The Standup with ThePrimeagen - DHH Omarchy Quattro Announcement

Episode Date: August 22, 2026

DHH joins The Standup to talk about Omarchy Quattro, the new $8M Omarchy Foundation, and why he thinks 2026 is finally the year of the Linux desktop. Plus, AI agents, Linux security, developer burnout, Mac vs. Linux, and what building software looks like in the age of AI.

Stingers on your bros like that, man. I was like, we know you're the primogen, bro. You know what you're just. I just want you guys to know who you're dealing with. Okay, buddy. He starts every company meeting with, we have to all say hello, the primogen. It's crazy. Yes.

Yeah, yeah, yeah, yeah, yeah, yeah. Anyway, sorry. You might get a code of conduct flag here. You better be careful. Yeah. Well, the good news is I also wrote the code of conduct. I have stated that this is okay behavior. It's actually the first commandment of the code of conduct is we all must refer to

him as the, yes. Yes. This is true. This is what I have, this is what I have demanded of them is that I, if they say, hey, prime, I say, excuse me, definite article, please. It's like how you have to call doctors, doctor. You have to call the primogen, the primogen. You know, like, he earned it, you know. You got to come up with your own character on the keyboard, like, like just a simple, a prime simple.

Yeah, that's, that's, that's a tough one because, you know. Oh, yeah, Prince, uh, his own simple. He was trying to push on to typewriter makers that like the official keyboard shouldn't include a print simple. It was amazing. Yeah. He was no, literally as the symbol confidence. Hey, Windows did it with co-pilot, so.

True. I won't buy one. And Dell It's actually not Aipo key It's an Amachi menu key It's very important Yeah, I mean, Dell has

I am looking at mine right now Del have an omachi key I'm not sure what you're talking on TJ What's this? That's how they're off now right? Yeah, what's this co-pilot Again? Is co-pilot the down detector

service? Yes, it is. You have a computer from the future prime But it's going to be generally distributed soon enough Oh, that's what I like to hear. So obviously, can I do an intro today?

Yes. Yeah, if you are to. I mean, I would prefer to do one if that's okay with you. Yeah, go for it. But we do have some things we want to kind of, you know, we kind of want to talk about today. I don't know if you guys know this, but we have a special guest on today. Legendary. I call everyone legendary.

I'm trying to think of something that makes me super legendary. I thought, Prime, I thought you were going to give us the mix. up and go legendary snack consumer trash dev. Oh, that's true. I am snacking. Okay, anyway, so here's the deal. Today we have on legendary pusher of operating systems, specifically making the year of Linux 2026 actually happening DHH.

We have also with us, Trash Dev, amazing snack connoisseur and teach. And today we are going to be talking something extremely special. I don't know if everybody knows this, but it turns out that we have this right here. We have some OMachi news coming in. So I'm looking right here at the board,

at the Conbon board. What are we doing today on the standup? We have to talk about Omachi. And I am saying that correct, by the way. And so with DHH on, we figured that maybe he could do a little bit of setting the stage on why we're talking about Omachi specifically,

because that kind of seems silly. It's been out for so long at this point. I still have my transitions on. But it's, we've had, we, it's been out for so long. Why, uh, why are we talking about it? Well, we're going to talk about Amachi because it was basically rewritten from scratch and the latest version Umachi Quatro, which is the first version that's gotten its own moniker just dropped on Friday and is ushering in nothing less than the future of the personal computer.

We are making computers malleable, personal. changeable, changeable, and it's all thanks to agents solving the biggest problem we've had in Linux that they're a pain in the ass to set up. They're a pain in the ass to configure, they're a pain of the ass to diagnose, yet thanks to the wonder of magical premium tokens, that is no longer true, we have AI to do everything for us. And not only is Amici Cuadro virtually fully AI developed, I spent about three months. together with Ryan Hughes and others, prompting our way to this incredible distribution that we have today.

And then the experience of actually using it once you set up your default agent is just nuts. You can make anything happen. We've rewritten the whole thing in QuickShall. We've dumped about 17 different dependencies. We've shrunk the ISO by 30%. We've increased the install time by like 40%. It now is up and up.

running on a fast computer in about 45 seconds. I lost my world record. You had Omachi 3 world record, which is, it might have even been Omachi 2 world record. It was very fast. True. Hey, Brian, don't forget. Strong planning's later.

No. I can't do this. I can't put you on call. I can't do, I can't do it. I can't do on call. You're on call? Lucky? What are you saying?

With Century, On Call is so much fun. It has all of our errors, logs, spans, metrics, replays, and even releases in one spot. It has all the context we need to solve any bug. Even if half the time Century beats me to the fix anyway. Just because software development's torture doesn't mean On Call has to be. Wow, that's amazing. Does Century also attend my meetings? That's the best part. Century can't actually make on-call fun, but it can make it less scary.

Everything you need to fix real issues in one place. Buceroo. Got that 6pm post-mortem planning. Don't forget. It was already ridiculously fast, but we're making it ridiculously lear fast. Because we've just optimized everything, and it's just it's tighter, it's better, and it's completely malleable. We now have a plugin repository. There's a community-led version on. on Umachi plugins.com that launched six days ago, and it almost has a thousand plugins now. We're uncharted at least to get a thousand plugins

within the first week. People have made everything happen. There's a million games. There's AirPods controls. There is flight radar thingsies. There is just all sorts of, there's even a plugin for finding Baxriac girlfriend. I just saw that.

Whoa. That's a good plugin. Okay? It just crashes or what? It turns off, it turns off arch. It installs windows and you have to go out. But as amazing as all this is, this launch is just nuts. We're running at like 5,000 ISO downloads a day right now.

Even crazier, just dropped. I don't know if you caught this prime because you've been on your stream, but literally an hour ago just dropped a huge announcement that I'm incorporating the Amacon Foundation. with $8 million in pledged funding from just like a few people you might have heard of. Maybe you heard of Michael Dell, maybe you heard of Toby Ludke, maybe you've heard of Patrick Collison or Matthew Prince or Brendan Nareep or Jason Freed or Jack Dorsey. A few of the people we've signed on for this insane mission to actually make Linux on the desktop 2026 happen and beyond.

I mean, this is also an insane raise to do for this kind of mission. Initially, I just asked, floated the idea with Toby on Sunday for this foundation. I'm like, do you know what? I should just put a million dollars into this. And immediately he replies, I'll match. And I go like, what do you mean? You don't have to think about it. You'll just match a million dollars.

And then I went like, well, maybe there are other people who would also just go like, yeah, I'll give you a million dollars to go make Linux totally awesome. And so we just read that, we just announced that. And then we also announced that we are straight away spending a bunch of that money on Vaxri, becoming the exclusive sponsor for him, since that he does not have to think about any commercial development for the next three years, plus a two-year option. So potentially for the next five years, we just set with VaxRee doing hyperland development

and posting Wifu's on X. So that, and then there's a bunch more announcements in. the pipeline. I mean, I'm running this thing as the executive director, but I mean, I'm putting my money in because I don't need the money. So all of the money is going to go towards developers who are integral to the development of Umachi and the subsystems and everything else that we have to promotion of the whole thing, to the infrastructure. And this is going to be amazing. So, uh, yeah, that's my out of breath spiel here for just a few crazy things that's happened this week. I have to say, wow. I'm going to first.

It was such a big announcement that TJ's camera turned off and he had to leave Riverside there for a moment. I was trying to look up on my laptop to show the XPS and say, thank you, Mr. Dell. And then while I did that, I accidentally unplugged the thing that connects my camera. So that's on me. That wasn't a Linux problem.

I just literally physically unplugged it, okay? Amano Machi Quatro, it plugged back in and worked just fine. Take that. Let me just mention one more thing to the announcement because this is, the important thing is here is Linux is fucking happening. We are going to make this happen. I have not been this determined

about almost anything else in my goddamn life about making Linux happen. And this couldn't have happened if we were still stuck with the hardware we had three, four years ago. What's making it happen now is that we finally have MacBook-level quality, competitive hardware.

