John Coogan (Co-Founder of Soylent): Cargo Cults, Exploit vs Explore, Self-help vs Stealth-help, ...

Today we're joined by John Kogan, the co founder of Soylent and Lucy. Now, the first time we talked outside of this podcast, we originally planned to just talk for 30 minutes, but we ended up having so much fun and had a three hour long convo talking about everything from his company Soylent to mind viruses to concepts in AI like exploration versus exploitation and even faster than light travel. At that point I knew I had to get him on the pod. And here we are. And a little more background him. He is the co founder of Soylent, which is a startup that aim to replace all food with just a drink. And the moonshot vision of that startup is to prevent world hunger as the earth's population increases and making food consumption more efficient and cheaper. Solent has raised over $70 million from venture capital funds like Google Ventures or Jreessen Howard's during just his time at Solent, they have sold over $100 million worth of products. And I would say it's the first company that can kickstarted the biohacking community and also brought tech and food together. So I'm very excited for this episode. Let's just get right into it.

Good to be here. Thanks so much for that introduction. It's really funny because when I saw the calendar invite for this podcast, I saw that I hadn't seen the name of the podcast and I realized that it was called Mental Models of Successful People. And I was cracking up to myself because a, I don't think of myself as successful yet. I think that there's a lot more that I have to do in my career and I also don't think that I have a really strong mental model for anything in particular. I feel like I've basically just winged it for most of my life. It should be an interesting discussion today, but hopefully we can pull some things out and just discuss mental models maybe more broadly and come up with some cool stuff. So excited to be here and excited to talk to you. Yeah. Where do you want to start?

Yeah, actually I find what you just said so interesting. To most people you would be seen as massively successful, yet you still don't view yourself that way. Is that an intentional decision that you made?

Yeah, I think it's to some degree. It's like moving the goalposts, right. As you, as you succeed at one thing, you're immediately keeping your eye on the next thing and you want to continue to improve. So there's that, but there's also just like quantitatively, like Sorling has not sold the business and neither has my latest venture, Lucy. So in terms of, like, traditional entrepreneurial success, I haven't hit that particular milestone, although that milestone is somewhat arbitrary. There are plenty of people that still own their companies and are wildly successful, and there are people that exit their companies and it doesn't look that good for them. It's all over the place. But, yeah, I think. I don't know, I like to think that I've always had a bigger ambition for what success looks like. And I think that. I don't know. I don't know if anyone can really be successful by age 30, 31. I think that there's so much more to success than just starting a company. I think, like, it's much more holistic. Like, you need to have a successful family, successful friend group, successful personal life, and all of that is an ongoing process. So I definitely don't see success as a binary. In any case, I feel it's a track that you want to be on and you want to maintain that track and not get knocked off. And there are lots of people that have found tons of success, made lots of money and lost it all, or become depressed and retreated into their inner circle and never to be seen again, and find it very difficult to think that I'll ever hit some sort of, like, binary and flip and be like, oh, yes, now I'm successful.

Wow, that's a really interesting mental model. Just always moving the goalpost and choosing to never call yourself as successful. Ironically, I think that is one mental model of successfully.

Yeah. I also think that there is a big problem in Silicon Valley and business more generally around the culture of hustle. I don't know if you're familiar with this, but, yeah, Gary Vee does a lot of this, but I think. And like, Alexis Ohanian, the co founder of Reddit, has criticized him a lot for this, but it's broad, it's beyond him, and it's something that's vague and it's not always crystal clear. Obviously, Gary Vee is, like, an extreme example of this, but I think that there is a. There is something. There is like, a deleterious effect to hustle porn in the sense that if you're constantly putting out content that anyone can do anything, like, you just gotta grind and, like, work harder. Like, sometimes that can be used in, like, manipulative ways to, like, get someone to work, like, too hard in service of some goal that actually benefits you more than them. And obviously, like, more broadly, like, the inspirational speaker is often that, or even, like, just the cult leader, if you Want to go more, more, even more broad. It's like the key to success is listening to me and then it's paying me more and subscribing to me and then all of a sudden it becomes like, pay for my executive training course. That's $10,000 a month. It's really, they're just extracting value from you. But there are some good critiques of that. So I feel like people should be very wary of get rich quick schemes and guides to success. I also think like, yeah, I also think like in terms of entrepreneurship, like one thing that, one thing that clicked for me earlier in my career was like entrepreneurship is, it's often seen as like the social network. Like it's Zuckerberg, it's this venture backed route where you become a billionaire and that's it. But there are, Entrepreneurship is like kind of an end game for most career paths. Think about it, if you go to dentistry, dentistry school and you learn to be a dentist, like at some point you're probably going to wind up opening your own practice and just becoming and owning your own business. You're going to hire a receptionist and help and probably manage that and grow that. And like most successful dentists are entrepreneurs in that way. They do their, they have a marketing plan and maybe that's mostly positive reviews and referrals through networks in their city. But a lot of people, lawyers are another great example. Like all the lawyers at major firms, there's a partnership track. They become partners in these limited liability partnerships and they are seeing equity from the fund. Any profit that goes through the partnership gets distributed at the end of the year and they all get checks from that. So it is very entrepreneurial. Although no one would ever say, oh yes, the best entrepreneurs are lawyers. They're antithetical to entrepreneurship. They're the opposite. But in some ways that is just as valid of an entrepreneurial, you know, experience. And a lot of the top lawyers that I've interacted with who are partners have some of the same characteristics. Like they're managing people, they're growing their business, they're thinking about things the same way. But so I again, this comes down to seeing success like more broadly than any one specific definition. And I think that when we put too much focus on, oh, success looks like this or like that's where people can, it's bad on both sides. Like people can wind up doing things that they're not interested in or you can wind up with a bubble in a particular area. Because everyone thought that the only definition of success was to learn to code. Then there's too many tech startups. Then you get the dot com bubble and you get people who aren't there for the right reasons because they're not actually interested in the tech, they're just interested in being successful, quote unquote. So they get. They have a bad experience with all this stuff. I want to be very wary that we're cautioning people against taking anything that we say, like too literally or applying anything too directly. Like your mileage may vary on all of this stuff.

I watched this video way back and it's about how to brainwash yourself for success. And the main thesis of the video is that if you don't intentionally brainwash yourself, you will just default to societal brainwashing. And I think that connects a lot with where we're currently talking about. It's whether you push this narrative of entrepreneurship too far on people, whether that's a negative thing.

Yeah.

So I don't know exactly what the answer is, but right now my thoughts are that this has a much more positive impact than it is negative.

Oh, 100%. Yeah.

Yeah. I think that we can push it and still have a positive impact. We don't. We'll default towards less exploration or less Elon Musk in the world, et cetera. I think we need more of that. What do you think?

Yeah, we definitely have like pretty fundamental problems as humanity that we need to solve and we need smart people to go and solve those. And that definitely correlates to entrepreneurship. I'm a big fan of entrepreneurship. Obviously. I would just encourage people to take in as many viewpoints as possible.

I see. Yeah, that makes a lot of sense.

Cargo cults tend not to work out too much. Like the idea a cargo cult, it's the.

What's a cargo cult?

It's a historical example of these islands where some cargo would wash ashore. And I'm not explaining it well, but basically people who had never seen these things would worship them and they'd become a cult to this cargo. But a cargo cult in technology means like you're building your company and you happen to be in love with Google or Facebook, so you go and you steal as much of their culture and everything from them and you apply everything rigorously and you're using their frameworks for everything and all of a sudden it doesn't work and you don't know why. And it's because you over indexed on something that was built with a specific purpose of making that company work well. So you see this a lot with CEOs who get obsessed with certain entrepreneurs or certain figureheads and then they start adopting not only their positive traits, but their negative traits as well. I don't know, there's probably some good examples out there from Elizabeth Holmes at Theranos. Like wearing the turtleneck. Like the turtleneck isn't what made Steve Jobs successful. What made him successful is obsession with the product and design and actually shipping things on time and making sure that the products were delightful and in customers hands by Christmas. And she wore the turtleneck but didn't ship a single product for 10 years. You've adopted the wrong thing at that point.

