The Weekly Dev's Brew

Scaling Vite: Responsibility & Community (with Capeletto Matias)

• Jan-Niklas Wortmann • Episode 3

Share episode

In this episode, Jan-Niklas Wortmann interviews Capeletto Matias, a key figure in the open-source community and a member of the Vite core team. They discuss Matias's journey into open source, the evolution of Vite, and the importance of community in the success of modern web development tools. The conversation highlights the technical aspects of Vite, its rapid growth, and the challenges of maintaining a vibrant community as the project scales. They also touch on the future of Vite and the potential competition from emerging tools.

Our Fantastic Guest
Capeletto Matias
Matias (patak) is a Vite core team member. He works on Open Source at StackBlitz, focusing on the Vite Ecosystem.

BlueSky

Chapters

00:00 - Guest Introduction and Background
00:58 - Matthias's Journey into Open Source
02:31 - Contributions to Open Source Projects
04:22 - Community Building and Collaboration
15:22 - The Evolution of Vite
21:33 - Concerns and Future of Vite
34:38 - ViteConf and Community Engagement

Links and Resources
Talk by Bret Victor
ViteConf

Thank you very much for listening!

We are also pretty much on all social media platforms, so make sure to like and subscribe!

Homepage - https://www.weeklybrew.dev/

BlueSky - https://bsky.app/profile/weeklybrew.dev
Instagram - https://www.instagram.com/weeklydevsbrew/
TikTok - https://www.tiktok.com/@weeklybrew.dev
YouTube - https://www.youtube.com/@theweeklydevsbrew
LinkedIn - https://linkedin.com/company/the-weekly-dev-s-brew

