"I don't like Programming" | Prime Reacts

0:00 I don't like programming. I never liked programming. >> Look at that cheeky smile he already has. Bro is baiting us like I've never been baited before in my entire lifetime. Don't even Don't even lie to me. >> In fact, programming. Now, I'm being a little facicious here, but I'm going to tell you what I mean. And that's going to include a few themes, including our I like his Montana based background going on right here. Very nice. I guess this could also be Colorado, but I would assume this is Montana, the greatest state in the union.

0:29 >> The things that we learned as programmers or the things that maybe you should learn if you're getting into it. Are they useless? Does it matter if you know what a hashmap is or like a lot of those fundamental concepts and stuff? I also want to talk about the type of programmer that I am and a bit about my like programming career trajectory. And also I want to talk about the whole transition uh to AI coding and also how in some ways at least people are saying the joy of programming has been taken away from them. And I hear that a lot.

1:00 Anyone press one in the chat one in the chat if the joy of programming has been taken away. By the way, this is why uh Odin this is why I actually went back to this uh because Odin is the C alternative for the joy of programming. And I feel like in certain ways because of the type of programmer that I am that maybe I've adapted a little bit differently than most people. When I was in school in college, I took an intro to C class back in 2016 and I learned all sorts of things. Of course, a lot of which I used in my career. I learned about loops, if statements, logic. Of course, I used those things on the job, but I also learned about memory management, right? And I don't think professionally I ever needed to do manual memory management even using a language like C++. There's smart pointers now. Why does anybody need to know how to manage memory? Maybe there's some niche. We did a whole bunch of manual man memory management but the large part of like even the television app was being all upgraded to smart pointers cuz we are largely on like a half version of C++ 11. Apparently there's two versions of C++ 11. The version A and version B. Version B has all the sweetnesses, but version A does not. And we were on version A for a long time and then slowly got up to >> areas that it's like extremely important. But most people, they're never going to need that. And also, what about DSA, data structures and algorithms? I spent a lot of time doing that, learning that, and I enjoyed it.

2:24 But did I ever really use it on the job? I mean, not really. I mean the fundamentals obviously like hashmaps, loops, like efficiency and general optimizations, but you can get that, you know, at mostly just the leak code easy level. You're not going to need those graph traversals on the job. Why would you? The only time I remember something like interesting coming up at Google, you know, the company, I'm not trying to mention that in like a bragging way. I'm just mentioning it because it's kind of the company that's known for like DSA interviews, right? And then you get in and you don't really use that. And I remember there was a case where uh Google has CL's which is basically like poll requests and then each CL will get like a unique number and I forgot what the exact detail was but I think it was like each CL will have like a unique ID of course that's autogenerated and then uh when it's actually merged in Google they have like a mono repo so every piece of every CL every pull request is merged into the exact same repo. Um and I was asking like a question about like something about the CL number conflicts.

3:21 I don't remember the exact details but basically what he said was yeah these are monotonic increasing so that there won't be conflicts or issues and you'll know which CL was merged before the other one and when I heard that I was like oh holy did he just say and this was actually like right before I left and he said it and I was like monotonic increasing like hey DSA right but for the most part like those concepts really were not needed and I've you know I've run into a surprisingly large amount of them throughout my time I have written multiple binary searches over ranges. Uh I've had a lot of like just things like that throughout my time. Why? Just because we were I mean in 2012, a lot of the stuff I was doing uh this was even when I wasn't at Netflix. This was my previous company. There was a lot of stuff I had to do that was just simply uh difficult, right? And you had to just be like, "Okay, we can't just simply linear scan through all this stuff."

4:11 >> Uh maybe nowadays we don't ever even need to know how to use Git. But I mean, you might still want to because it's so damn painful to watch these LLMs do it. And I don't know who made the judgment call to not just allow maybe like a text input or something and I can just type my own damn commit message and get this done with. I know I can use the terminal. This must be how Vim users feel watching normal people type. But it really is being a Vim user. Everything else is so painful just watching. I mean, it's why I can bar I could barely stand pair programming. It is just the worst experience in the universe.

