Svelte: Why You Should Take a Look


I’ve been working with design for the web since 2006, and I saw the frontend born and grow. Back there, the hardest part of the web development was the cross-browser part, because we had to handle a bunch of different browsers and the weirdest one, the Internet Explorer 6.

Of course, we didn’t have any fast javascript renderer at that time yet, so we were kind of limited on that part. Luckily things grew fast, and a little time later, we’ve had jQuery, Ajax, and a new frontend era begin.

Today we have a lot of amazing frameworks, lots of fantastic browser features on a new fertile terrain to explore, and that cross-browser complexity almost disappeared.

But, as I worked on this new era, using some of these modern frameworks, I started to think that we’ve lost the most significant part of the frontend purpose — the user.

These new frameworks are amazing, but this is just tools to achieve some goal. And I feel that at some point we stopped to discuss the front architecture supporting the user experience and started to consider it to support our tools. Not entirely, but I saw teams using the most of the time to tuning tools instead of tuning the user experience.

I’m an enthusiast, and I already have a part of a system wrote with ReasonML, and that part never broke in production, and I think this is amazing. Still, I believe that frontend development must be simple, easy for anyone who wants to test some usability fast.

We are on a stage where we need to split the frontend development from design because our tools are too specific and complex. We need high trained developers to work with it, and this creates a layer of complexity that can make our user experience development really slow.

For a long time, I’ve had this feeling about frontend, participating on plannings and thinking that we’ve maybe gone in a strange direction. I just couldn’t understand this feeling entirely until I saw a Rich Harris’s talk about reactivity and Svelte.

The frontend Goal is not to connect with the backend or make the developer’s life more comfortable. The frontend goal is always the user experience and everything around it is just a means to an end.

I’ll not cover any technical development with Svelte on this post, because everything that I could write at this point is already written. I don’t know if Svelte will be the next frontend tool, but I believe that its foundation and philosophy is what we need on frontend development.

Do you a favor, and watch the Rich Harris talk: