Latest / The MonkCast / Seth Webster on How Expo and the React Foundation Are Shaping Modern Development

Transcript

Seth Webster on How Expo and the React Foundation Are Shaping Modern Development · The MonkCast

• speaker-0: Hello, Kate Holderhoff here for another exciting conversation. I'm a senior analyst at Redmonk and with me today I have Seth Webster. He is the executive director of the React Foundation and chief developer evangelist at Expo. Seth, thanks so much for joining me on the MonkCast. ⁓ man, this is gonna be a good conversation. I can't wait. All right, so let's just begin with these two very exciting hats that you're wearing right now. Executive director and chief. speaker-1: It's great to be here. I always love doing these. speaker-0: developer evangelist. First off, how are you finding time to sleep? Most importantly. And second, I want to hear all about these roles because I know you're creating some exciting stuff these days. speaker-1: sleep question. It's honestly a little chaotic right now. I just got back from from China ⁓ yesterday. I don't know what day is. ⁓ I'm excited about doing both these roles because this is where you know right now I'm sitting at the center of where software is going. And the two hats at Expo, I think my job is ⁓ clear conceptually. It's helping developers go from idea to production as fast as humanly possible. And that you know means sweating the details, know, getting the small stuff right from the build systems, the tooling, the workflows, all of the stuff that historically slows people down. And the React Foundation is kind of like conceptually the opposite, right? For me, the React Foundation is about durability and longevity. It's about, you know, making sure that React is around for the next 10 or 20 years and continues to serve people and, ⁓ you know, continues to make itself an indispensable tool even in this age of AI. And I think that's really exciting. speaker-0: Fantastic. Okay. All right. Yeah. I'm looking forward to digging into all that. So let's, I guess, just begin with the React Foundation. ⁓ my God. So ⁓ at a high level, what is the RF and why does React need a foundation? speaker-1: You know, I think the big thing for me is that React over the years, even though nobody really intended for this to happen, you know, it's one of those happy accidents. React has grown from a piece of like niche UI development software into, you know, infrastructure. It's like the pipes in the walls, right? You know, it's used everywhere, at least one in two, you know, we can only measure as much as we can measure, but at least one in two websites uses React. A large majority of e-commerce websites use React. And so the amount of like GDP and like market cap that's flowing through React is just absolutely mind boggling, right? And making sure that that continues to be something that people can rely upon to, you know, make a living, to bring their ideas to life. I think that there's a huge amount of like economic trajectory that can be changed by leveraging React and ensuring that React continues to be something that people can use is incredibly important. And it largely outgrew its home. mean, as big as Meta is with its billions of users, you know, it has many, many people that need to be served across the planet. Like I said, I just got back from China and it's absolutely incredible the amount of stuff that they're doing with React in China that I had absolutely no idea about because we've not done a good job of reaching out to Asia. And so super excited to expand into that market and they're as excited as we are to. to have us. It's really exciting to think about what it means to ⁓ bring that many more developers into the fold and that many more ideas into the fold and so forth. ⁓ And so, you know, that's what the foundation is for is to create a more open ecosystem that is ⁓ listening at more listening posts and, you know, thinking a lot more about the durability of React across many, many use cases and many, many platforms while continuing to be ⁓ serving The folks have served very well for the last over a decade. speaker-0: Yeah, when I was thinking about transitioning into becoming a front-end engineer many years ago, the advice was just to learn React. Like, don't even learn JavaScript. I mean, all you need to know is React to get a good job. So the fact that it truly is the elephant in the room when it comes to these single-page applications, ⁓ I haven't seen that slow down at this point. Just huge importance in the app development space. speaker-1: Yeah, and know, for better or worse, we're far more than a front end only library now, right? We're looking at the full stack at this point, and I say for better or worse, because I think that this has unlocked a huge amount of incredible performance opportunities on the positive side. And then of course, this is the first year that React has ever been, ⁓ at least in my memory, that React has ever been at the center of... CVEs because we're now on the server that represents a new surface area for attack and some people, some brilliant folks out there, some brilliant engineers found a way to exploit vulnerabilities that we had never thought about. And so that's, you know, it's a growing pains for us to become a full stack system. And so I think it's to be expected. But as far as I know, there's not been any realized damage from that, but it's definitely an interesting evolution for us. speaker-0: Yeah, you're not the only folks right now grappling with this new batch of CVEs. mean, not only with Glasswing, but also just in general. I've spoke with people about the AI slop issue, but no, there's a lot of really great tooling out there that is finding these security issues. speaker-1: That's right. If humans have proven anything over the time we've been on the planet, we are nothing if not adaptable. And I think that, you know, as we figure out new ways to exploit vulnerabilities, we'll find new ways to defend against those probabilities. speaker-0: Okay. So we talked to a lot of foundations here at Redmonk, so I'm always interested in some of the brass tacks. So talk to me about the history of the React Foundation. How long has it been around? speaker-1: The React Foundation has been around officially, like doors open, since February 28th. But we announced the React Foundation's intent to launch back in October, like October 14th, 15th last year we did ReactConf. But before that, a number of us have been working on this for over four years. Myself, Matt Carroll, Eli White, Paul O'Shaughnessy, ⁓ just a bunch of folks at Metta who really believed in the fact that it was becoming time for React to become ⁓ more open and to live in a foundation. But we discovered very quickly when we first started to paper the foundation and figure out how would we launch it, that the organization internally