Hey, everyone. Welcome to another episode of the Weekly Dev's Brew, where we sit down with the minds shaping modern web development. Today, we're joined by someone who's helping revolutionize how we build for the web. If you've ever felt the excitement of seeing your changes instantly reflected in the browser without painful rebuilding and stuff, you might have had Matias Capeletto to thank for this. Known as Patak, in the community, Matias is a Vite Core team member and works on open source at StackBlitz. Bolt? We'll talk about that in a minute. Yeah. Hey. How are you doing today, Matias? Ah, I'm great. That was an amazing intro. Like, one of the best ones I've seen so far. Thank you. I appreciate this. Yeah, very happy to be here in the coffee shop with you. I should have called this coffee shop. This is so perfect. Perfect. You've been with the community for a while now. I've seen you doing Vite stuff for a couple of years now online. How did you get into this? Yeah. So, I was doing open source a long time ago, like some C++, like, let's say like 20 plus years ago. Yeah. Like, I'm old. And for a while I was doing some, like, web engine, like graphic engine that was not open source. And then I started working in an agency that, like, it was growing very fast. And we started using open source. I was in charge of culture and trying to do education also. And so we started to use open source as a way to, like, level up, like, the developers that were entering. Like, I think it is amazing when you do open source right, you can, not only your, like, technical skills, but you interact with maintainers, your communication skills, like social skills. That is, at the end of the day, it is what a lot of us lack. And, like, that should improve. So, I was doing that. And, like, this open source, we started to open source some applications. And then, like, as a way, we had this little open source club. And as a way to, like, show that it's possible to do PRs, I started to do PRs myself and saying, like, well, this is how you do it. And I started to say, oh, I remember why I really like it to do this before. And then it was, like, first view use. I was, like, we released an application that was a vue grid generator, CSS grid generator. So that was, like, very easy to extract some utilities, like, push it to vue-use. I met Anthony Fu there that I imagine, like, a lot of the audience know. And then the press. And then I started, like, talking, I remember saying to these folks, like, go and choose four open source projects you like and try to collaborate to them. Join our Discord. Like, do a documentation PR. Like, help them triage. And I chose four. One of them was Vite. Because, like, Evan was starting to do the refactoring from Vite1 to Vite2. That was, like, a really, really big sprint of, like, three months he did there. And so that was one of the projects. And I started doing that myself, too. And, like, send, as normally it happens, you send some documentation PRs. There was a lot of Rollup plugins that needed to be compatible at that point. And they weren't compatible. Because it was, like, just starting and the code was changing a lot. So I was reporting these compatibility issues. Like, very, like, mechanical work. But that is still helpful. Like, going through each of the plugins and saying, okay, like, this works. This doesn't work because of this. Like, giving a minimal reproduction. And Evan, at that point, was just, like, in, like, beast mode. And, like, closing issues as soon as I hit the open button. And it was really, like, rewarding to participate in that. And from there, one thing I really enjoy doing also, like, I was doing it also, like, kind of professionally in this agency. That it was working, like, generating spaces to collaborate between the people. Like, I don't know, a JavaScript channel where people can help each other. Like, this was, like, 150 developers. So, like, it's good to have these things. So, in the same way, there was no Discord for Vite. So we proposed it. Hey, like, let's do a community. Like, it seems like the project is ready for that. Evan created the community. He made me and Anthony Fu at that point, an admin. And I was, like, wow. Like, he's really trusting me more than I should maybe be trusted at this point. But I think, like, this is something that maintainers have a good eye in when to, like, encourage others to get involved. And become also maintainers and not only contributors. So, yeah, like, I started doing a lot of community there, like, in Discord, like, helping others. And when Evan created, after Vite2 was released, Evan needed to refocus on Vue because this Vite sidetrack, like, was taking a lot of, like, time from him. So he decided to create a team for daily maintenance. And, again, like, Anthony Fu, me, and underfin were the first team members. underfin now is working on Rolldown. So, like, he's still involved. And from there we started, yeah, working with Anthony, like, proposing new team members, like, trying to expand this team because it was a lot of responsibility, to be honest. I didn't have the experience at that point to, like, be in charge. Like, it's not that I ever was in charge alone of, like, Vite because luckily we had so many people, like, Anthony was there, obviously, that is, like, very helpful. But at the same time we got a lot of very, very good people, frameworks starting to use Vite. And from each framework that joined kind of, like, brought someone from there that, like, ended up being part of the team. One thing I'm wondering, because if I look at open source projects, mostly I pick the ones that I'm particularly interested in. And that might be my bias, but, like, bundlers is such a different thing from what you normally do as, like, JavaScript developers. It's, like, basically the complete opposite. Like, you have mostly no UI. So you don't have, like, this immediate feedback loop. I mean, with bundlers, you basically just look, does it build the thing correctly? Very high level, right? I don't want to, like, diminish any, like, your effort at all. But, like, from, like, a mindset perspective, why was Vite, like, on your top