Yeah. It's almost like overfitting.

Yeah, totally.

Models.

Exactly, exactly.

Speaking of Theranos, I know you made a video on Theranos and you had quite a contrarian viewpoint on it.

Yeah. So I see a lot of. There's a couple big channels on YouTube that just document like business disasters and I'm a big fan of those. I love reading the books when they come out about startup failures as well as startup successes. But the failures are always really salacious and there's all these like scandalous details and you get to think what would you do in that situation? Or how they allow that to happen. But I saw that there was just like too much negative context around Theranos and that was the only take was Theranos was terrible, ubiquitously. So I wanted to try and create a devil's advocate position where I defended the company and argued for where were the places. Obviously the fraud is terrible and obviously the. Actually the faking the lab results is like extremely harmful and they could have killed people. Like that's unforgivable. But in terms of some of the other things, like she, she deepened her voice and I don't really have a huge problem with that. I think that's fine. People like, there's a lot of sexism in the world and a lot of it is biologically wired around the tone of your voice. There's been a bunch of studies around this. She probably just read the study that said that CEOs there's some study that says like CEOs who have a deeper voice, have a longer tenure, they manage larger teams and they're worth and they make more money. And so she was probably like, people aren't taking me seriously because I have this high pitched voice. So I'm just going to artificially lower that. And that's fine. I know plenty. There's plenty of people that like wear glasses that don't have actual prescriptions because Glasses are associated with intelligence, so that's like fake. Or even the person who like wear, wears a business suit to a meeting to impress a client when they don't need to wear a business suit to their office. Like they're trying to tell you something and they're trying to communicate that it's on the job of the recipient of that information to process it. It's a little disingenuous. It's not great. But I prefer if everyone was just their natural authentic self. But there was something that I just thought was like it was overblown. Like people made it out like she's a psychopath for doing. No, it's rational. There's sexist people out there. There's plenty of people who are not, who are not complete frauds who might have done something like that. But of course it fits like a larger narrative. But yeah, it was just fun. I don't know if I'm gonna do more of those there, but I do. Trying to find the edge of where the debate because at a certain point when something crazy happens, there tends to be this pile on effect where the stakes get higher and higher. And then all of a sudden people on social media are like, she should be given the death penalty. And it's okay, let's back up, let's think about what's going on. The investors that put their money in these are wealthy people. They're taking a bet on something really ambitious. They back the wrong person, clearly. But people make these types of bets and we have to live with that. That's okay. But yeah, it was fun. I actually made a more recent video about Moderna and had a very interesting take in that one because four years ago the press actually was comparing Moderna to Theranos because they were being very secretive, they hadn't published any data and there was a bunch of turmoil at the company and there was high turnover rates. Top people were leaving their leadership team and they were getting called out by just one media outlet, wrote this article. And then one of the big industry insiders had a longer piece that kind of loosely compared them to Theranos. Now it wasn't so head on, but I just thought it was fascinating that four years ago they were being called the next Theranos and then this year they got their COVID vaccine approved, which is an insane like turnaround. But Moderna is completely different company than Theranos. Way more experienced leadership team, way more just solid foundation on the fundamental technology. They're really commercializing something that is real. But they did Hit a ton of stumbling blocks and a ton of hurdles. It's still interesting to think about where the company will go. It's not super clear to me that the company is going to be the next Genentech or something.

Yeah. I think the most interesting points in the Theranos video is how that connects with taking bets.

Yeah.

And I think when most people think about taking methods or think about experiments or building something, it tends to be pretty binary. It's think about one sample like I failed or like I succeeded. But I think generally it makes more sense to look at the entire sample space finding like an optimization strategy. And then sometimes that optimization strategy you have the trade off. It's like how VCS work. Right. You have 99% success rate.

No, 99% failure rate.

Like 99% failure rate, 1% success. But it turns out that usually works much better than something that's more spread out.

Yeah. There was a very interesting similar article in the New Yorker a couple weeks ago about Silicon Valley and whether or not the VCs had pushed fundraising too far, if they were putting too much money behind bad companies. And the primary example that they gave was WeWork and kind of the funding rounds that were raised through SoftBank. It was very interesting because the article, it tries to contrast SoftBank's approach of being very founder friendly and hands off and just giving a ton of money and letting them run with John Doar's approach. Is it John Doar? No, it was Tom Perkins approach. At Kleiner Perkins. At Kleiner Perkins, I think Tom Perkins took a board seat at, at Genentech very early on and was working in the office one day a week. And he really helped them build out their accounting and their financial strategy. He was really involved in the company and it was a wild success and it made his career and it made Kleiner Perkins a massive name in venture capital. And he's a legend and his reputation is basically untouchable. So the New Yorker article keeps comparing the two and saying, look, the Tom Perkins approach works really well. The VCs need to get in there, roll their sleeves. They can't do this founder friendly stuff because when you do the founder friendly stuff, you get a wework and things go really poorly. But what they leave out is that I did some digging and in that first Kleiner Perkins fund that, that they made a billion dollars or more off of the Genentech investment, I think it might have been like 60 billion or something. It was insane. In that first fund, they also put money into a company that Was called the Snow Job. That was. Which is a terrible name. And they made a. And they made a snowmobile motorcycle conversion kit so you could take your motorcycle and convert it into a snowmobile motorcycle and ride it around in the snow. And it was the most frivolous company. And then the oil embargo hit and gas prices spiked, so they went out of business and it was just a terrible flop. And if the press had been on Tom Perkins back then, everyone would have been saying, like, this makes Juicero look rational or whatever. Like, the Snow Job's the worst investment. Like, he sucks. That's terrible. But lo and behold, like survivorship bias, like we, we remember Genentech and synthetic insulin. No one remembers the Snow Job. So Tom Perkins is a legend. And I agree with that. He is a great VC. But great VCs have both wins and losses. This happens. That's the name of the game. So you can't look at someone's losses and say they're bad. You have to look at the whole picture. And that's what's. So it can be very frustrating because it's so. It's very satisfying to cherry pick and say, oh, you funded this company that did something bad. Like you are bad. But that's just not how the industry works really. And that's not how the model works. So it's just very funny to go back and when you dig into it, it's, oh yeah, like even the most legendary fund ever, the one that brought us synthetic insulin and Genentech, they made a ridiculous bet on a snowmobile motorcycle that went straight into the ground and made no money. Yeah, it's just a funny. It's a funny world.

Yeah. And I think there's some deep connection between that and the exploitation, exploration trade off. Maybe I should do like a primer on that just to make sure everyone's on the same page. So essentially, let's say you're engineering this, like, reinforcement learning agent, trying to beat this game. Essentially you have to make this trade off of exploiting, which just means taking all the data points that you currently have and making the best decision possible versus exploring, which is just assuming that you don't have all the data points in the world and you have to explore to get new data points. And that might unlock more rewards. And the most optimal strategy has to be some sort of combination of both, because you might be going down one path, it might currently be working, but that could be like a local minimum. If you just imagine there's eight different doors you want to Find the optimal balance between when do you want to explore or when you want to exploit. And if you keep exploring, you're just running in circles. But if you keep exploiting, you're just always going the same path. And by definition you're not exploring the whole solution space, which likely there's a better path. And there has a lot been a lot of interesting algorithms. The whole area of AI has just been how can we find the most optimal optimization problem given, you know, limited data points.

