Bun 1.0, Restarting JavaScript, and Darth Vercel
I am just curious because is there just gonna be a future where we just never see Adam? How's it going?
Adam:Good, I think.
Dax:You think?
Adam:I think. I think good.
Dax:You can't tell how you feel?
Adam:I don't know. Just something about you you started this you said you'd throw it on Twitch. And it occurred to me, like, how long it's been since I've even been on Twitch. And and it's it's almost like I don't know. Did I just have a little stint as a content creator and then I decided that wasn't for me?
Adam:Is that what happened? I can't decide.
Dax:I've thinking I've been wondering that too. Is he
Adam:About me or you?
Dax:No. About you. He's just been slowly retreating. You're not around as much.
Adam:I'm really not.
Dax:And you're just a fading memory in all of our minds.
Adam:The way you put it, they're just it stings a little. I got yeah. I got on Twitter yesterday for, like, fifteen minutes, and I sent, a fury of messages and replies and whatever. Yeah. I don't know.
Adam:I don't know if I'm coming back. I don't know. Maybe this is just a gradual it was a what's the it was a soft landing. No. What's the phrase I'm looking for?
Adam:There was a phrase in pop culture and I can't think of what
Dax:it is. Soft landing? Oh, like fifteen minutes of fame?
Adam:No. I wasn't going there. I was just No. Like, instead of like an abrupt, I'm leaving, I'm not coming back to the Internet, it's been kind of a slow fade. Oh.
Adam:A slow burn away.
Dax:I don't like that. I don't like that you're doing that if that's what you're doing. If that's what you end up doing. But then what are you gonna do? You're just gonna you're just gonna be alone?
Adam:Well, no. I mean, I have my family.
Dax:You don't have any friends.
Adam:Okay. This is getting okay. I was thinking about it yesterday, actually. Like, what what did I want out of doing any of that? Like, anything I did publicly, what was I what did I really want out of it?
Adam:And I can't think of a single thing, like, that I actually want out of it in life. Like, I I think meeting all of you, people I consider friends now, that was probably the only thing I really that will stick that that I actually, like, lasts and I care about.
Dax:Yeah. Like, I
Adam:I'm excited to go to TwitchCon, meet with all of these friends I've made on Twitch. But I yeah. I don't know how much I care about doing any of the things that you do as a content creator. I don't even like the phrase content creator. I don't even like saying it.
Adam:I don't like it coming out of my lips.
Dax:Well, I don't think of myself as a content creator.
Adam:I don't either. You're not. You're a developer. But you talk about things.
Dax:You don't have to be a content creator. It's funny though because I get, I understand there's like it's unclear. I think people are confused by exactly what the parameters of what I do are. Because I get messages all the time now from people asking me to do like content creator type things. And it's kind of weird because I'm like, I actually don't do that kind of thing.
Dax:Because it would be weird if I just started to like because it's always people that are like, hey, you wanna like we're working on this product, like, would you like to like talk about it or like it would just be so random for me to just to start talking about some product or reviewing it. It's like, that's actually not anywhere close to anything I do. Like, I'm just out there trying to sell my own product. Like, I don't I'm not I don't have time to also sell your product, you know. Yeah.
Adam:That would be weird if you did that, by the way.
Dax:It would
Adam:it would feel weird. I'm muted on Twitch, by the way. Guests, no audio. Is it possible that they're all trolling us?
Dax:No. They're not because I don't I don't see your bars moving.
Adam:I'm still talking. This is me I'm an idiot. Linux is There the
Dax:we go. No. I was just clicking the wrong thing. I was just changing the input on my mic thinking that would I'm dumb. Okay.
Dax:There we go.
Adam:You are dumb.
Dax:Here we go.
Adam:We agree. I'm just kidding. You're not dumb.
Dax:Cutting the podcast is gonna have a hard time. Thankfully, we don't rely on our own competence to do the podcast recordings. That's just all Riverside.
Adam:I'm sure our podcasts are great because Chris is great.
Dax:Yeah. Where are you? Do you think you'll ever find do you think it was just like a phase?
Adam:I yeah. I don't know. I have this thing where I go through phases. I don't know if you know this about me.
Dax:Yeah. It's like you're defining characteristic.
Adam:Yeah. What what did I do with that? Like, what what what good is that in the world to be, like, really good at having really intense phases and then never visiting it again?
Dax:All of life is just one big So, you know, it's not any different.
Adam:I don't understand.
Dax:You live and you die and then you can say, what good was it all? But, you know
Adam:Well, it is very different. Like someone who's like, I'm gonna be a lawyer for thirty five years and then retire. And I'm like, I'm gonna be this for two months really hard and then I'm not gonna do it anymore. What do I do with that? I mean, I have stuck with software development.
Adam:Like, I've done that for fifteen years now. So I do have some staying power, I guess, when it comes to, like, things that pay the bills. But when it comes to hobbies Yeah.
Dax:But I mean, the hobbies are gonna rotate. Feel like I think even for even for me, like, I'm I haven't really been on Twitch as much. But I am just curious because is there just gonna be a future where we just never see Adam?
Adam:That's what wonder. I wonder if I will completely fade out of public, sphere 100%.
Dax:And I'm less interested in that from, like, an external point of view. I'm more interested in that if you're gonna, like, what are you gonna be doing with all your time? Like I said, you have no friends.
Adam:I listen. We'll still do the podcast. So, I mean, can guy who is on a podcast twice a week really say he's faded out of the public?
