this looks too good. This looks suspiciously good. Like, where the fuck are all the semicolons? Where are all the line noise that I'm used to in basically every other programming language ever? And I just didn't like this is just really curious. And it woken in me just that desire for peak aesthetics because I'd been working with PHP, I'd been working with some Perl, I'd done a bit of Python, I'd done a fair bit of Java, I'd done a lot of other programming language, all fine in their own regard. But when I saw Ruby, I saw something different. And I thought that was really interesting. And then the other thing I would credit Ruby with is after I dove in, Ruby really taught me metaprogramming. And that's one of those defining moments when you realize, oh, a program can program itself. That is really next level. Like it takes your understanding of what programming is to a higher point. And I think whoever teaches you that first, that next level, that next tier, you're going to have some degree of a bond with. And I certainly developed the bond with Ruby. And then I also just thought, do you know what? Why not? Like if, if I like how this looks, I like what it's teaching me. I like how it feels to be a programmer writing this programming language. Why would I not pick it? Why would I even care whether it's popular or not popular, whether other people like it or they don't like it? At least at the time, and I'd say probably still, I didn't really mind the fact that there wasn't a lot of tooling, that there wasn't a lot of frameworks, there weren't a lot of libraries, because you know what, I could just write it myself. Now, that's not entirely true. I mean, Ruby obviously did have a fair number of libraries and frameworks. They didn't have as many as it does now, and it didn't have as many as other languages did, but it had enough. I think there is some basis where a new program language comes out, then there's literally nothing. There's just the primitives. That's a tough hill to climb by, I think. Ruby wasn't like that. There was just enough that I felt like I could bootstrap what I wanted to do off that and get going really fast. I mean, I didn't go away in a cave, spending five years meditating on Ruby to come up with Rails. It was basically like two weeks and I had some HTML talking to a database and just the early inklings of what Rails was going to be came about pretty quick. And then just a couple months into it, the shape of Rails as it exists today was visible. So that's also part of it. That was part of the appeal for me about Ruby, that Ruby wasn't just any other programming language. It was a programming language that kind of super turbocharged everything I was doing. And I was able to create results much quicker than I had with any other programming language. And some of that, I think, is inherited Ruby. I think it is a highly productive programming language. But part of it was also just my reaction to the programming language. When you find something that just really fits and you want to use it all the time. In photography, they have this saying, well, one saying, the best camera is the one you have with you. and the other thing is the best camera is the one that makes you want to go out and take photos. Ruby made me want to go out and do programming or go in and do programming, yes it is, and I think that part of it was a huge push for my career that I just went from, oh, I've been doing programming mostly because I wanted the programs like I wanted the output. Now I found a place I found a programming language where I just want to do the programming, where the programming itself is the reward. Now I still pursue programs. It's not like I'm just writing katas or whatever for the sheer joy of it, but there is a sheer joy in the act of programming with Ruby. I hadn't felt that. I hadn't seen that in any other programming language. And in the early days, I thought this is totally unique to Ruby. This is a property of Ruby. This is how Ruby makes programmers feel like they are superheroes in this way, and therefore everyone should use Ruby. And I spent quite a few years earnestly trying to persuade as many people as I possibly could that they should all use Ruby all the time for everything, certainly within web development. And it took me a little longer to figure out, do you know what, that feeling I have to Ruby, other people have to other languages. And that is the lion's share of the superpower. Now, again, I actually think there's a there, there too, that Ruby itself is intrinsically more productive and allows us to do things and certainly aesthetically more pleasing. But the larger chunk of it is helping programmers find that experience I found with Ruby. And you know what, if they find that in PHP, or if they find it in Python and that they find it in Elixir, whatever, wonderful, I'm so happy for you. That is really the main progression point. And it took a while for me to discover that, but even having had discovered that, I kept looking over the past 25 years I've been programming, 20 some years since Ruby, could I find another Ruby? Because I always thought like, oh, I if I was doing say PHP, and then I find Ruby, and I go, holy shit, there's so much better. I'm so much more productive. I'm so much happier doing it. Could I find another thing that would have that same feeling beyond Ruby? And I just never did. And I came to do Realization in part, like, the world isn't that large. We think of like, oh, my God, there's an unlimited number of programming languages. No, there's not. I mean, how many have there ever been made 1000 maybe? How many have ever been used really in anger? 50? How many are currently viable options that most people would consider today for web programming 1020 like it's not that many, you can actually survey the world, and you'll find that like, it's a sphere and you can get to the other side and you've been all the way around and then you can look at all those options and go which one is the one that creates that connection for me and for me it was just it not just was but is ruby that that after having sampled pretty much all of it and I just I like sampling right I like going to just like I like to travel I like to go to Japan and like oh it's just a foreign culture they're totally different everything is different that's what's about it. And then also, I'm glad I'm home. And that's what Ruby feels to me. I'll go explore some other environments. Sometimes I have to do it. We can talk about JavaScript perhaps later. I need it. I need it to do some things that Ruby won't let me do. And then I go like, oh, this is great. This is a wonderful little vacation here. And then I come back to Ruby and I go like holy shit. This is good to be home What was the question first