4:49 >> You know, mentioning a few things and I'm not going to act like I have the answer to these things, but I'll tell you for my opinion, my very humble opinion that I don't think the things that I learned and the things that I worked on that I didn't directly use, I don't think they were wasted. I really don't because I did a lot in my life and so have many of so has everyone, right? And what I mean by that is I I did so much things in school. We all learned so many things, so many subjects, so many concepts. And um do we really need that?

5:24 I mean, we all remember sitting in class thinking when am I ever going to need this knowledge that's right in front of me? And many times you don't. But does it mean it's useless? I'm not convinced. Holistically, we are like people that have all this like knowledge and I don't think it's worth like cutting out fundamental things. I think that's very like dangerous. It sets like a really uh scary precedence because okay, maybe maybe it is useless. Maybe nobody needs to ever know math anymore. We have calculators. But I think learning is extremely important. Uh because >> All right, so I have to jump in. I was waiting for him to kind of wrap up this idea. Uh well, one reason why I think that a lot of these things are super important is that the act of actually doing something is not just that you've okay, you now know how to say program a hashmap. I don't think that that's like the it's not going to be something that directly impacts you. It's not going to be the thing that's like, oh, when I go to my job, my job is going to be writing hashmaps. Like that's just not going to happen. You're probably never in fact you're you're going to never write a hashmap ever. Okay? It would be stupid of you to do that. You should just use a uh a library to do that. uh hashing is actually extremely complicated. Just let it go. But nonetheless, it's not even it's it's about the journey. Always has been. Okay. So, it's not about the journey either because that that that that entire statement, the journey is better than the destination has this really really wicked implication. Which means that if it's about the journey, you start off here. It's about the journey, not the destination. It either means that you stay the same or life is always progressively slightly getting worse because you're always finishing part of the journey. Okay, that's not it at all. uh just in case you're wondering uh the destination in fact has to be better than the journey because if you're not you're going downhill. Okay, so now that we got that little last bit out of the way uh the thing is is that if you are working on something and you solve a difficult problem or you go through some amount of of challenge trying to implement some sort of algorithm in which you're never going to use again. You also got to remember like every time I did one of those processes, it like I could feel my brain go I have a new way to look at a problem and fix things. Like I can I have grown in my ability to look at things that are difficult to program and now wrap them up. The reason why I am good at programming is because I also did hundreds if not thousands of these really stupid kind of thought experiments on programming. And by doing that over and over and over again, it's made my ability just to do regular ass programming really really good. And it's because you had to solve this over like it's very very difficult. All of those little things, they add up every day that you go and program even a big project. It's it's not like when you go and program a project, you're somehow getting smarter at programming. But if you do the same project for a long enough period of time, organizing that information, thinking about it, slowly building up the project, you do actually start gaining new skills cuz you have to refactor, you have to rethink things.

8:10 You will like gain new ways of looking at something. And then of course, bada bing, bada boom, you are now better at programming and you don't really like realize it. There's not like a certain moment in time that you're like, I went from A to B. It's like the constant churning through it and doing stuff. That's why I always worry about like the vibe coded future. It's because I don't understand how you get that experience. Like I truly don't. Like there's only two there's only two possibilities in the vibe coded future which is vi like a you never have to know programming ever again or b you do and you're going to be sad if you never know programming again.

8:42 Right? Like there's not really like some sort of middle ground that's going to be like oh you just don't have to learn how to program except if you're doing this right? Like that just I just don't buy it. I just don't purchase that. It's either like things are going to be solved or they're not going to be solved. Vive coders are betting on a vibe coders are betting on a but it's there's still so much to a like anybody with any level of experience can go through and look at it and you can see like you you will be driven nuts by some of the things and you know that given enough time and situations it's going to get worse and worse and worse and it's going to cause so many just annoying problems. It's really trending towards a and it could very well be a you know honestly if it's a I'm going to be okay.