four list for, like, open source projects to dive into? That's a very interesting question. And I will say that this, like, idea you have of Vite, like, is more about ES build or about Rollup. Right, that's true. Or about rolldown. Like, these are extremely complex pieces of, like, engineering that, like, generates, like, these, like, massively efficient bundling. And I was not interested in that part. Yeah. That sounds very interesting to me, to be honest. I will still honestly say that, like, I'm amazed by the people that do that kind of engineer. But what Vite, like, why it attracts it to me, like, Vite was part of this movement, like, other projects like Snowpack, WMR, Web Dev Server. Where all these projects were, actually, what you say, like, there isn't this feedback loop. That was exactly what attracted me to it. Because we were used to, for other, like, Webpack-like bundlers, where you started a dev server and it had to bundle the whole application, your application, before sending that to the browser. And you will, like, had to wait. And then, like, the model, hot model replacement was fast, but it wasn't instant. It was not the same as grabbing, like, a color in the dev tools and, like, exploring and say, like, oh, let's see how these, like, background will look with different colors. And it's very nice. Like, I don't know if you saw some of the talk from Brett Victor or, like, this kind of, like, the sign by exploration where... I should probably. Yeah, they are very good. And we can put some links, I think, at one point. Yeah, I'll put it in the show notes. It is, to me, it was very interesting. Like, the model that Vite was proposing was let's not bundle during dev. Like, in build time, yes, like, let's put all the resources. Like, we use it, like, you could think of Vite as an opinionated Rollup setup. Like, working out of the box, like, doing the more performance sync. And performance there was not the most important part for us. Because the idea is that you will develop most of your time in dev mode. And in dev mode, Vite will only, like, start a smart dev server that knows how to serve files that normally the browser will just, like, choke with. So, for example, if you have a TS file, a Vue file, a JSX. So, if the browser will, like, you just send that assist to the browser, it explodes. So, what Vite was doing is, in the middle, like, saying, okay, like, this is a TS file. I will strip the types. I will give that. And then I will, like, handle the imports so, like, the browser can still request, like, the proper sync. Because you cannot have, like, in the same way of, like, importing relative, like, to the file system. Like, in the browser, like, it should be, like, in the root of the... So, the idea there was that this smart dev server was an unbundled setup that was completely different from what was coming before. And it was extremely fast. And when you... Not only in the startup, but later on, when you started to play, like, with this, like, color picker in CSS, you could see the same. That you had the same experience you had in the dev tools, now in your editor. And that was really magical. Like, it is interesting because after four years or five years, now maybe it doesn't feel magical anymore. That is kind of double stake. And that's good. But that was very attractive at that point. No, I get that. Like, I do remember, like, Webpack, just the... Even just the hot module reload taking, like... Okay, let me grab a cup of coffee. I'll be back here in a minute. I would like to talk a little bit more about the community aspect that you mentioned before. Because if I look at my Discord, I get an immediate panic attack from, like, 500 channels with, like, all, like, a badge icon of, like, 99 plus something. Absolutely. Like, um... So this might just be me because I'm, like, I'm very interested in everything going on. So I'm very eager to just, like, oh, I should sign up to this other new channel that just popped up or something. so the list is growing and growing how do you keep on top of all those things? Maybe it's just me and I'm not able to but Yeah, you can't so then you can relax when you can not, you can say like okay yeah then I will do like whatever and it will be fine it will help I think the first difference is that you are not alone Discord only works when you manage to get a group of people that wants to help each other it's not like if you structure Discord in a way that you are in the middle of a star that every message goes through you or even like if the team is in the middle then it completely breaks and everybody burns out and stop looking at it and this is there is a lot of details in that regard but people starting to ping like you know like like so you can take a lot of measurement to try to teach like people how to properly use this and it is a lot easier to be honest at the start of the project when you are only like talking with early adopters that they know the right etiquette and they are very eager to help each other so like it's very different new Discord that maybe you have 100 people that everybody is like very passionate and building on top of this new tool that's completely different to what later happened for example in Vite that nowadays is not like to throw shade on like anyone because this is like the normal way things evolve but a lot of what is happening in the public channels right now is our new users joining like asking for help and there you still need to like for it to work still need to be that there are like good community members that wants to help and maybe it's a completely different set from the contributors doing code because there are different skills even so so yeah I think it it scales when you have like a good group of people then later on for like you can create also private spaces because if not like what what happened is that when the public channels are too noisy because there is too much stuff too many questions people will resort to DMs so like to try to move that DMs into like a place where you can still have that builder to builder communication you can still create like more private spaces that feel like that original like