at Meta was not ready to operate as a foundation. Basically meaning that we did not know how to ⁓ effectively ⁓ manage the number of PRs we would receive. ⁓ effectively know how to you know, open the table and involve more voices. You know, it was, it was very much focused on like the, the way that meta operates and, and for better or worse that served React really well for so many years. I mean, when you think about the battle hardening that React has seen before it ever reaches the hands of an independent developer or like, you know, the many companies, myriad companies that use React. it's gone through so much testing at Meta first that you can almost rest assured there's going to be no issues with React. And we do see very few bugs and like React is the source of very few sebs and so forth because of that ⁓ amazing infrastructure and the amazing testing it sees at Meta. But on the other side, that means it moves glacially slow, right? So people would always comment about React itself or React for Web and just be like, wow, you guys do release like every two years, right? Like that's so slow. And I think that's, ⁓ you know, it's a double-edged sword, right? The innovation feels slow, but then when we do release, there's a lot of innovation, right? You get a lot of features. But at the same time, ⁓ React on the web has done an amazing job with backwards compatibility. And so I think that's led to, you know, people feeling like you can really bet on React. You know, you can just, to your point about when you were learning to be a front-end engineer, Someone told you just learn React, right? It's that just, never doubt React. ⁓ And that's great, but at the same time, there's plenty of companies that would have liked to see us work on certain different things. And there was never a way for people to get their voice into the pipeline. And the React Foundation is gonna serve ⁓ here and forward as that place for people to get their voice into the conversation. Early conversations have been amazing. I'm really excited to share some things soon about, you know, things we're gonna do to stabilize the ecosystem and help people move faster, ⁓ make React available on more platforms. I I think it's just as exciting as this AI boom is, I'm equally excited about the future of React and the things that we're working on to take advantage of the fact that we're now a foundation. speaker-0: Yeah. Are you thinking of, ⁓ you know, changing the release cadence at this point? Is that on the table? We mentioned how quick everything is moving these days. speaker-1: think anything and everything is on the table. think we hold dear some pretty significantly, I don't know, important ⁓ principles around the quality of those releases. And so I don't think that React is going to turn into like a six or seven or 10 or 12 releases a year framework. But I do think that there's an opportunity to ⁓ make sure that we're releasing on a cadence that feels like it's in accordance with the speed at which everything else is moving while also continuing to hold that high. high bar. The thing I would say is on the other side, like React Native side of things, we did improve the release cadence there from ⁓ doing like one release a year. Over the time I was at Meta, we were improved from like one release a year to two releases a year to four releases a year to six releases a year. And I think it's important there because React Native is still on its march to 1.0, right? And one of the things I mentioned in terms of ⁓ stabilizing the ecosystem that I'm really excited to work on this year is that like stable API layer to help us get to 1.0 so that consumers or vendors like Microsoft are not struggling every time we do a release for two months to catch up, but also offers us the ability to start moving more quickly and round out decisions that we made early in React Native's history and sort of smooth those over into a place where you can be more dependent on React Native. and feel like it's slowing you down every time we do a release. ⁓ I think there's dual opportunities there to increase the cadence on React Web to a place where React Core, to a place where you feel like it's moving at the speed it should, and to stabilize the releases of React Native such that as fast as they move, they aren't disruptive. speaker-0: So I'm always interested in how open source leadership is thinking through their philosophy around open source writ large, right? So I guess I'm just curious, how are you thinking about governance and how it should look in this era of AI 2026? speaker-1: Well, here comes Kate with the easy questions. ⁓ I love the softballs. You're welcome. Yeah. I think so much is changing so quickly. It's important to come back to what I think is the foundation of, for as long as I can remember, it's been the foundation of how I think about these kinds of things, which is to have really strong principles, to be very principled about how you're operating. I think that That allows you to endure any of these big sweeping changes if you're operating with principles that you know are in the best interest of your community. And so we don't have a publishable set of principles yet. We've had our first board meeting and like those are things we're working through. We just had our first board meeting. As I said, we opened the doors February 28th. We had our first board meeting in March and so forth. And that was really about operationalizing or ratifying the operational sides of React, like what's our budget and. What are we going to meet? How often are we going to meet? Who's going to be on the technical steering committee? And so that's all still coming together. I think the thing I would think about is the first principle that we hold super dearly is that we're always serving the community, the best interests of the community. I think that there was a time in React's history where we were much more serving the priorities of meta. at any given time. I think that, again, to the point that I made earlier, I think that's really served React well for many years. But with so many people with such a huge vest and interest in React, the first thing we have to hold dear is that the choices we're making, the decisions we're making are taking care of all of those stakeholders. ⁓ And so I think about, when I say stakeholders, I'm not just talking about the big companies that are betting big on React. I'm talking about the indie developers. Because at scale, when you think about the opportunities that exist to refine the economic trajectory of individual humans' lives and that of entire ⁓ markets. The individual developer is the one who's putting their hands on the keyboard or commanding a fleet of agents ⁓ using React to build the next big thing or solve an incredibly important problem for their community. And so I think from the smallest developer to