And two reasons of that. One, Intel, Panthalake, just absolutely mind-blowingly, efficient and performant CPU and Dell revamping that XPS lineup and for the

first time in my career giving me a laptop I would pick over any other I'm not just saying that because Michael Dell just gave me a million dollars although I would have said it now if I hadn't said it already I've already said it fair fair I've been picking the XPS for a long time because it simply is my

favorite computer so I feel good about It's in that order. Otherwise, you're like, you're fucking chill. But now it is my professional duty to talk nicely about Dell. So I'm very happy that that lined up with the fact I just really like what not only Dell is doing, what Intel's been doing. Also, AMD, by the way, even though they're a little bit behind on efficiency, they're also putting out some absolutely cracking products. I know I'm now have to hawk Dell all the time.

But the nice people from AMD also just sent me this AMD 3-9-4-4-4-4. plus laptop from HP, which is the same chip that's in the framework desktop, absolute freaking beast. So if you don't need the battery life that you get out of the XPS, then that's a wonderful option too. So all

the pieces, all the pieces are coming together at the same time. Did you know we're even getting fucking Fortnite by the end of the year? I finally enjoyed Tim Sweeny enough. That's actually crazy. That's actually crazy to get Fortnite.com. After two years of haggling him,

on Twitter, like, at regular intervals. I mean, I'm not going to take credit for the whole thing. Maybe Val also deserves a little bit because, like, Steambox and Steam Dick and whatever is sending. The millions of dollars go into Proton and all that. Yes, they also deserve just maybe a touch. Maybe just a touch. Shout out, Gabe.

Again, look at how many different factors are coming together and conspiring this year, 2026 is when it all happens. And it's the year of the Linux laptop even. Not even the year of the Linux desktop. Yes, that is actually key Because when we talk about Linux on the desktop I mean, nobody buys desktops anymore I mean, I'm talking broadly here

If you look at the app and sales It's all MacBooks. The MacBook was really The number one target And until you had something that was competitive with it It was going to be uphill to sell it Right? Because you were going to ask people to do compromise You're going to either tell them that

Well, the battery life isn't great But you should use Linux anyway Which by the way, great argument I've been making that argument for two years. I totally think Linux was worth the drop in battery life. But now you don't even have to make that argument. Did you see the Dell XPS 14 with the LCD display went 42 hours in a battery life test? I'm like, what do you mean?

42 hours? It's just absolutely nuts. Yeah, to throw this on there, I have the slightly older Dell XPS, the one that we got at Omicon. And that one, I've been able to, I went on a flight to California. to Long Beach just a little bit ago. And during that flight from Rapid City to Denver, I programmed and did stuff. And then from Denver to Long Beach,

I programmed and did stuff. I shut my laptop and then hung out on the beach. And then at the end of the week, I opened up my laptop, still at half battery left. And this is just running OMachi. Like, there's nothing fancy going on here.

This just never has happened in my life of, so the fun, fun side story. When I started using Linux on my laptop was in 2016, And in 2016, I had 45 minutes of battery life. I couldn't make it through one meeting at Netflix every single time. It was this game where I could not open it. I had to keep it turned off until like midway into the meeting so I could open it

and then do some sort of presentation. I had to come up with reasons to stall me presenting. So it wouldn't die. And then during the fans kick on probably to Prime. And it's like, I can't hear you. I can't hear the fans. Are you prompting too? Is it still lasting a long time even with like running?

like LLMs? Oh, yeah. Yeah, to prime. Well, I don't, I typically all my LLM stuff, if I'm doing anything runs in the cloud, I don't use, I almost never use local LLM stuff anymore. But yes, on that flight, I happen to be like, hey, explain to me stuff about Apollo. And I started playing with Apollo, and I just kept on playing with it a whole bunch. So, yeah.

This, by the way, is also key. We're talking about battery live. We're talking about all sorts of things coming together. But the agent thing is really a thing that completely rewrite the game. Because on battery, it was a game of catch up, right? Like we had to get to the MacBook level. And then that was a little bit of a level. Maybe we're a little bit ahead, whatever.

They still make great chips. DM chips are awesome. So it's not like, all right, we just trumps them. But when it comes to agents working on your desktop and turning it into your personal favorite computer, Apple's nowhere. The Mac OS operating system is complete trash when it comes to having your dream computer produced by agents on your behalf

because you can't change anything of the core of how the OS works because that's Apple's domain. I saw a fun tweet the other day where it flagged Codex app as being malware and drag it. This whole idea that I own a computer and for me to get software on there in a reasonable way That producer, first I should go beg Apple, can I please make software and get that sign? It's just mind-blowing insane.

We do not have to live like this. With Agent, we're getting access to the mallable computer, this dream that computer makers and programmers have had since the fucking dawn of time here. And Mac doesn't have an answer. There is no answer. Yeah, you can make some little apps within the confines of what Apple, allows you to do, but who wants to live like that? You want to live with freedom. And can I just

jump in here for a second? It's not just Apple that's bad. It's also just like so much of the programs, they don't even have a way for you to edit the config other than launching the program and click, like the culture of developing an app is also bad. So it's not

just, you know, culture's downstream of the creator, but the creator is bad. And the culture is bad. And so it's just like both of them make it impossible to you, like I hate setting up an Apple because I know I have hours of pushing menu buttons. I don't want to push menu buttons ever.

I want to have one, I want to download JSON and put it right into the system. This is so crazy. I had an experience maybe four months ago. I was going to some place. I thought I had a PC that I was going to install a MacGon.

And like I hadn't been there for a while. So what I had instead was a fucking Mac Mini. So I was stuck for three days. and I did not have a PC because I'd left the laptop of home. So I just had that fucking back mini. And I was like, do you know what? Let me just give it go. It can't be that bad.

And I tried to set it up like I would want my ideal computer to do. And first of all, it took goddamn forever. But exactly as you say, the entire operating system is just not built for agents. It's not built for accessibility of the tooling we want to use. Raycast, great example. So Raycast, great launcher. Lots of people love it, right? The only way you can get your config in and out of that

is exactly as you say. You've got to go to a menu. You've got to scroll down. There's a little export config. And then you get this config. And then to get it in there, you've got to go back and just an import config. Like with your little mouse,

like a fucking caveman. It is absolutely mind-blowing how bad it is. And the only reason it's not like shoot yourself bad now is because of HomeBrew. HomeBrew has done amazing things for the Mac. but it's still limited in what it can do. Like, you can't set hot keys. There's none of the Mac OS hotkeys are credible proclamatically.

The only way to set up your workspace hotkeys, by the way, is going to a menu and manually click from row to row like a fucking pecking bird. It is just absurd. And it becomes trivily absurd in this age of agents where they could just do it for you. You can't check in your conflict. do anything. It is actually mind-blowing now, having lived on Linux for two and a half years and have

built a mod sheet that people just accept that this is how they live. Now, I understand that in the past, even quite recent past, you just went like, okay, yeah, that totally blows, but like I get better battery life. Or, I don't know, build quality, or it gets something else.

That's fair, but this is why it's a complete game changer now. Now that we've leveled out those things, thanks to Intel, thanks to Dell, thanks to other makers, who finally caught up and are competitive on the hardware side, we can finally press those advantages that Linux just has that beats the Mac

bloody. And then the agents entered a picture and you can make anything happen. You can make your computer to do anything. It is just unbelievable that all the stars should line up in a perfect constellation for this moment right now and then we raise fucking

$8 million for the Yamacom Foundation. We're going all in with Amachi. Oh my God. And you got Vaxri. I mean, you got Vaxri going full-time, like just doing what Vaxry does best. Like, that is, that's some real momentum. So this is just a deal that got signed before the announcement. Oh, my God, there's shit coming.

You have no idea. So much shit coming. The pipeline, man, it is bulging with amazing stuff that's going to come out of this. So I have a question. So the $8 million you just got, how is that being used? And I don't know if you can say this. How is this being used to making, like, Linux the desktop or age of Linux, being the stock thing.

Where's that money going? What's your vision, was it? Three main things. A, fund the developers who create the key subsystems for Amachi, people like Backstreet. So doing deals like that such that developers like Backstreet

don't have to think about money. They could just be like awesome Hyperland, which by the way, even though we're now the exclusive sponsor, HyperLine is still open source, free for any other distro could put it in. It's not like, oh my God, now it's off the market. This isn't proprietary software, bro. This is all MIT license stuff.

