DHH на Omacon: «pro‑computer», Omachi и почему год Linux на десктопе действительно наступает

I love by the way that we have not just one screen, we have a tiled screen set up with four tiles that will pop in one by one in honor of our Hyperland goodness. The first thing I want to focus on is why we are here, why I have decided to embrace Linux and the experience that I hope we can share with far more people. And that is a celebration of computers, which sounds perhaps a little odd, are we using computers all day long? Isn't computers used by everywhere in all circumstances? Yes, they are. And how many people are grateful for that incredible privilege? Not very many these days. And it seems like the rhetoric and the narrative from almost every angle is how computers are the problem and the root to all evil in the world. And I find that to be very curious because I have played with computers now for about 40 years. I got my first computer back in 80 and more than 40 years, 85 I think was when I got my first Armstrong. And as much as I got it for something specific to play video games, I also found very quickly that I simply fell in love with the computer as an object. It was not just about a purpose. It was not just about doing something with it, achieving something with it, having fun with it. No, it was also the computer itself. It was the keys, it was the aesthetics, it was simply being in front of that screen and having a direct interaction with this medium that allowed me to do seemingly everything. And that has been the spellbinding experience of the last 40 years, both as a career and as a hobbyist. And I find it curious that there's not more of an appreciation now. When I came up with the internet in the mid 90s, we had all these great tech magazines, we had Wired, we had Red Herring, we had all of these areas of the media celebrating computers, celebrating people doing amazing things with computers. And now it seems we have the opposite more of the time. We have people telling us all the bad things that can happen with computers. We got to get away from that. We need a full reboot and reset and a celebration of computers as objects. And not just because it's computers, but because humans create incredible objects that are far more than just about achieving some end. If you take the humble wristwatch, in my opinion, this is the greatest wristwatch of all time. This is the Rolex Daytona. It has about 300 tiny little parts in it that add up to an experience of telling the time that could be done with an also very nice $20 Casio. Computers are like that in many ways, too. And the way we use them are like that in many ways, too.

That the mission we're on when we're trying to use it to do something could be done in a million different ways. Often far simpler, often far less costly or less cumbersome, already prepackaged, but then it wouldn't be ours. Part of the fun of playing with computers in the way that we do, us in this room, is that we get to take the case back off. We get to tinker with 300 little balance wheels and springs and escapements and feeling like we're part of the clockwork, we're not just consumers of it. Another object that I love is the Leica Rangefinder camera. I've been an avid family photographer for 13 years as my oldest now. And most of my favorite shots I've taken with this camera. This is, in many ways, not a good camera, because it requires a lot of you. You have to line up this weird ghost image inside of that viewfinder to get a focus. You're manually moving a little dial underneath to get it all to line up, and that seems anachronistic. It seems odd. It seems out of fitting. You could just use your phone. Yeah, but it wouldn't be anywhere near as fun. And I'm doing this in part because I want to have fun with that engagement. I want to handle beautifully crafted objects that have been created by people with a deep love for photography, with a deep love for watchmaking. These are the parallels that we should be drawing on as computer users, as people who love the computers as objects. The final example I use here is the car. I love cars. I love beautiful cars. This, in my opinion, is one of the most beautiful cars made in the last 15 years. It's a fancy Toyota, also called a Lexus. And I like this car so much I bought it two times. So this is why it's up here. It is an incredible machine that has a purpose, just like any other cars. It can get you from A to B. But all the interaction points that you touch, all the little dials, have been crafted by someone who cared far more than I knew anyone could care about a volume dial. And that level of appreciation for creating beautiful mechanical, electronical, or digital products, I want to surround myself with that. I want to be in that mold where as much of what I do and I interest myself in is created by the kind of people who would care about this small stuff. And when it comes to computers, we should aspire to exactly the same thing, sweating every pixel, sweating every interaction, and making it something that's truly at the light for its own sake. Not just because it's more productive, not just because it can achieve some goal faster, but because it's simply a joy to use. We need to surround ourselves with beautiful things that make us aspire for more. Now, I've shown you three luxury objects. The great thing about computers is that luxury in a digital form is available to nerds like this, nerds like us who are not going to spend tens of thousands of dollars on a watch, but will happily spend maybe three times as much pouring in 200 hours at what could have been a billable rate of $200 an hour, AKA $40,000 rising our Linux setup and think nothing of it. That level of luxury engagement is really endearing. And the second thing that's endearing here is that this is a shared endeavor. We're not making just these digital watches and aspirational computers for ourselves in isolation. No, we're making it together. And I also don't think we have quite enough of this. This is one of the reasons why we're here in this room right now, because some of the fondest memory I have of my entire existence was land parties in the mid to late 90s. Having to get together to simply enjoy both computer games, but also computer demos. Also setting up BPS is exploring the early internet. That is a special kind of love. I've been reading this beautiful book by CS Lewis called The Four Loves, and I'll read just one small segment.