9:20 I am learning to accept and be okay with A and be okay with B. Uh that's why uh that's why I mean I I keep on always trying to do both right. I do handwriting and I do vibe coding. And when I say vibe coding now I really am pushing really hard into uh what's it called? Seeing if can I can I get can I with a prompt can I make a prompt do the small change I want to do. And man, I spend so like something that would take me 3 minutes of hand coding is taking me like 30 minutes of just like trying to coax this machine into producing the code I would produce. It is a little bit frustrating. I'll tell you that much.

9:52 Not going to lie to you. I do feel I feel I feel things emotionally going through it. I'm trying not to be angry. Like I kid you not, this function right here, this is me regenerating it for like the fourth time. I was like, "Brother, here's what you're gonna do. You're gonna use the function. You're gonna use the function sprite draw. Okay? I expect you to use Sprite Draw minimally. And that's that. And I and I even gave it this interface right here.

10:13 And I'm like, here's the interface you're going to use right here. All right. And I was just like, you're going to use this interface and you're going to use Sprite Draw. Doesn't use Sprite Draw. Used it just didn't. It used it. It copied the code right out of Sprite Draw. And I'm just like, I'm I'm going to I'm going to lose my marbles. I'm just trying to figure out how to move quick. I just [ __ ] just want to destroy just want to kick a computer.

10:34 It's not just about like the concrete thing. Okay. I learned this random framework or random tool and now I got to start a different job and I have to learn a different tool now. Everything I learned before is completely useless. I don't think that because of uh this uh concept that I haven't invented but it's called like a mental model and we all learn a lot of mental models just from school. If you pay attention, for example, in math, you learn these logic rules, these rigid rules, and they're consistent, and they teach you how to think. If you take a math person who's super good at math, and then you try to put them into a programming career, they excel, dude. Like, it's just a thing. I can't tell you exactly which math concept directly, you know, made them a super good programmer or not, but it's a phenomenon. This is a thing. And there's so many different mental models in the world of so many different concepts.

11:33 >> Okay, I do have to say I really did enjoy my math classes as I got older. I really, really, really enjoyed them and I really excelled at them and I loved it. I despised Oh, yeah. Like math, I did not enjoy math for a long time, but I forced myself to to enjoy it. I I spent all summer school doing what 3 hours a day in math class 4 hours a day in math tutoring until I got good at it.

11:57 And then once I got really good at it, I didn't need the tutoring anymore and I could just accelerate, right? I I didn't I wasn't I didn't, you know, I kept it humble. I ate some humble pie and went to stupid tutorings and then I got great at it. It was good. It was fun. I was I really enjoyed it once I got good. It was awesome. >> Many in the programming world, many in psychology, many in nature. There's like the concept of evolutionary biology. And the the important thing about these mental models is they allow us to reason about things that we do not understand.

12:29 Like in physics, there is this concept of laws of nature. Newton's laws of nature. They explain things. And when you don't have these things, you do lose something. I strongly believe that because I remember, I don't know about you guys, but I remember being a kid, I remember thinking, why does the moon go around the earth? It's so strange, right? And of course, an adult who knows is going to look at the kid and say, "Oh, it's gravity. It's gravity."

12:55 Gravity is just a word. That doesn't really explain anything. Newton's laws, Newton's laws explain it. An object in motion will stay in motion. And the beautiful thing about these mental models, >> I was really hoping he was going to drop the generalized gravitational formula. Okay. I thought it was Dude, I I honestly I was hoping Dude, I was hoping he was I forget what exactly it is, but I know there's two G's in there plus a universal constant with something being divided. Dude, I thought we were going to get the universal gravit gravitational law, and he was just going to throw it on us. I was so excited.