You go to the Umachi repo. You can fucking take anything and build it into your next dream or something, if that's what floats your boat. So we're going to take a bunch of money. We're going to put it towards that. That's the bulk of it. The bulk of it is going to go towards funding of that. Now, I mean, I raised $8 million, like, fucking just in the last few days, just what? I'm going to make it last, Right? So it's not like by next Tuesday, we've blown it all on hookers and coke here, right? Like, it's got a laugh. In my head, we're spending this money over the next five years. Like, that's, I got a commonest five-year plan here going. So there's five years, and we're going to spend a big lead forward. I'm thinking Michael Scott right now.

Dude, don't go to there. Too soon. Too soon. Too soon. So for a great leap forward or Michael Scott, which one are we talking about? Yeah. Too soon by Michael Scott. Scott. Yeah, too soon. So it's going to go towards that.

Then it's going to go through all the infrastructure required to build a top-tier Linux distribution. Package servers, mirrors, agent runners, QA servers, all the infrastructure it takes to operate

a top-tier Linux distribution, we're going to build it. We're going to run it ourselves. And we're going to take over more and more and more of it until at some point, we'll see how far it goes. I mean, my delusions of grandeur going to the sky. So getting to eventually core levels of Linux distribution,

that's the plan. And then finally, we're going to do something that I know nerds are goddamn allergic to. We're going to market. We're going to tell people that Umachi is the greatest fucking thing ever to... Let's go. Encounter their computer. We're going to spend money on merchandise. We're going to spend money on events.

We're going to spend money on activities that make people aware that Linux is now fantastic and that the best version of Linux for the age of agents is Amachi. So part of this is also like both, it's both extremely ambitious in what we're trying to do, yet also kind of narrow, right? Because I'm not giving up on the keyboard. We're going to be keyboard first. I'm not going to go like, oh, what does it take to get grandma to use this machine? Can we make a watered down version of Ubuntu?

That's good. Distribute a Timo version of Windows. Fuck that. No, no, no. The mission remain. I'm building OMachi to be the perfect computer for me. And then I'm giving you all the tools so that you can start from a system that's already amazing. And then it makes a perfect computer for you.

Right? So what was the question? Oh, how we're going to spend the money. Those are the three main pillars. right funded developers fund the infrastructure fund the excitement can i okay i have a question a very important question especially in in our day and age the i'd say the biggest risk factor to any of these type of projects right now is supply chain attacks i think we see them nonstop rust is under a current one right now with a ray ref i think it was or whatever it was every single one of these

is effectively a worm ready to be wormed on everybody's machine and being a part of a distribution you download stuff and it really does get executed and people really will take all of your stuff and spread even more worming. So with that in mind, what is Omachi's plan to be different? Because AUR, it's like, I've been scared to update my system lately because I know AUR has been also having a bunch problems. And so it's like, how am I going to have a system that I can trust? And so how does Omachi make a universe that I can trust?

I don't want one more NPM. what's the answer? I'll tell you, it's real simple. We're going to own the fucking universe. The entire Umachi installation and distribution is going to be sovereign. Now, we're leaning at the moment on the AUR for a handful of optional installation. There's nothing on Amachi itself that comes from the AUR. Everything is compiled on our built machines and our built servers and then go into our package repository.

I totally get, OMachi's only been around for a year. So if you are basing your current opinion on what a match is on that first version that's shipped as a git checkout in your home directory and ran straight off the AUR, you are about 17 light years out of date with where we are today. We run our own builders. We run our own package repository. We run our own mirrors. All of it lives behind Cloudflare. This is why we don't have to do vintage stuff like torrents and mirrors and shit. We just use modern tools like CDNs to ensure that OMachi has heard of those.

You can download anywhere. I've heard of those. We're going to get rid of the AUR entirely for anything OMachi presents for you as an option. I mean, we're almost already there. There's a handful of things left, and they're almost all from trusted vendors. It's like, oh, you can install Brave, the browser. We get that from the AUR. We don't have to.

Everything that I've been packaging recently, like T3 code, right? I just packaged that for Theo yesterday. And I didn't go to the AUR. I did the original package. Like, I'll go to where T3 is distributed. I'll get it from that. I will package it up. I will host it.

I will control the updates just that I can ensure that it's happening in a secure way. So the AUR is going to basically be reduced to an optional thing that people can use if they choose to, but we're going to keep deemphasizing it. Because we're going to keep getting more and more and more coverage such that there there's less and less reason for you to get to the AUR. Now, I actually think the AUR is amazing. I'm not saying this to slack on the AUR in any way, shape, or form. I do think that it is from a vintage of the Internet where we were more hopeful and happy

people who can trust each other slightly more. But there's 117,000 packages in the AUR. We're not hosting that tomorrow, right? But with everything, it's a power law. If you look at the top, whatever, 2% of packages, 1% of packages on the AUR is probably 80% of the downloads, right? So we just have to cherry pick off that 1%, and then we will shrink the scope that is available for these kinds of attacks. And then we will build out our own repository. We'll build our own build server. We'll build out everything. I mean, we're going to run a

fucking fork at the kernel for the things that we need. We already do. So the reason in part that your Dell XPS laptop runs so well on Amachi is that we run a forked version of the kernel that has patches for that. This is what Spencer Bull, who's by the way, on the newly announced Umachi core team. He's a, uh, he's a distinguished engineer, I believe it is at Dell, who's been on the Umachi train almost since day one. He spoke at Amacom, all in amazing research. He's been doing the liaison for Intel patches that we needed for the XPS to run at its absolute peak efficiency to go into our fork at the kernel. We're going to do that for the whole thing. I'm, I'm talking to everyone right now, including,

Linux kernel developers to come help us out to secure things. One of the next steps we've got to do is this amazing plugin repository, right? That's the same problem. That's the A.U.R. if you download a code from the internet, if you run it, it's code from the internet. It could do stuff to your computer. We're going to fix it. We're going to fix everything.

The plugins are going to get hosted on our setup. They're going to run through our agent-powered security vetting, which, by the way, all of this security drama right now, a lot of it is agent accelerated, right? Like people are finding exploits with agents. Yeah. You know what? That also means these agents are incredible defenders.

And that defense really goes in depth. So if we arm our side with the best, smartest, latest agents, and we basically have unlimited token budgets. Note on that announcement coming later for endless tokens for Amarchi, then we will be able to apply all the best intelligence from this new age towards securing Omachi to. the maximum degree possible. Now, we're going to be liable to all the same things. Everyone is liable to. We're going to be liable to the throat. They exploit in the kernel and whatever. But, but because we are a rolling distro off arch with a one month lag, just that we don't screw people's installations at the middle of the night on a fucking Tuesday. That's how we started. I woke up with Ryan Hughes at 3 a.m. more than one because some package got pushed out and people updated their installation. It was broken. We ended all of that nonsense by just being one month behind on default.

But then as soon as there's anything, a CVE or whatever, we can just update right there. We can ship the latest version, the latest patches, the latest everything, literally next moment. We can have agents monitoring all of it, monitoring the CVEs, opening the PRs, pinging people. We can afford to have on-call staff. We fucking raised $8 million. There's nothing we can't do. you can't um you can't buy cursor with that kind of money uh so you can't do that just throwing that out there it's going to take a little bit more than eight million but eight million is insane

twenty dollars a month i do i think it's very cheap comparatively but i do think there's something to be said about this this insane world that we live in um i forgot who i think it was sam lambert a planet scale said like they run stuff with their databases and they have all this uptime and all this performance and they're a team of 30 engineers or some small hand single double digit more engineers but yeah and i forget the exact number but amazon and their one database has a thousand effective people working on this their database offering and you just have to think the same thing with windows or with mac like how many thousands of engineers product developers all these people putting all this effort into an operating system and then you have vaxry just like

one dude who's making an experience that's just dramatically better. Like, how does this even, how are we even in this world in which one person is defeating a thousand? This can't, like, something has to be fundamentally broken. And it's not just the power of AI because he's been doing this before the power of AI came out. So it's something, it's something more than just that. And so that's why I'm very excited about Omachi, because I want you guys to somehow capture

the eight people who can defeat a trillion dollar organization. And that's all you need. We're absolutely doing it. And it's exactly as you say. All the great leaps forward, I'm not quoting fucking Mao here. I was like, whoa, we're back there again already, DHS. That was fast.