I'm if you become a man's friend without knowing or caring whether he is married or single or how he earns his living. What of all these unconcerned things, matters of fact, to do with the real question, do you see the same truth? That is the foundation of friendship. That is the foundation of Omicron, and that is why we're here today. This specific incarnation of Omicron came together as a curious throwaway tweet when I was excited back in July of last year, just a few weeks after Omicron had been released for the first time. And I thought, wouldn't it be fun just to get together? And as things often happen on the internet, it doesn't take much more than one reply to suddenly turn a fleety idea into okay, we're doing this. It's happening and then just some random stranger, Sammy, saying "that sounds a little silly. Shouldn't it just be Omicron?" and just going, "Yeah, it should." That small interaction of like just airing an idea, it would be fun to have a bunch of nerds together to talk about this stuff, to turning it into a conference with at least the first speaker confirmed, and then someone revising what the branding should be is incredible.

But that was not the first part of my Linux pilgrimage. That started a little while ago, about two and a half years I guess at this point, when I first had enough of Apple and their shenanigans to the point that I was willing to throw out every beautiful piece of MacBook that I had in my office and try something else. And the first version of that turned out to be Omacube, which was kind of a prototype for what Umachi eventually came to be. It is, by the way, still around. I'd like to do a version two of it. I think it's very lovely, but it served a greater purpose in my mind now as the kickoff point, as the starting point of getting into this stuff. Because I don't stand here as a 20-year expert on Linux, on Rising, on NeoVim, on any of the wonderful tools that now consume such a huge part of what I enjoy about computers. I literally discovered all of those things two years ago, 35 years into a career of computers and playing with computers and loving computers. Only two years ago did I discover basically what I now spend an enormous amount of my time on. I think that is unique and very special. Now, this whole thing also came to be as a fun side quest while doing something else. So, Omachi came to be in the downtime of last year's 24 hours of Le Mans, when you spend two weeks in France getting ready for this amazing race and then you have all these days where nothing happens. Well, I had my computer and I had spent a year on Amaku and I thought, you know what, there's room for something more. And I spent a lot of my time on a website that if you showed it to mere strangers, they'd go like, whoa, whoa, whoa, calm down here. This is an HR violation. This subreddit seems somehow illicit. But Unix porn on Reddit is an incredible place of creative demonstration. One of the things I really loved about computers in the late 80s, early 90s was the demo scene. The Amiga and the Commodore 64 demo scene. This engagement with the computer to create art, to create beautiful things, vivid colors, just all for the sake of it. And I found that on this subreddit on Unix porn. And a lot of what's going on here, well, all of what's going on here is Linux rising. People building bespoke, beautiful systems that look nothing like the computing experience that most people use when they use a major vendor. And I thought, this is fascinating. Now, I didn't actually want all of this exactly, but I wanted some of this in some form. And the way it became more concrete was actually with YouTube. YouTube is this other incredible gift, I would say, to the geeks of today. We can find experts who have explored terrains on all sorts of esoteric topics. And Chris Power, who runs the Typecraft channel, did that for me with Hyperland and the plethora of auxiliary projects that you need to actually create a full system. And I was spending my time at Le Mans watching YouTube videos up in my little French hotel room and getting expired. I thought, oh, I should try that and I should try that thing. And I had this long list of all these projects I wanted to try. And then I was also watching this other dude on YouTube, Prime, who I'd been following for a while. And he had this one video from, I think, four years ago detailing how his setup worked. And what I really took away from this was, oh, my God, I am not cool enough to use a split keyboard like that yet. That would be amazing to graduate to that at one point. But in the meantime, I can pick up some pointers here. And one of the pointers I picked up that really became a cornerstone of how Omachi worked was the idea that every program is accessible on a single key command. That you jump directly to everything. That it's not even the graduation to something like Raycast or other launchers where you hit a . Then you start typing and then you jump to something, which is a big step up from finding something in the dock and clicking on it. But still nothing like having a single command that, oh, super return, just boom, pops off a terminal.

