StrapiConf Fireside Chat with Pierre Burgy and Guillermo Rauch

I'm pleased to introduce Guillermo Rauch, creator of Next.js and co-founder and CEO of Vercel, to discuss the future of web development. Guillermo, thank you so much for being with us today. Could you give us a quick intro and story on how you ended up working on Next.js and starting Vercel?

Absolutely, I'm really glad to be here. So, uh, funny thing is, 4 years ago I knew that I wanted to build and enable developers to build better websites and build a company around that. But when we're working on the website for the company, which at the time was called Zeit, it was really hard, believe it or not, to even build a page and publish it online that was taking advantage of all these innovations in the frontend world, like React components. React components came out of Facebook. They were really great for building widgets, like an autocompleter type widget, but it was really hard to build entire pages and applications with these innovative technologies. So even for that one page that I needed to publish online, which was our main website, which it was a startup at the time, just one page, it was really hard to get it started on two fronts. One, with a framework that made it really easy to generate pages and publish them online and making them reliably autoscale fast, but then that other piece of like the deployment platform. I wanted to just push and self-serve to all this incredible infrastructure that was all those innovations that were happening in the cloud. I didn't want to spend a year setting up a Kubernetes cluster infrastructure, having to monitor it and so on. I was thinking more like this should work more like a CDN. So those two things led to the great of Next.js. And the Vercel platform. And the target developer was the frontend developer that was innovating on, uh, with these new tools, technologies, frameworks, and creating innovative experiences like on top of a headless CMS like Strapi.

Thanks a lot, um, and congrats on this amazing achievement. Next.js is now the fastest growing JavaScript frontend framework according to the 2020 JavaScript Rising Stars. With a recent $40 million Series B led by top VC, you've done pretty well for yourself, but in your mind, what has been your greatest accomplishments?

I think there's been a number of things that I'm really proud of that we've accomplished with our team. Uh, it's important to say that even though I did have influence over the initial ideas of Next.js, we hire a ton of people from our community. We worked a lot, but now people call it working in public because of the open source nature of our projects like Strapi and Next. I feel like we've always been working in public. So I think one of the things that I'm really proud of is having built a team that came from that open source community that care very deeply about these problems. We didn't just recruit engineers that were, I don't know, working on, you know, backend or other kinds of things, Java developers, we're just working with our community and building this incredible project. And actually I would say like more concretely something that when I think about Next.js and I look back, one of the things that I'm proud of is that we stayed minimalistic. So we stayed, we didn't add tons of stuff. We could have added, we could have turned it into like a full stack we could have built a database for, we could have done so many things, but we chose to be very mindful of the surface of the problems that we're gonna take on to have an opportunity to excel. And that's panned out really well for us because for example, when we introduced Next.js 4 years ago, it was really focused on server rendering. And then we realized like developers wanted to have the power to choose static or dynamic. With great flexibility. And then we were actually able to add capabilities for that. I think about an alternative universe where like we would have bloated it with lots of APIs and lots of features and wouldn't have been able to respond to market needs and customers' needs as fast as we've been able to.

Yeah, I definitely agree. I mean, hiring from the community is such a amazing things to do, to build a team of passionate people and being focused on what users want is definitely the way to build something that is extremely helpful for thousands of people. So I would like to talk a little bit about the CMS industry. So headless is pretty much mainstream by now with traditional CMSs trying to reposition and rebrand themselves as headless. In your opinion, why is the headless architecture paradigm becoming so popular and where do you see it going next?

Yeah, I think there's— with revolutionary technologies, what I find is that we can't just always pinpoint to a specific reason. And what I— even reflecting on the success of Next.js, I always find myself, you know, finding all these waves that were coinciding at the same time. So for Next.js and Vercel, The rise of frontend, the rise of React, the rise of serverless architectures, like the rise of static-dynamic hybrid architectures were all reasons for its success. And actually would even say the rise of decoupled architectures is a huge reason because I mentioned earlier, I could have been tempted to bind Next to MySQL, right? Like that's how WordPress works is that you install it and there is an assumption it is bound to a specific database. With Next.js, because we knew that technologies like headless CMS were rising, we said, you know what, we're not gonna build that. We're gonna really specialize on the frontend side of things, work on performance, work on giving developers the best tools for really great Web Vitals and Lighthouse scores and all these things, flexibility. I could even say more directly, the success of Next is because of the success of technologies like Strapi. Like we didn't have to build that. Now on your side of the world, you're not only empowering technologies like Next.js, but imagine that you go to a company that has hundreds of engineers. And I think one of the common technologies that is being built with Strapi And Next is a very popular device manufacturer. And I could imagine this company, and I know has hundreds of front-end initiatives and they're serving multiple kinds of devices. They can't just lock themselves into a technology that is coupled. This is okay, the CMS is coupled with the front-end, even if it gives you a short-term advantage. Now, if you now have an API that serves content, you can serve your new Next.js initiative, your legacy PHP initiative, your iOS initiative, your Android initiative. So I think the rise of multi-device, I think the rise of front-end development more generally and the fact that any given company has dozens if not hundreds of concurrent initiatives that could all be consuming data. You could even argue that the rise of microservices and the microservice thinking is what's also powering this headless CMS momentum. So I think it's, like you said, it's mainstream and it's here to stay.