Hold on. The military independence. All the great leaps forward, they did not come out of a goddamn committee. They did not come out of a thousand people replicating your chart. They came out of cracked Polish Linux kids like Backstreet who in their goddamn bedroom at, I don't know, when did he start Hyperland three years ago?

So he was 19, 18, just go like, hey, you know what, I fucking know better. Like, these Mac people have been working on Mac OS and have $3 trillion in market cap behind him, I can beat them. And then he fucking does, right? Like, world is moved forward by delusional, unreasonable people who with enough determination become right.

That's the group I'm tapping into. That's what we're doing. Can I ask another real question? Of course, trash. So you guys have been prompting this new version, Quatro. Like, what's the, like, what's the QA process here? Like, obviously you're dog feeding your own OS, but do you have like a big group of like alpha users? Do you have a QA department?

Like, what's that look like? Like, how can we trust you that this is actually not slop, right? Yeah. So the first is you have a goddamn obsessed maniac, me, who has at this point, I mean 17 lap, and 400 PCs connected to everything. I have installed Omachi Quatro, and I kid you not, probably about 500 times, maybe more. Because just running through the installation,

making sure everything is tight, making sure it all works on all these different permutations. That's just me. Like, that's the first layer of defense here. I am not going to ship anything unless it is goddamn delicious and awesome and works. Then second tier of defense here is we have an incredible community. already. I think we got like 40,000 people on the Discord now or something like that. We have multiple channels. So we have the dev channel. That's people running straight off Git.

Praise be those brave souls. Then we have the Edge channel, which gets the latest packages, and that system has gotten really nice. One of the key, by the way, just quick aside. When I started Amachi, it was a git checkout in your home directory, right? That works great for me. a little brittle if the user starts meddling, and why wouldn't they? It's in their goddamn home directory, right? So we had problems at time with overriding things, and we were sharing configs, and it was a little bit delicate.

Let's just say that, right? Now we've switched to a completely package-driven system. Umachi is a system package that installed straight into all the other system packages, and therefore its own configuration files and so on. They're not in user space. We're giving you access to Omachi and your. you can extend it, but we're not stepping on each other's toes anymore. That it raises like 70% of the bug reports that we had for former versions of Quatro,

just for Romachi, just separating those two things. But then we have the channel system. So we have the edge, which is sort of people who want to live on the edge. They get the latest packages, and they get, I think we're pushing edge packages every six hours by default. Sometimes I push it more often when I'm testing things. But otherwise, you get whatever get is six hours behind. That's great. If you want to run on the latest, there was actually a little issue this morning on some of this.

I fixed that. That's what Edge is for. Then we have RC. So when we get closer to a new release, I go through a release candidate process where we push it out to RC. And that has a whole dedicated group on Discord, like Omachi Dash release candidate. And we have a large number of people really stress it out and can test it on like a million other machines than what I have. Like the 17 I have does not cover the range of Linux computers available. And then when all that is great, then we have stable.

And by the way, a reminder, stable is by default, one month lacking the latest arch packages. And we do that so you don't get that limini update in the middle of the night because someone just pushed something out. And we got bitten by that in the early days of much as well. So we really just like speed run this process of what does it take to build a resilient yet up-to-date system. It's very easy to build a resilient system if you use packages. They're four years old. That's Ubuntu. If you want that, that's available.

Great. People have solved that problem for people want that. That's not what Amachi is. I want the fucking latest version of NeoVim right when it drops. I want the latest version of Codex. I want the latest version of my key developing tools, especially now when like the AIA are updating 17 times a day. I mean, I think that's cracked that it literally was.

was 17 times on a given day that open code updated, right? I want that right now, which, by the way, also super cool. So we have all this set up for all these packages, right? Like, that's using Pac-Man. We have our own mirrors. We have our own package repository, but it's still Pac-Man that does it. And you don't want to churn through that all the time. So you have your system packages, that's so much itself and all the other core you deal like QT and core library, Hyperland itself and Quillard lives over here in system packages. Then you have your tools like OpenCol. that changed 17 times a day. It can't live over here.

If you shove open code into your repository that also keeps your system packages, you now have to update all your system packages when you update open code. No, no, brittle, no good. So we have a two-tier system that runs at separate speed. Pac-Man is running system packages at a reasonable page, you got it, you got R-C, you got stable.

Okay. Then we have Meese, and Meese is goddamn amazing. Means, a lot of it. allows you to install any of the tooling you need for Ruby, Go, Python, all domain development environments, run them in different versions for different projects, and then it allows us to install all the high-churn developer tools, like Codex, like Plod, like OpenCode, like the H. CLI, the Basecamp CLI, all of these things can run on an update chain that is super fast and low criticality and low, what, impact area, right?

Something goes wrong on me's with it like, all right, so Codex has an issue, they'll pick out or push out something quickly. It doesn't fuck up your system. And then finally, I mean, there's a lot here. We're really in depth with it. Finally, we have a snapshot system. Whenever you go to update Omachi, we take a snapshot. We're using BTRFS as our default file system so we can make cheap snapshots. We make a snapshot before you update.

Then when you update, let's just say something crazy happened. You're unstable. We blew our minds and it messed up. right your computer's fucked your arch is now broken nothing works what do you do no biggie you just reboot you get a list of your earlier snapshots you pick the one from one back you go into that and you live happily ever after it is at this point quite difficult to brick a new installation and let's just say that's really nice you manage you manage you manage your fucking brick it you send your agents a muck and they just rip out everything

and they rewrite all your comments. Couldn't be me. Totally screwed, right? Totally goddamn screwed. Here's chef's kiss. At the bottom layer of all these snapshots that happen intermediate, we keep five and so on. You just fuck it all up, right?

We have factory reset now. Factory goddamn reset. You can factory reset. That's a revolutionary idea. Entire goddamn thing to the moment you installed it and I know it worked when you installed it because that was off an ISO that I bet it, right?

So you bring it back to ISO level. Not only will that now restore your computer to its original install set, you can fucking gift it. We have full, like, Mac levels of, like, I want to hand off my computer thing. I've got that on a lot of distros. I mean, I'm sure someone has done it, right?

Like, that's not a standard feature that just works well out of the box with encryption, where you can trust that your data is not getting handed over to wherever you found on fucking Crax list who's going to take over your machine, right? You want that stuff to feel secure, and we now have factory reset built-in. So there's about 17 layers of that resilience stack, and we're going to keep adding more to it,

until it's so got that bulletproof that you need to be a terrorist to blow it up. I'll say I've used these before, and it's really nice. Like, it's a tool that actually we're using in a bunch of repos before, and I like it a lot for, yeah. Quick call out. JDX who runs We're sponsoring him too Oh We're doing all of it

We're doing all of it Everything that sort of fits into this And whatever I mean Linux is I think on my installation I think a thousand and 87 packages We're not doing a thousand any Yet I should say right At the time my budget is 800 million

We're doing fucking everyone right But I'm starting with eight So with eight We gotta make it last a little We can't do everything We're gonna do the key components and it's going to be great. I find Meese to be like a nice middle ground that isn't,

it isn't quite like I feel like I have to have Knicks and then I spend a lot of time like arguing with people on the internet about which companies are allowed to use my software. But instead I just like, I just have my team all use the same version of whatever executables we have and then it's like pretty good for most things, you know. Can I say something? Yeah, of course. This is your show to say things here. Nix is actually part of the reason why we're doing an institution.

That is not just like a one-off thing here. I'm sorry to say, but Linux is fucking infiltrated with goddamn maniacs and just clowns. Just clowns who want to control what you think and who you can associate with and whose opinions are allowed to use what software. it's goddamn insufferable and Nix is perhaps worse of all. Now, there's amazing technology in Nix, amazing technology, right?

They figured some shit out here. But the goddamn cancer that is around both Nix and so many of the other institutions in the Linux world is one of the reason we're like, we need a counterweight. We need a place over here

that's not just a full of fucking goblins who are crazy. We need just some people who are like, Do you know what? We make fun computers. Do you like computers? Then come have fun with us. I don't give a flying. Fuck what you think about other topics or what your politics are or I don't know, whatever you support. You're welcome. You're welcome. You can come over here. If you just want to have fun with computers and find a competent group of people who are responsible for distribution, we're here for you. And you are invited. I don't. We got to get away from this nonsense that like all this shit is teams based. and it's team based along these fault lines of culture war bullshit.