off you answered everything you needed to answer that was? If there is any piece of advice any programmer ever needs to hear it's the programming. It's the side project It's the experiment you're running that's gonna make you want to do it again tomorrow It's probably like single-handedly the thing that's gonna make so many people's lives change and I just man

from the bottom of my heart I'll love that that is like that is my mantra in life And you know what's funny this used to be a big theme. I remember in the mid-2000s Both Ruby was advancing this because that's intrinsic to the language. Literally, Matt's motto for Ruby is optimized for programmer happiness. So that makes it pretty explicit. But plenty of other environments were following this at the time. And then I think we fell into a bit of a bad way in the 2010s, when suddenly it almost became a little dirty to say that you liked programming or that programmers should be passionate. In fact, that was somehow almost discriminatory to people who hate their jobs. And I was like, wait, that seems like not a right angle here. I totally understand that you can be gainfully employed doing programming as a job and not want to do it like 24 seven outside of that. Fine. Okay. But also, like, it's easier. If you like it, it's easier if it's just enjoyable. It's easier if you want to do it most of the time. And I will say a few programs I've met have not at least had a phase where that's been true. If they were to arrive at some point of notoriety, that they contributed something to the community, that they built something. Not a lot of programmers, I would say, who are just like on the nine to five thing, ended up doing that. Now, that's also fine. Like, whatever, we have millions of programmers, do they all need to create things all the time? Maybe not. But again, I don't think that reality is actually something to celebrate. And it's a little bit like that fucking dog, that meme with the dog, I don't know what the fuck I'm doing, that I just absolutely hate, unload even, this idea that knowing things, getting better at things, being competent is somehow that doesn't exist. And I remember that discourse was a lot about programmers who I knew knew things fainting that they didn't. And that was the kind of bullshit that I just went like, oh, comp the fuck on. I don't know everything. I'm totally happy to admit ignorance on a wide variety of topics. But also I do know some things about some of the things that I do. That's why I do him. That's why I get paid to do him. That's why there's impact from it and so forth. So let's not fall into this trap that no one fucking knows anything. The competence doesn't exist. And actually the highest calling is just to have a fucking nine to five that you don't really like because reasons.

Yeah, I think one of the like, for me, it's such a, it's not only a bad meme, because it sort of throws away, like all the things that you've learned. And like, clearly, like, if you were going to make site a website and Ruby on Rails, and I was going to write one in like, see, like, I think you would be done first. And there's something like, about that trade off. But I also think it really like misleads juniors and people entering the industry because their end goal now is like, Oh, my end goal is I don't have to know anything, never have to struggle to learn anything, never have to overcome and accomplish anything to get to mastery. Look at all these people. I respect who shipped valid things, who have done awesome stuff, and they're just saying,

well, I don't know anything.