the biggest company, we have to be thinking about what serves the greatest good and what serves the most important need at any given moment. And today that need seems to be empowering people to use agents to bring ideas to life more quickly. And that's where I think we're spending a lot of time is figuring out what is React's role in the world where agents are in agentic development or at the tip of the spear or at the forefront. And what we're finding already is that, you By default, I think there was even an article that came out in like, I'm going to say it was the Atlantic, may have been, but it one of these big magazines looking at like how agents, when you ask them to build like an e-commerce site, they build react sites and often they choose Next.js just by default. Well, that's going to make it really hard for other libraries and frameworks to take hold if agents are never even like offering to select them, right? And so we need to think about what does it mean to educate agents on the opportunities that exist for them to use, right? And give these models the ability to choose the right framework for the right job. Because next isn't always the right choice. Sometimes there's a better choice. And so I think that's an interesting problem to solve that we're excited to work with, like the model vendors on and independent model developers on, is helping to still create a proponents of choice and allow people to let a thousand flowers bloom. to the end that we help people bring ideas to life. The goal is always to empower developers to go from idea to product as quickly as possible. And obviously at high quality. mean, the biggest question is, like you said this a minute ago, is about confronting AI slop, right? We don't want to create a situation where it's super easy for us to deliver substandard and subpar experiences to the world. I'm envisioning a world now where every single market niche gets saturated with. apps. And I think it's going to happen. And so the differentiator is going to be your ability to create something that feels amazing, that taste is going to be the bottleneck, that design is going to be the bottleneck. And like, what are we delivering in the analog world is going to be more important than I think than just the digital side. Like, the fact that a hotel has a great app experience or website experience is going to be table stakes. And it's going to be about how do we translate that into an experience that feels very differentiated. And so To me, that's about continuing to provide choice. That's continuing to provide people with the opportunities to focus on the stuff that really matters, accelerate the boring parts, and make easier the parts that allow you to shine and be delightful. speaker-0: And for folks who aren't as familiar with all the frameworks and meta-frameworks, just want to point out here that when you refer to Next, you're thinking of ⁓ the meta-framework that Vercell runs that is built on top of React. speaker-1: Or lowercase ⁓ meta. speaker-0: Well, yes, good to differentiate this context. Yes, I like that. speaker-1: The number of times we had to say that internally when I was at Metta was, ⁓ man. speaker-0: Okay, well, that's some good inside baseball there. That's ⁓ interesting, I can imagine. So thank you for indulging me with all my questions on your open source philosophy. I know that's a tough thing to think about and try to sum it up in a sentence or two. That's challenging, I get it. ⁓ Okay, well, let's go ahead and pivot then to talking about your other hat, which is of course the work that you're doing at Expo, okay? But before we get too far in the weeds here, these hats are very much related. In fact, you could stack your hats, I imagine, right? But for folks who maybe aren't as familiar with native development, and I know I keep having to educate my own self about ⁓ mobile development and all the nuance around there, especially as a front-end engineer who had hubristically thought that I understood it all, ⁓ talk to me about the relationship between React Native and React. speaker-1: ⁓ So in the simplest way, React is the foundational layer that provides the reconciler and things like ⁓ JSX parsing, which is the syntax, the language of React is very much grounded in JSX. ⁓ And then React Native is the mobile projection, the native projection of that. Basically meaning it provides an opportunity for you to take the mental model and the programming model of React and use that via some really brilliant work that's React Native team and teams, companies like Expo and others, many software mansion and call stack and all these amazing pillars of the community have done to make it possible to use that same mindset, that same mental model and build brilliant, beautiful ⁓ React or native mobile apps. And one of the sort of key differentiators between React, React Native and many of the other like mobile abstraction flavors or native abstraction frameworks out there, is that one of the core principles we hold dear in the React Native world is that the resulting application is truly native. That it feels like it belongs on the platform on which it runs. Where a lot of the other approaches taken out there is to create like a pixel perfect render across all platforms. Basically meaning that the way it looks on Android is going to be the exact same way it looks on mobile and generally sort of the exact same way it looks on web if they have a web component to them. And what that results in is, in my opinion, is sort of an uncanny valley experience where you're like, this may feel pretty good, it may look pretty good, but it doesn't feel like it belongs on the platform. And so you always feel like you're using something from somewhere else. iOS users, Android users have come to expect certain like gestures and interactions and the way something looks and the way it feels and the way animations move and so forth. And so you end up with an experience that doesn't feel like it's germane to the platform. And so one of our core philosophies is that the net result of something as it's developed on React Native is that it uses the core native primitives at the end of the day and that it feels very much at home on that. You mentioned a lot there. You asked like, I don't know, it was like 11 questions in one. You rolled them all together. The way I think about like these, ⁓ the ecosystem for React and what makes it healthy is that when React was developed, wasn't ever, ⁓ first of all, like when it was developed, it wasn't necessarily initially conceived to be something that was ⁓ used outside of Meta. Quickly people realized, hey, this is actually really amazing and useful and like we're so productive and it allows hundreds of engineers to work on the same application at the same time. it became obvious that open-sourcing it was something that was going to