Microsoft Teams. Microsoft Teams every time. I'm not doing goddamn land acknowledgments here before I talk. We're not doing code of conduct fucking persecution shit. We're just running a Linux distribution for people who want to have fun with computers and share that hobby with others. And that's a really broad tent. I think there's a ton of people who would love.

to run Linux if they felt like there was a home for them that wasn't full with crazy clowns. My wife would like Linux. She doesn't even know that there's people arguing about this on the internet. So. Bless her soul. Please keep her in blissful ignorance on that point. Yeah, I'm not keeping her up to date on next package drama. It is a serious problem, not just for normies, so to speak, but also for developers.

There are plenty of people who look at the madness. of what's going on in some of these Linux distribution and just go like, I guess Microsoft and Windows is not so bad. And that's really a missed opportunity here. We can do better. We can just collect a group of same people who want to make good shit and then let everyone use it regardless of what they're going to use it for.

I hope you use it for nice thing. But like if I'm selling forks here, I'm not going to try to peep in your bedroom window and see if you're going to stab yourself in the leg with it. I think that's just not like we've ended the exchange here. I'm going to give you a matchy. Great fucking system. You make it yours.

You write whatever manifestos you want to write on it. You fucking try to take over political. Whatever. Not my goddamn business. Mm-hmm. You know? It's a big tent. It's a crazy world we live in that the, you know, I mean, I personally feel this is a crazy world we live in that we actually have to say all these things that that has to happen.

because I often choose technology, you know, call me kind of crazy. I don't choose the best technology at time. Often I pick technologies because I like the person, right? Like I like Hyperland, and I don't like, I don't like anime wifus on my timeline, okay? I don't want anything remotely pornographic on my timeline. And so when I see these kind of things, you know, like I don't follow Vaxri for that exact reason because I don't, I just don't want it on there, but I love Vaxri. If I could follow him without the accidental wifu titty slip, that I would go follow.

the man, but that's just like not a thing for me right now. And so it's just like, okay, hey, Baxter, I love you. I want to cheer for you. Trash, don't do it, dude. Trash, I can see you typing back through right now. Because I think he's, you know, like, I just like him and what he's trying to do. He just wants to build something cool. And I just don't understand how this is like, we just can't have this world that exists

where you can be like, oh, yeah, I really like his mission statement. This is great. I really like a window manager that's easy to use. Like, this is great. This is a great world coming from I3, which was just not as fun to you, especially now that we have Lua. Lua is so good. You can get Composer 2-5 to write you Lua,

and it's like pretty much spot-on. Like, Lua is dead simple to configure. It crushes it. It's so, it's so good. Gosh, I love it. Do you know what? I love thriller. You know, Michael Jackson.

Okay, good. I was about Michael Jackson. Right. I like the category too. Do you know what? I can listen to that without just listening to the music. Like, I don't have to do a full referee

on the artist here. We've got to normalize that. We've got to normalize this idea that, like, I can buy things, I can enjoy things, I can use things that are made by people who do not share my fucking opinion on every goddamn political issue, cultural

issue, religious issue, whatever issue, wifu issue. Do you know what? I can use things made by people I don't agree with perfectly. Like, this was how the entire world existed. minutes ago. We existed in that world where you would buy a pair of jeans. You didn't know what the

fucking CEO of the company thought about, I don't know, the conflict in the Middle East. And you'd be, you'd be happy. You'd be so happy in those jeans because they were great and they were good fit and like, it was the color was just right. Like, not everything has to be a goddamn referendum on who makes it. Is there perhaps a limit somewhere at the very far extreme? Maybe. Maybe. but none of the shit I see on the daily, on the timeline is even within a trillion miles of that goddamn line. It is all petty, partisan team bullshit. And we've got to get over it. So I have like a side step question here that's a little bit different.

I saw this. And I think it's really kind of, it's a very strong cultural moment we live in right now, especially since you're right now you're up here preaching a bunch about AI and building all these things. And I think one of the something that like a video that just came out on YouTube nine days ago said, I'm done coding with AI. And it has a half million views, 25,000 likes. There's a bunch of people right now in this world that are feeling really burnt out, really unsure, really like a whole bunch of just unstable waters right now, shall we say. And you're just seeing this over and over again. Scott from FM just, or Syntax FM, sorry, not just FM.

Syntax FM just dropped a video and a lot of people connected with it as well about kind of the complications of AI and all this. And then you are sitting here driving and pushing on a project nonstop that I think a lot of people would end up having burnout on, right? Because it's really tedious. There's a lot of tedium on it. It's, you know, tinkering around with OSs are great, but tinkering around with OS is 24-7 can become easy to burn out. I'm just saying, I'm just throwing that out what people think. Okay. just I'm telling you what people think here. And so anyone working on any deeply difficult problem for a long time,

plus mixing agent decoding, have been expressing a lot of burnout. And so how would you kind of respond to this? Because I think a lot of people would be very curious because you've obviously built something. You've gone transition from, I don't think agents are that great to, hey, agents are pretty dang good.

By the way, like same camp, we arrived at the same conclusion about at the exact same time. And so it's like, okay, so tell me, how would you speak to, people in this case. I get it. This is the biggest transformation we've seen in our industry since

inception. Oh, also include burnout. Sorry, I didn't say the word burnout, but a lot of people are experiencing going to get that. So the jump from punch cards to fucking executable code was less than what's happening right now and what's going to happen. There is no parallel for what AI is currently and shortly going to do to the shape of our industry, to the shape of how we use computers, it would be wild if a transformation of that magnitude did not produce some anxiety, did not produce some uncertainty about what the world is going to look like five minutes from now. I 100% get it, and there is no time in history where a transformation of that magnitude has not produce those things. Can you imagine you were living in the 1800s? 97% of the population is out

in the field with a fucking hoe and an ox. What do those have to do with this? Keep going. Keep going. Keep going. No here I can't help you tools, you got modernist. So they're out there, right? Hacking away And then suddenly the fucking steam engine goes by. Like what? Like smoke is coming out. It's on tracks. And it's going to cross from New York to California in like three days instead of three months.

Like what? Crazy transformation, right? They went a little crazy. Who wouldn't go a little crazy? The Luddites, who's like movement, whose word is now what we have a stand in for this. like there were these nice weavers and they were all the clothes was nicely weaving by craftsmen it's really nice but it's also weird fucking expensive and very rare and all these other things and then we came automated weavers and then suddenly clothes good new clothes could be bought by more people who now didn't have to have holes in their goddamn pockets um because they couldn't afford like the handcrafted stuff right you're running you're running uh whatever diligence uh in the wild west with your little hoarfs with your little hoarrow horses here, right? And suddenly the goddamn Model T goes by. Like, what the fuck is that? Where's the horse?

Are you hiding it under the body? What transformation do you think that was like? This is what progress looks like. And I'm not saying that to be glip. Because in every single one of those cases, there were true despair, actually. I hope we don't get to that level of despair. But there was despair, because one day you had a nice livelihood and you thought you knew which way the world turned and the next fucking days, maybe you're out of a job, maybe you've got to learn something new

and suddenly the world is completely different, right? We're at the cost of one of those things. And I get why people are fed up with this kind of talk. Because when you're in the transition, first of all, you got boosters and I mean, I'll own it. I've turned into a bit of an AI booster

because I've seen things, I've built things, I've made things, and my mind is just blown with what's possible. Quatro is the outcome of this. What's happening right now? What's Quatro? Wasn't going to happen without agents. Never going to have the level of ambition I have now without being agent accelerated.