Yeah, definitely. And you mentioned something that is really, really interesting, that maybe Next wouldn't be there without headless CMSs like Strapi. Absolutely. The opposite is very true as well. I think this is why we believe so much, as mentioned, previews. Previously in these keynotes. It's an entire ecosystem that is extremely exciting. It's growing really, really fast. And the good thing about this ecosystem is that we are all growing together. And that's—

and we're sort of collaborating at a very large scale. Yes. Asynchronously, right? It's like the epitome of remote work. The Strapi team and the Next team are collaborating but the binding glue is one data fetching hook that I mentioned. We took React and we added as little as we could to make it really useful to build websites. And one of the key things that my co-founder Naoyuki added was this data fetching hook, getInitialProps. And that hook is the one that has been able to connect our companies and connect our communities Now it's been broken down into getStaticProps and getServerSideProps but the point has been that we're entering this new phase of large-scale flexible cooperation that, yeah, it's just fascinating. We didn't know each other but we were collaborating and then we met each other in San Francisco one day and now we work remotely together. It's been great.

Exactly, exactly. And so with all of this microservices and Jamstack architectures, They can sometimes lead to confusion or complexity if not thought out properly. What advice would you give to companies who are getting ready to migrate from traditional to a more modern API-first tech stack?

Yeah, that's, that's a great point. I've always— I spent a lot of time thinking about this, right? Because coupling has these short-term advantages. You get up to speed, uh, it feels like you get up to speed more quickly. And, and, uh, and I think that vertical integration has benefits. I think the biggest piece of advice I would give is embrace the API, right? The Supreme Court just ruled in the US that you can't copyright APIs. I think it's a good time to say like embrace the API, embrace the openness of APIs, uh, and APIs are getting easier all the time, right? Like with REST, with GraphQL, with TypeScript-based autocompletion of the clients, I think we're working to the point where it's gonna be the Sun Microsystems sort of model of the network is the computer. It's becoming more true every day that it feels like these things are coupled. So I think you can take that leap of faith into this world of decoupled Because paradoxically, every day that passes, we're integrating our technologies better and they feel like they're coupled but they're retaining all these benefits of decoupling.

Exactly. And APIs allow all of the things to connect things together. And I'm very curious, from conversations you can have with your prospects and customers that are moving to Next.js or Vercel and from your discussions with your peers in that industry, What are the main benefits and opportunities for companies who embrace this modern approach to app development?

Yeah, I think the number one is delighting end users. I think really like as an industry that is building on top of web, our number one opportunity really is that you create better experiences for your users. So if you're in e-commerce and you're betting on Strapi, Next.js, and Vercel to render your content pages, literally what you're doing is you're giving your customers a better brand experience that is custom to you and that is authentic. You're giving them better performance. Better performance leads to better conversion for e-commerce, but it also leads to better SEO. So at the end of the day, my, My sort of design goal, my North Star has been whatever it is that we do, we're building a better product for end users. And the other thing I think that there are two other sort of important things that follow and go very close with that. One is developer experience. Yeah. My recommendation to most companies in the world as they make a transition toward digital would be bet on the developer empowerment. Don't try to like postpone it for the next century, like, "Oh, one day I'll become a tech company." I think it's embracing the fact that you can buy great tools for your developers, best-of-breed tools for each category, the best CMS, the best framework, the best platform. And ultimately that if developers are happy, then it becomes like this virtuous cycle of users are happy, developers are happy, just, you know, you bring new products to market, new initiatives, new campaigns. And we saw that really well. I mean, you mentioned our Series B financing. Throughout 2020, a lot of companies were forced to go digital really fast. And that's where it was great validation of our work in that, you know, companies that needed to adapt their business model to COVID were able to do that in record time. On top of our platforms.

Yeah, I can't agree more. And this is the exact same benefits with the former CMS point of view. And, you know, as you know, the Jamstack industry is quite fragmented with many CMS players, JavaScript frameworks, static website generators, e-commerce solutions. Why do you see the main areas of collaboration differentiation in this ecosystem?