Yeah. And best example that I know of the Explore exploit like problem is when you move to a new city and you're trying to find delicious food, you will naturally do this for the first couple months that you're living there, you'll be trying new restaurants every week that you go out, and then six months to a year later, you'll just go to your five favorite restaurants and you'll very rarely branch out. That is Explore exploit in practice. And everyone does this, do it, we do it more in certain areas, but that's one that I think everyone can wrap their head around very easily. But yeah, the Explore Exploit model is a great mental model. And I think Chris Dixon, who is our board partner at Soylent, from Andreessen Horowitz, he, he has a great blog post about like climbing the wrong hill, suggesting that like, that like in there's different career paths that you can choose and each one of those is like a hill. And so he was reflecting on the fact that he saw a lot of his friends go into finance and start climbing the finance hill, which is obviously very steep, you have to work very hard. But the entrepreneurship hill is actually much, much higher. And if you look at like the world's most successful billionaires, like the majority of them are entrepreneurs, not finance people. And that is because. And so his argument was like, you need to explore multiple hills, you need to make sure you're climbing the right hill and then start grinding on it like you're still going to be climbing either way. But you don't want to get to the top of your local hill, find the local maxima, and then look around and realize that, oh, I would have preferred to be doing something else. So yeah, that's a great mental model. There's. And I see that blog post like once a month on Twitter, like everyone's posting it all the time because it's great, it's accurate.

Yeah, yeah. The hard question then is how can you make that optimal trade off? Right. Even just in AI, we have all these different algorithms. Like one actually I was looking into recently. It's called ant colony optimization. I find it really interesting and it describes a lot of natural processes as well. But how it works is how ants find food. And you can see food as the thing that you want to optimize. How they find it is they go down this trail and they leave a smell on their path. So that the shorter distance to that food, the more stronger that smell is until they go there and come back.

Sure.

That also incentivizes more ants, ants to go in the same.

Oh, interesting. Oh yeah, that makes sense.

Yeah. So basically what then happens is you have a piece of food, ant leaves the smell, and then if the food is really close. Okay. It'll get more ants there. If the food is further, more ants are more likely to go down this path and that's exploiting. But also there's a probability of going down. Random hats as well.

Sure.

And that's like the exploration.

Yep.

And then, yeah, as the food gets depleted slowly more ants come here and gets like, stops going there and starts exploring again. And I think that can explain a lot of exploitation and exploration that exists in the world. Maybe even restaurants as well.

You're talking restaurants? Yeah. There's a great book about this called Algorithms to Live by Brian Christian and Tom Griffiths. It goes through, I think, six to 10 different algorithms and they're computer algorithms, but then they explain how to implement those in your life or how they might. So there's. There's one called like the Secretary problem, which probably has a better name nowadays, but it's the same idea of explore, exploit and how that can be used in hiring or looking for a house. There's a rule of thumb when you're looking for a house that you should like, look, you should passively look at, I think 10 or 20 houses, something like that. And then you should buy the next house that's better than all the ones that you've seen previously. And it's a great book. I highly recommend it. There's an audiobook too, that you can just cruise to. It's not code. It's. It's very much like. Like a nonfiction. Just like easy read. But great book. But yeah, that kind of takes me into like my career. I was definitely in that world of climbing the wrong hill. I studied economics in college and was on that finance track. Then started getting interested in programming and software development. And I didn't really have any job opportunities for programming because I didn't have any credentials. I had only taken one computer science class in college and was Otherwise, just self taught. So starting a company was really the only way that I could create enough space and time for myself to go and really get good at coding. So it was very much a. I definitely had some interest in entrepreneurship, but it was also just a great way to go and explore something new that I knew that I would have no way of doing if I went and got a job.

What was that decision making process? Was it just one day, hey, I want to try to learn a code? Or was it a rigorous decision making?

I mean it was all over the place. Like when I was a kid I learned basic, like HTML in order to put up like funny websites. Then I learned a little bit of like Flash animation and really basic motion graphics stuff for my Counter Strike Clan because we wanted to have a cool introduction on our website that was very trendy at the time, this is like 1999 or something probably. And then built a website for the team and was always hacking on things like very loosely installing like checks and cheats for different games and like learning how, oh, you have to debug this because you downloaded off of a torrent and it doesn't work or whatever, stuff like that. And then in college, like a lot of economics is quantitative, so econometrics uses a lot of statistical tools. So you're using Excel and then pretty soon you're using R or Stata or matlab. And then Python starts showing up when you Google things like, how do I run this regression faster on more data? Oh, use Python or use Julia. So picked up Learn Python the Hard Way by Zed Shah. Loved that book. Did that book like really fast and really enjoyed that. And then, and then started like building web apps. And on the financial side, like I was interning in finance throughout college and I literally went from the biggest possible application of finance working in life insurance. And the most, very boring to me. And then slowly went down to, okay, now I'm working at a hedge fund that trades big equities. And then I'm working at a VC firm that does medium sized deals. And then I worked at this seed fund that was super tiny and only had a couple million dollars under management. Really small. And at that point I started seeing entrepreneurs in the earliest stages and actually meeting the people who are going to go start companies and they would come in and they would give us this pitch about this technology that we were using and I would go learn about the technology and then ask them more questions like, oh, are you thinking about using this implementation of Hadoop or this implementation since your Deck says big data, which was a big trend back then. And they would freeze up sometimes. And I realized that there was a lot of fake it till you make it. A lot of people that were. That were not really. They didn't have all the answers. And that was okay. They still got funded. And I'm pretty sure some of them are successful. But. But I realized that, okay, if this person who walked in here and just got a check for $50,000 or $100,000 to start this company is going, it has a plan of figuring out the software side of this in six months, I could probably do the same thing. So at that point I started thinking a lot more seriously. Started going to more like startup meetups, startup weekends, hackathons, these types of things, like actually getting experience, like building things. Quickly built a bunch of things and then put the first company together with my co founder. We moved out to Silicon Valley, got funded by this incubator that was now part of YC, but it wasn't at the time. Yeah, imagine K12, which was focused on education technology. And then that company worked on that for three months, four months, five months. Dragged on past demo day, but did not get any more investment. Was not very good or polished. Only a few hundred users, never got it anywhere. And so that kind of flopped. And out of the ashes of that, we joined up with another team and Soylent kind of came out of that, which is the story I've told a bunch of times. But essentially the short version is we're running out of money. We're trying to build software, we look at our burn rate, we're spending money on food. There's no real good options to get cheap food in San Francisco. Can't grow it in an apartment, can't go out to eat because every restaurant's expensive. And you can't go to Costco because we don't have a car. So there's no way to get cheap food. Order on Alibaba and Amazon and make it yourself. So that's where the Soylent idea came from.

Can you talk more about your first edtech start? You also talk about it dragging on. I find that interesting.

It's very hard to tell, like, when you're failing because we had no money. So that's, I guess technically we were bankrupt, but we weren't like legally declaring bankruptcy. We just had no money in any of the accounts. And so the money's been dwindling down. We don't really have any more money. We're stuck and we're just not Getting any traction. Like there's no curve, but it's unclear. It's if we change one more thing, could we get better? So then at a certain point we set it to the side, started working on other things. There were also like a number of pivots in there and stuff, like the very early stages, like we're iterating really quickly. But yeah, post demo day. Post demo day, meet some investors, pitch those investors over the next week or two and then it becomes pretty clear, like we don't have any more leads. There's no one else in our funnel. We can go back to working, but we're out of money. We got to pack it up. And fortunately the other team that we were living with was in YC proper and had more money, so they were able to give us some food and let me like live there for free for a couple months and just kind of COVID the real basics. And that allowed us to keep working on stuff. We were programming other apps and websites, but none of those really went anywhere until Soylent took off. Yeah.

So about Soylent. So you had that idea and what was the entire process like? Because Soylent is like a moonshot. Right. Was it from the beginning you thought of getting rid of food? Is that part of the vision?