It just wasn't going to happen. We were going to have something else. I mean, I built all of Amundji fucking handwriting my little bash scripts, right? Last summer. No, no, no, no, no, no, no, me typing out bash. That's just how it came to me. It was very nice. I really liked it.

mean, I have no regrets about that. But what we have now is a lot better. That's why it's resonating with so many more people. It's why it's blowing up right now. It's a better product. Why is it a better product? Because suddenly I had an army of goddamn agents, 100xing. What I could do, where my ideas could go, right? So I totally get, you look at this transformation, you go like, fuck. Well, what does that mean? Am I going to have a job? Or is the world going to look like? Do I even like this? Now, when I spoke to, Lex Friedman last summer on his podcast. I said I did not like

AI writing my code. And I really didn't. And the reason I didn't like that was the mode we had at the time was not agents. It was AI tap completion. And it fucking sucked. I hated it with the passion of a thousand sons. It was like constantly someone trying to interrupt me as I'm trying to complete a thought and it was just like

pull my head off. I'm not going to do this shit. I'm going to farm potatoes, I think is what I said, rather than subject myself to this shit. So I thought this is where we are. This is say I. And then agents happen. And agents got good. Not only did they get good, they got amazing. Agents are better than me at almost all forms of programming at this point. I mean, I like to think I add some value here and I'm steering it and I'm reviewing the code and I'm telling it if it's

too complicated and I sell to simplify it and whatever. But like, in terms of comparative intelligence here, I'm accepting that I'm becoming a bit of a pointy-haired boss here telling it what to do. I'd like to, I mean, maybe I'm kind of, I speak nicely to my agents, by the way. I don't have them call me the DHS, like some, the five. Like, so. Thank you for the proper article, by the way. I'm trying to treat the clankers nicely because, you know what, in five minutes, they're going to be in human form.

And, like, I'd like them to remember in their pre-training that I was nice. So, anyway, I mean, just to add more to the existential despair here that Skynet is five minutes from happening, which Dario keeps telling us, right? Like, oh, I don't know. Maybe by the end of the year, we're all. all dead. It's not the best sales pitch. I find it much nice. It's crazy how that's not a good sales pitch. I find it much nicer to do a sales pitch that's based around. You know what? Agents make all your dreams come true. All your ideas are now within reach. Everything you ever wanted out of a computer you

ever wanted to build, you can, even if you don't know all of it. Now, I also like competence. So I hope along the way you level up. You don't just let agents do everything to the point that you don't learn anything. But it's truly an amazing time that we're living in. And I've just chosen to lean into optimism. Because you know what? If SkyNet is launching at the end of the year and we're all dead by fucking February, I'm going to be dead whether I thought that was terrible that was coming or whether I was delusional about optimism. Right? Dead the same way. If Skynet does not happen and we're on the cusp of an age of abundance here and everything's going to be amazing. Holy shit, what a waste of worrying if I sat around me like, oh my God,

it's all going to end. Save your worries for the actual end of the world. Like on that eve, when you can see the mushroom clouds and the Terminator and all their shiny metal walk and crush the skulls at the opening of the movie, you get, right? Because you know what? You have any fucking influence here. Like, this is the other just base reality. Whether you think this is bad, good, or whatever, you're not going to change it. Even goddamn Elon has come around on this point. I saw an interview with him recently where it was like, hey, five minutes ago, you were advocating that we should have six months of memorandum on AI because what if it all goes, tits up?

And he goes like, yeah. And then I sat down and realized that wasn't going to work. So now I just lean in. How fucking based. So we got to come to grips with our own limited ability to direct the future. And if Elon, of all people, is throwing in the towel on personally being able to steer this thing, no one's going to be able to steer this thing. This is simply not steer a ball. Straight out of game theory here.

Like, American can do whatever it wants. If China and the rest of the world does not follow suit and they will not, then that's just not how it plays out. So just comes to the time with reality. Then lean into the reality with optimism and cheer. and then let their dice fall where they may. Yeah, it's kind of like, I sometimes think of, like it stinks that I can't fly. But like, okay, like that just is the way that.

They're flying. Oh, they're air flying. Trash, can you get on that app next for me? I know you're making a digging game, but can you also make a VR flying one? That would be really helpful for me. On it. Yeah, so it's just been interesting.

Like I've had just been having a ton of fun over the past few trying out a bunch of different stuff with like AI and trying stuff that I definitely would literally have never tried to accomplish before we were talking just before we got on of how I I have like this music player that has a bunch of like terminal songs loaded on it stuff from different streams we've done and all these other things right was first off I would have never made a website to host the music and like put it on figure out how to get it on cloud flares that it streams and what else I need to do I wouldn't have done it but secondly I have the repo and I was like, I bet I can do this. And right after I installed a Machi Quatro, I said,

I think I can get this to play like in the little task bar, start a system thing that always runs and turns on. And whenever I open it, always opens the same persistent thing. And like, oh, I bet I can tap my Bluetooth headphones on Linux and have that work. And it was like, sure enough. And it was like literally one prompt. I was just like, I didn't even, I was like, can you give me this? Yeah. Yes. And this is, you are experiencing the Malibal computer. And once you've experienced the malible computer, you're not fucking going back into that little cage that Apple will have to

it's just not happening. Right? Like that is a one-way door to liberation. And I mean, I know you weren't going to go back anyway, teach. Let me say, whatever, much longer than I have. I mean, I've got in torts, right? Like, I showed up about five minutes ago and just got my mind blown and then got my

heart set on complete determination to evangelize this to the whole to the whole world. So, yeah. Can I, can I, I got to throw out something.

So I do want to echo at least minimally the optimism side of things. So, you know, I've had some experiences in the past that have made some difficult, that I made some difficult decisions that I had to then live through. You know, I haven't always been the wisest person. They could probably call me the not so wise prime.

And so that, that's been my past. And recently I had just an event in my life in which me, which made me really deeply consider what is good or bad. I don't really want to talk about the event itself, but let's just say that it was traumatic enough that I had to really, I had to think about truly what is worth it and how hard are you willing to pursue it. And so with that kind of in mind,

I also kind of really came to the conclusion that I love programming. I've always loved programming. I always, I always really, really did deeply love it. And I still love it a whole bunch today. I still think there's a lot of value knowing what you're doing. Most certainly, more value. You will get 10 times further if you know what you're doing versus when you don't know what you're doing. And this whole like read the code, don't read the code argument, all these things,

they're rather, they're rather boring to have because the fundamental argument here is that there's a good way and there's a bad way. And you need to pick the side because you don't want to be on the quote unquote wrong side of history. you need to be into this like certain bucket because if you're not in the certain bucket you're actually in the bad bucket and the bad bucket's going to be filled with bad and sad people because of the future and what I realize is that I just like I have to look at life in such a different perspective because if I don't like come to the terms and just

strive every day to enjoy it more and more like at some point that that switch gets turned off whether you like it or not whether Brian Johnson likes it or not that switch is getting turned off and all the little things that you got yourself so worked up for in those last moments, they don't even come to mind. Like you don't even have the opportunity to think about whether these things are going to be good or bad. And so having that sweet, sweet experience myself, I realized that the things that came to mind weren't, oh man, am I going to read the code or not?

It was like, none of that. It was crazy. None of that came to my mind. Instead, it was more, holy shit, I should spend more time with my kids. like I'm gonna I should enjoy what I'm doing because I got this like one shot and so it's like I better like focus on the good of things as opposed to just like whatever is the most upsetting thing and so I think the the side of optimism whether your situation is ideal or not because I think a lot of people in the chat are going to be like oh this is easy for DHH to say because he makes more an interest than I make in you know working 40 hours a

week sure that there's people that get that opportunity that's just that is just how life works no matter what system we have some people get a better they have a better current means than i do or somebody else does that's just fair but that doesn't mean you have to choose the negative side of things like negativity is always been a choice from the very very beginning in a lot of situations and it's coming from someone who has probably experienced a decent amount of negativity in their life and so i'm just saying that it's so easy to pick it and i don't want to pick it like at all anymore so Yeah, just wanted to write on that. No, I could not agree more.

And I think this is core to that idea that you're going to be who you think you are. If you're pessimistic and nervous and anxious, you're going to be a pessimistic, nervous, anxious person. Who the fuck wants to be that? Like, choose to be someone else. Choose to be an optimistic person who does good work, is fair, is honest. develops competencies, is filled with virtue, like lives a good, honest life. And do you know what? Sometimes that pans out exceptionally well, financially, materially. By the way, I know plenty of people where things have panned out very well materially. And they don't leave envious lives. I don't want to

trade. And I don't think a lot of people would want to swap positions with that. You can live a good honorable life without hitting goddamn jackpot. And that's what you got to lean into. Now, I understand that, like, the uncertainties of the near future are more pressing if, like, you don't have a huge cushion. Yep. You're working on a harder level. I get it. Like, what are you going to do about it? Are you going to live up to the challenge or you're going to tap out?