Dax:Depends how well the podcast is doing, but yeah.
Adam:Yeah. I guess so. Yeah. It does feel though it feels like the more I drift away from Twitter and I haven't streamed for months and I have no intentions of ever creating another YouTube video, it does feel like there's a world where I'm really not on the Internet much. And I don't know that I feel that bad about it.
Adam:I just I just don't know what I'm giving up. I think, like, friends. But, like, I can still be friendly and hang out on Discord and talk with people and go to events occasionally to hang out. Like, there's still, like, plenty of opportunities for touch points with people I actually wanna interact with. Yeah.
Adam:Yeah. So I just don't know what I'm missing. Like if you did what I'm doing, if you just faded out of all these social media places, would you feel like you're missing a certain something? And what is that something?
Dax:Yeah. Because I think for me, it's all roped up into what I'm trying to do career wise. Like I'm trying to have Uh-huh. Certain impact, be able to pull certain levers like and I've just been building that up over time. So for me, all that stuff is highly connected.
Dax:I definitely have it's just again, one of those things where when you first start doing something, it has to be like the thing you think about all the time. Then you get better at it and it's like less of like the primary thing you're doing. So I think I've definitely shifted out of, okay, like, you know, establishing a presence online is like my primary focus, like, that's like no longer something I have to think about. It's a little bit automatic. Yeah.
Dax:But it's not gonna ever go to zero or I can't imagine it because it's it's it's so tight in. Like I I can't do any of what I wanna do if I'm invisible, you know.
Adam:Yeah. No, that makes sense. So my I guess my work is pretty detached from anything I do publicly. The public stuff was always just sort of like a bolt Yeah. It was optional.
Adam:Yeah. Yeah. Yeah. I see how it works for you. Like, you've kinda got this whole thing going.
Adam:It's just it's a nice flywheel. Yeah. I don't have that. Yeah. I do miss just seeing like names in Twitch chat.
Adam:I do miss the hanging out. Mhmm. Just the aspects of but I mean, like, I it was it was never sustainable in the sense that I wasn't really working when I was streaming. I don't know if I'm capable
Dax:finally of admits it. After after, like, years of denying it, the auditing thing is just finally
Adam:admits was really eating into my amount of productive hours in a day.
Dax:Yeah. Likewise. Yeah.
Adam:But maybe yeah. Maybe there's a slower I'm just not good at, like, slower burn. I'm not good at, like, I'll just do it once a week. I'm just not good at, like, dipping my toes in things.
Dax:No. It makes sense. It's it's hard. Like I said, as soon as I lost the habit, it just went to zero. It didn't go to, 20%.
Dax:So Yeah. I think based on this stuff, it it is. So let's let's change gears. Yeah yeah yeah. And talk about the one thing we have to talk about because Bun one point o came out last week and that's
Adam:Oh, right right right.
Dax:That's been everywhere?
Adam:It's everywhere.
Dax:Well, mean, have you I mean, you not being on Twitter as much as it as it shown up
Adam:in some of face? I've been on Twitter, like, twice in the last week, and it has shown up in my face quite a lot. And I I read the blog post. I mean, I read the release, just all the details around the release. And impressive.
Adam:I'm I'm gonna let you start though.
Dax:Yeah. I have like infinite thoughts on this. I can probably talk about this for a while. Yeah. So BUN one point o came out.
Dax:I guess to me, I didn't really see initially going into it, knew it was gonna come out. I didn't know I like, was kind of like, this doesn't feel major to me because I've been trying to use it like pretty much once a week I would like check-in and and see how it works, see like what part like what part of SST would it fail on. Yeah. And every week I checked that in, it would get a little bit further. And I was like, at one point I was just like, you know, it must be the same thing.
Dax:But when I watched their video that they put together for their, for the official release announcement. It was actually pretty exciting, because I think I kinda got desensitized to some of what they were doing because it's been following it for so long. But then putting it together and laying out like, here's every single thing we're trying to do, here's how much progress we made on it, here's all the little nice things that we fixed. It was cool to see that all together and I see and even I got a little bit more hyped and I see why the reaction from the community was insane. Like people are going crazy.
Dax:Everyone's super excited and happy and it's kind of all I saw people tweeting about. So overwhelmingly positive, like very few things are received as well. Of course, there's the random haters that are always there but really very very few of them. I think everyone just sounds super excited.
Adam:Yeah. It's like I didn't realize they were taking on all these I didn't realize it'd be getting rid of my package manager and it'd be getting rid of my bundler potentially. It kinda makes the Turbo Pack and the, like, weirder extent Roam just look like like, what were they what were they doing? What are they do why why are they still doing stuff if if this was happening and it happened so quickly? Yeah.
Adam:I mean, I say so quickly. I don't know how long they've been working on it, but it was just Jared for
Dax:For a year at least. I think I was following him for like a year when he first started talking about it and then he raised money and that was another year. So it's probably like a little over two years, I would guess. Yeah. Obviously very impressive.
Dax:Yeah. So I saw that and I saw what we're talking about and if and I got more hype than I was. And then I went to go, okay, I'm like, okay, now that it's out, people people have been prodding me to do this but obviously a lot more people started to ask, hey, can we deploy, bun functions bun fungs bunctions, whatever you wanna call them through SSE.
Adam:Is it really bun funds? Is that That's what I've
Dax:been saying just because it's fun to talk about. Funds.
Adam:Yeah. Bun funds.