Yeah. So Rob's a computer scientist, the CEO, but David, our other co founder, has training. He did his undergrad in biology and he was on the track to do PhD in biology. I think the two of them got together and through a lot of late night chats shared what they knew. Rob got really interested in biology, but from a computer science perspective started seeing it as a computer science problem. And out of that he had this idea to start researching biology. And he has a unique skill for just he will try anything. Like he's endlessly ambitious would be the term. It's hard to put a finger on it, but it doesn't matter that he didn't study biology. He was going to go and learn it and he didn't need to run a trial because he was going to test it on himself. He is, he was that dedicated. So he just started whipping up these prototypes and testing them. And I was super confused at the time because I didn't think that there was a path, I didn't think that had anything to do with entrepreneurship. Like I saw starting a business and entrepreneurship is like you build a website, right? Not like you make a physical thing. I didn't really see that there was like a pro, a way to turn that into a company. But I thought it was a cool experiment and I liked the name that he had come up with it for it because I thought Soylent was really funny and interesting. So he, but then once he started blogging about it and he just like, it was just one thing after another. He'd have 10,000 people sign up on his email list. Checked his Twitter one day and he had 10,000 followers. Like he was just like, he was just like blowing up. And then there were like press people coming around and like wanting to talk to him. No reporter had any written, have ever had ever written anything about us. So it was just like, okay, this is clearly a thing. So then it took us like a couple weeks to like maybe a few months to figure out like who was going to be on the team, who was going to fit where, what everyone was going to do. I was really excited to just do the E commerce and technology portion. So I came on as like the CTO and we formalized things and then launched the crowdfunding campaign which did really well. Yeah, it was a wild experience.

Yeah. That also connects with the whole exploration thing. I think successful entrepreneurs in general tends to have this bias towards exploration.

Yeah, yeah, totally, totally.

Yeah. I think that's what we need more in the world. People tend to just go by their own data points.

Yes and no. Like we need both. Right. That's the definition of the exploit problem. Like we need people to explore, we need people to exploit. And as a society we need people to go and discover new problems, but we also need people to be ready to work and refine those problems and help and be the expert that can actually solve some sort of a very fundamental problem. There's also the same trade off in within sub problems that aren't entrepreneurial in nature but are fundamental problems. If you look at some of the AI research being done at Google by Jeff Dean and stuff. Like, although there's all this whole mythology around Jeff Dean and he's like this insane programmer, but he's not an entrepreneur. But I have no doubt that he's a genius and has the ability to reason in a very unique way about really abstract problems that most people could never solve. The fact that he's been at Google for years, it doesn't really change like who he is in my mind. I don't know.

Yeah.

But yeah, you should look up some Jeff Dean memes. They're, they're hilarious. For a while he was like, I don't know if you ever heard of like the Chuck Norris meme. This is like a very old meme Whereas it was like making fun of Chuck of how tough Chuck Norris is. Like when Chuck Norris jumps in the ocean, he doesn't get wet. The ocean gets Chuck Norris. Stuff like that. And so people would make the all these jokes about Jeff Dean being like the best programmer ever. Like he doesn't write code. Like code writes Jeff Dean.

I've never seen this. Oh, he's the head of AI at Google.

Yeah. Yeah. But he wasn't doing AI 10 years ago. I think he was like leading like software teams and like doing some like more like low level stuff or something. But yeah, I mean he's strong. Like probably like one of the best programmers of our time. Yeah.

Regarding exploring sub problems, another parallel I think that's interesting is how a lot of discoveries tends to be the intersection of multiple fields.

Sure.

Like you also talk about swelling. Right. It's seeing biology from a computer science perspective. Do you think that's a really good strategy to be able to solve problems? Do you think that people should be actively trying to put their toes in all these different areas and then trying to connect ideas? That's like the whole generalist versus the specialist debate.

Yeah, I think like curiosity is a very fundamentally entrepreneurial trait. So if you're super curious, you will probably go down rabbit holes and become like a pseudo expert in multiple things. And so you might find someone who's they're a programmer, but before they were a programmer they worked as a pizza delivery person and then all of a sudden they build a delivery app. So I think curious people do well as entrepreneurs. I think it's hard to force that. If you weren't curious and then you were just like, I want to follow some sort of pattern. Like I think that would be very hard to force yourself to do. It might be possible. I would just say if you do find yourself interested in a different topic, try and get over like that imposter complex. Like I, for my latest video on Moderna, I don't have a biology background. I don't think I took biology in high school and I wasn't even in the advanced class. Like I don't have an, I don't have experience there, but I just did enough research and kept it light on the science and was able to defer to the experts. And I actually clipped some of Moderna's videos so that they would explain it in their words and I would not get it wrong. And then I talk mostly about the business case, which is something that I do know about. And I got a lot of really positive Feedback from that on that video, including like entrepreneurs who had started multimillion dollar companies in the biotech space. So I think there's, and I think that I was a little bit worried that like that would be an ov. I would be overextending myself and I should stick to what I know. But I got over that hump and I think that was great. And I think that the best entrepreneurs are curious but also like confident that they can learn enough and they maintain a very high, a very fine tune level of understanding of what they understand and what they don't. So it's not enough to just know 80% of a topic, you also have to know which 20% you don't know generally and the type of people that can solve that for you and can reliably do that. Great examples like when you're building a company, I'm sure you'll go through this like eventually you will be solving computer science problems that you haven't seen before. Like you might be dealing with like security or hackers or cryptography or something that some topic that you just don't have the time anymore. But you need to be able to read the general stuff, get the basics, understand enough to evaluate whether or not someone is an expert and then employ them and work with them effectively. You don't need to become the expert in everything.

Yeah, there's a lot of good mental models in there. And you talk about curiosity and the imposter complex which is synonymous with this bias towards discomfort.

Yeah.

And that is pretty much the determining factors of exploration.

Right.

How curious someone is and how like how much of bias they have towards allowing discomfort and seeking discomfort about the finding what you don't know. I find that interesting. How do you go about finding things that you don't know when you're dealing with unknowns?

If you're in a topic, it's very easy. It's the textbook that has been sitting on your bookshelf for a month that you're not reading. This happens to me all the time where I will be diving into some sort of topic and I'll collect all the resources because that's the easy part. I'll organize all the stuff and for me usually that's like any advanced mathematics. So like I love programming, I'm okay at math, but as soon as you start talking about things beyond linear algebra, it's really hard for me to focus on that. So I know that if we're doing, if we're building something really complex in computer science, you're probably going to need someone who knows their math really well. And I think it should be pretty apparent where like the frontier is because you can traverse the topic pretty quickly. You should be able to, you should get really good at traversing topics. So if you want to learn about computer science, you start with the basic overviews. Pretty soon you're figuring out, okay, who actually built this stuff, who, who are the core. Who's Guido van Rossum, the creator of Python, for example. Okay. And then you traverse backwards to, okay, who's doing like the fundamental research in this stuff? Oh, there's some researchers in top computer science departments. Oh, you're looking at like Stanford, Waterloo, mit. Okay, who are these names? These people are probably the experts. Those papers that they're writing in the last couple of years, that's probably going to be the frontier. Who's citing the papers, who's implementing them on GitHub, who's doing that stuff like that, that's where the frontier is. And, and if you don't have any ability to really sit down and enjoy that stuff, then you probably know that that's something that you're, it's not in your wheelhouse. But that's okay because you know that, okay, that's a skill set that I don't have and, and probably don't even really have the ability to, you know, gain like without a lot of extra work.

Have you looked at your research? Is that something that's genetic?