You're like, no, you do. You're lying to them. You're tricking them. You're putting them on the wrong path. That's

100%. And to me, it's almost like a kind of luxury belief. By the time you reach some competence that is marketable, and you can actually use it to get a job and so on, you feigning that like it doesn't really matter. Everyone looks everything up on the Internet, whatever, just pushing it so far out that your demeaning competence as a valuable virtue to pursue. It's a luxury good. Like you're not paying the penalty for that. Right. People who embrace it and think that they can just ask AI for everything or they can just look everything up, but they don't have to understand the copy past that they slam into the project. They're the ones who are going to suffer because they're the ones who are going to have their career prospects curbed. And that's no. And that's what gets me is at the beginning of that whole trajectory of that narrative, I was actually on board. I was on board in as far as to say, you know what, we should make it a lot easier for people to get started. There's no fucking reason. It has to be that difficult that you have to spend literally days before you get to Hello World. the whole premise of early rails. I had been working in a Java shop. I saw what happened when we started a new project that someone had to go off for a fucking week to do XML sit ups so that we could have whatever the spring configuration that we needed for our web app server something something. And I just went like this is stupid. This does not need to be like this. This is actually a sign of lack of sophistication. We can solve this problem. We can solve the initial trajectory or angle of the curve and we can soften it a lot such that people who are not already trained wizards in XML systems can come in and get the hello world dopamine hit. To me, there's so much that is just bound up in that. Can you do a thing and then see a thing? If you can get into that loop, you're well on the way to become a programmer. But far too many environments, certainly at the time, and I argue still so, we've recreated some of that madness, especially in the JavaScript world, where it had to be difficult, right? And we mistook that that like the early difficulties, actually somehow that's where the competence in the mastery is no, fucking no. That's just the boilerplate shit that we're setting up. But then patch diverge, right? I was all about let's making the RAM soft, so that we can keep going. And I want you to have a 10 year career, constantly getting better, constantly learning more things, constantly getting more competent. Just that by the end of it, you're going to look like a fucking God to your former self. Like that's the value of it. Like we can make the computer do all these things and we can make him do it because we understand things, right? The ramp needs to be soft,

but it needs to go to infinity. Yeah. And I think something that, uh, and I think we'll probably come back to this when we talk about NeoVim, I know prime wants to hit a few things before we go like deep on the NeoVim, because I think between the three of us, we may accidentally like forget to go eat dinner and just talk about NeoVim for 12 hours. So we'll like hold off on that. Well, one of the things that, that I think is important to what you were saying is like that aspect of getting a dopamine hit along the path of learning. And I think it's really undervalued. And it's also undervalued in sort of like, the tooling or the syntax or the result or some other way that like whatever it is that you're doing can surprise you to motivate you to learn the next thing. Right. And, and like, when you're thinking at a time scale of my career, right, not like this week, but my career, you need some like positive reinforcement to come back to make you feel good. So the next time you sit down in front of the computer, it's a happy place. It's not Pavlog's dog in like getting, you know, a sad consequence, but I'm ready to eat you know and I think that aspect is so missed by a lot of people thinking in this longer time horizon.