Dax:Everyone's upset of bun for not using bun time as the runtime thing. My argument was if they called it bun time, I think they would delay enterprise adoption by, like, three years. Like, already with the
Adam:You know something about delaying enterprise adoption.
Dax:Yeah. Exactly.
Adam:With extracurricular activities.
Dax:But even just the logo, I feel like something back at least two years.
Adam:Oh, yeah. It is really cute, though. And did I see they made cookies? Yeah.
Dax:The the cookie looked really good.
Adam:They did look really good. Yeah.
Dax:The the whole event there.
Adam:I'm cutting right now. Like, I'm trying to lose another five pounds in a week. I've lost five pounds a week for, like, three weeks. Yeah. I wanna eat everything.
Adam:And everyone keeps posting all these pictures on Twitter of, like, delicious looking stuff. Sorry. Continue.
Dax:Yeah. Even if that, like, random cookie they got is getting you excited, I'm sure.
Adam:Mhmm.
Dax:Yeah. It sounds like you've been on your diet for a little bit.
Adam:Yeah. It's been a bit.
Dax:I started to implement so I oh, okay. Let me just try the AWS Lambda custom runtime, get bun in there. I talked to, I think his name Max Maxim? He's the one that has a really cool site that does like the daily benchmark Yeah. Runtime.
Dax:Yeah.
Adam:The benchmark's there.
Dax:Yeah. He's gonna get bun added in there. I think I've gone through this exact hype cycle of bun before too. Like, I realized like, like, when I talk about this, I don't wanna sound negative on the project. Like, I'm very excited and optimistic for it but I just need to like lay out some of realities.
Dax:There's a lot in the way and the end result might not actually be as exciting as we want. There's like so many different factors to this. So we can't really use it on AWS Lambda because we I put it on the custom runtime. It's like a five hundred millisecond cold start. It's not really viable.
Adam:But but I saw, like, where Jared quote tweeted your your tweet and Beznick what's his name? One of the serverless DAs responded that they'll take it to the product team. Do you think there's it's like, why couldn't AWS
Dax:They say that every single time with everything. So Yeah.
Adam:Do you so do you think, like, there's there's real, it's a long shot that they would actually add bun as a supported runtime
Dax:Yeah.
Adam:First party.
Dax:I think it's a long shot. But if I was someone at AWS and I was in charge, I would be like, we need to figure out how to be the first to ship native support for this and like really compete with Cloudflare Workers. Cause that's like been like the thing that's in the background being like, are they gonna supply AWS one day? Yeah. And I can see them really doubling down on bun, maybe even like acquiring it and like just being Interesting.
Dax:The best place to run. Again, you don't have to change your applications. That's a main issue with Cloudflare Workers. Amazing platform, but it's like completely next generation. You have to build all new stuff for it.
Adam:This would be just node compatible everything.
Dax:Yeah. Yeah, exactly. If like a large percentage of node applications could just invisibly get upgraded, that's like I can't think of any other company in the world better positioned than AWS to like really take advantage of the work being done here. Yeah. I don't I'm skeptical that they will put this stuff together.
Dax:I feel like they don't really bet in this direction. I think they like more react to we have customer demand and it's reached at a certain threshold and then we'll do something.
Adam:They're doing all their all their betting on the AI space right now, I think. Right?
Dax:Yeah. Probably they're And that's not even a bet. Right? That's just like already
Adam:That's kinda reacting to it. Yeah. Yeah.
Dax:It's I wouldn't say so but Yeah. So I I just don't feel like this is the type of thing that they really do. And additionally, I and I said this a few times on Twitter, whenever AWS adds like an adapter pattern somewhere where they're like, you can customize it or you can like write your own plugin, these never ship another native feature ever again.
Adam:Oh. Has there not been another a new runtime since
Dax:Nope.
Adam:They allowed custom interesting.
Dax:So they launched custom runtimes and they they launched Docker container function support, and they've just never done anything native since then.
Adam:Do do you have any kind of maybe this is a stupid exercise, but do you have any ability to kinda, like, guess what the benchmarks, what the cold start performance would look like with Bun?
Dax:Yeah. A lot of the work Bun put into was to improve the cold start time. So they like put in the work I think again, it's really hard to say because I haven't tried with the root application. But I it's it's it's just kinda hard to say because Yeah. The there's already a floor from like the firecracker startup, so I'm only talking about what happens after then.
Dax:Yeah. I think if you look at node 18, it's around like a hundred and eighty milliseconds. I wouldn't be surprised if bun got to like sub 50. I don't think it's gonna get to sub 10 like Go or Rust.
Adam:Or Rust. Yeah. Okay. So it's not getting to that level but it would be like quite a bit faster than node potentially.
Dax:Yeah, exactly. So again, that is kind of good, but is it like enough to make it all worth it? Maybe not.
Adam:Yeah.
Dax:So then you have runtime performance. And obviously, they have like really insane benchmarks, like crazy crazy crazy benchmarks, which are all accurate. Can't really argue with it. I think they're like fair benchmarks. But again, the reality is, the biggest bottleneck that most people have is they just wrote bad code And Bunn can only do so much for that.
Dax:And a great example of this is someone from the CDK team got CDK working under Bunn. It's not like a five x improvement. It's not a three x improvement. It's not a two x improvement. It's like a 1.3 x improvement or something.
Adam:What are you implying about CDK code?
Dax:I'm just saying bun even bun can only do so much, right? So there's just a reality of like, okay, for real world library application like CDK, okay. Is that worth kind of the upheaval? Maybe not. Then you have the serverless functions world where all of my functions pretty much do a tiny bit of logic and then schedule IO work, whether that's IO to the database, IO to whatever.