No, Think like when you get to that, when you get to that with a lot of topics, you're looking at, you're looking at an 8020 rule where you're like, you can get a feel for a lot of the topic for a little bit of investment, but in order to become an expert you have to spend way more time on something. So I think that it's just a factor of like true experts spend that time. There's probably some genetic component I don't really know, but I think anyone could probably develop the skill to traverse a topic and figure out where the, where the cutting edge is. Yeah, I don't think that, I don't think that's like a unique skill. I think that's something that just takes practice. Wikipedia, a couple academic journals, the top books on the topic, finding out.

I was debating with someone about whether curiosity and these abilities are genetic and the other person I was talking to, he thinks that it is genetic. I think that almost that itself is like a self fulfilling prophecy. I don't exactly know the science behind this specific case, but if you think about that study about willpower, for example. It's like people who think they have infinite willpower have way more willpower than the people that don't believe that. And I think that applies to a lot of other things as well.

Yeah, that sounds intuitively correct. I feel like a lot of that, like, genetic stuff winds up in like pop science where it's one study and then someone writes a book about it and they have some good anecdotes to support it so it gets picked up and parroted, but who knows if it's actually right? Like, how would you even quantify, like, curiosity effectively? Like, it's going to be extremely hard to quantify on any large scale. And so it's. And also what are we going to do with that information? Anytime someone says something like, oh, is this genetic or not? You can jump straight to, okay, what are you going to. What are you planning? What are you going to do if it is genetic? Are you going to treat people differently? Because that's not cool. No, let's just encourage curiosity and lots of people and reward it when it shows up and we'll be good. Like, everyone will be happy then. Like, I. I'm skeptical of. I feel like we definitely see these cycles where it's like, there's one paper wasn't really peer reviewed, but it made itself. It's made its way into a New York Times bestseller nonfiction science book that wasn't very rigorously reviewed. Now everyone's talking about it and now we take it as fact when we don't really have that evidence. But I don't know. In general, curiosity is good and fun and if you're a curious person, go and enjoy that. Don't sit in a cubicle if you can afford to. Yeah.

And also think about beliefs as what's more useful to believe. Personally, I think anything is achievable if you just believe in it. But I think that's the most useful thing to believe because 90% of the time, let's say that it is true or whatever probability you want to assign. If it is true and you believe it because of the placebo effect, it will increase the probability your mind actually rewiring yourselves towards whatever outcome it is. Then let's say that if you are wrong about it, it doesn't hurt to believe in that. And at the end day, you're dealing with uncertainty. So if you just think about the raw utility of these, like which one to believe. I think the higher utility one is the one where you push your mind as Far as possible, generally.

Absolutely true. But in specific cases, there are people that just have overwhelming disadvantages, socioeconomic disadvantages, any sort of disadvantage, structural, like racism, sexism. Like, there are things that there are ceilings and walls and stuff. So sometimes this goes back to the hustle porn stuff. If you're telling someone like, oh, yeah, you can do anything, and then it's okay. What if they have never been able to afford an education and they can't afford an education and so they just wind up getting taken advantage of. Did you really wind up helping them? I think the outcome matters to some extent, and you need to look at that. That's not to say that we shouldn't encourage curiosity or entrepreneurship in all sorts of people. That would be like, the worst. But I just think that, like, when you're looking at these types of things you like, you do need to be a little bit aware of. I want to avoid a situation where people say, oh, like, all you need to do, the only thing that you need to be successful is like, change your mindset. Because all of a sudden it's, you don't need roads, just change your mindset. You don't need a fire department. Change your mindset. Learn to be a firefighter. And then you'll put your own house out when it's on fire. And so I worry about that stuff going too far. Sometimes when I hear that, I'm like, okay, I see where this is going. And I want to be. I want to be careful. But yes, in general, if you're young, you're educated, get online, learn more things, explore, have a good time with it. Like, it's amazing.

Yeah, I think you can do almost everything, but I think the way you convince someone to do it isn't telling them directly they can do anything.

What's nice is that it feels like the floor for being in the group. In the group that can do anything if they put their mind to it, is lowering. With the Internet, with open access to information, open access to educational tools, like years ago, it would be much, much harder to start an enterprise. And the Horatio Algor myth, I don't know if you're familiar with that, but it's the classic idea of pulling yourself up by your bootstraps. It's this idea of the person who starts from nothing and then becomes wildly successful. It's never been fully true, but it feels like it might be getting a little bit more tr. Hopefully at the same time, we have wealth inequality growing. So maybe even though we're moving the goalpost down a Little bit and moving that hurdle rate so that more people can jump in and increase their being, their wellbeing. Maybe more people are being pushed down at the same time, which isn't great. I don't know. There's a lot of things to consider with these things. They're all societally important.

Yeah, I think when we think about making good decisions is first defining this utility function and then maximizing that. And just for clarify, because a lot of people, when they think about utility functions like utilitarianism, like maximizing, let's say money or maximizing amount of people, you also have to account for like emotional factors as well. When I say utility function is not just money or like people, it's also emotions or let's say inequality, et cetera. But that's the hard part. It's like defining that utility function and then it's like the exploitation versus exploration trade off of what type of strategies you can use to use your limited data points that you have in the world and make the optimal decision to optimize for whatever utility function you have. That's the two fundamental things in making the most rational decision. Is there any other thing that plays into making the right decision rather than taking utility function and then figuring out the right strategy?

I mean, yes, I guess that's correct. But it doesn't get you very far in the real world because you can pick your own utility function and you can have your own opinion. But because we, because there are so many other actors in the global political space or even the, you know, U.S. political dimension, so many different voters and politicians, it becomes a game of game theory for what you like. It's not just what you think is best. It's like what you can actually get done within the, the framework of debate with everyone else or who you can win over. Yeah, it's very interesting. There's only been a limited amount of research done on the effects of inequality. I think obviously there's some sort of, I don't know, like psychological, like negative output to higher inequality. But people tend to focus on like absolute gains. Imagine magical being comes down and says, I will give everyone 100 extra years on their life, but only if I'm giving one person like 100,000 years extra on their life. Like no one, everyone is better off in that scenario. Like there's Pareto improvement is what you call it. Like everyone's better off but. But it would increase inequality. But no one would say no to that, I think. But all else equal, I guess people would want to distribute that extra 100,000 years, equally amongst everyone. It's. It's complex. I feel like there are both, like, they're like the median as well as the standard deviation should be considered in these things. It's interesting, but we don't have that problem in America right now. Like, in America, like, people's lives are getting worse. Right. So there's no option, there's no policy right now that is a Pareto improvement where everyone gets better, some people get even more improved. It's like, people are. People's lives are getting worse. What can we do to stop that? And how painful is that going to be for the people who are at the top, essentially? But it's. Yeah, it's a very complex problem and I don't know if anyone has, like, a strong answer. Yeah, yeah.

The game theory connects with that really complicated scenario because you can run into the prisoner's dilemma where you're incentivized to essentially, like, fuck someone over. How do you even decide when you have a scenario?

Yeah, I suspect that there are a number of political actors that are using that to their advantage, essentially. They will tell you a story of a utopia which is maybe unachievable, but secretly they are satisfied with the intermediate state that we would have to go through and probably ultimately stagnate in in order to reach that utopia. So I will sing you a song about how amazing complete deregulation of everything will be, even though we might never get there, or complete authoritarian control of everything. And I'll tell you how amazing that would be if it was absolute. But in order to get there, oh, we got to go through the. Because this is the American political system, like, it's not just going to transform overnight. We have to go through these iterative processes. And will you be happy in a state that is just one notch towards that utopia? Because it's not going to be that utopia on either side. It's just going to be one notch closer. And so I think that's what you have to think of when you hear someone talking about how this will. How X ideology will lead to a utopia, is will you be satisfied with living your life on the road to that utopia? Because you'll probably never get there. And is this person secretly optimizing for some pit stop on that road? And they have no. And they have no aspirations of ever getting to that utopia. They only want to get you on the road to that utopia so they can pull you over on the side of the road and keep you at that pit stop forever, which is maybe Slightly more regulation or slightly less regulation, but they, but they actually have no ambitions of doing the entirely regulated or entirely deregulated political system or whatever. I don't know.