And I think that is something I knew so well because I tried to learn programming three times and failed earlier as I was younger. I just I wanted programming because I wanted to make games and wanted and I just couldn't get it because the initial thing it wasn't a curve, it was a step function. And there was no one there to give me a fucking hand and pull me up. So I just didn't get it. And I just gave up. So by the time I rolled around to Ruby on Rails, I was like, do you know what? We're not gonna make it this. We're gonna make it soft and we're gonna get people on board, but then we're gonna keep going, right? So I'll take the dog, he doesn't know anything. Great, that's your first day. That's your first day. But in a year from now, you can't be that fucking dog. You can't be a copy pasta monkey who doesn't know what you're inserting. At least some of the time you have to pull the thread and try to understand what the fuck is going on. Because it just becomes more enjoyable too. You have far more agency when you understand what you're working with, the tools and the materials and all of it. And then you go like, yeah, do you know what? I'm stumped. I'm stumped all the time. Every programmer is. We hit a bug. It seems like magic, but then we go like there is no magic. There's just fucking sand shoved together into silicon, and that's the chip. And it's going to be logical, I just don't understand it yet. And right now it seems like magic. I'll just keep cracking at it. And in the end, you'll figure it out and you go like, yeah, it wasn't magic. It was that thing. And having the developing the stamina and the perseverance to do that is just crucial. it's only gonna accelerate. The more we get the fucking AI slop in the just like spewing shit at us that we sort of barely understand we shove it in right. The more or the easier it almost becomes to just embrace that dog. I don't know where the fuck I'm doing. I'm a prompt engineer now. Like, I want you to be ready. I want you to be ready with some basic virtues and tactics for figuring shit out because it applies not just to programming. I've applied that principle of a programmer's mind, I'm going to figure things out, to about a million other domains. I've had bugs in my house with my ventilation system and I go like ooh, this is curious I don't understand why it's doing these things. But if I figure all the things out, oh, here's the variables I can clarify There's five of those it's a closed-loop system. We're not getting the up to right. This is great you just become a more capable human when you learn to understand thing and develop competence and Having seen the cycle a few times. I mean race car driving is another example I often use, is I went into race car driving quite late in my life, competing against kids who've been go-kiding since they were six years old. I had to figure out that system and figure out like what are the main moving levers, how do I go faster, what's the main thing? Oh, it's the brake trace, it has to look like a squiggly thing when I'm in this car and when I'm in this card, it has to look like a table breaking. And just diagnosing all that. It was the same fundamental sort of human inquiry principles I was applying in programming. So as you develop those in programming, you get them for the rest of your life. I mean, that's the other thing. This isn't just an investment in fucking whatever JavaScript framework du jour that transfers nothing to your rest of your life. If you approach this as like, competences is important, understandings is important, pulling the thread is important, sometimes struggling a bit is important, these are skills you take forward and they just make everything better, they make everything faster, they make everything more fun. It is just more fun to be competent. Yeah, I published a thing

about this a little while ago. And I think Prime will probably echo this idea too, that in the preface to the argument, I said, suppose AI can write all of the code possible in the future. Is it still today worth thinking about and learning code? And for exactly the same argument, my case was like, yes, today you can pay your bills with code. That's good. And all of the skills you get, even if they don't, even if I learn a bunch of things about OCaml, OCaml mentioned, it's like, I can still take that thinking, the brain muscles that I'm learning, the way of solving problems, decomposing into smaller things. I can just use that same thing for whatever the next task is when AI writes all of my code for me. And I think that's really underappreciated as well in the programming space. Hugely. Yeah.

Okay.

So

I do, I have to hear about this. So you tried to learn program programming three times. Now, I pretty sure that nobody in this chat thought at any one point you struggled learning programming. Yeah, here you are admitting, hey, I actually struggled learning programming. I've seen this a lot of times that when people meet their heroes, they end up saying something along the lines of, oh, yeah, actually, I struggled on this thing. And it just feels like a world rocker. Uh, why, why did you struggle, I guess, with programming, especially three separate times? Well,