be good for the world. And so it was open-sourced and people began to use it and we know the history is that it took off, right? But it wasn't necessarily meant for that, ⁓ know, originally conceived for that sort of broad distribution to use. So as we think about ⁓ the fact that React was developed, it wasn't something that was necessarily going to be used by, we didn't know that it was going be used by lots of people. The ergonomics of like, the data fetching and how do you do routing and how do you do these incredibly important things to any application, like across the board, every application, almost every application does data fetching and routing and animations and things like that. What became clear was that React could provide, like the team at React was not a super big team for a long time, and so the team there could provide these core primitives, these core... ⁓ abstractions like the language and the programming model and so forth, and that the community could fill the holes on data fetching and animations and those kinds of things. But what this ended up resulting in eventually was that you started to see the emergence of meta frameworks, things like Next.js and others, Redwood and now Tanstack, is amazing. There's lots of choices. ⁓ But these meta frameworks, what they do is they provide this layer above React. that gives you these amazing ergonomics and consistency and data fetching and beautiful animations and routing and so forth, which are so important to React and so important to creating an amazing developer experience and an amazing delightful user experience, right? That's the end goal here is if none of this serves a delightful user experience, then what are we doing? On the other side, it took a lot longer for us to see a beautiful selection here, a beautiful choice here for ⁓ developers doing native development. And a few years ago, was probably three and a half years ago, Expo reached this point where it was just absolutely amazing. It was just like the secret sauce in native mobile development. it was the, at least for me when I saw it, the biggest mobile accelerator, the biggest accelerator I'd ever seen. And I've used Cordova and I've used a number of these, MetaFramework and Xamarin and others. And I was just like, this is it. Like I was so sold on it when I saw it. as, as they've continued to evolve, that's only gotten more true, become more true. And so about three and a half years ago, we sat down with Meta and Expo and we met and we said, you know, here are a number of things we at Meta would really love to see from the Expo team. And if we can work these things through, we would love to make and recommend Expo as the default Meta framework for mobile development. And. Obviously, we've worked through those things and last year at ReactConf, we announced that we were ⁓ wholeheartedly, full-throatedly recommending Expo as the default framework for doing React Native development in the same way we had recommended that everyone use a meta framework for web development. And at the time, ⁓ Next was the clear leader there. ⁓ But what we did was we created a set of principles and guidelines that... Meta frameworks had to meet in order to be recommended. And whoever was going to do the best job, whoever was doing the best job at meeting those principles and guidelines, you know, that's from the highest quality, that was going to be the one that was going to sit at the top of the list, right? And there is no, at this point, there is no like competition to Expo. Like Expo has just led the way. They continue to be at the forefront. The number of innovations that are coming out. If you look at like the Expo Twitter feed and what other people are saying on Expo, about Expo, it's just innovation after innovation after innovation, creating more and more delightful experiences that again, feel amazingly native and at home on every platform on which you can render Expo today. And I expect the number of platforms also to increase as people are looking for more cohesive experiences across many things. so for me, working at the kind of bringing full circle, for me working at the React Foundation to create this durable, you know, React and React Native, to me they're the same thing, by the way. Like, we talk about React and React Native, and they are very different pieces of technology, which I hope that we can find a way to bring them closer together. But I think of it all as React because the mental model is all just React. ⁓ I want to create this world where this durable piece of technology, ⁓ almost fades into the background and you're just really focused on the delight you can create with them. For me, that's what I'm doing at the foundation is making sure that for the next decade, that React is staying relevant, staying at the forefront, becoming even more the default for agents and giving agents the tools as we think of agents. If we think of agents as another cohort of developers, a different type of developer, which I can come back to in a second. We want to give them the tools to be able to build the best React and React Native sites or best React and React Native apps ever. That's what I'm doing at the foundation is creating that space for all the engineers in the world, whether they're at big companies or independent developers. And then what I'm focusing on at Expo is helping to make sure everyone understands the value and delight that is possible with Expo that is much harder to do with React Native by itself. You have in Expo some of the most developer obsessed people I've ever seen, delight obsessed people. I mean, the number of times that, you know, Evan Bacon has sent out a tweet where you're just like, ⁓ my God, look at how beautiful this app has been, that has been created is. And that's as a result of the whole Expo team being so obsessed about the little tiny things, right? Sweating the details, sweating the small stuff. And that's, that's why I want to like work here and like, evangelize, if you will, ⁓ this work because I want people to be able to bring their ideas to life very quickly. I don't want people mired in the pain that is getting animations right on Bear React Native or even Swift and other frameworks. It's very hard to get that exactly right. I want to make that so easy so that the apps are beautiful and delightful and people love using them in order that these engineers are successful in delivering an app to the world that people use and hopefully solve some of the world's most important problems, right? It's not just about Candy Crush and it's not just about another, you know, streaming media streaming app. It's about people who are building apps to help people set the market prices for their, the agricultural stands that are in, you know, developing countries and so forth. And so like we've seen that over and over. If you give people these