Adam:Yeah. Right.
Dax:So even there, like, it's not gonna give me a crazy amount of benefit, for like single functions execution time. They put in a lot of work to make like, if you're running BUN in a container and it's like directly handling HTTP requests, they put in a lot of work for that to be able to handle like a large amount of concurrent requests, like way more than the competitors like Node and Yeah. Node and Deno. But again, it doesn't matter for
Adam:That doesn't really yeah. That doesn't apply to your usage of
Dax:Yeah. And anyone doing serverless stuff. Like, is that routing concurrency part is already handled by something outside of your runtime in a native language. So, yeah, just not that exciting from a serverless function point of view. I don't think it's gonna move the needle there.
Adam:Yeah. What about from a, like, developer workflow standpoint, like your local machine, your your dev feedback loop?
Dax:Yeah. So in terms of being a better package manager, I think that is definitely there. It's like pretty crazy how fast it is at copying files, like, it's like you I thought p I I was skeptical that it would be much faster than p n p m because the p n p m avoids copying files.
Adam:Yeah. Right.
Dax:But somehow it's faster than something that's avoiding copying files. So I think that is cool. Is that really enough, like, using BUN as just your package manager? Yeah, maybe. I think I could see myself doing that one day.
Dax:I'm actually trying to do that now just to force myself to get Yeah. SST working better in BUN because if I'm using it myself, it'll be in some device too. But that's cool. Bundling ES build is already so Yeah.
Adam:Where does the bundling fit in? Like because, I mean, everybody just uses one of these big meta frameworks like React or I'm sorry. Next. Js or Yeah. Astro or whatever.
Adam:Remix. Somebody's using some meta framework that they chose to build their application. Like, do do they have the option of using bun as a bundler instead of whatever that framework uses as a bundler? Like, is that how
Dax:it works? Yeah. So buns, the bundler portion of bun, they actually kept, they actually mirrored ESBL ESBL. ESBL. ESBL in a lot of ways.
Dax:So one they kept Like plug in
Adam:API and stuff.
Dax:Yeah. Plug in API. They have a very similar API in general. So I'm like very familiar with ESBL because we do a lot of that stuff for SSD. Switching that over to BUN would be very easy for us to do.
Dax:Also, they mirrored the implementation. If you look at the, the parser or like the TypeScript parser
Adam:Mhmm.
Dax:And the compiler, I'm pretty sure it's like almost like a part of ES build is doing, if I remember correctly. So it's very similar to ES build, which makes it good because you get predictable behavior in both. But ES build is already so fast. It's like already Yeah. Crazy crazy fast and I have the same issue with Turbo Pack.
Dax:I'm like, why is Next. Js like rebuilding the whole world when it's like basically gonna be the same speed as ES build? Like, so I'm kind of questioning that so many of the frameworks are already built on top of VEET plus ES build, replatforming those. I just don't know if it's gonna be worth it. I'm sure people will do it, like someone will be really motivated to go and do it.
Dax:Is it gonna stick? I don't know. Veep themselves might switch it under the hood, then kind of everyone kind of transparently switches. But Yeah. Yeah.
Dax:It's just like, don't know That'd be big move,
Adam:I guess.
Dax:Yeah. I don't know if that's enough of a game changer. It's just like there's such burden that comes with switching something like that, like the new bugs and not having as many maintainers or eyeballs as he has built. So yeah.
Adam:Yeah. So what is still slow? Like, I feel like type checking
Dax:Type checking.
Adam:There is the effort to like make type checking faster. Danny poor Danny.
Dax:Donnie. I don't
Adam:know if he's still working on Donnie's Danny's
Dax:in a deep depression now.
Adam:Yeah. So they're not taking that on. They did do something. Transpilation. What are they what did bun do related to TypeScript?
Adam:Like, you don't need TS node anymore. You can just run TypeScript files, but they don't do type checking.
Dax:Yeah. It it's kinda similar to Deno in that the transpiler is built into the runtime. So it's just doing the transpiling on the fly and it's so fast that there's not that much of an advantage to pre transpiling it. Yeah. Again, nice like little d x win but it it's it is so weird because it's so obviously better but then now if you start to like drill it down in all these different ways, it's kind of like
Adam:It's hard to fit it in Yeah. What's for it's the really yeah. The path forward.
Dax:Yeah.
Adam:And and you talked on Twitter about the like there is no business model for a runtime yet. There hasn't been one created so they'll have to figure that part out.
Dax:Yeah. That's the whole question on like where like, if there is demand somewhere, then like how do you build now a business model on top of the demand? Yeah. It's it's a it's a lot to figure out. But the thing I was saying because I wrote a few like posts that were kind of, I would say on different sides.
Dax:Like I one, wrote about difficulty of a business model, but two, also wrote about why even though this is so hard, you kind of have to attempt something like this and a VC has to fund something like this because on the off chance it's successful, it's like an alias sized opportunity. Like like, if it is, most likely, it's probably not going to work out. But if it does, it's going to overwhelmingly dominate everything. Like not just the jobs with ecosystem, it'll probably pull people in from other ecosystems. It's probably an opportunity to build like one of the biggest, like, just dev companies that's ever existed.
Dax:So the opportunity is there but, you know.
Adam:So so for Evan, for for the company behind Bun, you think it's still it's a it's a long shot, you would say.
Dax:Yeah. And I'm sure they would be the first to agree.