part of it was the environment in which I tried to learn it. Like now, and I'm going to sound really old, right? Like things were difficult back in my day. Um, the first time I tried to learn programming was literally like 1985 or 86. I was like five or six years old. I really liked video games and I wanted to figure out could I do something myself. And there were these magazines print on like dead trees that would come out and in the back of the magazine, like literally code listings would be there. And you could type in the code listings, like just the lines. And I'd spent like an hour and a half typing that shit in. And like something wasn't right and it wouldn't work. And I tried at that early phase to like, could I figure it out? Which is sort of a good encapsulating, like I was no fucking savant, right? Like six, seven years old. I'm just, I'm not getting it. Then again, I tried it, like, I don't know, 12 or 13. This was EC Amos. I remember a little bit more about this. And I remember thinking, what's the point of a variable? Like, why do you want to change it? Like, if I just wrote that it was five, why does it... There's just some basics of programming I just couldn't grasp. And I think part of it was, there really wasn't a lot of help, I'd say. I mean, you could get some books, you could get some magazines, there was no Stack Overflow, there was no Google, there was none of these other things, there was no AI. So I didn't get it. And then I tried, or I sort of looked into it once more when I was like about 15, 16, I had a bunch of programmer friends at that time, who were all deep into a simpler programming. And I looked at that for about five minutes and went like, whoosh. I remember thinking like, wait, I got to do vector math to do fucking games programming. Yeah, okay, I'm out. I'm not doing vector math. That's just not my forte. So it wasn't until, I mean, I guess I was like 20 years old or something like that, when I had been working with the web for some time I'd sort of been doing some HTMLs, some CSS, and then around 2000, I helped develop a gaming website called daily rush to decay. I had a couple programmer friends, I was working on the project with me, and we were writing in PHP. And I just kept helping out. And then I helped out a little more. And it really clicked in a different way because there was actual real programmers on the project. So if I fucked something up or it didn't work, there was someone to ask, there was someone to help, there was some other code to look at. And then it kind of started to click. And yeah, it was just a slow ramp. of, I wish I'd learned programming when I was 12. I would have liked to play those games, whatever the hell they would have been. That would have been fun. But it wasn't, it wasn't really until I found the right level, the right level of abstraction for me. And the right level of distraction for me is not vector programming. I have accepted that fact. I'm not going to be a games programmer. I'm not going to write a 3D engine. That's just not where I thrive. I thrive at a level of abstraction where it's more about concepts, words, picking the right things. That's why Ruby was so appealing to me, because after you boil away almost all the line noise that most broken languages have, you're left with essentially a form of English, like a very regimented form of English, but also Ruby, not that regimented, like Ruby allowed sort of all these little alternatives, like, oh, you could do whatever, command unless condition. What the fuck? You've changed the, isn't the if supposed to come first, and then there's supposed to be a not, then there's supposed to be a condition and then you do the thing. Ruby to me was just blowing my mind by the flexibility it had. And that was what helped accelerate things where I'm like, Oh, I'm actually good at that part. I can't do the vector math stuff, but I can do this stuff. I can build information systems I can do object oriented programming. I know object-oriented programming have sort of fallen a little bit out of vogue at the frontier here, like, oh, functional programming, like, that's the thing. No, I'm an object-oriented programmer through and through. I like functional programming paradigms. I like aspects of it. I like shoving that into my little grubby objects. as a way of thinking about the world, I love object oriented programming. That really just made sense to me. I could see whole systems. I could see all the elements. I could see the main models, and then I can have some functional programming elements into it, but I don't see like functions and data separate. So that helped me too. Just object oriented programming, high level programming languages like Ruby, and then information technology. What were we creating for some reason? I don't know. Why? Like, why do these things happen? I just really like crud. Like, I like dealing with a database. Like often it sounds disparaging. Oh, you're just making another fucking to-do list. Yeah, that's the internet. The internet is just a fucking collection of people making to-do lists. That is crud apps, right? Almost everything on the internet is some fucking database getting inserts, updates, rights and deletes. That's what it is. Let's just embrace it with joy and go like, yes, that's what I do. I fucking hammer on a database all day long. I shape it into little blocks and they can make virtually everything that exists on the internet. That's pretty damn incredible. I mean, it's not games programming. I kind of wish I was a games programmer. That is significantly more rock and roll than making another to-do app. I will accept that. But hey, you gotta work with the blessings you've got. And that's the blessing I got. Object-oriented programming at a high level in Ruby. Okay, that's that's my cards. That's my flush play it It's so funny

that you say that because I don't know what it is about programmers and this this look of game programming But I think every last one of us that couldn't keep up with actual game programming like man You know one day one day I know that I could be that guy if I just tried hard enough. I know I could. I

know I couldn't. And do you know why? I just tried recently. And so I think about a year ago or something, I took some time off, and I was like, do you know what that dream about games programming? Maybe I was just too young, maybe I'm so much smarter now. And I sat down like, let's start with something simple. And I picked up like Commodore 64 assembler code, like there's actually a great community, people making new Commodore 64 games. I'm like, that's got to be the simplest possible thing. Here, we're emulating literally a CPU with one megahertz, one, not two, not three, one megahertz. So how hard can it be? I mean, I could almost probably keep up with that CPU if I was running real fast, right? Like how many instructions per second can you execute with one fucking megahertz? So I sit down and I start like, all right, crack the knuckles, let's get going here, I'm going to do some game programming. And I lasted about fucking 45 minutes until I was like, yeah, oh shit, this is why. Yeah, this is not me. I cannot do that thing. And do you know what it reminded me about? So I talked about race cars and I really like cars, right? I like driving and I get all this stuff. And then I thought, multiple times thought, well, maybe boats, maybe boats are also fun. Like there's the ocean, I like the ocean, I like looking at the ocean. And every fucking single time I go out in a boat, it takes about 15 minutes before I go, yeah, yeah, I'm not into boats. Like, I like the image of boats. I like the idea of winning your hair and saltwater in your face kind of thing. But that's just not me. I want to get back to shore. And that's what I felt when I went into this Commodore 64 tutorial. I'm just like, yeah, gotta get back to shore. Like I gotta get some Ruby into these fucking Finkers again because this isn't me. I can't do the mob the thing the XORs all