tools and technology, they will solve their own problems. And I want to see that at scale. And I think AXPO is a huge lever. speaker-0: Okay. Well, to your point, the way that I learned the most about Expo, I think, was from the React docs where I did see that it was recommended ⁓ as the way to do native development. you you're right on there. That's a great signal, I think, to many folks who are in that space about how established you folks are and the fact that you're, I guess, trying to remove friction is really what I'm hearing from your description there, right? ⁓ But I do want to... talk a little bit more about that because again, you know, this it gets complicated very quickly. So for folks who aren't following the situation around cloud builds, Xcode wrestling, I guess idea to app store, you know, all the pain that that typically involves. ⁓ What do tell folks is Xpo's core value prop? speaker-1: You answered the question already for me. It's about reducing friction, Yeah. Okay. The thing we see a bunch of is like, know, ideas are cheap. Execution is everything. Being able to go from idea all the way into a customer's hand is the full journey, right? It's not enough that it just lives on your computer or lives in a GitHub repo or maybe lives on your phone. Although that is, think, to come back to the AI boom we're going through now, I think there is a... We're about to see the largest and are already seeing the largest emergence of personal software development in the history of software, right? You're going to see millions of billions probably of apps developed that will never see the light of day for another person that are just yours. And we're also going to see, I think we can come back to this if we want to dive into more AI topics, we're going to see the emergence of a lot of ephemeral software, software that exists only for a few minutes and then goes away. ⁓ But coming back to the core value prop is like for things that are more durable and things that are meant to be seen for others and even your personal software, you want it to be awesome, right? It's about reducing that friction between the idea and being able to have something you can use that that's beautiful and delightful and fun and so forth. That's also not like incredibly painful to develop and arduous, but to the point about software that like sort of never sees the light of day going from, your code to a build that works on everybody's phone. has typically, you that's available in the marketplace, has typically been really difficult. And Expo solves at every single point along the way, especially with the emergence of these new technologies just in the last couple of weeks that'll allow you to programmatically and declaratively set your App Store metadata such that you don't even have to sit and wade through that awful app, sorry, Apple, that awful App Store. metadata entry form. It's just so painful. mean, it's so daunting that people, especially people that like struggle to complete lots of different tasks, where you have to like really go through these huge forms. People get stuck there so often. I mean, how many I don't even know how many great ideas have died at the Apple and Google Play Store metadata entry forms. Solving that full cycle problem is what Expo is trying to do so that so that it's no longer the limiting factor. of bringing amazing ideas that solve real world problems into the world so that they can be used by others. speaker-0: Could you talk specifically about which expo product that is? Is that EAS or that does the metadata thing? speaker-1: It's all of it. EAS, we just released tooling this last week that allows you to declaratively set up the metadata stuff. But it is everything from ⁓ the Expo framework during your development cycle all the way through to EAS build and EAS submit. Allow you to do ⁓ the submissions to the app stores and set your metadata program. speaker-0: Yeah, mean, when I have done research on, well, how developers are grappling with the challenges associated with mobile, the App Store is always at the top of that list. mean, always. speaker-1: I feel exactly the same way. I think I have like 97 apps that I've developed and I haven't shipped any of them. Actually I shipped like three of them. But most of it has died at that App Store thing because I'm just like, I just don't have the time for this. It's an app I want to use and it's like, I've developed it for myself. And that's what I'm so excited about with this AI boom is like, I know that a lot of folks are sitting there. you know, in fear of what's happening. But for me, someone who's like North Star is helping other people bring their ideas to life. I'm so excited that like, it's not just going to be me developing my personal software anymore. It's going to be my wife who's an attorney. ⁓ You know, she's going to be able to develop her software. And my, you know, my father-in-law, who's an astronomer, he can develop, you know, at 80 years old, he can develop an app to track his. to track his discoveries and so forth. so I'm just so excited about this era of personal software that I can't even begin to tell you. speaker-0: Yes, yes. Well, I am 100 % with you. I self-identify now as a vibe analyst. I'm vibing all the things. I'm excited. Vibe analyst. speaker-1: was my first Vibe Analyst. Did you coin that? I don't know. Is that a thing? speaker-0: did coin that. Maybe I shouldn't have. I don't know how good it is, but yes, that is, it's what I'm running with. ⁓ Yeah, so ⁓ I'm with you. ⁓ I'm jazzed. I get to create all these silly little apps ⁓ and quickly, you know, so it's great. And as someone who has sort of a design background, I actually get very excited about the design part of this. That's awesome. Yeah, well, I'm saying 2026 is the year of design. I have been on record saying that because it's ⁓ coming back. speaker-1: Design's coming back. speaker-0: It has to. But yes, you have mentioned AI several times. We're deep in, so let's just keep going there. So I didn't give you that heart of a time about what I think of as like the spa wars. And of course, there's a very vocal subset of front-end developers who complain about React being a little heavy on JavaScript, that maybe we're all just going to create our own apps that are going only have the libraries that we absolutely need, et cetera, et cetera. These squeaky wheels are all over. You can find them. What I think is cool about ⁓ our current era is that code is becoming cheap. And so I'm interested in how you're thinking, maybe with both hats, about where we sit with like the Flutter versus React Native wars or the Meta Framework wars. How are you approaching that idea? Because it is so easy now to change frameworks or to rewrite code in a different language or to... update from Java 8 to Java whatever. This is the era we're living in where the bottleneck