early builder adopter discord I'm wondering I'm wondering now do what you think like because like Vite is the most popular bundler these days more or less every one major project offers support for it do you think it's more like that the success that Vite has is coming from like the technical aspects of things of like it's fast it has great support it offers like everything developers are looking for more or less or do you think like this community aspect is there because from my perspective with Webpack like 10 years ago it was just very painful there was like getting help was painful that's what my final statement so what if you would have to attribute the success of Vite to one of those things what would you pick can I pick both or like can I pick both or I need to I need to pick it's a very political correct answer yeah well the thing is that I seen that Evan set the correct pieces on like when it went from Vite 1 to Vite 2 there were like several changes for example like from WMR we got the idea of having a universal Rollup plugin API that works with the same plugin that you write it works in build and that and that allows so many exploration and a very intuitive API that framework just like went there and created like a whole world on top of it so I think that the technical aspects were good a lot of very good developers and passionate developers joined the Vite team and contributed being contributors also not only the smaller team but they polish so much the tool because that is if you have like a good idea but at the end you have like a thousand paper cuts it doesn't work and at the same time there were a lot of people that really like to also do like community ecosystem work we didn't mention the word ecosystem but I think that was very important in that I always say that maybe we will be talking about Snowpack at this point like it could be any of them because what we needed at that point is that frameworks were ready for a new like common shared base where we can say okay like let's fix let's work on these problems at this level so we don't need to replicate them all around and like right now you have maybe 15 20 different frameworks on top of it and if each of that framework should do like this work of like following with like node versions and rollup and TypeScript and make it sure that all works like out of the box without any problem and like DX is like it is it will be like such such a waste so at one point like SvelteKit was using Snowpack they move it to Vite the team that was doing Snowpack is starting to do Astro and they move it to Vite and then at one point we say WMR for example was that was from the preact and they also deprecated WMR and move it to Vite and at one point like the ecosystem say okay Vite will be this base and and I see like as I said we got so much from every one of this framework joining like you will start to get issues written by Rich Harris and a PR going with it and and you like as I said like I was gaining experience as like we were maintaining and for me it was like yes I will merge the PR of Rich Harris yes whatever you say thank you Rich and so and it was like I always love like all these connections between the teams and like ecosystem because Rich Harris started Rollup for example that was like the basis from where we build this and then like Lucas continued to develop it but there are so many examples of like all these tools like at one point because we had this scale and like a lot of projects and to use it like we could justify bigger investment in saying like okay maybe we can do a tool like Vitest like it was unthinkable at first to say let's do a new testing framework let's compete with Jest it was completely unthinkable and then at one point because the story of Jest and Vite was not good and we had all this momentum you get like someone like Anthony Fu inspired and saying okay I have a proof of concept working in four hours why do I see him like sitting somewhere and be like watch me I think I'm very glad you brought this up because the thing with Vitest is if you think like the jump from pretty much any building tool to Vite is crazy like the jump from any test tool to Vitest in terms of developer experience is just so good like even even if you're not using Vite I would always use Vitest because like the minute you're you use Vitest once and then use something else again you notice how painful everything is and Vitest drives everything works it just it just works yeah there is so much work shout out to Vladimir Ari Hiroshi for all the work that they have been doing now that's also like why it works very good people also joining that team oh I'm sure that it's just not like oh it works now I can tell that that four hours were all that yes so one thing I'm a little bit worried is a wrong word but like because it's such a fast moving space like everything is in JavaScript are you to some extent concerned that like tomorrow someone vibe codes a new web dev tool with Rust probably that is like crazy fast and I'm not saying it's realistic but like I think it is do you see what I'm saying like if speed is such a fundamental driver for the success of this project yeah so I was I was thinking that that was going to be the life of Vite as it's usual with any project you know like they have like an adoption curve and it is very rare to have something like TypeScript that just grows and then takes complete over and then there is like start to be built into everything and there is no more even discussion about like it's like competing with TypeScript at this point is like impossible and I don't want to like if you are competing with TypeScript like go because maybe you can inspire them to be better you know but it's still like it is like a very high uphill battle but so I like I seen that with the life of Vite it was like extremely interesting to me something that we didn't discuss but like I should say like I work at the Stackblitz and part of part of like writing that grow that we were having was that Stackblitz hired me because there was a lot of alignment between Vite and what Stackblitz needed like they were working on web containers that allows