13:28 Yeah, there's distance involved. Oh my gosh. GM. Yeah. Yes. Yeah, that looks that looks that looks that looks like it. I was so excited. I was so excited. We didn't get it. Very disappointed right now. is Newton's laws, they don't just explain why the moon goes around the earth. It explains so many things. And when Einstein, you know, built went on top of the things that Newton did, he had to deeply understand the mental models that Newton did, right? That that he created. And so it turned out that certain things that Newton Newton's certain things about Newton's theory of gravity were wrong. But was it useless?

14:05 Obviously not. But anyways, I could ramble about this literally all day. So, I'll move on to the type of programmer that I am. And >> I feel like some of you will be will be happy. I know who. Yeah, Einstein. Um, there you go. >> I'm about to make a very offensive statement to a lot of people. I know that. >> Okay, everybody. They have warned us that trigger warnings in fact don't help. You will be now more triggered.

14:33 Okay, now that I'm saying it, you're going to get more triggered. All right, so let's get triggered together. >> And I'm going to say it. I was the type who got into programming for the money. >> For those that don't know, this is totally an age problem. That's it. It's an age problem because you may not know this, but in 2005 when you got into programming like me, I was in college in 2005. So 11 years 11 years before him.

14:57 When I got into when I got into programming, you didn't make money at programming. Programming was not taking off. Like programming was just at the time like maybe in about five to five more years it really started taking off as a career you could make a lot of money in but you couldn't make money at that point. Yeah. When I graduated college I got a $50,000 a year job. And I was like bros we're eating Red Lobster. Okay. I'm eating Red Lobster.

15:22 This is this is crazy. 50 $50,000 right? Like and so it wasn't even an option. There was no option like oh hey you're going to go into programming. Oh man, you're getting rich, bro. Bro, get the money out. Right. And so that was crazy. That that was completely different. In 2016, it was a very different feel. By the time you were in the mid 20ou mid 2010s, people were like, "Hey, guess what everybody? You're making money." Okay. Oh, you're going to come be a little little fang boy. Hey, guess what fang boy? You about to get a whole bunch of money, [ __ ] We're going to rain on top of you. Also, there might be some weird things that happen, but don't worry about that.

16:01 Those are two different worlds. And so, when someone says, "Hey, I'm in it for the money." Like, I'm not shocked. Now, if somebody graduated in like 2002 and was like, "Yo, I'm in it for the money." I'd be like, "How the hell did you choose this career then? Why didn't you choose like being a doctor? What are you talking about?" Yeah. The mid the the mids it wasn't it it it was just okay. It wasn't fantastic. You weren't making a lot of money. You weren't making no money. So, you know, like, so when I see when I see these things and people say this, they're like, "Hey, I'm in it just for the money." It's like, "Yo, dog."

16:32 Yeah, that's totally like, you got to make that choice. When I was a kiddo, I didn't get to make that choice. It wasn't, "Oh, I'm going to do it for the money." I There was only, "Oh, you're going to become a programmer. You're in it for the love of the game. That's it. There's no other option because this thing's not going to pay as well as being a lawyer or a doctor or probably even a nurse. Like you're just gonna be at a computer and everyone kind of thinks you stink and are stupid. Very very different world. Very very different world. Right. So it's just different. It's totally different world.