Super shift return, boom. Now you've got a browser. And that experience and seeing it on YouTube was really inspiring. So those two influences was absolutely key to forming this, to forming Omachi. And which is why it's so very lovely that we're going to hear from both Chris Power and from Prime later today. Which, by the way, just quick aside, I love so much that we still use handles. This is one of the things I remember from the old Amiga demo scene day. Mine was Webster. I don't know why it was Webster, but that was my handle back in '94. We got Prime. We got Vaxery, whose real name I actually literally only learned, like, I think two weeks ago when you had to put it into something and reveal your hidden identity. That kind of fun, that kind of levity, that kind of blending of an art world of halfway Banksy mysteriousness, I think is just really fun. That we can do that and at the same time produce these products that then take off. Because what was so interesting with Omachi for me was I felt like I'd already done a version of this with Omacube. And there were people who were interested in that and they started using it, but it was modest. Omachi was not modest. Very quickly after this launched, tons of people started using it right away. Quite to my surprise, because I was thinking, do you know what? This is a very personal system, actually. This is my setup. It has all these programs pre-installed that perhaps is not something that everyone is going to want to use. And in that encapsulation, that understanding, I think is a realisation.

This is what we respond to, at least in this space. We do not respond to generally products that start from the beginning trying to be everything for everyone. Trying to appeal to every computer user out there. No, we respond to something far more targeted. And when we do, we can end up with something that's a much better fit. It's not a one size fit all glove that kind of sits shabbily on your hand. No. It's a number nine. I use a number nine. You use a number nine. Wonderful. It just fits so much better. Today, we're doing about 50,000 ISO downloads a month. Over the period that we've even had an ISO, which was not something that was there from the beginning, I think we've done almost around half a million downloads. Which is actually kind of a lot and far, far more than what Omacube achieved. Now, all of this, by the way, is served up deliciously by Cloudflare. I think we've done about almost five petabytes worth of ISOs, which is actually a sizable chunk. And this interaction with Cloudflare was a good illustration, too, of what I thought I could bring to the Linux world. Just like a little bit of modernity. Like, mirrors? Are you fucking kidding me? Like, we're still finding people who can fit a little server in their university's basement, and that's how we're going to distribute all the software. Why? We've had these content distribution networks for about 25 years. Could we adopt them?

Could we use them? Yes, we could. And then, of course, also, Discord, which actually was not something I had used at all before starting with Omachi. And very quickly thereafter, we had about 30,000 users on there. But even more than that, what I loved seeing, both from the early days of Omachi and now, is all the creative setups, all the beautiful setups people are putting this operating system into. There is an enormous amount of care by an enormous amount of people for their computing environment, for the computing object, that it is more than just a terminal. It is more than just a tool to get something done. No, it is an environment in and of itself. We should be celebrating these kind of personal cathedrals. We should be building more of them. We should be encouraging them. We should be egging each other on to do more of this. Surround ourselves with more beautiful things. Now, what I really appreciate, too, about what's going on with Omachi is that it is not just speaking to the already converted choir. This is a list of operating systems accessing omachi.org. And you can see it's actually a quite neat, even three-way split between Windows, existing Linux users, and the Mac. And I think that is one of the key elements of this new influx, this new vibe, is that it is not just the same scene. We are introducing new blood, new perspectives, new expectations about how things are supposed to work. One of the things people often celebrate about Arch is the Arch Wiki. It is an amazing resource. There are all these little recommendations and tips and tricks on how you can get esoteric software or hardware working with Arch. And wonderful, isn't that great? But couldn't we just solve all this shit so no one had to read the fucking manual? What if you just bought a new computer, slapped on an ISO, and everything worked? Even the . Even the Ethernet adapter. Just that you did not have to search through some fucking wiki to figure out the basics of your machine. We can do this. We can solve this if we choose that that is something that we want to care about. If we choose that the right of passage to starting to use Linux does not need to go through this needless drudgery of just getting the core hardware up and running. Now, that's great. And that's our little bubble as it's growing. But what's neat about this time now too is that it's growing outside of that too. Linux is on an incredible run right now. A run it has not seen on the client side in all the time that I've been watching. The joke literally since '97 or '98 whenever it was coined that the year of Linux of the desktop was going to happen actually might happen. After 25 years of prophecy, we are nearing an absolution here. We are nearing the fact that this is happening. When I saw that combined with my own kind of foray here I went like you know what part of the problem here is that too many organizations are kind of pussyfooting it. They are like yeah that's neat we will try a little bit. So I went when I come and say like fuck it we are burning the boats. Let's go. We are all just moving into this. Yeah it's not going to work quite right.