you to run Node in the browser they have this IDE that it was like instantly giving you like a playground but then you have like all their build tool taking minutes and it was like completely breaking the magic so the time when Eric Simmons the CEO like saw Vite it was like yes this is what we need or if not we had to build it so they started to put resources there then Anthony Fu got hired by NuxtLabs Jorn got hired by Astro and then like we like we could sustain the growth as we were going and still I thought at one point that yeah like there could be other tools coming and there were like TurboPack Vercel is starting to work on TurboPack for Next that Next is like the project that is not using Vite at this point RSPack starting also to work on like from ByteDance to work on an alternative to Webpack in Rust and and ESBuild kept growing like that's crazy but ESBuild was always very very like scope limited and it didn't want to create a whole ecosystem like and open up all the like plugin API that you need if you want to create like a whole ecosystem for frameworks that is like a good decision that is like a one man show like Evan Wallace is like amazing doing that and so I saw that like it was completely possible that like Vite will go on and then maybe be replaced or maybe like be an option maybe you want to use Vite for certain projects you know like you can find niches but but Vite kept growing and what happened at one point was that Evan You like said okay like Vite is at a level at this point where we need to start taking like more responsibility and and this I think was like a natural progression with with the scale because at the beginning you can see Evan closing issues saying and keeping like a very like like laser focus scope for Vite that helped it a ton saying like we are not going to support Electron for example because we are a front end tool we are not going to support that use case just like there are other tools go use Webpack like so this was like go use Webpack was like something normal to say at the beginning but but at this point at one point it started to be hard to tell others jump out of the ecosystem if Vite is not enough for you so one like I think there were two main issues that we had with Vite and this was a little bit worried for me in the sense of like how we are going to like three issues like how we are going to manage the scale so first issue was funding and like how we continue to scale because we will need more people working on Vite another one was speed for very large applications like not like Vite working perfectly when you have a thousand modules or less that will be like most projects but if you have these like like I don't know how to describe them but like it's like 10,000 20,000 module massive project inside an enterprise then you still like if you start to send all the modules unbundled or even rollup during build it takes too much time and some of the other tools like RSPack too RSPack at one point they will be ready to tackle this large project in a better way and we could still say like okay like if your project at one point reached that scale move away but it's it's very hard to like do a do a project that doesn't like scale until the end with you so that was the second one and the third one was that we needed a more flexible API because we only had like this client and then SSR like environment like default the two were like kind of hard coded in it that took us very far away with the framework game but frameworks are starting to be a lot more complex and they have like three four environments like something in node in a node server something in the edge in like cloud for worker like a client and so they needed more flexibility and this reason could have been what like took a slowly bit down with other alternatives like taking over but what happened at the end is that Evan created VoidZero and and now you have I see like 10 plus extremely like good full-time developers working to tackle first like the funding and the like scale is solved the first issue and and it's not like other companies are like backing up for example Stackblitz is still hiring me and and not only that we after VoidZero was announced we hired Ari Perkkiö that is working half-time on Vitest and the other time in in Bolt so so like we we still like are like very like we even invested in VoidZero like Stackblitz invested and so there is very good relationship there and they are working on rolldown that will replace eventually it's like a rollup port in Rust and will eventually replace both rollup and ESBuild that ESBuild is in Go so Vite always had these two bundlers that were like if you need speed you use ESBuild if you need flexibility you use rollup and we combine them like trying like that it doesn't show that actually you have two completely different things inside and rolldown will have the scope of both like be as fast as ESBuild have things like JSX like TS like type striping directly inside built-in and at the same time have the flexibility of the rollup plugin API like being like a best effort and being compatible so that that is already like in the next minor the documentation of it already will tell people how to try it out in their project sapphi red that is one of the VoidZero maintainers have been doing an amazing job on a new like fork of it it's called rolldown bit that already replaces the both pieces so that was the second and it's tackled and the third one we spent like the last two years actually reworking like everything internally in Vite with environment API and now you're able to create as many environments as you need and like you are able to run them in whatever runtime you need and so now it's also flexible so like it with these three things now I'm it feels that Vite is just starting this is like my feeling right now interesting real quick about the environment API because that's probably like one of my favorite release announcement blog posts because it's basically saying like oh we're so excited to announce the environment API it's huge game changer but if you're a normal user you probably don't care so I was just like okay and it's like a little bit