Adam:I guess most startups are. Yeah. Right?
Dax:Yeah. But it's like on the scale of like, this is gonna be a $100,000,000,000 company if it succeeds.
Adam:I mean, they're just so like, Jared, they're so good at executing. I feel like what is their pivot out of this if it just doesn't work? Like, if a runtime just can't be a a successful startup unicorn?
Dax:Probably like a like a modest acquisition probably.
Adam:Okay. They're just so good at shipping stuff. I mean, clearly, they've worked
Dax:Yeah.
Adam:Very fast, very effectively.
Dax:Yeah. Exactly.
Adam:And it sucks because like you said, it's like stuff we all want. Like, oh, you're making JavaScript way better. All all the fronts. It's great. I'm excited.
Adam:But to your point, like, in the real world, it has to, like, find its its path into everyone's life and that's a challenge.
Dax:And I think that's why they focus on the package management stuff because that's like a very easy way to at least get a foot in the door. Again, I don't know where it goes from there but if that becomes a de facto package manager because it's just so fast, that's pretty good pretty good start.
Adam:Yeah. The problem for me with the package manager front is like, I feel like as a community, we all just collectively decide like, oh, PMPM's so nice. This is great. We just made this transition. It's like
Dax:Timing is tricky.
Adam:Feels cozy. The people that make PMPM, we all love them. They're like I don't know. It just feels like, oh, now we're switching off PMPM? Well, now that one hurts a little bit.
Adam:Like
Dax:Yeah. And there's just like a whole set of things that comes with this. One of the reasons I like pnpm is they stop the default behavior is not to hoist. And I feel like there are like an insane long tail of frustrating pain and bugs that come from hoisting. And I was so glad we got away from that.
Dax:We're like, okay, we're back we like stopped using this weird hack where like lifting packages up technically still works. But bun, you know, is hoisting based. So that's kind of what's preventing me from switching to it as a package manager. I'm just like it just introduces so much pain I finally got away from.
Adam:It does seem like the bun philosophy, maybe it's the Jared philosophy, is such that like just like make anything that they could see work. Like, take as much burden off the user, like everyone's dum dums. And they I feel like he's done a lot of work, created a lot of features to, like, make stuff work that maybe shouldn't work.
Dax:Yeah. Exactly.
Adam:It's just, like, surprising. Oh, that actually works. Like, you don't have to do all this explicit configuration and stuff will just work. That seems like a kind of philosophy, but that does sort of, like, to your point, it's different. I mean, like the PMPM going more strict was nice.
Adam:It sucks when your stuff doesn't work but it's like pointing it out ahead of time.
Dax:Yeah. And it's funny because the the block so as a BUN one point o, I ran it against SST and we the first area we ran into, it's actually a weird case of us doing something correct, bun compensating for people doing it incorrect and then breaking our correct approach. So there's always like stuff like that that comes up. So now we have to add we have to add like some really weird code to again, the root issue is Babbel. Babbel doesn't ship ESM.
Dax:We import it correctly, but bun doesn't like it when we import it correctly. So we have to import it incorrectly just in case it's being run-in bun and then import it correctly if that fails. Yeah, like, technically it's better for the end user that's doing all kinds of weird stuff like mixing CJIS and ESM, but then for people like us that do ESM correctly, we now have the burden. So yeah, it's just really hard to get that stuff right where like and we actually with SST, I think we have to run this balance as well, like, and I think different people on our team have different approaches. Like Frank is more on the extreme of we will find a way to fix it for the user.
Dax:Like we'll do something weird and fix it for the user. And I'm on the other extreme of like, really being conservative about doing stuff like that. So we end up somewhere nicely in the middle, I think. But yeah, it's like that that balance is is hard to strike.
Adam:Yeah. So what about this? Maybe maybe this world you've talked about where JavaScript needs to restart, like, clean break. Just burn it all down. Start over.
Adam:Maybe bun could be the thing we all just start over with. Could we could you create your little package certification thing and that coupled with bun, coupled with you're the new JavaScript dictator. I don't know what Jared is. Jared's too busy to be a dictator. Could we do it?
Adam:Could we just build up from the ashes? Burn the whole thing down?
Dax:There's just so there's just so much because I feel here's what I feel like. I feel like we are close to making JavaScript on the back end decent. There's just like a there were just like a few missing pieces and I feel like those are coming together, like we're close. And now that bun exists, it kind of resets that in some ways because if we do want to re platform and do everything on bun, so much doesn't work again. Right?
Adam:Yeah. Yeah. You're back to missing things. And from parties that you can't rely
Dax:on Exactly.
Adam:To like
Dax:Like I can't ask you to do a support bun. So that just is like a non starter at that point. I mean,
Adam:you can ask.
Dax:Oh, I've asked.
Adam:Hashtag AWS wishlist.
Dax:Oh, yeah. Did you see what I posted when Jerry quote tweeted it? I was like No. I was like, I'm willing to stop talking shit for one year if AWS supports them. If that's
Adam:take that deal. Oh, they gotta take it. Yeah. That's a good deal.
Dax:I'm sure that has a good financial value in terms of like the amount of stress that comes off of the people.
Adam:They'd have to acknowledge it though.
Dax:Yeah. That's an issue. Yeah. They'd have to acknowledge it.
Adam:What about formatting? Who's gonna make formatting fast? Did Rome figure that one out? What is up for any of these?