Yeah, and that explains a lot, people going to the extremes in order to convey an idea because that is the most optimal.

Yeah, it's also just a much. It's much more aspirational. It's much easier to. It. It allows you to like hand wave a lot of practical problems out of the way. Oh, that wouldn't happen because we're in an absolute version of this system. But yeah, there are political realities, which is why everyone likes following like political pundits and they don't really like following like politicians. There aren't that many people that are like reading, watching C Span, but everyone's watching Fox News and CNN and msnbc.

Then you got the people that completely use that to disagree with them as well. Now I'm just realizing that it's almost completely missing that point. So specifically, example I'm thinking about, for example, Democrats would say let's move towards clean energy. And then Steven Crowder would say that's completely unrealistic and completely using that to essentially discredit that entire argument. I think there's more depth than that and I find that to be interesting model. It's seeing that as a system, not just, okay, it's like logical, this is irrational, this is rational, but rather it's every single thing that you say has some sort of impact on people's psychology in the world or people's beliefs. And sometimes to make that sort of impact, the impact is not direct towards what you say.

Right.

There's like subconscious processes, these psychological factors. And when you're pointing something out as irrational, sometimes that's actually not a useful thing to do all the time. Because let's say a case like this where you have to push something to the extreme in order to get people to slowly shift, shift their mindset. Sure, that might be, quote unquote irrational, but if you do that, that completely cancels out the positive effects of that. Right. All humans actually are irrational. And it's about figuring out how you can take this current system of irrational humans and optimizing everyone towards to optimize some sort of goal.

Yeah, it's interesting because I heard a different conservative take on the clean energy thing, which was that while this particular pundit was not a fan of making a jarring change to the economy that would shift everything to like solar and wind, which is the gold standard, clean energy, they were fine with the idea of moving along that path as long as there was like a pit stop in like the nuclear world, because they really liked nuclear energy as clean energy, but also something that would get us towards clean energy. And I thought that was interesting because I see on the left, I see a lot of clean energy advocacy, obviously for solar and wind, but I don't see as much for nuclear power because it's still controversial. But then I do see some liberal left wing people saying, no, we got to do nuclear power as well and get there. It's interesting. I feel like, yeah, it's, I feel like that maybe there's a world where if the, if the left gives ground on the nuclear stuff, then they can find more of a consensus and say, look, we're all aiming the same direction towards more clean energy sources, but we're not going to be an absolutist about that. Maybe that would be more politically viable. I don't really know. I know the Green New Deal has gone through a number of iterations and, and every iteration seems to be increasingly hated by the right. Who knows how successful that will be, especially if the Democrats don't control the Senate. It seems like it would be unlikely to pass. But obviously we need to solve our energy problem. One of the particularly difficult things with the energy problem was a couple of years ago, America was not energy independent. So energy independence was a huge justification that both sides could rally around. We were dependent on foreign oil, we were dependent on the Middle east specifically. And that made us very vulnerable because OPEC could manipulate the price of oil or there could be wars in the Middle east. All these things were bad. But now we have developed fracking technology and tapped extra oil reserves. So I believe we're a net exporter of energy now, or at least we're like way above where we were before. So that argument, which was, which spoke very strongly to the right wing of the American populace who want to be independent, who want to be secure, and they care about national defense. These are all issues. So you could go to them and say, hey, look, this isn't necessarily about putting jobs, like ruining oil and gas jobs. It's really just about like our national security. We need windmills and solar panels and nuclear reactors and we need to be able to generate electricity at home. So if something crazy happens in another part of the world, we aren't paying $8 a gallon for gas and our economy comes grinding to a halt. But because of fracking, that hasn't really been as much of an issue. So I think it'll be a hard. I think it's going to be an extremely hard battle. Fortunately, all the numbers in solar are going in the right direction. It's becoming more and more economically viable. And I think that within the next decade, solar is going to be fully economically viable, sans any government sponsorships. Hopefully it's a tidal wave that we can't put that no one can push against because it's a good technology and everyone likes it. And Obviously Tesla and SolarCity and all that stuff, it's just very trendy on either side of the aisle. There's really no political bent to that technology. So hopefully it just becomes, like, popular and people like it.

One mental model I'm seeing is the way you're viewing politics is essentially, how can you, like, speak to the problem that people have? And I think we're missing a lot of that in the world where it's always like, either people like Ben Shapiro, Simon Crowley, just like, this is irrational.

Yeah, they're both, like, very aggressive and that's definitely what gets clicks. And that's their business model. They're not optimizing for passing legislation, they're optimizing for entertainment. And what they do is clearly very entertaining and working very well. But, yeah, obviously, like, they're not. Not getting legislation passed, but that's not their goal, so that's understandable. Yeah, it's interesting. There isn't a lot. There isn't that much value to thinking that way. It's academically interesting to think about what would actually work. And in some ways it can be. Politics is the game of compromise. So in some ways, if you're trying to predict what will actually happen in the world and you care about the outcome more than the ideal, that could be a useful mental model. I really like the work of Ian Bremmer at the Gzero Media Group, because they're primarily a political research institution that sells information to hedge funds and financial brokers and they also put out public stuff. But essentially their goal is not to tell you that a political person did something bad and you should be angry. Their goal is to tell you what is likely to happen, what just happened, and the implications of what just happened, so that you can make shrewd investments, but you can also just use it to steer your life. If there's a particular social issue that you care about and will have a dramatic effect on your life, maybe pertaining to your sexual orientation or your immigration status, and you follow the law and you see that politically something is going to happen that will either be very good for you or very bad for you in a certain area or a certain region, it can actually make sense to you if you have accurate information. It can be very useful to look at politics to try and distance yourself emotionally from politics and try and understand what is likely to happen and try and just get a view of the way politics are likely to change in reality because as an entrepreneur there might be a change coming that could affect your business. There's a lot of regulatory arbitrage startups, but there's also just if you're building a YouTube related company, if, if we had good in good data that suggested that, that the government was going to shut down YouTube, you might want to reconsider your strategy. And whether or not you feel like that's a good thing to have happen or not, if you think it's the worst thing ever, if you, if you're not clear headed about the likelihood of that actually happening, you could be in for a world of hurt. So I try and distance myself from the emotional aspects. That being said, I'm in a very lucky position where there, there aren't a lot of political actors that are coming for me personally. Whereas if I was in a questionable immigration status situation, like obviously it'd be much more close to home. It'd be much harder to distance yourself emotionally. So I want to be empathetic to those people. But yeah, in general I think that it can be a lot more valuable to, to try and understand what's actually going to happen in politics and what's likely to get passed and what actually got passed and the way things, the way the world actually is as opposed to only thinking about the way, the way the world should be in some ideal, some ideology.

Yeah. And the reason I brought up the mental model earlier about seeing the world as it is and then figuring out what's the best way to change people's psychology. The reason I bring that up is I think that's a perfect segue into what we talked about last time about Lucy. And for context, Lucy is John Kogan's second venture. It raised over $10 million and was also part of Y Combinator. What it was a clean nicotine alternative. And what's interesting is it didn't market itself as many of the alternative products like hey, you should quit smoking and instead use our product because it's better for your health. Instead it just made it a net better tasting product.