they used to surround switching is diminishing. The aperture is widening not only in terms of who is invited to join this party, but also accelerating folks who maybe want to refactor or to re-approach this thing. How are you thinking about that entire issue? It almost feels quaint now to think about... wars? Do you just throw that off or are the Native wars still going on? speaker-1: I I love people. I think by nature I'm probably an introvert, but I also love people. And people are going to find reasons to argue and debate. it's a core part of the engine that drives innovation forward is these debates. And for me, I have always thought that the competition drives innovation. And so for me, it's all net positive that we have all these frameworks competing for eyeballs, competing for users and so forth, because they have to do so with some kind of value proposition. Otherwise people aren't going to choose them. How does AI change that? In my mind, it doesn't much. It does lower the cost of switching. But what this also means is if the cost of switching is lower, the cost of experiencing these other frameworks and trying them out and seeing what the effect is on the idea you have is also cheaper, which means that differentiation is more important than ever. know, having some way to say, I like this better because is more important than ever. And so I think that it's going to drive people to be more innovative and more creative. And so I'm not feeling like that's a net negative. think that drives the overall sort of the rising tide of salt boats cliche. I think applies here because you're going to see that differentiation. I also think that I don't, I just don't get too mired in the like Flutter versus React Native and all this stuff. It's to me, it's about the power of choice and being able to like say, this is the right tool for my job. For what I'm going to do, this is the right tool. And if Flutter is that thing for you, then it's great. You know, I think the Flutter team made a bunch of really amazing choices and innovations along the way. There's also things I don't agree with. think the approach that the Flutter team took on web, of course I'm very biased, didn't result in applications that feel germane to the web. They feel like they're running in a WebAssembly sandbox. I think there's lots of choices people make all the time as they're developing their apps that come back to bite them later. Also, a bunch of choices people make that turn out to be the best thing that they did during the development. And to your point, like AI in this world of being able to drive that cost of software development really low, you're not quite as stuck behind your past decisions as you used to be. It's much easier to unwind and undo them, which I think is just, that's just going to accelerate the people's ability to deliver delight. yeah, just don't, my team will tell you, like there used to be these wars going on on Twitter. My advice always was don't engage. Like you can engage if you want to like, if you want to learn more and you want to be like, Hey, help me understand where that's great. But don't, don't argue for react. Like it's just not, it's not worth it. It's not going to help result in people feeling like, you know, we're a team that listens and wants to participate in the global community and so forth. Even if somebody, and this happens all the time, people level things against react and, and flutter and others. They're just completely untrue. Like they're not based in fact. And. we could argue about those things, but instead I'd rather devote all that energy to figuring out what kernel of truth is in that argument and resolve that than, you know, go out and try to convince people with an argument. Just convince them by shipping great things. And that's been the philosophy I think we continue, we've had and we continue to try to extol. speaker-0: Right. Very diplomatically put. I can tell why they elected you executive director. I'm very impressed. Thank you. All right. Yes. Okay. Well, so I'm an analyst, right? I like to talk a little bit about business. So I'm interested in this particular moment at Expo. Very exciting things happening there. So y'all just closed a series beat. That's pretty exciting. speaker-1: That's awesome. speaker-0: So we probably have some tech journalists and investors listening right now. What would you, you know, what's your sense about what that says about what's going on at Expo? Like what, you know, what's the scuttle? speaker-1: Well, I mean, there's a lot we can say and a lot we're still figuring out. But I think the most important thing is that this empowers Expo to move faster and do more with the goals that we've set. Expo started as a humble idea to remove, as you mentioned earlier, the friction of developing mobile apps. And it has become an indispensable piece of technology for hundreds of thousands of apps. big burden and Expo is still a really small team relative to the size of the market we serve. And so I think that the most important thing is that this provides a layer of, to keep using this world, durability and stability for Expo that we now know with this investment that we can continue growing and continue delivering huge value to the technology that people are using every day. I think the most exciting part for me is that this opens up the opportunity to invest in where the market is going. And I think as you've already seen a huge amount of investment in the AI side of the business and allowing us to do really great things with the technology that's been built over the years. I think one of the most exciting things for me, and I'll speak sort of personally here for a second, is that James Eday and Charlie Cheever and Evan and the entire team, I won't list them all, but have built these ⁓ foundational pieces of technology that sort of span horizons, but also build upon each other for a decade now that make Expo sort of uniquely positioned at this moment to be a technology everyone relies upon in this age of idea to production through AI, through agents. ⁓ Like nobody is more well positioned, I don't think, Expo to to take advantage of this moment or serve people in this moment because these technologies are incredibly declarative. So agents use them really well and incredibly performant and they deliver delightful outcomes and so forth. And so I feel like as you see many of these like vibe coding tools, many of them are using Expo, right? Because of this foundation that's been laid over many years of sweating the details and sweating the small stuff that it becomes this inevitable choice to use something like Expo. And at this point, don't really know who would rival Expo in terms of the level of quality and then end to end ⁓ pipeline that you get from Expo, ability to go from literally you can go from an idea to the app store in days. And