Dax:Was thinking about this this morning. I was like, you know, everyone always complains about fragmentation in the JavaScript world and like how we can't stick to a tool. But you know what? Prettier freaking did it. Prettier got in there and everybody uses Everybody
Adam:uses Prettier.
Dax:We all see that thing Prettier does where we have a long template string and we have like a little variable in the template Yeah. String and it breaks into three lines and it sucks. But we all just do it, use it anyway. No one's out there complaining.
Adam:It's true.
Dax:And they yeah. They just got in there and they're the default. And it it's slow. It's always annoying to get integrated with the editor. But, hey, we found at least one thing we've all standardized on and we can do it.
Adam:Do you think there would have been more outrage if a major open source project got rid of Prettier than TypeScript?
Dax:Got Got rid rid of of Prettier. Prettier. Actually, Because
Adam:Something that a lot of people contribute to.
Dax:No? That does seem yeah. That does seem in lot of ways even more trivial to, like, care about that. Yeah. I don't know.
Adam:That's DHH's next move. He's gonna rip prettier out of the turbo
Dax:repo or whatever. I don't know if they're using it right now.
Adam:Oh, maybe not.
Dax:But, yeah, the only thing Rome ever shipped was their formatter.
Adam:Yeah. So can we take that formatter? Was it good? Can we is it fast? Yeah.
Adam:Is it written in Rust?
Dax:It's fast.
Adam:Don't know. Blazingly fast? Okay. Can we just rip that out?
Dax:There's a few. Right? There's a few that I've seen. There's at least two
Adam:Well, or Turbo Pack, they're doing it, aren't they? Aren't they also doing the formatting stuff?
Dax:I don't know.
Adam:I thought they were doing a native formatter. It's like there's like four different groups working on like, there's gonna be one tool
Dax:Yeah.
Adam:For all of your JavaScript development needs. And like, at the end of it, we're gonna have four different one tools.
Dax:By the none of them are gonna ship except for Bunn. I'm I'm pretty I'm very convinced.
Adam:Bunn has proven that it can be shipped in a finite number of, like, number of years, amount of time. Now the pressure's on, yeah, do they does Vercel stick with TurboPak? I don't know. Maybe that's very stupid and presumptuous of it.
Dax:Oh, I have a lot of thoughts on TurboPak as well.
Adam:Oh, please talk about
Dax:it. Yes. Yeah. So I posted the other day being like why, someone ran into some weird issue. I forgot what it was.
Dax:But I was talking about how it's really unfortunate that all of these other frameworks have, consolidated on Vite, which has made Vite very very good. Because not only had this set to support Nuxt, it also has to support Svelte and also support Yep. Solid Start and also so there's like this crazy diversity that just hardening it and making it super, just will work for everything with like really solid APIs. And there's a world where Next. Js, restarted and just built on top of Vite.
Dax:It's not realistic because you can't do that while being backwards compatible and that's like a big deal for them. So we'll to be a new framework effectively.
Adam:And while employing the webpack creator.
Dax:Yeah. Wait, does he work there? Yeah. Or one of the I I I know it's disputed. I feel like
Adam:Oh, really?
Dax:Some people referred him as a webpack creator but someone else claims it was a webpack. There's something wrong I see.
Adam:I don't know.
Dax:Maybe. Okay. Yeah. So their only path was to okay, we're stuck with webpack and webpack API. How do we make this better?
Dax:So I think they went in this weird route where they try to smash together way too many ideas. Like, if we need a successor to turbo to Webpack, let's build something called TurboPag, let's write it in Rust. Oh, but we also have this turbo repo thing, which is also a very good tool. Let's like, kill it for no reason and then rewrite it in Rust. So not only are we like rebuilding the entire everything, like, the ten years of webpack, not only are we rewriting that, we're also like trying to rewrite turbo TurboRepo and integrate it.
Dax:It's like a lot to take on and TurboRepo was already working and it was already written in Go and
Adam:it already fast. It's beyond that, isn't it? The scope of TurboPac, doesn't it their vision is this, like, one thing that does all of the different.
Dax:But isn't that everything that I just mentioned?
Adam:Well, there's other stuff. Right? I mean, all you mentioned was you got rid of TurboRepo and Webpack, but it also gets rid of, I don't know, Babbel transpilation. Right?
Dax:Oh, well, yeah. That's because they're using SWC. Formatting. So they've already done
Adam:Oh, right. Right. They're they've already done that work. Yeah. But then formatting I thought formatting and linting and like all this different stuff
Dax:was mean, also they thought so. Throw anything in there at this point because it's it's just so early.
Adam:It's the Rome Vision. Yeah. It's the Rome Turbo Pack. Well, how do I find the website for Turbo Pack?
Dax:Well, I was talking to some people that about this and people were saying that, they've seen Lee Robs, he's like the head of, DevRel for Brazil. His tone around Turbo Pak has shifted, where instead of referring to, like, when Turbo Pak comes out, it's being shifted more to, like, if Turbo Pak becomes Oh, the
Adam:interesting. So
Dax:I don't I don't I I'm like, it's just a weird, crazy, ambitious thing. Feel like it might have been started during zero interest rate era. And now it's kind of like, do you really wanna do this?
Adam:Straddling that's it's projects that straddle that line. Yeah. It's interesting.
Dax:Yeah. They're like kinda trapped. It's it's it's kinda funny to see that. And now there's this new thing called rs pack, which is just another effort to rewrite
Adam:In Rust.
Dax:Pack in Rust. Wow. But like less less crazy. They're trying to like maintain the same API and then just make up internals faster, which I think is like a reasonable iteration. I think that's like probably maybe what Vercel should try to do.