Exactly. Yeah. I tend to think about this dichotomy between self help and what I call stealth help. So stealth help is all about helping someone in a stealthy manner. So helping them without telling them that you're going to help them. I think Tesla is a really great example of this. It's an electric car. Obviously it helps the environment. But the first electric cars had the same promise of helping the environment. They were zero emissions, but they didn't sell well because they were ugly, they were expensive and they didn't work well as cars. They weren't fast and they couldn't get you going very far. But Tesla came out with a car that has the same range as a normal car. It's faster than a normal car, it looks great, has all the best features and they've been really successful. People don't buy Teslas because emotionally they want to make an impact on the environment. They buy Teslas because they're cool cars and they want to have fun driving them and it's a status symbol. And then, oh, by the way, they're helping the environment. I think it's very important to never make the mistake of thinking that having a high quality mission or being like an impact driven business will automatically generate customers for you. You have to deliver a product that's better than the competition, regardless of the mission. And then the mission becomes a bonus. It becomes something that people feel good about and it's a reason why they share and it definitely becomes additive. There are certainly people that stretch and spend a little bit extra money on the Tesla because they want to have that impact. But I don't think that Tesla sales are driven 90% by the clean energy movement or the zero emission vehicle. I think it's probably 90% driven just by the fact that they're the number one rated car by car and driver or they're just the other great features that people want. So I would say to entrepreneurs like never sacrifice in the product experience. Never lose sight of the fact that you're competing with other products and that having a, like a virtuous mission is not enough. That doesn't mean that you can have that. You need to sacrifice your desire to have an impact. Because if you get to scale with something that has a small impact, you will actually wind up having a much bigger impact. Obviously Tesla has taken way more gasoline cars off the road than that first EV1. That was a complete flop. I think 10 celebrities had those and that was it. No one bought them. So they had zero impact. Like they didn't do anything for the environment. Tesla's obviously done a lot more. I think they, it, it's okay, but it requires swallowing your pride because it means that you're not going to be on the TED talk stage every day talking about what an amazing thing you're doing. You're going to be on the stage talking about how fast your car drives or whatever, but you'll still be able to sleep at night because you're having a good impact.

It connects a lot with batched optimization and AI. Essentially the idea is non batch optimization is you take all this data points and you just do everything in one step. Okay, this is the most optimal you get right there. Whereas batch optimization is you take limited data points, you make small strides and then you just eventually iterate to the right optimal solution and that is connects directly with the whole exploitation trade off once again that's usually a much better strategy. People tend to just take their existing data points. Okay, the world needs to have clean energy. We know this. Sure that's true. And the exploitation of approach could be like they're just telling everyone like you need to use this car. But I think the batch optimization approach is like analogy for kind of what Tesla has taken. You start with what makes sense now. Making a nice looking car that the rich people will definitely buy regardless of whether it helps environment or not. And it's slowly iterating. But I find the self help and stealth help idea interesting. Did you make that term up?

Yes. Yeah, I think so. Like me and my co founders were talking about it for a while. It was like during reflections on Soylent, like the idea that like we were never marketing it as a diet product. We never told people like hey, use this to lose weight. That's not why we created it. We created it because we wanted something that was convenient and affordable mostly. And, and then we started getting you know, emails from people like oh wow, I've lost 60 pounds using this product. And I think that we wound up having probably a bigger, a bigger impact on those like kind of traditional health metrics and of those like aesthetic things because we weren't so in your face about it. If we had gone straight to people and said you're overweight, you need to lose weight, we have a miracle cure. Come and buy this product. Like no one would have bought it, people wouldn't have enjoyed it and we would have sacrificed on the like the important factors of like cost and convenience, all that stuff. So yeah, it's, we shoved the, that kind of like mission and we never really branded ourselves like a mission driven company. Although like I think we had a positive impact on a lot of people's lives. You could say that yeah, it's interesting like the mission. There's a whole industry around like mission driven startups. It's interesting because what startup doesn't have like a mission? And there are a lot of companies that like wind up having big impacts. Look at like Microsoft, right? So many people have learned more effectively or like grown businesses because of like Windows. And that was never like a mission company that was trying to help people with any of that. It was like they just built a cool technology and then people use it for everything. So I'm not, I feel like people, I feel like people like sell themselves short when they over optimize for. I need to have an impact. Like you build something huge like you could have a really positive impact. Careful. Obviously there are definitely drawbacks some of the bigger companies and you could have a really bad effect on the environment if you're some huge company that's not taking that seriously.

That's another ventilator. Overfitting versus underfitting or overfitting. It's choosing like having this mission and then coming up with a strategy to directly do that.

Yeah, a lot of the mission stuff works really well as marketing. There's a lot of companies that use it really effectively with a one on one model where you buy one of their products and they donate one or it just helps them stand out from the competition because there's no one who's like really charitable in the space or no one who's seen as clean or healthy or something that can totally work. It works for a lot of consumer products. But yeah, it's just something that like I think you need to be like clear eyed about because it's not like a, it's not a golden ticket. Still have to build a great product.

Yeah. So the reason I brought that up is this connects with what we talked about last time about these emergent properties. The example you gave was Code Academy versus Minecraft. So for context, basically I started my first company in the hacking space and that's how like I made my first like $3,000. And I learned to code so much faster that way than any other way I could have conceived of. And marketing like project management, etc. So with this specific example, Minecraft was actually a much better way of for me to learn a code and learn to build something than Code Academy and all these other resources is. And that's extremely interesting. It's Minecraft was never meant to learn to code. It's underfits the mission of, you know, helping people code.

Yeah.

But for my case specifically and I'm guessing nowadays a lot of other cases.

Lots of people, pardon me, like lots of people, probably like a million people have like learned to code because of Minecraft or probably hundreds of Thousands, what like 100 million monthly actives, A lot of young kids. The right time. I learned, I basically learned to code because of counter strike, like modding and like level editing and like installing hacking software, building websites around it. It was the motivation. It's not to say that you don't need like the tools and the resources. Like probably wound up on a lot of like Stack Overflow or whatever the equivalent was at the time. You probably wound up on Stack Overflow a lot. And GitHub. Yeah, you need those. But it's just interesting that something that wasn't explicitly designed with that in mind wound up having that effect. And I think that' that's something that people should like, I don't know, think about and highlight and also just accept that it's that these emergent phenomenon are sometimes uncontrollable, sometimes they're bad. A lot of people are very upset about Facebook and misinformation and political stuff and extremism and that's a real problem and they should solve that stuff. And it seems like a really hard problem. I don't know how they're going to solve it, but they should work on that, obviously. But there's also like really cool, positive emergent things that come out of these things as well. Like with social media, obviously, like the Arab Spring is a big one that everyone cites. But there are other, obviously without social media, like we wouldn't have met and now we're having a fun conversation. There are positive externalities as well as negative externalities and you can't really look at one without the other. And you can't and you shouldn't, you know, forget about these entirely. And I think we should probably, I don't know, accept that these are going to happen, but then also be super rigorous about identifying them early and being able to like, like think ahead and deal with the really negative ones. If people are going to get hurt by something, let's build that out of the product as early as possible.

Yeah, I wonder if that can be intentionally predicted, like whether you can predict, let's say like building Minecraft with the intention of teaching about a code.

No, I think it's happened a million times. I think there's so many people that have built education games and games to, to teach people to code specifically and I don't think any of them are nearly as successful or impactful as like a Minecraft or something like that. Like it happens all the time. People are always building stuff like that. I generally think it's like a little bit. It's like too on the nose, it's too direct and I don't know, but maybe that's just education, who knows? It doesn't necessarily apply to every industry.

Do you think there's a correlation? Like, let's say you map two ends of it, like things that underfit and things that overfit.

Right.

Like full on the student mission, like Code Academy and the things that completely don't care about the mission, like Minecraft. If impact is on the Y axis.

Yeah, yeah, yeah. Is it like this impact and attempt, like how hard they attempted to do that? It's certainly, certainly seems inversely correlated in that specific case. I bet it's not the same for learning languages or something. Even within the education space. I wouldn't be surprised. I know some people that like learn Japanese because they wanted to play video games that were first edition from Japan or something. But I have a feeling like Duolingo is probably doing a pretty good job at teaching people just directly. I really don't know and I definitely don't know if it applies outside of education. I just that the education thing is interesting because I know so many people who are like, I learned to code because of, you know, something that was not designed to learn to teach me to code. Yeah.