that is something I don't think anybody can really compete on ⁓ unless they're using Expo, which is you know, than his expo. So I'm like really excited about that. To me is like what this, what this like fundraise actually, actually means for me is that more and more people for a longer period of time will get the opportunity to leverage expo. speaker-0: Fair enough. ⁓ Yeah, and so when I've spoken with you folks, you introduced me to a new product that's in beta right now, which is the Expo Agent. And this jumped out at me because I speak to so many vendors who are thinking about moving into the agentics space. And, ⁓ you know, we've moved from the chatbot era into actually creating products around agents. So could you just tell us what that is, what it does, and like, what was the thought process around building it? speaker-1: I mean, the easy answer is like, this is where the market is going, right? The market is going in non-certain terms towards agentic development. There's sort of like a, you're gonna accept and get on board with this idea, or you're gonna resist it and life is gonna be tough. I think the easy answer is like, that's where the market is going. I think the more nuanced answer is that the team is so, again, back to like the focus on quality and sweating the details. I think the team felt incredibly passionate about the fact that we are uniquely positioned to be someone who can deliver a set of agentic tools for native mobile development or native development that will give people the ability to deliver more delight and add a faster clip than I think a lot of the other choices out there. I think that's the more nuanced answer. But I would think about how it's shaken out. You asked about what was the seed that that led us to think about that. that's what I think about. But I think as Expo Agent has been developed, we discovered a number of other opportunities that weren't so obvious in the beginning. So there's the basic chat my way to an app using an agent workflow that so many are going after. But I also think that the amount of experience this team has delivering applications natively at scale. working through the app store is understanding what kind of telemetry leads to delightful apps, what kind of telemetry or what kind of information or what kind of approaches lead to stickiness with customers, that what we've ended up with is a set of best practices, a set of knowledge that is unique in the industry. And what this means is that we can, instead of thinking about this just as a, do you know from chat to app, we can think about this as how do you go from idea to, you know, Apple App Award winning sticky app that sticks with your customers. And the nuance there is that we can treat, you can think of Expo Agent as something like a forward deployed engineer that sits with your team and is able to instruct you on the next most important things to work on. How do you optimize for performance? How do you optimize for delight? How do you optimize for engagement and so forth in really healthy ways, right? And so I think that's something we've discovered. That's one of the things we've discovered is that this wealth of knowledge that we have is being able to bake that into a product that people can actually use and getting it out of our individual engineers' heads and into the hands of developers in a way they can actually apply and use without having to go to Expo and hire one of our engineers to come help you. I think that's a huge opportunity and one that is uniquely ex- speaker-0: And so talk to me about what the agent does exactly. Like, are you running LLMs in the background or do they bring their own LLM or like, what does it look like? speaker-1: Right now we're reusing, know, the Expo agent sits on top of Claude with a bunch of skills and tools and things that we give Claude to up level it. A bunch of those have been released ⁓ independent of Expo agent. So you can go and install the Expo skills into Claude and get a lot of ⁓ that stuff. If Expo agent is not something you want to work with, you know, we're never have been a company that wants to lock you into using one, you the only way to do this is our product, right? So we're releasing a lot of that independent of ExploAgent, but the experience of using ExploAgent is so wonderful that I think a lot of people are going to be drawn to that, to do that anyway. But yeah, I think the big thing there is like just, you know, thinking about ⁓ ExploAgent as a... ⁓ bringing together all of the knowledge of the Expo team into a single agentic workflow that you can use to deliver your apps from the idea all the way to production, including something we launched not long ago. Sorry, I'm pretty redundantly saying the word launch, but the product is called Launch, which takes a lot of the headache out of getting your idea. You can literally click a button in Expo Agent, launch this to the app store, it's head it to test flight, and it's on your phone, and just, you know. 20 or 30 minutes, depending on how backed up ⁓ Apple servers are. But the idea here is that, it's a set of agents, it's a set of LLM calls that are building your app using the best practices that Expo understands. And then with this idea before, deploy to engineer, something that could be sitting within your repo and on check-ins and... on changes to the code, it's actually looking at it from that wealth of knowledge perspective and saying, here are the opportunities for you to improve this. ⁓ Asking questions later about, why do people bail out at the third step of my login flow or by onboarding flow is things that ExploAsian is going to be able to help with as well. speaker-0: Last question for you here, Seth. I want to zoom way out. So, you know, I talked to lot of developers. There's this real undercurrent that I'm hearing of, I guess you just even call it fear, right? So the developer community is going through this tumultuous time. A lot of folks are out of work. Whether you want to blame this on the post-zero interest rate phenomena, economic era that we're in, or on AI, I'll leave that to folks and their personal preferences here. ⁓ But the fact of the matter is that some developers I talk with have a lot of anxiety around this entire moment. And so you're deep in, you have a good sense of what folks are seeing. So I want to hear the advice that you're giving to developers right now. What do you say to folks who are on the inside, who've been doing this for decades and are suddenly seeing that the work that they have done for the past decades of their career suddenly shifting in this transformative way. What advice are you giving them? speaker-1: You know, I think it's changing really quickly. even my own thoughts on this are changing really quickly. The one thing I would say is I want to speak first to companies considering what to do with their engineers