Live up to the fucking challenge. Wish, don't wish for things to be easier. Wish for you to. Fuck, now I'm butching it. How does it go? You're almost there. I'm almost there, right? hard things.

I think I got it. I think I got it. Don't wish for things to be easy. Wish for more skills, right? Wish for more competence. Wish for more strength. Like, you should be leveling up. Don't try to change the world.

Like, did that, I mean, I'm saying that as a fucking megalomaniac sitting here, right? So you can't try to change the world. I'm just saying, like, even I am not going to change, like, whether AI is happening or not. Even Elon is not changing whether AI is happening or not. It's happening. So these are the cards. Play.

Play and don't fucking whine. Play and don't whine. Play better. And if this hand sucks, you're going to be dealt another one in like five minutes and then try again. And over the long run of your life, if you keep playing as well as you can,

you're going to be able to get the best of what you could. And that's all we can never hope for. Yeah. I just had this discussion with my son. So he's five. We went on our longest bike ride. recently he's been getting pretty good at biking and everything. And we have a few big hills on the way back home.

And I was telling him, Neil, you know, we're at the bottom of the hill. I want you to try and pedal all the way to the top without getting down. If you think you're not going to make it, I promise you, you won't. So just like keep pedaling. And man, when he gets to the top of the hill, he's stoked. I'm stug. It's not much better, not much better than that, especially after getting some ice cream, which was awesome too. So, yeah, I often think about that.

Was probably the peak life experience that are available to humans. And it is not contingent on a billion dollars. It just isn't. The very best things in life are free. The second best things like baganis, Lamborghinis, whatever, very expensive. The best things, free, free. Like that experience, no amount of money could buy. and again

I know people who made whatever billion dollars or whatever who does not have that and I would never in a trillion years ago like do you know what you should trade

like whatever that nice experience with your kid on the bike what about a little more money what have you had a little more money you want to give it up bro I was thinking the same thing while we were on the trail TH one quick question

did Neil go like this of course Neil got to go fast no he well Well, he has seen the ad that we did that in. He likes the Sonic shirt that I wore in that ad too. So he doesn't, he hasn't transferred that idea yet that while I play myself as my son in that ad, he doesn't know that that's what's happening. That's a few, too many layers of irony removed for his five-year-old brain.

By the way, I do want to jump in for a second because I see some various chats being like, oh, this is so disconnected or some version or various form of that. I think the thing is is that a lot of people don't recognize the fact that this isn't just saying, hey, life will be great if you do these X things. Like there's just no guarantee. And some people have it way harder. I've met people who are just, I mean, damn, like my, my editor, I'll flip, but just passed just recently. And he had a lot of hard things that he had the struggle with and fight for.

His life just wasn't easy. You know, these things weren't free and, and simple for him. And like, dude, he, he was, there's so many. positive things about him and what he, uh, what he did. And so it's just like, even in when I meet somebody that has like truly hard circumstances,

he's still like, there's like a lot of things you can take from that. It's just like you don't have to. Um, it's so easy to look at someone who has things and just assume life is easy. I've met plenty of people that have nothing that are happier than me. And I've met plenty of people that have more than I do that are much, much, much sadder than me. Like this isn't what I'm,

what we're trying to say or what I'm trying to say. What I'm trying to say is that at the end of the, the day, choosing to be happy can help. It won't solve any problem. It's not going to make life suddenly go good. You're not going to suddenly have no medical issues or whatever, right? Like, all those things can still exist. And all you could, you could totally be screwed out of jobs. You could be betrayed by friends, no matter how good and happy you are, the worst things in the world can still happen to you in your life. Could, you know, on paper be bad. But still choosing the good and trying to find the good in things is going to be a better, more fulfilling life than it is

looking and lamenting the bad. Like, that's, that's it. There's nothing here that I'm trying to say is, like, deeply, you know, I'm not preaching the prosperity gospel here, right? There's no process. I don't, I, I, I know how statistics work. I know that the, the 2% richest are not that rich comparatively, right? Like, it is, it is shocking how the, the scale works in life.

And so, like, I can't buy the tariff. like Elon Musk could literally buy a $100 billion building. Like that's insane. I don't like I don't I can even think in these kind of proportions. I'm over here getting upset about Pokemon card packs. Yeah. My kids just want Pokemon trash. Yeah, who would do that?

That's weird. Yeah. DHH, I got a maybe, uh, I know, uh, we're ending tour or getting near towards the end of the time. I got a question that kind of goes with what we've been talking about here from someone that was in the chat, which was pretty much like, you know, as we're experiencing a bunch of this like AI change and we're trying agents and all this stuff. We're doing all these different things.

Like do you have any good like practical tips for people of how they can continue to level up like while they're doing that? Because I think that's one of the big things right now. This is one of the person, you know, the question was pretty much like, I feel like I'm just atrophying in the skills and that I'm just delegating them to claw. How do I like continue to get stuff done but also pursue competence and also continue like something in the right direction? I think that's a really big challenge for people right now.

now. Yes, and it's when I've felt myself. I mean, I talked about the competence dripping from my fingers moment when I over-relied on AI for a skill I actually wanted to learn. I think this is the first thing you got to decide. What do you want to learn? You're not going to be able to learn all of it. Like, I've used AI to write goddamn rust. Do you know how much I hate rust? You don't even like type script. You want to go typeless off. I am typeless. When I have to look at the code, I'm writing it without types. When the clanker is writing it and I'm not having to look at it, it can write Rust all day long. I wrote several C++ applications. I've never written C++ in my life. I know the shape of it more than I know the shape of Rust, but I'm not going to review it.

I went in, eyed wise open. Do you know what? This is an exercise where I am not going to look at the code. This is a black box. I'm asking the agent for an outcome and I'll be happy with it. when I get the outcome I like. Great. Then I have other areas. If the agent's writing Ruby for me, you damn well better believe that I'm reviewing every goddamn line of that code. And I keep chisling it, not by hand anymore. I'm trying to, like, I'm trying to teach the next generation. And the next generation happens to be an agent. So my teaching is agents MD. Like, don't do this. Don't do that. Do it this way. Formatted in this way. Until I end up with an agent that produces code that I like, and that doesn't always perfectly happen. But we're so much close.

are now than even what we were last December, right? Like the jump from Ocus 4-8 to Babel is wild. So try to subdivide. What do you care about? What do you want to learn and learn something? You cannot be a healthy, happy person who just leans back and goes, accept, except, by the way, you should never be accept, except, except, you should running and bypass all permissions all the time. That's the only way to live. So you're not even clicking upset here. But just this idea that like I don't have any competence anymore. If that's where you've ended up, like, yeah, bro, you're done. Because that job, that can be done by someone else than you.

If you don't need to do anything to sit just and click accept, yeah, a fucking Muppet can do that. You're not needed. So you've got to bring something to the table, which, by the way, it's also where meaning and happiness lies. There's no meaning. There's no happiness and just going like, tick, tick, right? That's actually a bird in Zimson, by the way. Oh, the Byrne Simpson's. You're talking about the Byrne Simpson that prevented the nuclear, yeah.

Yeah. Cause the nuclear meltdown. It doesn't it? Yeah, caused it. Anyway, you got to have something. You got to have something where you're leveling up. You've got to have something where you're getting better. Like, I'm spending so much time with agents on Linux subsystems and diagnosing weird error codes.

And I've learned more about Linux and the internals of how that thing works than I could ever have dreamed of doing just on my own because it's a guided experience. Because as good as AI is at doing stuff for you, it's arguably an even better tutor. It's an absolutely incredibly patient teacher who will, I mean, all the time just go like, I don't understand that. And it'll go like, explain it. And then sometimes like, I don't understand the explanation. You try again. Bro.