And you are going to have to figure some stuff out. And there's going to be some gripes. I was just listening to gripes last week about someone who couldn't get on a hotel Wi-Fi network with Amachi because we haven't solved that problem fully. And I was like great. Here's a new mission for you. You should solve this and reap the glory of curing this problem once and for all for everyone. And Linux is full of these quests. It's incredible. It's like one of those RGB's you show up and there's like there's a little gnome over there and just like oh if you go off here to the dragon you can get all this if you just beat the old witch with the stick. And you go like oh that sounds like a good quest. And then there's a fucking ogre over there like oh we should go over there and we should beat them with the big whatever ships. You're like oh that's a good quest. There are endless quests. Every day there's a new quest and you are invited to take any which one of them and you can absolutely solve them and hop on board. Now that led to a really curious experience in the early days of Amachi where I saw a lot of this over there. A lot of love. A lot of people being really excited.

And I was like ah you know what it's not right. This is way too positive. This is not the internet that I know. There is not balance in the universe yet. And then lo and behold three months later some of this showed up. And I was like ah now I know it's real. Now I know we have something. When the haters finally arrive and start shitting on your project you can bask in the glory that you have made a small dent. And the first thing folks got really excited about was the definition of the word distro. What is a distro? What's just a script? And actually is bash kind of bad? These were these amazing discussions we got to have. And I just basked in it. In the glory that we had provoked this existential distro. And they also had a point. Let's be fair. The first version of Amachi was a arch post installed script. Now by the way so was the first version of Ubuntu as I'm told. So was the first version of Fedora. These also build on pre-existing packages and pre-installed afterwards. This is how most things bootstrap. This is how almost everything starts. It starts as this rickety, barely functioning promise of something better. And so did Amachi. So we quickly had to learn a lot of things. One thing was like how the hell do you build an ISO? What is actually needed to go into this thing to get something to boot and install a new operating system in a smooth, easy manner? I didn't know that. Amachi was well down the line and I was using it. I had no clue how ISOs were built. Well, to figure it out. And Ryan and I did. And then suddenly there was an ISO and then like, oh shit. Now one of the main attack points, this is not a distro. It doesn't have an ISO. Boom. It has an ISO. So what now, bro? Oh, well, I mean, you don't have your own package mirror. Well, we had to figure out how to do that too. In part because, as I'll get into with the packages, the AUR, the Arch user repository, which is an incredible public resource, right when Amachi was taking off, was getting attacked by Chinese hackers for some reason. Still haven't figured out quite why. But it was down for quite a long time and therefore you could not install Amachi because we were depending on AUR packages. And everything was just break. Shit. Now we've got to figure out how to get a proper mirror. Also because we wanted things to be fast. And Arch is still stuck in little computer under the university desk mirror setup. And we wanted to use something faster. So, we got a mirror. Then finally we started building packages too. So that we would not have to rely on the AUR. The AUR.