But yeah, perhaps it's more about the motivation aspect and like educational products specifically.

Yeah, totally. With coding and specifically, like a lot of it is like at the end of it you get a product like you get a script or a piece of code that does something and if that something is cool and unique, that's gonna be a lot more motivating and really gonna be rewarding. Maybe with language it's less so because you're probably not. It's gonna take you a really long time until you're like writing like truly unique sentences that is like poetry or something novel. So I don't know. It's interesting. But yeah, with coding it's super satisfying to have. Oh, I made this product and it's. And it does something that nothing else does and it's unique. It's like the concept in chess of going off book. Like the sooner that you can get off in chess, like there's something called like the book of Moves, which is like every chess game follows like a pattern of like moves that have been done before. So if I move my pawn to e4. Like that has been done before, obviously. But at a certain point in the game, as we get more and more complex, there will be a moment where no chess game has ever been played with those exact same set of moves before. And that's what's called going off book. And that's because the number of chess games that are possible is 1 times 10 to the 1000 or something like that. It's like some huge number. So you. So there's so many possibilities that you quickly enter into entirely new territory. And with coding, it's like that as well. Like you go through some basic tutorials and then you build a script that, that allows you to hack Minecraft and very quickly you have written code that has never been written before. It's like entirely unique. Right. Like, think about the first Minecraft script that you wrote. Like no one had ever written that exact code. That's really satisfying. But if I'm going through like a language class and I'm learning like Spanish and I'm just learning the basics, like it's going to take a long time until I'm like off book and writing and creating sentences that have never been uttered before in that new language. So. So it's probably just a steeper learning curve or something like that. I don't really know. I'm just winging it.

Maybe it's. Finding new off book moves is like exploring the solution space that like most people have never explored before. So things that are designed to be underfitting the exploration of the solution space usually is much larger. There's so many possible. It usually isn't even for coding like Minecraft. There's like servers, like running game servers, which you don't need to code. But like for a small subset of people like myself and many others, they might go down that coding path.

Yeah.

So it's things that underfit tend to offer more exploration of solution space and that allows you to go off book more. Whereas something like Code Academy, which is too.

Yes, yes. Okay. Yeah, I totally understand what you're saying now and I get that. Yeah, yeah, that does make sense. Yeah, it's interesting. There's something about. Yeah. Education that like when you get to a point where you're able to create something new with that skill, at least in me, that triggers something very satisfying that it feels like a breakthrough and it feels okay. Like I was able to do this and it's not just copying a template or copying like a format. So yeah, it feels like when you're learning something new, like A lot of the, a lot of the hill is getting to that portion where you're like, okay, now I have created something that's unique and that uniqueness is extremely satisfying. Therefore now I'm willing to pour a lot more time into it. Whereas if there's a certain thing where you have to spend years and years doing things that are not unique, like the longer that curve is, the harder it's going to be to get people to self teach through that. Which is probably like why becoming a doctor requires like 10 years of education. Because you don't really want to go off book there. Like going off book is something that like only the most advanced medical people should be doing really. Otherwise it's just identify these things things really understand this like cold be the book. But yeah, I don't know.

It creates that curiosity feedback loop when.

You for sure build something new for sure. And then once you get that going, it's really easy to keep researching and.

Correlated with a lot of like ambitious people. I think most ambitious people tend to have, have that loop when they're young. Like for example with Lego. For me, it's actually one interesting thing I can show you. I built this way back, so LEGO VR headset.

Oh, that's awesome. You put a phone in there or something? Yeah.

So it actually works.

Let's see. That's amazing.

But that's it.

Oh, that's awesome. That's really cool.

Yeah, yeah. I used to be so into this stuff. I think this curiosity feedback loop seems to be one very strong factor in terms of developing these mental models.

Right.

Developing curiosity.

Yeah.

All right. To wrap it up.

Yeah, yeah.

Is there anything that you want to mention?

Just go and subscribe to my YouTube channel. It's YouTube.com John Coogan/C O O G A N I have a bunch of fun videos about like business and startups and basically put a new one out every week. So that's a lot of fun.

Yeah, I'll link that in the description. And for anyone who sticked all the way till end now, here is the bonus section. Like productivity hacks, morning night routines. So yeah, just give some like rapid fire.

In general, I'm a firm believer in moderation, so I don't do any of the extreme things. I feel like when people go down those like insane rabbit holes, they just burn out and don't have a great time. So like I work out, but I'm not a workout fanatic. I eat healthy, but I'm not a health nut. I get a decent amount of sleep But I'm not like obsessed with my sleep. I. I don't really do a ton of quantified self. I'm not obsessively tracking anything in my life. Go to the doctor, make sure I'm healthy, eat, sleep, drink some water. I don't know. I like. I quit drinking a little bit, but I don't really have a problem with drinking some alcohol every once in a while. It's not a big deal to me, but wouldn't advise anyone to drink a lot of alcohol. That's not good. But also there's a big trend of people going completely sober and I feel like that's kind of a reaction to a lot of things. So I don't know. I'm pretty moderate with email. I try and do inbox zero, try and get rid of all my, all my emails. That's a pretty old technique. If I need to manage a product, use some software I'm not crazy about. I'm definitely in the exploit phase of my tooling where, you know, in the early stage I tried all the different tools. Now I know I like and I'm getting comfortable with those so I'm explaining them. So I just use like Google Docs. I don't even use notifications. Notion. Google Docs is fine for me. I've seen Notion. Notion is probably better, but why do I need to learn anything? So I just use Google Docs and it's fine. Airtable looks awesome. I'm sure it's great. Google Sheets is fine for me. So I'm not the type of person that's bouncing around from like new thing to new thing. I feel like a lot of the problems I've seen this with like companies that scale, where a lot of people are like, we're having a problem managing this product, this project or whatever, we need a new tool for this. And then the new tool will solve this and then they go and bring in the new tool and the problem's still there because the problem was a human one. The problem was it's boring or it sucks or person A doesn't want to talk to person B so they're not communicating and there's no amount of tool that will make them like each other. So yeah, I'm not a huge believer in like always get the best tools and do that stuff. If you're in the young, if you're young and you're. You don't have familiarity with anything, go and find the best tool and then stick with that. But like Julia came out, that's A better language. Go is really good. React Native. I write Python. Like, I. It's fine. It works good enough. If I was ever in a problem, if I was ever in a place where I really didn't work for me, I'd probably consider switching. But I'm not at the phase where, like, I switch just because it's fun. But I used to. There was a time when I was, like, writing closure, which is, like, way too.

I've never even heard of that.

It's a script. It's a lisp. But it was very, like, it was very trendy to write lists because that's what PG writes. And closure was like. It was cool. It was like a cool way to think about things differently. But I'm not going to become, like, a closure guy. But, yeah. So productivity just. I'm a big believer in moderation. Moderation is not sexy, so it doesn't sell well. So you won't see a lot of people who are advertising a lifestyle of moderation. You won't see a lot of fitness influencers who are like, hey, maybe just work out three times a day or three times a week and you'll be fine. Get your resting heart rate up, make sure you're fine. Like, that'll probably be the most sustainable thing for you. They're gonna say, beyond the most aggressive diet. Work out every day. Perfection is the. But I think perfection is the enemy of the good. And you just want to. Just want to make sure you have a balanced life across everything. Don't become the insane fitness freak who is spending 10 hours in the gym and doesn't have a family and doesn't, like, can never read a book because they're working out all the time and don't do the opposite. So, yeah, moderation, that's pretty much it.

Yeah. It's been a great time talking to you.

Awesome. Always fun talking to you. Thank you.