the stuff I'm surprised you didn't try out something like dragon Ruby. I don't I don't know if you've heard of it It's a game engine in Ruby. So no, I want to check that one out. That might be a little bit

You know what though? I think it's not the syntax. That's not my problem. My problem is the actual domain, right? And a lot of it is about its collision detection, its math, its things and like, yeah, that's just not. I mean, I willfully flunked my high school math final year, because at the beginning of the year, I remember getting the first assignment and I went like, this is too much fucking work. I got other things I want to do with my life. I can't do this. So I made a deal with a buddy in my class like, hey, can I just copy your homework the entire year? It's fine. I'll just introduce some bugs in it so it doesn't look too obvious. And I just did that. And I arrived at the final point where we had to take the exam and I was like, you know what, I could probably wing it. You only need 26% correct to not get an F, at least in my school. I was like, that's gotta be possible. And I just realized, no, it's not fucking possible. Math is very unforgiving. You don't fucking know it. You can't do it. There's not like, oh, I'm gonna figure it out in the exam room here. Like, I just, uh, integral equations here. They're just gonna come to me through divine intervention. No, they're not. And then I learned, do you know what? That's not true of all the mains. I went to my exam in like in English, I'm already from Denmark, and I went to my exam in Danish, and the Danish one is the best one. So I walk in and you draw your question, right? Like, this is what you're gonna get examined on. And I draw fucking, I don't know, middle-aged poetry or something. And I go like, I don't know fucking anything about this. I don't remember anything about it at all. I walk into the exam room, you've got 20 minutes to prepare. I open this book and there's like a checklist of like the five factors you should look at. Oh, the magic. Anyway, whatever, right? Then I walk into the fucking exam room. And I just start like bullshitting at 200 miles an hour. Like so fast, they didn't even have a chance to interject the question. And there's only 15 minutes of examination. And as you might have realized by now, I could fucking talk for 15 minutes without letting anyone get another word in. And I'm like, boom, A plus, boom. I was like, I gotta find a way to commercialize this aspect of what I'm good at. Not the fucking math part. And now a lot of people go like, oh, that makes sense. That's why Ruby and Rails are so fucking slow. And like, yes, to some degree, yes. But also to some degrees, not because by the time Ruby and Rails got off the ground, I did happen to attract enough people into the big tent of this who actually did like math, right? And who didn't fucking flunk their high school thing because they thought it was too much work. And they made it better. And I think this is what's so wonderful about open source. You can show up and you go like, I am good at these two things. Do anyone want to trade? I can make these parts and then you make the other parts and then we put our parts together and it's more than the sum. That's really nice. That's really great. Ruby on Rails is as good as it is, as fast as it is, used in as many high profile places as it is, because literally like 7,000 people contributed code to the code base. I wasn't gonna do all that shit on my own. I mean, I was gonna make the to-do list work on my own, but I wasn't gonna make Shopify do a million requests a second on

my own, right? That needed some of the math folks to come in and assist. Yeah, speaking of open source, something that I really resonated with from your writing is talking about how open source isn't a community or a democracy, right? It's a gift exchange. I don't know if you want to sort of elaborate on that idea because I really, so I'm a core maintainer of Neovan. I wrote telescope and a bunch of other stuff. So like, I definitely, when I read that, I was like, yes. Give it up for DHH on that one. Neovim, by the way, yes, Chad, I know you're already gonna spam it, that's fine. But go ahead, take us away on sort of that aspect. Yeah, so what's really interesting

about open

source

is it has sort of the shape and the outline as something that