more verbose than that right like it's explaining it a little bit but I was reading it and was like it sounds exciting I can tell that the team is excited about it but I have no idea what this is doing but now the way you talk about it like oh yeah okay like like tools like TanStack start SolidStart are massively going to benefit from things like that yeah exactly the idea would be that frameworks will actually use this that is a more like low level API that we directly promote like like Vite at one point is a toolkit for framework authors yes it was built by Evan that is a framework author like with all like his own needs in like in there so like normally like we take like pride like if a user upgrades a bit mayor and say like wow everything just works like if they don't notice that Vite is there then Vite is doing its job I think that is a good way for open source projects to look at things because I mean to some extent also like often open source work is like a way for like a creative outlet that's at least how I use it a lot these days so if I have 10 users I don't care if I break 10 users and it's a little harsh but for like bigger projects I love that there's the diligence to say like the ideal scenario would be to not break our users obviously and there is so much going on behind the scenes for that to happen like at one point for example like we were doing releases and like user count was growing and like pressing that release button was starting to be so scary like really scary and what we did at one point is that Dominik G created this structure called Vite ecosystem CI where we can it's a repo that like together with some GitHub actions that is connected to our repo and we get reports like weekly on the main of how Viteeco is working with all the CIs of the ecosystem I think there are like 25 CIs right now and also every time that we need to merge a PR for example like if we want to merge it for a patch we will first see like are we going to break someone or not with this and like before pressing the release button we now see that everything is green so at least we can say like hey like if we break someone like you're not testing for that so that was like so so much helpful yeah I talked with Daniel Roe about that because he was also like this is such a game changer for us to know like there's going to a breaking change that we might need to handle or something so that's a fantastic idea for an ecosystem particularly with like such a fundamental technology like using such a broad system I mean we I think we met for the first time and like walking past each other like two weeks ago at this point so probably by the time this goes live a little later but you we were in Amsterdam and we you no I cannot talk and you partially you announced ViteConf yeah in person ViteConf so the thing is ViteConf has happened for the last three years yes we are this will be like the four edition this was one of the very interesting things that happened like with having that extra funding from Stackblitz so like one of the like first conversation I had with Eric when I joined the Stackblitz was he said to me look let's do like what do you think if we do a Vite conference and I said like we don't have experience of doing that what are you talking about but at the same time there was already some discussions in the community like let's do a meetup let's do like a small thing and there was the intention of like there was enough movement for something like that but we didn't have like experience we didn't have the funding and like to go and do this kind of things and basically Eric said like you know like you have the marketing team of Stackblitz you have like we are going to hire producers we are like yeah let's go and it was really interesting like they even put like designers like web designers to do web page and like all like they use it Stackblitz we use it our mailing list like to invite people so like actually grow and we ended up with that first year I think like 24,000 people like created like going in that online conference 44 speakers I think like and it was like a 24 hour marathon that we did and it was so nice for the ecosystem I think that first one was so special like to like canalize all that energy there and it was online it was like this 24 hour because we wanted to be around the globe I loved that part of it that was really well done it was really really nice and then we did the second edition and we did the third edition and we are starting to get like hey are you going to do one in person one day like this and we said the same as before like we don't have the experience for that it's a different game to do a whole different thing yes but it happened this year that Evan was looking to do an in person conference and so like we were already connected with the JSWorld folks that are doing like this amazing conference for Vue and the ecosystem in the theater in Amsterdam that is so beautiful I love that conference so much like it's like 1000 people so it's not like these massive things like you still get to talk a lot to all the attendees and like there is a lot of speaker communication it's really really nice and single track that I love also about conferences probably my favorite stage I've been on too because it's just like it kind of feels like a theater so you see literally everyone you're not like standing above the audience like you're actually below so everyone looks down on you and it's like a great atmosphere and you have those massive screens behind you where you're like so tiny it's a little bit scary to be honest it is intimidating but in the photos it looks so good it's so cool and yeah I also I completely love it and so we like as we do with this like again like the relationship with Evan and VoidZero like Eric established like it's very good so like we talk it and say look like we were also thinking on something like going in person but we didn't like the experience so why instead of having like two smaller conferences we like join forces do a very big one and like continue the tradition of like that bit conf started and say this year like bit conf person so this year we are organizer