Incredible. And suddenly it was a distro. It's delightful what can happen when you just don't stop. When you start and you just keep going. And there's like, well, it doesn't do that. Okay. Well, let me go away for two minutes. Now it fucking does. And it doesn't do this. Well, okay. We can fix that too. And if you do that enough times, eventually you fix everything. But why now? Why is there this surge right now around Linux? Not just about Amachi, but as I've shown in other parts of it. Well, one reason is that the existing offerings in the operating space have gotten shittier. Amazingly so, given all the time, money and millions of man-hours poured into it. Microsoft and Windows is seemingly worse than it's ever been. So, that's nice if you're trying to do something else. Like there's a comparative advantage that just keeps improving just by the fact that your alternatives just go sliding down the slop rain into something shitty where your fucking start menu is made in React. And that's why it takes two and a half seconds to open. What the fuck? Great. Thank you. Is there a mole? Is there a Linux mole inside of Microsoft getting them up to all this stuff? Yeah, yeah. You should build your startup menu in React. That'd be great. And then everyone else goes like, no, you really shouldn't. You should just make it fast and not crash and not be full of fucking ads. All right. So, this slide, this Microsoft slob slide, has inspired a bunch of people who have been using Microsoft and Windows for a long time to give Linux another try. This is Linus, tech review, who four years ago tried to switch to Linux and it wasn't ready. It wasn't good enough. So, he didn't stick with it. Well, slob has just inspired him to try again. And this time, it's gone better. We fixed more of the things on the list of things that were broken. In a short period of time, Linux has become a credible alternative for gaming. Tens of thousands of games are now running awesomely on Linux. It's gotten so bad that Microsoft is apologizing for how bad it is and promising to do better. Now, I don't give them high odds of actually putting this to fruition, but if it did, wonderful too. This is the wonders of competition. If Linux is suddenly comparably better and it forces Microsoft to become better, everyone gets to use better computers. Yay! But here's the one I personally care a little more about. Mac OS Tahoe. Because I've been a Mac user or was a Mac user for damn near 20 years. What the fuck is this?

Is someone dancing on Steve Jobs' grave and pissing on his tomb just for the sheer fun of it? How did the company that cared about the Radeye to the nth degree and celebrated it, ended up with fucking this? It looks worse than the inconsistencies between them. Terrible. It's gotten so bad that even the biggest superfans of Apple, like John Grouper over at Daring Fireball, is saying it would make more sense if we found out that the team behind redesigning the UI for Mac OS Tahoe was hired by Meta a year ago. They sabotaged their work. The Mac looked clownish and amateur. This is the biggest superfan for Apple. Oh, and here's another good one from John Terry. Tahoe is the worst user interface update in the history of the Mac. Every change is either wrongheaded, poorly executed, or both. The bad ideas embodied in Tahoe reveal an Apple design team. Of human-computer interaction. Ouch. These two guys were part of the Six Colors report card that's been going out every year. This is the rating for Apple's OS quality at D+. Now, that sounds bad, but wait until you hear how Apple's relationship with developers is going. That's a D-minus. And it's not like a D-minus where, well, I used to be failing and now I'm on the way up. No, it's a long slide down the same slob route that Microsoft was on. Now, the great thing about this, as I said with Microsoft, is if you're trying to do something else, we can hardly believe our luck. There are now so many people who are willing to give alternatives a try because the two main titans of the industry are fumbling so hard that even their biggest fans are giving them a D or D-minus on the report card. Now, this is one of the techniques I really tried to study from one of the absolute saints of letting your opponents just fuck their own stuff up.

It's Saint Gabe of Valve who have managed to bring Steam into the leadership position largely by just not screwing over players. A really good strategy. Now, meanwhile, all this is happening that Microsoft and Apple are doing their best to create opportunity for Linux. Other people are actually helping in a positive direction. This is Li Bhutan from Intel celebrating the fact that the Panther Lake chips made in Arizona are fucking awesome. I will co-sign that. There's finally something happening on the other side of the chip equation. Apple has been able to walk away with every accolade for the last five years since the introduction of the M chips for very good reason. They were incredible chips and they were so far ahead of what everyone else was doing. It was frankly embarrassing and embarrassing for Intel most of all that they managed to squander this. But in 2021, Pat Gelsinger, CEO at the time, set off on a daring mission called 18A to bring Intel back in the game. And five years later, here's the wafer that proves it. One of the reasons a lot of people disliked PCs for a very long time was the fact that the battery life was trash. Absolute trash.