Dax:So now there's that. I'm sure Next is looking at that being like, well, they we can just let them do it and feel the benefit. So yeah. I I don't know. Don't know if that's gonna go anywhere.
Adam:I'm so curious. I'm so curious how it all plays out. Yeah. The the just broadly. I mean, you're hitting on something I'm very curious about, which is the impact of zero interest rates over a decade.
Adam:Still waiting for that shoe to drop. I keep saying that like it's some ominous prediction. I don't know. I just feel like it's gonna take years, isn't it? It's gonna be slow, kinda like my exit from the Internet.
Dax:I don't I don't know. I don't know. I don't think it's gonna be dramatic. I'm just convinced. I've been thinking more about how, there's a class of founders right now.
Dax:And I'm really excited for this class to graduate because I feel like it's gonna be some of the best founders that have been around for a long time because if you have succeed if you have succeeded in the next five years, you did it during a relatively hard time compared to the previous five years. I gotta be honest, the founders that came out of the previous five years all sucked. Like, I just feel like no one is impressive or like people that are like like at the level of like people like Gizos or Gates or Steve Jobs, like none of that really happened in the last five or even like ten years. But I think there's like an opportunity for that to happen again, just given given the level of pressure and difficulty there is What's right
Adam:the the cycle the the phrase,
Dax:Good times.
Adam:Like tough times. Yeah yeah yeah. That whole thing.
Dax:Yeah yeah. Exactly.
Adam:So Our our listeners have no idea what it is, but we know. So that's okay.
Dax:I don't think any single person could not know. This this is like such this
Adam:gets
Dax:coded so much.
Adam:I only learned it in like the last year. So apparently, I've been living in the Ozarks. Oh, have been living.
Dax:Fun, Veat, Turbo oh, Veat also has an interesting future ahead of itself as well. I think they're like positioned in insanely well. I think we're gonna see some very interesting things come out from them.
Adam:What's the veet? Yeah. What's the beat before I go into this, I have a question. But there's something I wanted to tie a bow on. You know what?
Adam:I forgot. So it doesn't matter. There's one of the oh, yeah. I was just gonna say on the downturns thing. I guess when I think about historically, downturns are very short.
Adam:Like,
Dax:the dot years. Yeah.
Adam:Bubble, like, bursting was like, it it burst. It's fast, furious, and then it was over. So, yeah, I guess it's unrealistic to think that we're in some, like, long term bad, I don't know, depressed state.
Dax:Things are just so different too. Like, if you look at some of the stocks from the.com or even 02/2008, it took those a very, very long time to get back to the price they were at, like, decade, maybe, in some And I was looking at my portfolio and I was like, yeah, everything crashed and I was like, oh, I'm so sad and sucked. But then it just got back to where it was in two years. I can't imagine waiting like ten years Yeah. To get back to where I was ten years before that.
Dax:It's it's really crazy.
Adam:Oh, yeah. Just the timing. Like, if you're, like, someone traditional retirement age, like, that math, that would have sucked. Yeah. Okay.
Adam:So the question I wanted to ask you was, oh, what's the Veep business model? Well I mean, I know how Evan makes money. Like, Evan
Dax:Well, has had own some information that I can't share publicly.
Adam:Oh, okay. Well, that's fine.
Dax:I guess we'll say there's gonna let's just say there's gonna be stuff Veed's future is gonna be interesting.
Adam:Oh. Oh.
Dax:We'll probably get out
Adam:it in
Dax:the next, like
Adam:You'll have
Dax:six months or so.
Adam:You'll have to tune in to the next episode to know. Yeah.
Dax:But they are positioned insanely well because the smartest people in this space build on top of Veat. You have Ryan Carnado, you have Rich Harris, all the people that work on Nuxt. And then like the Nitro thing that came out of Nuxt, like Solid Start is gonna be it's built on top of that. So Astro is built on top of Veats. It's like, you have this one island of Webpack plus Vercel and all the people that Vercel was able to fund.
Dax:Yeah. Then you have, like, the whole decentralized version of it, which is, like, all the people doing stuff in a decentralized way but, like, kinda collaborating.
Adam:Yeah.
Dax:It's yeah. I, like, cannot It's empire. Yeah.
Adam:It's the empire and the rebellion. Mean, you just if you've ever watched Star Wars, there's never been a better, like, depiction of the two sides, good and evil in Star Wars in the dev community. It just feels like this band of, like, ragtag rebels taking on the empire. Right?
Dax:Yeah.
Adam:It's so fixed. It fits so well.
Dax:Because they all them just wearing black in every photo.
Adam:I know. Everything's black and white, like, it's there is a death star somewhere in Vercel headquarters, I promise you.
Dax:You know, sometimes I see pictures from their team where it like, it seems like they're hanging out, it's like an event or out of work or like they're like traveling for car and they're all wearing black and I'm just like, what would happen if I worked at Versailles and I showed up to one of those wearing like like a pink shirt? Would I get in trouble?
Adam:Yeah. Or one of your really cool, like, one of your cool outfits.
Dax:Yeah. Yeah. The really loud noisy ones.
Adam:Yeah. Uh-huh.
Dax:Like, what would would would like someone pull me aside and talk to me?
Adam:Probably. There's surely a dress code. That doesn't just happen by accident. Right?
Dax:I wonder if it's entirely social pressure. I might be entirely social pressure driven. Unspoken. Someone like me, like, that has the reverse effect on because I just wanna be annoying. So someone like that has to work in Versailles.