partners so like I'm going to like the last editions I was very behind organization together with other people from from stack blitz but right now we are going to be more in a consulting role like I just met today with Evan and Jos and it's so nice like already I'm getting excited about how good it's going to be and tickets are going to be sold soon so so just for like Jos is the person organizing the JS world because I feel like we're dropping names left and right a little bit yeah yeah sorry sorry sorry no you're good I know who you're talking about but I can see like people watching this and be like Jos who is Jos I I missed the last three characters like there yeah so I I think it's going to be really nice ViteConf has always been this place where also projects like storybook in the first one they went there and say you can now run a storybook with Vite and no webpack because until then if you wanted to run a storybook you were running webpack and Vite at the same time like that was the state of the ecosystem at that point and like we always had like teams choose to do like big announcement roll up like Lucas chose to do like the mayor like release on that day for example Evan chose to do the Void Zero announcement that day last year we announced it bolt that completely changed stackblitz and now we are kind of bolt like yeah both things and so I'm expecting this year to also be kind of big I can imagine but it's still like from a like a similar because how do I say this but usually it's often that like the last couple of Vite Confs they were basically like project maintainers doing the big announcement so it's more like very technical talks and then somewhere sprinkled in like I remember like I've seen a playwright talk for instance that was more pragmatic for actual developers yeah what is like what you're trying with because with an in-person conference you cannot do like 24 hours no no no no I mean you can it would be pretty fucking cool but it it yeah no so we are we are aiming right now for two days and like it's interesting because it's going to be around like half the speakers that we have like normally because we had 24 hours before and online is a lot easier for like preparing talks beforehand we are going to keep an online stream because we still want to give anyone the opportunity to like if you want to be with us like of course like came there but if not like we want to keep that online part and yeah like I see the vibe if someone has gone to this the theater before we are aiming like that experience like around a thousand people like something that is more like still on the personal side you know and also the content wise they're probably going to be like we have a lot more to show from the bit side maybe than other times because Void Zero has been like working like crazy and now we have a lot of people working on this so I will expect like some more talks about rolldown OXC OXC OXC but at the same time there will be a lot of slots that are for the ecosystem because this was always what ViteConf was about like Vite to us it shines through others like through the people that are building on top of it so from tools like Vitest or Storybook to frameworks like Sveltekit Nuxt TanStack start and other projects that are not directly building on top of Vite but like Nitro for example that Vite but for servers and so there is to be honest there is too much for a single conference but this will just be the first year and like hopefully this starts like a tradition of like okay ViteConf 26 will be even bigger it's kind of terrible because at the beginning of the year I was like okay I should really scale down a little bit with conference and then end of last year Josh Goldberg did Squiggle Conf where I was like oh this is super exciting now there's ViteConf and I'm like my travel schedule is already like but I'm generally excited for you I think this is a great thing for the ecosystem and Amsterdam is just such a perfect spot for this we already talked about this yeah yeah and one thing that to me is important like this idea of meeting in person later on helps so much when you work online with others and two years ago in Vue JS in Amsterdam Jos managed to get 10 people from the Vite team together for the first time it was the first time I saw Evan the work for like two years and it was so nice and we actually decided the environment API like with paper and pen in the venue like we were sitting in a table with Vladimir one of the maintainers of Vitest that had a strong influence on environment API and it was yeah so the venue had a lot of like really history for us and now it may be a really good opportunity also for like the new people that join from Void Zero that like I don't know a lot of them like I have never met Boshen for example the author of OXC and so like I'm really looking forward to like like meeting like all of them yeah I can imagine I think we should I will put the link in the show to because I would some trip but I'm very excited for it so is there anything else that you would like to plug at this point otherwise I think we're at a good point to slowly wrap things up anything that you want to give a shout out to let's do a general shout out to everybody the sheer amount of people that is behind all these tools because to maintain something like Vite and if you consider a whole ecosystem it would be nice to have this kind of graph where all the people and projects are connected and all the relationship because you will see how much work there is behind all these tools so shout out to everybody that is from people doing it in their spare time from companies funding people to do the work that is really important to everybody that is just funding the other developers that are doing this and users doing nice reports with minimal reproductions everybody collaborates to create something in this awesome thank you so much for joining me today this was great I love seeing your perspective on how those things come together and really thank you so much for joining us thanks a lot it was a great conversation thanks and then thanks everyone for listening today and see you next time bye bye