Now, it got better in the last few years. There was Luna Lake last year and AMD had something that was sort of passable, but Apple was far ahead. Isn't competition amazing? Isn't it just great that this is a test of the XPS 14 absolutely trouncing Apple on one of these benchmarks? Even if it wasn't trouncing it, just cashing up would have been a huge achievement and key to what we needed. It's not just about battery life, though. Panther Lake's new chips are absolutely competitive. They are about on par with an M3 in single-core performance and about on par with an M5 in multi-core performance. We did not used to have this. We didn't have this five minutes ago, a chip that was both battery efficient and really damn fast. Now we do. As I mentioned with St. Gabe, he has tirelessly been improving the situation of gaming on Linux to an enormous degree. And this is absolutely key to why there is this new surgence. Much of the improved statistics for Linux market share come from this. But there's also this guy. I don't know if you've noticed, but AI has sort of been occupying some folks over the last few years. And I know that things haven't always been great between NVIDIA and Linux. But do you know what operating system the new DDX Spark runs? Oh shit, it runs Linux. So even though Linus gave him the finger, he still didn't hold that much of a grudge that he's not leaning into this. All the AI workloads in the world are running on Linux. And NVIDIA doesn't hold a grudge at all. Even their consumer facing services like GeForce Now are being rolled out for Linux. Making the small number of games that you can't run natively on Steam available through cloud gaming. And this is before we've even talked about what's coming from Snapdragon. Who managed to poach one of the key teams from Apple who apparently had all the secrets to how you built M chips. And took it over there and is now delivering some amazing performance with the Snapdragon stuff. Closing the gap on the final part of the equation where Intel and x86 isn't quite there yet. Single core performance. Snapdragon x2 elite chips that also run ARM are on par with the cores that Apple are putting out. Incredible.

So that's what's happening on the hardware side. On the software side, something very interesting has been happening too over the last 10 to 15 years. The total near death of key important desktop applications that aren't on the web or in the terminal. Basically, every piece of important new software created in the last 15 years ended up as a web app. Now, there's a lot of people who didn't like that for various reasons. Linux folks should be thrilled. The fact that the greatest application platform in the world works flawlessly on Linux and always has is a huge part of the reason why Linux is now a viable alternative to these other giants. And then, of course, the rise of the terminal. Brought on by the rise of AI. What a curious set of circumstances that it just so happened to be that last year with the introduction of Cloud Code and all the other CLIs for AI, they were all terminal driven. Tailwind, tailwind, tailwind. All this adds up to the fact that we can now create a new kind of computer that's competitive in even a new way by a bunch of geeks who look like this. Us. Us. Together.

Create a new kind of computer that is not trapped in this old fight between Stallman on the one side being a radical, single-mindedly dogmatic individual who has done wonderful things for open source but holy shit I do not want him to be our emperor B. Or, on the other hand, this strange looking enterprise sales guy who represents another part of software that I don't care for much either. Not even this guy who I otherwise would put in the pantheon of heroes of the beautifully designed computer represent exactly what we're trying to build. Now, part of that is that all those guys had a single-mindedly obsession with owning everything. About a decade ago I wrote a piece called "Reconsider" that I'll just read from here. "Part of the problem seems to be that nobody these days is content to merely put their dent in the universe." No, they have to fucking own the universe. To be in the market, they have to dominate it. It's not enough to serve customers, they have to capture them. Well, we don't have to do any of those things. If we just want to build beautiful computers mainly for ourselves and others who love them, we can put that small dent in the universe and leave these other three guys to wrestle for world supremacy. Now, the kind of computer I want to build is a pro-computer and I'll explain what that means. It's a computer made for and by people who love them. It's going to be pragmatic, ergonomic, aesthetic, and sovereign. Like peas. The pragmatism comes in large part by open licensing. This is open source, not the kind of open source that Stallman wants us to put out where it infects everything and all of it has to be that way. But a base that's that while inviting great commercial services.

There's nothing evil about commerce. There's nothing evil about capitalism in the way that it provides us with free options to spend our money. Great. I even put my own fucking commercial apps straight into the operating system because I use them every day. And also, do you know what? That seems fair. I built the fucking thing. Dell just showed up. Well, it didn't just show up. I've had a relationship with Dell for 15 years buying their computers on the server side. But Mr. Dell, Michael, showed up not too long ago and said, like, hey. Well, he didn't say. I reached out to Michael and said, please, can I have your awesome new Panther Lake machine? It looks great. And Michael replied, sure. Let's do something. And now not only have we done something, we have built an incredible feat of releasing the first distro that is fully compatible with Panther Lake with the XPS series. And that's amazing. But we're also just embracing all computers, even Apple, who I do not generally care for in their current incarnation as a company. I love their vintage. We can bring all these computers into this world. Second, these computers have to be ergonomic. And a key part of the ergonomics is they have to be keyboard friendly. This is probably one of the biggest insights I had after watching Prime demonstrate his setup, is that everything should be reachable instantly from the keyboard in a way where I'm just dancing through it all. Now that requires a trade off where usability sits higher than learnability. It's going to take a while. This is pro. This is fine.

