out frameworks and stuck with them. That isn’t to say another framework couldn’t come along, but I’d be surprised if in 12 months the focus isn’t still on the frameworks that we’re using at the moment.
They’re my thoughts on what we could see happen over the next 12 months or so, and I’d be keen to hear what everyone else thinks. I’m pretty sure that I’ll get things wrong, too! There are also other things I’m interested to see more of once they are released, including Facebook’s work on Relay and GraphQL. It’s hard to say much about the tools when they have yet to be fully open sourced, but I’ve no doubt they will have an impact when they are.
My thanks to Max Murdoch for his reviewing of this post.
If you enjoyed this article please feel free to share it on Twitter.