I would say while it's really fragmented, I would say relative to how fragmented it's going to be in the next 10 years in terms of like the explosion of services that you're going to be able to choose from, we're at the very forefront. So we're seeing every day everything is being API-fied. It's almost like we opened the Pandora's box and now every day it seems like I'm finding a new exciting API that I can use as a frontend developer to make my product better. So I think embracing that is just gonna be like kind of the key thing for the decade. I think the approach to how we build these pages is continuing to evolve. It's gonna be a combination of static, dynamic, on-demand rendering, edge rendering. We're putting computation everywhere in the world. And sometimes, you know, one of the core tenets of Jamstack, which is the J which is putting a lot of JavaScript in the client. I'm not so sure that that's gonna stay the case. I think putting a lot of a burden on the device, especially in growing emerging markets, it might be too much of a burden and we're gonna move that more and more to the cloud. I think it's been one of the big things that we learned with Next is companies sometimes to start static, like we can call that like pure Jamstack, and then grow into more and more dynamic workloads and use the cloud computational power more and more. So I think we're gonna continue to see an explosion in like the tools that we have as developers to build these pages, uh, not only on the framework level, but how that framework literally does the rendering, how it performs the computation, how it delivers the content to end users. And finally, the, the biggest trend that I'm seeing as well is personalization, geo-based personalization, basically user-based personalization. What are the pieces of content— speaking of CMS— what are the pieces of content I can recommend to you today at this very time in Paris? We're going to continue to see more and more tools to bring the content to the user that's most relevant.

Yeah, I 100% agree. I mean, everything will be about, uh, for fragmented ecosystem that will be all connected together. And I do believe as well that we will see more and more this hybrid approach between JavaScript frontend, like server-side rendering, and static website generators. We'll see some hybrid. And as for the customized experiences, this is really something we want to invest in as a CMS to display different content according to some parameters.

Absolutely, content targeting, I think it's going to be fascinating.

Yeah, definitely, definitely. And there are already a number of Strapi templates and starters for Next.js. What could the Strapi and Vercel teams and communities do to improve the integration points between the two frameworks and where should they go to continue the conversation?

I think the biggest thing that I see there is non-technical teams want to be enabled to feel like they're directly manipulating the content. If I go to Strapi.io and Vercel.com, it's like, you know what? Like, I wanna replace the heading. I wanna, I'm done with this headline. I have a new idea. I think we have to work on an integration such that we continue to bring down the friction that it takes for anyone to propose, preview, and ship such a change. And I think the technology is already all there because this has been possible in other ecosystems that are not, let's say, as future-proof, right? Like if I use the likes of like Wix, Squarespace, and all that, I get the direct manipulation, I get the access. It feels like it's a little bit easier and just right there, but obviously I don't get the future-proof, I don't get the developer empowerment, I don't get all this great benefits, but I think these two worlds are going to collide. I think with technologies like incremental static regeneration in Next.js and Vercel, you can preview content And then you can dynamically push it and get that, you know, real-time content editing experience without having to wait for a full rebuild. So technology is already there. Now what we're missing is facilitating the UIs to make this really seamless. If I'm a non-technical user, I can just press edit, you know, whether it's a drawer or a sidebar that comes in and start editing the content. Propose it as a change, preview it, share hyperlink, ship it. That has to be brought down to like 30 seconds. And I think for a lot of setups, a lot of proficient users of our technology is already there. Like a lot of people are probably listening to this and they already have that, but just democratizing access to everybody in the world that wants to use Strapi and Next.js, it should be a matter of seconds.

Exactly, and all of this headless API-first approaches added a lot of complexity, so it can be a bit harder for content editors to preview what they update in the CMS. And as you mentioned, the technologies are already there, so I think it's just—

yeah, it's exciting—

integrations. And yeah, I can't wait to, to, to make more integrations in the future on other things. I have one question for you. So you've been following the development of Strapi for some time now, and already had a bit of visibility on what we just announced. What do you see as Strapi's key strengths and opportunities for the future?

Easy. Open source, I think, like Strapi's ethos has been to build in an open manner. I think that's an ethos that a lot of companies are still like eager to adopt, right? Like we're still finding lots of customers that are in like legacy proprietary stacks. That feel locked in, that feel like they can't manipulate and understand the software in a deep way. So I think that continues to be the number one strength that I see in this shared ecosystem because Vercel is a business and there's a Strapi company, but we're building in the open. So, you know, it's like Brendan Eich always says, like, always bet on JavaScript. I think always bet on open source is pretty much a no-brainer.

I'm glad we are very well aligned on this vision and passion for open source. Well, thank you so much for your insight, Guillermo. It was extremely, extremely interesting.

Awesome to be here.

All right, it's time for us to wrap up this keynote. Friendly reminder to share your feedback, questions, and favorite moments on Twitter with the #StripeEConf hashtag. If you haven't already, join us on Discord for live Q&A, sponsor booth, and giveaways. Once again, thank you so much for being part of this journey. Enjoy the rest of the conference.