17:01 I did I got into fang for the cash. I really liked math. And >> to be fair, I joined Netflix for the cash. And uh to be actually the the honest part is I left my job that I liked because I wanted to take the risk. I wanted to try to do something risky and it was very I was very worried about it. They say they fire fast. The problem was is that I was always like the smart kid in all the jobs I was at, but they're all in Bosezeman, Montana. And I wasn't sure if I was smart at all. Maybe it was just like, you know, big fish, little pond versus, okay, I'm going to go out to the Silicon Valley where everybody out there is a good programmer and I want to give it a shot. I wanted to really try to do something good. And so I was like, okay, I'm going to get I'm going to give it here we go. I'm going to give it a I'm going to I'm going to give it a good one. And so there we go. And it turns out I did okay also out there. So I was very happy. It was very scary though. I I was actually pretty worried. My wife was uh 30 weeks pregnant or 32 weeks pregnant for the first our first kid when we got the first interview and we moved out there at like 36 weeks pregnant. Had no doctor, knew not a single person, didn't even have a place to live and she was about to have a baby at any moment. It was kind of it was generally pretty stressful >> physics. So, I actually studied uh electrical engineering when I first started in college, but they required me to take a programming course, which I wasn't happy about, and it was intro to C. And I really wasn't excited about it.

18:21 I really struggled with it. Even just print F was really hard for me to wrap my head around. I don't know why. I'm kind of of a slow learner. But once like I think a month or two into the class, you know, you learn about variables, you learn about the different types, you learn about if statement, loops, functions, reuse code, you learn about strrus. And then there comes a moment where for me something just clicked. I realized that wait a minute, these small number of primitives that are very logical and very consistent actually it's it's so few of them yet they're so extremely powerful. They have the the power of infinite complexity. These small primitives can create incredible software operating systems, right? And that's what I loved about it. The same thing I loved about Newton's laws that a small number of things can actually explain so much that a small number of fundamentals have an incredible amount of power. So actually I loved it. I loved memory management. But then I had to get practical. And then when I was looking at the job market, I was like, "Okay, well, I got to learn web development, right? Got to learn JavaScript and I got to learn what an XHR is." And an XHR, an XML HTTP request, is named that. Yet, it's not necess. By the way, if you haven't, if you haven't done that, if you haven't done an XML HTTP request, I just want to let you know, you need to go back to the basics, buddy. Okay? You need to go back to XHR business. Okay? He must be trolling too. Yeah, I'm a little bit surprised why he was doing XHR in 2016.

19:55 Like I I don't really believe that at that point you had I mean dollar sign.get.jax.post those were all around for some time at this point. Like those that was jQuer that was the jQuerious day and I was using that in like 2009 2010, right? There's no way that jQuerious was not had at that time. I just I refuse to believe I refuse to believe you're hand rolling your own format and it's not necessarily HTTP and it's not necessarily a request. It could be a response to an XHR could mean response.

20:25 So, as a person who who likes logic and likes consistency, oh my god, you cannot believe how much web development pissed me off. I hated programming to an amount I cannot articulate. As much as I loved C and the primitive nature of it, I hated web development just as much for the opposite for the same reasons because it was kind of the opposite, right? So, I did get a little bit jaded. I will say that the web environment is just a piece of crap. And I tweeted out earlier today that I feel justified every single day in in my disgust of the web ecosystem. And I feel it more and more. And with the non-stop shy hallooting where you're getting constantly wormed all the time, I have never felt more uh more exonerated in my life than saying, "Hey, it's really not a good idea to have like this e this is not good. What we have created is not good. We should not be living like this people. I think it's dangerous." When I started using rust, what was my one my first major complaint of rust? Cargo looks like npm to me. I don't like it. I don't like that we're breaking out every last little thing. I don't like that. I don't like it. And guess what? I was right. I don't think it's good. I >> I've talked about this before, but I was unemployed for like a year after I uh left Amazon. Uh and during that year, let me tell you something. Let me tell you something. I did not do any sort of real development. All I did for that year was make leak code videos in Python, only Python. And then I I did I was lucky enough to get into Google. And when I got into Google, I did not do any Python. I did not do any TSA. But I also did pretty well at Google. I got promoted within a year and I think obviously the fact that I have a CS degree and I didn't cheat through that degree. I actually did learn the fundamentals and the important stuff that did help me a great deal and the fact that I was still thinking program pro like a programmer um doing DSA uh videos and also kind of communicating my thoughts. I think that helped me a lot.