Dax:There must be someone annoying that works in Versailles that just refuses to wear black.
Adam:I mean, they've hired some pretty big names, so I would think somebody's their own person. And they're like, whatever. I don't do the whole evil empire thing.
Dax:Yeah. Well, and it's funny because, Richarra's works ever so and he's technically on the other one. So he's like Oh. Who's the who's he? He's like the spy or like the, the guy that changes sides.
Dax:Is he Darth Vader? Because Darth Vader turns out to be good at the end, you know?
Adam:Darth Vader too. I don't I don't think
Dax:Darth Vader
Adam:it to true. Be
Dax:Darth Vader. The emperor down that they'll Oh.
Adam:Okay. I forget. There is a more dark character than Darth Vader. Okay. Then sorry.
Adam:I'm working out in my head
Dax:who's who. So Ritter is gonna pick up Guillermo and throw him down.
Adam:Oh my god. You said it. You said it out loud. I mean, obviously, I guess, Guillermo, if I'm saying that they're the empire, then there's a figurehead. Interesting.
Adam:Now I I really wanna map this out. You know what? If our listeners care enough and are nerdy enough to map out who is who in this Star Wars universe Yes.
Dax:Someone do it. Please. Someone do it. That would be amazing.
Adam:Yeah. Someone would do this. Someone will make a little image, a meme. It can be low effort. Just make something.
Dax:Who's the one that profits off both sides? That's I think that's me.
Adam:Wait. In Star Wars, is there someone Is it
Dax:like Jabba the Hutt?
Adam:Boba Boba Fett or something? Yeah. Okay. Jabba the Hutt.
Dax:Or like Lando, maybe.
Adam:Yeah. Just an independent mercenary. You're just making your money off of all the conflict.
Dax:Yeah. I'm just I'm just benefiting off the chaos. I'm just trying to optimize for maximizing the chaos. Yeah.
Adam:I love it. I love it so much. Oh, okay. Evan is Luke Skywalker. Just saw that in, chat.
Dax:Repelling. Well, then then you have to you have to fit Jared and Bunn in there somewhere too because they're
Adam:like Yeah. How does Bunn Yeah. Fit How does Bunn fit in, like, on a serious note, in the, like, Vite landscape? Like you said, if Vite doesn't they're not if they don't abandon ES Build and they don't use Bun as a bundler, how does how does Bun make its way into that community?
Dax:I mean, I think you would have to see new frameworks that come up that are Bun native, which already is a thing. Like, I'm already seeing some of that. I think Alesia is
Adam:one of them.
Dax:And but those have to become a lot more popular than the Vite powered framework. So yeah. It's again just a hard path that we can't really envision.
Adam:At the end of the day, it it always like, I think probably like you think, which is, like, it's gotta run somewhere. And for us, like, we're running all our stuff on AWS Lambda functions. It's like, at the end of the day, if it doesn't run faster there
Dax:You can't do to sell.
Adam:Yeah. Yeah. So that's tough. Yeah. They gotta figure out how to and there's there's no picture for Cloudflare and Bun.
Adam:Is there like Cloudflare is its own thing.
Dax:Yeah. It's Cloudflare is so tightly integrated that it doesn't make any sense. Like, their stuff is already, like, crazy fast because the level of integration they do with Yeah. How they deploy things.
Adam:So if I'm Jared and I'm trying to figure out how do we how do we get the integrations with deployment partners, cloud vendors, or whatever, what is his best path forward?
Dax:There is a path is building something as big as Cloudflare or AWS. It's Oh, jeez.
Adam:Yeah. That's But
Dax:that's the type of thing as a VC you have to invest in.
Adam:Yeah. I mean, they can execute. So, like, you get a team that's proven like that. I feel like sick I'm on a huge problem. But man, that that one scares you.
Dax:Yeah. The thing I pointed out is name one hosting provider that has gotten big in the last ten years. Like, extremely rare. It's very hard to do.
Adam:So is Fly in the last ten years, like, they're
Dax:They're gonna go in. They're like so so pretty early and, they're doing everything right.
Adam:Okay. So you're saying like gotten big, like, top three scale, like, the big big players in the last season. I gotcha.
Dax:And if anyone will do it, it's gonna be Fly. And that's another problem, Fly becomes a competitor. Right? If it really is, like, this big opportunity, Fly can't let that go to someone else. They need to be the place to host BUN.
Dax:They've already They've they already had like a blog post like a few weeks ago, like they're doing the work to be the best place to run run BUN already.
Adam:Well, there you go.
Dax:Yeah. So why is got some Again, if if you're running containers, if you're not in a serverless model, a lot of the work that Bund did can be realized today. Yeah. That's why Fi makes sense.
Adam:But Interesting.
Dax:I don't know. Like I said, the future of this company is a big question mark and
Adam:Exciting. There's some exciting stuff on the horizon. I know there's
Dax:something interesting to
Adam:talk Yeah. So much interesting stuff. Did we already talk about all of it? Are we done? Is that it?
Dax:I think so. Wait. I'll tell you the secret of VEET once we get off.
Adam:Yeah. Offline. Yes.
Dax:That's awesome. Audience, you don't get to know
Adam:Sorry, everybody.
Dax:You don't get to know the secret of
Adam:We're gonna hang up so I can learn the fun stuff.
Dax:Alright. We should probably get up here. Okay. Alright. See you.
Dax:See you.