during this time, where these new agentic opportunities, opportunities to use agents, may feel like an opportunity to save money. And I want to speak to the folks thinking about that. I think this is an opportunity to better leverage your engineers. I don't think this is an opportunity to ⁓ save money. think saving money is an easy choice in the short term, but it's very much cutting off your own nose to spite your face. I think your shareholders are going to love it, and I think probably it's going to be great for the short term. But over the long term, you're going to hire a lot of these people back. So instead, figure out how to use AI to my second point, which is that to speaking to the engineers out there, I would very much encourage you to not shy away from these tools, to not resist and ⁓ not focus too much on retaining the old way of working. The fact is this shift is happening. I would say that it has happened. Even though we are not stable and we have not locked in on what the actual day-to-day workflow is going to be forever and right now it's all kind of kludgy and you're typing into the terminal or you're typing into cloud code or codex or whatever. I think these tools are going to continue to evolve and not participating is just like the company is cutting off your own nose to spite your face. I would say learn how to be. a director of agents, you know, it's very much like the transition a lot of software engineers have gone through moving from like individual contributor to manager. Figure out the best way to manage a fleet of agents and use what is uniquely you, your design tastes, your idea about how things should work. Learn, you know, go back to the old architecture books and learn how to better design an architect software by yourself. You don't have a team of people working with you making those architectural decisions, figure out how to be a one-person band using a fleet of agents. It is not only incredibly rewarding when you unlock the best way to do it. It feels really good and so forth. It becomes a huge amplifier of your ability, where you used to have to turn to a design team and say, how do I make a beautiful login for you? Now have this person on your team and they work for you. when you used to have to say, do I secure encryption across the wire that's my own and that, you know, and so forth, ⁓ that used to be, you'd have to hire a security and encryption expert. And now you have this person on your team and they work for you. I think this is an incredibly amplifying moment. If you choose to sort of, it's a mission impossible. If you choose to accept this mission, ⁓ this is an incredibly amplifying moment for every individual on the planet. And those who, engage and choose to figure this out for themselves will be the ones who continue to spine gainful employment and those who don't are going to struggle. But in the same way, we used to walk everywhere until we figured out horses and we used to ride horses everywhere until we figured out the combustion engine and we figured out cars and so forth. think we're in this moment where we're figuring out who we are and how we adapt to this new time. And so it's understandable that people feel that sense of identity question, right? Like, who am I in this new world? But I think that the answer is, is that the more you can adapt and prove, just like as humans did thousands of years ago, we proved that now we walk for pleasure. don't, know, many parts of the world still walk because they have to, but we do walk for pleasure. We ride horses for pleasure and we drive for pleasure because there are all these alternatives. I think that we will figure out that at some point you're going to, maybe people will code. by hand as sort of an exercise of craft. I continue to encourage people to do that as well, because understanding how software works and how these agents are building software will only make you a ⁓ manager of these agents in knowing that they're building software that is durable, that is not full of security holes and, you know, going to lead to impossible maintenance burdens later. But the cost of developing a software is reaching an all time low. And if you don't get on board with that piece of it, then it's going to be incredibly hard for you to maintain this career. And I don't want it to be bleak. I really, really do want to say that it's like, for me personally, it's been a huge enabler of my focusing on my backlog of ideas that's like 15 years old. I've gotten so many of them done in the last six months. And that's amazing. You know, think of this as an opportunity to accelerate and amplify yourself because each engineer brings a unique set of ideas and taste and ⁓ approaches to the table. Have the agents wield your taste and wield your approach and wield your ideas on your behalf. And hopefully that enables you to not work 80 hours a week. You can get a lot more done in a shorter period of time. speaker-0: All right, that's fantastic. I am gonna wrap us up there. I could chat with you for hours, Seth. This is certainly within my wheelhouse here. So for folks who are interested in hearing more from you, what social media channels do you typically go to? Are there any links that you wanna share for Expo or the React Foundation? Yeah, just direct us to how to continue this conversation. speaker-1: You can find the React Foundation at react.foundation.expo is at expo.dev of course. The team does an amazing job of sharing blogs and so forth. Pretty frequently like updates and stuff there really frequently. You can find me at Seth Webster on X. I'm not a super active social media user. I personally find it to be a really big distraction. I really want to be focused on building things most of the time, but I'm always overjoyed to hear from people. So if you want to chat about any of these ideas or whatever, feel free to... slide on into the DMs. If you want to find me on Instagram, can find me as Seth WebsterPhoto. ⁓ There's lots of pictures of birds there. That's my other side passion is photography and I love wildlife. so lately the easiest to access wildlife is all the birds in my neighborhood. feel free to check that out as well. speaker-0: Very cool. All right. Bird photographer. feel like we buried the lead here. is, we talk about that earlier. There we go. All right. speaker-1: Thank so much for having me, Kate. speaker-0: Yes, it's been an absolute pleasure having you on. Okay, again, my guest today was Seth Webster. He is the React Foundation ⁓ Executive Director, and he is Chief Developer Evangelist at Expo. Again, my name is Kate Holterhoff. I'm a Senior Analyst at RedMonk. If you enjoyed this conversation, please like, subscribe, and review the MonkCast on your podcast platform of choice. If you're watching us on RedMonk's YouTube channel, please like, subscribe, and engage with us in the comments.