22:18 And then so at Google, I was a what I would call a copy paste programmer. And I'm embarrassed to admit it, and I know I'm not doing myself any favors by admitting that, but I was just so lazy because a lot of the work at Google is not that interesting. You're not doing a lot of the the the fun interesting computer science concepts that I'm I enjoy develop. >> People don't realize that Fang is really boring. Like fang is just generally super boring. A lot of what you're going to do at larger companies is a lot of nothing. You're going to argue. It's going to take you weeks upon weeks to like change a button color. This is why like I mean how many articles have we read? I had that one private message of somebody dumping me their actual day.

23:01 It's mostly meetings and then you have to go like everything I even at Netflix even all the stuff I had to do. I had to like write a backing document. I had to come up with the reasons why we're going to do it. I had to go and get really arguy for the things. Luckily, I picked a bunch of big projects that would take time to actually program, but nonetheless, I had to like go and first fight for my right to program and then I could program, right? Like it it is it is a different world. Startups, yeah, startups are much much different because startups the ship can be moved really easily. At a bigger company, it's hard.

23:30 I don't necessarily even blame them. The thing is you can't at a small you can at a small company say I don't like X. I'm going to change X from the front end to the middleware, from the middleware to the back end, from the back end how the data looks. Like you just can't do that at a bigger place because you're actually intersecting so many different teams, right? And so it's just like it's not the same. You know what I mean?

23:53 Anyways, I like this. >> Just looks very different there. And I wasn't really enthusiastic about learning all those internal tools and stuff like that, but I still did it. Basically, uh, most of the code, you know how people say that with AI coding, most people aren't handwriting code anymore? Baby, I was not handwriting code for a long ass time. I mean, not literally, but um for example, like you know, back in a hundred years ago when Stack Overflow was a thing, a lot of people were the types of programmers that they would just, you know, if they had a question, they just blindly open up Stack Overflow, copy and paste a code snippet into their code, maybe change some variables around, and then they're good to go. Maybe they didn't even fully understand that piece of code. I kind of did that uh at Google. They don't have Stack Overflow, but they have like their internal mono.

24:39 >> Did you guys do that? Who did that? Type one in the chat if you did that. Type one. You guys disgust me. Okay. I would go and I I would certainly copy it and paste it into my editor, but then I would debug it. I would have to understand why it worked the way it did. I would then want to change a bunch of stuff about it. I never liked the shape that it was in. So, I always wanted to put it off into functions and different ways I put it or spread it out or change it. But rarely did I ever just copy, run, see it work, and move on like that.

25:08 That's crazy. that I I don't understand that at all. Gross. Prime debugging is odd. True. True. >> Which is like in code search. And so I would just, okay, I needed to use this library. I could use the document. I could go read the documentation. The documentation sucks ass at Google. So screw it. I'm just going to go into the codebase, find somebody else who used this API, copy and paste their code, try to understand it a little bit, change my variables around, and I'm good to go.

25:32 And so even though I'm the type of person who likes to go deep into things, I like to do a small number of things and do them extremely well and extremely thoroughly and extremely deep, but that's not the nature of being a programmer in most jobs including at Google because let's say you know you have like >> I'd say that's one benefit of doing your own games like writing writing up stuff and and going through that is that you do get to go deep. You get to go deep as all you want. And there is something that's kind of beautiful, I guess, about AI in this sense. If you're like, I don't care about how pathing works and I don't want to really care about physics and stuff. I'm just going to be like, yo, make the monsters walk the path. No mistakes. Like, you can totally do that.

26:12 You could spend all your time just on shaders because you're like, "Dude, I love shaders." Or maybe you don't like shaders at all and you don't want to learn anything about shaders and so you can just invert that and be like, "Yo, make the grass kind of look wavy and then move on." Like, to me, that's okay. That's one kind of thing I do like about AI is you can kind of like skip certain steps if you want to and just focus on the things you like. That's the part of what makes LM's great.