The Future of Svelte (Interview with Rich Harris)

- Hey everyone, I'm Lee, the Head of Developer Relations for Vercel, and I'm joined today by Rich Harris. We are going to talk about all things Svelte, all things Rich and how he ended up here on the Vercel team. So Rich, do you wanna kick things off by introducing yourself? - Sure, I'm Rich. I've been at Vercel for the last three weeks. Prior to that, I worked at The New York Times and I've been working in journalism my entire life, so this is a big, exciting change for me. - Well we're really, really excited to have you here, and to kick things off, a fun trivia fact

that I just recently found out, I didn't actually realize that you also created Rollup, which is something that I had used in the past, so I was thinking about this a little bit and I wanted to ask you, what was the first popular open source library you created and what was that transition like between, I'm just gonna build this little project for fun, just to see where it goes, and oh wow, there's actually a decent chunk of people using this. How does that change the dynamic

of building an open source? - Yeah, I mean popular is definitely a relative term, an open source, but the first big project that I worked on, was actually Svelte's predecessor, it was a thing called Ractive, which I released the first version of in 2013. It actually came out about the same time as React. I remember I was working on it, I was getting ready for a public release, and the React announcement appeared on Hacker News like a month before we were ready, and I remember thinking, ugh, there's no point now,

what a waste of time. But launched it and it had a pretty good reception. React was not yet the thing that everyone had to do, and so the market hadn't decided yet, and so for a long time, I sorta harbored this fancy that Ractive could be a competitive project to React, and I really love that project, it is my first big open source endeavor, and I poured my heart and soul into it,

and it got users, and what is kind of hard to see as an open source maintainer is that, or sorry, if you're not an open source maintainer is that a project doesn't have to be huge in order for the amount of activity on the project to become overwhelming for a maintainer. And so this thing that I had spend, I was spending all my evenings and weekends on, it was starting to get users, it was even starting to get used by companies, like at one point we were used by the Yellow Pages

and Delta Airlines and other people like that, in addition to my own employer. It was becoming kind of unsustainable for me to work on, to the point that I got eventually just totally burned out on it. And I think that was my first introduction to the realities of being an open source maintainer, and it's what sort of taught me how to deal with having users and also how to approach problems of like,

how do you get contributors on board? And how do you get them to think in the same direction and work on the project with a unified front? And also how do you say no to feature requests? Which is a huge thing. As soon as you start getting any level of adoption, you need to be able to start turning around to people and saying, actually, it's great that, you're excited about this project and it's great that you're contributing, but that's not what we're gonna do for the following reasons. And so, in every way it was a baptism of fire

in the inner workings of open source, and it laid the foundation for Svelte several years later. Although, as you say, Rollup came in between those two projects. - That's fascinating. I actually didn't know a lot of that history, so that explains a lot I guess. Then, so you've been thinking about building front ends for a long time now. I mean, that's, it's going on almost 10 years,