And it'll cut that simpler. Simpler. It'll just keep doing it. And you've got to find a ratio here where like every day you're a little bit better. If you end every day a little bit worse, bro, it's going to compound and you're done. You got to end every day a little bit better. Not a lot. I'm not telling you, like, you got to learn everything, right?

You got to end every day a little bit better. By the way, totally possible. The magic of agents in part, I think, once we get to this golden era of abundance, is that you're not running eight hours a day, just flat out. Like, I've been doing that now for a little bit too long. I mean, I don't know if you can tell. I'm a little delirious. And some of it is coming from the fact of just trying to blow my gasket for eight hours a day running a lot of agents, right?

Like, that's a lot of decision-making load. I want to get to a place where I'm working three hours a day. Like, I'm doing intensive work for three hours a day, and then I'm fucking chilling and learning the guitar and I don't know, reading the fucking great novels of time, right? I think we can get to that now. So did Keynes, right? He predicted in the 20s. He was so close. And we ended up with something quite different.

And I do understand, by the way, right now, we are in this insane transformation. And I know a ton of people who are working harder than they've ever worked in their life. That the acceleration you get by entering into this mex suit that can run faster than you've ever been able to run, that can jump higher than you've ever been able to run, never really jump, they can fire fucking missiles out his ass is intoxicating to the point that you

want to be in the mex suit all the time, all day long, right? I think this is perfectly natural during the transition phase as we're entering this new age, and it is a new age. It can't be the sort of standard. We keep forever. That's not going to be sustainable, and we're all going to pop our skulls if that's true. So I'm sitting like, all right, I'm sprinting on maximum velocity right now. By the end of the year, I don't fucking want to work like this. I want to chill. I want to have more time left over, right? Now, the other thing, I mean, I just did it, but don't try to predict the future.

Don't try to predict where this is going. I guarantee you you're going to get an AI psychosis. If you try to chess play where we're going to be 12 months from now or 24 months from now, it's not going to work out well for you. You've got to deal with reality as it is today. Make the most of it. Lean in with optimism. lean in with curiosity, lean in with goddamn strength and honor.

And you're going to come out all right. I often tell my wife when we're thinking like, you know, just thinking about how do we raise our kids and all this other stuff, right? And then one of the phrases we often use is we're going to do the best we can with the tools we've got, you know? So I'm not a perfect dad, but I'm going to do the best I can with the tools I got. You know what I mean? And I think that's like, that just reminds me of that.

I'm like, I don't know. I don't know what my son's going to be like when he's 10. I'm going to try and do my best right now. Right. Yes. Yeah. Well, I'll tell you this much, as somebody who has slightly older kids than you,

I've done a significantly better job raising my fourth kid than I did my first kid. You know, like RIP first kid, pour one out every single year of his life. I'm kind of in that situation. I'm just like, yo, I was much worse father for you than I am with Livy.

Like, I'm sorry, it's just like, that's just the world that we live in is that. It's complicated. I have to, I have to learn. Dude,

Dude, I learned from Bluey how to be a dad, to be honest. That dude's the best dad. Okay, well, that's a little scary. But I appreciate that. I appreciate that you learned on the job by watching TV with your kid. Like, that's a good move. That's pretty cool. It's working out good.

Like, all these things, this is like the most natural and normal human experiences that you have to learn. You know, we learn by doing. I mean, AI learns by doing. They just do it one million times to get slightly good. And we just have to do it a couple times. times you get pretty good. And so it's like, oh, just learning the right things. It's hard. It's hard. But also. But exciting. But exciting. You don't want it to be easy. You want the challenge. And some people have more challenges and they wish maybe they were playing at a slightly lower difficulty

level. Yeah, you don't get to pick. That's just the kind of world that it is. The level is set for you. And then you try your dambis to got them beat it. And if it's on ultra mega hard, well, all right. Man, you're going to get good at the time you beat it. And just, what else are you going to do? What else are you going to do? You can fucking lean back and give up, lean back and just succumb to despair? No. Then you're left with the only serious question in philosophy is Albert Kermu saying.

Like, are you going to kill yourself? Like, fuck, no. No, you're going to live. You're going to live. You're going to live. You're going to be good. You're going to get better. It's going to be all right.

Well, you do know what is the sign above hell? Burgatory? No. No, the sign above hell. Oh, no, no, I don't know, Fron. Abandon hope, all ye who enter. This is straight out of Dante, right?

Like, the one thing you don't get in hell is hope. Yes. Yes. You got to have hope. And if you don't have any, you pretend until it emerges. There's a great study on smiling that literally, when people are forced to smile, they're producing endorphins. Like, you can literally go like this.

I don't fucking have you. I'm just fucking smiling. Like, you know what? It manifests. Like, not all of it, right? Just like a little bit of extra endorphins and, I don't know, dopamine in your brain. You pretend until you make it. Yeah.

Take it to make it. What's it called? The guy who, the famous racer, I forget his name. There's an article written by the Mittlify. Yes, that guy. I think he won some sort of amateur Le Mans. I believe an announcer said, this guy's racing like a race car driver. But what's it called?

I love that clip. It's one of my favorites. But there's this racer, first off, there was this article talking about how you shouldn't smile, don't smile, right? We talked about this one. But the funny part is that he gave an example of long distance runners don't smile. And the most successful long distance runner of all time said, hey, you should smile. And a bunch of institutions came about and said, actually, yeah, we did testing. And if you smile, you actually just perform better in physical exercise, just by having some form of joy. You actually just do better, which is just hilarious.

I don't do well at running because I've never smiled while I'm running. I'm never going to be there. I hate it. I'm running around my neighborhood like an insane person. Honey, go inside. Go inside. I'm exercising. No, he's saying the neighbors are pulling their kids inside.

Like, it's the crazy guy running around the neighborhood smiling. Yeah. All right. Well, hey, Omachi, Quatro, feeling pretty good. I'll try it. I'll put on the old. My book. Yeah, I just installed and I noticed that the key, when you hold down, the key rate is faster than it was on the previous versions.

And that was confusing for me the first time. Oh, it was? Yeah. I think that's the actual change. Maybe those are default change there. Some defaults somewhere change. I was surprised by it. The whole thing should be faster.

I did a lot of runs with agents to optimize things. Finding all the sort of little polling systems we had in all the subsystems that pinged the CPU to wake it up all the time. I found a bunch of that stuff, got rid of it. Quatro should really buck the trend that things always get slower and more bloated and whatever. I mean, again, we shrunk the ISO tremendously. We majorly sped up the installation. Battery life should be better. Everything good is in there.

Go get it. I will say one thing I've been really enjoying with agents that I would super recommend to like anybody trying them out is there's a bunch of changes that are just too painful for people to make by hand. least for me, where it was like, hey, go find every place that we do this thing. We have to change it like a little bit differently in all the spots, but it's like kind of the same, but not exactly the same. So you couldn't just like, yep, let's work through the first example. Cool. Nope, you did it wrong.

This. Blah, okay, do the second one. Nope. That's good. No, blah, blah, blah. Okay, cool. Now go find every single other one in my project and fix every single one. Okay, cool.

Go. And I'll come back later. It feels amazing. You can actually, like, change the whole code base without having to tear all of your air out. So try that one. That's another practical one to try on chat. There you go.

Dude, if it wasn't for agents, I wouldn't be on Linux, to be honest. Having Linux with in agents, like, it just unblocks all the pain points and I can just get what I want

like immediately versus I would probably quit in like five minutes. That's why TJ and I have always historically made the joke that there's fathers against Arch Linux. Because the moment you have like kids in a full-time job, like back when I was streaming,

plus working on Netflix, plus having four kids using arts Linux was like hey bro you got to go read the docs for like four hours to get like even remotely the correct idea of the shape of the things to come it's just like man this is such a hard commitment for me to get you know fathers against arts Linux is you know I resonated with that joke because it was so hilarious but then as things got easier obviously then it got a lot you know hyperland made a lot of these things easier but you know as time went on reading the reading the manual in like sniper rifle fashion feels really good like hey I want to learn about this one thing.

Give me all the links. Oh, nice. Good job, TJ. That's my sign. We have to go, guys. Okay, bye, bye, everybody. Bye. Is it actually over?

Yeah, we're ending it now. All right. I'm out of the day. Fibco. Errors on my screen. Terminal coffee. And hell living.