Spend an afternoon getting familiar with it. Every other computer is chasing learnability first and foremost and will put it before usability at any point of the day. So for me, there's a direct correlation between beauty and truth. There's a direct correlation between the environment that I am in and the comfort and joy I take out of it. And do you know what? When it came to computers, we used to know this. This is a fucking gorgeous machine. This is a Danish computer designed in 1981 to just do, I don't know, some math stuff. It has an amber phosphorus display that looks absolutely gorgeous. How can we buy these computers today? Why don't computers look like this now? Even the Commodore 64, the first computer that I use, look at those browns, man. Look at those browns. Look at the goddamn interface you had when you started it up. The color coordination. This wasn't just black on white DOS shit. This was someone who knew something about colors. My beloved Armstrong. Look at that fucking keyboard. A red escape key, a blue enter key, green keys, integrated cassette player. Everything a boy could ever want. These are some VHS tapes from the 80s.

How did we lose the ability to do this? We know that this was possible, that everyday ordinary things could look this fucking cool. Holy shit. This is a scene out of Aliens. One of my favorite movies, Alien. Aliens is also my favorite movies, but Alien. Computers in the movies look amazing. Why don't ours? Why can't they look like this? Why can't they? We don't have sound on this thing? Maybe we don't have sound. I want this running through my veins when I'm using my computer. I want this mix of the 70s and the future retro going together. This is Space Bolt's demo for the Party 2 in ORS called the State of Art. This was done in 880 kilobytes. Fucking incredible. We used to do this. We have not lost neither the power, the wisdom, or the skill. The future should look like the future. That was the operating principle for the design of the Cybertruck. Our computers should look like the fucking movies. They should look like they belong on a set. And they should be gorgeous. Absolutely gorgeous.

Absolutely gorgeous. These were three themes by the three contributors. We have Taha, Biana, Oljebovo. Incredible. Finally, sovereignty. The last S&Ps here. Now, sovereignty is really important because it gives you the fortitude to tell people to fuck off. And to say, "I'm not doing that." And that's really important. I am a big believer and supporter and frequent users of the words "fuck off." Mostly just in my head, but sometimes also on Twitter. But even more important than that kind of sovereignty is the kind of sovereignty Cathy Sierra talks about. That we should be upgrading our users, not just the product. That value is less about the stuff and more about the stuff it enables to build better cameras, build better photographers. That applies directly to the Linux experience as it was and as it is in a lot of communities. That you have a scenario where nothing fucking works. And then almost an impossible wall to climb before you get your bespoke masterpiece. Do you know what? It's very commendable. The kind of people who were able to set up Hyperland from scratch with the entire ecosystem of everything and spend 150 hours doing that. Jesus, you have my respect. That is hard work for a very small niche of individuals.

And there's a much, much broader set of people who would have fun getting to the same place on less of a wall and more of a ramp. And a big part of that is omakase. That you got to start with something that is approachable, that is delicious, that is appetizing off the gate. When you don't know how all the things are supposed to fit together. The other thing is, when you want to change things, it's got to be approachable. Umachi is built in bash. It's basically all bash. Bash has gotten bashed a lot over the years by programmers who think it's a shitty programming language. I think it's amazing. It is one of my all time favorite programming languages. Exactly because it's so fucking shitty at doing something as basic as designing an array. Because then you just don't use arrays. And there's a lot of software that doesn't need arrays. There's a lot of software that just needs if, else, installed this piece of software. Bash is great. Do you know what else is great? AI. AI is incredible with Linux for getting things off the ground and changed. Update my way bar to include a calendar when hovering over the clock. That's a prompt that I saw someone did.

And then their OS just changed and it did that. What? That's not how this works in Mac OS land. That's not how this works in Windows land. You can't just fucking change your operating system with a single prompt. On Linux you can. Finally, you should be able to make it look amazing. Barely known quality coordination. Just going like, hey, that's a really cool backdrop. If I get Bjarne's ether here to do some stuff, my whole computer looks different. All of it? Huh. I'm game for that. Now, you're going to break it. You're going to break your computer when you do it like this. That's great. This is Kintsugi. When this cup breaks, you put it together and it looks better. It's got fucking gold now. And the gold is your sweat, blood, tears and learnings that go into it. That's the kind of computers we should have. That's the kind of piece we should put in the pot. And that's what Omacom is all about. Thank you. Thank you. Thank you. Thank you.